Economists differentiate between preferences (what you prefer – cookies or cakes) and beliefs (flat earth or evolution). People can have wrong beliefs but they can not have wrong preferences. People like what they like. It is important to understand the distinction as it plays a key role in understanding why people back certain policies and adhere to maintain specific political affiliations. It is important to know this as it helps one put other people’s views in perspective. In the time of social media echo chambers, it is easy to disregard opinions different from your own by calling them uninformed. The truth is there’s more to trending opinions than just people’s preferences.
In one of the chapters of the book, Good Economics For Bad Times by Esther Duflo and Abhijit Banerjee, a fascinating concept of changing preferences has even discussed. In 1977, two economists Becker and Stigler argued that preferences are a part of who we are. It suggests that preferences are stable and not open to influence. This became a foundational idea in economics. The concept is that once we have considered all the information at hand, we decide to do what we decide to do based on our intrinsic nature. There is some nuance here. They understand that we can always dismiss any and every choice made by any individual as their preference. That’s not what they want. They want us to take a deeper look at whenever we bump into a seemingly irrational choice or behavior, to understand it, rather than just dismiss it as a matter of preference.
A quick sidebar on choices. We are told not to judge other people’s choices and assume they are making the right choice after having considered all possible options, but we don’t really believe in this when it comes to the poor. Which is why policies like food stamps are so popular. We think the poor can’t be just given money. We assume they will waste their money on things they want rather than need. There is enough research by Abhijit and Esther in this field and they have enough data to explain how this line of thinking is totally flawed. One of the more famous examples of this was a person they met in rural Mongolia who when chanced upon some extra money, bought a TV with it when it could have gone to better use for food and other essential household items. The researchers gasped at this and interviewed the person about this. They learned that there was absolutely nothing to do in the village. The TV gave the family something to do and offered respite from the mundane life in a village that did not offer any other avenues for leisure. The family explained how for them the TV was an essential commodity offering a significant increment to the quality of life and now that they have it, subsequent spending would go towards food and other essentials. This is something we can keep in the back of our heads when we find ourselves judging other people’s choices.
Coming back to preferences. Abhijit came up with a model for herd behavior. It highlights the importance of “random first events” and how they lead to the creation of information cascades. This model explains how people collectively make choices that seem like fads to an untrained observer, but in reality, has more to it. He takes the example of a train station where people take their place in a queue. The first person decides to stand and wait for the train by standing at a particular place on the basis of their preference (often unrelated to where the train usually stops). The people who follow, even if they have their own preference, think there is some information that the person who is standing there already, possesses. They make a rational decision to follow that person thinking they have some info they don’t and thus increasing the number of people in the queue. This way, rational decisions taken by people, end up creating longer queues (which may or may not be right).
To dig in further, a study on an online review site was conducted to see if these “random first events” can be used to create trends. An artificial upvote was added to some comments and a downvote was added to others. Even after millions of views (which should have nullified the effect of those one extra votes), it was still observed that the posts that had been given the first artificial upvote, ranked much higher than the other posts that got the downvote. Even if people liked the posts that got the artificial downvote, they felt influenced by it and thought something was missing and that persuaded them into not voting for them.
The lesson is simple. Human beings are open to influence and when you chance upon a choice or a preference you can’t explain, don’t dismiss it as intrinsic human nature. Dig in further to understand what triggers it. Remember, preferences change and are not stable. What is popular or acceptable now, will not stay that way.
I run a startup called Harmonize. We are hiring and if you’re looking for an exciting startup journey, please write to jobs@harmonizehq.com. Apart from this blog, I tweet about startup life and practical wisdom in books.