14Y& ( Human Resources Development and Operations Policy HRO Working Papers Swimming Against the Tide: Strategies for Improving Equity in Health Nancy Birdsall Robert Hecht May 1995 HROWP 55 Papers in this series are not formal publications of the World Bank. They present preliminary and unpolished results of analysis that are circulated to encourage discussion and comment, citation and the use of such a paper should take account of its provisional character. The findings, interpretations, and conclusions expressed in this paper are entirely those of the author(s) and should not be attributed in any manner to the World Bank, to its affiliated organizations, or to members of its Board of Executive Directors or the countries they represent. Swimming Against the Tide: Strategies for Improving Eqiiity in Health by Nancy Birdsall Robert Hecht Nancy Birdsall is Executive Vice President, Inter-American Development Bank. Robert Hecht is a Principal Economist at The World Bank. The views expressed here do not necessarily reflect those of either the Inter-American Bank or The World Bank nor of their members. Please do not quote without authors' permission. Abstract The evidence shows that government spending for health in many developing countries benefits the well-to-do more than the poor. However, a combination of favorable political forces and sound public policies can shift the focus of government expenditures toward the poor. Doing this is an essential part of any effective poverty reduction program in developing countries. The better-off gain disproportionately in part because most public spending goes toward curative, high-cost hospital care in urban areas, where income is usually above average and where the wealthy have better access to care. In addition, health care subsidies for the middle classes, in the form of tax relief and government transfers to social security-based health insurance, skew total public expenditure further toward the better-off. This inequity can be understood in the context of "public choice" theory: politicians and bureaucrats seek not to optimize economic efficiency, but instead to maximize their chances of getting reelected or staying employed. This paper suggests that to implement health policies that favor the poor, it is necessary to "swim against the tide." Costa Rica, Korea, Malaysia and Zimbabwe provide examples of success stories with lessons for other countries. High and sustained economic growth has set the stage for redistributive health spending. So have pro-poor political forces. In Zimbabwe, for example, these forces were represented in post-independence groups eager to redress earlier discrimination. In Korea, it was a modernizing authoritarian regime in search of greater legitimacy. In political environments like these, pro-equity governments have reduced, eliminated or avoided health care subsidies that blatantly or covertly favored the better-off. Public spending has been targeted toward initiatives serving primarily the poor, such as rural health facilities and basic primary and preventive care. In middle-income countries like Costa Rica, Korea and, to a certain extent, Chile, increased public spending has universalized health insurance. In these circumstances, general tax revenues have come to subsidize insurance for the poor more than for the middle class. The authors argue that donor involvement can help to implement policies and programs to improve equity in health. They also advocate expanded donor backing for applied research to measure existing inequalities and to analyze the political dynamics behind the distribution of public resources for health. I Contents I. Introduction ............................................. . 1 II. Measuring Equity in Health ................................... 2 III. Evidence of Inequity: Public Expenditures and Physical Access .5.... . . . . . . . . S IV. The Politics of Inequity .................................... . 10 V. Strategies for Reform ...................................... . 13 Macroeconomic and Political Determinants ..... .................. . 13 Instruments for Promoting Equity ............................ . 14 Bibliography .............................................. . 18 I. Introduction A major objective of developing country governments and of donor agencies, stated repeatedly in their policy documents and speeches, is achieving greater equity in health. Chile's constitution, for example, states that the government has an obligation to "protect free and egalitarian access to actions that promote, protect, restore health and rehabilitate the health status of individuals" (Ministry of Health, 1992). The World Health Organization's most recent report on implementation of "Health for All by the Year 2000" concludes by advocating a new policy framework for "ensuring equity in health through more effective inter-sectoral health promotion and protection; and pursuing equality in access to primary health care... " (WHO, 1992). This concern for achieving greater equity in health is based largely on the view, now widely held in most societies, that everyone should have access to basic health care independent of their ability to pay for it.' Put another way, the tolerance level for inequality in health is lower in most societies than the tolerance level for inequality in income. It is probably lower than the tolerance level for inequalities in education, too, since differences in health are so much more, literally, a matter of life and death. But the reality in virtually every society is far different from the apparent objective: the poor die earlier and more often. In this paper, we aim to show that this cruel reality is not that surprising: it is the predictable outcome of the usual alignment of economic and political forces, which like a powerful ocean current, constrains the extent to which public resources can and will go to the poor. Awareness of this powerful current should not, however, be a cause for pessimism and inaction. On the contrary, awareness is the only sensible starting point for designing and implementing realistic policies and programs to swim against the tide. The first issue we face is the measure of equity itself. There is no agreement on how to measure equity: whether by actual health outcomes across different groups, by utilization of services, or by access itself - and there are currently few data for any of these measures, within or across countries. We argue that the simplest and most useful measures of equity from a policy-making point of view are access and public expenditures; these measures are conservative, in that they generally register less inequality than do indicators of health outcomes (Section II). We then review what little evidence there is within countries on equity using these measures (Section III); and discuss why public policies in health (as well as in education) usually favor the wealthy at the expense of the poor (Section IV). We close (Section V) with an analysis of the cases of certain countries that have achieved a more equitable distribution of public resources for health, and identify some lessons for reform strategies elsewhere. ' It may also stem, of course, from the growing recognition that better health care can accelerate economic growth by improving productivity at work and learning in school, and can help reduce poverty directly and indirectly. In this paper, we start from the premise that there is also an independent political and social objective of achieving equity. 