Labour Market Policies and Poverty Reduction Strategies in Recovery from the Asian Crisis

Transcription

1 Labour Market Policies and Poverty Reduction Strategies in Recovery from the Asian Crisis Report of The ILO Japan Government of Indonesia, Tripartite Sub-Regional Seminar Jakarta, Indonesia, 29 April 1 May 2002 International Labour Organization Bangkok

2 Copyright International Labour Organization 2002 ISBN August 2002 The designations employed in ILO publications, which are in conformity with United Nations practice, and the presentation of material therein do not imply the expression of any opinion whatsoever on the part of the International Labour Office concerning the legal status of any country, area or territory or of its authorities, or concerning the delimitation of its frontiers. The responsibility for opinions expressed in signed articles, studies and other contributions rests solely with their authors, and publication does not constitute an endorsement by the International Labour Office, or by the collaborating governments or organizations, of the opinions expressed in them. Printed in Thailand

3 Preface Employment is widely accepted as one of the best routes out of poverty. As the incidence of poverty increased sharply in several East and Southeast Asian countries during the financial crisis, it is important to find ways and means of making economic recovery sufficiently employment-intensive so that it can at the same time contribute to poverty reduction. Reorienting employment and labour market policies towards achieving this goal would be important. Labour market policies can also be geared towards facilitating the process of restructuring and reforms that would be part of the recovery process itself. The aim of this Seminar was to assess whether economic recovery in East and Southeast Asia is being translated into gains in the labour market so as to contribute to poverty reduction. Moreover the Seminar investigated how the more affected regions within the countries are responding in terms of economic recovery, employment expansion and poverty reduction. On the basis of discussions among the participants of the Indonesian experience and the experience of their own countries, the Seminar examined how to enhance the role of labour market policies for employment and poverty reduction as well as strengthen the institutional capacity for more effective labour market policies. Amongst the main conclusions arising from the Seminar, a few might be worth mentioning. These include, the need to adopt a forward-looking approach and develop preparedness to face the social dimensions of economic crisis and capacity to respond rapidly and effectively, especially given the frequency of sharp economic downturns. The Seminar highlighted the need to integrate employment concerns into overall policies especially during economic crisis and recovery. It was also concluded that active labour market policies can play a significant role in responding to the adverse social effects of economic crisis as well as when improvements in labour markets lag behind economic recovery. In addition a strong labour market information system can provide support to the task of monitoring the social dimensions of crisis and recovery. The issue of tripartite dialogue mechanisms was repeatedly raised as critical in facilitating the process of responding to the social dimensions of economic crisis and strengthening economic recovery. I would like to take this opportunity to thank all the representatives from the governments, the employers and workers organisations of Indonesia, Japan, Malaysia, Philippines, Republic of Korea and Thailand, and the Japanese government for providing financial support under the ILO/Government of Japan bilateral programme. I would also like to acknowledge the contribution of the Recovery and Reconstruction Department at ILO Headquarters Geneva in providing technical support and the ILO Area Office in Jakarta for assistance in organizing the meeting. Last, but not least, I should like to thank the Indonesian Ministry of Manpower and Transmigration for hosting the event. Yasuyuki Nodera Regional Director, ILO ASIA Pacific Region, Bangkok.

4 CONTENTS Part I Proceedings of the Seminar Background Inauguration Session Technical Session I Recent Trends in Growth, Employment and Poverty in the Sub-region Technical Session II Employment-Poverty Linkages in the Context of Indonesian Crisis and Recovery Technical Session III Sub-national Differences in the Context of Indonesian Crisis and Recovery Technical Session IV Strengthening the Labour Market Information System in Indonesia Technical Session V Policy and Programme Approaches to Promote Employment and Reduce Poverty during Recovery from the Crisis in East and Southeast Asia Working Group Discussions Closing Session

5 Part II Papers Presented at the Seminar 1. Rizwanul Islam and J. Krishnamurty Employment, Labour Market and Poverty during Economic Recovery and Slowdown in East and South East Asia 2. Iyanatul Islam Poverty, Employment and Wages: An Indonesian Perspective 3. Diah Widarti A Study on Sub-National Differences in Employment, Income and Poverty Levels in the Context of Economic Crisis and Recovery in Indonesia 4. Shafiq Dhanani Strengthening the Indonesian Labour Market Information System Annex 1: List of Participants Annex 2: Seminar Programme

6 PART I PROCEEDINGS OF THE SEMINAR

7 Background The Seminar, which was supported by the Japanese government and attended by tripartite delegations from Indonesia, Japan, Republic of Korea, Malaysia, the Philippines, and Thailand, was a continuation of previous workshops held in Tokyo in 1998 and Manila in The major purpose of the seminar was to enhance understanding of the impact of the Asian crisis on poverty and in particular to identify solutions within a labour market policy framework. In this context, the seminar aimed to identify ways and means of strengthening labour market policies, poverty reduction approaches, and employment promotion efforts, in order to achieve more sustainable economic and social development. The Seminar aimed to address the following questions/issues: (i) Is economic recovery in the East and Southeast Asia region being translated into gains in the labour market and is it contributing to poverty reduction? How are the more affected regions within these countries responding in terms of economic recovery, employment expansion and poverty reduction? (ii) Taking into account the experience of the East and Southeast Asian region, how can the role of employment and labour market policies for poverty reduction be further enhanced? (iii) How can the institutional capacity for more effective labour market policies be strengthened?

8 Inaugural Session The inaugural session began with opening statements by Mr. Alan Boulton (Director of ILO Area Office of Jakarta), Mr. Yasuyuki Nodera (Regional Director of the ILO Bangkok) and Mr. Makato Ogawa (representing the Japanese government), followed by an opening address by Mr. Yacob Nuwa Wea, Minister of Manpower and Transmigration, Indonesia. In his statement, Mr. Alan Boulton welcomed the Minister of Manpower and Transmigration, the Regional Director of ILO ROAP and the participants. He expressed his gratitude to the Japanese government for the support given to this Seminar and to the ILO Regional Office of Asia Pacific for organising the Seminar. In his speech, Mr. Yasuyuki Nodera welcomed the Minister of Manpower and Transmigration and the representatives of the Japanese government, as well as all the participants. Moreover, Mr. Nodera thanked the Japanese government for the financial support to the Seminar and explained the purpose of the Seminar as a continuation of previous seminars conducted in Tokyo and Manila. Mr. Makato Ogawa, representing the Japanese Government, expressed his sincere gratitude to the Minister of Manpower and Transmigration for hosting this important seminar. He also thanked the International Labour Organization for its professional work in preparing all aspects of the meeting. He stated that the project has demonstrated a growing collaboration between the Ministry of Health, Labour and Welfare of the Japanese government, the ILO and the World Bank aiming to improve economic and social conditions in East and South-East Asia. In his statement, Mr. Ogawa noted that this Seminar was the follow-up activities of the Tokyo Seminar held in 1999 and the Manila Seminar of The Tokyo Seminar was held in the midst of the Asian financial crisis when unemployment and poverty increased while real wages decreased, creating social tension and unrest. This hardship raised pressing short-run concerns about job creation and poverty alleviation, whilst highlighting the need to deal with longer run concerns related to the development of labour market institutions and policies and to human resource investments. Moreover, the crisis taught us that relying solely on economic growth, as the primary safety net is not

9 enough to ensure sustained poverty reduction. As economic recovery takes hold in the region, it is important to evaluate economic outcomes but also equally important social outcomes. This is the aim of this Seminar. The strength of this Seminar lies in the combination of country-based experiences, cross-country analysis and the working group discussions scheduled towards the end of the Seminar. Mr. Makato Ogawa also provided a brief description of the institutional background of labour market policy in Japan. Labour market information is provided through the internet and there are about 12,000 public employment service staff and 587 employment service offices. Public employment offices are the centre of the employment policies providing placement services, unemployment benefit and other grants for job creation. The institutional integration of the employment service, unemployment insurance service and other ALMPs enhances the efficiency of labor market policy. In 2001, the unionization rate was 20.7%, although this rate has been decreasing since In Japan, labour laws are discussed in the tripartite advisory committee before deliberation by the Parliament. Furthermore, the Japanese Government has played a significant role in assisting countries of East and South-East Asia to respond to the crisis. Since the crisis, the bulk of grants and loans have been in support of safety nets, including the maintenance of social expenditures in the wake of the crisis. Moreover, the Ministry of Health, Labour and Welfare is providing technical support in labour market policies by sending technical advisors to the East and South-East Asia region. Finally, Mr. Ogawa stated that the Japanese Government wishes to continue working with the ILO. The Indonesian Minister of Manpower and Transmigration officially opened the Seminar, welcoming the distinguished representatives of the Japanese government, the regional director of ILO Bangkok, the director of ILO Jakarta office and the participants from Japan, Republic of Korea, the Philippines, Thailand, Malaysia, and Indonesia. Mr. Nuwa Wea stated that this Seminar would provide an opportunity to identify the current progress or shortcomings in efforts to improve employment and reduce poverty. He also expressed his appreciation of the topic of the Seminar in view of its relevance to the current economic and manpower situation in Indonesia.

10 Mr. Nuwa Wea added that since the economic contraction of 1998, Indonesia s economy has not recovered to pre-crisis levels. Companies have reduced their production, thus negatively affecting the employment level. In response to the crisis, the Ministry of Manpower and Transmigration has simultaneously launched both rescue and rehabilitation programmes. For the rescue programme, the Government allocated 1.5 trillion rupiahs for labour intensive projects in the forestry sector, urban and rural areas, and skilled labour improvement projects including the promotion of small and medium enterprises. Rehabilitation programmes consisted of improving the labour market information system, employment services and crisis centres in addition to promoting entrepreneurship and agro-business. The current priority for the Ministry of Manpower and Transmigration is to improve the overseas employment management programme, which has potential to reduce unemployment. A number of activities, addressing amongst others the private recruitment agencies, are now aiming at giving better protection for Indonesian migrant workers. Technical session I Recent Trends in Growth, Employment and Poverty in the Sub-region In his presentation (section 1-3 of R. Islam s and J. Krishnamurty s joint paper), Mr. Riswanul Islam described how the economic slowdown of the major engines of growth (USA, European Union and Japan) in 2001 caused fluctuations in the global economy and in particular after the Sept. 11 th attacks. This downturn has caused setbacks in Asian economic recovery, weakening the recovery momentum of 1999, and has had severe consequences on employment and consumption poverty. The countries of East and Southeast Asia are facing problems of decreasing global export demand. Moreover, the fluctuations have had severe repercussions on underemployment, poverty and living standards, leading to social tension. The labour markets have been adversely affected adjusting mainly through increased

11 unemployment, especially declining manufacturing employment, leading to sharp increases in poverty and declining real wages. During the discussion, a number of participants suggested looking into some additional factors affecting economic recovery of the region. For example, demographic factors might be examined. Again apart from the USA, European Union and Japan, China could also affect the economies of Southeast Asia and the Pacific. AFTA (ASEAN Free Trade Agreement) which will come into effect in the near future may enable Southeast Asian countries to be less dependent on USA, EU, and Japan. The question of social protection for the informal sector workers was also raised during the discussion. Other comments related to the non-economic situation. A view was expressed that in Indonesia, the transition to democracy is yet to pass through its consolidation period. It was also stated that there were the issues of a lack of law enforcement, and the lack of a well-established dialogue between workers representatives and the employers. In response, Mr. Riswanul Islam expressed his agreement with the importance of demographic factors on the labour supply and the employment situation, and acknowledged that the analytical framework presented does not specifically cover such demographic factors. However, ILO does consider population as a critical factor. Mr. Islam appreciated the comments related to non-economic factors, which can become particularly important in the context of social dialogue, at least when judging from the Republic of Korea s experience in dealing with the process of democratisation. With regards to AFTA, Mr. Islam agreed that such an agreement will be important for the region s economy in the future. He also proposed a close examination of the trade and other economic relationships between China and the Southeast Asia and Pacific economies in order to better understand the role of China in the region. Finally, Mr. Islam clarified that the lack of discussion on the issue of social protection could be attributed to the fact that this issue had been extensively examined in the previous seminar held in Manila in 2001.

12 Technical Session II Employment-Poverty Linkages in the Context of Indonesian Crisis and Recovery Mr. Iyanatul Islam s presentation focussed on employment-poverty linkages in Indonesia in the context of the Asian crisis and discussed a number of possible solutions to the poverty problem. The presentation illustrated that by using a modified poverty measurement, the incidence of poverty in Indonesia is found to be much more widespread than had been initially thought. For instance, indicators of capability deprivation (malnutrition, literacy, access to health services and low life expectancy) show that in 1999 more than 50 percent of Indonesians did not have access to health services, while the incidence of child malnutrition was 30 percent. Furthermore, vulnerability is more widespread than poverty and needs to be urgently addressed. Indicators of powerlessness suggest that poverty has deep-seated social political and institutional roots: 34 percent of Indonesians are vulnerable to episodes of poverty and this type of poverty worsened more sharply then consumption poverty during the crisis. Indeed, even those whose expenditure is 20% above the International Poverty Line (IPL) are vulnerable to an episode of poverty. From 1990 to 1996, Indonesia recorded significant urbanization and formalization of the work force, with the proportion of the urban workforce increasing from 26% to 36% and the share of the formal sector rising from 28% to 38%. Rapid growth in the pre-crisis period had led to a deepening of skills of the workforce, indicated by a decrease in the proportion of less educated/skilled workers from 87% in 1990 to 79% in Since 1990, the manufacturing sector has played a more significant role in employment creation by drawing less-skilled workers from the rural and informal economy into the formal economy. This rapid growth in the pre-crisis period brought about rising real wages. From , real wages grew 5% per annum on average, while the proportion in poverty declined by 29%. The crisis wiped out several of the major gains achieved during the rapid growth period in the pre-crisis era: poverty proportions shot up during the peak of the crisis, real wages collapsed by 40% in 1998 and remained 10% below their pre-crisis level in

13 2000. In fact one recent study claims that currently as much as 50% of Indonesians are poor. There was involuntary reallocation of labour to the agricultural sector, and an enlargement in the size of the urban informal sector. Mr. Iyanatul Islam emphasized a number of key elements necessary to deal with poverty and employment. Firstly, considering the importance of macroeconomic policies in stimulating economic growth it is crucial that employment concerns are integrated into policy making. Secondly, there is a need to create enough durable jobs annually to absorb new entrants to the workforce and clear the backlog of the unemployed and the underemployed. Secondly, labour empowerment is a necessary ingredient for the workers and the poor to cope with the problem of vulnerability. Thirdly, and integral to the above, is a credible industrial relations system which represents the interests of both formal and informal sector (or marginalized) workers and thus contributes to alleviating the sense of voicelessness and powerlessness. The presentation led to several comments and observations. A number of participants raised issues concerning the problems of protection of the large unorganised and marginalized labour force, which is outside the reach of labour standards and coverage of social insurance. The modern industrial sector, it was felt, could not absorb much labour. Participants also noted the existence of large numbers of Indonesians in Malaysia, often facing poor terms and conditions of work and not well informed of their rights. Furthermore, such workers receive little help from the trade unions and the government. One criticism of the paper was its neglect of the reality of a large number of domestic helpers that are unprotected by the definition used in the paper. The participants also agreed on the importance of social dialogue in giving voice to the unprotected labour force in the informal and marginalized sector. Moreover, the participants noted that education and skill enhancement are as important as giving protection to those in the informal and marginalized sector. Based on their country experiences, several participants felt that a meaningful dialogue between government, employers, and workers was required when a crisis hits the region, and to some extent this has happened in Malaysia, where there is now a closer relationship amongst the tripartite elements. Thus, the Malaysian workers now have more involvement in all stages of the decision-making processes. Similarly,

14 in Thailand and the Republic of Korea, tripartite dialogue has been strengthened. However, in Indonesia the view was expressed that the dialogue mechanism has yet to be strengthened, along with the improvement of industrial relations. Mr. Iyanatul Islam responded that social protection is particularly important for the excluded majority in Indonesia and agreed that enhancing the skills of these groups is important as well. He emphasised the need to broaden the social dialogue instrument in Indonesia providing the labour force (including the excluded groups) the means to voice their rights, a critical factor for healthy industrial relations and eventually for poverty reduction. In his view, it was critical to have a coalition between the trade unions and civil society in order to empower the workers in the informal sector. The crisis gives Indonesia an opportunity to learn from the experiences of the Republic of Korea, Malaysia, and Thailand in dealing with early stages of democratisation which, despite seeming somewhat chaotic at the beginning, could create a trust and meaningful co-operation in the end. The participants also mentioned their country experiences with regards to training, education and small & medium enterprise (SME) promotion. The Republic of Korea is currently promoting an active labour market policy, providing training and retraining for workers. Malaysia provides grants for young people to take courses in mathematics, English, and a number of vocational training courses. Furthermore, support for SMEs in the form of soft loans is given in Thailand, Malaysia, and the Philippines acknowledging the potential role of SMEs in job creation and reduction of dependence on foreign direct investment (FDI). Concerns were expressed regarding having an employment strategy that is overreliant on cheap labour. There was the danger of relocation as wages began to rise as in the case of Malaysia, where as a result many of the IT industries migrated to China. Despite some caution with the fact that a number of countries may not be ready to take such a route of technology and capital-based development, participants agreed not to overlook technology and international trade issues in employment policy. Some participants stressed the need to consider the international context and link Indonesian issues to the rest of ASEAN, especially in respect of labour migration.

15 Technical Session III Sub-national Differences in the Context of Indonesia Crisis and Recovery In her presentation, Dr. Diah Widarti focused on the impact of the financial crisis and the adjustment processes in different regions of Indonesia especially on the employment, income, and poverty situation. Overall, the country faced the financial crisis at a time when it was already experiencing dwindling exports, a slowdown in overall economic growth, and sharply reduced rice production leading to imports of rice and prospects of a sharp rise in food prices, not to mention the much needed political changes that were sweeping the country. Economic growth in Indonesia has been heavily concentrated in parts of Java (centred on Jakarta) and Sumatra (centred on Riau). Provinces located outside Java are mainly dependent on the agricultural sector, whilst the majority of provinces in Java are more heavily dependent on the manufacturing sector. A few provinces however have different patterns. In Aceh and Riau for example - the mining and oil sectors constitute an important component of GDP. The data also demonstrates that the growth of per capita GDP varied among regions, but in 1996 almost every province showed relatively high growth. In only a few provinces such as Bengkulu, Southeast Sulawesi and Aceh per capita GDP was low. Briefly, in 1996 the growth of regional per capita GDP in all provinces was still positive. Moreover, in recent years, Jakarta and Riau have been pools for population migration. Due to its prominent mining sector, East Kalimantan has also received an influx of people from its border regions. Indonesia s weak and varied economic recovery can be attributed to several factors. These included: regional differences in initial/baseline conditions, economic effects, and subsequent recovery processes as well as social dimensions of employment, income, and living standards, particularly within the poorer and vulnerable segments of the society. There were also considerable regional differences in history, ecological conditions, settlement and migration patterns, production conditions, political and social structure, and economic performance levels.

16 The presenter also drew attention to an increased rate of unemployment from 5.0 percent in 1996 to 6.4 percent in Given the structure of employment in the country, sectors like agriculture, trade, transport, and other informal economic activities had substantially higher rates of underemployment. With regard to the impact on poverty levels, the scenario has been loss of jobs and labour migration to low-productivity work, yielding lower incomes. The classic mechanism of high food inflation leading to an increase in poverty appears to be have worked already, thus implying a significant addition to the already 35 million people in the country recorded below the poverty line in Several participants shared the experiences of their respective countries. Whereas in Indonesia, the rural sector was badly hit by the crisis, in Malaysia, the urban sector was severely affected following the downturn of the construction and manufacturing sectors. In the Republic of Korea, the crisis occurred shortly before the election, so the elected government focused on banking restructuring, together with a social safety net programme and most importantly initiated tripartite dialogue to reach consensus. In the Philippines, investment in the real estate sector came to a halt. The participants suggested that the paper should also consider the fact that Indonesia was faced with a multi-dimensional crisis: economic downturns in the time of political transition. Another issue being discussed was the phenomenon of international migration including the undocumented (illegal) migration from Indonesia to Malaysia, which has been somewhat important in terms of economic benefits at national and sub-national levels. While it is difficult to find solutions in the near future, the participants tended to believe that international migration should be handled in a way so that it benefits the involved countries. It was suggested that the paper might have used post-1999 data. Fortunately, recent data on sub-national level would be available in the near future. With regards to questions on domestic savings, it was explained that Indonesia s domestic savings reached 24 percent of GDP before the crisis, and declined to 17 percent during the crisis. Recently, the savings level has increased to 20 percent. This is in contrast to Malaysia, where the level is around 40 percent, which may have made her much less dependent on international assistance.

17 Technical Session IV Strengthening the Labour Market Information System in Indonesia Mr. Shafiq Dhanani began his presentation by stating that Active Labour Market Policies (ALMPs) and an effective Labour Market Information System (LMIS) are necessary to ensure that analysts, job seekers, employers and training providers have access to up-to-date information on the supply and demand for different types of labour, thereby leading to better matching of skills and opportunities. Mr. Dhanani identified four main users of LMI: (1) job seekers and students or trainees, (2) firms and others employers, (3) education and training providers, and (4) policy makers and labour market analysts. In his opinion, labour market information for the first three groups of users are far from adequate. Labour market information mainly addresses the needs of policy makers and labour market analysts. In fact, whilst labour market information has been targeted at policy makers on a relatively regular basis, the utilization of the existing data for policy-making has lagged behind. Trends in unemployment and employment have been available on a six-monthly basis. Trends in earnings of agricultural workers have been available on a monthly basis, including a monthly indicator of the cost of living, allowing for the monitoring of real wages. Earnings of manufacturing workers in medium-large scale industries have been available on a quarterly and annual basis. Earnings for most sectors now have been available on a six-monthly basis. While there is always scope for improving the frequency and coverage of labour market information, policy makers, especially in the Ministry of Manpower and Transmigration, have not made adequate use of the existing labour market information. Policy makers could have used the existing data far more effectively for negotiating the minimum wage with the employers and workers, for education and training planning, and for monitoring living standards, particularly during the crisis. Furthermore, the national labour force survey Sakernas was able to track a number of important labour market changes before and during the crisis, particularly in relation to the labour market supply, open unemployment, discouraged workers, under-

18 employment and the employment structure. A careful analysis of existing data was able to reveal a number of real changes in the labour market situation by gender, urban-rural location, age and region. Thus, there are number of weaknesses and critical gaps in the present Indonesian LMIS. The Central Bureau of Statistics (CBS) has not been aggressive enough in publicizing its large and valuable sources of labour market data to the four sets of users of LMI. In the same way that CBS disseminates consumer price index (CPI) data in press conferences, Ministry of Manpower and Transmigration (MOMT) should hold regular press conferences distributing press briefings to disseminate key labour market information on issues such as unemployment and quarterly and annual wages. The speed of processing, and publication of the annual socio-economic survey Susenas and quarterly market wages should be increased to provide more timely indicators of poverty incidence and market wages. Dissemination, information flows and data utilization have room for improvement as well. Moreover, the nation-wide network of public employment services should be re-oriented from data collectors to service providers for job seekers, trainees, retrained workers, employers and firms. Public employment services needed to strengthen their service orientation. Finally, Mr. Dhanani highlighted the dangers of decentralised data collection, which had been proposed in the current euphoria of regional autonomy. The participants were in general agreement with the presenter and suggested additional factors affecting quality of labour market in the country. For example, it is compulsory by law for a company to report vacancies but enforcement is poor. Furthermore, following regional autonomy measures, regional governments are not obliged to report to the national government. This creates problems of data availability and the CBS faces budget shortages. The discussion moved into exchange of experience across countries. In Malaysia, labour market information (LMI) has received higher priority since the crises. The government has invested in strengthening LMI infrastructure including information technology. Moreover, data on salary and fringe benefits in the private sector are now available on CD-ROM. Like in Indonesia, in Malaysia there is a lack of information on vacancies, as it is not compulsory for companies to report them.

19 In Thailand, the government is now trying to improve timeliness of labour market information by conducting labour force surveys every month instead of three times a year, but this has been traded off with a reduction of sample size. Other measures include extending data linkages amongst government and non-government organisations in order to better forecast the economic situation. There is scope for the government to improve its monitoring capacity and win the trust of the public. Similarly, as in Malaysia, the government does not compel companies to report vacancies. At present, the Ministry of Labour relies on its employment service agents throughout the country 5 in Bangkok and 75 outside the city to obtain LMI. The Philippines had a manpower registry service. The statistical system had been decentralised but was centrally coordinated. Quarterly labour force surveys were conducted. The importance of estimating invisible unemployment and the labour market situation of domestic helps were stressed. In the Republic of Korea, a private initiative exists to institute on-line job seekers and vacancy information. In fact, it is much cheaper for the employers to post vacancy on the internet than in the press. In Japan, the Ministry of Labour and Health and the Ministry of Social Affairs collect labour statistics. Japan Industry and Labour compile the reports from all relevant ministries, and disseminate these throughout the country. In his response, Mr. Shafiq Dhanani noted that some countries faced problem of lack of vacancy data in their LMI. In Indonesia, public agencies of employment service have turned into data collectors and performed very limited job canvassing, somewhat neglecting their mandate as job service providers. In addition, public employment agencies have failed to update their data on job seekers and vacancies. This has worsened LMI for job seekers (students, high school leavers, training institutes).

20 Technical Session V Policy and Programme Approaches to Promote Employment and Reduce Poverty Judging from the experience of the countries of East and Southeast Asia in terms of periodic fluctuations, Mr. Krishnamurty (section 4 of R. Islam s and J. Krishnamurty s joint paper) pointed out the importance of preparedness and adoption of policies and programmes that enable a quick response to crisis. He highlighted that a revival in economic growth does not automatically lead to an improvement in the employment, and the labour market and poverty situation. A rapid and effective response mechanism must include a combination of effective macroeconomic policies and active labour market policies (ALMPs). This combination would promote sustainable growth of output as well as employment and income growth, minimize the impacts of the crisis on employment and directly alleviate the distress of the most affected groups. ALMPs have been an important component in the response to the Asian economic crisis and the recent economic slowdown. ALMPs include training and re-traning policies and programmes, labour market information services, credit and livelihood promotion programmes, and special employment or public works. Through effective ALMPs the labour market will function better for workers in large-scale establishments, in the informal sector, in self-employment as well as for those in intermittent or casual employment including those subject to poverty. Furthermore, a monitoring system is required to identify the impacts of policies and programmes on the economic condition of vulnerable and poverty affected groups. Monitoring is necessary to provide early warning indicators. In examining policy options, Mr. Krishnamurty emphasised the importance of considering both short and long-term impacts of macroeconomic policies (in particular impacts on employment and labour market, poverty and inequality). Due to their quick impact and strong multiplier effects, expansionary fiscal policies have played a particularly important role in East and Southeast Asia in dealing with the slowdown. And although the countries of East and South East Asia had budget

21 deficits at the time of the crisis, there was room for such expansionary policies. Moreover, due to the important role of SME s in employment creation and income distribution, it is essential to create a favourable environment for the development of these enterprises including their access to credit/financial assistant. The participants raised several key issues. These covered such issues as social and political risk and moral hazard in macroeconomic policies, and the question of good governance. Policy prescription is only one aspect; the implementation of policies or programmes constitutes another important aspect. One participant asked whether the underemployed or the unemployed should be given higher priority. Furthermore, the participants noted that some countries like Indonesia face economic and political crisis, let alone ethnic conflicts in several areas. Thus, policy prescription will need to incorporate these problems in the solution package. It was pointed out that output growth might lead to employment growth though there may be a time lag. Another participant pointed to the importance of preparedness. The experience of earlier programmes was often lost. Moreover, funds had to be managed in such a way that rapid response was facilitated. The presenter agreed that governance is critical in this respect. In fact, it is essential that different groups of employers, workers, and government reach agreement on policy options. With regard to the issue of quick disbursement of funds, the presenter maintained that there should be some flexibility in national budgeting procedures allowing for the possibility of having rapid response funds to meet unforeseen contingencies. In response to the remarks made by some participants, the presenter also agreed that more resources are necessary for ALMPs. This might require more government spending and more taxation to finance training and re-training programmes.

22 Working Group Discussions During the third day, the seminar took the form of working group discussions on the following issues: Employment and poverty reduction during economic crisis and recovery: lessons learnt and strategic directions for the future. Employment-friendly growth for poverty reduction during economic downturn and recovery: policies and programmes. Strengthening labour market information systems to make them more responsive to economic fluctuations. Group A: Employment and Poverty Reduction during Economic Crisis and Recovery: Lessons Learnt and Strategic Directions for the Future The discussion in this group focused on what happened during the crisis, lessons learnt from the crisis, and strategic directions. The participants identified a number of effects of the crisis. There was drastic depreciation all the countries that had their national currencies pegged to the US dollar suffered drastic depreciation in their currencies. All countries experienced massive capital outflows, followed by massive lay offs and retrenchments of workers and increased unemployment because of economic contraction in varying degrees. Moreover, the banking system was badly hit by high non-performing loans (NPL) and many countries were faced with social and political instability. The participants found that the governments were heavily preoccupied with banking restructuring and fiscal policy in responding to the crisis, while employment concerns tended to lag behind, with the exception of a few training and re-training programmes. The participants also felt that a sustainable economic growth rate of about 5 percent would be necessary to create sufficient employment. Support to micro enterprises, because of their evident role in job absorption, was repeatedly suggested.

23 Furthermore, in order to withstand shocks, market economies must learn how to adjust to trade liberalization and intensified competition in a technology driven global market. Within ASEAN, every member country must strive for cooperation in adjusting to the impacts of the implementation of the AFTA-CEPT. Meanwhile, the issue of cross-border labour migration must be handled in such a way that both the sending and receiving countries benefit from the exchange. The tripartite dialogue mechanism was debated in Group A, in particular in relation to NGOs. The participants found that tripartite dialogue gains momentum, as the dialogue becomes an instrument enabling the countries to reach a broadened commitment to the policy and programme. The most debated issue concerned the role of NGOs and civil society organizations in general. The participants agreed that NGOs should not intervene with established tripartite institutions. However, they acknowledged that the existing tripartite partners represent the formal sector, which constitutes only a small fraction of the entire economy. By representing workers in the unorganised and informal sector, NGOs may become important dialogue partners. In this respect, the participants envisaged the promotion of a broadened dialogue forum where all relevant social partners participate in formulating a consensus to deal with recovery. Such a strategy has already been initiated in Indonesia with an informal forum called tripartite plus. In Thailand, a committee consisting of various parties was established to generate a commitment to creating employment for all. For the developing economies which have large groups of informal sector workers, policy and resources must be formulated and mobilized in order to cover these workers most of whom are marginalized and poor with safety net protection. The participants agreed that it is urgent to strengthen the SMEs considering their role in employment creation and infusing economic dynamism (especially in the countryside). Thus, a set of policies should be designed enabling SMEs to get easy access to credit and markets. Employers can help by establishing linkages with them through contracting and franchising along the supply chain. Furthermore, inflation

24 must be managed at relatively low levels in order to benefit both employers and workers and at the same time contribute to competitiveness. The participants also mentioned the need to develop an early warning system to monitor the recovery processes and to detect possible crises created from shocks or economic downturns. Another aspect discussed amongst participants was the role of international institutions during the recovery processes. The participants acknowledged that the IMF created some positive impacts especially with regards to bank restructuring, yet the degree of this contribution to overall economic recovery was difficult to single out. However, a number of IMF recommendations relating to job retraining and security, and assets selling with considerable discount were not found useful. Many participants maintained that the international institutions should be more sensitive to specific conditions and difficulties faced by each country. Group B: Employment-friendly Growth for Poverty Reduction during Economic Downturn and Recovery: Policies and Programmes The discussion of Group B focused on three major issues: a) conduciveness of macroeconomic and sectoral polices to employment-friendly growth, b) actions needed to strengthen ALMPs, and c) experience of labour market programmes. With regard to the first issue, the participants stressed that: SMEs role in employment creation should not be underestimated. It was necessary to promote SMEs and create an enabling environment for foreign direct investment (FDI). Monetary and fiscal policies should be geared to create demand and should be designed in consultation with social partners. Tripartite dialogue should be part of the dialogue package, especially for its potential contribution to healthy industrial relations, political stability, and eventual economic recovery. Participants referred to the experiences of the Republic of Korea, Malaysia, and Thailand.

25 The macro policy framework should integrate skill enhancement and productivity improvement. Several steps were proposed by the participants to strengthen active labour market policies. A sound support of labour market information, including monitoring and evaluation systems, and an IT-job net in collaboration with other institutions were raised repeatedly, together with the need for closer coordination between the government and the private sector. As in Group A, the participants brought up the importance of tripartite mechanisms as an essential ingredient in strengthening labour market policies. Training and education were common policies in response to the crisis. In Malaysia, the government set up a Human Resource Development Fund, a training grant for workers and those who were going to be retrenched in order for them to obtain new skills and enhance their employability. Training grants have also been given to unemployed graduates and school-leavers. In the Philippines, the government provided training for workers in the informal sector and rural areas aimed at improving their wages and their potential to become self-employed. The Ministry of Industry, Republic of Korea, provided financial assistance to IT venture capital, which had the potential for employment creation. In Japan, there exists an employment insurance system (EIS) and training for workers through the EIS fund. Group C: Strengthening Labour Market Information Systems to Make Them More Responsive to Economic Fluctuations With the exception of Japan, the participants identified gaps in the existing LMI in all countries, in terms of meeting the needs of users as well as in terms of the speed at which the data becomes available for the users. There was general agreement that there should be some improvement in employment data (including unemployment and underemployment) in terms of sample size, dissemination, and the speed of the existing LMI. In this respect, the participants also brought up the problem of reliability of data concerning workers in the informal sector. Another issue related to the heavy focus of LMI for policy-making purposes, rather than for employers and job

26 seekers. Furthermore, the problem of coordination between the statistical office and the labour ministry was mentioned repeatedly. Related to this issue was data duplication and inconsistency between the two organizations. While acknowledging budget constraints, it was also obvious that the IT aspect of LMI has received little attention. A number of suggestions came up with regards to improving LMI. The participants mentioned the possibility of wider data dissemination, including a regular newsletter on employment issues. Furthermore, data on vacancy and other information on students and high-school leavers were required for the purpose of career counselling. Information on informal sector needed considerable improvement. Better coordination among the relevant organizations was also stressed. Data on undocumented migration needed to be developed. The participants also suggested that the statistical office or labour ministry should consult the users or tripartite elements in designing or revising the questionnaire. Closing Session The closing session began with some concluding remarks by Mr. Rizwanul Islam. He pointed out that the concluding remarks were not meant to substitute the thoughts and opinions expressed by the participants. Furthermore, the purpose of the discussion, which took place during the seminar was simply to identify problems and potential solutions, not to reach a broad-based consensus. With these qualifications, Mr. Islam presented the following points which, he felt had emerged from the discussions of the Seminar. 1. By the year 2000, recovery seems to have been proceeding well in the countries that were affected by the Asian economic crisis of Nevertheless, as the global economy went into a recession in 2001, the recovery of crisis-affected Asian economies also suffered a setback, thus showing both the positive and negative aspects of globalisation.

27 2. The economic crisis of had severe adverse social consequences in terms of increases in unemployment and underemployment, fall in real wages, and increases in poverty. And the economic recovery during did not bring about a full recovery in terms of social dimensions. 3. Unemployment rates declined in , but remained well above their pre-crisis levels. In Indonesia, unemployment continued to rise even during the recovery period, although the incidence of poverty returned to its pre-crisis level. In other countries (e.g., Philippines and Thailand), the incidence of poverty did not register any decline. 4. In some countries, a good deal of the new employment generated was of the non-standard variety i.e., casual, temporary, and part-time. 5. It is important to monitor the social dimensions i.e., employment, labour market and poverty aspects, not only during a crisis but also during economic recovery. Particular attention needs to be paid to problems relating to the informal sector. 6. Given the frequency of sharp economic downturns, it is necessary to adopt a forward-looking approach and develop preparedness to face the social dimensions of economic crisis and capacity to respond rapidly and effectively. 7. The first precondition (although not sufficient) for generating employment and improving the labour market situation is to put the countries back on the path of high growth on a sustained basis. Moreover, to achieve that objective it is essential to put in place a conducive-environment for investment. 8. Recognizing the centrality of employment in reducing poverty, it is important to integrate employment concerns into overall policies especially during economic crisis and recovery. A consensual approach to policy making needs to be followed.

28 9. While macro-economic stability is essential from the point of view of achieving sustained growth, during periods of sharp economic downturns counter-cyclical economic policies are needed in order to avoid deterioration in the employment and poverty situation. 10. Active labour market policies (consisting of training and retraining of workers, employment services, and job creation programmes like promoting the growth of micro and small enterprises and labour based approaches in infrastructures) can play a significant role in responding to the adverse social effects of economic crisis as well as when improvements in labour markets lag behind economic recovery. Capacity to formulate and implement such programme needs to be strengthened. Indeed they can play the role of safety nets during a crisis. Thus safety nets do not need to be entirely welfare oriented. 11. Given the significant degree of cross-border movements of labour across the region and the impact of economic crisis and downturn on such movements, it is essential to look at the interaction between the labour markets of the various countries and work towards maximizing the benefits of such labour movements for all partners. 12. Closer cooperation, partnership and dialogue among governments and employers and workers organizations can facilitate the process of responding to the social dimensions of economic crisis and strengthening economic recovery. 13. A strong labour market information system can provide strong support to the task of monitoring the social dimensions of crisis and recovery. It is necessary to improve the processing, publication, dissemination and utilization of labour market data already collected (e.g., data on unemployed workers, cost of living index, etc.). Employment services need to be improved by making them more service oriented. Career guidance, counselling and job matching should from important components of such services. The efficiency and effectiveness of employment services can be enhanced through the use of modern technology.

29 Resources allocated for LMI system should continue to receive high priority even when budget cuts are warranted during a crisis. Following Mr.Islam s remarks, Mr. Alan Boulton - Director of the ILO Jakarta Office and Mr. Djoko Sidik Pramono, the Secretary-General of the Ministry of Manpower and Transmigration made their closing remarks. In his statement, Mr. Alan Boulton expressed his appreciation to the Japanese government for its support as well as to the contributions made by all participants during the Seminar. He noted that some issues had been repeatedly raised: tripartite dialogue mechanisms as a critical way to find solutions, and the need of social protection for the workers in the unorganised and informal sector. In that context, he acknowledged the achievements made by the Indonesian government, and expressed ILO s willingness to assist in improving industrial relations and labour rights in Indonesia. In his closing remarks, Mr. Djoko Sidik Pramono expressed his deep appreciation of the Seminar topics, in particular due to their relevance to Indonesia s efforts in improving linkages between labour market information system and overseas and domestic employment. Moreover, he thanked the participants for inputs during the Seminar, which would remain valuable reference for all participants with regards to national action programmes. The Secretary-General also took the opportunity to update the participants with regards to the plan to improve overseas employment management, which has received full support from the President of the Republic of Indonesia. On behalf of the government, the Secretary-General finally stated his appreciation to the ILO, the Government of Japan, and especially all delegations attending the seminar for their contribution and deliberation.

