One number does more to your retirement savings than almost anything else you pick: how much of it is in the stock market. Most people can name every fund they own and cannot name that number. This gets you to it.
You will split a contribution between two funds. Then you will do it again with a different pair. Answer the way you actually would, because what matters here is what your own instinct produces. Nothing is scored and there is no right answer waiting at the end.
Round two only unlocks once round one is locked in.
Split your contribution.
You can move the slider as much as you like before locking.
Same task. Different pair.
You can move the slider as much as you like before locking.
Your result
What you actually chose
Only one number here matters, and it is not the split. It is how much of your money ended up in the stock market once you look through the funds to what they hold.
| Round | To Fund A | To Fund B | Stocks you own |
|---|---|---|---|
| 1 · stocks vs bonds | — | — | — |
| 2 · stocks vs balanced | — | — | — |
What the original 180 did
Shlomo Benartzi and Richard Thaler mailed this questionnaire to University of California employees and got 180 back. Each person saw one version only, so nobody was being caught out on their own consistency. The three groups were given three different pairs of funds, and the money landed in three different places.
Fifty-fifty was the most popular single answer in every group. It just means something different each time you say it, and the paper reports what it should have taken to stay consistent: the second group's allocation to the stock fund implied 21%, and they put in 46%.
The same thing shows up in real plans
A questionnaire is a questionnaire. So the same two authors pulled 162 real 401(k) plans and sorted them by how much of the fund menu was made up of stock funds.
In the same paper the authors work out what someone genuinely weighing risk against return would do across an even wider spread of menus. That person barely moves.
The study that cuts the other way
There is a serious challenge to all of this and it belongs on the same page. Gur Huberman and Wei Jiang went through the records of more than half a million people across 600-odd 401(k) plans, which is far more evidence than 162 plan averages. On the headline claim they landed somewhere else: how much people put into stocks was not very sensitive to how much of the menu was stock funds.
So treat the menu effect above as contested rather than settled. What Huberman and Jiang did confirm is narrower and harder to shrug off. People spread their money roughly evenly across the funds they use, and the number of funds they use is three or four, whether the plan offers four options or fifty-nine.
What to do with this
Not much, and it is cheap. Log in to your retirement account and get to one number: the percent of it sitting in stocks. Fund names will not get you there. Then ask whether you would have picked that number on purpose. If yes, you are finished and an accident just became a decision. If no, you now know something specific about your own money that you did not know this morning.
Where these numbers come from
The questionnaire, the 54/73/35 result, the 21% implied allocation and the 162-plan figures are from Shlomo Benartzi and Richard H. Thaler, Naive Diversification Strategies in Defined Contribution Saving Plans, American Economic Review vol. 91, no. 1 (2001), pp. 79–98. One hundred and eighty questionnaires were completed, a 12% response rate, across three groups of 53, 66 and 61 people. The 401(k) figures are the three terciles of a 162-plan sample as of 30 June 1996, with a mean allocation to equities of 48.64%, 59.82% and 64.07% as the share of equity options rose from 0.37 to 0.81; the difference across the three groups is significant at the 0.01 level. The 50% to 53% optimizer comparison is the authors' own calculation in the same paper.
The challenge is Gur Huberman and Wei Jiang, Offering versus Choice in 401(k) Plans: Equity Exposure and Number of Funds, Journal of Finance vol. 61, no. 2 (2006), pp. 763–801, covering more than half a million participants in over 600 plans offering between 4 and 59 funds.
The fund descriptions above are shortened paraphrases of the wording used on the original questionnaire. Your own result is arithmetic, not a finding: round one stock exposure is what you put in Fund A, and round two is what you put in Fund A plus half of what you put in the balanced fund.
This page is general educational information, not individualized investment advice. Past performance does not guarantee future results. Sirmium Capital is a registered investment adviser.