2 II. Measuring Equity in Health For most policy-makers and for Figure 1: Child Mortality in Rich and Poor the general public, "improving equity" Neighborhoods in Selected Metropolitan Areas, means working toward greater equality in Late 1980s health outcomes or status among all the individuals in a country, regardless of the Percentage Devlation from National Mean income group to which they belong. Such equality is far from being a reality in 0.41- - - developing countries today - on the 0.2 ................... .. ... contrary, health status differs dramatically according to income level. In Indonesia, 0 India and Kenya, child mortality is higher .0.2 .... . . . . ... in states or provinces with larger proportions of poor people. Within cities, -0.4 ... .... there are large differences in child 0.8 survival between rich and poor Cairo, Egypt Colombo. Sri Lanka Llma, Peru neighborhoods (see Figure 1). In U Poor Neighborhoods IRich Neighborhoods Madurai, the second largest city in India's S W orld BNk, hsRhio d Tamil Nadu State, children in the poorest households were more than twice as likely Note: Child mortality indexes for each neighborhood are to suffer from serious physical or mental calculated by dividing the observed number of deaths disabilities as children from slightly among children of women in the sampled households of a better-off households. In Porto Alegre, distributionofwomen by the legcth of inmber (given the Brazil, child mortality in poor households bearing children and the national average child mortality in 1980 was twice the level for wealthier levels at each duration of childbearing). Percentage families. deviations from the national average are obtained by subtracting 1 from a neighborhood's index and multiplying the result by 100. Neighborhoods in each city were ranked Differences in health outcomes according to the proportion of houses with concrete floors. across income groups, however, are likely Poor neighborhoods were the lowest 25 percent in this to exaggerate differences in inputs of ranking; rich neighborhoods were the top 25 percent. public resources across income groups to health. Health outcomes are in fact "produced" by households, using a combination of inputs, including benefits of public health services but also housing, food and nutrition, sanitation practices, exercise, smoking and other habits, and so on. Many of these inputs are in turn related to the education of household members; many are accidental inputs to health in that other objectives may largely govern their use or lack of use. For many reasons, high-income households are likely to "produce" more and better health than low-income households, since their members tend to be better educated and with higher income can purchase more food and better water and sanitation.2 In high-income countries in some periods, the very rich, with more obesity and smoking, may have suffered 2 Behrman (1988) presents an economic model that incorporates the household production of health. 3 some health disadvantages. But today, with more widespread understanding of how health is produced, the rich virtually everywhere combine more education and more income with better access to information about the causes of good health to ensure better health outcomes than the poor enjoy - and this would be true even were there no systematic difference across income groups in access to health services per se. Indeed, in a few countries the rich may not rely on public services at all, instead purchasing medical services directly from private providers. In these countries, even if all public resources for health were spent entirely on the poor, the rich could still end up with better health outcomes. If the problem with measuring differences by health outcome is that many other factors affect outcomes besides health services, why not measure outcomes once individuals are sick? Do individuals with the same health problems fare systematically differently depending on their income? Again, there is strong evidence of major gaps between rich and poor in developing countries in treatment for identical health conditions. In rural parts of C6te d'Ivoire, Ghana, and Peru, individuals in rich and poor households have roughly equal chances of being ill at any given time; but of those who fall ill, individuals in wealthy families are about twice as likely to obtain care (see Table 1) (Baker and van der Gaag, 1993). In rural Peru in 1984, children in families in which the mother had secondary or university education (with education status closely correlated with income) were 3-6 times more likely to be vaccinated than in families in which the mother had no education (see Table 2) (Musgrove, 1986). Table 1: Percent Seeking Care for Illness or Injury Rural Quintiles 1 2 3 4 5 All Cote d'Ivoire 23 35 49 39 44 36 Ghana 26 39 41 46 46 41 Peru 20 20 30 34 39 29 Urban Quintiles 1 2 3 4 5 all CMte d'Ivoire 49 58 63 65 64 60 Ghana 40 46 55 58 59 52 Peru 35 44 48 53 57 48 Source: Baker and van der Gaag, 1993. These measures are also, unfortunately, likely to exaggerate differences across income groups in access to health care. Even for a given sickness, the poor are less likely to acknowledge they are sick; when sickness is acknowledged, the poor are still less likely to use 4 available services than the rich.3 The tendency of the poor to minimize sickness (or of the rich to exaggerate it) is generally assumed to reflect different expectations of what is normal and of the efficacy of individual actions. The tendency of the poor to utilize available services less, even when sickness is acknowledged, is not surprising since use of services, even if the services themselves are free, usually involves other costs, e.g. time lost from work and transportation. Table 2: Vaccination Coverage of Children Under 5 in Peru by Mother's Education, 1984 Some Complete None Primary Primary Secondary Higher National Total 16.2 26.7 42.2 55.8 65.3 Urban 34.4 43.5 50.2 58.5 65.5 Rural 11.8 15.1 26.8 36.0 61.0 Source: Musgrove, 1986. In short, for reasons that have nothing to do with health services, the poor are less able to "produce" good health outcomes, and less likely to utilize health services that are available to them. These two realities lead to two conclusions: * Differences in health outcomes across income groups exaggerate differences in policy "effort" by an amount which is unknown and changes over time. Therefore measures of equality in terms of public resources, to be comparable over time or across countries, must be restricted to simple measures of public expenditures per person or of physical access, independent of demand. Such measures also have the advantage of being relatively conservative, i.e., they minimize inequality. * At the same time, it is clear that equality in health outcomes requires much higher public spending on the poor per person than on the rich. The second point has interesting implications for policy. For example, the World Bank's 1993 World Development Report places substantial emphasis on the point that many 3Baker and Van der Gaag (1993) report systematic differences in the extent to which different income groups report sickness, using data from household surveys in several different countries. Though not new, their finding of the same pattern in different countries is convincing. It is also well-known that women are systematically less likely to report themselves sick than are men - another apparent instance of differences across groups. 