We are once again live at the *Economist* by invitation: http://www.economist.com/economics/by-invitation/guest-contributions/there_absence_evidence_expansionary_case">
There is an absence of evidence for the expansionary case
Brad DeLong our guest wrote on Aug 3rd 2010, 17:54 GMT
LONG ago Martin Feldstein, Andrew Abel, Olivier Blanchard, Thomas Sargent, Charles Kindleberger, and Barry Eichengreen taught me that what we did—what made us economists doing analysis rather than ideologues hearing voices in the air or propagandists trying to please our political masters—was that we economists (i) generalised from historical patterns (ii) to build a family of self-consistent models that we then used (iii) to guide our analysis.
But I cannot fit the arguments that short-term expansionary effects will follow from transitory contractionary fiscal policy in the core economic powers—the U. S., France, Germany, and Japan—into that framework, no matter how hard I try.
Start with what I am now told that all economists believe: that contractionary monetary policy in the form of an open-market sale of government bonds for cash that raises interest rates will reduce production and employment in the short run. First, the government sells bonds for cash. Second, the greater supply of bonds on the market diminishes their prices—which means higher interest rates. Third, higher interest rates make it more costly for businesses to finance investment spending, and also reduces asset values so that households feel poorer and are less willing to spend on consumption goods and services. Fourth, the flow of spending then drops. Fifth, general-equilibrium considerations then attenuate or amplify this drop. And, sixth, this drop in spending is what diminishes production and employment in the short run.
The caveats are that perhaps the open-market operation won't raise interest rates—as in a liquidity trap, or if the open-market operation is taken as a credible signal of a change in long-run policy that diminishes risk and so makes investors more confident holding assets. Then there will be no rise in the coat of capital to businesses, no fall in household wealth, no change in either flow of spending, nothing for general equilibrium considerations to attenuate or amplify, and so no contractionary effect on production and employment.
Bit these caveats are just that—caveats: if you want to argue that they apply, you have to make the case that the current situation is a special case. You can't just wave your hands and say that nobody knows. That, at least, is how I thought the game of economic analysis was played according to the rules. And that is the analysis of the short-term effects of contractionary monetary policy
Now let's turn to fiscal policy. I am also now told that nobody knows whether a transitory cut in government purchases that finances a long-run reduction in taxes will be contractionary or not. I am now told that there are perfectly reasonable and competent economists on both sides.
But when I rerun the steps of our argument, this time for fiscal policy, I don't see that. What I see is this: First, the government cuts back on government purchases. Second, the flow of spending then drops. Third, general-equilibrium considerations then attenuate or amplify this drop. And, fourth, this drop in spending is what diminishes production and employment in the short run.
There are, once again, caveats. This time:
1. Perhaps the central bank responds by cutting interest rates substantially, which boosts spending by businesses and households and so offsets the fall in government purchases.
2. Perhaps without central bank intervention financial market reactions lead to a substantial fall in interest rates, which boosts spending by businesses and households and so offsets the fall in government purchases.
3. Perhaps there is no such thing as a transitory cut: perhaps all cuts in government purchases are and are expected to be permanent—in which case the fall in expected future taxes is an order of magnitude larger than the fall in current government purchases and it is possible that the resulting boost to private spending offsets the fall in government purchases.
It happens that right now I believe that case (3) applies to Greece and perhaps Italy, and that case (2) applies to Portugal, Italy, and perhaps Spain. And I hope without much reason that case (1) applies to Britain. It happens that two years ago I believed that case (2) would by now apply to the core economies of the North Atlantic. But right now I see no evidence and no arguments that it does.
This absence of evidence matters. For these caveats are just that—caveats: if you want to argue that they apply, you have to make the case that the current situation fits into one of these special cases. You can't just wave your hands and say that nobody knows.
Or, at least, that was how I thought the game of economic analysis was played according to the rules.