Abstract
This Selected Issues paper on the United States analyzes problems in the measurement of output and prices. The paper examines income versus expenditure measures of national output. Sources of consumer price index and findings of the Boskin Commission are discussed, and mismeasurement of output and productivity is analyzed. Developments in productivity across industries in the United States are described. In particular, the paper focuses on the slowdown in aggregate productivity growth that began in the mid-1970s and examines whether this slowdown has continued in recent years and is common across industries.
VII. The Insurance Role of Social Security1
1. The Old-Age and Survivors Insurance portion of the Social Security system is primarily a pay-as-you-go pension plan, in which payroll taxes of current workers are used to finance the benefits of current retirees.2 As a result of a sharp rise in births from roughly the mid–1940s to the mid-1960s (the baby-boom generation), a long-term decline in fertility rates, and improvements in life expectancy, the number of retirees per worker (the dependency ratio) is expected to rise sharply over the next three decades (Chart 1). This aging of the population compromises the longer-term finances of the Social Security system, which has become the main vehicle for retirement savings in the United States. In 1994, Social Security was the major source of income for approximately 60 percent of the U.S. population over 65 and was responsible for keeping almost 40 percent of the elderly out of poverty. Based on the system’s current terms (i.e., contribution and benefit rates), projections suggest that Social Security will not have sufficient funds to pay all promised benefits on time starting around 2030. A low level of confidence in the system, because of this long-term imbalance, was one of the key problems identified by the Social Security Administration Advisory Council (1997).
UNITED STATES: PROJECTED DEPENDENCY RATIO 1/
Citation: IMF Staff Country Reports 1997, 097; 10.5089/9781451839494.002.A007
Sources: Day (1996).1/ The number of retirees per worker.2. Past analyses of the Social Security system have examined its adverse effects on savings and the supply of labor (see for example Feldstein (1974) and (1996)) and, in turn, options to reform the system are often assessed in terms of the effect they would have on the level of output or aggregate savings. Alternatively, reform options are evaluated on the basis of the ratio of the present value of expected benefits to the present value of contributions (see Gramlich (1996)). However, these analyses generally neglect to take into consideration the value of the insurance that Social Security provides against old-age poverty.
3. This paper illustrates the old-age insurance value of a social security system and evaluates proposed reforms by comparing their insurance value. The analysis is based on a life-cycle general equilibrium model. Parameters in the model are chosen to be consistent with the observed structure of the U.S. economy and the Social Security system, together with the age distribution and life-expectancy characteristics of the U.S. population. The model assumes that the economy is populated with a large number of heterogeneous, risk-averse, rational individuals. These individuals are uncertain about how long they will actually live and how long retirement will last. They can save to provide for retirement, but it is assumed that there are no private annuity markets (in reality this market is thin) to provide some protection against the risk of life-span uncertainty. Results of the model need to be qualified because not all of the potential distortions associated with a social security system are fully incorporated, and including these distortions would likely lower the net welfare gains from social security implied by the model. Also, the sensitivity of the results to changes in model parameter values, functional forms, and the importance of bequests has not been fully explored.3
4. The model illustrates that risk pooling made possible by the introduction of a social security system can provide welfare gains despite the adverse effects it may have on aggregate savings, employment, and output. These gains reflect the insurance the system provides to the elderly against the risk that they will live much longer than expected and outlive their savings. At the same time, adverse selection problems can be so severe that they prevent the wide use of voluntary private retirement insurance, making it necessary for social security to be mandatory.4 The model also suggests that, given the value of social security as old-age insurance, the age of eligibility could be raised without substantially reducing the net welfare gains from the system. On the other hand, the insurance value of the system is diminished to the extent that confidence declines in the ability of the system to meet future obligations.