30 Part II Papers Presented at the Seminar

31 EMPLOYMENT, LABOUR MARKET AND POVERTY DURING ECONOMIC RECOVERY AND SLOWDOWN IN EAST AND SOUTH EAST ASIA BY RIZWANUL ISLAM & J. KRISHNAMURTY EXECUTIVE SUMMARY The Asian economic crisis of and the global economic slowdown of 2001 (which was exacerbated by the events of 11 th September 2001) serve to indicate that market economies continue to remain vulnerable to fluctuations. Such sharp fluctuations in economic activities can have serious adverse social consequences in terms of higher unemployment and under-employment, erosion of gains made in reducing poverty, and increased hardship for the poor. The purpose of the paper is to see how the employment, labour market and poverty situation in the countries of

32 Indonesia, the Philippines, Malaysia, Republic of Korea and Thailand (ESEA) developed during the brief period of recovery and the subsequent slowdown (in 2001), and to discuss policies needed to respond to the adverse social effects of economic downturns. In the aftermath of the Asian economic crisis, the countries of ESEA were able to attain economic recovery fairly quickly. In fact, the rates of economic growth were quite substantial already in 1999 (in particular in the Republic of Korea and Malaysia), and by 2000 all five countries were firmly on the path of economic recovery. Yet, with the weakening of the major economies (USA, Japan, Germany and France) in 2001 and the events of 11 th September 2001, there appears to have been a reversal in the economic recovery of the countries of ESEA the worst affected being Republic of Korea, Malaysia and Thailand. In an increasingly globalized world, the adverse affects of slowdown in one region spreads more rapidly today than in the past. Due to their export-oriented nature and dependence on a narrow range of export markets (USA, EU and Japan) the economies of ESEA are particularly vulnerable to linkages with other countries. The paper provides data indicating that in 1999 and 2000, the countries of ESEA benefited from the growth of world trade, whereas in 2001 export growth for all of them suffered sharp declines. Apart from trade another channel through which the world s economies have been integrated is foreign direct investment (FDI), and in the countries of ESEA private capital flows were already stagnating below their pre-crisis levels prior to Sept.11 th The social impact of the crisis demonstrated itself in the form of increases in unemployment, underemployment and poverty, declines in living standards, and an upsurge in social tensions. In response to these shocks the labour market adjusted itself through a variety of mechanisms: Open unemployment, underemployment, sectoral shifts in employment (from industry to agriculture), declining wages and earnings, changes in labour force participation, migration and informalization. As for the global slowdown of 2001, evidence suggests that not only did economic recovery of the countries of ESEA suffer a setback, their labour markets were also adversely affected. In response, it appears that labour markets have adjusted mainly through increases in unemployment, and especially declines in manufacturing employment. In terms of poverty, the Asian economic crisis eroded significant gains in poverty reduction, a trend that only Indonesia appears to have reversed during the recovery period of The experience of the countries of ESEA points to the need to be prepared for periodic fluctuations in economic activity (often due to events beyond their control) by adopting policies and programmes that make it possible to rapidly respond to crisis. A rapid response mechanism would include some combination of macroeconomic policies that continue to promote growth of output as well as employment and income growth; labour market policies that minimize the impacts of the crisis on employment, and measures to directly alleviate the distress of the most affected groups. In terms of macroeconomic policies, several policy instruments are available to fight an economic downturn: monetary, fiscal, trade and exchange rate policies. Yet, when

33 examining their policy options, it is important for countries to consider both short and longer-term impacts of policies, and in particular the impacts on employment and labour markets. In the countries of ESEA the role of expansionary fiscal policies in the context of economic crisis is well recognised (due to their quick impact and strong multiplier effects), and data on the budget balance shows that at the time of the Asian economic crisis all the countries of ESEA had room for the adoption of expansionary policies. In addition to macroeconomic policies, active labour market policies (ALMPs), especially training and retraining policies and programmes, labour market information services, credit and livelihood promotion programmes, and special employment (or public works) programmes offer considerable scope for expanding wage and selfemployment and raising income levels. The recent experience of the ESEA countries suggests that active labour market policies have been an important component of their response to the Asian economic crisis, and, one would assume to the more recent global economic slowdown. The paper reviews the components of ALMPs, which are particularly relevant to the situation and needs in East and South East Asia. ALMPs can be used to make the labour market function better for workers in largescale establishments, in informal sector enterprises, in self-or wage-employment, as well as those in intermittent or casual employment including those subject to income poverty. In fact the labour market information system (LMIS), which is one of the major and comprehensive sources of information and analyses on which ALMPs are based, can be used to identify and target specific groups who are poor and vulnerable and who may be facing particular difficulties in being absorbed in productive employment. The paper explores ways in which this can be done. In addition, the paper points to the importance of a monitoring system that helps to identify the impact of policies and programmes on the economic condition of vulnerable and poverty affected groups. The paper points out that it is important to remember that a revival in economic growth does not automatically lead to an improvement in the employment, labour market and poverty situation. As the adverse social consequences of sharp economic downturns usually involves loss of employment opportunities and other related repercussions on the labour market, the appropriate way to minimise these consequences would be to work through a decent work agenda. In the decent work approach, social dialogue assumes particular importance in ensuring that necessary steps are taken in an environment of human dignity and in a democratic and participatory manner. Employment, Labour Market and Poverty during Economic Recovery and Slowdown in East and South East Asia By Rizwanul Islam 1 1 The authors would like to thank Ms. Nina Torm for the valuable research assistance that she provided. Responsibility for the paper and inadequacies that may be remaining in it lie solely with the authors.

34 Krishnamurty 1. Introduction The Asian economic crisis of and the global economic slowdown of 2001 serve to indicate that market economies continue to remain vulnerable to fluctuations. And such sharp fluctuations in economic activities can have serious adverse social consequences in terms of higher unemployment and under-employment, erosion of gains made in reducing poverty, and increased hardship for the poor. In situations where only small proportions (if any) of labour force are covered by unemployment insurance, loss of jobs may mean not only loss of income but also a lack of access to other social benefits (e.g., health services), and there is very little by way of social safety nets, the adverse impact of economic downturns on labour markets can create social tensions. It is, therefore, important to monitor the social dimensions of sharp economic downturns and adopt appropriate policies to counter the adverse social effects. While counter-cyclical economic polices are important from the point of view of responding to economic fluctuations, policies focusing on employment and labour markets are also essential to prevent increased hardship for the poor. Countries of East and South East Asia were only coming out of the economic crisis of when the global economic slowdown of 2001 caused a setback to their economic recovery. The purpose of the present paper is to see how the employment, labour market and poverty situation in these countries developed during the brief period of recovery and the subsequent slowdown (in 2001), and to discuss policies needed to respond to the adverse social effects of economic downturns. Section 2 provides an overview of recent economic growth in these countries, focussing on how their recovery suffered a setback due to the global economic slowdown in Section 3 examines how the employment, labour market and the poverty situation in these countries developed during their recovery period and in 2001 when recovery was hampered. Policies for employment promotion and poverty reduction during economic downturns and recovery are discussed in Section 4 while the final section presents a few concluding remarks. 2. Global Economic Slowdown and Setback in the Recovery of East and South East Asian Countries It is true that the countries of East and South East Asia (ESEA) that were affected by the economic crisis of were able to attain economic recovery fairly quickly. With the exception of Indonesia, the rates of economic growth were quite substantial already in And in 2000, the economic crisis appeared to have become history. But in the wake of the global economic slowdown of the process of recovery seems to have suffered a setback. Data on GDP growth presented in Table 1 shows the extent of recovery achieved by the five countries of ESEA during as well as the setback suffered by them in In 1999, recovery was indeed spectacular in the Republic of Korea and Malaysia. And in 2000, all five countries were firmly on the path of economic recovery. But in 2001, there were reversals in all these countries (see Figure 1). As 2 While the economic slowdown started already in early 2001, the terrorist attacks of 11 September on USA have exacerbated the slowdown. On this, see Islam (2001).

35 with economic recovery, the rates of decline also varied - the worst affected being the Republic of Korea, Malaysia and Thailand. Table 1: Summary of Economic Trends, East and South East Asia, GDP Growth (Average annual percentage change) GDP per capita ($) a) b) 2003 b) Indonesia Republic of Korea Malaysia Philippines Thailand Notes: a) GDP per capita at market prices (current US dollars), b) Forecasts from ADB, Asian Development Outlook Sources: ADB, Asian Development Outlook EIU Country Forecasts, 2000, 2001 & Figure 1: Quarterly Growth Rates of GDP in Countries of East and South East Asia ( ) Growth rate Thailand Malaysia Philippines South Korea Indonesia Qtr (2000) 2 Qtr 3 Qtr 4 Qtr 1 Qtr (2001) 2 Qtr 3 Qtr Quarter In order to have a proper perspective on the reversal of economic recovery in countries of ESEA, it is essential to understand the root causes of the global economic slowdown in 2001 and how ESEA countries were affected by that slowdown. It may be recalled in that context that the world economy had actually started showing signs of weakness in early And the decline in the growth of world output reflected a weakening of the three major engines of growth of the global economy. The US economy, after growing at breakneck speed for five years, started showing signs of fatigue towards the end of 2000 when growth in the manufacturing sector faltered. The technology bubble had also burst by then. Japan s economy entered into another recession during the first half of Within the EU region, the major economies like Germany and France had initially thought that they could weather the storm and

36 maintain their growth. But by the middle of 2001, their growth rates also started declining. Then came the terrorist attack on America on 11 th September, which inflicted a severe shock on the already weak economy of the country. And the direct and indirect effects of the shock reverberated quickly through the global economy. The US economy entered into an outright recession. Likewise, growth in the major EU countries declined further. 3 And given the high degree of integration of today s global economy, and more particularly, the dependence on America s economy, the adverse effects quickly spread to other countries. In an increasingly globalized world, the adverse effects of slowdown in one region spreads more rapidly today than in the past. As trade and capital flows are the major channels through which the world s economies have been getting integrated, declines in these flows create ripple effects and contribute to a spreading of contagion. And that is what actually happened in 2001 when growth of world trade declined (from 7.9 per cent in 2000) to 2.2 per cent. More particularly, it is the developing countries that suffered a sharper decline compared to the developed countries. A significant aspect of global integration is the high degree of dependence, especially of developing countries, on a few developed countries (notably, USA and EU countries) for their exports. And amongst the developed countries, dependence on USA is particularly noticeable. According to one estimate, US accounted for twofifths of world GDP growth over the past five years, either directly or indirectly by absorbing imports form other countries. 4 Indeed, American imports now amount to six per cent of the GDP of the rest of the world - twice as large as in And in the first half of 2001, American imports fell at an annual rate of 13 per cent; imports of IT equipment fell at a rate of almost 50 per cent. 5 Coming to the countries of ESEA, the export-oriented nature of their economies is quite well known. Data on exports as percentage of GDP presented in Table 2 confirms this, although the variation in the degree of export orientation is also clear. Table 3 provides an indication of their dependence on USA, EU and Japan for export markets. It appears from this table that it is only the Philippines who have been able to reduce its dependence on US market - but from a high level, and even after the reduction, the share of exports going to USA remains nearly 30 per cent. For Indonesia and Thailand, the dependence on USA has actually increased. Likewise, the degree of dependence on EU appears to have increased for all five countries. Table 2: Exports as a Percentage of GDP Indonesia Malaysia Philippines Republic of Korea Thailand For detailed figures, see Islam (2001). 4 Estimate by Morgan Stanley, reported in The Economist, August 25 th The Economist, August 25 th 2001.

37 Source: World Bank World Development Indicators Table 3: Destination of Asian Exports (as % of total exports) Year USA Japan EU Indonesia Malaysia Philippines R. Korea Thailand Sources: UNCTAD Handbook of Statistics 2000 for 1995 &1998 figures. ADB, Asian Development Outlook 2001 and 2002 for 1999 and 2000 figures. EIU Country Forecasts for 2001 projections (figures released after Sept. 11 th 2001). The result of such high dependence on a narrow range of developed country markets for exports is reflected in Table 4. In 1999 and 2000, the countries of ESEA benefited from the growth of world trade as reflected in the high growth of their exports. In 2001, export growth for all of them suffered sharp declines. Table 4: Annual Average Percentage Growth Rates of Exports Year a) 2003 a) Indonesia Malaysia Philippines Republic of Korea Thailand Notes: a) ADB, Asian Development Outlook 2002 forecasts. Source: ADB, Asian Development Outlook 2001 & 2002.

38 Apart from exports, another channel through which developing countries are being affected is foreign direct investment (FDI). According to the UN Department of Economic and Social Affairs (DESA), FDI was expected to decline in 2001, even before the 11 th of September event. And the decline must have been exacerbated by the changed environment. For the countries of ESEA, private capital inflows were already stagnating well below their pre-crisis levels. And in the current scenario, an improvement in the short run looks unlikely. Global integration and linkage with developed countries thus work in two ways. During good times, such interlinkage allows other countries to benefit from the spread effects of economic growth via healthy growth in trade and capital flow. This positive aspect was demonstrated in 1999 and 2000 when the crisis affected countries as Asia could recover quickly, thanks to the high rates at which the US and EU economies were growing and the consequent strong demand that was being generated for exports from the Asian countries. On the other hand, when the US economy started slowing down, the negative effects quickly spread, especially to the economies which are closely linked to it. Thus, it is not surprising that the countries of ESEA suffered such sharp declines in their GDP growth as shown by Table Social Dimensions of Economic Recovery and its Subsequent Reversal in East and South East Asia In terms of the social impact of the crisis, it is well known that the crisis caused increases in unemployment, underemployment and poverty, declines in living standards, and an upsurge in social tensions that endangered the fruits of the rapid economic growth achieved during the past two decades. Despite quick recovery, social and economic hardships continued for a large number of workers. And the reversal of economic recovery in 2001 must have led to a worsening of the situation in the labour markets. The literature on the impact of the crisis on labour markets in the region shows that the adjustment to shocks took several forms: open unemployment, underemployment, sectoral shifts in employment, declining wages and earning, changes in labour force participation, migration, and informalization. An important question to address in this regard would be how the labour markets performed during the period of recovery and in 2001 when the process of recovery suffered a setback. During the recovery (i.e., in 1999 and 2000), open unemployment declined only in the Republic of Korea and Thailand. And in these countries also the rates of open unemployment remained at levels much higher than the pre-crisis levels. In Indonesia and Malaysia, unemployment continued to increase, and in the Philippines it increased in 2000 after a small decline in Underemployment levels remained very high despite some decline. On the whole, it thus seems that the labour market situation did not improve much during despite economic recovery. And in 2001, Indonesia, Malaysia and Thailand experienced increases in unemployment (see Table 5). Table 5: Summary of Unemployment, Underemployment and Employment Trends,

39 Unemployment Rate (as % of Total Labour Force) Underemployment Employment a) b) % Change % Change % Change Indonesia c) N/a R. Korea d) N/a N/a N/a Malaysia N/a N/a N/a Philippines e) 10.1 e) 9.8 e) Thailand N/a N/a N/a Notes: a) EIU estimates, b) EIU forecasts, c) Figure from Dhanani 2002, d) Figure from National Statistics Office e) Figure from National Statistics Office Sources: ADB Key Indicators EIU Country Forecasts 2000, 2001 and National Statistics Office Websites: Republic of Korea: Philippines: It needs to be noted, however, that simple figures on unemployment may not reflect the true picture of the labour market. In some countries of South East Asia, notably in Indonesia and the Philippines, and to some extent in Thailand also, underemployment remains a serious problem. Also, declines in unemployment may actually mask new phenomena like atypical forms of employment (indicative, in turn, of the precarious nature of the new jobs that are created), which characterize the labour markets. In order to examine whether economic recovery is leading to real improvements in the labour markets, it would, therefore, be important to have a good understanding of the extent to which such phenomena characterize the labour markets in the recovering economies. Underemployment had increased (albeit in varying degrees) in all the crisis affected countries in Unfortunately, comparable data for recent years are not easily available for all countries. Data for Indonesia and the Philippines indicate declines in underemployment both during and respectively. However, despite small declines, the levels of underemployment remain very high in both these countries. In terms of the sectoral distribution of employment (Table 6), there was a substantial increase in the share of agriculture and a decline in the share of industry in Indonesia during the crisis period. In 1999, the trend got reversed for both the sectors indicating positive developments in the labour market. In 2000, however, there was movement back to agriculture, although the share of industries continued to rise. The decline that year was in the service sector. During , however, the shares of all three sectors moved in the right direction (i.e., declined for agriculture and increased for industry and services). But for the other four countries, 2001 was a bad year for industrial employment - with the share of this sector in total declining in all. It thus appears that with the global economic slowdown of 2001, not only did economic recovery of the countries of ESEA suffer a setback, their labour markets were also adversely affected.

40 Table 6: Employment Change by Industrial Sector, & % Change in Employment % Change in Employment Agriculture Industry Services a) Agriculture Industry Services Indonesia b) Korea c) Malaysia N/a -3.8 d) N/a Philippines e) Thailand f) Notes: a) Includes construction, b) Figures are from Dhanani 2002, c) Figures are yearly averages d) Figure represents % change from Sept to Sept. 2001, e) Figures are from October 1999,2000 and 2001 f) Figures are from May 1999, 2000 and 2001 Sources: ADB Key Indicators National Statistics Office websites: Korea: Philippines: Thailand: Indonesia: Real wages (Table 7), however, fared better - except, of course, in Thailand, where except for a small recovery in 1999, the decline continued through In Indonesia, which had suffered a steep fall in real wages in 1998, there was substantial recovery in both 1999 and 2000, so that they went back fairly close to the pre-crisis level. In Korea, Malaysia and the Philippines, recovery in real wage rates continued through Table 7: Real Wage Trends in Crisis Countries, % Change in Real Wages Year a) Indonesia b) N/a N/a Korea Malaysia Philippines c) Thailand Notes: a) EIU forecasts, b) Figures are from Dhanani 2002, c) EIU estimates. Sources: ADB Key Indicators EIU Country Forecasts 2000, 2001 and The experience of 2001 thus seems to offer some interesting contrasts with that of the earlier crisis period in respect of the manner in which the labour markets adjusted to the economic downturn. Of course, the downturn of 1998 was very sharp, and the labour markets adjusted through a variety of mechanisms. On the other hand, the economic slowdown of 2001 was less sharp compared to the earlier crisis; and the labour markets adjusted mainly through increases in unemployment, and especially declines in manufacturing employment. The services sector also appeared to have played a role, as employment in that sector recorded positive growth in 2001 in all the countries of ESEA. Real wage rates do not appear to have played a role in facilitating labour market adjustments in 2001.

41 An important aspect of the social effects of an economic crisis is the impact on poverty. It is well known that the Asian economic crisis of has had a severe adverse impact on the poverty situation of the affected countries. While Indonesia was the most severely hit (not only in terms of the economic downturn but also in terms of the sharp reversal of the achievement made in reducing poverty), the incidence of poverty increased in the Republic of Korea, Philippines and Thailand also. 6 The important question in this respect would be whether economic recovery succeeded in reducing poverty and putting the countries back on the path of poverty eradication. Although lack of up to date data on poverty makes it difficult to arrive at a firm conclusion in this regard, data compiled and presented in Table 8 does provide some picture. Table 8: The Incidence of Poverty in Selected Countries of South East Asia, Countries and data source Indonesia BPS data a) 17.7 N/a b) Malaysia ADB data N/a N/a N/a Philippines NSCB data ADB data World Bank data Thailand ADB data Mahmood and Aryah (2001) 35 c) N/a N/a N/a N/a N/a N/a N/a N/a d) N/a N/a Notes: a) The figures are based on the so-called new poverty standard based on the 1998 standard. b) This is a preliminary figure for February 2000 obtained from World Bank s Indonesia Country Update. c) This figure is for d) This figure is from World Bank: Thailand Economic Monitor N/a N/a In Indonesia, the head-count ratio of the incidence of poverty increased sharply between 1996 and 1998, but declined thereafter, falling back to the level of 1996 by August Thailand provides a different experience where the declining trend in the incidence of poverty did get reversed during , and that trend continued till 1999 when the incidence of poverty increased further. Data on Philippines provides a mixed picture. While figures provided by the Government (National Statistical Coordination Board) indicate an increase in poverty between 1997 and 2000, ADB data show the opposite. World Bank figures on the other hand indicate an increase between 1997 and 1998 and a slight decline in although the level of 1999 is still higher than that of The weight of evidence thus seems to indicate that while Indonesia has been able to reach its pre-crisis level of poverty, Philippines and Thailand are yet to be able to do so. 6 See the country studies on Indonesia and Thailand in Betcherman and Islam (2001), Lee (1998), and World Bank (2000). 7 This is based on Islam (2002). But World Bank s preliminary figure from the Country Update put the figure for February 2000 at a higher level (22 per cent).

42 Several conclusions seem to follow from the above discussion. First, although the crisis affected countries were able to achieve economic recovery fairly quickly, that was not followed by a full-fledged recovery of labour markets. While the labour markets had adjusted to the crisis through increases in unemployment and underemployment, changes in the sectoral shares of employment, and declines in real wages, the opposite processes did not take place immediately (with the exception, perhaps, of wages) when economic recovery started. Second, while earlier gains in poverty reduction were eroded significantly during the crisis, during the recovery period the process of erosion appears to have stopped only in Indonesia. Third, the setback in economic recovery suffered by the crisis-affected countries in 2001 has led to a further deterioration in their labour markets. 4. Policies for Employment Promotion and Poverty Reduction during Economic Downturn and Recovery Fighting External Shocks The Asian crisis and the global economic slowdown of 2001 (which was exacerbated by the events of 11 September 2001) bring home the point that in an increasingly global economy, sharp fluctuations in economic activity in one part of the world can have fairly quick adverse impacts on other parts of the world. When this happens to countries like those in East and Southeast Asia, still recovering from an earlier crisis, it points to the need to be prepared for periodic fluctuations in economic activity, often due to events beyond their control. Such countries have to adopt policies and programmes that make it possible to rapidly respond to crises, and to mitigate their impacts without retreating from their commitment to maintaining an open economy. A rapid response mechanism for this purpose would include some combination of macroeconomic policies that continue to promote growth of output as well as employment and income growth; labour market policies that minimize the impacts of the crisis on employment, and measures to directly alleviate the distress of the most affected groups. As all these mechanisms need to be activated at short notice, they call for preparedness on the part of the concerned governments. Again, periodic external shocks must not be permitted to derail long-term social and economic policies relating to income growth, poverty and inequality reduction. The difficult employment situation over the medium term makes it essential to combine the creation of a larger number of new jobs and the consequent fall in unemployment with a reduction in the number of the working poor and an increase in the quality of employment. Macroeconomic Policies Globalisation has reduced the autonomy of individual countries in macroeconomic policy and yet it has increased the necessity of sound macroeconomic policy choices. More rapid growth requires macroeconomic stability, which will provide the climate for expanding investment and growing employment. As Professor Stiglitz has recently pointed out:

43 In framing macro-economic policies, we need to keep our eyes on the ultimate objectives, not on intermediate variables on employment growth, living standards, not interest rates, inflation rates, or exchange rates, variables that are important only to the extent that they affect the variables of fundamental importance. 8 Stabilisation measures in response to a crisis, especially in the short run, can adversely affect job creation. They can also result in damaging cutbacks in public expenditure on education, health and other social services, on which the future growth of employment and incomes depend. These measures may also accentuate income decline, inequality and poverty, contributing thereby to the processes already set in motion by the economic crisis. Hence, when countries examine their macroeconomic policy options, they need to consider both short and longer-term impacts of policies and, in particular, the impacts on employment and labour markets as well as poverty and inequality. Several policy instruments are available to fight an economic downturn. These include monetary, fiscal, trade and exchange rate policies. They vary considerably in terms of the nature and speed of their impacts. Monetary policies are generally more effective in limiting a boom than in stimulating a recovery and hence fiscal policies are needed which are quick in impact and have strong multiplier effects. Within the region, the role of expansionary fiscal policies in the context of economic crisis is well recognised. It is worth stressing that although most of the countries of the region were having budgetary deficits, the magnitudes were within reasonable limits except in Malaysia, and to some extent, Philippines (Table 9). The budgetary situation, thus, did have room for the adoption of expansionary policies. Table 9: Budget balance (% of GDP) a) 2001 a) Indonesia Malaysia Philippines Republic of Korea Thailand Notes: a) ADB, Asian Development Outlook 2002 estimates Source: ADB, Asian Development Outlook Table 10: Policy Responses to the Global Economic Slowdown Country Monetary Fiscal Trade Policy 8 Stiglitz J. (2001).

44 Indonesia BOI has maintained a tight monetary policy. Average interest rates remain at 14%, yet are expected to fall modestly in the second half of Interest rates will have to remain relatively high to enable the government to issue securities to finance the budget deficit and to prevent the Rupiah from depreciating further. In 2001 budget deficit met the target of 3.7% of GDP. In 2002 deficit target revised from 2.5% to 2.7% of GDP. The EIU predicts a deficit of 2.9% of GDP for In 2003 further decline in deficit with stronger economic growth & reduction in domestic interest rates. In 2001 the Rupiah averaged :US$1 down from an average of Rp 8.422:US$1 in In 2002 Rupiah expected to remain under pressure due to decreasing tourism revenue and a possible oil price fall. Stronger export growth in the second half of 2002 will support the Rupiah. Malaysia On Sept. 20 th 2001, BNM lowered the intervention rate by 0.5% to 5%. In October the BNM lowered public lending rates by 0.5% and eased access to funds for small and medium-sized industries. It is expected that the rate will be cut again during the first half of Philippines Since October 4 th 2001 overnight borrowing and lending rates have been lowered to 7.25% and 9.5% respectively. Official rates are now at their lowest level since Sept.1 st In Sept the government announced M$ 4.3 bn (US $ 1.1bn) stimulus package. M$1 bn to be spent on small rural and selected urban development projects. The 2001 budget deficit was estimated at 6.5% of GDP, yet narrowing to 5% in 2002 and 3.3% in In 2001 the budget deficit rose to an estimated P147 bn (US$ 2.9bn) (3.8% of GDP). The government is targeting a budget deficit of P 130 bn for 2002 and P 88 bn for The Ringgit (pegged to the US dollar) was not devalued after September 11 th The currency fell by 11% in both 2000 and 2001, averaging P51: US$1 in A further depreciation to an average of P52.35: US$1 has been forecast for Republic Korea of In October 2001, BOK lowered the overnight call rate by 50 basis points to 4%. The EIU predicts that the BOK will maintain an accommodating monetary policy until end Table 10 (Contd.) Thailand On January 21 st 2002 the BOT cut interest rates for the second time in a month. The average interest rate for savings accounts is 1.75% and the primelending rate is 7.2%. As the bench mark 14 day repurchase rate stands at 2%, there is still a positive interest rate differential On Oct 8 th 2001 a $1.5 bn injection package was announced. The budget deficit for 2001 was estimated at 2% rather than 0.1% of GDP. In 2002 and 2003 the government is expected to run budget deficits of 1.9% and 0.9% of GDP. Fiscal spending rose by 19% in the first two months of fiscal year 2001/02. Including a Baht 58 bn reserve fund earmarked for economic stimulus (to support labour intensive projects in the agricultural sector). Budget deficit of 3.6% of After Sept. 11 th financial support was given to exporting companies. BOK allowed the Won to depreciate. In the Won may appreciate modestly (from W1.291: US$1 in 2001 to W1.170: US$1 in 2003). (Cont.) BOT bought US dollars in order to curb the appreciation of the Baht. The Baht s buoyancy and concern about its negative impact on export competitiveness was one of the factors supporting the BOT s decision to cut interest rates again on January

45 with the US. Strength of the Baht, low inflation and need to stimulate growth are main justifications for lowering interest rates. GDP, or Bt 203 bn for the fiscal year 2001/ Source: Compiled from various media reports. (Adapted from Islam, 2001 and updated.) Table 10 presents in a summarized form information on policy instruments used by the countries under discussion to respond to the economic slowdown of It is quite clear that they applied a combination of monetary, fiscal and trade policies a strategy which makes good sense. While the impact of tax reductions on aggregate demand may be mixed, the effects of increased public spending are likely to produce quick and positive impacts on employment and incomes. Further, through the multiplier effect, future incomes and employment are likely to grow, assuming that crowding out effects on private investment are not significant. 9 Apart from increased public spending, other measures are needed to create a favourable environment for micro and small and medium enterprises, which are likely to be the main source of job creation. While reduction in interest rates may also be useful, specific measures are needed to increase employability by raising the productivity and skills of the working poor and the unemployed, and by improving the access of these enterprises to credit and labour and product markets. Active Labour Market Policies Active labour market policies, especially training and retraining policies and programmes, labour market information services, credit and livelihood promotion programmes, and special employment (or public works) programmes offer considerable scope for expanding wage and self-employment and raising income levels. The recent experience of the East and Southeast Asian countries suggests that active labour market policies have been an important component of their response to Asian economic crisis and, one would assume, to the more recent global economic slowdown. As we have argued elsewhere 10, elements of active labour market programmes (ALMPs) were in operation in the East and Southeast Asian countries well before the Asian crisis. However, they were not viewed as ALMPs, but as part of the normal battery of policies and programmes to improve skills, create jobs and place job seekers. The Asian crisis brought into the limelight the need to use these instruments to fight crises and this meant that they needed to be considerably strengthened, better co-ordinated, and much quicker in terms of response. We briefly review below some of the important components of ALMPs particularly relevant to the situation and needs in East and Southeast Asia. A comprehensive labour market information system (LMIS) is needed to enable all the major economic agents in the labour market - employers, the workers and job 9 See Islam R. (2001) for a discussion. 10 Islam R. et.al. (2001).

46 seekers, and the governments to be better informed of the current labour market situation and likely changes, so as to be able to respond quickly to them. This would make possible a better match between labour supply, comprising persons with different skills, experience, location and expectations and labour demand, for different skills, experience, location and desired wage rates. The LMIS would have to be closely linked to the training and retraining system. It would not only take into account on the supply side of the likely output of trained/retrained persons, but it would also provide information and analyses needed by the training/retraining system to revise and update its plans in the context of the changing labour market situation. It is important to emphasise here that in the context of a volatile external environment retraining may be just as important as training. Again, it should be noted that the training and retraining system in most countries covers a wide variety of training sources, institutional and on the job, public and private. A rather different ALMP measure is the promotion of self-employment and enterprise development through credit and marketing support programmes. This has the potential to create a large volume of work opportunities (both self and wage employment) especially in micro and small and medium enterprises. Enterprise promotion programmes should be seen as an important instrument in crisis response. There is the danger that in the economic stringency following a crisis, this type of programme may be cut or credit sources may dry up. Many of the ALMP measures work well only if work opportunities are growing in the labour market. In the context of a sharp decline in work opportunities due to a crisis which may be nation-wide or restricted to a few regions- an important and highly relevant ALMP measure is the public works or emergency employment programme. Such programmes can be used to quickly create jobs, sustain purchasing power or jump-start a sagging local economy. Again these are not new in many East and Southeast Asian countries. Several countries have large public infrastructure programmes using labour-based methods, but their role as means of rapid response to crises is perhaps inadequately appreciated. As an immediate response to a crisis situation, emergency employment programmes work through directly creating jobs and, through the injection of purchasing power, starting the process of economic recovery. Emergency employment programmes are often viewed as unproductive, but this is not necessarily so, provided there is adequate preparation (even before the crisis). A short list of works could be maintained which could be implemented quickly in different parts of the country to strengthen infrastructure. These programmes should be positioned to respond to early warning signals of sharp declines in employment in specific areas or for the country as a whole. Given that there are frequent seasonal declines in work opportunities as well as area-based gluts in labour supply, emergency employment programmes can be kept operational irrespective of whether or not at any particular moment, there is a major crisis. It is much easier to upscale or downscale activities depending on the labour market situation, rather than totally stop and again start these activities. ALMPs, Poverty and Employment

47 ALMPs can be used to make the labour market function better, not only for workers in large-scale establishments. The battery of instruments available makes it possible to deal with the problems of persons working in informal sector enterprises in self-or wage- employment, as well as those in intermittent or casual employment, including those subject to income poverty. Some ALMP instruments, like the public employment service system may be focusing on the demand and supply of labour in respect of large establishments, but this is not true of other instruments. In fact, the labour market information system, which is one of the major and comprehensive sources of information and analyses on which ALMPs are based, can be used to identify and target specific groups who are poor or vulnerable and who may be facing particular difficulties in being absorbed in productive employment. We explore below some ways in which this can be done. In the countries under discussion, labour force surveys provide enough disaggregated information to identify groups facing problems in the labour market. They may, in some instances, provide information on their earnings and nature of their employment. The latter might provide some evidence on the growth of insecure or atypical forms of employment as well as seasonal or year-round unemployment and underemployment. Another useful set of information relates to the prevalent quarterly or monthly wage rates of selected categories of wage earners. Movements in wage rates over time, and differences across space may be quite revealing of what is happening to relatively poorer sections of the population outside the formal sector. As Shafiq Dhanani shows 11 such data on wage movements of unskilled and low skilled categories of workers, working in the agriculture and the informal sector, is available in Indonesia. Perhaps similar data is available for other countries as well. Sharp declines in this indicator could point to serious difficulties faced by these groups which could be seasonal or perennial in nature and which may be localized or region-specific or nationwide. While most countries collect a wide array of data relating to income and/or personal consumption expenditure, and the magnitude and characteristics of poverty are generally analysed, the interface between poverty and labour force characteristics should be more fully explored. This would enable us to formulate better policies and programmes that take into account the labour market situation and behaviour of specific groups facing income poverty. The difficulty often is that poverty data and employment data are not collected in the same survey, but it may still be possible to get labour force information of individuals, collected through a labour force survey, cross-classified by some measure of average household consumption expenditure obtained through the same survey. It may turn out that the incidence of open unemployment and poverty status are not closely associated. On the other hand, underemployment may be better related to poverty. It would also be useful to know whether the poor are mainly employees or the self-employed, and the sector and region in which they are currently engaged. Profiles on this basis would help to improve targeting by identifying the labour force groups suffering from poverty and providing clues to the kinds of programmes are likely to be of assistance. 11 Dhanani S. (2002).

48 LMIS and Programme Response Monitoring In the case of the countries under discussion, considerable national and international attention has been devoted to monitoring the recovery process from crisis. Crises tend to recur and early warning indicators are needed not only to predict crises, but also to track the recovery process and, as and when required, to revise policies and programmes or, at least, tailor them to suit emerging needs. However, this has mainly taken the form of monitoring relatively easy to obtain estimates of aggregate output growth and investment expansion, trends in trade and the balance of payments, and prices. While these are no doubt useful, they do not help to identify the impact of policies and programmes on the economic condition of vulnerable and poverty affected groups. Unfortunately, poverty is not easy to measure accurately and it is difficult to obtain estimates without significant time lags. This is also true to some extent of employment, but it is still possible to monitor the impacts of policies and programmes by looking at employment and labour market indicators. Some of these have been discussed already. They include monthly or quarterly monitoring of trends in real wages for different categories of workers and for different locations. It may also be possible to arrange for special priority tabulations to provide quick results on a limited number of variables from labour force surveys. Another approach would be to collect on a quarterly basis information on employment in the formal and large-scale enterprise sector. An abridged schedule might be used for this purpose. The aim should be to ensure that at any given time the policy maker has in hand information on employment growth for the period up to the previous quarter. Looking at recent patterns in the growth of employment in larger enterprises, and combining it with information on recent wage trends, the policy maker would be in a better position to assess the employment situation and to adjust policies and programmes based on this information. While this would complement, not substitute for, poverty monitoring, the suggested indicators would help to identify labour market trends for a wide variety of groups within the labour market, including women and men and the rural and urban poor. 5. Concluding Remarks The major engine of growth in the global economy the US appears to be coming out of its brief recession. Along with the prospect of revival in the growth of the US economy, the prospect of growth in the export oriented economies of East and South East Asia has also brightened up. Indeed, projections released recently by the Asian Development Bank reflect such an optimistic outlook (growth forecast of 5.2 per cent and 5.2 per cent for 2002 and 2003 respectively). It is important to remember, however, that a revival in economic growth (especially if the sustainability of such recovery remains doubtful) does not automatically (and immediately) lead to an improvement in the employment, labour market and poverty situations. At least, this is the lesson from the experience of the short-lived recovery of Policies to fight economic downturns, therefore, need to take this into account. That, in turn, would require simultaneous and integrated application of economic and social policies.

49 As the adverse social consequences of sharp economic downturns usually involve loss of employment opportunities and other related repercussions on the labour market, the appropriate way to minimize these consequences would be to work through a decent work agenda. The decent work approach is based on the promotion of opportunities for women and men to obtain decent and productive work, in conditions of freedom, equity, security and human dignity. This approach would be highly relevant in the current global economic climate of periodic fluctuations and the associated uncertainties as well as possible social consequences. In such an environment, it would be necessary to undertake measures to revive economic activities that can generate new productive employment, to help enterprises survive so that jobs can be preserved, and to ensure that workers continue to be protected against risks posed by ill health and supported in old age. And as there can be tensions inherent during periods of economic difficulties requiring adjustments in the economy as well as at the enterprise level, social dialogue assumes particular importance to ensure that necessary steps are taken in an environment of human dignity and in a democratic, participatory manner. References

50 Asian Development Bank (2001) Asian Development Outlook 2001 Update. (WWW) Asian Development Bank (2001) Key Indicators. (WWW) Asian Development Bank (2002) Asian Development Outlook (WWW) Betcherman G. and Islam R. (2001) East Asian Labour Markets and the Economic Crisis: Impacts, Responses & Lessons. World Bank: Washington. Dhanani S. (2002) Strengthening the Indonesian Labour Market Information System. Paper prepared for the Seminar on Strengthening Employment and Labour Market Policies for Poverty Alleviation and Economic Recovery in East and Southeast Asia (Jakarta, 2002). ILO: Geneva. (published in this volume). Economist Intelligence Unit (2002) Country Forecasts (WWW) Islam R. (2001) Employment Implications of the Global Economic Slowdown 2001: Responding with a Social Focus. ILO: Geneva. Islam R. et al. (2001) Active Labour Market Policies in East and Southeast Asia: What has been done and what can be done? Paper in the Report from the Seminar on Labour Market Policies: its Implications to East and Southeast Asia. Lee E. (1998) The Asian Financial Crisis: The challenge for social policy. ILO: Geneva. Mahmood M. and Aryah G. (2001) The Labour Market and Labour Policy in a Macroeconomic Context: Growth, Crisis and Competitiveness in Thailand in Betcherman G. and Islam R. (2001) East Asian Labour Markets and the Economic Crisis: Impacts, Responses & Lessons. World Bank: Washington. Stiglitz, J (2001). Keynote address to the ILO Global Employment Forum. (WWW) UNCTAD (2000) Handbook of Statistics. UNCTAD: Geneva. World Bank (2000) Philippine Poverty Assessment Volume I: Main Report. World Bank: Washington. World Bank (2001) World Development Indicators. World Bank: Washington.

51 POVERTY, EMPLOYMENT AND WAGES: AN INDONESIAN PERSPECTIVE BY Iyanatul Islam EXECUTIVE SUMMARY It is widely recognised that Indonesia was the worst hit country in the tragic episode of the 1997 Asian crisis, and after numerous debates on how badly the poor were affected, a consensus emerged that poverty shot up for a period before stabilising

52 again. Yet, innovations in poverty measurement and analysis have suggested a disenchantment with simple notions of poverty and a number of studies have emerged suggesting that the magnitude of poverty in Indonesia is much more widespread than was initially thought. These findings are emerging precisely at a time of decreasing fiscal resources, prospects of prolonged slow growth against a faltering global economy and political uncertainties. Faced with such harsh realities, it is imperative that Indonesian policy makers develop a deeper appreciation of the sources of the manifold manifestations of poverty and identify cost-effective and credible policy instruments that could lift the many millions who are poor today and those who are likely to become poor tomorrow towards a reasonable secure and prosperous future. This rethinking on poverty in Indonesia should entail a greater understanding of the interaction between the labour market and those who are poor. The purpose of this paper is precisely to explore this linkage. Reflecting current intellectual trends, the paper highlights multiple dimensions of Indonesian poverty in both the pre- and post-crisis period. This involves an empirical synthesis of trends in income/consumption poverty, capability deprivation encompassing the UNDP s human poverty index, vulnerability and powerlessness. According to the income/consumption approach, the percentage of people living below the poverty line fell from 40% in 1976 to 18% in 1996, yet rose again to 37% in September 1998 before falling to its pre-crisis level of 18% by August By contrast, the capability deprivation indicators (malnutrition, literacy, access to health services and low life expectancy) suggest that in 1999 more than 50% of Indonesians did not have access to health services while the incidence of child malnutrition was 30%. Similarly, vulnerability (the risk of falling into poverty or getting deeper into poverty in the future) in Indonesia is more significant than current consumption poverty, recent estimates suggesting that the incidence of vulnerability is 34%. Thus despite decades of sustained economic and social gains prior to the 1997 Asian crisis, Indonesia has still to grapple with the manifold manifestations of poverty. The paper explores the extent to which the evolution of consumption poverty (and consumption-based indicators of vulnerability) can be reconciled with developments in the labour market. During the s, there was a reallocation of labour from low productivity, low-paid jobs in the rural economy to better paid employment in the urban and formal economy. Moreover, rapid growth in the pre-crisis era was accompanied by a deepening of skills of the workforce and a corresponding increase in the share of the educated/skilled workforce. In addition, casualisation of the workforce and involuntary underemployment (major attributes of poverty) were not widespread. These labour market features are consistent with the sustained decrease in consumption poverty as documented above, and suggest a strong link between labour market developments and progress in poverty reduction. Another important aspect of the poverty-labour market nexus in Indonesia is the role that the employment-intensity of rapid growth played in the pre-crisis period. Higher employment elasticity translates to the fact that for any given growth rate, the job creation rate is higher. The available evidence suggests that employment elasticity was quite low (0.3) in the 1970 s, rose appreciably in the 1980 s (0.8) and tapered off a little with the onset of the 1990 s. These estimates are consistent with conventional interpretations of the evolution of the employment-poverty linkage in the Indonesian context.

