Crises are rarely good news for anyone, but they can be devastating for poor people who are rarely able to hedge in advance or borrow to protect their living standards. In financial crises it is more often the rich who receive compensation as bank bailouts occur. Moreover, the poor suffer more as financial crises spread to the real economy. At times of crisis, wealth transfers take place not only between rich and poor, but also between domestic and foreign investors (and between investors with and without access to foreign financial systems). And redistribution takes place between uninformed and informed investors.
It is important to distinguish absolute impacts from relative impacts. Absolute poverty is very likely to rise in a crisis, though how much so will depend on the extent of the aggregate economic contraction and the rise in inequality. There is little or no clear pattern in how past aggregate economic contractions affect relative inequality within countries—the ratios of incomes, such as between the richest 10% and the poorest 10%. Indeed, one of the stylized facts about growth and distributional change in developing countries is that there is little or no correlation between changes in inequality and economic growth. In growing economies, relative inequality falls about as often as it rises during periods of rising average income, though absolute inequality—the absolute gap between “rich” and “poor”—tends to rise. Similarly, in contracting economies, relative inequality increases about as often as it falls, although unusually large contractions in mean household income are associated more often with rising inequality. Absolute inequality tends to fall when economies contract.
It is also important to recognize that an aggregate poverty or inequality measure does not tell the whole story when attempting to understand the social impacts of crises. There are likely to be both gainers and losers at any level of living, including among the poor. And there may well be adverse impacts on important non-income dimensions of welfare, including the nutrition and schooling of children.
Diverse welfare impacts can be expected
Initial conditions in specific developing countries and their domestic policy responses to the crisis will jointly determine how its impact is distributed. More export-oriented economies tend to suffer less from homegrown recessions but are likely to suffer more from recessions stemming from abroad, like the current one; the more open the country, the worse the hit. The extent of trade diversification will probably also influence how vulnerable an economy will be to external shocks. Ironically, the poor living in high-inequality countries will tend to be better protected from an aggregate economic contraction than those living in low-inequality countries.
An aggregate shock at the country level will have heterogeneous impacts across households, depending on (inter alia) household demographics, education attainments and location. The largest proportionate income losses need not be among the poorest initially, some of whom will be protected by the same factors that have kept them poor, namely geographic isolation and consequently poor connectivity with national and global markets. However, cumulative effects stemming from nonlinearities and multiple equilibria (whereby similar shocks to two identical households can have very different long-run outcomes) can mean that even a middle-income household falls into destitution after a sufficiently large shock.
Research on the poverty impacts of Indonesia’s severe economy-wide crisis of 1998 found sharp but geographically uneven increases in the poverty rate, reflecting both the unevenness in the extent of the economic contraction and the differing initial conditions at the local level. Proportionate impacts on the poverty rate were greater in initially better-off and less unequal districts of Indonesia. Another study of the welfare impacts of the same crisis found that most households were impacted, but that it was the urban poor who fared the worst; the ability of poor rural households to produce food mitigated the worst consequences of the high inflation. By contrast, the rural poor bore a heavier burden of the macroeconomic shock in Thailand around the same time, in part because of their greater integration with the urban economy than was the case in Indonesia.
Drawing on thirteen rounds of annual labor force data in conjunction with two waves of a household panel, another study examined the impact of the financial crisis in Indonesia on labor market outcomes. The study found that fears of a large collapse in employment rates as a result of the crisis were unfounded and that there was only a modest increase in unemployment. However, real wages in Indonesia did decline markedly as a result of the crisis, with the largest declines observable in the urban sector. Earnings from self-employment were also found to decline significantly for women and in the urban sector. But earnings from self-employment among men in rural areas remained stable. As a result, given the importance of self-employment in the Indonesian labor force, conclusions of the effect of the crisis on wages that focus only on the market wage sector are likely to overstate the magnitude of the crisis. There are many ways in which Indonesian families manages to protect family incomes despite declines in wages, although such strategies were most effective among richer households. Poor people in Indonesia bear the brunt of any deleterious medium and longer term effects of the crisis.
