Business cycles, interest rates, and NGDP

Almost every day there’s something of interest over at Paul Krugman’s blog.  Sometimes I like to re-frame these arguments from a NGDP targeting perspective.  For instance, here’s Krugman mentioning a longstanding puzzle in international macro:

And I should say that there is a long-standing puzzle concerning world business cycles: economies move in synch more than can easily be explained via concrete linkages in the form of exports.

I’m going to take a stab at this, even though I know less international macro than he does.

One might think of exports as a “real linkage.”  Belgium exports a lot to the rest of Europe, so it would be hard for Belgium to avoid recession with expansionary monetary policy, if the rest of Europe fell into a deep recession.  At a minimum, the “PSST” issue discussed by Arnold Kling would be a huge problem.  I’m sure that Belgium manufactures lots of goods that are either inputs for the production of final goods in other European countries, or final goods with demand that is highly sensitive to the European business cycle.  That’s a sensible plan for Belgium, but it’s also a “real” problem for Belgium when Europe screws up its stabilization policy.

Krugman indicates, however, that the business cycle linkages seem much stronger than what you’d expect from exports alone, and that sounds right to me (based on casual empiricism.)

In that case I’d want to break the problem down into two components, real linkages, and nominal linkages produced by monetary policy synchronization.  For simplicity, assume that stable NGDP growth is the optimal monetary policy.  Also supposes that among developed countries both RGDP and NGDP are strongly correlated.  Then it’s quite possible that part of the business cycle synchronization is due to similar monetary policy failures in various countries.  The most famous example is a fixed exchange rate regime, such as the gold standard.  When the global real interest rate fell (due to weak investment demand) there was a tendency for people to increase their real demand for gold.  That tended to transmit deflation and depression to all countries on the gold standard, regardless of trade linkages.

So in principle it’s not hard to explain why business cycles are somewhat correlated.  Indeed we have a pretty complete explanation for why that occurs under fixed exchange rate regimes.  The tougher problem is explaining the correlation in floating rate regimes.  What makes NGDP cycles correlated?  To a market monetarist, that’s just another way of asking; “Why is monetary policy correlated between various countries?”

There could be lots of reasons.  Central bankers in various countries probably tend to think alike.  If we assume capital markets are integrated, then these central bankers might well respond to various economic shocks in much the same way.  Take mid-2008 as an example.  For various reasons there was a negative shock to investment demand in 2008.  Housing slumped in the US and elsewhere.  The sharp rise in oil prices reduced demand for cars, trucks and RVs.  These are industries that often rely on credit.  So as demand for loans declined, the Wicksellian equilibrium interest rate fell all over the world.  Trade patterns tended to transmit this effect to even those countries where there was no housing bubble.  Yet we are told that the trade patterns along cannot explain the degree of synchronization.

Now suppose that central banks were slow to respond to the fall in the Wicksellian equilibrium interest rate (which is the rate necessary to keep NGDP growing at a steady rate.)  Why might they have been slow?  As I said, central bankers tend to think alike, and they were all concerned about the upsurge in headline inflation during mid-2008, due to rising commodity prices.  So they all tended to tighten money, according to the definition used by market monetarists.  That is, they let NGDP growth expectations fall.

People that aren’t market monetarists aren’t likely to see this as tight money.  But in any mainstream model, a shock that reduces NGDP growth will tend to reduce RGDP as well.  So even if you don’t call it “tight money,” the effect is the same.  The failure to respond aggressively to changes in the global S&I relationship led to much lower NGDP all over the world.  Call it weaker “animal spirits” if you prefer.  But whatever you call it, it provides a rationale for highly synchronized business cycles that does not solely rely on export linkages.

This isn’t to say trade plays no role in what I’ve just described.  Suppose the export problems reduced RGDP growth by one percent, even assuming a stable NGDP growth path.  This would reflect the usual PSST issues.  This 1% growth slowdown may reduce the Wicksellian equilibrium rate.  If the central bank doesn’t respond fast enough, monetary policy will unintentionally tighten, which might reduce NGDP growth by another 3% or 4%.  In the short run that will be mostly in the form of reduced RGDP growth, so you might get 1% less inflation, and 2% to 3% less RGDP growth.

A naive observer will only see the export demand shock; the monetary shock will look invisible.  A skilled monetary economist will notice the fall in NGDP growth, which can’t be explained solely by less exports.

In my view this is why monetarism looks so out of touch with reality for many people with feet firmly planted in the real world.  They see some “obvious” problem like a housing bust, which really is a problem.  It’s capable of reducing RGDP growth by say 1%.    Then they see a demand-side recession follow soon after.   They see no sign of tight money (recall that rates are low.)  So they can’t imagine how this mysterious “tight money” could have caused the recession.  But as we saw with Australia, if the central bank keeps NGDP growing (at least in terms of two year averages) then RGDP can hold up pretty well, despite a big drop in exports.  Australia had the growth slowdown that we and the Europeans should have had, if we’d been following a NGDP level targeting approach.

[Ryan Avent has some related comments.]

Part 2. Here’s another interesting Krugman argument:

So what is happening when we’re at the zero lower bound? One way to think about it is to draw the supply and demand for savings that would prevail if the economy were at full employment. They look like this:

The policy problem is that for whatever reason “” in current conditions, mainly the deleveraging taking place after an era of debt complacency “” the interest rate that would match savings and investment at full employment is negative.

This is a part of a longer post that criticizes Bill Gross, who recently advocated higher interest rates.  I mostly agree with Krugman, but would quibble with one aspect of his argument.  I’m not sure why he regards those saving/investment schedules as representing “full employment.”  Krugman and I both think the SRAS is fairly flat right now.  This means that any monetary policy powerful enough to push the economy to full employment would have to generate extremely rapid NGDP growth, because the flat SRAS means that inflation would probably rise by less than RGDP, and we’d need lots more RGDP to get to full employment.