53 In addition to employment, wages is another dimension of the poverty-labour market nexus, which affects the welfare of workers. The paper reviews the long-run behaviour of real wages in Indonesia, highlighting episodes of falling/stagnating real wages, with specific reference to the 1997 financial crisis. Moreover, the relationship between real wages, productivity and poverty is explored, in particular the role that changes in productivity have played in underpinning wage growth. The findings suggest apparent co-movements in real wages, productivity and consumption poverty with substantial declines in poverty being accompanied by rising productivity and real wages. For instance from 1978 to 1997, real wages and productivity grew by 5% p.a. on average, yet in 1998 real wages declined by 40% and remained 10% below their pre-crisis level three years later in A debate that has emerged under these circumstances is the extent to which employment and wage adjustments during the crisis period reflect the virtues of labour market flexibility. The paper argues that whilst useful, the notion of labour market flexibility can only represent the beginning rather than the end of any analysis of the linkage between poverty and the structure of the labour market. In attempting to develop a policy framework for dealing with issues that directly and indirectly impinge with the evolution of poverty, it is important to be aware of the key labour market challenges. These include; the need to create enough durable jobs annually to absorb new entrants to the workforce and clear the backlog of the unemployed and the underemployed. Moreover, there is a need to empower workers both in the formal and informal economy to cope with the problems of vulnerability to future episodes of poverty. In addition, there is a need to develop a credible industrial relations system that can represent the interests of both formal sector and marginalized workers and thus contribute to alleviating the sense of voicelessness and powerlessness of the poor. As for the elements of a policy framework that one could devise for dealing with these labour market issues, the paper provides some thoughts as follows. Firstly, there is a need for the return of rapid and sustainable growth. Secondly, employment concerns must be reflected in macroeconomic policy, enabling the government to track the employment intensity of growth through estimates of employment elasticity. A third policy challenge involves ensuring that barriers are not erected against broad-based participation by the poor in the education and training system. A fourth policy element is concerned with the strengthening of more formal systems of social protection in order to deal with labour market risks, particularly those triggered by economic shocks. A fifth component is concerned with the building of appropriate labour market institutions, which provide voice and representation for ordinary workers. This involves a process of social dialogue (between government, employers and worker organisations), which builds on a labour rights-driven industrial relations system. Poverty, Employment and Wages: An Indonesian Perspective By

54 Iyanatul Islam 1. Introduction What a difference a crisis makes! Not too long ago Indonesia was heralded as a rare exemplar of a low-income economy that has managed, over the space of two decades, to reduce mass poverty significantly. In the 1970s, this was attributed to rapid growth and the judicious use of revenues derived from Indonesia s oil wealth to fund public infrastructure, especially in agricultural and rural sectors. Such an activist policy eventually fuelled demand for unskilled workers in agriculture and constructionrelated activities and had a positive impact on real wages. In the mid-1980s, sustained poverty reduction was seen as the product of broad-based employment growth fuelled by the implementation of a trade liberalisation-cum-deregulation reform agenda. Certainly, the available poverty statistics and pertinent social indicators pointed in the direction of a country that, under the tutelage of the authoritarian Suharto regime, delivered both political stability and sustained prosperity shared by many. Not surprisingly, these achievements bred a sense of complacency on the battle against poverty. It was felt that the latter as a generic phenomenon was largely tamed. The government needed to focus on pockets of poverty that marked the archipelagic terrain of Indonesia. 12 The 1997 financial crisis and its aftermath have shaken perhaps forever- the sense of optimism that characterised the Suharto era. It is widely recognised that Indonesia was the worst hit country in the tragic episode of the 1997 Asian crisis. After an initial bout of robust debate on how badly (or mildly) the poor were affected by the terrible recession of 1998 that followed the financial crisis, a consensus seems to have emerged that poverty shot up for a period, but the worst seems to be over. 13 Under normal circumstances, such a finding would have inspired cautious optimism. Instead, several developments have overshadowed the onset of even cautious optimism. Innovations in poverty measurement and analysis have suggested a disenchantment with simple notions of poverty such as insufficient income to acquire the basic necessities of life and a growing a recognition of its multidimensionality. It is increasingly being couched in terms of deficient capabilities (lack of basic education, adequate health, nutrition etc), vulnerability (the risk that people can move in and out of poverty) and powerlessness (the notion that the poor feel helpless and unable to influence the institutional, social and political circumstances that affect their daily lives and their future) Breman (1999) offers a critique of the way poverty statistics were used in Indonesia under the Suharto regime. This, does not, however, undermine the genuine progress against poverty that was made in the high-growth era prior to the 1997 financial crisis. 13 For a thorough review of this debate, see UNSFIR (1999). See also Booth (1999, 2000). 14 For a comprehensive statement on the changing views on global poverty, see World Bank (2000a), World Development Report, 2000/2001. See also Kanbur and Squire (1999). The notion of poverty as capability deprivation has been forcefully argued by Sen (1999) and is one of the core planks of the UNDP s global reports on human development. UNDP has also made the point that capability deprivation is equivalent to deprivation of human rights. See, for example, UNDP (2000). The notion of powerlessness is a key feature of a landmark voices of the poor study drawn from the responses of a large sample of poor individuals and households spread across a large number of countries in the

55 Based on this notion of multidimensionality, findings in the Indonesian context have emerged that suggest that the magnitude of poverty is much more widespread than was initially thought. At least one study claims that as much as 50 per cent of Indonesians are poor based on the notion of vulnerability (vis-à-vis a little over 20 per cent based on conventional definitions of poverty). 15 Another study has suggested that the incidence of capability deprivation was more than twice the rate of poverty (as conventionally defined using consumption expenditure data) that was officially recognised to have prevailed in the Suharto era. 16 Survey data also suggest that approximately 87 per cent of those who are in the bottom quintile of households (in terms of expenditure distribution) do not have the incentive or the capacity to participate in political and civic activities. 17 Such sombre findings are emerging precisely at a time of dwindling fiscal resources, prospects of prolonged slow growth against a faltering global economy and political uncertainties. 18 To compound these challenges, Indonesia has embarked on what one author has aptly called a systemic transition. It is engaged in the quest for democratic consolidation; it has embraced decentralised governance after decades of highly centralised rule; it is struggling to implement a broad-based economic reform agenda. 19 Faced with such harsh realities, it is imperative that Indonesian policy makers develop a deeper appreciation of the sources of the manifold manifestations of poverty and identify cost-effective and credible policy instruments that could lift the many millions who are poor today and those who are likely to become poor tomorrow towards a reasonably secure and prosperous future. 20 A salient aspect of this rethinking on poverty in post-suharto Indonesia should entail a greater understanding of the interaction between the labour market and those who are poor. To what extent has poverty reduction strategies in the past been a reflection of broad-based employment growth? What has been the role of productivity-driven wage increases in influencing the living standards of ordinary Indonesians? How have employment and real wages adjusted during the Indonesian crisis of 1997 and what has happened since then? What are the corresponding normative and policy implications? What is the way ahead? These, then, are the issues that are explored in this paper. Section 2 offers an account of Indonesian poverty before, during and after the crisis. Reflecting current intellectual trends, a distinction is made among the multiple dimensions of poverty. In developing world. See Narayan et al (2000). For a critique of the broadening of the notion of poverty, see McCawley (2000). 15 See World Bank (2000b) that draws on Pritchett et al (2000). 16 See Dhanani and Islam (2000, 2002). 17 As reported in World Bank (2001, table 4, p.9). 18 Growth in 2001 is now expected to be below 3.5 per cent after being close to 5 per cent in The central government budget deficit is expected to increase to 6.0 per cent of GDP in the absence of countervailing measures. See World Bank (2001: 4-6). 19 See Mishra (2001). 20 Media reports claim that the Indonesian government appears to be aloof to the theme of a renewed engagement with the battle against mass poverty that is now a major part of the agenda of the international development community. For example, media reports allege that the Indonesian government made a low-key representation at the recently concluded UN Conference on Financing for Development held in the Mexican City of Monterrey which re-affirmed the UN Millenium Goal for halving global poverty by See the Jakarta Post, March 25, 2002.

56 particular, trends in consumption poverty, capability deprivation, vulnerability and powerlessness are highlighted. This sets the context for a discussion of the relationship between poverty and different indicators of employment (section 3), of the relationship between poverty and real wages (section 4), as well as the different ways in which employment and real wages adjusted during the Indonesian crisis and developments since then. Finally, section 5 summarises the key findings, while section 6 discusses some policy options. 2. The multiple dimensions of poverty in Indonesia: pre- and post-crisis perspective This section will briefly review trends in consumption poverty, capability deprivation and vulnerability and powerlessness. All of them represent important aspects of impoverishment in Indonesia. Consumption poverty Perhaps the most researched, and publicised, aspect of poverty in Indonesia draws on the so-called income/consumption approach. The latter entails the use of expenditure/distribution data to derive a poverty line below which people are classified as poor, that is people who have inadequate purchasing power to acquire the basic necessities of life. When expressed as a proportion of the population, such a statistic becomes the well known and widely used head-count ratio. In the case of Indonesia, current consumption expenditure, rather than current income, is used to estimate the head count ratio. Estimates of consumption poverty are routinely provided by the Central Bureau of Statistics (CBS) and may be regarded as the official estimates for Indonesia. Longterm trends in pre-crisis poverty in Indonesia based on the income-consumption approach, is shown in Figure 2.1. Both CBS and alternative estimates are shown. 21 The sharp fall in poverty between 1976 and 1996 is evident, whether the CBS or the alternative approach is used. The only exception seems to be the period when the alternative estimates suggest a slight increase in the head count ratio, but with the CBS measures showing a discernible decline The alternative estimates take the 1996 poverty line and projects it backwards and forwards using the prevailing inflation rates. The procedure is described in Dhanani and Islam (2002). The CBS poverty lines in Figure 2.1, on the other hand, pertain to poverty lines derived for each period of observation. 22 Note that the estimates presented in figure 2.1 vary in one significant respect from figure 2.2. In figure 2.1, the head count ratio is shown to be per cent for 1996, while it is shown to be 18 per cent for 1996 in figure 2.2. This discrepancy is due to the fact that CBS has used two different poverty standards, the so-called old poverty standard based on 1996 data (that is embodied in figure 2.1) and a new poverty standard based on 1998 data (that is embodied in figure 2.2). The numbers depicted in figure 2.2 are readily comparable across time as they are based on a common poverty line. On the other hand, the statistics in figures 2.1 are less comparable across time because they are based on periodspecific poverty lines.

57 Consumption poverty, Indonesia, , CBS and alternative estimates Poverty incidence (%) CBS Alternative estimates Figure 2.1 The most conspicuous aspect of more recent poverty statistics in Indonesia, that is, immediately prior to and after the crisis, is its volatility. These are captured in figures 2.2 to 2.4. Figure 2.2 presents data on consumption poverty that is based on CBS estimates between February 1996 and August 1999, while Figure 2.3 shows alternative estimates of consumption poverty between February 1996 and February In both cases, consumption poverty rises sharply between 1996 and 1998, signifying the height of the crisis, but then begins to taper off quite quickly. For example, the head count ratio in September 1998 according to the CBS figures doubles from its 1996 level of 18 per cent to 37 per cent, but by August 1999, the value of the head count ratio has apparently returned to its pre-crisis level. Consumption poverty in Indonesia, , CBS estimates Poverty incidence (%) (Feb) 1998(Feb) 1998 (Dec) (Feb) 1999 (Aug) Series1 Figure 2.2

58 Consumption poverty in Indonesia, , alternative estimates Poverty incidence (%) (Feb) (Sep) (Dec) (Feb) (Aug) (Feb) Series1 Figure 2.3 Figure 2.4 shows trends in the poverty severity index for Indonesia for the period (derived from SUSENAS 1996, 1998 and 1999). 23 The data as presented shows that severity of poverty shot up between February 1996 and December 1998, but tapered off by February 1999, mirroring the trend in the overall head count ratio (see Figures 2.2 and 2.3). While the poverty severity index apparently returned to their pre-crisis levels by August 1999 in urban areas, this was not the case in rural areas, with the value of the severity index for rural Indonesia remaining a little higher than its pre-crisis level. Figure 2.4 Trends in Poverty Severity Index (P 2 ), Urban Rural 0.00 Feb '96 Dec '98 Feb '99 Aug '99 23 Poverty severity indices capture both the depth of poverty and the degree of inequality among the poor. See Foster et al (1984).

59 Income poverty using international poverty lines An alternative way of depicting income/consumption poverty is to use so-called international poverty lines (IPLs). This method has been popularised by the World Bank and entails the use of dual poverty lines, a one dollar a day threshold and two dollars a day. In other words, people earning either one dollar a day or two dollars a day would be classified as poor. The former may be considered to be a measure of extreme poverty, the latter a measure of the incidence of relative poverty, assuming a generous threshold by developing country standards. 24 IPLs have been criticised for their lack of reliability vis-à-vis national poverty lines, 25 but the point is that IPLs provide a basis for global comparisons of poverty. Subject to this caveat, what has happened to the evolution of income poverty in Indonesia using the World Bank standards? Figure 2.5 depicts some pertinent information that is derived from data maintained by an international agency. 26 According to both thresholds, poverty fell sharply between 1984 and 1996, and rose conspicuously between 1996 and 1999 under the impact of the 1997 crisis, but is projected to have stabilised since then. Obviously, in terms of the two dollars a day benchmark, poverty in Indonesia is widespread, being around 60 per cent in Poverty incidence (%), Indonesia, International poverty lines, Figure Poverty incidence (%) $2 a day (% below ) $1 a day % below ) Capability poverty The income/consumption approach measures inadequate purchasing power as a measure of deprivation. Even if one could guarantee adequate purchasing power, it does not guarantee the elimination of actual deprivation. Thus, for example, people in a particular community could have the potential capacity to acquire basic education, but because of, say, lack of access to appropriate facilities may not be able to acquire basic education (at least at reasonable cost). Hence, while people in this hypothetical community may be non-poor based on non-monetary dimensions, they would still be deprived in their inability to be educated. The converse may also be true. Someone could be deemed poor in a monetary sense, and yet, because of, say, an activist 24 Agenor (2002:4). One could argue that the higher IPL captures estimates of the near-poor. 25 See David et al (1999). 26 Asia Recovery Information Centre Data Base. Unfortunately, the Centre does not make it clear whether estimates beyond 2001 are projections. The study assumes that this is the case.

60 public policy that provides free basic education, may have the capacity to acquire rudimentary schooling. As the above admittedly contrived example shows, it has now become commonplace to view poverty as encompassing not only material deprivation but also low achievements in education and health. 27 A more general statement of this notion is that poverty should be conceptualised as capability deprivation, that is the severe restriction on capabilities that a person has the substantive freedoms he or she enjoys to lead the kind of life he or she values. 28 The ability to be literate and healthy is thus not only a means to an end, enabling people to escape from the trap of material deprivation by being productive members of society, but are ends in themselves. Indeed, such a line of thinking regards the elimination of capability deprivation as being at par with the acquisition and consolidation of core human rights. The right to be literate and healthy is as fundamental as the right to freedom of expression. The current international development goals recognise the notion of poverty as going beyond material deprivation. 29 Thus, the international development community, based on discussion at various United Nations conferences in the 1990s, declare the need to:! Reduce by half the proportion of people living in extreme income poverty (living on less than $1 a day)! Ensure universal primary education! Eliminate gender disparity in primary and secondary education (by 2005)! Reduce infant and child mortality by two-thirds! Reduce maternal mortality rates by three-quarters! Ensure universal access to reproductive health! Implement national strategies for sustainable development in every country by 2005, in order to reverse the loss of environmental resources by Not all of these goals directly relate to poverty eradication goals (for example, the reference to environmental management), but the declarations make it clear that simply reducing income poverty is not enough. Furthermore, it is not merely an issue of focusing only on the education and health indicators prescribed in the international development goals. One could argue that capability deprivation should also take account of such indicators as malnutrition, access to safe drinking water, access to sanitation facilities and access to shelter of acceptable quality. One issue that emerges in measuring capability deprivation is that the components of it are so diverse that they cannot just be added up in the way that monetary units can. UNDP has, over a period of time, attempted to offer a composite measure of capability deprivation. Known as the human poverty index (HPI), the framework shares the spirit of the UNDP s well known human development index (HDI). The HPI is an aggregation of several indicators of deprivation: the percentage of people not expected to survive to 40; the illiteracy rate; percentage of people without access to water and health services; percentage of underweight children under the age of five. One caveat that should be noted in interpreting estimates based on HPI is that the framework assigns arbitrary weights to the different components of capability 27 World Bank (2000), World Development Report 2000/2001, p Sen (1999). 29 See, for example, the OECD (1996).

61 deprivation by assuming that all the components are equally important. Furthermore, the HPI does not give a measure of the proportion of people living in human poverty in a way that could be compared with the proportion living in income/consumption poverty. Table 2.1 Key indicators of capability deprivation, Indonesia, HPI Population not expected to survive to age 40 (%) Adult illiteracy rate (%) Population without access to safe water (%) Population without access to health services (%) Under-nourished children under the age of five (%) Source: BPS/BAPPENAS/UNDP (2001: annex 1.2, p.15) Table 2.1 offers an illustration of the use of the HPI framework in Indonesia. From the above table, one could infer that, the HPI improved between 1990 and 1996, but remained stable between 1996 and 1999, despite the onset of the 1997 financial crisis. In terms of individual indicators, the degree of capability deprivation varies from 52 per cent (more than half of the Indonesian population do not have access to safe water) to 12 per cent (this being the proportion of the adult population that are deemed to be illiterate). The extent of child malnutrition is also quite high (30%), particularly when judged against the benchmark that a well-nourished population typically has around three per cent of children who suffer from some form of malnutrition. 30 There have been improvements in the various indicators between 1990 and 1999, with the conspicuous exception being the deterioration of the population without access to health services (it jumped from 10.6 per cent to 21.6 per cent between 1996 and 1999). Other important indicators of the well being of mothers and children pertain to primary schooling, infant and maternal mortality. Target reductions in these indicators are now part of the international development goals. The evidence, as at 1999, suggests that net enrolment rates at the primary level is around 95 per cent and infant mortality is around 47 (per thousand live births). One weak spot is the maternal mortality rate, which at 450 per 100,000 live births, is one of the highest in South-east Asia Sadaah et al (1999:2). 31 BPS/BAPPENAS/UNDP (2001: ch.3).

62 How do these figures compare with estimates of consumption poverty? At around 18 per cent of the population, consumption poverty in 1999 is broadly aligned with some indicators of capability deprivation, but is notably below the levels recorded by other indicators (especially child malnutrition and access rates to safe water). Vulnerability There is a growing realisation that people can move in and out of poverty. 32 This has in turn inspired greater attention on the notion of vulnerability as an important dimension of poverty. Vulnerability may be defined as...the probability or the risk today of being in poverty or to fall into deeper poverty in the future. 33 Such vulnerability to a decline in well-being may be triggered by shocks at the micro /household level (e.g. illness, death of breadwinner in the family), at the meso/community level (e.g. a bad harvest, a collapse in the price of key products, environmental degradation) and the macro/economy wide level (e.g. the 1997 Asian financial crisis). Until the onset of the Asian financial crisis in 1997, vulnerability was not a widely recognised phenomenon in Indonesia. It was simply assumed that cross-section surveys offered an adequate description of the phenomenon of poverty. Certainly, this assumption appeared quite plausible during the long economic boom of the Suharto era, when popular expectations of uninterrupted improvements in living standards held sway. In the more sombre circumstances of the post-crisis period, when many from the emergent middle class joined the ranks of the poor, even if temporarily, it became evident that it was necessary to assess the risk of individuals and households falling into poverty or getting even poorer. Reflecting this mood to capture movements in and out of poverty, a spate of studies has explored the prevalence of vulnerability in Indonesia. 34 Some of these studies have been made possible by the fact that various surveys were carried out that tracked the same set of households over time. For example, the 100 Villages survey interviewed a panel of over 10,000 households four times from August 1998 to October A mini-susenas (that is, a smaller version of the full SUSENAS) was carried out in December 1998 and about 7,500 households were re-interviewed in August An alternative technique that does not rely on panel data has also been developed. It relies on regression techniques to estimate a predicted mean consumption level, predicted consumption volatility and vulnerability to poverty for each household. 37 The estimates are derived from cross-section data on expenditures and a number of household and community characteristics. In the case of Indonesia, this has entailed the merging of SUSENAS and PODES (village potential) data, the latter being a 32 Jo and Ellwood (1983) are among the pioneering studies that examined the dynamics of poverty. 33 Coudouel et al (2001: 32). 34 See Skoufias et al (2000), Pritchett et al (2000), Chaudhuri et al (2001), Suryahadi and Sumarto (2001), Widyanti et al (2001). 35 The survey is conducted by CBS and funded by UNICEF. It should be noted that the 100 Villages survey is not meant on a representative sample of the country. It was designed (in 1994) to offer a representation of various parts of the rural economy. Hence, the conclusions of any study based on this survey needs to be treated with some caution. 36 This was conducted by CBS and funded by UNDP. 37 Chaudhuri et al (2001: 1).

63 complete enumeration of villages in terms of such characteristics as size, population, infrastructure and local industries. 38 An illustration of the key findings on vulnerability is provided in Tables 2.2 to 2.4. The first table draws on poverty transition matrices based on the 100 Villages Survey, the others on regression techniques applied to SUSENAS. Table 2.2 Persistent, chronic and transient poverty, Indonesia, 1999 Poverty category Incidence (%) Extent to which mean real per capita consumption is % below (-) or above (+) poverty line Variance as measured by standard deviation Persistently poor Chronically poor Transiently poor Never poor Source: Adapted from Wideband et al 2001, Table 5, and p.11. The original data source is the 100 Villages Survey. As Table 2.2 shows, 17.5 percent of the poor are persistently poor, that is, they are always poor over the periods observed are chronically poor, that is, they are usually poor over the period observed per cent transiently poor, that is, they are sometimes poor over the periods observed per cent are never poor over the periods observed. The vulnerable population would include the transiently poor implying that the prevalence of vulnerability is larger than the prevalence of poverty per se. It is of some interest to note from the table that income variability (as measured by the standard deviation) is in fact highest for those who are never poor. It is also of some interest to note that the persistently poor and the chronically poor who form the poorest of the poor in terms of this framework have average consumption that are 10 to 30 per cent below the poverty line. Table 2.3 revisits the above poverty categories using SUSENAS-cum-PODES data. It pertains to two years: 1996 and 1999.Both current consumption poverty and vulnerability jumped between the two years. As can be seen, the head count ratio goes up from 15.6 per cent to 27.4%. 39 At the same time, the prevalence of vulnerability went up from 18.1per cent to 33.7%. This suggests that vulnerability is not only significantly higher than current consumption poverty, but it also went up more sharply than the former with the onset of the 1997 financial crisis. Table 2.3 Poverty and vulnerability, Indonesia, Poverty category Change Transient poor Chronic poor Total Total vulnerable group Source: Adapted from Suryahadi and Sumarto, 2001, Table 1, p Suryahadi and Sumarto (2001). 39 The discerning reader will note that these estimates vary from the CBS estimates presented earlier. This is due to the use of a different poverty line and, to some extent, the merging of two datasets.

64 Table 2.4 presents a simple matrix in which the correspondence between poverty and vulnerability is shown more explicitly. Clearly, there is a great deal of overlap between the proportion in poverty and the proportion that are vulnerable, but the two do not coincide perfectly. Hence, while 78 per cent of the currently poor are vulnerable, 32 percent of the non-poor are also vulnerable. 40 It is the inclusion of the non-poor in vulnerability estimates that explains why in Indonesia, as in many other developing countries, the proportion of the population that faces a non-trivial risk of future poverty is significantly higher than the proportion currently observed to be poor. Table 2.4 Poverty and vulnerability, a simple matrix, Indonesia, 1999 Non-poor Poor Non-vulnerable 64% 22% Vulnerable 32% 78% Source: Adapted from Chaudhuri et al (2001), p.1. Note that the prevalence of vulnerability is 45% according to this study. The original data sources are mini-susenas December 1998 and 7,500 households re-interviewed in August Some qualifications need to be highlighted as part of concluding the discussion of vulnerability in Indonesia. The data provided here do not directly address the issue of seasonality as a distinct facet of vulnerability. There would be households that would be vulnerable only in bad years (e.g. a bad harvest), and others that are vulnerable irrespective of seasonal effects. Such a subtle distinction is not really captured by the available data suggesting the need for further research along this direction. 41 Powerlessness Recently, attempts have been made in Indonesia as part of a worldwide study that has tried to listen to the voices of the poor. 42 The purpose is to elicit responses from the poor through small-group discussions. This in turn allows one to obtain perceptual data on direct experiences of poverty. An important theme that emerges from the above exercise is that the poor in Indonesia, as in many other developing countries, feel! A strong sense of exclusion from decision-making processes that directly affect their lives! A strong sense that women have no voice in decision-making processes! A strong sense of distrust of state institutions It is worth noting that different studies generate different estimates of vulnerability. In Pritchett et al (2000), the prevalence of vulnerability varies from 30% to 50%. In Chauduri et al (2001) it is 45%. In Sumarto and Suryahadi (2001), it is estimated to be 34%. The differences in the estimates stem from different definitions of vulnerability and use of different data. 41 See Dercon and Krishnan (2001) for a study of the impact of seasonality on poor households in Ethiopia. 42 Narayan et al (2000). For Indonesia, the relevant sources are: Consultations with the poor in Indonesia as compiled by Mukherjee (2000) and Bangsal and Sari (2000). 43 The discussion draws on Mukherjee (2000) as reported in World Bank (2000, pp24-25, 55).

65 Exclusion from local decision-making processes Respondents in poor communities that were surveyed were resigned to the fact that they were excluded from local-level decision-making processes. As one person in East Nusa Tengarra noted: When the Village Development Council calls a meeting, all the decisions about a project or program have already been made Even an old, aged and blind man will be invited along to the meeting, along with rest of the village men. They would all remain silent and listen Elsewhere, someone observes: Those who are able to have influence are only people with high social status, the village officials or the rich. Lack of voice of women in local decision-making processes Here are some typical observations:! Women s groups everywhere confirmed that women are neither invited nor expected to attend village meetings which are often conducted at male-only events and places such as Friday post-prayer meetings or the Village Forum! Where women have their community gatherings and activities, they are often there as conduits for implementing development programs or self-help initiatives, with little or no decision-making roles! When women are invited to the meeting (of the Village Development Council), they are only given the task to prepare and serve the refreshments! Both women and men seem to concur that community decisions are rights and responsibilities of menfolk. Women s role is only to accept and implement them. Lack of trust of state institutions The poor trust an institution that is perceived to be transparent, fair, sticks to commitments made, offers assistance without hidden agendas and believes in the capacities and convictions of the poor. When judged against such criteria, a number of faith-based NGOs were singled out, but none of the state institutions were identified. In particular, rural women consistently excluded village councils and government programs for poverty alleviation as worthy of their trust, and as being open to community influence. While listening to the voices of the poor in Indonesia is, in and of itself, a fascinating exercise and corroborates what has been widely recognised as key attributes of poverty, the challenge lies in translating such perceptual data into meaningful indices that can be readily tracked. Currently, one lacks voice and power indicators that can be readily compared to consumption/income indicators of poverty, capability deprivation and measures of vulnerability. Perhaps greater attention needs to be paid to the design of national household surveys in the future that would

66 routinely incorporate components that seek to elicit information on the civic and political activities of the poor. Certainly, some data that are available in this form suggest that well over 80per cent of low-income households in Indonesia do not participate in political and civic activities. 3. Poverty and the labour market: the role of employment The review of the multiple dimensions of poverty in Indonesia has shown that:! There was an impressive long-run decline in consumption/income poverty over two decades. This conclusion is valid irrespective of the estimation method used.! Consumption poverty shot up during the crisis, but it was transient, with the head count ratio tapering off to its pre-crisis level by mid-1999.! Indicators of capability deprivation exhibit greater stability and were apparently unaffected in any major way by the 1997 financial crisis.! Vulnerability in Indonesia is more significant than current consumption poverty, and worsened more sharply than the latter during the crisis.! Perceptual data gleaned from direct experiences of poverty suggest that the poor feel alienated and excluded from local decision-making processes a phenomenon particularly affecting women and that they lack trust in state institutions. The purpose of this section is to explore the extent to which the evolution of consumption poverty (and consumption-based indicators of vulnerability) can be reconciled with developments in the labour market, where such developments are assessed from the perspective of a range of employment indicators. The discussion is reinforced by developing a cross-section profile of the employment /skill and demographic characteristics of the poor. Evidence based on capability deprivation and powerlessness are more difficult to accommodate within the framework of labour market analysis, as they suggest that poverty has deep social, political and institutional roots. Finally, the section revisits some important debates in the Indonesian context. Do employment adjustments during the crisis represent a high degree of labour market flexibility that, in turn, mitigated the impact of the crisis on consumption poverty? Does the liberalisation-cum-deregulation reform agenda in a context of regional decentralisation that has gained momentum in the wake of the 1997 crisis entail adverse consequences for the employment prospects of ordinary Indonesians? Changes in employment structure: overview and implications for poverty Drawing on a variation of key indicators of the labour market (KILM) devised by the ILO, a brief account is offered of the following employment indicators in the Indonesian context: (a) labour force participation rate; (b) employment-to-population ratio; (c) status in employment; (d) part-time workers; (e) the size of the urban informal sector; (f) indicators of unemployment and underemployment; (c) the sectoral composition of employment (g) the skill-levels of the work-force. 44 These indicators enable one to develop an understanding of the structure of the Indonesian 44 See Irawan, Ahmed and Islam (2000) for an application of the ILO s KILM framework to Indonesia.

67 labour market. This in turn sets the context for appreciating the poverty-employment nexus. Table 3.1 captures the various ways in which one can track the transformation of the Indonesian labour market over the 1990s. The Indonesian labour force grew from 75 million in 1990 to 96 million in 2000 entailing an average annual growth rate of 2.5 per cent. In terms of absolute numbers, this suggests an addition of two million members to the workforce annually. Hence, the challenge is to ensure that the economy through a combination of the natural processes of growth and policy interventions is able to generate at least two million jobs to maintain flow equilibrium in the labour market. Table 3.1 Overview of the Indonesian labour market, Size of the labour force (mil) Labour force participation rate (%) Employment to population ratio Female labour force (%) Youth labour force (%) Urban labour force (%) Formal employment (%) Urban informal sector employment (%) Part-time employment (%) Female parttime employment (%) Time-related underemploy ment rate (%) Youth unemployme nt rate (%9 Unemployme nt rate (%) Source: Estimated from SAKERNAS, the national labour force surveys, various years. Some indicators also extracted from Puguh, Ahmed and Islam (2000), statistical appendix. Some indicators for 2000 (part-time employment, time-related underemployment, youth unemployment rate, size of the urban informal sector) are estimates based on historical trends. Direct computations of these indicators will require special tabulations based on unpublished SAKERNAS data.

68 In some respects, one cannot detect a great deal of change in the labour market; in other respects, the changes are evident. Thus, the labour force participation rate as well as the employment-to-population ratio - remained relatively stable, the former entailing modest fluctuations between 66 and 68 per cent. The gender composition of the workforce was also steady, being around 38 per cent as was the proportion of young workers among the total labour force; it fluctuated between 21 and 23 per cent. As a contrast to the stable features of the labour market noted above, one observes significant urbanisation and formalisation of the workforce. The relative size of formal sector employment rose from 28 per cent to 38 per cent on the eve of the crisis (1996), while the relative size of the urban labour force increased from 26 per cent to 36 per cent. These developments are consistent with the sustained decrease in consumption poverty documented in section 2 of this study. A reallocation of labour from lowproductivity, low-paid jobs in the rural economy to better-paid employment in the urban and formal economy is one well-known route via which ordinary workers and their dependents can improve their living standards. As far as different indicators of under-utilisation and casualisation of the workforce is concerned, the picture is mixed. Part-time employment remained roughly steady between 1990 and 1993 (28 to 29 per cent) and rose moderately thereafter (33 per cent). The relative size of the urban informal sector hardly changed (gravitating around 45 to 43 per cent). There is also no clear trend of an increase in the proportion of women who worked part-time. Thus, these measures of casualisation of the workforce do not show any consistent deterioration in the pre-crisis period. There is little evidence as well that, prior to the crisis, involuntary underemployment (or timerelated underemployment) was widespread, being around eight to nine per cent. This suggests that the sustained decline in consumption poverty in the rapid-growth Suharto era represented genuine social gains. What about unemployment and its implications for poverty? In the pre-crisis period, it was below five per cent. The conventional wisdom is that the aggregate unemployment rate is an inappropriate indicator of the state of the Indonesian labour market, given its stage of development. In the absence of a comprehensive unemployment benefits scheme, joblessness for protracted spells is an option that only few can afford. Not surprisingly, the long-term unemployment rate (defined as those unemployed for a year or more) was around one per cent between 1990 and The available statistics to be discussed more fully at a later stage - typically suggest that unemployment in Indonesia is largely a voluntary phenomenon, concentrated among educated youth. Certainly, as can be seen, the youth unemployment rate was far above the overall unemployment rate. Nearly, 70 per cent of the unemployed are represented by young members of the workforce. The current debate on youth unemployment has tried to disentangle choosy youth from unwanted youth. 45 While some commentators suggest that the youth 45 See Manning and Junankar (1998).

69 unemployment rate largely reflects a case of young graduates queuing for reasonably secure public sector jobs, 46 others have warned that not all the unemployed young members of the workforce should be classified as choosy youth coming from reasonably affluent backgrounds. There is a cohort who are the victims of penurious socio-economic circumstances. Table 3.1 also enables one to focus on the impact of the crisis on the Indonesian labour market. A scrutiny of the information embedded in table 3.1 would suggest that, in terms of some indicators, the impact of the crisis on the labour market was modest and, hence, seemingly at odds with the view that there was a sharp increase in poverty when the crisis was at its worst in Thus, the unemployment rate rose only moderately (from 4.9 per cent in 1996 to 5.7 per cent in 1998), although there was a more discernible increase in 2000 (over 6 per cent). 47 The incidence of involuntary underemployment actually went down between 1996 and 1998, while the incidence of part-time employment hardly changed. On the other hand, the size of the urban informal sector went up noticeably - from 43 per cent to 46 per cent between 1998 and 1996 and has not yet (that is, by 2000) reverted to the pre-crisis level. At the same time, there has been an alarming increase in the incidence of youth unemployment between 1996 and 1999, probably reflecting constrained job opportunities in the public sector induced by the growth slow-down in the post-crisis period. A clearer perspective on the impact of the Indonesian crisis on the labour market can be gleaned from table 3.2. The latter exhibits the sectoral composition of the workforce over the 1990s. The employment share of the agricultural sector, after falling on a sustained basis from 55 per cent in 1990 to 43.5 per cent in 1996 (and 41 per cent in 1997), experienced an abrupt reversal: its employment share went up to 45 per cent (approaching the level achieved in 1995). Employment in the manufacturing sector contracted as well: the employment share of the sector declined from 12.6 per cent to 11 per cent between 1996 and Thus, the impact of the crisis was reflected not through a big jump in unemployment, underemployment or even parttime employment, but in an involuntary shift of the workforce into agricultural activities as well as into the informal economy of the urban sector. Table 3.2 Sectoral composition of the workforce (%), Indonesia, Agriculture (Food crop) Industry (Manufact.) Services (Trade) (Non-trade) Source: See Table 3.1 It remains to be seen to whether some of these sectoral reallocations of labour will be persistent. As the estimates for 1999 show, the changes in the employment share of agriculture have resumed their historical pattern, but estimates for 2000 show that it 46 Dhanani (2001a). 47 Preliminary estimates for 2001 suggest that the unemployment rate is 8.1 per cent (Dhanani, 2002, table 2.1, p.3).

70 has slipped back to 45.1 per cent. 48 On the other hand, manufacturing employment has grown to the point where its employment share has gone up to 13 per cent. It is also noteworthy that by 2000 (table 3.1), the share of formal sector employment has not yet reached its pre-crisis level, suggesting the muted, but lingering, impact of the 1997 crisis on the Indonesian labour market. A discussion of the Indonesian labour market would be incomplete without some reflection on a key aspect of the employment structure: the extent to which rapid growth in the pre-crisis era was accompanied by a deepening of the skills of the labour force. Table 3.3 enables one to offer a commentary. While recognising that the process skill formation emanates from multiple sources, table 3.3 uses the educational attainment of the workforce as a credible approximation of the skills embodied in the workforce. It is clear that the rapid-growth, pre-crisis era was accompanied by substantial upgrading of the educational profile of Indonesian workers. The proportion of less educated/skill workers (defined as those who have primary education and less) fell from 87 per cent to 79 per cent between 1990 and 1996, with a corresponding increase in the share of the educated/skilled workforce (defined as those with at least secondary education) from 14 per cent to 21 per cent. These structural changes in the Indonesian labour market juxtaposes rather well with the sustained decline in poverty witnessed during the pre-crisis period. As a subsequent discussion will demonstrate more explicitly, there is a strong correlation between educational attainment and poverty incidence at the household level. Thus, as more Indonesians gained access to the education system, it provided them with an escape-route from impoverishment. Table 3.3 Educational attainment of the workforce (%), Indonesia, Less educated/skil led workforce No schooling Primary - incomplete Primary Lower secondary Educated/sk illed workforce Upper secondary Tertiary Source: See Table Preliminary estimates for 2001 suggest that the employment share of agriculture is 43.8 per cent while the employment share of manufacturing is 13.3 per cent (Dhanani, 2002: table 2.8, p.12).

71 Changes in employment elasticity: overview and implications for poverty An important question pertains to the extent to which the employment-intensity of rapid economic growth, especially in the manufacturing sector, played a role in influencing trends in poverty in the Suharto era. Provided a stable relationship exists between output growth and the job-creation rate a variation of so-called Okun s law observed in industrialised countries it should be possible to estimate employment elasticity. 49 The latter, as is well known, is a useful summary statistic for computing the employment intensity of economic growth. From the perspective of poverty analysis, what matters is not growth of output per se, but the extent to which it is employment-friendly. Higher employment elasticity translates to the fact that, for any given growth rate, the job creation rate is higher. However, care should be taken in interpreting the normative implication of this proposition. It is possible to have excessive employment creation that is at the expense of growth of productivity. For example, an employment elasticity that exceeds unity means a proliferation of lowproductivity jobs. Thus, employment-friendly growth should really mean the expansion of durable and productive work opportunities. Bearing such analytical caveats in mind, consider table 3.4, which shows employment elasticity in manufacturing for several sub-periods: , and Employment elasticity was quite low (0.3) in the 1970s, rose appreciably in the 1980s (0.8) and tapered off a little with the onset of the 1990s (0.7). These estimates are consistent with conventional interpretations of the evolution of the employmentpoverty linkage in the Indonesian context. In the 1970s, Indonesia s oil wealth was used to fund public infrastructure and rural development, but the manufacturing sector remained small, inefficient and protected. This constrained its capacity to act as a locomotive for employment creation, although the oil wealth-driven public investment policy of the 1970s did reap some social dividends, given that it heralded the onset of a remarkable decline in mass poverty. As the oil price boom dissipated in the early 1980s, the Suharto government was forced to turn to outward-oriented industrialisation strategy that began in earnest from the mid-1980s. The much higher values of manufacturing employment elasticity in the 1980s support the notion that the manufacturing sector was beginning to play a more significant role in spearheading employment creation by drawing in less-skilled workers from the rural and informal economy to the more formal economy. Some studies have documented the fact that manufacturing employment elasticity declined appreciably in the latter half of the 1990s, raising the possibility that the capacity of industrialisation to sustain large-scale job creation was becoming constrained. Table 3.5 conveys pertinent information. As can be seen, employment elasticity in the formal sector (represented by large and medium-scale enterprises) was an average of 0.5 for the period, but fell from 0.8 in the period to 0.3 in the period. Whether this represents the onset of a durable trend remains to be seen. Table 3.4 also provides useful information on employment elasticities on sub-sectors within manufacturing. For the period as a whole, the sub-sectors with high employment elasticities (above 0.5) are: garments, 49 See Padarino and Vivarelli (1997).

72 furniture, electrical goods and footwear. It is noteworthy that these are also subsectors with a good deal of export-orientation. Current research on the employment-intensity of economic growth in Indonesia has also reflected on methodological complications entailed in the estimation of employment elasticity. One study has shown that the conventional way of estimating employment elasticity, which is simply an ex-post procedure, may engender unreliable estimates. 50 They seem to fluctuate widely depending on the time periods used. Econometric estimates of employment elasticity using pooled province-level data generate more stable estimates and suggest an employment elasticity of approximately 0.7 for the period. Table 3.4 The evolution of employment elasticity in Indonesian manufacturing Period Source: Extracted from Islam, R (2001: table, p4) Employment elasticity in manufacturing Table 3.5 Sectoral employment elasticities based on medium and large-scale enterprise survey (three-digit ISIC classification), Indonesia, Sectors All medium and large-scale manf Food Manf Textile Garment Furniture Electrical goods Transport equip Metal Footwear Source: Islam and Nazara (2000a), table 7, p.18. Linking poverty to the employment characteristics of households: a profile The overview of changes in the Indonesian labour market, as captured in a range of employment indicators, suggest that formalisation and urbanisation of the workforce together with an upgrading of the educational attainment of workers are fully consistent with the sustained declines in poverty in the pre-crisis period. At the same time, there is little evidence that casualisation of the workforce became pronounced, while other indicators, such as the gender composition of the workforce and the relative employment share of young workers remained roughly stable. There is also some evidence that growth under the Suharto regime became more employmentfriendly. The 1997/1998 crises reversed the poverty-reducing structural changes in the Indonesian labour market, with a forced movement of workers into the agricultural sector and into the informal segment of the urban economy. It remains to be seen whether these reversals will turn out to be persistent, with data for 1999 showing a 50 See Islam and Nazara (2000a). The ex-post procedure really entails dividing observed GDP growth with observed employment growth.

