Journal Issue

Reforming Social Security Systems

International Monetary Fund. External Relations Dept.
Published Date:
January 1993
  • ShareShare
Show Summary Details


Many social security systems worldwide are currently in a crisis, providing inadequate protection while imposing a heavy fiscal burden. A crucial policy issue, therefore, is how lives and livelihoods can be made more secure against deprivation, particularly during a period of structural adjustment. While industrial countries have begun to deal with these problems, most developing countries and postsocialist countries undergoing market-oriented transition need to reform their social security systems as part of econo-mywide adjustment programs. Accordingly, policymakers need to explore fiscally sound yet socially effective approaches for the provision of social security—that is, social insurance (public pensions, unemployment compensation, health care) and social assistance (primarily targeted benefits).

From promise to crisis

Over a period spanning nearly a century, earnings-related social insurance (mainly public pensions and unemployment compensation) and limited forms of social assistance were established first in Europe, followed in the Americas, and later, to a lesser extent, in other continents. Until the 1970s, social insurance schemes prospered financially and gained considerable popularity, as the first generations of enrolled retirees and unemployed were able to benefit in amounts that often far exceeded their contributions during relatively short service periods.

Accumulated reserves of the social security system were often used to finance government budget deficits, at practically no interest yield, with little regard for the long-term sustainability of the system. In addition, reserves of public pension funds were partly earmarked to cross-subsidize health care and other noncon-tributory schemes, or were invested in projects with highly questionable rates of return. Given the continued expansion of the contributing workforce, this approach may have seemed justifiable for partially funded programs; the main preoccupation was with maximizing coverage of the enrolled population rather than with long-term financial stability and intergenerational equity. Whatever financing difficulties emerged, there was scope for solving them with stepwise increases in payroll-based contribution rates.

Provision of social assistance, including health care, for the poor was uneven across countries. In industrial countries, a more equitable income distribution was sought primarily through highly progressive personal income taxation and selected categorical subsidies. In many developing countries, price controls on key foodstuffs and other “essential” products, as well as interest rate ceilings, proliferated on equity grounds. The accompanying subsidies were financed from general budget revenue.

It was largely the first oil shock that exposed the financial imbalances that underlay the social security programs of many developing countries, in the context of severe macro-economic disequilibria. Mounting inflationary pressures, rampant commodity shortages, external payments difficulties, increasing indebtedness, and a marked slowdown in growth revealed major structural rigidities. Social security institutions came under financial stress. The real value of benefits was eroded by inflation, while new rounds of payroll tax rate hikes plus budgetary injections were needed to ensure financial solvency, in effect on a pay-as-you-go basis.

These financial problems have been compounded by generous eligibility for benefits granted on the basis of political expediency rather than past contributions or genuine need. Some countries started to face the potentially most costly social security provision—universal access to health care, with few or no qualifications. Excessive claims for sick pay, partial disability pensions, early retirement pensions, length-of-service pensions, and health care benefits have been on the rise in a number of countries; in particular, they placed an unsustainable burden on developing countries facing major macroeconomic imbalances. Meanwhile, the contribution base in the latter countries began to shrink, partly because of a slowdown in real income growth in the formal sector. Further, the rise in the benefit-contribution ratio has been exacerbated by the turnaround in demographic trends, driven by declining birth rates and rising life expectancy, the full repercussions of which have yet to be felt. The typical policy response in many countries—sometimes as a conditioned reflex—has been to raise statutory contribution rates further.

These developments, along with a lack of transparency and accountability for social security finances (often combined with corrupt and inefficient administration), weakened the perception of any linkage between benefit eligibility and contribution record and, thus, undermined compliance with what has come to be regarded increasingly as an onerous form of regressive payroll taxation. In turn, incorporation of high payroll tax rates in labor costs contributed to erosion of the taxing country’s international competitiveness—absent border tax adjustment for such taxes, under the destination principle, in foreign trade—imposing yet an additional burden of adjustment on the exchange rate.

