NBER Reporter 2013 Number 1: Research Summary
Saving in Developing Countries
Eswar S. Prasad *
The evolution of national savings in developing countries (a broad term that I use here to refer to middle-income emerging markets, as well as less developed low-income economies) has received considerable attention in discussions of global current account imbalances. In the run-up to the global financial crisis, these imbalances were characterized by large and rising current account deficits in the United States, United Kingdom, and a few other advanced economies, matched by corresponding surpluses in many emerging markets and a few oil-exporting economies. Rising saving rates in China and many other Asian economies began to receive increased attention from researchers around this period, and Federal Reserve Chairman Ben Bernanke's 2005 speech arguing that the "savings glut" in emerging markets was a proximate cause of the imbalances gave further impetus to that research. 1
Economists have been more successful in explaining changes in saving rates within specific countries over time than in explaining differences in saving levels across countries. 2 The fact that Asian economies traditionally have had higher saving rates than developing and industrialized economies in other regions has received some attention, but there is no persuasive explanation for this phenomenon. Economists have had to rely on weak non-economic explanations, such as the argument that Asians are culturally predisposed towards saving. This hypothesis has been formally tested using data from the U.S. Census to examine whether immigrants to the United States from high-saving countries tend to save more than immigrants from low-saving countries. The results show that there are significant differences in immigrants' saving behavior by country of origin, but those differences do not match up with the differences in national saving rates. In particular, immigrants from high-saving Asian countries do not save more than other immigrants. 3
Saving in Asia
Given their high and rising saving rates, Asian economies have been the subject of considerable research. In an early contribution focusing on the region, Susan Collins looks at rising national saving rates in nine Asian developing economies (plus Turkey) over the period 1960-84. She concludes that high growth rates, a low dependency ratio, and high income levels are all positively associated with saving rates. She argues further that there are structural differences between low-income and middle-income countries in the determinants of savings. 4
Charles Horioka and Akiko Terada-Hagiwara find that domestic saving rates in developing Asia rose during the period 1966-2007. They conclude that the main determinants of those trends were the age structure of the population (especially the elderly dependency ratio), income levels, and the level of financial sector development. 5 They forecast that over the next two decades the domestic saving rate in developing Asia as a whole will remain roughly constant, despite rapid population aging in most of those economies, in part because the negative impact of population aging on the domestic saving rate will be largely offset by the positive impact of higher income levels.
National saving comprises saving by households, corporations, and the government. Household savings typically has attracted most of the attention of researchers because it is more amenable to theoretical modeling than the other components of nations saving, and because its determinants can be analyzed using household-level survey data. Corporate saving (retained earnings) has received less attention, but in fact has been the key driver behind the surge in national savings in major Asian emerging markets during the latter half of the last decade. 6
While household saving rates have also trended up in most major Asian economies, one prominent Asian economy where the household saving rate has fallen quite significantly over the last two decades is Korea. Young Jun Chun evaluates the effects of population aging and fiscal policies on national saving in Korea. 7 Using a life-cycle model that incorporates a generational accounting approach, he argues that rapid population aging and the long-term budgetary imbalance have and will continue to drive down the national saving rate in Korea.
The sheer scale of China's saving, which now exceeds 50 percent of GDP, has drawn considerable research attention. Dennis Yang, Junsen Zhang, and Shaojie Zhou look at determinants of all three components of saving in China and conclude that economic, demographic, and policy trends in the internal and external environments of the Chinese economy are likely to lead to a decline in national saving in the foreseeable future. 8
With greater access to household-level datasets, there has been an intense focus on explaining the rise in China's household saving rate. From 1995 to 2005, the average urban household saving rate in China rose steadily by 7 percentage points, to about one quarter of disposable income. The urban saving rate has continued to rise since then, driving the national household saving rate higher as well. Marcos Chamon and I use data from China's Urban Household Surveys to explain why households are postponing consumption despite rapid income growth. 9 Tracing cohorts over time indicates a virtual absence of consumption smoothing over the life cycle. Saving rates have increased across all demographic groups, although the age profile of savings has an unusual pattern in recent years, with younger and older households having relatively high saving rates. We argue that these patterns are best explained by the rising private burden of expenditures on housing, education, and health care. These effects and precautionary motives may have been amplified by financial underdevelopment, as reflected in constraints on borrowing against future income and low returns on financial assets.