73 resumption of historical trends, but 2000 indicating otherwise. The unemployment rate has gone up moderately, largely reflecting a high incidence of youth unemployment. Involuntary underemployment was not widespread in the Suharto era. The crisis does not seem to have brought about a lasting worsening in the incidence of underemployment. The overview of the structural changes in the Indonesian labour market thus sets the context for further analysis of the poverty-employment linkage at the household level. Pertinent information on this issue is displayed in tables 3.6 to 3.8. In table 3.6, one is able to discern the incidence of poverty by main sector of occupation and the relative importance of each sector of occupation in nation-wide poverty. Table 3.7 replicates the estimates, this time from the perspective of the educational level of households. Table 3.8 draws together a wide array of information on the demographic, skill and employment status of poor households and compares it with non-poor households. The information in the above tables is supplemented by figures 3.1 to 3.4 that depict the employment/skill characteristics of households at different points on the expenditure distribution. More specifically, a quintile approach is used in which the employment/educational attributes of households in different quintiles poorest 20 per cent to richest 20 per cent are compared. The figures are undergirded by the plausible assumption that the poorest quintile largely coincides with the population of households officially classified as poor. The information embedded in the various tables does not hold surprises. The bulk of the Indonesian poor can be found in the agricultural sector. For example, in 1999, the poverty incidence in agriculture was approximately 40 per cent vis-à-vis a poverty rate of only five per cent for household heads who identified finance as their main sector of occupation. Furthermore, 58 per cent of nation-wide poverty can be accounted for by the agricultural sector. Thus, any strategy of poverty reduction in Indonesia will have to recognise the widespread existence of poor households in agricultural activities. Table 3.7 examines the linkage between poverty incidence and the educational attainment of household heads. There is a strong correlation between the two variables, although one should hasten to add that correlation based on cross-section data do not necessarily imply causality. The bulk of poverty in Indonesia resides among household heads who have attained primary education or less. For example, in 1999, poverty rates varied between 39 per cent and 48 per cent for groups with primary education or less. Such groups in turn accounted for the bulk of nation-wide poverty in Indonesia. At the other extreme, poverty among household heads with tertiary education was two per cent in 1999, while for those with secondary education or more, the incidence varied between five and eight per cent. It thus seems that the upgrading of the Indonesian workforce during the 1990s was an important mechanism in bringing about the impressive decline in poverty prior to the crisis. This also means that ensuring broad access to the education and training system, particularly at the secondary level, is going to remain a major policy goal for the Indonesian government. Table 3.8 compares the demographic, employment and skill characteristics of poor households with non-poor households. The former have a somewhat larger household

74 size (4.6 persons vs. 4.0 persons), but there is no evidence that poor households are either substantially younger or older than non-poor households. There is also no clear evidence of feminisation of poverty in the sense that a moderate 16 per cent of households headed by females are classified as poor. In terms of employment status, it appears that informal self-employment is a major characteristic of poverty. Thus, 56 per cent of household heads who were selfemployed were classified as poor vis-à-vis 30 per cent of household heads who were classified as employees were regarded as poor. In terms of hours worked per week, there is some difference between poor and non-poor households. Thus, poor households in 1999 worked an average of 33 hours per week compared with non-poor households who worked an average of 36 hours per week. Table 3.8 shows a discernible gap between poor and non-poor households in the area of education. Mean years of schooling for poor households were 6.0 years in 1999 vis-à-vis 7.4 years for non-poor households. Table 3.6 Poverty incidence by main sector of occupation (%), Indonesia, Feb 1996 Feb 1996 Feb 1999 Feb 1999 Poverty incidence Contribution to Poverty incidence Contribution (%) total poor (%) (%) total poor (%) Agriculture Trade Manufacturing Civil, social and private services Transport and Communication Construction Receiving transfer Mining & quarrying Others Finance Source: Pradhan et al, Unpublished estimates. Original source is the national socio-economic survey or SUSENAS. Table 3.7 Poverty incidence by educational level of household head (%), Indonesia, Illiterate and incomplete primary education Incomplete primary education but literate Completed primary education Completed junior secondary education Completed senior secondary education Completed tertiary education Feb 1996 Feb 1996 Feb 1999 Feb 1999 Poverty incidence Contribution to total Poverty incidence Contribution to total (%) poor (%) (%) poor (%) to

75 Source: As in table 3.6. Table 3.8 Demographic, skill and employment status of poor and non-poor households, Indonesia 1999 Poor households (HH) Average family size % Female headed HH Mean age (years) of HH head Mean yrs of schooling % HH head as own-account worker and unpaid family worker (%) Non-poor households (HH) % HH head as employee Average hrs worked/week Source: Sutanto and Irawan, Unpublished estimates. The original data is from SUSENAS. Figure 3.1: % workforce in informal sector by quintile, Indonesia, Figure 3.2: White collar workers by quintile (%), Indonesia, 1999 (%) Series Series1 0 Q1 Q2 Q3 Q4 Q5 Total 5 0 Q1 Q2 Q3 Q4 Q5 Total Figure 3.3: Educational attainment (mean yrs of schooling) by quintile, Indonesia, 1999 Figure 3.4: Hours worked per week (<20) by quintile, Indonesia, 1999 Yrs Q1 Q2 Q3 Q4 Q5 Total (%) Series1 0 Q1 Q2 Q3 Q4 Q5 Total Figures 3.1 to 3.4 revisit the linkage between poverty and employment/educational attributes of workers from the standpoint of quintiles of households. They are generated from estimates derived from the 1999 national socio-economic survey (SUSENAS). These estimates were compiled by a Jakarta-based research agency using special tabulations. 51 The figures add to the information embodied in the previous tables in the sense that they enable one to see the relationship between poverty and employment characteristics from the perspective of households at different points on the expenditure distribution. As can be seen, there is a 51 See Insan Hitawana Sejahtera (2000). In figures 3.1 to 3.4, Q1 = poorest 20% Q2=richest 20%.

76 preponderance of the poorest 20 per cent of households in the informal sector (Figure 3.1), while the proportion of white-collar workers is rather low for the bottom quintile (Figure 3.2). Furthermore, in terms of mean years of schooling, the gap between the bottom quintile and top quintile is quite substantial: the poorest 20 per cent have a little over five years of schooling vis-à-vis ten years for the richest quintile (Figure 3.3). It is noteworthy that the average educational attainment of the poorest 20 per cent is well below the long-held goal of the Indonesian government that all Indonesians should have at least nine years of schooling. Thus, upgrading the educational attainment of poor Indonesians at least to the national standard remains a major policy challenge. Figure 3.4 shows the distribution of households by quintile in terms of low hours of work per week (less than 20 hours). A discernible pattern emerges: the incidence of low hours of work diminishes as one moves up the richer quintile, this tendency becoming more marked for the top two quintiles. Thus, inadequate hours of work are one attribute of Indonesian poverty. Unemployment and poverty What the above figures and the previously discussed tables do not show is the extent to which unemployment can be related to poverty. Recall from the previous discussion that the conventional wisdom is that unemployment has limited relevance to the understanding of poverty in Indonesia. How much is this contention borne out by the available data? It is clear from table 3.9 that unemployment in Indonesia is largely concentrated among educated workers. In 1999, secondary school leavers had an unemployment rate of 16 per cent, the highest in the country. This is followed by tertiary graduates who exhibited an unemployment rate of 13 per cent. At the same time, the poverty rate for this group is well below the national average and ranges from 13 per cent (secondary school graduates) to two per cent (tertiary graduates). This experience contrasts sharply with the case of less educated workers encompassing those with primary education or less. The unemployment rate for this cohort varies from five per cent (primary school graduates) to less than one per cent (no schooling). The poverty rates for this group are highest in the country (37 per cent to 48 per cent). Thus, the contention that unemployment has limited relevance in understanding poverty in Indonesia seems to be borne out by the available evidence. In general, the poor cannot afford to stay unemployed for too long and have to seek work of any kind to eke out an existence. It would be useful, as part of future research strategy, to probe more fully the relationship between poverty and unemployment. For example, it should be possible to undertake a quintile analysis by identifying unemployment rates for different quintiles and highlighting the extent to which unemployment is a luxury for richer quintiles, but is unaffordable for poorer quintiles. A more careful scrutiny of unpublished SUSENAS and SAKERNAS data along such directions is worth exploring.

77 Table 3.9 Unemployment rate and poverty incidence by educational attainment (%), Indonesia, 1999 Unemployment rate (%) Poverty incidence (%) No schooling Less than primary Primary Secondary Tertiary Source: based on previous tables in this section Employment, labour market flexibility and poverty: revisiting the debate An important debate that has emerged in the Indonesian context pertains to the role that labour market flexibility has played in the pre-crisis period in facilitating sustained improvements in the living standards of workers and in mitigating the social consequences of the Indonesian crisis. It is important to understand this debate in order to develop a fuller appreciation of the poverty-employment nexus. Labour market flexibility may be regarded as a synonym for the textbook case of a competitive or neoclassical labour market. The virtue of the latter is that workers are free to adjust the workforce in response to shifting relative wage and employment opportunities, while firms are free to adjust the workforce in response to shifting profit opportunities. In such a framework, collective bargaining driven by trade unions, strongly enforced hiring and firing decisions, unemployment benefits, minimum wages etc are regarded as undesirable institutional arrangements as they constrain the free choice of workers and firms. More importantly, arrangements meant to protect the interests of workers paradoxically hurt employment opportunities by raising the price of labour above what the market will bear. Despite its appealing simplicity, the analytical construct of labour market flexibility in pre-crisis Indonesia was adapted to the requirements of paternalistic authoritarianism. Thus, the emphasis was on the provision of centrally mandated benefits for workers (especially minimum wages, and later, a formal social security system). This was juxtaposed with a political framework that tightly circumscribed labour rights, while informal systems of social protection through the network of friends and families were seen as playing an adequate and complementary role for workers in the informal economy. The rapid growth of the pre-crisis period and its ability to absorb new entrants to the workforce appeared to justify such a strategy. 52 The notion of labour market flexibility re-emerges in post-crisis Indonesia. Some observers argue that such flexibility enabled workers to cope with the crisis with a degree of resilience that ensured that the social impact of the crisis was contained. Had labour markets been rigid, as in the case of a so-called Keynesian labour market, one would have witnessed large-scale unemployment and poverty Manning (1998), Wiebe (1996), Mason and Baptist (1996) are prominent examples of studies on Indonesia that work within the analytical tradition of labour market flexibility. 53 As Manning (2000: 108) observes: I conclude that labour markets remained highly flexible despite the rapid economic transformation. I suggest that this flexibility is a key explanation for why unemployment and poverty did not rise more than they did during the crisis (emphasis added). See also Tubagus (2000).

78 Both the pre-crisis characterisation of the Indonesian labour market and its post-crisis resurrection overlook a number of problems. To start with, some observers, while sympathetic to the notion of labour market flexibility, drew attention to the stresses and strains afflicting the pre-crisis period. Thus one study highlighted the limits of a paternalistic system that offered substantial centrally mandated minimum wage increases in the 1990s to appease workers. 54 The consequence was a threat to employment opportunities and falling international competitiveness although subsequent evaluations have not reached a clear-cut position on this issue. 55 Another study highlighted the problems caused by the lack of a credible industrial relations system. The mid-1990s were infamous for a sharp increase in labour unrest. 56 Both these studies argued for the revamping of the industrial relations system and recommended democratic forms of enterprise-level bargaining. As far the use of the notion of labour market flexibility in explaining the evolution of poverty both during and after the crisis is concerned, one can express reservations about its validity. 57 In a so-called Keynesian labour market, the primary vehicle for a rise in poverty following a demand shock (such as the one induced by the Indonesian crisis) is a rise in unemployment (given that wages are rigid). In a neoclassical labour market operating in a dualistic, developing economy presumably the Indonesian case the primary vehicle for a rise in poverty as a consequence of a demand shock is a decline in real wages reflecting a reallocation of labour to low productivity activities. Indeed, employment adjustments during the Indonesian crisis followed this classic pattern (as was discussed previously) and was accompanied by a steep fall in real wages (as will be discussed at a later juncture). Whether the unemploymentdriven increase in poverty (the Keynesian case) will be higher than the real wagedriven increase in poverty (the neoclassical case) is ultimately an empirical issue that cannot be determined a priori. A better way of coming to grips with the understanding of the evolution of poverty during the crisis is to eschew pre-conceived notions of the virtues of labour market flexibility and carefully re-visit the current state of knowledge on what could have happened to the poor. It now appears that the sharp increase in consumption poverty and the plummeting of real wages during the crisis period was a tragic manifestation of a collapse in the purchasing power of Indonesians as a result of the inflation shock of If one combines the 1998 episode with the fact that a significant proportion of Indonesians are clustered around the poverty line, then relatively small shifts in the latter may lead to large changes in the incidence of consumption poverty. However, the sharp rise in poverty (and the steep fall in real wages) during 1998 turned out to be transient as inflation abated (and became deflation for a while), reflecting the restoration of macroeconomic stability. In addition, the government launched a comprehensive, donor-supported, social safety program. Some elements of the 54 See Aggrawal (1995) 55 See Islam and Nazara (2000b) for an evaluation of the evidence on the impact of minimum wages on employment in Indonesia. They argue that minimum wage increases in the 1990s were benign in terms of their impact on employment, but SMERU (2001) reaches an alternative conclusion. 56 See Edwards (1996) 57 See Dhanani and Islam (2001)

79 programme worked well, others did not. On balance, it appears that the social safety net interventions managed to reinforce the resilience of ordinary Indonesians in the face of unprecedented adversity. The evidence suggests that social safety net interventions probably prevented an additional 7-11 per cent of the population from sliding below the poverty line. 58 It is clear from the preceding discussion that the notion of labour market flexibility, while useful, can only represent the beginning rather than the end of any analysis of the linkage between poverty and the structure of the labour market. The paradigm of labour market flexibility is essentially silent on, or pays insufficient attention to, a range of complex issues. How does one give primacy to employment concerns in macroeconomic management? Should such primacy be given? Are there basic workers rights? If so, how does one embed them within the industrial relations system? Is there any role for public deliberation in the formulation of labour market policy? What role is there for the state in dealing with labour market risks, such as sub-optimal employment adjustments and falling real wages, triggered by economic shocks? These issues or at least some of them - are revisited in the discussion on policy options. The Indonesian reform agenda, regional decentralisation and their implications for employment and poverty In the aftermath of the 1997 financial crisis, Indonesia s introduced a series of reform measures. These entail:! Privatisation of some state-owned enterprises! Deepening of trade liberalisation and economic deregulation, a process initiated in the mid-1980s! Regional autonomy. These measures were in addition to other IMF-recommended measures to restore macroeconomic stability and rehabilitate the banking and financial sector. What has been the employment impact both actual and prospective of these initiatives? Before proceeding to a discussion of the Indonesian case, it would be useful to provide a comparative perspective. One study, entailing a large-scale survey of international experience, suggests the following. 59 First, the job destruction effects from trade liberalisation varied from country to country, being substantial in some and minimal in others. Second, the downsizing of the public sector following a privatisation program has turned out to be most dramatic in the transition economies of East Europe where, in some cases, up to 50 per cent of public sector workers were considered to be redundant. In general, public sector downsizing has meant a loss of earnings and benefits for retrenched workers, this loss being quite significant for those with seniority in the public sector. Women have also experienced bigger losses than men, partly because women are more likely then men to withdraw from the labour force after downsizing. Third, trade liberalisation seems to impose a decline in real wages in the short-run (up to three 58 See Dhanani and Islam (2002) for a thorough review. 59 Rama (2001)

80 years), with econometric estimates suggesting a 3.5 per cent decline in real wages for every 10 per cent increase in the foreign trade-to-gdp ratio. Thus, it would appear that trade liberalisation and privatisation can impose short-run hardships on ordinary workers and contribute to a worsening of poverty. Fortunately, there are some positive consequences that offset the costs imposed on ordinary workers. As noted, the wage decline is temporary. In addition, an inflow of foreign direct investment can actually raise wages, with every one per cent increase in the foreign direct investment-to-gdp ratio raising wages by 2.5 per cent. It should be noted that the job destruction effects of privatisation often affect privileged workers in the protected sectors and not usually workers hovering near or below the poverty line. Returning now to the Indonesian case, what conclusions can one draw about the employment implications of its reform agenda? One study has offered a preliminary assessment. 60 Here are the key conclusions. The privatisation of state-owned enterprise are driven by the twin objectives of raising funds for budgetary support from asset sales and, in the longer term, to turn loss-making enterprises into efficient, privately owned and managed entities. However, none of the sixteen SOEs intended for privatisation in 2001, and employing some 50,000 employees in all, was privatised by the target date. 61 Can one speculate on the likely employment impact of this program, assuming full implementation at some stage? Based on the premise of over-staffing levels of per cent, the employment impact of this program would, at most, have entailed some 7,500 persons in a workforce of 90 million. In any case, most of this would have been achieved through natural attrition and early voluntary retirement. Trade liberalisation has not led to a decline in manufacturing employment so far. Employment in medium and large-scale manufacturing establishments may have grown a little in 2000 and Import liberalisation probably had an impact on certain importcompeting industries and led to specialisation in other industries. For example, the number of domestic market-oriented radio and audio equipment manufacturers has declined, but the number of producers of loudspeakers has increased. Overall, consumer good imports in dollar terms regained their pre-crisis level by 2000, despite an economy that was only two thirds of its 1996 original size, again in dollar terms. Imports of textiles, clothing, television, radio and audio equipment, and motorcycle parts were either equal to or higher than before the crisis. If these trends continue, the corresponding domestic industries will either suffer or will not fully benefit from the economic recovery underway. At the same time, trade liberalisation offers the opportunity to Indonesian manufacturers to identify and develop market niches in the world market, and thereby create employment in new industries. Why has manufacturing employment held up in the presence of a trade liberalisation program, even in the short-run? The decline in manufacturing activity was induced by the terrible recession of 1998, but this was soon offset by export production that exceeded pre-crisis levels by 1999, before achieving record levels in 2000 in response 60 See Dhanani and Widarti (2001). 61 Recent media reports suggest a good deal of resistence to the privatisation program by workers of the state-owned enterprises. See, for example, The Jakarta Post, 21 March, 2002.

81 to strong demand for electronics and other goods in the US market. Certainly, a substantially undervalued rupiah facilitated this process. Very few large manufacturing establishments faced permanent closure, and some that had halted operation in resumed activity in As a result, employment in garment and electronic establishments was higher in 1999 than before the crisis. Unfortunately, these positive developments in the manufacturing sector are now under threat. In 2001, the US economy slowed down considerably and the information and communication technology boom came to end. In addition, the horrific terrorist attacks on September 11 (2001) have reinforced the slowdown of the US economy. These developments have led to a drastic decline in the demand for Indonesian export and to significant lay offs. The financial restructuring measures, including the closure and mergers of private and public banks, did not, it seems, reduce employment in the banking and financial sector. At the same time, the government s anti-inflation strategy, fashioned within the framework of the IMF s financial assistance package, managed to rein in inflation, which declined from 78 to 2 per cent between 1998 and 1999, before rising again to 9 per cent in This minimised the further erosion of real wages of low-income employees and self-employed workers and farmers. Since the poor are always disproportionately affected by inflation, the return to a low inflation regime protected them a point that was highlighted at a previous juncture. What about the employment impact of decentralisation and regional autonomy? Such developments did not result in civil service redundancies. The government engaged in an administrative reshuffling rather than large-scale downsizing. The transfer of central government employees in January 2001 to local government administrations really involved those who were already located in provincial and district offices. On the other hand, the recruitment freeze, in place since 1995, was continued. As a result, the number of civil servants fell from 4.0 million to 3.9 million, or by 80,000, between 1999 and 2000 due to natural attrition. It is likely that the recruitment freeze has also affected the youth unemployment rate to the extent that it has worsened the queuing process for public sector jobs that is characteristic of youth unemployment in Indonesia. A significant threat to employment generation in an age of decentralisation stems from revenue-generating levies and taxes on businesses and movement of goods imposed by a number of regional administrations, and their negative impact on farmers and other producers in distant markets. While the corresponding employment and income losses are difficult to estimate, the uncertainty raised by such ad hoc levies and taxes has already resulted in a freeze in all mining investments in the country, and perhaps of investment in other areas as well. Furthermore, regional administrations have so far proven notoriously unable to handle labour relations reinforcing the image that Indonesia has become a country prone to unrest and industrial strife. It is likely that such an image is dissuading the inflow of foreign direct investment preventing the latter from acting as a vehicle for enhanced employment opportunities and higher wages.

82 4. Poverty and the labour market: the role of wages 62 So far, the discussion of the poverty-labour market nexus focussed primarily on issues in employment. There is, of course, another critical dimension that affects the welfare of workers, namely, the rewards to their labour services in the form of real wages. Accordingly, this section will:! Review the long-run behaviour of real wages in Indonesia! Highlight episodes of falling/stagnating real wages, with specific reference to the 1997 financial crisis! Suggest a framework that can explain the behaviour of real wages in Indonesia and highlight the role that changes in productivity have played in underpinning wage growth! Explore the relationship between real wages, productivity and poverty at both the sectoral and aggregate level. Trends in nominal and real wages: overview Nominal earnings of wage employees increased by 15 per cent per annum on average in the period before the crisis, according to the national labour force surveys (table 4.1). Earnings did not grow evenly during this period however, slowing down to respectively 10 per cent and seven per cent per annum during and During this 21-year period, consumer prices rose by an average of 10 per cent per annum, resulting in real growth in earnings of 5 per cent per annum. During the two lower nominal growth periods, real earnings stagnated in , while declining by 2 per cent p.a. during This section draws on Dhanani and Islam (2001b). See Papanek (1999) for a seminal study on wages and poverty in pre- and post-crisis Indonesia.

83 Table 4.1 Trends in Nominal & Real Earnings, Indonesia, Index (1997 = 100) Average annual growth rate 1 (Per cent) Pre-crisis Nominal Earnings Indonesia Male Female Urban Rural Urban CPI Real Earnings (1997=100) Indonesia Male Female Urban Rural Source: Earnings: Labourer/Employee Situation in Indonesia (National labour force survey Sakernas), annual publication for , CBS. Indikator Tingkat Hidup Perkerja [Living standards Indicator], and Indikator Social Wanita [Women s social indicators], various years, for , CBS. Special tabulations from raw data produced by CBS for 1982 and Consumer prices: weighed average consumer price index of 43 cities (27 cities before 1994), annual average indices, Monthly Bulletin, CBS. Note: a Growth rates for , and are point-to-point compound growth rates. Growth rates for and are average annual growth rates. The East Asian financial crisis, which began in Indonesia in mid-1997, had a major impact on real earnings. In the first year, consumer prices rose by 100 per cent, while nominal earnings grew by 20 per cent, causing real earnings to decline by over 40 per cent. In the following two years, nominal earnings continue to grow by 20 per cent p.a., while inflation was brought under control, rising by just 5-10 per cent per annum. Real earnings began to climb slowly again to around 90 per cent of their pre-crisis level by Published earnings data from the labour force surveys by gender and urban-rural location are available only since The real earnings of females grew faster than for males in the period before the crisis (six vs. four per cent p.a.). This acceleration took place during and Rural real earnings also grew faster than urban earnings during this period (four vs. three per cent p.a.), particularly during the earlier period. These differential growth rates led to the narrowing of gender and urban-rural earnings, suggesting tighter labour market conditions during these two periods of relatively rapid wage growth.

84 Real wage trends by sector Real earnings of employees grew more rapidly in agriculture and manufacturing than in other sectors before the crisis. They grew by five per cent p.a. in these two sectors compared with four per cent p.a. in utilities, construction and transport and services, three per cent p.a. in trade and just two per cent in mining and finance between 1976 and 1997 (table 4.2). Since earnings in the agricultural and manufacturing sectors were lower than average, while those in mining and finance were higher than average, these differential growth rates lead to the narrowing of earnings differentials observed in the previous section. Table 4.2 Real Earnings by Sector, (Annual average growth rate) (Deflated by urban consumer price index) Pre-crisis Agriculture Mining, quarrying Manufacturing Utilities Construction Trade, hotel, restaurant Transport Finance and banking Services Other All sectors Source: same as table 4.1. Note: 1 Other included mining, construction, transport and finance sectors in one category in the questionnaires. Growth rate is point-to-point compound growth rate. : Not available. In general, real earnings moved in the same direction across sectors, though at different rates, during the sub-periods of rapid growth of , and Similarly, the overall decline observed during was shared by most sectors. The exception was the earlier period, which registered real earnings increases in some sectors such as manufacturing, transport, and services, and declines in others such as agriculture and trade. Real wage trends by education The labour force survey data is the only source of information on wages by education level. This data show that the real earnings of employees with primary school education and below rose relatively more rapidly than those with junior and senior secondary education (three per cent vs. one per cent and two per cent respectively), leading to the narrowing of wage differentials by educational level before the crisis noted in the previous section (table 4.8). Their real wages also fell less rapidly during the crisis period At higher education levels however, the real earnings of diploma and university graduates rose in line with the average during periods of rapid growth, while continuing to grow in the periods of constant overall earnings.

85 Table 4.8. Real Earnings by Education Level, (per cent change p.a.) Pre-crisis No schooling Did not complete primary Primary school Junior secondary school Senior secondary, general Senior secondary, vocational Diploma I and II Diploma III University All levels Source: same as table 4.1. Note: 1 No schooling and did not complete primary school together until Senior secondary general and vocational together. 3 Diploma and university graduates together. : not available. Long-run trends in real wages: summary The long-term trends in real wages over the period can be summarised as follows. Following a period of relative wage stagnation during most of the 1970s, which was otherwise a period of rapid economic growth, real wages grew by five per cent per annum on average during the 20-year period between 1978 and 1997, and where equally shared by the agricultural and non-agricultural sectors. Real wage improvements took place in all years except the mid-1980s, paradoxically also otherwise known as a period of rapid economic growth. The financial crisis of had a major impact on real earnings. Real wages declined by 40 per cent in the first year, remained at 10 per cent below their pre-crisis level three years later in Before the crisis, females and rural areas improved their wages relative to males and urban areas. Workers in the agricultural and manufacturing sectors, where wages were lower than in other sectors, improved their position relative to workers in other sectors. Workers with less than primary education improved their position relative those with junior and secondary school education. This led to the narrowing of wage differentials. Wage differentials were unaffected by the crisis. Understanding the long-run behaviour of real wages in Indonesia: towards a demand-supply framework Although changes in definitions, questionnaires and coverage over such a long time frame introduce many comparability problems in employment data from different time periods, a careful assessment of the labour force survey data does suggest that the demand for labour grew somewhat more rapidly than its supply to produce the observed real wage increases during most of this period. The key factors responsible for the moderate tightening of the labour market may be summarised as follows. On the supply side, two important measures were undertaken. First, the nationwide family planning programme launched in the early 1970s succeeded in slowing down the growth of the working age population, from around 4.0 per cent to 2.6 per cent per year between the late 1970s and the late 1990s. Second, the large-scale school building programme and education campaign led to increased enrolment at double the

86 rate of the working age population. Enrolments were particularly rapid in the late 1970s and first half of the 1980s, averaging respectively 14 per cent and 8 per cent per annum, and at a higher rate for female youth. At its peak, the population attending school increased by 0.7 million children and teenagers per annum during , and by million per annum in the 1980s. This resulted in the slower growth rate of the labour force relative to the working age population over this whole period (2.4 vs. 2.8 per cent per annum). On the demand side, strong and sustained economic growth in labour-intensive sectors, particularly manufacturing and construction, led to a corresponding demand for labour. Non-oil GDP growth averaged 8 per cent per year in this 21-year period, led by the manufacturing and construction sectors, which grew at respectively 13 per cent and 10 per cent per annum. Most other sectors grew by 7-10 per cent per annum, with the exception of the agricultural sector, which averaged just over 3 per cent per year. The windfall revenues from the first and second oil booms of the mid-1970s and late 1970s were ploughed back into an ambitious public infrastructure programme which included the rehabilitation and construction of new irrigation canals, and the construction of school buildings, roads and health centres, thereby creating substantial demand for construction labour. Agricultural production grew as a direct result of improved rural infrastructure and investment in disseminating the green revolution package of IRRI rice varieties and fertilizer, providing ample demand for rural labour. In addition, strong, sustained and broad-based economic growth, particularly of relatively labour-intensive sectors such as manufacturing, construction and transport, created additional demand for unskilled labour in both urban and rural areas. Real wages first increased in the late 1970s, when agricultural production grew faster than in earlier periods. This coincided with rapid expansion of the construction sector when some wage employees were drawn away from agriculture, even though the end of the oil boom in led to a sharp drop in construction labour thereafter. After momentarily stagnating due to the further contraction of construction labour in the second half of the 1980s, the demand for wage labour once again picked up in the construction sector which, coupled with continued strong demand for unskilled wage labour in the manufacturing and other sectors, allowed the agricultural sector to gradually reduce its work force during most of the 1990s, leading to sustained real wage increases. Agricultural and non-agricultural wages moved together, because wage movements in different sectors reinforced each other due to a high degree of labour mobility and integration between sectors. The moderate tightening of the labour market led to narrower wage differentials by gender and education attainment, and between urban and rural areas. Some observers have highlighted the point that institutional factors have also played a major role in influencing the behaviour of wages in the 1990s. 63 Beginning in 1992, the Suharto regime began an aggressive pursuit of minimum wage policy. Such a policy stalled for a while during the crisis, but seems to have picked up with renewed vigour since then. Minimum wage increases, it is argued, pushed up average wages as 63 See SMERU (2001a).

87 well. Statistically, it has been difficult to substantiate this proposition. While no one doubts that minimum wages have gone up substantially in real terms, the circumstances of pre-crisis Indonesia also need to be taken into account. A general tightening of the labour market led to a real wage boom that in turn provided a conducive environment for minimum wages to be pushed up. The wage-productivity relationship: overview Table 4.9 offers an overview of the trends in real wages for various sub-periods covering the and compares them with changes in productivity. For the economy as a whole, real wage growth has kept pace with productivity growth for the period. Furthermore, periods when productivity growth has been buoyant has also been accompanied by real wage growth, but this has not always been the case. For example, rapid productivity growth in the period stands in stark contrast to the decline in real wages. Considering the fact that this period also coincides with the trade liberalisation-cum-deregulation reform agenda, one could raise doubts about the short-consequences of such an agenda for the welfare of workers. However, by the beginning of the 1990s, and until the eve of the 1997 financial crisis, real wages grew a little above productivity growth. Note that there were other sub-periods ( ) in which real wage growth exceeded productivity growth sometimes by quite substantial margins. At the sectoral level, one can discern significant variations. Real wages grew at roughly twice the rate of productivity growth in agriculture in the period, but the converse was the case in manufacturing. In the trade and services sector, real wage growth also exceeded productivity growth during the pre-crisis period. Table 4.9: The wage-productivity relationship in Indonesia, (% change) Sector All GDP/worker Real wage Agriculture GDP/worker Real wage Manuf. GDP/worker Real wage Trade GDP/worker Real wage Services GDP/worker Real wage Source: National Accounts and Labour Force Survey Sakernas, CBS Note: worker by sector: total employment minus family workers. All sectors employment: 15+. Other sectors, 10+ until 97, 15+ after.

88 Wages, productivity and poverty: identifying the linkages Table 4.10: Real wages, productivity and poverty: (% change) Sector All GDP/worker Real wage Poverty Source: As in previous tables in this section Table 4.10 show the trends in real wages, productivity and poverty. The information is also depicted in figure 4.1. Co-movements between real wages, productivity (GDP/worker) and poverty incidence, Indonesia, % change All GDP/worker Real wage Poverty Figure 4.1 Over the pre-crisis period ( ), real wages and productivity have grown at five per cent annually while poverty declined by 29 percentage points. At the same time, in the crisis period and its aftermath ( ), the increase in poverty of six percentage points is accompanied by a collapse in both real wages and productivity. Yet, there are sub-periods during which the linkage between indicators of consumption poverty, real wages and productivity diverge. This is most conspicuous in the period when poverty fell at a steep rate, but real wages actually declined at an annual rate of two per cent. It is also conspicuous in the period when real wages barely grew (at about one per cent), but consumption poverty fell rather steeply. Apart from these anomalous episodes, the co-movements in real wages, productivity and consumption poverty are quite apparent with the substantial declines in poverty being accompanied by rising productivity and real wages and vice versa. An important study on real wages in Indonesia has also explored the wage-poverty linkage. Pertinent information derived from that study - is provided in table As can be seen, there is, on the whole, a reasonably good correspondence between the movement in real wages for agricultural workers (in Java) and the changes in poverty incidence. The conspicuous exceptions seem to be and when poverty fell even when real wages declined Regression analysis using quarterly Indonesian wage data by Papanek (1999) suggest that real wages respond positively to output growth and negatively to inflation.

89 Table 4.11: Comparison of changes in poverty and real wages in Indonesia Changes of number of poor based on official poverty line (%) Agricultural real wages in Java (%) Source: Adapted from Papanek, 1999, table 2. The link between poverty and wages at the sectoral/occupational level is exhibited in table 4.12 using data for 1999 and As can be seen, agriculture, which is the least paid (with wages being 54 per cent of the national average) and also among the least productive sectors, had the highest concentration of poverty and accounted for nearly 60 per cent of the total population in poverty. At the other extreme, finance, in which average wages were 75 per cent above the national average and contained a high incidence of white-collar workers, had the lowest poverty incidence in the country. However, the relationship between wages and poverty at the sectoral level is by no means exact. For example, mining and quarrying has relative wages that are 52 per cent above the national level. Yet, it has a poverty rate (29 per cent) that is higher than the manufacturing sector (23 per cent), which has relative wages that are 7 per cent below the national standard. Table 4.12: Relationship between poverty and relative wages by sector of occupation Feb Poverty incidence Contribution to Relative wage (%) total poor (%) (national =100) Agriculture Trade Manufacturing Civil, social and private services Transport and Communication Construction Receiving transfer Mining & quarrying Others Finance Electricity, gas and water Source: See previous tables in this section.

90 5. Summary of key findings This paper has navigated a wide and diverse terrain. Hence, it would be useful at this juncture to bring together different strands in the analysis and to highlight the key findings. These findings in turn are guided by the primary goal of studying the linkage between poverty and the structure of the labour market. This has entailed a study of the evolution of employment and wages in pre- and post-crisis Indonesia and the various ways in which such an evolution has influenced the nature and scale of poverty. The multiple dimensions of poverty in Indonesia Reflecting current intellectual trends, the paper highlighted multiple dimensions of Indonesian poverty in both the pre- and post-crisis period. This has involved an empirical synthesis of trends in income/consumption poverty, capability deprivation encompassing the UNDP s human poverty index (HPI), vulnerability and powerlessness. The stylised facts (on which there appears to be a reasonable degree of consensus) are as follows.! There was an impressive decline in consumption poverty over two decades. The national estimates suggest that poverty fell from per cent in 1976 to 11 per cent by Alternative estimates generally support this optimistic conclusion, as do estimates of income poverty based on international poverty lines. On the other hand, a generous definition of poverty ($2-a-day) suggests that approximately 60 per cent of Indonesians still live in penurious circumstances.! Consumption poverty shot up during the crisis, but it was transient, with the head count ratio tapering off to its pre-crisis level by 1999.! Indicators of capability deprivation (malnutrition, literacy, access to health services, low life expectancy) suggest that Indonesia made impressive progress that testifies to the genuine social gains embedded in indicators of income/consumption poverty. The value of the HPI fell from 28 to 25 between 1990 and 1996 and was apparently unaffected in any major way by the 1997 financial crisis. Nevertheless, capability deprivation is a significant problem in Indonesia. For example, in 1999, more than 50 per cent of Indonesians did not have access to health services while the incidence of child malnutrition was 30 per cent.! Vulnerability the risk of falling into poverty or getting deeper into poverty in the future - in Indonesia is more significant than current consumption poverty. Recent estimates suggest that the incidence of vulnerability is 34 per cent and that it worsened more sharply than consumption poverty during the crisis. Indeed, even those whose expenditure is 20 per cent above the poverty line are vulnerable to an episode of poverty.! Perceptual data gleaned from direct experiences of poverty suggest that the poor feel alienated and excluded from local decision-making processes a phenomenon particularly affecting women and that they lack trust in state institutions.

91 Thus, the harsh reality is that, despite decades of sustained economic and social gains, Indonesia still has to grapple with the manifold manifestations of poverty and identify cost-effective and credible policy instruments that could lift the many millions who are poor today and those who are likely to become poor tomorrow towards a reasonably secure and prosperous future. The poverty labour market nexus: the role of employment This paper has argued that a salient aspect of this rethinking on poverty in post- Suharto Indonesia should entail a greater understanding of the interaction between the labour market and those who are poor. At the same time, the limits of this analytical framework need to be recognised. Indicators of powerlessness suggest that poverty has deep-seated social, political and institutional roots. The finding that more than 30 per cent of Indonesians are vulnerable to episodes of poverty suggest that a mere expansion of current employment might ease the pain and suffering of current poverty, but would be inadequate in dealing with the more complex challenge of vulnerability. In order to deal with the latter, appropriate policy interventions would be required to ensure the economic security of at least a third of Indonesians. Such interventions would certainly encompass labour market policies, but cannot be limited to them. Similarly, the challenge of dealing with capability deprivation cannot be kept within the conventional confines of labour market policies. Determined public action in the domain of public health and basic education rather than a mere expansion of employment - engenders healthy and literate people. They are also transformed into healthy and productive workers who make a contribution to the productive capacity of an economy. Bearing the above caveats in mind, what can one conclude about the key features of the poverty-labour market nexus in Indonesia? It is clear that developments in the labour market are consistent with the broad finding that there was significant progress in poverty reduction in the pre-crisis period. There was significant urbanisation and formalisation of the work force in the 1990s. The share of formal sector employment rose from 28 per cent (in 1990) to 38 per cent on the eve of the crisis (1996), while the relative size of the urban labour force increased from 26 per cent to 36 per cent over the same period. At the same time, rapid growth in the pre-crisis era was accompanied by a deepening of skills of the work force. The proportion of less educated/skilled workers (defined as those who have primary education and less) fell from 87 per cent to 79 per cent between 1990 and 1996, with a corresponding increase in the share of the educated/skilled workforce (defined as those with at least secondary education) from 14 per cent to 21 per cent. These developments are consistent with the sustained decrease in consumption poverty documented in section 2 of this study. A reallocation of labour from lowproductivity, low-paid jobs in the rural economy to better-paid employment in the urban and formal economy is one well-known route via which ordinary workers and their dependents can improve their living standards. This is consistent with household-level data that show that the incidence of poverty drops from 40 per cent to 23 per cent as one moves from the agricultural sector to the manufacturing sector. Furthermore, there is a significant correlation between educational attainment and

92 poverty incidence at the household level. The evidence shows that 87 per cent of Indonesia s poor have primary education or less. Thus, as more Indonesians gained access to the education system during the 1990s, it provided them with an escaperoute from impoverishment. Measures of casualisation of the work-force (in terms of part-time employment and growth of informal sector employment) do not show any consistent deterioration in the pre-crisis period. This is significant because household-level data show that employment in the informal sector is a major attribute of poverty, with 70 per cent of the poorest 20 per cent of households identifying informal sector employment as their major occupation. There is little evidence as well that, prior to the crisis, involuntary underemployment (or time-related underemployment) was widespread, being around eight to nine per cent. Although the open unemployment rate was low in the pre-crisis period, not much significance should be attached to this statistic. The link between unemployment and poverty is tenuous in the Indonesian case. Household-level data show that unemployment is concentrated among educated graduates. At the same time, the poverty rate for this group is well below the national average and ranges from 13 per cent (secondary school graduates) to two per cent (tertiary graduates). Another important aspect of the poverty-labour market nexus in Indonesia is the role that the employment-intensity of rapid growth played in the pre-crisis period. Higher employment elasticity translates to the fact that, for any given growth rate, the job creation rate is higher (subject to the important qualification that the value of employment elasticity should not exceed unity). The available evidence suggests that employment elasticity was quite low (0.3) in the 1970s, rose appreciably in the 1980s (0.8) and tapered off a little with the onset of the 1990s (0.7). These estimates are consistent with conventional interpretations of the evolution of the employmentpoverty linkage in the Indonesian context. In the 1970s, Indonesia s oil wealth was used to fund public infrastructure and rural development, but the manufacturing sector remained small, inefficient and protected. This constrained its capacity to act as a locomotive for employment creation, although the oil wealth-driven public investment policy of the 1970s did reap some social dividends, given that it heralded the onset of a remarkable decline in mass poverty. As the oil price boom dissipated in the early 1980s, the Suharto government was forced to turn to outward-oriented industrialisation strategy that began in earnest from the mid-1980s. The much higher values of manufacturing employment elasticity in the 1980s support the notion that the manufacturing sector was beginning to play a more significant role in spearheading employment creation by drawing in less-skilled workers from the rural and informal economy to the more formal economy. The poverty-labour market nexus: the role of wages Real wages grew by five per cent per annum on average during the 20-year period between 1978 and 1997, and were equally shared by the agricultural and nonagricultural sectors. Real wage improvements took place in all years except the mid- 1980s, paradoxically also otherwise known as a period of rapid economic growth. Before the crisis, females and rural areas improved their wages relative to males and urban areas. Workers in the agricultural and manufacturing sectors, where wages were

93 lower than in other sectors, improved their position relative to workers in other sectors. Workers with less than primary education improved their position relative those with junior and secondary school education. Not surprisingly, these developments meant a narrowing of wage differentials in the pre-crisis period. For the economy as a whole, real wage growth has kept pace with productivity growth for the period. Furthermore, periods when productivity growth has been buoyant has also been accompanied by real wage growth, but this has not always been the case. For example, rapid productivity growth in the period stands in stark contrast to the decline in real wages. Considering the fact that this period also coincides with the trade liberalisation-cum-deregulation reform agenda, one could raise doubts about the short-run consequences of such an agenda for the welfare of workers. However, by the beginning of the 1990s, and until the eve of the 1997 financial crisis, real wages grew a little above productivity growth. There were other sub-periods ( ) in which real wage growth exceeded productivity growth sometimes by quite substantial margins. Over the pre-crisis period ( ), real wages and productivity have grown at five per cent annually while poverty declined by 29 percentage points. Yet, there are subperiods during which the linkages between indicators of consumption poverty, real wages and productivity diverge. This is most conspicuous in the period when poverty fell at a steep rate, but real wages actually declined at an annual rate of two per cent. It is also conspicuous in the period when real wages barely grew (at about one per cent), but consumption poverty fell rather steeply. Apart from these anomalous episodes, the co-movements in real wages, productivity and consumption poverty are quite apparent with the substantial declines in poverty being accompanied by rising productivity and real wages. Other studies support these findings. The poverty-labour market nexus, the Indonesian crisis and its aftermath Did the Indonesian crisis of 1997 wipe out all the major gains that were made during the rapid growth era of the Suharto period? It appears that for a while it threatened to do so. As poverty shot up during the peak of the crisis, there was an involuntary reallocation of labour to the agricultural sector and an enlargement in the size of the urban informal sector, thus creating a phase of de-industrialisation. Real wages also collapsed by about 40 per cent. By 1999, however, some form of recovery seems to have taken place. The forced reallocation of labour to the agricultural sector and the urban informal sector has been reversed, but estimates for 2000 suggest a less certain picture. Real wages are now about 10 per cent below their pre-crisis values. A debate that has emerged under these circumstances is the extent to which employment and wage adjustments during the crisis period reflect the virtues of labour market flexibility. Some see it as a testimony to the resilience and coping mechanisms of ordinary Indonesian workers that tempered the rise in poverty and prevented a blowout in the unemployment rate. Others suggest that the evolution of poverty during the crisis period were driven by factors that were essentially of a macroeconomic nature and, at least partly, influenced by social protection policy. Thus, the ability of the government to restore macroeconomic stability and rein in the inflation shock of 1998 together with the reasonable effectiveness of some of its social safety net initiatives reversed the collapse in the purchasing power of ordinary Indonesians and paved the way for a

94 tenuous social recovery. The government benefited from the assistance that it received from the international donor community in the sphere of macroeconomic policy and the design of social safety net schemes. Considerable interest has also centred on the extent to which the crisis-induced reform agenda of intensified trade liberalisation, deregulation and privatisation may impair employment prospects. At the same time, speculation has emerged on the employment impact of the Indonesian government s ambitious regional decentralisation agenda that was launched in A preliminary assessment suggests the employment displacement effects of the trade liberalisation program have so far been benign. This is largely fortuitous. Prior to the growth slow-down of the US and the World economy, the period turned out to be favourable to Indonesia. Buoyed by US export demand and fuelled by a massively depreciated Indonesian Rupiah, domestic firms and its multinational subsidiaries managed to boost export production in sufficient volumes. This sustained employment in the export-oriented sectors and offset the deflation of domestic demand. Now that a worldwide growth slowdown (worsened in turn by the horrific September 11 th terrorists attacks on the US) has dissipated the export boom of , it remains to be seen how the trade liberalisation agenda will influence future developments in the Indonesian labour market. The privatisation program has, so far, not been launched. Nevertheless, on the assumption of full implementation at some point and a premise of 15 to 20 per cent overstaffing in the 16 state enterprises designated for privatisation, the overall employment impact is likely to be some 7,500 persons in a workforce of 90 million. In any case, most of this would have been achieved through natural attrition and early voluntary retirement. The fear of large-scale redundancies of public sector employees as a result of a smaller central government following the implementation of regional decentralisation has not materialised. This is largely because the transfer of central government employees in January 2001 to local government administrations really involved those who were already located in provincial and district offices. On the other hand, the recruitment freeze, in place since 1995, was continued. As a result, the number of civil servants fell by 80,000 between 1999 and 2000 due to natural attrition. It is likely that the recruitment freeze has also affected the youth unemployment rate to the extent that it has worsened the queuing process for public sector jobs that is characteristic of youth unemployment in Indonesia. A significant threat to employment generation in an age of decentralisation stems from revenue-generating levies and taxes on businesses and movement of goods imposed by a number of regional administrations, and their negative impact on farmers and other producers in distant markets. While the corresponding employment and income losses are difficult to estimate, the uncertainty raised by such ad hoc levies and taxes has already resulted in a freeze in all mining investments in the country, and perhaps of investment in other areas as well. This is likely to have a negative impact on growth and employment.