The social security crisis is nowhere more acute than in the postsocialist economies of Eastern Europe and of the former Soviet Union. In these economies, social security schemes have become grossly dysfunctional, owing to rigid ideological constraints. In the past, unemployment was hidden through large-scale redundancies in state-owned enterprises and through relatively easy eligibility for various social insurance benefits (especially low retirement age, and generous sick pay and disability benefits), given the lack of unemployment compensation. The primary vehicle for poverty alleviation took the form of price subsidies to satisfy merit wants—available to all, regardless of need.

At present—even after the official recognition of unemployment and poverty and the creation of specific schemes to deal with them openly—the dysfunctional treatment of unemployment in most transition economies seems to have been aggravated by increased recourse to early retirement and disability benefits. At the same time, these economies—lacking the necessary administrative capacity—are singularly incapable of delivering social assistance to the needy. This is because social security programs were run by state-owned enterprises; as these enterprises become increasingly exposed to market discipline and shed redundant labor, a rapid increase in poverty occurs.

The case for reform

In an effort to correct macroeconomic imbalances—reflected in external current account deficits and high inflation—a number of developing countries embarked in the late 1970s on comprehensive adjustment programs that, besides short-term stabilization, were aimed at bringing about structural reform. The latter consisted mainly of liberalizing commodity prices, interest rates, and the exchange rate; financial and fiscal reform; overhauling or privatizing state-owned enterprises; and liberalizing foreign trade and payments. Governments recognized that the magnitude and nature of such structural changes, in combination with fiscal and monetary restraint, could have a significant short-run adverse impact on incomes and employment. Such an outcome is inherent in postsocialist economies undergoing market-oriented transformation.

Clearly, the removal of subsidies to loss-making public enterprises and the phaseout of consumer subsidies (especially those involving foodstuffs) to households can create an immediate deterioration in social conditions as compared to the status quo ante. Arguably, however, under most adjustment programs living conditions improved in comparison to the relevant counterfactual situation—characterized by widespread shortages, hoarding, falling real incomes, and black market activities—which was unsustainable given domestic financial disintermediation, output contraction, and capital flight. Moreover, market-oriented structural reform combined with stabilization would lead to sustainable economic growth in the medium term.

Beginning in the second half of the 1980s, the IMF sought to deal with the social consequences of adjustment programs in an active and systematic manner. In essence, the underlying rationale was the necessity to buttress the social and political acceptance of the adjustment effort with an explicit effort to contain adverse short-run distributional implications. As part of its appraisal of macroeconomic policies, the IMF staff began to examine the possible consequences of the stance and mix of key adjustment measures on socioeconomic conditions in member countries, and to identify ways of strengthening the social safety net for adversely affected groups. This initiative focused primarily on easing the temporary impact of such measures on the poorest strata in low-income countries.

More recently, increasing attention has been paid to comprehensive social security reforms in the context of structural adjustment. At the request of member countries, the IMF provides technical assistance to support such reform. Since 1989, at least a dozen member countries received technical assistance for de. signing and administering social security schemes—public pensions, unemployment compensation, health care, and social assistance. In addition, social security reform measures, along with other structural fiscal reform measures, have been increasingly incorporated in IMF-supported adjustment programs.

Options and constraints

The main goal of social security reform is cost effectiveness—meaning “the biggest social bang for the fiscal buck”—based on the premise that social security can no longer be viewed simply in terms of the magnitude and coverage of benefits, but must be assessed also in terms of fiscal cost and allocative efficiency. In other words, a cost-effective social security system provides maximum protection at the least cost in terms of fiscal resources and allocative distortions and contributes to macroeconomic balance. This fundamental criterion for reform encompasses a range of considerations, with implications for the design and implementation of the system. More specifically, it allows for a variety of reform options while being subject to a number of limiting constraints.

“The social security crisis is nowhere more acute than in the postsocialist economies of Eastern Europe and of the former Soviet Union”

First, it is necessary to take into account explicitly the relevant cultural and social characteristics, as well as historical antecedents, of each country. Obviously, these characteristics shape behavior, attitudes, and aspirations with regard to the provision of social services. At one end of the spectrum, there are societies where the extended family or village structure still operates rather actively as an informal social security scheme, obviating the urgent introduction of large-scale public pensions and assistance schemes. The example of certain Asian countries (such as Indonesia, under the goton royong principle), and to a lesser extent, Mediterranean countries, illustrates this approach. At the other end, in countries where households have been atomized and informal self-help arrangements have been weakened or broken, there is an immediate need to provide an extended social safety net. This is the case, for instance, in much of the former Soviet Union, where the individual has been made totally dependent on the state, through his or her workplace—while facing virtually insurmountable barriers to mobility.

