Why we tend to prefer spreading limited resources evenly across options.

Naive Allocation

, explained.
Bias

What is Naive Allocation?

Naive allocation, otherwise known as naive diversification, or the diversification bias, refers to our tendency to equally divide our resources among the options available to us, regardless of whether the options themselves can be considered equal.

Where this bias occurs

The concept of naive allocation is particularly relevant when it comes to investments. In this case, people tend to invest equivalent sums of money in various investment options. Importantly, this decision is made based on instinct, rather than through the use of mathematical models. Arguments exist both for and against this strategy, but one important point put forth by Shlomo Benartzi calls us to err on the side of caution when making investment decisions. He points out that, because our reliance on naive allocation can be influenced by the way certain options are presented to us, we may be heavily influenced by the possibilities offered to us by a stockbroker.1

Individual effects

Naive allocation influences the way we make decisions. It can cause us to spread our resources too thin, or to invest too much or too little of a given resource into a certain option. This reliance on gut instinct, instead of reason, can cause us to make illogical decisions, which may yield negative consequences down the road.

Systemic effects

This concept is of particular importance for so-called “choice architects”, who are people tasked with presenting choices to decision-makers. The role of a choice architect is present in many different domains, including economics, shopping, charities, and even the restaurant industry. Naive allocation comes into play when the choice architect is designing the options available to the decision-maker. The way choice architects present options to decision-makers can greatly influence the choices they make, which can add up to have a significant impact on their respective fields. If people are arbitrarily sharing resources among equal options, bad options will end up getting more than their fare share.

Why it happens

One factor that gives rise to naive allocation is the way the options presented to us are described. If they’re depicted as falling into different categories, we’re more inclined to diversify our selections.2 This is not the only way categorical organization influences our decision-making. It has been implicated in other biases as well, such as the category size bias. It is also more likely to occur when we’re faced with simultaneous decisions – related decisions that must be made at the same time – than it is when we are faced with sequential decisions – related decisions that are made at separate time points.3 

Partitioning of options

Naive allocation is influenced by the way in which the options available to us are presented. It is one of the key concepts that choice architects must keep in mind when describing options to decision-makers. The specific term for the type of presentation that gives rise to naive allocation is “partitioning of options”.4

Take investments, for example. Suppose that a stockbroker provides us with a single list of our investment options: stocks, bonds, and real estate. This list format is not likely to elicit a naive allocation response. However, if a stockbroker gives us three separate lists, one of the different stocks we can invest in, one of the possible bonds we can invest in, and one with our real estate investment options, we are more likely to rely on naive allocation in our decision-making. When the investment options are listed together, we view them as all fitting into the same category. When they are listed separately, we view them as distinct categories. We then try to diversify our selections by choosing an option from each category.5 This does not guarantee that the best options will be selected but rather than poor options will get selected more than they should, in the name of diversity.

Simultaneous vs. sequential choice

An interesting thing about naive allocation is that it only occurs during simultaneous choice. Simultaneous choice is when we are given multiple decisions to make at the same time, whereas sequential choice is when we make multiple decisions one after the other. The first evidence of this was provided by Itamar Simonson in 1990. In their study, college students were asked to choose between six snack foods, such as chips and candy bars. Half of the participants were in the sequential choice condition. They were asked to pick one of the six snacks at three separate meetings over the course of three weeks. The participants in the simultaneous choice condition were asked to select three snacks at the first meeting and were told that they could eat one snack at each of the three meetings held a week apart. Of the participants in the simultaneous choice condition, 64% chose three different snacks. This was significantly different from the 9% of participants in the sequential choice condition who selected three different snacks.6 These results show that simultaneous choice is an important factor in eliciting naive allocation in decision-makers.

Why it is important

It’s important to understand the biases that can affect our decision-making. Naive allocation in particular may cause us to spread our resources out over all available options when it would perhaps be better to focus on just one or two options. Furthermore, this equal allocation of resources may lead us to over or under-invest in a given option. For example, naive allocation would have us invest the same amount of money in two different stocks, even if it may be more advantageous to invest slightly more in a better stock and less in a poorer-performing one. By knowing about the effects of naive allocation, we can overcome it to make more logical decisions.

How to avoid it

Understanding the concept of naive allocation is a good starting point for avoiding this bias. It can be tempting to just “go with our gut” when faced with a decision, but it’s often more effective to put in the work to come to a conclusion based on logic and reason. We can become better decision-makers by slowing down and taking a rational approach to the choices that come our way.

Additionally, making choices sequentially, instead of simultaneously, can help diminish this effect.7 Instead of trying to make multiple choices at once, taking things one by one, and giving ourselves some time between each decision, can help ensure that we come to a rational conclusion.

How it all started

One of the most influential papers on naive allocation was written by Shlomo Benartzi and Richard H. Thaler in 2001. It is titled “Naive Diversification Strategies in Defined Contribution Savings Plans”.8 Not only is it one of the earliest papers written on this topic, but it is also one of the most cited. One of the key concepts introduced by Benartzi and Thaler is what they called the “1/n heuristic”, which describes the phenomenon of an individual dividing their resources equally among the options available to them, with “n” denoting the number of options. This naive diversification strategy closely resembles our definition of naive allocation.

