THE RELATIONSHIP BETWEEN DEMOGRAPHIC CHANGE AND INCOME INEQUALITY IN AGING SOCIETY OF THAILAND PAPUSSON CHAIWAT *, and SAWARAI BOONYAMANOND The incidence of poverty in Thailand has been continuously decreased while income inequality still remains high. This phenomenon reflects the unevenly distributed benefits of development to all individuals in the society, especially the elderly. This work attempts to investigate the relationship between demographic change ( as measured by the proportion of the elderly population) and income inequality ( as measured by inequality between the elderly and other population age groups). The correlation test is used to determine the relationship between these two factors while the Shorrocks index of order 2 is used to measure inequality between groups. It is found that the correlation coefficient is about 0.9746 significantly, meaning that both factors have a strong positive relationship. Accordingly, the continuous increase in the proportion of the elderly is the main source of the inequality problem in these recent years. Despite the government has implemented indirect taxation and social policies aiming to alleviate this problem, the rate of indirect taxation collected is too low to use as an efficient tool to distribute earnings among populations. This result confirms that the problem of income inequality, especially in the elderly group, will continue to rise when the country becomes an aging society. Keyword: Income Inequality; Aging Society; Shorrocks Index; Social policy JEL Classification: J1, O23, R1 1. Introduction Since 2010, Thailand has become an aging society. The elderly population grew from 1.7 million in 1970 (or about 4.9 percent of the total population) to 7.5 million in 2010 ( or about 11.8 percent of the total population) [Figure 1] and has continuously risen. This phenomenon largely is the results of declines in both the birth rate and the death rate and longer life expectancy. According to Institute for Population and Social Research (IPSR), it is forecasted that the elderly in Thailand will reach the number of12.9 million or about 19.8 percent of the total population in 2025. This implies that the problem of the old will pose an important question to the country s economy in the near future. * Graduate School of management and Innovation, King Mongkut s University of Technology Thonburi, Thungkhru, Bangkok 10140, Thailand Email: papusson.cha@kmutt.ac.th Corresponding authors Faculty of Economics, Chulalongkorn University, Phayathai Road, Bangkok 10330, Thailand Email: sawarai.b@chula.ac.th
Figure1Thailand s population projection 80 70 60 50 40 30 20 10 0 63.4 66.1 67.4 66 62.2 56.2 52.2 50 59.3 38.3 43.2 45.1 29.2 24.4 20.7 17.2 22.7 26.9 4.6 4.8 5.5 7.4 9.5 11.8 15.1 8.5 1960 1970 1980 1990 2000 2010 2020 2030 2040 Chidren (Age 0-14 years) Working Age (Age 15-59 years) Elderly (Age 60 years up) Source: Pattama Wapanawong (2006), cited in Chuenruthai Kanchanachitra (2007) Changes in the population structure inevitably have resulted in Thailand s declining economic growth, lower labor force, and higher income inequality. Sarntisart (2004) suggests that comparing with other countries in the same region, Thailand has a high level of economic growth but its benefits are not distributed evenly to all people especially to the poor or those with lower incomes. Moreover, Boonyamanond (2008) explain s that both taxation and government expenditures policies that Thailand has been implemented are not efficient tools to distribute equal incomes. This causes income inequality as measured by the Gini coefficient to remain high and exhibit an unpredictable trend. Since whenever the number of the elderly increases, households ability to earn income tends to decline with higher dependency burden and lower level of income on average. As a result, the income inequality problem will rise rapidly with a change in demographic structure. The purpose of this research is to examine the relationship between demographic change and income inequality in the aging society of Thailand. An analysis of the inequality of each age group is also conducted. 2. Data and Methodology Thailand s socio-economic surveys conducted in 1986, 1988, 1990, 1992, 1994, 1996, 1998, 2000, 2002, 2004, 2006, 2007 and 2009 are used. Population are categorized into 4 groups according to their age: the first group is the young working-age population aged 15-29 years, the second group is the middle working-age population aged 30-44 years, the third is the old working-age population aged 45-59 years), and the last group is the elderly60 years and over. The methodology is as follow: first, this study uses the Gini coefficient cient to explain the pattern of income inequality in each age group. Next, the Shorrocks index of order 2 is applied to decompose the within group and between group inequality among age groups. Finally, correlation coefficient is calculated to test the existing relationship between demographic change and between group inequalities. The 5% significance level is employed to indicate statistical al significance.
