“
Many models are constructed to account for regularly observed phenomena. By design, their direct implications are consistent with reality. But others are built up from first principles, using the profession’s preferred building blocks. They may be mathematically elegant and match up well with the prevailing modeling conventions of the day. However, this does not make them necessarily more useful, especially when their conclusions have a tenuous relationship with reality. Macroeconomists have been particularly prone to this problem. In recent decades they have put considerable effort into developing macro models that require sophisticated mathematical tools, populated by fully rational, infinitely lived individuals solving complicated dynamic optimization problems under uncertainty. These are models that are “microfounded,” in the profession’s parlance: The macro-level implications are derived from the behavior of individuals, rather than simply postulated. This is a good thing, in principle. For example, aggregate saving behavior derives from the optimization problem in which a representative consumer maximizes his consumption while adhering to a lifetime (intertemporal) budget constraint.† Keynesian models, by contrast, take a shortcut, assuming a fixed relationship between saving and national income. However, these models shed limited light on the classical questions of macroeconomics: Why are there economic booms and recessions? What generates unemployment? What roles can fiscal and monetary policy play in stabilizing the economy? In trying to render their models tractable, economists neglected many important aspects of the real world. In particular, they assumed away imperfections and frictions in markets for labor, capital, and goods. The ups and downs of the economy were ascribed to exogenous and vague “shocks” to technology and consumer preferences. The unemployed weren’t looking for jobs they couldn’t find; they represented a worker’s optimal trade-off between leisure and labor. Perhaps unsurprisingly, these models were poor forecasters of major macroeconomic variables such as inflation and growth.8 As long as the economy hummed along at a steady clip and unemployment was low, these shortcomings were not particularly evident. But their failures become more apparent and costly in the aftermath of the financial crisis of 2008–9. These newfangled models simply could not explain the magnitude and duration of the recession that followed. They needed, at the very least, to incorporate more realism about financial-market imperfections. Traditional Keynesian models, despite their lack of microfoundations, could explain how economies can get stuck with high unemployment and seemed more relevant than ever. Yet the advocates of the new models were reluctant to give up on them—not because these models did a better job of tracking reality, but because they were what models were supposed to look like. Their modeling strategy trumped the realism of conclusions. Economists’ attachment to particular modeling conventions—rational, forward-looking individuals, well-functioning markets, and so on—often leads them to overlook obvious conflicts with the world around them.
”
”