Steve Williamson has a new post on NGDP targeting that gets off to a bad start:
Nominal GDP (NGDP) targeting as a monetary policy rule was first proposed in the 1980s, the most prominent proponent being Bennett McCallum.
A number of interwar economists, including Hayek, advocated NGDP targeting. (Actually nominal income targeting, as the term ‘NGDP’ was not yet widely used.) People should read George Selgin’s work on the history of NGDP targeting.
Then things get much better, as Williamson has an excellent discussion of the evolution of the NGDP. And then I start to have some problems with his analysis:
One might imagine that Sumner would have the Fed conform to its existing operating procedure and move the fed funds rate target – Taylor rule fashion – in response to current information on where NGDP is relative to its target. Not so. Sumner’s recommendation is that we create a market in claims contingent on future NGDP – a NGDP futures market – and that the Fed then conduct open market operations to achieve a target for the price of one of these claims.
Imagine buying a boat and being told by the sales person that the steering works fine 99% of the time, and only fails in wild and raging typhoons. I don’t know about you, but a stormy day is when I most want the steering mechanism to work. And it’s during periods where NGDP falls far below trend (1932 and 2009) where I most want my monetary policy instrument to work. Given rates are at zero, and we need more stimulus, it seems an odd time to question my dissent from interest rate targeting orthodoxy.
More importantly, I rarely discuss NGDP futures targeting in my blog, as I know it’s not going to happen anytime soon. I certainly have plenty of advice on other techniques the Fed could use. But later in the post Williamson makes this questionable claim:
The only policy instrument that currently matters is the interest rate on reserves (IROR).
I can think of several other options. For instance, the Fed could do lots more QE. The purpose of QE is to raise inflation and/or NGDP growth expectations. I’m sure there are lots of models where QE doesn’t work, but in the real world is does work. Fed policy statements hinting at QE had an obvious and unmistakable effect on asset prices in late 2010, including TIPS spreads. Now you might argue that market participants were foolish to believe in QE (as Krugman once seemed to imply.) But even Krugman eventually conceded that higher inflation expectations are higher inflation expectations, and it doesn’t much matter whether they occur for the right or wrong reason. In fairness, even I have some doubts about whether QE is the best way to go. It’s a clumsy tool. And although (I believe) I was the first to publish a paper advocating negative IOR, I also have doubts about whether that’s the appropriate policy. Instead, I think by far the best option is to set a NGDP target, and more importantly, to engage in level targeting. This sort of policy announcement would immediately boost NGDP growth expectations (or reduce real interest rates if you prefer to model things that way.) If the Fed had done that back in early 2008 we never would have hit the zero bound in the first place, as expected NGDP growth would have been far higher, and nominal interest rates are strongly impacted by changes in expected NGDP growth (but also deviations from trend.) That’s why countries like Australia don’t have to worry about the zero rate trap, they keep NGDP growing fast enough to avoid it.
So, what do we make of this? If achieving a NGDP target is a good thing, then variability about trend in NGDP must be bad. So how have we been doing? The first chart shows HP-filtered nominal and real GDP for the US. You’re looking at percentage deviations from trend in the two time series. The variability of NGDP about trend has been substantial in the post-1947 period – basically on the order of variability about trend in real GDP. You’ll note that the two HP-filtered time series in the chart follow each other closely. If we were to judge past monetary policy performance by variability in NGDP, that performance would appear to be poor. What’s that tell you? It will be a cold day in hell when the Fed adopts NGDP targeting. Just as the Fed likes the Taylor rule, as it confirms the Fed’s belief in the wisdom of its own actions, the Fed will not buy into a policy rule that makes its previous actions look stupid.
That makes no sense to me. During the 1980s and 1990s the Fed moved toward a rule (call it the Taylor Rule, inflation targeting, whatever) that made their previous high inflation policy of the 1970s look stupid. In fact if you speak to Fed officials they love to tell you how misguided Fed policy was during the Great Inflation, and how they’ve done a much better job since about 1983.
There’s another interesting feature of the first chart. Note that, during the 1970s, variability in NGDP about trend was considerably smaller than for real GDP. But after the 1981-82 recession and before the 2008-09 recession, detrended NGDP hugs detrended real GDP closely. But the first period is typically judged to be a period of bad monetary policy and the latter a period of good monetary policy.