Benartzi and Thaler aimed to identify whether people use naive diversification strategies when making asset allocations, specifically in the context of savings plans. The take-home message of their research was that asset allocation is positively correlated with the number of funds presented to savings plan participants. They stated that, while it has the potential to work out in participants’ favor, this strategy in no way guarantees “sensible or coherent decision-making” (p.96).9

Example 1 - Healthy eating

Interestingly enough, it has been shown that naive allocation can be used by restaurants to encourage patrons to order a healthier meal. If healthy foods are dispersed throughout the menu, in several different categories, and unhealthy foods are restricted to one category, people will tend towards ordering more healthy food options and fewer unhealthy food options.10

As an example, a restaurant menu may be split into the categories of “appetizers”, “entrees” and “desserts” or the categories of “appetizers”, “entrees”, “cookies”, “cakes”, and “ice creams”. The former will encourage less unhealthy eating than the latter. This is because, as is explained by naive allocation, we like to divide our resources among the categories presented to us. Therefore, while the first menu might prompt us to order one item from the “dessert” section of the menu, the second menu might prompt us to order multiple desserts, so that we can sample something from each of the categories.

Similarly, this strategy can be used to encourage patrons to order more healthy meal options by further dividing healthy foods into many categories. As opposed to “appetizers” and “entrees”, a menu using titles such as “salads”, “soups”, “vegetables”, “fruits”, “seafood”, and “chicken” will encourage people to order a variety of healthy foods, instead of filling up on dessert.

Example 2 - Trick-or-treat

When faced with several related decisions at once, we tend to diversify.11 One example of this comes from a Halloween study, which is described in Benartzi and Thaler’s paper on naive diversification. Trick-or-treaters in the sequential choice condition were sent to two separate houses, where they were given the choice between a Three Musketeers candy bar and a Milky Way candy bar. In the second condition, the simultaneous choice condition, children were sent to one house, where they were presented with a pile of Three Musketeers bars and a pile of Milky Way bars and were told to take two chocolate bars of their choosing.12

The results showed that 48% of participants in the sequential choice condition chose different candy bars, whereas 100% of participants in the simultaneous condition chose different candy bars.13 This not only demonstrates naive allocation at work, but it also provides evidence for its underpinnings, specifically, simultaneous versus sequential choice.

Summary

What it is

Naive allocation describes our tendency to divide our resources evenly among the set of possibilities presented to us.

Why it happens

Naive allocation occurs as a result of how possible options are presented to the decision-maker, as well as from simultaneous decision-making, which is when we are faced with making multiple small decisions at the same time.

Example 1 – Healthy eating

Restaurants can encourage people to order more healthy food and less unhealthy food by spreading healthy food across several menu categories and isolating unhealthy food to a single menu category, because we like to spread our resources across the available options, and thus sample something from each category.

Example 2 – Trick-or-treat

One study on the topic of naive allocation split young trick-or-treaters into two conditions. In the simultaneous-choice condition, participants were presented with Milky Way and Three Musketeers candy bars and told to take two bars of their choosing. All the trick-or-treaters in this condition chose one of each. In the sequential-choice condition, trick-or-treaters went to two houses where they were asked to choose if they wanted a Three Musketeers or Milky Way bar. 48% of trick-or-treaters in this condition chose a different bar at the second house than they did at the first. This demonstrates how simultaneous decision-making can give rise to naive allocation.

How to avoid it

We can avoid naive allocation by making decisions in an effortful, deliberate manner. It can be tempting to just “go with our gut”, but we can become better decision-makers by slowing down and taking a rational approach to the decisions that come our way. Additionally, making our choices one at a time, instead of simultaneously, can reduce the influence of this bias.

Références

  1. Bloch, B.J. (2019). Naive Diversification vs. Optimization. Investopedia. https://www.investopedia.com/articles/stocks/11/naive-diversification-vs-optimization.asp#citation-9
  2. Johnson, E.J., Shu, E.B., Dellaert, B.G.C., Fox, C., Goldstein, D.G., Larrick, R. P., Payne, J.W., Peters, E., Schkade, D., Wansink, B. and Weber, E.U. (2012). Beyond nudges: Tools of a choice architecture. Marketing Letters. 23(2), 487-504.
  3. Benartzi, S. and Thaler, R. H. (2001). Naive Diversification Strategies in Defined Contribution Savings Plans. American Economic Review. 91(1), 79-98. DOI: 10.1257/aer.91.1.79
  4. See 2
  5. See 2
  6. See 3
  7. See 3
  8. See 3
  9. See 2
  10. See 3
  11. See 3
  12. See 3
  13. See 3
Notes illustration

Vous souhaitez savoir comment les sciences du comportement peuvent aider votre organisation ?