5 factors outside the health sector, especially household income and education, influence health outcomes.' At the same time, the Report concludes that countries can greatly improve health outcomes by making available a minimum package of highly cost-effective public health and clinical services - the costs of which are not presumed to vary at all across households within countries nor across countries (except for differences in supply costs). The estimated costs ($12 in low-income and $22 in middle-income countries) are underestimates to the extent they are based on average costs of current supply, and fail to take into account the additional costs of generating demand and subsidizing utilization by poor households. In the remainder of this paper, however, we focus on the first point: comparable measures of health equity across countries, and the implications of this more conservative measure of equity for policy. III. Evidence of Inequity: Public Expenditures and Physical Access The increase in household Figure 2: Travel Time, CMte d'Ivoire (minutes) surveys in the last two decades in developing countries provides a rich new Household (percent) source of systematic. data on differences *oo in physical access to health services for different income groups. These data tell so the same story virtually everywhere: access to basic health care is unequal, Im with the poor having much less access to 40 simple, cost-effective services. Data gathered through the World Bank's Living 20 Standards Measurement Survey project (LSMS) show this result for 6te 0Bonom 20% Top 20% Bottom 20% Top 20% Botlom 20% lbp 20% d'Ivoire, Ghana, and Peru (see Figures 2- Hospital Dispensary Physician 4). The differences between rich and E s 5mins. 1540 mins. SWO4O mihs. * > 00 mine. poor are most dramatic in Peru, where Nationally-based expendfture decile nearly two-thirds of low-income families have to travel for more than an hour to reach a primary care provider; by contrast, more than 95 percent of wealthy households are located less than half an hour from their primary provider. In C6te d'Ivoire, the disparities are also striking. Even in Ghana, where primary care centers are more widely distributed throughout the country, about half of the poor are more than an hour from such facilities, compared to less than 20 percent of the better-off. Data on public expenditures can also provide a good measure of health equity. Early studies by Meerman on Malaysia (1979) and Selowsky on Colombia (1979) showed how household survey data on utilization of services and household expenditures on services could ' An entire chapter (Chapter 2) is devoted to this point. 6 be used to study the incidence of public expenditures on health and other public programs by income group. W'hile these studies discuss the role of demand in utilization of services across income groups, estimations of public subsidies nevertheless are based on utilization of services, a measure which is highly sensitive to differing levels of demand. Figure 3: Travel Time, Ghana (minutes) Household (percent) 100 20r Bottom 20% Top 20% Bottom 20% Top 20% Bottom 20% Top 20% Hospital Health Center Physicion < 30 mins. * 30-60 mins. E > 60 mins. Nationally-based expenditure docile Figure 4: Travel Time, Peru (minutes) Household (porcent) Bottom 20% Top 20% Bottom 20% Top 20% Bottom 20% Top 20% Hospital Hoalih Center Physician |< S0 min. * 30 60 min. E1> 00 mIn.| Netlonally-baad expenditure deolie 7 Recent data from Indonesia show a highly skewed distribution of public monies in 1990: the bottom 20 percent of households received only 12 percent of public spending for health, while the top fifth obtained nearly 30 percent of public expenditures (see Figure 5) (World Bank, 1993b). As with the earlier studies, however, these results reflect in part differences across income groups in the demand for and use of publicly financed services. A result from a 1991 national household survey in Indonesia (Table 3) is more revealing: families in the top income deciles reside on average in areas where the density of primary care centers and private doctors is 3 to 5 times greater than in areas inhabited by the poor. During the 1980s, the Figure 5: Public Spending for Health Government of Indonesia had made a in Indonesia, 1990 major effort to reduce even greater earlier inequities by building more health centers Percent of Public Expendfture and health posts in low-income areas of the country. As a result, the share of the / poor falling ill in rural Java who used modem health providers (doctor, hospital, 24 30 primary care center, polyclinic, . . .. . paramedic) had risen from 47 percent in 16 20 1 1978 to 55 percent a decade later. Possibly because of the factors affecting 10 health production in the household, and o itm E demand for health services, however, the gap between rich and poor remained 1 2 3 4 5 large: the share of those from wealthy Income Quintile rural households falling ill who used Source: World Bank, 1993b. modern providers was nearly 73 percent in 1987. And for Indonesia as a whole, individuals in the highest rural income decile made more than twice as many annual visits to modem providers than those living in the lowest decile (3.00 versus 1.41), including one and a half times as many visits to a primary health center (1.17 versus 0.77) (van de Walle, 1992). Comprehensive national data on the incidence of government spending by income groups is scarce. Two data sets available (Costa Rica and Malaysia) show government spending that favors the poor, but the more common pattern is of bias toward the wealthy. * In South Africa, public subsidies to the wealthiest 15 percent of families covered by private health insurance, in the form of tax relief, amounted in 1990 to nearly a fifth of all public spending for health - without counting direct government expenditures for the wealthy (Broomberg, 1992). * In Zambia, more than 20 percent of the Ministry of Health budget in the late 1980s went to a single teaching hospital serving the population of the capital city, whose inhabitants had incomes far above the national average. 