Financing the health sector in Peru

LSM - 31 A-9s APRIL 1987 Li\ ing Standards NlMasuremrent Stid., \Vorking Paprr \o, .. I Financing the Health Sector in Peru Ruben M. Sua9e-Berf  LSMS WORKING PAPER SERIES No. 1. Living Standards Surveys in Developing Countries. No. 2. Poverty and Living Standards in Asia: An Overview of the Main Results and Lessons of Selected Household Surveys. No. 3. Measuring Levels of Living in Latin America: An Overview of Main Problems. No. 4. Towards More Effective Measurement of Levels of Living", and "Review of Work of the United Nations Statistical Office (UNSO) Related to Statistics of Levels of Living. No. 5. Conducting Surveys in Developing Countries: Practical Problems and Experience in Brazil, Malaysia, and the Philippines. No. 6. Household Survey Experience in Africa. No. 7. Measurement of Welfare: Theory and Practical Guidelines. No. 8. Employment Data for the Measurement of Living Standards. No. 9. Income and Expenditure Surveys in Developing Countries: Sample Design and Execution. No. 10. Reflections on the LSMS Group Meeting. No. 11. Three Essays on a Sri Lanka Household Survey. No. 12. The ECIEL Study of Household Income and Consumption in Urban Latin America: An Analytical History. No. 13. Nutrition and Health Status Indicators: Suggestions for Surveys of the Standard of Living in Developing Countries. No. 14. Child Schooling and the Measurement of Living Standards. No. 15. Measuring Health as a Component of Living Standards. No. 16. Procedures for Collecting and Analyzing Mortality Data in LSMS. No. 17. The Labor Market and Social Accounting: A Framework of Data Presentation. (List continues on the inside of the back cover)  LSMS Working Papers No. 31 FINANCING THE HEALTH SECTOR IN PERU Ruben M. Suarez-Berenguela The World Bank Washington, D.C. 20433, U.S.A.  This is a working document published informally by the Development Research Department of The World Bank. The World Bank does not accept responsiblity for the views expressed herein, which are those of the authors and should not be attributed to the World Bank or to its affiliated organizations. The findings, interpretations, and conclusions are the results of research supported by the Bank; they do not necessarily represent official policy of the Bank. The designations employed, the presentation of materials, and any maps used in this document are solely for the convenience of the reader and do not imply the expression of any opinion whatsoever on the part of the World Bank or its affiliates concerning the legal status of any country, territory, city, area, or of its authorities, or concerning the delimitation of its boundaries, or national affiliation. The LSMS working paper series may be obtained from the Living Standards Measurement Study, Development Research Department, The World Bank, 1818 H Street, N.W., Washington, D.C. 20433, U.S.A. Ruben Suarez-Berenguela is a consultant to the Living Standards Unit of The World Bank. April 1987  LIVING STANDARDS MEASUREMENT STUDY The Living Standards Measurement Study (LSMS) was established by the World Bank in 1980 to explore ways of improving the type and quality of household data collected by Third World statistical offices. Its goal is to foster increased use of household data as a basis for policy decision making. Specifically, the LSMS is working to develop new methods to monitor progress in raising levels of living, to identify the consequences for households of past and proposed government policies, and to improve communications between survey statisticians, analysts, and policy makers. The LSMS Working Paper series was started to disseminate intermediate products from the LSMS. Publications in the series include critical surveys covering different aspects of the LSMS data collection program and reports on improved methodologies for using Living Standards Survey (LSS) data. More recent publications recommend specific survey, questionnaire and data processing designs, and demonstrate the breadth of policy analysis that can be carried out using LSS data.  Acknowledgments I am grateful to Jacques van der Caag, Avi Dor for comments on an earlier version of this paper. I benefited, in addition, from comments and suggestions from Phil Musgrove. Any remaining errors and omissions are, of course, my responsibility. I also thank Carmen Martinez for typing the various drafts and Farah Ebrahini and Bruce Ross-Larson for editing the manuscript.  ABSTRACT This paper reviews the health status of The Peruvians and the financing and spending patterns of Peruvian health institutions. Between 1975-85 declining income per capita of the population has been accompanied by unsuccessful attempts to reduce total government expenditures. However, government expenditures in social programs declined from one third of the budget in 1973-75 to less than one fifth in 1981. Expenditures in health programs as a proportion of the government budget declined from 6.4 percent in 1970 to around 4.5 percent in 1980-85 (to 0.6 percent of the CDP). This amounts to approximately 5.0 U.S. dollars per capita. Central government revenues the main source of financing of public health institutions. Expenditure by corporate health institutions and private institutions and individuals account for approximately 90 percent of total resources spent on health related goods and services. Corporate institutions spent around 100 U.S. dollar per "affiliated" member, private individuals spent between 10 and 20 U.S. dollars per capita. These findings call for the need to explore the actual scope of health government programs and the role that nongovernment institutions may play in implementing health programs. Child and infant mortality and the high incidence of environmental related diseases appear as the most critical health problems in Peru. Observed mortality and morbidity patterns and the tendency of Peruvian health institutions to concentrate on curative rather than preventive services, suggest that there might be a misallocation of resources. Analysis of health related indicators also show a high degree of inequality in the spatial distribution of health resources and sanitation services.  Table of Contents Introduction ............ .. .....* * .... * .....*.....* * .... * ** 1 1. The Peruvian Economic Situation, 1970-85.. . .. ....2 1.1 Growth, Inflation, and Income...o.e.o... s..oo .............2 1.2 Government Finances: Revenues, Expenditures, and Deficits ............4......... o ..... oo ... oo- oo......... 4 1.3 Government Revenues and Expenditure Patterns.........o ... 6 2. The Public Health Sector Expenditureo-o.. ....-...............10 2.1 The Peruvian Health Sector: An Overview........... 10 2.2 Government Expenditures on Health Programs............ ...14 2.3 Public Health Sector: Financing and Expenditure Patterns ..... . .... . ... o. .... o.... .. .17 2.4 Corporate Health Sector: The Social Security System... . ..... 20 2.5 Private Expenditures on Health-Related Goods and Services.o ... _....... . .... .o27 3. Health Indicators: Distribution of Manpower and Facilities..... .. --.....oooo_ 32 3.1 Health Status: Morbidity and Mortality Rates.32 3.2 Coverage and Coverage-Related Indicators.oo.......... ....38 3.3 Poverty, Coverage, and Regional Disparities ... 41 4. Summary and Conclusions...o . ... . .. .. 48 Appendixo.,*.. .. .. ... ... -. . .... 549 References~~~~~.0 o: : : : : . . . 55  INTRODUCTION This paper reviews the health status of the Peruvians and the financing and spending patterns of Peruvian health institutions. The paper analyzes the main trends and relative importance of institutions providing health services and explores such alternative financing policies as users' fees, cost-recovery programs, and community participation. It also analyzes the potential for implementing risk-sharing programs. Although the emphasis is on public institutions, the study presents data on corporate health institutions -- institutions that include the Peruvian Institute of Social Security (IPSS), the for-profit and nonprofit organizations of the private sector, and the health services of the army, corporate government, and private institutions. The paper has four sections. Section 1 describes the evolution of the Peruvian economy during the last fifteen years. Section 2 analyzes the relative importance of each part of the health sector in terms of coverage, number of facilities, and financial resources. Section 3 presents the main indicators of the health status of the population, the major health problems affecting the Peruvian population, and the issues associated with the distribution of facilities, manpower, and financial resources. This section also reports on some of the findings from studies using data from the 1984 Nutrition and Health National Survey (ENNSA) in Peru. Section 4 summarizes the findings on the major problems affecting the health sector.  -2- 1. THE PERUVIAN ECONOMIC SITUATION, 1970-85 1.1 Growth, Inflation and Income The dynamics of Peru's economic growth during the last twenty years have been shaped by two completely different approaches to managing the economy: a period of nationalistic popular reforms from 1968 to 1975 followed by a period of stabilization, structural adjustment, and liberalization from 1975 to 1985. During the first period, a self-proclaimed Revolutionary Government of the Armed Forces seized power and promised to implement drastic social reforms such as nationalization, agrarian reform, educational reform, worker participation in the management of firms (Comunidad Industrial), and promotion of cooperatives and "social property." The government undertook large investment projects and nationalized certain strategic industries. It introduced subsidies for oil, gasoline, and basic staple foods; prices of other basic products were also controlled and/or heavily subsidized. The government severely restricted foreign trade by increasing the existing tariff rates and by introducing many nontariff restrictions. The second period began in 1975 with a coup d'etat against the reformist military government. A group of more conservative militarists seized power and called for a return to a more orthodox management of the economy, with more reliance on the private sector. In an attempt to reduce government deficits and disequilibrium in the external sector, the new government drastically reduced subsidies and dismantled several social reforms of the first period. The government was committed to a program of stabilization and the reorientation of the economy toward a free-market strategy of growth. A new civilian government came to power in 1980. Without  - 3 - changing the basic orientation in the management of the economy this government began consecutive drastic but unsuccessful stabilization programs. It also began a medium-term strategy of structural adjustment leading toward trade liberalization. Data in Table 1.1 show the behavior of some of the macroeconomic aggregates during the two periods. During the 1970-75 period of popular reforms, average yearly rate of gross domestic product (GDP) growth was 4.8 percent, slightly below the historically high rate-of 5.5 percent a year in the fifties and sixties. With a population growth rate of 2.7 percent a year, per capita income increased at an average rate of 2.04 percent. Also during this period inflation rose from a historically low yearly rate of about 5 percent to 13 percent in 1975 and 30 percent in 1976. During the 1975-85 adjustment and liberalization period, while the population growth rate declined to 2.6 percent, the average rate of GDP growth dropped to 1.23 percent a year; per capita income declined by an average yearly rate of 1.23 percent. Inflation skyrocketed from 30 percent during the mid-seventies, to 60 percent in 1980, 110 percent in 1984, and 170 percent in 1985. Economic recession was particularly severe during the last five years of the structural adjustment and trade liberalization programs. Between 1980 and 1985 the economy remained almost stagnant; GDP fell at a rate of 0.7 percent a year, and per capita income declined even faster at 3.4 percent a year. In 1985 income per capita was about 6 percent lower than at the beginning of the seventies.  - 4 - TABLE 1.1: Peru: Gross Domestic Product, Inflation, Population, Per Capita Income, 1970, 1975, and 1980-85 1970 1975 1980 1981 1982 1983 1984 1985 Real GDP,(bill. 1980 US$) 6.2 12.2 14.5 16.7 14.4 11.1 10.0 14.2 Rate of growth (p.a. real terms) 5.0 -0.5 0.1 3.1 0.6 -12.5 4.4 1.9 Inflation rate 5.0 13.0 59.2 75.4 64.4 111.2 110.2 169.9 Population (millions) 12.8 14.6 16.6 17.0 17.4 17.9 18.4 18.9 Income Index (1970=100) 100.0 110.7 113.9 114.6 112.8 95.9 97.4 96.7 (real terms) Source: Elaborated data from INE (1986); Suarez, R. (1986) 1.2 Government Finances: Revenues, Expenditures, and Deficits Since the mid-seventies, management of government finances and control of government deficits have been among the key issues of Peru's stabilization programs. Data on the evolution of government finances are presented in Table 1.2. During the first phase of the military government, after two years of austerity measures and policy reforms (1969-70), the government pursued expansionary fiscal policies. Government expenditures, as a proportion of the GDP, increased from an average of 16-18 percent during previous years to above 20 percent during the early seventies. Revenues did not increase in proportion, and domestic and foreign borrowing were heavily used to finance rising government deficits.  - 5 - During the stabilization and liberalization period, efforts to reduce government expenditures were unsuccessful. At the beginning of the seventies, government expenditures represented about 17 percent of GDP; in the mid- seventies, they rose to almost 20 percent and have remained around this level in the eighties. For examples, in real terms, government expenditures increased considerably. In 1981 government expenditures in real terms were almost twice the level observed in 1970. Between 1975 and 1980, government revenues also rose substantially. From 1980 on, however, while government expenditures remained high, government revenues began to fall. TABLE 1.2: Peru: Public Sector Finances, 1970, 1975, 1980-85 (millions of intis) 1970 1975 1980 1981 1982 1983 1984 1985 GDP 1980 intis (000') 3364.7 4247.4 4971.8 5123.3 5168.9 4549.6 4765.8 4842.4 Real Rate of growth 5.0 -0.5 0.1 3.1 0.6 -12.5 4.4 1.9 GDP Current Prices 240.7 550.2 4971.8 8375.3 13777.0 25334.0 57114.2 155293.5 Deflactor GDP 7.2 13.0 100.0 161.5 266.5 556.8 1198.4 3207.0 Central Government Total Expenditure 42.1 106.7 1046.7 1830.2 2634.0 6048.0 10728.3 23869.0 Total Revenue 38.8 88.6 1008.4 1509.7 2459.6 3732.0 5228.1 21667.0 Deficit -3.3 -18.1 -38.3 -320.5 -174.4 -2316.0 -5500.2 -2202.0 Expend. (1980 intis) 588.5 823.7 1046.7 1133.1 988.2 1086.1 895.2 744.3 Revenues (1980 intis) 542.4 684.0 1008.4 934.7 922.8 670.2 436.2 675.6 Expend. Index (1970=100) 100.0 140.0 177.9 192.5 167.9 184.6 152.1 126.5 Revenue Index (1970=100) 100.0 126.1 185.9 172.3 170.1 123.6 80.4 124.6 Total Government Expend/GDP (%) 17.5 19.4 21.1 22.1 19.1 23.9 18.8 15.4 Deficit/GDP (%) -1.4 -3.3 -0.8 -3.9 -1.3 -9.1 -9.6 -1.4  -6 Government deficits, around 3-4 percent of GDP between 1971 and 1974, increased to 6 percent in 1976 and 7.5 percent in 1977. At this time the government attempted to control government expenditures. Favorable export prices, resulting in additional export tax earnings, were used in part to balance the budget. Thus government deficits, as a proportion of GDP, were reduced to 4.7 percent in 1978, 0.5 percent in 1979, and 0.8 percent in 1980. During the eighties, amidst a process of structural adjustment and liberalization, inconsistent expansionary fliscal and monetary policies were pursued. From 1980 to 1984 high government expenditures continued; in 1984 total government expenditures represented almost 24 percent of GDP. An unsuccessful reform of the tax system and the economic recession resulting from falling terms of trade led to a drastic reduction of government revenues. Deficits rose sharply from 2.8 percent of GDP in 1980 to more than 9 percent in 1983 and 1984. Recurrent government deficits were financed by borrowing in domestic and foreign financial markets. Deficits were financed by an expansion of the money supply and intensive use of foreign financing, leading to severe inflation and an acute external debt. This evolving pattern of government finances explains the most important changes in the makeup of government expenditures, analyzed in the following section. 1.3 Government Revenues and Expenditure Patterns Tables 1.3 and 1.4 present data on the composition of government revenues and expenditures for the years 1973, 1975, 1980, and 1982. Changes in revenue and expenditure patterns suggest that some important changes in income distribution might have occurred during these years. Reform of the tax  - 7 - system during the eighties reduced the usually progressive income tax and increased the usually regressive taxes on goods and services. Overall, during 1973-82, tax revenues were the main source of government financing; taxes represented around 90 percent of total government revenues. Non-tax revenues, comprising fees on government services and grants, represented only 10 percent. Government revenues from income taxes declined from 24.4 percent in 1973 to 15 percent in 1982; between 1973 and 1981 revenues from income taxes on individuals declined from 5.9 percent to 1.6 percent. Revenues from the general tax on goods and services for domestic and international transactions increased from 54.4 percent in 1975 to 71.4 percent in 1982. Revenues from taxes on payroll and from non-tax sources remained relatively constant. There are no revenues from the social security system because in Peru social security contributions are not paid to the government; they are directly paid to the Peruvian Institute of Social Security (IPSS), an autonomous institution. Increases in government expenditures have not been uniform for all government functions. Between 1973 and 1981 the most important changes were reductions in the share of government expenditures on such social programs as health, education, housing and community activities. Expenditures for these social programs declined from approximately one-third of the total budget in 1973-75, to less than one-fifth in 1981. The budget for education was reduced the most; less significant reductions were made in defense and general public services. In 1981 a single item--other purposes--absorbed the largest proportion of total government expenditures. This item comprises mainly the  -8- interest and amortization payments on domestic and foreign public debt. Debt- related payments increased from 10 percent of government expenditures in 1973 to 21 percent in 1981. Estimates for 1984/85 show that these payments represented 25-27 percent of total government expenditures. 1/ TABLE 1.3: Peru: Structure of Government Revenues, Selected Years (millions of intis) Government Revenues 1973 (%) 1975 (%) 1980 (%) 1982 (%) Total Rev. and Grants 51.9 100.0 88.6 100.0 1008.4 100.0 2459.6 100.0 Tax Revenues 47.3 91.0 79.5 89.8 936.7 92.9 2232.7 90.8 Income Tax 12.7 24.4 21.0 23.7 267.9 26.6 369.6 15.0 Individual 3.1 5.9 5.0 5.6 16.4 1.6 Corporate 8.8 17.0 11.1 12.5 247.8 24.6 Social Security 0.0 0.0 0.0 0.0 0.0 0 0 0.0 0.0 Employers' Payroll Tax 1.8 3.4 3.4 3.8 33.9 3.4 108.8 4.4 Taxes on Property 1.9 3.6 3.5 4.0 56.2 5.6 146.3 5.9 On Goods and Services 18.6 35.8 31.8 35.9 380.4 37.7 1125.7 45.8 On Intntl. Transac. 9.6 18.6 20.8 23.5 283.0 28.1 630.8 25.6 Other Taxes 2.7 5.3 0.8 0.9 5.6 0.6 25.1 1.0 Non-Tax Revenues 4.5 8.7 9.0 10.2 70.5 7.0 221.3 9.0 Grants 0.1 0.2 0.1 0.1 0.9 0.1 5.7 0.2 Source: Elaborated from IMF (1984) p.645-47.  - 9 - TABLE 1.4: Peru: Government Expenditure Patterns, by Functions (millions of intis) Expenditure by Functions 1973 (%) 1975 (%) 1980 (%) 1981 (%) Total Expenditure 64.2 100.0 106.7 100.0 1046.7 100.0 1830.6 100.0 General Public Service 8.8 13.6 13.1 12.2 129.2 12.3 191.2 10.4 Defense 9.8 15.2 16.7 15.7 130.7 12.5 252.9 13.8 Education 14.7 22.9 21.9 20.5 116.0 11.1 207.6 11.3 Health 3.5 5.5 5.5 5.1 47.4 4.5 97.1 5.3 Social Sec. & Welfare 0.2 0.2 0.3 0.2 0.0 2.8 0.2 Hous. & Commut. Act. 2.5 3.9 2.6 2.4 15.6 1.5 16.9 0.9 Other Community Serv. 0.0 0.0 0.0 23.0 1.3 Economic Sectors 14.4 22.4 26.3 24.6 0.0 0.0 Other Purposes 7.2 11.2 20.5 19.2 202.1 19.3 389.8 21.3 (public debt payments) Total Social Programs* 20.9 32.6 30.2 28.3 179.0 17.1 324.4 17.7 * Education, Health, Social Security and Welfare and Housing. Source: Elaborated from IMF (1984) p.645-47. In summary, the more intensive use of the general sales tax, rather than the income tax, and the reduction of government expenditures on social programs seem to have adversely affected income distribution. There is general agreement on the effects of this type of change in the system on income distribution; however, the effects on distribution that result from changes in government expenditures on social programs are less clear. Data on the changing patterns of government revenues and expenditures also suggest that even when maintaining the regressive general sales tax, more revenue can be collected from individual and corporate income taxes. The problem seems to be one of political feasibility: increasing the tax burden of the politically strong high- and middle-income groups. Also, redistribution of expenditures toward social programs requires reducing the budget for defense and general services, reducing expenditures in the economic  - 10 - sectors, or reducing payments on the outstanding public debt, thus reducing the "other purposes" expenditures. The alternative for the current government has been to limit the amount of resources to be used for amortization payments to foreign creditors and to reallocate these resources for social and economic programs. A short- term consequence of limiting such payments has been to have the Peruvian government declared "ineligible" for new loans by international organizations. Access to international capital markets has also been severely curtailed. Medium-term consequences of this situation are not yet foreseeable. 