LSM 1A q
Living Standards
Measurement Study
Working Paper No. 123
The Demand for Medical Care
Evidence from Urban Areas in Bolivia
Masako Ii
-..-.=- --= .-- .=.. --- .-~   -  ii


LSMS Working Papers
No. 49    Scott and Amenuvegbe, Sample Designs for the Living Standards Surveys in Ghana and
Mauritania/Plans de sondage pour les enquetes sur le niveau de vie au Ghana et en Mauritanie
No. 50    Laraki, Food Subsidies: A Case Study of Price Reform in Morocco (also in French, 50F)
No. 51    Strauss and Mehra, Child Anthropometry in C6te d'Ivoire: Estimates from Two Surveys, 1985
and 2986
No. 52    van der Gaag, Stelcner, and Vijverberg, Public-Private Sector Wage Comparisons and
Moonlighting in Developing Countries: Evidence from Cte d'lvoire and Peru
No. 53 Ainsworth, Socioeconomic Determinants of Fertility in C6te d'Ivoire
No. 54    Gertler and Glewwe, The Willingness to Pay for Education in Developing Countries: Evidence flom
Rural Peru
No. 55    Levy and Newman, Rigidit6 des salaires: Donnes microeconomiques et macroeconomiques sur
l'ajustement du march du travail dans le secteur moderne (in French only)
No. 56    Glewwe and de Tray, The Poor in Latin America during Adjustment: A Case Study of Peru
No. 57 Alderman and Gertler, The Substitutability of Public and Private Health Care for the
Treatment of Children in Pakistan
No. 58    Rosenhouse, Identifying the Poor: Is "Headship" a Useful Concept?
No. 59    Vijverberg, Labor Market Performance as a Determinant of Migration
No. 60    Jimenez and Cox, The Relative Effectiveness of Private and Public Schools: Evidence from Two
Developing Countries
No. 61    Kakwani, Large Sample Distribution of Several Inequality Measures: With Application to C6te
d'Ivoire
No. 62    Kakwani, Testing for Significance of Poverty Differences: With Application to C6te d'lvoire
No. 63 Kakwani, Poverty and Economic Growth: With Application to C6te d'Ivoire
No. 64    Moock, Musgrove, and Stelcner, Education and Earnings in Peru's Informal Nonfarm Family
Enterprises
No. 65    Alderman and Kozel, Formal and Infbrmal Sector Wage Determination in Urban Low-Income
Neighborhoods in Pakistan
No. 66    Vijverberg and van der Gaag, Testing for Labor Market Duality: The Private Wage Sector in
C6te d'lvoire
No. 67    King, Does Education Pay in the Labor Market? The Labor Force Participation, Occupation,
and Earnings of Peruvian Women
No. 68 Kozel, The Composition and Distribution of Income in C6te d'Ivoire
No. 69    Deaton, Price Elasticities from Survey Data: Extensions and Indonesian Results
No. 70    Glewwe, Efficient Allocation of Transfers to the Poor: The Problem of Unobserved Household Income
No. 71 Glewwe, Investigating the Determinants of Household Welfare in C6te d'Ivoire
No. 72    Pitt and Rosenzweig, The Selectivity of Fertility and the Determinants of Human Capital
Investments: Parametric and Semiparametric Estimates
No. 73    Jacoby, Shadow Wages and Peasant Family Labor Supply: An Econometric Application to the
Peruvian Sierra
No. 74    Behrman, The Action of Human Resources and Poverty on One Another: What We Have
Yet to Learn
No. 75    Glewwe and Twum-Baah, The Distribution of Welfare in Ghana, 1987-88
No. 76    Glewwe, Schooling, Skills, and the Returns to Government Investment in Education: An
Exploration Using Data from Ghana
No. 77    Newman, Jorgensen, and Pradhan, Workers' Benefits from Bolivia's Emergency Social Fund
No. 78    Vijverberg, Dual Selection Criteria with Multiple Alternatives: Migration, Work Status, and Wages
No. 79 Thomas, Gender Differences in Household Resource Allocations
No. 80    Grosh, The Household Survey as a Tool for Policy Change: Lessons from the Jamaican Survey
of Living Conditions
No. 81 Deaton and Paxson, Patterns of Aging in Thailand and C6te d'Ivoire
No. 82 Ravallion, Does Undernutrition Respond to incomes and Prices? Dominance Tests for Indonesia
No. 83 Ravallion and Datt, Growth and Redistribution Components of Changes in Poverty Measure: A
(List continues on the inside back cover)


The Demand for Medical Care
Evidence from Urban Areas in Bolivia


The 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 statistical offices in developing countries. Its goal is to foster increased
use of household data as a basis for policy decisionmaking. 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 policymakers.
The LsMs Working Paper series was started to disseminate intermediate prod-
ucts from the LsMS. Publications in the series include critical surveys covering dif-
ferent aspects of the LSMS data collection program and reports on improved
methodologies for using Living Standards Survey (LSS) data. More recent publica-
tions recommend specific survey, questionnaire, and data processing designs and
demonstrate the breadth of policy analysis that can be carried out using LSs data.


LSMS Working Paper
Number 123
The Demand for Medical Care
Evidence from Urban Areas in Bolivia
Masako Ii
The World Bank
Washington, D.C.


Copyright © 1996
The International Bank for Reconstruction
and Development/THE WORLD BANK
1818 H Street, N.W.
Washington, D.C. 20433, U.S.A.
All rights reserved
Manufactured in the United States of America
First printing February 1996
To present the results of the Living Standards Measurement Study with the least possible delay, the
typescript of this paper has not been prepared in accordance with the procedures appropriate to formal
printed texts, and the World Bank accepts no responsibility for errors. Some sources cited in this paper may
be informal documents that are not readily available.
The findings, interpretations, and conclusions expressed in this paper are entirely those of the author(s) and
should not be attributed in any manner to the World Bank, to its affiliated organizations, or to members of its
Board of Executive Directors or the countries they represent. The World Bank does not guarantee the accuracy of
the data included in this publication and accepts no responsibility whatsoever for any consequence of their use.
The boundaries, colors, denominations, and other information shown on any map in this volume do not
imply on the part of the World Bank Group any judgment on the legal status of any territory or the endorsement
or acceptance of such boundaries.
The material in this publication is copyrighted. Requests for permission to reproduce portions of it should
be sent to the Office of the Publisher at the address shown in the copyright notice above. The World Bank
encourages dissemination of its work and will normally give permission promptly and, when the repro-
duction is for noncommercial purposes, without asking a fee. Permission to copy portions for classroom
use is granted through the Copyright Clearance Center, Inc., Suite 910, 222 Rosewood Drive, Danvers,
Massachusetts 01923, U.S.A.
The complete backlist of publications from the World Bank is shown in the annual Index of Publications,
which contains an alphabetical title list (with full ordering information) and indexes of subjects, authors, and
countries and regions. The latest edition is available free of charge from the Distribution Unit, Office of the
Publisher, The World Bank, 1818 H Street, N.W., Washington, D.C. 20433, U.S.A., or from Publications,
The World Bank, 66, avenue d'I6na, 75116 Paris, France.
ISSN: 0253-4517
Masako Ii is an associate professor in the faculty of economics at Yokohama National University, Japan.
Library of Congress Cataloging-in-Publication Data
Ii, Masako, 1963-
The demand for medical care: evidence from urban areas in Bolivia
/ Masako Ii.
p. cm. - (LSMS working paper; no. 123)
Includes bibliographical references.
ISBN 0-8213-3565-0
1. Medical care-Bolivia-Finance. 2. Ambulatory medical care-
Utilization-Bolivia. 3. Maternal health services-Utilization-
Bolivia. 4. User charges-Bolivia. I. Title. II. Series.
RA410.9.B515 1996
362.1'0984-dc2O                                     96-927
CIP


Contents
Foreword   .................................................vii
A bstract  .. ... ......  ... ..................          ......   ....  ... .. ...  ix
Acknowledgment    ............................................                       xi
Chapter  I.  Introduction........................................
Chapter I. The Health Care System in Bolivia  ......................... 3
Background on the Bolivian Health Sector ..................... 3
Health  Care  Financing  .................................                6
Chapter III. Empirical Specification ................................. 7
Chapter    IV.  The  Data  .......................................                ...  11
Chapter V.  Empirical Results  .................................... 13
Adult Outpatients Aged 16 Years or Older . . . . . . . . . . . . . . . . . . . . . 13
Child Outpatients Aged 15 Years or Younger ................... 19
Infant Outpatients Aged Between 0 and 4 Years Old . . . . . . . . . . . . . . . 24
Obstetrical Care for Women between 15 and 49 Years Old . . . . . . . . . . . 30
Prenatal Care for Women between 15 and 49 Years Old   ............ 33
Chapter VI. Conclusion  ........................................37
A nnexes  ......  ...  ... ............      ....  ........     ......   .. ......41
A-1: Descriptive Statistics (Adults Aged 16 Years or Older) .................. 41
A-2: Descriptive Statistics (Children Aged 15 Years or Younger) ............... 42
A-3: Descriptive Statistics (Children Aged 4 Years or Younger) ................ 43
A-4: Descriptive Statistics (Women Between 15 and 49 Years Old) . . . . . . . . . . . . . . 44
Bibliography  ...............................................45
Text Tables
1:  Money Cost of Obtaining Medical Care ........................... 3
2:   Opportunity Cost of Obtaining Medical Care ........................ 4
3:   Consulting Fees for Formal Medical Care by Monthly Income Level .........       5
4:   Cost of Delivering a Child by Monthly Income Level ................... 5
5:   Cost per Visit of Prenatal Care by Monthly Income Level ................        5
6:   Structure of Government Revenue, Bolivia, 1989  ..................... 6
7:   Disbursements of External Grants and Credits to the Health Sector, 1985  ......  6
v


8a: Choice of Provider, Multinomial Logit Coefficients and Standard Errors    14
8b: Probability of Choosing a Provider (The Effects on Probability of Changes in
Individual and Household Attributes, Adults Aged 16 Years or Older) ........ 15
8c: Probability of Choosing a Provider (The Effects on Probability of Changes in
Health Facility Attributes, Adults Aged 16 Years or Older) ............... 17
8d: Elasticities of Probability with Respect to User Fee .................... 18
9a: Choice of Provider Multinomial Logit Coefficients and Standard Errors (Children
Aged  15  Years or Younger)  ..................................20
9b: Probability of Choosing a Provider (The Effects on Probability of Changes in
Individual and Household Attributes, Children Aged 15 Years or Younger) ..... 22
9c: Probability of Choosing a Provider (The Effects on Probability of Changes in
Health Facility Attributes, Children Aged 15 Years or Younger) ............ 23
9d: Elasticities of Probability with Respect to User Fee (Children Aged 15 Years of
Younger)  ..............................................25
10a: Choice of Provider Multinomial Logit Coefficients and Standard Errors (Children
Aged  4  or Younger)  .......................................26
10b: Probability of Choosing a Provider (The Effects on Probability of Changes in
Individual and Household Attributes, Children Aged 4 or Younger) .......... 27
10c: Probability of Choosing a Provider (The Effects on Probability of Changes in
Health Facility Attributes, Children Aged 4 or Younger) ................. 28
10d: Elasticities of Probability with Respect to User Fee (Children Aged 4 or
Younger)  ..............................................29
1 la: Choice of Provider Multinomial Logit Coefficients and Standard Errors (Women
Between 15 and 49 YearsOld) ................................30
1 1b: Probability of Choosing a Provider (The Effects on Probability of Changes in
Individual and Household Attributes, Women Between 14 and 49 Years Old) . ... 31
1 1c: Probability of Choosing a Provider (The Effects on Probability of Changes in
Health Facility Attributes, Women Aged Between 15 Years and 49 Years) ...... 32
1ld: Elasticities of Probability with Respect to User Fees (Women Between 15 and 49
Years  Old) .............................................33
12: Linear Regression for Women Who Sought Prenatal Care ................ 34
13: Tobit EstimationResults .........                                        36
14: Price Elasticities for Prenatal Care Visit ........................... 36
vi


Foreword
User fees usually play a minor role in generating resources for social services such as
health in developing countries. Tightened budgets have forced many developing countries to
make painful choices regarding the financing and provision of their major social services. Some
governments are beginning to reevaluate the policy of providing uniform and heavy subsidies
from public purses, and are considering whether to introduce user fees so that consumers pay
a larger share of the cost. Whether enough revenue can be raised from patients depends on the
price elasticity of demand for medical care. This research examines the possible trade-offs
between a cost recovery policy such as user charges and the demand for and use of health
services.
Lyn Squire, Director
Policy Research Department
vii



Abstract
This research analyzes the determinants of demand for medical services in urban areas
of Bolivia. It also examines the possible trade-offs between cost recovery and the use of health
services for different age, sex, ethnic, and income groups. The data used are from the third
year of the Encuesta Integrada de Hogares (EIH), a multipurpose household survey conducted
by the statistical office in Bolivia and the World Bank.
The above data are used to estimate a multinomial logit model, and a nested multinomial
logit model (choice of medical facilities by patients). The main empirical result of this research
is that the demand for medical care is responsive to changes in price, but price elasticities are,
in general, very low. This finding is comparable to that of research for other countries.
Moreover, the price elasticity of demand falls as income rises. For children, the price
elasticities are lower than for adults.
Price elasticities do not vary much by ethnic group or gender, but estimation results show
that Aymara speakers (an Indian group) are more likely to care for themselves. Probably there
are cultural barriers that prevent Aymara speakers from seeking formal care.
Our results also show that income and education are also important determinants of
demand for medical care. For children, mother's education is far more influential than father's.
Since price elasticities are so low, it suggests that there is potential for the Bolivian
government to raise revenues by charging user fees. If additional revenues are not used to
expand primary health care or to improve quality, imposing user charges on services may not
substantially reduce inefficiency and/or inequity in the health sector.
ix



Acknowledgment
This paper is a part of my Ph.D thesis submitted to University of Wisconsin, Madison.
I am very grateful to my thesis advisors, Ralph Andreano, Arthur Goldberger and Brad Barham.
This research would not have been possible without the opportunity to work under Margaret
Grosh, Emmanuel Jimenez, John Newman, and Jacques van der Gaag at the World Bank. They
provided an excellent example of research on living standards in developing countries. My
research experience with Jere Behrman, University of Pennsylvania, at Unidad de Analisis de
Politicas Sociales in Bolivia was also invaluable. His generosity and insights deepened my
understanding of the Bolivian economy. Jim Shafer kindly provided his help in editing this
paper.
xi


4


Chapter I. Introduction
Throughout the developing world, the health sector is in trouble. Policies are usually ill-
defined, systems are highly centralized and expenditures tend to be biased toward curative care
rather than preventive care. For example, hospitals are built in urban areas, while rural areas
lack basic health care clinics. Governmental budgetary support for health is faltering and in
some cases actually declining in real terms, as countries struggle to exercise fiscal restraint in
the face of poor economic performance and burgeoning debt. At the same time, the belief that
health is a basic human right underlies ambitious promises such as the declaration of Alma Ata,
which aims at the World Health Organization's "Health for All by the Year 2000". Promises
for rapid improvements in health conditions continue to be made and these call for substantial
increases in spending.
User fees (which will often be referred to as prices in this paper) usually play only a
minor role in generating resources for social services such as health services and education in
developing countries. Instead, the public sector, the most important provider of these services,
has traditionally relied on general revenues for financing. These revenues, in turn, are financed
through direct or indirect taxes, inflation, or budgetary deficits supported by domestic and
foreign borrowing.
Tightened budgets have forced many developing countries to make painful choices
regarding the financing and provision of their major social services. Some governments are
beginning to reevaluate the policy of providing uniform and heavy subsidies from the public
purse, and are considering whether to introduce user fees so that consumers pay a larger share
of the cost. This study examines the possible trade-offs between cost recovery and the use of
health services for different age, sex, and income groups.
Whether enough revenue can be raised from patients depends on the price elasticity of
demand for medical care. If small price increases greatly reduce utilization, the extra revenue
may be too small to justify the policy. A further issue is whether price sensitivity varies among
groups in the population. If the poor, the aged, women, or children are more price sensitive
than, say, relatively well-to-do adult males, user fees may have undesirable consequences for
the distribution of health care.
This study addresses these issues by analyzing the demand for medical care -- outpatient,
prenatal, and obstetrical services in urban communities in Bolivia. Bolivia is well suited for
studying these issues because the population of Bolivia has the worst health indicators in the
continent. Life expectancy averages 51 years for men and 55 years for women and infant
mortality is twice the average of the region, at 110 per 1,000 live births. Maternal mortality,
at 48 per 10,000 live births, is almost twice the average for the region, and leads to 1,400 deaths
a year. It is associated with high fertility, poor health conditions of mothers, and lack of
prenatal and delivery care.
In the early 1980s, Bolivia experienced a devastating economic crisis. The decline in
economic activity was substantial. Official estimates show a fall in real gross domestic product
1


(GDP) of over 10 percent from 1980 to 1985, 24 percent in per capita terms. Per capita
consumption is estimated to have dropped by over 16 percent, and private fixed investment
averaged only 3 percent of GDP between 1983 and 1985, compared to 7 percent in 1980.
Inflation reached an annual rate of 28,000 in the first 9 months of 1985. In 1984, because of
a lack of funds, the Bolivian government suspended payments to foreign creditors.
Since 1985, macro stabilization policies have succeeded in bringing about positive growth
rates in gross national product (GNP), lower inflation rates, and reduced external -'ebt, yet the
effects of these macro policies on investment in human capital in sectors such as health and
education could be serious, especially in the long run. Bolivia's austere structural adjustment
programs put severe constraints on the government budget. Most hospitals and clinics in Bolivia
have started to charge user fees, although the government has not directed them to do so and has
no policy on the issue.
This research analyzes the determinants of demand for medical services. A common
finding of several recent studies is that utilization of health services tends to be relatively price
inelastic; that is, a unit increase in price leads to increased revenues. However, these are point
estimates at the mean along the demand curve. Many of the same studies show that demand
may be relatively more price elastic at lower income levels.
This study provides estimates of the factors affecting individual choices among different
modes of treatment for illness and injury.  For the principal combinations of characteristics
(e.g., expenditure level, age, and sex) I calculate elasticities of demand with respect to the user
fee, and to waiting and travel time.
Price elasticities do not vary much by ethnic groups and gender, but estimation results
show that Aymara speakers (an Indian group) are more likely to care for themselves. Probably
there are cultural barriers that prevent Aymara speakers from seeking formal care.
Our results also show that income and education are also important determinants of
demand for medical care. For children, mother's education is far more influential than father's.
Chapter II provides background information on the health care system in Bolivia.
Chapters III explains the empirical specification of demand for medical services. Chapter IV
describes the data set. Chapter V reports the empirical results. Chapter VI presents the
conclusions.
2


Chapter II. The Health Care System in Bolivia
Background on the Bolivian Health Sector
Bolivia has a mix of public and private health care. The major provider of public health
care is the Ministry of Social Welfare and Public Health (the Ministry), which operates
hospitals, health centers, and health posts. Health centers offer limited services, have some
laboratory equipment, and are staffed by a doctor. Health posts have little equipment and are
attended primarily by nurses' aides. The next largest provider is the Bolivian Institute for Social
Security (Social Security). It operates hospitals only for insured workers, pensioners, and their
dependents. Benefits include free diagnosis, therapy, and drugs. Approximately 300 non-
governmental organizations (NGOs) constitute the third provider of health care services. Based
on 1992 estimates from a Ministry expert, the Ministry covers 25 percent of the population,
Social Security approximately 20 percent, and NGOs 10 to 25 percent (in urban areas they reach
about 10 percent of the population and in rural areas about 25 percent). The private for-profit
sector covers 5 percent. An estimated 30 percent of the population in Bolivia does not receive
formal medical care. Some of these people consult traditional healers known as yatiri,
curandero, callawaya, and naturista. This informal medical care is still prevalent in Bolivia.
The number of people who consult traditional healers varies with economic trends and increased
during the economic crisis of 1985-87.
This paper focuses on the demand for medical care in urban areas, where a large private
sector exists. For example, our sample data show that among the people who sought formal
medical care, 33.1 percent received medical care at for-profit private institutions, 23.9 percent
at the Ministry, 27.3 percent at Social Security, and 15.7 percent at NGOs. Price data can be
constructed for each health institution in terms of both money and opportunity costs. Tables 1
Table 1: Money Cost of Obtaining Medical Care
(in Bolivianos, Bs: 1 US Dollar = 3.40 Bolivianos, December 1990)
Number      Consulting   Travel Cost
User Fees    of Visits  Fees Per Visit  (one-way)     Medicine Cost
N    Mean  S.D. Mean   S.D. Mean    S.D.  Mean   S.D.   N    Mean   S.D.
Public Health  374  20.44 72.12  3.00  2.89  9.40  30.44  3.31  10.66 340  77.49  239.24
Social Security 448  3.33 47.70  2.93  2.78  0.58  5.06  3.56  11.14 416   43.44  202.66
Non-         222   21.35 112.95  3.19  3.40  5.56  16.71  4.26  12.44 213  41.02   79.71
Governmental
Organization
Private      548   31.85 73.29  2.45  2.65  12.82  26.46  2.87  9.39 518   81.96  155.88
Self-Care   1151    0.90  12.37  0.00  0.00  0.00  0.00  0.00   0.00 767   14.31   39.59
Total       2743   11.80 58.33  1.64  2.59  4.40  17.91  1.95   8.28 254   47.39  153.86
Source: Encuesta Integrada de Hogares (1990).
Notes: Data are only for adults 16 years or older. S.D. = standard deviation.
3