8 * In the Northwest Frontier Province of Pakistan, nearly 27 percent of the provincial health budget for 1991-92 was earmarked for two teaching hospitals (Smithson, 1993). * In Brazil in the mid-1980s, nearly 80 percent of all public spending for health was devoted to largely curative, high-cost hospital care concentrated in urban areas and especially in the affluent southern part of the country. Households in the top income quintile received about 38 percent of public subsidies for health (McGreevey, 1988). * In Peru in 1984, the Lima health region consumed nearly 47 percent of the government's budget for patient-related care, even though the relatively affluent capital region had only 32 percent of the country's population (Musgrove, 1986). * In many Latin American countries, ministry of health spending appears to be fairly progressive because only the poor use subsidized government health centers and hospitals. But when this distributional pattern is combined with the large public subsidies to social security-based health care for the middle classes, total government expenditure for health is again weighted strongly toward the better-off (Mesa-Lago, 1991). Table 3: Distribution of Health Facilities by Income Decile, Indonesia - 1986 (percent of villages in district of individual's residence with health facility) Income Decile Type of Health Facility 1 2 3 4 5 6 7 8 9 10 Hospital 2 2 2 2 3 3 3 5 7 10 Health Center 10 10 10 11 11 11 14 16 23 32 Private Doctor 10 11 12 14 15 18 21 26 35 48 Source: World Bank, 1993b. All figures in these tables have been rounded off. Recent analysis of data collected from household surveys in rural Kenya (Dayton and Demery, 1994) tell a story similar to the one from Indonesia (see Figure 6). Overall, the distribution of public subsidies to health care is regressive: even within rural areas, the bottom quintile receives only 14 percent of the total health subsidy, compared to 24 percent for the top rural quintile. The average subsidy for households in the bottom 10 percent of rural income is 9 Figure 6: Share of Health Subsidy Targeted to the Poor, Rural Kenya - FY93 less than half the subsidy for households in the top rural decile. This skewed Share of Subsidy pattern is the result of significantly higher 30 use by the upper-income rural households 26 of hospital-based services, for which the 2S---------_ unit subsidy is also much greater than for 20 ------- 19 19 clinic or dispensary-based care. By contrast, the distribution of government - - ----- ......... health subsidies through the lower-level 10 : - ^ _ - facilities (health centers and dispensaries) ... - ^ --- ~ ~ 1shows a more progressive pattern. The incidence of health center subsidies, for 0 Health Hospitals HealthCenters Dispensai example, is estimated to be 24 percent for the bottom rural quintile and 13 percent |iE Richest 20% * Poorest 20% for the top rural quintile. Source: Dayton and Demery, 1994. Data indicatingdisproportionately large hospital expenditures in major cities do not necessarily signify inequity, for two reasons. First, most such expenditures potentially benefit all parts of the population, including the rural poor who can in principle travel to exploit hospital care when their medical needs require it. Second, most of the most cost-effective health programs government can finance do not require hospital stays. Once all these cost-effective programs are fully available to the entire population, they may reasonably take up only a small portion of overall public spending - and the remainder of such spending, for critical referral care, may reasonably occur in urban hospitals and with high unit costs could constitute a large portion of total spending. However, wherever the most basic services are not fully accessible to the rural poor - as is the case in Kenya, Pakistan, and Peru - a high proportion of public resources going to urban hospitals should suggest a problem of inequity. Coupled with other information of the type reported above, indicating lower demand and utilization of all kinds of public health services among the poor and much worse health outcomes among the poor, the data on heavy concentrations of public spending on urban hospitals, at the least, amounts to a strong warning signal of inequity. As mentioned above, data from Costa Rica and Malaysia show a different pattern - one of public expenditures favoring the poor. In Costa Rica in 1986, 28 percent of government health spending accrued to the poorest fifth of households, while just 11 percent went to the wealthiest quintile (World Bank, 1990). Even when these shares are adjusted for the fact that the average poor household had more members than the average wealthy family, public spending for health still favored the poor. In Malaysia, data from both 1978 and 1990 show a similar continuing progressive distribution of public funds for health (Meerman, 1979 and World Bank, 1992a). In Section 4 below, we examine some of the underlying factors that help to explain this "pro-poor" bias in Costa Rica and Malaysia. 10 IV. The Politics of Inequity5 The predominant pattern of unequal distribution of public resources for health in developing countries is no mere accident. On the contrary, it is predictable, given the related distributions of economic resources and political power in these countries. The explanation for this inequity can be understood in the context of public choice theory, which provides a positive model of what the government will do, under the assumption that the chief agents act to maximize individual utility rather than social welfare.' According to the theory, politicians and bureaucrats do not seek to optimize economic efficiency but rather to maximize their own chances of getting reelected and staying employed; similarly, individuals use governments to maximize their own income by creating and protecting market positions and capturing publicly- financed services and transfers. Politicians and political parties have some discretionary power because of barriers to entry and because they are in a position to shape as well as respond to people's tastes and preferences. As the same time, they must act in a way that deters threats from potential competitors, and this limits the scope of their monopoly power. Where democracy does not exist in developing countries, a similar process can occur with even fewer political checks on the use of government resources to benefit the already powerful. The allocation of resources resulting from public choice politics is often inefficient, for several reasons: - In a setting of imperfect information, people may not know the degree and direction of redistribution going on. If well-defined groups know they are "losers," they are more likely to mobilize and foment opposition to existing policies; therefore the "gainers" benefit from perpetuating a "veil of ignorance." - The real costs of publicly-produced goods may be above minimal levels, because government imposes costs of bureaucracy and red-tape, often lacks competitive pressures for internal efficiency, and uses distortionary tax financing. * The diversion of entrepreneurial energies toward extracting a surplus from public agencies rather than toward productivity-enhancing market activities can impede private sector efficiency and growth. This is not to say that there will be no redistribution to poorer groups under public choice theory. The extremes of poverty and socio-economic immobility may raise fears of crime or revolution which will ultimately hurt the rich; historically, the provision of basic education, The discussion in this section is from Birdsall and James (1993). 6 This "public choice" approach (see Buchanan, Tollinson and Tullock, 1980) is somietilnes referred to as "new political economy" (NPE). 