2. PUBLIC HEALTH SECTOR EXPENDITURE, 1970-85 2.1 The Peruvian Health Sector: An Overview In theory Peru's National System of Health Services (Sistema Nacional de Servicio de Salud) is in charge of coordinating the health programs of government and nongovernment health institutions: the Ministry of Health (MOH), the Health Services of the Army and the Police, and the private health sector. But, as stated in the last National Health Plan, these institutions act independently of the National Health Services, and their actions lack coordination. In practice they are autonomous institutions with their own programs. The only institution under direct government policy intervention is MOH. 2/ To analyze the financing of the health sector, Peruvian health institutions are classified into three groups: (1) public health sector, comprising all institutions providing both preventative and curative health services to the general public. Access to these services is determined mainly  - 11 - by space availability and quotas; in some cases these institutions charge a nominal fee. (2) The corporate health sector, comprising institutions providing essentially curative health services to associated members or employees; payment for services are included as fringe benefits or retained from wages or salaries as contributions to a risk-sharing fund. Only the insured or affiliated members have access to these services. (3) The private health sector, comprising those individuals and institutions providing formal and informal curative care by implicit or explicit price-rationing schemes; access to the system is based on space availability and price rationing, or ability to pay. This classification distinguishes among the types of health services provided by the various health institutions and identifies factors determining supply of (provision) and demand for these services. The first broad distinction is preventative and curative health services. Preventative includes those actions - that by their nature - can be classified as public goods. Marketable preventative actions will be considered as curative services. They can be thought of as a composite commodity consisting of medical care and instructions, drugs, and other factors affecting a particular treatment. In Peru, preventative actions are the responsibility of public health institutions: MOH and local government institutions. Curative services are provided by all the institutions of the health sector. Public and private health institutions provide only medical care; corresponding drugs and other health-related goods and services are the patient's responsibility (for outpatients). The corporate health sector, however, provides both medical care and prescription drugs. These are important issues that should be considered in modeling health  - 12 - services and in evaluating the efficiency of each health institution. Table 2.1 presents rough estimates of the relative importance of the three health sectors in the Peruvian economy by the number of centers, for example, hospitals and sanitary posts, and by the proportion of the population covered by each of the sectors. 3/ MOH is the most important institution of the public health sector both in the number of centers under its supervision and the number of people served. Sociedades de Beneficencia Publica and the local governments have fewer facilities; however, they provide services to relatively large segments of the population. As a whole, public health institutions have about one- third of the total number of hospitals, more than 70 percent of the health- care centers, and about 90 percent of the sanitary posts. These institutions serve an estimated 56.5 percent of the total population. The corporate health sector, consisting of public- and private-funded institutions, provides health services to affiliated employees, members, and their relatives. These include IPSS; the Health Services of the Army and the Police (Sanidad de las Fuerzas Armadas y Fuerzas Policiales); and such state-owned health institutions and private enterprises such as Minero Peru, Southern Peru Copper Corporation, Marcona, Centromin, and Collective Land Ownership Organizations (SAIS). With the exception of IPSS and the Army, most corporate health facilities belong to private, state-owned, or collective enterprises. Although these services are not open to the general public, people covered by specific corporate health programs might have access to IPSS services and to services from public and private health institutions. Although some of these corporate health institutions might have some extension programs that cover a broader group of people, they basically provide medical services to their associates or affiliated workers.  - 13 - TABLE 2.1: Peru: National Health System, Institutions and Coverage: 1983-84 Health Sanitary Covered Health Sectors Hospitals Centers Post Other Population (%) (000) Public Health Sector 116 463 1405 13 10844.6 56.5 Ministry of Health 109 451 1402 9 10046.6 52.3 Soc. Benefit Public 5 1 1 500.0 2.6 Other Public Institutions 2 7 2 3 48.0 0.3 Local governments 4 1 250.0 1.3 Corporate Sector 98 149 130 0 3190.8 16.6 Social Security 18 67 17 2680.0 14.0 Army and Police 13 54 65 340.0 1.8 State Owned Firms 10 4 7 60.0 0.3 Cooperatives, Agricultural 13 7 16 56.0 0.3 SAIS, Agricultural 1 11 19.8 0.1 Private Firms 43 17 14 35.0 0.2 Other Ministries 25.0 0.1 Private Sector 116 18 3 4 346.9 1.8 Private Institutions 111 9 4 251 1.3 Other Non-Gov. Institutions 6 0.0 Non Profit, Private 5 3 3 23.4 0.1 Private Insurances 72.2 0.4 TOTAL 330 630 1538 17 14382.3 74.9 Total Population 19197.9 100.0 Non-covered 4815.6 25.1 Source: Elaborated from BCR (1984) p.7; ANSSA (1985).  - 14 - The private health sector includes for-profit and nonprofit institutions and individuals who provide health services to the general public through a price-rationing system. Although the number of private hospitals is similar to the number of hospitals in the public health sector, it is estimated that the private sector serves a relatively small proportion of the population, in contrast to the relatively high segment covered by the corporate and public health institutions. 2.2 Government Expenditures on Health Programs A large proportion of the expenditures on health in countries like Peru comes from government expenditures on health programs. Furthermore, even though demand or expenditures for health services might be dependent on income, on the supply side, government decisions about provision of health services are influenced mainly by policy decisions on the role of the government in the economy, priorities assigned to social programs, and the financial constraints faced by the public sector. On the demand side, because services are provided free of charge, the main issue becomes the distributive impact of government expenditures on health, that is, distribution of health centers and accessibility of public health services to different social groups. Table 2.2 shows the evolution of public sector expenditures and the amount of resources assigned to the MOH budget. Data show that total government expenditures as a proportion of GDP increased from 18.6 percent in 1970 to an average of 20 percent between 1975 and 1984. In 1985 the percentage declined to 15.4. In real terms, expenditures throughout the  - 15 - eighties have been significantly higher than expenditures at the beginning of the seventies. Public expenditures on health show a different pattern evolving. The amount of resources devoted to MOH, as a percentage of the government budget, declined from 6.4 percent in 1970 to around 4.5 percent during the last ten years. As a proportion of GDP, government expenditures on health declined from 1.2 percent during the seventies to 0.6 percent in 1985. Although total government expenditures show a slight increase in relation to that of the seventies, expenditures on health programs in real terms, after significantly increasing during 1980-81, declined approximately 23 percent from the level in 1970. The evolving pattern of government expenditures on health programs is reflected in the real public health expenditure priority index (see figure p.16). The index increased from a base of 100 in 1970 to 122 in 1981 and 128 in 1982, then declined to 77 in 1985. 41  - 16 - TABLE 2.2: Peru: Government Revenue and Expenditure in Public Health Programs, 1970-85 Central Government 1970 1975 1980 1981 1982 1983 1984 1985 Total Expenditure 44.8 110.7 950.9 1859.8 3071.8 5166.5 10728.3 23869.6 Total Govern. Expend/GOP (%) 18.6 20.1 19.1 21.9 21.6 19.5 18.9 15.4 Government Expend (1980 intis) 626.3 854.6 950.9 1122.2 1113.2 879.7 889.0 741.3 Index, 1970 = 100 100.0 136.5 151.8 179.2 177.7 140.5 142.0 118.4 Ministry of Health, Total Expenditure (billion intis) 2.9 4.9 32.1 81.2 141.9 211.8 472.3 995.8 Total Expenditure (million US$) 74.2 108.9 93.9 160.2 143.34 93.3 82.9 91.4 (% of GDP) 1.2 0.9 0.6 1.0 1.0 0.8 0.8 0.6 (% of Total Government Expend.) 6.4 4.4 3.4 4.4 4.6 4.1 4.4 4.2 Gov. PHS Expend. (1980 intis) 40.1 37.8 32.1 49.0 51.4 36.1 39.1 30.9 Index, 1970=100 100.0 94.3 80.0 122.1 128.2 89.9 97.5 77.1 PHS Expend. per capita 3.1 2.6 1.9 2.9 3.0 2.0 2.1 1.6 Per capita index (1970=100) 100.0 82.7 61.7 92.0 94.3 64.3 67.9 52.3 Sector Priority index 100.0 69.1 52.7 68.2 72.1 64.0 68.7 65.1 (base 1970=100) PHS Expenditure Indicators (Index 1970=100, 1980 intis) 170- 1 &-_I - './ 140 O 1,3 /_ elk ~ ~ ~ en 110 \ / +7 70 -~V ' 0 \ ___ 'I70 9:5 19- 191 1S.5193 184 18 'N.. / .~~Y-.:3N Evp. Indome + PHSE R~nr.-mp 1rr-J--x Priarii~y irN  - 17 - Per capita public expenditures on health, assuming universal coverage, declined from a high of US$9 during 1980-81 to less than US$5 over 1984-85. By excluding those not covered (see Table 3.1) and by assuming a constant coverage rate, per capita expenditures would have been US$12 in 1980- 81 and US$6.6 in 1984-85. Excluding the population receiving curative services from corporate and private health institutions and assuming a constant rate of coverage during 1980-85, per capita expenditures on the 52 percent of the population covered by MOH would increase to US$17 for 1980-81 and US$9.6 for 1984-85. Although data on the proportion of government expenditures on preventative health care is not readily available, such expenditures do not seem to be an important part of the government budget; instead a large proportion of the resources are spent on curative services. 2.3 Public Health Sector: Financing and Expenditure Patterns The relative importance of government finances in the budget of MOH is presented in Table 2.3. The central government is the main source of MOH financing. During the last two years the central government provided more than 85 percent of the financial resources for MOH. Foreign borrowing was the next source, providing about 7 percent of the funds. Ddnations and funds from "other" services represented about 3-5 percent of MOH resources. Borrowing from domestic sources and revenues from service fees were minor, representing about 3 percent of the total budget. Aggregate estimates of the sources of financing, based on actual MOH revenues, show that central government financing and MOH revenues represent slightly higher proportions than those contained in the planning budgets; they also show that borrowing figures were usually overestimated. 5/  - 18 - TABLE 2.3: Peru: Ministry of Health, Sources of Financing, 1984-85 (billions of Soles) Sources 1984 % 1985 % Central Government 402.7 85.3 853.2 85.7 Revenues 13.9 2.9 31.9 3.2 Borrowing Domestic 0.9 0.2 - - Foreign 39.5 8.4 63.2 6.3 Donations 7.4 1.5 13.1 1.3 Others 7.8 1.6 34.4 3.5 TOTAL 472.3 100.0 955.8 100.0 Source: Ministerio de Salud (1985), Table 45. These data imply that despite the severe reduction in the amount of real resources transferred from the central government, the public health sector has been unable to obtain or use alternative sources of financing, for example, user fees, financing from local governments, contributions from institutional users. Although introducing such alternative sources of financing is difficult in periods of economic recession, these sources might have contributed to sustaining higher revenue levels. The composition of MOH actual expenditures is presented in Table 2.4. The major change over time is the increase in current expenditures at the expense of capital expenditure accounts. Current expenditures increased from 85 percent in 1980 to around 90 percent in 1982 and 1984.  - 19 - TABLE 2.4: Peru: Ministry of Health, Actual Expenditure Patterns, 1980, 1982, 1984 (z) Expenditure Items 1980 1982 1984 Current Expenditure 84.1 91.0 89.6 Wages and Salaries 52.5 70.0 69.7 Supplies and Materials 25.1 21.2 16.6 Services 3.1 3.5 3.6 Transfer 14.8 1.6 1.4 Pensions 4.4 3.6 8.7 Capital Expenditure 15.9 8.9 10.4 Research 2.6 3.5 3.3 Constructions 84.8 82.5 56.9 Equipment and Durables 7.5 14.0 39.0 Transfers 5.1 3.4 0.8 TOTAL 100.0 100.0 100.0 (Real terms, millions of intis, 1980) (49.0) (61.5) (58.0) Source: ANNSA (1985), Tables 4.7.4, 4.7.6, pp. 100,102. Among the current expenditures, the payroll, including wages, salaries, and pensions, is the most important increasing from 52.5 percent of total expenditures in 1980 to around 70 percent during 1982-84. The more drastically reduced current expenditures are those of supplies and transfers. 