Human Resources Development and Operations Policy
Working Papers
Social Security Issues
And Elements of Reform
Nguyen X. Nguyen
May 1994
HROWP 28



Papers in this series are not formal publications of the World Bank. They present preliminary and unpolished results of analysis
that are circulated to encourage discussion and comment; citation and the use of such a paper should take account of its
provisional character. The findings, interpretations, and conclusions expressed in this paper are entirely those of the author(s)
and should not be attnbuted 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.



Social Security Issues
and Elements of Reform
by
Nguyen X. Nguyen



The author works in the Office of the Vice President for Human Resources Development and
Operations Policy. The views represented in this paper are those of the author and are not to
be interpreted as necessarily indicating the position of the Bank or its Executive Board. This
paper would not have materialized had it not benefitted from the constant support and guidance
of George Psacharopoulos. I am also indebted to many World Bank staff members, in
particular, Estelle James, Diane Steele, Johanna Coenen, Chad Leechor, and Jose Sokol for their
comments.



Abstract
This paper documents the main problems many social security systems are currently facing. as
well as various elements for their reform suggested in the literature.
The current problems are classified into inadequate and inequitable protection, economic and
political environments, demographic patterns, labor patterns, administration, other inefficiencies,
financial management, and benefits inflation.
The paper suggests various constraints and general principles of reform; and proposes a menu
of elements of reform that includes fostering private sector and non-governmental organizations;
reducing health care costs, administrative costs, benefit costs; enhancing coverage and financing
(financing short-term and long-term benefits); and improving portfolio management.
The suggestions are primarily intended to generate further discussions on this important topic.
Moreover, recognizing that unique conditions require corresponding remedies, the list of reform
elements enumerated in this paper is meant to be neither exhaustive nor definite. Rather, it is
to suggest a menu of reform options, a frame of reference or, simply, a practical starting point
for policy makers who, in the final analysis, will devise a plan of action adaptable to the
characteristics of the individual economy in question.






Contents
I. INTRODUCTION .......................................... 1
II. DEFINMON OF SOCIAL SECURITY AND SOCIAL SAFETY NET  ... ..... 2
III. CURRENT ISSUES    ..................................                                                   3
1. Inadequate and Inequitable Protection   .........................  3
2. Economic and Political Environments  .........................  4
3.  Demographic Pattems ...................................  5
4. Labor Patterns  .......................................  5
5. Administration   .......................................  6
6.  Other Inefficiencies  ....................................   6
7. Financial Management  ..................................   7
8. Benefits Inflation  ......................................   8
IV. SOME ELEMENTS OF REFORM  .......                                 ......................                  8
1. Constraints and General Principles of Reform   ................... .                             9
2. Private Sector and Non-Governmental Organizations                            .        .         10
3. Health Care Costs ..................................... 10
4. Administrative Costs                                      ..11
5. Benefit Costs  .................                          ..................  12
6. Coverage and Financing                     .................... 13
a. Financing short-term  benefits.    ....................   14
b. Financing long-term  benefits ........................... 14
7. Portfolio Management ..................................  16
V.  SOCLkL SECURITY  AND  POVERTY  REDUCTION  ................... 17
BIBLIOGRAPHY                           .......................... 19






I. INTRODUCTION
Over one billion people around the world suffer from acute and chronic deprivation every day for various
reasons, including the loss of the family's income. The loss could be permanent as in the cases of lasting
injury or old-age, or it could be temporary such as unemployment, or short-term disability including
maternity. For people in the developing countries, this disruption of income could yield grave
consequences, given the existing low standards of living (World Bank, 1986, 1990; Cornia et al., 1987;
Dreze and Sen, 1989; UNICEF, 1989). For people in the developed countries, however, the loss,
although also common, is relatively less unpleasant (Danziger and Weinberg, 1986; Sawhill, 1988;
Atkinson, 1989).
Social security programs account for an important part of that difference. In most developed countries,
governments provide or support pensions, unemployment benefits, family income support, education,
health services, and facilities for the disabled (Barr, 1987; Atkinson, 1989). The level and extent of
coverage may vary across economies, but, in general the society attempts to help the least fortunate.
Public social security systems, common in the developed world and the Latin American and the Caribbean
region, are rare in many developing countries, especially those in Africa, the Middle-East, and Asia.'
Unemployment and pension benefits cover only a small portion of the populace, mainly, the workers in
the formal sector (Midgley, 1984; McGreevey, 1990; Mesa-Lago, 1991a). State support for health care
is weak, with resources disproportionately directed toward curative rather than preventive care (Halstead
et al., 1985; Cadwell, 1986; World Bank, 1993d). State support for education is not much different
either, with varying degrees of educational efficiency (Schultz, 1988; Psacharopoulos, 1993;
Psacharopoulos and Nguyen, 1991).
Grave incidence of poverty and deprivation, a low degree of development of formal social security
systems, resource constraints, inadequate institutions to provide social security benefits, the fragile
economic conditions which characterize economies during transition or adjustment, and the powerlessness
of the most vulnerable populace make the issue of providing adequate social security in the developing
countries both urgent and difficult.2
Coverage is, however, only one side of the problem. Many social security systems are burdened with
heavy financial difficulties (McGreevey, 1990). In particular, countries with relatively more mature
social security systems are facing greater financial challenges (Mesa-Lago, 199 lb; Lloyd-Sherlock, 1992).
This is particularly burdensome on developing countries already short on resources, or on economies in
the midst of adjustment or transformation. Even in most industrialized countries, there is concern that
the present social programs cannot be maintained at their current levels but must be reformed.
Social security commands a large proportion of public sector revenues and expenditures in many countries
(Grosh, 1990). The sheer size of these systems implies two major consequences for public finance.
First, the reserve funds are an attractive source for financing government deficits rather than being
invested to further capital accumulation. Second, to the extent that the funds are mismanaged, financial
insolvency and crisis will follow. Many countries already find themselves in this muddle, and have
resorted to increasing taxes, sometimes to excessive levels. In addition, an aging population, slow
economic growth, and benefits inflation are all pushing many systems toward bankruptcy, thereby
contributing to the country's fiscal burden and macroeconomic instability.
'In these regions, the old have long been cared for by the extended families, community and other informal
arrangements. As will be presented later, with the unavoidable demographic, socio and economic shifts, the
challenge these countries face is how to efficiently involve more the formal systems of social insurance and
assistance without accelerating the decline of the informal systems.
2In brief, social security in the developing countries is limited due to both supply and demand factors. Resource
constraints and lack of effective administrative institutions restrict the supply side. Effective demand is limited by
the absence of political clout of the populace most in need of coverage.



2
The need for system reform is, consequently, ever more pressing. The principal objectives of a reform
program are twofold: increase coverage and ensure solvency. The elements of reform, however, would
depend on the specific conditions of each country. As a general rule of thumb, economies with limited
formal social security systems (as found in many countries in Africa, Asia) would focus on strengthening
coverage. Those with more mature systems (as found in the developed and the Latin American and the
Caribbean regions) would highlight the solvency issue. Many formerly socialist economies in transition,
unfortunately, have the daunting task of confronting both problems.
This paper defines social security, identifies the main problems currently faced by many social security
systems, and documents various elements of reform suggested in the literature. The suggestions are
primarily intended to generate further discussions on this important topic. Moreover, recognizing that
unique conditions require corresponding remedies, the list of reform elements enumerated in this paper
is meant to be neither exhaustive nor definite. Rather, it is to suggest a menu of reform options, a frame
of reference or, simply, a practical starting point for policy makers who, in the final analysis, will devise
a plan of action adaptable to the characteristics of the individual economy in question.
II. DEFINITION OF SOCIAL SECURITY AND SOCIAL SAFETY NET
In its broadest sense, social security is the response, by social means, to an aspiration for security, for
confidence that one's quality of life would not be totally eroded by any social or economic eventuality.3'4
More practically, social security means the protection which society provides for its members against
economic and social distress caused by loss or substantial reduction of earnings resulting from illness,
maternity, injury, unemployment, old age, and death.5
Defining social security by objectives (as done above) is broader than the common definition by
instruments whereby social security is described in terms of specific public programs involving social
assistance, social insurance and redistribution (Kotlikoff, 1987; Atkinson, 1989).6.7
For practical purposes, however, a working definition which identifies the main components of a social
security system would be more convenient. A typical social security program provides five types of
benefits (Puffert, 1988). These benefits, in order of most common usage, are:
1.     Work-injury benefits (earnings replacement and medical expenses).
2.     Pensions for old-age, disability, and death of the family's main income earner.
'The term 'social security' should be clarified because of its ambiguity in usage. In the U.S., it means
retirement pensions; in the United Kingdom, it refers to the entire cash benefit system; and in mainland Europe it
refers to all cash benefits plus health care in accordance with the usage of the International Labour Organisation
(ILO, 1984b). This paper adopts the ILO usage. However, the largest component of most social security systems
is pensions.
'By restricting to social means, one is mainly concerned with measures which include redistribution, direct
intervention, and possibly alterations in market functioning.  These measures, however, are not the sole
responsibility of the state; they can also be rendered by the community or the family.
'In essence, the system is a broad-scale risk sharing arrangement among members of a community (society).
'Several objectives of social security include: to prevent deprivation and adversity (Burgess and Stern, 1989),
to insure against social risks, to provide a social safety net, to achieve better targeting of limited benefits to those
most in need, and to achieve reforms within the context of a more limited role for the state (William McGreevey,
May 2, 1990, Office Memorandum, the World Bank).
7Ihere are sound arguments for adopting the definition by objectives (Burgess and Stern, 1989). First, many
public instruments, so familiar to the developed countries, do not exist in the developing world; therefore, a
definition based on these instruments would be too narrow. Second, focusing on public programs would undermine
the role of community and family support. Third, even within the sphere of public action, the instruments of the
developed economies might not be applicable to the developing economies.