Table 2: Qpportunity Cost of Obtaining Medical Care
Traveling Time
Waiting Time (Minutes)    (Minutes, One-Way)
Mean     S.D.              Mean   S.D.
Public Health           69.21    84.16             31.54  89.99
Social Security          71.25  104.14             31.62  83.40
Non-Governmental        57.05   101.95             40.38  102.22
Organization
Private                  31.73   61.19             37.57  108.52
Self-Care                0.00    0.00               0.00  0.00
Total                    32.03   72.37             20.24  75.70
Source: Encucata Integrada de Hogarce (1990).
Note: Data are only for adults aged 16 years or older.
and 2 show average cost and waiting and traveling time for the adult outpatients reported in the
survey. The sample faces an average consulting cost of about 20 bolivianos at the Ministry and
at NGOs, about 31 bolivianos at private clinics, and about 3 bolivianos at Social Security.
Medicine costs are, on average, almost twice as high at the Ministry and private clinics than at
Social Security or NGOs. The same sample faces an average waiting time of about one hour
at the Ministry, Social Security, or NGOs, and about a half an hour at private clinics.
There has been a significant increase in efforts to recover costs in the Bolivian health care
system over the last decade. About 50 percent of health facilities, however, still provide free
health care. Provision for fee waivers or fee reductions for the poor are made by a hospital's
social worker or by the institution. Criteria frequently used include income, size of household,
and number of employed household members. The final evaluation and decision is subjective
and strictly the responsibility of the individual (World Bank 1989a, p.53). In our sample of
adult outpatients, 45.3 percent were treated free of charge. The percentage receiving free care
does not vary much with income level although the mean consulting fee increases with income
level (Table 3).
However, in the sample of pregnant women the percentages of those receiving free care
decreased markedly as income rose (Tables 4 and 5). More than 50 percent of women from the
poorest 20 percent of households received free care while only 9.3 percent from the richest 20
percent of households received free care when they delivered babies (Table 4). Almost 40
percent of women from the poorest 20 percent received free prenatal care while less than 20
percent of women from the richest 20 percent received free prenatal care (Table 5). Cost of
delivery and cost per visit for prenatal care increased as income rose.
In the survey, more than 50 percent of adult patients from the poorest 10 percent of
households answered that expense was a major reason why they did not seek medical care,
whereas only 3 percent of adult patients from the richest 10 percent of households gave the same
4


Table 3: Consulting Fees for Formal Medical Care by Monthly Income Level (in Bolivianos)
Income Group                          Consulting Fee
Percent Receiving
Quintile                           N    Mean       S.D.                    Free Care
1: 13.71 - 79.38                 205     4.97     12.64                     45.9
2: 79.40 - 121.66                236     9.19      25.84                    47.9
3: 121.67 - 180.17               327    11.86      36.40                    43.1
4: 180.21 - 288.42                373   16.79      43.91                    47.5
5: 288.56 - 6296.33              450    40.00     127.49                    43.3
Source: Encuesta Integrada de Hogares (1990).
Notes: 1. Quintile of per capita income measured by per capita monthly household consumption. Quintile I is the lowest.
Quintiles are created for the full sample.
2. Data are only for adults aged 16 years or older.
Table 4: Cost of Delivering a Child by Monthly Income Level (in Bolivianos)
Income Group                          Consulting Fee
Percent Receiving
Quintile                           N    Mean       S.D.                    Free Care
1: 13.71 - 79.38                 243    25.84     44.22                     54.32
2: 79.40 - 121.66                 176   54.45      63.02                    35.23
3: 121.67 - 180.17                160   78.04     102.80                    35.63
4: 180.21 - 288.42                111  160.16     234.20                    32.43
5: 288.56 - 6296.33               86   657.47    727.59                      9.30
Source: Encuesta Integrada de Hogares (1990).
Notes: 1. Quintile of per capita income measured by per capita monthly household consumption. Quintile 1 is the lowest.
2. Data are only for women aged 13 years or older.
Table 5. Cost per Visit of Prenatal Care by Monthly Income Level (in Bolivianos)
Income Group                          Consulting Fee
Percent Receiving
Quintile                           N    Mean       S.D.                    Free Care
1: 13.71 - 79.38                 205     4.97     12.64                     45.9
2: 79.40 - 121.66                236     9.19     25.84                     47.9
3: 121.67 - 180.17               327    11.86     36.40                     43.1
4: 180.21 - 288.42               373    16.79     43.91                     47.5
5: 288.56 - 6296.33              450    40.00     127.49                    43.3
Source: Encuesta Integrada do Hogares (1990).
Notes: 1. Quintile of per capita income measured by per capita monthly household consumption. Quintile I is the lowest.
2. Data are only for women aged 13 years or older.
5


reason. As everywhere else in the world, Table 6: Structure of Government Revenue,
health care agencies in Bolivia have a difficult Bolivia, 1989
time reaching the poor groups in the society.
Source of           Percent of Total
Health Care Financing                      Government Revenue  Government Revenue
TorAL INcomE TAX         5.8
The public health sector is financed  TOTAL INDIRECr TAX     73.4
from different sources from year to year, but  Value Added Tax     23.4
typically 50 percent of the finances come   Specific Taxes          3.9
from the national treasury, 25 percent from  Transaction Tax        3.6
external aid, and 25 percent from user fees.  Direct Taxes         42.5
(Bolivian Petroleum
Corporation)
The national treasury gets the majority  TOTAL INTERNATIONAL TAX  11.0
of its budget from tax revenues, which are  Import Duties           11.0
broken down by type of tax in the table    PROPERTY TAxEs           3.1
below.                                     OrHER TAXES              3.4
ToTAL TAX REVENUE        96.7
The heavy dependence on the value   NON-TAx REvENUE           3.3
added tax (VAT) and the fact that the VAT
rate is a uniform 10 percent (it was increased                     loo.o
to 13 percent in March 1992) has raised
complaints that the system  is regressive. Source: World Bank, Bolivia Country Economic Memorandum
However,    the   largest  proportion  of (1989c).
government revenue (42.5 percent) is
generated through an excise tax on petroleum. The Bolivian government imposes a direct
taxation and a transfer system on the Bolivian Petroleum Corporation. Specific taxes are
primarily imposed on luxury goods and those with the lowest incomes are exempt from some
taxes and evade others by working in the informal sector.
Other sources of finance for health care are external aid and the social security system.
In 1985, external aid was from USAID (35 percent), UNICEF (25 percent), and PAHO\WHO
(19 percent). Additional sources of external aid were channeled directly through the Ministry.
Table 7: Disbursements of External Grants and Credits to the Health Sector, 1985 (in millions US$)
Source of Financing                    Amount                       Percent
UNICEF                                  1.95                         24.9
PAHO/WHO                                1.45                         18.5
USAID                                   2.71                         34.6
UNFPA                                   0.60                          7.7
World Food Program                      1.12                         14.3
Total                                   7.82                        100.0
Source: The Ministry of Social Welfare and Public Health (1986).
Notes: PAHO-Pan American Health Organization; UNFPA-United Nations Fund for Population Activities.
6


Chapter II. Empirical Specification
Behavior in medical markets is distinguished by the roles that physical need and life-cycle
patterns play in determining demand. Circumstances, such as accidents, pregnancies, and
infections, often dominate health care consumption decisions. Many medical needs are age and
sex specific, such as immunizations early in life, pregnancy care during fertile years for women,
and the onset of degenerative diseases late in life.
The major policy issues for demand analysis in high-income countries are the rising costs
of medical care provision and the possible overuse of physician services. however, in low-
income countries, the core problems relate to the access to facilities, the capturing of true
demand patterns, and demand creation - in general, how to assure that basic health care
services are used.
Anthropologists, sociologists, and geographers have had a considerable amount of success
finding negative correlations between medical service use and so-called "barriers to utilization."
These barriers include such items as physical distance from households to facilities, cultural
distance between patients and providers, the unavailability of drugs, the length of time spent
waiting at facilities, and the unavailability of transportation.
From an economic perspective, only a few have approached the issue of demand analysis
of medical care (Heller 1982 and Akin et. al. 1985), Gertler and van der Gaag 1990). Heller
found very little sensitivity to prices for total annual medical visits made in a rural Malaysian
sample.  Akin et al. also found that prices and distance are not nearly as important as
determinants of demand for medical services in a rural region of the Philippines as has usually
been assumed by planners. Gertler and van der Gaag found that utilization of health services
tends to be relatively price inelastic. That is, a unit increase in price leads to increased
revenues. Again, however, these are point estimates at the mean along the demand curve. The
same study showed that demand may be relatively more price elastic at lower income levels.
My research builds on the studies done by Gertler and van der Gaag (1990). However,
their studies are about demand for medical care in rural areas where the sample frames consist
entirely of farm households, with very limited choice in health facilities. In Cote d'Ivoire, at
the time of the surveys (1985), no fee was charged. The price of medical care was the
opportunity cost of time spent in obtaining care. The opportunity cost of time was calculated
as the product of the round-trip travel time and the individual's wage rate.' For children, the
opportunity cost of the mother's time was used. The round-trip travel time for each individual
to each alternative provider comes from the community survey, and the appropriate village-level
agricultural wage rate for males and females is taken as the unit opportunity cost of time. In
1985-86 in Peru, user fees for public medical care were very low. Gertler and van der Gaag
1. An additional cost is the amount paid for transportation. In rural Cote d'Ivoire the vast majority of people
reported walking, which suggests negligible transportation costs. Another potential time price is waiting and
treatment time. These data are not available in the survey.
7


used the department's median clinic and hospital fee paid by individuals in their sample as the
monetary price.
My study focuses on the demand for medical care in urban areas where about 50 percent
of the total population are wage earners. Urban patients face a wider choice of health facilities
than those living in rural areas. In Bolivia, as seen in the previous section, there has been
considerable cost recovery by user fees at both public and private health facilities. This study
focuses on child and maternal health as well as on the general population's health, since children
under 5 and women of reproductive age (15-49) are particularly vulnerable groups.
I develop a demand model under the assumption that sick individuals face a choice among
self-treatment and several professional treatments. The model is formulated as follows.
Individual consumer i experiences an accident, illness, or pregnancy. The individual must
decide whether to obtain medical care, where to obtain care, and what quantity of services to
consume. The decision is made in order to maximize utility.
For the ith individual faced with J + 1 choices, suppose that the utility of choice j is
(1) UY = a'XV + PZ + er
Utility depends on -XO, the observed attributes of the choice j perceived by individual i, and Zi,
the observed characteristics of individual i. A disturbance term Pj captures unobserved
variation in tastes and in the attributes of alternatives, and errors in the perception and
optimization by the consumer. If the consumer makes choice j, then UV must have been the
maximum among the J + 1 utilities. Hence, the statistical model is driven by the probability
that choice j is made, which is
(2) Prob [ UV > UL1]  for all other k j.
The model is made operational by choosing a distribution for the disturbances. Let Y be a
random variable indicating the choice made. It takes one of the value 0,1,2,...,J. McFadden
(1973) has shown that if and only if the J + 1 disturbances are independent and identically
distributed with the type I extreme value distribution:
(3) P ( eu <a) = EXP( -e-),
8


then,
EXP( aX
(4) Prob[ Y =1 j   EP                , j = 0, 1,2,...J
[ EXP( a'XV + p;Z)
J-0
This leads to what Greene (1993) calls the mixed model (a combination of the conditional logit
model and multinomial logit model) or what Maddala (1983, p.44) calls the general model.
Utility depends on Xi,, the attributes of the choice j for individual i, and Z,, the characteristics
of individual i. The attributes of the choice j, Xj, vary with choice and individual, while the
individual characteristics Z1 are the same for all choices and vary only across individuals.
In my study, the three attributes for each choice j are the consulting fee, waiting time at
the health facility, and traveling time to the health facility. I assume each individual knows
which health care facilities are available, the expected costs of the facilities, waiting time at the
facilities, and traveling time to the facilities.2 Not everyone will have access to all existing
health facilities; for example, individuals who are working in the informal sector and do not pay
social security taxes may not receive care from Social Security hospitals. As explained in
Chapter 2, I include a dummy variable to indicate if an individual is from a family with a
member employed at a company which registers Social Security. Individual characteristics are
age, sex, per capita consumption, the number of sick days in the previous month, education,
language spoken, working status, number of adults aged 16 years or older in the household, and
number of children aged 15 years or younger in the household. For the demand by infants aged
between zero and four, the relationship between household structure and demand for medical
care will be examined. For example, if a child lives with a grandmother, he or she may be well
taken care of within the home.
Vector a gives the vector of "implicit prices" for the health facility attributes.' By
splitting the sample, a different set of "implicit prices" could have been obtained for each
characteristic, for example, Spanish-speaking male, Spanish-speaking female, Aymara-speaking
male, and Aymara-speaking female. These coefficients give the relative valuations of the three
attributes - consulting fee, waiting time, and traveling time - for each of these groups. Also,
if given the attributes of a new facility, as perceived by an individual, then the estimated
coefficients can be used to predict the probability that this individual will use the facility.
Therefore, the probability of choosing the category considered in the estimation procedure can
be predicted given a new set of attributes X4. The number of elements in vector a, is the
2. This assumption is crucial in the multinomial logit model specification. Bitran (1990), however, found that "most
respondents claimed either not to know anything about the choices not made or to have only a vague idea about the
choices' characteristics" in the survey done in Zaire. Bitran also said that "Some respondents, however, were able
to provide some cardinal information about distances and prices of the choices not made, relative to those options
as well as to the choice made, e.g., provider X is farther away and less expensive than Y".
3. I normalize this vector by setting the first element of a equal to 1.
9


number of attributes of a health facility, namely three, as well as the number of alpha parameters
to be estimated. On the other hand, the number of elements in vector P, is the number of
individual characteristics, L. The number of beta parameters to be estimated is equal to L
multiplied by J-1, where J is the number of health facilities. With the estimated parameters and
a new individual with specified characteristics, the probability that the individual will choose one
of the J health facilities considered can be predicted.
Let choice 0 be the self-care alternative, let choices 1,2,3, or 4 be the various provider
alternatives, and let Xe=0.4 Therefore, the self-care demand function (that is, the probability
of choosing self-care) is
(5) Prob[ Y, 0 ]
1 +  EXP(a'X, + P,)
i-1
and the demand for provider i is
EXP( aX + pZ,)
(6) Prob[ Y, j  -                      .____a_I___+_P; _
i
1 + E EXP( a iX, +ps
The log likelihood for this problem is
(7) In L, E  D. In Prob[ Y,j],
J-0
where DU is a dichotomous variable that takes on the value 1 if individual i chooses alternative
j, i.e. if Y, =j.
4. Xj. are the consulting fees, waiting time and traveling time for self-care.
10


IV. The Data
The data used in this study are drawn from the third year of the Encuesta Integrada de
Hogares (EH), a multipurpose household survey conducted by Instituto Nacional de Estadistica
(INE) in Bolivia with the financial support of the World Bank. The survey was conducted in
departmental capital cities and in a sample of cities with populations greater than or equal to
10,000. The data covered 6,346 households and 30,350 individuals from nine departments of
Bolivia in 1990. A stratified random sample of 6,346 households was chosen to reflect the
population in 10 cities. In the stratification, barrios (neighborhoods) were randomly selected in
each of the statistical zones used by INE and then households were randomly selected within the
selected barrios. I consider such barrios to be the communities for which I control for
unobserved community effects. There are 319 barrios in the data set. This multipurpose survey
collected information on family background and the resources available to households. Thus,
it gathered data on migration, education and training, employment (for both first and second
occupations and for both formal and informal sectors), unemployment (for both active and
inactive people), health, fertility, children's health, housing (including housing materials, water
service and sewerage service), income, and expenditure.
Detailed information on quantity and type of health care used was obtained from all
household members who reported an illness or injury during the four weeks prior to the
interview. The arguments of the demand functions are household monthly consumption, the cost
of obtaining medical care, and individual and family characteristics that may affect the efficacy
of medical care.
The measure of total household consumption includes imputed rents from their own
houses. The total cost of obtaining medical care is the sum of monetary prices and opportunity
costs. The monetary price is the sum of consulting fees, cost of medicines, and traveling costs.
The opportunity cost of time is the sum of round trip travel time and waiting time.
Variables that may influence the efficacy of medical care include age, the number of days
in the previous month the individual was sick, education, language spoken, regions, working
status (whether work is in the informal or formal sector), the number of other adults in the
household, and the number of children in the household. There are three major languages
spoken in Bolivia (plus a number of minor ones): Spanish, Quechua, and Aymara. Almost
everyone in the survey claims that they can speak Spanish. Since the two prominent Indian
languages are associated with different socioeconomic histories and options than the European
language, I include the language variable to see whether there are differences in demand for
health care among those who claim that they can speak one or the other of the two major Indian
languages and those who claim that they can speak only Spanish.
There have been wide regional variations in Bolivia, in part because of the varied
topography. In recent decades, for example, the central region stretching from La Paz, the
capital of Bolivia, through Cochabamba to Santa Cruz has been the center of activity. I divide
the sample into three regions: Central (La Paz, El Alto, Cochabamba, Santa Cruz), North
(Cobija, Trinidad), and South (Sucre, Oruro, Potosi, Tarija).
11


The informal sector is large, both in terms of share of employment and, especially, in
terms of number of economic units. Many in the informal sector sell small amounts of their
goods on the street, or labor in tiny family run workshops. The informal sector, as defined in
Horton (1990), includes workers on own account (excluding professionals), domestic employees,
and employees, employers, and unpaid family workers in establishments with less than five
workers.
The sample statistics for adult outpatients reveal the following characteristics:
approximately 50 percent seek formal medical care, the average age is 42.4 years, the average
education is 7.7 years, and 40 percent of outpatients are men (Table A. 1).
Estimation of a discrete choice model requires information about provider characteristics,
such as price, traveling time, and waiting time, for each individual and treatment option,
regardless of the choice actually made. The definition and measurement of prices in the health
care market pose difficult theoretical and empirical problems. The problems arise because health
care services vary in quality and, thus, their individual prices are not comparable.
Two main schools of thought regarding the appropriate price can be found in the
empirical literature on health care demand. Researchers such as Akin et al. (1985) conducted
health facility surveys and used the provider-reported price of a visit for each type of health
service. Such a price is a fixed datum which is independent from patient characteristics.
On the other hand, other researchers such as Gertler et al. (1987) and Gertler and van
der Gaag (1990) use self-reported prices from the household survey. Gertler et al. (1987) use
as price the predicted price from an estimated hedonic price equation with individual and
provider characteristics as independent variables. Gertler and van der Gaag (1990) replace the
missing prices with median expenditures based on the experiences of other users.
In Bolivia, no health facility survey has been done, therefore I have to rely on household
level expenditure data. The method that uses provider-reported prices has its advantages and
problems. One advantage is that the survey work is facilitated considerably by not having to
gather price information from  consumers.  Also, recall problems are avoided.  The
disadvantages are important, however, and may outweigh the advantages. Firstly, it is well
known that some health care providers price discriminate, often in an ad-hoc manner. Recording
of a single price thus ignores the presence of price discrimination. Secondly, some providers
have complex price schedules, with various prices for different types of health problems and
with prices varying depending on the types and amounts of drugs and other supplies provided
to the patient. Gathering this information as well as the criteria that would allow the analyst to
assign the appropriate price to users can be very difficult. Thirdly, some consumers have health
insurance that allows them to receive care free of charge, or at reduced prices from certain
providers. Provider-reported prices would thus have to be adjusted based on the insurance
information provided by users.
For each alternative, I use the median reported payment and opportunity cost in each city
as the expected price of that provider.
12


