Between 1969 and 1972, real U.S. social security retirement benefits rose by 20%. The rise was unanticipated and followed over 15 years of relatively constant real benefits. This paper proposes a model of retirement behaviour in which workers respond differently, although in a theoretically consistent manner, to the anticipated and unanticipated components of the social security benefit they can receive upon retirement. The retirement age decision in the presence of unanticipated benefit changes is shown to be a special case of utility maximization under a nonlinear budget constraint. The model is estimated using the Longitudinal Retirement History Survey.

What Do We Know About the Risk of Individual Account Pensions? Evidence from Industrial Countries by Gary Burtless. Published in volume 93, issue 2, pages 354-359 of American Economic Review, May 2003

Most economists believe that legislation obligating the government to increase expenditures in the future will have some current stimulative effect on private spending; tax increases scheduled to take place in the future will depress current spending. Few economists, however, have considered the effects of budgeting or actuarial rules that oblige the government to change current spending patterns or tax laws. If such rules are followed, future taxes and expenditures will turn out to be different from current ones. Will private behavior be affected solely by currently legislated tax rates and expenditures? Or will it be affected instead by the taxes and expenditures that current budgeting rules appear to require? This question is central to interpreting the fiscal implications of additions to Social Security reserves. Social Security revenues, including interest, will exceed expenditures by $52 billion in 1989. The annual surplus will reach $98 billion by 1994 and $450 billion by 2020. These shortand medium-term surpluses are projected to be dwarfed by even larger deficits starting around 2030. Using a 2 percent real interest rate and other detailed economic and demographic assumptions, the Social Security actuary now projects that expenditures will exceed revenues over the next 75 years by just under 5 percent of the present discounted value of expenditures. Furthermore, the long-run deficit, measured in present-value terms, will grow larger each year. As each year passes, one year of surplus passes into history and all the future years with large deficits move one year closer. The present value of future benefit obligations will consequently grow relative to the present value of future revenues. Ironically, the annual surpluses in the Social Security accounts will swell to unprecedented levels just as the long-run deficit is rising. Which is the better guide to the effect of Social Security on aggregate demand-the annual surpluses, whose growth suggests that Social Security's contribution to fiscal policy is now restrictive; or the long-term deficits, which are also growing larger and suggest that Social Security is stimulative? One's answer to this question determines one's interpretation of recent budget policy. Consider the effect of the Social Security Amendments passed in 1983. Because the OASDI program faced severe financing problems in the early 1980s, Congress reduced benefit entitlements and raised Social Security taxes, but delayed full implementatDiscussants: Robert M. Ball, National Academy of Social Insurance; Lawrence H. Thompson, U.S. General Accounting Office; John Hambor, Social Security Administration.

A model of labor supply is formulated which takes explicit account of nonlinearities in the budget set which arise because the net, after-tax wage depends on hours worked. These nonlinearities may lead to a convex budget set due to the effect of progressive marginal tax rates, or they may lead to a nonconvex budget set due to the effect of government transfer programs such as AFDC or a negative income tax. The nonlinearities affect both the marginal wage and the "virtual" nonlabor income which the individual faces. The model is estimated on a sample of prime-age males from the Gary negative income tax experiment.

A model of labor supply is formulated which takes explicit account of nonlinearities in the budget set which arise because the net, after-tax wage depends on hours worked. These nonlinearities may lead to a convex budget set due to the effect of progressive marginal tax rates, or they may lead to a nonconvex budget set due to the effect of government transfer programs such as AFDC or a negative income tax. The nonlinearities affect both the marginal wage and the "virtual" nonlabor income which the individual faces. The model is estimated on a sample of prime-age males from the Gary negative income tax experiment.