Robert J. Gordon writes,

The evolution of the economy after 2000 was, of course, entirely different than after 1929, and we have previously attributed this to the aggressive easing of monetary policy that sustained a major boom in residential construction and in sales of consumer durables sufficient largely to offset the decline of investment in equipment and software. A second difference was the aggressive easing of fiscal policy in 2001-03 through a succession of reductions in Federal income tax rates, including the tax rates on capital gains and dividends. A third major difference was that equities could be margined up to 90 percent in the late 1920s, compared to 50 percent in the 1990s, raising both the level of speculative frenzy in 1927-29 and the extent of wealth destruction when the crash finally came.

I found the entire paper fascinating. I think that both periods represent interesting examples of the Kindleberger theory of booms related to displacement. I think that the question of what made the economy less unstable in 2000 than in 1930 is an important one, and Gordon’s answer is plausible (he looks at more than simply fiscal and monetary policy, although that is what I quoted above).

I really think that empirical macroeconomics ought to be economic history. The benefit-cost ratio of econometric equations in macro has been quite low, in my opinion.