Chapter V. Empirical Results
Adult Outpatients Aged 16 Years or Older
Maximum-likelihood estimation results of the multinomial logit model for adult
outpatients aged 16 years or older are presented in Table 8a. Patients face five health care
choices: Ministry of Public Health, Social Security, NGOs, private, and self-care. All the
estimated coefficients are relative to self-care. The provider's characteristics are user fee,
waiting time, and travel time. The user fee coefficient is negative and the coefficient associated
with the fee-times-per capita consumption variable is positive. These signs suggest that relative
to lower-income people, higher income individuals are more likely to choose outside providers
and less sensitive to price changes.
For the individual's characteristics, age, education, gender, per capita consumption, work
status, language,s duration of illness, and other household attributes have statistically significant
coefficients for some providers. For example, relative to self-care, age and education also have
positive effects on the choice of providers. Working in the informal sector has significantly
negative effects on all formal care provider choices relative to self-care.
Using the median values of all independent variables as a benchmark, Tables 8b and 8c
provide the results of changing one individual, household, or provider characteristic. The effects
on probability of changes in individual and household characteristics are explored in Table 8b.
The effects on probability of changes in provider characteristics are analyzed in Table 8c. Each
row in Tables 8b and 8c adds up to 100 percent.
The findings emanating from an analysis of Table 8b are as follows:
Age. A person's age has a positive effect on the decision to seek care. In particular,
the elderly are much more likely to seek care in the presence of a health problem than
are other age categories. At age 20, the probability of a person to care for himself is
57.3 percent, while at age 60 it is 41.2 percent. Young people may have a higher
opportunity cost of time. Age also affects the choice of provider, with people aged 60
years or older showing a marked preference for Social Security and private facilities.
Since Social Security facilities are well-equipped for curative care, the elderly may prefer
them.
Education. The level of formal education of the person has a positive effect on the
probability to seek care, suggesting that the more educated place a higher value on
treatment. Among the providers, the more educated choose private facilities. Educated
people might regard the quality of care provided at private facilities to be better.
5. The omitted language is Quechua and other minor languages.
13


Table 8a: Choice of Provider, Multinomial Logit Coefficients and Standard Errors (Adults Aged 16 Years
or Older)
Ministry of
Public Health       Social Security         NGOs                Private
Percentage
Choosing:                13.6                16.3                 8.1                 20.0
Standard             Standard            Standard            Standard
Variables        Coefficient  Errors  Coefficient  Errors  Coefficient  Errors  Coefficient  Errors
Constant           -3.2975c   0.4896   -3.6133c   0.4996    -1.6896c   0.5591   -2.4418c   0.4891
User fee           -0.0198    0.0172   -0.0198    0.0172    -0.0198    0.0172   -0.0198    0.0172
User fee times per
capita consumption  0.0030    0.0052    0.0030    0.0052     0.0030    0.0052    0.0030    0.0052
(100 Bs.)
Waiting time       -0.0029    0.0024   -0.0029    0.0024    -0.0029    0.0024   -0.0029    0.0024
Travel time         0.0002    0.0026    0.0002    0.0026     0.0002    0.0026    0.0002    0.0026
Age                 0.0438b   0.0188    0.0624c   0.0182     0.0180    0.0220    0.0096    0.0155
Age squared        -0.0004b   0.0002   -0.0003'   0.0002    -0.0002    0.0002    0.0001    0.0002
Education (years)   0.0139    0.0141    0.0405c   0.0133     0.0144    0.0169    0.0348c   0.0123
Gender (1=male)   -0.0397    0.1281   -0.0490    0.1321    -0.9282    0.1563   -0.3426c   0.1180
Per capita cons.    0.2144c   0.0488    0.0982b   0.0460     0.1116'   0.0590    0.2903b   0.0937
(100 Bs.)
Work status        -0.2268'   0.1317   -1.2142c   0.1612    -0.4525c   0.1630   -0.1996'   0.1205
(1= informal)
Speaking Spanish    0.3542b   0.1535   -0.1161    0.1458    -0.2565    0.1706    0.1639    0.1320
(1 =yes)
Speaking Aymara     0.2466    0.2116   -0.5895b   0.2446    -0.8208c   0.2773   -0.6116c   0.2183
(1 =yes)
Number of adults
in household (aged  0.0234    0.0401   -0.1573c   0.0434    -0.1484c   0.0546    0.0251    0.0358
16 or older)
Number of
children in         0.0903b   0.0375   -0.0405    0.0427     0.0515    0.0475    0.0821b   0.0353
household (aged
15 or younger)
Duration of illness  0.0572c  0.0064    0.0439c   0.0071     0.0306'   0.0085    0.0417c   0.0063
(days)
Social Security     0.1638    0.1488    2.1948c   0.1459     0.3797b   0.1757    0.4336c   0.1280
(1= yes)
Log-likelihood        -3546.568
Number of               2743
observations
Source: Encuesta Integrada de Hogares (1990).
Notes:
a. Significant at 10 percent level (two-sided).
b. Significant at 5 percent level.
c. Significant at 1 percent level.
14


Table 8b: Probabilfty of Coosing a Provider (7he Effeas on Probability of Oanmges in Individual and
Household Attributes, Adults Aged 16 Years or Older)
Minitry of
Public Healkh  Social Security  NGOs     Se -Care     Private
~~.                                                       ........c
Age (Years)
20                                  12.4          4.0           9.8        57.3        16.5
40                                  15.4          8.0           9.7        48.1        18.7
60                                  14.1         12.7           8.7        41.2        23.4
Education
0                                    15.3          6.6          9.6         52.5        16.0
2                                    15.3          6.9          9.7         51.3        16.8
4                                    15.4          7.4          9.7         50.0        17.5
10                                   15.4          8.7           9.8        46.2        20.0
Gender
Male                                16.0          8.2           9.6        51.9        14.3
Female                              15.4          8.0           9.7        48.1        18.7
Per capita consumption quintiles (median)
Quintile 1: 57.87                    13.7          8.1          9.6         53.7        15.0
Quintile 2: 99.00                    14.4          8.1          9.7         51.4        16.5
Quintile 3: 146.38                   15.2          8.0          9.7         48.8        18.3
Quintile 4: 223.69                   16.1          7.9          9.8         44.6        21.7
Quintile 5: 404.88                   19.2          7.2          9.3         34.0        30.3
Work status
Informal                            14.6          2.8           7.3        57.1        18.2
Not informal                        15.4          8.0           9.7        48.1        18.7
Language
Spanish                             15.4          8.0           9.7        48.1        18.7
Aymara                              17.1          6.1           6.8        59.3        10.6
Quechua and other languages         11.2          9.3         13.0         49.9        16.5
Number of adults aged 16 or older
0                                   13.3         11.9          14.1        44.6        16.1
1                                   14.0         10.4         12.5         46.0        17.0
2                                   14.7          9.2          11.1        47.2        17.9
3                                   15.4          8.0           9.7        48.1        18.7
Number of children aged 15 or younger
0                                   13.6          9.2           9.3        51.0        16.8
1                                   14.5          8.6          9.5         49.6        17.8
2                                   15.4          8.0           9.7        48.1        18.7
3                                   16.3          7.4           9.9        46.6        19.7
Duration of illness
1                                   14.4          7.7          9.6         50.4        18.0
3                                   15.4          8.0           9.7        48.1        18.7
5                                   16.5          8.3           9.9        45.9        19.4
10                                  19.3          9.1         10.1         40.4        21.0
Social Security
Yes                                  10.0         39.6           7.9        26.6        16.0
Source: Encuesta Integrada de Hogares (1990).
15


Gender. Females appear to be more likely to seek care than males. The probability of
treating themselves is 48.1 percent for women, while it is 51.8 percent for men. Among
health care providers, women are more likely to choose private providers. The
probability of choosing private facilities is 18.7 percent for women and 14.3 percent for
men.
Per capita consumption. People from wealthy families seek more formal care, and
among formal care options, the Ministry of Public Health or private facilities are
preferred. For the highest income group, 30.3 percent choose private facilities.
Work Status. People working in the informal sector are more likely to choose self-care
(57.1 percent) and less likely to go to Social Security (2.8 percent).
Language. Compared to Spanish speakers or Quechua speakers, Aymara speakers are
more likely to choose self-care and are less likely to go to Social Security, NGOs, or
private facilities.
Number of adults aged 16 years or older. With more adults in the household, people
seek less professional care. They may receive relatively more support from family
members than those who live with fewer adults.
Number of children aged 15 years or younger. The probability to seek care when ill
increases with the number of children aged 15 or younger in the household.
Duration of illness. As expected, the duration of the health problem is positively related
to the probability of seeking care. Generally, in the presence of an illness, people tend
to wait a few days before seeking care hoping that the problem will correct itself.
Social Security. Among those eligible for Social Security, almost 40 percent of the
patients go to Social Security facilities.
Table 8c shows how changes in user fees and waiting times at Public Health and Social
Security facilities have an effect on the probability of seeking care and the choice of provider.
Increases in user fees from 5 Bolivianos to 20 Bolivianos at Ministry of Public Health facilities
result in the decrease of patients by 2.7 percent. Some people (1.6 percent) have to give up
formal care, 0.6 percent switch to private care, 0.2 percent switch to Social Security, and 0.3
percent to NGOs. People regard paying 10 Bolivianos and waiting 30 minutes at Public Health
facilities, almost the same as paying 5 Bolivianos and waiting 60 minutes. Similarly, people
regard paying 5 Bolivianos and waiting 30 minutes at Social Security facilities almost the same
as paying nothing but waiting 60 minutes. The effects of the price changes are explored in
Table 8d.
16


Table 8c: Probability of Choosing a Provider (The Effects on Probability of Changes in Health
Facility Attributes, Adults Aged 16 Years or Older)
Ministry of
Public Health Social Security  NGOs     Private    Sef-Care
P       PwdictedPerentage Choosing- 15.4             8.0           11
User fee - Public Health
0                                   16.4         7.9          9.6        18.5       47.6
5                                   15.4         8.0          9.7        18.7       48.1
10                                  14.5          8.1          9.8        18.9       48.7
20                                  12.7          8.2         10.0        19.3       49.7
User fee - Social Security
0                                    15.4         8.0          9.7        18.7       48.1
5                                    15.5         7.4          9.8        18.8       48.4
10                                   15.6         6.9          9.8        18.9       48.7
Waiting time - Public Health
15                                    16.0         7.9          9.7        18.6       47.8
30                                    15.4         8.0          9.7        18.7       48.1
60                                    14.3         8.1          9.9        19.0       48.8
Waiting time - Social Security
15                                    15.4         8.3          9.7        18.7       48.0
30                                    15.4         8.0          9.7        18.7       48.1
60                                    15.5         7.4          9.8        18.9       48.5
Baseline (Median Values)
Age                                             40 years old
Education                                       7 years of education
Gender                                          Female
Household per capita consumption (monthly)      157.36 Bolivianos
Work status                                     Not working in informal sector
Language                                        Spanish
Number of adults aged 16 years or older         3 persons
Number of children aged 15 years or younger     2 persons
Duration of illness                             3 days
Social Security                                 No
Ministry of
Public Health Social Security  NGOs     Private
User fee                                5             0           4           15
Waiting time                            30           30           30          15
Traveling time (round trip)             30           30           30         30
Source: Encuesta Integrada de Hogares (1990).
17


Table 8d: Elasticities of Probability with Respect to User Fee
Own- and Cross-Price          Ministry of
Elasticities                 Public Health  Social Security  NGOs       Private       Self-Care
Price of Ministry               -0.064         0.012         0.011        0.012         0.012
of Public Health
Price of Social Security         0.001         -0.007        0.001        0.001         0.001
Price of NGOs                    0.006         0.006        -0.054        0.006         0.006
Price of Private                 0.042         0.043         0.042       -0.183         0.042
Own-Price Elasticities        Quintile 1     Quintile 2   Quintile 3    Quintile 4   Quintile 5
Price of Ministry               -0.079         -0.072       -0.065       -0.055        -0.031
of Public Health
Price of Social Security        -0.008        -0.008        -0.007       -0.006        -0.004
Price of NGOs                   -0.066        -0.060        -0.056       -0.047        -0.028
Price of Private                -0.228         -0.210       -0.189       -0.155        -0.081
Own-Price Elasticities         Spanish         Aymara      Quechua        Male        Female
Price of Ministry               -0.064         -0.063       -0.066       -0.063        -0.064
of Public Health
Price of Social Security        -0.007         -0.007       -0.007       -0.007        -0.007
Price of NGOs                   -0.054         -0.056       -0.052       -0.054        -0.054
Price of Private                -0.183         -0.221       -0.206       -0.211        -0.183
BIAsid*ties6f Frobibiliywt RWpei to waliigTi
Own- and Cross-Time           Ministry of
Elasticities                 Public Health  Social Security  NGOs        Private      Self-Care
Time of Ministry                -0.074         0.150         0.012        0.014         0.014
of Public Health
Time of Social Security          0.000         -0.079        0.000        0             0.007
Time of NGOs                     0.002         0.008        -0.080        0.002         0.009
Time of Private                  0.005         0.008         0.080       -0.035         0.008
Source: Encuesta Integrada de Hogares (1990).
Table 8d shows elasticities of probability with respect to user fees and times (waiting time
in this sample and travel time as well in other samples). Elasticities are defined as
aP,  xk
aP   x
18


where xj is the median value of facility characteristics (price, waiting time, or traveling time)
and P. is a probability of choosing j* facility evaluated a vector of median values x of individual,
household, and facility characteristics."
Similar to the findings of several other studies of health care demand in developing
countries, (e.g., Akin et al. 1986 and Gertler and van der Gaag 1990) our results show that
demand for care in urban Bolivia is relatively inelastic. Elasticities are highest at private
facilities, -0.183, and lowest at Social Security facilities, -0.007. The estimates show that the
price elasticities of demand fall with income, so that poor individuals are more price sensitive
than wealthy individuals. Lower price elasticities imply that user fees can be a source of income
for the health care system. Higher price elasticities for poor individuals, however, indicate that
further increases in user fees will be regressive and may reduce the use of medical care by the
poor unless some protective measures are provided for the poor. Aymara-speaking people have
a slightly more elastic demand than Spanish- or Quechua-speaking people except in the choice
of Ministry of Public Health services. Price elasticities do not vary much between male and
female except for private facilities. At private facilities, males have more a elastic demand (-
0.211) than do females (-0.183).
Elasticities of probability with respect to waiting time are also low. Elasticities for
Public Health, Social Security, and NGO facilities are between -0.074 and -0.080, while
elasticity for private facilities, at -0.0035, is slightly lower than for the other three types of
health facilities.
Child Outpatients Aged 15 Years or Younger
Table 9a shows the maximum-likelihood estimation results of the nultinomial logit model
for child outpatients aged 15 years or younger. Patients face five health care choices: Ministry
of Public Health facilities, Social Security facilities, NGOs, private, and self-care. All the
estimated coefficients are relative to self-care. The coefficient of the traveling time is negative.
This implies that the probability that people will seek care outside the home from any provider
decreases with the traveling time to the provider. The user fee coefficient is negative and the
coefficient associated with the fee-times-per capita consumption variable is positive. These signs
suggest that relative to lower-income people, higher income individuals are less sensitive to price
changes. The magnitude of the coefficient (0.0200), however, is so large that at per capita
consumption of 100 Bolivianos, the price effect becomes almost zero. The price elasticities by
different income levels are explored later in Table 9d.
For individual characteristics, relative to self-care, gender and father's education have
no statistical significance for any providers. Age, mother's education, per capita consumption,
language, duration of illness, and other household attributes are statistically significant on the
choice of provider. For example, age has a negative effect on the choice of professional
6. Since the median value of the user fees at Social Security facilities is zero, the probability of choosing Social
Security facilities is evaluated at 0.5 Bolivianos.
19


Table 9a: Ooice of Provider Multinomial Logit Coefficients and Standard Errors (Children Aged 15
Years or Younger)
Ministry of
Public Health      Social Security        NGOs                Private
Percentage
Choosing:              10.2               10.2                 7.7                 12.5
Standard            Standard           Standard              Standard
Variables        Coefficient Errors  Coefficient  Errors  Coefficient  Errors  Coeffcient  Errors
Constant          -1.4763'  0.3830   -3.3043'   0.4449   -2.0410'   0.4187    -3.24100   0.4740
User fee          -0.0201   0.0255   -0.0201    0.0255   -0.0201    0.0255    -0.0201    0.0255
User fee times per
capita cons.       0.0200'  0.0114    0.0200'   0.0114    0.02009   0.0114     0.0200   0.0114
(100 Bs.)
Waiting time       0.0047   0.0031    0.0047    0.0031    0.0047    0.0031     0.0047    0.0031
Travel time       -0.0069'  0.0022   -0.0069'   0.0022   -0.0069'   0.0022    -0.0069c   0.0022
Age               -0.1842'  0.0615   -0.0527    0.0659   -0.0317    0.0690    -0.0353    0.0590
Age squared        0.0116'  0.0044    0.00878   0.0045    0.0046    0.0049     0.0045    0.0042
Mother's          -0.0025   0.0186    0.0382b   0.0182   -0.0029    0.0217     0.0444b   0.0173
education (years)
Father's          -0.0248   0.0164   -0.0065    0.0167    0.0009    0.0193     0.0168    0.0152
education (years)
Gender (1=male)   -0.0916   0.1527    0.1366    0.1664    0.0285    0.1706    -0.1240    0.1458
Per capita cons.   0.1688b  0.0846   -0.0812    0.0846   -0.0399    0.097     -0.1907    0.1294
(100 Bs.)
Spanish-speaking   0.1839   0.1926    0.1147    0.2021   -0.1772    0.1926     0.4575b   0.1984
head (1=yes)
Aymara-speaking   -0.1032   0.2709    -0.3514   0.3129   -0.86280   0.3087    -0.0038    0.3024
head (1=yes)
Number of adults
in household       0.1047   0.0643   -0.0818    0.0720   -0.0062    0.0768     0.1895c   0.0578
(Aged 16 or
older)
Number of
children in       -0.1596'  0.0526   -0.0809    0.0588    0.0439    0.0542    -0.0311    0.0495
household (Aged
15 or younger)
Duration of illness  0.07630  0.0131  0.0412b   0.0160    0.06300   0.0142     0.0774'   0.0125
(days)
Social Security    0.0558   0.1764    2.9017c   0.2275   -0.9686    0.2029    -0.1079    0.1661
(1 =yes)
Log-likelihood       -2252.77
Number of
observations           2089
Source: Encuesta Integrada de Hogares (1990).
Notes:
a. Significant at 10 percent level (two-sided).
b. Significant at 5 percent level.
c. Significant at 1 percent level.
20


provider. Mother's education has a positive effect on the choice of Social Security facilities and
private facilities. Number of children in a household has a negative effect on the choice of
Ministry of Public Health facilities.
Using the median values of all independent variables as a benchmark, Tables 9b and 9c
provide the results of change on an individual, household, or provider characteristic. Each row
in Tables 9b and 9c adds up to 100 percent. The findings are as follows.
Age. Children are more self-cared for than are adults. Seventy four percent of children
are self-cared (Table 9b) while 48 percent of adults are self-cared (Table 9b). Older
children are more likely (40 percent for 15 year olds) to be treated at formal health
facilities than younger children (29 percent for one year olds).
Mother's education. Mother's education has a positive effect on seeking professional
care. Among professional care options, Social Security and private facilities are more
likely to be chosen as a mother's education level becomes higher.
Gender. As in the adult case, females are less likely (72.9 percent) to treat themselves
than are men (73.9 percent). Among formal care options, women show a preference for
Public Health (8.2 percent) and private facilities (9.3 percent).
Per capita consumption. Children from wealthy families are more likely to be treated
professionally. Among professional care options, the probability of seeking care at the
Ministry of Public Health and the probability of seeking care at private facilities are both
positively related to their consumption levels. The probability of seeking care at NGOs
or Social Security decreases as the consumption level increases.
Language. If a head of household speaks Aymara, his or her children are more likely
to receive self-care. Among health facilities these children are less likely to be treated
at Social Security facilities or NGOs. If a head of household speaks Quechua or other
languages, among health facilities his or her children are most likely to be treated at
NGOs.
Number of adults. With more adult members in the household, the probability of
receiving formal care increases. When there is only one adult in the household, the
probability of receiving formal care is 24.6 percent while the probability goes up to 27.8
percent when there are three adult members.
Number of children. The probability of receiving self-care increases with the number
of children aged 15 or younger in the household. Adults may be less likely to take their
children to formal health facilities if they have gained experience by treating other
children themselves.
Duration of illness. The duration of illness is positively related to the probability to seek
care.
21