3
3.     Sickness and maternity cash benefits, and medical care.
These are considered social insurance benefits. Social assistance benefits include:
4.     Family allowances -  benefits for raising children -  which are generally available in the
industrialized countries, countries in Africa with a French colonial heritage, and much of South
America.
5.     Partial earnings replacement during short spells of unemployment is common in the industrialized
countries but not elsewhere.
In conjunction with providing social security, which is more or less a formal and permanent arrangement,
many countries establish temporary proactive employment and income generating activities (transfers,
retraining, public works) as part of their social safety nets. These programs are intended to provide relief
to the vulnerable populations following major shocks (such as transformation of a centrally planned
economy into a market-based economy, major macroeconomic adjustments, natural disasters, etc.). The
importance of the social safety net topic deserves a separate, hence, more elaborate treatment.' Most
of the components of a social safety net, nevertheless, can also be dealt with within the domain of social
security.
HI. CURRENT ISSUES
1. Inadequate and Inequitable Protection
Despite remarkable progress in expanding coverage in the past decades, the existing systems still exclude
many poor people, not to mention that these systems sometimes serve the more affluent at the expense
'The term 'social safety net' has been used in different ways, from describing a comprehensive model to a
minimal and temporary system.
The minimal and temporary model has been advanced in response to the plight of the poor affected by structural
adjustment (UNICEF, 1987), or by the transition toward a market economy (Development Committee Meeting,
1993). This model calls for maintaining and targeting social services while cutting other government expenditures
over the adjustment period. Sectoral policies (such as retraining and income generating activities) to protect the poor
and unemployed would be encouraged. Chile, Ghana, and Kenya created job programs to provide work and pay
for many poor people who were laid off in the course of adjustment efforts. Specifically, countries are advised to
direct nutrition, basic medical care, and primary education programs to the most vulnerable populace.
At the other end, the comprehensive model comprises a set of measures to help protect vulnerable groups such as
workers at risk of unemployment, the unemployed, pensioners, single parent families, children, etc. The system
would include: 1) income support such as unemployment benefits, family allowances, and pensions; 2) employment
services; 3) proactive employment and income generating activities such as training and small business assistance;
and 4) funding and provision of health care (Ralph Harbison, April 18, 1990, Office Memorandum, the World
Bank; World Bank, 1990; World Bank, 1993b).
In between the two extreme definitions, the term can take many forms. For instance, according to India's Fifth
Five-Year Plan (1980), the social safety net, called the Minimum Needs Program, guarantees a core of 12 essential
social services which include primary and adult education, rural health, rural water supply and sanitation, rural
roads, rural electrification and housing, domestic cooking energy, urban slum improvement, nutrition and food
security for poor families, employment and income-generating programs.
Although the term has been used in different ways, there is general agreement about its main components: income
support, employment services such as job opening information and job counseling, programs to enhance labor
quality such as retraining, direct employment such as public works, and health services (Barr, 1992). This
definition overlaps a great deal with the LO's definition of social security used in this paper.
In fact, the relationship between the usage of the terms 'social security' and 'social safety net in this paper is:
Social Security = Pensions + Social Safety Net - Proactive employment and income generating activities.



4
of the poorest.9"0
In Asia, the Middle East and North Africa, and the Sub-Saharan Africa, for instance, substantial gaps
in social security coverage remain. For the aged and vulnerable populations, the support of the traditional
family and the immediate community remains the principal means for filling in the void of a broader-scale
risk pooling arrangement. Industrialization and modernization, which induce migration and weakening
of the system of extended family support, has put increasing strain on the traditional system of support.
In India, for example, the formal sector employees (mostly public servants) are covered by sickness,
maternity, pension and survivor benefits, and even, subsidized housing. Workers in the unorganized
sector, who constitute 90 percent of earners, do not have any social insurance or social assistance (Guhan,
1988). The state-level pensions, designed for the aged below the poverty line, cover probably only one-
fifth of that population.
Even in economies with more advanced social security systems, such as those in Latin America,
inadequate and inequitable protection remains (McGreevey, 1990; Mesa-Lago, 1991a). In many areas,
poor people have limited or no access to social insurance, risk sharing, and health care services.11'12
Only about two-thirds of the population is eligible for coverage. Health care benefits tend to cover the
middle-income groups, and pensions are most available to those with higher income (World Bank,
1993b).
2. Economnic and Political Environments
Economic and political conditions carry grave consequences on the social security systems. By the end
of the 1980s, the world economy was in a recession. The negative impacts on many developing countries
were severe, prompting governments to undertake strong measures to alleviate their budgetary difficulties.
In addition, profound changes in Central and Eastern Europe and in the former Soviet Union compelled
these countries to adapt their existing programs to new circumstances.
The slowdown in economic growth shed doubt about the social security systems being able to sustain
long-term financial health and undertake benefit improvements. The recession results in higher numbers
of unemployed persons, fewer contributors to the programs, and in increasing stress as formerly
employed contributors are drawing benefits from social security. Many economies found their hands tied
in the wake of this budget imbalance. They fear that increases in taxes and social security contributions
would not only be unpopular, but would also lead to economic distortions such as evasion from taxes,
and discouragement from work and saving.
'he payroll tax financing mechanism, in the long run, shifts the tax incidence to consumers (this assumes that
labor supply is inelastic), many of whom are poor. Therefore, the system (characterized as partial coverage because
it generally covers solely the modem sector workers) results in transferring some resources from the poor to the
middle class.
'oMany current systems fail to cover situations of dependency when marriages break up, or situations which
arise independently of marriage (ILO, 1984a). Upon divorce, the woman may no longer share in her husband's
pension benefits. Moreover, a non-working woman with dependent children who cohabits with a man may have
no rights to benefits.
"In 1985, for the two poorest groups in Brazil, which accounted for 41 percent of the population, only 18
percent of the heads of households were social security affiliates. These two groups only received 3 percent of tho
social security benefits and 18 percent of benefits paid out by all social programs (McGreevey, 1990; and World
Bank, 1993c).
'"Health care services offered to rural workers and their families in Brazil, Ecuador, and Mexico, which began
in the 1970s, are the exceptions (Mesa-Lago, 1991). Few programs provide services to workers in the growing
informal sector (World Bank, 1993c).



5
In brief, the financial viability of social security institutions in many developing countries is being
challenged. Real revenues have declined due to the drop in real wages, open unemployment, expansion
of the informal sector, evasion and delay of payment, and growth of public debt. On the other hand,
expenditures rose because of accelerating medical costs, increased salaries and benefits, and growing
welfare and unemployment outlays.
In the 1980s, almost every Latin American country suffered economic crisis (Lee, 1992). For many, the
1980s is a lost decade.13 In Panama, the social security funds turned negative, creating a large portion
of the budget deficit in the early 1980s. The deficit was 12 percent of GDP by 1989, a contrasting
situation to 1982, when social security savings financed nearly one-fifth of public capital spending. In
Uruguay, in 1982, the social security system, responsible for 67 percent of the public deficit which
reached 18 percent of GDP, was a major factor leading to the country's macroeconomic collapse later
that year.
The situation is particularly alarming in the formally socialist economies in the midst of macroeconomic
imbalances during the transition. The universal social security approaches (financed mostly by transfers
from state-owned enterprises, SOEs) are quickly becoming inappropriate in a market-oriented
environment. On the one hand, revenues are short due to increasing bankruptcy and privatization of the
SOEs. On the other hand, the systems have to cope with mass unemployment, sharp increases in the cost
of living, moral hazard (increased recourse to early retirement and disability benefits), and the lack of
administrative capacity to provide assistance to the needy (many programs are run by state-owned
enterprises).
3. Demographic Patterns
The aging population, driven by declining birth rates and rising life-expectancy, is an unavoidable
pressure to social security. Pension is the area most affected by demographic shifts. In the developed
countries, the number of persons aged 65 or older will increase far faster than other age groups in
proportion to total population (OECD, 1988). In the European Economic Community, one half of the
members are experiencing zero or negative population growth. In many developing countries as well,
life expectancy at birth is now about 75 years.
In addition, workers (at least in the developed countries at the present) seem to retire earlier (Bureau of
the Census, 1989). Three reasons have been suggested to explain this trend: 1) a lowering of the
retirement age, 2) technological progress that gives younger workers a comparative advantage, and 3)
pension policies that make it economically feasible to retire early.
In the long run, an aging population, will not only tilt the benefit-ontribution balance, it will increase
overall expenditures because per capita public social spending is much higher for the elderly than for the
young. For example, in the OECD countries, per capita public spending is two to four times higher for
the elderly than for the young (Maguire, 1987). This is particularly evident for health care (ILO, 1987).
4. Labor Patterns
High unemployment is projected to persist, especially in the industrial countries."4 Unemployment tends
to increase outlays for social security while reducing its revenue base.
Most developed countries have some form of early pension program intended to encourage laid-off
workers to retire early in order to make room for the younger ones. This job replacement effect has been
'3In Mexico, for example, industrial real wages in the beginning of the 1990s are only 60 percent of what they
were ten years ago.
"M'he OECD projected that the average OECD unemployment rate will remain above 8 percent for a while
(OECD Observer, 1988).