3. Results of the Study Income inequality in each age group: Gini coefficient The Gini coefficients calculated on individual s income are shown in figure 2. It is found that during 1986-2009, the elderly group (represented by the light blue line) displayed the highest Gini coefficients than other groups. This means that individuals aged 60 years and over exhibit different levels of income: while some may own businesses that generated high revenues, a huge number of them may have no income and live on transfers from their descendants. In addition, a rapid increase in inequality among the elderly especially in 2009 reveals a more serious problem in this group of population. Figure2 The Gini Coefficient measured by individual income of each age group 0.95 0.90 0.85 0.80 0.75 0.70 0.65 0.60 0.55 0.50 0.927 0.853 0.830 0.803 0.799 0.804 0.802 0.785 0.788 0.790 0.783 0.791 0.765 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 2006 2007 2009 All groups 15-29 Years 30-44 Years 45-59 Years 60 Years up Likewise, the older working-age population group also experienced an increasing inequality trend, meaning that the ability to earn income of this group had suddenly declined. Of interest are changes in income inequality among the population aged 15-29 years, which had relatively high inequality at the beginning, then constantly declined. This may cause by higher starting wages, more demand for labor, and higher education for the young. The middle working-age population group had the lowest inequality except in 2006, 2007 and 2009, which may due to higher employment rates and wages than other age groups. Overall, the Gini coefficientscients indicate a worsening situation of income inequality in Thailand. Despite the government s attempt to implement higher direct taxation to alleviate this problem, the rates of direct taxation are still too thin to be used as an efficient tool for distributing earnings among the population. So considering the source of disparity in each age group will provide a more appropriate policy direction to the policy-maker. Decompose of income inequality: Shorrocks index of order 2 When inequality is decomposed by employing the Shorrocks index of order 2, the results demonstrate that, on average, about 98% of inequality arose from difference in income among individuals within the same age group, while about 2% inequality arose from difference in income between age groups. These proportions were fluctuated from year to year. In 1986 and 1988,the
Shorrocks indices were relatively low at 2.8829 and 2.5538 respectively [Figure 3], which were largely influenced by the proportion of within group disparity of 96.7% and 96..5%. This means that the average income of each age group people was quite different during the period. As the Shorrocks indices had been decreased constantly from 5.3758 in 1990 to 2.1335 in 2004 representing a more equal distribution of income, the proportions of income inequality within the same age group went up, but slowly decreasing. 100% 99% 98% Figure3 The Shorrocks indices of order 2 and The proportions of within and between group inequality 0.03 0.04 0.01 0.01 0.01 0.01 0.02 0.02 0.02 0.02 0.01 0.01 0.01 5.38 4.78 6.88 6.18 7.50 6.50 5.50 97% 96% 95% 2.88 2.55 3.94 3.66 3.16 2.85 2.96 2.13 4.01 4.50 3.50 2.50 94% 0.97 0.96 0.99 0.99 0.99 0.99 0.98 0.98 0.98 0.98 0.99 0.99 0.99 1.50 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 2006 2007 2009 % within % between Shorrocks Nevertheless, the proportions of income inequality within the same age group were high again during 2006-2009. In 2006,the Shorrocks index went up to a very high level at 6..8755, while the ratio of income disparity within the same age group and between each age group was about 99.4:0.6.However in 2007, the Shorrocks index dropped to 4.0132before went up again in 2009 to the level of 6.1786, with the ratio of 98.8: 1.2 and 99.2: 0.8 respectively. These findings indicate that inequality within age group had a huge impact on overall income inequality in Thailand. To be more precise, it is likely that differences in income among individuals in the same age group are the main source of inequality, rather than differences in average income of each age group. Table 1 shows that during 1986-2004, inequality arose mostly from inequality within the middle and the old working-age population, which may due to the fact that individuals who lived in the urban tended to have more ability to earn income than those in the rural. This is also applicable to different regions across Thailand, which may benefit differently from the development of the economy. However, inequality had decreased as a result of government policies such as the public officers salary adjustment in 1992, the poverty reduction policy in 1997, the Thailand village fund portal in 2004, all of which seemed to have favorable impacts on those middle and old working-age population. As for the young working-age population and the elderly, they had relatively minor impact on the overall inequality due to their smaller proportion to the total population as well as the lack of demand for labor for these two age groups.
