THE INTERNATIONAL EXPERIENCE OF SERVICE SECTOR DEVELOPMENT, AND SOME KEY LESSONS FOR KAZAKHSTAN AND DEVELOPING ASIA

JOINT GOVERNMENT OF KAZAKHSTAN and ASIAN DEVELOPMENT BANK KNOWLEDGE AND EXPERIENCE EXCHANGE PROGRAM THE INTERNATIONAL EXPERIENCE OF SERVICE SECTOR DEVELOPMENT, AND SOME KEY LESSONS FOR KAZAKHSTAN AND DEVELOPING ASIA 21 September 2014 This report is one of the reports under the umbrella study Services Sector in Kazakhstan as an Engine for Diversified Economic Growth. The study is undertaken under TA-8414 KAZ: Joint Government of Kazakhstan and the Asian Development Bank Knowledge and Experience Exchange Program (KEEP). The report is prepared by Donghyun Park and Gemma Estrada, Economics and Research Department. The overall study s task manager is Lyaziza Sabyrova, Principal Regional Economist, ADB. 1

1 Introduction The purpose of this paper is to extrapolate policy lessons from the international service sector development experience as they pertain to Kazakhstan. The Central Asian country is currently seeking to diversify its economy away from a disproportionate dependence on oil, gas, and other natural resources. The recent development of an export-oriented beef industry in the country s vast steppes is just one example of Kazakhstan s concerted efforts to balance its economy. Though natural wealth, currently Kazakhstan s comparative advantage, has made the country easily the richest in Central Asia, with relatively sound policies and an open, outward-looking regime also contributing to strong economic performance, prosperity rooted in finite resources is unsustainable. Kazakhstan s medium- to long-term economic challenge is to build a robust, resilient foundation based on a diverse mix of self-sustaining firms and industries. However, diversification for its own sake achieves little, and leaves a costly and wasteful legacy of white elephants. Of course, the government can and should use the natural resource bonanza to enable a dynamic private sector to emerge and thrive, an especially important government role in transition economies like Kazakhstan s. For example, public investment in infrastructure such as roads, airports, and telecommunication networks raises the productivity of the economy as a whole, rather than a few firms or industries. Another important example is a sound and effective regulatory framework that encourages open and fair competition. Kazakhstan is more fortunate than most other transition economies that lack natural wealth and hence fiscal 2

resources since, in principle, the government can invest those resources in infrastructure. The extent to which the government should oversee the allocation of scarce resources in countries with underdeveloped private economies is a subject of heated debate. It is uncertain whether the highly activist industrial policy associated with Japan and the Republic of Korea is suitable for Kazakhstan. What is clearer is that, regardless of the intensity and direction of industrial policy, there is plenty that Kazakhstan can do to foster a more vibrant, non-resource-based economy. It is equally true that the sustainable dynamism familiar to advanced economies requires productive manufacturing as well as productive services. While manufacturing and industrialization receive most of the attention among policymakers in developing countries, the service sector is no less important to economic development. This paper s central objective is to survey Kazakhstan s current services industry in terms of its strengths and weaknesses, and set forth policy recommendations accordingly. Services in Kazakhstan suffer from low productivity, but it is hardly alone in this regard. Rather, an underdeveloped, inefficient service sector is a common problem in the rest of developing Asia. For example, the PRC has become a manufacturing powerhouse the factory of the world but it has a relatively inefficient service sector. Even success stories such as the information and communication technology business process outsourcing (ICT BPO) exports of India and the Philippines are isolated enclaves with limited linkages to the rest of their economies rather than examples of a broadly and deeply dynamic service sector. Services play a smaller role in developing Asia relative not only to advanced economies, but also to other parts of 3

the developing world (Figure 1). Therefore, it is best to assess Kazakhstan s services development within the broader context of developing Asia. Figure 1: Sector Shares by Country Groups, 2012 (in % of GDP) 80 Agriculture Industry Services 75.1 60 61.3 63.3 % 50.4 40 39.3 29.9 31.5 23.3 20 0 10.3 8.8 Developing Asia Developing Europe Latin America and the Caribbean 5.2 1.6 OECD OECD = Organisation for Economic Co-ordination and Development. Source: ADB estimates using data from the World Bank, World Development Indicators online database (accessed 22 July 2014). The rest of this paper is organized as follows: Section 2 reviews service sector heterogeneity and measurement problems. Section 3 examines the current state of services in Kazakhstan and developing Asia, while Section 4 extends this to the future through the prism of international structural transformation patterns. Section 5 analyzes the most fundamental problem facing services in Kazakhstan and Asia: low productivity. Section 6 delves into modern service sector underdevelopment, and Section 7 sets forth policy options. Section 8 concludes the paper with final observations. 2 Service Sector Heterogeneity and Measurement Problems 4

Compared to agriculture, mining, and most of all, manufacturing, services has long occupied a diminished place in both the public imagination and economic research. One reason is its sheer diversity, encompassing an enormous range of industries and activities and discouraging simple mental imagery or easy encapsulation (Table 1). 1 In the case of Asia, services intrinsic heterogeneity is compounded by the enormous differences among countries development levels. Not surprisingly, this heterogeneity has far-reaching sector implications for policies that necessarily must be country- and industry-specific. Eichengreen and Gupta (2009) argued that the broad aggregation of services obscures two distinct growth waves : the first occurs early in the development process at relatively low income levels in traditional sectors, such as personal services, while the second occurs later at higher incomes in sectors that are heavily reliant on information technology and possess greater scope for cross-border tradability, such as communication, computer, technical, and business services. For some purposes, it may be useful to focus on a set of service activities, such as business services, where the prospects for high-wage employment and cross-border trade appear relatively high and political sensitivities may be less acute than what are encountered with, for example, education or health. These possibilities may be considerable: Jensen (2011) pointed out that in the United States (US) in 1960, business services employed less than half as many workers compared to manufacturing, but by 2007, business services employment was more than double that of manufacturing. 1 Furthermore, the definition of services is not always clear cut. For example, potable water, electricity, and other public utilities are defined as part of industry, rather than services. In many Asian countries, a critical issue in economic development is the lack of access to public utilities. [is this sentence unfinished?] 5