In subsequent work, Chamon, Kai Liu, and I examine the role of precautionary saving motives in explaining both the increase in China's household saving rate since the mid-1990s and the interesting fact that the age-savings profile has become U-shaped during the 2000s. 10We find that, in addition to the factors identified in our earlier research, rising income uncertainty and pension reforms help to explain both of these phenomena. Using a panel of Chinese households covering the period 1989-2006, we document that strong average income growth has been accompanied by a substantial increase in income uncertainty. Interestingly, the permanent variance of household income remains stable while it is the transitory variance that rises sharply. A calibration of a buffer-stock savings model indicates that rising savings rates among younger households are consistent with rising income uncertainty and that higher saving rates among older households are consistent with a decline in the pension replacement ratio for those retiring after 1997. We conclude that rising income uncertainty and pension reforms can explain more than half of the increase in the urban household savings rate in China since the mid-1990s, as well as the U-shaped age-saving profile.
Other researchers have used less disaggregated data to provide complementary perspectives on household saving behavior. Horioka and Junmin Wan conduct a dynamic panel analysis of the determinants of the household saving rate in China using a life-cycle model and panel data on Chinese provinces for the period 1995-2004. 11 They find that the main determinants of variations over time and over space are the lagged saving rate, the income growth rate, (in many cases) the real interest rate, and (in some cases) the inflation rate. They find little evidence that variables relating to the age structure of the population have the expected impact on the household saving rate. Their results provide mixed support for the life-cycle hypothesis and the permanent income hypothesis, and are consistent with the existence of inertia or persistence in household saving behavior.
Other research on China has emphasized demographic factors as one of the main determinants of the rising household saving rate. Chadwick Curtis, Steven Lugauer, and Nelson Mark undertake a quantitative investigation using an overlapping-generations model. 12 In their model, dependent children's utility enters into parents utility so that parents choose the consumption level of the young until they leave the household. Working agents give a portion of their labor income to their retired parents and save for their own retirement, while the aged live on their accumulated assets and on support from their children. These researchers take future demographic changes, labor income, and interest rates as exogenously given. They argue that their calibrated model accounts for much of observed increase in the household saving rate from 1963 to 2009.
While evidence of conventional demographic factors, such as an aging population, in driving household saving rates has been mixed, there are other aspects of changing demographics in China that have been the subject of research as well. Shan-Jin Wei and Xiaobo Zhang propose a novel and unorthodox explanation based on competitive saving resulting from unbalanced sex ratios (tilted in favor of males) in China. 13 As the sex ratio rises, Chinese parents with a son raise their savings in a competitive manner in order to improve their son's relative attractiveness for marriage. The pressure on savings spills over to other households. Both cross-regional and household-level evidence supports this hypothesis. They conclude that this motive potentially can account for about half of the actual increase in the household savings rate during 1990-2007.
Abhijit Banerjee, Xin Meng, and Nancy Qian exploit the changes in China's demographics caused by its family planning policies to study the effects of changes in the demographic structure on savings and wealth. 14 They find that children provide a substantial amount of support for elderly parents and that sons provide more support than daughters. Their empirical estimates support the predictions of a simple life-cycle model, based on which they conclude that the exogenous reduction in fertility because of family planning policy caused a significant increase in household savings, and that all of the increase is driven by parents who have a daughter as their only child.
As in other Asian economies, corporate saving was a principal driver of the rising national saving rate in China. 15 During 2003-7, the share of household saving in GDP actually declined, even though the household saving rate (saving as a share of disposable income) continued to rise. This apparent anomaly is the consequence of a greater share of national income going to capital than to labor. If households effectively own the firms in an economy, either directly or indirectly, this should not matter because firms' profits will increase household disposable income. However, in China, a majority of firms are still state-owned and most of them don't pay dividends to the state.