6
lower than anticipated because of the lack of skills in the younger workers that did not have the benefit
of steady employment (Haanes-Olsen, 1989). This group, called the "new poor," is an emerging
troublesome trend in the developed countries. They are likely to be net users of rather than net
contributors to social security. Another cause of high unemployment (particularly in Western Europe)
is the lack of labor mobility partly due to unions and business regulations."I
In many formerly socialist economies, early retirement is posing a threat to the financial viability of
pension systems.
The situation in developing countries, although different in nature, is also grave. Rapid population
growth in the past decades imposes an increasing demand for jobs; some additional 38 million persons
will seek employment in developing countries throughout the 1990s every year.  Since formal
employment creation, in many countries, is expected to fall short of the rise in the labor force, the
resulting open unemployment, underemployment, and employment in the informal sector will severely
limit efforts to broaden social security coverage in these countries.
5. Administration
Inadequate and inefficient administration, in addition to financial constraint and an unsubstantial cash
economy, has hindered the broadening of social security coverage in the developing countries.
Consequently, the focus, understandably, has been to improve management techniques, data processing,
financing, and funds investment.
Administrative costs are high for many countries. In half of the Latin American countries, for instance,
they range from 10 percent to 32 percent of total expenditures (McGreevey, 1990; Mesa-Lago, 199 la)."
6. Other Inefriciencies
Other inefficiencies include: excessive personnel, personnel with lack of administrative experience, high
remuneration,'7 overlapping administrative institutions, poor information and accounting, complex legal
system, incentives to retire early, and inefficient health care management and provision. The last two
deserve some elaboration.
The benefit structure of retirement nurtures some major inefficiencies. First, pension schemes in many
countries encourage early retirement (often when the worker is at his/her highest productive years) by
providing rich benefits (sometimes 100 percent replacement rate) for time of service regardless of age."
Second, high pension benefits usually go to the high income groups even though contributions are capped
and fall short of benefit expenses.
The provision of health care is both inefficient and inequitable in many cases. Many countries have dual
systems of health care, resulting in duplication of services and bias toward high cost medicine. Many
social security systems focus mostly on high-cost medicine which benefits only a few while leaving the
provision of basic health services to the poorly financed ministries of health. In addition, cost sharing
arrangements with the Ministry of Health to provide services to non-contributing people have been costly
1"With the possible exception of the Nordic countries of Denmark, Iceland, Finland, Norway, and Sweden which
have a common labor market, in some European countries, a number ofjobs are not available to nonunion workers.
Many countries have regulations that restrict business operation (such as regulated business hours).
'>This compares to three to four percent in the institutes in the U.S. and Canada.
'7In Panama, for example, salaries, the main component of expenditures, account for more than 21 percent of
total outlays.
"For example, Brazil offered retirement benefits based on time of service regardless of age. Consequently,
some workers retire before age 45 (McGreevey, 1990).



7
to many social security funds."9
The social security system does not have a strong built-in mechanism for health care cost containment,
There are no effective restraints on demand for health care services, and minimal incentives for providers
to moderate their supply of services.
7. Financial Management
The financing mechanism of many social security systems contributes to their revenue shortfalls. In Latin
America for instance, the insured do not finance more than one-third of benefit costs, leaving the rest to
the employers and the state (investment yields are rninimal as will be explained later). Employers'
evasion and payment delays are widespread; in some countries, they range from 23 to 60 percent. The.
state becomes the principal debtor.'
Adding to the problem, pension funds are used to finance the budget deficit (by being invested in public
sector debts yielding low or even negative returns). These funds are also invested in projects with
questionable rates of return.21 This practice, coupled with high inflation, erodes the capital base of the
funds.' It is fair to recognize that the absence of efficiently functioning financial markets presents a
severe constraint on better investment decisions. Pension funds are also used to cross-subsidize health
care and other non-contributory systems (for instance, child allowances).
The pay-as-you-go method of financing commonly employed in many systems separates contributions
from benefits, hence creates incentives to evade contributions. This is especially acute in the Latin
American countries. Most of them began with fully funded systems with initial surpluses in the trust
funds (Jimenez and Puffert, 1988). The funds eroded over time mainly due to forced holdings of
'9n Panarna, for instance, this cross subsidization has contributed to the fund's deficit. By the end of the 1980s,
despite a mandatory contribution rate of 18 percent of wages, social security accounts for half of the budget deficit
which reached 12 percent of GDP (Lee; 1992).
This poses, yet, another inequity in the economy; some resources are taken from the poor (in terms of general
taxes) and transferred (via the state's paying the social security deficit) to the insured who are generally urban
workers.
2,In Latin America, pensions funds have been invested in: 1) loans to the governments, often non-negotiable
and not indexed to inflation (which are practically forced loans to cover the budget deficits); 2) personal and
mortgage loans (generally not indexed) for the insured people; 3) loans to the maternity-sickness programs to
develop their infrastructure and to cover their deficits; 4) administrative building and housing projects (with very
low returns); and 5) some low-profit commercial projects such as stores with state subsidized prices, cinemas,
sports, etc. In the non-Latin Caribbean, most of the funds are invested in government stocks, bonds, or treasury
bills (Mesa-Lago, 199 lb). Consequently, in eight countries documented by Mesa-Lago, only three countries reaped
positive returns in 1980-87: Chile (13.8 percent), Bahamas (2.7 percent), and Barbados (0.7 percent).
Funds are also used to finance government budget deficits in other parts of the world as well (as in the U.S.). In
Malaysia, the Employees Provident Fund was required to invest at least 70 percent of its surplus funds in
government securities (94 percent in practice in 1975). In Singapore, on the contrary, the Central Provident Fund
was used to develop the housing infrastructure.
'1n Latin America, six factors responsible for low yields are (Mesa-Lago, 1991b): 1) inflation, 2) lack of
portfolio diversification and concentration in government and other low yield instruments, 3) investment in short-
and medium-term (instead of long-term) instruments, 4) high proportions of reserves in net-current assets (often
unpaid state obligations), 5) government intrusion (in Barbados and Jamaica, for instance, the Ministry of Finance
controls investment decisions), and 6) age of the program (the more mature the program, the more benefit demands
are placed upon it - refer to Lloyd-Sherlock, 1992). It should be mentioned had the funds been invested in long-
term instruments, the capital base would have been depleted even more due to high inflation unless the real yield
is higher for the longer-term instruments than it is for the others.



