Tag: Macroeconomics

Must-read: Martin Sandbu: “Free Lunch: On Models and Making Policy”

Posted on by Brad DeLong

Must-Read: Superb from the extremely sharp Martin Sandbu! Only three quibbles:

  1. There are indeed “three great economists” in the mix here, but their names are Summers, Krugman, and Blanchard…
  2. This isn’t really a conversation that would have taken place even in an academic setting. If I have ever been in the same room at the same time with Larry, Paul, and Olivier–let alone all of Olivier’s coauthors, Michael Woodford, Danny Vinit, and Lukasz Rachel and Thomas Smith–I cannot remember it. And discussions and exchanges in scholarly journal articles are formal and rigid in an unhelpful way.
  3. Do note that Keynes was on Summers’s side with respect to the importance of maintaining business confidence: cf.: General Theory, ch. 12, “The State of Long-Term Expectation”

Martin Sandbu: Free Lunch: On Models and Making Policy: “The internet has… open[ed] up to the public…

…discussions… that previously took place mostly in face-to-face gatherings or scholarly journal articles. Neither medium was particularly accessible….

Summers posted a characteristically succinct statement on why he disagreed with the Federal Reserve’s decision to begin tightening… His analysis is well worth reading in full, but the trigger of the ensuing debate was his explanation for why the Fed thinks differently: ‘I suspect it is because of an excessive commitment to existing models and modes of thought. Usually it takes disaster to shatter orthodoxy.’… DeLong expressed doubts that the Fed’s analysis was indeed compatible with existing models; Krugman asserted that conventional models straightforwardly showed the Fed to be in the wrong, and that… policy was driven… by… ‘a conviction that you and your colleagues know more than is in the textbooks’….

Summers then responded… showed a fascinating divergence…. DeLong and Krugman think the Fed erred by ignoring… models…. Summers thinks the Fed erred by ignoring things that such models do not capture…. Summers is also much more comfortable with the notion that policymakers should aim to underpin market confidence. That notion has often been derided by Krugman…. Two quotes rather nicely capture the methodological disagreement here. Summers writes: ‘I think maintaining confidence is an important part of the art of policy…. Paul is certainly correct in his model but I doubt that he is in fact.’ DeLong responds: ‘Larry, however, says: We know things that are not in the model. Those things make Paul’s claim wrong. My problem with Larry is that I am not sure what those things are.’…

What is a policymaker to do if she thinks this is the case in reality, even if no extant model captures it? Surely not wait for 30 years in the hope that new mathematical techniques enable economists to model that reality. In his willingness to listen to those who may have an untheoretical ‘feel’ for the market, and in his intellectual respect for the limits of his own knowledge, Summers comes across as the most Keynesian of these three…

Brad DeLong and Jan Hatzius on the macroeconomic situation

Posted on by Brad DeLong

An interview I did last fall with Jan Hatzius:

Brad DeLong is a professor of economics at UC Berkeley, where his research focuses on financial crises and 20th century macroeconomics, as well as the political economy of monetary and fiscal policy. He has taught at Harvard University and served as Deputy Assistant Secretary of the Treasury for Economic Policy under the Clinton Administration. Below, he and Goldman Sachs Chief Economist Jan Hatzius discuss risks around liftoff and the structural downshift in rates.

The views stated by Brad DeLong herein are those of the interviewee, and do not necessarily reflect those of Goldman Sachs:

Allison Nathan: Has the US economy recovered from the Great Recession?

Brad DeLong: Yes and no.

It has not recovered from the large loss in productivity and potential output. The Great Recession knocked down our level of output by 8% compared to where we thought it would be now in 2007. We will probably never recover that loss. That is very unusual for the United States, which had a substantial recovery in lost output even after the Great Depression.

But as far as the labor market, we have mostly recovered–but not fully recovered.

Jan Hatzius: I believe that the labor market is mostly recovered and that the overall economy has come a long way toward recovery. To Brad’s point about disappointing output, the question is how much of that is exogenous weakness versus some form of hysteresis, with the effects of the Great Recession weighing progressively on economic activity. It’s difficult to know the answer, but I’ve become more sympathetic to the idea that we underestimated the extent of the exogenous slowdown. We’ve substantially revised our views on potential growth, and I don’t think it is all or even mostly due to the aftereffects of the Great Recession.

Allison Nathan: Has the time come to raise rates?

Jan Hatzius: Given how far the funds rate is below normal, how close the economy is to full employment, and my expectation of gradual increases in wage and price inflation, now seems like an appropriate time to move. But I would say there is still a good case for waiting on risk-management grounds because the future of the economy is uncertain. It seems more dangerous to lift off too early and find that the economy can’t tolerate tighter policy than to end up a bit high on inflation for a while. That said, I feel less strongly about the risk-management case than I did three or six months ago.

Brad DeLong: There are four reasons why it is not yet time to begin normalizing monetary policy.

First, we are only 300bp away from the equilibrium funds rate. Given how close we are, why not just wait until we get to full employment?

Second, there is an unknown amount of slack left, and it might be substantial.

Third, to Jan’s point on risk management, I think there’s a 50% chance that the Federal Reserve will be really sorry that it raised rates when it did. And there is no upside for proceeding with rate hikes now. The Fed could delay hikes another six months and then just raise them faster and still get to the same place. But if it raises rates now, it will have no way to catch up because rates would presumably still be very low.

Finally, we are likely to find ourselves at the zero lower bound sometime in the future again—and when we do, we want market participants to feel very certain that there will be overshooting coming out of it—more inflation and lower real interest rates over the medium term that will boost growth. We need a reputation of coming off of the zero lower bound with a roaring economy, and I believe we need to stay at zero longer for that to happen.