Table 1: United Nations International Standard Industrial Classification, Divisions 45 99 G - Wholesale and retail trade; repair of motor vehicles and motorcycles 45 - Wholesale and retail trade and repair of motor vehicles and motorcycles 46 - Wholesale trade, except of motor vehicles and motorcycles 47 - Retail trade, except of motor vehicles and motorcycles H - Transportation and storage 49 - Land transport and transport via pipelines 50 - Water transport 51 - Air transport 52 - Warehousing and support activities for transportation 53 - Postal and courier activities I - Accommodation and food service activities 55 - Accommodation 56 - Food and beverage service activities J - Information and communication 58 - Publishing activities 59 - Motion picture, video and television program production, sound recording and music publishing activities 60 - Programming and broadcasting activities 61 - Telecommunications 62 - Computer programming, consultancy and related activities 63 - Information service activities K - Financial and insurance activities 64 - Financial service activities, except insurance and pension funding 65 - Insurance, reinsurance and pension funding, except compulsory social security 66 - Activities auxiliary to financial service and insurance activities L - Real estate activities 68 - Real estate activities M - Professional, scientific and technical activities 69 - Legal and accounting activities 70 - Activities of head offices; management consultancy activities 71 - Architectural and engineering activities; technical testing and analysis 72 - Scientific research and development 73 - Advertising and market research 74 - Other professional, scientific and technical activities 75 - Veterinary activities N - Administrative and support service activities 77 - Rental and leasing activities 78 - Employment activities 79 - Travel agency, tour operator, reservation service and related activities 80 - Security and investigation activities 81 - Services to buildings and landscape activities 82 - Office administrative, office support and other business support activities O - Public administration and defense; compulsory social security 84 - Public administration and defense; compulsory social security P - Education 85 - Education 6

Q - Human health and social work activities 86 - Human health activities 87 - Residential care activities 88 - Social work activities without accommodation R - Arts, entertainment and recreation 90 - Creative, arts and entertainment activities 91 - Libraries, archives, museums and other cultural activities 92 - Gambling and betting activities 93 - Sports activities and amusement and recreation activities S - Other service activities 94 - Activities of membership organizations 95 - Repair of computers and personal and household goods 96 - Other personal service activities T - Activities of households as employers; undifferentiated goods- and services-producing activities of households for own use 97 - Activities of households as employers of domestic personnel 98 - Undifferentiated goods- and services-producing activities of private households for own use U - Activities of extraterritorial organizations and bodies 99 - Activities of extraterritorial organizations and bodies Source: United Nations Statistics Division. International Standard Industrial Classification of All Economic Activities, Revision 4. http://unstats.un.org/unsd/cr/registry/regcst.asp?cl=27 (accessed 10 September 2012). The analytical challenges created by this diversity are compounded by the output of many sectors being hard to measure (public education, for example). In many countries, the service sector is insulated from competition, and subject to administered or otherwise highly regulated prices (again, think public education). If neither outputs nor quantities are measurable, assessing productivity becomes difficult. Needless to say, these conditions contrast starkly with those prevailing in agriculture, mining, and manufacturing, where output is standardized and enormous attention has been devoted to understanding productivity determinants. These analytical challenges are even further compounded at the level of the firm, with many of today s major multinational corporations that originated in manufacturing, such as General Motors or General Electric, having large services divisions. Indeed, part of the apparent activity intensification may reflect the changing nature of the firm, 7

specifically outsourcing and offshoring, with the latter also affecting productivity measurement in service-using sectors, such as manufacturing (Yuskavage, Strassner, and Medeiros 2008; Houseman et al. 2011). These cross-sector connections are key: after surveying numerous studies, Francois and Hoekman (2010) concluded that performance in services may be a major growth factor, and that both domestic and cross-border policy manifestations may be a key driver in economic development. 3 Current State of the Service Sector in Kazakhstan and Developing Asia Historically, economies have followed similar structural transformation patterns. Early in development, agriculture s dominance vis-à-vis production and employment gives way to industry, particularly manufacturing. As industrialization proceeds, manufacturing labor productivity improves, which increases incomes and demand for services. Service sector employment and output thus rises as the economy becomes postindustrial. The expansion of the middle class fuels the demand for services, e.g., finance, health care, education, leisure, and entertainment, and thus further augments growth. This pattern has characterized the advanced economies structural transformation; developing Asia is following their lead. The service sector share in GDP and employment grows as a country develops, and tends to be higher in richer countries. For example, in 2012, services GDP share in the PRC, India, Indonesia, and the Republic of Korea were 44.6%, 56.3%, 38.7%, and 59.5%, respectively, whereas the corresponding figures for Germany, Japan, the United Kingdom (UK), and the US were 68.7%, 73.1%, 78.7%, and 78.6%, respectively. In Kazakhstan, services accounted for 55.8% of GDP in 2012 (Figure 2). For services employment in 2011, the figures for the PRC, India, Indonesia, and the Republic of 8