95 In sum, this comprehensive account of the interactions between poverty incidence and the labour market suggests that a reasonable degree of consensus has emerged in understanding the multiple ways in which such interactions take place. Important structural changes in the Indonesian labour market entailing urbanisation, formalisation and skill upgrading of the work-force, rapid real wage growth underpinned by productivity growth - are essential in understanding how they have favourably affected the evolution of poverty in pre-crisis Indonesia. These structural changes were partly the result of the natural processes of sustained rapid growth and partly the product of policy interventions. At the same time, the crisis reversed the poverty-reducing developments in the Indonesian labour market, but by 1999 a tenuous recovery has been in progress. It remains to be seen whether the recovery process will be consolidated. The current reform agenda and the regional decentralisation program appear, until now, to have had a benign impact on employment, but a more comprehensive judgement must await the test of time and the accumulation of more knowledge through further research. 6. Discussion of policy options 65 Any attempt to develop a policy framework for dealing with issues that directly and indirectly impinge with the evolution of poverty should start with a clear recognition of the key labour market challenges. These are:! The need to create enough durable jobs annually (at least two million according to current estimates) to absorb new entrants to the work-force and clear the backlog of the unemployed and the underemployed! The need to empower workers both in the formal sector and in the informal economy to cope with the problems of vulnerability to future episodes of poverty that recent studies have highlighted! The need to develop a credible industrial relations system that can represent the interests of both formal sector and marginalised workers and thus contribute to alleviating the sense of voicelessness and powerlessness of the poor that recent studies have found. The notion that one can simply rely on the virtues of a flexible labour market to respond to the afore-mentioned challenges is misplaced. The framework of labour market flexibility relies largely on the decentralised interaction of employers and workers to engender improvements in collective welfare. This means an insufficient appreciation of the current labour market challenges that are the product of the weaknesses of the pre-crisis period. Admittedly, the Suharto government was successful in mitigating the incidence of mass poverty through rapid, employmentoriented growth, but it was also conspicuous for its failure to respect labour rights, build up a credible industrial relations system and invest in institutional arrangements to deal with labour market risks. These inadequacies and failures of the past have, in turn, been amplified by the Indonesian crisis. What are the elements of a policy framework that one could devise for dealing with labour market issues that impinge on the manifold manifestations of poverty in Indonesia? Here are some preliminary thoughts. 65 This section draws on Islam (2001).

96 The need for the return of rapid and sustainable growth There has to be a return to reasonably rapid, and sustained, growth. Current forecasts suggest that the growth rate in the short to medium term may not exceed four per cent. 66 This is barely enough to create jobs to cope with the two million entrants to the workforce and by no means enough to cope with the backlog of the unemployed and the underemployed. Based on current estimates of employment elasticity, a growth rate of at least five per cent in the short to medium term appears critical in underpinning any strategy for mitigating the incidence of poverty in post-crisis Indonesia. 67 The return of growth in post-crisis Indonesia means a re-engagement with the paraphernalia of economic and institutional reforms that the international development community has been urging the government to adopt over the last five years. Such reforms, it is argued, will once again make Indonesia an attractive place to invest (both from the perspective of domestic and foreign investors) and thus rekindle the engine of growth. At the same time, a mere return of growth will not be enough. In the haste to embrace new investment in manufacturing and mining activities, the government cannot afford to neglect the informal segments of the economy and the agricultural sector where the bulk of the Indonesian poor reside. A strategy of reigniting growth in post-crisis Indonesia needs to be accompanied by appropriate measures that ensure the health and buoyancy of the agricultural sector and that provides ample scope for small and medium-sized business enterprises (SMEs) to play a major role in productive employment opportunities for ordinary Indonesians currently languishing in the informal sector. 68 These initiatives in turn are also likely to improve income distribution and thus reinforce the capacity of the growth-process to reduce poverty. Indeed, cross-country data confirm such an interpretation. 69 Reflecting employment concerns in macropolicy It is difficult to be serious about employment-led recovery from the 1997 financial crisis and its terrible aftermath unless employment concerns are reflected in macroeconomic policy. Indonesian monetary authorities are currently entrusted with 66 Asia Recovery Information Centre (2002). 67 The five per cent figure was highlighted in the 1999 ILO Employment Mission to Indonesia (ILO, 1999a). Note that the required growth rate to absorb new entrants to the labour force is sensitive to the particular value of employment elasticity used. Islam and Nazara (2000a) find that the required growth rate to maintain flow equilibrium in the Indonesian labour market varies from 3.5 per cent to 4.7 per cent. 68 There is now a sizeable literature on SMEs in Indonesia. For a recent evaluation, see Berry et al (2001). They argue that SMEs fared better than large firms during the crisis, while in the pre-crisis period their productivity grew at rates quite similar to their large counterparts. The authors advocate the creation of a business environment conducive to SMEs and recommend the provision of business development services by the private sector. 69 See Agenor (2002:7). The study suggests that poverty could decline rapidly with distributionally neutral growth, but it only takes small deviations from such neutrality to wipe out the gains. The study proceeds to use cross-country regressions to show that output growth and real exchange rate depreciations tend to reduce poverty, while illiteracy, income inequality and macroeconomic volatility tend to increase it.

97 the task of reigning in inflation to an agreed target. 70 At the same time, fiscal policy is also severely constrained by the need for fiscal consolidation in order to reign in a crisis-induced explosion of public debt. This has understandably created tensions between an anti-inflation strategy and the need to nurture an employment-friendly recovery. 71 The medium-term inflation target is of the order of 3-5 per cent. 72 Yet, the central bank also recognises the possible consequences of its actions. Thus, it draws attention to the policy dilemma (emphasis in original) in the monetary decision-making process at Bank Indonesia. 73 In other words, how does it meet the predetermined inflation target without inhibiting the resurgence of growth? Bank Indonesia has not clearly delineated how it intends to resolve the dilemma that it faces. Indeed, some external observers, including a senior ex-imf official, have expressed concern that a preoccupation with reigning in inflation to a set target could push Bank Indonesia to become too inflation-averse. 74 Several possible suggestions may be made in dealing with the policy dilemma facing the Indonesian monetary authorities. To start with, the primary goal is the return of full employment driven by broad-based growth. In trying to preserve this fundamental goal, why not aim for a modest benchmark where the monetary authorities try to ensure that the inflation rate stays, on average, below double-digit levels? After all, in pre-crisis Indonesia, widely regarded as a successful case of economic development, the average long-run inflation rate ( ) was around 12 per cent. 75 What are the ex-ante social gains of keeping inflation persistently below five per cent or, conversely, the social costs of allowing inflation to drift above five per cent in the medium term? 76 Unless there are convincing answers to such a basic question, inflation targeting in a stringent fashion will cause more problems than they will resolve. Recasting the current inflation targeting procedure to the more modest approach of avoiding the entrenchment of double digit inflation rates at least until the recovery is consolidated is only the first step in incorporating employment concerns in macroeconomic management. A complementary step is to enhance the government s 70 The challenges and constraints of implementing inflation targeting in Indonesia are explored in Alamsyah et al (2001). 71 The government is being urged to run a primary surplus (equal to two per cent of GDP) over the next few years. See World Bank (2000c). 72 Note, however, that in the latest letter of intent (signed on August 27, 2001) between the Indonesian government and the IMF, the inflation target was set at 9-11 per cent a recognition that too tight an inflation target is probably not realistic at this stage. See ARIC (2002: 48). 73 See Bank Indonesia (2000) 74 The external observer is Paul Krugman, while the ex-imf official is Hubert Neiss. These comments were made at a seminar organized by Strategic Intelligence ( The Second Indonesia Economic Forecast The Road to Recovery: A Strategic Intelligence Seminar Conference with Professor Paul Krugman, August 29, Grand Hyatt Hotel, Jakarta) and received prominence in the local media (e.g. see The Jakarta Post, August 30, 2000). The commentary by Philip Bowring in the International Herald Tribune is also worth highlighting. He notes: Most of East Asia and Southeast Asia needs some inflation to spur consumer demand, reduce real interest rates and spark a revival in asset prices ( Asian Economies Should Ride on the Fuel Price Storm, International Herald Tribune, September 26, 2000) 75 Estimates kindly made available to the author by Zulfan Tajoeddin, Research Associate, UNSFIR, Jakarta, Indonesia. 76 See Akerlof et al (1996) who suggest that the social costs of moderate inflation have been exaggerated.

98 capacity to monitor developments in the labour market. While there is a wealth of short-run data on inflation, even quarterly data on employment does not exist, thus constraining the capacity of the government to regularly monitor the labour market. Yet, generating quarterly data on employment is not too difficult and can be reconstructed out of existing sources. 77 An approach that attaches primacy to employment concerns should entail clear statements by the government on the rate of job creation that would be consistent with maintaining equilibrium in the labour market (that is, the number of jobs created should absorb the growth of the labour force and make a dent on the current stock of the unemployed and the underemployed). The advantage of integrating employment concerns in macropolicy is that it enables the government to track the employment intensity of economic growth through estimates of employment elasticity. Higher employment elasticity will translate into lower growth rates required to reach threshold employment creation targets (that is, the two million jobs noted above). Several pertinent questions naturally emerge at this stage. What has been the historical behaviour of employment elasticity? How is it going to be influenced by the current reform agenda of further domestic deregulation, trade and FDI liberalisation, and privatisation in the short term and medium term? What will be the likely impact of decentralisation on the employment intensity of economic growth in the short and medium term? 78 Once again, clear policy positions on these issues are not yet evident. However, thorough evaluations by the government of these issues are critical as part of developing a coherent and credible strategy for employment-led recovery. Assuming that tracking employment elasticity is a legitimate exercise, what scope is there for the government to enhance employment elasticity, thereby reducing the burden on economic growth alone to generate the necessary jobs to absorb new entrants to the labour force? One possibility is to highlight the role that certain sectors/sub-sectors play in employment creation. For example, table 3.5 in this study identified garments, footwear, furniture and electronics as sub-sectors within manufacturing with high employment elasticities (above 0.5 for the period). It is encouraging that these are also sub-sectors with a significant degree of export-orientation. There is scope in these activities for small and medium-sized enterprises (SMEs) to play an important role in job creation. The government should seek to reduce entry barriers to sectors/sub-sectors with high employment elasticity both by reducing regulatory impediments and restraining the restrictive practices of dominant firms. ILO studies have consistently argued that there is another direct form of intervention that the government could exercise more assiduously that could influence employment elasticity. This pertains to the proposal that the government ought to consistently incorporate the use of labour-based rather than equipment-based production methods in its public investment policy. 79 The public sector in Indonesia, 77 It should be possible to generate quarterly data on employment by combining the national socioeconomic survey (SUSENAS) which is held in February of every year and the national labour force survey (SAKERNAS) which is held in August of every year. This will require the use of consistent definitions across both the surveys to underpin labour market data. 78 The ILO Area office in Jakarta has completed a preliminary study of the short-run employment consequences of the current reform agenda. The results have been reported in section 3 of this study. 79 ILO (2000a).

99 as in many other developing countries, is a major player in infrastructure investment. One ILO study has shown that labour-based production methods in infrastructure investment in Indonesia can generate as many as 1.2 million durable jobs over four years without compromising standards of quality that one associates with equipmentintensive production techniques. This means that, if general growth-induced employment creation is of the order of two million annually, it has the potential to rise to 2.3 million annually with the adoption of labour-based production methods in public investment policy. 80 This point is of particular importance because infrastructure maintenance and investment will grow significantly as economic recovery takes hold. Access to the education and training system An important area of poverty-focused policy intervention pertains to the education and training system. It does not fall directly within the domain of labour market policies, but has a major bearing on developments in the labour market. As this study has shown, there is a strong correlation between educational attainment and poverty incidence at the household level. More specifically, the available evidence suggests that 87 per cent of Indonesia s poor have primary education or less. Simply upgrading a household head s education to a junior secondary level is associated with a drop in the poverty rate from 30 per cent to 17 per cent. These are significant figures and suggest that all Indonesians need to have educational attainments at least to a junior secondary level. This is a target that the Indonesian government has recognised. It is also a target that is line with the international community that argues the need for education for all. 81 This in turn highlights a major policy challenge: how to ensure that barriers are not erected against broad-based participation by the poor in the education and training system. Failure to respond to this challenge will engender two long-term problems: Indonesia will be robbed of much-needed human capital; there will be entrenchment of privilege (as the poor can t get in). 82 Some qualifications need to be added as part of concluding this brief discussion on the education and training system and its relevance in fashioning a poverty-reduction strategy. The study of the education-poverty nexus in the Indonesian context relies exclusively on cross-section data at the household. Such evidence provides confirmation of correlation, not causality. Expanding education alone will not solve the multi-faceted nature of poverty! It remains to be said that sustainable poverty reduction entails a complex array of policy interventions. The interaction of investments in education with complementary investments as in the provision of 80 ILO (1999a). 81 Mingat and Winters (2002) draw attention to the need for education for all by 2015 this being the goal set by 180 countries at the World Education Forum in Dakar, Senegal in Such a goal in turn is a continuation of an agenda set by the 1990 World Conference on Education For All held in Thailand. 82 SMERU (2001b) has released a newsletter in which the difficulties faced by the poor in gaining access to secondary schooling are highlighted. The author, Oey-Mayling Gardiner, maintains that there are two barriers facing the poor (a) insufficient number of state-funded junior secondary schools; (b) exhorbitant fees, even in the so-called state-funded sector, that discriminate against the poor. Recent media reports, focusing on the experience of a particular Indonesian district (Banjarnegara), also highlights the rather difficult barriers that ordinary families face in educating their children at a junior secondary level. The drop-out rate for this locality at the junior secondary level is nearly 50 per cent! See Jakarta Post, 25 March, 2002.

100 agricultural extension, the establishment of micro-credit schemes etc will go a long way towards ensuring that the well-documented social returns to basic education are realised and the conditions of the poor improved. 83 Furthermore, the challenge is to provide high-quality basic education in terms of adequate retention rates, appropriate curricula, well-trained teachers, textbooks and other learning materials and proper physical infrastructure. Social protection, vulnerability and labour market risks A key element of making labour market policies in Indonesia more poverty-focused is the recognition of the fact that at least a third of Indonesians are vulnerable to future episodes of poverty. The 1997 financial crisis has demonstrated how externally driven economic shocks can force millions of Indonesians to move, at least temporarily, into low-paying, low-productivity jobs in the agricultural sector and the informal economy of the urban sector. There is a growing professional opinion that informal systems of social protection (the variety of assistance flowing from private organisations and the informal network of friends and family) are not necessarily efficient and cannot cope effectively with economy-wide shocks (such as the 1997 crisis). 84 Thus, more formal systems of social protection need to be nurtured and strengthened in dealing with labour market risks, particularly those triggered by economic shocks. Two pertinent policy options in this domain are unemployment benefits and public works, although there are many other initiatives (e.g. microfinance) that are also pertinent. Indonesia could consider the feasibility of setting up an unemployment benefits scheme to cater to the needs of formal sector workers. At the same time, a welldesigned public sector scheme could serve as an employment protection scheme for workers in the informal economy and rural areas the so-called excluded majority. 85 There are, of course, a number of objections that one could raise against the operation of an unemployment benefits scheme such as fiscal affordability, administrative complexity, and disincentive effects that lure prospective participants to stay voluntarily unemployed in order to access the benefits. Recent research suggests that some of these concerns have been exaggerated. 86 Thus, the so-called disincentive effects can be, and have been, mitigated by taking appropriate counter-measures (such as strict eligibility criteria). The fiscal affordability critique has also been questioned by others. A recent ILO study has made a strong plea in favour of introducing unemployment insurance schemes in East Asian economies. 87 Estimates suggest that an average required contribution rate of between 0.3 to 0.4 per cent of payroll between 1991 to 2000 would have been sufficient to provide all insured job losers over this period, including during the current crisis, with 12 months of benefits 88. In the case of Indonesia, the required contribution is 0.44 per cent. Another 83 Studies typically show that the social returns to education in low-income countries are highest at the primary and secondary level. See Mingat and Winters (2002). 84 See Murdoch (1999) 85 The literature on social protection in developing countries now focuses on the challenge of dealing with the excluded majority. See, for example, Beattie (2000), Ginnekan (1999), ESCAP (2000), ILO (2000b) and World Bank (2000b). 86 See Acemoglu and Shimmer (2000). 87 Lee (1998: chapter 4). 88 Lee (1998: 83). The calculations assume that the coverage of the scheme would be the same as existing social security provisions and would provide 12 months benefit at a replacement rate of 50 per cent of previous earnings.

101 independent study claims that most of the Asian economies should be able to operate (an unemployment insurance programme) with OECD generosity utilising an average payroll tax rate of 1.0 per cent. 89 Whatever the future shape of unemployment benefits scheme in Indonesia, it will leave out a large fraction of the workforce in the informal sector, those in selfemployment and in rural areas. Hence, discussions of alternative forms of employment protection are necessary. Perhaps the best-known example is public works. Public works schemes have been quite common in all the Asian economies affected by the 1997 financial crisis. The general conclusion seems to be that they have not been particularly successful in meeting their goals. The employment creation effects were modest and intended beneficiaries often failed to gain access to the potential benefits of the programme. 90 One factor that undermined the effectiveness of public works as a safety net measure during the Asian crisis is the haste with which the measures had to be put into action. In Indonesia, for example, labour-intensive job creation programmes (or Padat Karya PK - programme) that were terminated in 1994 were quickly resurrected without any careful reflection being given to an overall strategy. Numerous ministries and levels of government were involved causing serious coordination problems. The public works projects were also characterised by a low wage bill and the absence or at least insufficient participation of women. A more recent evaluation of the PK programme that compared it with other interventions in the social safety net domain such as the rice subsidy programme however offers a more subtle conclusion. It draws attention to the fact that a social safety net programme can be targeted primarily to the permanently poor, but it can also be judged on the basis of its ability to target beneficiaries who suffer the most shock during a systemic crisis, irrespective of their pre-crisis status. Using survey data, the study concludes that the worst hit were 300 per cent more likely to participate in the PK programme, while the worst hit were only 58 per cent more likely to participate in the rice subsidy programme. Nevertheless, the study concludes that when other dimensions of programme design are considered (such as administrative complexity, waste and malfeasance etc), it is likely that the cost per dollar of benefits delivered to poor beneficiaries would be significantly higher for the PK programme than for the rice subsidy programme. 91 The government has recognised the problems plaguing the PK programmes and has issued new guidelines in an effort to rectify them. It has explicitly recognised the need to increase the participation of women workers, to have sustainable infrastructure development and to have community and civil society involvement. Nevertheless, the government still conceptualises the PK programme as a key instrument for providing 89 Vroman (1999:37) 90 Direct employment creation programmes typically generate temporary employment equivalent to less than 1 per cent of the workforce. See Jorgensen and Domelan (1999:16). Lee (1988: 55) estimates that if one includes all social expenditures directed towards employment creation, then their effects may be summarized as follows: seven per cent of the unemployed in Thailand; 10 per cent of the unemployed in Indonesia; 24 per cent of the unemployed in Korea. 91 See Sumarto et al (1999).

102 short-term unemployment relief and assumes that they will be phased out once sustainable economic recovery takes place. 92 The government should seek inspiration from international evidence, which shows that well-designed public works schemes can be an important part of the tool-kit of long-term social protection policies in developing countries. 93 Dealing with voicelessness and powerlessness as dimensions of Indonesian poverty: the role of social dialogue and labour rights The problems of voicelessness and powerlessness have been shown to be important dimensions of contemporary Indonesian poverty. This demonstrates that poverty has deep-seated institutional roots, suggesting in turn that a focus on labour market policies alone would be inadequate in dealing with the plight of the poor. Nevertheless, much can be accomplished by paying attention to the building of appropriate labour market institutions that provides voice and representation for ordinary workers. Perceptions about growing labour unrest in Indonesia have understandably caused considerable concern among both domestic and international investors. 94 This in turn is likely to have a negative impact on investment, hinder employment prospects and adversely affect the incidence of poverty. One appropriate instrument for dealing with labour unrest is social dialogue or consensus building through deliberation and negotiation between employers and workers. The notion of social dialogue has built on the ILO s time-honoured principle of tripartism entailing the participation of the government, employers and worker organisations. This principle might have to be creatively adapted to respond to the constraint, opportunities and challenges that confront the promotion of social dialogue in Indonesia. How should one proceed? To start with, the cause of social dialogue should be championed by the strategic ministries rather than just the ministry dealing with labour affairs. 95 Employers associations need to be more committed to the cause of social dialogue and more coherent and united in terms of their agenda. Perhaps one of the biggest challenges facing the development of social dialogue in Indonesia is that the trade union movement is still at a nascent stage both in terms of its capacity and degree of representation. A good deal of work will need to be done on various fronts to bring about a cordial industrial relations climate through a culture of dialogue and discussion. One should also recognise the structural constraints operating on a putative trade union movement. This reflects the current stage of Indonesia s development in which own-account workers and the informal sector represent more than 40 per cent of the workforce. How does one develop credible voice representation for this excluded group? This is where trade unions could work creatively with non-workplace civil 92 See Parikesit and Hudayana (1999). 93 See Ravallion (1998), Subbarao et al (1997). 94 A 2001 survey by the Japan Bank for International Cooperation (JIBC) notes that Indonesia continues to fall further behind China and Thailand in the competition for Japanese investment and it could be overtaken by Vietnam and India in this decade. The JIBC has urged the Indonesian government to take immediate steps to improve its investment climate. It is noteworthy that Japan is the largest source of private direct foreign investment for Indonesia. See Jakarta Post, 22 March, See Campbell (1999) who emphasises this point in his account of the Asian experience with social dialogue in the post-crisis era.

103 society organisations with a credible record of representing the poor. 96 Linking the trade union movement to emerging pro-poor associations could turn out to be a promising route for a broad-based coalition that embeds voice representation for the poor and the vulnerable in Indonesia. Changing the nature and scope of dialogue partners may also invigorate the nature of the social dialogue itself. There is always the risk that social dialogue could become a residual forum for managing the consequences of particular policy actions or even as a supplement to bread-and-butter issues that form the substance of collective bargaining mechanisms. To be effective, social dialogue ought to become a forwardlooking forum that can set the agenda for future labour market issues. It would be fair to maintain that Indonesia faces the dual challenge of moving up the technology ladder while simultaneously trying to deal with standard labour market risks of fluctuating real wages, unemployment and underemployment that confront a typical developing economy. Technological and industrial transformation means the revisiting of constraints and problems afflicting the education and training system, while dealing with labour market risks entail a focus on the governance of social protection. A broad-based coalition of employers, worker organisations and public intellectuals may represent a promising development in providing the intellectual energy to facilitate Indonesia s transition to a more dynamic and yet fair society. No process of social dialogue can be effective and durable if it does not build on a labour rights-driven industrial relations system. There is now a consensus in the international community that the recognition of basic worker rights is the foundation on which labour market institutions have to be built. The Indonesian government has made a promising start in this direction by ratifying all the core ILO conventions that pertain to the fundamental principles and rights at work. 97 It has thus acquired the enviable distinction of being the only country in the East Asian region to accomplish such a task. 98 Nevertheless, ratification is only the beginning. The next challenge is the sustained realisation of basic worker rights for all Indonesians. Only then would social dialogue work, labour unrest mitigated and the problem of powerlessness that afflict the poor at least partially addressed. 96 The ILO area office in Jakarta supports a project in which innovative attempts are being made to develop voice representation for the informal and unorganised sector. It is too early to say how effective this project is likely to be, but it is certainly a start in the right direction.

104 References Acemoglu, D and Shimmer, R (2000) Productivity Gains from Unemployment Insurance, European Economic Review, vol.44, pp Agenor, P (2002) Macroeconomic adjustment and the Poor: Analytical Issues and Cross-Country Evidence, January 14, mimeo, World Bank: Washington DC Agrawal, N (1995) Indonesia: Labour Market Policies and International Competitiveness, September, background paper to the World Development Report, Washington DC: World Bank Akerlof, G, Dickens, W and Perry, G (1996) The Macroeconomics of Low Inflation, Brookings Paper on Economic Activity, 1: 1-76 Alamsyah, H et al (2001) Towards Implementation of Inflation Targeting in Indonesia, Bulletin of Indonesian Economic Studies, 37(3) Asia Recovery Information Centre (ARIC) (2001) Asia Recovery Report, September 2001, Manila: Asian Development Bank Asia Recovery Information Centre (ARIC) (2002) Asia Recovery Report, Update, December 2001, Manila: Asian Development Bank Bangsal and Sari (2000). Indonesian Study on Voices of the Poor as cited in World Bank (2000) Bank Indonesia (2000) Economic, Monetary and Banking Developments, 2 nd qtr, mimeo, Jakarta Beattie, R (2000) Social protection For All: But How? International Labour Review, vol.139 (2), Berry, A, Rodriguez, E and Sandee, H (2001) Small and Medium Enterprise Dynamics in Indonesia, Bulletin of Indonesian Economic Studies, 37 (3) Booth, A (2000) Poverty and Inequality in the Soeharto Era: An Assessment, Bulletin of Indonesian Economic Studies, 36(1) Booth, A. (1999). Survey of Recent Developments. Bulletin of Indonesian Economic Studies, 35 (3), BPS/BAPPENAS/UNDP (2001) Indonesia Human Development Report 2001, Towards a New Consensus: Democracy and Human Development in Indonesia, Jakarta 97 These are: freedom of association, right to organise and bargain collectively, the prohibition of forced and child labour, elimination of discrimination of employment. 98 See ILO (1999b).

105 Breman, J. (2000). The Impact of the Asian Crisis on Work and Welfare in Village Java. Dies Natalis 2000 Address, delivered on October 12 on the occasion of the 48 th Anniversary of the Institute of Social Studies. The Hague: Institute of Social Studies. Campbell, D (1999) Globalization and Change: Social Dialogue and Labour Market Adjustment in the Crisis-Affected Economies of East Asia, mimeo, ILO/EASMAT, Manila Chaudhuri, S; Jalan, J and Suryahadi, A (2001) Assessing Household Vulnerability to Poverty: A Methodology and Estimates for Indonesia, paper presented to Workshop on Poverty and Vulnerability, Third Asia Development Forum, Bangkok, June Cloudeaul, A; Henstchel, J and Wodon, Q (2001) Well-being Measurement and Analysis, April, draft prepared under the PRSP, April, Washington DC: World Bank David, I.P., Asra, A. & de Castro, M. (1999). Poverty Incidence in the Asian and Pacific Region: Data Situation and Measurement Issues. Mimeo. Manila: Asian Development Bank. Dhanani, S (2002) Strengthening the Indonesian Labour Market Information System, February, paper prepared for the Recovery and Reconstruction Department, ILO: Geneva Dhanani, S (2001) Unemployment and Underemployment in Indonesia Before and During the Financial Crisis, May, Background Paper prepared for In-focus Socio- Economic Security Programme, ILO, Geneva Dhanani, S and Diah Widarti (2001) Indonesian Reform Agenda and Employment Prospects, December, mimeo, ILO Area Office, Jakarta Dhanani, S and Islam, I (2002) Poverty, Vulnerability in a Period of Crisis: The Case of Indonesia, World Development, forthcoming Dhanani, S. & Islam, I. (2000). Poverty, Inequality and Social Protection: Lessons from the Indonesian Crisis. UNDP/United Nations Support Facility for Indonesian Recovery (UNSFIR) Working Paper 00/01. Jakarta: United Nations Development Program. Dhanani, S and Islam, I (2001a) Labour Market Adjustment to Indonesia s Economic Crisis: A Comment, Bulletin of Indonesian Economic Studies, 37 (1): Dhanani, S. & Islam, I. (2001b). Indonesian Wage Structure and Trends, Background paper prepared for the Infocus Socio-Economic Security Program (ILO/SES). Geneva: International Labor Organization. Dercon, S and Krishnan, P (2000) Vulnerability, Seasonality and Poverty in Ethiopia, Journal of Development Studies, 36(6), pp.25-53

106 Edwards, S (1996) Labour Regulations and Industrial Relations in Indonesia, paper presented to the Ministry of Manpower-World Bank workshop on Indonesian Workers in the 21 st century, Jakarta, April 2-4 ESCAP (2000) Social Security and Safety nets in Economic and social survey of Asia and the Pacific, Bangkok Foster, J., Greer, J. & Thorbecke, E. (1984). A Class of Decomposable Poverty Measures. Econometrica, 52, ILO (1999a) Indonesia: Strategies for Employment-led Recovery and Reconstruction, Jakarta and Geneva: ILO ILO (1999b) Project document for technical support and training targeting the creation of 1.2 million jobs, in collaboration with AusAID, Geneva and Bangkok, ILO and Canberra, AusAID ILO (1999c) Demystifying the core conventions of the ILO through social dialogue: the Indonesian Experience, Jakarta and Geneva: ILO ILO (2000a) Employment-intensive investment in infrastructure: jobs to build society, Geneva: ILO ILO (2000b) World Labour Report: Income Security and Social Protection in a Changing World, Geneva: ILO Insan Hitawasana Sejahtera (2000) Provincial Poverty and Social Indicators SUSENAS , March, Jakarta, prepared for Asian Development Bank Islam, I (2000) Beyond Labour Market Flexibility: Issues and Options for Post-Crisis Indonesia, Journal of the Asia Pacific Economy, 6(3): Islam, R (2001) Poverty Alleviation, Employment and the Labour Market: Lessons from the Asian Experience, paper presented at the Asia and Pacific Forum on Poverty: Reforming Policies and Institutions for Poverty Reduction, Mania: ADB Islam, I and Nazara, S (2000a) Technical Note on the Indonesian Labour Market: Estimating Employment Elasticity for the Indonesian Economy, Jakarta: ILO Area office Islam, I and Nazara, S (2000b) Minimum Wages and the Welfare of Indonesian Workers, Occasional Discussion Paper No.3, Jakarta: ILO Country Office Jo, B and Ellwood, D.T. (1983) Slipping into and out of Poverty; The Dynamics of Spells, Working Paper No. 1199, Cambridge, M.A.: National Bureau of Economic Research Jorgensen, S.L. and Domelan, J.V. (1999) Helping the Poor Manage Risks Better: The role of Social Funds, paper at IADB conference on Social protection and Poverty, February 4-5. Forthcoming chapter in Lustig, N (ed) Shielding the Poor:

107 Social Protection in the Developing World, Washington DC: Brookings Institution and Inter-American Development Bank Kanbur, R. & Squire, L. (1999). The Evolution of Thinking About Poverty: Exploring the Interactions. Mimeo. Washington DC: World Bank. Lee, E (1998) The Asian Financial Crisis: The Challenge for Social Policy, Geneva: ILO Manning, C (2000) Labour Market Adjustments to Indonesia s Economic Crisis: Contexts, Trends and Implications, Bulletin of Indonesian Economic Studies, vol. 36 (1), Manning, C (1998) Indonesian Labour in Transition: An Indonesian Success Story?, Cambridge: Cambridge University Press Manning, C and Junankar, P.N. (1998) Choosy Youth or Unwanted Youth? A Survey of Unemployment, Bulletin of Indonesian Economic Studies, 34 (1) Mason, A and Baptist, J (1996) How Important are Labour Markets to the Welfare of the Poor? paper presented to the Ministry of Manpower-World Bank workshop on Indonesian workers in the 21 st century, Jakarta, April 2-4 McCawley, P (2000) Poverty in Indonesia: The Role of the State in the Post-Soeharto Era, paper presented to the 50 th Anniversary Conference, Faculty of Economics, University of Indonesia, Jakarta, 4-5 October Mingat, A and Winter, C (2002) Education for All by 2015, Finance and Development, 39 (1) Mishra, S (2001) History in the Making: a Systemic Transition in Indonesia, Journal of Asia Pacific Economy, forthcoming Morduch, J (1999) Between the Market and the State: Can Informal Insurance Patch up the Market? Stiglitz summer research workshop on poverty, July 6-8, Washington DC: World Bank Mukherjee, N (2000) Consultations with the Poor as cited in World Bank (2000b) Narayan, D et al (2000) Voices of the Poor: Can Anyone Hear Us? New York: Oxford University Press for the World Bank OECD (1996) Shaping the 21 st Cooperation, DAC, Paris:OECD Century: The Contribution of Development Padalino, S and Vivarelli, M (1997) The Employment Intensity of Economic Growth in G-7 Countries, International Labour Review, vol.136, no.2, pp Papanek, G (1999) The Impact on the Poor of Growth and Crisis: Evidence from Real Wage Data, paper presented to Conference on Economic Issues Facing the New Government, August 18-19, LPEM-FEUI/USAID/PEG

108 Parikesit, D and Hudayana, B (1999) Socio-economic summary of review of labourbased programmes, Phase 2, Working Paper No. 1, in collaboration with AusAID, Geneva and Bangkok, ILO and Canberra, AusAID Pradhan, M., Suryahadi, A., Sumarto, S. & Pritchett, L. (2000). Measurements of Poverty in Indonesia, 1996, 1999, and Beyond. Social Monitoring and Early Response Unit (SMERU), Jakarta. World Bank Working Paper Washington: World Bank Pritchett, L., Suryahadi, A. & Sumarto, S. (2000). Quantifying Vulnerability to Poverty: A Proposed Measure, Applied to Indonesia. Social Monitoring and Early Response Unit (SMERU), Jakarta. World Bank Working Paper Washington: World Bank Puguh, I, Ahmed, I and Islam, I (2000) Labour Market Dynamics in Indonesia: An Analysis of 18 Key Indicators of the Labour market, , Jakarta: ILO Ravallion, M (1998) Appraising workfare programs, Working paper No.1955, Washington DC: The World Bank Rama, M (2001) Globalization and Workers in Developing Countries, mimeo, July 24, Washington DC: World Bank Sadaah, F; Waters, H and Haywood, P (1999) Indonesia: Undernutrition in Young Children, East Asia and the Pacific Region, Watching Brief, January Issue 1, Washington DC: World Bank Sen, A.K. (1999) Development as Freedom. New York: Alfred Knof Skoufias, E; Suryahadi, A and Sumarto, S (2000) Changes in Household Welfare, Poverty and Inequality During the Crisis, Bulletin of Indonesian Economic Studies, 36(2), pp SMERU (2001a) Wage and Employment Effects of Minimum Wage Policy in the Indonesian Urban Labour Market, October, mimeo, Jakarta: SMERU Research Institute SMERU (2001b) Education Difficulties for the Poor: If They can be Exploited, Why Not? Newsletter, No.03, May-June Subbarao, K et al (1997) Safety Net Programs and Poverty Reduction: Lessons From Cross-Country Experiences, Directions in Development, Washington DC: World Bank Suryahadi, A and Sumarto, S (2001) The Chronic Poor, The Transient Poor and the Vulnerable in Indonesia Before and After the Crisis, Working Paper, May, Jakarta: SMERU Research Institute Suryahadi, A, Suharso, Y. & Sumarto, S. (1999). Coverage and Targeting in the Social Safety Net Programs: Evidence from 100 Village Survey Research working paper, Social Monitoring and Early Response Unit (SMERU). Jakarta: SMERU.

