Chapter 2 Worldwide decline in growth expectations

Chapter 2 Worldwide decline in growth expectations Key Points of Part 1, Chapter 2 In developed countries, the economic growth has slowed down since the global economic crisis because of weak growth in total demand. With their real growth rate staying lower than their potential growth rate, developed countries have stayed in a state of negative GDP gap. In addition, there have been discussions about the relationship between this situation and such factors as widening income inequality, expanding household debts and the growing role of information technology in industry. In the European Union (EU), for example, the polarization of employment is proceeding due to the loss of routine-type jobs, which previously absorbed middle-class. The potential growth rate has also been trending downward in the medium to long term because of the aging of society coupled with low birthrates, among other factors, so the need for productivity improvement is growing. Section 1 Weak growth in total demand in developed countries 1.Economies of developed countries since the global economic crisis Since the global economic crisis, the economic growth of developed countries has slowed down compared with the pre-crisis growth. A comparison concerning the average annual economic growth rates 95 before and after the economic crisis shows that the growth rate slowed down from 2.2% to.8% in the euro area, from 3.% to 1.9% in the United Kingdom, and from 2.7% to 2.1% in the United States (Figure I-2-1-1). 95 The simple average of the real GDP growth rates between 2 and 27 and between 21 and 215 extracted from the World Economic Outlook Database October 215 (IMF). Regarding the euro area, the calculation was made using the aggregation of the growth rates of the individual countries in the area weighted on the basis of the purchasing power parity. 29

(%) 3.5 3. 2.5 2. 1.5 1..5. Figure I-2-1-1 Economic growth rates before and after the economic crisis 2-27 3.1 21-215 2.65 2.22 1.9 2.14.82 Euro area United Kingdom United States Note: Figures are the simple average of the values calculated by the IMF as the annual growth rate for each period. Source: World Economic Outlook Database October 215 (IMF) According to Japan s Monthly Economic Report, in December 215, the U.S. economy was continuing to recover in the United States, was recovering at a moderate pace in the euro area and was recovering in the United Kingdom. However, since the global economic crisis, entering an expansionary phase or a similar expression has not been used in describing the economic situation in these regions (Table Ⅰ-2-1-2). Table I-2-1-2 Economic outlook on the United States and Europe in the Monthly Economic Report (issue of December of each year) United States Europe The economy is expanding. The economy is recovering at a moderate 25 pace in the euro area. The UK economy is recovering. The expansion of the economy is slowing The economy is recovering. 26 down. The economy is continuing to recover at a The economy is recovering. 27 moderate pace due to the decrease in house construction. 28 The economy is shrinking. The economy is shrinking. Although the US economy is still in a Although the European economy is still in a serious condition with the unemployment serious condition with the high 29 rate hovering around 1%, it is recovering unemployment rate, the economy seems to at a moderate pace thanks to the effects of have bottomed out thanks to the effects of the policy measures. 21

the policy measures, including the one for supporting the upgrading of automobiles. Although there are still some negative Although the economy is recovering in factors, including the lingering high Europe as a whole, there are significant 21 unemployment rate, the economy is gaps among the countries. The German moderately recovering thanks to the effects economy is recovering at a moderate pace. of the policy measures. The economic recovery is becoming weak. The economic recovery is slowing down. The German economy is recovering at a moderate pace, although some sectors show 211 signs of weakening. The French and UK economies are treading water, with some sectors weakening. While the economy is continuing to recover The economy is weakening. Although the only at a slow pace, it seems to be gaining German economy is picking up moderately, 212 some strength as well. there are some sectors showing signs of weakening. The economy is recovering at a moderate Although the economy still remains weak, 213 pace. there are some signs of recovery. The German and UK economies are recovering. The economy is recovering. The euro area continues to see the signs of 214 economic recovery. The UK economy is recovering. The economy is continuing to recover. The euro area is seeing a moderate economic recovery. The German economy 215 is recovering at a moderate pace, although some sectors are weakening. The UK economy is recovering. Source: Monthly Economic Report (Cabinet Office) Forecasts of the world economic growth that are periodically published by international organizations have been repeatedly revised downward 96. According to the October 21 edition of the IMF s World Economic Outlook report, the real economic growth rate in developed countries in 215 was forecast at 4.4%, but the forecast was revised downward to 2.% in the October 215 edition of the report (Figure Ⅰ-2-1-3). 96 While economic forecasts by the IMF (World Economic Outlook) are cited here, forecasts by other organizations, including the OECD and the United Nations, have also been repeatedly revised downward. 211

