Americans Need a Simple Retirement System

In a recent column in the Sunday New York Times, Richard Thaler (a pioneer in behavioral economics) reported on two studies regarding financial literacy – one demonstrating the extent of the problem and the other documenting the lack of success to date of educational efforts to improve financial literacy.  The important implication is that we should never have let our retirement system evolve into one where individuals are responsible for all of the complicated decisions of saving and investing for retirement.  But given where we are, we need to make the system we have as easy and automatic as possible.  

I am going to repeat the questions asked of a nationally representative sample of Americans, which come from a study by Annamaria Lusardi of George Washington University’s School of Business and Olivia S. Mitchell of the Wharton School of the University of Pennsylvania. 

1) Suppose you had $100 in a savings account and the interest was 2 percent per year.  After five years, how much do you think you would have in the account if you left your money to grow?  More than $102, exactly $102, or less than 102?

2) Imagine that the interest rate on your savings account was 1 percent per year and inflation was 2 percent per year.  After one year, how much would you be able to buy with the money in the account?  More than today, exactly the same, or less than today?

3) Buying a single company’s stock usually provides a safer return than a stock mutual fund.  True or false?

The first two questions are aimed at fundamental economic concepts related to saving; the third evaluates the respondent’s understanding of risk diversification, a concept crucial for investing.  The answers seem so simple – more than, less than, and false.  Yet among those ages 51-65, the percent answering the three questions correctly was 69 percent, 78 percent, and 60 percent, respectively.  Only 41 percent got all three right. (The results were worse for those younger and older!)  The findings of this study are consistent with a lot of other evidence showing that Americans lack basic levels of financial literacy.

On first being exposed to results such as these, most people’s immediate response is that we need more financial education.  Indeed, that was a common response when 401(k) plans first came on the scene.  But a “meta” analysis of 168 studies by three business school professors finds that interventions to improve financial literacy have virtually no effect on financial behavior, with particularly weak effects for low-income groups.  The problem is that, like all educational efforts, financial education decays over time and has negligible effects on behavior after 20 months.  The authors suggest that, given this decay, “just in time” financial education, which focuses on what people need to know when taking on student loans, applying for a mortgage, or making initial 401(k) investment decisions, might be a more effective way to proceed.     

For those of us interested in retirement security, the message is clear.  People need a simple system.  Unfortunately, with respect to Social Security, deciding when to claim benefits has now become a mini-specialty as the options have multiplied.  The public needs a clear message to wait as long as possible, assuming they are relatively healthy.  And in the 401(k) arena, employees need to be auto-enrolled; they need auto-escalation in their default contribution rate; they need clear guidance that they and their employer should be putting aside a total amount of about 15 percent; they should be enrolled in low-fee target date funds; and they need very simple mechanisms for withdrawal.  

Given the state of play, better design of retirement systems seems like it will trump education every time.