Table 1 Thailand s inequality for each age group Age Within Group Inequality s Between Group Shorrocks Year Group Inequality Impact Inequality Index 1986 Total 2.789 0.0939 2.8829 15-29 Y 0.2434 8.73% 30-44 Y 1.0654 38.20% 45-59 Y 1.3048 46.78% 60 Y up 0.1754 6.29% 1988 Total 2.4983 0.0895 2.5538 15-29 Y 0.1693 6.78% 30-44 Y 1.1837 47.38% 45-59 Y 0.9199 36.82% 60 Y up 0.1976 7.91% 1990 Total 5.3106 0.0652 5.3758 15-29 Y 0.7119 13.41% 30-44 Y 1.4028 26.42% 45-59 Y 1.1471 21.60% 60 Y up 2.0487 38.58% 1992 Total 4.7098 0.0682 4.7779 15-29 Y 0.275 5.84% 30-44 Y 2.7993 59.44% 45-59 Y 1.3566 28.81% 60 Y up 0.2789 5.92% 1994 Total 3.8832 0.0574 3.9406 15-29 Y 0.2152 5.54% 30-44 Y 2.1638 55.72% 45-59 Y 1.3065 33.65% 60 Y up 0.1977 5.09% 1996 Total 3.6058 0.0522 3.658 15-29 Y 0.3835 10.64% 30-44 Y 1.6521 45.82% 45-59 Y 1.3434 37.26% 60 Y up 0.2267 6.29% 1998 Total 3.0966 0.0596 3.1562 15-29 Y 0.1807 5.84% 30-44 Y 1.2105 39.09% 45-59 Y 1.459 47.12% 60 Y up 0.2463 7.96% 2000 Total 2.7999 0.0486 2.8485 15-29 Y 0.3006 10.74% 30-44 Y 0.8209 29.32% 45-59 Y 1.4692 52.48% 60 Y up 0.2092 7.47% 2002 Total 2.9087 0.0502 2.9589 15-29 Y 0.1004 3.45% 30-44 Y 1.1814 40.62% 45-59 Y 1.3766 47.33% 60 Y up 0.2503 8.61%
Age Within Group Inequality s Between Group Shorrocks Year Group Inequality Impact Inequality Index 2004 Total 2.0865 0.0470 2.1335 15-29 Y 0.0893 4.28% 30-44 Y 0.7418 35.55% 45-59 Y 1.0557 50.60% 60 Y up 0.1997 9.57% 2006 Total 6.8355 0.0400 6.8755 15-29 Y 0.1694 2.48% 30-44 Y 1.7377 25.42% 45-59 Y 2.6584 38.89% 60 Y up 2.27 33.21% 2007 Total 3.9665 0.0462 4.0132 15-29 Y 0.1029 2.59% 30-44 Y 2.0834 52.53% 45-59 Y 1.5327 38.64% 60 Y up 0.2483 6.26% 2009 Total 6.1307 0.0479 6.1786 15-29 Y 0.1598 2.61% 30-44 Y 1.885 30.75% 45-59 Y 2.7228 44.41% 60 Y up 1.3632 22.24% In the corresponding periods, Thailand faced with the deteriorating distribution of income. The Shorrocks index significantly increased from 2.1335 in 2004 to 6.8755 in 2006, of which 6.8355 represented within age group inequality and 0.04 represented between age group inequality. Of the total inequality within age group, 38.89% were inequality within the old working-age group, 33.21% were the elderly group, 25.42% were the middle working-age group, and 2.48% were the young working-age group. Consistent with income inequality as measured by the Gini coefficient, the Shorrocks index also shown a rapid increase in income inequality among the elderly. Such increase was triggered by a high level of growth in the export sector, in which businesses are owned by some experienced individuals in their later working-age and their old age, who will receive even higher income and profit. Later on in 2007, when the Thai export sector was negatively affected by the Hamburger crisis, many trading companies were suffered and entrepreneurs thus earned lower income and profit. At the same time, workers were faced with lower income. The overall income inequality as measure by the Shorrocks index of order 2 thus decreased to 4.0132, which mostly came from a fall in income disparity within the group of elderly, despite a slight increase in between group inequality. Finally, when the government implemented numerous policies aiming to stimulate labors consumption and the overall economic performance, such policies did not turn out as successful as they were meant to be. The country experienced a high level of income inequality again in 2009 as measured by the Shorrocks index of 6.1786, of which 6.1307 were arose from differences in income among individuals in the same age group and 0.0479 were arose from differences in average income between age groups. Policies such as a 2,000 baht financial aid for low-income Thai nationals, a 500 baht monthly allowance for the elderly, and a cost of living reduction including electricity, water supply, and mass transportation, the elderly and low income workers,unfortunately. were introduced adargely lived in the rural area and hwho needed help the most l lower accessibility to these aids.