The extremely rapid NGDP growth required for full employment would almost certainly push equilibrium nominal interest rates above zero.  Thus if monetary policy was expansionary enough to truly generate full employment, it would also generate higher interest rates. If I’m not mistaken, this is Nick Rowe’s argument that the IS curve is upward sloping, whereas I presume Krugman is making the standard assumption that it’s downward-sloping.

I found the Bill Gross article to be maddeningly vague.  Bill Gross seems to believe that prosperity is associated with higher nominal rates, which would be correct.  The question is whether he has an effective plan for getting from here to there.  It can’t be done by simply raising the fed funds target (which is Krugman’s point, and he’s right.)

A commenter named Cthorm sent me this quotation from Bill Gross and a comment afterward:

“The transition from a levering, asset-inflating secular economy to a post bubble delevering era may be as difficult for one to imagine as our departure into the hereafter. A multitude of liability structures dependent on a certain level of nominal GDP growth require just that – nominal GDP growth with a little bit of inflation, a little bit of growth which in combination justify embedded costs of debt or liability structures that minimize the haircutting of or defaulting on prior debt commitments. Global central bank monetary policy – whether explicitly communicated or not – is now geared to keeping nominal GDP close to historical levels as is fiscal deficit spending that substitutes for a delevering private sector.”

Any doubt he is a market monetarist?

Yes, I have some doubts.  The discussion of NGDP certainly sounds promising.   But the last sentence could be referring to a constant NGDP, a constant growth rate, or returning to a previous trend line.  Those are three very different policies.  Bill Gross needs to do much more if he’s going to convince others that he has a coherent plan for generating higher interest rates without driving us right back into recession.  Don’t forget, both the Japanese and Europeans tried to raise rates before the economy was ready.  And they both had to cut them soon after.


Tags:

 
 
 

44 Responses to “Business cycles, interest rates, and NGDP”

  1. Gravatar of dwb dwb
    7. February 2012 at 07:12

    I think Bill Gross confuses money and credit, and also supply and demand (both for credit and money). ack. The yield premium that lenders demand to lend money is the credit spread and as i posted before its actually somewhat wider than the 2005-2007 period. Banks generally hedge interest rate risk and transform market risk into credit risk. Banks do not make money buying T-bonds and borrowing fed funds (thats a maturity mismatch problem that the asset-liability risk committe would bust their butt for). They make money buying (selling) t-bonds from(to) Sally and selling (buying) them to/from Joe (and they make money on the fees and credit difference between the two). If they make a loan and keep it on their books they are looking at credit spread not UST rate. they’ll be more than happy to help you swap your fixed rate debt from fixed to floating or vice versa, or refinance or help you buyback your existing debt, or hekp you buyback your debt. thats not a statement about rates, it a statement that they make money on fees selling you products you may or may not need just like any other company. all those things i mentioned are generally 0-NPV transactions done at-market, but may help (for example) rating agency metrics – and serendipitously the bak gets a small fee for helping you do this. great work if you can get it!

  2. Gravatar of Morgan Warstler Morgan Warstler
    7. February 2012 at 07:20

    More and more I’m focusing on the upside of people being unable to get a return on loaning money to the gvt. or buying mattress CDs.

    That to me is as moral as the promise that their money will not be inflated away.

    1. no risk-less investment
    2. stable money (0-2% inflation)

    These are both good forces for generating the kind of small government, small business world we want to have.

    Note that neither is justified because they aid in promoting employment, we shouldn’t accept unemployment, we should auction off the unemployed.

    One last note, perhaps you’d think risk-less investment aids in capital formation? I suspect that near-riskless investment is an acceptable substitute.

  3. Gravatar of Steve Steve
    7. February 2012 at 07:34

    Why did Trichet tighten whenever Richard Fisher was a voting member of the FOMC? THE TRADE OF BAD IDEAS IS MORE IMPORTANT THAN THE TRADE OF REAL GOODS.

    All major central banks engage in COMMODITY PRICE TARGETING, even if it’s just pass through, i.e., “second-round effects.”

    And Bill Gross is a fiduciary and a sales person. He has to advocate for the results his clients want, not the policies that would get us there. That makes him painful to listen to.

  4. Gravatar of tim tim
    7. February 2012 at 07:42

    This is from Milton Friedman’s Theoretical Framework for Monetary Analysis (journal of political economy 1970). I think it’s interesting regarding the S&I schedule where the equilibrium rate is negative:

    One way to summarize his argument for that proposition is in terms of a possible conflict between the “market” and the “equilibrium” rate of interest. If investment opportunities were sparse, yet the public’s desire to save were strong, the “equilibrium” rate of interest, he argued, might have to be very low or even negative to equate investment and saving. But there was a floor to the “market rate” set by liquidity preference. If this floor exceeded the “equilibrium rate,” he argued, there was a conflict that could only be resolved by unemployment that frustrated the public’s thriftiness. The fallacy in this argument is that the introduction of money not only introduces a floor to the “market rate”; it also sets a floor to the “equilibrium rate.” And, in the long run, the two floors are identical. This is the essence of the so-called Pigou effect (Friedman 1962, pp. 262-63).

  5. Gravatar of D.Gibson D.Gibson
    7. February 2012 at 08:07

    Has Krugman ever directly commented on NGDP targeting?

    If I am an investor in another country (eg, my Brazil ETF) I care more about their RGDP growth than their NGDP growth. I move my investments accordingly. If many people do that, does it have a reduction on S=I for a country? Or does everything balance out through accounting?