11 employment or medical insurance have been ways of combatting these problems. Also, since there are more poor people than rich, the desire to constrain the popularity of opposition political groups in a democracy leads to some redistribution to lower-income groups on grounds of expediency. The social services sectors, including health and education, are arenas in which many of these forces play themselves out, as they involve a variety of quasi-public goods with different mixes of public and private benefits. The fact that health and education services generate social as well as private benefits, and their frequent designation as "merit" goods provides justification for government intervention along welfare theory lines. Yet once this intervention begins, ostensibly to correct for market failures and to benefit poor consumers, it is often seized by producer groups and the allocation of resources diverted to a more private service mix that predominantly benefits the rich. While rhetoric stresses the importance of avoiding price- rationing in order to preserve access for the poor and thereby garner their political support, alternative rationing mechanisms emerge, such as proximity to hospitals, knowledge of how to use the health system, and selection tests for admission to free public universities which are equally income-biased. Three politically influential groups often work actively to protect the flow of government funds toward health services that benefit them directly, and resisting efforts to reallocate public resources for health to the poor. First, government officials and politicians stand to gain from construction projects for large hospitals. These major civil works are highly visible and popular, often seen by the general public as evidence of government commitment to medical care. At the local level, they are seen as evidence of the commitment of local politicians to local needs, and often of the ability of these politicians to command central resources for local programs. Those seeking reelection will point to new hospitals as among their major accomplishments. High-ranking civil servants also benefit disproportionately from access to free medical services in these major hospital facilities. At the extreme, many developing countries allow senior officials and their families to obtain sophisticated medical care abroad, in Europe or North America, at government expense. Such medical "evacuations" may even have a line in the ministry of health budget. Second, various middle-income consumer groups object strongly to any erosion of their public subsidies for health services. These groups include professionals from the private sector and labor organizations. British colonial policies in Ghana and Zambia, for example, dictated that public hospitals in Accra and Lusaka, respectively, should provide free or heavily subsidized medical care to the colonial elites. Once established, it has been extremely difficult for policy- makers in those countries to shift this demand for medical care to private hospitals or to charge full costs to the wealthy for care in the government teaching hospitals. Labor unions seek to protect public subsidies in the middle-income countries of Latin America, where governments extend financial support to social security-based health care through some combination of tax relief, public contributions to insurance premiums, and direct budgetary transfers to social 12 security agencies (Mesa-Lago, 1991). This social-security based care benefits only those in the formal sector.7 Third, the health workers and their respective labor organizations themselves are a major source of resistance to change. A shift in public funding to basic care would require that doctors and nurses be redeployed from large urban-based hospitals to smaller peripheral facilities in poor urban neighborhoods and rural areas. Living and working conditions in these may well be more difficult than in the central hospitals; it is thus not surprising that health workers oppose these changes. Similarly, a reallocation of government spending for health would reduce the demand for publicly-financed services from medical specialists. The specialists, through their professional associations, can be vocal and influential lobbyists for continued spending on sophisticated equipment (such as diagnostic imaging machines) and hospital facilities. As Hausmann (1993) points out in the case of Venezuela, "(Public sector) budgets have become an entitlement of producers, not the purchase of a service on behalf of consumers.... Centralization breeds unionization and the consequence is a bilateral monopoly in which the government is bound to be the weaker member and will compromise the future to get over the present." For the above pattern of public subsidies to dominate, it is not necessary that all politicians and officials pursue only their own individual interests, nor that service providers and middle-class consumers have complete control of public revenue and expenditure patterns.8 It is only necessary that these tendencies occur widely enough and often enough to minimize countervailing efforts to reach the poor - that is, to prevent what we have referred to above as "swimming against the tide." The more equal the overall distribution of income, and the stronger the political voice of the poor, the easier it should be to design and implement pro-poor programs. And indeed, as discussed below, pro-poor bias has and can emerge, particularly in countries with relatively low income-inequality or political systems which rely on the bottom third of the income distribution to maintain political legitimacy. Malaysia and Costa Rica, discussed below, as well as Sri Lanka, Korea, and the Indian state of Kerala, are examples of economies with relatively low income inequality and histories of broad-based social programs to reach the poor. Communist countries, including China and Cuba, are examples of economies in which the political legitimacy of the ruling party has relied heavily on support of the working and peasant classes. Taiwan, Korea, Hong Kong, Singapore, Thailand, and Malaysia are also examples of countries in which political leaders, in the face of external and internal communist threats in the ' Several countries in Latin America, including Costa Rica, discussed below, and Brazil, have recently extended some forms of social security-based care, such as emergency hospital services, to all, in an effort to eliminate at least the most glaring forms of a two-tier system. Colombia is currently studying this option. 8 Toye (n.d.) for example, in a lengthy discussion of the NPE, argues that many leaders have demonstrated powerful social consciences, and that the idea of the state as embodying common social goals has and will continue to affect resource allocation patterns. 