Figure I-2-1-3 Transition Downward revisions of developed countries growth rates by the IMF s World Economic Outlook reports (%) 3.5 3. 2.5 2. 1.5 1..5. 21 211 212 213 214 215 216 217 218 219 22 IMF WEO Oct 21 2.7 2.2 2.6 2.6 2.5 2.4 IMF WEO Oct 212 3. 1.6 1.3 1.5 2.3 2.6 2.6 2.6 IMF WEO Oct 214 3.1 1.7 1.2 1.4 1.8 2.3 2.4 2.4 2.3 2.3 IMF WEO Oct 215 3.1 1.7 1.2 1.1 1.8 2. 2.2 2.2 2.2 2. 1.9 IMF WEO Oct 21 IMF WEO Oct 212 IMF WEO Oct 214 IMF WEO Oct 215 Source: World Economic Outlook (IMF) of the relevant years Among the likely factors behind the slowdown of the real economic growth rate in developed countries are weak growth in total demand and supply trends associated with a decline in the potential growth rate. The GDP gap that represents the difference between the real and potential growth rates in the OECD member countries has remained negative since 29 although the negative gap has been shrinking in recent years. This suggests that means of production, including capital stock and labor, are not being sufficiently utilized due to a total demand shortage. (Figure Ⅰ-2-1-4). 212

2 21 22 23 24 25 26 27 28 29 21 211 212 213 214 215 216 217 GDP gap Figure I-2-1-4 Changes in the GDP gap in the OECD member countries 4 3 2.3 2.9 2 1 1.5.3.6 1.3 1.2-1 -2-3 -.4 -.5-2.4-2. -2.2-2.5-2.2-1.8-1.2 -.6-4 -5-3.8 Note: The GDP gap is defined as real GDP/potential GDP-1. The above figures are the values calculated by the OECD. Source: OECD.Stat Regarding the background to the weak growth in total demand, a variety of explanations have so far been offered. The secular stagnation theory advocated by former U.S. Secretary of Treasury Lawrence Summers maintains that one background factor is the fact that as a result of a decline in the propensity to invest caused by a rise in the propensity to save in developed countries, the natural interest rate that achieves full employment has fallen and demand has declined 97. In this theory, increased inequality, growing uncertainty over the future, reduction in the ability to borrow, a greater accumulation of assets by foreign central banks and sovereign wealth funds are cited as the reasons for the increased propensity to save. As the reasons for the reduced propensity to invest, slower growth in the labor force, the availability of cheaper capital goods, tighter credit (due to strengthened regulation of borrowings) and the new economy s tendency to conserve capital 98 are cited. As for the contributing factors of the stagnant economic growth since the global financial crisis, explanations other than the secular stagnation theory have been provided from the viewpoints of the expansion of cumulative debts 99, a decline in productivity growth 1, capital inflow from countries with 97 Summers, L. H. (216), The Age of Secular Stagnation, Foreign Affairs, Volume 95 Number 2, March-April 216, Council on Foreign Affairs 98 As examples of the impact of the new economy that tends to conserve capital, Summers cited the restraining impact of the arrival of the sharing economy on hotel construction and automobile demand, the restraining impact of the rise of e-commerce on construction of malls, and the shrinkage of demand for copiers, printers, and office space due to the development of information technology. 99 Refer to Reinhart, Carmen M. and Rogoff, Kenneth S. (211), This Time Is Different: Eight Centuries of Financial Folly, Princeton University Press, and Mian, Atif and Sufi, Amir (214). 1 Refer to Gordon, R. (216), The Rise and Fall of American Growth: The U.S. Standard of Living Since the Civil War, Princeton University Press. 213