Table 9b: Probability of Choosing a Provider (The Effects on Probability of Changes in Individual and
Household Attributes, Children Aged 15 Years or Younger)
Ministry of
Public Heakh     Social Security  NGO      Private    Sey-Care
Per.ena . Choang. .            1 .0.5            10.2        7.7        12.         59.
Predicted Percentage Choosing:       7.6               2.1        8.2         8.3         73.9
Age (Years)
1                                 10.6              2.3         8.1        8.3         71.0
10                                6.4               3.0         9.6        9.5         71.4
15                                9.4               5.9        11.4       12.1         60.2
Mother's education
0                                  7.9               1.7        8.2         6.5         75.7
2                                  7.8               1.8         8.2        7.1         75.1
4                                  7.7               1.9        8.2         7.7         74.5
10                                  7.4              2.4         8.1         9.8        72.4
Gender
Male                               7.6               2.1         8.2        8.3         73.9
Female                             8.2               1.8        7.8         9.3         72.9
Per capita consumption quintiles
(median)
Quintile 1: 57.87                  6.7               2.2         8.3        6.9         75.9
Quintile 2: 99.00                  7.3               2.1        8.2         7.9         74.5
Quintile 3: 146.38                 8.1               2.0        8.1         9.3         72.6
Quintile 4: 223.69                 9.5               1.8        7.8        11.9         69.0
Quintile 5: 404.88                13.1               1.3        6.8        20.5         58.3
Language of head of household
Spanish                            7.6              2.1         8.2         8.3        73.9
Aymara   .                         6.3              1.4         4.6        5.8         81.9
Quechua and other languages        6.5              1.9        10.0        5.4         76.1
Number of adults aged 16 or older
0                                  6.4               2.5         8.6        5.9         76.6
1                                  7.0               2.3        8.4         7.0        75.4
2                                  7.6               2.1        8.2         8.3         73.9
3                                  8.2               1.9        7.9         9.8         72.2
Number of children aged 15 or
younger
0                                 11.6              2.5         6.8        8.7         70.3
1                                 10.1              2.3         7.3        8.6         71.7
2                                 8.8               2.2         7.7        8.5         72.9
3                                 7.6               2.1         8.2        8.3         73.9
Duration of illness
1                                 8.0               2.1         8.5        8.8         72.5
3                                  9.0              2.2         9.3        9.9         69.6
5                                 10.0              2.3        10.1        11.1        66.6
10                                12.8               2.5       12.1        14.3        58.3
Social Security
Yes                                6.0              28.0       55.1         5.6        55.0
Source: Encuesta Integrada de Hogares (1990).
22


Table 9c. Probability of Choosing a Provider (The Effects on Probability of Changes in Health Facility
Attributes, Children Aged 15 Years or Younger)
Ministry of Public Health Social Security  NGOs  Private    Se- Care
Pred~4~vabbod,i e e gChsg.7.......
User fee - Public Health
0                                   7.5                2.1         8.2       8.3          73.9
5                                   7.6                2.1         8.2       8.3          73.9
10                                   7.7                2.1        8.2        8.3         73.8
20                                   7.9                2.1        8.1        8.3         73.6
User fee - Social Security
0                                    7.6                2.1        8.2        8.3         73.8
5                                    7.6                2.1        8.2        8.3         73.8
10                                   7.6                2.1         8.2        8.3         73.8
Travel time - Public Health
15                                   8.3                2.0         8.1        8.3         73.3
30                                    7.6               2.1         8.2        8.3         73.9
60                                    6.2               2.1         8.3        8.5         74.9
Travel time - Social Security
15                                   7.5                2.4         8.1        8.3         73.6
40                                    7.6               2.1         8.2        8.3         73.9
60                                    7.6                1.8        8.2        8.4         74.1
Baseline (Median Values)
Age                                                 4 years old
Mother's education                                  6 years of education
Father's education                                  8 years of education
Gender                                              Male
Household per capita consumption (monthly)          114.21 Bolivianos
Language of head of household                       Spanish
Number of adults aged 16 years or older             2 persons
Number of children aged 15 years or younger         3 persons
Duration of illness                                 0 days
Social Security                                     No
Public Heath       Soc. Sec.     NGO       Private
User fee                                5                 0           3          10
Waiting time                            30                30          30         15
Traveling time (round trip)             30                40          30         30
Source: Encuesta Integrada de Hogares (1990).
23


Table 9c shows how changes in user fees and waiting times at Public Health facilities and
Social Security facilities have an effect on the probability to seek professional care and the
choice of provider. Compared to the adult case, the effects on probabilities of changes in user
fees are small for both Public Health and Social Security facilities. For example, by reducing
the user fee from 5 Bolivianos to zero we increase the demand at Public Health facilities by only
0.1 percent (Table 9c). On the other hand, by the same reduction in user fee, we can increase
the demand by adult patients by one percent (Table 8c). At Social Security, the probability has
not been changed at all by increasing the user fees from zero to 10 Bolivianos. The probabilities
are a little more sensitive to travel time than to user fees. By reducing the travel time from 30
minutes to 15 minutes, the probability of choosing Public Health facilities increases 0.7 percent.
By reducing the travel time from 40 minutes to 15 minutes, the probability of choosing Social
Security increases by 0.3 percent.
Table 9d shows that children's demand for professional health care is less price elastic
than adult demand. Indeed, price elasticities are positive in the highest consumption quintile,
which may indicate that the richer individual would be willing to pay for better quality care
regardless of its cost. The travel time elasticities are large relative to the price elasticities. The
opportunity cost of time may be a large portion of the total price of medical care for children
in urban Bolivia.
Infant Outpatients Aged Between 0 and 4 Years Old
Maximum-likelihood estimation results of the multinomial logit model for infant
outpatients aged between zero and four years old are presented in Table 10a. Patients face five
health care choices: Ministry of Public Health facilities, Social Security facilities, NGOs,
private, and self-care. All the estimated coefficients are relative to self-care. The coefficients
of waiting time and travel time are negative. This implies that the probability that people will
seek care outside the home from any provider decreases with the waiting time at the provider
and travel time to the provider. The user fee coefficient is negative and the coefficient
associated with the fee-times-per capita consumption variable is positive. These signs suggest
that relative to lower-income people, higher income individuals are less sensitive to price
changes. Compared to the user fee coefficient (-0.0040), the magnitude of the coefficient
(0.0145), however, is so large that at per capita monthly household consumption of only 27.5
Bolivianos, the price effect becomes almost zero. The price elasticities by different income
levels are explored later in Table 10d.
For individual characteristics, gender and presence of older females in the household are
not significant for any providers. Age, education, type of illness,7 and female-headed household
are statistically significant for some providers. For example, the age coefficients are negative
for all the providers. Infants less than 1 year old are more likely to be treated at health
facilities. Female-headed households have a negative effect on the choice of formal care.
Children from female-headed households are less likely to seek formal care.
7. The omitted disease is diarrhea.
24


Table 9d. Elasticities of Probability with Respect to User Fee (Children Aged 15 Years of Younger)
Ministry of
Own- and Cross-Price Elasticities  Public Health  Social Security  NGOs PPvvate            Sey-Care
Price of Ministry of Public Health  0.013           0.000        -0.001         -0.001       -0.001
Price of Social Security            0.000            0.001        0.000         -0.004        0
Price of NGOs                       -0.001           0.000        0.010         -0.001       -0.001
Price of Private                   -0.003           0.000        -0.002         0.029         0.003
Own-Price Elasticities            Quintile 1     Quintile 2     Quintile 3    Quintile 4   Quintile 5
Price of Ministry of Public Health  -0.040          -0.001        0.043         0.113         0.271
Price of Social Security           -0.004           0.000         0.005         0.012         0.031
Price of NGOs                      -0.031           -0.001        0.034         0.092         0.234
Price of Private                   -0.079           -0.003        0.009         0.219         0.494
Own-Price Elasticities             Spanish         Aymara        Quechua       Male         Female
Price of Ministry of Public Health  0.013           0.013         0.013         0.013         0.013
Price of Social Security            0.100            0.010        0.002         0.010         0.001
Price of NGOs                       0.010            0.080        0.007         0.010         0.008
Price of Private                    0.029           0.029         0.029         0.029         0.029
Elasticities of Probability with Respect to Travel Tne
Ministry of
Own- and Cross-Tune Elasticities  Public Heakh  Social Security  NGOs         Private      Se-Care
Time of Ministry of Public Health  -0.190            0.015        0.018         0.014         0.015
Time of Social Security             0.011           -0.272        0.010         0.005         0.005
Time of NGOs                        0.020           0.015        -0.188         0.014         0.017
Time of Private                     0.020            0.015        0.018         -0.191        0.017
Source: Encuesta Integrada de Hogares (1990).
The effects on demand of changes in individual and household characteristics are explored
in Table 10b. The effects on probability of changes in provider characteristics are analyzed in
Table 10c. Each row in Tables 10b and 10c adds up to 100 percent.
Age. About 50 percent of infants before age 1 were treated professionally. About 70
percent of 1 year old children received no formal treatment. This result is similar to the
result given in Table 9b for 1 year olds.
Mother's education. Mother's education has a positive effect on seeking professional
care. When the mother has no education, 82 percent of children are not treated
professionally. When the mother has 10 years of education, 74 percent of children are
treated professionally.
Gender. Unlike the situation for adults or children, infant males are more likely to be
treated professionally (23 percent) than infant females (14.3 percent). This might be
some evidence of gender bias.
25


Table 10a: Choice of Provider Multinomial Logit Coefficients and Standard Errors (Children Aged 4 or
Younger)
Ministry of
Public Health      Social Security        NGOs               Prvate
PeeatgChoig:1,$836.7                                                   63.7_______
Standard           Standard            Standard            Standard
Variables              Coefficient  Errors  Coefficient  Errors  Coefficient  Errors  Coefficient  Errors
Constant                -1.1597c  0.3437    -3.9580c  0.5022   -1.6480'  0.4442    -2.97081  0.4807
User fee                -0.0040   0.0257    -0.0040  0.0257    -0.0040   0.0257    -0.0040   0.0257
User fee times per capita  0.0145*  0.0115   0.0145c  0.0115    0.0145'  0.0115     0.0145'  0.0115
cons. (100 Bs.)
Waiting time             -0.0027  0.0030    -0.0027   0.0030   -0.0027   0.0030    -0.0027   0.0030
Travel time              -0.0041  0.0033    -0.0041   0.0033   -0.0041   0.0033    -0.0041   0.0033
Age                     -0.8541'  0.2484    -0.5159*  0.3024   -0.8345c  0.3098    -0.7595c  0.2597
Age squared              0.1327b  0.0625     0.0965   0.0727    0.15490  0.0758     0.1117   0.0643
Mother's education       0.0394   0.0261     0.08950  0.0292    0.0022   0.0346     0.0633b  0.0262
(years)
Father's education (years)  -0.0678b  0.0264  -0.0399  0.0303  -0.0206   0.0352     0.0142   0.0260
Gender (1=male)          0.0463   0.1996     0.3434  0.2430     0.1431   0.2449     0.1804   0.2105
Female head              -0.7785'  0.3888   -1.4067'  0.6102   -0.2889   0.4721    -0.4697   0.4306
Presence of old female  -0.3222   0.3446    -0.4334   0.3995   -0.5313   0.4662     0.1314   0.3247
Per capita cons. (100 Bs.)  0.2893  0.1084  0.0149   0.1259   -0.344P   0.1842     0.3366b  0.1657
Cold (1 =yes)            0.1592   0.2155     0.5037   0.2856    0.635(0  0.3008     0.4259P  0.2378
Measles (1=yes)          -0.4634  0.6429    -0.2481   0.6976    1.4459  0.5080     0.3301   0.5244
Other diseases (1=yes)   0.3924   0.4666     1.33210  0.4775    2.07801  0.4281     0.6796   0.4737
Duration of illness (days)  0. 1265'  0.0214  0.0689'  0.0350   0.1038  0.0247     0.13530  0.0218
Social Security (1= yes)  -0.1340  0.2424    2.95871  0.3432   -0.1497   0.2995     0.0223   0.2375
Log-likelihood              -1155.78
Number of observations        1187
Source: Encuesta Integrada de Hogares (1990).
Notes:
a. Significant at 10 percent level (two-side).
b. Significant at 5 percent level.
c. Significant at 1 percent level.
Female-headed households. When the family is headed by a female, the percentage of
self-treatment goes up to 85 percent from 77 percent. Because these females support the
family through income-generating activities, female heads of household may face tighter
time constraints on non-market activities such as seeking medical care.
Presence of an Older Female. When a woman who is older than 45 is present in the
household, the percentage of self treatment goes up to 81.2 percent from 77.3 percent.
This also decreases the probability of choosing formal care, except in the case of private
facilities where the presence of an older female increases the probability of choosing
these facilities.
26


Table 10b: Probability of Choosing a Provider (The Effects on Probability of Changes in Individual and
Household Attributes, Children Aged 4 or Younger)
Ministry of
Public Health Social Security  NGOs     Private     SeJ-Care
Percentage Choosing:                 10.8          8.3         6.7          10O6        63.7
P..i.cd Percentage Chooing:           8.3          2.0        .5 ..977.
Age (Years)
0                                  18.7         2.6        13.0          12.0         53.7
1                                 11.9          1.2         8.6           8.2         70.1
3                                  7.1          1.9         6.4           5.0         79.6
Mother's education
0                                   6.7          1.1         6.0           4.1         82.2
2                                   7.1          1.3         6.1           4.5         80.9
4                                   7.6          1.5         6.3           5.0         79.5
10                                  9.0           2.5         6.8           6.9        74.8
Gender
Male                                8.3          2.0         6.5           5.9         77.3
Female                              8.1          1.4         5.8           5.1         79.5
Head of household
Male                               8.3          2.0          6.5          5.9         77.3
Female                             4.2          0.5         5.4           4.1         85.7
Presence of old female
Yes                                 6.3          1.3         4.0           7.1         81.2
No                                  8.3          2.0         6.5           5.9         77.3
Per capita consumption quintiles (median)
Quintile 1: 57.87                   7.0          2.0         7.7           4.8         78.5
Quintile 2: 99.00                   8.1          2.0         6.7           5.7         77.5
Quintile 3: 146.38                  9.9          1.9         5.7           7.0        75.5
Quintile 4: 223.69                 11.8          1.9         4.3           9.7        72.3
Quintile 5: 404.88                 18.5         15.6         2.0          19.0        58.9
Illness
Diarrhea                           7.6          1.3         3.7           4.2         83.2
Cold                               8.3          2.0         6.5           5.9         77.3
Measles                            4.3          0.9        14.3           5.2         75.2
Other disease                       8.2          3.5        21.7           6.0        60.6
Social Security
Yes                                5.4         28.1         4.2           4.5         57.7
Source: Encuesta Integrada de Hogares (1990).
Per Capita Consumption. Children from wealthy families obtain more formal care.
However, while only about 30 percent of the adults from the highest income group care
for themselves, nearly 60 percent of children from the same income group do so.
Illness. For diarrheal illnesses, 83 percent of children aged 4 and under are treated at
home.
27


Table 10c: Probability of Choosing a Provider (The Effects on Probability of Changes in Health Facility
Attributes, Children Aged 4 or Younger)
Ministry of     Social
Public Heath   Security     NGOs        Pivate      Sey-Care
User fee - Public Health
0                                     7.9           2.0         6.6         6.0          77.6
5                                     8.3           2.0         6.5         5.9          77.3
10                                    8.7           1.9          6.5        5.9          76.9
User fee - Social Security
0                                     8.3           2.0          6.5        5.9          77.3
5                                     8.3           2.1          6.5        5.9          77.2
10                                     8.3           2.2         6.5         5.9          77.1
Travel time - Public Health
15                                     8.8           1.9         6.5         5.9          76.9
30                                     8.3           2.0         6.5         5.9          77.3
60                                     7.4           2.0         6.6         6.0          78.0
Travel time - Social Security
15                                     8.3           2.2         6.5         5.9          77.1
40                                     8.3           2.0         6.5         5.9          77.3
60                                     8.3           1.8         6.6         5.9          77.4
Waiting time - Public Health
15                                     8.6           1.9         6.5         5.9          77.0
30                                     8.3           2.0         6.5         5.9          77.3
60                                     7.7           2.0         6.6         6.0          77.8
Waiting time - Social Security
15                                     8.3           2.0         6.5         5.9          77.2
30                                     8.3           2.0         6.5         5.9          77.3
60                                     8.3           1.8         6.6         5.9          77.4
Basefine (Median Values)
Age                                               2 years old
Mother's education                                7 years of education
Father's education                                8 years of education
Gender                                            Male
Head of household                                 Male
Presence of old female                            No
Household per capita consumption (monthly)        107.14 Bolivianos
Illness                                           Cold
Duration of illness                               0 days
Social Security                                   No
Ministry of     Social
Public Health   Security     NGO         Private
User fee                                 5             0           3          10
Waiting time                            30            30          30          15
Traveling time (round trip)             30            40          30          30
Source: Encuesta Integrada de Hogares (1990).
28


Table 10c shows that an increase in fees at Public Health or Social Security facilities does
not discourage patients from seeking formal care, but an increase in waiting time or travel time
does discourage some patients, and keeps other patients out of the formal health care system
completely. For example, by increasing travel time from 30 minutes to 60 minutes, the
probability of choosing Public Health facilities is reduced by 0.9 percent and the probability of
choosing self-care increases by 0.7 percent. By increasing waiting time from 30 minutes to 60
minutes, the probability of choosing Public Health facilities is reduced by 0.6 percent and the
probability of choosing self-care is increased by 0.5 percent.
As discussed in the previous section, children have lower user fee elasticities than adults.
Table 10d shows that infants between 0 and 4 years old have even lower price elasticities than
children aged 15 and under. User fees play only a minor role in determining health care demand
for infants. On the other hand, time seems to play more of a role than user fees although both
waiting time and travel time elasticities are low. Waiting time elasticities range between -0.018
and -0.077 and travel time elasticities ranges between -0.112 and -0.164.
Table 10d: Elasticities of Probability with Respect to User Fee (Children Aged 4 or Younger)
Ministy of    Social
Own- and Cross-Price Elasticities  Public Health  Security  NGOs         Private     Sely-Care
Price of Ministry of Public Health   0.053      -0.005      -0.005        -0.005      -0.005
Price of Social Security             0.000       0.006       0.000         0.000       0.000
Price of NGOs                       -0.002      -0.002       0.033        -0.003      -0.002
Price of Private                   -0.007       -0.005      -0.006         0.108      -0.007
Own-Price Elasticities            Quintile 1  Quintile 2  Quintile 3    Quintile 4   Quintile 5
Price of Ministry of Public Health   0.020       0.047       0.078         0.127       0.226
Price of Social Security             0.002       0.005       0.009         0.014       0.028
Price of NGOs                       0.012        0.029       0.049         0.083       0.017
Price of Private                    0.042        0.098       0.161         0.259       0.448
Elasticitisu o Probability with Re. to W.itin Time
Ministy of    Social
Own- and Cross The Elasticities  Public Health  Security    NGOs         Private     Sey-Care
Time of Ministry of Public Health  -0.072        0.000       0.092         0.005       0.007
Time of Social Security              0.000      -0.077       0.005         0.000       0.002
Time of NGOs                        0.004        0.000      -0.018         0.005       0.005
Time of Private                      0.002       0.000       0.002        -0.038       0.002
Elasticities ofProbability jAth Relij tt Travel Tm
Minisny of    Social
Own- and Cross Tone Elasticities  Public Health  Security   NGOs         Private     Self-Care
Time of Ministry of Public Health  -0.112        0.015       0.014         0.010       0.010
Time of Social Security              0.005      -0.164       0.006         0.000       0.003
Time of NGOs                        0.007        0.015      -0.115         0.005       0.008
Time of Private                     0.007        0.015       0.009        -0.116       0.007
Source: Encuesta Integrada de Hogares (1990).
29