  6. Gravatar of 123 123
    7. February 2012 at 08:09

    Cthorm, sitting in close to Gross, sees him as a Keynesian: “he is also saying that IOR is a problem.” – i.e. the problem is that interest rates, for example interest rates on reserves, are too high.
    I, looking from another continent, see Gross as a monetarist who argues that the problem is low M3, low NGDP, and perhaps low M0.

  7. Gravatar of StatsGuy StatsGuy
    7. February 2012 at 09:09

    Ugh, there are so many arguments here… Think of this from a statistical perspective:

    Why are X and Y correlated?

    1) There’s mutual causation – the “real exports” argument

    2) There’s some sort of auto-correlated error. For example, a “herd behavior” where all monetary policy makers (or CEOs or others) make the same decision errors because they are not making those decisions independently.

    3) There’s spurious correlation driven by a mutual causal factor –

    a) Price of inputs (e.g. oil), which are internationally traded commodities, affect everyone simultaneously

    b) Other real shocks

    4) Expectations of future equalization (error is endogenous with policy outcome). I would expect this to manifest primarily through future-discounting mechanisms, which essentially means the capital markets (and financial flows DWARF real trade flows).

    Overall, a sloppy question from Krugman. Not his best thinking.

    RE: Bill Gross – dude, he’s arguing FOR HIS BOOK. The guy SHIFTED BACK INTO BONDS six months ago, and he wants the value of those bonds to GO UP. Anything he can do to limit expectations for NGDP growth and to curtail or box-in policy response by presenting the Fed as unreasonable/lax/dangerous/whatever he will do. Remember the “Ring of Fire”? Remember the US bond market was doomed, right until he bought US bonds? Bill Gross is a shill, the most cynical of the cynical. I respect his ability, but I would never take national policy advice from him, nor investing advice since he will only advocate a position AFTER he has bought into it.

  8. Gravatar of Benjamin Cole Benjamin Cole
    7. February 2012 at 09:09

    Another great post by Scott Sumner.

    Lars at Market Monetarist also noted the strange Bill Gross comment. Here is what I commented there:

    “I am beginning to wonder if Bill Gross needs to retire. I have actually talked to him a few times, many years ago, and he is friendly and smart. But lately, his pronouncements have been off the mark.

    He seems to be grappling here, nearly incoherently so, with the idea of nominal GDP targeting. He makes some accurate assessments about the need for moderate inflation in modern economies, and then surmises that central banks are not hip to that need. Okay, going in the right direction then….

    How this fits with Gross’s calls for a tighter Fed policy I can’t tell you. We need a blood transfusion, so apply the leeches.

    Maybe Gross is converting to Market Monetarism, but like many, will have to do so in cathartic and flatulent excretions, such as this. A few more epiphanies, and we will have another convert!”

    It is difficult to overcome lifelong shibboleths and partisan standards. For generations, certain members of the economic mainstream have ever called for tight money and balanced federal budgets, until the two ideas have become conflated, certainly in their inchoate rhetoric.

    Now we need a bullish monetary policy (Market Monetarism) and balanced federal budgets. Join us Bill Gross, we will be happy to have you on our team!!!

  9. Gravatar of sean sean
    7. February 2012 at 09:21

    I think Bill Gross is just throwing out arguments that are in his best interest and refusing to be intellectually honest. I do not know if he is doing this on purpose (knows his arguments are bad) or is unable to get to the truth in his own head. Bottom line being at zero for the long-term is very bad for a bond fund manager. He knows that 2% 10 years are not a good investment. They may be appropriate monetary policy, but they are not ideal for PIMCO especially when he has to charge a fee for his services on that 2%. It puts him out of business over time. The initial move from 4% to 2% was quite profitable for bond fund managers, but after the move you are just stuck with low rates.

  10. Gravatar of Cthorm Cthorm
    7. February 2012 at 10:42

    Scott, thanks for the H/T.
    “I found the Bill Gross article to be maddeningly vague.” I would hardly disagree, it jumps from point to point and mixes language from several different economic camps.

    “The question is whether he has an effective plan for getting from here to there.” I don’t think he does…but you and the other MMs do, and elements of your arguments are starting to show up in his statements. I’m doing my best to make sure the ideas keep reaching him.

    @123 – I hardly “sit close to Bill Gross.” I’m an analyst in the Compliance department, I don’t even sit in the same building. Nor do I see him as a Keynesian. As far as I can tell he is not a strict adherent to any particular school of economics, he is primarily an investor and trader. There are other Managing Directors who are clearer about their economic roots; El-Erian and (former MD) Paul McCulley are clearly Keynesians, and Chris Dialynas is more of a Monetarist (certainly a fiscal conservative).

    @Stats – what book? He manages dozens of portfolios and they all have defined risk profiles and investment objectives. Treasury holdings might be hurt by higher interest rates (assuming market-driven and not fed funds rate hikes), but StocksPlus would benefit from higher equities and any TIPS holdings from higher inflation. In any case talking your book is the norm in the financial press, but there is also an element of “putting your money where your mouth is”. I don’t think whether or not he benefits from a policy prescription should matter, the only thing that matters is if the policy would benefit the economy as a whole.

  11. Gravatar of Major_Freedom Major_Freedom
    7. February 2012 at 10:49

    “One might think of exports as a “real linkage.” Belgium exports a lot to the rest of Europe, so it would be hard for Belgium to avoid recession with expansionary monetary policy, if the rest of Europe fell into a deep recession.”

    They can export money like we do!

    “At a minimum, the “PSST” issue discussed by Arnold Kling would be a huge problem. I’m sure that Belgium manufactures lots of goods that are either inputs for the production of final goods in other European countries, or final goods with demand that is highly sensitive to the European business cycle. That’s a sensible plan for Belgium, but it’s also a “real” problem for Belgium when Europe screws up its stabilization policy.”

