Starbucks Equilibrium

November 10, 2015 • John Vandivier

This article argues that economics should begin to think in terms of real variable maximization, as opposed to utility maximization, using a lighthearted Starbucks analogy.

Take a look at this:

I'm a fan of coffee. I drink it for two reasons. First, it tastes good. Second, it has caffeine. Today the thought crossed my mind: I can adjust my own preference to favor one over the other.

This can be thought of as the "economics of getting used to it". As an individual, I can increase my income to maximize my utility, or I can increase my utility at lower levels such that the place I am at becomes the maximum utility point.

If I choose to prefer the taste of coffee less relative to its caffeine content, I can improve my work productivity by consuming more caffeine for the same price.

Pike Place tastes a little better.
Blonde Roast (Verdana) has more caffeine.
They cost the same amount.

Better yet, I could work to cultivate an acquired taste for Verdana.

What if everyone in the economy switched from Pike Place to Verdana?

Productivity would increase
Real income would go up
Utility could go, up, down, or stay the same.
- That's not to say it's indeterminate: It will be determined by the degree of the shift in aggregate preferences.
- An increase to real income is expected to increase utility for a reason other than the preference shift. So even if preferences don't shift favorably, utility may still increase due to the income effect.

There are quite a few markets like coffee. The food market at large is mostly analogous. People could consume tasty food or healthy food. They often prefer the former. Economists largely argue that preferences are exogenous and not to be judged as good or bad. I many times disagree.

What if economists were more concerned with maximizing real variables instead of utility? I think economics would be a much better field for it. In that case we would seek to maximize productivity and advocate for light roast coffee over dark roast.

There is also the idea that economists should describe what is rather than what ought to be. I think economists should do the same thing they have been doing, which is discuss both positive and normative ideas. I also think that markets do tend to better themselves, so I think that a real-variable oriented approach to economics is better for both normative and positive assertions. To be clear, both services are very much in demand and supply with respect to economists.

Of course, some preferences are harder to change. Some preferences we don't want to change. I think by market logic the latter group will eventually perish by the presence of more productive competitors.

While <a href="http://econlog.econlib.org/archives/2015/10/superforecastin.html">reviewing a book called Superforecasting by Philip Tetlock and Dan Gardner, Bryan Caplan notes one important finding from that book, "The strongest predictor of rising into the ranks of superforecasters is perpetual beta, the degree to which one is committed to belief updating and self-improvement. It is roughly three times as powerful a predictor as its closest rival, intelligence."

Perpetual beta is the constant willingness to adapt one's own beliefs. I propose that an improved ability to forecast can be viewed as or similar to an increase to productivity. I also propose that preferences can be viewed as or similar to beliefs. Under this view we may take the evidence collected in favor of perpetual beta to hypothesize that the constant willingness to adapt one's own preferences can lead to increased productivity.

I'll wrap up with a joke: Why is Starbucks named Starbucks? Because the prices are so high.