8
government bonds, inappropriate investments, and overly rich benefits.2  Consequently, the schemes
are now mostly pay-as-you-go (except, to some extent for Chile and, potentially, Argentina, Colombia,
and Peru). With increasing expenditures, many systems had to raise taxes or rely on general revenues.2'
8. Benefits Inflation
While the contributing side is being depleted, the benefits side keeps expanding, both in terms of the
richness of the benefit package (driven by political motives) and the number of beneficiaries (driven by
economic downturn, labor and demographic patterns).'
Benefits such as partial disability pensions, early retirement pensions, length-of-service pensions, sick
pay, and additional health care benefits have continuously risen in many countries, creating increasing
resource strain on social security systems.
In many countries, the dual problem of benefits that are indexed to inflation or wage increase, and
revenues that are not indexed, creates tremendous fiscal hardship in an inflationary environment (Mesa-
Lago, 1991a; McGreevey, 1990).
IV. SOME ELEMENTS OF REFORM
The numerous problems with social security have induced many countries to attempt system reform. The
reform programs are designed to increase the solvency, efficiency, and equity of the systems, while
expanding coverage to the vulnerable population, often within tight budget constraints. In short, the
reforms call for cost effectiveness (i.e., reducing expenditures in many cases) while providing wider and
better protection.
Addressing social security reform is one of the most politically difficult undertakings. Yet it is required,
for many countries, to stabilize the budget deficit and to restore the long-term viability of their social
security systems. Moreover, many reforms are to be carried out while the economy is undergoing
various austerity policies. The timing, however, does seem to cooperate; the world economy is projected
to come out of the slow growth pattern by the middle of this decade, holding out better prospects for the
developing countries (World Bank, 1993a).'
It should be emphasized that not every social security system is in financial trouble. In fact, the more
recent the system, the more likely it is to generate a surplus because benefit outlays have not caught up
with revenues (Mesa-Lago, 199 lb; Lloyd-Sherlock, 1992). However, the basic structural problems are
there (the demographic shifts in particular).' Without a constant effort to contain these problems,
'Historically, Argentina, Uruguay, and Brazil began as fully funded pension systems that became pay-as-you-go
mainly because of the depletion of their capital reserves.
2'With the combined tax rates (from workers, employers, and state contributions) in excess of 30 percent in
many countries, further recourse to a higher payroll tax may not be desirable. Moreover, tax evasion reduces the
tax base, and adds to the revenue shortfalls.
'In Panama, the number of beneficiaries to contributors rose by one between 1981 and 1984, and will probably
double in two decades. On the other hand, the benefit level is unsustainable; many retirees receive 100 percent
replacement of their last salary (Lee, 1993).
2"The World Bank projected that the prospects for the developing countries hold out the promise of higher
growth rates for the remainder of the 1990s. Real GDP per capita, mostly stagnant through the 1980s, is to rise
to about 2 percent after 1996 (The World Bank, 1993d). Growth rates vary by region, however, with East Asia
and the Pacific leading the pact and Sub-Saharan Africa at the bottom.
'An often overlooked structural problem is the shift in the composition of production in a country. It can be
shown that if the real rate of interest exceeds significantly the growth rate of real wages, the system does not require
a high contribution rate or a high active worklife ratio (ratio of working life to retirement life) (Vittas, 1993).



9
sooner or later, the system will have to face the challenge of reform discussed in the following.2'
1. Constraints and General Principles of Reform
1.     Stabilize the economy and keep it on the growth path. As the economy develops, the formal
sector is expanded, producing two beneficial consequences: 1) access to social insurance increases, 2)
the tax base, therefore revenue, increases. The rate of economic growth will determine the level of
security the society can effectively provide to its citizens.
As a paradox, however, growth sows the seeds of future financial difficulty for a number of reasons.
With growth, fertility decreases, life expectancy increases, and the population ages more quickly
(assuming constant immigration). Health care and other benefits with high income elasticity of demand
will push up outlays. In addition, the real rate of interest tends to fall (as capital is becoming less
scarce), the growth rate in real wages tends to accelerate; these factors will make it more difficult to
provide a high pension rate unless the contribution rate and the retirement age are raised (Vittas, 1993).
2.     Take into consideration the cultural and social characteristics of each country, because these
elements shape behavior, demand and supply conditions of social services.
For example, it is essential to set up social security systems to reinforce and not replace existing
institutions such as the extended family and community-based arrangements for old age, sickness, and
disability.2 There are four reasons for this: 1) the infeasibility of providing universal formal social
security in most developing countries, 2) stringent fiscal constraints, 3) the need to design effectively
targeted formal security mechanisms that reinforce support for households in the rural and informal
sectors where traditional support mechanisms are still workable, and 4) the fact that the family and the
local community may possess informational advantages in order to deal effectively with the problems of
adverse selection and moral hazard.'
3.     Tailor the social security system to the country's level of economic development and labor market
conditions. This implies that excessive payroll taxes should be avoided in order to promote labor
participation in the formal sector. Moreover, contributions should be low in low-wage, labor-intensive
economies. Benefit levels should also reflect the wealth of the economy, and its relative international
competitiveness.
4.     Focus the reform on eliminating excess consumption and waste of scarce resources. In general,
means-tested transfers should be preferred to open-ended subsidies. However, due to administrative
constraints, the developing countries may consider categorical transfers.
5.     Pursue an equitable distribution of costs and benefits between present and future generations. An
overly rich benefit package would undermine the system's viability (resulting in redistributing toward the
A country with a recent system is likely to be a developing country characterized by scarcity of capital and
abundance of labor. This implies that the real rate of interest is high and wages are low. With time and with
development, the balance tilts, inducing wages to rise faster than the rate of interest, hence, exposing the system
to more financial risks.
'Policy reform, in theory, involves determining welfare augmenting directions relative to the status quo, with
respect to a set of policy options, subject to various constraints posed by public supply and excess demands. In the
following, the important constraints and general principles are identified and elements for reform are described.
'For example, the system should encourage, whenever possible, home-health care rather than institutional care
for the aged and disabled.
3'This implies transferring assistance through these institutions. In turn, it should help with population control
by alleviating the need for raising a large family for security reasons in the old age. In addition, other arrangements
such as charitable, religious, or other organizations should be supported. For a list of issues facing countries
seeking to set up or develop pension funds, consult Davis, 1993.



10
current generation at the expense of future ones). In particular, public pension programs should be made
self-sustaining over long periods of time capable of weathering transient and unavoidable cycles.
6.     Recognize that the credibility and accountability of management is critical to any reform. Without
public trust and support, the system would most likely be marred by inertia (resistance of privileged
groups and the bureaucracy, unwilling politicians) and evasion.
2. Private Sector and Non-Governmental Organizations
7.     Encourage the participation of community and non-governmental organizations in the planning
and running of local health and social welfare services (for instance, to avert hit-and-run practices, cream
skimming, etc.).
8.     Support voluntary private insurance and retirement funds. The government does not need to get
involved in the financing of these institutions, although it should regulate them.3" These schemes could
be substitutes for, or merely supplements to the public systems.'
9.     Increase the role of the private sector in providing social and health care services. The resulting
competition between private and public providers should increase the efficiency in the delivery of services
(a multipi'lar model supported by this paper would make room for the private sector to compete). For
example, private supplementary insurance, on top of the statutory insurance, should be encouraged to
provide a higher level of protection.
10.    Regulate competition among private insurers in order to provide protection against fraud and to
minimize risk avoidance (bias selection or cream skimming) by insurance plans. Other provisions such
as funding rules, ownership of surpluses, portability would also increase the attractiveness of private
insurance (Davis, 1993).
3. Health Care Costs
Controlling health care costs is critical given that health care spending is often the second (behind
pensions) most expensive item of most social security systems. In Latin America, for example, the share
of social security spending allocated to health care is over 60 percent for Colombia, Costa Rica, El
Salvador, Honduras, Mexico, Dominican Republic, and Venezuela in 1983 (McGreevey, 1990).
11.    Redirect resources toward more cost-effective interventions (in most cases, this means
emphasizing primary care and moving away from high-tech medicine), improve preventive care, reduce
administrative costs, and amplify cost-consciousness in the decision making of both demand and supply
agents. These are some of the primary candidates for cost containment.
One option for containing health care costs to the social security institutes, and at the same time,
emphasizing preventive care away from curative care, is to support a 2-pillar system. The first pillar
would guarantee and provide a minimum benefit package that serves the community as a whole. This
package of public health and essential clinical services would include: immunizations, some school-based
health services, programs to reduce alcohol and tobacco consumption, AIDS prevention, prenatal and
delivery care, family planning services, management of the sick child, treatment of tuberculosis, and case
management of sexually transmitted diseases (STDs), etc. (World Bank, 1993d).3 The provision of this
3'The regulation is necessary to correct two major market failures in an insurance market, notably, the adverse
selection (also called biased selection) problem and the moral hazard.
3'he Argentina's Pensions Reform ratified by parliament on 10/01/1993 allows for both options. Chile's
reform in the early 1980s did the same to a large extent.
33The World Development Report 1993 estimated that the per capita annual cost (in 1990 dollars) of the package
is $7.80 in low-income countries, and $14.70 in middle-income countries (World Bank, 1993a).