current account surplus, the expansion of the global imbalance 11 and the liquidity trap. For example, there is the view that the restriction on borrowings will be tightened because of the expansion of household debts caused by increased borrowings and as a result, consumption will be curbed and a total demand shortage will persist for a long period of time 12. The next paragraph will analyze the growth model that has been driving the economic development of Japan and other developed countries and the state of the middle class. 2. Changes in the employment structure Developed countries hitherto had realized an increase in per-capita income due to a continuous increase in the proportion of middle-class in particular who constitute the core stratum of persons employed and this supported the growth of their economies as a whole. However, in recent years, it has been pointed out that the proportion of middle-class has continued to decline. 13 On the other hand, the proportions of low-income and high-income have been rising. The decrease in middle-class and the increase in low-income and high-income strongly suggest that the employment structure is polarized between low-income and highincome. For developed countries trying to cope with downward pressure on demand due to a population decline by increasing per-capita income amid the shrinkage of the working-age population, this change in the employment structure may have a considerable impact on their growth model. Below, this white paper closely examines changes in the employment structure in developed countries, including Japan, and considers the contributing factors of the changes. (1) Changes in the employment structure in the United States, Germany and Japan Figure I-2-1-5 shows changes in the number of employees by occupation type in the United States, Germany and Japan. It also shows the most recent wage levels concerning specific occupations. 11 Refer to the theory of the global saving glut that was proposed by former Federal Reserve Board Chairman Ben Bernanke. 12 Kobayashi, K. (216), Persistent Demand Shortage Due to Household Debt, RIETI Discussion Paper Series, 16-E-12, February 216, Research Institute of Economy, Trade and Industry 13 For example, according to a report issued in May 214 by the Federal Reserve Bank of Dallas (Cheremukhin, A. (214)) the share of middle-skill jobs, which provide the foundation of the middle class, has declined steeply in the U.S. labor market since the 198s, whereas the shares of low-skill jobs and highskill jobs have grown. This report noted that the share of middle-skill jobs in the total number of employees, which was 58% in 1981, was down to 44% in 211 and that the shares of low-skill jobs (13% in 1981) and high-skill jobs (29% in 1981) rose to 17% and 39%, respectively, by 211. 214

Figure I-2-1-5 Changes in the number of employees and wage level by occupation type United States 6 5 4 3 2 1-1 -2-3 -4 (Million people) 2.4 55 637 Service Transportation and carrying 651-2.2 Clerical 646-2.9 Production process 75 Sales 756 -.1 Construction and mining 821 Workers engaging in the installation, maintenance and repair of equipment (Dollars/week) 1,2 1,78 1, 4.9 Professional 8 6 4 2 Changes in the number of employees (from 2 to 214) Wage level (214) (right axis) Source: Website of the United States Department of Labor Germany (Increase/decrease, million people) 3 2 1 2..4 63 65 93 2.1 (1, dollars/year) 111 12 1 2.8 8-1 54 62 -.7-1.9 6 4-2 2-3 Sales Clerical and assistant Machine operation Construction and production process Engineering Professional Increase or decrease in the number of employees (from 1991 to 212) Wage level (21) (right axis) Source: EUROSTAT 215

4 3 2 1-1 -2-3 -4-5 (Increase/decrease, million people) 2.7 21.9 217.6 219. -.8 -.5 Japan 236.2 243.5-4.3 Service Sales Clerical Production process -.1 Transport and machine operation Increase or decrease in the number of employees (from 1992 to 212) Wage level (214) (right axis) (1, yen/month) 272.9 2.3 Professional and engineering 3 25 2 15 1 5 Note: Figures for 1992 are based on the old occupational classification before the amendment (December 29) and those for 212 are based on the new occupational classification. Production process in 1992 refer to production process and laborers under the old occopational classification and those for 212 refer to the total of production process, construction and mining and carrying, cleaning, packaging, and related under the new occupational classification. Transport and machine operation are equivalent to transport and communication under the old occupational classification. Source: Employment Status Survey (Ministry of Internal Affairs and Communications) and Basic Survey on Wage Structure (Ministry of Health, Labour and Welfare) This indicates that in all of these countries, the numbers of employees engaging in occupations whose wages are at around the medium level, such as production process and clerical occupations, are falling although there are differences in the period of observation of changes in the number of employees due to data constraints (the period is from 2 to 214 for the United States, from 1991 to 212 for Germany, and from 199 to 212 for Japan). The figure also shows that the number of employees engaging in services occupations, whose wage level is relatively low, and the number of employees engaging in professional and technical occupations, whose wage level is relatively high, are increasing. However, the wage difference between occupations in Japan is very small compared with the difference in the United States and Germany. While the wage difference between professional occupations, the wage level of which is the highest, and service jobs, the wage level of which is the lowest, is very large in the United States and Germany, the difference between the two types of occupations is small in Japan. Therefore, although it is presumed that a change is occurring in the employment structure in terms of difference between occupation types in Japan, it is not appropriate to 216