Korea were 35.7%, 27.3%, 40.8%, and 78.0%, whereas the corresponding figures for Germany, Japan, the UK, and the US were 70.1%, 70.6%, 79.6%, and 81.6%, respectively. In Kazakhstan, services employed 54.5% of all workers (Figure 3). The service sector plays a larger role in advanced economies than in developing Asia, which suggests that the latter can learn from the experiences of the former as it develops its own service industries. The same is true for Kazakhstan, which closely follows developing Asia s pattern. Figure 2 Services Share in GDP, 2012 (in %) Indonesia PRC 38.7 44.6 Kazakhstan India Republic of Korea 55.8 56.3 59.5 Germany Japan United States United Kingdom 68.7 73.1 78.6 78.7 0 20 40 60 80 100 % PRC= People s Republic of China. Source: World Bank, World Development Indicators (accessed 30 July 2014). 9

Figure 3 Services Share in Employment, 2011 PRC= People s Republic of China. Source: World Bank, World Development Indicators (accessed 30 July 2014). Figures 4 and 5 show services share in GDP and employment as it evolves for a broader range of developing Asian countries. With only a few exceptions, the whole region fits the general pattern of services being underdeveloped. Though Asian countries trail advanced economies in terms of both output and employment, services share across the entire region is growing. Furthermore, although Asia s service sector lags behind the advanced economies, it is a major source of output and employment. Asia thus conforms to the historical pattern of agriculture declining and giving way to industry and services. In sum, the sector is already participating in Asian economies; in Kazakhstan, services GDP share rose from 28.7% in 1992 to 55.8% in 2012, although employment share remained fairly stable. Figure 4 10

Services Share in GDP (in %) 1980 1990 2012 Central Asia East Asia South Asia Southeast Asia The Pacific Highincome Armenia Azerbaijan Georgia Kazakhstan Kyrgyz Republic Tajikistan Turkmenistan Uzbekistan PRC Hong Kong, China Republic of Korea Mongolia Taipei,China Bangladesh Bhutan India Maldives Nepal Pakistan Sri Lanka Brunei Darussalam Cambodia Indonesia Lao PDR Malaysia Myanmar Philippines Singapore Thailand Viet Nam Fiji Kiribati FSM Palau Papua New Guinea Samoa Solomon Islands Timor-Leste Tonga Germany Japan United Kingdom United States 0 20 40 % 60 80 100 FSM =Federated States of Micronesia. PDR = people s democratic republic. PRC = People s Republic of China. 11

Sources: Asian Development Bank (2012); Asian Development Outlook database; CEIC Data Company; World Bank. World Development Indicators online database (databases accessed 22 July 2014); ADB estimates. Developing Asia Hong Kong, China Singapore Republic of Korea Maldives Malaysia Taipei,China Kazakhstan Philippines Mongolia Azerbaijan Kyrgyz Republic Armenia Sri Lanka Indonesia Thailand Bangladesh Georgia PRC Pakistan Bhutan Viet Nam India Cambodia Figure 5 Services Share in Employment (in %) 1990 2011 High-income Germany Japan United Kingdom United States 0 30 60 90 % PRC = People s Republic of China. Sources: Asian Development Bank (2012); Asian Development Outlook database; CEIC Data Company; World Bank. World Development Indicators online database (databases accessed 22 July 2014); ADB estimates. 12

In 2000 2012, services accounted for more than one-half of GDP growth in most economies in the region (Figure 6), particularly in India and Sri Lanka, where over 60 percent of the growth in 2000 2012 was due to services. However, in East Asia, particularly the PRC and Taipei,China, the story is still industry driving overall growth. As noted in ADB (2007), the sector has played an important role in countries where the pace of industrialization has been slow, such as in South Asia and the Philippines. Furthermore, for South Asia, modern services drove overall growth (see Bosworth and Maertens 2010, and Ghani 2010). In Kazakhstan, which grew by a robust 7.98% during this period, the contributions of agriculture, industry, and services to growth were 0.27%, 3.14%, and 4.57%, respectively. This implies that services accounted for over half of growth. Figure 6 Contributions to Annual GDP Growth, 2000 2012 (in %) 13

2000 2012 Azerbaijan PRC Turkmenistan Bhutan Tajikistan Kazakhstan Mongolia Cambodia Armenia Lao PDR India Uzbekistan Viet Nam Georgia Bangladesh Singapore Sri Lanka Indonesia Philippines Malaysia Thailand Republic of Korea Pakistan Nepal Kyrgyz Republic Solomon Islands Taipei,China Samoa Brunei Darussalam Tonga Fiji FSM Agriculture Industry Services -3 0 3 6 9 12 15 Percentage points FSM =Federated States of Micronesia. PDR = people s democratic republic. PRC = People s Republic of China. Sources: ADB estimates based on data from the World Bank, World Development Indicators online database (databases accessed 22 July 2014). 14

4 Structural Transformation and the Future of Service Sector Development in Kazakhstan and Developing Asia The data show an improved service sector in developing Asia, but will this continue? A well-known pattern of economic development and growth has the share of services in output and employment rising as a country grows richer. This rise of services is actually the tail-end of evolution from agriculture to industry and on to postindustrial pursuits, which has been repeated so often and widely as to appear inevitable. Typically, as industrialization takes hold, along with associated services, agricultural employment slackens. Low labor productivity in agriculture early on means that its workers can shift to manufacturing without substantially affecting output. These laborers migrate from rural areas to cities and find work in factories and shops, boosting industry as a share of total economic activity. In the second phase, agriculture continues its decline, but industrial expansion also slows. Employment in services continues to rise, however, as the economy moves into its postindustrial phase, with higher overall incomes stimulating their demand. Global data show that services share in GDP (Figure 7) and employment (Figure 8) both correlate positively with per capita income. That is, service sector output and employment tend to be more important to the economies of richer countries. Figure 7 Percent Share of Services in GDP against Log GDP per capita, 2012 (in $) 15