11
package falls in the realm of public action because it would be inadequate if left to the private sector due
to the presence of externalities.
The second pillar provides consumers the option to purchase supplemental coverage, at their private cost,
in the private markets. This would encourage the private sector's development in this endeavor, facilitate
a multipilar social security system discussed below, and introduce some cost consciousness into the
demand for health services.34
12.    Choose the most suitable form  to structure the health care financing and delivery system
according to the country's historical and social givens. The objective is to make the system simple and
efficient within the constraints of the country.
There are two main polar strategies to organize the health care system: pure competition and regulation.
The former is characterized by private insurers and private providers of care, while the latter, by national
insurance and public providers. The cost containment effect of these models is still unsettled. Most
countries adopt a certain blend of the two models depending on their historical and social conditions.
There are also two polar mechanisms of payment: capitation and fee-for-service. Under capitation, each
beneficiary is enrolled in a provider network (composed of physicians, hospital(s), and sometimes,
laboratories and pharmacies) that receives a fixed payment every period (monthly or bi-weekly) on behalf
of the beneficiary whether or not services are provided. In turn, the network would guarantee the
beneficiary of any necessary health care services should the beneficiary become ill. This method of
payment is argued to be able to contain costs by giving providers incentives to control utilization.
Under fee-for-service, patients can choose their doctors and hospitals, and payments are made only for
services rendered on behalf of the beneficiary. Containing costs under fee-for-service, in the presence
of minimum cost-consciousness of both patients and providers, usually requires 2 elements: regulating
payment (for example, payment to hospitals based on Diagnostic Related Groups, and to physicians based
on a Resource Based Relative Value Scale), and controlling utilization (for example, the Volume
Performance Standards or Global Budgets).
4. Administrative Costs
13.    Unify administrative institutions. Many countries have a dual system of government health care
which results in duplication of services and inequity because of differential access to funds. In Colombia
in 1980, for example, the social security system used 62 percent of all public health spending (the rest
was left to the Ministry of Health) on 18 percent of the population. Brazil and Costa Rica have taken
steps to abolish the barriers between health care under social security and health care provided by the
Ministry of Health. Costa Rica has virtually solved the problem of dual systems by integrating the
delivery of health services of its ministry of health with the social security system.
14.    Standardize claims processing.  Claims processing can be very expensive and inefficient,
especially in a system of multi-payers, each using a different claim system. There are two basic steps
in this task. First, devise a uniform billing form (this could be done in conjunction with insurance
companies and health care providers). Second, in time, introduce it into an electronic data processing
environment. The advances of computer technology provide great promises in this endeavor.
15.    Other elements for reducing administrative costs include: eliminate redundant personnel, cut back
on privileged labor conditions, simplify legal system, improve accounting and managerial efficiency.
This includes establishing a system of unique identification for every contributor and/or beneficiary; and
computerizing registration and payments.
34This multipillar model could be applied beyond health care benefits. The paper supports this model for all
social security benefits.



12
5. Benefit Costs
16.    Avoid promising and granting lavish benefits relative to the wealth of the economy. Benefits can
be phased-in over a long period which parallels the growth of the economy. Indexation of benefits has
merits based on equity grounds, but it may instill further inflation (demand-pull inflation) and harm the
financial condition of the system (especially when wages, hence payroll contributions, are not indexed).
If indexation is politically necessary, then adjust benefits according to salary increases rather than cost-of-
living increases.
17.    Support, whenever feasible, a multipillar system which includes: 1) a minimum subsistence
package provided by the state, 2) a compulsory fully-funded defined contribution system which could
be publicly or privately run, and 3) a voluntary private program to supplement the other two schemes.
This system would minimize the burden on the state, while at the same time, encourage the private sector
to be involved in the provision of social security benefits (Development Committee Meeting, 1993; World
Bank, 1994).
The minimum subsistence benefit package would be designed to assure limited but effective protection
against the risks most costly and most prone to the poor. Coupled with wider coverage (possibly financed
in part by a tax with a broader tax base than that of the payroll tax), this strategy would gear the social
security system toward playing an even more active role in alleviating poverty.
18.    Reduce the incentives for early retirement, hence the potential loss of production, by either: 1)
raising the minimum retirement age;' 2) restricting the length-of-service retirement benefit or, at least,
tying it to age (for example, a person may retire after 30 years of service if he is at least 60 years old);
or 3) setting the replacement rate conditioned on age and length-of-service. In particular, raise the
replacement rate according to increasing ages of retirement, in order to encourage a longer work life.
19.    Because of administrative simplicity, and also, of consistency with the proposed multipillar
pension, consider a flat benefit structure. Under the public fund, everybody receives the same minimum
benefit package regardless of earnings. This system would be progressive given that contributions are
wage-based. The better-offs, however, would still be able to afford supplemental coverage.
Under this structure, people may still have incentives to retire early. Therefore, the conditions for
retirement that link age with minimum length of service have to be well specified. Another drawback
of this system is that it disconnects benefits from contributions, therefore, encourages high income earners
to evade and invest in private pensions instead. One way to circumvent this problem is to require a proof
of enrollment in the public plan as a prerequisite for buying private insurance.
20.    If a flat benefit structure is not politically feasible, retain the earnings-related benefit structure,
but set the replacement rate at an economically feasible level. A low replacement rate would, not only
reduce the incentives to retire early, but also encourage personal savings (to maintain the same level of
income after retirement), and reduce outlays.36
In particular, conforming to a multipillar pension system, set the replacement rate such that it would
provide a subsistence level to the lowest wage earner. Another option would be a vary the replacement
rate by income class, with the highest replacement rate applied to the poorest populace.
21.    Make benefits (which include eligibility, benefit levels, and contribution rates) uniform across
35Argentina's Pensions Reform, ratified by parliament on 10/01/1993, would, over the next nine years, raise
the retirement age from 60 to 65 for men, and from 55 to 60 for women. This is much different from Brazil which
offered retirement benefits for time of service regardless of age. Some workers retired before age 45.
'Argentina, Brazil, Panama, Uruguay, and Costa Rica, as some extreme cases, have replacement rates that
could reach 80 to 100 percent.



13
occupations and economic activities. For example in Panama, special pensions for selected public
employees - military personnel and teachers - created a large deficit because of low contribution rates
and high benefits. In particular, many policy analysts recommend that countries move toward unified
systems of disability benefits based on degree rather than on cause of disability.
22.    Recognize all kinds of couples who share income in determining rights to social security.
Persons, whether married or not, who live together should each have benefits in their own right rather
than be treated as dependents (this is particularly applicable to the publicly provided minimum package).
23.    Devise and enhance the administrative capacity and the legal framework to support an efficient
operation of a multipillar system. This includes simplifying, whenever possible, both the current
administrative and legal structure.
6. Coverage and Financing
Many countries are facing grave financing constraints posed by their social security system's deficits.
On the one hand, raising payroll contributions runs the risk of driving workers to the informal sector and
employers to evade. On the other hand, using other revenue sources undermines the principle that one
should pay for one's own interest.7   Moreover, borrowing breeds inflation or macroeconomic
instability.
Expanding coverage under these conditions is difficult. Access and financing are entwined because,
unless the financing side is stabilized and costs controlled, increasing coverage will only aggravate the
financial problem.
24.    As a general principle of financing the system, it is crucial to make a distinction between the
insurance aspect and the assistance aspect of social security.' Social insurance should rely on wage-
based contributions (one should be pay for one's insurance), while social assistance on general tax
revenue."
25.    Weigh the disadvantages and advantages of alternative means of financing - general taxation or
contribution - against the individual country's condition. Financing by contributions has been contested
"As will be presented in the following, both payroll and general taxes, however, can be used to finance various
aspects of social security.
W"he principal elements of social insurance are (ILO, 1984b):
insurance is financed by contributions normally shared by employers and employees, with, perhaps, some
form of contribution from the state;
contributions are accumulated in funds used to pay the benefits with surplus funds invested to earn further
income;
participation is mandatory;
a person's right to benefit is determined by his/her contribution;
work-injury insurance is usually financed in whole by employers.
The main elements of social assistance are:
it is financed by the State (i.e., general revenues);
benefits are paid as a legal right;
benefits are paid based on needs;
the level of benefit is determined by bridging the beneficiary's income to a community-set minimum.
'According to this Bismarckian model, achieving universal social insurance coverage is not immediately feasible
for most developing countries because the tax base is narrow due to the fact that the majority of the labor force is
made up of the self-employed, agricultural or unpaid family workers (i.e., the informal sector). Consequently,
innovative means of financing have to be found if coverage is to be offered beyond the formal sector. It is
interesting to note that the reform in Uruguay was partly based on increases in general consumption taxes.