Obstetrical Care for Women between 15 and 49 Years Old
Table I la presents maximum-likelihood estimation results of the multinomial logit model
for women aged between 15 and 49 years old. Patients face four health care choices: hospital,
private, health center or health post, or home. All the estimated coefficients are relative to
home. The coefficient for user fee is positive and the coefficient associated with the fees-times-
per capita consumption variables is negative. These signs suggest that relative to high-income
people, lower-income individuals are less sensitive to price changes.
For individual characteristics, age, education, per capita consumption, language, work
status, and prenatal visits are statistically significant for some providers relative to self-care.
But marital status and household structure (number of adults and children in the household) are
not significant determinants of provider choice.
Table 1 lb provides the results of changes in individual and household characteristics
using median values as a benchmark. The effects on the probability of changes in provider
characteristics are analyzed in Table 11c. Each row in Tables 1 lb and 1 lc adds up to 100
percent.
Table 11a: Choice of Provider Multinomial Logit Coefficients and Standard Errors (Women Between 15
and 49 Years Old)
Health Center
Hospital            Private           or Health Post
Percentaga ChDosing:                       48.                 13.4                   5
Standard           Standard              Standard
Variables                            Coefficient  Errors  Coefficient  Errors  Coefficient  Errors
Constant                              -3.2556     1.9190  -9.8862c   2.7720    -9.5193*    3.0450
User fee                               0.0003     0.0020   0.0003    0.0020     0.0003      0.0020
User fee times per capita cons. (100 Bs.)  -0.0008  0.0009  -0.0008  0.0009    -0.0008     0.0009
Age                                    0.0934     0.1221   0.320Y    0.1852     0.4392b     0.2043
Age squared                           -0.0010     0.0020  -0.0048    0.0032    -0.0070b     0.0035
Single (1 =yes)                        0.3794     0.3391   0.5950    0.4750     0.3186      0.4679
Education (years)                      0.0960w    0.0291   0.138(Y   0.0386     0.1383'     0.0394
Per capita cons. (100Bs.)              0.79360    0.1921   1.2467    0.2411     0.588"      0.2250
Informal working (1=yes)              -0.4359     0.4891   0.7727    0.6702    -0.2559     0.6123
Not working (1=yes)                    0.2412     0.4623   1.2887"   0.6336     0.0508     0.5636
Speaking Spanish (1= yes)               0.2618    0.2524  -0.0350    0.3682    -0.2049      0.3563
Speaking Aymara (1=yes)                -1.8853c   0.3839  -0.0996    0.4593    -0.7592     0.4679
Number of adults in household         -0.1079     0.0944  -0.0499    0.1270     0.1628     0.1124
Number of children in household       -0.0220     0.0716  -0.0651    0.1060    -0.7478     0.1021
Number of prenatal visits              0.3169*    0.0474   0.3184    0.0579     0.2203'     0.0617
Log-likelihood                            -741.08
Number of observations                       776
Source Encuesta Integrada de Hogares (1990).
Notes
a. Significant at 10 percent level (two-sided).
b. Significant at 5 percent level.
c. Significant at 1 percent level.
30


Table 1 1b: Probability of Choosing a Provider (7he Effects on Probability of Changes in Individual and
Household Attributes, Women Between 14 and 49 Years Old)
Hospital       Private    Health Center      Home
Percentage Choosing:                       48.6          13.4            9.5           28.5
Predicted Percentage Choosing:             64.6          10.3           10.1            15.0
Age (Years)
15                                       64.9           4.4           3.2            27.4
20                                       65.4           7.3           6.5            20.8
40                                       73.2           8.6           5.7            12.4
Marital status
Single                                    66.5          13.1           9.8            10.6
Not single                                64.6          10.3          10.1            15.0
Education
0                                         57.9           6.6           6.5            29.0
2                                         60.3           7.5           7.3            24.9
4                                         62.2           8.4           8.2            21.2
10                                        65.1          11.3           11.1           12.5
Per capita consumption quintiles (median)
Quintile 1: 57.87                        60.2            7.9          10.5            21.4
Quintile 2: 99.00                         63.6          9.6           10.2            16.5
Quintile 3: 146.38                        66.3          11.8           9.8            12.1
Quintile 4: 223.69                        70.1         15.0            8.3            6.6
Quintile 5: 404.88                        64.6          27.8           5.9             1.7
Work Status
Informal                                 53.4          10.0          12.1            24.4
Not working                              64.6          10.3          10.1            15.0
Work formal                              64.9           3.6          12.3            19.2
Language
Spanish                                  64.6          10.3          10.1            15.0
Aymara                                   19.9          25.4          15.3            39.5
Quechua and other languages              56.6          12.1          14.2            17.1
Number of children aged 15 or younger
0                                        60.0          12.4          12.6            14.9
1                                        61.6          11.7          11.7            15.0
2                                        63.1          11.0          10.9            15.0
3                                        64.6          10.3          10.1            15.0
Number of prenatal visits
0                                        50.8           8.0          10.6            30.5
5                                        70.8          11.3           9.1             8.7
10                                       79.0          12.7           6.3             2.0
Source: Encuesta Integrada de Hogares (1990).
Age. Younger women are more likely to deliver babies at home. As women get older,
they are more likely to deliver babies at a hospital. The probability of delivering a baby
at private clinics or health centers increases with age until around age 40 when the
probability starts decreasing. Women over 40 years old who have work experience in
31


Table llc: Probability of Choosing a Provider (The Effects on Probability of Changes in Health Facility
Attributes, Women Aged Between 15 Years and 49 Years)
Hospital       Private     Health Center     Home
User fee - Hospital
0                                         65.2           10.1            9.9          14.7
25                                         64.9           10.2          10.0           14.9
50                                         48.6           13.4            9.5          28.5
User fee - Health Center
0                                         64.5           10.3           10.3          15.0
25                                         48.6           13.4           9.5           28.5
50                                         64.7           10.3          10.0           15.0
Baseline (Median Values)
Age                                                     27 years old
Marital status                                          Not Single
Education                                               8 years of education
Household per capita consumption (monthly)              113.95 Bolivianos
Work status                                             Not working
Language                                                Spanish
Number of adults aged 16 years or older                 2 persons
Number of children aged 15 years or younger             3 persons
Number of prenatal visits                               3 times
Public Health    Private     Health Center     Home
User fee (Bolivianos)                          50            190            25                  0
Source: Encuesta Integrada de Hogares (1990).
the formal or government sector or who have some family member who has work
experience in the formal sector or government sector are eligible for medical care at a
government hospital.
Marital Status. Single women are more likely to seek obstetrical care outside the home
(90 percent).
Education. Educated women are more likely to deliver babies at formal health facilities.
Per Capita Consumption. Women from wealthy families are more likely to deliver
babies at formal health facilities and, among formal health facilities, private clinics are
preferred.
Work Status. Women who work in the formal sector are least likely to deliver babies
at private facilities. These women may be able to go to Social Security hospitals where
costs are much lower.
32


Table 11d: Elasticities of Probability with Respect to User Fees (Women Between 15 and 49 Years Old)
Own- and Cross-Price Elasticities  Hospital  Private    Health Center     Hone
Price of Hospital              -0.010        0.019          0.020          0.020
Price of Private                0.012       -0.092          0.019          0.127
Price of Health Center          0.007        0.000         -0.012          0.002
Own-Price Elasticities       Quintile I    Quintile 2     Quintile 3    Quintile 4   Quintile 5
Price of Hospital              -0.002       -0.009         -0.014         -0.023       -0.053
Price of Private               -0.024       -0.079         -0.145         -0.228        -0.396
Price of Health Center         -0.002       -0.010         -0.020         -0.034       -0.068
Own-Price Elasticities        Spanish       Aymara        Quechua
Price of Hospital              -0.010       -0.002         -0.001
Price of Private               -0.092       -0.082         -0.094
Price of Health Center         -0.012       -0.011         -0.012
Source: Encuesta Integrada de Hogares (1990).
Language. Aymara-speaking women are most likely to deliver babies at home. In
Bolivia it is common, especially among Aymara Indians, for males to help females
deliver babies. Once they seek formal care, Aymara women are more likely to deliver
babies at private facilities but least likely to deliver at hospitals or health centers.
Number of children. The number of children has little effect on the choice of
obstetrical care.
Number of prenatal visits. The probability of delivering babies at home is about 30
percent for women who did not seek prenatal care at all. This probability decreases as
the amount of prenatal care increases. For women who made 10 prenatal care visits, the
probability of delivering babies at home is less than two percent.
Price elasticities are very low for obstetrical care as well as for medical care. The price
elasticities of demand fall with income, but stay almost the same among different ethnic groups.
Prenatal Care for Women between 15 and 49 Years Old
Compared with outpatient medical care discussed in sections A through D of this chapter,
prenatal care can be considered a homogeneous good. In this section, I set up the demand
specification for prenatal care using the number of prenatal care visits as the quantity
demanded.'
8. The demand specification for prenatal care using the timing of the first prenatal visit might be another approach.
In this case, using a multinomial logit model or ordered probit model, or using the indicator of the first prenatal
visit during the first, second, or third trimester would be a dependant variable.
33


I use a Tobit model since, as shown in Table 12, many pregnant women receive no
prenatal care, while the other observations spread over positive values. The Tobit specification
is:
(10)   y," = xilp + u,i=1,2,...,n
Yd J8  f Y>0
= 0 if y: < 0.
where u, is assumed to be i.i.d. drawings from N(0,ol). It is assumed that yj and xi are observed
for i=1,2,...,n., but yi' is unobserved if yi'<50. In the demand equation for prenatal care, yi*
is the number of prenatal care visits and x, is for age, education, per capita consumption, user
fee for prenatal care, work status, number of adults, number of children, and language.
For the 32 percent of women who did not seek prenatal care, I estimated the user fees
for prenatal care for use in the Tobit model. I used a linear regression for the women who
sought care, reported in Table 12, to assign prices for the women who did not seek care. I used
per capita consumption, language, and region to estimate user fees since, as discussed in Chapter
3, these variables are relatively reliable indicators of user fees. Average estimated cost for the
women who did not seek parental care is 3.02 Bolivianos while the average actual cost was 4.20
Bolivianos.
Table 12: Linear Regression for Women Who Sought Prenatal Care
Dependent Variable = Consultation Cost per Visit (Total Sample =530)
Variable                                  Coefficient          Standard Error
Constant                                    5.8030b              2.7000
Per capita consumption                      0.0140'              0.0017
Speaking Spanish (1= yes)                  -4.60518              2.6182
Speaking Quechua (1= yes)                  -5.4971a              2.7207
Speaking Aymara (1= yes)                   -4.1815               2.8310
Living in South (1= yes)                    0.5955               0.9039
Living in Central (1=yes)                   0.5885               0.9039
R2: 0.152
Source: Encuesta Integrada de Hogares (1990).
Notes:
a. Significant at 10 percent level (two-sided).
b. Significant at 5 percent level.
c. Significant at 1 percent level.
34


Estimated results for the Tobit model are presented in Table 13. Older and more
educated women seek more prenatal care. Women with more children are less likely to seek
prenatal care. The coefficient on cost is negative and significant. These results hold in Poisson
or Negative Binomial regressions. Since the result from the Tobit model is similar to the result
from the Poisson or the negative binomial model, the result from the Tobit model is used for
further analysis.
In order to obtain user fee elasticities, we also need the marginal effects on the number
of prenatal care visits as affected by the change in the price. In the Tobit model there are three
marginal effects (Maddala 1983, p.160): one for the latent variable, yi*, and two for the
observed variable yi (with and without the condition that the observed variable is greater than
zero).
Denote z= fl'x/a, and let 03 be the jth component of 0. Drop the subscript i, which
refers to the ith observation. The following are the three marginal effects:
aE(y *
(11)         ),
(12) y E(Y)  _ 0(z)p (
(13) M6  -*O)_ ,1    z "(Z  (Z)2
(1)              4D(z) zD(z)
where 4 is the standard normal density function and 4' is the standard normal distribution
function. I report three marginal effects in Table 13: E(y) is the mean of potential y's, E(y)
is the mean of all observed y's, positive and zero, and E(y I y > 0) is the mean of the positive
y's. All the marginal effects are evaluated at the median values. I use (12) to calculate price
elasticities, since I am interested in predicting how y, the number of prenatal care visits, changes
with the price of care for women who have or have not had prenatal care visits before.
Price elasticities are presented in Table 14. Price elasticities for prenatal care are very
low. This implies that women are not price sensitive when seeking prenatal care. Quechua or
Spanish-speaking females have slightly higher price elasticity than Aymara-speaking females
though the difference is very small (less than 1 percent).
35


Table 13: Tobit Estimation Results
Dependent Variable = Number of Prenatal Care Visits (Total Sample = 781)
Coefficient
or Marginal Effect                  Marginal Effect  Marginal Effect
(11)         Standard Error       (12)            (13)
Constant                                  -4.5373'           2.4500          -3.1053        -2.1815
Age                                        0.3988"           0.1734           0.2729          0.1917
Age squared                                -0.0067b          0.0030          -0.0048         -0.0034
Education (years)                          0.2520'           0.0340           0.1724          0.1211
Per capita consumption                     0.0045'           0.0010           0.0031          0.0022
Cost of prenatal care                     -0.0442'           0.0238          -0.0303        -0.0213
Work status (1= informal)                 -0.1567            0.3316          -0.1072         -0.0753
Number of children                        -0.1809,           0.0996          -0.1238         -0.0870
Number of adults                           0.0116            0.1175          -0.0079         -0.0056
Speaking Spanish                           -0.2286           0.3513          -0.1564         -0.1099
Speaking Aymara                           -1.7930*           0.5021          -1.2272         -0.8621
a                                      3.6686'           0.1210
Log-likelihood: -1646.761
Source: Encuesta Integrada de Hogares (1990).
Notes:
a. Significant at 10 percent level (two-sided).
b. Significant at 5 percent level.
c. Significant at 1 percent level.
Table 14: Price Elasticities for Prenatal Care Visit
Quintile 1    Quintile 2    Quintile 3    Quintile 4   Quintile 5      Total
Spanish                 -0.057        -0.023        -0.031        -0.027       -0.039       -0.025
Aymara                  -0.043        -0.017        -0.024        -0.021       -0.032       -0.019
Quechua                 -0.059        -0.023        -0.032        -0.027       -0.040       -0.026
Source: Encuesta Integrada de Hogares (1990).
Note: Quintile of per capita household consumption. Quintile 1 is the lowest.
36


Chapter VI. Conclusion
Models of choice of medical care providers were estimated using data from urban areas
in Bolivia. The alternative providers included government facilities (Ministry of Health and
Social Security), non-governmental organizations, private clinics, and self-care.  This
specification reflects the actual choices available to the population. The models were estimated
separately for adults, children, and infants.
Consulting fees have negative effects on the demand for medical care but demand
elasticities are very low. Waiting times have negative effects on the demand for adults and
infants, and travel times have negative effects for children and infants. Since someone in the
family usually has to accompany children or infants to a health facility, travel time has a
negative effect on their medical care demand. Adults are more price sensitive in their demand
for care than are parents for the care of their children or infants.
Children are more self-cared for than are adults. Seventy four percent of children are
self-cared while 48 percent of adults are self-cared.' Among adult patients, the elderlylo are
much more likely to seek care in the presence of a health problem while among infants those
who are younger than one year old are more likely to seek care. Small children and older
individuals have weaker immune systems and are more prone to complications or death from
illness.
The level of formal education of the person has a positive effect on the probability to
seek outpatient medical care." For children or infants, mother's education has a positive effect
on the demand for medical care although father's education has no, or even a negative, effect
on demand. Among professional care, private facilities are more likely to be chosen as the
formal education level becomes higher.
Among adult and child outpatients, females appear to be more likely to seek care than
males. However, in the case of infants, males are more likely to be treated professionally than
are females. This might be some evidence of gender discrimination.
Income (measured by per capita consumption) has a substantial effect on the demand for
medical care. Patients from wealthy families seek more formal care, and prefer private
facilities.
Work status has some effect on the choice of medical facility. People working in the
informal sector such as selling small amounts of food and daily products on the street or working
9. All the probabilities are evaluated at the median values.
10. In 1991, life expectancy at birth in Bolivia was 59 years old.
11. The level of formal education of the women has a positive effect on the probability to seek obstetrical and
prenatal care (li 1993).
37


in small family run workshops are more likely to choose self-care and less likely to go to Social
Security facilities. Women who work in the informal sector are more likely to deliver a baby
at home than women who do not work or who work in the formal sector. Women who work
in the formal sector are least likely to deliver babies at private facilities. Many of them may go
to Social Security facilities, which provide care cheaper than do private facilities.
Compared to Spanish speakers or Quechua speakers, Aymara speakers are more likely
to choose to be self-cared for and less likely to go to Social Security facilities, NGOs, or private
facilities. For obstetrical care, Aymara-speaking women are most likely to deliver babies at
home. For those Aymara women who decide to deliver babies outside the home, they are more
likely to deliver babies at private facilities.
With more adults in the household, people seek less professional care. They may receive
relatively more support from family members than those who live with fewer adults. Children's
demand for medical care increases with more adults in the household. Since there are more
adults in the household, they are able to take their children to health facilities more often.
The probability of receiving formal care for children decreases with the number of
children in the household. Adults may be less likely to take their children to formal health
facilities if they have gained experience by treating other children themselves.
The number of children does not have a substantial effect on the demand for formal
obstetrical care. Women with more children do not necessarily switch to informal care such as
delivering a baby at home. However, the number of prenatal visits has a substantial effect on
the choice of formal care. The probability of delivering babies at home is about 30 percent for
the women who do not seek prenatal care at all while for women who seek 10 prenatal care
visits, the probability is less than two percent."
Caution must be drawn in interpreting low price elasticities. Coefficient estimates have
a downward bias when variables suffer from measurement error (Judge et al. 1988, p.582).
Thus, measurement error in user fee, travel time, and waiting time variables may have
contributed to lower elasticity estimates.
My primary purpose in estimating models of medical care provider choice is to evaluate
the effect of charging user fees for government medical care.  In evaluating the effect of
charging user fees, I need to know if the effect of user fees on utilization is uniform across
income groups or ethnic groups. If poorer individuals' decisions to use medical care are more
price elastic than richer individuals', user fees will be regressive in that they will reduce poorer
individuals' utilization by more than that of richer individuals.
Any evaluation of the proposal to institute user fees requires knowledge of the demand
function from which price elasticities can be calculated. Price elasticities provide information
12. All the probabilities are evaluated at the median values.
38


about how user- fees will affect utilization and revenues. Overall, my estimates show that the
price elasticities are low. However, demand for outpatient medical care is more elastic for
individuals in the lower income groups than for those in the upper income groups and quite
inelastic for individuals in the higher income groups.
User fees have a great potential for cost recovery, but care must be taken in
implementing them. Uniform user fees can generate substantial revenues but are very likely to
reduce the utilization of medical care by the poor.
39