10 0 Developing Asia Rest of the world HKG PAL Share of services in GDP (%) 80 60 40 MLD GEO FIJ SAM IND SRI TON KGZ PH I KAZ AFG PAK BAN MON N EP ARM MAL TAJ UZB THA PRC VIE CAM INO BHU TKM LAO AZE KOR SIN BRU 20 4 6 8 10 12 Log GDP per capita (Current $) ARM = Armenia, AZE = Azerbaijan, BAN = Bangladesh, BHU = Bhutan, BRU = Brunei Darussalam, CAM = Cambodia, PRC = People s Republic of China, FIJ = Fiji, GEO = Georgia, HKG = Hong Kong, China, IND = India, INO = Indonesia, KAZ = Kazakhstan, KIR = Kiribati, KOR = Republic of Korea, KGZ = Kyrgyz Republic, LAO = Lao People s Democratic Republic, MAL = Malaysia, MLD = Maldives, MON = Mongolia, NEP = Nepal, PAK = Pakistan, PNG = Papua New Guinea, PHI = Philippines, SAM = Samoa, SIN = Singapore, SOL = Solomon Islands, SRI = Sri Lanka, TAJ = Tajikistan, TAP = Taipei,China, THA = Thailand, TON = Tonga, UZB = Uzbekistan, VAN = Vanuatu, VIE = Viet Nam. Source: World Bank. World Development Indicators online database (accessed 22 July 2014). Figure 8 Percent Share of Services in Employment against Log GDP per capita, 2011 (in $) 16

10 0 Developing Asia Rest of the world Share of services in employment (%) 80 60 40 20 PH I MON KGZ ARM SRI THA AZE INO PRC PAK BHU VIE CAM IND MAL KAZ KOR HKG 6 8 10 12 Log GDP per capita (current $) SIN ARM = Armenia, AZE = Azerbaijan, BAN = Bangladesh, BHU = Bhutan, BRU = Brunei Darussalam, CAM = Cambodia, PRC = People s Republic of China, FIJ = Fiji, GEO = Georgia, HKG = Hong Kong, China, IND = India, INO = Indonesia, KAZ = Kazakhstan, KIR = Kiribati, KOR = Republic of Korea, KGZ = Kyrgyz Republic, LAO = Lao People s Democratic Republic, MAL = Malaysia, MLD = Maldives, MON = Mongolia, NEP = Nepal, PAK = Pakistan, PNG = Papua New Guinea, PHI = Philippines, SAM = Samoa, SIN = Singapore, SOL = Solomon Islands, SRI = Sri Lanka, TAJ = Tajikistan, TAP = Taipei,China, THA = Thailand, TON = Tonga, UZB = Uzbekistan, VAN = Vanuatu, VIE = Viet Nam. Source: World Bank. World Development Indicators online database (accessed 22 July 2014). The rise in services may not be linear. Eichengreen and Gupta (2009) argued that there are two distinct waves of services growth. The first wave sees output rise at a decelerating rate until it levels out at about $1,800 per capita (2000 purchasing power parity dollars). Output then accelerates in the second wave to reach about $4,000 per capita before leveling off again. The first wave is the rise of traditional services such as lodging, meal preparation, housecleaning, and grooming services. The second wave is dominated by modern services like banking, insurance, computing, communication, and business services. This two wave pattern is robust using more recent data (Box 1). 17

Box 1 Two waves of growth in services According to Eichengreen and Gupta (2009), the two waves of services growth can be characterized mathematically by a quartic relationship between income and the output share of services, defined as where, and are the service sector value added, GDP, and log per capita GDP, respectively, for country i at time t. is a period dummy, which is included to allow different intercepts for different time periods. But how robust is this relationship? Replicating the 2009 Eichengreen and Gupta estimations using a larger sample size (157 countries as compared with 80 countries) and more recent data (1960 2010 as compared with 1950 2005), the basic two wave pattern emerges again (box table). All GDP per capita terms from the first to fourth order are highly significant for both GDP and share of employment, confirming the quartic relationship between income and services development. Quartic Relationship between Log GDP per capita and Services GDP and Employment Shares Variable Services Share in GDP Services Share in Employment Log income per capita 414.668*** 1,013.291*** [5.472] [4.173] Log income per capita, squared 72.132*** 177.987*** [ 5.050] [ 4.210] Log income per capita, cubed 5.453*** 13.694*** [4.623] [4.220] Log income per capita, quartic 0.149*** 0.386*** [ 4.132] [ 4.169] Dummy for 1970 1989 1.069*** [2.927] Dummy for 1990 2010 4.929*** 4.345*** [12.604] [13.117] Observations 5,402 2,222 Number of countries 157 139 R-squared 0.249 0.439 18

Note: T statistics are in square brackets. *** indicates that the coefficient is significant at 1%. The estimation allows the intercepts to differ in the periods 1970 1989 and 1990 2010 for column 1, and in 1990 2010 for column 2. Country fixed effects are included in both estimates. Source: Park and Shin, forthcoming-a. ***** End Box ***** These global trends apply broadly to developing Asia (Noland, Park, and Estrada 2013). Richer Asian countries tend to have larger service sectors, as do better-educated countries in the region. In all Asian countries, the sector expands as per capita income grows. Results from cross-country analyses hold when scrutinizing individual developing Asian economies. In 12 major Asian economies, the evolution of the share of agriculture, industry, and services in GDP and employment largely mirrors international historical experience (Park and Shin 2013a). Based on these historical trends, the already-large service sector will likely expand significantly as Asian economies continue to grow. Within the regional expansion exists a great deal of heterogeneity among countries regarding the relative importance of services, as emphasized by Ghani (2010). To some extent, such heterogeneity reflects Asia s wide income and development range (as explained above, services share in GDP and employment tends to rise with per capita income). However, income and development can explain only part of this heterogeneity. For example, services share in India s GDP was 54.7% in 2010, while the corresponding figure for the PRC was only 43.1% despite a substantially higher per capita income. Further, there is also a great deal of heterogeneity with respect to how quickly services shares in GDP and employment grow. For example, in 1980 services share in employment was similar in Indonesia and the Philippines, but by 19