discuss the situation in the country in parallel with the polarization of employment between income classes based on the large wage difference that is observed in the United States and Germany. Why is the change in the employment structure being observed universally across various countries? (2) Contributing factors of the change in the employment structure Table I-2-1-6 shows the corresponding relationship between the occupational classification and the task classification. The task is the minimum unit of an occupation and defines the contents and nature of specific work required by the occupation. Jobs are classified into routine and non-routine tasks and also into manual and cognitive tasks. 14,15 Routine Table I-2-1-6 Occupational classification and task classification Cognitive Clerical General clerical Accountancy clerks Production-related clerical Sales clerks Outdoor service Transport and post clerical Office appliance operators Sales Merchandise sales Quasi-sales Sales Manual Production process Product manufacturing and processing (metal products) Product manufacturing and processing (except metal products) Chemical product manufacturing Ceramic, earth, and stone product manufacturing Food manufacturing Beverage and cigarette manufacturing Spinning, weaving, apparel, and fiber product manufacturing Wooden and paper product manufacturing Printing and bookbinding Rubber, plastic product manufacturing Other product manufacturing and processing (except metal products) Machine assembly Machine maintenance and repair Product inspection Machine inspection 14 Cognitive abilities as referred to here are abilities of perception, judgment, imagination, inference, decision, memory and language understanding, for example. 15 The classification of tasks here (routine tasks, non-routine tasks, manual tasks and cognitive tasks) is based on Autor, Levy, and Murnane (23) and Autor and Price (213). Autor and Price (23) put a clear division between skills and tasks, and define a task as the minimum unit of a job (occupation) and skills as a set of abilities accumulated by individual in order to perform tasks. They argue that devote their skills to tasks in exchange for receiving wages. The classification here focuses on tasks. 217

Nonroutine Administrative and managerial Management government officials Officers of companies and organizations Other administrative and managerial Professional and engineering Researchers Engineers Data processing engineers Other engineers Health care Doctors (except dental surgeons and veterinary surgeons) Nurses (including assistant nurses) Other health care Social welfare specialist professionals Legal Management, finance, and insurance specialists Teachers Workers in religion Authors, journalists, editors Artists, designers, photographers, film operators Musicians, stage designers Other specialist professionals Manufacturing-related and quasimanufacturing Construction and mining Construction and civil engineering Electric construction Mine Carrying, cleaning, packaging, and related Carrying Cleaning Packaging Other carrying, cleaning, packaging, and related Service Domestic support service Care service Healthcare service Domestic hygiene service Food and drink preparatory Customer service Residential facilities, office buildings and other management personnel Other service Transport and machine operation Railway drivers Motor vehicle drivers Ship and aircraft operators Other transport Stationary and construction machinery operators Security Agriculture, forestry and fishery Agriculture Forestry Fishery Source: Japan Standard Occupational Classification (Ministry of Internal Affairs and Communications), Autor and Price (23), etc. Routine tasks are those performed repeatedly in accordance with prescribed procedures. Assembly work in a manufacturing process and clerical work, for example, are routine tasks. As such tasks are performed in accordance with detailed, meticulously prepared work manuals in many cases, the contents of assembly work described in the manual, for example, can be easily coded into a computer program, so human labor in the manufacturing process is likely to be replaced by a 218