109 Suryahadi, A., Sumarto, S., Suharso, Y. & Pritchett, L. (2000). The Evolution of Poverty During the Crisis in Indonesia, 1996 to 1999 (Using Full Susenas Sample) Research working paper, Social Monitoring and Early Response Unit (SMERU), World Bank Research Working Paper Jakarta: SMERU. Sutanto, A and Irawan, P.B. (2000) Regional Dimensions of Poverty: Some Findings on the Nature of Poverty. Paper presented at the International Conference on Poverty Measurement in Indonesia, 16 May. Jakarta: Central Bureau of Statistics & World Bank. Tubagus, F (2000) The Impact of the Crisis on the Labour Market in Indonesia Report prepared for the Asian Development Bank, March 22, Manila UNDP (2000) Human development report: human rights and human development, New York; UNDP UNDP (2000). Human Development Report 2000, New York: UNDP UNSFIR (1999) The Social Implications of the Indonesian Economic Crisis: Perceptions and Policy. Discussion Paper No. 1, project of United Nations Development Program. Jakarta: United Nations Support Facility for Indonesian Recovery (UNSFIR). Van Ginnekan, W (ed.) (1999) Social security for the excluded majority, Geneva: International Labour Office Vroman, W (1999) Unemployment and unemployment protection in three groups of countries, Social Protection Discussion Paper no. 9911, May, Washington DC: World Bank. Widyanti, W; Sumarto, S and Suryahadi, A (2001) Short-term Poverty Dynamics: Evidence from Rural Indonesia, Working Paper, September, Jakarta: SMERU Research Institute. Wiebe, F (1996) Income insecurity and underemployment in the Indonesian informal sector, paper presented to the Ministry of Manpower-World Bank workshop on Indonesian workers in the 21 st century, Jakarta, April 2-4 World Bank (2000a) World Development Report, 2000/2001, New York and Oxford: Oxford University Press World Bank (2000b) Poverty Reduction in Indonesia: Constructing a New Strategy, Draft for Discussion, Jakarta: World Bank Country Office World Bank (2000c) Indonesia: Managing Government Debt and its Risks, May 22, East Asia and the Pacific Region, Washington DC: World Bank World Bank (2001) Indonesia Update, October, Washington DC: World Bank

110 A STUDY ON SUB-NATIONAL DIFFERENCES IN EMPLOYMENT, INCOME AND POVERTY LEVELS IN THE CONTEXT OF ECONOMIC CRISIS AND RECOVERY IN INDONESIA DIAH WIDARTI EXECUTIVE SUMMARY Scholars suggest that the financial crisis hit the Indonesian economy at a time when it was already facing dwindling exports, a slowdown in overall economic growth, and sharply reduced rice production leading to imports of rice and prospects of a sharp

111 rise in food prices, and not to mention the much needed political changes that was sweeping the country. It provides a critical appraisal on both the national and subnational levels of the impacts of the economic crisis in parallel to regional-specific differences in baseline conditions, such as demographic situation and economic production. Furthermore, the paper offers outcomes of subsequent post-crisis recovery management, again on both the sub-unit level and the country as a whole. In short, this paper attempts to present the effects of the financial crisis and economic recovery in different sub-national regions of the country in comparison to the overall composite national picture with specific focus on employment, income, and poverty situations. The higher levels of unemployment and most likely underemployment in the informal sectors and underemployment in formal segments might have led to declines in Indonesia s per capita GDP. Most provinces in Indonesia experienced significantly lower per capita GDP growth rates in 1999 than in The data seem to suggest that during the process towards recovery inter unit differentials of per capita GRDP has narrowed down. On the contrary, inter unit differentials in unemployment and underemployment seem to widen after the crisis (1999). The provinces of Maluku, Irian Jaya, Aceh, and Jakarta gave substantial contraction in earning potentials. All these provinces, except Jakarta, had some sort of social disputes whether they would be religious or ethnic. The loss of jobs and the decrease of employment opportunities play a major role in decreasing one s income-generating potential. Observation is made in that although the country s overall economic growth levels in 1999 have closely recovered to those levels experienced in 1996, and that no significant increases in open unemployment rates were transparent, the country s poverty incidence prove otherwise. Unemployment and underemployment rates may certainly have expanded in more informal sectors of the country s economy where they are difficult to measure, thus leading to an overall increase of the country s poor population. A direct correlation can be deduced between employment and income as well as income and poverty. It seems that the economic crisis has directly enlarged the poor and needy in the country. Consequently increasing poverty affected human development in Indonesia. The level of human development, reflected in human development index has declined from 1996 to Most provinces shared such a decline The paper believes that there should be good coordination frameworks in the country for understanding the challenges faced by a democratic Indonesia in the post-crisis period. Such challenges entail the need for nurturing employment-intensive recovery, developing a cordial and credible industrial relation system, and empowering workers to cope with the problems of vulnerability and risk that are inherent in any globallyoriented market economy. Moreover, the paper develops a number of propositions in conjunction with the outline of a packaged programme above mentioned. First, macro-economic policy should incorporate employment issues with more emphasis given on enhancing

112 employment intensity to support economic recovery and growth. Second, formal systems of social protection need to be nurtured and strengthened in dealing with labour market risks, particularly those triggered by economic shocks.

113 A Study on Sub-National Differences in Employment, Income and Poverty Levels in the Context of Economic Crisis and Recovery in Indonesia By DIAH WIDARTI 1. INTRODUCTION Up to mid 1997, East and Southeast Asian countries had the distinction of being amongst the world's fastest growing economies. Southeast Asian countries in particular were then dubbed as the Tiger Economies. The Asian Financial Crisis of drastically changed the region s economic status. It consequently drove governments of these countries to restructure their entire economic and political systems from one that were mostly autocratic to one that is now more transparent in line with the demands of market globalisation. The Indonesian economic recovery after the crisis has been very modest compared to its surrounding neighbours, and is expected to weaken in and beyond as the adverse ripple effects of the current global economic slowdown begins to mushroom out to the export-oriented economies of Southeast Asia when the region s economy has just begun to improve as recent as a year ago. With regard to Indonesia, it was quite clear then that this financial crisis stemmed from deeper economic malaise, and it is clear now that it is leading to immense socioeconomic problems in the country. However, the country s weak and varied national and sub-national economic recovery is largely contributed by several important issues that cannot be ignored, such as; regional differences in initial situations, economic effects, and subsequent recovery processes; social dimensions of employment, income, and living standards; and regional diversities in history, ecological conditions, settlement and migration patterns, production outputs, political and social structure, and economic performance levels. There are important social dimensions to such problems in the county including employment, income, and living standards, particularly within the poorer and vulnerable segments of the society. So far, economic growth in Indonesia has been heavily concentrated in parts of Java and Sumatera particularly within the provinces of Jakarta and Riau respectively. However, such growth has not been transparent in the wider context of the country, especially in its eastern parts such as the province of East Nusa Tenggara that is measured as the second poorest in the country. Although on a national level the Indonesian economy recorded gradual growth, it is observed that different regions and provinces responded variably to the crisis 99, and are therefore expected to progress disparately to one another and to the national standard. Reductions in unemployment and underemployment, increases in income, 99 Regional profiles prepared by the author are available on request.

114 and drops in poverty rates based on the country s overall GDP growth rates cannot be expected as there are striking differences on the national and the sub-national levels of baseline conditions, economic crisis effects, and subsequent recovery processes that are further complicated by each region or province continuing diversity in history and culture, political and social structure, geography and demography, and economic performance and production patterns. The result may therefore be that the expected impact of the economic downturn in 1997 and its subsequent recovery processes on different regions and provinces may not be consistent with the national average as growth rates at the national level may be accompanied by very different rates at the sub-national level. The diversity of Indonesia provides an interesting case study on the impacts of areaspecific differences in initial conditions, economic levels, and subsequent recovery management on the overall composite national picture. This paper addresses these issues. It deals with the impacts of the economic crisis and adjustments in different regions of the country with specific focus on employment, income, and poverty situations. Objectives of the Study Ultimately, this research study aims to provide a definitive picture of sub-national units (hereafter referred to as units ) dimensions of economic crisis effects and following recovery phases in Indonesia to produce a summary of critical issues and findings, and to generate appropriate recommendations to better take into account policies and programmes relating to economic growth, employment expansion, income creation, and poverty reduction. Specifically, the study s objectives are to: 1. Identify initial characteristics of different units within the parameters of: per capita income/consumption levels; production patterns; inequality and poverty; labour force size, structure and growth; employment, underemployment, unemployment and earnings levels. Provide information on political, social, and cultural differences where appropriate to explain units differences in crisis impacts and recovery responses. 2. Present major indicators of units including demography, economy, employment and income, and poverty incidence, and highlight key sectional differences between units in terms of various factors such as degree of dependence on specific industries and their respective production output. 3. Isolate dissimilarities in different units experiences in parallel to general impacts of the economic downturn and subsequent recovery processes across the country by identifying for example: units that did not experience economic downturn or where the impact was slight; concentration of economic impacts; differences in units recovery efforts from the national average; and distinct forces at work specific to some areas that have not had a nation-wide impact.

115 4. Describe impacts of the economic meltdown and subsequent efforts of crisis management and recovery responses of different units through analyses of recent trends in labour supply, employment, and earnings of different groups in the labour force. Relate units trends to their baseline characteristics. 5. Analyse the contribution to or mitigation of past and current government policies at national and sub-national levels on the effects of the economic crisis and the needed recovery process. Deliver recommendations to better address units dimensions of economic impacts and recovery management in the country particularly in relation to employment, income, and poverty levels. Scope and Methodology of the Study 100 Analyses of pre- and post-economic crisis periods only cover the Republic of Indonesia s 26 provinces after the province of East Timor was abolished subsequent to its independence from the country in Research and analyses within this paper are based on available secondary data, which are gathered from relevant government bodies and other institutions at the time. 102 To establish a baseline/benchmark for measuring the various factors previously stated, different units initial characteristics in the first half of the 1990s are identified and explored. These include: per capita income/consumption levels; production patterns; inequality and poverty; labour force size, structure and growth; employment, underemployment, unemployment and earnings levels. Political, social, and cultural differences are elaborated to clarify the differences in economic impacts and recovery responses. Subsequently, to measure the consequential effects of the economic downturn and efforts at recovery, recent trends from the second half of the 1990s (i.e. 1995) in labour supply, employment, and earnings of different groups in the labour force are 100 It is noted that this paper was completed on April 2002, and is based on information available to that period. 101 Although as part of the decentralization and de-concentration process (developed in 1995) that came with the fall of the New Order Government ( ), the country presently has 32 provinces which were the result of 6 new provinces being added onto the existing 26 provinces, only the original 26 provinces are included in this research study. The country s 26 provinces are grouped into the following primary regions; Sumatera D.I. Aceh, North Sumatera, West Sumatera, Riau, Jambi, South Sumatera, Bengkulu, and Lampung (8); Java D.K.I. Jakarta, West Java, Central Java, D.I. Yogyakarta, and East Java (5); Bali & Nusa Tenggara Bali, West Nusa Tenggara, and East Nusa Tenggara (3); Kalimantan West Kalimantan, Central Kalimantan;, South Kalimantan, and East Kalimantan (4); Sulawesi North Sulawesi, Central Sulawesi, South Sulawesi, and Southeast Sulawesi (4); Maluku & Irian Jaya Maluku and Irian Jaya (2). 102 These data are not completely accurate as there may be inadequate data gathering and surveying processes involved. The study only analyses the government s social and economic policies with regard to economic crisis and recovery management. The private sector s influence and its effects on economic recovery are not taken into account. The study takes into consideration regional differences prior to the 1997 economic crisis and their differences after the meltdown. Conclusions and recommendations of the study are based upon available data only and should not be taken in whole without additional research in the field.

116 evaluated. This assessment provides a foundation for remaining analyses and findings of the rest of the paper. The impact of the crisis and the recovery of each unit are traced and assessed. Focus is given on the economic crisis impacts on employment, income and earnings, and poverty levels. These changes are related to the baseline situation and the specific operational, political, social, and economic factors as well as the general economic impact and subsequent recovery process across the country. Analyses of sub-national economic impact and recovery process differences to the overall composite picture of the national experience are modelled based on these findings. These analyses may bring about issues such as: units that were not affected by the economic downturn; concentrations of the economic impacts on several units; differences in recovery processes between the national and sub-national levels; and specific regional or local factors. Government policies on both the national and sub-national levels are examined to observe whether they have contributed to, or mitigated effects of the economic crisis and whether they have contributed positively or negatively to the recovery process in the country. Finally, recommendations to better manage the impacts and recovery processes of the economic crisis on both the national and sub-national dimensions are produced in which they are particularly focused on the aspects of employment, income, and poverty. 2. Baseline 103 To measure any effects of the economic recession on Indonesia s employment, income, and poverty levels and to evaluate the country s recovery after such crisis, it is imperative that a baseline picture (pre-crisis condition) is established for use as benchmark. This chapter focuses on founding relevant baseline conditions of the above said variables by initially elaborating on the Indonesia s demography then subsequently relating the same variables to the country s economic performance prior to the crisis. Pre-crisis period used is between 1990 and Demography Indonesia is the largest and one of the most diverse archipelago nations with various ethnic groups, religious beliefs, and dialects. Inter-regional differences are the consequence of considerable variation in some characteristics that are important to securing the country s national stability: geophysical and economic condition, distribution of natural resources, and the social and cultural attributes of the various 103 Detailed tables are presented in Annex 1

117 ethnic groups. Provincial details of areas, administrative units, and population in 1990 and 2000 are presented Table The total population of Indonesia in 2000 reached million. This figure represented a population growth rate of close to 14 percent from the 1990 number of million. Broken down into regional areas 105, Figures 1 and 2 show that Kalimantan, Maluku and Irian Jaya, and Sumatera alone make up more than 80 percent of total Indonesia but only hold about 26 percent of the country s population in Whereas Java, being the most populous region with close to 60 percent of Indonesia s population in 2000, makes up only a mere 6.6 percent of the country s land-mass area, and therefore possessing the most numbers of regencies, municipalities, and villages in the country. Consequently, Java holds the highest population density in the country with 12,500 persons/km 2 and Irian Jaya maintains the lightest density at 5-persons/ km 2. The geographic condition of the country alone plays a large part in creating a substantial population imbalance. This is further augmented by the fact that most of Indonesia s long yet remarkable economic development in has been predominantly concentrated in the western part of the country than in its eastern region, consequently leading to heavy migration of people from less economically developed regions to higher economically active provinces for work, education, and family reunion reasons, among others. Figure 1. Percentage Land Area According to Primary Region, Indonesia Maluku & Irian Jaya 26% Sumatera 25% Sulawesi 10% Kalimantan 28% Bali & Nusa Tenggara 4% Java 7% 104 Prior to the decentralized government system, the Republic of Indonesia was divided administratively into 26 provinces (since 1999 East Timor were no longer part of Indonesia). The provinces were further subdivided into 268 regencies, 73 municipalities, 4,049 sub districts, and 69,050 villages (Central Board of Statistics, 2001). As part of the decentralization and de-concentration process (developed in 1995) that came with the fall of the New Order Government ( ), the country presently has 32 provinces, which were the result of 6 new provinces being added onto the existing 26 provinces. 105 The data refer to 1999 figures and do not reflect figures for new provinces. At the time of writing data separation for new provinces was not yet available.

118 Figure 2. Percentage Population According to Primary Region, Indonesia 1990 & Percentage of Total Populatio Sumatera Java Bali & Nusa Tenggara Kalimantan Sulawesi Maluku & Irian Jaya Pop. of 1990 (%) Pop. of 2000 (%) Primary Regions Migration As a country with many islands and a considerable amount of inter-regional disparities, one would expect that population mobility is rather high. The available data however shows that migration, and in particular recent internal migration had been relatively low 3.3 percent in 1990 declining to 2.5 percent in It seems that population mobility in the form of commuting is more prevailing. 106 Figure 2 also supports this finding where there are negligible population ratio differences in 1990 and in 2000 in the primary regions. Migration is usually associated with some possible reasons such as employment, education or others like family reunion. Table 2 shows that Jakarta as the country s capital received the highest proportion of recent migrants, 13.1 percent in Following are the provinces of West Sumatra (4.9%), Yogyakarta (4.6%), East Kalimantan (4.5%), and Riau (3.5%). Just before the crisis, migration to all these provinces, with the exception of Riau, declined. An increase in migration to the latter region is partly explained by one of Riau s area popularity, Batam Island Industrial Zone, in providing new employment opportunities and better quality of life. The figures stated above support previous deduction that internal migration in the country dominantly occurs from less economically developed provinces to more developed ones, especially those that enjoy high mining and quarrying activities such as oil and gas production as well as minerals exploration. In terms of international migration, this phenomenon has been becoming increasingly important in Indonesia (Adi 1995; Hugo 1995; Spaan 1994). Government 106 A large percentage of the population are observed to commute to and from work, especially in the case of those who live in the province of West Java but work in metropolitan Jakarta.

119 intervention in regulating international migration in Indonesia is well known. 107 International migration outflows from Indonesia is expected to continually increase in the future due to limited job opportunities in the country, increasing average level of education, proliferation of mass media, easy and cheap access to overseas countries, and developments in global and national transport systems. As Indonesia has indeed experience significant social changes, Table 3 demonstrates that international migration outflow from the country has also increased. The average level of education of the majority of Indonesian adults is increasing. In 1980, the percentage of population that completed lower or upper secondary schooling was only 12.4, but in 1990 this number increased to 24.5 (CBS, 1983 and 1993). Such change indicates that more Indonesians have experienced a reorientation of values and attitudes. Another change is related to the country s transport and mass communication. Indonesia has experienced enormous improvements in its transportation infrastructure. This has made remote villages accessible to the wider world. Additionally, the government has made enormous progress in disseminating information to the most remote villages in the country via television. 108 Improvements in these infrastructures have partly increased the likelihood of migration both on national and international levels. Hugo (1995) noted that the official data on legal outflows of Indonesian workers has shown a significant increase since the First Five Year Plan ( ). In (First Five Year Plan) there were 5,423 international migrants to the Middle East, Malaysia and Singapore. This number increased to 19,332, 94,921, 291,181 and 600,163 in (Second Five Year Plan), (Third Five Year Plan), (Fourth Five Year Plan), and (Fifth Five Year Plan), respectively. The official target for the Sixth Five-Year Plan ( ) is about 1.5 million workers. Meanwhile the undocumented system refers to the bulk of illegal workers movement from Indonesia to Malaysia that is not covered by official statistics. This movement occurs along two major routes from East Java-South Sumatra to Peninsular Malaysia and from Flores-South Sulawesi to Sabah and is catalysed by syndicates, recruiters (calo) and middlemen (taikong). Through these illicit systems, the chance of exploitation is high. Despite this drawback, many migrants still prefer these informal recruitment channels. The government has attempted various means to discourage the use of informal and illegal immigration channels as difficulties often arise when assistance are needed by these migrants should problems arise in the destined country apart Research on international migration in the country is still limited due to inadequate data collection systems and undocumented migration records. The latter has become the main constraint to conduct further research. 108 The ownership of television has increased from less than one in ten households in 1970 to one in three in 1990 (CBS, 1993). 109 Migrants preferences in the use of these informal passages might be due to the fact that they are unaware of safer and legal instruments in place or simply because they consider these unofficial recruitment channels to be faster and more effective (Spaan, 1994)

120 Household Number and Size In the past Indonesia had serious population growth problems, which were further, enlarged by the already existing dilemmas of unequal population distribution and high urban population densities. The government implemented several actions that include re-transmigration of people from heavily populated islands and provinces to ones that are less populated and development of a family-planning programme to control the growth in family sizes. The two were carried out quite successfully over the years, with more accomplishment encountered through the latter plan. The country s population still grew nevertheless at a more reasonable rate. In , provinces increased their number of households varyingly from one to the other Maluku and Irian Jaya grew the highest at 25.6 percent while Java grew the least at 12.8 percent. However, in general all provinces in Indonesia experienced a positive growth in the number of households yet a negative growth in the size of households. In other terms, as reflected by the growth of the number of households in each province, the population of each province grew within the given period of time. Nevertheless, the negative growth shown by the size of households suggests that the major number of households in each province encountered a diminishing size of household inhabitants. This pre-crisis trend is observed to be largely due to the success of the country s family planning program. The decline in household sizes can again also be attributed to the fact that there is increasing population mobility within households for matters such as education, work, and family reunion. Refer to Figures 3 and 4 as comparison between 1990 and It is observed that Indonesia s average household size in 1990 was 4.5 persons per household and declined to 4.3 persons per household. Varying household size growth rates were apparent in the provinces of the country with declines mostly encountered in provinces where their inhabitants have averagely higher education levels. Decreases in family size were also observed more in urban areas than in rural counties, and that family members movement from households for issues such as education or work also contribute to the decline in this factor.

121 Figure 3. Number of Household and Average Household Size According to Province, Indonesia, ,000 8,000 7,000 6,000 5,000 4, , ,000 1, Household ('000) Average Household Size Figure 4. Number of Household and Average Household Size According to Province, Indonesia, , ,000 8,000 7,000 6, , ,000 3, ,000 1, Household ('000) Average Household Size It is likely that some changes have occurred in the structure and functioning of families that may directly or indirectly affect household size and number. As stated by Hugo (1995), such changes include: (1) a tendency to move from extended towards nuclear family; (2) a decrease in the strength of patriarchal control structures; (3) a transition from marriage partners selected by parents to selection by individual partner; (4) a decrease in polygamy; and (5) an increase in marriage age and a decline in fertility rate of about 50 per cent in the last 20 years. He argues that such social change can influence the international migration decision-making inside the family in Indonesia.

122 Life Expectancy and Infant Mortality Rate The Indonesian government has been successful in improving the quality of life for its people; seen by a remarkable increase of life expectancies and a substantial decrease in infant mortality rates (IMRs). Most of these developments occurred during the first five years of the 1990s. Indonesia s life expectancy grew from 63 to 66 years old (5.3%) during the decade. A modest increase, but an increase nonetheless. Most regions stayed within 2 to 3 percentage points of the country s numbers. Readily available medical attention may play the most important role in both life expectancy and infant mortality rate. Easy access seems to be more important than affordability in the case of Indonesia. Balanced nutrition is most likely not the standard for most of the rural provinces, making it difficult for infants to reach a certain age or for mothers to give birth to healthy children. Organizations trying to reach people in rural areas will certainly encounter problems, leaving these people with no outside help and few crops to grow for their subsistence. Severe droughts and civil unrest in some parts of the country have also attributed to lowering life expectancies and increasing infant mortalities. Although life expectancy and infant mortality have enhanced over time, there seems to be a distinct disparity between the western and eastern parts of Indonesia, with the latter region lagging behind the national average. Human Development Index Table 4 proves that Indonesia has been successful in increasing the quality of its human development. Literacy rates in the country reached 85.5 percent in The highest literacy rates are encountered in the western part of Indonesia, while the lowest is discovered mostly in the eastern part of the country, namely Irian Jaya (67.4% in 1996). Another province that deviated negatively and considerably from the country average is West Nusa Tenggara. As expected more or less the provinces in Java hold the highest literacy rates in the country followed by those provinces in Sumatera. The country s literacy rate is of course correlated with its population mean years of schooling which reached close to 7 years between 1996 and In 1996, the Human Development Index (HDI) for the country is measured at The government s accomplishment in increasing this index can be directly correlated to the improvements in the average level of education and the increases in adjusted real per capita expenditure of its inhabitants. Economy The New Order government that took power in 1968 implemented a centralized economic system. This included in a raft of equalisation policies subsiding regional governments to help overcome inter-regional and inter-household economic inequalities. Such policies resulted in high rates of national economic growth and

123 sustained relatively low levels of income disparity. However, Indonesia s economic horizon in the 1990s changed significantly, with the already transparent economic problems having impacts on the country s overall economy. Production: Origin and Growth In 1996 Indonesia s gross domestic product (GDP) economic sectors in order of largest output are manufacturing (24.7%), tourism and trade (16.8%), agriculture (15.4%), mining and quarrying (9.1%), services (8.9%), business services and finance (8.8%), construction (8.0%), transportation and communication (7.2%), and power and utilities (1.2%). 110 Refer to Table 5. This has not always been the case. At the beginning of the First Five Year Plan ( ), the country s economy was largely dominated by the agriculture industry followed by the mining and quarrying sector. Indonesia experienced a change of direction in its economic mission with the start of foreign investment pouring into the country during the latter parts of that period, primarily within the mining and quarrying sectors namely for oil and gas exploration and minerals extraction projects. The government further altered the country s economy in the Second Five Year and Third Five Year Plans (i.e and respectively) by drawing further foreign direct investments in the country. Government subsidy, deregulation, and circular nepotism became the main attraction for foreign and domestic investors alike. In these periods, the manufacturing sector transpired to become the country s largest economic contributor. From the late 1980s to early 1990s, the country s economic expansion was remarkable by any world standard. Indonesia s gross domestic product (GDP) development or economic growth has been recorded to be consistently around 8 to 12 percent per annum. Nevertheless, as stated previously, economic growth and expansion were mostly concentrated in the western parts of the country than in the eastern ones. There were only a number of provinces in the eastern region, namely East Kalimantan and Irian Jaya that saw the benefits of the country s economic development mainly due to the fact that they had some sort of high mining and quarrying potentials. Despite this, these gains were very small, as administration centres in the respective region often funnelled profits for personal gains with the majority retraced back to Jakarta. Table 7 shows that in the pre-crisis period, Java as expected was the highest contributor to the country s national gross domestic product (GDP), supplying close to 60 percent. Following Java was Sumatera (22.5%), Kalimantan (10.1%), Sulawesi (4.5%), Bali and Nusa Tenggara (3.5%), and Maluku and Irian Jaya (2.7%). The western part of Indonesia contributes approximately 80 percent to the national GDP, while the eastern part supplies the remaining 20 percent. 110 GDP/GDRP is broken down into the following economic sectors; 1) Agriculture; 2) Mining and Quarrying; 3) Manufacturing; 4) Power and Utilities; 5) Construction; 6) Tourism and Trade; 7) Transportation and Communication; 8) Business Services and Finance; and 9) Services.

124 Investment The country s economic boom from mid 1980s to mid 1990s witnessed an unprecedented upsurge in both domestic and foreign investments. Foreign investments up to this period were increasing in value and were consistent. International companies were encouraged to invest in Indonesia via governmentinduced attractions such as subsidies, joint-venture partnerships, and lower taxes. The appeal of a low-cost base features offered by the country was not ignored by international corporations. Table 9 confirms that in 1994 there were 823 domestic projects valued at Rp.53.3 billion and correspondingly there were 449 foreign investments valued at Rp.23.7 billion. Both domestic and foreign investments concentrated in high-margin sectors such as mining and quarrying, manufacturing, and tourism and trade. The unequal distribution of investments in all the other sectors proves to be one of the main causes of the country s prolonged recession. Employment The country s labour force participation rate has always been higher within the male population than its female counterparts. Table 10 shows this case. In 1996, Indonesia s labour force stood at 85.7 million people, in which it was divided into 53 million males and 32.7 million females. The Central Board of Statistics (CBS) in 1996 estimates that that from the total male working age population only 83.5 percent were actively participating in that they were either employed or actively seeking employment. With regard to the female working age population, the estimate is lower at 50.7%. 111 It is clear from this comparison that male workers are more active (or more successful in finding employment) than female workers. Nevertheless, these figures are considered not to be an accurate measure for overall employment. The available data suggests that in this same year, the active population from the country s total labour force was only 66.2% implying that a higher non-economically active population level than 34.0 percent may indeed be in existent. As per Tables 11.a., 11.b., 11.c., and 11.d., in 1996, males between the years of (30.9%) dominated male employment in urban areas followed by those in the years age group (26.8%). In rural areas, this segmentation remained equal. Women in the years age group were also dominant in female employment in urban and rural areas. These tables also exhibit that there were no rural workers in the region of Jakarta as the whole area is considered to be urban (metropolitan). Rural areas median employment age group was years old, while urban areas median age group was younger at years old. Traditionally, employment was considerable in the agricultural sectors with the majority of the country s overall working population having less than primary school 111 Working age population in 1996 is referred to as population aged 10 years and above, while in 1999 it is population aged 15 years and above.

125 education (<6 years) about 50 percent and 75 percent for urban and rural areas. In urban areas the country s employed population were dominated by certificate holders of primary school education (<6 years), senior secondary school education (<12 years), followed by junior secondary school education (<9 years) and university education (<17 years). In rural areas, this ranking changed in that the employed population remained to be dominated by primary school education holders, followed by junior secondary school education holders, senior secondary school education holders, and university graduates. Tables 12.a., 12.b., 12.c., and 12.d. exhibit this information. The discrepancy during this baseline period indicated that although primary school certificate holders dominated the employment force, more and more of the employed segment were at least trying to complete their senior secondary school education for better employment opportunities or due to the different cultures enforced in urban areas that put more emphasis on education and its benefits. Tables 13.a., 13.b., and 13.c. show that Indonesia s agricultural sector in 1994 and 1996 took most of the country s actively employed population. Other industries that followed include tourism and trade, manufacturing, and services. Such domination by the agricultural sector can be party explained by the fact that most of the country s employed population have less than 6 years of formal education (primary school) in which informal work in this type of sector can effectively absorb at least in comparison to other industries. Although an increasing number of the country s inhabitants are taking up higher education, the transition of an agricultural-based economy to one that is say more manufacturing-orientated will consume some time. 112 The small percentage of the employed population education in university level schooling depicts the small proportion of the labour force that are employed in sectors such as banking and finance, power and utilities, as well as mining and quarrying. The latter two economic sectors heavily utilize plants and machineries thus depleting the need of real manpower. Open unemployment was recorded at 4.8 percent (4.3 million people) of the country s labour force in 1996 (Table 15.a and 15.b.). Nevertheless, a country like Indonesia accounts two-thirds of its employment through the informal sector (approximately 57 million people), in which underemployment that is more suited to describe less than ideal working status in this sector was recorded at 30 million people. This is supported by the fact that in 1996, as per Tables 14.a., 14.b., and 14.c., a large population of the country s labour force s status are as follows employees (32.5%), self-employed with family assistance (23.6%), unpaid family workers (21.5%), selfemployed without any assistance (21.1%), and employer (1.3%). Self-employment with and without family assistance as well as unpaid family workers can be considered informal sectors of employment in which underemployment often prevails. 112 Despite a large employed population in the agricultural segment, Indonesia s highest GDP contributing sector is the manufacturing industry higher employee productivities and output are profound in these types of industries coupled by increased efficiencies by the use of new technology in plant, equipment, and machinery.

126 Income The Indonesian government has been successful in increasing its per capita gross regional domestic product (GRDP). Higher per capita GRDP are mostly in the western part of the country, namely within the provinces of Java, and in those provinces that are highly active in the mining and quarrying sector such as oil and gas production as well as minerals extraction. Higher GRDP in major urban centres such as Jakarta and Yogyakarta is tied to higher cost of living. Per Capita Gross Regional Domestic Product (GRDP) With Oil & Gas The growth rate of per capita Gross Regional Domestic Product (GRDP) in Indonesia with oil and gas in 1996 was 6.2 percent. In the same year, East Kalimantan followed by Jakarta, Riau, and Irian Jaya, Aceh, and Bali possess significantly higher per capita GRDP than the rest of their peers. The country s average per capita GRDP stood at Rp. 2.2 million for that year where those above stated provinces per capita GRDP were recorded at Rp. 9.1 million, Rp. 7.9 million, Rp. 4.9 million, and Rp. 3.7 million, Rp. 3.1 million, and Rp. 2.4 million respectively for that year. Those provinces except Jakarta and Bali have high per capita GRDP due to their richness in natural resources, predominantly in oil and gas. Jakarta s high per capita GRDP is not surprising as it is the country s capital, and prior to the economic crisis of 1997/1998 and the government decentralization program subsequently after that, the province enjoyed much of the country s financial privileges. This particular province high per capita GRDP is also a consequent of the metropolis high living costs but not necessarily its living standards. Bali s higher than average per capita GRDP is largely contributed by the fact that it is an attraction for international tourists bringing into the country large amounts of foreign currencies that have significantly higher values than the Indonesian Rupiah. Per Capita Gross Regional Domestic Product (GRDP) Without Oil & Gas Interestingly enough, the country s per capita Gross Regional Domestic Product (GRDP) growth rate without oil and gas in 1996 was 6.5 percent, which is higher than its per capita GRDP growth rate with oil and gas in the same year (6.2%). Observation deduces that this may be the result of re-allocation of appropriate government concentration and efforts in those areas outside the country s mining and natural resources revenue generating industries. 113 This appropriation of government efforts led to the growth of other industries besides oil and gas. This course was probably taken as the Indonesia s national oil company (PERTAMINA) was one out of the many public sector companies (Badan Usaha Milik Negara BUMNs) that were targeted for privatisation as part of the country s agreement with the International Monetary Fund (IMF) and the World Bank, which therefore led to its significant restructurings. These restructurings led to many pre-planned projects to be postponed, further reviewed, or simply cancelled. Observations of steep declines in 113 Note that during the period , OPEC underwent substantial restructuring that resulted in lower market prices for oil and gas produce.

127 foreign investments and projects that were made in previous sections support the findings above. Data illustrates that in 1996, Jakarta, East Kalimantan, and Irian Jaya possess significantly higher per capita GRDP than the rest of their peers and the country s average per capita GRDP which stood at Rp. 1.9 million for that year. Their per capita GRDP are Rp. 7.9 million, Rp. 4.9 million, and Rp. 3.6 million respectively. This analysis although leaving out oil and gas still shows that East Kalimantan and Irian Jaya to hold the second and third highest per capita GRDP amongst all the country s other provinces. Similarly as before, it can be explained that Jakarta s high per capita GRDP is a result of it being the country s central province and capital enjoying many business and financial privileges that were previously inherent with the New Order Government ( ). Its high per capita GRDP is also a consequent of the metropolis high living costs. Poverty Table 18 obtained from SUSENAS of CBS illustrates that with regard to lowering poverty, the Indonesian government has been successful. Poverty in urban and rural areas in the country dropped from a combined 54.2 million in 1976 to 22.5 million in However on the same issue Table 19 sourced from the CBS estimates that in 1996, 13.6 percent of the urban population lived below the poverty line, and nearly 20 percent of the rural population were considered poor. It is observed that poverty remains to be a rural and agricultural problem in the country. This phenomena is partly explained by the fact that rural workers obtain less pay than urban employees, and that the agricultural sector have a high proportion of informal employment and underemployment leading to a high proportion of this population measured to be in poor conditions. Such factor has also attributed that much of the country s vulnerable groups lay in rural areas and particularly within the agricultural industries. Through cross-tabulation of various tables, it is observed that those types of workers that are considered to be poor are those that are self-employed in the agriculture, trade and manufacturing sectors, and to a certain extent, those that are wage-workers in the agriculture, manufacturing and construction industries. As anticipated, it is further observed that the incidence of poverty is more transparent in the eastern parts of the country than in the western regions. This is largely due to Indonesia s failure to equally distribute economic growth, wealth, and benefits a centralized, and less transparent government is partly to blame. 114 Using different estimate that was based on February 1998 SUSENAS standard however, the number of population below poverty line in 1996 was higher, i.e million.

128 Summary In summary this section reveals some of the following points. Indonesia possesses significant population imbalance, with some provinces being more heavily populated than others, leading to substantially over-dense urban areas than rural ones. The country s population grew at an average of 13 percent between 1990 and 2000, with varied growth recorded in the provinces. High population growth in certain provinces such as Riau and Yogyakarta are attributed by the regular flow of migrants to these regions for matters such as seasonal work, permanent employment, education, and family reunion, among others. In relation to its population size, the level of internal migration is not considerably high. It was only 3.4 percent in 1990 and declined to 2.5 percent in Nevertheless, international migration from Indonesia is expected to increase due to factors like limited job opportunities, economic hardships, increasing average level of education, explosion of mass media, easy and cheap access to overseas countries, and improvements in global and national transportation and communication systems. A large proportion of illegal workers migration is undocumented as workers pass through unofficial and illegal channels such as syndicates and middlemen, although the risk of exploitation is very high the reason is believed to be workers perception of higher effectiveness of these said avenues. With the country s economic boom during the late 1980s and early 1990s as well as with the gradual population growth, most provinces experienced increased growth in household numbers with Maluku and Irian Jaya growing the highest at about 26 percent and Java expanding the least at 13 percent. In all, nearly all provinces underwent decreases in average household size the country s average household size is 4.5 persons per household. This is clearly a result of the country s successful family planning programme. The country s social changes such as higher average education levels and movement of family members away from households have also attributed to the decline of this factor. Indonesia has been quite successful in increasing its people s life expectancy and reducing infant mortality rates (IMRs). Enhancement of these factors is directly correlated with the improvements of the country s human development index (HDI) that include level of education, literacy rates, and per capita expenses on personal development. Life expectancy and IMR are also strongly influenced by healthcare access and affordability, daily diet and

129 nutrition enrichment, and natural disaster incidence. Although, life expectancy and infant mortality rate have improved over time, there seems to be a disparity between the western and eastern parts of Indonesia with the latter region lagging to the national average. Pre-crisis periods saw a centralized system of governance and finance with stages of equalization policies subsidizing regional governments to help overcome inter-regional and inter-household economic inequalities resulting in higher rates of economic growth and low levels of income disparity. Like the national pattern, the most dominant economic sectors that contribute to the formation of the gross regional domestic product (GRDP) in the regions are the agriculture, manufacturing, and tourism and trade sectors. Pre-crisis, Java and its 5 provinces was the highest contributor to the country s national gross domestic product (GDP), supplying close to 60 percent. Following Java was Sumatra (22.5%), Kalimantan (10.1%), Sulawesi (4.5%), Bali and Nusa Tenggara (3.5%), and Maluku and Irian Jaya (2.7%). The western part of Indonesia contributes 80 percent to the national GDP, while the eastern part supplies the remaining 20 percent. The country s economic boom from mid 1980s to mid 1990s witnessed an unprecedented upsurge in both domestic and foreign investments, especially within the manufacturing, mining and quarrying as well as tourism and trade sectors. Indonesia s labour force participation rate slightly increased but nevertheless change was negligible. Traditionally, employment was considerable in the agricultural sectors with the majority of the country s overall working population having less than primary education (<6 years) about 50 percent and 75 percent for urban and rural areas. Rural areas median employment age group is years old, while urban areas median age group is younger at years old. Open unemployment was recorded at 4.8 percent (4.3 million people) of the country s labour force before the crisis. However, Indonesia accounts twothirds of its employment through the informal sector, in which underemployment that is more suited to describe less than ideal working status in this sector was recorded at 30 million people. The Indonesian government has been successful in increasing its per capita gross regional domestic product (GRDP). Higher per capita GRDP are mostly in the western part of the country, namely within the provinces of Java, and in those provinces that are highly active in the mining and quarrying sector such as oil and gas production as well as minerals extraction. Higher GRDP in major urban centres such as Jakarta and Yogyakarta is tied to their higher cost of living.

130 With regard to poverty, again the Indonesian government has been successful in lowering the country s poverty incidence from 54.2 million in 1976 to 22.5 million in In the country, poverty remains to be a rural and agricultural problem. Those types of workers that are considered to be poor are those selfemployed and wageworkers in the agriculture, trade, manufacturing, and construction sectors. Given unequal distribution of economic growth, wealth, and benefits, poverty incidences are more transparent in the eastern parts of the country than in the western regions. 3. CRISIS Clearly in 1997/1998 when the Asian Financial Crisis descended upon the region, Indonesia was one of the first countries that were severely affected by it. During the peak of the crisis, it became evident to foreign observers that the country may prove to suffer the worst of these consequences due to its unsound economic and political foundations. This section elaborates on the impacts of the crisis on the country s national and sub-national levels by analysing the crisis effects on those economic variables that were presented in the previous section. In this section, analyses make comparison of these variables in baseline conditions in pre-crisis period ( ) to their trends during post-crisis phase ( ). Demography Since population distribution in Indonesia is highly disproportionate, uneven population densities particularly in urban areas pose major problems that are encountered in any major and highly dense cities in other parts of the world. These include high unemployment and underemployment levels, crime rates, and poor housing conditions. Indonesia s population growth mainly occurred in the first half of the 1990s and had a downward trend in latter part of the decade. Population growth in the regions varied, but it was observed that higher growth occurred in those regions that have higher gross regional domestic product (GRDP) such as Riau (+43.3%), Southeast Sulawesi (31.3%), East Kalimantan (29.8%), Central Kalimantan (29.1%), Irian Jaya (28.1%), and West Java (+23.1%) offering better employment opportunities and life styles and therefore becoming targets of national (internal) migrations. Higher population growth rates in some regions were also caused by a migration of people due to ethnic and religious conflicts. In all, the economic crisis that may also have induced religious and ethnic differences, contributed to population movement from one province to the other, namely from one that is less prosperous to one that is able to offer more opportunities to alleviate some of the hardships that were caused by the crisis. Figure 5 below illustrates that the percentage population growth rates during the period ranges from 1.5 percent in Jakarta to 43 percent in Riau, with the country s overall population growing just shy of 14 percent during the same period. The small growth rate in Jakarta reflects the province s success with the family

131 planning program and higher education levels of its inhabitants assuming to lead to better understanding of smaller family sizes. However, this rate may also reflect the fact that Jakarta s many inhabitants may have lived in the outskirts of Jakarta such as in Bogor, Bekasi and Tangerang (Jabotabek areas) and are being recorded as West Java s population instead of Jakarta s. West Java population in that decade has grown by over 23 percent. The most likely scenario may also be that the city s population may have left the area during the period, especially after the economic crisis, largely in part due to economic hardships and a lack of employment opportunities. Those provinces with above average population growth rates such as Riau (43.3%), Southeast Sulawesi (31.3%), East Kalimantan (29.8%), Central Kalimantan (29.1%), and Irian Jaya (28.1%) are observed to have high mining and quarrying activities, many of were the target of foreign and domestic investments during the country s economic boom in the first quarter to the third quarter of last decade, and thus they have become primary migration areas for people from other provinces that have fewer economic activities and offer less employment opportunities. 115 Figure 6 shows that only the provinces of Riau, West Java, Bali, and East Nusa Tenggara had higher annual population growth rates in the period than in the period In East Nusa Tenggara, it is expected that this higher population growth rate between those two periods were caused by a migration of people from the ethnic and religious conflicts that were occurring in its neighbouring province, East Timor. It is also anticipated that the higher annual population growth rate in the province of Riau may also be caused by population migration from other neighbouring provinces in the island of Sumatera given the fact that Riau is the wealthiest province on the island due to its oil and gas production. This population increase may also most likely be brought about by the influx of people entering one of Riau s offshore area namely Batam island, known as an Industrial Zone, for work reasons and employment opportunities. 115 However, note that these provinces growth rates indicate not only the population migration from one province to the other but also the growth rate of each province s existing inhabitants.