Allison Nathan: Why isn’t the Fed more concerned about the lack of inflation, and should it be?

Brad DeLong: Janet Yellen and Stanley Fischer really believe in their models, which predict that inflation will rise to and above 2% a year in 2017 and 2018. They see this inflation path as a tangible reality, but it is in fact only a shadow cast by their assumptions. The Fed should be much more concerned about model uncertainty and, in turn, the lack of inflation right now.

Jan Hatzius: I have a slightly more positive view on inflation. My approach is “trust but verify,” which Ronald Reagan used to say on arms control negotiations. I trust the inflation models more than Brad does, but I’d like to see more verification in terms of the numbers picking up. A recent pick-up in wage growth is somewhat encouraging; our broad measurement of wage growth has risen to 2.6% yoy—still low but the highest rate we’ve seen in the recovery. We’re only at 1.3% for core PCE, but that is roughly what we expected at the start of the year, not a downside surprise, which would be more worrying.

Allison Nathan: Do you worry about financial imbalances?

Jan Hatzius: Not really. Asset markets don’t look particularly frothy to me today, debt growth is reasonably muted, and the private sector is still running a decent financial surplus of a little over 2% of GDP. So I just don’t see sources of worries over financial imbalances in the United States. But there are certainly other places around the world that show greater cause for concern, China being among them.

Brad DeLong: I generally agree. The aggregate numbers don’t seem to suggest anything like the troubling financial imbalances we have seen in the past. People worried about imbalances today tend to say that their concern centers on who is bearing risk in the economy or the markets, especially given that risk-bearing capacity on Wall Street has declined and that low interest rates have continued to generate a “search for yield.” But we don’t have the data to know how many people are unprepared to bear the risks associated with their positions.

Allison Nathan: Given current excess liquidity, will the Fed be successful in actually lifting the fed funds rate?

Brad DeLong: Yes. It will be very interesting to see what it has to do to be successful. But if the Fed wants the rate to get somewhere, it will get it there.

Jan Hatzius: Agreed.

Allison Nathan: Where will the Fed’s communication from the December meeting leave market expectations for future rate hikes?

Jan Hatzius: Markets will think that January is firmly off the table, and that March will be on the table if the data cooperates. The probability for March now priced into the market is roughly 50%, and my guess is that this will rise, but probably not above 60-70%. How exactly the Fed gets the market there is a bit of a dance. They will emphasize data-dependence, but the idea of March being a real possibility could cause a tightening of financial conditions and set expectations for hiking at every meeting, which the Fed wants to avoid. But I think they’ll find a way to manage this.

Brad DeLong: I agree they will emphasize data dependence and the need to assess the effects of the first hike, which is likely to push expectations to March at the earliest.

Allison Nathan: In the last few rate-hike cycles, the fed funds rate rose faster and ended up higher than the Fed initially projected. Will this time be different?

Brad DeLong: This time will be different, because in all recent hiking cycles, the Fed started out well behind the curve.

In the mid-2000s, the Fed was unhappy that its short-term rate increases were having so little traction on the long end of the bond curve, which led them to hike more rapidly than they initially intended.

In 1994, the tightening cycle began after Alan Greenspan had cut Bill Clinton slack for an entire year to get deficit reduction accomplished, so Greenspan was very eager to start raising interest rates once the Fed actually began to hike.

In 1989, Greenspan had delayed hikes out of fear that stock market crash in October 1987 would cause a recession. It didn’t, so the Fed ended up needing to catch up to where it thought it should be.

And before then, Paul Volcker definitely believed that the Fed was far behind the curve at the end of the 1970s when he became Fed Chair.

Given this pattern of the past four major tightening cycles, this time really is different.

Jan Hatzius: This time should be different, but the market is priced for something too different. If the recent trends continue, we will be at full employment by the end next year and inflation and wage growth will also likely be higher. In that environment, I don’t see the Fed taking a six-month break. I see them hiking a quarter-point every quarter, which is twice as much as what the market currently expects.

Allison Nathan: What’s the risk that the Fed will need to return to more accommodative policy?

Jan Hatzius: I’d give it a roughly 15% probability, which is not insignificant and still a good reason to delay, or at least to go more slowly and be very responsive to new information about the economy, especially in the early days of the normalization process. You could call that a second-best approach to delaying liftoff, but I do think that is where they are now. I see the odds of a shallower path than projected by the dots, but not an outright reversal, as higher, roughly in the 30% range.

Brad DeLong: I think there’s a greater than 50% chance that the rate path will be shallower than the current dots. I see a roughly 50% chance the Fed will end up wishing that they had stuck at zero or at least hiked even more gradually than what the market is currently expecting, but I am not sure they would actually dare reverse course. I could see them staying at 1% for a while, wishing they hadn’t hiked but not daring to go back.

Allison Nathan: Will the ECB and BOJ be less likely to ease further once the Fed lifts off?

Jan Hatzius: The extent to which the Fed, ECB, and BOJ are driven by each other gets very overplayed in the market. These are all economies with flexible exchange rates, and foreign monetary policy decisions almost always have offsetting effects on their own desires to move in one direction or another. For example, if the Fed tightens and that leads to a stronger dollar and a weaker US economy, the implications for the ECB are pretty ambiguous. So I don’t see a big spillover, and it is very hard to prove any such spillover empirically.

Brad DeLong: I agree that internal monetary politics are overwhelmingly primary in both the Euro area and Japan. What the Fed does is only very small noise compared to their focus on their own political-economic configuration. The only places in the developed world where monetary policy is tightly linked to the Fed are Britain and Canada.

