Mixed messages
Keynesians face a dilemma. If they go to Congress and say the following:
The best way out of the recession is to boost AD. The Fed can do that best, even at the zero bound. They simply need a higher target for whatever nominal aggregate they are targeting, or they can do level targeting. But the Fed won’t do those things because they don’t think we need more AD. They interpret their mandate differently than you guys do. So instead you guys need to create more AD and inflation the hard way, through spending hundreds of billions of dollars and imposing large future tax burdens on our economy.
I think it’s fair to say the message wouldn’t go over very well. So instead they say to Congress that we need fiscal stimulus because monetary policy is out of ammunition. But if they are debating with economists, or at least the dwindling number of economists who understand how monetary policy operates at the zero bound, they won’t be taken seriously propounding that sort of crude 1938 Keynesianism. Sometimes they play around with “expectations trap” models, although as I discussed here, those are even more applicable to fiscal policy. In any case, the expectations trap is a bit far-fetched, so the normal argument is; “Yes, the stubborn conservatives at the Fed, ECB and BOJ could do a lot more, but they refuse to set higher inflation targets and thus we need fiscal stimulus.”
I think this two-faced approach has really hurt their credibility. It helps explain why they aren’t being taken seriously in policy-making circles. But I don’t have any good alternatives. The economics profession is split in so many ways that almost any suggestion (including my own argument for monetary stimulus) is not going to get support from enough economists to convince the skeptics.
BTW, I also think this explains why smaller countries seem even less receptive to the fiscal stimulus argument than larger countries. The zero bound isn’t much of a problem for small countries that want to inflate—they can always depreciate their currencies. So why would they even consider fiscal stimulus? Of course small countries that don’t want to inflate (Switzerland) don’t have this option, but if they don’t want more inflation then their problem is not a lack of AD.
I got thinking about all this while reading some posts by DeLong, Cowen and Kling. Here is Tyler Cowen expressing skepticism about fiscal stimulus:
1. The monetary authority moves last anyway.
Brad DeLong responds:
Yes, the central bank can neutralize any additional fiscal stimulus by raising interest rates. (It is not clear that it can undo any fiscal contraction by some combination of lowering interest rates and quantitative easing: it may be able to.) What is clear is that the U.S. Federal Reserve and the Bank of England are right now definitely not in a place where they would neutralize any additional fiscal stimulus by raising interest rates. And my bet is that the ECB is also not in such a place–although it is much harder to figure out what they think and what they will do. That the central bank moves last is important and relevant, but not determinative when you are in the neighborhood of the zero lower bound on interest rates.
I can’t make heads or tails of this, but perhaps you commenters can set me straight as to what I am missing. As I read DeLong, he seems to concede that the Fed “may” be able to offset fiscal restraint, presumably with some unconventional policy like QE or higher inflation targets. I don’t want to put words into DeLong’s mouth, but as the Fed obviously can’t take the conventional step of cutting rates once they have fallen to zero, I don’t see any other interpretation. So let’s assume DeLong is alluding to unconventional monetary stimulus. Note that there is no zero bound for unconventional monetary stimulus. Perhaps it wouldn’t work, but it would fail for reasons other than the zero bound.
But in that case I have no idea what DeLong means by the “What is clear . . .” sentence. The argument that fiscal policy can be offset with monetary policy is not based on moving interest rates, as indeed (I believe) DeLong implicitly admitted in the previous sentence. Just as nobody thinks the Fed would respond to fiscal contraction by cutting rates to minus 1%, nobody thinks they’d raise rates to 1% the day after Congress passed a fiscal stimulus. But they don’t have to. The way monetary policy works at the zero bound is by changing inflation expectations. If the expected amount of inflation over the next 5 years doesn’t change after fiscal stimulus is announced, then the fiscal stimulus is expected to fail.
Why might it be expected to fail if the Fed didn’t raise rates right after the stimulus was passed? Simple; the Fed’s exit strategy would be adjusted. If you think this possibility is far-fetched then you haven’t been paying attention. A couple months ago all the chatter out of the Fed was about how the economy was recovering strongly, and that before too long we needed to be contemplating an exit from stimulus. (Yes, they believe they are actually being stimulative, as they haven’t read their Milton Friedman recently.) Now just two months later all the chatter is about how the exit will be delayed in 2012. What’s happened in between? The Greek and Spanish crises, the strong dollar, higher than expected unemployment claims, lower than expected private job creation, low core CPI inflation, etc, etc. In other words, the Fed most definitely does respond to signals about AD.
I know that Congress doesn’t think of stimulus in terms of inflation, but let’s suppose that the amount of stimulus Congress wants would lead to 12% (total) expected inflation over 5 years and the Fed wants to do a policy that will lead to 7% inflation over 5 years. If the Congress uses fiscal stimulus to bump up the expected inflation rate, the Fed can offset that without raising current interest rates. All they need to do is continue hinting that if AD rises to X level, they will exit their zero interest rate policy.
Now I am not saying this will definitely happen. I can think of several alternative scenarios. The Fed might be intimidated by Congress’s action, and move toward a more accommodative policy stance. Or the fiscal deficits might raise expected inflation by increasing fears that the debt will eventually be monetized. But what I don’t understand is how DeLong’s argument addresses Cowen’s observation that the central bank moves last.
The left likes to claim the fiscal stimulus “worked.” The right likes to point out that growth was less than predicted by Obama. The left responds that “other bad things” happened in late 2008 and early 2009, which explain the disappointing growth. My view is that the “other bad things” were mostly an excessively tight monetary policy in late 2008 and early 2009, which was due to Fed passivity in the face of expectations that fiscal stimulus would carry the load.
