رشد بهره وری، افزایش نابرابری درآمد و بیمه های اجتماعی : مورد چین ؟
|کد مقاله||سال انتشار||مقاله انگلیسی||ترجمه فارسی||تعداد کلمات|
|7308||2001||14 صفحه PDF||سفارش دهید||5082 کلمه|
Publisher : Elsevier - Science Direct (الزویر - ساینس دایرکت)
Journal : Journal of Economic Behavior & Organization, Volume 46, Issue 4, December 2001, Pages 395–408
This paper builds a simple dynamic general equilibrium model to mimic two striking stylized facts observed in China’s reform: productivity growth contributes significantly to output growth, and income inequality increases dramatically over time. Calibration exercises broadly matches the data. The economic growth rate, the aggregate productivity and income inequality increase as the coverage of the social insurance decreases. Perfect insurance is shown to be sub-optimal. With empirically plausible parameters, the level of income inequality under different degrees of social insurance can be very similar, even though their welfare implications are not. Social insurance cannot effectively reduce inequalities.
This paper studies two striking stylized facts about China’s reform: that the output growth in the early stage of the reform is mainly due to productivity growth, and that the income inequality is increasing over time.1 The first point can be illustrated by the agricultural sector. According to Gelb, Jefferson and Singh (1993, p.109), “almost half of the 42.2 percent growth of output in the cropping sector in 1978–1984 was driven by productivity change caused by reforms. Specifically, almost all of the productivity change was attributable to the changes resulting from the introduction of the household responsibility system”, a scheme that essentially eliminated social insurance and privatized the agricultural sector (Table 1). Note that the average productivity growth rate in the agricultural sector before the reform is negative.2,3 The significant positive productivity growth is also in sharp contrast with most of the East Asian countries.4 Thus, it seems interesting to investigate how China could be an exception. Another unusual fact about China is the continuing increase in income inequality. For instance, Li et al. (1997) examines the Gini coefficient for 49 developed and developing countries covering the period 1947–1994. They find that “90 percent of the total variance in the Gini coefficients can be explained by variation across countries, while only a small percentage of the total variance is due to variation over time…. On the other hand, seven of the countries in the sample have annual changes in excess of 1.0 percent … in China the index was increasing during our sample period at a rate of 3 percent a year, the largest rate of change that we observed.” This finding seems to be very robust.5 This paper is a preliminary attempt in understanding these two facts in an unifying framework. Roughly speaking, it embeds a static moral hazard problem into the AK model developed in Rebelo (1991). Labor effort is unobservable6 and generates disutility. Under complete social insurance regime (“communism”), agents exert a low level of effort and hence the productivity is low and the difference across agents’ wealth is small. Switching from the “communism” regime to a market economy, the income inequality, the “aggregate productivity”, and the economic growth rate will increase even without any alternation in the underlying technology. The idea is simple. Agents operate stochastic production technology individually. Only increases in effort can increase the probability of good outcome. A cut in the insurance coverage induces the agents to put more efforts and hence increases the expected rate of return in the aggregate.7 Since some agents receive higher productivity shocks, the income inequality increases. In the model, the increase in “aggregate productivity” is a mis-measurement as the unobservable labor efforts are not taken into account. Clearly, the original perfect insurance regime is sub-optimal. With realistic parameter values, calibration exercises estimate that the optimal coverage lies in between 10 and 25 percent. Imperfect insurance coverage ex ante inevitably leads to an income inequality ex post. Therefore, this model predicts both productivity growth and rise in income inequality, as China experienced. This paper also finds that a “large” difference in the insurance coverage may generate “small” difference in income inequality. It suggests that social insurance per se might not be an effective policy instrument for inequality reduction. This paper is not the first attempt to understand the transition economies within a dynamic general equilibrium framework.8 The focus of this paper is different from theirs. This paper attempts to investigate how the change of social insurance can simultaneously explain both the growth of “measured aggregate productivity” and the income inequality.9 For tractability, this paper sacrifices the investigation of other issues. The organization of this paper is simple. After presenting the basic model, it will be shown that departing from communism will increase the “productivity”, economic growth rate and income inequality. In addition, some calibration results will be presented. The paper will then conclude.10
نتیجه گیری انگلیسی
According to the conventional wisdom, social insurance would erase the incentive to work. Therefore, transiting from communism to market economy may rejuvenate the enthusiasm of work. Recent experience in China is consistent with this view. This paper suppresses many important elements of China’s success and highlights the importance of social insurance in an endogenous growth model. On the positive side, it is found that both economic growth rate and income inequality increase when the extent of social insurance is decreased. With reasonable parameter values, the model here matches the magnitude of productivity growth, economic growth and change in Gini coefficients observed in China. Interestingly, it is also found that a “big” difference in the extent of risk-sharing might have “small” implications to the income inequality. This may suggest that while social insurance can improve the welfare, it might not be effective in eliminating the income inequality.28 On the normative side, it is found that perfect insurance cannot be the optimal arrangement. Numerical simulations show that imperfect insurance regime can be superior to the one without. A coverage of 10 to 25 percent might be close to the optimal level. It may suggest that there can be some merit for social insurance even with moral hazard.29 For further investigations, many assumptions here need to be relaxed. For instance, the degree of capital market imperfectness can be very important in determining the form of wage contracts and should be endogenized. A discrete occupational choice (such as entrepreneurs versus employees) might also change the evolution of the income distribution. The possibility of bankruptcy should be allowed and the notion of “risk aversion” should be taken more seriously.30 Future research should also allow the possibility for agents to switch to technologies with lower risk and yet lower return as the social insurance coverage is decreased.31 With income inequality, the possibility of “jealousy” should also be considered seriously in computing the “social optimum.”32 International trade is another potentially important factor suppressed in this paper. Extensions to include different government policies such as pension and unemployment insurance would also be interesting.33 Future research along these directions should be encouraged.