    It’s impossible not to screw it up, if the standard is totally unhampered private sector price system.

    “Krugman indicates, however, that the business cycle linkages seem much stronger than what you’d expect from exports alone, and that sounds right to me (based on casual empiricism.)”

    “In that case I’d want to break the problem down into two components, real linkages, and nominal linkages produced by monetary policy synchronization. For simplicity, assume that stable NGDP growth is the optimal monetary policy. Also supposes that among developed countries both RGDP and NGDP are strongly correlated. Then it’s quite possible that part of the business cycle synchronization is due to similar monetary policy failures in various countries.”

    OK, now given virtually no countries have in fact adopted NGDP monetary policy, explain the real world correlation. Oh that’s right, it’s because they didn’t target NGDP. My bad.

    “The most famous example is a fixed exchange rate regime, such as the gold standard. When the global real interest rate fell (due to weak investment demand) there was a tendency for people to increase their real demand for gold. That tended to transmit deflation and depression to all countries on the gold standard, regardless of trade linkages.”

    If the context is a global decline in interest rates, which is a global decrease in the rate of time preference, which means there is a global increase in the preference for more future goods and less present goods, which of course should reduce current spending on and current production of consumer goods, then of course it will show a nominal “deflation” and “depression” in the statistics you mechanistically track.

    This outcome however is a GOOD thing. When people consume less and save more, then current production SHOULD fall. Resources and labor that otherwise would have gone into current production of consumer goods, are freed up so that people can go work in producing goods that won’t be ready for consumption until the future.

    That’s exactly what the people want! How does this justify printing and spending money to stretch the economy towards current output when the people clearly want less current output? Is the purpose of monetary policy to help people, or go against them?

    If I am on an island with one other person, and we produce and consume 5 coconuts a day, and then one day we both decide to reduce our time preference, and thus produce and consume fewer coconuts each day, say from 5 to 3, so that we can devote more of our time and resources to building a house that won’t be available as “output” until a month from now, then OF COURSE current production will fall. Does that mean that some magical fairy state has to come along and consume 2 coconuts a day to “make up” for the “lost” production and consumption of daily coconuts, so that the statistical number pseudo-monetarists have a fetish over, doesn’t fall from one period of time to the next?

    What about the people themselves? Aren’t they more important than ensuring the silly statistic you observe doesn’t fall?

    “Indeed we have a pretty complete explanation for why that occurs under fixed exchange rate regimes. The tougher problem is explaining the correlation in floating rate regimes. What makes NGDP cycles correlated? To a market monetarist, that’s just another way of asking; “Why is monetary policy correlated between various countries?”

    “There could be lots of reasons. Central bankers in various countries probably tend to think alike. If we assume capital markets are integrated, then these central bankers might well respond to various economic shocks in much the same way. Take mid-2008 as an example. For various reasons there was a negative shock to investment demand in 2008. Housing slumped in the US and elsewhere. The sharp rise in oil prices reduced demand for cars, trucks and RVs. These are industries that often rely on credit. So as demand for loans declined, the Wicksellian equilibrium interest rate fell all over the world.”

    For various reasons. “For various reasons.” For a monetarist to say that is like a robber saying a car went missing “for various reasons” while he is driving away in it.

    “Trade patterns tended to transmit this effect to even those countries where there was no housing bubble. Yet we are told that the trade patterns along cannot explain the degree of synchronization.”

    The housing bubble was mere one facet of a much larger economic bubble. The reason why we call it the housing bubble and the Nasdaq bubble, etc, is because busts tend to start with one sector of the economy going through correction before other sectors go through a correction. What usually happens is that the Fed reinflates once again to reverse the economy wide corrections before they occur. If the Fed didn’t do anything post 2008, then we would have been talking about a housing bubble, automotive bubble, manufacturing bubble, etc. But the Fed reinflated (and the Treasury spent money) before this took place. So we just see 2008- as a housing bubble and bust.

    “Now suppose that central banks were slow to respond to the fall in the Wicksellian equilibrium interest rate (which is the rate necessary to keep NGDP growing at a steady rate.) Why might they have been slow? As I said, central bankers tend to think alike, and they were all concerned about the upsurge in headline inflation during mid-2008, due to rising commodity prices. So they all tended to tighten money, according to the definition used by market monetarists. That is, they let NGDP growth expectations fall.”

    Which investor do you know of, which seller do you know of, bases their decisions on NGDP? Why would they if the Fed isn’t even targeting it? What difference does it make to an individual investor or seller that some aggregate spending statistic is different? Unless their own dealings and circumstances are expected to change, NGDP is irrelevant. If you tell me that NGDP will fall by 10% next year, but I have an expectation that my sales will rise by 1% next year, then I am going to base my current decisions on my own dealings and expectations concerning my business, not your NGDP forecast.

    “People that aren’t market monetarists aren’t likely to see this as tight money. But in any mainstream model, a shock that reduces NGDP growth will tend to reduce RGDP as well.”

    That’s a good thing. When there is an NGDP “shock”, it means that people are spending less money and want prices to fall so that they have more purchasing power. Responding to that by printing and spending money is the exact opposite of what people want and can only short circuit their plans.

    “So even if you don’t call it “tight money,” the effect is the same. The failure to respond aggressively to changes in the global S&I relationship led to much lower NGDP all over the world. Call it weaker “animal spirits” if you prefer. But whatever you call it, it provides a rationale for highly synchronized business cycles that does not solely rely on export linkages.”

    You’re just presuming your model is true using circular logic. You’re saying economies fell together into recession because there was a collective decline in NGDP, as a premise to show that a collective decline in NGDP explains the correlated fall into recession.

    That’s funny.

    “This isn’t to say trade plays no role in what I’ve just described.”