Second, for the sake of allocative efficiency, administrative simplicity, and increased labor mobility, social security provisions should be uniform across occupations and economic activities, including the government workforce. Unless justified by clearly differentiated risk, eligibility for benefits, benefit levels, contribution rates, and the degree of subsidization, if any, should be uniform. Moreover, social insurance schemes (including Chilean-type government-mandated private pension funds) can be supplemented usefully with voluntary private insurance and retirement funds. In addition, nongovernment initiatives—by charitable, religious, or other organizations—that provide valuable assistance to the indigent, deserve support.

Third, there is a need to tailor the social security system to a given country’s level of economic development and labor market conditions. To promote labor force participation in the formal sector, it is necessary to remove impediments and disincentives to employment, including high payroll tax rates, in that sector. Besides strengthening the links between benefits and wage-based contributions, contribution rates should be kept relatively low in a low-wage, labor surplus economy. Furthermore, inadequate administrative capacity makes the provision of means-tested social assistance in many developing countries difficult.

It cannot be overemphasized that social security reform should be geared to the eradication of excess consumption and waste of scarce resources. With few exceptions, open-ended consumer price subsidies should be abolished and replaced by means-tested transfers. Given administrative limitations, such as those prevailing in some developing or postso-cialist economies, it may be necessary to resort to various forms of categorical transfers to well-identified groups (for instance, households with children and elderly), possibly in the form of benefits in kind. In addition, automatic indexation of benefits for inflation is fully justified for equity reasons. Health care provision is another area of considerable potential for saving through various cost containment techniques. In particular, improved management practices, stepped-up preventive programs and education, increased information about medical and hospital costs for both health care users and providers, cost-sharing, and user fees and supply incentives, should help reduce abuse and waste.

Fourth, there is a need for a clear institutional distinction among social security schemes by function or purpose: old-age, disability, and survivors’ pensions; unemployment compensation; health-care services; and poverty alleviation. Separation of social insurance programs and social assistance programs, notwithstanding some inevitable overlaps, helps to differentiate their characteristics so that the public may better understand the purpose of each program, and promotes accountability for each program. Such a functional distinction implies that public pensions should not be used as a form of unemployment relief—through a low retirement age or easy access to early retirement benefits and disability benefits. This differentiation also has implications for the sources of financing. Social insurance programs (old age, disability, unemployment) should be financed primarily with wage-based contributions by the insured employee and by the employer on his or her behalf, supplemented by income from reserves created with such contributions. By contrast, the main source of financing for social assistance programs should be general tax revenue.

Fifth, besides ensuring social protection and fairness for the present generation, social security must seek an equitable distribution of benefits and costs between present and future generations. This intergenerational aspect is often neglected in both industrial and developing countries owing to political expediency or concern mainly with short-term macroeconomic disequilibria. Public pension programs, in particular, should be financially self-sustaining over long periods of time, capable of withstanding considerable demographic and economic fluctuations. Viewed as an intergenerational contract, a public retirement scheme commits future generations that do not have an opportunity to influence or vote on the design of the present system, to support preceding ones. An overly generous benefit structure inherited from the past usually can be corrected only through a gradual and painful reform, including costly grandfathering of existing benefits, following a protracted national debate.

Finally, it is imperative to observe certain constraints on the speed and scope for implementing reform. The size of the formal sector imposes an outer limit on the coverage of insurance-type programs. The capacity to administer the collection of contributions and the disbursement of benefits imposes an even tighter limit in the short run. Although not bound to the formal sector, the delivery of social assistance is determined chiefly by fiscal resources and administrative capacity. Equally important for the success of the reform—especially against a record of unfulfilled expectations—is the institutional transparency of social security operations and accountability of management. In this respect, the creation of well-administered public trust funds (one for each social insurance program) can make an important contribution to credibility and popular acceptance. A binding constraint, particularly on health care and social assistance programs, as well as on the extent of grandfathering during the transition to a new system, is the availability of fiscal resources.

Other Resources Citing This Publication