A. Insurance Role of Social Security
5. Although all individuals in the model are assumed ex-ante to save sufficiently to provide for retirement based on their life expectancy, those who ex-post live much longer than expected can end up with little savings and, hence, will be able to consume only minimal amounts in the later stages of their lives. The introduction of a social security system provides insurance against this outcome and can therefore raise welfare. On the other hand, the social security system eliminates a precautionary motive for savings associated with longevity uncertainty and encourages early retirement. Therefore, after its introduction, individuals save and work less and aggregate savings, employment, and output fall. The tabulation below shows the model’s steady-state estimates of the effect on aggregate economic variables of the introduction of a social security system. However, the model results illustrate that the welfare gains from the old-age insurance that social security provides can more than offset the welfare losses from lower output, employment, and savings. The model also is useful in examining the distributional and welfare effects of social security. Chart 2 shows the distribution of consumption by population group, and Chart 3 shows the distribution of the lifetime utilities of the population, with the average of this utility distribution representing the measure of economy-wide welfare used here.5
UNITED STATES: CONSUMPTION DISTRIBUTION 1/
Citation: IMF Staff Country Reports 1997, 097; 10.5089/9781451839494.002.A007
Sources: Model simulations.1/ The x-axis represents values for consumption.UNITED STATES: UTILITY DISTRIBUTION 1/
Citation: IMF Staff Country Reports 1997, 097; 10.5089/9781451839494.002.A007
Sources: Model simulations.1/ The x-axis represents values for utility.Impact of Social Security on Aggregate Variables
(In percent deviation from steady-state baseline)
Impact of Social Security on Aggregate Variables
(In percent deviation from steady-state baseline)
| Output | −6.1 |
| Consumption | −3.3 |
| Investment | −14.0 |
| Capital | −14.0 |
| Employment | −1.3 |
| Wages | −4.8 |
Impact of Social Security on Aggregate Variables
(In percent deviation from steady-state baseline)
| Output | −6.1 |
| Consumption | −3.3 |
| Investment | −14.0 |
| Capital | −14.0 |
| Employment | −1.3 |
| Wages | −4.8 |
6. When there is no social security, the long lower tails in the consumption and utility distributions reflect individuals who live for many years, have used up most of their wealth, and are able to consume very little. When social security is available, these tails disappear, since these individuals are protected from the effects of long life on consumption. The clipping of the lower tail in the utility distribution increases average utility, and therefore, welfare.
Distributional and Welfare Impact of Social Security
(In percent change from steady-state baseline)
Distributional and Welfare Impact of Social Security
(In percent change from steady-state baseline)
| Wealth distribution | |
| Mean | −14.0 |
| Standard deviation | −12.2 |
| Consumption distribution | |
| Mean | −3.3 |
| Standard deviation | −60.0 |
| Utility distribution | |
| Mean | +26.7 |
| Standard deviation | +44.8 |
| Value of Social Security | |
| As a percentage of aggregate consumption | 4.5 |
| As a percentage of GDP | 3.1 |
Distributional and Welfare Impact of Social Security
(In percent change from steady-state baseline)
| Wealth distribution | |
| Mean | −14.0 |
| Standard deviation | −12.2 |
| Consumption distribution | |
| Mean | −3.3 |
| Standard deviation | −60.0 |
| Utility distribution | |
| Mean | +26.7 |
| Standard deviation | +44.8 |
| Value of Social Security | |
| As a percentage of aggregate consumption | 4.5 |
| As a percentage of GDP | 3.1 |
The tabulation above shows how disparities in wealth and consumption are reduced by introducing a social security system. Note, however, that disparities in the utility distribution security is reported in the final two rows of the table. According to the model, individuals would willingly give up around 3 percent of GDP in order to obtain the insurance that social security would provide.
Need for a mandatory system
7. The basic model can be modified by replacing a government-run social security system with a voluntary private retirement insurance scheme. The modified model suggests that young individuals would prefer not to purchase retirement insurance. There are two main reasons for this. First, because individuals discount future consumption, the young would prefer to consume more early in their lives, and they would have more disposable income to do so if they do not buy retirement insurance. Second, given uncertainty about the time of death, the young know that they may die before retirement, so they run the risk of buying insurance that they might not end up needing. Both of these effects diminish as individuals advance toward retirement age.