2010 it was notably higher in the Philippines. Such differences notwithstanding, the sector is expanding in both absolute and relative terms throughout the region. Urbanization is a powerful driver of structural change behind services inevitable expansion in Asia, which added more than 1 billion people to its cities from 1980 to 2010 and will add another 1 billion by 2040 (ADB 2012). Rising urbanization and the resulting population density increase spur demand for an array of services. With its diverse offerings, the sector thrives on the large range of urban business activities. The location advantage of urban areas offers proximity to firms and suppliers alike. For services such as retailing, education, health care, and other community and personal services, face-to-face interaction with clients is important, giving urban locations a further advantage (Kolko 2010). Large concentrations of people and firms in urban centers are ideal incubators for services. The evidence in Asia shows that services in more urbanized economies make up a greater share of employment (Figure 9). The rapid urbanization of Asia is therefore another reason to expect that services will become more important as a source of future growth and jobs. 20

Figure 9 Urbanization Rate and Services Share in Employment in Developing Asia, 2011 (in %) 10 0 Developing Asia Rest of the world Share of services in employment (%) 80 60 40 20 SRI CAM KGZ THA IND BHUPAK VIE KAZ PH I AZE INO PRC ARM MON MAL KOR HKG 0 20 40 60 80 10 0 Urbanization rate, % SIN ARM = Armenia, AZE = Azerbaijan, BAN = Bangladesh, BHU = Bhutan, CAM = Cambodia, PRC = People s Republic of China, GEO = Georgia, HKG = Hong Kong, China, IND = India, INO = Indonesia, KAZ = Kazakhstan, KOR = Republic of Korea, KGZ = Kyrgyz Republic, MAL = Malaysia, MLD = Maldives, MON = Mongolia, PAK = Pakistan, PHI = Philippines, SIN = Singapore, SRI = Sri Lanka, TAP = Taipei,China, THA = Thailand, VIE = Vietnam. Sources: CEIC Data Company; World Bank. World Development Indicators online database (both accessed 22 July 2014). Overall, Asia s service sector has been large and growing, and trends point to services becoming an even bigger growth force. The expansion of the middle class and their purchasing power should also boost services share in the economy. In addition, some countries in East and Southeast Asia are becoming postindustrial vis-à-vis growth and development, with the concomitant increase in the share of services. 5 The Fundamental Service Sector Problem in Kazakhstan and Developing Asia: Low Productivity 21

Services contribution to GDP, employment, and GDP growth in developing Asia is significant, and is set to become so even more as the region continues to augment incomes. However, if the sector is to become a dynamic engine of regional growth, it must grapple with a fundamental structural problem: very low productivity. Developing Asia s labor productivity in services, defined as service output per worker, lags behind that of Organisation for Economic Co-operation and Development (OECD) countries by a wide margin (Figure 10). While Singapore; Hong Kong, China; and Taipei,China are fast approaching the OECD average, the rest of the region languishes well below. Perhaps most strikingly, the Republic of Korea, a highly successful industrializer like the PRC, has failed to achieve services labor productivity of even half that of the OECD (Box 2). Other countries lag even further behind. Among the larger economies, only Malaysia is near 20%, and the others hover at 10% or less. While there is considerable diversity, labor productivity in developing Asia as a group clearly has a lot of catching up to do with the advanced economies. 22

Figure 10 Services Labor Productivity (in $ 000) 1990 2000 2011 OECD Singapore Hong Kong, China Taipei,China Republic of Korea Malaysia Kazakhstan Georgia India PRC Sri Lanka Thailand Armenia Philippines Indonesia Pakistan Azerbaijan Mongolia Viet Nam Cambodia Bangladesh Kyrgyz Republic 0 20 40 60 80 2005 constant $ ('000) OECD = Organisation for Economic Co-operation and Development, PRC = People s Republic of China. Sources: ADB estimates using data from CEIC Data Company; International Labour Organization. Key Indicators of the Labor Market; World Bank. World Development Indicators online database (all accessed 16 April 2012). Box 2 Stark Lessons from the Republic of Korea on Improving Services Labor Productivity The Republic of Korea is one of the few developing countries to successfully dodge the middleincome trap and join the high-income club. Its remarkable economic success has largely been built on export-oriented industrialization, which gave rise to a world-class high-tech manufacturing industry. 23