Allison Nathan: Where will rates end up in this cycle?

Brad DeLong: There has been a structural downshift in rates. A decade ago, we wondered if rates would end up at 6%; a decade before that, it was 8%. Now, virtually everyone would be surprised if they ended above 4%. I’m more pessimistic than most in that I see them most likely to end at 3% or below, with a fairly small chance they’ll end up higher and a one-in-three chance that in retrospect we won’t really see this as a tightening cycle at all because rates will be in the 0-1% range three or four years down the road.

The primary factor behind my relatively pessimistic view is people’s loss of confidence in the capacity to bear and understand risk, and in the idea that large downward movements in asset prices are once-in-a-generation episodes like the Great Depression. In the 2000s, we suddenly had three episodes: the bursting of the dotcom bubble, the US real estate collapse, and the 2008-2009 equity and risky debt market crash, with the most sophisticated players blindsided by one or two if not all three of those events. This eroded confidence, widening the risk spreads between safe and risky assets.

Jan Hatzius: I forecast the terminal rate at 1-2% in real terms and 3-4% in nominal terms, which puts me on the slightly more optimistic side of this debate. In my view, the labor market is a much better indicator of cyclical progress than real GDP, especially in an environment where potential growth has slowed. And I am struck by the amount of labor market improvement we’ve seen. So I do think rates will end substantially higher from here.

Must-read: Stan Fischer: “Monetary Policy, Financial Stability, and the Zero Lower Bound I”

Posted on by Brad DeLong

Must-Read: Very disappointing to me that both nominal GDP targeting and price path level targeting appear to be completely off of Stan Fischer’s radar:

Stan Fischer: Monetary Policy, Financial Stability, and the Zero Lower Bound I: “Are We Moving Toward a World With a Permanently Lower Long-Run Equilibrium Real Interest Rate?…

…Research that was motivated in part by attempts that began some time ago to specify the constant term in standard versions of the Taylor rule has shown a declining trend in estimates of r*. That finding has become more firmly established since the start of the Great Recession and the global financial crisis…. A lower level of the long-run equilibrium real rate suggests that the frequency and duration of future episodes in which monetary policy is constrained by the ZLB will be higher than in the past. Prior to the crisis, some research suggested that such episodes were likely to be relatively infrequent and generally short lived. The past several years certainly require us to reconsider that basic assumption….Conducting monetary policy effectively at the ZLB is challenging, to say the least….

The answer to the question ‘Will r* remain at today’s low levels permanently?’ is that we do not know….

Raising the Inflation Target…. The welfare costs of high and variable inflation could be substantial. For example, more variable inflation would make long-run planning more difficult for households and businesses….

Negative Interest Rates: Another possible step would be to reduce short-term interest rates below zero if needed to provide additional accommodation. Our colleagues in Europe are busy rewriting economics textbooks on this topic as we speak…. It is unclear how low policy rates can go before cash holdings rise or other problems intensify, but the European experience has certainly shown that zero is not the effective lower bound in those countries….

Raising the Equilibrium Real Rate: An even more ambitious approach to ease the constraints posed by the zero lower bound would be to take steps aimed at raising the equilibrium real rate. For example, expansionary fiscal policy would boost the equilibrium real rate…. The Federal Reserve’s asset purchases… have reduced the level of the term premium….

Eliminating the ZLB Associated with Physical Currency….

None of these options for dealing with the difficulties of the ZLB suggest that it will be easy either to raise the equilibrium real rate or to mitigate the constraints associated with the ZLB…

A semi-platonic dialogue about secular stagnation, asymmetric risks, Federal Reserve policy, and the role of model-building in guiding economic policy

Posted on by Brad DeLong
Sanzio 01 jpg 3 820×2 964 pixels

Sokrates: You remember how I used to say that only active dialogue–questions-and-answers, objections-and-replies–could convey true knowledge? That a flat wax tablet covered by written words could only convey an inadequate and pale simulacrum of education?

Aristoteles: Yes. And you remember how I showed you that you were wrong? That conversation is ephemeral, and very quickly becomes too confused to be a proper educational tool? That only something like an organized and coherent lecture can teach? And only something like the textbooks compiled by my lecture notes can make that teaching durable?

Aristokles: But, my Aristoteles, you never mastered my “dialogue” form. My “dialogue” form has all the advantages of permanence and organization of your textbooks, and all the advantages of real dialectic of Sokrates’s conversation.

Sokrates: How very true, wise Aristokles!

Aristokles How am I to take that?

Xanthippe: You now very well: as snark, pure snark. That’s his specialty.

Hypatia: This is all complicated by the fact that in the age of the internet real, written, permanent dialogues can spring up at a moment’s notice:

Sokrates: And with that, let’s roll the tape:

Other things linked to that are highly relevant and worth reading:

Things I did not find and place outbound links to, but should have:

  • Polya
  • Dennis Robertson
  • Donald Patinkin

Looking at the whole thing, I wince at how lazy people–especially me–have been with their weblog post titles. I should find time to go back and retitle everything, perhaps adding an explanatory sentence to each link…

Must-read: Olivier Blanchard et al.: “Macro Effects of Capital Inflows: Capital Type Matters”

Posted on by Brad DeLong

Olivier Blanchard et al.: Macro Effects of Capital Inflows: Capital Type Matters: “Some scholars view capital inflows as contractionary…

…but many policymakers view them as expansionary. Evidence supports the policymakers. This column introduces an analytic framework that knits together the two views. For a given policy rate, bond inflows lead to currency appreciation and are contractionary, while non-bond inflows lead to an appreciation but also to a decrease in the cost of borrowing, and thus may be expansionary….

