Researchers from Utah State University, California State University, Fullerton and George Mason University used the national Health and Retirement Study to measure retirement timing uncertainty directly as the standard deviation of the difference between self-reported retirement expectations and actual retirement dates.

They found deviations ranging from 4.28 to 6.92 years, depending on the sample.

The researchers note that an individual who draws a retirement shock at age 60 instead of age 65—approximately one standard deviation earlier than expected—would lose multiple years of prime wage earnings, putting a significant dent in the individual’s lifetime budget. This loss is amplified by the need to spread available assets over a longer retirement period.

They also note that uncertainty about the date of retirement helps to explain consumption spending near retirement and precautionary saving behavior.

The research found individuals would give up 2.6% to 5.7% of total lifetime consumption to fully insure the risk of retirement date uncertainty, and 1.9% to 4% of lifetime consumption just o know for sure their actual retirement date at age 23.

Given the magnitude of the welfare cost, the researchers question whether existing social insurance programs help to mitigate retirement timing risk. At a very basic level, the objective of Social Security is to prevent poverty in old age by helping retirees maintain a minimum standard of living. Because benefits are paid out as a life annuity that lasts as long as the individual lives, and because replacement rates are more generous for the poor than for the rich, Social Security is commonly thought to meet its objectives.

However, according to the research, a Social Security retirement program that is calibrated to match current U.S. policy provides only a small amount of timing insurance. Social Security can partially insure timing risk as an early retirement shock leads to a lower total Social Security tax liability and to a higher replacement rate through the progressive benefit-earning rule. Moreover, the payment of Social Security benefits as a life annuity boosts the individual’s expected wealth, which makes him less sensitive to timing risk. However, to adequately insure against timing risk, a program would need to provide individuals with a large payment if they unexpectedly retire early and a small payment if they retire late. Social Security does just the opposite: individuals who suffer early retirement shocks have low average earnings and benefits, while individuals who retire late have high average earnings and benefits.

The researchers note that in some public pension systems such as in Japan, the UK, Spain and other European countries, part of retirement benefits are independent of the individual’s earnings history. A component of retirement benefits is fixed regardless of when retirement occurs. “This feature can mitigate up to one-third of the welfare costs of retirement timing uncertainty,” the researchers say. Having a component that is unrelated to earnings can significantly increase the amount of timing insurance provided by Social Security.

The research report, “The Welfare Cost of Retirement Uncertainty,” can be purchased from the National Bureau of Economic Research at www.nber.org.