Besides, tax incentive policies such as a reduction in tax rates for community enterprises, tax refund for real estate consumers, aiming to help the business sector did not span to the population in all age groups. Correlation test between demographic change and between group inequalities Demographic change does not only raise the proportion of the elderly, but it also reduces the proportion of efficient workers in the labor force. When the number of the elderly increases, the number of people who have ability to earn higher income decrease. This makes the household income as well as the real income on average to decrease. Considering only those population age 15 years old and over who are not in the educational process, as can be seen in figure 4, the proportions of the middle working-age group, the old working- in particular had age group, and the elderly had been steadily increased. The old working-age group the highest population growth, increased from 19.86% in 1986 to 30.20% in 2009. Following by the elderly group, which contributed to 12.06% of the population age 15 years old and over in 1986 and went up to 20.88% in 2009. Likewise, the proportion of the middle working-agto 32.33% in 2009, but at a slower pace than other groups. The results group had been increased from 30.55% in 1986 show that since2007, the old working-age population and the elderly together were accounted for more than half of the population. On the contrary, a proportion of the young working-age population had been rapidly decreased from 37.71% in 1986 to16.59% in 2009. Consequently, these results indicate that Thailand has to face the problem of aging society where the majority of workers are gettingg older and older. Figure4 The proportion of population aged 15 years old and over by age group 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% 12.06 12.48 12.511 13.35 14.73 15.80 16.56 17.63 17.37 18.71 19.19 20.27 20.88 19.68 19.63 19.711 20.07 21.46 21.90 22.35 22.75 25.33 26.69 28.11 28.98 30.20 30.55 31.71 31.57 33.61 33.87 35.11 35.63 36.25 35.98 34.91 33.93 33.30 32.33 37.71 36.18 36.21 32.97 29.94 27.19 25.46 23.37 21.32 19.69 18.77 17.45 16.59 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 2006 2007 2009 15-29 Y. 30-44 Y. 45-59 Y. 60 Y. up Not only inequality within each age group, between-group inequality is also of interest. Even if its impact on overall income inequality is less pronounced, it demonstrates differences in ability to earn income and average income of each age group accordingly. When between-group inequality increases, this means some members in one group have lower ability to earn, which resulted in lower income.
