The key to happiness is low expectations

And a simple model that seems to agree

Last week I moved to Paris. Yes, that Paris. The one in France. Coming from Houston Texas where everything couldn’t be bigger, I was destined for a place where space is a luxury. My “cozy”, studio apartment includes a shower whose cold water knob I habitually bump into, and a kitchen designed only for one chef. The coffees here only take a couple of sips to finish, and there is hardly any breathing room on the metro. Needless to say, so far it’s been unlike anything I’ve ever experienced. But I’m thrilled. Who knows what surprises could be around the corner? What interesting people would I meet? Would my French get any better? (It couldn’t possibly get worse.)

Going in, I didn't know what to expect.

The only thing I did expect was that I would make my fair share of mistakes along the way. And believe me, so far I have. I’ve locked my clothes in the washing machine, needing to rerun it twice before the door unlocked itself. I’ve searched about 5 different supermarkets for protein powder before figuring out you can only get it at a specialty nutrition store. And I’ve missed my exit on the metro several times.

But, I think I’m starting to get the hang of it. The smell of a French gym is only jarring the first few times (spoiler alert, it smells like just that, a French gym.)

It can be easy to get frustrated, making so many silly mistakes simply because you aren’t familiar with how another country works, and don’t speak the native language well. But my expectations about messing up seems increase my patience.

The key to happiness is low expectations

-Barry Schwartz, TED Talk

I’m sure you’ve heard the phrase before, and I’m not an expert on the philosophy of psychology, but I will say in economics, expectations certainly matter. Milton Friedman conjectured that when people expect higher levels of inflation, they choose to spend now and face a lower price rather than a higher one in the future. The increase in demand then causes prices to rise, leading to a self fulfilling prophecy. These “adaptive expectations” have become common in macroeconomic models of inflation. That is, the level of inflation depends on what people think the level of inflation will be.

Put differently, when faced with some uncertainty your expectation matters for the decision you choose.

This probably sounds familiar. Say, you feel like you’ve done well on an exam. You were prepared, felt confident in your answers, and walked out of the test expecting to have done well. There is still uncertainty around how you may have actually done on the test, but you’re confident and so you spend less time studying for the next one. Take the alternative, where you expect to have done poorly. You sulk, come out of the test room light headed, maybe a little angry, maybe with a few tears, but you pick yourself back up and study harder for the next exam. Your grade is unknown, but your decision on how to study changes.

A tangential example which comes to mind is the well studied phenomenon of “framing” in a negotiation.

Taken from a Metropolitan Diary NYT article

The expectation here is that the slices would be the same size, and so when they weren’t, the decision, or frame of mind, was to be unhappy. Imagine if I had thought everything would go smoothly moving 5000 miles around the world!

Happiness is an arbitrary term. But to an economist it is roughly equal to a utility function. These functions typically take the form of Cobb Douglas, or logarithms, such that the marginal utility is inversely proportional to the quantity. This means the more you have of a good, the less an additional unit of the good will have an impact on your happiness. If you gave $100 to a college student, or Jeff Bezos who would you expect to be more happy? Mathematically is looks like

dU/dx = 1/x

Where U = utility and x = a good. Take the integral to solve for U, and you get

U(x,C) = ln(x) + C,

Where C is a constant that will likely change by each person. Now you have a function for your happiness, which depends only on good x. It is a positive relationship, with decreasing marginal returns. But notice that utility here doesn’t depend on expectations.

So why not build that into the model?

dU/dx = 1/x - y(x)

Where y = expected happiness after a unit of x, which we might expect depends again inversely on the quantity of x. We think expectations can differ by good and person, and so let y(x) = k/x where k is a constant.

dU/dx = 1/x - k/x

The subtraction indicates directionally how your happiness changes. If the pizza meets or exceeds expectations, you get more happiness. If it falls short, maybe you’re less happy than you were previously. Such as, for example the pizza gave you food poisoning, you are now worse off than before. Integrating again we get:

U(x,k,C) = ln(x)-k*ln(x) + C

Which, from our logarithm operations, we know is equal to

U(x,k,C) = ln(x/x^k) + C

U(x,k,C) = ln(x^1-k)+C

So lower your expectations? (That is, lower your value of k?) You just might find yourself increasing your utility.