Believe those who are seeking the truth. Doubt those who find it. Andre Gide

Wednesday, September 4, 2013

The magic fairy NGDP wand

It's been a good day of blogging. And it seems like the day has not ended yet. Scott Sumner posts his most recent reply to me here: Second Reply to Andolfatto.

Alright, time to take a step back and put things in some perspective. What exactly is my beef with NGDP targeting anyway?

The somewhat surprising answer is: nothing, per se. I am, however, somewhat perplexed with many of those who strongly advocate NGDP targeting. I believe they are overstating the case for NGDP targeting. And in doing so, I believe that they are diverting attention away from the real economic and political forces that are potentially holding growth back. It may be comforting to believe that most of our major economic problems can be solved by having the Fed simply wave a magic fairy NGDP wand. Yes, well, I'm sorry, but I remain skeptical.

Back in April 2012, I asked David Beckworth to provide me with what he viewed as a theoretical foundation for NGDP targeting (see here). David pointed me to a very nice paper by Evan Koenig, which emphasized the role of nominal (unindexed) debt.

That led me to believe that the rationale for NGDP targeting rested on the idea that such a policy would smooth out shocks to the price-level in a way that inflation-targeting would not. This led me to write down a simple OLG model with nominal debt here and here. What I found was a case for stabilizing the price-level, but not necessarily the NGDP.

Now why did I find this result and why does it seem to contradict the prescription offered by Scott and others? I think I have traced the source of the discrepancy.

Scott et. al. like to organize their thinking around a static textbook AS-AD model. The friction is some sort of nominal price rigidity. Consider a negative AD shock. That has the effect of depressing the P and increasing the real debt burden (e.g., in the context of a sticky nominal wage, it increasing the real wage bill for the business sector, leading to layoffs, and a decline in Y). An NGDP target would stabilize NGDP by stabilizing both P and Y. O.K., good.

Now consider a negative AS shock. This causes P to rise and Y to fall. As NGDP (=PY) may not be very much changed, the policy would advocate little if any intervention. However, a strict PL target would require reducing P. Reducing P would mean increasing real debt burdens, which would further decrease Y. Conclusion: a PL target is destabilizing.

Now, I know that Scott finds mathematical models "annoying" (to me, this is like saying one finds musical scores annoying and that one can always play music better by ear). But please, go take a look at my model (academic economists should find it very simple and straightforward).

My model is explicitly dynamic--you know--kind of like the way reality is. The shock I consider does not fit easily into either the AS or AD category. The event is a "bad news" shock--a downward revision in the forecast of future capital return (or the after-tax return to capital). The impact effect of the shock is like a negative AD shock, because it depresses the demand for investment (and leads to a flight to government money/debt, which causes a surprise decline in P). The contemporaneous AS remains unaffected.

On the other hand, the decline in current capital spending manifests itself as a smaller future productive capital stock, so the that future real GDP is expected to decline. (Whether it actually declines depends on the realization of the shock: it may be either higher or lower than expected). So in this sense, my shock looks like a negative AS shock too.

Now, let's imagine that this bad news is persistent. Then absent intervention, P remains depressed and Y remains depressed. Moreover, because debt is not indexed, and because a surprise drop in P hurts the initial group of investors in my model, the amount of capital investment that occurs on impact is too low (relative to a world in which debt could have been indexed). The PL target rule corrects this inefficient reallocation of purchasing power (away from investors, toward consumers, in my model).

What is the effect of targeting NGDP in my model? To stabilize NGDP, capital spending has to be stabilized and, to the extent that future productivity is lower (in according with earlier expectation), capital spending has to be increased. We can obviously think of policies that achieve this result. In fact, my model is consistent with the proposition that higher inflation leads to higher output (via a Tobin effect). But whether such a policy is desirable is open to debate. In particular, is it really a good idea to encourage more investment in a sector (e.g., housing) in a sector where the returns have fallen? Moreover, people are heterogeneous (my model takes this into account). There would be winners and losers. I don't here very much talk about this prospect from the NGDP targeting crowd.

So there you have it. Please stop telling me that NGDP targeting is "obviously" and unambiguously a good thing. I agree that it is -- if you insist on organizing your thinking with Econ 101 tools. And you know what? Maybe this toolkit is the correct toolkit to use. But again, forgive me if I just do not see this as obvious. This is supposed to be science, not religion.

Finally, maybe someone can speak on the following issue (see here). There is a difference between wishing for a NGDP target before the crisis, and wishing for the policy to be implemented right now. I have some sympathy for the idea that it would have been nice to have the policy in place earlier (I think it would have largely been innocuous). But I am having a harder time seeing the benefits of such a policy imposed right now by the Fed with the tools it has available. In particular, even if "debt overhang" was a major drag on the economy following the end of the most recent recession, what evidence do we have that it is still a *major* force holding the recovery back (especially, as I pointed out in my earlier posts, the price level seems to be close to its long-run trend path). What is the mechanism that people have in mind?

