TCS Daily

Dueling Studies Square Off on Causes of Economic Growth

By Kevin Hassett - July 17, 2001 12:00 AM

In my previous entry, I argued that the economic data were beginning to suggest that the economy was turning a corner, and that a traditional recession may well have been averted. Since that time, jobless claims jumped up, suggesting that some risks are still there, but other things, retail sales in particular, lined up pretty well with the picture of the economy that I described last time. Economic growth looks low, but positive. Things feel so bad because they were so astonishingly good a short time ago.

So these bad times look like they will be better than they have ever been. Going forward, the key question is, have good and bad times ratcheted up permanently? Or did we just live through a boom and bust period that was a one-time-only experience? A new study from the National Bureau of Economic Research suggests that the arguments on the side of better times forever are stronger than ever.

In the study, "Is Growth Exogenous? Taking Mankiw, Romer, and Weil Seriously," Professors Ben Bernanke and Refet Gurkaynak of Princeton contrived a very clever experiment that may help change the way economists think about economic growth. As you can tell by its title, there is a little lingo involved in the debate, so here's a little background.

Robert Solow, the Nobel prize-winning economist, shocked the profession in 1956 when he provided an elegant proof that economic growth over long periods may well be exogenous. An exogenous variable is one that we really have no control over, such as weather. An endogenous variable is one we can affect, like government spending or taxes. Solow showed that while the economy certainly can have ups and downs, over time these may well cancel out each other. Under Solow's view, what drives economic growth in the long run is really technological change and innovation. Those advances come in such fits and starts that they may well be approximately exogenous.

Subsequently, Paul Romer of Stanford and others developed models within which growth was endogenous. The logic behind their work was pretty straightforward. Innovations occur because of investments we make. We buy computers so we can study things faster. Once we have a computer, we can find new things more quickly than we could before. The investment in the computer permanently raises the growth rate. Growth is, according to this view, endogenous.

The debate stayed mainly theoretical until 1992, when a path-breaking paper written by N. Gregory Mankiw, Romer, and Phillippe Weil demonstrated that Solow's growth model described the variation in growth across countries fairly well once one controlled for the level of skill of the workers in a given country. Ever since I've read their paper, I have felt strongly that the case against the endogenous growth models is fairly strong, if not 100 percent convincing.

Enter Bernanke and Gurkaynak. They update the Mankiw, Romer and Weil paper by extending the Solow model to also include endogenous growth aspects. Having done that, they then ran a horserace. Sure, the Solow model fits pretty well, but can we improve the fit by using the endogenous growth models to influence our specification? Solow is good, but does he really beat Paul Romer?

To perform their test, Bernanke and Gurkaynak gathered data from many countries and let the horses run. Their conclusion: "...long run growth is significantly correlated with behavioral variables...and ...this correlation is not easily explained by models in which growth is treated as the exogenous variable."

In other words, there is more than blind luck to economic growth in the long run.

The approach adopted by Bernanke and Gurkaynak will certainly launch a thousand papers and their conclusions may well end up being overturned. But if they are right, here is what it means for us today.

We have been investing an enormous amount of resources in capital and education. These investments permanently increase the level at which we innovate, and this permanently raises the level of growth around which our economy fluctuates. We may have just experienced a new economy recession with low but positive growth, but that does not mean that the high growth of the past few years is a thing of the past. Our investments, Bernanke and Gurkaynak's work suggests, will help the country return to a higher growth path quickly.

TCS Daily Archives