6/ Reductions in expenditures on supplies and medicine suggest that the health service provided by this institution is basically medical care. Drugs and materials that are part of the treatment are the responsibility of patient.  - 20 - 2.4 Corporate Health Sector: The Social Security System The most important corporate health institutions in Peru are IPSS, SAID, the Health Services of the Army and Police, hospital and clinic cooperatives, private health institutions, and decentralized state-owned enterprises. The social security system in Peru dates back to 1850, when retirement and unemployment pension plans were created for civil servants and army personnel. In 1911 a risk-sharing plan for blue-collar workers was introduced. 7/ The first social security program that covered medical assistance and maternity for blue-collar workers was created in 1936. In 1948 this system was expanded to include government and private-sector white-collar workers, and in 1950 to include army personnel. During the sixties new social security plans were created that extended coverage to other workers. These offered different types of contributions and benefits, and their funds were administered independently of IPSS. Special pension plans and contribution requirements were created for special interest groups; for example, employees from the executive, legislative, and judiciary branch of the government; other elected representatives; workers from the central bank; foreign service employees; and workers from the decentralized government institutions were all under different contribution and benefit plans. Also, special benefit plans were created for individual groups of blue-collar workers, for example, taxi drivers, stevedores, domestic workers, fishermen, independent and self- employed workers. A unification of the social security programs began in the early seventies. In 1972 and 1973 retirement and pension plans of all branches of  - 21 - the army and those for the blue- and white-collar workers from the public and private sectors were unified. In 1979 health-care programs and maternity care for blue- and white-collar workers from the public and private sectors were also integrated. 8/ In theory by 1980 all the social security programs, with the exception of the army's, were unified into IPSS. However, despite the efforts at unifying the system, the old differentiation between white-collar and blue- collar health institutions persists. Although important gains have occurred in making medical care and maternity programs uniform, Peru's social security system is still highly stratified, and its coverage is limited. Recent reforms of the current government are directed toward effective unification of the system and toward provision of medical care and other social security programs to all the social groups; however, little progress has been made toward this goal. Studies on income distribution in Peru show that the social groups covered by the social security system are mainly urban groups who have formal jobs in the modern sector of the economy. They include military personnel, public and private sector white-collar workers, and blue-collar workers from the largest modern-sector firms. They constitute the middle- and high-income groups in the upper quartile of the income distribution. 91 Data on the budget of the social security system, its proportion of GDP, and the growth of the system's financial resources in real terms are presented in Table 2.5.  - 22 - TABLE 2.5: Peru: Social Security System, Total Expenditures, 1961-82 (Millions of intis) Med. & Maternity IPSS Real Terms % of Expenditure** % of Year Total Budget Index* GDP % GDP 1961 1.7 2.3 N.A. 1965 3.3 2.5 N.A. 1970 - - - - 1975 19.4 100.0 3.1 56.7 2.0 1980 179.3 137.7 3.2 65.2 2.4 1981 337.9 160.6 3.6 66.9 2.7 1982 521.9 140.0 3.4 63.1 2.3 * Index 1975=100.0 ** Includes monetary reimbursements. Source: Elaborated from Cepal (1985) p.332-333, and IMF (1986). In real terms IPSS's financial resources increased from 19.4 millions in 1975 to 27.2 millions in 1982. The medical care and maternity program was the most important in terms of expenditures absorbing approximately two-thirds of the total resources. During 1980-82 the IPSS budget was more than four times that of MOH. Overall expenditures on maternal and medical assistance programs for this period were around 2.7 times that of the MOH budget (see Table 2.2). The social security system is financed by contributions from workers, employers, and the government. Workers' contributions are retained from their wages and salaries by their employers who are responsible for paying their own and their workers' social security contributions. The government contributions to the social security system are as an employer and through transfers to the system.  - 23 - The share of contributions to the social security system by employees, employers, and the government are presented in Table 2.6. In 1983 workers' contributions represented 5 percent of the "basic wages or salaries," employers' shares were 14 percent, and government contributions were 2 percent (as transfers). Independent and self-employed workers' contributions were 15 percent of their reported monthly income (minimum and maximum income levels are established). By law, all wage-earners or salaried employees from cooperatives and public or private enterprises have to be registered in the social security system. For independent workers, participation in the social security system is optional.  - 24 - TABLE 2.6: Peru: Contributions to the Social Security System, by Programs, 1983 (percentage of wages) Self-Employed & Programs Employees Employers Government Independent Total Pensions 2.5 5.0 1.0 7.5 8.5 Medical & Maternity 2.5 5.0 1.0 7.5 8.5 Professional Risk - 4.0 c/ - - 4.0 TOTAL 5.0 14.0 2.0 15.0 21.0 a/ Variable rates, with minimum and maximum contributions. b/ Excluding independent workers. c/ National average, rates range from 1 to 12.5 percent, depending on type of risk. Source: From Cepal (1985) p. 328 and SAA (1985) p.56-57. The medical and maternity programs provide coverage for the spouse, concubine, and children under one year of age. But, even the system's coverage for dependents is limited. Table 2.7 presents gross estimates of the coverage of the medical and maternal assistance social security programs. Health services to those affiliated with IPSS is provided directly through IPSS's medical posts, clinics, and hospitals, and through contracts for services from private providers. For some groups of white-collar workers there is a "free-option" system by which medical care can be received from private practitioners or clinics. Reimbursement rates for health-related goods and services are established by IPSS. As in most Latin American countries, health services for military personnel are separated from the rest of the social security system. Within the military, there are separate arrangements for the air force, navy, army, and the police; each has its own private hospitals, medical services, and programs.  - 25 - TABLE 2.7: Peru: Coverage of the Social Security System, 1961-83 (millions) % of Economically Economically Total Active IPSS IPSS % of Total Active Year Population Population Actives Total b/ Population Population 1961 10.2 3.3 0.8 0.9 8.8 24.2 1965 11.5 3.7 1.2 1.3 11.3 32.4 1969 12.8 4.2 1.5 1.6 12.5 35.7 1975 15.2 4.9 1.7 2.3 15.3 34.7 1980 17.3 5.6 2.3 2.9 17.3 41.1 1981 17.7 5.8 2.4 3.1 17.4 41.4 1982 18.2 6.0 2.4 3.1 18.1 40.0 1983 18.7 6.2 2.5 3.2 18.7 40.3 1984 19.2 6.4 2.6 3.4 17.7 40.6 1985 19.7 6.8 2.7 5.2 26.3 39.7 a/ 1961-1969 coverage of pension program. Excludes fishermen and jockeys. 1975-1980 coverage of health programs including estimates of army and fishermen. From 1980 on, includes domestic workers, affiliated members (optional) and pensioners. b/ 1961-1969 includes only spouses of white collar workers with maternity assistance. Since 1975, includes spouses or concubines and children less than one year old with access to the benefits. Inactive 1961-1969 includes white and blue collar workers in pension programs, since 1975 includes workers with professional risk coverage, excluding army, fishermen and jockeys. Source: Elaborated from Cepal (1985) p.325-326 and INE (1986) pp.33, 66, 138.  - 26 - The limited number of people served by the social security system in Peru is explained in part by the relatively low proportion of workers engaged in formal wage- or salary-stipulated contracts, by the relatively high cost of the contributions required of independent or self-employed workers who constitute a large segment of the workers in urban areas, and by the large proportion of workers who earn below the stipulated minimum wage used for calculating the minimum legal contributions. Finally, for workers in rural areas who form about 30 percent of the labor force, only workers of large cooperatives and SAID receive regular wages, and most of these cooperatives have their own medical posts or clinics for regular medical assistance. The large proportion of the rural population, however, is out of the scope of the social security system. Combined data from Table 2.6 on the coverage of the medical assistance and maternity program and data from Table 2.5 on the IPSS budget show that between 1975 and 1982, total expenditures in these programs represented between 2.0 and 2.7 percent of GDP. In real terms, expenditures on these programs increased from about 85 millions in 1975 to nearly 120 millions in 1982. Expenditure per affiliate and dependent members, after a significant increase from US$106 in 1975 to US$136 in 1981, declined to about US$100 in 1982. Although little is known about expenditures in the other corporate sectors, it is generally perceived that their services are better because members choose the services even when given the option of using the IPSS facilities or public health sector services. The "better" services can be attributed to higher per capita expenditures and to more efficient provision of resources. These hypotheses can be investigated to compare the cost of  - 27 - services among the various corporate health institutions and to identify ways of improving the efficiency of the system. Assuming that per capita expenditures in the other corporate health institutions are similar to that of IPSS, total expenditure of the corporate health sector would be about 15 percent higher than the public health sector, representing about 2.7 percent of GDP. In summary, people covered by the social security system receive the equivalent of almost five times the medical care received by people in the public health sector programs. Part of the difference might be explained by the inclusion of prescriptions and materials provided by the social security and corporate institutions. Another explanation may be that social security provides a different type of health service than does the public health sector. Hospitals of corporate health institutions, especially those of the social security and the army, are known for having better and less crowded facilities and more modern equipment, which enables them to provide more sophisticated and expensive medical care. 2.5 Private Expenditures on Health-Related Goods and Services Because services provided by the public and corporate health institutions are free, household expenditures for health-related goods and services could be used to estimate the amount of resources spent on privately produced health goods and services. In this section the results of household surveys are