14
as being labor-cost-increasing and also as being regressive because of its burden on low wage earners.
General tax financing is opposed for its high levels of taxation. The main advantage of contributions is
when benefits are earnings-related. Tax financing is more efficient to address universal coverage of flat-
rate benefits. The choice is to be made based on the social conditions of each country (further discussion
below).
26.    Distinguish the functions of different social security schemes. It is imperative to differentiate a
social security system into 2 functions: 1) long-term benefits or long-term social insurance (mainly
pensions), and 2) short-term benefits composed of short-term social insurance (mainly health care
benefits), and social assistance (mainly unemployment benefits and family allowances).
Such functional distinction has many implications. First, the public may better understand the purpose
of each program. Second, accountability for each program is easier. Third, cross-subsidization across
functions is discouraged (for instance, pensions should not be used for unemployment assistance in the
form of easy early retirement and disability benefits).
27.    Employ different models of financing for different social security functions. In particular, finance
short-term benefits with a pay-as-you-go system. For long-term benefits, make a two-step transition:
1) move from a pay-as-you-go to a partially-funded system; then, gradually, 2) extend the equilibrium
period if possible. The transition is made possible by reducing outlays, increasing revenues, and
streamlining the benefit package.
a. Financing short-term benefits.
In the case of short-term benefits, the annual cost is expected to be stable in relation to the insured wage
bill. Therefore, the method of financing is the pay-as-you-go method (also called the annual assessment
system or pure assessment method), under which revenues raised each year balance the expected cost of
that year. A margin is usually adopted in order to build up a contingency reserve.
b. Flnancing long-term benefits."'
In general, there are two methods of financing long-term benefits (especially pension insurance). The
first method is to cover the whole future cost of each year's crop in the year of their reward. The second
method relies on the accumulation of capital to provide additional income in the future. The choice of
the method or any blend thereof depends, again, on the conditions of individual countries.
The first method, called the 'assessment of constituent capitals system," involves raising each year's
revenue to balance the capital value of the pensions awarded in that year. For example, if the capital
value of a pension is 15 times its annual amount, the contribution needed in a given year would be 15
times the total amount of pensions awarded in that year. While popular in employment injury insurance,
this method is not used for pension insurance, most probably because of the high level of annual
contribution it requires (ILO, 1984b).
The second method, called capital accumulation system, has been favored in practice. Basically, it
arranges that current contribution plus interest on capital reserves will cover current cost at all times.
The attractiveness of this method is that it could require low contributions (if the interest on capital
reserves is high - this requires a well functioning capital market and efficient portfolio management).
There are two versions of this method.
'Financing long-term benefits with the pay-as-you-go method could yield unfortunate consequences. In duo
course, as an increasing portion of the population becomes eligible for pensions, the annual cost would rise steeply
to the point where the required annual wage contribution would be exorbitant. A major restructuring of tho
financing scheme would be needed.



15
i.     The first version, called the general average premium system, consists of determining at the outset
a contribution rate (as a percentage of the insured wage bill) which ensures the financial viability of the
system to infinity.4" This system is also called the full capitalization model or general fixed-premium
with full reserves.' This version still prescribes high contribution rates which might not be politically
feasible for many countries.
ii.    The second version, called the scale premium system or partially-funded system, only requires
that the financial solvency of the system be guaranteed over a certain "equilibrium period" (10 to 25 years
for example) instead of ad infinitum. The contribution rate in relation to the wage bill, consequently,
is smaller. As with the above version, this system may only use the interest on capital, never the
principal of the accumulated funds itself, for benefit outlays. When current resources (contribution and
interest) can no longer cover expenses, the contribution rate is scaled (i.e., raised) to a level sufficient
for another equilibrium period.
28.    Abolish (if any) low ceiling on employers' contributions. A relatively low ceiling discourages
part-time jobs by encouraging firms to prefer overtime to acquiring more workers. In particular, abolish
the contribution ceiling for earnings-related benefits (given that this ceiling serves the very rich).
29.    Raise revenues for the following purposes:
* to cover the fund deficit, thereby, minimizing the negative impact on the macroeconomy; or
* to build up capital in order to: 1) make a transition from pay-as-you-go to a partially-funded system;
or 2) to extend the equilibrium period if the system is already partially-funded; or 3) to make a
transition from a partially funded system to a full capitalization system (i.e., to extend the equilibrium
period to infinity) if desirable.
This can be achieved through numerous means.
* Raise the contribution rate in those countries that do not have a very high rate. The extent of the
increase would take into consideration the potential negative effects such as evasion, payment delays,
transfer to the informal sector, substitution of capital for labor, etc.
* Strengthen the inspection and collection activities by enhancing the collection capabilities (sometimes,
reducing corruption), prosecuting and imprisoning debtors and corrupted collection officials, setting fines
and interest at a higher rate than inflation and commercial bank rates.
* Consider an ad-hoc, and even, temporary tax. A VAT tax might be regressive. Sin taxes are good
candidates, so are luxury commodity taxes (again, keeping in mind that broad-based taxes are more
suitable for social assistance).
Many analysts propose a profit tax to avoid a possible bias against labor-intensive industries induced by
a payroll tax (Lloyd-Sherlock, 1992). Brazil and Argentina have had difficulties implementing such
taxes.
* Raise the active worklife ratio by raising the retirement age to accommodate population aging. This
4'The fixed-premium is actuarially calculated to finance estimated future pension liabilities based on
demographic, economic, and other factors. Because the premium is fixed, there is no tmnsfer of that burden to
future generations, and the contributors know in advance what the burden is going to be (defined-contribution). The
conditions to ensure viability of this system must be met at all times: 1) benefits and entitlements may not increase
without a corresponding increase in revenue; 2) there is neither evasion nor payment delays; 3) funds are invested
efficiently and profitably; and 4) actuarial evaluations are done regularly to make any necessary adjustments.
0The fully-funded model could, in theory, be either the assessment of constituent capitals system, or the general
average premium system (also called the full capitalization system). In practice, it is usually referred to the latter.



16
is probably one of the most powerful means of revenue generating for many financially distressed pension
systems (Vittas, 1993).
Other measures which save administrative costs in the long run, and increase collection at the same time,
include:
* Establish a single ID for everybody (social security number or citizenship number, for example), not
only for social security, but also for taxes, elections, visits to public health centers, etc.
* Computerize registration, payments, individual account reports, and identify the delinquents.
* Conduct periodic actuarial evaluations of the funds in order to determine whether there is a need to
revise the equilibrium period or to raise more revenues. This should restrain the politically driven
exercise of introducing new benefits.
7. Portfolio Management
Since social security funds (especially public pension funds) account for a large pool of capital in many
resource-scarce economies, their efficient management is important for many reasons. First, for fully
and partially funded plans, the higher the expected return, the lower the taxes, and the better the chances
that the plans will stay funded. Second, given its share of GDP, a mismanaged system entices harmful
consequences to the economy (increasing the budget deficit). Third, a well managed system will help
develop an efficient capital market by redirecting savings into profitable undertakings (Mesa-Lago,
1991b).
30.    Enhance the portfolio management of pension funds in order to maximize the interest on the
accumulated capital. This is critical to the financial viability of systems which rely on the capital
accumulation system of financing (such as the fully funded or partially funded systems).
The most important element is probably to suppress general inflation which erodes assets invested in
government securities (which is the case for most pension funds) or fixed-income funds. For instance,
in Latin America in 1980-87, the countries with the highest average annual real yields in their social
security funds had the lowest inflation rates, and vice versa (Mesa-Lago, 199 lb).'3 This is yet another
reason for social security system reform to be carried out in conjunction with macroeconomic
stabilization.
31.    Ensure the autonomy of the social insurance agency. The agency should coordinate its investment
policies with other public agencies; however, it should be solely responsible for its investment choices.
The creation of the Investment Board in Jamaica provides an exemplary step in this direction.
This autonomy is necessary to restrict government intrusion to use pension funds for other purposes such
as pressuring institutes to lend to public agencies or to hold government securities with fixed low interest.
In Panama, for example, the government failed to pay its cash contribution for public sector workers and
interest on the government bonds held by the social security agency.
32.    Enhance the financial management skills of personnel in charge of investment. This step is
increasingly important as social security institutes break away from government-directed investment
policies, and develop their own with the intention to maximize returns (given a tolerable level of risks).
This necessitates investing in foreign markets as well, especially in markets which have the tendency to
be countercyclical to the performance of the country's economy. International agencies can help train
and upgrade the skills of the institutes' personnel.
33.    Reduce as much as possible the proportion of total assets which are either fixed or net-current
"One exception was Chile which had the highest positive yield but an inflation rate similar to that of Ecuador
and Costa Rica who had negative yields.