I have several problems with this. It’s been a long time since I studied detrending, but I recall McCallum pointing out during the great “unit root” debate that it’s very difficult to detrend GDP. Look at how NGDP falls from about 3% above trend in 2008 to 3% below trend in 2009, a swing of 6% relative to trend. We know that NGDP fell 4% in absolute terms between 2008:2 and 2009:2, which I assumed was at least 9% below trend. His graph seems to imply that trend NGDP during that 12 month period was 2% at most (6% – 4%.) Correct me if I’m wrong, but could the HP filter approach be distorted if the dramatic collapse in NGDP after 2008 was allowed to impact estimates of trend? And I’d completely reject that claim that NGDP was as far above trend in 2008 as it was below in 2009.
But even if I’m wrong on this technical question, I don’t think he’s drawing the right conclusions. The 1970s were bad for a number of reasons. One was the unusually large supply shocks, such as the oil embargo and the off and on wage/price controls. These made output more unstable, and lower, that what would have been implied by NGDP shocks alone. And the Fed had no control over those factors. Another huge problem was the rapidly rising rate of inflation, and also inflation expectations. Evan Soltas has a new post pointing out that this dramatically raised the real tax rate on capital, and hence the stock and bond markets did very poorly. That was what most people are talking about when they blame the Fed for bad monetary policy. There was also some instability of NGDP growth, but Steve’s right that it was not far out of the ordinary. But even here I’d be careful with the data. In the early 1980s we had some severe RGDP and NGDP instability, which was due to Fed policies aimed at squeezing inflation out of the economy. This could be regarded as a legacy of the bad policies of the 1970s. Since about 1984 there has been some reduction in NGDP instability, until 2008 or course. Marcus Nunes has looked at the data much more closely than me, and can document this fact.
So, if variability about trend in NGDP is a bad thing, why should we not worry about the seasonal variability? I can’t see how any answer that the NGDP targeters would give us to that question could make any sense. But they should have a shot at it.
I think this shows just how far apart we are in terms of thinking about the problem of business cycles. I may be wrong in the answer I’m about to give, but quite frankly I’m surprised that Williamson didn’t anticipate it, indeed wonders whether there is an answer. Most mainstream economists, both Keynesian and monetarist, distinguish between optimal output fluctuations, roughly those that would occur if all wages and prices were completely flexible, and suboptimal output fluctuations, which are presumably generated by the interaction of demand shocks and wage/price stickiness. (BTW, I know the term ’demand’ is not clearly defined, I just mean NGDP shocks.) Most economists assume that season output fluctuations (in industries like construction, retailing, agriculture, etc) are efficient. The fluctuations in hours worked in those industries that are due to seasonal factors are predictable, and do not cause huge welfare losses. In contrast, RGDP fluctuations at cyclical frequencies are assumed to be suboptimal, and are assumed to generate large welfare losses—particularly when generated by huge NGDP shocks, as in the early 1930s. That may be wrong, and obviously one would need a lot of labor market modeling to formalize this conjecture, but I’m pretty sure it’s the assumption that 90% of mainstream economists have in the back of their minds. But now I’ll give Williamson “a shot” at my conjecture.
The monetary models we have to work with tell us principally that monetary policy is about managing price distortions. For example, a ubiquitous implication of monetary models is that a Friedman rule is optimal. The Friedman rule (that’s not the constant money growth rule – this comes from Friedman’s “Optimum Quantity of Money”) dictates that monetary policy be conducted so that the nominal interest rate is always zero. Of course we know that no central bank does that, and we have good reasons to think that there are other frictions in the economy which imply that we should depart from the Friedman rule. However, the lesson from the Friedman rule argument is that the nominal interest rate reflects a distortion and that, once we take account of other frictions, we should arrive at an optimal policy rule that will imply that the nominal interest rate should be smooth.
I certainly don’t agree with that, and I’m not sure Friedman would either. I had thought his argument was that a zero rate minimizes the opportunity cost of holding base money, which can be produced at zero cost. I thought he was critical of interest rate smoothing. But it’s been a long time since I read the article. In any case, once you add wage/price stickiness it’s not at all clear that stable interest rates are important. One could argue that the welfare cost of fluctuating short term risk free rates is trivial compared to the welfare costs of suboptimal employment fluctuations caused by sticky wages.