How can we reconcile the models and reality? [Our] answer… relies on… allowing for both ‘bonds’ (the rate on which can be thought of as the policy rate) and ‘non-bonds’… which are imperfect substitutes…. Capital inflows may decrease the rate on non-bonds and reduce the cost of financial intermediation…. Capital inflows may in this case be expansionary even for a given policy rate. In emerging markets, with a relatively underdeveloped financial system, the effect of a reduction in the cost of financial intermediation may dominate, leading to a credit boom and an output increase despite the appreciation. In more advanced economies, the appreciation may dominate, and capital inflows (even into non-bonds) may be contractionary (e.g. the Swiss case)….

The appropriate policies vis-à-vis capital inflows depend very much on the nature of the inflows.

Sterilised foreign exchange (FX) market intervention, if done through bonds (as is usually the case), can fully offset the effects of bond inflows…. When, however, sterilised foreign exchange intervention is used in response to non-bond inflows… FX intervention… by reducing upward pressure on the currency… increases capital inflows, and thus increases the effects of inflows on credit and the financial system….

The policymaker may have several objectives in mind – with respect to credit growth (given the risk of financial crisis); the currency (given the risk of Dutch disease); and output (given nominal rigidities in the system). The issue is really one of matching the policy instrument (there are three) to deliver on the most important objectives, without too much cost in terms of the other objectives…. If the central bank is worried about both appreciation and unhealthy or excessive credit growth, FX intervention or capital controls are preferable to the use of the policy rate in response to an increase in bond inflows…. In response to non-bond inflows, our framework suggests that if the goal is to maintain exchange rate stability with minimum impact on the return to non-bonds, capital controls do the job best, followed by FX intervention, followed by a move in the policy rate….

[We] use global flows to all emerging market countries together with the VIX as instruments…. We find that, while bond inflows have a negative effect on economic activity, non-bond inflows have a significant and positive effect. We also find that non-bond inflows (excluding FDI) have a strong positive effect on credit…. FDI inflows, while they increase output, have a negative impact on credit, perhaps because some of the intermediation which would have taken place through banks is replaced by FDI financing…

(Early) Monday DeLong Smackdown: Larry Summers on how we know more than we write down in our lowbrow (or highbrow) economic models

Posted on by Brad DeLong

Larry Summers: Thoughts on Delong and Krugman Blogs: “I think the issue is more on the supply side than the demand side…

…If I believed strongly in the vertical long run Phillips curve with a NAIRU around five percent and in inflation expectations responsiveness to a heated up labor market, I would see a reasonable case for the monetary tightening that has taken place. I am sure Paul and Brad are right that a desire to be ‘sound’ also influences policy.  I am not nearly as hostile to this as Paul. I think maintaining confidence is an important part of the art of policy.

A good example of where market thought is, I think, right and simple model based thought is I think dangerously wrong is Paul’s own Mundell-Fleming lecture on confidence crises in countries that have their own currencies.  Paul asserts that a damaging confidence crisis in a liquidity trap country without large foreign debts is impossible, because if one developed the currency would depreciate, generating an export surge.

Paul is certainly correct in his model, but I doubt that he is in fact. Once account is taken of the impact of a currency collapse on consumers’ real incomes, on their expectations, and especially on the risk premium associated with domestic asset values, it is easy to understand how monetary and fiscal policymakers who lose confidence and trust see their real economies deteriorate, as Olivier Blanchard and his colleagues have recently demonstrated.  Paul may be right that we have few examples of crises of this kind, but if so this is, perhaps, because central banks do not in general follow his precepts.

I do not think this is a pressing issue for the US right now. But the idea that policymakers should in general follow the model and not worry about considerations of market confidence seems to me as misguided as the view that they should be governed by market confidence to the exclusion of models.

Larry is taking the side that our economists’ models are primarily filing systems rather than truth-generating mechanisms. As I have already written twice in the past week, the question is: Are models properly idea-generating machines, in which you start from what you think is the case and use the model-building process to generate new insights? Or are models merely filing systems–ways of organizing your beliefs, and whenever you find that your model is leading you to a surprising conclusion that you find distasteful the proper response is to ignore the model, or to tweak it to make the distasteful conclusion go away?

Both can be effectively critiqued. Models-as-discovery-mechanisms suffer from the Polya-Robertson problem: It involves replacing what he calls “plausible reasoning”, where models are there to assist thinking, with what he calls “demonstrative reasoning”. in which the model itself becomes the object of analysis. The box that is the model is well described but, as Dennis Robertson warned,there is no reason to think that the box contains anything real. Models-as-filing-systems are often used like a drunk uses a lamp post: more for support than illumination.

In the real world, it is, of course, the case that models are both: both filing systems and discovery mechanisms. Coherent and productive thought is, as the late John Rawls used to say, always a process of reflective equilibrium–in which the trinity of assumptions, modes of reasoning, and conclusions are all three revised and adjusted under the requirement of coherence until a maximum level of comfort with all three is reached. The question is always one of balance.

The particular fight Larry wants to pick this weekend is over Paul Krugman’s Mundell-Fleming lecture, and Paul’s claim that a floating-rate sovereign that borrows in its own currency and is in a liquidity trap should not worry about maintaining “confidence”. Paul’s argument is that, in the model, pursuing aggressive expansionary policies will eat first to currency depreciation, then to an export boom, then to full employment, and only then will any downsides emerge. Thus the process can be stopped before it begins to generate risks. And it is only after full employment is attained that policy concern should shift to avoiding such risks.

Larry, however, says: We know things that are not in the model. Those things make Paul’s claim wrong.