Annex Table A-l: Descriptive Statistics (Adults Aged 16 Years or Older)
Facility Used
Ministry of Public
Heakh          Social Security      NGOs              Private            Sey-Care
Numberof people                   2743               374               448               222               548                1151
Percentage                         100               13.6              16.3              8.1               20.0               41.9
Variables                    Mean     S.D.     Mean      S.D.     Mean     S.D.    Mean      S.D.    Mean      S.D.      Mean     S.D.
User fee (Bolivianos)         11.80   58.33     20.44    72.12     3.33    47.70    21.35   112.95   31.85     73.29      0.90    12.37
Transportation cost (Bolivianos)  3.9  16.56     6.63    21.32     7.12    22.27     8.53    24.88     5.74    18.77      0        0
Cost of medicine (Bolivianos)  38.86  140.64    70.44   229.16    40.34    195.59   39.36    78.49    77.47   152.69      9.54    33.01
Waiting time (minutes)        32.03   72.37     69.21     84.16   71.25    104.14   57.05   101.95    31.73    61.19      0        0
Travel time (minutes) (round trip)  40.48  151.40  63.08  179.98  63.24    166.80   80.77   204.44    75.14   217.04      0        0
Age (years)                   42.44   17.54     41.67     16.32   47.83     17.93   40.41    16.66    45.37    18.92     39.59    16.56
Gender (1 =male)              0.40     0.49      0.42     0.49     0.42     0.49     0.39     0.49    0.35      0.48      0.41     0.49
Education (years)             7.70     5.55      7.51     5.11     9.18     6.12     7.64     5.20     8.47     5.76      6.83     5.22
Person informally working      0.35    0.48      0.36      0.48    0.15     0.35     0.32     0.47     0.32     0.47      0.44     0.50
(1=yes)
Duration of illness (days)     6.65    9.32      9.75     10.57    7.11     9.55     6.83     9.21     7.92     9.84      4.82     8.11
Number of adults in household  3.08    1.55      3.17      1.59    3.14     1.65     2.82     1.36     3.15     1.55      3.04     1.54
(aged 16 or older)
Number of children in household  1.97  1.75      2.14      1.88    1.73     1.76     2.09     1.70     1.79     1.68      2.07     1.73
(aged 15 or younger)
Per capita monthly consumption  217.55  209.04  220.57   228.51  232.20   239.36   198.67   181.49   315.85   255.96    167.70   142.30
(Bs.)
Speaking Spanish (1 =yes)     0.59     0.49      0.62     0.49     0.59     0.49     0.55     0.50     0.67     0.47      0.54     0.50
Speaking Aymara (1 =yes)       0.13    0.34      0.16     0.36     0.08     0.27     0.09     0.29     0.07     0.26      0.19     0.39
Speaking Quechua (1=yes)      0.25    0.43      0.21     0.41     0.29     0.45     0.32     0.47     0.22     0.42      0.25     0.43
Living Central (1 =yes)        0.45    0.50      0.48     0.50     0.36     0.48     0.36     0.48    0.41      0.49      0.51     0.50
Living North (1=yes)          0.16     0.37      0.22     0.42     0.15     0.36     0.11     0.32     0.19     0.39      0.14     0.35
Living South (1 =yes)          0.39    0.49      0.30      0.46    0.49     0.50     0.52     0.50     0.40     0.49      0.35     0.48
Possession of Social Security  0.34    0.47      0.26     0.44     0.71     0.45     0.30     0.46     0.34     0.47      0.23     0.42
(1=yes)


Annex Table A-2: Descriptive Statistics (Children Aged 15 Years or Younger)
Facility Used
Public Heath     Social Security      NGOs               Pivate            Se-Care
Number of people                   2089              212               213               160               260                1244
Percentage                         100               10.2             10.2               7.7               12.5               59.6
Variables                     Mean     S.D.    Mean      S.D.    Mean     S.D.     Mean     S.D.     Mean      S.D.     Mean      S.D.
User fee (Bolivianos)          4.02    17.55     8.75    25.51     1.08     6.16    4.66      8.99     20.22    38.22    0.25       3.38
Transportation cost (Bolivianos)  1.99  12.11    4.79     17.94    7.96    26.61    3.48     12.23     3.43     13.79    0          0
Cost of medicine (Bolivianos)  10.08   31.13    25.80    52.50     8.32    30.18   14.82     41.35     30.38    51.07    2.86       7.66
Waiting time (minutes)        22.35    68.54    63.56    89.00    53.65    55.15   79.34    137.99     34.97   102.36    0          0
Travel time (minutes) (round trip)  24.91  109.42  46.83  86.86   68.43   186.00   68.73    176.12     63.58   188.13    0          0
Age (years)                    5.02     4.29     4.60     4.44     6.07     4.70    5.43      4.44      5.55     4.51    4.75       4.08
Gender (1 =male)               0.52     0.50     0.50     0.50     0.54     0.50    0.53      0.50      0.48     0.50    0.53       0.50
Mother's education (years)     7.03     5.40     6.78     5.50     9.33     6.02    6.46      4.88      9.19     5.81    6.30       5.01
Father's education (years)      7.87    5.81     7.18     5.66     9.60     6.34    7.34      5.78      9.82     6.53    7.34       5.43
Duration of illness (days)     2.29     5.41     3.58     5.88     2.54     5.80    3.43      6.41      3.89     6.96     1.55      4.54
Number of adults in household  2.53      1.16    2.60      1.17    2.53     1.13    2.51      1.07      2.77     1.37     2.48      1.13
(aged 16 or older)
Number of children in household  3.24    1.62    2.84      1.43    3.00     1.48    3.53      1.96      2.97     1.48     3.37      1.62
(aged 15 or younger)
Per capita monthly consumption  156.11  132.30  187.70   161.32  167.84   110.97   138.22   111.09    243.28   170.69   132.80    113.43
(Bs.)
Spanish speaking head (1 yes)  0.60     0.49     0.65     0.48     0.67     0.47    0.56      0.50      0.74     0.44    0.55       0.50
Aymara speaking head (1 yes)   0.15     0.36     0.13     0.33     0.09     0.29    0.09      0.29      0.08     0.27    0.19       0.39
Quechua speaking head (1= yes)  0.22    0.42     0.20     0.40     0.23     0.42    0.33      0.47      0.16     0.37    0.23       0.42
Living Central (1 =yes)        0.47     0.50     0.50     0.50     0.34     0.47    0.49      0.50      0.47     0.50     0.48      0.50
Living North (1= yes)          0.18     0.39     0.22     0.42     0.17     0.38    0.13      0.34      0.15     0.36     0.19      0.39
Living South (1=lyes)          0.35     0.48     0.28     0.45     0.49     0.50    0.38      0.49      0.38     0.49     0.33      0.47
Possession of Social Security   0.34    0.47     0.30     0.46     0.88     0.33    0.26      0.44      0.35     0.48     0.26      0.44
(1=yes)                     I                I                 I                 I                 I     _I


Annex Table A-3: Descriptive Statistics (Children Aged 4 Years or Younger)
Facility Used
Ministry of
Public Heafkh   Social Security      NGOs             Private          Sey-Care
Numberofpeople                           1187              128               98              79                 126             756
Percentage                               100               10.8              8.3             6.7               10.6             63.7
Variables                          Mean     S.D.    Mean     S.D.     Mean    S.D.    Mean     S.D.     Mean     S.D.    Mean      S.D.
User fee (Bolivianos)               3.03    11.68    6.48     8.66     0.87     2.90    4.52    5.63    17.65    29.84     0.14     1.75
Transportation cost (Bolivianos)    1.3      7.96    5.16    18.64     4.14    12.29    3.06   11.85     1.84     2.9      0       0
Cost of medicine (Bolivianos)       9.33    29.73   28.26    58.39     6.45    14.69   18.87   57.1     27.08    35.49     2.54    8.49
Waiting time (minutes)             21.31    75.84   60.58    96.78    60.15    61.89   82.95   160.46   40.40    133.17    0       0
Travel time (minutes) (round trip)  24.24  110.1    50.91   108.21    68.41   133.47   59.77   134.08   85.92   260.05     0       0
Age (years)                         1.88     1.37    1.51     1.36     1.90     1.45    1.72    1.43     1.60     1.39     2.01     1.32
Gender (1 =male)                    0.53     0.5     0.53     0.50     0.60    0.49     0.54    0.50     0.56     0.50     0.52    0.50
Mother's education (years)          7.18     5.17    7.11     5.22    10.23     5.63    6.65    4.62     9.52     5.28     6.46    4.89
Father's education (years)          8.02     5.56    7.24     5.53    10.47     5.74    7.41    5.53    10.41     6.23     7.49    5.24
Number of adults in household       2.52     1.16    2.57     1.21     2.52     1.03    2.39    0.97     2.79     1.46     2.48     1.12
(aged 16 or older)
Number of children in household     3.21     1.62    2.70     1.44     3.09     1.63    3.48    1.95     3.13     1.64     3.29     1.59
(aged 15 or younger)
Per capita monthly consumption (Bs.)  144.14  120.15  172.57  126.39  155.47  105.49  111.91   75.52   236.91    168.77  125.76   105.92
Female head (1=yes)                 0.10     0.30    0.10     0.30     0.04     0.20    0.10    0.30     0.09     0.28     0.11    0.31
Presence of old female (1=lyes)     0.10     0.31    0.09     0.29     0.10     0.30    0.08    0.27     0.13     0.34     0.10    0.31
Duration of illness (days)          1.70     4.61    3.17     5.39     1.45     3.08    2.95    5.52     3.58     7.48     1.04    3.65
Diarrhea (1 =yes)                   0.36     0.48    0.36     0.48     0.22     0.42    0.23    0.42     0.26     0.44     0.41    0.49
Cold (1=yes)                        0.54     0.50    0.56     0.50     0.64     0.48    0.52    0.50     0.63     0.49     0.52    0.50
Mcasls (1= yes)                    0.04     0.19    0.02     0.15     0.03     0.17    0.09    0.29     0.05     0.21     0.03    0.18
Other disease                       0.05     0.23    0.05     0.23     0.10     0.30    0.16    0.37     0.06     0.24     0.04    0.19
Living Central (1 =yes)             0.46     0.50    0.45     0.50     0.36     0.48    0.47    0.50     0.48     0.50     0.47    0.50
Living North (1=yes)                0.18     0.39    0.27     0.44     0.14     0.35    0.11    0.32     0.18     0.38     0.18    0.39
Living South (1 =yes)               0.36     0.48    0.28     0.45     0.50     0.50    0.42    0.50     0.34     0.48     0.35    0.48
Possession of Social Security (1 =yes)  0.32  0.47   0.23     0.43     0.88     0.32    0.23    0.42     0.35     0.48     0.26    0.44


Annex Table A-4: Descriptive Statistics (Women Between 15 and 49 Years Old)
Facility Used
Heath Center
Hospital              Private          or Health Post          Hone
Number of people                         776                      377                  104                 74                  221
Percentage                               100                     48.6                  13.4                9.5                 28.5
Variables                        Mean        S.D.        Mean       S.D.       Mean       S.D.     Mean      S.D.      Mean       S.D.
Cost of delivery (Bolivianos)    132.31     325.23      147.77     329.50     350.47     547.92    77.66    161.99      21.56     39.62
Age (years)                      27.41        6.40       27.20       6.40      27.20      5.77     27.18      5.51      27.96      6.94
Single (1=yes)                    0.12         0.32       0.12       0.33       0.11       0.31     0.14      0.34       0.11      0.31
Education (years)                 7.90        5.11        8.82       4.76      10.43       5.29     9.45      5.41       4.62      3.79
Number of adults in household     2.58         1.25       2.55       1.25       2.62       1.18     3.01       1.65      2.48      1.09
(aged 16 or older)
Number of children in household   3.18         1.70       3.10       1.66       2.63       1.29     3.04       1.67      3.64       1.84
(aged 15 or younger)
Per capita monthly consumption   158.31      165.19     166.00     138.76     291.98     301.06    139.69   108.82      88.53     57.41
(Bs.)
Speaking Spanish (1 =yes)         0.63        0.48        0.75       0.44       0.67       0.47     0.61       0.49      0.42      0.49
Speaking Aymara (1=yes)           0.14        0.34        0.03       0.18       0.13       0.34     0.12       0.33      0.32      0.47
Speaking Quechua (1=yes)          0.22        0.42        0.21       0.41       0.16       0.37     0.27       0.45      0.25      0.43
Not working (1 =yes)              0.66        0.48        0.69       0.46       0.68       0.47     0.62       0.49      0.60      0.49
Informally working (1 =yes)       0.25        0.43        0.19       0.39       0.24       0.43      0.23      0.42      0.35      0.48
Living Central (1=yes)            0.43        0.49        0.30       0.46       0.63       0.49     0.55       0.50      0.50      0.50
Living North (1=yes)              0.18         0.39       0.23       0.42       0.05       0.21      0.04      0.20      0.21      0.41
Living South (1= yes)             0.39         0.49       0.46       0.50       0.33       0.47      0.41      0.49      0.29      0.45
Seek prenatal care (1 yes)        0.68         0.47       0.81       0.39       0.83       0.38      0.80      0.40      0.36      0.48
Prenatal visits (times)           3.19        3.07        3.94       3.01       4.60       3.37      3.24      2.39      1.23      2.17


Bibliography
Abu-Zeid H.A.H., and Dann W.M. 1985. "Health Utilization in Ismaila, Egypt." Social
Science and Medicine, 21:451-461.
Akin, John S., C.Griffin, D.K.Guilkey, and B.M.Popkin. 1984. The Demand for Primary
Health Care in the Third World. Totowa, N.J.:Littlefield, Adams.
. 1986. "The Demand for Primary Health Care Services in the Bicol Region of the
Philippines." Economic Development and Cultural Change, 34(4):755-82.
Akin, John S., Hazel Denton, David K. Guilkey, Ronald J. Vogel and Annemarie Wouters.
1991. "Health Care Costs, Demand and Cost Recovery in Ogun State, Nigeria."
Washington, D.C.: The World Bank.
Alderman, Harold, and Paul Gertler. 1989. "The Substitutability of Public and Private Health
Care for the Treatment of Children in Pakistan." Living Standards Measurement
Study Working Paper 57. Washington, D.C.: The World Bank.
Amemiya, Takeshi. 1985. Advanced Econometrics. Cambridge, Mass.: Harvard University
Press.
Amemiya, Takeshi, and Keiko Shimono. 1989. "An Application of Nested Logit Models to
the Labor Supply of the Elderly." Economic Studies Quarterly, 40(1):14-22.
Baker, Judy L., and Jacques van der Gaag. 1993. "Equity in Health Care and Health Care
Financing: Evidence from Five Developing Countries." in E. van Doorslaaer, A.
Wagstaff, and F. Rutten (eds). Inequity in the Finance and Delivery of Health Care:
An International Perspective. London, England: Oxford University Press.
Behrman, Jere, Masako li, and David Murillo. 1992. "Correlates of Urban Schooling in
Bolivia in 1990: Gender, Family Background, and Region." La Paz, Bolivia:
UDAPE/Grupo Social.
-_     . 1992. "Household School Demands in Urban Bolivia." La Paz, Bolivia:
UDAPE/Grupo Social.
Bitran, Ricardo. 1991. "Health Care Demand in Developing Countries: A Model of
Household and a Market Simulation Model of Health Care Financing." Unpublished
Ph.D dissertation, Department of Economics, Boston University.
Brown, Stephen, and David Sibley. 1986. The Theory of Public Utility Pricing. New York:
Cambridge University Press.
45


Cameron, A.C., P.K. Trivedi, Frank Milne, and J. Piggott. 1988. "A Microeconometric
Model of the Demand for Health Care and Health Insurance in Australia." Review of
Economic Studies, 85-106.
Creese, Andrew. 1991. "User Charges for Health Care: A Review of Recent Experience."
Health Policy and Planning, 5:260-270. Switzerland: World Health Organization.
de Ferranti, David. 1985. Paying for Health Services in Developing Countries. World Bank
Staff Working Paper Number 721. Washington D.C.: The World Bank.
Dor, Avi, and Jacques van der Gaag. 1987. "The Demand for Medical Care in Developing
Countries: Quantity Rationing in Rural Cote d'Ivoire" Living Standards Measurement
Study Working Paper 35. Washington, D.C.: The World Bank.
Dubin, Jeffrey, and Daniel McFadden. 1984. "An Econometric Analysis of Residential
Electric Appliance Holdings and Consumption." Econometrica, 52:345-362.
Ellis, Randall P., and Germano M. Mwabu. 1991. "The Demand for Outpatient Medical
Care in Rural Kenya." Boston University, Processed.
Gelbard, Alene H., and Lieselotte de Barragan. 1991. "Policy Implications of the 1989
Bolivia DHS." Paper prepared for DHS World Conference 1991, Washington, D.C.
Gertler, Paul, and Harold Alderman. 1989. "Family Resources and Gender Differences in
Human Capital Investments: The Demand for Children's Medical Care in Pakistan."
Washington. D.C.: International Food Policy Research Institute.
Gertler, Paul, L. Locay, and W. Sanderson. 1987. "Are User Fees Regressive? The Welfare
Implications of Health Care Financing Proposals in Peru." Journal of Econometrics,
36(supp.): 67-88.
Gertler, Paul, and Jacques van der Gaag. 1990. The Willingness to Pay for Medical Care:
Evidence from Two Developing Countries. Baltimore: Johns Hopkins University
Press.
Gertler, Paul, John Strauss, Omar Rahman, Nga Vuong, and My Vuong. 1993. "Health Care
Utilization and Provider Choice in Jamaica: An Interim Report." Rand Corporation.
Glewwe, Paul. 1990. "Improving Data on Poverty in Third World: The World Bank's Living
Standards Measurement Study." World Bank Working Paper 416. Washington, D.C.:
The World Bank.
Goldberger, Arthur S. 1991. A Course in Econometrics. Cambridge, Mass.: Harvard
University Press.
46


Greene, William. 1993. Econometric Analysis. New York: Macmillan.
Grosh, Margaret E. 1991. "The Household Survey as a Tool for Policy Change." Living
Standards Measurement Study Working Paper 80. Washington, D.C.: The World
Bank.
__     . 1990. Social Spending in Latin America: The Story of the 1980s. World Bank
Discussion Paper, no.106. Washington, D.C.: The World Bank.
Heller, Peter. 1982. "A Model of the Demand for Medical and Health Services in Peninsular
Malaysia." Social Science and Medicine, 16:267-84.
Horton, Susan. 1990. "Labour Markets in an Era of Adjustment: Bolivia." University of
Toronto. mimeo.
Ii, Masako. 1993. "Willingness to Pay for Medical Care:Evidence from Urban
Areas In Bolivia." Ph.D Dissertation, Department of Economics, University of
Wisconsin, Madison.
Inter-American Development Bank. 1991. Economic and Social Progress in Latin America,
Special Section: Social Security. Washington, D.C.
Jimenez, Emmanuel. 1987. Pricing Policy in the Social Sectors: Cost Recovery for Education
and Health in Developing Countries. Baltimore: Johns Hopkins University Press.
. 1990. "Social Sector Pricing Policy Revisited: A Survey of Some Recent
Controversies." Proceedings of the World Bank Annual Conference on Development
Economics 1989.
Judge, George G., R. Carter Hill, William Griffiths, Helmut Lfitkepohl, and Tsoung-Chao
Lee. 1988. Introduction to the Theory and Practice of Econometrics. 2nd ed. New
York: John Wiley & Sons.
Lavy, Victor, and John M. Quigley. 1991. "Willingness to Pay for the Quality and Intensity
of Medical Care: Evidence from Low Income Households in Ghana." Living
Standards Measurement Study Working Paper 94. Washington, D.C.: The World
Bank.
McFadden, Daniel. 1973. "Conditional Logit Analysis of Qualitative Choice Behavior." In
P.Zarembka (ed.), Frontiers in Econometrics. New York: Academic.
__     . 1977. "Quantitative Methods for Analyzing Travel Behavior of Individuals: Some
Recent Developments." Cowles Foundation Discussion Paper no.474.
47


. 1978. "Modelling the Choice of Residential Location." In A.Karlquist et al. ed.,
Spatial Interaction Theory and Residential Location, 75-96. Amsterdam: Noth-
Holland.
_ . 1981. "Econometric Models of Probabilistic Choice." in C.F. Manski and D,
McFadden, eds., Structural Analysis of Discrete Data with Econometrics Applications,
pp.198-272. Cambridge: MIT Press.
McGreevey, William. 1990. Social Security in Latin America: Issues and Options for the
World Bank. World Bank Discussion Paper, no.110. Washington, D.C.: The World
Bank.
McPake, Barbara. 1993. "User Charges for Health Services in Developing Countries: A
Review of the Economic Literature." Social Science and Medicine, 36:1397-1405.
Maddala, G.S. 1983. Limited-Dependent and Qualitative Variables in Econometrics.
Cambridge: Cambridge University Press.
Ministry of Public Health and the World Bank. 1990. "Salud Urbana: Segun la Encuesta
Integrada de Hogares 1989." La Paz, Bolivia.
Morales, Juan Antonio, and Jeffrey Sachs. 1988. "Bolivia's Economic Crisis." NBER
Working Paper No.2620. Cambridge MA.
Mwabu, Germano. 1986. "Health Care Decisions at the Household Level: Results of a Rural
Health Survey in Kenya." Social Science and Medicine, 22:315-319.
Mwabu, Germano, Martha Ainsworth, and Andrew Nyamete. 1993. "Quality of Medical
Care and Choice of Medical Treatment in Kenya: An Empirical Analysis." Kenyatta
University and The World Bank. mimeo.
Schacter, Mark, Margaret Grosh, and Steen Jorgensen, 1992. "What? Why? How? A Primer
on the ESF." in Steen Jorgensen, Margaret Grosh, and Mark Schacter (eds.),
Bolivia's Answer to Poverty, Economic Crisis, and Adjustment: The Emergency Social
Fund. Washington, D.C.: The World Bank.
Sommerfelt, A. Elisabeth, J. Ties Boerma, Luis H. Ochoa, and Shea 0. Rutstein. 1991.
"Maternal and Child Health in Bolivia: Report on the In-Depth DHS Survey in
Bolivia 1989." Columbia, Maryland: Institute for Resource Development/Macro
Systems, Inc.
Vogel, Ronald. 1988. Cost Recovery in the Health Care Sector, Selected Country Studies in
West Afica. World Bank Technical Paper Number 82. Washington, D.C.: The World
Bank.
48