    This is where the “subsidiary arguments” I referred to earlier make their appearance, in order to make it seem like this isn’t really an NGDP fetishism. The following could have taken place and could not have taken place, it won’t matter for the NGDP story:

    “Suppose the export problems reduced RGDP growth by one percent, even assuming a stable NGDP growth path. This would reflect the usual PSST issues. This 1% growth slowdown may reduce the Wicksellian equilibrium rate. If the central bank doesn’t respond fast enough, monetary policy will unintentionally tighten, which might reduce NGDP growth by another 3% or 4%. In the short run that will be mostly in the form of reduced RGDP growth, so you might get 1% less inflation, and 2% to 3% less RGDP growth.”

    “A naive observer will only see the export demand shock; the monetary shock will look invisible. A skilled monetary economist will notice the fall in NGDP growth, which can’t be explained solely by less exports.”

    A naive monetary economist will only notice the fall in NGDP growth. A skilled recalculation economist will notice the fact that the collapse in prices tends to always be greater in the higher stage capital goods than for consumer goods.

    “In my view this is why monetarism looks so out of touch with reality for many people with feet firmly planted in the real world. They see some “obvious” problem like a housing bust, which really is a problem. It’s capable of reducing RGDP growth by say 1%. Then they see a demand-side recession follow soon after. They see no sign of tight money (recall that rates are low.) So they can’t imagine how this mysterious “tight money” could have caused the recession.”

    In my view this is why recalculation looks even more out of touch with reality for many people with heads not firmly up their keesters. They see some “obvious” problem like the fall in aggregate spending, which is really not a problem. It’s capable of reducing RGDP growth by say 1%. Then they see unemployment follow soon after. They see no sign of previous loose money (recall that rates are, and were, low) So they can’t imagine how this mysterious “monetary manipulation” could have caused the recession.

    “But as we saw with Australia, if the central bank keeps NGDP growing (at least in terms of two year averages) then RGDP can hold up pretty well, despite a big drop in exports. Australia had the growth slowdown that we and the Europeans should have had, if we’d been following a NGDP level targeting approach.”

    The Australian central bank had to do less than the Fed to keep their NGDP growing because the Australian economy wasn’t as distorted as the US economy. NGDP not falling in Australia signalled a healthier economy, not a more informed central bank.

  12. Gravatar of Bill Woolsey Bill Woolsey
    7. February 2012 at 11:01

    I think Friedman was mistaken.

    First, the Pigou effect assumes outside money. Gold, or government fiat currency subject to a quantity rule might work. It doesn’t apply to purely private money, and it might not work with government money with a price level or nominal GDP target. (It won’t work with recardian equivalence.)

    Second, the natural interest rate is a real interest rate, and with inflation, it can be less than zero.

    Third, the problem is “short run.” The problem is that level of the real interest rate needed to keep real expenditures at the full employment level right now is negative. There is no requirement that this condition be expected to last for a decade, much less pemanently. What would permanently negative real interest rates do to the value of beachfront property? Famous paintings, or the like?

    In other words, if the framing of the problem is a permanent unemployment _equilibrium_, then the negative real interest rate won’t do.

    It is rather a characteristic of a disequilibrium state. And it might last a few years.

    Similarly, if you are arguing that even if prices and wages are perfectly flexible, the negative natural interest rate will prevent equilibrium, is pretty implausible. Pure inside money, inflation targeting, maybe flexible prices won’t help.

    But who cares about that? Prices and wages aren’t perfectly flexible. How much energy should we put into thinking about what would happen if prices and wages could all fall 60% over the next three months?

    Of course, I must admit that considering the “unrealistic” scenario of an all deposit monetary system so that nominal interest rates on money might turn negative, and so, other short term nominal interest rates could be negative, and then what would happen to saving/consumtpion and investment if that happened, is exactly what makes the real world scenario where zero interest currency is what is keeping that from happening. The zero nominal bound keeps real interest rates from falling enough to clear markets.

  13. Gravatar of 123 123
    7. February 2012 at 11:02

    @Cthorm

    “sit close to Bill Gross.” – that was a joke that has some basis in reality…

    “Nor do I see him as a Keynesian. ” I am not putting him in a narrow one-economic-school box either. But your interpretation of Gross’ statement about IOR (IOR is harmful) makes sense in the Keynesian context only – as it is the Keynesian theory that says that too high real interst rates are the source of all evil. On the other hand, Krugman has criticised the monetarist Gross, who says that more money should be printed, so the interest rates can be higher. This is the contrast I wanted emphasize with my previous comment.

  14. Gravatar of Cthorm Cthorm
    7. February 2012 at 11:13

    @123

    Cheers on the joke, sorry I didn’t pick it up. I just don’t want to misrepresent myself.

    I don’t see how IOR being harmful is a purely Keynesian opinion. It is a sop to financial institutions, and one that is contractionary in the monetarist sense because it siphons funds out of circulation. From an optics point of view I think it’s “double plus ungood” because it offsets expansionary policies like QE, but does not offset the political costs thereof.

  15. Gravatar of Pulman Pulman
    7. February 2012 at 11:38

    From this post it’s quite clear why Krugman got the Nobel Prize and you did not.

  16. Gravatar of 123 123
    7. February 2012 at 11:43

    @Cthorm

    Regarding IOR, please see my reply in the Australia thread.

  17. Gravatar of John John
    7. February 2012 at 12:12

    If the Wicksellian rate is what promotes stable growth, wouldn’t you also expect negative consequences from below Wicksellian rates? That’s what Mises was talking about when he developed his theory of the business cycle (the most correct one) over 100 years ago.

  18. Gravatar of Brendan Brendan
    7. February 2012 at 12:19

    Gross, Charles Schwab and countless other investment pro’s have made essentially the same stupid claim. They take the true fact that rates and velocity are positively correlated. They note that velocity is low right now. And conclude the fed must raise rates. They fail to distinguish between a rise in rates generated by higher growth expectations, versus a contraction in the monetary base caused by the fed.

