If you’ve ever eaten at a buffet, you probably know what the naive diversification heuristic is without even realizing it. Let me explain.
When I visit a buffet, I end up taking a little bit of everything so I don’t miss out on the restaurant’s best dish. It always seems like a reasonable way to avoid missing out. Nobody likes missing out, it’s comparable to losing. As humans, we cannot stand losing. This leads to a behavioral bias known as loss aversion. Even when performing simple tasks, like picking food at a buffet, loss aversion kicks in. In the buffet scenario, it drives us to diversify. After diversifying my food selection, I usually discover that about 75% of the food I picked is just okay. The other 25% is what I end up truly enjoying. At the buffet, the naive diversification heuristic isn’t a big problem because I can return to the line and get more of what I like. However, this behavioral issue can create serious problems for retirement plan participants.
Daniel Read and George Loewenstein were the first to uncover the diversification bias. They performed a study on college students that led them to recognize this behavior. The college students were given a choice of six common snack foods. Over the course of three weeks, group one was told to pick snacks at the beginning of each week for only the week ahead. Group two was told to pick snacks that they would eat for the following three weeks. Just 9% of group one chose three different snacks, but 64% of group two chose three different snacks. Group two decided to diversify.
When we have to make more complex decisions and are given multiple options, the tendency is to diversify out of fear that we’ll make the wrong choice. Sometimes, like with a buffet, this works out. Other times, like in workplace retirement plans, it can cause issues.
The naive diversification heuristic in 401k‘s is often described as the 1/N strategy. The 1/N strategy involves dividing contributions evenly across all of the funds offered in the retirement plan. If the plan has 10 funds, the 1/N strategy puts 10% into each. Here’s the problem: the 1/N strategy often leaves investors with an asset allocation that’s heavily dependent on their retirement plan’s structure. This may or may not be appropriate for them.
Richard Thaler and Shlomo Benartzi researched naive diversification in defined contribution savings plans, and found that allocations to stocks and bonds are significantly correlated to how many offerings each asset class has in the plan. If a retirement plan has more stock mutual funds than bond mutual funds, there will likely be a heavier stock allocation. Additionally, if a plan has more large cap stock mutual funds, there will probably be a heavier large cap stock allocation.
This is potentially dangerous to retirement plan investors because each individual participating in a plan has different goals and varying tolerances for risk. These factors are what should be determining their asset allocation. If the 1/N strategy (naive diversification) is being used, they may be very inappropriately invested depending on their plan’s mutual fund components.
I’m all for keeping things simple when it comes to investing, but Albert Einstein did say to, “Make things as simple as possible, but not simpler”. The naive diversification heuristic might be making things a little bit too simple in 401k’s and other workplace retirement accounts.
Sources:
http://faculty.chicagobooth.edu/Richard.Thaler/research/pdf/DiversificationStrategies.pdf
http://www.cmu.edu/dietrich/sds/docs/loewenstein/DiversificationBias.pdf