17
assets in order to maximize invested assets. Fixed assets include real estate, hospitals, office buildings,
recreational facilities, etc., which, in most cases, are owned and operated by the institutes. Sell fixed
assets which are neither essential nor profitable.
34.    Diversify the portfolio in order to minimize risks and to compensate low yields with high yields.
It has often been the case that government bonds, mortgage loans, rental buildings, and other low-yield
instruments have made up a large share of many portfolios. Government securities are particularly risky
in countries suffering political instability or upheaval.
35.    At the same time, pay more attention to high yield and relatively safe instruments. These include
fixed-term deposits (short-term deposits in inflationary situations, and long-term deposits in low-
inflationary economies), deposits in hard currency (in hyper-inflation situations), investment in emerging
markets, etc.
This, however, does not imply that fund managers should stay away from government sponsored projects.
The bottom line is to invest in high yield means. Singapore has successfully used its pension plan to
develop its housing infrastructure coordinated by the government." Likewise, even though personal
and mortgage loans are to be generally avoided, the fund could be used to develop the secondary market.
V. SOCLAL SECURITY AND POVERTY REDUCTION
Demographic and social factors necessitate the development of social security systems. Chief among
these are the shift in demographic profiles - the aging population; the weakening of the traditional
extended family and other community-based social security arrangements due to industrialization and
urbanization; and the rise of a large industrial labor force.' These trends are expected to amplify the
need for social security in the future.
Social security could play an instrumental role in the development process of a country.' Foremost,
it could be a significant means of raising national saving."' In the 1970s, several institutes in the Latin
American region enjoyed large surpluses that added several percentage points of GDP to national savings
(McGreevey, 1990). Given its size, it is a powerful fiscal instrument that could be used to further
economic progress. For example, the fund could be used to develop and strengthen the country's capital
market. It could also be used to finance social undertakings with high rates of returns (the public housing
project in Singapore for example). By protecting low-income workers and the unemployed, it could
improve labor mobility, and hence, the efficient operation of the labor market.'
"This contrasts sharply with Panama's use of the fund to promote public housing construction, an act which
contributed to the fund's financial trouble. The difference could be that Panama invested in public housing
construction, while Singapore allowed members to withdraw their funds to invest in private housing, approved
securities, and other purposes such as education (Vittas, 1993).
4'With increasing industrialization, certain risks could best be met through the application of social insurance
principles. The risks (mainly injury, disability, unemployment, old-age, and death) are pooled; premiums or
contributions are paid by employees and employers; and benefits are paid as an earned right.
'Social security has been criticized for various potential negative effects: 1) bias toward capital-intensive
technologies in the modem sector due to the tax on labor, 2) moral hazard (people decide to supply labor less than
they would otherwise), 3) crowding out private insurance and interhousehold gifts, and 4) reduced national saving
by a pay-as-you-go system, in particular. The empirical validation, however, has been mixed (Atlinson, 1987; Cox
and Jimenez, 1990, Arellano, 1985; Break, 1981; Koskela and Viren, 1983; Mackenzie, 1988; Feldstein, 1974,
1982; Barro, 1974). For a more extensive survey of the theoretical and empirical issues, consult Wallich, 1983;
Shome and Squire, 1983; or Munnell, 1986.
'7For instance, in Belgium and the Federal Republic of Gerrnany, social security receipts account for more than
20 percent of GDP in 1977.
'To increase the probability that the system could fulfill these potentials, this paper supports a three stp



18
Social security, moreover, could make a significant contribution to poverty alleviation. Poverty reduction
has always been a cornerstone of the World Bank's mission. The World Bank's approach to poverty
reduction over time has mirrored the evolving development thinking and the conditions of the developing
countries. During the 1970s, the foundation for poverty alleviation was strengthened with three building
blocs: economic growth, income redistribution, and basic needs. The 1980s highlighted the importance
of investment in human resources. Human capital was recognized, not only as the main source of rising
income, but also, of equitable distribution. The increasing burdensome external debt coupled with
rampant inflation which characterized the large part of the Third World in the same decade added another
dimension to the World Bank's approach: stabilization and adjustment. Adjustment, in most cases,
however, entails austerity policies which could involve adverse short term consequences to the affected
population. Therefore, cushioning the impacts of these adverse effects on the populace is critical to
poverty alleviation, as well as to securing the local support for these policies. Social security is
considered an obvious vehicle. Consequently, it has become an integral part of structural adjustment
programs supported by the World Bank and the IMF.
Social security plays a vital role in supporting the vulnerable populace at a time when, around the world,
unemployment is stubbornly high, when many workers are being forced into early retirement, when
increasing marriage break-ups are preventing married women with family responsibilities from
participating in the labor force, and when the speed of modernization is weakening the traditional means
of security based on intra-family and community support. The need for effective social security systems
has, therefore, never been greater.
transition: 1) support a multipillar system, with the government providing the minimum benefit package; 2) finance
social assistance with pay-as-you-go, and social insurance with a capitalization method; 3) reform the public-pillar
social insurance system, by generating savings and revenues in order to make a transition from a pay-as-you-go
system to a partially funded system; and 4) find ways to extend the equilibrium period of the partially funded
system, if possible. (If the equilibrium period is extended to infinity, the system has become fully funded. This
paper does not make a blanket recommendation of making a transition to a fully funded system because of the
resource constraints and demographic patterns faced by many countries -see also Mckensie, 1988.)
The private-pillar system would be financed by provident funds which collect contributions from workers and
employers into personal accounts from which workers (or their survivors) can draw in the events of old age,
disability, death, or other contingencies.



19
BIBLIOGRAPHY
Arellano, J. P. 1985. "The Impact of Social Security on Savings and Development," in C. Mesa-Lago,
ed., The Crisis of Social Security and Health Care: Latin American Experiences and Lessons. Latin
American Monograph and Documents Series, No. 9, University of Pittsburgh, Pittsburgh,
Pennsylvania.
Atkinson, A. B. 1987. 'Income Maintenance and Social Insurance," in A. J. Auerbach and M.
Feldstein, eds., Handbook of Public Economics, Vol. 2. Amsterdam, New York: North-Holland.
_____. 1989. Poverty and Social Security. Hemel Hemstead, England: Harvester Wheatsheaf.
Barr, N. 1987. 7he Economics of the Welfare State. London: Weindenfeld and Nicolson.
. 1992. "Income Transfers and the Social Safety Net in Russia."  Studies of Economies in
Transformation. World Bank, Washington, D.C.
Barro, R. 1974. "Are Government Bonds Net Wealth?" Journal of Political Economy 82.
Break, G. F. 1981. "The Economic Effects of the OASDI Program," in F. Skidmore, ed., Social
Security Financing. Cambridge: MIT Press.
Bureau of the Census. "An Aging World." International Population Reports, Series P-95, No. 78.
Department of Commerce, Washington, D.C.
Burgess, R., and N. Stern. 1989. "Social Security in Developing Countries: What, Why, Who and
How?"  Working paper, Development Economics Research Programme, London School of
Economics, London, England.
Cadwell, J. C. 1986. "Routes to Low Mortality in Poor Countries." Population and Development
Review 12.
Cornia, G., R. Jolly, and F. Stewart, eds. 1987. Adjustment with a Human Face. Oxford: Clarendon.
Cox, D., and E. Jimenez. 1990. "The Connection between Social Security and Private Transfers in
Peru." Public Economics Department, World Bank, Washington, D.C.
Davis, E. P. 1993. "The Structure, Regulation, and Performance of Pension Funds in Nine Industrial
Countries." Policy Research Working Paper No. 1229, Financial Sector Development Department.
Washington, DC: The World Bank.
Danziger, S. H., and D. H. Weinberg, eds. 1986. Fighting Poverty. Cambridge: Harvard University
Press.
Development Committee Meeting. September 27, 1993. "Issues Paper on Social Security Reforms and
Social Safety Nets in Reforming and Transforming Economies." World Bank, Washington, D.C.
Dreze, J. P., and A. K. Sen. 1989. Hunger and Public Action. Oxford: Clarendon.
Feldstein, M. 1974. "Social Security, Induced Retirement, and Aggregate Capital Formation." Journal
of Political Economy 82.
_   1982. "Social Security and Private Saving: Reply." Journal of Political Economy 90.
Grosh, M. E. 1990. Social Spending in Latin America: the Story of the 1980s. World Bank Discussion
Paper No. 106. Washington, D.C.
Guhan, S. 1988. "Social Security in India: Looking One Step Ahead." Bulletin of the Madras
Development Seminar Series, 18.