    I work in the investment industry. Few have any clue about monetary policy, yet beliefs are so strongly held and hawkishness is nearly unanimous. Even a smart, modest, extremely rigorous economist/fund manager like John Hussman gets crazy when he starts talking about the Fed.

    http://www.hussmanfunds.com/wmc/wmc110912.htm

  19. Gravatar of ssumner ssumner
    7. February 2012 at 12:30

    dwb, Perhaps, I couldn’t figure it out.

    Morgan, If you don’t like low risk bonds, why do you like low real yields on those bonds? Wouldn’t that encourage the government to borrow more?

    Steve, Interesting.

    Tim, Good quotation.

    D. Gibson, He did sort of endorse it. I don’t recall the exact wording, but he had a couple posts that offered increasing support for the general idea. That’s not to say he views it as the ideal policy, but better than what we have now.

    If you invest in Brazil it might tend to make US saving more than US investment. But more people are going the other way, so we still save less than we invest.

    123, I think he sees him as a market monetarist, not Keynesian. The critique of IOR came from market monetarists, not Keynesians.

    Statsguy, The interesting question is the extent to which NGDP targeting could shield you from global recessions. In my view the larger the economy the more a NGDP target would help smooth out the cycle.

    I do not take policy advice from Gross.

    Ben, I’m glad I wasn’t the only one who found that unclear.

    Sean, Good point.

    Cthorm, I appreciate that.

    Major Freedom, You said;

    “When people consume less and save more, then current production SHOULD fall.”

    Why? You seem to assume that current consumption equals current output, but it doesn’t.

    Bill, I agree about outside money, and I think Friedman also agrees. I think you misinterpret Friedman; he wasn’t arguing that the Pigou effect was a useful way to find macroeconomic equilibrium.

    Pulman, You said;

    “From this post it’s quite clear why Krugman got the Nobel Prize”

    So you think people are awarded Nobel Prizes for sketching out someone else’s ideas on a napkin?

  20. Gravatar of ssumner ssumner
    7. February 2012 at 12:32

    John, Yes.

    Brendan, I agree that many people in the investment industry are clueless about macro. Not all, but many.

  21. Gravatar of Ben Wolf Ben Wolf
    7. February 2012 at 12:33

    “Second, the natural interest rate is a real interest rate, and with inflation, it can be less than zero”

    Even without inflation the natural inter-bank interest rate is zero. Without CB intervention government spending ensures the rate falls to nothing or just above.

  22. Gravatar of 123 123
    7. February 2012 at 14:01

    Scott: “The critique of IOR came from market monetarists, not Keynesians.”
    Speaking about personalities, you are right. For example, DeLong has supported IOR in June 2008 as a stimulus measure. (and I agree with him). On the other hand, Bernanke has implemented DeLong’s policy suggestion, and I’d say Bernanke is a monetarist, not a Keynesian.

    But I was discussing theory, not personalities. From the Keynesian perspective, IOR is a crime against humanity, raising the flood of the liquidity trap by 25bps. From monetarist perspective, there are various justifications for IOR (Chuck Norris effect, stressing the importance of policy targets vs. instruments, “macroeconomics of doing nothing” etc.).

  23. Gravatar of Morgan Warstler Morgan Warstler
    7. February 2012 at 14:19

    Scott, I mean gvt. bonds should legally never pay more than inflation…. precisely so no one wants to loan to them… unless there is a crisis like now.

    In that way, gvt. debt taking really only exists during a flight to safety…

    So you get a gvt. that can’t borrow in good times, and only borrows when people are ok with no return.

    That’s TRUE Keynesian thinking isn’t it – because it enforces the the shrink gvt. during good times – the piece DeKrugman always forgets about.

    One step past that, I’m quite alright with ending all true risk-less investments – it is a hack to having Natural Money.

    We want to people to stare at their pile, and see it not grow, unless they get out there and take risks.

  24. Gravatar of Morgan Warstler Morgan Warstler
    7. February 2012 at 14:20

    Woolsey,

    Auctioning the Unemployed (my Guaranteed Income plan) is perfectly doable.

    We could easily have super flexible wages.

  25. Gravatar of Ben Wolf Ben Wolf
    7. February 2012 at 14:37

    “Scott, I mean gvt. bonds should legally never pay more than inflation…. precisely so no one wants to loan to them… unless there is a crisis like now.”

    The U.S. government can always sell its bonds, regardless of interest rates or economic situation. The system is designed so that auctions never fail.

  26. Gravatar of Doc Merlin Doc Merlin
    7. February 2012 at 14:48

    ” It can’t be done by simply raising the fed funds target (which is Krugman’s point, and he’s right.)”

    It /could/ have been done by allowing defaults to proceed at pace, to rest the economy, which would have after the defaults finished, increased the demand for credit.

  27. Gravatar of dwb dwb
    7. February 2012 at 16:45

    if govt bonds “never legally pay more than inflation” someone clever will just find a way around it so that the yield will reflect the correct economics. either they will sell them at a discount, or as with so-called Islamic bonds (sukuks or sharia bonds), someone will find a clever way around the law. It in no way prevents the govt from running deficits.

  28. Gravatar of John John
    7. February 2012 at 16:55

    dwb,

    They could just do what t-bills do and sell at a discount from par in the primary market. That gets around paying interest while incorporating the time value of money (aka the source of interest).

  29. Gravatar of Ben Wolf Ben Wolf
    7. February 2012 at 17:51

    @dwb

    It wouldn’t even be necessary to employ clever tricks. The Fed ensures the primary dealers always have sufficient reserves on hand to take down the entire auction, and they’re required by law to create a market for the bonds. Even in the extremely unlikely situation that U.S. bonds become totally unpalatable, the game would continue.

