Build Yourself a Theory of Happiness

When planning for the future, one of the greatest challenges is that we're bad at predicting what will actually make us (un)happy. In other words, we lack a theory of happiness. 

A person from an average-sized town leaves everything behind and moves to London, only to find that they are crushed by the loneliness, high costs and stress of the big city life. Someone who achieved sudden wealth buys a big mansion in a gated community, wherein she starts to feel lonely and alienated, and finds herself missing the communal living of her student years. Another takes up an important job at a fancy firm, where the stress and long hours drive them to depression. 

In these examples, the problem is not with the choice itself. Some folks may end up happier moving to London, buying a big mansion or getting the fancy job. The problem is with the process, which is often faulty. All these choices happen to score high on prestige and monetary reward. Sometimes, being unable to identify the predictors of our happiness, we resort to the substitution principle:

If a satisfactory answer to a hard question is not found quickly, System 1 will find a related question that is easier and will answer it. (Kahneman, p. 97)

Hence we unwittingly default to proxies which are easier to measure, such as money or prestige. The question "which job will make me happiest?" becomes "which job will give me the most money or prestige?" Many important dimensions are lost in this conversion. 

The other problem is that we are bad at simulating how our life will actually be like under the new circumstances and how we will feel about it. When we try to simulate the future, we tend to fall on the inside view: we roughly imagine what the future would look like and see what feelings these images evoke in us; then we use these feelings as a cue to our future feelings and satisfaction. This is an understandable heuristic, but it's ineffective. 

Here's a better approach. 

First, you need to test your ability at predicting your happiness. It may help to set up a diary where you try to predict your happiness at different points in time, or conditional on certain events occurring or not, and then rating yourself on the predictions. I may set up a framework for this in a future post. 

Then take a scientific approach. Gather the experiences of others and the results of previous research. Formulate hypotheses, such as 'I would be happier with job X than job Y'. Devise experiments to test them, such as getting internships for X and Y and tracking your satisfaction. Finally, use the results to update your beliefs and formulate better hypotheses. This is how you build a theory of happiness. 

It's true, we can't always afford the whole process. Experiments can be costly, difficult to implement, and ambiguous in their interpretation. For example, if you were unhappy during internship X, is it because you actually don't like the job, or is it because of contingent factors, e.g. you didn't like the company, or the project you worked on was unattractive? 

Nevertheless, I find it useful to keep this in mind as the golden standard of pursuing happiness. And to be conscious of the heuristics I use to approximate it and their limitations. 

Whenever I make plans for the future, I try to at least implement the first step: putting aside easy proxies such as money and prestige, taking the outside view and gathering objective information. This is already a long way from the standard approach of simulating outcomes in my imagination.