Thanks very much to all of you who have commented and pointed me to readings. I don't always have the time to get to them, owing to the demands on my time here at work, but I do appreciate it! And thanks to Scott for his thoughtful replies to my posts. It's been a fun discussion.


  1. David, been busy and hope to respond to you and Miles Kimball in the future. For now, here a theory paper on NGDP level targeting and its implications for debt contracts.

    1. Thanks, David. Looking forward to a good discussion.

  2. This is not a formal argument, however, Lars Svensson just released a paper showing that tightening monetary policy increases leverage (at least temporarily):

    “Leaning Against the Wind” Leads to a Higher (Not Lower) Household Debt-to-GDP Ratio
    Lars E.O. Svensson
    August 2013

    "“Leaning against the wind” — a tighter monetary policy than necessary for stabilizing inflation around the inflation target and unemployment around a long-run sustainable rate — has been justified as a way of reducing household indebtedness. But it actually has the opposite effect; it leads to higher real household debt and a higher household debt-to-GDP ratio. The reason is that a tighter policy than a baseline induces a very slow fall relative to the baseline of total nominal (mortgage) debt but a faster fall in the nominal price level and nominal GDP. There is then first a rise in real debt and the debt-to-GDP ratio relative to the baseline, a rise that is almost as large and as fast as the fall in the price level and nominal GDP. Then, real debt and the debt-to-GDP ratio very slowly fall back to the baseline, something that could take more than a decade. Therefore, “leaning against the wind” as a way of reducing the household debt-to-GDP ratio is counterproductive."

    It's not that difficult to imagine a scenario where an economy is hit by the third largest negative AS shock since WW II:

    And the central bank passively tightens monetary policy in order to stabilize the price level causing leverage to rise high enough to cause a financial crisis.

    1. Mark...did you mean to say AS shock?

    2. Yes, the Robert Gordon "food-energy effect" (FEE) peaked at 1.9% in July 2008. This is the third largest year on year FEE after the 1974 and 1980 commodity price shocks respectively.

    3. Alright, good. Sure, take a look at my original post and you see the sharp spike in energy prices.

      So do I take it that you disagree with the "AD shock" interpretations of what has happened since 2008?

    4. "So do I take it that you disagree with the "AD shock" interpretations of what has happened since 2008?"

      Not at all, NGDP (AD) obviously fell below trend.

      But more importantly I'm hardly the only person to have made the observation that the negative AD shock was preceded by a negative AS shock in 2008. At least Scott Sumner and Marcus Nunes have explicitly stated as much.

      This is why Inflation Targeting (IT) or Price Level Targeting (PLT) can be so hugely destabilizing. Anytime a negative AS shock occurs the central bank tightenss policy leading to a negative AD shock. Thus real output declines not only in response to the negative AS shock but also the negative AD shock. And of course this is fully symmetric. Positive AS shocks will cause a central bank practicing IT or PLT to increase AD leading to excessive booms.

      Incidentally, the advantage of NGDPLT in handling AS shocks over IT or PLT has been repeatedly emphasized by Jeffrey Frankel and many other advocates. I'm puzzled that you seemingly have only just noticed that this is probably the main argument in its favor.

    5. Mark,

      Yes, OK. But from an operational point of view, the Fed does not let its policy be governed by short-run movements in the price-level that are perceived to be transitory (hence it's focus on core PCE). So the adverse supply shock you allude to above need not result in tightening under a PLT.

      As for your concluding paragraph, well maybe I'm just slow.

      Actually, I think the gist of my argument has been lost. It is this. Suppose a negative AD shock. Bad. Why? Nominal debt, debt overhang. Solution? Stabilize price level. What happened? Price level measures are stable. And so, now what? You want higher NGDP too? On what grounds? The NGDP target was meant to mitigate the adverse consequences of nominal rigidities. That has been accomplished (if you take the price level measures seriously).

    6. David, in practice is not always easy for the Fed to distinguish transitory vs. non-transitory changes. See the September 2008 FOMC decision. Fed put a lot of weight on inflation (despite being in the midst of imploding economy)and thus failed to lower federal funds rate. As you note, though, this inflation surge turned out to be temporary. A NGDP targeting CB would not respond this way.

    7. "Actually, I think the gist of my argument has been lost. It is this. Suppose a negative AD shock. Bad. Why? Nominal debt, debt overhang. Solution? Stabilize price level. What happened? Price level measures are stable. And so, now what? You want higher NGDP too? On what grounds?"

      Traditionally the burden of nominal debt is measured with respect to nominal income (i.e. leverage) not with respect to the price level. For example, see the above paper by Lars Svensson.