But services lag far behind the economy s dynamic manufacturing sector. In 2010, industry s labor productivity was 118% of the OECD figure, but services labor productivity languished at 43%. Of particular concern is the weak performance in modern areas such as business services. This jeopardizes future growth, as the country is in the midst of a structural transition into a more services-oriented, postindustrial phase. Services employment share has risen much more sharply than its output share. From 1980 and 2008, services employment share soared 30 percentage points, from 37% to 68%, but its GDP share rose less than 15 percentage points, from 47% to 61%. The Republic of Korea s exceptionally rapid deindustrialization pace has resulted in services becoming the source of many marginal, low-wage jobs for workers who cannot find good manufacturing jobs. This drags down the entire sector s labor productivity. Some policy options to diagnose weaknesses in services are uncontroversial, such as setting up training programs to help workers gain the skills needed for modern services, and augmenting investment in research and development and in ICT. Yet some factors holding services back, such as government regulations and restrictions, tariffs, and impediments to foreign direct investment, can be more sensitive. In particular, serious political considerations designed to protect small and medium-sized enterprises and their large workforce complicate regulatory reform. Further, regulatory barriers tend to be high for business and professional services such as finance, law, education, and health care. This holds down services labor productivity even as modern services GDP share approaches OECD levels. The Republic of Korea thus faces big obstacles to developing its modern service sector. It also has some valuable assets, notably its highly skilled, well-educated workforce, but the transition will still require a lot of hard work. This example shows that manufacturing success does not necessarily translate into services success. The stark lesson for developing Asia that should not be underestimated is that even a highly successful economy can find it challenging to build up a dynamic, highly productive service sector. Reference: Park and Shin (2013b). ***** End Box ***** The literature, as summarized in Eichengreen, Perkins, and Shin (2012), contains arguments outlining services labor productivity growth obstacles: First, services being labor intensive, not capital intensive, makes it difficult to benefit from the innovation embodied in physical capital. Second, services firms are typically too small to invest in research and development or to risk new production techniques. Third, international competition is generally weak because most services are not tradable. 24

Finally, much of services employment masks the underemployment of those unable to find better jobs. It has long been argued for these reasons that, as economies become more services-oriented, growth slows. As manufacturing automates and resources are reallocated to services, achieving labor productivity growth becomes more challenging. Park and Shin (2013a) performed a cross-country econometric analysis to discern the major services labor productivity determinants. The coefficient of the initial per capita GDP is negative and highly significant. This means that the lower the initial per capita GDP, the higher the subsequent services labor productivity growth rate. This result bodes well for developing Asian countries, where incomes are still relatively low. The coefficient of services trade as a percentage of GDP is positive and significant at 1%, suggesting that trade in services improves labor productivity. This is plausible as exporting services exposes domestic providers to competition, forcing them to become more efficient (Francois 1990). The cross-country empirical results suggest a number of policy implications. For example, since services labor productivity grows more readily in poorer countries, reform is especially important in those countries. This applies all the more for developing Asian countries since they generally still have lower incomes. Similarly, the relationship of services trade to productivity suggests barriers should be lowered. It should be noted that low labor productivity also plagues industry in developing Asia. As is the case for services, labor productivity in Asia lags far behind the OECD level (Figure 11). In some countries, the productivity gap with OECD is wider for services than for industry, but in others the reverse is true, revealing no clear pattern (Figure 12). Most Asian economies face the daunting challenge of closing the 25

productivity gap in both industry and services. Meeting this challenge will require tackling a wide range of structural and policy impediments. In addition, synergies between industry and services have implications for policies that can attend to productivity concerns simultaneously. 26

Figure 11 Labor Productivity in Industry (in $ 000) OECD Singapore Republic of Korea Taipei,China Hong Kong, China Azerbaijan Malaysia Kazakhstan Thailand Georgia Armenia PRC Sri Lanka Philippines Indonesia Mongolia India Cambodia Pakistan Viet Nam Bangladesh Kyrgyz Republic 1990 2000 2011 0 20 40 60 80 100 120 2005 constant $ ('000) OECD = Organisation for Economic Co-operation and Development, PRC = People s Republic of China. Source: ADB estimates using data International Labour Organization. Key Indicators of the Labor Market (8th edition) and World Bank. World Development Indicators online database (databases accessed 22 July 2014). 27

Figure 12 Labor Productivity Compared to OECD, 2011 (in %) Services Industry Singapore Hong Kong, China Taipei,China Republic of Korea Malaysia Kazakhstan Georgia India PRC Sri Lanka Thailand Armenia Philippines Indonesia Pakistan Azerbaijan Mongolia Viet Nam Cambodia Bangladesh Kyrgyz Republic 0 30 60 90 120 150 180 % of OECD's productivity OECD = Organisation for Economic Co-operation and Development, PRC = People s Republic of China. Source: ADB estimates using data International Labour Organization. Key Indicators of the Labor Market (8th edition) and World Bank. World Development Indicators online database (databases accessed 22 July 2014). 28

6 Underdeveloped Modern Service Sector in Kazakhstan and Developing Asia In Asia, an important reason for low services productivity is the dominant role of traditional services. Modern services, such as information and communication, finance, and professional business services, generally have a smaller presence in Asia than they do in advanced economies (Table 2). In 2010, they constituted only 8% 12% of the economy in the PRC; India; Indonesia; Taipei,China; and Thailand, while in advanced economies such as France, Japan, and the US they were twice as large, accounting for 17% 25%. Only three countries in developing Asia Singapore, the Republic of Korea, and Hong Kong, China have modern service industries comparable in size with those of the OECD. Advanced economies have shifted toward modernizing services in order to enjoy higher productivity and offer better wages than with traditional services. Eichengreen and Gupta (2009) found a positive correlation between services output and income share per capita, but this relationship holds only for services that combine both traditional and modern features that are consumed mainly by households, such as education and health care, and for modern services provided to households and businesses alike, such as finance and ICT. Furthermore, modern services enjoy the highest productivity growth among service industries, and their output share tends to rise rapidly as high income is achieved. These findings further support that the key to improving Asia s overall services productivity lies in shifting from traditional to modern service industries. Table 2 29