20
Halstead, S. B., J. A. Walsh, and K. S. Warren, eds. 1985. Good Health at Low Cost. New York:
Rockefeller Foundation.
Haanes-Olsen, L. 1989. 'Worldwide Trends and Developments in Social Security, 1985-1987." Social
Security Bulletin 52(2).
International Labour Office. 1984a. Into the Twenty-first Century: The Development of Social Security.
Geneva, Switzerland: ILO.
. 1984b. Introduction to Social Security. Geneva, Switzerland: ILO.
. 1987. "Demographic Developments and Social Security (Report II)." 4th European Regional
Conference, Geneva. Geneva, Switzerland: ILO.
Jimenez, E., and Puffert, D. 1989. "The Macroeconomics of Social Security in Brazil."  LAC
Discussion Papers N. 0043. World Bank, Washington, D.C.
Koskela, E., and M. Viren. 1983. "Social Security and Household Savings in an International Cross
Section." American Economic Review 73(1), March.
Kotlikoff, L. J. 1987. "Social Security," in Eatwell, J., M. Milgate, and P. Newman, eds., The New
Palgrave: A Dictionary of Economics, Vol. 4. London: Macmillan.
Lee, K. W. 1992. "Interaction of Public Expenditures and Social Security Financing in Latin America."
Social Security Documentation, American Series, No. 15. International Social Security Association,
Geneva, Switzerland.
Lloyd-Sherlock, P. 1992. "Social Insurance Reform in an Ageing World: The Case of Latin America."
Working Paper, Development Economics Research Program, London School of Economics, London,
England.
Mackenzie, G. A. 1988. "Social. Security Issues in Developing Countries: The Latin American
Experience." IMF Staff Working Papers, 35(3), Washington, D.C.
Maguire, M. 1987. "Making Provisions for Ageing Populations." OECD Observer 148.
McGreevey, W. 1990. Social Security in Latin America: Issues and Optionsfor the World Bank. World
Bank Discussion Paper No. 110. Washington, D.C.
Mesa-Lago, C., ed. 1985. The Crisis of Social Security and Health Care: Latin American Experiences
and Lessons. Latin American Monograph and Documents Series, No. 9, University of Pittsburgh,
Pittsburgh, Pennsylvania.
1991a. Social Security and Prospects for Equity in Latin America. World Bank Discussion
Paper No. 140. Washington, D.C.
1991b. Portfolio Performance of Selected Social Security Institutes in Latin America. World
Bank Discussion Paper No. 139. Washington, D.C.
Munnell, A. H. 1986. "The Impact of Public and Private Pension Schemes on Saving and Capital
Formation." International Social Security Review 39.
Organization for Economic Cooperation and Development. 1988. Ageing Populations - the Social Policy
Implications. Paris, France: OECD.
Psacharopoulos, G. 1993. "Returns to Investment in Education: a Global Update." Policy Research
Working Paper No. 1067. World Bank, Office of the Director, Latin America and the Caribbean,
Washington, D.C.



21
Psacharopoulos, G., and N. X. Nguyen. 199 1. 'The Age-efficiency of Educational Systems: a Summary
Measure." New Education 13.
Puffert, D. J. 1988. "Social Security Finance in Developing Countries." Policy, Planning, and
Research Working Paper No. 36. World Bank, Background Paper for World Development Report
1988, Washington, D.C.
Ricketts, J. M. 1986. 'Worldwide Trends and Developments in Social Security, 1983-1985." Social
Security Bulletin 49(9).
Sawhill, I. V. 1988. "Poverty in the U.S.: Why is It so Persistent?" Journal of Economic Literature
26.
Shome, P., and L. Squire. 1983. "Alternative Mechanisms for Financing Social Security." Staff
Working Papers No. 625. World Bank, Washington, D.C.
Schultz, T. P. 1988. "Education Investments and Returns," in Chenery, H., and T. N. Srinivasan, eds.,
Handbook of Development Economics, Vol. I. Amsterdam, New York: North Holland.
Thompson, K. 1986. Survey of Social Security in Developing Countries in Asia and the Pacific.
Bangkok: International Labour Organization.
UNICEF. 1987. Adjustment with a Human Face: Protecting the Vulnerable and Promoting Growth.
Oxford: Oxford University Press.
. 1989. The State of the World's Children 1989. Oxford: Oxford University Press.
U.S. Social Security Administration. 1986. Social Security Programs Throughout the World, 1985.
Washington, DC: Government Printing Office.
1991. "Social Security Programs in the United States." Social Security Bulletin, September.
1 1992. Social Security Programs Throughout the World, 1991. Washington, DC: Government
Printing Office.
Vittas, D. 1993. "The Simple(r) Algebra of Pension Plans." Policy Research Working Paper No. 1145.
World Bank, Financial Sector Development Department, Washington, D.C.
Wallich, C. 1983. 'Savings Mobilization Through Social Security: The Experience of Chile during
1916-1977." Staff Working Papers No. 553. World Bank, Washington, D.C.
World Bank. 1986. Poverty and Hunger: Issues and OptionsforFood Security in Developing Countries.
Washington, DC: World Bank.
1990. World Development Report 1990. New York: Oxford University Press.
1993a. Global Economic Prospects and the Developing Countries. Washington, DC: World
Bank.
. 1993b.  Human Resources in Latin America and the Caribbean: Priorities and Action.
Washington, DC: World Bank.
_   1993c. Poland: Income Support and the Social Safety Net during the Transition. Washington,
DC: World Bank.
1993d. World Development Report 1993. New York: Oxford University Press.
1994. Income Security for Old-Age. Draft.






Human Resources Development and Operations Policy Working Paper Series
Contact for
Title                        Author                  Date             paper
HROWP1    Social Development is          Nancy Birdsall           March 1993       L. Malca
Economic Development                                                  37720
HROWP2    Factors Affecting              Eduardo Velez           April 1993        B.
Achievement in Primary       Ernesto Schiefelbein                     Washington-
Education: A Review of the   Jorge Valenzuela                         Diallo
Literature for Latin America                                          30997
and the Caribbean
HROWP3    Social Policy and Fertility    Thomas W. Merrick       May 1993          0. Nadora
Transitions                                                           35558
HROWP4    Poverty, Social Sector         Norman L. Hicks          May 1993         J. Abner
Development and the Role of                                           38875
the World Bank
HROWP5    Incorporating Nutrition into   F. James Levinson       June 1993         0. Nadora
Bank-Assisted Social Funds                                            35558
HROWP6    Global Indicators of           Rae Galloway            June 1993         0. Nadora
Nutritional Risk (11C                                                 35558
HROWP7    Making Nutrition               Donald A.P. Bundy       July 1993         0. Nadora
Improvements at Low Cost    Joy Miller Del Rosso                      35558
Through Parasite Control
HROWP8    Municipal and Private Sector   Donald R. Winkler       August 1993       E. De Castro
Response to Decentralization    Taryn Rounds                          89121
and School Choice: The Case
of Chile, 1981-1990
HROWP9    Poverty and Structural         lshrat Husain           September 1993  M. Youssef
Adjustment: The African Case                                          34614
HROWP10   Protecting Poor Jamaicans      Margaret E. Grosh       September 1993  M.E. Quintero
from Currency Devaluation    Judy L. Baker                            37792
M. Rodriguez
30407
HROWP1 1   Operational Education         George Psacharopoulos   September 1993  L. Malca
Indicators                                                            37720
HROWP12   The Relationship Between the  John Clark               October 1993      P. Phillip
State and the Voluntary                                               31779
Sector
HROWP1 3   Obstacles to Women's          Joseph Kutzin           October 1993      0. Shoffner
Access: Issues and Options                                            37023
for More Effective
Interventions to Improve
Women's Health
HROWP14   Labor Markets and Market-      Arvil V. Adams          October 1993      S. Khan
Oriented Reforms in Socialist                                         33651
Economies



Human Resources Development and Operations Policy Working Paper Series
Contact for
Title                         Author                  Date              paper
HROWP15   Reproductive Tract Infections,  May T.H. Post            October 1993      0. Shoffner
HIV/AIDS and Women's                                                     37023
Health
HROWP16   Job Security and Labor           Ricardo D. Paredes      November 1993   S. Khan
Market Adjustment in                                                    33651
Developing Countries
HROWP17   The Effects of Wage              Luis A. Riveros         November 1993   S. Khan
Indexation on Adjustment,                                               33651
Inflation and Equity
HROWP1 8   Popular Participation in        Philip R. Gerson        December 1993   L. Malca
Economic Theory and Practice                                            37720
HROWP1 9   Economic Returns from           Edwin Mansfield         January 1994      I. Dione
Investments in Research and                                             31447
Training
HROWP20   Participation, Markets and       Deepak Lal              January 1 994     L. Malca
Democracy                                                               37720
HROWP21   Safe Motherhood in               Patricia Daly           January 1994      0. Shoffner
Francophone Africa            Michael Azefor                            37023
Boniface Nasah
HROWP22   Indigenous People and            George Psacharopoulos   February 1994    I. Conachy
Poverty in Latin America      Harry Anthony Patrinos                    33669
HROWP23   Is Grameen Bank Sustainable?   Shahid Khandker           February 1994    S. David
Baqui Khalily                             33752
Zahed Khan
HROWP24   Concepts of Educational          Marlaine E. Lockheed    March 1994        M. Verbeeck
Efficiency and Effectiveness    Eric Hanushek                           34821
HROWP25   Scientific Research for          Erik W. Thulstrup       March 1 994       L. Malca
Development                                                             37720
HROWP26   Issues in Education Finance      Stephen P. Heyneman    April 1 994        B. Cassorla
and Management in ECA and                                               37172
OECD Countries
HROWP27   Vocational Education and         Julian Schweitzer       April 1994        A. Gonzalez
Training: The Role of the                                               37799
Public Sector in a Market
Economy
HROWP28   Social Security Issues and       Nguyen X. Nguyen        May 1994          M. Espinosa
Elements of Reform                                                      37599