8. The fact that young workers opt not to purchase insurance is an example of adverse selection captured by the model. In reality, there are additional informational problems affecting the pricing and availability of private retirement insurance, because the purchasers of this insurance are likely to have more information about then* expected life-spans than the insurance companies. The model suggests that even when there are no such informational asymmetries, the adverse selection problem arising from the preference of the young not to buy retirement insurance can be so severe that it prevents a nonmandatory retirement insurance plan from working. In these circumstances mandatory enrollment in social security would overcome the adverse selection problem.6
Old-age insurance instead of retirement insurance
9. The welfare gains derived from the basic model arise because of insurance against old age. However, the social security system protects all retirees, not just the very old. To illustrate the effects on the insurance role of social security of increasing the age at which benefits are received, the model is modified. Individuals are assumed to retire at age 65, but receive social security benefits only after they reach some age greater than 65. Hence, social security in effect becomes an insurance scheme that provides protection only against the “catastrophe” of living much longer than expected.
10. The welfare gains predicted by the model for different ages of benefit payout are reported in the tabulation below. Almost 90 percent of the welfare gains of the case when the age of entitlement to benefits is set at 65 can be obtained with less than half the level of social security taxes, if benefits are paid only to those above age 75.
Old-Age Insurance 1/
A 43 percent income replacement rate for social security benefits is assumed in all cases.
Old-Age Insurance 1/
| Age at Which Benefits are Received | Contribution Rates (Percent) | Value of Social Security (Percent of Aggregate Consumption) | Value of Social Security (Percent of GDP) |
|---|---|---|---|
| 65 | 13.25 | 4.5 | 3.1 |
| 75 | 6.11 | 3.9 | 2.7 |
| 80 | 3.38 | 3.3 | 2.2 |
A 43 percent income replacement rate for social security benefits is assumed in all cases.
Old-Age Insurance 1/
| Age at Which Benefits are Received | Contribution Rates (Percent) | Value of Social Security (Percent of Aggregate Consumption) | Value of Social Security (Percent of GDP) |
|---|---|---|---|
| 65 | 13.25 | 4.5 | 3.1 |
| 75 | 6.11 | 3.9 | 2.7 |
| 80 | 3.38 | 3.3 | 2.2 |
A 43 percent income replacement rate for social security benefits is assumed in all cases.
Low confidence in Social Security
11. One problem with Social Security today is the low level of confidence in the system’s ability to pay promised benefits in the future. The basic model can be modified to examine how the risk that the government will default on promised benefits would affect a social security system. The model shows that the risk of default lowers the insurance value of social security. Therefore, raising the confidence in social security can raise welfare by making it a more effective insurance program. Hence, it would be possible to raise economic welfare while lowering social security benefits, provided doing so improves the financial position of the social security system and the system’s ability to meet its future obligations becomes more credible (see tabulation below).
Effect of Default Risk on Welfare
Effect of Default Risk on Welfare
| Probability of Default (Percent) | Value of Social Security (Percent of Aggregate Consumption) | Value of Social Security (Percent of GDP) |
|---|---|---|
| 0 | 4.5 | 3.1 |
| 5 | 4.2 | 2.9 |
| 10 | 4.0 | 2.7 |
Effect of Default Risk on Welfare
| Probability of Default (Percent) | Value of Social Security (Percent of Aggregate Consumption) | Value of Social Security (Percent of GDP) |
|---|---|---|
| 0 | 4.5 | 3.1 |
| 5 | 4.2 | 2.9 |
| 10 | 4.0 | 2.7 |
B. Options for Reform
12. The report of the Social Security Administration Advisory Council proposed three different approaches to dealing with the long-term financial problems to of the Social Security system: the maintenance benefit (MB) plan, the individual accounts (IA) plan, and the personal security account (PSA) plan.
13. The MB plan proposes to continue the current system, subject to some modifications. To meet the system’s financial needs, income taxation of Social Security benefits would be increased, coverage would be extended to state and local government employees, and the payroll tax rate would be raised by 1.6 percentage points in the year 2045. The plan also recommends studying the possibility of the Social Security trust fund investing a portion of its assets in a stock index fund, to raise its return and avoid raising taxes in 2045.