Economy Total Services Trade Developing Asia Services Share in Value Added, 1990 and 2010 (in %) Hotels and Restaurants Transport and Storage Real Estate and Dwellings Public Administration, Community, Personal, and Other Services Communication, Finance, and Business Services 1990 2010 1990 2010 1990 2010 1990 2010 1990 2010 1990 2010 1990 2010 PRC 31.5 43.4 6.8 8.5 1.6 2.1 3.8 4.9 2.1 7.3 7.9 11.2 9.4 9.4 Hong Kong, China 87.2 92.9 21.8 24.0 3.0 3.3 7.7 8.1 5.1 5.2 30.3 27.9 19.4 24.4 India 46.1 54.7 11.8 15.1 1.0 1.4 6.4 6.4 5.0 6.1 13.3 14.5 8.8 11.2 Indonesia 42.4 37.7 13.5 10.9 3.2 2.8 6.1 3.4 2.9 2.6 10.1 10.2 6.5 7.8 Republic of Korea 51.5 58.5 11.8 8.6 2.4 2.3 4.7 4.2 6.5 7.2 14.8 20.1 11.2 16.1 Malaysia 44.9 46.0 10.9 11.9 2.2 2.3 3.8 3.3 5.4 4.1 8.3 9.7 14.4 14.6 Philippines 50.8 55.1 14.7 17.4 3.2 3.9 5.8 6.5 15.7 13.4 11.5 13.9 Singapore 67.8 71.7 13.1 16.5 3.5 2.2 11.4 8.6 3.6 4.1 9.6 10.7 26.6 29.6 Taipei,China 55.0 66.2 13.4 18.8 1.7 2.0 4.6 3.3 6.4 8.9 17.5 20.8 11.4 12.4 Thailand 50.9 43.0 17.8 13.1 5.4 4.7 4.5 4.1 2.2 1.4 9.7 12.0 11.3 7.7 OECD France 69.2 79.7 11.8 10.6 2.3 2.6 4.6 5.0 9.8 13.4 21.7 26.1 18.9 22.0 Japan 59.8 72.6 12.8 12.3 4.9 4.5 9.4 13.0 19.1 25.7 13.6 17.2 United States 73.4 80.2 12.9 11.6 3.4 3.8 3.0 2.8 12.1 12.2 23.0 24.8 18.9 25.1 Latin America Chile 49.8 53.9 14.7 9.4 5.0 5.9 4.9 4.4 12.7 16.2 12.5 17.9 Mexico 61.1 64.2 15.7 11.5 6.9 5.7 8.8 10.0 14.9 20.4 12.4 13.1 = data not available or combined with other services, OECD = Organisation for Economic Co-operation and Development, PRC = People s Republic of China. Note: Initial data for Malaysia and Hong Kong, China are from 2000; for Indonesia and Mexico, from 1993; and for the Philippines, from 1998. Latest data for the PRC and Japan are from 2009. Source: ADB estimates using data from CEIC Data Company (accessed 25 April 2012). The key question then is why is developing Asia s modern service sector relatively underdeveloped? One factor is the strong education and skills prerequisites needed for employment in services such as ICT, finance, professional services, and management support. These requirements are hurdles to entry into such services, but the industry compensates with high average earnings. Table 3 shows the high share of workers in business-related services in the US with college and advanced degrees 44%, compared with 23% in manufacturing and earnings premiums of 20% over manufacturing and more than 65% over traditional service industries. 30

Table 3 Education and Earnings in Selected Sectors in the United States, 2007 Workers with a College Degree (%) Workers with an Advanced Degree (%) Average Earnings ($) Sector Manufacturing 23 7 49,081 Business services 44 14 59,096 Personal services 36 16 35,261 Wholesale and retail 19 3 35,819 Source: Jensen (2013). Figure 13 shows business services total employment share in selected economies in developing Asia. (It excludes large parts of the PRC because data are available only for urban areas, which account for 15% of the workforce, and of India, as detailed industry data are available only for the formal sector.) The business service industries of most countries in developing Asia (excepting Singapore and the Republic of Korea) are small and undeveloped compared with those of the US or the UK. It is notable that the PRC has a relatively small business service industry even in urban areas. Formal India has a respectably large business service industry, but it is likely unrepresentative of the country. 31

Figure 13 Business Services Share in Total Employment, 2007 (in %) United Kingdom United States Republic of Korea Singapore India, formal sector Taipei,China Palau Kazakhstan PRC, urban Malaysia Philippines Sri Lanka Kyrgyz Republic Armenia Azerbaijan Indonesia Mongolia Georgia Thailand 0 5 10 15 20 25 % PRC = People s Republic of China. Source: Jensen (2013). The prospects for business services growth in developing Asia depend, therefore, on raising the region s skills endowment through higher education. Figure 14 shows the average educational attainment of two age cohorts in select countries in 2010. The countries are plotted by average number of years of schooling and per capita GDP, with the size of the bubble indicating the relative workforce size. Ample stocks of human capital, including well-educated older workers, have contributed to developing 32

business services in advanced economies, while lower educational attainment, especially among older workers, has impeded business services in developing Asia. Figure 14 Two Cohorts Average Years of Schooling against GDP per capita, 2010 (in $ 000) AUS = Australia, AUT = Austria, BEL = Belgium, CAN = Canada, EGY = Egypt, BRA = Brazil, FIN = Finland, FRA = France, GDP = gross domestic product, IND = India, INO = Indonesia, ITA = Italy, JPN = Japan, KOR = Republic of Korea, MEX = Mexico, PRC = People s Republic of China, RUS = Russian Federation, SPA = Spain, TUR = Turkey, UKG = United Kingdom, USA = United States of America. Note: Bubble size indicates the size of the workforce. Source: Jensen (2013). 33