13 post-war period, employed a variety of mechanisms to ensure that urban workers and the rural poor shared in economic growth. These mechanisms included public housing programs (Hong Kong and Singapore), extensive investment in rural infrastructure (Indonesia and Thailand), land reform (Korea and Taiwan), and heavy emphasis on universal access to basic health services and primary education.9 In all these countries of East Asia, the political situation required some swimming against the tide. These as well as other examples discussed below demonstrate that the public choice view of the world need not and does not always dominate. V. Strategies for Reform Faced with these powerful forces that maintain an inequitable distribution of public resources for health, some countries such as Costa Rica, Malaysia, and Zimbabwe have nevertheless achieved greater equity. How have they done so? We argue below that certain kinds of macroeconomic policies and political arrangements are important in determining a country's successful equity orientation in the health sector. We then point to the specific strategies and policy and program instruments that some developing countries have adopted to enhance equity in health. Macroeconomic and Political Determinants One fact stands out from the beginning: the countries that have been most successful in achieving equity have experienced moderate to high economic growth rates, and have been able to tap that expanding resource base to improve health care for the population as a whole, including the poor. Such a "shared growth" policy stance ensures that all segments of the population, including the poorest segments, benefit from and are able to participate in their country's economic growth. While growth does not automatically lead to a redistribution of basic public services, it is extremely difficult to achieve such a redistribution without it. Ghana, Peru, and Zambia are examples of countries that suffered from stagnating or even declining national income in the 1970s and 80s. During this period, government spending for health was severely constrained. There was no incremental public funding for health that might have been allocated to primary care for the poor. By contrast, Costa Rica's economy grew at an average of nearly 6 percent a year in the 1970s and by 3 percent in the 1980s. Malaysia did even better, growing at average of 7 percent a year during the two decades. Zimbabwe grew by 6 percent annually in the first half of the 1980s, when the government's redistributive efforts in health were most vigorously pursued (World Bank, 1992b). Other countries that made strong advances in improving equity in health, such as Chile, China, and Korea, also have been among the fastest growing economies in the developing world. 9 For discussion of this approach to "shared growth" in East Asia, see World Bank (1993c). For the implications of income inequality, see Birdsall, Ross and Sabot (1994). 14 As mentioned above in the previous section, political dynamics have been a key factor in those developing countries that have moved strongly toward, or struggled to maintain, equity in health. These pro-poor political forces take different forms, including the post-independence drive to redress earlier discrimination, as in Zimbabwe, and the search for legitimacy by a modernizing authoritarian regime, as in Korea (and in Chile, too?). In Malaysia, the government's consistently progressive approach to health was prompted by its objective of assisting the relatively disadvantaged ethnic majority and maintaining its important political base in the country's rural areas, where lower-income households are concentrated. In Zimbabwe, the main impetus for change was national independence and majority democratic rule, starting in 1980. The newly-independent government recognized that its political support came from low-income rural households that had previously been disenfranchised. Members of these rural households had joined the guerilla movements that fought against the previous regime. The new government sought to reward their supporters and to redress earlier imbalances favoring the urban middle-class by investing heavily in the rehabilitation and construction of over 500 health centers, in the training of thousands of nurses to staff these facilities, and in basic programs of immunization, antenatal care, and infectious disease control. There was a concomitant decision to place a moratorium on new investments in the central hospitals, which had benefitted mainly the urban populations and especially the country's white minority. At the same time that these governments shifted health spending toward the poor, they explicitly sought to preserve existing high quality health services for middle and upper-income groups, thus maintaining broad political support for their reform efforts. In Zimbabwe, the two highest quality tertiary care hospitals, in Harare and Bulawayo, respectively, received generous operating budgets from the government during the 1980s, even though new capital spending was severely restricted. At the end of the decade, the two hospitals still accounted for over 10 percent of the Ministry of Health budget. In Costa Rica and Malaysia, middle-class families continued to have access to well-funded health care in public facilities, even as coverage was extended to poor households. And the wealthy enjoyed the choice of opting out of the public system and utilizing private health care with its associated amenities - services financed privately with private insurance and/or out-of-pocket payments, and not with public funds. Instruments for Promoting Equity In this political environment favorable to more equitable access to health services and to more equitable distribution of public resources for health, the pro-equity governments have used several instruments to achieve this objective. First, at the same time that the wealthy were encouraged to shift much of their consumption of health services to the private sector, these governmnents have sought consciously to reduce, eliminate, or altogether avoid public subsidies to private financing and delivery of health care to the better-off. These subsidies can take many "hidden" forms, including government budget transfers to social security-based insurance and tax deductions on employer and employer contributions to insurance schemes. In Zimbabwe, the government has gradually reduced large subsidies to the better-off by cutting down their 15 deductions from income tax for premiums paid to private health insurance. In Costa Rica, the wealthy must contribute to the social security-based health fund, even if they are served by private doctors and hospitals. Second, pro-poor governments have targeted public spending toward health interventions and facilities serving primarily, but not exclusively, the poor. Zimbabwe's focus on rural health facilities and district hospitals is a good example of this type of targeting using simple geographical criteria. Costa Rica's emphasis on basic primary and preventive care - immunizations, control of diarrheal disease, safe childbirth services - in the 1970s also effectively targeted the poor, who suffered greater disease burden from vaccine-preventable illnesses, diarrhea, and childbirth complications than did the rich. At the same time, these primary and preventive services also benefitted the middle-class and wealthy, thus helping to maintain political support for these initiatives. Third, at least in the case of middle-income countries, equitable access and public spending for health have been pursued through the effective universalization of health insurance. In Costa Rica, for example, the democratically-elected government decided in the early 1980s to expand the social security-based health system to the entire population. This meant covering the 20 percent of Costa Ricans who had not previously been covered, and especially the poor. Such a decision required the government to subsidize health services for the poor, since their employment-based contributions to the social security fund would not be large enough to meet the cost of services for them. At the same time, ministry of health and social security hospitals were unified, and all Costa Ricans became eligible for care in the same public hospitals, managed by the social security agency. Improved equity in health through the universalization