14. Under the IA plan, individual accounts would be created alongside the current Social Security system. A 1.6 percentage point increase in the payroll tax rate would fund these accounts. Workers could select a number of investment options, but the accounts would be administered by the government. At retirement, the funds in the individual accounts would be converted to an annuity to supplement the Social Security benefits Additional measures to improve the financial position of the Social Security system would include increasing income taxation of benefits, expanding coverage to state and local government employees, speeding up the already scheduled increase in the age of eligibility for full benefits7, and reducing the growth of benefits for middle-and high-wage workers.
15. The PSA plan proposes to establish an individual account for each worker. Five perentage points of payroll taxes would be allocated to these accounts, and these funds could be invested according to the workers’ choices but would be privately managed. The rest of the taxes would finance a minimum retirement benefit to all eligible retirees. In addition to this minimum benefit, a worker would receive at retirement the funds accumulated in his PSA. Under this plan, coverage of state and local government employees also would be expanded, the age of eligibility would be raised faster than currently envisaged, and there would be changes to benefits and their income taxation. The 5 percentage points of taxes diverted to the PSAs would not be available to pay for current retirees, and to pay for these liabilities, a 1.5 percentage point increase in the payroll tax rate would be required.
16. A key result from the analysis of this paper is that the pooling of risk through a pay-as-you-go social security system can deliver higher welfare. This suggests that the MB plan might provide the highest welfare and the PSA plan might provide the least.8 It also suggests that, if a defined contribution approach is used (as is the case in the IA or PSA plans), it would be preferable to convert the balance in these accounts to an annuity upon retirement. This is because those individuals who live for many more years than expected could withdraw ‘too fast’ from their saving accounts and could end up in poverty.
17. Another result from the model is that the welfare gains from social security largely stem from protecting the very old. Hence, raising the age at which benefits can be collected may be an effective way to restore financial balance while minimizing the welfare losses due to lower benefits. All plans involve raising this age, but perhaps a more rapid increase, as proposed under the IA and PSA plans, might be a better way to restore balance and limit the increase in payroll taxes.
List of References
Day, Jennifer Cheeseman (1996), “Population Projections of the United States by Age, Sex, Race and Hispanic Origin: 1995 to 2050,” U.S. Bureau of the Census, Current Population Reports, pp 25–1130, U.S., 1996.
Feldstein, Martin S. (1974), “Social Security, Induced Retirement, and Aggregate Capital Accumulation,” Journal of Political Economy, Vol. 82, No.5.
Feldstein, Martin S., (1996), “The Missing Piece in Policy Analysis: Social Security Reform,” National Bureau of Economic Research Working Paper 5413, 1996.
Gramlich, Edward M. (1996), “Different Approaches for Dealing with Social Security,” Journal of Economic Perspectives, Vol. 10, No. 3, 1996.
Schulz, James H. (1995), The Economics of Aging, sixth edition, Auburn House, Westport, Connecticut, 1995.
Social Security Administration Advisory Council (1997), Findings, Recommendations and Statements, U.S. Government Printing Office, Washington D.C., 1997.
Social Security Administration (1996), Fast Facts and Figures About Social Security, U.S. Government Printing Office, Washington D.C., 1996.
Valdivia, Victor H. (1997), “The Insurance Role of Social Security,” forthcoming Working Paper of the International Monetary Fund.
Prepared by Victor Valdivia. A complete description of the model and the simulation results is given in Valdivia (1997).
In the remainder of this paper, Social Security refers only to Old-Age and Survivors Insurance. The disability and hospital insurance parts of the Social Security system are not considered.
In this model, parents care about their offspring and bequeath wealth to them. Most other life-cycle models do not have these features.
Adverse selection occurs when only those individuals who have good reason to believe that they will live for a long time buy annuities The cost of annuities is therefore high, and many people cannot afford them; see Schulz (1995).
The utility measure reflects the value of leisure. The use of average utility as an economy-wide welfare measure assigns equal weight to all individuals.
Social Security also has the advantage over traditional private pensions in the United States in that it is fully portable from one job to another, and protects workers who change jobs before they are vested in private pensions plans.
The retirement age is already scheduled to rise gradually from 65 to 67 between the years 2000 and 2022.
The insurance value of the proposed reform plans is not fully examined explicitly. For example, the insurance value of the two tiers in the PSA plan was not considered in the model.