The idea that it is important to have the central bank target a NGDP futures price as part of the implementation of NGDP targeting seems both unnecessary and risky. Current central banking practice works well in the United States in part because the Fed (pre-financial crisis at least) is absorbing day-to-day, week-to-week, and month-to-month variation in financial market activity. Some of this variation is predictable – having to do with the day of the week, reserve requirement rules, or the month of the year. Some of it is unpredictable, resulting for example from shocks in the payments system. I think there are benefits to financial market participants in having a predictable overnight interest rate, though I don’t think anyone has written down a rigorous rationale for that view. Who knows what would happen in overnight markets if the Fed attempted to peg the price of NGDP futures rather than the overnight fed funds rate? I don’t have any idea, and neither does Scott Sumner. Sumner seems to think that such a procedure would add extra commitment to the policy regime. But the policy rule already implies commitment – the central bank is judged by how close it comes to the target path. What else should we want?
I have all sorts of problem here. First of all I’ve published two different papers discussing how the Fed could continue targeting the fed funds rate under a futures targeting regime, for periods of 6 weeks at a time if the Fed wishes. You set up a number of NGDP futures contracts auctions, each contingent on a different setting of the fed funds target. You announce that the contracts will only be exercised in the auction that, ex post, most nearly balances the long and short positions of NGDP futures speculators (which means the Fed takes little risk.) And then you set the fed funds rate at that level for the next 6 weeks. It’s not my favorite approach (remember the zero rate bound) but it’s doable.
Keep in mind we’ve operated for decades under other monetary regimes (such as the gold standard) which theoretically made the interest rate endogenous and hence uncontrollable. The question of whether to stabilize overnight rates for a few weeks at a time seems like a trivial footnote to me; there are much bigger issues to worry about. But yes, it can be done under NGDP futures targeting.
I have even more problem with the last part of the paragraph. If the Fed had an explicit target I wouldn’t be so single mindedly pursuing an NGDP target. Suppose they had a Charles Evans-type target minimizing the deviation of inflation from 2% and unemployment from 5.6%. In that case I’d criticize the Fed for falling absurdly short of their target, and then not lifting a finger to make up the difference. And I’d call for level targeting, to hold them accountable. Indeed while experts like Woodford (and Bernanke in pre-FedBorg days) correctly note that level targeting has all sorts of advantages at the zero rate bound, the biggest advantage is that central banks can’t go year after year saying ”oops, we fell just short of our target, we’ll try harder next time.” Imagine if the BOJ had its 1% inflation target converted to price level targeting. There’d be panic—”Oh my God we are actually going to have to raise the price level! We can’t keep stringing them along with promises.”
Williamson concludes as follows:
Making promises about future NGDP cannot help the Fed do a better job of making promises about the future path for the IROR, so NGDP targeting appears to be of no use in our current predicament.
It can help because it’s hard to make credible promises about the future path of interest rates, unless they are conditional on various macro outcomes (inflation, NGDP, etc) otherwise the price level has no anchor. Presumably Williamson is assuming the Fed has a credible target, which I’ve already argued is not the case. Since we don’t have that, a higher NGDP target can generate monetary stimulus in a less risky way than a long open-ended promise of low rates.
I’m kind of perplexed as to why Williamson calls himself a “New Monetarist” (although perhaps it’s just my petty jealousy that he got to the term first.) It used to be that the sine qua non of being a monetarist was viewing monetary policy as being about changes in the quantity of money, not interest rates. Indeed when I was at the UC in the late 1970s anyone claiming that a fiat money central bank might be unable to create inflation by printing money (even at zero rates) would be laughed at. Now you have UC profs like John Cochrane basically teaching that view. Maybe I shouldn’t be such a reactionary; after all, some market monetarist ideas are also anathema to strict monetarists. Times change and macro theory never stops evolving. Still, I’d really like to know what stylized fact made all those US economists who were dismissive of the idea that the BOJ was out of ammunition in the late 1990s, suddenly come to the conclusion that the Fed is out of ammo. Is there some fact that disproved the previous conventional wisdom that I somehow overlooked? I suppose that’s unfair to Williamson, as at least he admits that IOR still works.
PS. Mark Thoma links to Williamson, and adds this comment:
I’ll be interested to hear the responses to his questions (assuming he has more luck than David Andolfatto in getting advocates of NGDP targeting to repond).
I’m surprised by this. I had a lengthy response, and I recall some other NGDP targeters did as well.
PPS. I still have some projects to finish, so I may not post much in the near future.
HT: Bill Woolsey.