of insurance was also the course followed by Korea during the 1980s. In less than 10 years, the Korean government created a comprehensive national health insurance system from scratch, forging together several hundred regional and industry-based social insurance funds (modeled on the German "sickness funds"). To achieve truly universal coverage, the government chose to subsidize the insurance funds for the roughly 8 percent of the population that is disabled or indigent (Yang, 1991). Chile has also pursued greater equity in health in recent years by establishing a single national health fund (FONASA), into which both payroll deductions for social insurance and a general revenue subsidy for health care are deposited. All Chileans are eligible for health care financed from the FONASA, whether in the form of payment vouchers to private service providers or capitated and diagnostic-related payments to public sector providers. In this way, Chile has managed to reach the roughly 15 percent of the population that until recently was not covered by social insurance (Bossert, 1992). The successful experience of these reforming countries reveals several important lessons for other developing countries: the need for sustained economic growth to underpin policies for reallocation of public financing; the crucial political conditions that allow for fuller and more equal participation by all segments of the population; and the use of a gradual approach in which 16 services for the wealthy and middle-class are not undercut, even when the share of public spending for these groups is declining. These examples of successful countries also show that maintaining a more egalitarian health system is a constant struggle to "swim against the tide." There are always interests that would skew essential health services and public funds for health away from the poor. In Costa Rica and in Zimbabwe, for example, pressures from government doctors to concentrate in the main tertiary hospitals in the largest cities and to acquire complex diagnostic equipment have only been partially resisted. China is a good example of a country where the tide has been so strong in recent years that important egalitarian features of the health system have been eroded. In the 1960s and 1970s, China experienced one of the most dramatic advances in health of any developing country: child mortality, for example, declined from 210 to 85 per 1000 live births between 1960 and 1975. Much of this progress was due to broad-based provision of public health services in the areas of insect vector control, immunization, improved hygiene, and family planning (Jamison, 1985), backed by well-targeted public spending. At the same time, China's unique rural health insurance system, which covered about 500 million persons in the 1970s, guaranteed adequate funding for basic clinical services (e.g., treatment of tuberculosis and respiratory infections and safe pregnancy and delivery care) throughout the country. As an unfortunate consequence of China's economic liberalization prograrm. of the past decade, government funding for public health has declined and the rural insurance system has now largely disintegrated. A recent study (World Bank, 1992c) suggests that these new health policies have made the distribution of government spending for health in China more unequal and may be contributing to an increased incidence of easily treatable diseases such 4s tuberculosis. The experience of the few reforming countries and of the much larger number of countries that have not yet shifted the balance of public resources for health toward low-income groups also points to the many serious obstacles to equity-oriented reforms. Politically- influential groups that stand to lose from a change in the status quo will block changes. Achieving lasting reforms requires a combination of political enfranchisement, skillful coalition- building and negotiations, and enlightened leadership. The generation and dissemination of information - for example, on differences in health status, service utilization, total health spending and government expenditures among different income groups in a given country - can be a crucial element in achieving reforms that improve equity. Policy-relevant information becomes an especially powerful force for change in societies with democratic political institutions, a broadly-educated population, and a diverse and critical press. Under these circumstances, the analysis of health spending levels and patterns across geographical and income groupings can be a potent tool in the hands of reformers, and can be used to couinteract the efforts of others (e.g., politicians in power, elite civil servants, professional associations) to obtain a disproportionate share of public resources for health. This 17 has been the case in Chile and in some OECD countries such as the Netherlands where equity- enhancing health reforms are currently taking place. The international community can play a catalytic role in this process of change, by providing information and extending financing to soften the transitional costs of reform. External development agencies have stimulated information dissemination and debate on equity issues in health through their sponsorship of sectoral studies and policy seminars. Recent donor- supported studies of government spending for the social sectors in countries as geographically diverse as Indonesia, Kenya, and Uruguay are helping to shape the health policy debate in each of these countries. When these studies are embedded within broader country "public expenditure reviews" that analyze both the incidence of taxation and spending, as in Argentina and the Philippines, they further enrich the debate. Financial backing from development institutions has also facilitated and strengthened equity-oriented health reforms. In Costa Rica, the Inter-American Development Bank and the World Bank have financed improvements in the Ministry of Health and the social insurance agency that are designed to sustain the country's long-standing emphasis on primary care for all population groups, including the poor, and to ensure that all workers contribute their stipulated share of social insurance costs. In Zimbabwe, a coalition of European bilateral donors and the World Bank have helped to finance a substantial share of public investments in community and clinic-based services for poor rural groups. These international institutions and others should continue to support public policies and programs for improved equity in the health sector in developing countries. In the end of the day, however, the impetus for sustained reform will come mainly from within the developing countries themselves. 18 Bibliography Baker, Judy L. and Jacques van der Gaag. 1993. "Equity in Health Care and Health Care Financing: Evidence from Five Developing Countries." In Eddy van Doorslaer, Adam Wagstaff and Frans Rutten. Equity in the Finance and Delivery of Health Care: An International Perspective. Oxford: Oxford University Press. Birdsall, Nancy. 1993. "Pragmatism, Robin Hood, and Other Themes: Good Government and Social Well-Being in Developing Countries." In Lincoln C. Chen, Arthur Kleinman, and Norma C.P. Ware, eds. Health and Social Science in International Perspective. Cambridge, MA: Harvard University Press. Birdsall, Nancy and Estelle James. 