The financial crisis in Argentina was also found to have had a dramatic effect on real income of workers and households, with 63% of urban households experiencing real income falls of 20% or more between October 2001 and October 2002. In general, households were not found to offset falling real wages by sending more members to the labor market and working more hours: instead, total household labor hours per week declined on average for all quintiles. In contrast with the evidence found in Indonesia, research in Argentina found that self-employment did not play much of a role in allowing households to mitigate the effects of the crisis.
Following a period of stormy economic and political reforms in the 1970s and 1980s, inequality in Chile stabilized and remained relatively unchanged. The structural reforms of the Chilean economy, which started in 1974 and were largely completed by the late 1980s, had important implications for, among other things, the regulation of labor markets. The reforms included trade liberalization, privatization of state-owned assets, deregulation of various markets, and reforms in the structure of taxes, subsidies, and benefits. The impact of these changes on income distribution was expected to be profound – and to result generally in a worsening of inequality. However, Chilean inequality remained remarkably stable, albeit high, during most of the period 1987-94.
The period 1974-94 in Brazil was characterized by persistent macroeconomic disequilibrium, the main symptoms of which were stubbornly high and accelerating inflation and a GDP time series marked by unusual volatility and very low positive trend. In addition, substantial structural changes were taking place in Brazilian society, with rapid population growth, high urbanization rates, significant improvements in education outcomes, and also growing open unemployment. Casual observation of Brazilian income data suggested, however, that between 1976 and 1996 little changed in the Brazilian urban income distribution. While measured inequality declined slightly between 1976 and 1996, extreme poverty increased significantly. The existence of a group excluded both from the productive labor markets and from any substantive form of safety net was identified, and points to self-targeted labor programs or other safety nets for policy implications. Considerable effort among the urban population in Brazil to maintain living standards in the face of falling returns in the formal labor market and in self-employment was needed.
In another study using household panel data we examined the impacts of Russia’s financial crisis in 1998 and the response of the public safety net. The response of the public safety net helped reduce the impact of that crisis. The research found that the incidence of income poverty would have been two percentage points higher without the changes in the safety net. The same study also found that protecting public spending on the safety net to its pre-crisis level could have prevented the rise in poverty.
In the wake of the 1997 financial crisis in Asia, we undertook a study on the impact of the crisis on agricultural households in Indonesia and Thailand. Using detailed household-level survey data collected before and after the crisis began, the research concluded that although the natures of the shocks in the two countries were similar, the impact on farmers' income (particularly on distribution) was quite different. In Thailand, poor farmers bore the brunt of the crisis, in part because of their greater reliance on the urban economy (through seasonal off-farm work), than did poor farmers in Indonesia. Urban-rural links are much weaker in Indonesia, and for that reason poorer farmers there were more insulated from the ramifications of the crisis (by the same token, they did not share as much in the fruits of the earlier boom years). Farmers in both countries, particularly those specializing in export crops, benefited from the currency devaluation. Although there is some evidence that the productivity of the smallest landholders declined over the period in question, it is difficult to attribute this directly to the financial crisis. At least in Thailand, a rural credit crunch does not seem to have materialized. In Thailand, the view that the smallholder sector can serve as a safety net that will absorb poor urban workers who are of rural origin was proven wrong, as the smallholder sector could not have high return for the extra labor from returning migrants.
It can be hard to predict the welfare consequences of crises
In one study we developed a top-down macro-micro model of the Brazilian economy, and used this model to investigate the link between macroeconomic shocks and the distributions of employment, earnings and household incomes. The study examined to what extent counterfactual distributions generated by the model for 1999 compared with the distributions actually observed in 1999. While the shock was expected to have a negative impact, the massive devaluation did not result in a collapse of the financial sector with attendant devastating effects on the credit market and real economy. Even so, incomes fell and poverty rose. Unemployment also rose across the board, but predominantly in urban areas and for more skilled workers. The predictive performance of the top-down macro-micro model was mixed. The model made a number of mistakes even in the direction of employment changes, and the errors were often large. All in all, the study is cautious in suggesting that this approach can predict distributional outcomes of macroeconomic shocks or policy packages with great accuracy. But it suggests that these tools do hold promise.