      In your example the price level falls because of a negative shock to AD. If the CB has policy of stabilizing the price level it restores the price level by increasing AD back to its previous level. This results in both the the price level and the nominal income level being returned to their previous levels. In this scenario the effects of PLT and NGDPLT are identical.

      Imagine instead a negative shock to AS. The price level rises but initially nominal income remains the same. If the CB is practicing PLT it reduces AD to restore the price level and this reduces nominal income. The real value of debt remains unchanged but the ratio of debt to nominal income rises. The burden of debt increases. This is destabilizing.

      If instead the CB is practicing NGDPLT it does nothing in response to a negative AS shock. The price level increases but nominal income remains the same. True, the real value of debt falls, but since the ratio of debt to nominal income does not change the burden of debt remains the same. This is stabilizing.

    8. Mark:

      Yes, I understand. I guess that I am just questioning the quantitative importance of what appeared to be a transitory AS shock (together with the fact that the Fed did not tighten, given its focus on core PCE).

      What we have now, according to standard interpretation, is a shortfall in AD. And as you say above, PLT and NGDPT accomplish the same thing in this event. So I repeat, if this is true, then what is the case for jacking up NGDP at this stage?

    9. "Yes, I understand. I guess that I am just questioning the quantitative importance of what appeared to be a transitory AS shock (together with the fact that the Fed did not tighten, given its focus on core PCE)."

      By what standards? The only coherent way of measuring whether a policy that mainly works through AD is expansionary or contractionary is the rate of change in AD (NGDP) relative to trend.

      NGDP growth averaged 5.5% from 1997Q1 through 2005Q4, dates at which the unemployment rate was equal to NROU. In 2003Q4 through 2006Q2 yoy NGDP growth was 6.3% to 7.1%. It slowed to 5.1% to 5.3% in 2006Q3/2006Q4 and to 4.3% to 4.8% in 2007. So the rate of change of NGDP was consistently below trend from 2006Q3 on forward. And it was significantly below trend throughout 2007:

      And in 2008Q1 it fell to 3.1%, in 2008Q2 it fell to 2.7%, and in 2008Q3 it fell to 1.9%. By the time Lehmans Brothers filed for bankruptcy the yoy NGDP growth rate had been below trend for over two years and was down to one third its trend rate.

      If your eye was on the wrong target, namely core the PCEPI inflation rate everything looked terrific. Yoy core PCEPI was consistently between 1.6% and 2.4% from January 2004 through November 2008. Yoy headline PCEPI was in the 2.5% to 4.2% range from September 2007 through October 2008, which encouraged FOMC hawks to advise putting the monetary brakes on despite the fact that NGDP had been wailing alarm bells with increasingly louder volume for two years.

      "What we have now, according to standard interpretation, is a shortfall in AD. And as you say above, PLT and NGDPT accomplish the same thing in this event. So I repeat, if this is true, then what is the case for jacking up NGDP at this stage?"

      Standard measures of factor utilization, such as the unemployment rate relative to NROU, and the CBO's estimate of potential GDP, suggest that employment and real output could be much higher without sparking a significant rise in the inflation rate. Furthermore empirical estimates and major models suggest that an increase in NGDP sufficient to close the unemployment and output gap would only result in an increase in the price level one third as great.

    10. By what standards? The only coherent way of measuring whether a policy that mainly works through AD is expansionary or contractionary is the rate of change in AD (NGDP) relative to trend.

      I honestly have no idea what you are talking about. I'm going to read that Koenig paper. Maybe it will help.

    11. Here’s a very simple way to frame this.

      The following link is to a dynamic AD-AS diagram, and which can be found in “Modern Principles: Macroeconomics” by Tyler Cowen and Alex Tabarrok:

      This is the elementary picture I am thinking about when I think about monetary policy.

      You’ll note that the rate of change in the aggregate demand (AD) curve is equal to the sum of the inflation rate and the rate of change in real GDP (RGDP), and so is precisely equal to the rate of change in nominal GDP (NGDP).

      The central bank affects the economy by setting the position of the AD curve. The central bank can target the inflation rate by setting the level of AD. By focusing on core inflation it seeks to strip away most of the effects of shifts in the Short Run AS (SRAS) curve. But regardless of what is the central bank's target, and the other forces affecting the AD curve, I view a shift in the AD curve to the right to be "expansionary" and a shift to the left to be "contractionary" monetary policy.

      From 1997-2005 the AD curve was set such that NGDP increased at an average rate of 5.5%. It fell below this rate from 2006Q3 on, and its rate of decline picked up speed in 2008. This constitutes a steady shift in the AD curve to the left, so I view that as contractionary monetary policy regardless of what the inflation rate was.

    12. 1) Is the "AD curve" in that diagram simply the equation of exchange, MV = PY? If so, then this is just an identity; it is empty of theoretical content.