China's high corporate saving rate has received attention in policy circles, but has been the subject of only limited research so far. Tamim Bayoumi, Hui Tong, and Wei examine firm-level data and conclude that it indicates a global trend of rising corporate saving over the period 2002-7. 16 Chinese state-owned firms only recently were required to pay out dividends to the state, and these payments are still quite low relative to profits. However, these authors conclude that there is no significant difference in the savings behavior and dividend patterns between Chinese majority state-owned and private listed firms. Other evidence reported by Loukas Karabarbounis and Brent Neiman suggests that China is not special and that declining labor shares and the rise of corporate saving are global phenomena. 17 One factor behind these phenomena is the global decline in the cost of capital beginning in the 1980s, which has led firms around the world to shift away from labor and towards capital, financed in part with an increase in corporate saving.
More Work Ahead
With developing economies playing an increasingly important role in the global economy, there is growing interest in explaining saving behavior in these economies from both micro and macro perspectives. Increase in access to household and firm-level datasets in China and other developing economies has set off an exciting research program, although a number of questions have not yet been conclusively answered. For instance, the micro evidence suggests a range of plausible explanations for the rise in China's household saving rate, although there is no easy way to distinguish among these different hypotheses in a unified framework. Integrating the micro and macro perspectives to explain the determinants of saving-investment balances in these countries is likely to remain a fertile area of research.
* Eswar Prasad is a Research Associate in the NBER's International Finance and Macroeconomics Program and the Tolani Senior Professor of Trade Policy and a Professor of Economics at Cornell University.
1. U B. Bernanke, "The Global Saving Glut and the U.S. Current Account Deficit," Sandridge Lecture, Virginia Association of Economists, Richmond, VA, March 10, 2005.
2. See S. Edwards, "Why are Saving Rates so Different Across Countries? An International Comparative Analysis," NBER Working Paper No. 5097, April 1995, and Journal of Development Economics, Vol. 51, no. 1(October 1996), pp. 5-44. That paper also has a good survey of the literature that preceded it.
3. C.D. Carroll, B. Rhee, and C. Rhee, "Does Cultural Origin Affect Saving Behavior? Evidence from Immigrants," NBER Working Paper No. 6568, May 1998, and Economic Development and Cultural Change, Vol. 48, no. 1 (1999), pp. 33-50.
4. S.M. Collins, "Saving Behavior in Ten Developing Countries," in National Saving and Economic Performance, B. D. Bernheim and J. B. Shoven, eds. January 1991, pp. 349 - 76. Also see S. Edwards, "Why are Saving Rates So Different Across Countries? An International Comparative Analysis," NBER Working Paper No. 5097, April 1995, and Journal of Development Economics, Vol. 51, no. 1(October 1996), pp. 5-44.
5. C.Y. Horioka and A. Terada-Hagiwara, "The Determinants and Long-term Projections of Saving Rates in Developing Asia," NBER Working Paper No. 17581, November 2011, and in Japan and the World Economy, vol. 24, no. 2 (March 2012), pp. 128-137.
8. D.T. Yang, J. Zhang, and S. Zhou, "Why Are Saving Rates so High in China?," NBER Working Paper No. 16771, February 2011, and in Capitalizing China, J. Fan and R. Morck, eds. forthcoming from the University of Chicago Press.
9. M. Chamon and E.S. Prasad, "Why Are Saving Rates of Urban Households in China Rising?" NBER Working Paper No. 14546, December 2008, and American Economic Journal: Macroeconomics, January 2010, Vol 2(1), pp.93-130.
11. C.Y. Horioka and J. Wan, "The Determinants of Household Saving in China: A Dynamic Panel Analysis of Provincial Data," NBER Working Paper No. 12723, December 2006, and Journal of Money, Credit and Banking, vol. 39, no. 8 (December 2007), pp. 2077-96.
13. S. Wei and X. Zhang, "The Competitive Saving Motive: Evidence from Rising Sex Ratios and Savings Rates in China," NBER Working Paper No. 15093, June 2009, and Journal of Political Economy, June 2011.
14. A. Banerjee, X. Meng, and N. Qian, "The Life Cycle Model and Household Savings: Micro Evidence from Urban China," forthcoming as an NBER Working Paper.
15. E.S. Prasad, 2011 (op. cit. endnote 7)