Heterogeneous impacts can also be expected on human development
Assessments of the effects of aggregate economic shocks have often presumed that these crises would have a negative impact on education and health outcomes. Empirical findings reveal no such simple regularity. Some recessions have led to reductions in school enrollment, as in Costa Rica in 1981-82, while others have led to increases, as in the United States during the Great Depression. Similarly, negative covariate shocks in Zimbabwe associated with the 1982-84 drought led to persistent losses in height-for-age for young children, while infant mortality declined in the United States during the Great Depression.
What explains this variation across countries? In predicting the direction of the short-term impact of economic crises, note that aggregate economic shocks have income and substitution effects which, in the absence of policy interventions, determine whether schooling and health outcomes would improve or deteriorate. A contraction of GDP has a negative income effect which, all else being equal, can lead to lower investments in education and health. In the case of schooling, the substitution effect derives from a decline in child wages due to decreased labor demand during such a contraction. This has the effect of reducing the opportunity cost of schooling, thereby leading to higher school enrollment. In the case of health, the substitution effect arises because the decline in labor demand reduces adult wages which, in turn, frees up parents’ time for time-intensive health-promoting activities (collecting clean water, cooking better meals, or taking children to health check-ups). The aggregate effect of a crisis on education and health outcomes cannot therefore be signed ex ante—it is an empirical matter.
In practice, recessions in developed countries are generally associated with better health and education outcomes. In the United States, economic downturns have generally decreased infant mortality. In the countries of the former Soviet Union, declining income was associated with dramatic increases in adult mortality, particularly from alcoholism and suicide, but there was no obvious change in child health. This pattern implies that the substitution effect dominates the income effect.
In the poorest developing countries, however, both health and education outcomes deteriorate during economic crises, evidence that the income effect dominates. This pattern is consistent with evidence from Côte d’Ivoire, Ethiopia, Malawi, Tanzania, Zimbabwe and (less clearly) Indonesia, as well as (for health) India. In middle-income countries, health outcomes deteriorate during crises (as found in Mexico, Peru, Nicaragua), while schooling is unaffected or improves (as found in Brazil, Mexico, Nicaragua, and Peru).
For example, in Peru, a middle-income country with high levels of school enrollment, the deep economic crisis in the late 1980s could have decreased schooling outcomes because public expenditures and household incomes fell. However, these reductions seem to have been offset by the lower opportunity cost of attending school. Hence, although public education spending fell by almost 50 percent, children were more likely to be enrolled and less likely to be working during the crisis than in other years. But time series household data show that the infant mortality increased by 2.5 percentage points during the crisis.
In Indonesia the 1997 financial crisis decreased enrollment rates among children aged 8-13 and increased enrollment rates among children aged 14-19, but these changes were small, just one percentage point of enrollment.
We have studied the relationship between log per-capita GDP and infant mortality using data births and deaths from 123 Demographic and Health Surveys (DHS) covering 59 countries conducted between 1986 and 2004. A 1-percent contraction in per-capita GDP was associated with an increase in infant mortality of between 0.18 and 0.44 deaths per thousand children born. This finding means that there were about one million excess deaths during the period 1980-2004 in countries experiencing large economic contractions (10 percent or greater).
Psychological well-being is also affected
The financial crisis in Indonesia adversely impacted psychological well-being. Several different dimensions of psychological distress were evident among male and female adults across the entire age distribution. In addition, the imprint of the crisis can be seen in its differential impacts on low education groups, the rural landless, and residents in those provinces that were hit hardest by the crisis. Elevated levels of psychological distress persist even after indicators of economic well-being such as household consumption had returned to pre-crisis levels, suggesting long-term deleterious effects of the crisis on the psychological well-being of the Indonesian population.