      2) What does the AD curve look like in a "cashless" economy (or as an economy approaches a cashless economy?)

      3) Assuming we settle on an empirical counterpart for M, why shouldn't external shocks causing a shift in V shift the "AD curve?" And, in particular, if M is better interpreted as government liabilities (including treasuries), I'm not sure what sort of control a central bank is supposed to have over "AD."

      4) Consider a neoclassical model absent any nominal frictions. Imagine a negative productivity shock. Then it is possible for V to change (via change in interest rate). And you would ascribe any change in NGDP as a tightening or loosening of MP. Just makes no sense to me.

      Just a few questions that pop to mind.

    13. 1) That's actually my least favorite part of Cowen and Tabarrok's dynamic AD-AS Model diagram.

      2) "Cashless" implies barter to me. As Caroline Humphrey in the definitive anthropological work on barter put it, “No example of a barter economy, pure and simple, has ever been described, let alone the emergence from it of money; all available ethnography suggests that there never has been such a thing.” ("Barter and Economic Disintegration," Man 20.1, 1984, pp. 48–72) However, there are economies which are dominated by barter. For example among the Lhomi of north-east Nepal, although the exchange of common produce, as opposed to rare valuables, is most likely to approximate to a notional "equilibrium price", the practice of barter with no established measures of weight and volume means that there can be no underlying index of value/numeraire. So I don't think we can talk about an AD curve in such a case. But are the Lhomi of north-east Nepal truly relevant?

      3) The velocity of money is of course unstable. But who's talking about targeting M?

      As for monetary aggregates that include government liabilities, in the US case the only such aggregates were M4 and L, which included T-bills, and I believe they were only officially calculated from 1975 through 1980. Only the Center for Financial Stability still estimates an M4, but it's a Divisia measure so it's hard to say how important the T-bill component is.

      I estimate that a simple sum M4 would be equal to about $19.1 trillion today. Since the outstanding amount of T-bills is about $1.56 trillion, this means government liabilities would be about 8.2% of a simple sum M4, and an even smaller percent of L, were these aggregates still in existence. So given how small a component T-bills are of M4 and L, I don't think that presents much of an impediment over the Fed's ability to affect the size of those aggregates.

      4) While its easy to imagine a neoclassical model without nominal frictions it's hard to imagine an real life economy without one. It makes no sense to me to imagine hypotheticals that have no correspondence to reality.

    14. I see from your latest post that you mean "cashless" in a Woodfordian sense. Thus, although my initial response is mostly oblique, the gist of it is still the same. As long as there is a index of value/numeraire we can still talk about a standard AD curve.

    15. David:

      You say, what is the point of raising nominal GDP "at this stage."

      If the price level falls because demand falls, increasing demand back to where it started should get the price level back to where it started.

      If the price level rises because supply falls, then nominal GDP targeting leaves nominal GDP the same. It doesn't increase it. The price rises and the quantity falls.

      In your model, the price level falls because the demand for capital goods falls. If this somehow reduces the demand for consumer goods too (which I don't think it does,) then their prices fall too. To get the price level back up, the demand for capital goods and/or consumer goods must rise so that either the price of capital goods recover or the price of consumer goods rises enough to offset the lower capital goods prices.

      This happens when the demand for consumer goods and capital goods both rise. Either the demand for capital goods rises back to the initial level, or the demand for consumer goods rises enough to offset a decrease in the demand for capital goods.

      Now, where nominal GDP targeting is relevant is next period. If the productivity of the capital goods really is lower, then there will be fewer consumer goods, and if nominal GDP remains on target, then the price of the consumer goods rise next period.

      But nominal GDP isn't being increased. It is staying the same. It is that there aren't many consumer goods produced.

      Given the nominal interest rate the period before, the higher prices of consumer goods makes buying the capital goods more attractive. Of course, in the period before, a lower nominal interest rate on the government debt/money would also motivate people to purchase the capital goods.

      Now, you say, is it really a good idea to purchase these capital goods. Why produce more houses when we already have too many.

      What is the opportunity cost of producing these capital goods? It is true that they are expected to be less productive, but don't the higher prices of consumer goods next period (when there will be less of them) properly signal the need for them at that time?

  3. Nick Rowe responds:

  4. Okay, so let's say the NGDP level targeting was employed. From my reading, this would suggest the central bank should have increased money supply even more than it already did. But how would that fix the issue of banks not lending all those excess reserves? Wouldn't it just end up with more excess reserves than already are in place?

  5. David,
    The risk of real income deviations from trend is captured in the credit risk premium. The effect of price level deviations from trend is captured by the term premium. Therefore:

    1) PLT reduces the term premium over a cycle

    2) NGDPLT reduces the credit risk premium over a cycle, but it may also increase the term premium.

    It is not clear in 2) that the net effect is a reduction in risk premia.