Keynesians gave up by the 1970s. They saw that fiscal programs took too long to implement. They especially disparaged temporary measures, which would not stimulate the consumption that classic Keynesians thought was important to stimulus. Every undergraduate text has repeated these conclusions for at least 40 years. I learned this view from Dornbusch and Fisher’s undergraduate text, taught by Bob Solow, in the 1970s. Even the optimistic projections by the Obama economic team say that fiscal stimulus will not really kick in for two years, validating the durability of this view.
The equilibrium tradition which took over professional academic economics in the mid-1970s has even less room for fiscal stimulus. Some “equilibrium” analyses do say fiscal stimulus can increase output – but by making us feel poorer, work harder at lower wages, and consume less.8 That’s not what advocates have in mind! A large fiscal program can affect prices, wages, and interest rates with all sorts of interesting general-equilibrium implications, but these analyses haven’t really converged on anything solid, much less the large “multipliers” necessary to make traditional fiscal stimulus attractive.
More deeply, macroeconomics was revolutionized starting in the 1950s, by the realization that what people think about the future is crucial to understanding how policies work today. As I have emphasized, the effects fiscal stimulus will have now depends crucially on whether people expect the new spending to be paid back by future taxes or whether they expect it to be quickly monetized. Classic Keynesian analysis analyzed policies by treating each point in time separately. We do not have to agree if expectations are formed “rationally,” all we have to agree is that expectations of the future matter crucially for how people behave today, and the classic Keynesian analysis of fiscal stimulus falls apart.
In textbooks and graduate curriculums across the country, stimulus is presented at best as quaint history of thought with no coherent defense that one should believe it in the context of modern economics. (For example, David Romer’s classic graduate text Advanced Macroeconomics.) At worst, it is presented as a classic fallacy. (My view of the treatment in Tom Sargent’s Dynamic Macroeconomic Theory and Sargent and Ljungqvist’s Recursive Macroeconomic Theory.)
“New-Keynesian” thought is devoted to defending the importance of monetary policy, and incorporating specific frictions in the equilibrium tradition, not to rescuing the ancient view that fiscal stimulus is important and abandoning that tradition. Mike Woodford’s New-Keynesian opus Interest and Prices has no mention at all of fiscal stimulus. More deeply, new-Keynesian economics is completely devoted to the proposition that expectations of the future matter centrally for how the economy behaves today. Its central thesis is that central bankers must manage expectations, not manage “demand.” It has no room at all for the sort of analysis in which one adds up “consumption,” “investment,” and “government” demands, without considering alternatives for those demands or expectations of the future, to determine output.
These ideas changed because Keynesian economics was a failure in practice, and not just in theory. Keynes left Britain 30 years of miserable growth. Richard Nixon said, “We are all Keynesians now,” just as Keynesian policy led to the inflation and economic dislocation of the 1970s--unexpected by Keynesians but dramatically foretold by Milton Friedman’s 1968 AEA address. Keynes disdained investment, where we now all realize that saving and investment are vital to long-run growth. Keynes did not think at all about the incentives effects of taxes. He favored planning, and wrote before Hayek reminded us how modern economies cannot function without price signals. Fiscal stimulus advocates are hanging on to a last little timber from a sunken boat of ideas, ideas that everyone including they abandoned, and from hard experience. If we forget all that, we could repeat the economics of postwar Britain, of spend-and-inflate Latin America, and of bureaucratic, planned India.
There has been no grand empirical reevaluation of fiscal stimulus either. Empirical work is hard, since governments try fiscal stimulus in bad times. If you bleed with leaches when you have a cold, empirical work might say that the leaches cured you. Empirical work has to find fiscal stimulus events that were applied randomly, without regard to the state of the economy. Harder still, it has to find stimulus spending that people expected to be paid off rather than inflated away. Most current empirical work does not make this distinction, and therefore is in danger of measuring the slope of the Phillips curve rather than the fiscal multiplier. Finally, empirical work without a plausible mechanism is hard to believe. Even so, doing the best to surmount these problems, nothing in recent empirical work on US data has revised a gloomy opinion of fiscal stimulus.9 Looking across the world, large government deficits and spending programs are clearly not the keys to economic health, and evidence of stimulus effects over time in the US needs to be reconciled with this supreme lack of evidence across countries.
I've heard that so many times - almost forgot to challenge.
