More Bathtubs

James Kwak

Recently I criticized David Brooks for not understanding the difference between stocks and flows (that is, between your paycheck and your bank balance). (Paul Krugman instead criticized the Tax Foundation, the source for Brooks‘s error – I wonder why?)

It turns out that a lot of people make this kind of mistake. Difficulty understanding stocks and flows may be a fundamental cognitive error such as anchoring or availability bias. In one experiment by Cronin, Gonzalez, & Sterman http://www.sciencedirect.com/science/article/pii/S0749597808000447 it was shown that more than half of a group of students at MIT Sloan – one of the top business schools in the U.S. – could not figure out, from a chart of entrances to and exits from a department store, when the most and fewest people were in the store. These errors turn out to be robust to different framing stories, different ways of presenting the data, and even when getting the questions wrong meant you had to stay in the room for an hour.

The underlying issue seems what they call the correlation heuristic: people think that the behavior of a stock (the amount of water in the tub) should be similar to the behavior of its inputs (the rate at which water pours from the faucet). This is especially a problem when it comes to understanding climate change. In another experiment http://jsterman.scripts.mit.edu/docs/Sterman-2007-UnderstandingPublicComplacency.pdf most people thought that stabilizing emissions was sufficient to stabilize the level of carbon dioxide in the atmosphere; if you think about it, though, you should realize that if you want the level to be stable, inflows have to equal outflows (and right now inflows are about double outflows).

This fallacy may be one thing that leads people to adopt a wait-and-see approach to climate change. On a correlation heuristic-influenced view, we should wait until bad things start happening in year X and then reduce emissions (because reducing emissions will reduce the level of greenhouse gases in the atmosphere). But if we want to stabilize greenhouse gas concentrations at the year X level, we would have to immediately reduce emissions to the rate at which they are absorbed by plants and oceans—which right now would require a 50 percent reduction in emissions. (And this leaves aside the fact that climate change itself lags behind greenhouse gas concentrations, so keeping concentrations stable will not prevent climate change from continuing.)

The national debt situation is complicated (and helped) by the fact that (according to most people) what matters is debt as a percentage of GDP, not in nominal terms. So spending is the inflow, but there are two outflows: tax revenues and economic growth. (Or, to put it another way, the bathtub is constantly getting bigger.)

Still, I don‘t think this gets David Brooks off the hook. Thinking that the impact of a tax increase on the national debt will equal one year‘s incremental tax revenues goes well beyond the correlation heuristic. That‘s more like thinking that a salary increase and a bonus are the same thing, or confusing the speed and the range of a car.

Source: Baseline Scenario http://baselinescenario.com/2011/10/18/more-bathtubs