Unemployment Insurance Savings Accounts

Unemployment Insurance Savings Accounts (UISAs) could substantially reduce UI's adverse effects on incentives without decreasing protection for those who suffer unemployment.

Unemployment Insurance (UI) in the United States supports the income of people made unemployed but also provides an incentive for people to become, or stay, unemployed. In Unemployment Insurance Savings Accounts (NBER Working Paper 6860) , NBER President Martin Feldstein and co-author Daniel Altman show that a system of unemployment protection based on individual savings accounts - Unemployment Insurance Savings Accounts (UISAs) - could substantially reduce UI's adverse effects on incentives without decreasing protection for those who suffer unemployment.

Under such a system, people would be required to save a fraction of their wages (up to 4 percent) in special accounts. If they became unemployed, people would benefit from their accounts rather than taking state UI benefits.

Positive accounts would earn a market rate of interest. If the accounts were exhausted, then the government would lend money to tide people over, again charging the market rate. Positive UISA balances would be converted into retirement income or bequeathed if the individual died before the retirement age. Negative account balances would be forgiven at the retirement age, or if the individual died.

The virtue of such a system is that when a person becomes unemployed, drawing from a UISA with a positive balance reduces personal wealth by an equal amount. This means that the costs of unemployment are fully borne by the individual. UISAs would provide the same protection as the existing state UI system, without the adverse incentives. The only adverse incentive effect is for individuals who expect to retire (or die) with negative balances in their UISA; they face the same adverse incentives as under the traditional system.

Feldstein and Altman use data from the Panel Study on Income Dynamics to simulate a UISA system over a 25-year historic period. Their analysis indicates that almost all individuals have positive UISAs balances and therefore remain sensitive to the cost of unemployment compensation. They find that around 5 percent of employees, based on the historical data, would retire or die with negative account balances and that only about half the benefits under the UISA system would be paid to such individuals. Even among individuals that experience unemployment, most have positive account balances at the end of the unemployment spell.

Although about half the benefit dollars would go to individuals whose accounts are negative at the end of their working life, less than one third of the benefits go to individuals who also have negative account balances when unemployed.

The cost to the taxpayer of unrecovered loans in the negative accounts is substantially less than the cost of the current UI system. Under a system of UISAs, the government could cut the payroll tax. Thus, as well as reducing the cost of providing unemployment protection, UISAs would permit a reduction in payroll taxes, reducing distortionary effects in both cases. Feldstein and Altman show that a system of UISAs, combined with a government safety-net, is a viable economic option.

-- Andrew Balls