132 Figure 5. Population Growth According to Province, Indonesia, (%) INDONESIA Irian Jaya Maluku S outheast Sulaw esi S outh Sulawesi C entralsulaw esi N orth Sulaw esi East Kalim antan South Kalim antan C entralkalim antan West Kalim antan East Nusa Tenggara West Nusa Tenggara Bali East Java D.I.Yogyakarta C entraljava West Java D.K.I.Jakarta Lam pung B engkulu South Sum atera Jam bi Riau West S um atera N orth S um atera D.I.A ceh Percentage Growth Figure 6. Annual Population Growth Rate According to Province, Indonesia, & (%) East Timor Irian Jaya Maluku Southeast Sulawesi South Sulawesi Central Sulawesi North Sulawesi East Kalimantan South Kalimantan Central Kalimantan West Kalimantan East Nusa Tenggara West Nusa Tenggara Bali East Java DI Yogyakarta Central Java West Java DKI Jakarta Lampung Bengkulu South Sumatera Jambi Riau West Sumatera North Sumatera D.I.Aceh Percentage Growth

133 Figure 7 and Table 3 reveals that the number of Indonesian migrant outflows overseas increased in the period , and is expected to further incline with the globalisation of international markets. The figure and table stated present information on Indonesian nationals that are known to be working in the four major world regions of Asia Pacific, America, Middle East, and Europe between 1995 and It is noteworthy that although in total there was an increase of close to 400 percent from 120,603 overseas nationals in 1995 to 457,876 nationals in 2000 in more detail these increase were only consistent in Asia Pacific and the Middle East, whereas the number of nationals overseas actually declined in America and Europe in the same period. This may be due to more difficult access, less official and unofficial channels, tougher immigration laws, lower demand for Indonesian workers (unskilled labour) or simply the inadequacies of the Indonesian government in documenting and monitoring the number of migrant outflows to these countries. Figure 7. Number of People Working Overseas According to World Regions, Indonesia, 1995, 1996, 1997, 1998, 1999 and 2000 Total World Regions Europe Middle East America Asia & Pacific 0 100, , , , , ,000 Asia & Pacific America Middle East Europe Total ,894 1, , , ,803 3, ,636 1, , ,574 2, ,937 1, , , , , ,070 1, , , ,436 3,483 47,524 1, ,603 Number of People Household Number and Size In general in the period , as per Figure 8, all provinces in Indonesia experienced a positive growth in the number of households yet a negative growth in the size of households.

134 Figure 8. Number of Household and Average Household Size According to Province, Indonesia, ,000 6 Household Number ('00 12,000 10,000 8,000 6,000 4, Household Size (Person 2, D.I.Aceh North Sumatera West Sumatera Riau Jambi South Sumatera Bengkulu Lampung DKI Jakarta West Java Central Java DI Yogyakarta East Java Bali West Nusa Tenggara East Nusa Tenggara West Kalimantan Central Kalimantan South Kalimantan East Kalimantan North Sulawesi Central Sulawesi South Sulawesi Southeast Sulawesi Maluku Irian Jaya Household ('000) Number of People/Household 0 Figure 9 shows that the highest and smallest household growth during the period were experienced by the regions of Maluku & Irian Jaya (25.6%) and Java (12.8%). However, during the period , Maluku & Irian Jaya which during the previous period ( ) experienced the highest growth in household number amongst the six primary regions of the country, drastically altered its trend by incurring the smallest increase (0.71%) in household number while the largest increase was encountered in Kalimantan (15.5%). From this, it can be deduced that during the early part of that decade where the country experienced economic boom, all provinces experienced an increase in household numbers but in the latter part of the same decade, particularly when negative economic conditions started to prevail in the country after the financial crisis, all provinces with the exception of Riau, Central Java, and North Sulawesi decreased the size of their household numbers. Figure 10 on the other hand illustrates that all provinces incurred steep decreases in the number of household size, with Maluku (-11.1%) holding the most drastic decrease in household size during the period and West Java (-17.8%) experiencing the most drastic decrease in this segment during the period The latter deduced that West Java being the closest province to Jakarta was largely affected by the country s economic crisis or otherwise has done extremely well with the country s family planning program.

135 Figure 9. Growth in Household Number According to Province, Indonesia, & (%) D.I.Aceh North Sumatera West Sumatera Riau Jambi South Sumatera Bengkulu Lampung DKI Jakarta West Java Central Java DI Yogyakarta East Java Bali West Nusa Tenggara East Nusa Tenggara West Kalimantan Central Kalimantan South Kalimantan East Kalimantan North Sulawesi Central Sulawesi South Sulawesi Southeast Sulawesi Maluku Irian Jaya Percentage Growth Figure 10. Growth in Household Size According to Province, Indonesia, & (%) Percentage Growth D.I.Aceh North Sumatera West Sumatera Riau Jambi South Sumatera Bengkulu Lampung DKI Jakarta West Java Central Java DI Yogyakarta East Java Bali West Nusa Tenggara East Nusa Tenggara West Kalimantan Central Kalimantan South Kalimantan East Kalimantan North Sulawesi Central Sulawesi South Sulawesi Southeast Sulawesi Maluku Irian Jaya Life Expectancy and Infant Mortality Rate As illustrated in Figure 11, provinces data for life expectancy are very similar to that of their respective region. An exception is West Nusa Tenggara which had no growth between , but grew an unsurpassed 18.4 percent in the first four years of the decade. Unfortunately even with that growth, West Nusa Tenggara still has the lowest life expectancy in Indonesia, 49 years old in 1990 and 58 years old in It should be noted that East Nusa Tenggara also had no growth during the latter part of the decade. Jakarta (71 years), Yogyakarta (71 years) and Bali (70 years) were the provinces with the highest life expectancies in Is it possible that the standard of

136 living in big cities, although polluted and over-stressed, might be better than that of the countryside? Or could it be that one s standard of living and one s life expectancy are not correlated. Figure 11. Growth in Life Expectancy According to Province, Indonesia, and Percentage Growth Aceh during the period experienced a 1.5 percent increase in life expectancy. This factor further enhanced in the following period ( ) by another percentage point to 3 percent. Out of all the eight provinces in Sumatera, only the province of South Sumatera underwent similar pattern in which its inhabitants life expectancy grew from 1.8 percent during to 3.3 percent in the period Figure 12 illustrates that Infant Mortality Rate (IMR) had a downward sloping trend during the late 1990 s. Again, as with life expectancy, West Nusa Tenggara has the worst numbers, a 19.8 percent decrease was only enough to bring it down to 81 deaths per 1000 live births, from 101. East Nusa Tenggara is the only province whose IMR increased (3.7%); only three other Provinces have a higher IMR, Central Sulawesi (60), South Kalimantan (67) and the above mentioned West Nusa Tenggara (81). The Province of South Sulawesi had a remarkable 32.1 percent decrease in its IMR (from 52 deaths per 1000 live births to 36), followed closely by Central Java (-30.8%) and Yogyakarta (-30.6%). Not only do Jakarta and Yogyakarta have the highest life expectancy rates but they also have the lowest incidence of IMR in the country (24 and 25 deaths per 1000 live births, respectively). Neither provinces nor regions deviated too much from the average decrease in IMR for Indonesia (-18.1%).

137 Figure 12. Infant Mortality Rate According to Province, Indonesia, 1990, 1996, 1997 and D.I.Aceh North Sumatera West Sumatera Riau Jambi Mortality per 1000 Live Births South Sumatera Bengkulu Lampung DKI Jakarta West Java Central Java DI Yogyakarta East Java Bali West Nusa Tenggara East Nusa Tenggara West Kalimantan Central Kalimantan South Kalimantan East Kalimantan North Sulawesi Central Sulawesi South Sulawesi Southeast Sulawesi Maluku Irian Jaya Human Development Index In 1996, the Human Development Index (HDI) for the country is measured at Between 1996 and 1999 the HDI fell averagely by 2 points with the steepest decrease suffered by North Sulawesi (-2.6) and the least decline experienced by East Nusa Tenggara (-1.1). Jakarta, Yogyakarta, East Kalimantan, and Riau rank the highest in HDI and West Nusa Tenggara, Irian Jaya, West Kalimantan, and East Java possessing the lowest ranks in the country. Table 4 presents the information mentioned above in more detail. Economy The crisis significantly changed Indonesia s economic horizon in the latter part of , as there were already transparent economic problems within the country s overall economy. Production: Origin and Growth With regard to the overall picture during the 1990s (1996 and 2000 data), as per Figure 13, Indonesia s Gross Domestic Product (GDP) in 1996 vs is predominantly made up of these economic sectors; agriculture (15.4% vs. 16.7%), mining and quarrying (9.1% vs. 9.4%), manufacturing (24.7% vs. 26.2%), power and

138 utilities (1.2% vs. 1.7%), construction (8.0% vs. 6.0%), tourism and trade (16.8% vs. 16.0%), transportation and communication (7.2% vs. 7.4%), business services and finance (8.8% vs. 6.9%), and services (8.9% vs. 9.6%). It is observed that the nine categories of economic sectors in proportion to the country s overall GDP during the period of remained constant with the manufacturing, agriculture, and tourism and trade sectors being the third largest significant contributors to Indonesia s total GDP. 116 Figure 13. Gross Domestic Product (GDP) at Constant 1993 Market Price According to Economic Sector With Oil & Gas, Indonesia 1996 & 2000 (%) Percentage of Total GDP Economic Sectors However, in more detail the figures from 1996 to 2000 show that the country s economic activities slightly changed within that period in that the agriculture, mining and quarrying, manufacturing, power and utilities, and services sectors contributed more to the country s overall GDP, while the construction, tourism and trade, transportation and communication, and business services and finance industries played a lesser part in the country s total GDP. Tables 5 and 6 present available data on these sectors. The highest increases were observed in both the manufacturing and agricultural sector hence inferring that the country may have increased more informal activities than those of formal ones. Informal activities are less difficult to achieve in the two sectors as they can be rural-based and do not require extensive working capital or research and development costs GDP/GDRP is broken down into the following economic sectors; 1) Agriculture; 2) Mining and Quarrying; 3) Manufacturing; 4) Power and Utilities; 5) Construction; 6) Tourism and Trade; 7) Transportation and Communication; 8) Business Services and Finance; and 9) Services. 117 The increase in the manufacturing sector as depicted in Figure 13 may be a reflection of increased informal activities in the home-based industries and small and medium enterprises (SMEs) activities that were widely encouraged by the government subsequent the country s economic crisis.

139 As stated early in the paper, for analyses purposes, Indonesia has been segregated into six (6) primary regions and twenty-six (26) provinces. Table 7 presents the country s GRDP according to provinces and regions, and shows that Java contributed approximately 57 percent to the country s overall economy in This however, is a decrease of about 3 percent from 1996 figures when this region alone provided nearly 60 percent towards the entire economy. Consequently, the remaining regions of Sumatera, Bali and Nusa Tenggara, Kalimantan, Sulawesi, and Maluku and Irian Jaya, in 2000, although very modestly, increased their inputs to the country s overall economy. In that year these said regions contributions are; Sumatera (22.5%); Kalimantan (9.9%); Sulawesi (4.6%); Bali and Nusa Tenggara (3.8%); and Maluku and Irian Jaya (2.6%). Through further examination, it is discovered that within Java, four out of five provinces contributed nearly equally (11.0% to 15.0%) towards the region s economy with the lowest contributor being Yogyakarta at 1.3 percent. Furthermore, it is also noted that the two most common contributors to the region s economic well-being is held by the manufacturing industry as well as the tourism and trade sector. Briefly, other provinces that had given significant contribution to Indonesia s overall economy are given as follows; North Sumatera (6.0%); East Kalimantan (5.5%); and Riau (5.4%). Table 7 illustrates that in all, the western part of Indonesia supplied over 80 percent to the country s economy while the eastern region gave the remaining 20 percent. The figures below, Figure 14 and 15 explain the country s overall consistency of economic activities as a decrease in the agricultural and mining sectors and an increase in the manufacturing sector in the western part of Indonesia in 2000 is balanced by an increase in the agricultural and mining sectors and a decrease in the manufacturing sectors in the eastern part of the country in the same period. Figure 14. Gross Regional Domestic Product (GRDP) at Constant 1993 Market Price According to Economic Sectors With Oil & Gas, Western Indonesia 1996 & 2000 (%) Percentage of Province Total GRDP Economic Sectors

140 Figure 15. Gross Regional Domestic Product (GRDP) at Constant 1993 Market Price According to Economic Sectors With Oil & Gas, Eastern Indonesia 1996 & 2000 (%) Percentage of Province Total GRDP Economic Sectors Figure 16 exhibits that in general Indonesia s five regions (Sumatera, Java, Bali and Nusa Tenggara, Kalimantan, and Sulawesi) GRDP rate of change out of the total six correlated quite closely to one another for the period Only in the regions of Maluku and Irian Jaya did GRDP movements sway away from the rest. Further analysis discovers that Maluku is the main culprit for the region s difference in economic growth rates as its region s economy contracted by almost 25.0 percent in 1998/1999. Figure 16. Gross Regional Domestic Product (GRDP) Rate of Change at Constant 1993 Market Price According to Primary Region With Oil & Gas, Indonesia 1996/1997, 1997/1998, 1998/1999, & 1999/2000 (%) 15.00% 10.00% Rate of Chang 5.00% 0.00% -5.00% % % % 1996/ / / /2000 Sumatera 3.98% -7.53% 1.96% 4.54% Java 4.42% % 1.67% 4.22% Bali & Nusa Tenggara 5.63% -3.53% 1.70% 10.99% Kalimantan 5.24% -2.95% 3.64% 2.58% Sulawesi 4.68% -4.56% 3.44% 5.12% Maluku & Irian Jaya 6.23% 7.29% -8.80% 1.36% INDONESIA 4.49% % 1.69% 4.33%

141 The stated figure also depicts that in 1996/1997, prior to the country s economic crisis that hit in 1997/1998; Indonesia s economy was expanding by close to 4.5 percent with which the country s highest growth rates have been recorded at around 8 to 12 percent in the early part of the 1990s. Subsequently, in the period 1997/1998 all the country s primary regions except Maluku and Irian Jaya experienced negative growth rates that ranged between -3 percent in Kalimantan to -16 percent in Java (Indonesia as a whole contracted by percent). Maluku and Irian Jaya witnessed a positive growth rate of 7.3 percent attributed to Irian Jaya s economic development of about 12.7 percent. A specific study on the latter province proved that an economic boom in the mining sector did indeed occurred in Irian Jaya in the period It can also be deduced from the figure above that in order of severity caused by the country s recession during 1997/1998; Java was most affected, contracting by nearly 16 percent. Others that follow are; Sumatera (-7.5%); Sulawesi (-4.6%); Bali and Nusa Tenggara (-3.5%); and Kalimantan (-3.0%). To date (1999/2000), Indonesia s economy that grew at 4.3 percent has rebounded close to where it was in 1996/1997 (+4.5%). The praise for much of this growth can be given alone to the region of Bali and Nusa Tenggara that grew nearly 11.0 percent in the same period. Additional examination uncovers that West Nusa Tenggara s economy during this period expanded by close to 35 percent accredited to a significant increase in economic activities (+24.0%) in the mining and quarrying sector during Figure 17 and Table 5, 6 and 8 demonstrate this case. Figure 17. Gross Regional Domestic Product (GRDP) Rate of Change at Constant 1993 Market Price According to Province With Oil & Gas, Indonesia 1996/1997, 1997/1998, 1998/1999, & 1999/2000 (%) 40.00% 30.00% 20.00% Rate of Change 10.00% 0.00% % % % Nort West DI h Sum Aceh Sum atera Riau Jam bi Sout Lam DKI Cent DI Beng West East h pun Jaka ral Yogy kulu Java Java Sum g rta Java akar West East West Cent Sout East Nort Cent Sout Sout Bali Mal Irian Nus Nus Kali ral h Kali h ral h heas uku Jaya a a man Kali Kali man Sula Sula Sula t 1996/ / / / / / / /2000

142 Investment Table 9 shows both approved domestic and foreign projects and investments in all provinces of Indonesia. On a national level, the number of domestic projects in Indonesia declined from 823 units in 1994 to 237 units in However, the actual investment value in those same years remained nearly equal with only 0.4 percent increase from 1994 value of Rp. 53,289 million to 1999 value of Rp. 53,502 million. Domestic investments decreased in 14 provinces out of the 26 provinces in Indonesia, with the highest investment value decline of 91.8 percent in Yogyakarta followed by 91.1 percent in North Sulawesi. Other significant domestic investment deterioration from the period was recorded in Central Java (-85.3%), East Java (81.2%), Jakarta (-78.9%), Lampung (-64.3%), and South Sumatera (-58.6%). Domestic investment declines were balanced by sharp rises in domestic investments in other provinces of the country. Two significant increases in domestic investments during the period were observed in Central Sulawesi (1,660.2%) and Central Kalimantan (307.9%). The provinces of Maluku and Irian Jaya also saw a drastic increase in domestic investments from Rp. 748 million in 1994 to Rp. 8,436 million in 1999, representing an incline of about 1,027 percent. Although domestic investments grew modestly in the period , this was not the case with foreign investments. On an overall scale, foreign investments in Indonesia decreased by over 54 percent from a value of Rp. 23,724 million in 1994 to a value of Rp. 10,891 million in It is clear that after the economic crisis hit the country, many foreign projects and investments were either postponed or cancelled. Nearly all foreign investment values dropped in all Indonesia s 26 provinces, negative percentage changes in the period range between percent to over percent. Nevertheless, there were two provinces in that period that experienced a substantial increase in foreign investments; Yogyakarta (5,150.0%) and West Kalimantan (1,224.7%). Abnormally high percentage increases in these two provinces are explained by the initial lack of foreign investments in the base period of It is most likely that foreign investments in West Kalimantan are allocated mostly in the mining and quarrying sector; logging or tourism industry. Employment Indonesia experienced a small overall increase in labour force participation rate between 1996 and 1999 of 1% points taking it up to 94.8 million people. However, the total active male working population (working or actively seeking work) in this year remained the same as in 1996 at 83.5 percent. Similarly, the female working population also recorded a figure of 51.2 percent of active workers in its segment. The increase in LFPR combined with a stagnant employed population unfortunately might just be an insight into the unemployment increment that occurred during this period. Since LFPR is comprised of the employed population, the population actively looking for work and of course the underemployed, it can be alleged that an increase in LFPR simply means an increase in either unemployment or underemployment. Similarly as in 1996, it is deemed that the level of Indonesia s non-economically

143 active population is much higher than the figure reported especially since the country continues to feel the grip of the economic crisis. There were two provinces that enjoyed a significant increase in worker participation rates during They were Irian Jaya (39.9%) and South Sumatera (27.0%). The increase in employment in the province of Irian Jaya may be explained through the rise of mining and quarrying activities in that province during However, the expansion of employment in Sumatera cannot be readily justified as the level of economic activities in this province remained quite constant in the same period, with only minor growths in the agricultural as well as tourism and trade sectors. Further analyses discovered that the levels of combined labour participation rate from for both males and females decreased in twenty (20) out of the total of twenty-six (26) provinces. Notable decreases in workers participation levels were observed in Aceh (-5.3%), Central Kalimantan (-4.8%), East Nusa Tenggara (-3.0%), Lampung (-2.8%), North Sulawesi (-2.4%), and West Kalimantan (-2.1%). The 5.3 percent decline in Aceh s active participation levels from 1996 to 1999 was contributed by a significant 10.3 percent decrease in female labour force participation rate in the region. This is also the case for at least half a dozen provinces, whose female LFPR had a significant decrease, more so than that of male LFPR. This corresponds to the negative levels of economic growth that has swept the region from due to political instabilities, although in the period 1999/2000 the economy has been able to grow at a very modest rate of 0.5 percent (refer to Table 8 and Figures 16 and 17). On the other hand, the labour participation rate decline in Kalimantan was due to the increased level of both male and female workers in the region, although an economic growth of about 3 percent was recorded in 1999/2000. The decrease in female LFPR might be due to various factors, amongst those, the female population simply giving up on seeking employment and therefore not being part of the LFPR anymore and/or population aged 15 and under is no longer part of the LFPR. Most of Indonesia s working population is between the ages of In 1996 the group trailed for second place by about 4 percentage points but by 1999 the gap had narrowed to approximately 2.5 percentage points. There appears to be an increase in the number of older people working. The percentage of population aged 65+ increased by about half of a percentage point for Indonesia as a whole but it had regional increases of up to 4 percentage points, such is the case of North Sulawesi. Although Yogyakarta still has the highest population aged 65+ working, in urban areas from 6 percent to 8 percent and in rural areas from 9 percent to 11 percent. In Indonesia s urban areas Riau and East Kalimantan experienced significant changes in the components of its employed females by age group. Riau for example had a drop in the female employed population aged from 35.5 percent to 22.4 percent between 1996 and 1999, with no accounting in the other age groups which did not experience the same dramatic changes. East Kalimantan experienced similar changes in that age group, however, looking closely it is obvious in East Kalimantan that the

144 work load is shifted to those aged which increased by over 12 percentage points between 1996 and Higher education might explain why the younger population is no longer forming part of Indonesia s LFPR. By 1999 Indonesia had seen an increase in higher education of its employed population, particularly within its urban population and even more so for females in urban areas. In a time of financial crisis the more education a person has the better prepared that person is for any type of changes (Indonesians may have realized that during the crises). Even though unemployment rises, those with more education are likely to replace those people that are less educated, many times they will be over qualified for the job but they will have a job nevertheless. During 1996 to 1999 the percentage of Aceh s urban female population with a university degree jumped from 12.9 percent to 28.2 percent, the largest increase of any province for any education category. In urban areas Aceh is consistently one of the provinces with the highest education levels. Also consistent is West Nusa Tenggara, however, on the other extreme with the lowest levels of education within its employed population. The education outlook for Indonesia s rural employed population does not look promising, even though there was a slight increase in the percentage of those with higher levels of education. On average only about 1 percent of Indonesia s rural employees have university level education. A few provinces decreased the percentage of rural employees with university level education between 1996 and A good example is Irian Jaya who by 1999 had 0 percent of female employees with university level education. This decrease in highly educated employees might be due to a migration from urban to rural areas because of difficulty in finding work that requires (or at least values) university education. Agriculture the principal economic sector in Indonesia has experienced a decline from 46.2 percent in 1994 to 43.3 percent in The stronghold that Irian Jaya held in 1994 and 1996 by having the highest percentage (76.2% and 73.1%, respectively) of employed population under this sector was lost to East Nusa Tenggara by The decrease in agriculture employment might be indicative of a shift to mining and quarrying or possibly an even more extreme transition to business services & finance and/or simply the services industry, all of which experienced increases in the number of people employed between 1996 and Power & utilities, construction, and transportation & communication all increased, although slightly the percentage of people they employ between 1996 and Tourism and trade employees jumped between 1994 and 1996; however after the crisis by 1999 the numbers went back down to below 1994 numbers. Lack of tourism and desire to trade because of Indonesia s domestic political problems (i.e. riots in 1998) may be to blame for the decrease in employment in such sector. The rate of open unemployment in Indonesia experienced a slight decrease between 1996 and 1997 from 4.8 percent to 4.7 percent and again between 1999 and 2000 from 6.4 percent to 6.1 percent (Table 15.b.). However, during the crisis open unemployment was affected just like everything else with an increase of just under 2 percentage points between 1997 and 1999 from 4.7 percent to 6.4 percent. This might

145 seem small considering the country s financial troubles but the problem in Indonesia is not open unemployment but underemployment. The vast majority of people that lost their jobs or businesses due to the crisis cannot afford to stay at home and wait for a full time job with at least minimal wages to come along. This means that by working a few hours a week or worse working many hours with pay so little that it would be impossible to live on, they will not be part of the open unemployment figures. Open unemployment figures them seem as though Indonesia s unemployment rate is very low and therefore poverty levels are just the same. Income Per Capita Gross Regional Domestic Product (GRDP) With Oil & Gas The growth rate of per capita Gross Regional Domestic Product (GRDP) in Indonesia with oil and gas in 1996 was 6.2 percent. However, as recorded in 1999 the country s per capita GRDP was -0.5 percent. This in part is a reflection of the country s slow and painful economic recovery. Despite this fact, as per Figure 18 and Table 16, the picture on the sub-national level shows a different perspective in that not all the provinces had a negative per capita GRDP. Provinces that grew negatively in terms of their per capita GRDP during 1999 were Maluku (-27.4%), Aceh (-6.1%), Irian Jaya (-6.1%), and Central Kalimantan (-2.8%). The significant negative growth in these provinces may be contributed by the fact that these were politically troubled areas associated with many ethnic conflicts that have been embedded into the provinces for so long. Clearly this has hindered those provinces economies from growing appropriately or at least recover at the same rates as the rest of the country. Some of the country s other provinces such as Jambi, Bengkulu, Jakarta, West Java, East Java, Bali, and Southeast Sulawesi grew negatively between -0.1 percent to -1.5 percent; the remainder, although very modest recovered positively in 1999.

146 Figure 18. Growth Rate of Per Capita Gross Regional Domestic Product (GRDP) at Constant 1993 Market Price According to Province With Oil & Gas, Indonesia 1996 and 1999 (%) W estern Part Eastern Part Irian Jaya Maluku SoutheastSulawesi South Sulawesi C entralsulawesi North Sulawesi EastKalim antan South Kalim antan C entralkalim antan W estkalim antan EastNusa Tenggara W estn usa Tenggara Bali EastJava DIYogyakarta C entraljava W estjava DKIJakarta Lam pung B engkulu South Sum atera Jam bi Riau W estsum atera North Sum atera D.I.A ceh Percentage Figure 19 shows per capita GRDP with oil and gas absolute values in Rupiah. The figure illustrates that in 1996, East Kalimantan followed by Jakarta, Riau, and Irian Jaya, Aceh, and Bali possess significantly higher per capita GRDP than the rest of their peers and the country s average per capita GRDP which stood at Rp. 2.2 million for that year. These provinces per capita GRDP were recorded at Rp. 9.1 million, Rp. 7.9 million, Rp. 4.9 million, and Rp. 3.7 million, Rp. 3.1 million, and Rp. 2.4 million respectively for that year. Those provinces except Jakarta and Bali have high per capita GRDP due to their richness in natural resources, predominantly in oil and gas. Jakarta s high per capita GRDP is not surprising as it is the country s capital, and prior to the economic crisis of 1997/1998 and the government decentralization program subsequently after that, the province enjoyed much of the country s financial privileges. This particular province high per capita GRDP is also a consequent of the metropolis high living costs but not necessarily its living standards. Bali s higher than average per capita GRDP is largely contributed by the fact that it is an attraction for international tourists bringing into the country large amounts of foreign currencies that have significantly higher values than the Indonesian Rupiah. The picture for 1999 does not appear to change much than 1996 s situation.

147 Figure 19. Per Capita Gross Regional Domestic Product (GRDP) at Constant 1993 Market Price According to Province With Oil & Gas, Indonesia 1996 and 1999 (in Rupiah) W estern Part Eastern Part Irian Jaya Maluku Southeast Sulaw esi South Sulaw esi C entralsulaw esi N o rth Sulaw esi East Kalim antan South Kalim antan C entralkalim antan West Kalim antan East N usa Tenggara West N usa Tenggara Bali East Java DIY ogyakarta C entraljava West Java DKIJakarta Lam pung B engkulu South S um atera Jam bi Riau West S um atera N o rth S um atera D.I.Aceh 0 1,000,000 2,000,000 3,000,000 4,000,000 5,000,000 6,000,000 7,000,000 8,000,000 9,000,000 10,000,00 0 Rupiah Per Capita Gross Regional Domestic Product (GRDP) Without Oil & Gas Interestingly enough, the country s per capita Gross Regional Domestic Product (GRDP) growth rate without oil and gas in 1996 was 6.5 percent, which is higher than its per capita GRDP growth rate with oil and gas in the same year (6.2%). Furthermore, the country s economic recovery in 1999 without oil and gas of -0.3 percent did better than its per capita GRDP growth rate with oil and gas in the same year (-0.5%). Observation deduces that this may be the result of re-allocation of appropriate government concentration and efforts in those areas outside the country s mining and natural resources revenue generating industries. 118 This appropriation of government efforts led to the growth of other industries besides oil and gas. This course was probably taken as the Indonesia s national oil company (PERTAMINA) was one out of the many public sector companies (Badan Usaha Milik Negara BUMNs) that were targeted for privatisation as part of the country s agreement with 118 Note that during the period , OPEC underwent substantial restructuring that resulted in lower market prices for oil and gas produce.

148 the International Monetary Fund (IMF) and the World Bank, which therefore led to its significant restructurings. These restructurings further led to many projects that were already planned to be postponed, further reviewed, or simply cancelled. Observations of steep declines in foreign investments and projects that were made in previous sections support the findings above. The picture on the sub-national level shows a totally different perspective in that all provinces experienced a positive per capita GRDP growth rate during 1999 in the range of between 0.2 percent in South Kalimantan to 4.3 percent in North Sulawesi, other than these provinces: Maluku (-27.5%), Irian Jaya (-5.2%), DI Aceh (-3.9%), Central Kalimantan (-2.8%), Jakarta (-1.5%), South Sumatera (-0.6%), Bali (-0.6), Bengkulu (-0.2%), and East Java (-0.1%). The first three provinces experienced a difficult economic recovery due to the simple reality that they were and still continue to be politically troubled areas with significant ethnic and religious conflicts. This was also the case recently with Central Kalimantan. The latter three provinces kept their negative per capita GRDP growth rate as consequential effects of the many lost jobs arising out of those provinces poor business conditions. Figure 20 shows this case. Table 17 and Figure 21 shows per capita GRDP without oil and gas absolute values in Rupiah. The figure illustrates that in 1996, Jakarta, East Kalimantan, and Irian Jaya possess significantly higher per capita GRDP than the rest of their peers and the country s average per capita GRDP which stood at Rp. 1.9 million for that year. Their per capita GRDP are Rp. 7.9 million, Rp. 4.9 million, and Rp. 3.6 million respectively. This analysis although leaving out oil and gas still shows that East Kalimantan and Irian Jaya to hold the second and third highest per capita GRDP amongst all the country s other provinces. Similarly as before, it can be explained that Jakarta s high per capita GRDP is a result of it being the country s central province enjoying many business and financial privileges. Its high per capita GRDP is also a consequent of the metropolis high living costs. Again the picture for 1999 does not appear to change much than 1996 s situation.

149 Figure 20. Growth Rate of Per Capita Gross Regional Domestic Product (GRDP) at Constant 1993 Market Price According to Province Without Oil & Gas, Indonesia 1996 & 1999 (%) W estern Part Eastern Part Irian Jaya Maluku SoutheastSulawesi South Sulawesi C entralsulawesi North Sulawesi EastKalim antan South Kalim antan C entralkalim antan W estkalim antan EastNusa Tenggara W estn usa Tenggara Bali EastJava DIYogyakarta C entraljava W estjava DKIJakarta Lam pung B engkulu South Sum atera Jam bi Riau W estsum atera North Sum atera D.I.Aceh Percentage

150 Figure 21. Per Capita Gross Regional Domestic Product (GRDP) at Constant 1993 Market Price According to Provinces Without Oil & Gas, Indonesia 1996 & 1999 (in Rupiah) W estern Part Eastern Part Irian Jaya Maluku SoutheastSulawesi South Sulawesi C entralsulawesi North Sulawesi EastKalim antan South Kalim antan C entralkalim antan W estkalim antan EastNusa Tenggara W estn usa Tenggara Bali EastJava DIYogyakarta C entraljava W estjava DKIJakarta Lam pung B engkulu South Sum atera Jam bi Riau W estsum atera North Sum atera D.I.A ceh 0 1,000,000 2,000,000 3,000,000 4,000,000 5,000,000 6,000,000 7,000,000 8,000,000 9,000,000 Rupiah Poverty The levels of poverty incidence grew in most provinces in Indonesia between 1996 and 1999 (Figure 22). 119 The most devastating effects of combined urban and rural poverty incidence augmentation were seen in East Java, Central Java followed by South Sumatra, North Sumatera, East Nusa Tenggara and Lampung. In general, Indonesia s urban poverty population grew by 5.8 percent from 13.6 percent in 1996 to 19.4 percent in On the other hand, the country s rural poverty population increased by over 6.0 percent from 19.9 percent in 1996 to over 26.0 percent in Tables 18 and 19 provide detailed information of poverty incidence in the country in percentage and in numbers. In 1996, the highest urban poverty incidence occurred in West Nusa Tenggara, East Nusa Tenggara, Lampung, Bengkulu, Central Java, East Java, and Jambi, whereas the lowest were experienced by Jakarta, East Kalimantan, Central Kalimantan, and Riau. In the same year, the 119 The poverty line in Indonesia is based on a minimum food group that equals to a daily intake of 2,100 calories plus other essential non-food expenses (Islam, 2001).

151 highest rural poverty incidence took place in such provinces such as Irian Jaya, Maluku, East Nusa Tenggara, Southeast Sulawesi, West Nusa Tenggara, East Java, DI Yogyakarta, and East Kalimantan. Between 1996 and 1999, the provinces of South Sumatera (+12.0%), West Java (+10%), West Sumatera (+9.0%), Central Sulawesi (+9.0%), Central Java (+7%), Maluku (+7.0%), Aceh (+7.0%), East Kalimantan (+6.0%), North Sumatra (+6%) and North Sulawesi (+6.0%) incurred the steepest inclines of urban poverty incidences. Only in Bengkulu did the urban poverty incidence decrease between the periods 1996 to Figure 23 exhibits this case. The picture does not seem that drastic in the rural areas, although increases in poverty incidence were also recorded in most provinces in the country (Figure 24). Significant increases in rural poverty incidence between 1996 and 1999 were suffered by Irian Jaya (+16.0%), Jambi (>15%), East Kalimantan (+15.0%), Central Kalimantan (+13.0%), Yogyakarta (+13.0%), East Java (+9%), East Nusa Tenggara (+9%), South Kalimantan (+8.0%) and to a lesser extent Central Java (+6.0%). Figure 22. Urban and Rural Population Below Poverty Level According to Province, Indonesia 1996 & ,000,000 Population Below Poverty Level 10,000,000 8,000,000 6,000,000 4,000,000 2,000,000 0 D.I.Aceh North Sumatera West Sumatera Riau Jambi South Sumatera Bengkulu Lampung DKI Jakarta West Java Central Java DI Yogyakarta East Java Bali West Nusa Tenggara East Nusa Tenggara West Kalimantan Central Kalimantan South Kalimantan East Kalimantan North Sulawesi Central Sulawesi South Sulawesi Southeast Sulawesi Maluku Irian Jaya

152 Figure 23. Urban Population Below Poverty Level According to Province, Indonesia 1996 & 1999 (%) Figure 24. Rural Population Below Poverty Level According to Province, Indonesia 1996 & 1999 (%) D.I.Aceh North Sumatera West Sumatera Riau Jambi South Sumatera Bengkulu Lampung DKI Jakarta West Java Central Java DI Yogyakarta East Java Bali West Nusa Tenggara East Nusa Tenggara West Kalimantan Central Kalimantan South Kalimantan East Kalimantan North Sulawesi Central Sulawesi South Sulawesi Southeast Sulawesi Maluku Irian Jaya D.I.Aceh North Sumatera West Sumatera Riau Jambi South Sumatera Bengkulu Lampung DKI Jakarta West Java Central Java DI Yogyakarta East Java Bali West Nusa Tenggara East Nusa Tenggara West Kalimantan Central Kalimantan South Kalimantan East Kalimantan North Sulawesi Percentage Central Sulawesi South Sulawesi Southeast Sulawesi Maluku Irian Jaya Percentage

153 Summary This part of the paper in summary uncovers some of the following highlights. Population imbalance has created problems such as health, employment and poverty levels. Despite varying population growth in the regions, this trend has been more prevalent in regions that have high GRDP and are targets of both domestic and foreign investments such as Riau, Southeast Sulawesi, East Kalimantan, Central Kalimantan, Irian Jaya, and West Java. Internal migration i.e. recent migration although recorded as declining (3.4% in 1990 to 2.5% in 1995), in fact is increasing in certain parts of the country, especially from lower economically active provinces to higher economically producing ones. International migration on the other hand is expected to rise dramatically particularly to Asia Pacific and the Middle East. Declining numbers of Indonesian workers in America and Europe is attributed to among others tougher immigration laws and lower demand for less and unskilled labours. However, inadequate record and monitoring by relevant Indonesian bodies may also be a factor. The economic crisis coupled with several ethic and religious tensions have lowered both household numbers and sizes in certain provinces such as Maluku and Aceh. With regard to Maluku, this is a quite drastic change as this province in the first part of the decade had the highest growth in household numbers. The impacts of the crisis are distinctly felt in provinces such as West Java that declined its household size by nearly 18 percent. Its close proximity with Jakarta is assumed to be the cause. Life expectancies in regions such as West Nusa Tenggara and East Nusa Tenggara (Western Indonesia) stayed stagnant after the crisis, whereas those provinces in the East were still able to improve this factor although at modest rates. In fact mortality rates were downward sloping in the second part of the decade after the crisis. This trend is seen in almost every province. Despite the improvements in life expectancies and infant mortality rates, disparities between the western and eastern parts of Indonesia are strongly transparent with the latter region lagging to the national average. The crisis has reduced the country s human development index (HDI) by over 5 percent from 67.7 in 1996 to 64.3 in Again besides the crisis, various ethnic and religious conflicts have also become factors in the lessening of the country s and some of its regions HDIs. Post-crisis periods experienced a change from a centralized system of governance and finance to one that is decentralized and de-concentrated. The economic crisis and unstable political foundation have not made this transition easy. The results recently have been higher inter-regional and inter-household

154 economic inequalities leading to increased levels of income disparities between regions. After the economic crisis, the three largest economic sectors contributing to the country s national gross domestic product (GDP) remained to be the same during pre-crisis; manufacturing, agriculture, and tourism and trade. Java also remained to be the highest producing region in the country. Again the western regions of the country continue to supply the majority of the national GDP. Domestic investments remained virtually stagnant as governmental subsidies and assistances were put forth to the necessary regions. Nevertheless, the economic crisis in 1997/1998 drastically altered the number and amount of foreign investments that entered the country thereafter. Foreign investors and world business confidence dropped dramatically following the country s recession. To date, although the country s economic growth has rebounded (+4.3%) to where it was it was in 1996/1997 (+4.5%); investment projects are still postponed and are progressing very slowly. All of Indonesia s economic growth rates correlate closely to one another with the exception of Maluku and Irian Jaya that experienced drastic economic contraction after the crisis but a closer observation reveals that this region s economic decline was caused more by ethnic and religious conflicts rather than its ability to progress growth. Agriculture remains to be the largest absorber of the country s labour force. Data indicate that the crisis decreased female s labour participation rates notably in provinces of Aceh, Central Kalimantan, East Nusa Tenggara, Lampung, North Sulawesi, and Bengkulu. Although there has been increasing trends in the level of average education in the employed population, a majority still only holds primary schooling education level (<6 years) and that the majority of the labour force are employed in rural areas. Median age groups employment remained the same for both urban and rural areas as in pre-crisis periods. Open unemployment increased to over 6 million people even though not a steep increase in proportion to the growth of the country s labour force, it is observed that there are more drastic expansion in underemployment levels in informal economic sectors of the country that accounts for nearly two-thirds of Indonesia s employment population. Declining per capita GRDP growth was recorded in some provinces after the crisis however inter-unit differentials show smaller variation. It seems that ethnic and religious conflicts were observed to be the main cause of this contraction rather than economic hardships although the latter has made the recovery more difficult. The success of the Indonesian government in increasing the earnings of its inhabitants is more prevalent in western provinces and in those regions with high mining and quarrying activities such

155 as Riau and East Kalimantan. Urban centres higher GRDP remain to be tied to their higher living costs. In spite of the government s success in lowering poverty incidence from the mid 1970 s up to pre-crisis periods, the recession has seen this factor expand in most provinces. It is observed that urban and rural poverty incidence increases occurred particularly in those provinces that bounded regions originally having more diversified economic activities. For instance, East and Central Java s poverty incidence grew as more and more people migrated back into their hometowns after encountering economic hardships in Jakarta and West Java. Again, poverty remains to be a rural and agricultural problem but in detail more prevailing in those that are self-employed in agriculture and those that are wage-workers in the agriculture, trade, construction, and manufacturing sectors. Similarly as pre-crisis periods, poverty incidences remained more distinct in the country s eastern regions. 4. Recovery This part of the paper investigates regional responses during the period of crisis to recovery and how the government has attempted to overcome or at least mitigate the impacts of the crisis on both the national and sub-national levels. Regional differences are analysed and presented as they may contribute to the country s and its regions progress to recovery. The section mainly focuses on the effects of the crisis on employment, unemployment, income, and poverty levels. In doing so, the reference time period is between 1995 and Labor Force and Employment As observed in the previous chapter, the performance of the labour force activity varies by region. In some regions the LFPRs have grown, in others they have declined. Although previously labor force has been growing at a stable rate, during and after the crisis increases in unemployment levels between 1996 and 1999 has led to the decrease of employment levels in some provinces. If there was hardly any change in labor force activity this was partly due to the fact that people had to work or at least find work in order to survive, as unemployment benefits or social security systems that are common in many developed countries do not exist in Indonesia. The growth of employment nevertheless has lagged behind that of the labour force growth. Employed population by age group has changed in the period Although the national figure shows that the composition was more significant for older people in 1999 than that in 1996, regional data show variations (Tables 11.a., 11.b., 11.c., and 11.d.). The composition of employed population by education has also been slowly changing. Individuals with an education of primary school and below made up a lesser percentage of the employed in 1999 than they did in The other three education

156 categories (junior secondary school, senior secondary school and university) have had an increase, even though individuals with no more than primary school education still make up about 50 percent of the urban employed and over 75 percent of the rural employed population (Tables 12.a., 12.b., 12.c., and 12.d.). There has been a shift in the sectoral composition of employment (Tables 13.a., 13.b., and 13.c.). This composition shift may not have been voluntary. In fact Indonesia s working population has had to move away from sectors such as agriculture where individuals, particularly women work (unpaid) for the family. Agriculture, which retains the major portion of Indonesia s labour force with the exception of Jakarta, has experienced a decrease in the last decade; the share of labour force in agriculture since the 1980s continued to decrease. In 1980, the share of labour force in agriculture was 55.9 percent, while in 1990 it declined to 49.2 percent. 120 Sectoral composition of employment during seems to virtually stay constant. The employment decline at the national level is seen in such sectors as tourism and trade, and manufacturing. A slight increase yet is shown in services, which might be a reflection of the increasing informal activities that is a consequence of necessarily accommodating that segment of the labour force that needs to find easy employment entry. As women s participation in the Indonesia s labor force is significant, the question of how women's employment is likely to be affected by the current crisis needs to be addressed. In this regard, it is important to note that the rapid growth of employment in the manufacturing sector prior to the economic crisis has created a very large number of job opportunities for women. Indeed, women have played a major role in the strong expansion of labour-intensive manufacturing industries such as export production and on the wage front, although inequalities exist, Indonesia has succeeded in reducing the gap between men and women. The gap in the field of education has also been narrowed down. It is important to ensure that the progresses mentioned above are not reversed in the wake of the current crisis. Although official data do not provide a gender breakdown of figures on retrenchment of workers, informal responses to questions on this appear to maintain that there is no gender discrimination in deciding who are to be retrenched. But the fact remains that women were employed in large numbers in a few export-oriented industries, e.g., textiles, garments and shoes. Such sectors were adversely affected (at least temporarily) by the liquidity crisis hence a large number of female employees are indeed facing the prospects of retrenchment if this short-term problem is not resolved effectively and operations of these said industries do not resume quickly. Similarly, home workers who are linked to export-oriented sectors are also likely to face difficulties. The 1999 data in fact demonstrate an interesting picture that there was a 120 Hugo (1995) confirms that such change was influenced by the practices of direct seeding of rice, use of pesticides, hand tractors, and mechanical hulling. He predicted that for the first time more Indonesians were working in non-agricultural than agricultural sectors, and in the future this trend is likely to continue.