My problem with Larry is that I am not sure what those things are. Paul notes that in normal times–away from the interest-rate zero lower bound–a loss of “confidence” in a country that floats its and borrows in its own currency an be contractionary:

Now,,, we can examine the effect of a loss of investor confidence… which we can view as a sudden stop… reduction in capital inflows at any given interest and exchange rate…. What would happen to a country with its own currency and a floating rate confronted with a foreign loss of confidence?… The depreciation of the currency at any given interest rate would increase net exports, and hence shift the IS curve out. This might be the end of the story. As I noted… however, the central bank might be concerned about the possible inflationary consequences of depreciation, and would therefore lean against it; in that case the MP schedule would shift up. So interest rates would rise due to rising demand for domestic goods and, possibly, tight money driven by inflation concerns. It is possible, if the latter motive is strong enough, that output could actually fall…

But, Paul says, at the zero lower bound things are different because the central bank has a cushion between what it wants the real interest rate to be and what the zero lower bound forces the real interest rate to be:

Right now the United States, the United Kingdom, and Japan are all stuck in the liquidity trap…. [with] policy… constrained by the zero lower bound…. This in turn means that any shift in the MP schedule, unless very large, won’t lead to a rise in interest rates…. All that matters is the rightward shift in the IS curve…. In short, under current conditions the much-feared loss of confidence by foreign investors would be unambiguously expansionary, raising output and employment in nations like the United States, the United Kingdom, and Japan…

It’s not clear what we know that is not in the model that would reverse this conclusion of Paul’s in the case of a floating exchange rate in a country that borrows in its own currency and happens to be in a liquidity trap. Paul notes that his conclusion goes against market wisdom:

Many people will, I know, object to this conclusion…. [It] seems very counterintuitive… and is very much at odds with what almost every policymaker and influential figure has been saying…. when I have tried laying out this argument to other economists, I have found that in general they recognize the point but argue that real-world complications mean that a sudden stop will nonetheless be contractionary even in countries with independent currencies and floating rates. Why? They offer a variety of reasons….

And Paul goes on:

The short-term/long-term interest rate distinction does not appear to offer any channel through which a nation with an independent currency can suffer a decline in output due to reduced foreign willingness to hold its debt….

Several commentators—for example, Rogoff (2013)—have suggested that a sudden stop of capital inflows provoked by concerns over sovereign debt would inevitably lead to a banking crisis…. [But] why, exactly… should [we] believe that a sudden stop leads to a banking crisis…. [It is] is only a problem for the banks if they have large liabilities denominated in foreign currency…. Inflation—and fear of inflation—is a potential channel through which sudden stops can end up being contractionary even for countries with independent currencies and domestic-currency debt…. Large foreign-currency debt can effectively undermine monetary independence, as can fears of depreciation-led inflation. However, the major nations with large debts but independent currencies don’t have large foreign-currency debt, and are currently quite remote from inflation pressure. So the crisis story remains very hard to tell.

I would say that the major nations with large debts don’t have large foreign-currency debt that we know of…

More musings on the current episteme of the Federal Reserve…

Posted on by Brad DeLong

Paul Krugman’s Respectable Radicalism politely points out (at least) one dimension along which I am a moron.

Let me back up: Here in the United States, the current framework for macroeconomic policy holds that the economy is nearly normalized, that further extraordinary expansionary and fiscal policy moves carry “risks”, and that as a result the right policy is stay-the-course. I was arguing that the Economist Left Opposition demand–for substantially more expansionary monetary and fiscal policies right now until we see the whites of the eyes of rising inflation–was soundly based in orthodox lowbrow Hicks-Patinkin-Tobin macro theory. That is the macro theory that economists like Ben Bernanke, Janet Yellen, and Stan Fischer taught their entire academic careers.

Paul Krugman points out—politely—that I am wrong.

The Economist Left Opposition framework contains at least one claim that is substantially non-orthodox: We claim that worries about the debt accumulation from expansionary fiscal policy right now are profoundly misguided. Under current conditions, the government’s borrowing money or printing money and buying stuff does not raise but lowers the debt-to-annual-GDP ratio. However large you think the influence of an outstanding debt burden on interest rates happens to be, interest rates in the future will be lower, the debt as a multiple of annual GDP will be lower, and thus the debt financing burden and all debt-related risks will be lower in the future with a more expansionary fiscal policy than baseline. This is definitely nonstandard. And it is embarrassing to note that this is my idea–or, rather, Larry Summers and I were the ones who did the arithmetic to show how topsy-turvy the macroeconomic world currently is with respect the fiscal policy. This was a really smart thing for us to do. And it is definitely not part of the standard orthodox policy-theory framework in the way that the rest of the Hicks-Patinkin Economist Left Opposition framework is.

As Paul writes:

Paul Krugman: Respectable Radicalism: “Hysteresis [in the context of very low interest rates]… is indeed a departure from standard models…

…But the [rest, the] case that the risks of hiking too soon and too late are deeply asymmetric comes right out of IS-LM with a zero lower bound… the framework I used….

Being an official… can create a conviction that you and your colleagues know more than is in the textbooks…. But… [at the] zero-lower-bound… world nobody not Japanese [had] experienced for three generations, theory and history are much more important than market savvy. I would have expected current Fed management to understand that; but apparently not.

I wrote about Rawls’s reflective equilibrium idea yesterday, so let me just cut and paste: Are models properly idea-generating machines, in which you start from what you think is the case and use the model-building process to generate new insights? Or are models merely filing systems–ways of organizing your beliefs, and whenever you find that your model is leading you to a surprising conclusion that you find distasteful the proper response is to ignore the model, or to tweak it to make the distasteful conclusion go away?