1993. "Efficiency and Equity in Social Spending: How and Why Governments Misbehave." In M. Lipton and J. van der Gaag, eds. Including the Poor. Oxford: Oxford University Press. Birdsall, Nancy, David Ross and Richard Sabot. 1994. "Inequality and Growth Reconsidered." Paper for the Annual Meetings of the American Economics Association, Boston, January 3-4. Bossert, T. 1992. "An Analysis of the Chilean Health Reform." Latin American and Caribbean Health and Nutrition Sustainability Project, Washington, D.C. Broomberg, J. 1992. "Aspects of Health Care Financing in South Africa." Unpublished. Buchanan, J., Tollinson and G. Tollick. 1980. Toward a Theory of a Rent-Seeking Society. Texas University A&M Press. Chile, Government of. Ministry of Health. 1992. Health Care Situation and Health Care in Chile. Santiago. Dayton, Julia and Lionel Demery. 1994. Public Health Expenditure and the Rural Poor in Kenya. Washington, D.C.: The World Bank. Hausmann, Ricardo. 1993. Sustaining Reform: What Role for Social Policy?. Paper for the IDB-OECD International Forum on Latin American Perspectives, Paris, 8-9 November 1993. Jamison, Dean. 1985. "China's Health Care System: Policies, Organization, Inputs, and Finance." In Scott B. Halstead, Julia A. Walsh, and Kenneth S. Warren, eds. Good Health at Low Cost. New York: Rockefeller Foundation. Meerman, Jacob P. 1979. Public Expenditure in Malaysia: Who Benefits and Why. 19 Mesa-Lago, Carmelo. 1991. Social Security and Prospects for Equity in Latin America. Discussion Paper 140. World Bank, Washington, D.C. Musgrove, Philip. 1986. "The Measurement of Equity in Health." World Health Statistics Quarterly. Number 39: 325-335. Rosero-Bixby, Luis. 1985. "Infant Mortality Decline in Costa Rica." In Scott B. Halstead, Julia A. Walsh, and Kenneth S. Warren, eds. Good Health at Low Cost. New York: Rockefeller Foundation. van de Walle, Dominique. 1992. The Distribution of the Benefits from Social Services in Indonesia, 1978-87. Working Paper No. 871. Washington, D.C.: World Bank. Selowsky, Marcelo. 1979. Who Benefits form Government Expenditure? A Case Study of Colombia. World Bank Research Publication. New York: Oxford University Press. Smithson, Paul. 1993. Public Sector Health Financing and Sustainability in Pakistan. London: The Save the Children Fund (UK). World Bank. 1984. China: The Health Sector. A World Bank Country Study. Washington, D.C. World Bank. 1988. Brazil: Public Spending on Social Programs: Issues and Options. Washington, D.C. World Bank. 1990. Costa Rica: Public Sector Social Spending. Washington, D.C. World Bank. 1992a. Malaysia: Fiscal Reform for Stable Growth. Washington, D.C. World Bank. 1992b. Zimbabwe: Financing Health Services. Washington, D.C. World Bank. 1992c. China: Long-Term Issues and Options in the Health Transition. Washington, D.C. World Bank. 1993a. Investing in Health: The 1993 World Development Report. New York: Oxford University Press. World Bank. 1993b. Indonesia: Public Expenditures, Prices. and the Poor. Washington, D.C. World Bank. 1993c. The East Asian Economic Miracle: Economic Growth and Public Policy. New York: Oxford University Press. World Health Organization. 1992. Implementation of the Global Strategy for Health for All by the Year 2000, Second Evaluation. Geneva. 20 Yang, Bong-Min. 1991. "Health Insurance in Korea: Opportunities and Challenges." Health Policy and Planning 6, no. 2: 119-129. Human Resources Development and Operations Policy Working Paper Series Contact for Title Author Date paper HROWP39 Notes on Education and Harry Anthony Patrinos September 1994 I. Conachy Economic Growth: Theory 33669 and Evidence HROWP40 Integrated Early Child Mary Eming Young October 1 994 0. Shoffner Development: Challenges and 37023 Opportunities HROWP41 Labor Market Insurance and Deepak Lal October 1 994 M. Espinosa Social Safety Nets 37599 HROWP42 Institutional Development in Alberto de Capitani October 1 994 S. Howard Third World Countries: The Douglass C. North 30877 Role of the World Bank HROWP43 Public and Private Secondary Marlaine E. Lockheed November 1 994 M. Verbeeck Schools in Developing Emmanuel Jimenez 34821 Countries HROWP44 Integrated Approaches to T. Paul Schultz November 1994 M. Espinosa Human Resource 37599 Development HROWP45 The Costs of Discrimination in Harry Anthony Patrinos November 1 994 I. Conachy Latin America 33669 HROWP46 Physician Behavioral Nguyen X. Nguyen December 1 994 M. Espinosa Response to Price Control 37599 HROWP47 Evaluation of Integrated T. Paul Schultz January 1995 M. Espinosa Human Resource Programs 37599 HROWP48 Cost-Effectiveness and Health Philip Musgrove January 1 995 0. Shoffner Sector Reform 37023 HROWP49 Egypt: Recent Cha-nges in Susan H. Cochrane February 1995 0. Shoffner Population Growth Ernest E. Massiah 37023 HROWP50 Literacy and Primary Kowsar P. Chowdhury February 1995 M. Espinosa Education 37599 HROWP51 Incentives and Provider Howard Barnum March 1995 0. Shoffner Payment Methods Joseph Kutzin 37023 Helen Saxenian HROWP52 Human Capital and Poverty Gary S. Becker March 1995 M. Espinosa Alleviation 37599 HROWP53 Technology, Development, Carl Dahlman April 1995 M. Espinosa and the Role of the World 37599 Bank HROWP54 International Migration: Sharon Stanton Russell May 1995 0. Shoffner Implications for the World 37023 Bank HROWP55 Swimming Against the Tide: Nancy Birdsall May 1995 A. Colbert Strategies for Improving Robert Hecht 34479 Equity in Health Human Resources Development and Operations Policy Working Paper Series Contact for Title Author Date paper HROWP23 Is Grameen Bank Sustainable? Shahid Khandker February 1 994 S. David Baqui Khalily 33752 Zahed Khan HROWP24 Concepts of Educational Marlaine E. Lockheed March 1994 M. Verbeeck Efficiency and Effectiveness Eric Hanushek 34821 HROWP25 Scientific Research for Erik W. Thulstrup March 1 994 L. Malca Development 37720 HROWP26 Issues in Education Finance Stephen P. Heyneman April 1 994 B. Cassorla and Management in ECA and 37172 OECD Countries HROWP27 Vocational Education and Julian Schweitzer April 1 994 A. Gonzalez Training: The Role of the 37799 Public Sector in a Market Economy HROWP28 Social Security Issues and Nguyen X. Nguyen May 1994 M. Espinosa Elements of Reform 37599 HROWP29 Health Problems and Policies Mary Eming Young May 1994 0. Shoffner for Older Women: An 37023 Emerging Issue in Developing Countries HROWP30 Language and Education in S.M. Cummings May 1994 M. Espinosa Latin America: An Overview Stella Tamayo 37599 HROWP31 Does Participation Cost the Jesko Hentschel June 1994 D. Jenkins World Bank More? Emerging 37890 Evidence HROWP32 Research as an Input into Harold Alderman June 1 994 P. Cook Nutrition Policy Formation 33902 HROWP33 The Role of the Public and Deepak Lal June 1 994 M. Espinosa Private Sectors in Health 37599 Financing HROWP34 Social Funds: Guidelines for Soniya Carvalho July 1 994 K. Labrie Design and Implementation 31001 HROWP35 Pharmaceutical Policies: Graham Dukes July 1 994 0. Shoffner Rationale and Design Denis Broun 37023 HROWP36 Poverty, Human Development Harsha Aturupane August 1 994 P. Cook and Growth: An Emerging Paul Glewwe 30864 Consensus? Paul Isenman HROWP37 Getting the Most out of Helen Saxenian September 1994 0. Shoffner Pharmaceutical Expenditures 37023 HROWP38 Procurement of Denis Broun September 1 994 0. Shoffner Pharmaceuticals in World 37023 Bank Projects