computer-controlled machine. Even performing tasks that require a certain level of cognitive ability, such as accounting, are likely to be replaced by low-wage in emerging countries now that it is possible to turn account books and accounting manuals into electronic data and instantly send the data abroad. Routine tasks are classified into routine cognitive tasks, such as accounting, and routine manual tasks, such as data input. Meanwhile, non-routine tasks are classified into non-routine cognitive tasks, which are advanced tasks that require problem solving, persuasion, intuition and imagination, and non-routine manual tasks, which require situational adaptability, a high level of visual and auditory sensitivities, language skills, and deftness of hand, for example. Non-routine cognitive tasks are performed by people engaging in professional and technical occupations, such as doctors and lawyers, people engaging in management occupations, such as business managers, and people engaging in creative occupations, such as writers and artists. All these occupations require a high level of education, analytical capability and imagination. On the other hand, non-routine manual tasks include tasks performed by people working in the service and transport industries, such as security guards, cooks and truck drivers. Of these two types of tasks, it is currently difficult to automate non-routine cognitive tasks through software coding because performing them requires many abstract elements. Meanwhile, in the case of non-routine manual tasks, which are mainly tasks related to services, such as security guarding, cooking, and driving, the provision of labor and consumption activity need to take place at the same time and in the same place, making it physically difficult to outsource such tasks abroad. The above classification of occupations into four types of tasks shows that many of the occupations for which the number of employees is decreasing as indicated in Figure I-2-1-5 correspond to routine tasks that can be easily automated or outsourced abroad. In contrast, many of the jobs for which the number of employees is increasing correspond to non-routine tasks that are regarded as being difficult to automate and outsource abroad. This strongly indicates that the change in the employment structure in developed countries has occurred as a result of the replacement of jobs characterized as routine tasks that can easily be automated or outsourced abroad by machines and low-cost foreign labor against the backdrop of the advance of IT and the diffusion of information and communications technology. Let us look at Figure I-2-1-7 as data that strongly corroborates this inference. 219

Figure I-2-1-7 Changes in the shares of employees by task type around the world (Changes in shares from 2 to 213, percentage point) 8 6 5.6 5.6 5. 4 3.2 3.5 2.4 2.4 3.4 2 1.1.5-2 -4-6 -8-1 -4.4 World -4.1 Developed countries (excluding the EU) -8.1 Developed EU countries -4.9-4.6 Countries of Central and Eastern Europe (excluding developed EU countries and CIS countries) -1. Middle Eastern countries -8.4 East Asian countries Source: "Trends Econometric Models, October 214" (ILO) 2.8-1.1-1.7 Southeast Asian and Pacific countries -8.3 3. 5.3 South Asian countries -3.9.5 3.3 2.7 Latin American and Caribbean countries Non-routine manual tasks Routine tasks Non-routine cognitive tasks -3..3 North African countries 1.2 -.3 -.8 Sub-Saharan countries This figure shows changes in the shares of employees engaging in routine tasks, non-routine manual tasks and non-routine cognitive tasks in regions around the world. Since 2, the share of employees engaging in routine tasks has declined steeply in developed countries but has grown significantly in emerging countries and regions, including East and South Asia and Central and East Europe. 16 (3) Current status of the change in the employment structure in Japan We will take a closer look at the current status of the change in the employment structure in Japan. Figure I-2-1-8 shows the long-term trend in the number of employees in Japan by task type. 16 The ILO observes that the steep decline in non-routine manual tasks in emerging countries/regions, including East Asia and South Asia, is due to a decrease in the number of engaging in agriculture, forestry and fishery. 22

Figure I-2-1-8 Trend in the number of employees by task type (Million) 6 5 4 3 2 1 1977 1982 1987 1992 1997 22 27 212 Routine tasks Non-routine manual tasks Non-routine tasks Non-routine cognitive tasks Note: Non-routine tasks = Non-routine manual tasks + Non-routine cognitive tasks Figures for 22 and before are based on the old occupational classification before the amendment (December 29) and those from 27 onward are based on the new occupational classification. Source: Employment Status Survey (Ministry of Internal Affairs and Communications) The figure makes clear that the decrease in the number of employees engaging in routine tasks started around the year 2 in Japan. This coincides with the start of acceleration of Japanese companies expansion abroad, including Asian countries. It also coincides with the start of the spread of the Internet worldwide. In contrast, the number of employees engaging in non-routine tasks, including both manual and cognitive ones, has consistently been rising. Figure I-2-1-9 shows the trend in the number of employees by occupation type more minutely classified. 221