Waddington, C.J., and K.A. Enyimayew. 1989. "Pricing of Pay: The Impact of User
Charges in Ashanti-Akim District, Ghana." International Journal of Health Planning
and Management, 4:17-47.
World Bank. 1980. Health Sector Policy Paper. Washington, D.C.
. 1987. Financing Health Services in Developing Countries: An Agenda for
Reform. Washington, D.C.
___      . 1988. Bolivia Population, Health, and Nutrition Sector Memorandum.
Washington, D.C.
. 1989a. Bolivia Public Sector Expenditure Review with a Special Emphasis on the
Social Sectors. Washington, D.C.
. 1989b. Staff Appraisal Report Bolivia Integrated Health Development Project.
Washington, D.C.
. 1989c. Bolivia Country Economic Memorandum. Washington, D.C.
. 1990. Bolivia Poverty Report. Washington, D.C.
. 1991. Bolivia Country Economic Memorandum. Washington, D.C.
. 1992. Bolivia Public Sector Investment Program Review. Washington, D.C.
. 1993. World Development Report 1993. New York: Oxford University Press.
Yoder, Richard. A. 1989. "Are People Willing and Able to Pay for Health Services?" Social
Science and Medicine, 29:35-42.
49



Distributors of World Bank Publications.
ARGENTINA                           DENMARK                              JAPAN                               SINGAPORE TAIWAN
Carlos Hirsc, SRL                  SamfundsLitteratur                   Eastern Book Service                Cower Asia Pacific Pte Ltd.
Galena Cuemes                       Rosenoerns All6 11                   Hongo 3-Chome, Bunkvo-ku 113        Golden Wheel Building
Florida 165, 4th Floor-Ofc. 453/465  DK-1970 Frederiksberg C             Tokyo                               41, Kallang Pudding #04-03
1333 Buenos Aires                                                                                            Singapore 1334
EGYPT, ARAB REPUBLIC OF              KENYA
Oficina del Libro Internacional     Al Ahram                             Afnca Book Service (E.A.) Ltd.      SLOVAK REPUBLIC
Alberti 40                          Al Galaa Street                      Quaran House, Mfangano St.          Slovart C.T.G. Ltd.
1082 Buenos Aires                   Cairo                                P.O. Box 45245                      Krupinska 4
Nairoin                             P.O. Box 152
AUSTRALIA, PAPUA NEW GUINEA,        The Middle East Observer                                                 852 99 Bratislava 5
FIJI, SOLOMON ISLANDS,              41, Shenif Street                    KOREA, REPUBLIC OF
VANUATU, AND                        Cairo                                Daejon Trading Co. Ltd.             SOUTH AFRICA, BOTSWANA
WESTERN SAMOA                                                            P 0. Boa 34                         Oxford Universitv Press
D.A. Information Services           FINLAND                              Yeoeida                                Southern Africa
648 Whitehorse Road                 Akateemmen Kirjakauppa               Seoul                               P.O. Box 1141
Mitcham 3132                        P.O. Box 23                                                              Cape Town 8000
Victoria                            FIN-00371 Helsinki                   MALAYSIA
University of Malaya Cooperative    SPAIN
AUSTRIA                             FRANCE                                  Bookshop, Limited                Mundi-Prensa Libros, S.A.
Gerold and Co.                      World Bank Publications              P.O Box 1127, Jalan Pantai Baru    Castello 37
Graben 31                           66, avenue d'Ikna                    59700 Kuala Lumpur                  25001 Madrd
A-1011 Wien                         75116 Pars
MEXICO                              Libera Internacional AEDOS
BANGLADESH                          GERMANY                              INFOTEC                             Consell de Cent, 391
Micro Industries Development        UNO-Verlat                           Apartado Postal 22-860              08009 Barcelona
Assistance Society (MIDAS)        Poppelsdortqr Allee 55               14060 Tlalpan. Mexico D F.
House 5, Road 16                    53115 Bonn                                                               SRI LANKA & THE MALDIVES
Dbanmondi R/ Area                                                       NETHERLANDS                         Lake House Bookshop
Dhaka 1209                          GREECE                               De Lindeboom/ InOr-Publikaties      P.O. Box 244
Papasotinou S.A.                     P.O. Box 202                        100, Sir Chittampalam  A.
BELGIUM                             35, Stournara Sir.                   7480 AE Haaksbergen                    Gardiner Mawatha
Jean De Lannoy                      106 82 Athens                                                            Colombo 2
Av. du Roi 202                                                           NEW ZEALAND
1060 Brussels                       HONG KONG, MACAO                     EBSCO NZ Ltd.                       SWEDEN
Asia 2000 Ltd.                       Private Mail Bag 99914              Fntzes Customer Service
BRAZIL                              46-48 Wvndhat Street               New Market                          Regenngsgatan 12
Publicacoes Tecnicas Internacionais  Witmug Centre                       Auckland                            S-106 47 Stockholm
Ltda.                             7th Floor
Rua Peixoto Gomide, 209             Central, Hong Kong                   NIGERIA                             Wennergren-Wlliams AB
01409 Sao Paulo, SP                                                      University Press Liuted            P.O. Box 1305
HUNGARY                              Three Crowns Building Jencho        5-171 25 Solna
CANADA                              Foundation for Market Economy        Private Mail Bag 5095
La Diffuseur                        Dombovan Ut 17-19                    Ibadan                              SWITZERLAND
151A Boul. de Mortagne              H-1117 Budapest                                                          Librairie Pavot
Boucherville, Quebec                     .                               NORWAY                              Case postale 3212
J4B 5E6                             INDIA                                Narvesen Information Center         CH 1002 Lausanne
Allied Publishers Private Ltd.       Book Department
Renouf Publishing Co.               751 Mount Road                       P 0. Box 6125 Etterstad             Van Diermen Editions Techruques
1294 Algoma Road                    Madras - 600 002                     N-0602 Oslo 6                       P.O. Box 465
Ottawa, Ontario KlB 3W8                                                                                      CH 1211 Geneva 19
INDONESIA                            PAKISTAN
CHINA                               Pt. Indira Limited                   Mirza Book Agency                   TANZANIA
China Financial & Economic          Jalan Borobudur 20                   65, Shahrah-OsQuaid+Azam            Oxford University Press
lishing House                     P.O. Box 181                         P.O. Box No. 729                    Maktaba Street
8, Da P Si Dong Jie                 Jakarta 10320                        Lahore 54000                        P.O. Box 5299
Bejing                                                                                                       Dar es-Salaam
IRAN                                 Oxford University Press
COLOMBIA                            Kowkab Publishers                    5 Bangalore Town                    THAILAND
Infoenlace Ltda.                    P.O. Box 19575-11                   Sharae Fasal                       Central Books Distnbution Co. Ltd.
Apartado Aereo 34270                Tehran                               P.O. Box 13033                      306 Silom Road
Bogota D.E.                                                              Karachs-75350                       Bangkok
IRELAND
COSTA RICA, BELIZE, GUATE           Government Supplies Agency           PERU                                TRINIDAD & TOBAGO, JAMAICA
-MALA, HONDURAS,                    4-5 Harcourt Road                    Editorial Desarrollo SA             Systematics Studies Unit
NICARAGUA, PANAMA                   Dublin 2                             Apartado 3824                       #9 Watts Street
Chispas Bookstore                                                        Lima I                          -   Curepe
75 Meters al Norte del Hotel Balmoral  ISRAEL                                                                Trinidad, West Indies
en calle 7                        Yozrmot Literature Ltd.              PHILIPPINES
San Jose                            P:O. Box 56055                       International Book Center           UGANDA
Tel Aviv 61560                       Suite 720, Ctyland 10               Gustro Ltd.
COTE D'IVOIRE                                                            Condominium Tower 2                 1st Floor, Room 4, Geogiadis Chambers
Centre d'Edition et de Diffusion    RO.Y. International                  Ayala Avenue. H.V. dela             P.O. Box 9997
Africaines (CEDA)                 P 0. Box 13056                         Costa Extension                   Plot (69) Kampala Road
04 B.P. 541                         1el Aviv 61130                       Makau, Metro Manda                  Kampala
Abidjan 04 Plateau
Palestinta Anuthonty/Middle East     POLAND                              UNITED KINGDOM
CYPRUS                              Index Information Services           International Publishing Service    Micronfo Ltd.
Center of Applied Research          P O.B. 19502 Jerusalem               Ul. Piekna 31/37                    P.O. Box 3
Cyprus College                                                           00-577 Waszawa                     Alton. Hampshire GU34 2PC
6, Diogenes Street, Engomi          ITALY                                                                    England
P.O. Box 2006                       Licosa CommissLonana Sansoni SPA     PORTUGAL
Nicosia                             Via Duca Di Calabna, 1/1             Livraria Portugal                   ZAMBIA
Casella Postale 5i2                  Rua Do Carmo 70-74                  University Bookshop
CZECH REPUBLIC                      50125 Firenze                        1200 Lisbon                         Great East Road Campus
National Information Center                                                                                  P.O. Box 32379
P.O. Box 668                        JAMAICA                              SAUDI ARABIA, QATAR                 Lusaka
CS-113 57 Prague I                  lan Randle Publishers Ltd.           Jan Book Store
206 Old Hope Road                    P.O. Box 3196                       ZIMBABWE
Kingston 6                           Riyadh 11471                        Longman Zimbabwe (Pte.) Ltd.
Tourle Road, Ardbennie
P.O. Box ST 125



LSMS Working Papers (continued)
Decomposition with Applications to Brazil and India in the 1980s
No. 84 Vijverberg, Measuring Income from Family Enterprises with Household Surveys
No. 85    Deaton and Grimard, Demand Analysis and Tax Reform in Pakistan
No. 86 Glewwe and Hall, Poverty and Inequality during Unorthodox Adjustment: The Case of Peru,
1985-90
No. 87    Newman and Gertler, Family Productivity, Labor Supply, and Welfare in a Low-Income Country
No. 88    Ravallion, Poverty Comparisons: A Guide to Concepts and Methods
No. 89    Thomas, Lavy, and Strauss, Public Policy and Anthropometric Outcomes in C6te d'Ivoire
No. 90    Ainsworth and others, Measuring the Impact of Fatal Adult Illness in Sub-Saharan Africa: An
Annotated Household Questionnaire
No. 91    Glewwe and Jacoby, Estimating the Determinants of Cognitive Achievement in Low-Income
Countries: The Case of Ghana
No. 92    Ainsworth, Economic Aspects of Child Fostering in C6te d'Ivoire
No. 93    Lavy, Investment in Human Capital: Schooling Supply Constraints in Rural Ghana
No. 94 Lavy and Quigley, Willingness to Pay for the Quality and Intensity of Medical Care:
Low-Income Households in Ghana
No. 95    Schultz and Tansel, Measurement of Returns to Adult Health: Morbidity Effects on Wage Rates
in C6te d'lvoire and Ghana
No. 96    Louat, Grosh, and van der Gaag, Welfare Implications of Female Headship in Jamaican Households
No. 97    Coulombe and Demery, Household Size in C6te d'Ivoire: Sampling Bias in the CILSS
No. 98    Glewwe and Jacoby, Delayed Primary School Enrollment and Childhood Malnutrition in
Ghana: An Economic Analysis
No. 99    Baker and Grosh, Poverty Reduction through Geographic Targeting: How Well Does It Work?
No. 100 Datt and Ravallion, Income Gains for the Poor from Public Works Employment: Evidence from
Two Indian Villages
No. 101   Kostermans, Assessing the Quality of Anthropometric Data: Background and Illustrated
Guidelines for Survey Managers
No. 102   van de Walle, Ravallion, and Gautam, How Well Does the Social Safety Net Work? The
Incidence of Cash Benefits in Hungary, 1987-89
No. 103   Benefo and Schultz, Determinants of Fertility and Child Mortality in Cate d'Ivoire and Ghana
No. 104   Behrman and Lavy, Children's Health and Achievement in School
No. 105   Lavy and Germain, Quality and Cost in Health Care Choice in Developing Countries
No. 106   Lavy, Strauss, Thomas, and De Vreyer, The Impact of the Quality of Health Care on Children's Nutrition
and Survival in Ghana
No. 107   Hanushek and Lavy, School Quality, Achievement Bias, and Dropout Behavior in Egypt
No. 108   Feyistan and Ainsworth, Contraceptive Use and the Quality, Price, and Availability of Family Planning
No. 109   Thomas and Maluccio, Contraceptive Choice, Fertility, and Public Policy in Zimbabwe
No. 110   Ainsworth, Beegle, and Nyamete, The Impact of Female Schooling on Fertility and Contraceptive Use:
A Study of Fourteen Sub-Saharan Countries
No. 111   Oliver, Contraceptive Use in Ghana: The Role of Service Availability, Quality, and Price
No. 112   Montgomery, Kouam6, and Oliver, The Tradeoff between Number of Children and Child Schooling:
Evidence from Cate d'Ivoire and Ghana
No. 113   Pradhan, Sector Participation Decisions in Labor Supply Models
No. 114   Beegle, The Quality and Availability of Family Planning Services and Contraceptive Use in Tanzania
No. 115   Lavy, Spratt, and Leboucher, Changing Patterns of Illiteracy in Morocco: Assessment Methods Compared
No. 116 Lavy, Palumbo, and Stern, Health Care in Jamaica: Quality, Outcomes, and Labor Supply
No. 117   Glewwe and Hall, Who Is Most Vulnerable to Macroeconomic Shocks? Hypotheses Tests Using Panel
Data from Peru
No. 118 Grosh and Baker, Proxy Means Tests for Targeting Social Programs: Simulations and Speculation
No. 119   Pitt, Women's Schooling, the Selectivity of Fertility, and Child Mortality in Sub-Saharan Afica
No. 120   Grosh and Glewwe, A Guide to Living Standards Measurement Study Surveys and their Data Sets
No. 121   van de Walle, Infrastructure and Poverty in Viet Nam
No. 122   Ravallion, Poverty Comparisons: A Guide to Concepts and Methods