25.00 20.00 15.00 10.00 5.00 0.00 Figure5 The relationship among the proportion of elderly group and between-group income inequality of the old and others 0.0348 0.0347 16.56 17.63 17.37 18.71 19.19 15.80 20.27 14.73 20 0.88 13.35 12.06 12.48 12.51 0.0277 0.0236 0.0208 0.0237 0.0230 0.0193 0.0175 0.0118 0.0088 0.0069 0.0041 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 2006 2007 2009 0.0400 0.0350 0.0300 0.0250 0.0200 0.0150 0.0100 0.0050 0.0000 Porportion of population age 60 years up Shorrocks Between group (Age 60 years up and others) Income of the elderly and population in other age groups are then analyzed. It is found that inequality between the elderly and other age groups, reflecting differences in average income among these two groups, had been increased. This result implies that the populations in other age groups have income higher than the elderly on average. Even if widening income inequality could perhaps be explained by other factors such as economic factors or social and cultural changes, it is found that between group inequality had been continuously increased with the number of the elderly as a proportion of the total population. Table 2Correlation test Correlation Between group inequality of the old and others Between group inequality of the old and others 1.000000 0.974613 1.000000 t-statistic Between group inequality of the old and others Probability Between group inequality of the old and others ----- 14.43702 ----- 0.0000 Between group inequality of the old and others ----- Between group inequality of the old and others -----
Table 3Granger causality test Null Hypothesis: Obs F-Statistic Prob Between group inequality of the old and others does not Granger cause 11 1.31435 0.3362 does not Granger cause Between group inequality of the old and others 6.04437 0.0365 The correlation test with a very high correlation coefficient of 0.9746 justifies the existing relationship between the proportion of the elderly and income inequality between the elderly group and other groups, suggesting that changes in the Thai population structure and increases in the proportion of the elderly strongly relate to a worsening income distribution. The Granger causality test is also employed to investigate the causal relationship between these two factors. It is revealed that an increasing number of elderly causes the between-group inequality to increase. 4. Conclusion Income inequality among the elderly population as measured by the Gini coefficient is relatively high compared with those in other age groups and tends to increase steadily. Therefore, in a situation where a larger amount of population is aging and no provision of additional income after retirement is sufficiently provided, income disparity is inevitably widen. The results from decomposing income disparity by using the Shorrocks index of order 2 indicate that differences in income among individuals in the same age group have a larger impact on overall income inequality than differences in average income between age groups. Such significant disparity in income arises from different in geographic, economic, and social characteristics. The relationship between the proportion of the elderly and inequality in income between this age group and others does exist with a relatively high correlation level, meaning that as Thailand transits to the aging society, income inequality will be deteriorated. Saving up more money for the old age, working after retirement, and buying private health insurance can help individuals to sustain their income or insure them against any losses in the future. In addition, the government should have proactive policies to distribute benefits of development benefits to the population in the country, especially to those helpless individuals in their later live. These actions together would alleviate the problem of income inequality in Thailand in the future. 5. References Boom, E. D. and Finaly, E. J. 2009. Demographic Change and Economic Growth in Asia. Asian Economic Policy Review 4, 1: 45-64. Boonyamanond, S. 2008. Interconnection between income distribution and economic growth: Crosscountry and Thai evidence. Ph.D. Dissertation. Bangkok: Chaulalongkorn University. Deolalikar, B. A. 2002. Povery, Growth and Inequality in Thailand. ERDWorking paper No.8. Asian Development Bank. Fofack, H. and Zeufack, A. 1999. Dynamics of income inequality in Thailand: Evidence from household pseudo-panel data. The World Bank.
Heerink, N. and Folmer, H. 1994. Income distribution and the fulfillment of basic needs: Theory and empirical evidence. Journal of Policy Modeling 6, 6 (December 1994): 625-652. Israngkura, A. 2003. Income inequality and university financing in Thailand. East Asian Development Network (EADN). Kanchanachitra, C. 2007. Thai Health 2007. Nakorn Pathom: Institute for Population and Social Research. [In Thai] Karoly, A. L. and Burtless, G. 1995. Demographic change, rising earnings inequality, and the distribution of personal well-being: 1959-1989. Demography 32, 3: 379-405. Kuznets, S. 1955. Economic growth and income inequality. American Economic Review 45, 1 (March): 1-28. Lerman, I. R. and Yitzhaki, S. 1985. Income inequality effects by income source: a new approach and applications to the United States. The review of Economics Statistics 67, 1 (February 1985): 151-156. Nkang, M. N. 2009. Demographic change, economic growth, income distribution, and poverty in Nigeria: A computable general equilibrium analysis. Poppov Research Network. Oyekale, A. S., Adeoti, A. I. and Ogunnupe, T. O. 2004. Sources of income inequality and poverty in rural and urban Nigeria. Nigeria: University of Ibadan, Department of Agricultural Economics. Sarntisart, I. 1997. Educational expansions and labor earnings inequality: The case of Thailand between 1988 and 1992. Chulalongkorn Journal of Economics 9, 2 (May):127-174. Sarntisart, Isra 2004.Growth, structural change, and inequality: The experience of Thailand. In Cornia, G. A. (ed.), Inequality, growth and poverty in an Era of Liberalization and Globalization,. 427-405New York: Oxford University Press.