Drawing on data from such surveys in Bosnia, Herzegovina, Indonesia, India, Mexico and Tonga, we have found that individuals who are older, female, widowed, and reporting poor physical health are more likely to report worse mental health outcomes. Individuals living with others in poor mental health are also significantly more likely to report worse mental health themselves. Economic and multi-dimensional shocks such as illness or crisis can have a greater impact on mental health than overall levels of poverty. This may have important implications for social protection policy. Mental health modules can be meaningfully added to multi-purpose household surveys in developing countries.
Crises can have severe long-term consequences for human development
In the short run, households might try to smooth their consumption levels by increasing their labor supply and drawing down their savings during the crisis. But when work opportunities are scarce and families have no access to credit markets, they may have to reduce food intake, even for very young children, or pull out children from school. Evidence from past crises show that children who experience short-term nutritional deprivations can suffer long-lasting effects. Civil war and drought in Zimbabwe seriously affected the lives of infants, with effects carrying over to later ages. Panel data indicate that these children had significantly lower height during adolescence, delayed school enrollment, and reduced grade completion equivalent to a 7 percent loss in lifetime earnings for them. In Ethiopian communities experiencing crop damage during a sustained drought, overall child growth was lower, especially among children who were less than two years of age during the drought.
Contact: Martin Ravallion, firstname.lastname@example.org, 202-473-6859
Most research documents cited in this summary are available through the World Bank’s research archives at http://econ.worldbank.org/docsearch or the Bankwide archives at http://www-wds.worldbank.org/.
1. M. Halac and S. Schmukler. 2004. “Distributional Effects of Crises: The Financial Channel.” Economia 5(1): 1–67.
2. J. Frankel and S. Schmukler. 1996. “Country Fund Discounts and the Mexican Crisis of December 1994: Did Local Residents Turn Pessimistic Before International Investors?” Open Economies Review 7: 511–34.
J. Frankel and S. Schmukler. 1998. “Crisis, Contagion, and Country Funds.” In Managing Capital Flows and Exchange Rates: Perspectives from the Pacific Basin, ed. R. Glick. Cambridge University Press.
D. Kaufmann, G. Mehrez, and S. Schmukler. 2005. “Predicting Currency Fluctuations and Crises: Do Resident Firms Have an Informational Advantage?” Journal of International Money and Finance 24(6): 1012–29.
See E. Levy Yeyati and S. Schmukler, and N. van Horen. 2008. “Emerging Market Liquidity and Crises.” Journal of the European Economic Association 6(2-3): 668-628.
3. F. Ferreira and M. Ravallion. 2008. “Poverty and Inequality: The Global Context.” In The Oxford Handbook of Economic Inequality, ed. W. Salverda, B. Nolan, and T. Smeeding. Oxford University Press.
4. M. Ravallion. 2004. “Competing Concepts of Inequality in the Globalization Debate.” Brookings Trade Forum: 1–38.
5. See Figure 1 in M. Ravallion. 2007. “Inequality is Bad for the Poor.” In Inequality and Poverty Re-Examined, ed. J. Micklewright and S. Jenkins. Oxford University Press.
6. N. V. Loayza and C. Raddatz. 2006. “The Structural Determinants of External Vulnerability.” Policy Research Working Paper 4089, World Bank, Washington, DC.
7. M. Ravallion. 2007. “Inequality is Bad for the Poor.” In Inequality and Poverty Re-Examined, ed. J. Micklewright and S. Jenkins. Oxford: Oxford University Press.
M. Ravallion. 1997. “Can High-Inequality Developing Countries Escape Absolute Poverty?” Economics Letters 56: 51–57.
8. M. Ravallion and M. Lokshin. 2007. “Lasting Impacts of Indonesia’s Financial Crisis.” Economic Development and Cultural Change 56(1): 27–56.
9. J. Friedman and J. Levinsohn. 2002. “The Distributional Impact of Indonesia’s Financial Crisis on Household Welfare.” World Bank Economic Review 16(3): 397–424.
10. F. Bresciani, D. Gilligan, G. Feder, T. Onchan, and H. Jacoby. 2002. “Weathering the Storm: The Impact of the East Asian Crisis on Farm Households in Indonesia and Thailand.” World Bank Research Observer 17(1): 1–20.