Arnold Kling also has an interesting take on this issue:
So the argument for using fiscal expansion instead of monetary expansion needs to be pretty compelling. Consider two arguments:
1. We should be happy about a larger public sector and a smaller private sector.
2. At low interest rates fiscal policy is very effective (no crowding out) and monetary policy may be ineffective (liquidity trap).
When Tyler says that (2) is a “red herring,” he is implying that it is a stalking horse for (1).
That probably plays into conservatives favoring tax cuts during recessions and liberals favoring spending increases.
Tags:
20. June 2010 at 12:33
“I can’t make heads or tails of this, but perhaps you commenters can set me straight as to what I am missing.”
It just sounds like he’s arguing that because inflation expectations are below where Bernanke and the other doves would want it, they would not neutralize a stimulus today by committing to an earlier exit in response. That could be a “maybe” if the Fed didn’t already own a TRILLION dollars of Fannie/Freddie MBS that the hawks are eager to unload ASAP.
20. June 2010 at 13:03
maybe delong believes in a kind of cowardly fed that will do the right thing only if it can stay within the space where it will not be blamed for inflation should it occur.
What I do find odd is that delong frequently makes cost benefit arguments based on where expected inflation is right now, but if deficit spending only works as stimulus by raising inflation it should at least evaluated at the inflation rate they would produce. Moreover since inflation expectations are fungible I dont think it makes a lot of sense to say something like fiscal stimulus works down here and then we can fill up inflation expectations with monetary stimulus.
It seems to me that if we are committed to returning to “normal” inflation expectations then spending should be evaluated on the cost benefit of those normal expectations, which we can get through monetary stimulus. Am I missing something?
20. June 2010 at 13:29
thruth, But then why wouldn’t the Fed adopt a more expansionary policy–say a lower interest on reserve rate? Or more QE?
e, Good points. I’ve complained that “depression economics” is misguided for very similar reasons to what you mention.
20. June 2010 at 14:03
“But then why wouldn’t the Fed adopt a more expansionary policy-say a lower interest on reserve rate? Or more QE?”
I think the narrative is that the Fed doesn’t have the political will to do any more QE. On reserves, isn’t the interest rate on reserves basically zero now? The argument against a negative rate is that banks would merely tilt their portfolios to slightly more risky assets that would not do much in the way of AD enhancement. (perhaps De Long et al imagine that after shifting to a riskier type of reserve asset, the banks would be less willing to do risky lending).
20. June 2010 at 19:19
Scott,
You believe that policy, fiscal or monetary, raises money expenditures and that this raises real output and inflation. Because of the convention of speaking about policy in terms of the impact on expected inflation, you just skip over the intervening steps and talk about policy impacting expected inflation. To you, expected inflation is an side effect of expected nominal growth.
But these neo-Keynesians have higher expected inflation making real interest rates more negative and so causing an increase in real expenditures, real output, and prices. The increase in real expenditures and prices implies an increase in money expenditures.
The higher expected inflation is needed to cause the increase in money expenditures.
But that isn’t needed with fiscal stimulus. They see fiscal stimulus working exactly like you see monetary stimulus working too.
I agree, with your point that if the Fed is really committed to keeping the price level from rising back to its past growth path, (2% inflation with no memories) then moving back up the aggregate supply curve is blocked.
I really think we just need to stick to talking about money expenditures. Inflation targeting is a mistake. The growth path of nominal expenditures should be targeted.
20. June 2010 at 21:37
‘I really think we just need to stick to talking about money expenditures. Inflation targeting is a mistake.’
I agree, Bill!
Inflation targeting has the problem that it sees effects from both demand and supply. Money_expenditures/ngdp/etc targeting is great because it seperates demand side and supply side issues automatically, because the total expenditures are volume*price.
The important part
Price targeting of all sorts miss that supply changes and demand changes when they cause the same direction of changes in price, cause opposite directional changes in volume. Money_expenditures/ngdp etc capture this. Purely looking and arguing from price doesn’t.
21. June 2010 at 07:23
thruth,
“I think the narrative is that the Fed doesn’t have the political will to do any more QE. On reserves, isn’t the interest rate on reserves basically zero now? The argument against a negative rate is that banks would merely tilt their portfolios to slightly more risky assets that would not do much in the way of AD enhancement. (perhaps De Long et al imagine that after shifting to a riskier type of reserve asset, the banks would be less willing to do risky lending).”
The rate is still higher than what banks can earn on T-bills. It doesn’t matter if they shift toward riskier assets or not. The point is to have them reduce their demand for base money–get the money out into circulation.
Bill, That is a very interesting point, and a complicated one to discuss. I would say higher expected inflation is needed in the new Keynesian model. But it is not needed in the “transmission mechanism” sense that you discuss, rather it is needed in the model consistency (aka rational expectations) sense. If the model predicts higher inflation from new Keynesian stimulus, and we both agree that it does, then you darn well better observe higher inflation expectations in the TIPS market (after an unanticipated fiscal stimulus) or else the stimulus isn’t expected to do any good.
Bill and Doc, I would love to have the term ‘inflation’ banned from economic discourse, and just talk about nominal growth. But I keep getting sucked back because everyone else talks about inflation.
21. June 2010 at 14:45
[…] In fact, the zero bound does not bind, in most cases anyway. Fiscal stimulus supporters might respond that while the zero bound does not bind, the Fed might nonetheless be reluctant to engage in the […]
21. June 2010 at 20:56
Actually whats clear is that since Scott Sumner showed up on the screen and exposed Krugman and Delong’s con game, they have been forced to retreat and tone down their screeching hysteria.
Thank the gods for Scott Sumner!
22. June 2010 at 06:06
Thanks Contemplationist, but I think you overstate my impact.
23. August 2010 at 15:56
[…] numerous previous posts I argued that the regional Fed presidents were sabotaging fiscal stimulus with all their chatter […]