The figure reveals a dramatic improvement in educational attainment in developing Asia in little more than a generation. As Asian attainment converges with that of advanced countries, business service industry size and productivity will likely converge as well. However, as average educational attainment cannot rise much in the short term, developing Asia s growth prospects will continue to lag for years to come. The obvious solution is to liberalize business service imports in the short term while strengthening human capital and other infrastructure necessary for producing business services in the long term. 7 Policy Options for a More Vibrant Service Sector in Kazakhstan and Developing Asia Diversity in services and across Asia means that barriers and policy options will largely be tailored to individual countries and industries. Some common themes regarding enabling services development nevertheless emerge, including gaps in regulations, human capital, and infrastructure. 7.1 Regulatory gaps and policy responses A well-functioning regulatory environment protects consumers and maintains competitive markets; by contrast, anecdotes abound about regulatory overload. India, for example, has some 13 bodies to regulate higher education, each functioning in isolation. In the PRC, private service providers bear the brunt of heavy regulation because the state s impartiality as the industry regulator is undermined by its simultaneous participation as a major competitor via state-owned enterprises (SOEs). Wölfl et al. (2010) developed what they call product market regulation indicators to quantify burdensome and potentially anticompetitive regulations. These indicators, 34

shown in Table 4, cover three domains that jointly influence regulation: state control, barriers to entrepreneurship, and barriers to trade and investment, each divided further into categories. A higher score denotes a heavier regulatory burden. Although country coverage is limited, the table shows that firms in developing Asia generally face heavier regulatory burdens than their counterparts in the OECD, with India, Indonesia and especially the PRC (largely due to the continued prominence of SOEs) all suffering a regulatory burden at least twice as heavy. Table 4 Comparative product market regulation, 2010 (Index scale of 0 6 from least to most restrictive) Indicator PRC India Indonesia 35 Republic of Korea OECD average Product market regulation 3.30 2.84 2.73 1.48 1.36 State control 4.63 3.58 4.36 1.99 2.04 1. Public ownership 5.33 4.00 5.10 2.76 2.93 Scope of public enterprise 6.00 4.91 5.73 1.75 3.08 Government involvement in infrastructure 5.48 4.65 4.83 2.65 3.30 Direct control over business enterprise 4.50 2.45 4.74 3.88 3.20 2. Involvement in business operation 3.94 3.15 3.63 1.22 2.42 Price controls 4.38 1.13 3.00 1.78 2.64 Use of command & control regulations 3.50 5.18 4.25 0.67 2.53 Barriers to entrepreneurship 2.89 2.73 1.86 1.14 1.42 1. Regulatory and administrative opacity 0.25 2.01 0.16 0.00 1.55 License and permits system 0.00 2.00 0.00 0.00 1.87 Communication and simplification of rules and procedures 0.50 2.02 0.32 0.00 0.91 2. Administrative burdens on startups 5.58 4.44 1.64 1.57 1.68 Administrative burdens for incorporation 5.25 4.50 1.00 2.75 1.36 Administrative burdens for sole proprietorships 5.50 5.50 2.25 0.75 1.53 Sector-specific administrative 6.00 3.33 1.67 1.21 1.55

burdens 3. Barriers to competition 2.83 1.74 3.79 1.85 1.77 Legal barriers 1.43 0.86 4.57 1.14 1.52 Antitrust exemptions 0.00 1.23 2.86 0.44 1.37 Barriers to entry into network sectors 5.39 3.56 3.92 3.52 1.57 Barriers to entry into services 4.50 1.33 0.00 2.31 1.76 Barriers to trade and investment 2.40 2.22 1.97 1.30 0.63 1. Explicit barriers to trade and investment 2.52 2.84 2.33 1.00 1.08 Barriers to FDI 3.36 2.52 2.88 1.01 1.34 Tariffs 2.00 4.00 1.00 2.00 1.31 Discriminatory procedures 2.21 2.00 3.13 0.00 1.08 2. Other barriers 2.27 1.60 1.60 1.60 0.79 Regulatory barriers 2.27 1.60 1.60 1.60 0.87 FDI = foreign direct investment, OECD = Organisation for Economic Co-ordination and Development, PRC = People s Republic of China. Source: Park and Shin (2013b) using data for the beginning of 2008 from OECD. Product Market Regulation database (accessed 14 March 2012). The heavy presence of SOEs in many Asian service industries militates against development, as regulations often protect them from domestic competition. In the PRC, SOEs still play a large role in rail transportation, education, health care, news and publishing, and television broadcasting. In India, railways and postal services remain government monopolies. More generally, vested interests that stand to lose from competition, regardless of whether competitors are public or private, exert political pressure on governments and abuse industry standards and codes of practices to limit market entry and competition Regarding competition from abroad, developing Asia maintains some of the world s most restrictive policies on services trade. As a result, the region s share of the growing global services trade remains limited, although it has grown in recent years (Figure 15 and 16). Borchert, Gootiiz, and Mattoo (2012) indexed the services trade 36

policy restrictiveness of 79 developing and transitional economies, as well as in 24 developed countries. The authors found fairly strong correlation between lower per capita income and restrictive barriers, identifying the PRC, India, Indonesia, Malaysia, the Philippines, and Thailand as having notably high barriers (Figure 17). Figure 15 Shares of Global Services Trade held by Asia-12 and Advanced Economies, 2000 and 2010 (in %) % 2000 2010 80 60 40 20 0 Asia-12 Advanced economies Note: In Asia-12 are the People s Republic of China; Hong Kong, China; India; Indonesia; the Republic of Korea; Malaysia; Pakistan; the Philippines; Singapore; Taipei,China; Thailand; and Viet Nam. The advanced economies are the EU-15 (Austria, Belgium, Denmark, Finland, France, Germany, Greece, Ireland, Italy, Luxembourg, Netherlands, Portugal, Spain, Sweden, and the United Kingdom), Japan, and the United States. Source: ADB estimates using data from World Bank. World Development Indicators online database (accessed 2 September 2012). 37