11. J. Smith, D. Thomas, E. Frankenberg, K. Beegle, and G. Teruel. 2002. “Wages, Employment and Economic Shocks: Evidence from Indonesia.” Journal of Population Economics 15: 161–93.
12. D. McKenzie. 2004. “Aggregate Shocks and Urban Labor Market Responses: Evidence from Argentina’s Financial Crisis.” Economic Development and Cultural Change 52(4): 719–58.
13. F. Ferreira and J. Litchfield. 1999. “Calm after the Storms: Income Distribution and Welfare in Chile, 1987-1994.” World Bank Economic Review 13(3): 509–38.
14. F. Ferreira and R. Paes de Barros. 2005. “The Slippery Slope: Explaining the Increase in Extreme Poverty in Urban Brazil: 1976-1996.” In The Microeconomics of Income Distribution Dynamics in East Asia and Latin America, ed. F. Bourguignon, F. Ferreira, and N. Lustig. Oxford University Press and the World Bank.
15. M. Lokshin and M. Ravallion. 2000. “Welfare Impacts of Russia’s 1998 Financial Crisis and the Response of the Public Safety Net.” Economics of Transition 8(2): 269–95.
16. F. Bresciani, D. O. Gilligan, G. Feder, O. Tongroj, and H. Jacoby. 2002. “Weathering the Storm: The Impact of the East Asian Crisis on Farm Households in Indonesia and Thailand.” World Bank Research Observer 17(1): 1–20.
17. F. Ferreira, P. Leite, L. Perreira da Silva, and P. Picchetti. 2008. “Can the Distributional Impacts of Macroeconomic Shocks be Predicted? A Comparison of Top-Down Macro-Micro Models with Historical Data for Brazil.” In The Impact of Macroeconomic Policies on Poverty and Income Distribution, ed. F. Bourguignon, M. Bussolo, and L. Perreira da Silva. Palgrave MacMillan and World Bank.
18. C. Ruhm. 2000. “Are Recessions Good for Your Health?” Quarterly Journal of Economics 115(2): 617–50.
19. E. Brainerd. 1998. “Market Reform and Mortality in Transition Economies.” World Development 26(11): 2013–27.
E. Brainerd. 2001. “Economic Reform and Mortality in the Former Soviet Union: A Study of the Suicide Epidemic in the 1990s." European Economic Review 45(4–6): 1007–19.
20. F. Ferreira and N. Schady. 2008. “Aggregate Economic Shocks: Child Schooling and Child Health.” Policy Research Working Paper 4701, World Bank, Washington, DC.
21. N. Schady. 2004. “Do Macroeconomic Crises Always Slow Human Capital Accumulation?” World Bank Economic Review 18(2): 131–54.
22. D. Thomas, K. Beegle, E. Frankenberg, B. Sikoki, J. Strauss, and G. Teruel. 2004. “Education in a Crisis.” Journal of Development Economics 74(1): 53–85.
23. S. Baird, J. Friedman, and N. Schady. 2007. “Aggregate Income Shocks and Infant Mortality in the Developing World.” Policy Research Working Paper 4346, World Bank, Washington, DC.
24. J. Friedman and D. Thomas. 2007. “Psychological Health Before, During and After an Economic Crisis: Results from Indonesia, 1993-2003.” Policy Research Working Paper 4386, World Bank, Washington, DC.
25. J. Das, Q. Do, J. Friedman, and D. McKenzie. 2008. “Mental Health Patterns and Consequences: Results from Survey data in Five Developing Countries.” Policy Research Working Paper 4495, World Bank, Washington, DC.
26. H. Alderman, J. Hoddinott, and B. Kinsey. 2006. “Long-Term Consequences of Early Childhood Malnutrition.” Oxford Economic Papers 58(3): 450–74.
27. T. Yamano, H. Alderman, and L. Christiaensen. 2005. “Child Growth, Shocks, and Food Aid in Rural Ethiopia.” American Journal of Agricultural Economics 87: 273–88.