Figure I-2-1-9 Trend in the number of employees by occupation type (Million) 25 2 15 1 5 1977 1982 1987 1992 1997 22 27 212 Production process and laborers (routine manual tasks) Clerical (routine cognitive tasks) Professional and engineering (non-routine cognitive tasks) Sales (routine cognitive tasks) Service (non-routine manual tasks) Agriculture, forestry and fishery (non-routine manual tasks) Transport and machine operation (non-routine manual tasks) Administrative and managerial (routine cognitive tasks) Security (non-routine manual tasks) Note: Figures for 22 and before are based on the old occupational classification before the amendment (December 29) and those from 27 onward are based on the new occupational classification. Production process from 27onward are the total of production process, construction and mining and carrying, cleaning, packaging, and related under the new occupational classification. Transport and machine operation in 22 and before are transport and communication under the old occupational classification. Source: Employment Status Survey (Ministry of Internal Affairs and Communications) From this figure, we can see that the decrease in the number of employees engaging in routine tasks is mainly due to a decline in the number of manufacturing process. The number of manufacturing process, which totaled 2 million in 1992, continued to drop rapidly thereafter, falling to 16 million in 212. The number of clerical and that of sales are also on a downtrend. On the other hand, we can see that the increase in the number of employees engaging in non-routine tasks is mainly due to a rise in the number of professional and engineering and that of service. In particular, the number of professional and engineering has been consistently rising since 1977. The number of professional and engineering, which was only slightly more than 4 million in 1997, reached 1 million in 212. Is the change in the employment structure in Japan proceeding universally in all regions across the country, or is there any difference between urban and provincial areas? Figures I-2-1-1 and I-2-1-11 show the trend in the number of employees by occupation type with 222

Rate of change Transport and machine operation respect to ordinance-designated cities (including Tokyo s special wards) and other cities (hereinafter referred to as non-ordinance-designated cities ). The horizontal axis represents the number of employees by occupation type, while the vertical axis represents the rate of change in the number of employees between 1992 and 212. The red horizontal line represents the rate of change in the total number of employees for all occupation types during the same period. Generally, the wage level (the median value in 214) by occupation type rises from the left end to the right end. Figure I-2-1-1 (%) 8 6 4 2-2 -4-6 71.6% Security 44.4% Service 1,312 Trend in the number of employees by occupation type (ordinance-designated 2,556-7.% Sales cities) (from 1992 to 212) Clerical 3.2% 3,425 2.9% -8.3% 3,641 Production process -2.1% Agriculture, forestry and fishery -22.2% Wage level (median value in 214) by occupation type Low 43.5% Professional and engineering 2,37 2, 4, 6, 8, 1, 12, 14, Number of employees in 1992 (1, people) 69 Administrative and managerial - 41.% High Total number of employees for all occupation types Note: Figures for 1992 are based on the old occupational classification before the amendment (December 29) and those for 212 are based on the new occupational classification. Production process in 1992 refers to production process and laborers and in 212 refers to the total of production process, construction and mining and carrying, cleaning, packaging, and related under the new occupational classification. Transport and machine operation are equivalent to transport and communication under the old occupational classification. Ordinance-designated cities refers to the cities designated by the ordinance as of 1992, which include Tokyo s special wards. Source: Employment Status Survey (Ministry of Internal Affairs and Communications) and Basic Survey on Wage Structure (Ministry of Health, Labour and Welfare) 223

