Scott Sumner, responding to a post by Tyler Cowen, says:
Expectations traps: They’re even more applicable to fiscal policy, by Scott Sumner: Tyler Cowen links to this post from Mark Thoma:
As for Tyler’s (and others’) call for monetary policy instead of fiscal policy, here’s the problem. It relies upon changing expectations of future inflation (which changes the real interest rate). You have to get people to believe that the Fed will actually be willing to create inflation in the future when it comes time to do so. However, it’s unlikely that it will be optimal for the Fed to cause inflation when the time comes. Because of that, the best policy is to promise that you’ll create inflation, then renege on the promise when it comes time to follow through. Since people know that, and expect the Fed will not actually carry through, it’s hard to get them to change their expectations now. All that credibility the Fed has built up and protected concerning their inflation fighting credentials works against them here.
Paul Krugman developed the idea of an expectations trap as a way of explaining the dilemma faced by the Bank of Japan. Except there is just one problem. Almost everyone agrees that Japan does not face an expectations trap. They can devalue the yen whenever they wish, as much as they wish. ...
But here’s the bigger flaw with the whole expectations trap argument. People think it applies to monetary policy, but they forget it applies equally to fiscal policy. (Indeed I never realized this until today.) Here’s why. Krugman’s model relies on rational expectations, indeed you can’t get the expectations trap without ratex. But if you have ratex in your model, then no policy can work unless it is expected to work. ...
This is relatively easy to dispense with. First, those of us who pushed for fiscal policy were told we were basing this on old-fashioned models, IS-LM at best, maybe even more outdated than that. So the New Keynesian theorists went to work and showed that within the modern models used for policy analysis, fiscal policy does, in fact, find support.
The model I've been using in particular is Eggertsson's and to some extent Woodford's. The difference is that Eggertsson looks at how multipliers vary across various tax and government spending policies, while Woodford is more concerned with the determinants of the size of the government spending multiplier. For example, Woodford says:
Much public discussion of this issue has been based on old-fashioned models (both Keynesian and anti-Keynesian) that take little account of the role of intertemporal optimization and expectations in the determination of aggregate economic activity. Yet discussions of monetary stabilization policy over the past several decades have been transformed by the development of a new generation of macroeconomic models that simultaneously consider the dynamic implications of intertemporal optimization on the one hand, and delays in the adjustment of wages and prices on the other. The implications of these models for fiscal stabilization policy have been much less fully developed than their implications for monetary policy. But this is not because the models do not have implications for fiscal policy. The present paper reviews some of these implications for one specific question of current interest: the determinants of the size of the effect on aggregate output of an increase in government purchases, or what has been known since Keynes (1936) as the government expenditure "multiplier."
Going back to the creditability issue raised by Scott Sumner, one of the points that Eggertsson makes is that government spending does not have the credibility problem that plagues monetary policy. He says:
As shown by several authors, such as Eggertsson and Woodford (2003) and Auerbach and Obstfeld (2005), it is only the expectation about future money supply (once the zero bound is no longer binding) that matters ... when the interest rate is zero. ... Expansionary monetary policy can be difficult if the central bank cannot commit to future policy. The problem is that an inflation promise is not credible for a discretionary policy maker. ...
This credibility problem is what Eggertsson (2006) calls the "deflation bias" of discretionary monetary policy at zero interest rates. Government spending does not have this problem. ... The intuition is that fiscal policy not only requires promises about what the government will do in the future, but also involves direct actions today. And those actions are fully consistent with those the government promises in the future (namely, increasing government spending throughout the recession period). ...
Even so, monetary policy might still work, it's a matter of being able to credibly commit to future inflation:
It seems quite likely that, in practice, a central bank with a high degree of credibility, can make credible announcements about its future policy and thereby have considerable effect on expectations. Moreover, many authors have analyzed explicit steps, such as expanding the central bank balance sheet through purchases of various assets such as foreign exchange, mortgage-backed securities, or equities, that can help make an inflationary pledge more credible (see, e.g., Eggertsson (2006), who shows this in the context of an optimizing government, and Jeanne and Svensson (2004), who extend the analysis to show formally that an independent central bank that cares about its balance sheet can also use real asset purchases as a commitment device). Finally, if the government accumulates large amounts of nominal debt, this, too, can be helpful in making an inflation pledge credible. However, the assumption of no credible commitment by the central bank, as implied by the benchmark policy rule here, is a useful benchmark for studying the usefulness of fiscal policy.
I think the assumption that the Fed cannot credibly commit to future inflation is a relevant benchmark in the present case since I am not sure that people believe that the government will actually create inflation in the future even if they promise to do so now. As I said in the original post, I think the inflation fighting credential the Fed has worked so hard to earn work against them in this instance. My point was that I didn't want to put all of my faith in one policy instrument -- monetary policy -- when theory and experience says that fiscal policy is the superior policy tool at the zero bound. Monetary policy alone might work, but again, if fiscal policy is available and these uncertainties exist, why take a chance? Why not use fiscal policy as well?
More generally, if people want to go back and use older or different models to make their arguments, that is fine, but it doesn't have a lot to do with the argument I was making. Perhaps they believe these models are superior to the models that Woodford and Eggertsson are using, that's certainly their prerogative, but which model provides better answers to the questions we are asking is a completely different argument. For the most part, the issues being raised about credibility have been considered and addressed within the New Keynesian framework.
Update: See Paul Krugman's comments, Policy And The Tinkerbell Principle.