Do Retirement Savings Policies Increase Total Retirement Saving?

The adequacy of retirement saving has become an area of increasing concern for policy makers and the general public, in light of a shifting private pension landscape that puts more responsibility on individuals to save for retirement and looming fiscal challenges that may lead to future cuts in public old age support programs. The U.S. and many other developed countries spend substantial amounts on policies to encourage individuals to save for retirement; for instance, the U.S. spends $125 billion per year on tax subsidies for retirement savings accounts. But the effect of such policies on financial preparedness for retirement may be limited if people either do not respond to the incentives or offset higher retirement saving by saving less in non-retirement accounts.

In Active vs. Passive Decisions and Crowdout in Retirement Savings Accounts: Evidence from Denmark (NBER Working Paper 18565), researchers Raj Chetty, John Friedman, Soren Leth-Petersen, Torben Nielsen, and Tore Olsen provide new evidence on the effectiveness of retirement saving policies.

The authors use a panel dataset with 45 million observations on savings in both retirement and non-retirement accounts for the population of Denmark for the period 1994–2009. The Danish context not only offers a large and rich data set, but also a series of policy reforms that the authors use to identify the impacts of retirement savings policies.

The authors first explore the impact of automatic contributions such as default settings on retirement saving, policies which in a neo-classical world should have no effect on savings. They use two quasi-experimental approaches. First, the authors track individuals’ savings rates when they switch to different jobs with higher or lower employer retirement contributions. These contributions are automatic in that they require no active choice by individuals. Workers could change their own retirement saving to offset any increase (or decrease) in employer contributions. But total savings rise by more than 85 cents when workers move to a firm that contributes an additional dollar to retirement saving. These changes in saving behavior persist for more than ten years after the firm switch and ultimately result in higher wealth balances at retirement. Second, to explore the effect of automatic contributions outside of employer-provided pensions, the authors examine the effect of the Mandatory Savings Plan (MSP). This policy required all Danish citizens to contribute 1 percent of their earnings to a retirement savings account from 1998 until 2003. The authors find that total saving rose by roughly 1 percent as a result of the policy, suggesting similarly little offset in other accounts.

Next, the authors study the impact of subsidies for retirement savings, again making use of a change in government policy. In 1999, the government reduced the subsidy for contributions to a popular type of retirement savings account for individuals in the top tax bracket by 14 percentage points, while leaving subsidies for those in lower tax brackets unchanged. Aggregate contributions from affected savers to this type of account fell by roughly 50% following the reform, but this substantial drop was due to the actions of relatively few savers. In contrast to a neo-classical model, in which all savers should respond, just 15% of affected savers reduced their pension contributions following the subsidy reduction. The authors then examine what happened to total savings for these 15%. For each dollar by which these savers reduced their pension contributions, the authors estimate that 98 cents went right back into taxable saving accounts. As a result of this shifting, each dollar of government expenditure on subsidies for pension savings generates only 1 cent extra of total savings.

Why are automatic contributions so much more effective at raising savings than price subsidies? The authors explore this question in the final section of the paper by examining the heterogeneity in response across individuals to the savings policies. They find that there are two different types of savers: 15% are "active" savers who think regularly about retirement and respond to incentives, while 85% are "passive" savers who are not focused on retirement and do not pay attention to relevant policy changes. Price subsidies induce active savers to shift assets across accounts but have no impact on passive savers' behavior. In contrast, automatic contributions raise the savings of passive savers. Passive savers tend to be less wealthy and financially prepared than active savers. As a result, automatic contributions not only have larger effects on aggregate savings than price subsidies, but also do more to increase the savings rates of those who are least prepared for retirement.

Overall, the study's findings suggest that price subsidies are less effective as a policy tool than automatic contributions in increasing savings. The authors conclude, "policies that influence the behavior of passive savers have lower fiscal costs, generate relatively little crowd-out, and have the largest impacts on individuals who are paying the least attention to saving for retirement."

The authors acknowledge funding from the Danish Council for Independent Research and from the U.S. Social Security Administration through grant #10-M-98363-1-02 to the NBER as part of the SSA Retirement Research Consortium.


National Bureau of Economic Research, 1050 Massachusetts Ave., Cambridge, MA 02138; 617-868-3900; email:

Contact Us