used to provide rough estimates of the magnitude of the private health sector. The ECIEL (Programa de Estudios Conjuntos pare Integraci6n Econ6mica Latinoamericana) household survey on income and expenditure patterns in  - 28 - metropolitan Lima found that the average share of private expenditures on health-related goods and services, excluding contributions to social security and private medical insurance, represented about 2.2 percent of the household expenditure. Income elasticity of expenditures on health-related services estimated from the ECIEL survey was found to be less than one: 0.55 (Figueroa 1974). Other findings from the ECIEL survey show an inverse relation between family size and share of expenditures on health-related services. A significantly higher share of health-related expenditures was found in households in which the head of household was more than 65 years old retired. A slightly higher share of health expenditures was found in households in which the head of household was "less educated." Also the survey showed that there was a marked seasonality for expenditure patterns, with higher expenditures during autumn and winter months (May-November), and that the number of children (under six) did not seem to affect health expenditure shares. The National Household Consumption Survey (ENCA) of 1971-72 showed higher private medical expenditures. For metropolitan Lima, medical expenditures, including prescriptions and materials, represented on average 3.45 percent of the total household expenditure; both the lowest and highest expenditure shares were found in the higher income groups: from 2.52 to 4.50. The National Multipurpose Household Survey (ENAPRON) of 1977-78 showed that for metropolitan Lima and twelve of the largest cities, the average share of expenditures on health-related goods and services was 2.54 percent of the total reported income. This survey showed the lowest and  - 29 - TABLE 2.8: Peru: Per capita Income and Private Expenditure on Health, 1970-85 (intis) 1970 1975 1980 1981 1982 1983 1984 1985 GDP per capita (intis 000' real) 262.9 290.9 299.3 301.3 296.4 252.1 256.0 254.2 Index GDP per capita 100.0 110.7 113.9 114.6 112.8 95.9 97.4 96.7 GDP, per capita (US$) 485.9 837.4 875.9 985.1 826.6 651.8 542.3 751.0 Index, 1970=100 100.0 172.3 180.3 202.7 170.1 134.1 111.6 154.6 Private Expenditure in Health (per capita) Hypothesis A (ENCA) ($) 9.1 9.6 9.8 9.8 9.7 8.9 8.9 8.9 Hypothesis B (ECIEL) ($) 5.8 6.2 6.3 6.3 6.2 5.7 5.7 5.7 Index, 1970=100 100.0 105.9 107.6 108.0 107.0 97.7 98.6 98.2 In US $ Hypothesis A (ENCA) 16.8 21.0 21.5 22.8 20.9 18.8 17.5 20.0 Hypothesis B (ECIEL) 10.7 15.0 15.5 16.8 14.9 12.8 11.4 14.0 Index, 1970=100 100.0 139.8 144.1 156.5 138.6 118.8 106.4 130.0 Share from Income ($) Hypothesis A (ENCA) 3.5 3.3 3.3 3.3 3.3 3.5 3.5 3.5 Hypothesis B (ECIEL) 2.2 2.1 2.1 2.1 2.1 2.3 2.2 2.2 highest expenditure shares to be in the middle-income groups, with expenditure shares ranging from 1.86 to 2.83 percent (see CEPAL 1984). A survey on economic activities conducted between 1978 and 1979 in eight Andean rural communities in the south of Peru, the poorest region, found that expenditures on health-related services, medicine, and materials 10/ represented an average of 2.42 percent of the total monetary exports. - Using data from the ECIEL and ENCA surveys on expenditure shares and income elasticity, changes in total and per capita private expenditures on health between 1970 and 1985 were estimated (see Table 2.8). Expenditure shares and income elasticity were used to estimate average household  30- Private Per-capita Heaulth Ex:[Den di t 1Jr e- (In LU.S. ci3iforz) 13 -~3 7 _- 19g70 19g75 1 9s0 1 9e1 1 iso 198E3 1 g<4 19835 )ecs Oncorrre an theat ErExpendtr  - 31 - expenditure on private health-related goods and services. The results presented under Hypothesis A were derived using the ENCA estimates of average expenditures for 1970; results for the remaining years were derived using the income elasticity value obtained from the ECIEL survey. Estimates based on the ECIEL survey (Hypothesis B) produced the lowest per capita expenditure rates. These can be taken as the low benchmark estimates that show average individual household expenditures on private health-related goods and services. An implicit assumption of these estimates is that no changes in the structure of relative prices in the economy has occurred; that is, the relation between prices of health services and other commodities has remained constant. In general, however, the estimates are rough and thus should be studied with caution. 1l/ Data from Table 2.8 show that as per capita income fell from its 1981 peak, private expenditures on health goods and services also declined. In 1985 per capita expenditures in real terms was approximately 2 percent less than expenditures in 1970 and 10 percent less than the highest level, US$22, attained in 1981 (see index 1970 = 100.0). Since 1981 health expenditures have declined continually from between US$17 and US$23 in 1981 to between US$14.0 and US$20.0 in 1985. Adding expenditures on health-related goods and services and the public, corporate, and household expenditures on private health-related goods and services provides the estimate of the total amount of resources devoted to health. Table 2.9 summarizes indicators of the relative importance of each of these sectors by coverage and expenditure.  - 32 - TABLE 2.9: Peru: Suinmary Results, Composition of the Health Sector (around 1980-84) Expenditure Total Expenditure Sectorial Sectors Coverage US$ per capita % of the CDP Share (M) PHS 56.5 10 to 17 0.6 - 0.8 - 10 CHS 16.6 100 to 130 2.3 3.1 - 45 PS ? 11 to 20 2.1 - 3.5 - 45 TOTAL 100.0 55 to 77 5.0 - 7.0 100 Although most policy debates on financing health programs concentrate on financing the public health sector, this sector's expenditures represent only about 10 percent of total health expenditures. Expenditures of the corporate health sector and private households on health-related goods and services represent about 90 percent of total expenditures. These results show a further need to explore the role that the corporate and private sectors could have in implementing health programs. 3. HEALTH INDICATORS: DISTRIBUTION OF MANPOWER AND FACILITIES 3.1 Health Status: Morbidity and Mortality Rates The main indicators of the health status of the Peruvian population are presented in Table 3.1. This table summarizes the evolving pattern of the mortality, life expectancy, and infant mortality rates from selected years from the fifties to 1986. In 1986 average life expectancy at birth is estimated at 60.8 years, which is below the average life expectancy rate of 61.2 years for other Latin American countries and the average of 71 years for the developed countries (from 1970-75 estimates).  - 33 - TABLE 3.1: Peru: Evolution of Health Status Indicators, Selected Periods (per thousand) 1950-55 1975 1980-85 1986 Crude Birth Rate 47.0 39.4 37.0 35.0 Crude Mortality Rate 21.6 12.2 11.7 9.7 Infant Mortality/Rate a! 156.0 106.6 99.0 90.5 Fertility Rate 5.6 4.9 4.7 Life Expectancy at Birth 44.1 56.5 58.9 60.8 a/ Live birth, up to one year. b/ Per woman in child bearing age. Source: Elaborated from INE (1986, 1986c) and Ministry of Health (1986). As in most developing countries, the evolution of the life expectancy rate shows that after a significant increase in the life expectancy rate during the fifties and sixties, the rate of increase leveled off during the last decade. Cumulative increases in the life expectancy rates declined from 13 percent between 1960 and 1970 to less than 5 percent during the last ten years (1975-85). For developed countries the life expectancy rate leveled only after it reached 70 years. 12/ Data also show that whereas both birth and mortality rates have been declining, infant mortality remains high, which is the most important factor explaining the relatively low life expectancy and still high crude mortality rates. Peru's infant mortality rate (infants born alive and living to one year) is about 90 per thousand; it is one of the highest among Latin American countries and is in sharp contrast to the infant mortality rates of the most developed countries, whose rates range from 10 to 20 per thousand. Estimates  - 34 - of the death rate by age group show that, in 1964, about one-third of registered deaths were of infants. More than 50 percent of deaths were children under five years. Table 3.2 presents 1981 estimates of percentages of deaths by age and gender. Comparing these figures with those for 1964 suggests that although important gains have been made in reducing infant mortality, this group still accounts for almost one-fourth of total registered deaths; the under five years age groups account for about 36.5 percent of total registered deaths. 13/ Table 3.2 shows that no major difference exists in the proportion of deaths in each age group by gender. For both male and female groups, more than one-third of deaths are children under five years. The proportion of deaths remains between 4.1 and 4.5 up to the 45-54 years group; thereafter it starts to increase by age. TABLE 3.2: Peru: Percentage of Deaths by Age and Sex Groups, 1981 Age Group Total Male Female 0 to 1 23.2 24.3 22.0 1 to 4 13.3 12.7 14.0 5 to 14 4.1 4.1 4.0 15 to 24 4.5 4.8 4.2 25 to 34 4.3 4.5 4.1 35 to 44 4.5 4.5 4.5 45 to 54 5.5 5.9 5.1 55 to 64 6.8 7.4 6.2 65 to 74 8.8 9.2 8.4 74 and more 17.7 15.1 20.4 Not specified 7.3 7.5 7.1 TOTAL 100.0 100.0 100.0 Source: Elaborated from WHO (1985), Table 13.  - 35 - Table 3.3 and 3.4 summarize morbidity cases and mortality rates by type of illness for selected periods. 14/ The data for 1980-84 show that per 100,000 people, an average of 803 cases of illness were reported. About one-third of the reported cases were intestinal parasitic diseases. Dysentery, parasitosis, and other infectious diseases, including typhoid fever and typhoid-related diseases, represented more than 50 percent of the reported cases per 100,000 inhabitants. Also, although some improvements have been made in reducing the incidence of tuberculosis, tuberculosis and malaria cases are still high. The pattern in the incidences of malaria shows that from the fifties to the early sixties cases of malaria declined from 209.8 per 100,000 population to 19.3. During the early sixties, reported malaria cases declined to less than 2,000; in 1976 malaria increased to 4,000 cases, and in 1977 to 32,000 cases. At the beginning of the eighties, reported cases declined to 14,000, but in 1983 it again increased to 28,000 and in 1984 the figure was 26,000, averaging 135.9 cases per 100,000 people. Data in Table 3.3 are reported cases of an illness. Although some of the variation over the various periods can be attributed to early under- reporting of an illness, there are no trends that indicate under-reporting to be the major cause of the differences.  - 36 - TABLE 3.3: Peru: Morbidity, Reported Cases by type of Illness Selected Periods and Last Available Year (rate per 100,000) Type of Illness 1950-55 1960-65 1970-75 1980-84 1984 Typhoid and Related 52.9 50.2 117.0 86.7 Dysentery 188.3 173.9 37.5 33.3 Infectious Hepatitis 25.8 32.3 31.8 35.9 Parasitosis 100.4 125.1 233.3 234.7 Tuberculosis 213.4 235,4 145.7 137.7 113.1 Malaria 209.8 19.3 66.7 114.2 135.9 Poliomyelitis a/ 1.5 0.9 1.7 0.9 Measles (Sarampion) a/ 82.9 59.6 72.7 85.0 Diphtheria a/ 0.5 0.5 0.9 0.8 Tetanus a/ 2.4 2.3 2.3 2.3 Tosferina a/ 101.7 83.6 54.0 48.7 a/ Corresponds to 1968-70. Source: Elaborated from Ministry of Health (1986), Tables 3 to 6. TABLE 3.4: Peru: Principal Causes of Death, 1981 Causes of Death Total Male Female Infectious and Parasitic Diseases 19.3 19.4 19.1 Tuberculosis 3.9 4.3 3.4 Malignant Neoplasms 6.8 5.9 7.7 Circulatory System 11.9 11.2 12.6 Pneumonia 14.4 13.9 14.9 Bronchitis, Emphysema and Asthma 3.2 3.1 3.3 Digestive System 3.6 3.9 3.2 Perinatal Condition 8.5 9.1 7.8 Injury and Poisoning 6.9 8.1 3.3 Other 21.5 20.9 24.6 TOTAL 100.0 100.0 100.0 Source: Elaborated from WHO (1985), Table 13.  - 37 - Vaccination programs have not significantly reduced the incidences of diseases such as measles, polio, tetanus, and diptheria. Reported cases of polio, measles, and diptheria per 100,000 persons almost doubled from the 1970-75 period to the 1980-84 period. With the exception of tosferina, dysentery, and hepatitis, the number of cases for all the illnesses that can be controlled by immunization programs almost doubled. Table 3.4 presents data on causes of death by gender and type of illness. These data are consistent with morbidity patterns presented in Table 3.3. Infectious and parasitic diseases, other intestinal infections, and pneumonia are the main causes of death in both men and women and represent about 42 percent of the total registered deaths. About 70 percent of these cases are deaths of children under age five years. Pneumonia is the cause of 46 percent of deaths in this age group. Perinatal mortality accounts for 8 percent of total deaths. Table 3.4 shows little difference in mortality patterns by gender. Although there are some minor variations in the number of deaths for each disease, the same group of illness is responsible for about 80 percent of the registered deaths. The high incidences of infectious parasitic diseases, and other intestinal infections like dysentery, the morbidity and mortality patterns, and the age profiles of morbidity and mortality reveal that infant and child mortality is the most critical health problem in Peru. Most of these types of illnesses are related to lack of sanitation, inadequate waste disposal, and inaccessibility of safe drinking water. The general goal of a health program is to reduce morbidity and mortality rates. The above patterns suggest that the major causes of  - 38 - morbidity and mortality are environmental-related diseases; therefore, health programs should emphasize preventative measures that address this problem. This objective requires the allocation of health resources for both preventative and curative services. Available information suggests that concentrating resources on curative services might be inadequate. But without a proper knowledge of the relative costs of preventative and curative services, it is difficult to assess the inadequacies of resource allocation. Further work is needed in this area. 15/ Another issue is the role of the market mechanism in providing the necessary preventative health services that fall within the typical classification of public goods. Although individual efforts can be made to solve the sanitation and environmental problems, an effective solution requires a broader based community effort. 3.2 Coverage and Coverage-Related Indicators Adequate housing, access to safe drinking water, and appropriate waste disposal are often cited as preconditions for reducing infectious parasitic diseases, which in developing countries like Peru are among the main causes of illness among adults and death in children under five years. Inadequate housing and sanitary conditions adversely affect curative health programs and result in the persistence and spread of communicable diseases.16/ Table 3.5 presents data on the Peruvian population's access to adequate water supply and excretal disposal facilities.  - 39 - TABLE 3.5: Peru: Housing and Population by Type of Services, 1981 Water, Sewer. Water and Water and Only Only Without Type of Services TOTAL Total and Electric. Sewerage Electricity Water Electricity Services TOTAL 3257.1 1659.2 915.6 66.3 153.4 102.8 421.1 1597.9 Percentages 100.0 50.9 28.1 2.0 4.7 3.2 12.9 49.1 Population a! 1775.6 9044.2 4990.7 361.6 836.4 560.2 2295.2 8710.3 Notes a/ Population by type of service has been estimated using the national average of 5.451 persons per house. Source Elaborated from INE (1986c), p.79. The data in this table show that only one-half of the total inhabited houses had some type of services: water excretal disposal electricity. The proportion of houses with the three types of services was only 28 percent. The proportion of houses with both water and sewage disposal systems was only 30 percent, while the proportion of population with access to safe drinking water was a little higher, 38 percent. Data on the population living in other housing conditions are not readily available. The figures in Table 3.5 assume a constant number of members per dwelling (the national average of 5.4) and should be taken cautiously. By using the national average, distribution of the population classified as having access to the different types of services will be similar to the distribution of houses. Limited access to safe drinking water and adequate excretal disposal facilities is more acute in rural areas. Also important disparities exist among the various regions (see Section 3.3). WHO researchers compiled data for a group of specific preventative and curative primary health services included in the "Health for All by the Year 2000" program. 17/ Coverage was defined as the ratio of population  - 40 - receiving certain types of services to the population in need of these services. 181 Table 3.6 presents data on the coverage of various health- related services. One-hundred-percent coverage is one of the goals of WHO's Health for All by the Year 2000 Plan. The above data show several areas in which Peru is still far from these goals. Access to medical services is often measured by the ratio of population to various types of health personnel. Indicators of TABLE 3.6: Peru: Coverage and Coverage Related Indicators in Peru % Coverage with Antenatal Care 59 Attendance of Delivery 43 Child Care Vaccination 65 Vaccination with Third Dose 24 Endemic Disease AT (?) 37 Activity to Population Ratio, as % Consultations 40 Hospitalization 54 (?) Rough estimate based on partial information. Data for years prior to 1978. Source: Montoya-Aguilar, C. (1985), p.2. health manpower resources show that Peru is not far from the minimum standards set by WHO. The ratio of population to medical personnel in Peru is about 1300 persons per physician and 2,400 persons per nurse; minimum goals set by WHO are 1,250 per physician and 2,220 per nurse.  - 41 - The above indicators are not only guidelines for establishing priorities, they are also essential for designing policy objectives and programming optimal allocations of resources among the various health programs according to specific objective functions. These functions, defined as quantitative goals, coverage goals, and estimates of the relative costs of the various preventative and curative health services, are essential for achieving optimal allocation of resources, particularly in a context of limited financial and human resources. 19/ To the extent known, such procedures are not incorporated into the health planning process in Peru; instead, the process involves basic programs, with the general objective of universal coverage and loosely defined quantitative goals. 3.3 Poverty, Coverage, and Regional Disparities Aggregate average indicators say little about the distribution of personnel and facilities or their availability to different social groups. Table 3.7 presents data showing disparities in the distribution of housing services, hospital beds, and health personnel among the various political departments. Data show that across departments there are large variations in the proportion of houses with different types of services and in the ratios of population to health personnel. In Lima and Callao more than 80 percent of the houses have at least some type of basic service, that is, water, sewer electricity. In the poorest department of the Sierra, the proportion of houses with some of the services ranged from a high of 22 percent in Ayacucho to around 15 percent in Amazonas, Apurimac, Cajamarca, and Puno and 12 percent in Huancavelica.  - 42 - Regional average ratios of population per medical personnel also vary greatly. Although national averages of population per hospital bed and per physician are 500 and 1,300 respectively, the ratio population/hospital beds is only 200 for Callao and about 300 for Arequipa, Lima, Moquegua, and Tacna. But in the poorest departments, the ratios are at least five times higher for Ayacucho, San Martin, and Apurimac; between five and seven times higher for Huancavelica, Puno, and Amazonas; and more than ten times higher for Cajamarca. A more skewed distribution exists for doctors; while Lima and Callao enjoy an average of about 500 persons per doctor, the ratio increases to almost 18,000 persons per doctor in Ayacucho and Cajamarca, 25,000 persons per doctor in Cajamarca, and more than 31,000 persons per doctor in Huancavelica and Amazonas.  - 43 - TABLE 3.7: Peru: Social Security Coverage, Regional Disparities, (thousands) Total Average b (% Department Population a! Income Insured Coverage Lima 4,746 100.0 1,256 26.7 Callao 443 100.0 109 24.6 Ica 434 73.1 89 20.5 Tacna 143 87.0 28 19.6 Arequipa 706 63.6 123 17.4 Lambayeque 674 63.8 101 15.0 Moquegua 101 90.0 14 13.9 Pasco 213 59.1 28 13.1 La Libertad 963 65.0 116 12.0 Ancash 818 54.3 80 9.8 Junin 852 58.5 81 9.5 Tumbes 104 75.9 9 8.6 Loreto 445 70.1 38 8.5 Piura 1,126 65.4 94 8.3 Ucayali 201 66.2 15 7.5 Madre de Dios 33 75.9 2 6.1 San Martin 320 45.9 16 5.0 Cuzco 832 48.6 39 4.7 Huanco 485 50.9 23 4.7 Huancavelica 347 38.6 13 3.7 Puno 890 41.7 33 3.7 Amazonas 255 42.7 8 3.1 Ayacucho 503 40.9 14 2.8 Cajamarca 1,046 39.1 28 2.7 Apurimac 323 40.7 8 2.5 TOTAL 17,005 2,374 14.0 a/ Census enumerated population. Excludes omitted and jungle populations (estimated at 762,226). b/ Medical attention and maternity programs. Army not included. c/ Average income of Lima index: Intis 74.88=100.0. Source: Elaborated from: INE (1983); Cepal (1985), p.327; BCR (1984); BCR (1984), p.44-45. Because only about one-third of the physicians and about 11 percent of the dentists work through MOH, distribution of health personnel across the country seems to be only marginally determined by MOH allocations. 20/  - 44 - Measures of inequality in the spatial distribution of health care resources and sanitation services have been estimated and reported in a study of the Central Bank of Peru (BCR 1984). Cini coefficients have been estimated for the distribution of houses with water and houses with sewers, and for the distribution of the population according to health care and sanitation services, Gini coefficients ranged from 0.32 for the distribution of dwellings with water to 0.51 (highest inequality) for the distribution of doctors. Second in inequality was the distribution of dwellings with sewerage (see Appendix Table A.4.1). Estimates on the coverage of the social security system by departments clearly show a positive correlation between average income of the department and level of coverage; that is, coastal departments with high urbanization and per capita income have the highest coverage. Social security coverage is more limited in departments in the poorest regions of Peru--those south of the Sierra with the larger proportion of the rural population. This is not surprising given the process by which the social security system was created in Peru. The data on the distribution of services provided by MOH and IPSS show that the number of doctor consultations per capita varied from less than 0.2 in the departments of Amazonas, Apurimac, and Huancavelica to more than 1.5 in Lima, Callao, and Ica. A similar pattern was found in the distribution of the number of per capita dentist consultations. A high correlation (0.82) was found between the number of hospital beds per capita and the number of hospital "egresses." The only variable that had a different pattern was the distribution of health care per capita per health or auxiliary personnel. In general, expensive resources seem to be more concentrated than cheaper ones (doctors versus nurses, and sewerage versus water supply). 