(Rate of change) Figure I-2-1-11 (%) 6 4 2-2 -4-6 38.6% 4,126 Agriculture, Security - 44.3% Low 59.1% Service 3,666 6,762 Sales - 8.5% Trend in the number of employees by occupation type (non-ordinancedesignated cities) (from 1992 to 212) Transport and machine operation 9,528 Clerical - 6.8% - 1.1% 1,843 16,751 Production process - 21.2% Wage level (median value in 214) 23.2% Professional and engineering 5,856-6.9% 1,687-39.3% 1, 2, 3, 4, 5, (1, people) Number of employees in 1992 Administrative and managerial High Total number of employees for all occupation types Note: Figures for 1992 are based on the old occupational classification before the amendment (December 29) and those for 212 are based on the new occupational classification. Production process in 1992 refers to production process and laborers and in 212 refers to the total of production process, construction and mining and carrying, cleaning, packaging, and related under the new occupational classification. Transport and machine operation are equivalent to transport and communication under the old occupational classification. Non-ordinance-designated cities refers to the areas other than the cities designated by the ordinance as of 1992 and Tokyo s special wards. Source: Employment Status Survey (Ministry of Internal Affairs and Communications) and Basic Survey on Wage Structure (Ministry of Health, Labour and Welfare) A look at changes in the number of employees by occupation type in ordinance-designated cities in Figure I-2-1-1 shows that the number of manufacturing process, which has the largest share in the overall number of employees, declined 2.1% over the 2-year period until 212. The numbers of agriculture, forestry and fishery and administrative and managerial also fell steeply, 22.2% and 41.%, respectively. The decrease in the number of agriculture, forestry and fishery presumably reflects a decline in the number of dedicated farmers due to the aging of such and a lack of successors. Meanwhile, the decline in the number of administrative and managerial has been 224

presumably caused in large part by an increase in business closures due to the aging of owner managers of small and medium-size enterprises and a lack of successors. On the other hand, another factor is an increase in cases where routine tasks performed in the past mainly by so-called middle managers, such as the function of providing information, issuing instructions and orders and reporting, are replaced by computers as a result of the advance of information technology. The number of clerical increased albeit slightly. However, in ordinance-designated cities, the numbers of service, and professional and engineering, both of which are non-routine tasks, increased significantly, by 44.4% and 43.5%, respectively. In ordinance-designated cities, it is notable that the number of professional and engineering, the wage level of which is high, increased at a similar pace to the pace of the rise in the number of service, the wage level of which is relatively low. The overall number of employees grew 2.9% by absorbing those including manufacturing process who had been made redundant as a result of job cuts. On the other hand, Figure I-2-1-11 shows that in non-ordinance-designated cities, the number of manufacturing process, which has the largest share in the overall number of employees, dropped 21.1%, almost the same rate of decline as the rate in ordinance-designated cities, over the 2-year period until 212. The number of clerical and sales also declined, 6.8% and 8.5%, respectively. All these occupations are routine tasks. As for occupations for which the number of employees increased, the number of service increase most (59.1%), much more than the number of security (38.6%) and that of professional and engineering (23.2%). All these occupations are non-routine tasks. Unlike in non-ordinance-designated cities, the number of employees increased mainly in the service industry, where the wage level is relatively low. This indicates that in non-ordinance-designated cities, the increase in non-routine tasks, such as professional and engineering tasks, failed to fully absorb made redundant due to the decrease in routine tasks such as manufacturing processing. As a result, the overall number of employees in nonordinance-designated cities fell 6.9% over the 2-year period in contrast to the increase in ordinancedesignated cities. The results strongly suggest that in Japan, the employment structure is undergoing the same change as in the United States and European countries, except for a decline in the numbers of engaging in agriculture, forestry and fishery and employees engaging in management jobs (owner managers of small and medium-size enterprises) due to a lack of successors, in that routine tasks are decreasing while non-routine tasks are increasing. (4) Summary The above analysis shows that the polarization of employment, a decrease in the number of middleclass on the one hand and an increase in low-income and high-income on the other hand, is proceeding in the United States and European countries. Meanwhile, it was confirmed that although it is not necessarily clear in Japan whether the 225

polarization based on wage differences between jobs is proceeding, a change in the employment structure is occurring in the form of a decrease in routine tasks taking place at the same time as an increase in non-routine tasks, as is the case in the United States and European countries. In particular, it was confirmed that in Japan, the change in the employment structure is becoming prominent in non-ordinance-designated cities. In non-ordinance-designated cities, the overall number of regional jobs is decreasing as a result of the change in the employment structure. On the other hand, in ordinance-designated cities, although the change in the employment structure is recognized in some aspects, the overall number of regional jobs is increasing as a result of the significant increase in the number of professional and engineering. In non-ordinance-designated cities, the challenge will be how to increase the number of professional and engineering, whose wage level is relatively high, in order to maintain and raise income per resident in the future. 226