Both can be effectively critiqued. Models-as-discovery-mechanisms suffer from the Polya-Robertson problem: It involves replacing what he calls “plausible reasoning”, where models are there to assist thinking, with what he calls “demonstrative reasoning”. in which the model itself becomes the object of analysis. The box that is the model is well described but, as Dennis Robertson warned,there is no reason to think that the box contains anything real. Models-as-filing-systems are often used like a drunk uses a lamp post: more for support than illumination.

In the real world, it is, of course, the case that models are both: both filing systems and discovery mechanisms. Coherent and productive thought is, as the late John Rawls used to say, always a process of reflective equilibrium–in which the trinity of assumptions, modes of reasoning, and conclusions are all three revised and adjusted under the requirement of coherence until a maximum level of comfort with all three is reached. The question is always one of balance.

What I think Paul Krugman may be missing here is how difficult it is to, as Keynes wrote:

The composition of this book has been for the author a long struggle of escape… from habitual modes of thought and expression. The ideas which are here expressed so laboriously are extremely simple and should be obvious. The difficulty lies, not in the new ideas, but in escaping from the old ones, which ramify, for those brought up as most of us have been, into every corner of our minds…

In this case, the old ideas with respect to the budget deficit are those of various versions of fiscal crisis and fiscal price level theory developed largely out of analysis of Latin American and southern European experience, and those of various versions of monetarist theory based upon the experience of the 1970s. How difficult this is is illustrated by one fact I find interesting about Paul Krugman (1999): Back then, his analysis of the liquidity trap and fiscal policy back in 1999 was… very close to Ken Rogoff’s analysis of the liquidity trap and fiscal policy today:

Paul Krugman (1999): Thinking About the Liquidity Trap, Journal of the Japanese and International Economies 14:4 (December), pp. 221–237: “The story… [of] self-fulfilling pessimism is… a multiple equilibrium story…

…with the liquidity trap corresponding to the low-level equilibrium…. Over some range spending rises more than one-for-one with income. (Why should the relationship flatten out at high and low levels? At high levels resource constraints begin to bind; at low levels the obvious point is that gross investment hits its own zero constraint. There is a largely forgotten literature on this sort of issue, including Hicks (194?), Goodwin (194?), and Tobin (1947))….
Thinking about the liquidity trap

Multiple equilibria… allow for permanent (or anyway long-lived) effects from temporary policies. There may be excess desired savings even at a zero real interest rate given the pessimism that now prevails… but if some policy could push the economy to a high level of output for long enough to change those expectations, the policy would not have to be maintained…. Balance-sheet problems… may involve an element of self-fulfilling slump: a firm that looks insolvent with an output gap of 10 percent might be reasonably healthy at full employment….

‘Pump-priming’ fiscal policy is the conventional answer to a liquidity trap…. In either the IS-LM model or a more sophisticated intertemporal model fiscal expansion will indeed offer short-run relief…. So why not consider the problem solved? The answer hinges on the government’s own budget constraint….

Ricardian equivalence… is not the crucial issue…. Real purchases… will still create employment…. (In a fully Ricardian setup the multiplier on government consumption will be exactly 1)….

The problem instead is that deficit spending does lead to a large government debt, which will if large enough start to raise questions about solvency. One might ask why government debt matters if the interest rate is zero…. But the liquidity trap, at least in the version I take seriously, is not… permanent…. [When] the natural rate of interest… turn[s] positive… the inherited debt will indeed be a problem….

Fiscal policy [is] a temporary expedient that cannot serve as a solution [unless]….

First, if the liquidity trap is short-lived… fiscal policy can serve as a bridge… after [which]… monetary policy will again be able to shoulder the load… a severe but probably short-lived financial crisis in trading partners… breathing space during which firms get their balance sheets in order….

[Second, if] it will jolt the economy into a higher equilibrium…. If this is the underlying model… one must realize that the criterion for success is quite strong…. Fiscal expansion… must lead to… increases in private demand so large that the economy begins a self-sustaining process of recovery….

None of this should be read as a reason to abandon fiscal stimulus…. But fiscal stimulus… [is only] a way of buying time… [absent] assumptions that are at the very least rather speculative…

Since 1999, Paul has changed his mind. He has become an aggressive advocate of expansionary fiscal policy as the preferred solution. Why? And is he right to have done so? Or should he have stuck to his 1999 position, and should he still be lining up with Ken today?

One part of the reason, I think, is–and I say this with whatever modesty I have ever had still intact–that DeLong and Summers (2012) has provided one of the very very few additions of conceptual value-added to Krugman (1999). We pointed out that with a modest degree of “secular stagnation”–a modest fall in safe real interest rates over the long run–and a slight degree of hysteresis, fiscal expansion in a liquidity trap does not worsen but improves the long-run fiscal balance of an economy in a liquidity trap. This was something that Krugman missed in 1999. It is something that people like Rogoff continue to miss today.

This has consequences: The more scared you are of some long-run collapse of the currency from excessive government debt relative to annual GDP, the stronger you should advocate for more expansionary fiscal policy when the economy is in a liquidity trap. The more you think that real interest rates in the long run are coupled to high values of government debt relative to annual GDP, the stronger you should advocate for more expansionary fiscal policy when the economy is in a liquidity trap. The more you worry about debt crowding-out useful and productive government spending in the long run, the stronger you should advocate for more expansionary fiscal policy when the economy is in a liquidity trap. This whole line of thought, however, was absent from Krugman (1999), and is absent from Rogoff and company today.

