Q/A | Published in September 2003

Choice Overload

Too many investment options hurt participation rates, Columbia professor finds

By PS | September 2003

Too many investment options hurt participation rates,Columbiaprofessor finds

Too much choice is not always a good thing—at least for 401(k) investment options. Offering employees lots of investment choices in a 401(k) plan actually leads to lower participation, according to a study by Columbia University's Sheena Iyengar in cooperation with The Vanguard Center for Retirement Research. The research finds that, on average, every additional 10 investment choices cuts participation rates by 2%.

"The chance of a worker participating in a savings plan declines as the number of funds increases," says a paper on the research issued by Vanguard in June. "For example, an employee with 5 funds in his or her plan has a predicted participation rate of 72%, while one with 35 funds in the plan has a predicted participation rate of 67.5%."

The role of choice in American life has, for several years, intrigued Iyengar, an assistant professor of management at Columbia Business School. She has examined its impact in contexts ranging from selecting jam at a grocery store to picking potential suitors at a speed-dating event. She recently talked with PLANSPONSOR's Judy Ward about her research on 401(k)s.

PS: How did you get interested in studying the role of choice in 401(k)s?

Iyengar: It was a pretty natural question that emerged. After you look at the effects of too many choices about jams or chocolates, jams and chocolates are fairly trivial decisions that do not have very heavy consequences associated with making a suboptimal choice. However, this is a context where you could imagine that you would be heavily invested in making an optimal choice.

PS: When did you start studying 401(k) plans?

Iyengar: In 1997, the last year Vanguard asked me if I would be willing to analyze a data set. There are 647 different firms in the data set—they range anywhere from about 30 employees to about 30,000 employees—and it has plans that offer anywhere from two to 60 options.

It is a large-enough sample that we can actually look at it in a more systematic manner. People could obviously say, "Even if you get an effect, how do you know it is not just because there is an industry effect going on? Or how do you know it is not because the companies are offering pension plans or other types of savings plans? Or how do you know that it is not a function of salary or things like that?" It is because the actual sample size is 100,000 people in 647 firms, so it allows us to control for a lot of different variables that ordinarily you would not be able to control for. We got the data in the fall, and have been analyzing it since.

PS: What are your key findings so far?

Iyengar: People, if they have more options, are actually less likely to invest. A sort of complementary finding to that is that, among those people who do choose to invest, and the options are increasing, they are more likely to invest in company stock.

PS: Why?

Iyengar: It is a familiar option. If you do not know where to invest, you kind of feel like you know more about your own company.

PS: What is the ideal number of options to get the most participation?

Iyengar: Once you get close to 10, it stops dropping.

PS: Why does more choice lead to lower participation?

Iyengar: In the case of 401(k)s, it is not all that easy to get advice within your firm. A lot of companies probably are worried about being liable if people make the wrong choices. I suppose you could spend the energy going out and seeking private brokers or something to help you with it, but most people probably are not going to put in that effort. So, it becomes easier to delay making the choice than to make the choice. I think they are just scared or nervous. They do not know how to go about actually choosing, and they really do not want to make the wrong choice.

PS: What can employers do to help?

Iyengar: A number of things. You could go the default route. [Two academics at the University of Chicago] have just come out with a paper on "libertarian paternalism" suggesting that people be given a good default [portfolio] option that did not really require them to do a whole lot of thinking or research. If they were given a good default, I imagine that people would consider it fairly reasonable, rational, and logical to actually invest in their 401(k)s.

You could offer some kind of investment consultant within the organization, or you could actually not offer that many options: At the companies only offering two options, participation rates were about 75%. The other thing you could do is offer people a small number of options, and those people who really care about lots of choices would pay a little extra to get more options. If you were forced to pay for extra options, you might be more willing to actually look into those options that you have paid for, do your homework on them.