By Daniel Richards, Manzur Rashid, Peter Antonioni

Partly because economists (micro and macro) can’t easily conduct lab experiments, and partly because statistical inference is complicated, they turn to building models — simplified versions of reality — in order to think through complex problems.

There are several advantages to using formal models:

  • Macroeconomic problems are complex: They’re so complex that trying to tackle them head-on is almost bound to fail. In this case, a model is like a roadmap. It doesn’t tell you every twist and turn or bump in the road. But it does give a good tool for thinking about how to go from point A to point B. For example, if the question is why average wages in the U.S. are much higher than average wages in Bangladesh, a macroeconomist can safely build a model that ignores the fact that within each country there is a lot of wage and skill variability across workers. That data would be relevant to explaining why different people have different wages, but not why the U.S. average is many times that of the Bangladesh average.
  • Modeling forces you to develop logically consistent hypotheses. For example, it’s likely that interest rates include an inflation premium. All else equal, lenders charge a higher interest rate in an environment of 10 percent inflation than in one of 0 inflation. If this is true, however, then a model that says the Federal Reserve can set the interest rate wherever it likes regardless of the inflation rate has a consistency problem even before one starts to test it with data.
  • Modeling forces you to make your assumptions explicit: Results in economics papers often read along the following lines: “If we assume X and Y, then Z must be true.” For example, “If we assume that households decide how much to spend on consumer goods today based on the income they expect to earn on average in the future, then their spending will be less sensitive to changes in income this period.”

Making assumptions explicit is good practice because it means economists can’t easily pull the wool over people’s eyes.’ In other words, it keeps economists honest.

Intuition can lead you astray: You can spend a lot of time thinking about an economic problem and come to a conclusion that modeling subsequently proves is wrong.

For example, your intuition may tell you that firms rather than workers should pay payroll taxes (the mandatory taxes due when someone works) so that ordinary people get to keep more of their income. But by modeling this problem, economists worked out that it doesn’t matter who officially pays the tax (the worker or the firm), the outcome is the same regardless. If the firm officially pays the tax, then it passes some of the tax onto the worker by lowering wages, and if the worker officially pays the tax, then she passes on some of the tax to the firm by only being willing to work for a higher wage.

  • Comparative statics: Don’t let the jargon scare you; comparative statics simply means comparing the outcome before and after some change. Modeling allows you to see what would happen if certain things within the model change. For example, after you’ve written down your model, you may want to see what would happen to the economy if government spending increased. The model allows you to see the impact without having to change government spending in the real world.