  30. Gravatar of Alex Theisen Alex Theisen
    7. February 2012 at 18:00

    Re: The 2nd Krugman point, I think the real difference between his model and yours and Rowe’s seems to be just a matter of what is being held constant. It seems like Krugman’s model is holding “real” expectations constant; given currently depressed expectations of real output and demand (and thereby depressed investment demand) the only real interest rate that will equate desired savings and desired investment and fully employ all resources is a very negative one. Since Krugman’s tacitly assuming that the central bank only operates through inflation expectations and changes in the nominal interest rate, his model basically takes “real” expectations (and therefore the real natural rate) as given, and so he gets a downward sloping IS curve. What’s weird is that if the economy were really at full employment, then expectations of growth and demand would obviously not be depressed, and so the natural rate would likely be higher. It’s probably not legitimate to keep expectations, investment demand, and the natural rate constant when you’re shifting around those variables.

    You and Nick Rowe seem to have a model where the central bank largely works through influencing expected NGDP, not the real interest rate or inflation expectations. Since in this model, the natural rate is a dependent variable and isn’t monetary policy-invariant, you get something like Nick Rowe’s upward sloping IS curve.

    Of course this all probably shows some of the issues with using a crude model like IS-LM when reasoning about the macroeconomy; a model isn’t independent of the underlying vision of what the policy regime is.

  31. Gravatar of Morgan Warstler Morgan Warstler
    7. February 2012 at 19:31

    “Even in the extremely unlikely situation that U.S. bonds become totally unpalatable, the game would continue.”

    Not the real point… after a Congress takes the outsized play (in make believe land) of removing the gvt. as competitive borrower from private markets…

    The meek Fed taunts them? Nah.

    I think I made a mistake, I should have said, “I’m drawn to the political argument the right could use to make people angrier about the deficit.”

    My goal (as always) is to end the idea of risk-less gains into our economic system. I routinely bring up Natural Money and then people with respond with holes in Islamic borrowing.

    The point is that we are a better form of capitalism when all capital has a mind frantically trying to actually invest in “please god, I hope the consumer wants to buy this”

    That is the daily prayer that all Americans should have to kneel down in front of the rest of us, and desperately try to divine what will make us all happy.

    Equity investors are closer to priests where the market is god.

    We want more priests. We ALWAYS want more priests.

  32. Gravatar of Jon Jon
    7. February 2012 at 19:46

    Scott,

    Krugman indicates, however, that the business cycle linkages seem much stronger than what you’d expect from exports alone, and that sounds right to me (based on casual empiricism.)

    In that case I’d want to break the problem down into two components, real linkages, and nominal linkages produced by monetary policy synchronization. For simplicity, assume that stable NGDP growth is the optimal monetary policy. Also supposes that among developed countries both RGDP and NGDP are strongly correlated. Then it’s quite possible that part of the business cycle synchronization is due to similar monetary policy failures in various countries.

    Isn’t one of the most significant factors here that many small currencies practice an exchange-rate stabilization policy that links their exchange-rate to a big currency such as the Euro?

    Now, there are two classes of this sort of activity generally, one is a country like Sweden where the SNB actively manages the exchange-rate from an export-driven perspective. i.e., they will always loosen at least as much as the big currency they are targeting.

    Conversely an emerging market economy such as Romania is pegged to the Euro. The NBR explicitly rejects inflation targeting (let alone NGDP targeting)–arguing that household liabilities are heavily denominated in Euros where as household assets are in RON.

    AFAIK, this is a much more dominate effect in terms of synchronizing monetary policies…

  33. Gravatar of Morgan Warstler Morgan Warstler
    7. February 2012 at 19:48

    “Given the crudeness of the tools the FOMC uses to set monetary policy, allowing for such a margin of error is no doubt prudent. For example, when the economy slowed in the first half of last year, inflation picked up, accelerating to a 6.1% annual rate during the second quarter. And, when the economic growth accelerated in the second half, inflation slowed. These results are the precise opposite of what the Fed’s playbook says are supposed to happen.”

    http://www.forbes.com/sites/charleskadlec/2012/02/06/the-federal-reserves-explicit-goal-devalue-the-dollar-33/2/

  34. Gravatar of Ray Ray
    7. February 2012 at 20:25

    Scott,

    just wanted to say I love reading the blog. While I some of it goes over my head as a college student, I can always catch enough to always make an interesting read of a perspective they don’t mention in econ classes.

  35. Gravatar of Major_Freedom Major_Freedom
    7. February 2012 at 21:18

    ssumner:

    I said: “When people consume less and save more, then current production SHOULD fall.”

    You said: “Why? You seem to assume that current consumption equals current output, but it doesn’t.”

    Why? It’s because production takes time.

    No, I am not assuming that current consumption equals current output. Current consumption makes up only a portion of current output. A reduction in current consumption and hence reduction in the production of current consumer goods that is brought about by an increase in savings and investment in longer period production processes that won’t be finished until the future, will generate a reduction in current output.

    You seem to assume that a fall in consumption spending and hence fall in production of consumer goods, brought about by a rise in savings and investment, will immediately and without any time delay result in a replacement of output to make up for the reduced consumer goods output. That assumption is wrong because it ignores the time element.

    I provided the coconut example to explain the intuition behind why a fall in current output is expected, and desired, if people reduce their consumption spending and save more, and why it makes no sense to call for an external entity to come along and bring about more consumption to “replace” the lost consumption.

  36. Gravatar of Bill Woolsey Bill Woolsey
    8. February 2012 at 04:10

    Major Freedom:

    The partially finished house and partially finished canoe are both current output.