A second part of the reason is that even modest “secular stagnation” does more than (with even a slight degree of hysteresis) reverse the sign on the relationship between fiscal expansion today and long-run government-debt burdens. It also undermines the effectiveness of monetary policy as an alternative to fiscal policy. Monetary expansion–in the present or the future–needs a post-liquidity trap interest-rate “normalization” environment to have the purchase to raise the future price level that it needs to be effective in stimulating production now. Secular stagnation removes or delays or attenuates that normalization.

Third comes the credibility problem.

Fourth, there is a sense in which Paul has not shifted that much. Look at his analysis of Japan…

Third comes the credibility problem. Back in the days of Krugman (1999), he at least had little doubt that a central bank that understood the situation would want to generate the expected inflation needed. That was the way to create a configuration of relative prices consistent with full employment. That was what a competent central bank would wish to do. And a central bank that wished to create expectations of higher inflation would have a very easy time doing so.

The mixed success of Abenomics, however, has cast doubt on the second of these—on the ability of central banks to easily generate higher expected inflation. Japan today appears to be having a significantly harder time generating expectations of inflation than I had presumed. And

With respect to the first—the desire to create higher expected inflation—Ben Bernanke, while chairman of the Federal Reserve, repeatedly declared that quantitative easing policies were not intended to produce any breach of the 2% per year inflation target upward. These declarations were not something that I expected, and were not something that I understood. They still leave me profoundly puzzled.

Fourth, there is a sense in which Paul has not shifted that much. Look at his analysis of Japan today. In his view, fiscal expansion today is needed to create the actual inflation today that will (i) raise the needle on future expected inflation, and so (ii) allow for a shift to policies that (iii) will amortize rather than grow the national debt. Inflation someday generated by the fiscal theory of the price level and high future interest rates generated by the risks of debt accumulation still have their places in his thought.

Must-read: George Evans and Bruce McGough: “The Neo-Fisherian View and the Macro Learning Approach”

Posted on by Brad DeLong

Must-Read: Via Mark Thoma: George Evans and Bruce McGough: The Neo-Fisherian View and the Macro Learning Approach: “[A] REE… needs an explanation for how economic agents come to coordinate on it…

…This point is acute in models in which there are multiple RE solutions…. The macro learning literature provides a theory for how agents might learn over time to forecast rationally… by updating their econometric forecasting models provided the REE satisfies ‘expectational stability’ (E-stability) conditions. If these conditions are not satisfied then convergence to the REE will not occur and hence it is implausible that agents would be able to coordinate on the REE…. For a wide range of models this gives plausible results… the basic Muth cobweb model… Lucas (1986) used an adaptive learning scheme to show that though the overlapping generations model of money has multiple REE, learning dynamics converge to the monetary steady state, not to the autarky solution. Early analytical adaptive learning results were obtained in Bray and Savin (1986) and the formal framework was greatly extended in Marcet and Sargent (1989). The book by Evans and Honkapohja (2001) develops the E-stability principle and includes many applications….

There are other approaches…. The ‘eductive’ approach of Guesnerie asks whether mental reasoning by hyper-rational agents, with common knowledge of the structure and of the rationality of other agents, will lead to coordination on an REE…. The Garcia-Schmidt and Woodford (2015) ‘reflective equilibrium’ concept… draws on both the adaptive and eductive strands as well as on the ‘calculation equilibrium’ learning model of Evans and Ramey (1992, 1995, 1998)…. The lack of a TE or learning framework in Cochrane (2011, 2015) is a critical omission…. In our… ‘Observability and Equilibrium Selection,’ Evans and McGough (2015b), we develop the theory of adaptive learning when fundamental shocks are unobservable…. Adaptive learning thus operates as a selection criterion and it singles out the usual RE solution adopted by proponents of the NK model. Furthermore, when monetary policy does not obey the Taylor principle, then neither of the solutions is robustly stable under learning; an interest-rate peg is an extreme form of such a policy, and the adaptive learning perspective cautions that this will lead to instability…. The learning approach argues forcefully against the neo- Fisherian view.

Musings on the current episteme of the Federal Reserve

Posted on by Brad DeLong

Larry Summers attributes the Federal Reserve’s decision to tighten policy, in what appears to him and to me to be a weakly-growing and high-slack economy, to four mistakes, which are themselves driven by a fifth, overarching mistake. The four mistakes are:

  1. The Fed has much too much confidence in its models that tell it that the unemployment rate takes the temperature of the labor market and the Phillips Curve now still has the slope it had in the 1970s.
  2. The Fed operates as though FOMC members are tased whenever inflation rises above 2%/year, with no countervailing painful consequences of low inflation, low employment, or low output.
  3. The Fed believes–without empirical support anywhere that I can see–that quick sharp moves up or down in interest rates have larger effects in total than the same interest-rate change made gradually and over a longer term.
  4. The Fed thinks–without theoretical support that I can see–that zero interest rates are not a reflection of an economy in a pathological state, but rather a cause of economic pathology that is dangerous now that the economy is once again “normal”.

And Summers sees the fifth, overarching mistake the Fed is making right now as:

(5) The Fed is excessively committed to “existing models and modes of thought… in the thrall of orthodoxy”:

Larry Summers: My Views and the Fed’s Views on Secular Stagnation: “First, the Fed assigns a much greater chance that we will reach 2 percent core inflation…

…than is suggested by most available data…. Second, the Fed seems to mistakenly regard 2 percent inflation as a ceiling not a target…. Third, the Fed seems to be in the thrall of notions that… do not… have analytic support premised on the idea that the rate of change of interest rates, as distinct from their level, influences aggregate demand…. I know of no model in which demand will be stronger in say 2018 if rates rise and then fall than if they are kept constant at zero. Nor… do I know of a reason why recession is more likely if the changes are backloaded…. The argument… is in the category with the argument that I should starve myself in order to have the pleasure of relieving my hunger pangs. Fourth, the Fed is… overestimating the neutral rate…. The desire to raise rates reflects… a sense that zero rates are a sign of pathology and an economy creating 200,000 jobs a month is not diseased….