157 decline of self-employed women with family assistance. Such a decline was compensated by the increase of self-employed female without assistance, either in the forms as employer, employee or even as unpaid family workers (Tables 14.a., 14.b., and 14.c.). 121 This picture was rather different from their male counterparts. The percentage of employed males in the two employment status categories namely selfemployed with family assistance and employer show an increase, while the decline was shown in such employment status categories as self-employed without assistance, paid employee and unpaid family worker. It seems that in response to the crisis, some segment of the population have reverted in self-employment or in establishing their own business. The primary working status of employed population varies according to region. There are a very low percentage of those people working as employers (category 3). This applies to Indonesia as a whole and to every province in the country without any exceptions. However, the number of people in this category grew, in percent of the population were employers by 1999 it had grown to 2.7 percent of the employed population in Indonesia. In 1996 Irian Jaya has less than half of 1 percent (0.4 %) of its population under this category, the lowest of any province in the country. Not coincidentally, Irian Jaya also has one of the highest percentages of family workers. This is consistent with most provinces; if a province/region has a high percentage of those working as employers then it will have a low percentage of unpaid family workers and vice versa. Similar to Irian Jaya in categories as employers and unpaid family workers are Bengkulu, East Nusa Tenggara, West Nusa Tenggara, Central Sulawesi and South Sulawesi. Only Jakarta had above 2 percent of the employed population as employers in 1996, by 1999 these figures had changed dramatically for some provinces. Although Bengkulu, Irian Jaya, East Nusa Tenggara and a few other provinces remained relatively unchanged there were three provinces that stood out. Aceh s percentage of employers grew from 1.5 percent in 1996 to 5.1 percent in West Nusa Tenggara and West Kalimantan grew at a much higher rate from 0.6 percent to 14.2 percent and from 1 percent to 16.8 percent respectively. It is a good change to move away from family workers since those provinces with higher percentages of family workers tend to also have higher incidence of poverty levels (Tables 14.a., 14.b., and 14.c.). Jakarta is one of the least diversified provinces in the country as far as primary employment status is concerned. Most of its work force is concentrated as paid workers, 66.2 percent in 1996 and 63.1 percent in 1999, the highest percentage of any province. Although West Java and East Kalimantan are not far behind they do seem to spread out into the other employment status categories more than Jakarta. As previously mentioned it is worth noting the role of women in the different employment categories and their change between 1996 and 1999, which were quite 121 Primary working status is categorized into 5 primary categories: 1) Self employed without external assistance; 2) Self employed with family assistance/impermanent work; 3) Employer 4) Manual labour worker/office clerical worker/employees; and 5) Unpaid family worker.

158 significant. In 1996 only three provinces, Jakarta, Central Java, and Bali, deviated from the norm of women being employed mainly by family (22 provinces had over 30 percent of its female workforce under category # 5 unpaid family worker). Those same three provinces are the only ones that had at least 1 percent of women being employer (category # 3) none reaching 2 percent, and most provinces had under half of one percent of employers. The next major category for women in 1996 was employees (workers) manual labour/clerical work, particularly in Jakarta where 68 percent of the women either worked as manual labours or clerks. By 1999 exactly half of Indonesia s provinces had achieved at least 1 percent of women employer. Surprisingly the growth that West Nusa Tenggara and West Kalimantan had in employer category is due mainly to women s participation. In 1999, 21.4 percent of West Nusa Tenggara s workingwomen were employer and the figure is 26.3 percent for West Kalimantan. Another change in women s employment status is from being a family worker in 1996 to working in manual labour/clerical in 1999; however women continue to dominate the family worker status. If we use working status as a proxy for informal activities, nearly two-thirds of total employment is still accounted for by the informal sector. 122 Indonesia s rural elderly work more than those in urban areas or at least more people over the age of 65 are actively working in rural areas than in urban ones. This situation seems to reflect rural working environment that is more flexible and more informal in the sense that old age does not inhibit for easy entry. In urban areas 25 to 34 years old seems to be the predominant working age whereas, in rural areas the highest percentage of the total workforce is between the ages of 35 to 44 years old, this applies to both male and female. The percentages of individuals working in each age group remained relatively unchanged from 1996 to Those people with education not exceeding primary school were the most affected in regards to employment. In rural areas the percentage of the workforce with university level education increased in almost every province from 1996 to 1999, as it did for Indonesia as a whole. The bulk of rural workers (biggest percentage) still remain with those who have less than primary education. Unemployment Table 15.a. shows that open unemployment rate has increased since 1996 just before the economic crisis. If in 1996 the total open unemployment population was 4.3 million, in 1999 the figure has increased by almost two million, to 6 million people. Males showed the larger proportion of open unemployment. Is this a reflection that women became discouraged to enter the labor market after the crisis? A deeper study needs to follow. 122 Informal types of activities are based on employment status. Those working as self employed on their own, self-employed with family assistance, and unpaid family workers are used for the proxy of working in informal sector.

159 The employment situation in the region attained immediately before the crisis may be summed up as follows. Regional unemployment however demonstrates different variations. In almost every region in 1999, open unemployment has increased. Growth in open unemployment between 1996 and 1999 was led by Jakarta, which grew 89 percent during that period (100% male and 75% female) followed closely by Irian Jaya (78%), East Kalimantan (65%), and West Java (61%). Not a good trend considering West Java already has the highest number of open unemployment in the country and Jakarta is not too far behind. Yogyakarta shows the least increase in open unemployment. However some provinces demonstrated differently in its trend in open unemployment, which showed a decline from that in Open unemployment in West Nusa Tenggara, West Kalimantan, South Kalimantan, North Sulawesi, and Maluku decreased by 40.1 percent, 38.9 percent, 22.3 percent, 8.8 percent and 1.5 percent respectively. The overall decrease is mainly due to a decrease in male open unemployment with the exception of West Nusa Tenggara, which had a 67 percent decrease in female open unemployment (-4.0 % male). The declining open unemployment in Maluku perhaps may be explained by the fact that this is a conflict area and was relatively so during that period this alone would have hindered job seekers to look for employment in the province. Looking at this factor from the point of view of gender, some differences occur in the regions. In 1996 before the crisis the number of unemployed males were higher than females in exactly half of Indonesia s provinces Jambi, South Sumatera, Bengkulu, Jakarta, West Java, Central Java, Yogyakarta, East Java, Bali, East Nusa Tenggara, West Kalimantan, East Kalimantan, and Maluku. During the recovery years of 1999, some regions still held this pattern, while others are showing changes. In 1999 sixteen of the provinces had more males than females unemployed. The overall number of open unemployment has been higher for males than for females. Underemployment As is well known, in a country like Indonesia where a sizeable segment of the population is still poor and there is no unemployment benefit, people cannot afford to remain openly unemployed (i.e., fulfill the conditions of not working and looking for work). Indeed, underemployment, both in the sense of working fewer hours than is regarded as full-time employment and of working long hours with very low incomes remains to be a pervasive problem in Indonesia. Underemployment rates, defined as the proportion of the workforce working less than 35 hours per week, have tended to be much higher than unemployment rates. It is most likely that the problem has tended to be more severe in rural than in urban areas, and particularly so in agriculture. It is also acute amongst a large section of those working in the urban informal sector. The national data on underemployment was recorded at around 30 million people in Table 21 demonstrates that after the crisis the magnitude of underemployment is substantial in most provinces. Inter unit 123 Ministry of Manpower and Transmigration

160 differentials in underemployment were wider during the recovery period, as was the case for unemployment. One should not, therefore, be surprised if open unemployment does not actually grow at the alarming rates compared to those of underemployment. A more likely outcome is substantially higher rates of underemployment in sectors like agriculture, trade, transport, and other informal activities (e.g., services). Since many of these activities are located in urban areas, they (except agriculture) can perform the role of sponge without requiring people to migrate on a massive scale. In 1999 the underemployment rates have gone up. This occurred in almost every region in the country. Human Development Index and Income 124 Income levels can be correlated to the human development index (HDI), which measures life expectancy, adult literacy rates, and mean years of schooling and adjusted real per capita expenditure. HDI also demonstrates a regions progress in relation to other regions in the country. In the case of Indonesia many provinces either fell or went up in HDI ranking from 1996 to Only the highest rankings Jakarta (HDI rank # 1) and Yogyakarta (HDI rank # 2), and the lowest rankings retained their positions. East Java, West Kalimantan, East Nusa Tenggara, Irian Jaya and West Nusa Tenggara were the lowest rankings in the country at 22, 23, 24, 25 and 26 respectively for both years, 1996 and 1999 (Table 4). Bali was amongst those provinces whose HDI rank was reduced (from # 8 to # 10), likely caused by a decrease in tourism. Many countries accompanied Indonesia in a recession causing decreases in discretionary income, which led to a high decrease in income levels for places such as Bali, which is heavily dependent on both domestic and foreign travelers. Riots in other parts of Indonesia may have deterred tourists from coming to Bali or Indonesia as a whole foreign travelers are often advised not to go to troubled countries. Indonesia s adjusted real per capita expenditure decreased by just under 2 percent during the period. However, not all provinces followed the country s trend; some actually had an increase in expenditure. Riau (0.2%), Jakarta (0.3%), West Kalimantan (0.1%), Southeast Sulawesi (0.5%), and Maluku (0.6%) all had a modest increase but an increase nevertheless in real per capita expenditure (adjusted to 1993 levels) during 1996 to Irian Jaya increased its spending by 2.3 percent in the same three-year period and East Nusa Tenggara had an unsurpassed growth in expenditure of 6 percent. East Nusa Tenggara had the lowest per capita expenditure in 1996, lamentably that increase was not enough to take it up to the average in On the other side of the spectrum is Bali, which decreased 3.5 percent from 1996 when it was one of the highest spending provinces (second only to Yogyakarta). 124 The index established by the UNDP in collaboration with BAPPENAS and BPS (see Indonesia Human Development Report 2001).

161 Literacy rates and education levels (both part of the HDI equation) in the country although growing are still low, particularly education levels in rural areas. There are areas such as East Java, West Nusa Tenggara, East Nusa Tenggara, West Kalimantan, and Irian Jaya where the mean years of schooling do not supersede 6 years. These are, of course, the same provinces, which also have the lowest life expectancies and the lowest literacy rates. Low education and literacy rates have detained the poorest regions from obtaining higher levels of income, which in turn has detained them from obtaining higher levels of education. While Indonesia is recovering it can very well remain stagnant should per capita income revert back to pre-crisis levels. This will continue the vicious circle of poverty. In spite of Indonesia s decreasing real income levels it appears as though it is making an astonishing recovery given the current circumstances. During all provinces with the exception of Aceh had growth in GRDP; Indonesia grew by 4.7 percent during that period. In the coming year ( ) the roles reversed and all but one province, Irian Jaya, had decreases in GRDP, some were as high as 17.8 percent. By Indonesia appeared to be back to normal with a 4.8 percent increase in GRDP, and as expected with normality, 25 out of the 26 provinces also had positive growth. West Nusa Tenggara had an unsurpassed growth of 34.2 percent during (Table 8). Poverty As noted in the previous section that in the area of poverty reduction, the performance of Indonesia has been quite impressive. Poverty has declined dramatically with economic growth. Gains in human welfare indicators have been as impressive as economic performance, especially in education. In two decades during , the incidence of absolute poverty (i.e., the percentage of population below the 'poverty line') declined from 54.2 million to 22.5 million (Table 18). The benefits of growth appear to have been distributed fairly wide, and all provinces have increased income (per capita expenditure) and employment although they have done so at different rates and starting from different levels. The impact of the crisis on poverty at the national level is significant. There has been an increase in the percentage of the population who are considered to live below the poverty line in the period Such increasing trend in poverty incidence occurred in both urban and rural areas. Despite the increasing levels of poverty incidence in the country, Indonesia is heading towards recovery, normality, and hopefully beyond. Government Measures In response to the economic crisis particularly regarding its impact on employment, the government ran a series of measures. For those people who experienced job

162 losses the program for retrenched workers in the framework of Social Safety Net (SSN) program. As noted, the short to medium term strategy in facing the economic crisis has been to focus on how to reduce the impacts experienced by the most affected segment of the population in particular those in the category of the unemployed, the recently retrenched workers, and the poor and vulnerable. There were two categories of workers that were considered unskilled and skilled workers. The measures that were introduced were as follows. Overcoming Drought Impact and Labour Problems The programme of Overcoming Drought Impact and Labour Problems locally referred to as Program Penanggulangan Dampak Kekeringan dan Masalah Ketenagakerjaan/PDKMK, was especially aimed to help unskilled labours. This is basically a relief programme which mainly recruits the unemployed in labourintensive activities for income generation. In the first implementation stage of the programme (April 1998-February 1999) various labour-intensive works were completed. They include: Cultivation of agricultural land for rice, corn, soy bean, peanut, onion, chilli and cassava fields Building ponds for prawn, fish and crabs Rehabilitation of irrigation/channels Establishment of trade markets Road Husbandry activities Fish ponds Small industry Road rehabilitation Gardening Plantation Overcoming the Unemployed Skilled Workers The programme of Overcoming the Unemployed Skilled Workers known as Program Penanggulangan Pengangguran Pekerja Terampil (P3T) aims to assist the skilled retrenched workers by retraining and placing them for re-entry into new employment. There are two schemes under the P3T that are as follows: Productive Economic Institution (Lembaga Ekonomi Produktif/LEP) New Entrepreneur (Wira Usaha Baru/WUB) The establishments of new entrepreneurs in various forms are expected from WUB model, including: Self-employment Sub-contracting or partnership Being an agency or representative for large companies Franchise Industrial supporting business

163 Other business related to regional comparative commodities In brief, the program for unskilled workers is generally a relief programme through labour-intensive projects, and the retraining programme for skilled workers via P3T, is a type of employment guaranteed scheme. Despite the drawbacks that might have been encountered in the field, this program may be maintained with some improvements. Through the P3T programme one tries to explore local potentials in terms of natural and human resources but at the same time, one should also try to enhance employment opportunities in the area. This section demonstrates that despite the impressive progress made by the Indonesian government in various aspects of the quality of life (i.e., life expectancy, infant mortality, literacy, and education); such progress has been hampered, if not reduced by the impact of the crisis (Table 4). The crisis has brought about some changes in the Indonesian employment picture, which is not always shared by every region. Some regions might follow the national pattern some might not. With the trend in the past that showed people moving away from the agriculture sector, the recovery process demonstrate that although not very significant the agricultural sector has some promising potentialities to be explored. Development in this sector should lead towards exploring more business potentials in the area of processing agricultural products and other agricultural-based businesses.

164 Summary In summary this part of the paper put forward some of the following points. Indonesia s labour force participation rate (LFPR) has grown steadily over the last two decades. However, the growth of employment in the country has always lagged behind the growth of LFPR. During the crisis, it was observed that LFPR remained unchanged. This is explained by the fact that there was a migration of workers to more informal activities that often involved lower or even no pays and wages. This trend is attributed by the mere need for one and one s family to survive and the simple fact that there is no welfare or social security net in existence in the country. As such one cannot afford to remain openly unemployed. The crisis saw employment movement from younger age groups to older ones this is especially so in rural areas. The employed population saw an increase in the percentage of educated workers although 50 percent of the urban employed population has no more than primary school training. This number is about 75 percent in rural areas. It is observed that the uneducated segment of the employed population and labour force was most affected by the country s economic crisis. It is found that there was a sectoral shift in employment from one industry to the other. Pre-crisis periods experienced a decrease in agricultural employment from the 1980s to the 1990s; however, during and after the crisis, particularly in the period , employment in this sector surged as it contained most of the country s informal sectors. Before the crisis, there was rapid growth in manufacturing employment that provided increased opportunities for women, although wage inequalities were apparent between males and females. The government, nevertheless, was successful in reducing this gap between males and females. It has furthermore lessened the gap between males and females in the field of education. After the crisis, there was a significant increase in home business activities and selfemployment with and without pay. Consequently, as pre-crisis periods, family workers continue to be dominated by the female population. After the crisis, in 1999, the unemployment population reached 6 million people from a figure of 4.3 million people in Data shows that there is a higher proportion of male in the new unemployment figure, however this infers more towards the fact that a larger proportion of the female population was discouraged to enter the labour force during and after the crisis. As per the variance analysis in Table 20, during the gap (inter-unit variance) of this factor increased slightly from 3.8 to 5.3. But as indicated above, significant difficulties lie within the underemployment population of the work force.

165 Table 21 exhibits that underemployment continues to be a major problem in Indonesia and through variance analysis it was measured to experience considerable escalation during the period from 29 to over 97. Although the percentage of the unemployed slightly increased between 1996 and 1999, it is apparent that there was a more severe increase in the number of the underemployed population. Hand in hand with the country s poverty situation, underemployment was found to be more severe in rural areas (agricultural sector) than in urban ones. This factor has now become one of the critical attributes to the country s growing number of the poor and needy. It is established that one s human development index (HDI) is correlated closely to one s real income. Prior to the crisis from the 1970s, the government has been successful in improving Indonesia s population quality of life, and hence the country s HDI. The crisis saw decreases in the population s discretionary and disposable incomes that led to decreases in its population s ability to spend towards personal human development activities such as education and health care. It is highlighted that although most provinces experienced decreases in per capita GRDP, some provinces actually increased per capita expenditure. Tables 16 and 17 illustrate this case. The country witnessed successful reduction of the poverty population, from 40 percent in 1976 to about 11 percent in Although the crisis caused both urban and rural poverty levels to increase between 1996 and 1999, averagely from 17 percent to over 23 percent, it is more evident in the latter areas Concluding Remarks It can be clearly seen that on a national basis, Indonesia s levels in economic activity, unemployment rate, income level, and poverty incidence had a downward trend-line from pre-crisis period (pre-1995) to post-crisis period (post-1995), although they have all modestly increased positively. The country s GDP was surely observed to have a downward trend from precrisis period to post-crisis time although a positive growth rate of 4.3 percent in 1999 came close to a growth rate of 4.5 percent that was experienced by the country in Reallocation of domestic investments in Indonesia was observed during , as values of this investment remained constant during the period with increases in some and decreases in others. Nevertheless, the picture for foreign investment was not so smooth. Foreign investment in the country more than 125 Urban poverty s variance (inter-unit differentials) did indeed increase from 57.4 in 1996 to about 62.9 in 1999; while rural increases from to in the same time frame.

166 halved during the same period, reflecting the lack of foreign investors confidence in the country. It is believed that foreign investors faith in the country is more dependent upon the government s ability to restructure its political foundation than its ability o reform its economic policies. Hence, it is anticipated that foreign investment will remain shy of the country until such time where political reformation is in order. Migration patterns identified direct links between economic needs, natural forces, and political volatility. Jakarta and Riau over recent years have been a pool for population migration from neighbouring provinces, as these two provinces remain to be on top of the economic concentration list. In other cases, East Kalimantan due to its prominent mining sector may also have received an influx of population from its bordering regions, particularly South Kalimantan where ethnic tensions have surfaced in the last decade. This scenario is well illustrated by Maluku s negative growth rates in population levels, employment opportunities, and earning potentials, and a large increase in poverty incidence. Although a significant increase in open unemployment was not substantially transparent, it is expected that unemployment figures in the informal sectors such as agriculture, trade and services, have increased substantially. The informal sector represents two-thirds of the country s total employment. Another result of the economic crisis may well be that significant increases in underemployment figures can be anticipated. Research indicates that there have been a shift in sectoral composition of employment, and that employment growth is inadequate to absorb labour force growth, making jobs scarcer to find for the country s working population. The economic crisis will lead to more displacements of the employed to the unemployed. It is speculated that the female segment of the working population will suffer more as women are largely employed in industry-based sectors such as textile and garment that is temporarily halted due to a transitory liquidity crisis. This will therefore lead to more retrenchments of women workers and an increase in informal sector (e.g. home-based industries) unemployment. A shift of employment from formal sectors to informal sectors is inevitable. The country s poverty level grew during the period 1996 to More drastic trends were observed in rural areas than urban ones, although both grew in the same period. The poverty population expanded quite substantially in some regions of the country than other suggesting that a large population of people within these areas were already close to border-line poor in pre-crisis periods, and that the economic crisis leading to job losses and migration of sector employment among others, resulted in pushing this population below poverty levels. This higher incidence of poverty levels have on the other hand, decreased the life expectancy and infant mortality rates (IMRs) in the country. Previously, Indonesia was applauded for having been able to significantly reduce its poor

167 population and subsequently increase its people s life expectancy as well as lessen infant mortality rates. This was the case until the first half of the 1990s, but changed in the second half of the decade. Higher life expectancy and lower infant mortality rates follow economic expansion, i.e. improvements in these two factors are observed to exist in mainly those areas targeted by the government s economic programs, in which better access to facilities and health care services comes with the territory. Most if not all the issues discussed in this paper were difficult to diagnose in the early stages of the country s economic crisis, as the consequential effects were still unfolding. It is known that the crisis hit Indonesia when the country was already experiencing strains from a decrease in export growth rates and a lower overall national GDP rate. Furthermore, severe droughts and forest fires that ruthlessly distressed the agricultural and tourism sectors added onto the problems faced by the country. Indonesia is represents that of both good and evil of East Asian economic developments. The country s early economic success in amongst other diversifying its economic engines from those that were agriculture-based to ones that are manufacturing- and service-oriented, have all been altered due to poor management and unequal attention of political power and thus economic concentration to only a few regions in the country. Traditionally, the agricultural sector in a country like Indonesia has been able to absorb the unemployed population in such crisis. Nevertheless, the lack of link that is witnessed by second-generation migrants in urban areas to their rural areas may have changed this sector s ability to act as a sponge of the unemployed. Can agriculture continue to provide refuge for the potentially unemployed as the number of this population will definitely rise sharply? 6. Recommendations The advent of Indonesia s economic crisis has brought about all that is wrong in East Asian economic development ranging from uneven concentration of economic progression in one area to the other to the lack of social security for the country s poor and needy. Government policies and reforms for development need to take into consideration all the issues elaborated in this paper. Although such policies need to be implemented in the longer-horizon of the country s economic well being, given the consequential effects of the country s economic crisis, such programs may need to focus on the intermediate terms and problems. Inclusion of Employment Issues in Macro-Economic Policies A major key indicator in traditional economic theory is inflation. Within the area of monetary actions, it is imperative that the government reduce inflation gradually but consistently in the country. Inflation pushes the prices of goods and services in the

168 country, but in a country such as Indonesia, wages remain constant. Hence this creates a tightening of the economic system. High inflation levels curtail the ability of the Indonesian people to afford basic human needs such as food and healthcare. Ultimately, high inflation rates without supportive government subsidies will result in ballooning the poverty population in the country. In this segment of macro-policies, the Indonesian government should also attempt to decrease interest rates to suitable levels to encourage economic spending this hopefully will jump-start the country s economic engine. Nevertheless, traditional economic theories in a country like Indonesia may not work appropriately, as the country simultaneously is also required to increase its inhabitants amounts of savings (for reserve purposes) and such target can only be achieved in part by implementing high interest rates to attract population savings. Fiscal stimulus by the government is also necessary. Public spending in the country should be increased. This can also provide a jump-start to Indonesia s economy while also offering new employment opportunities for a large proportion of its labour force that is unemployed or recently retrenched. Another means of fiscal policy is increasing taxation rates in the country, particularly on the private sector. One of the country s problems in the past is the lenient tax law which is not strongly enforced. Furthermore, taxes aimed at improving the country s welfare systems and increasing the country s reserve levels, are often mismanaged and misused. Taxation should be revised and income from it at this stage should go towards improving the country s social security and welfare systems. Subsidies to and Protection of Industries Concentration of the country s competitive advantages should be undertaken. It is clear that in a country like Indonesia, and as per dictated by its history, the country has been able to successfully diversify its GDP from various industry segments that were discussed in this paper. As might have been noted that from the 1970s to the 1990s, Indonesia has been triumphant in making the manufacturing industries contribute 25.0 percent to the country s total GDP. Analyses suggest that presently this segment of the economy is under pressure and temporarily halted due to a liquidity crisis in the country. Therefore, it is recommended that subsidies be bestowed to industries such as manufacturing, agriculture, and services that also play a significant role in creating employment in the informal sectors. Education and training must be given to the working population that is active in these said industries to enable them to increase efficiency and productivity. Further strategies and programs must also be implemented in these industries whereby active workers are able to obtain subsidies on certain products or services that may enable them to increase their productiveness and effectiveness. Such program may encourage more people to expand certain industries, create employment opportunities, and ultimately contribute more towards the country s total

169 production while at the same time minimizing the problems and critical issues that arise from the country s dwindling industry productions and exports as well as its augmented unemployment problems. At the same time, while subsidies are given to these industries, government should implement policies in place to protect these industries against cheap imports. During research, it is observed that despite Indonesia s attractiveness in the past as a low-cost base for operation, this horizon has changed in the interim. Other countries in Asia and particularly Southeast Asia are becoming increasingly competitive and some have proved in recently as a lower-cost base for international companies operation than Indonesia. Furthermore, their labour force has also become increasingly efficient and productive, thus escalating their attractiveness to foreign investors and international company. Hence forth, appropriate tariffs and quotas on imported goods and products should be effectively coordinated and employed to better protect the country s fragile industries. Without these protective measures, it will be more difficult for the industries to recover and this can lead to further destruction of the country s existing and frail employment, income, and poverty situation. Expansion of New Employment Opportunities This program has been and would be continued. This revolves around developing and expanding employment opportunities in the country. Given that the private sector influence is outside the scope of this paper, it is recognized that public-sponsored economic activities may well need to be developed. The amount of public spending on such activities will unavoidably bubble to an amount that the government itself may think to be excessive, but in any case will justify the need. Expansion of employment opportunities in the country may only be justified by the subsequent demands of the consequential production. These demands should be categorized into domestic and international needs. The government will need to conjure up what constitutes necessary domestic demands and what makes international demands. Domestically, the immediate problems will without doubt relate to the country s poor and needy population in which such necessities like basic food and health care services are essential. Hence, new employment expansion should lead to the fulfilment of these needs. Although easy in theory, the reality may indeed be extremely difficult or may be impossible. However, as stated before, this real need to resolve the problems of the country s poorer population does indeed validate the reason. During the crisis it is observed that new economic activities emerged, particularly through the government program of Training for Skilled Retrenched Workers. New ideas or new employment opportunities should be encouraged to absorb job seekers.

170 Maintenance of Human Resources Development Indonesia still has a long way to go in developing its human resources. It is suggested that in order to compete in the world economy and move to the next stage of development, within 10 to 15 years, Indonesia first has to provide all children with at least 9 years of quality basic education. 126 Apart from that, the country has to develop a more flexible and efficient post-basic education system that meets the needs of a rapidly modernising economy. When the economy begins to recover, the government needs to consider that the human resources developed over the decades will be the foundation of that recovery. Therefore the foresight of preventing investments in human capital from budget cuts is very important. Thus, the long-run strategy that needs to be maintained is human resources development. The burden in looking after existing human resources stocks particularly those that are still employed should be transferred to employers. It is particularly relevant because by doing so employers may enhance their employees productivity through training. This in turn will increase companies efficiency. Apart from that, the government should attempt to find or generate other employment opportunities that may help in reducing unemployment. As trends have transpired in some provinces, with the declining employment opportunities in other economic sectors, there is an inclination that the agricultural sector becomes more important as a last resort of employment. There are still potentials that can be explored in the agricultural sector, such as the agriculture-business and agriculture-processing areas. In line with agricultural employment that somehow remains to grow modestly during the crisis, the service and trade sectors are also slightly growing. It seems that their growth is more related to the growing of informal activities in these sectors, as a last resort of people who are not able to find jobs in other formal sectors. If the assumption is true in that most of the employed population exists in informal types of employment, it is likely that they are employed in micro-enterprises or small business establishments. Therefore skills upgrading for people working in the agricultural sector and training for SMEs development should be considered. The WUB (Wira Usaha Baru/New Business Enterprise) carried out through the P3T is one option. Focus should be given to micro-enterprises, particularly the kind that provides value-added benefits. Small enterprises dominate the non-agricultural sector not only in the number of enterprises but also in number of people employed. The vocational training sector will need to be upgraded and expanded in the mediumto long-term to maintain the skills stock and at the same time to meet the skills demands most likely to be associated with globalisation. It is imperative that the skill 126 Nine year Compulsory Basic Education has been underway since several years ago.

171 acquisition system be as responsive as possible to changing skill demands, while at the same time assuring acceptable quality levels to employers and trainees. The roles of central and regional authorities need to be strengthened and capacity building needs to be carried out. The implementation of decentralization and regional autonomy through the issuance of the Act No. 22 and Act No. 25 of 1999 and its supporting regulations should be executed with strong commitment by all levels of the government. At the central level, the coordination among concerned agencies should be well in place. Policy support to the target groups needs to be coordinated and specifically targeted to the neediest. Development of Relief Fund (Welfare) for the Poor and Needy A program will need to revolve around developing and maintaining a relief fund (welfare system) for the needy and poor. Such program is imperative given the projections presented in this paper anticipating that there will be a sizable number of people categorized as being needy and poor added onto the existing 35 million in the country that are already below the poverty line in This program will to a certain extent act as a sponge for this group of the population, in which the main aim is to provide basic needs such as food, shelter, and healthcare. Subsequently, such program must also continue to work in parallel with the previous recommendation ( employment expansion ) to enable a full-scale provision and further prevention of deterioration of the country s employment and poverty situation. It must be taken into account that all three major issues of employment, income, and poverty are intermingled with one another in a complex web that is also contributed by other social factors such as education, age, family extension, and other intricately ethnic and social cultures that are evidently strong in a diverse country like Indonesia.

172 REFERENCES Adi, R. (1995) Migrasi Internasional Tenaga Kerja Indonesia: Harapan dan Kenyataan. Pusat Penelitian Unika Atmajaya. Jakarta. Central Board of Statistics (1997) Labor Force Situation, Indonesia Jakarta. Central Board of Statistics (1998) Labor Force Situation, Indonesia Jakarta. Central Board of Statistics (2000) Labor Force Situation, Indonesia Jakarta. Central Board of Statistics (2000) Open Unemployment and Underemployment in Indonesia, Jakarta. Central Board of Statistics (1996) Statistical Yearbook of Indonesia, Jakarta. Central Board of Statistics (2001) Statistical Yearbook of Indonesia, Jakarta. Dhanani, S. and Islam, I. (2000) Poverty, Inequality and Social Protection: Lessons from the Indonesian Crisis. UNDP/United Nations Support Facility for Indonesian Recovery (UNSFIR) Working Paper 00/01. Jakarta. United Nations Development Program. Hugo, G.J. (1995) International labour migration and the family: Some observations from Indonesia. Asian and Pacific Migration Journal, Vol. 4, Nos. 2 3, pp Islam, I. (2001) Identifying the Poorest of the Poor in Indonesia: Towards A Conceptual Framework. Working Paper: 01/06. United Nations Support Facility for Indonesian Recovery. Jakarta. Islam, I. (2000) Employment, Labour Markets, and Economic Recovery in Indonesia: Issues and Options. Working Paper: 00/04. United Nations Support Facility for Indonesian Recovery. Jakarta. Manning, C. (2000) Labour Market Adjustments to Indonesia Economic Crisis: Contexts, Trends, and Implications, Bulletin of Indonesian Economic Studies, 36 (1): Ministry of Manpower and Transmigration (2001) Profile of Human Resource in Indonesia, Board of Manpower Planning and Development. Jakarta.

173 Spaan, E. (1994) Taikongs and Calos: The Role of Middlemen and Brokers in Javanese International Migration. International Migration Review. Vol. 28, No. 1, pp

174 ANNEX Detailed Tables LIST OF TABLES Table 1. Administration and Population According to Province, Indonesia 1990 & Table 2. Table 3. Migrant Status Population Aged 5 and Over on Residence 5 Years Ago and Present Residence, According to Province, Indonesia 1990 and Number of People Working Overseas According to World Regions, Indonesia, Table 4. Human Development Index According to Province, Indonesia 1996 and Table 5. Distribution of Gross Regional Domestic Product (GRDP) at 1993 Constant Price According to Province and Economic Sector With Oil & Gas, Indonesia 1996 (%) Table 6. Distribution of Gross Regional Domestic Product (GRDP) at 1993 Constant Price According to Province and Economic Sector With Oil & Gas, Indonesia 2000 (%) Table 7. Table 8. Table 9. Table 10. Table 11.a. Table 11.b. Gross Regional Domestic Product (GRDP) According to Province With Oil & Gas, Indonesia 1996, 1997, 1998, 1999 & 2000 (in 000 of Rupiahs) Gross Regional Domestic Product (GRDP) Growth Rates Based on 1993 Constant Prices According to Province, Indonesia 1996/97 to 1999/2000 Approved Domestic and Foreign Projects and Investments According to Province, Indonesia 1994 & 1999 (in 000 of Rupiahs) Labour Force Participation Rate According to Gender and Province, Indonesia 1996 & 1999 (%) Urban Male Employed Population According to Province and Age Group, Indonesia 1996 and 1999 Urban Female Employed Population According to Province and Age Group, Indonesia 1996 and 1999

175 Table 11.c. Table 11.d. Table 12.a. Table 12.b. Table 12.c. Table 12.d. Table 13.a. Table 13.b. Table 13.c. Table 14.a. Table 14.b. Table 14.a. Table 15.a. Rural Male Employed Population According to Province and Age Group, Indonesia 1996 and 1999 Rural Female Employed Population According to Province and Age Group, Indonesia 1996 and 1999 Urban Male Employed Population According to Province and Education, Indonesia 1996 and 1999 Urban Female Employed Population According to Province and Education, Indonesia 1996 and 1999 Rural Male Employed Population According to Province and Education, Indonesia 1996 and 1999 Rural Female Employed Population According to Province and Education, Indonesia 1996 and 1999 Employed Population According to Province and Economic Sector, Indonesia 1994 Employed Population According to Province and Economic Sector, Indonesia 1996 Employed Population According to Province and Economic Sector, Indonesia Employed Male Population According to Province and Status, Indonesia 1996 and 1999 Employed Female Population According to Province and Status, Indonesia 1996 and 1999 Employed Male and Female Population According to Province and Status, Indonesia 1996 and 1999 Open Unemployment According to Province and Gender, Indonesia 1996 and 1999 Table 15.b. Open Unemployment Rates, Indonesia Table 16. Table 17. Per Capita Gross Regional Domestic Product (GRDP) and Growth Rate at 1993 Constant Price According to Provinces With Oil & Gas, Indonesia 1996 & 1999 (in Rupiah) Per Capita Gross Regional Domestic Product (GRDP) and Growth Rate at 1993 Constant Price According to Provinces Without Oil & Gas, Indonesia 1996 & 1999 (in Rupiah)

176 Table 18. Table 19. Table 20. Number of Population Below Poverty Line According to Residence, Indonesia, and (Million) Percentage of Population Below the Poverty Line by Residence and Province, Indonesia, 1996 and Variance of Open Unemployment According to Province, Indonesia 1996 and 1999 (%) Table 21. Variance of Underemployment According to Province, Indonesia 1996 and 1999 (%)

177 Table 1. Administration and Population According to Province, Indonesia, 1990 and 2000

178 Table 2. Migrant Status Population Aged 5 and Over Based on Residence 5 Years Ago and Present Residence, According to Province, Indonesia 1990

179 Table 3. Number of People Working Overseas According to World Regions, Indonesia, 1995, 1996, 1997, 1998, 1999, and 2000

180 Table 4. Human Development Index According to Province, 1996 & 1999 Province L.E Adult Mean YearAdj.Real Per HDI HDI HDI Red. (years) Litrc. schooling Cap.Exp(Rp) Rank Fall D.I.Aceh North Sumatera West Sumatera Riau Jambi South Sumatera Bengkulu Lampung DKI Jakarta West Java Central Java DI Yogyakarta East Java Bali West Nusa Tenggara East Nusa Tenggara West Kalimantan Central Kalimantan South Kalimantan East Kalimantan North Sulawesi Central Sulawesi South Sulawesi Southeast Sulawesi Maluku Irian Jaya INDONESIA

181 Table 5. Distribution of Gross Regional Domestic Product (GRDP) at 1993 Constant Price According to Province and Economic Sector With Oil & Gas, Indonesia 1996 (%)

182 Table 6. Distribution of Gross Regional Domestic Product (GRDP) at 1993 Constant Price According to Province and Economic Sector Without Oil & Gas, Indonesia 2000 (%)

183 Table 7. Gross Regional Domestic Product (GRDP) According to Province With Oil & Gas, Indonesia 1996, 1997, 1998, 1999 and 2000 (in 000 of Rupiahs)

184 Table 8. Gross Regional Domestic Product (GRDP) Growth Rates Based on 1993 Constant Prices According to Province, Indonesia, 1996/1997 to 1999/2000

185 Table 9. Approved Domestic and Foreign Projects and Investments According to Province With Oil & Gas, Indonesia 1994 and 1999 (in 000 of Rupiahs)

186 Table 10. Labour Force Participation Rate According to Gender and Province, Indonesia 1996 and 1999 (%)

187 Table 11.a. Urban Male Employed Population According to Province and Age Group, Indonesia, 1996 and 1999

188 Table 11.b. Urban Female Employed Population According to Province and Age Group, Indonesia, 1996 and 1999

189 Table 11.c. Rural Male Employed Population According to Province and Age Group, Indonesia, 1996 and 1999

190 Table 11.d. Rural Female Employed Population According to Province and Age Group, Indonesia, 1996 and 1999

191 Table 12.a. Urban Male Employed Population According to Province and Education, Indonesia, 1996 and 1999

192 Table 12.b. Urban Female Employed Population According to Province and Education, Indonesia, 1996 and 1999

193 Table 12.c. Rural Male Employed Population According to Province and Education, Indonesia, 1996 and 1999

194 Table 12.d. Rural Female Employed Population According to Province and Education, Indonesia, 1996 and 1999

195 Table 13.a. Population 10 Years of Age and Over Who Worked During the Previous Week According to Province and Economic Sector, Indonesia 1994

196 Table 13.b. Population 10 Years of Age and Over Who Worked During the Previous Week According to Province and Economic Sector, Indonesia 1996

197 Table 13.c. Population 10 Years of Age and Over Who Worked During the Previous Week According to Province and Economic Sector, Indonesia 1999

198 Table 14.a. Employed Male Population According to Province and Status, Indonesia, 1996 and 1999

199 Table 14.b. Employed Female Population According to Province and Status, Indonesia, 1996 and 1999

200 Table 14.c. Employed Male and Female Population According to Province and Status, Indonesia, 1996 and 1999

201 Table 15.a. Open Unemployment According to Province and Gender, Indonesia, 1996 and 1999

202 Table 15.b. Rate of Open Unemployment, Indonesia, 1996, 1997, 1998, 1999, and 2000

203 Table 16. Per Capita Gross Regional Domestic Product (GRDP) and Growth Rate at 1993 Constant Price According to Province With Oil & Gas, Indonesia 1996 and 1999 (in Rupiah) No

204 Table 17. Per Capita Gross Regional Domestic Product (GRDP) and Growth Rate at 1993 Constant Price According to Province Without Oil & Gas, Indonesia 1996 and 1999 (in Rupiah) Notes: Inter unit variance analysis on per capita GRDP with oil and gas declined from Rp 4.3 to Rp 3.7 trillion while without oil and gas declined from Rp 2.2 to 1.8 trillion.