21/  - 45 - A health map of Peru was constructed by choosing a set of variables using a rank correlation coefficient criteria; that is, variables with correlations between 60 and 90 percent were chosen as indicators to organize a ranking of departments according to health status of the population, sanitation, availability of health personnel, extent of coverage of health programs, and accessibility of health services. Of the three groups of departments ranked, the group with the poorest health indicators comprised eight departments from the high mountains (Sierra): Apurimac, Puno, Ayacucho, Cajamarca, Amazonas, Cusco, Huancavelica, and Huanuco. These departments had the highest deficits of health resources and services and the lowest coverage of health programs. They represented about 28 percent of the population and had high illiteracy rates. They were also predominantly agricultural economies, with incomes well below the national average and high migration rates. The group with the highest health indicators and the highest coverage of health programs and availability of health facilities were the rich coastal departments of La Libertad, Arequipa, Lambayeque, Lima, Callao, Ica, Tacna, and Moquegua. The BCR study found that the distribution of health resources across the various regions of the country had a pattern similar to the distribution of productive activities. The study also stated that some improvements in coverage could be obtained by eliminating access barriers to some of the facilities of corporate health institutions. Indicators of distribution patterns of health resources, particularly of health personnel and health facilities, still have several shortcomings. They hide an important source of inequality derived from the method of calculation, which assumes that all resources, including those with restricted access, are available to the general public. In addition, it is assumed that  - 46 - needs or incidences of illness are uniformly distributed across regions and among the population groups and that existing facilities or resources correspond to population needs. Some of these issues have been addressed in data from the 1984 National Health and Nutrition Survey (ENNSA) and in studies by the National Health Sector Analysis Group (ANSSA), and ad hoc research groups formed to analyze the data from this survey. 22/ The results from ENNSA reported by the National Institute for Statistics (INE) (1986) show that the proportion of the population involved in accidents or having symptoms of disease was inversely related to level of income and education. A larger proportion of the poor and lower educated population reported having symptoms of illness. As the level of education and income increases, the proportion of the population with symptoms of illness or accidents declines. The proportion of the population with symptoms of diseases seeking professional medical attention was positively related to both education and income. Demand for traditional healers and auxiliary health personnel (lower level health personnel) was inversely related to income levels. ENNSA results on the prevalence and types of diseases and the distribution of population with symptoms seeking health care by age groups are consistent with morbidity and mortality patterns presented in Section 3.1. There are high incidences of parasitic infections and respiratory diseases. Although the prevalence of symptoms was relatively evenly distributed across political and ecological regions and rural and urban areas, distribution of population with symptoms seeking medical attention was higher in coastal regions and urban areas. The unconditional probability of seeking some type of medical care ranged from 15 percent in Lima to less than 5 percent in the rural mountains of Sierra (see Table 3.7).  - 47 - 4. SUMMARY AND CONCLUSIONS To analyze the financing of the health sector, Peruvian health institutions were classified into three groups: (1) the public health sector, which provides free services to the general public; (2) the corporate health sector, which restricts services to affiliated members; and (3) the private health sector, which provides services through explicit price-rationing schemes. The central government is the main source of financing for the Ministry of Health (MOH), the main institution of the public health sector. The resources devoted to MOH as a percentage of the government budget declined from 6.4 percent in 1970 to around 4.5 during the last ten years. As a proportion of GDP, government expenditure in health declined from 1.2 percent during the seventies to 0.6 in 1985. Per capita public expenditure on health declined from a high of US$9.0 during 1980-81 to less than US$5 during the last two years. The data presented in this paper (Table 2.8) show that while coverage by the public health sector is relatively high, the proportion of financial resources represents only 10 percent of the total resources spent in health- related goods and services. Expenditures of the corporate and private health sectors represent about 90 percent of total sectorial expenditures. This result calls for further studies to explore the roles that the corporate and private sectors can play in implementing health programs. While mortality rate has been declining infant mortaly rate remains high: around 90 per thousand (compare with between 10 to 20 per thousand in developed nations). Infections parasitic diseases, pneumonia, and other environmental related diseases appear as the main causes of morbility and  - 48 - mortality rates. With the exception of tosferina (whooping cough) incidence of illness that can be controlled by immunization programs almost double. Prevalence of environmental related diseases and high incidence of illness that could be prevented by vaccination suggest that concentration of Peruvian health institutions in the production of curative services rather than in preventive actions might be indicating a gross misallocation of resources spent in health. It also implies that there is a limited scope of the market mechanisms in alleviating the major causes of observed morbidity and mortality patterns. Further analysis of these issues is needed. Health related indicators also show that there is a high degree of inequality in the spatial distribution of health physical facilities, health personnel and sanitation services. Furthermore, there seen to be a higher degree of concentration of relatively more expensive resources.  - 49 - TABLE A.2.1: Peru: Ministry of Health, Actual Revenues by Sources, 1980-84 (Z) Sources 1980 1981 1982 1983 1984 Central Government 88.9 90.7 88.5 88.2 86.7 Revenues 6.7 6.2 8.2 7.1 7.2 Borrowing 4.0 2.8 1.5 2.8 4.6 Transfers 0.3 0.2 1.8 1.9 1.5 TOTAL 100.0 100.0 100.0 100.0 100.0 Actual budget (billions of soles: Current 49.0 83.2 166.3 341.5 725.6 Constant, 1980) 49.0 50.7 61.5 58.1 58.0 Source: ANNSA (1985), pp.96-97. TABLE A.2.2: Peru: IPSS, Covered Population: Active and Dependents, 1980-85 (thousands) Covered Population 1980 1981 1982 1983 1984 1985 Active a/ 2272.2 2373.9 2390.0 2497.6 2611.0 5242.5 Dependent Spouses 620.0 649.5 676.9 705.3 734.5 765.2 Children 45.2 47.2 49.2 51.2 53.2 1761.9 a! Active members: white collar, blue collar, domestic workers, affiliated members and pensioners. Source: INE (1986), p.66.  - 50 - TABLE A.3.1: Peru: Measures Related to Equity in Health Care, 1982 Health Care Resources and Sanitation Services (Z) Resources Dwellings Hospital With With Department Population Doctors Nurses Beds Total Water Sewerage Amazonas 1.5 0.0 0.1 0.3 2.4 0.6 0.4 Ancash 4.8 1.2 3.4 2.7 5.1 4.5 3.7 Apurimac 1.9 0.1 0.5 0.7 2.2 0.5 0.2 Arequipa 4.2 5.3 8.2 6.8 4.3 5.7 6.0 Ayacucho 2.9 0.2 1.3 0.9 2.6 1.5 0.7 Cajamarca 6.1 0.5 0.6 0.9 6.3 2.0 1.4 Callao 2.7 6.3 4.5 4.9 2.3 4.4 5.3 Cuzco 4.9 1.0 2.2 2.9 5.4 2.6 2.1 Huancavelica 2.0 0.1 0.4 0.7 2.4 0.6 0.4 Huanuco 2.8 0.6 0.6 1.5 2.9 1.0 0.9 Ica 2.5 2.8 3.8 3.7 2.5 3.1 2.4 Junin 5.0 1.5 4.0 4.5 5.3 4.3 3.1 La Libertad 5.6 4.9 5.5 4.9 5.5 6.5 6.1 Lambayeque 4.0 2.6 5.4 3.8 3.5 4.5 4.3 Lima 28.1 66.6 50.7 48.4 26.4 45.1 52.8 Loreto 2.6 0.8 0.7 1.5 2.2 2.0 1.8 Madre de Dios 0.2 0.1 0.1 0.2 0.2 0.0 0.1 Moquegua 0.6 0.7 0.2 1.1 0.2 0.7 0.7 Pasco 1.2 0.6 1.2 1.7 1.3 0.8 0.6 Piura 6.6 2.3 2.7 3.2 6.1 5.2 4.2 Puno 5.2 0.5 1.5 1.6 6.4 1.5 0.8 San 14artin 1.9 0.2 0.5 0.8 1.7 0.9 0.3 Tacna 0.9 0.6 1.3 1.4 0.9 1.5 1.4 Tumbes 0.6 0.2 0.3 0.3 0.5 0.1 0.1 Ucayali 1.2 0.3 0.3 0.6 1.0 0.4 0.3 TOTAL 100.0 100.0 100.0 100.0 100.0 100.0 100.0 Gini Coefficient of Inequality - 0.51 0.38 0.34 - 0.32 0.41 Source: Banco Central de Reserva del Peri&, Mapa de Salud del PerAi, Lima: December 1984. Graphs 1-5, pp.l9, 20, 23, 24 and 29, from Musgrove, P. (1986).  - 51 - TABLE A.3.2: Peru: Measures Related to Equity in Health Care, 1984: Morbidity and Medical Attention Prevalance (%) of Symptoms Percent Seeking Medical Attention Respira- All tory Parasite In Given Presence of Symptoms Department Kinds Disease Infection Total All Ages I Year 1-4 Yrs. 5 Yrs. Coast 34.89 16.11 0.21 12.67 30.86 51.41 30.68 28.64 Urban 34.87 16.34 0.20 13.36 31.54 55.74 32.69 29.96 Lima 36.57 17.68 0.15 14.88 33.49 59.43 35.87 31.62 Slums 37.32 16.85 0.19 14.14 32.05 64.84 32.85 29.78 Rural 35.00 14.38 0.28 7.50 18.76 27.14 16.52 18.71 Mountains 30.05 11.95 0.16 5.53 15.89 24.13 15.47 15.49 Urban 21.50 9.88 0.11 7.38 28.20 41.33 25.63 28.09 Rural 33.75 12.84 0.18 4.73 12.49 19.32 12.33 12.10 Jungle 36.03 12.06 1.78 7.65 18.39 27.33 17.20 17.94 Urban 33.72 11.59 1.36 10.36 25.59 42.63 22.34 24.97 Rural 37.35 12.32 2.02 6.11 14.69 18.42 14.46 14.41 National Total 33.31 14.20 0.36 9.60 24.18 38.89 23.83 23.35 Urban 32.51 14.93 0.26 12.15 30.77 52.64 30.90 29.46 Rural 34.62 13.01 0.53 5.45 14.01 20.55 13.51 13.69 Source: Encuesta Nacional de Nutricidn y Salud (National Health and Nutrition Survey), Peru, 1984. Tables 2.1, 2.2, 2.2A and 2.13 produced by the Instituto Nacional de Estadistica, September 1985, unpublished, from Musgrove, P. (1986).  TABLE A.4.1: Peru: Regional Disparities, selected Health Indicators, 1981 (M) Percentage of Houses with Type of Services Thousand of People Per Water, Water Water Persons Hospi- Sewer. and and Only Only Without Per tal Physi- Department TOTAL Total & Elec. Sewer. Elec. Water Elec. Services House Bed cian Nurse Dentist Amazonas 100.0 17.8 5.0 1.9 2.0 6.8 2.1 82.2 5.7 2.2 25.0 24.7 10.2 Ancash 100.0 38.7 24.9 6.7 1.0 0.7 5.4 61.3 5.2 0.7 5.2 3.5 21.7 Apurimac 100.0 17.4 2.3 1.0 5.5 6.6 1.9 82.6 4.8 1.6 31.3 11.1 29.4 Arequipa 100.0 68.7 36.0 2.9 8.8 2.3 18.6 31.3 5.4 0.3 1.0 1.3 3.6 Ayacucho 100.0 23.0 4.9 0.6 1.7 5.3 10.5 77.0 4.6 1.5 17.9 5.8 25.0 Cajamarca 100.0 15.8 7.0 3.2 0.2 0.2 5.2 84.2 5.4 3.3 17.9 27.8 29.4 Callao 100.0 90.3 64.4 1.2 4.9 1.0 18.9 9.7 5.9 0.2 0.5 1.4 2.0 Cuzco 100.0 27.3 9.7 1.0 3.4 4.1 9.2 72.7 5.0 0.8 6.0 5.6 7.6 Huancavelica 100.0 12.5 2.6 0.7 4.0 2.6 2.6 87.5 4.6 1.7 31.3 11.7 35.7 Huanuco 100.0 18.4 6.4 3.2 0.8 1.6 6.4 81.6 5.4 1.2 5.7 11.7 16.1 N Ica 100.0 65.6 28.3 1.5 11.6 8.0 16.1 34.4 5.7 0.4 1.1 3.0 2.2 Junin 100.0 49.5 16.0 1.5 4.1 5.2 22.7 50.5 5.2 0.6 4.0 2.6 7.9 La Libertad 100.0 52.8 28.4 3.4 7.2 6.1 7.5 47.2 5.5 0.6 1.5 1.8 7.1 Lambayeque 100.0 59.2 31.9 3.1 6.9 6.1 11.3 40.8 6.2 0.5 1.9 1.4 6.7 Lima 100.0 85.2 56.4 1.3 5.5 1.7 20.4 14.8 5.7 0.3 0.5 1.4 2.0 Loreto 100.0 41.5 20.8 0.9 7.9 2.8 9.1 58.5 6.9 0.4 3.7 9.3 6.8 Madre de Dios 100.0 34.0 5.6 0.4 5.9 0.5 21.5 66.0 6.9 0.7 3.7 4.5 11.1 Moquegua 100.0 62.3 31.8 0.6 5.2 1.7 23.0 37.7 4.8 0.3 1.1 5.6 3.2 Pasco 100.0 46.5 12.0 0.7 7.6 4.1 22.1 53.5 5.5 0.4 2.7 2.6 9.3 Piura 100.0 37.8 17.8 2.2 5.3 7.6 4.9 62.2 5.9 1.0 3.7 6.0 11.6 Puno 100.0 14.8 7.6 2.6 0.4 0.7 3.5 85.2 4.4 1.9 12.5 8.5 50.0 San Martin 100.0 36.2 5.7 0.6 13.9 2.8 13.1 63.8 6.0 1.5 10.4 8.9 15.2 Tacna 100.0 77.3 46.7 1.9 10.0 7.0 11.6 22.7 5.3 0.3 1.8 1.7 4.5 Tumbes 100.0 86.6 7.3 0.4 2.6 0.6 75.7 13.4 6.4 0.9 5.4 4.9 6.7 Ucayali 100.0 27.4 5.3 0.3 4.8 1.1 16.0 72.6 6.7 1.0 6.6 8.2 13.5 TOTAL 100.0 50.9 28.1 2.0 4.7 3.2 12.9 49.1 5.5 0.5 1.3 2.4 4.3 Source: Elaborated from census data reported in INE (1986c), p.79; INE (1986), p.24, 31 and Ministry of Health (1985), Table 7.  - 53 - NOTES 1. Figures refer to actual expenditure from January to October 1984 and 1985, see INE (1986) p. 32. 2. See Ministry of Health (1985) p. 3-10, 33-35. 3. These estimates should be taken cautiously. I have taken the distribution of population from ANSSA (1985). Distribution of institutions does not necessarily follow the same criteria. Data refer to coverage of a population by particular health group or institution. No detailed explanation of methodology used for the estimation is presented. For the private health group, coverage seems to refer to those having access to formal private institutions; it does not consider those receiving medical attention from private doctors, auxiliary health personnel, and traditional healers. 4. This priority index is defined as the ratio of total government expenditures to government expenditures on health programs. The index is independent of whether expenditures are increasing or declining; it only measures whether expenditures on health programs have been reduced or have increased proportionately to total government expenditures. (See Suarez 1985.) 5. ANSSA (1986) reported that during 1980-84 central government resources and MOH revenues represented approximately 88 and 7 percent respectively. Borrowing was between 1.5 and 4.6 percent and transfers between 0.2 and 1.8 percent (see Table 4.7.1, p. 96). 6. Transfers seem now to be considered within the general budget as a result of the integration of some decentralized health institutions to which these transfers were made. 7. In 1934 stevedore workers from Callao, Peru's main seaport, and jockeys were incorporated into the existing retirement and pension programs. 8. For a detailed description of the evolution of the social security system in Latin American countries and Peru see Cepal (1985), Messa-Lago (1981), and Roemer (1975). 9. See studies from Webb (1977) and Figueroa and Webb (1977). 10. The 2.4 percent share has been estimated by using as weights the share of medical expenditure of each of the eight communities and by corresponding share of the value of the monetary exports. Data has been taken from Figueroa (1983), Tables 3.4 and 3.5, pp. 63-65. Import shares ranged from 0.3 to 6.6 percent. 11. Although existing information shows the possibility of presenting more detailed calculations, they are out of the scope of the present paper.  - 54 - 12. For a summary of health status indicators of the various countries and regions of the world see World Bank (1980). 13. A recent report from MOH presents estimates that the under five years age groups accounted for more than 45 percent of total deaths. But no specific data on deaths by age group is presented. See Ministerio de Salud (1986), p. 3. 14. Data needed to evaluate health status of the population and efficiency of the medical care systems should combine age groups and morbidity and mortality rates; regrettably such statistics are not readily available. For Peru, knowledge of incidence of illness is based on cases of illness reported to health personnel and is reported in MOH annual reports. A detailed collection of data on causes of deaths by type of illness, gender, and age group is compiled by WHO. 15. A preliminary discussion of this issue in terms of cost per preventative and curative health service is presented by Musgrove (1986). 16. For references and discussions on the relation between access to drinkable water and sanitation facilities and improvements in health, see World Bank (1970), pp. 23-27. 17. See PAHO (1982). 18. A discussion of definitions and issues related to defining and selecting health coverage indicators is contained in a collection of yet unpublished articles from Montoya Aguilar and Marin-Lira from WHO. I would like to thank the first author for providing the material presented here. 19. This observation is derived from the analysis of the Ministry of Health Plan 1985-86, and from the way such plans have been conducted in the past. For a summary of the planning process see OIH (1977), p. 82-96. 20. BCR study estimates that in 1980 only 26.9 percent of doctors and nurses and 11.3 percent of dentists were employees; the rest were working for the corporate and private health sector (see BCR [1984], p. 30). Concentration of health personnel in urban areas and in coastal cities can be thought of as the result of optimal market allocations. Given the patterns of distribution of income and wealth, allocation of MOH resources seems to have done little in altering spatial distribution patterns. 21. This is one of the conclusions from Musgrove's interpretation of BCR data. See Musgrove (1986), p. 5. Data are presented in Table A.IV.2. 22. A description of the organizational and methodological aspects of the survey can be found in Musgrove (1985) and in INE (1986).  - 55 - References Amat y Leon, C. and Leon, H. (1982). Niveles de Vida y Grupos Sociales en el Peru, Centro de Investigaciones de la Universidad Catolica del Peru - Fundacion Friederich Ebert. Lirmia, Peru. ANNSA (1985). "Informe Preliminar," Documento de Trabajo, Proyecto de Analisis del Sector Salud, ANNSA-Peru, Lima, Julio. ANNSSA (1985). "...Estructura del Sector Salud, Responsabilidad Poblacional," II Seminario-Taller, Hotel el Pueblo, Mayo, Lima, Peru. Mimeo. BCR (1981). "Mapa de Pobreza 1981," Departamento de Estudios del Sector Social, Banco Central de Reserva del Peru, Lima, Peru. BCR (1984). "Mapa de Salud del Peru," Departamento de Estudios del Sector Social, Banco Central de Reserva del Peru, Marzo, Lima, Peru. BID (1982). Progreso Economico y Social de America Latina: Informe 1980-81, Banco Interamericano de Desarrollo, Washington, D.C. CEPAL, (1984). Estructura del Gasto de Consumo de los Hogares segun Finalidades de Gasto, por Grupo de Ingresos, Cuadernos Estadisticos de la Cepal, Naciones Unidas, Santiago de Chile. CEPAL, (1985). El Desarrollo de la Seguridad Social en America Latina, Estudios e Informes de la Cepal, Naciones Unidas, Santiago de Chile. Chirinos, 0. (1986) "Financiamiento y Gasto del Sector Salud," Estudio 6.2, Area Financiamiento y Costos, ANSSA-Peru. III Seminario-Taller, Lima, 6-8 February 1986. Couriel, A. (1981). Estado, Estrategia de Desarrolo y Necesidades Basicas en el Peru, DESCO, Centro de Estudios de Promocion y Desarrollo, May, Lima, Peru. de Ferranti, D. (1985). "Paying for Health Services in Developing Countries: An Overview," World Bank Staff Working Papers, Number 721, Washington, D.C., U.S.A. Figueroa, A. (1974). Estructura del Consumo y Distribucion del Ingreso en Lima Metropolitana, 1968-1969, Un Estudio de Presupuestos Familiares dentro del Programa ECIEL, Serie: Estudios Economicos #1, Pontificia Universidad Catolica del Peru, Deapartamento de Economia, Lima, Peru. IMF (1985). Government Finance Statistics Yearbook: 1984, Volume VIII, International Monetary Fund, Washington, D.C. IMF (1986). International Financial Statistics, Vol. XXXIX, #8, August, International Monetary Fund, Washington, D.C.  - 56 - INE (1983). Peru: Compendio Estadisitico 1982, Instituto Nacional de Estadistica, Lima, Julio. INE (1986). Informe Estadistico, Cuarto Trimestre 1985, Insituto Nacional de Estadistica (INE), Direccion General de Indicadores Economicos y Sociales, Direccion de Coyuntura, Lima, Enero. INE (1986c). Peru: Compendio Estadistico 1985, Instituto Nacional de Estadistica (INE), Direccion General de Indicadores Economicos y Sociales, Direccion de Estudios Socioeconomicos, Lima, Mayo. Ministerio de Salud (1985). Plan Nacional de Desarrollo Para 1985-1986: Plan Operativo Sectorial, Oficina General de Presupuesto y Planificacion, Mayo, Lima - Peru. Novelo, G. and Sanchez, C. (1978). "Prestaciones de los Sistemas de Seguridad Social en las Areas Rurales," Seguridad Social, Ano XXVII, Epoca V, Nums. 109-110, Tomo II, Enero-Abril, Mexico, D.F., p. 88-ff. Musgrove, P. (1986). "Measurement of Equity in Health," World Health Statistics Quarterly, #4, (forthcoming). . (1986a). "Cuanto Mas Vale Prevenir Que Curar?", Reflexiones sobre la Distribucion de los Castos en Salud, Julio, (mimeo). . (1985). "The National Health and Nutrition Survey of 1984 in Peru: Origin, Content and Uses," Paper prepared for the International Health Section of the 113th Annual Meeting of the American Public Health Association, Washington, D.C. (mimeo). OIH (1977). The Health Situation in Peru, Office of International Health (OIH), U.S. Department of Health, Education and Welfare, Rockville, Maryland, December. PAHO (1982). Health for All by the Year 2000: Plan of Action for the Implementation of Regional Srategies, Official Document #79. Washington, D.C. Roemer, M.I. (1975). "Organizational Issues Relating to Medical Priorities in Latin America," University of California, Los Angeles. . (1984). The Organization of Medical Care in Peru: Characteristics and Trends. (H0071746). SSA (1985). Regimenes de Seguridad Social en las Americas, 1983, U.S. Department of Health and Human Services, Social Security Administration (SSA), Office of Policy, Office of Research Statistics and International Policy, September, Washington, D.C.  - 57 - Suarez, R. (1985). "The Financing of the Public Health Sector in Central America and Panama," Pan American Health Organization, Division of Strategic Planning, August, Washington, D.C. . (1986). "The Labor Market in Peru: An Overview," Universidad Catolica del Peru, Labor Markets Division, The World Bank, (in progress). . (1986a). "Informal Sector, Labor Markets, and Returns to Education in Peru," Living Standards Units - Education Policy Division, The World Bank, Washington, D.C., July. Vega, J., Madrid, S., and Locay, L. (1986). "Recursos Humanos y Administracion en Salud," Mercado Fuerza de Trabajo (Localizacion Geografica e Institucional del Personel que Ofrece Servicios de Salud), Informe Preliminar, ANNSA-Peru, III Seminario Taller, Febreo, Lima, Peru. World Bank (1980). Health: Sector Policy Paper, The World Bank, Washington, D.C., February.  LSMS WORKING PAPER SERIES (continued) No. 18. Time Use Data and the Living Standards Measurement Study. No. 19. The Conceptual Basis of Measures of Household Welfare and Their Implied Survey Data Requirements. No. 20. Statistical Experimentation for Household Surveys: Two Case Studies of Hong Kong. No. 21. The Collection of Price Data for the Measurement of Living Standards. No. 22. Household Expenditure Surveys: Some Methodological Surveys. No. 23. Collecting Panel Data in Developing Countries: Does it Make Sense? No. 24. Measuring and Analyzing Levels of Living in Developing Countries: An Annotated Questionnaire. No. 25. The Demand for Urban Housing in the Ivory Coast. No. 26. The C8te d'Ivoire Living Standards Survey: Design and Implementation. No. 27. The Role of Employment and Earnings in Analyzing Levels of Living: A General Methodology with Applications to Malaysia and Thailand. No. 28. Analysis of Household Expenditures. No. 29. The Distribution of Welfare in the Republic of C8te d'Ivoire in 1985. No. 30. Qualilty, Quantity and Spatial Variation of Price: Estimating Price Elasticities from Cross-sectional Data. No. 31. Financing the Health Sector in Peru.  Suarez-Berenguela