Why is the Fed making these mistakes if indeed they are mistakes? It is not because its leaders are not thoughtful or open minded or concerned with growth and employment.  Rather I suspect it is because of an excessive commitment to existing models and modes of thought…

I do think–confidently–that Summers is absolutely 100% correct in his identification of the four component intellectual errors that the Federal Reserve is currently making. And it is certainly true that these are the result of an excessive commitment to some current modes of thought–there are, after all, a lot of people who join the Fed in thinking that zero interest rates are a cause rather than the proper treatment of pathology right now, that the Fed needs to raise rates now to give it the space to lower them if need be later, that it is dangerous for inflation to rise above 2%/year ever, that the Phillips Curve somehow has a steeper slope than the recent evidence of the past generation can justify belief in, and that the unemployment rate rather than the detrended employment-to-population ratio gives the temperature of the labor market.

But do these beliefs on the part of the Fed really reflect an excessive commitment to existing models? There I have my doubts. Or, rather, it depends on what you think the proper function of economic modeling is. Are models properly idea-generating machines, in which you start from what you think is the case and use the model-building process to generate new insights? Or are models merely filing systems–ways of organizing your beliefs, and whenever you find that your model is leading you to a surprising conclusion that you find distasteful the proper response is to ignore the model, or to tweak it to make the distasteful conclusion go away?

Both can be effectively critiqued. Models-as-discovery-mechanisms suffer from the Polya-Robertson problem: It involves replacing what he calls “plausible reasoning”, where models are there to assist thinking, with what he calls “demonstrative reasoning”. in which the model itself becomes the object of analysis. The box that is the model is well described but, as Dennis Robertson warned,there is no reason to think that the box contains anything real. Models-as-filing-systems are often used like a drunk uses a lamp post: more for support than illumination.

In the real world, it is, of course, the case that models are both: both filing systems and discovery mechanisms. Coherent and productive thought is, as the late John Rawls used to say, always a process of reflective equilibrium–in which the trinity of assumptions, modes of reasoning, and conclusions are all three revised and adjusted under the requirement of coherence until a maximum level of comfort with all three is reached. The question is always one of balance.

But it is clear to me that if you give even minor weight to the first–see well-founded models as a way of generating new insights rather than just organizing old beliefs–that the line of work into the economics of the liquidity trap that I see as well-represented by Krugman’s (1999) “Thinking About the Liquidity Trap” tells us, very strongly, that the Federal Reserve is on the wrong track intellectually right now. It tells us that what is out of whack has not been and is not the real money stock, but is instead the expected inflation rate. Or, rather, that because the expected inflation rate is too low, there is no value of the real money stock consistent with full employment equilibrium. If expected inflation were higher, the existing money stock would be ample, or even require shrinking.

This is the conclusion of Krugman (1999): that the economy needs higher expected and actual inflation, and that the free-market economy with full price flexibility would deliver that inflation. But the Fed does not appear to acknowledge either that the economy needs higher inflation, that the flexible-price benchmark would deliver this inflation, or that the job of the Fed is to mimic that full employment-generating price structure as closely as possible.

Moreover, the Fed does not engage in reflective equilibrium. It rejects the conclusions of what I regard as the standard Patinkin-style existing model of Krugman (1999). But it does not propose an alternative model. There seems to me to be no theoretical ground, no model even considered as a filing system, underpinning the “orthodox” modes of thought that the Fed believes. And it does not seem to feel this absence aaa a problem. I find that somewhat disturbing.

Must-read: Mark Thoma: “On Summers: My Views and the Fed’s Views on Secular Stagnation”

Posted on by Brad DeLong

Must-Read: I would say that if monetary policy makers wish to place limits on what can be expected from monetary policy, they need to also be making loud and constructive arguments about what will do the stabilization policy job if monetary policy is not going to push the envelope. I don’t think Plosser ever made loud and constructive arguments directing other policy makers to do a constructive job…

Mark Thoma: On Summers: My Views and the Fed’s Views on Secular Stagnation: “The Fed’s job would have been, and will be a lot easier if fiscal policy makers would help…

…I disagree with Charles Plosser’s view on monetary policy, but I have some sympathy for the view that many people have come to expect too much from monetary policy:

On the monetary policy side central banks have clearly pushed the envelope in an effort to stabilize and then promote real economic growth. The pressure to do so has come from inside and outside the central banks… raised expectations of what the central bank can do…. It is not clear that this is wise or prudent.  Many have come to fear that without substantial support from monetary policy our economies will slump into stagnation. This would seem to fly in the face of nearly two centuries of economic thinking…

If secular stagnation is real, the Fed cannot overcome it by itself. Fiscal policy will have to be part of the solution…. Monetary policy — and fiscal policy too — can have a permanent impact on the natural rate of output by helping the economy to recover faster. The faster the recovery, the less the natural rate is lowered. So I agree with Summers that monetary policy needs to take the possibility of secular stagnation into account, I just wish he’d put more emphasis on the essential role of fiscal policy — something he has certainly done in the past, e.g.:

I believe that it is appropriate that we go back to an earlier tradition that has largely passed out of macroeconomics of thinking about fiscal policy as having a major role in economic stabilization…