-   rqL                      г.  �ё          �(   •'гtг ''й ' ' '�  :7;� '_5'   ..��{.,1._`   ^•  � '-".У  '   4у,-'.   •    -,•:..,...,�/  .  уΡ,��.�.?\. �
'rL�i�M.     2.   f'  S.,.�.  .r}`'      V'"�      ri''  � r�� .�.`   у'r'�•![1`�   lу:� �   �N�.r,.      r.!r.�ibF:.r�,�_ц�q+,jsl�-•
�!'                                           r`      }3                          ,.� JJ.vr`d�Ri�7'RRYY  .�г r�l�
" г                              -                   � •�  '�ы,.
v�                   r�-'З : �
г�                „�';' ij•s`г!.�
ti                    -                                                          м                 rF�-'.�;
��                              �       ; zц'�'^и:".' i-:
�  -�   '.и         г �'                                                             _ _   ш   �?./�'� .д�'�?'.
ь                                                     _          ,   '    :�t:i   .�...   ,                   �     И     b���,',:
'                       ��  ,                                   .f.`5а�"        ,   �         ^у'   f  •      ti    М     �"г� }
, `�:.                                                                                          ь. �            М           ъ �  �f •''г
3S            М     ор         �
1                                                                                                        .,���,                   ..f  �..1`�:
�;                            о        �.._
�•                   _^'     Z   �' �,�:�1..
+                                                    -                               ч`r.s�°'.   ,е'       ���     СО   �'.,'.,'�.
у                                                                                               ,.               �           °�   и   'S%�-�g.c.r � ^,
!       -   Р        .:i�16y�h �
�   д.                        'М .,   �  ,н.                        � �' ;�.у,,+...�
.                       �          �, �`  ,ч�
'К    -                У    дцwi   yj.�.,                             �Г'•   .�
-           °а.   �•            _             ?лг      �     ..�, `rr_  ;;..,•,"^у,�.__   � ^�^       ! _., г.а., °. �. и'; •..аи   �� .:�т•.,
�•�.:      �         }��;�рΡа��,,,     �   ,         .,�       ,    �,��.w,���'.
't                           -               �,     ч   �r �,Af                      �::r,.; ',..' 1   �
�7+з          ���   f�ы Т.3 �+.>у   r               '1i�.:':�'ь гr.дlа:-   'f�di,'`'.+'ry.
о                       �          '"�   •�,  Kti,1� А_'.{:..   .     � и sвв�      S"      с�М �'  {   �, '
А   �'�'              #+уь   }�  r3'                     .'дд.::�l    ryc.r  !S- ry."ув� �
..   •                            v.,�b ,.4_ ц��,:  �г.- r.�r, f'�г:. ,   4 �                   г 7',;.?г`�'.'�'дУ`,я)� , .•   �- �.tv' t' �:
'                        :ti.   '�С'..;rl,м   �`        �;�ь(а     г....   '"�   .,                  .•      �qiit   у   •
г    у         1�'   �к                                q'!.    �_е:.lЧ'   .'   t              ...        .и.?�ы   л1�а�ё�'�
-     у  �ь � �  ':   � ц!    � �.��г',ми; 4"t�,.        *i  а";,   5.:с'  :  �,{,                     '{'^:г'.'.  •�.:  'й�.i7.л.ь,: У  JI.�   .   .�
r   ,,                       �W i   у':,;        :ii. '�I��':чг'. у'а"   i:1''r               �� ;!t,  °�   - �'.#�"      '             �
�•....   :     й'У,.�
w'ЧГ+              1�              ✓ У�у�?'�.."' :         г.:.     .л'� п г f               .с             "i
�   .'�'+, А" г.,           ..     ,1�� у.,�,„ 1! ,. ,!''^^  $'   'г   „ �г..                   qiг.,,; ' ,.   ,:�А,   �r  ,
: �                                           �г   к'.''т ,'.. �.+=;,                               `�,        '  ' ,= г
:,�.  л',it �р�� •�   A��f' ы�'     �   M       .�ч       .  а..�н`.},l'                      ..     �,ь .,�      .,fi   =�г�,,'
'а   г д,7'?   ��       м   i:��_,J�A  .,tr-y4"•."��.�  .;Г11   t�,»�.   jt;г��,q               -;    F': '�I�`•'J':o-^';'';Гf:й,ч  '
,, , '..мг.1�'  �j. ��,УΡg   �!i,'ц� � !   . r�y �д   iA v:,i �; ��.: G: г '`!F"  r в.�        - и.'1i        ц         .+..         .
�~/    �         j4    �д,:W    ..�:     R          .м7     i  � }y.Vi+                  }'1'   �     :   .'�`;•7'           ,
,Fs   +-   "'ТМ   i.     ;   ;    ь.i r      и°1�Fv61�„ц,и.:fi`+!S�"./                   f.'i"  'У', �  =Т:'�ii.A��,.:,;�,.г';.
�гi".�Х;'��^`�у ��   .'��А.:�!.7.   . ,�.,.4яt, �нт.�Lь �'г.-,'1 ``�.' `х•...Г г        � д - :''s.'.А,уг�уΡ{ гд. �  �! .r�  •    ,
_                                                                                   .�      .�,,,��j   l�.   `,д.  Г',��У-гг�С,�-  !"- sj[:;У,.  г._
�yS;'.i'...�уУв�,��   б i.:'.'г1�''  'уΡ� _'r`.'S''  г�,��г,���i,А'.V����•+'i  �4:?�l��r:,i4   :,�л'�'  ...,ufr:�+� Jн1r-��,b�.�;��-
�f' ''�j.   ::°i...   �' и�ь'  L,�' �',.   '`�*г• �Г �  *3✓"'wfд`��'9    К,�},''"�.. F  fцд,  .   г   • ь  ..;'4',{':г,ь,.. ��г��•ч'  r:
4,.            �Р               j�,   %i д��:А   '�'   ¢•ьУА�"F.�ч. ��YI��АΡ( tt•  215�              �      ly'у.   "5j^kД7ч:,  �!�% _
, iи         �,�        ��:,г.�"G� .   . ��ь`'�       л'       1-:{.fiN�•'                        i� ��'�'      �1.,�,  '  -  п   �
г�             :.. ���        ,_.      :�;    ��Y��S�'дl.:',.�,. '�":;=.                                                             �, 'r°,��`���'����х., �s�,.
д�}��'�    j�'' 4 ,
.                                  �                '.ЭU �iL�?д'^.��•��.,. �     'гf• ,r, ;F'т�t,  У У�!Ь!!'ь   ' ltirf"�•�ри;•'i'У,хг•.,м1  и��^cRi'.  �                            : �"       1�3 �              i
i�'а, .'�'г�gtтi:_'�L,   •�'�?ь�'.�=,'�!���:?•�.�'i^   :��;�У,3^,;;е                   г,�'�.,•�`'r`�диеf:��У  ."t
'             (i�$ %;�i-!��                       "�.�' ^., kз,           г  �. _  .   [�h': `х:�` ,. � _                                  .: �:'  и -  . У.г«
� �                            �i�.�y?0.'• h`�(.н�7 ,   l   ;е' �s;�� �  � А:        ëi f��.��'            �.s�8�'� i  :1 �                  ;� ��   i�      г,нG,1...,,     .  i
l             i.fЙS:   `       .�м:   М    �,}�   '   i`,,.Tiif гУ••   [,�S                       '!.'1•'гмь;:.i..:'          fC,'  ,    J
f ь  �                    д '   �ьj   G'',�1•^�   'С  � ,�   �-ь....             .б_ .  1                   .,У.г   Й                         'д•,�ч     ' �'      1
�   S�  ..  :1                 i+ ,�',.   `-•/;r, ,ц1., 4".  �ч.'�.�`        .ьг{;s��•* .�£,: �Г'��1.:�       ry;  э'                  ,.�eL.,'..:   :аг'i   'iг
�.�            "(�     ,     'У            '•`               �.г'. '.,fи: ij,.                                           `i-' �.  А'.•''.�..�.
�              г                        ir .   х.  е„ь  ь'   G-1.�  г,у��      •   .   `�.:7'�,   �.ь.   ,у j*   уΡл.,й�:.F                    " '.  -.R�'-   ....      I
��i'У   н��уF   1ц   �S�y- '?в�:.l.� L:�.ч;�,,�i  ^L.�l�::t:.ll��i.�"  "11'в',��ti.!   _^              е"�.('.ka_•  .�  ��ri 1.i "�       .;7
1     'i  `'}'       �д°   .2   д°ь   ,   у�',А}:  •   .    у   м��и�:у.-ч ч_:r'.r.C                          .di        ,,�-   'ё    �-
,} :   �.. .n i'�   �   .�%��5`ьt��.к.�чs3'у> ��.--��."дf."�;'.7��-��ги. .м:Li'�'�.'�.�                          ::-, :   в �;'Э+уlг' � _ �+'    _^�
д>                  ��g'��       �5.,н�а.,{     r    xfи  �.   J•'�гь. -ь'      _                    г•,r}l'�i'.оу�[Л.  +�,.-�'��а�_,  .
;jr               f�  j•�4.  �•'r'q�.       у.             '°„^'."      st,                                sc     +� ;`sг,  у.
•„{,с     �агл�.,�,� л.•   М{'�'=�`'я;.г�'�,1'k,3,'•т^`i'Yьr...,�f3s�rj�,.��j",_'Р-   ь.а,�'iч2�..,;�>�vF�г`��.-��.,•��
��•ii�   }М�'йJt;:.�i     '�Е'.*>��.п,.�        �� ^"b''  ;(с ч!'т�Н�.:•            °'F.'�   .   ��. р            ��~•'       �  .    .
�          ..���i�.r�' �     2-t  ..       �`f :� ♦                            .в�,':��-�^5н'К.:��r+�  "      �
�             2ул   "�.        5'           ь-.':г                      �"�    ",S.'.                9I.я,'"•'� ,     r'_:    'Nь.'.q4'�   _
'�гу�,;У    G;,:, J��;л'А.   � ч.                      ц"ь,о:.  •..5�'     а-�              ;и . �             ;. �,�   . ^  ,
j                  ��,r�;,s...--s,.  �.��ьt'-i Э, м   �'у;                              •. 4•.  - .у;4�у •у':ьр�+;,�<f ,
`�,�                                                                                                                                       p         �i;Ii,'�'�   `i'.r'Чв,: ....                               ю�.•�  ..        5�  ь    г чгz,.
'        -'7е,        а',��^'�''Tw•:'•.':'ксд.`.й           ,.а�,:°�t�k.   !   •!'Г                  -,ц��-.`    -                    �
,�)'�:�                '9�','            tl'                ��, �                              .        -     1 . �`t,,:=fi".:,i,� �.
,.�ь•У             J!Рь            �             r�.                                       . �.}; i..!:�
'�.                                •�л':.,                                   .. �      �,,r.             . �уΡ
�''�л           °';��:      �  -iG=  зr г:                       л :   �'� .`      У"�r.� : ' .
-.                        ТΡt                          �� �       ���    �..,,эЧ:'r'.А.
!' .                        •f к'+��r           �г                                         .�i^' �:•   �          .
�   у     ,.ьуΡ.  .. IУΡ.�                      f-   !�S(''  !   '/=1�,'и^,,,
{              п  •Т�^      i       '�;р:°   йL?�Т                   �'уΡ   ���r,i/{а    г}ч'•.    .'Г!в    •i
�                       уа.�,'�f��         rr  °r:.�.  i:     �''", r                                     i� г� '"°i1,�YYY".��   •�`" r.'�_
,                 н.     .Р'-а   �3ь •.��7�  7      :.!                                     J   �.в                   i`    ,.'  •
�r��   •             �          �    �L�  s .   � ...               ..��.�,•,�;�т  д ��;.  ,`:�'�:•,
.    ,   °i..          +  ,   {,   ��  �+(,�'{ .+��,-�,�,уΡ��                                   5'.,т,�,+">.#'г'и,���f�t'� а         �k�Sл-а
'.,е         ?�ь.�'� •'�й             '€��.•У,   £ли.+_'.Ar��%j,r                  �;i'!'}` � g '�?�t�'�'"+��� �г.
"�;  �!r  ��   _      ,�°�}'..t        у•   b7uiA�43'''-F'iл�'��•уΡ,,                                  `И'�,�i`'�dF�   У'�       �'.  .   .
}                    з                      :�ь   i-   г�•.  f_л        5' :
�`31��          лj �,k �t�, �.   �Ы.' GA      '�    §1j'  ; :°4'4�д_'�.,д�. Э� i' .             .F'.'S'�' F.f :j � r   3�I�{i: .'�Tf :.: .
..   у'  т  �,�дΡ   е7,..�•,,•    Yr, :'�.  �   ,     .t                          ,. ,  `,:К_'  +   ..�'
'      ':уΡ  t .д;,, 1�i  г'�.Iu.`1.,   Tk qX4     �"�   7;'                        ..io .:' f•   ''уь.'1�c�i   ,�т�5�..�, .,.
�.            +1,yi?s�I�S`т'  д��:�  . .        �.v::       l�:                                                     -.
�          у  "'      1'              . ^   .`'�и�     ,   . .}' 9у   `�х                      �' i лКt.У;� �ьихт�,��t�,3.' 3 '   . i
:�,�F,ь   ,j'                   7.>               .�г7Я�   ��  •гГ                                       `'   ё'       f-..��`;  `
.ч� .                  f,iгy'7;�'.�s�.      у�'i.,�dп.,.:s                                   'R��_     �к           �а�У�-_:
�                                                                                     v;             .tb,ч;        � 1:�r   t'i'£;1  #='г,�,;ц.  .', t".   .'     �{'�:�>]                    '9 �,,.<,   k�.. �  'Т`   r.T �   �
r.     •  ...`   '�  .� А  А�      -��•'_r"`.  +"„х,рг�5,  'д�� �,.i!�•�.`?К{f!              .+   ':��'s. ' й.'    А.   у'4"'r`.
�iR�•rt             У,;�. 4N �.у.�°j�'            д                      °'      •:ia��   �:r    ь`�',т'  ,1
:�        r                  '�!        � '�п•       х � 'С   *н        }�  F'.   r`�_  ��.I� s                        . 4.�..�.�.             А?   �
7                                                                                 +1   �         �•м�     ц.I�    4''.,• �'а. ..•_      ��    �w4C f.il  ti'{f   "YS'  �'.�                     �� Г     'А��S� �''�>.%:в
�                                                                                         �Ги        �     ��,�i4г          'жг   �:^          �      ,У'                                     �1  !г     i   У+�.  i�'.:�.'i
�            /�              уΡ   � r  ,�            �'г,   ��     �' �}'r�,�,й';п:1��•ч;                   �дг,.
~�   �   ''У  �,    '   �   ь �,.�Sл   �-Z.T� lv'�,�  t•G��     .�             }-,            °,г,�'^..г�•L7:  �'у        .;.
.,хвr°'.          .      и +�_. �,. �   .А�..�   !уΡ   F �    .. � i *1; �.                    �уΡ+!��   l ��уΡ4'   у.�.
�                                                                                           Г}.La'`.ь��-       -.��а+�'-'•ь��',�'.�° е*(..м�,  '(У�Л�iаАй�;.�,г!А)_�.-�.:в�. l�SiV -            .,3"'•l�д1^•��  л��� t
�{,.,                1  ,�+-� .      ^                    � г.^!;•..«  � =М'����r.zJ..,               _      �г",.�,      1
�'.  i      _'�'  Гу /�г,�}   I   А    �.уΡrlt                 •д..�'г                           •      .   К�      �S•�и��,'~. ."�
�                                                                д                          �7:Х�,��             w.;ifT.•."'у,,М'.ь.�f•гн�ч �.�!  ��r: `i.!  ,   �,-ь.J.ь:. !'vY,`r.J'��.'   �  5ь   '_Э А�.   ^�.  й�        .
.-.1                                                                                                           1"1у„   . $  �,_-..   '' й5 ,      �+     .tih'•`�if; .ь=.�:::,�"   .г'   ..�-`,. „   *Р: ' °lti...;       �. ...
�/                                                                                    i                �л79�iд"У'',               г fi,-,.д,i:,,,�`�'          мь�,д:.•.       ц,-;.•  х'j •R=      ��          �_'�'у�','�,?�`к,-.._�.::1'_.�:
Т."     �             �{          �ц9_        ;r,ry'".,.   #    r,�w. �,,ч�,`  .`.4Э;, ,.� 'ь�У;.�'   wi.                       �:�_l+д,4�..   .   S
г                        +-      t::        т г ,а 1{.                 �'           ..- ь ',{   w,�   ь''., ,r,,:,� �-....   � }.;r�                  4,  л � �;   . ,,� �;t`�? �'�• .•�
L'ч                                                !     �!     •t                    г' t:��                               � �i.��. �   У.`:i         У,   с+. •   •А'¢С�' ..lГ, 4          ��         м             М
р                                                               �                           ;r'kfi°г`'�        `''�'ti°-.t�,.,   ,.у��     .,Эыi_!�  �т   �,'  i.    ,t                 _.7';�г'с:'..'��r"3i_.          "�•;•:
�   ~   :11   т,   t .       �   F.    ry'.л           Jy`'        '��`v.Y.    �.n        . ;       '�*5�   �,; .' .,,  .J;�              -. �'              г   - У:
��                             �+  �:'i`r•С'!..-.   �5.:��',в``�гл-,^   ,.'йг;:,'•'t°`.-,   л'"'ко;.•   : =гу,и4           г     _ .'s:t'!          а �5-:,-.
�                                                         _�.                 -        у'�'�,{  '�w�.3'•��;.nr   ч�  'ч У��'i.,ь'� :tА,  г:".     'iУ�''   :�`t�',�'%1'•i1:i•       _      :;ь.,:д1':У� г'. i.q;,
у�  .!'I ,"}ум-          }n7,'.�,. �G:.   �." р�         :.,��� rJ  Э.�'�:i.;       г}х.=:   'э        _   �.;й       :+t.-`.�рΡ�  ~    '�
_                                      '1t  +'.,                 б    ; L.          �             t�..   "rv';     '                     _           '���'  3}�lr,�
'jй     е�,'в'мi"у'�..   ,   �..'г.Ъt,н,. �     ,7'°!';   '       1       �т...
�                      �у�.', ''~i• �"jr.'                   -           � 5i•°'             t•�                              ��         `Э'•У.�`.-    "
У`''         '+4    :,'�;             г!.°.,',у      i�": н'•   �,      •i�.  .: �  I5    i ' ��   .:        .        �,  г      �-.`  �'-уr�г.д.
Wвтв                                                                          .�         < yw.EуΡ�            ,� � �ar,'л      �  �;           �,,,^.�.                         ^�       х`'
,L                                           'У           :!.                           4.;R  1,�.°ro=?             S'���в      ;'!1. ,SlУ'4+�.,F�i•.'д.'"п  -  У�4'г: �'�.�      ..1, '�.;.,.�":f`�'ц�,��. �.г..-.:._  _
r�iCit�.     '�i•�� +ч�У''            �д���^�е:'r.%i1\i    F� уΡл'� l           ь�,-�        �w.'.ьд;�          д`'•. ;   .,t:   'k•`Х!'�ь'/..w.}Чг.
Е-�                                                       '��и.       �       уΡв    :,ы к�'?.�'�'''   es.9r.  1iд', г�•t,�'�.'�� rИ�r`�`'�-      -��5: '.:':                          �;.,;.�,� ,;Fё г,,.   г�=т.;ь.*,.г_
'�.           ^У  �4  yne   +.              �  �'.     ^;ц.+л..� ..ja:�•          •    °  .ь��.',                          �t.���„     �го.,;
i'            f              �7' ;   �,�у}Х е �  ��.i;     .4��,ун.,'цz¢''`   _   :ь       `',♦�  •,ч'_�'".;' �з.-''"           �:�':�С: ..��`;{'.�'i:,�.-..
��!�  вд��..лС%, iMhi+�Wl•          �'1т"� -w:,�,��   л�'i-�      .,<�..         - '        1.            �      "            .           "
•�.                         _                          i                          _j,,�       - УЧ:.�.   _   "    = �   '`�й   �   R���         ��4;�;'ь  ' �' Si�;.ь
'�           ';          � R7''.,"'�, �; k'•'^'�,",�.,,� � 1т.ч�..           т                      :r,'г, ig; •  t ;.•.>�, . - 4ti+," I`о .   �iг�   - ы^�'-   :. � •:�
Qt.��   �у,,,.г   -Г,   :'гь�(,UH✓�г.д     !•�а�'  �           д'... _                     ,у.,      •                        •i-      ,х..,п'�ууУ   {;.i,l...t,.•
г           �рΡ    �p    г_   ,i   �';�        "'r   r.+r�     '''":(,.�   °t�{       -        -      -   'гiyi;l!^     .:,✓'�лгvй�•'   -�i._.:  '('�.;,ix'•°Fwz
��          9�'                   -           `.?йг�' l�y  ��'  =iи,.дг..,,�:,д`-�                   ';,� _ .   .,.   .�гс^   .:��'; 'г�'�-,.рΡ `ц�,-.и��   • г'�-� :;L";.- 1.
'1р-     ,��q        "�   ��.s��'у%ti:_�''�"*,   х�.    'i•4^�г'•'   i-,f_' �'д'.ч.        _           ,+i,          'iA_  t�tN.i�   :,:�:чт�4�i...ц..�:.    -
®                        t�уΡ                       -          !6'                                      ч,                                     `                    ,,s^,':�        к`   ,�   .У3,£,°г.: ° .г
❑                                                                     %��'             ••�Т.•д•   г       .Чу"и"`               -     ,`1•у       _            "         -�' .              .
}М               .t-      ,Р   ,�г  ��'� .'"з':i�...F:,;;$;�н��,^;�     F-  `�::.   -          -         rf�,       :.Йti:�f�F;t..f� 1�п.атм;�   °:"�''_.ачСг5�1�, 7wи�,л
;�,I��,АΡ��'�           ,д   �,�}' -д•�`µт7:�,   ^�     :,►�,г�ы;;�"  .s,+;   -                                &•у�.•,...�3 г    -'t.   ;гS.+ич~i.r4?-�.
�.'               �ц �      ,У               5'дгN  ssr,e  ..У..  „у'..   ,,�,,   -     ''м.   _              -;'iF,.  _�:�   `.!"'�УΡ у�'      '�i�i_  �ь           ,fi.
�f  �у               ,:и':t�"`ыУ                                     _                                    �.�''��^i.д,,  •.t".�`Р1   .{:.,..,-S.I.' �';.     ,�';';,�:,.-:
.чл�г.   r7'i:''         '.;.�гн,'�'с-'•Sf.":'�"^'�-;`�   Г        _        -        .,1         �уΡ,7      ��'��3  f_.                      д,�.   ,
�Ь�г.Гт..  ;ч�:.  '         •��,.   •.-.�..  л:;:.l�.:�f,'г�   -'.г<,;.�'_      '   �р             .:l,114                .4..   з�.,   ц.   ,    �
-�З                      � 1'г  и     ч                                                     ; � �.,   •�+�в•,�•�  ..               ь       чк'   ' i,�Rt:i'•i. .   :
1  _        £ и`�'((..   ,��F.  r.?t�r,i.��` � �"�{ �',>.       .�:'.�::�r�        �  •          _      ьw�;'�..�3л,'             .1'� �   :f;�`   ��� А`�    �� �'F'Г�
-       '��gуΡ�г'�'�.�.�o-r�:i;         ,..�г;•��г�/,,и,ff:г...г.,.,   -`t^  !�',уΡ+                        �к�"•'Ч7+`,���?�'vC+i�        ?�           __,j�''=$'"?{.�`р�.;(�у;у.ё'   '�
У.L,iрΡ�..-t`,�•'н..  _   «'л�;�..�:!'п'`�р}��  i'                -                  --                 �'д.S7          �'�х"Г             -             ?f�'.i.
ivп �f'�с'S'л.   р1�        �.ц"д,  ..Р_•.��Ъ �ь  .'.�,      -.:L':И.�гУ"и.t. `      �Г�Г.ь^'S�^   fпд�7'1,   S:                 `У°   -.l�=  ьi��i-�=в�ь��,  ��1�`!
д�   .!' ,� i.A   •. Т}   �...{„. уΡвh"�`'-.` . � л   r1. �,9,\ г�    .   �н^!,                          -       Г й' .    _ У! �,    i'�.!       i:, �      _^'ч( .   -.а. 9
r�~      Г   '1"-   !'�Гь !е   :г!$�,у.. и '•ii                             _      .                                                  •'�К        ♦   yS.�    �.у'.       1'}`5l;ь�'�:�..
��д   .�t   .,"&i,r,��уΡ{''a.p�k•   ;r  t`,,�,£,              •.�.,;:;,f.-;s'   -                                   JC              h5�  .'.i�г?aF�`'�ч,            :Yf:,'
��4 -�{' .   .Ъ.  'А     ,:i     �  'Й   ''%        ".                                                           �1�   '.�+,�                       с                    ..!^i '�
�Е!�     У'Т•   5             {'                                           -                              ).      �,    �rJiY'1�{           .            'FP.                `,�
�       t у      ;,чZ�   иr "г� fir �.:�: `ч :5::                    _      �z'.,,                                        =л1С�.t'r  -
�'г.l�      �'�r���}.'�f.           �+г°                                                                                     ,��                                                                  �•Г'.'                   ,.
N.RFli   �           .вl     •'i•  i�'   .   .   �'r'i      .    � .   ^У.,  ь   .�`4:      �     � �..                                      чF�� !:.•.•:
..    .�  .,  ь!г'"а. �  ��FMi�     .�.�., �У 1�°��3--   �F! .   :'{,iц�Fi"�i   .  , .   .      .  .         н1:,,     _   - йt н   ..�  ' i   �'�   .  г     . .   �  '              ��
.              '""^,•'    '.1�:"а��.1й;�г*,'�.'..�•   �.�"i ., ,ч          :tИ