    In GDP figures,they are measured as “goods in process.”

    In reality, they aren’t very important. Much more of investment involves the production of tools.

    Suppose that every day people are gathering coconuts and making ladders. (The ladders increase the speed of coconut gathering.)

    It doesn’t take a long time to build a ladder, but that its time that could be used to gather coconuts (after climbing the tree.)

    Less consumption and more saving means spending more time constructing ladders today. So, there are less coconuts and more ladders. Total output is the same now, just different types of output are produced.

    The ladders allow more coconuts to be gathered, and so coconut production increases the next day.

    Unforuntately, ladders wear out so it is necessary to spend some time producing replacement ladders if labor productivity in gathering coconuts is to be maintained.

    Anyway, even if it takes days to complete a ladder, there is no reduction in output. Suppose the first day you make the sides. Then the next day, the steps. Then the third day assemple. You gather less cocunuts the first day to make the sides, total output unchanced. The next day, less coconuts are gathered to make have time for steps. Output is unchanged. And finally, the third day, less coconuts are gathered to assemble the ladder. Output is unchanged.

  37. Gravatar of UnlearningEcon UnlearningEcon
    8. February 2012 at 05:00

    Scott, OT question:

    Do you favour any capital controls to keep monetary stimulus in the domestic economy? Seems a large part of the problem with QE1/2 was that they seeped abroad.

  38. Gravatar of ssumner ssumner
    8. February 2012 at 06:06

    123, I don’t agree on either point. Bernanke is certainly not a monetarist.

    And monetarism says any policy aimed at raising the demand for reserves will reduce the money multiplier.

    Morgan, Never reason from a price change. The market includes both the supply and demand side. You are only looking at demand. The low yields will increase the supply.

    Doc Merlin, Maybe, but that’s not obvious to me.

    Alex, You said;

    “I think the real difference between his model and yours and Rowe’s seems to be just a matter of what is being held constant. It seems like Krugman’s model is holding “real” expectations constant; given currently depressed expectations of real output and demand (and thereby depressed investment demand) the only real interest rate that will equate desired savings and desired investment and fully employ all resources is a very negative one. Since Krugman’s tacitly assuming that the central bank only operates through inflation expectations and changes in the nominal interest rate, his model basically takes “real” expectations (and therefore the real natural rate) as given, and so he gets a downward sloping IS curve.”

    Maybe, but how can that be right? How can Krugman vary expected inflation, but not expected real output, when the SRAS curve is relatively flat? Perhaps I’m misunderstanding his model, it’s not a model I use. But those two assumptions seem inconsistent.

    I see your later comment suggests you see the same problem.

    Jon, I agree that with fixed exchange rates it’s easy to explain the synchronization. But the flexible rate case is tougher.

    Morgan, He forgets about supply shocks.

    Thanks Ray.

    Major Freedom, See Bill.

    UnlearningEcon, No, that’s not a problem at all. I don’t think much of the QE went overseas. And if it does, they can just print more.

  39. Gravatar of Nominal spending determines outcomes, not fiscal stimulus | Historinhas Nominal spending determines outcomes, not fiscal stimulus | Historinhas
    8. February 2012 at 08:02

    […] reading this quote from a recent post by Scott Sumner: In my view this is why monetarism looks so out of touch with reality for many people with feet […]

  40. Gravatar of Major_Freedom Major_Freedom
    8. February 2012 at 10:20

    Bill Woolsey:

    I thought GDP calculations excluded sales of intermediate goods and services to avoid the problem of double counting.

    Maybe you are using a particular GDP I am not familiar with that includes intermediate goods and services, in which case yes, I will agree that there would be no decrease in “output”, but it was my impression that GDP is biased towards consumption spending since it excludes a chunk of gross investment.

  41. Gravatar of Alex Theisen Alex Theisen
    8. February 2012 at 10:32

    “Maybe, but how can that be right? How can Krugman vary expected inflation, but not expected real output, when the SRAS curve is relatively flat? Perhaps I’m misunderstanding his model, it’s not a model I use. But those two assumptions seem inconsistent.”

    I think that just comes from the fact that, if you’re thinking in purely IS-LM terms, IS curve is a “real” equation, and LM (or the MP) curve is a “nominal” one. The way I learned IS-MP in my intermediate Macro class, monetary policy only affects the LM or MP curve, and fiscal policy affects the IS curve. When you have a model that keeps them so tightly segregated, it’s easy to think that expected inflation can go up without expected real output and investment demand going up as well. Reading Krugman’s posts, I think that’s why he seems sometimes to talk about NGDP targeting as if it were just a covert way of raising inflation expectations or lowering expected interest rates. If your mental model explicitly only gives monetary policy two bullets, you’re going to filter reality through that, even though when pressed, Krugman would probably admit that an NGDP target would raise growth expectations and the natural rate of interest. It’s really just a question of what intuitive model you use when thinking about an economy. Krugman thinks in IS-LM, you think in AS-AD.

  42. Gravatar of ssumner ssumner
    8. February 2012 at 11:58

    Alex, You may be right, but since Krugman clearly believes the SRAS is pretty flat, that seems like a very big weakness in his IS-LM approach.

  43. Gravatar of Opinion: Pick your horse in the grand recovery stakes Opinion: Pick your horse in the grand recovery stakes
    16. February 2012 at 08:28

    […] at The Money Illusion, in part 1 of this post the market monetarist, Scott Sumner, offers this explanation: … monetarism looks so out of touch […]

  44. Gravatar of ‘Free Market’ Double Standards 4.0 « Unlearning Economics ‘Free Market’ Double Standards 4.0 « Unlearning Economics
    23. March 2012 at 09:20

    […] 4. We need monetary stimulus, but no capital controls to keep it in the domestic economy. […]

Leave a Reply