This paper addresses whether or not TDSP eligibility actually increases measured wealth accumulation.
Summary
John Sabelhaus and Ken Ayotte
Demographic trends and projections of continued slow productivity growth have led to significant concern about the viability of Social Security and Medicare in the next century. In addition to these gloomy predictions about public retirement programs, fundamental changes that are occurring in employer-sponsored pension coverage are also becoming a source of concern. Although the overall level of employer-sponsored retirement plan coverage in the U.S. has remained high during the last few decades, an increasing number of working families are now receiving their primary employer-sponsored retirement plan coverage through 401(k) and other voluntary Tax Deferred Saving Plans (TDSPs) rather than traditional Defined Benefit (DB) pensions. If combined employer and employee saving under a typical TDSP is less than employer saving under a typical DB plan, the outlook for national saving and retirement-income adequacy in the next century is even worse than current (gloomy) projections indicate.
Comprehensive analysis of how the switch from DB to TDSP coverage will affect saving requires a series of tests. The first step is to assess whether or not families eligible for TDSP coverage save more than families without TDSP coverage. If TDSP eligibility is found to raise saving, the second step is to ascertain whether or not the increase is large enough to account for the fact that TDSP assets are measured before tax, while other non-pension assets are measured after tax. Finally, even if TDSP eligibility is shown to increase net after-tax wealth, it remains to be shown that the amount of saving being done through TDSPs is equal to the saving that would have occurred under traditional DB coverage.
This paper addresses only the first question--whether or not TDSP eligibility actually increases measured wealth accumulation. That first step seems straight-forward, and is obviously crucial in the overall analysis, but the answer is the subject of an intense debate between Poterba, Venti, and Wise (PVW: 1995, 1996) and Engen and Gale (EG: 1996). Using data from the Survey of Income and Program Participation (SIPP), PVW find a strong, positive effect on saving in financial assets. But using the same data and a similar technique, EG find no increase (and possibly a decrease) in broadly-measured net saving by TDSP eligible families.