Do OMOs “work” at the zero bound? That’s the wrong question.

Here are the right questions:

1.  If nominal rates are zero, and NGDP is expected to grow on target (let’s say 4.5%, for example), then what is the equilibrium base/GDP ratio. I.e., what’s the demand for base money at that expected NGDP growth rate?

2.  What is the ratio of T-bills to GDP?

3.  What is the ratio of Treasury securities to GDP?

4.  What is the ratio of “eligible assets” to GDP?

When Keynesians like Woodford and Krugman talk about the (non) effectiveness of open market operations at the zero bound, they tend to envision a situation where expected NGDP growth is inadequate, and then ask whether OMOs “work” in the sense of boosting expected NGDP growth.  But that’s looking at things backwards.  Everyone (I think) agrees that if the Fed bought up all of planet Earth, then OMOs would be “effective,” i.e. they’d boost expected NGDP growth. When Keynesians worry about a lack of effectiveness of monetary policy, they typically restrict the term ‘monetary policy’ to something narrower than buying up all of planet Earth.  Indeed they often call the purchase of non-government securities “fiscal policy.”  I think that’s wrong, the purchase of high quality corporate debt at going market prices by central banks is just as much “monetary policy” as the purchase of Treasury debt, or the lending of reserves to commercial banks.  But let’s put that issue aside for the moment, and agree that only the purchase of Treasury securities counts as monetary policy.

In that case here’s the right way to think about monetary policy ineffectiveness:  Assume that the Fed always stands ready to buy enough assets to keep expected NGDP growing at say 4.5% per year, level targeting.  They agree to start with the safest assets (say T-bills) then move on to slightly riskier assets, such as T-notes and bonds, and then move on to agency debt, municipal bonds, high quality corporate debt, etc.  In that case monetary policy is “ineffective” if the public’s demand for base money when expected NGDP growth is on target is greater than the amount of Treasury debt held by the public.  That’s it.  It’s that simple.  Monetary policy would be ineffective using the Keynesian definition (although it would still be effective using my broader definition.)

Treasury debt held by the public is around 70% of US GDP.  It seems unlikely to me that the demand for base money would ever exceed 10% of GDP, if NGDP was expected to grow at 4.5%, the Fed was doing level targeting, and the IOR was zero.  However I can’t say it’s impossible, after all, there’s been a surprisingly large secular decline in real interest rates since the early 1980s.  Whose to say that doesn’t continue?

Some might point to the fact that base demand is currently around 18% of GDP.  However that reflects two special factors; interest on reserves, and a very low rate of NGDP growth since mid-2008 (about 2% per year.)  If there had been normal NGDP growth and zero interest on reserves, then base demand would have been much smaller.

If we use a Svenssonian “target the forecast” approach, then OMOs are very unlikely to be ineffective, even when at the zero bound.  That’s because base demand in the US is unlikely to exceed 70% of GDP.  On the other hand one could easily envision a situation where monetary policy became ineffective in a country with a small national debt, such as Australia.  Of course in that case I presume the RBA would simple expand the definition of “monetary policy” to include the purchase of foreign government debt.  Perhaps they already do this.

And I really don’t see any problem with relying exclusively on monetary policy for demand stabilization in the US, even if the demand for base money exceeded 70% of GDP.  For God’s sake the US government already does insane things like bail out auto companies and banks.  Why in the world would we be squeamish about having the Fed buy high quality private debt at fair market value?  The EMH says they should not lose money on average, and they already make huge profits from seignorage.

As I pointed out a few years back, it all comes down to Yoda’s famous maxim:

Do or do not, there is no try.

When Keynesians worry about ineffective monetary policy, they are actually contemplating a situation where the central bank makes a half-hearted effort, with no intention to do whatever it takes to succeed.  In that case yes, policy might fail.  If the Fed sets the base at a level where the internal Fed forecasts calls for only 3.5% NGDP growth, then yes, the policy may fail to produce expectations of 4.5% NGDP growth.  But is that any surprise?

Here’s the right approach:  The Fed says; “We are going to do whatever it takes to succeed, so that our internal forecasts are on target.”

And here’s the right question:  “How much stuff do we have to buy?”

No one should ever ask whether monetary policy is “effective.”  As soon as you start thinking that way, you have begun to fail.

PS.  Because of the secular decline in real interest rates, we are very likely to hit the zero bound in future recessions.  Keynesians must give up this fixation on using interest rate targets, and shift to a policy of adjusting the the base to target NGDP expectations.  Or at the very least, convince the Fed and ECB and BOJ to shoot for a more rapid trend rate of NGDP growth (as in Australia), so that nominal rates stay above zero.

Update:  I just noticed an excellent Nick Rowe post, which does a great job of explaining why interest rates targeting must end.



39 Responses to “Do OMOs “work” at the zero bound? That’s the wrong question.”

  1. Gravatar of Saturos Saturos
    6. September 2012 at 05:16

    Scott, it’s really about time you updated the “key blog posts” list. For instance, this one should be added to the section on liquidity traps. Very illuminating.

  2. Gravatar of Alex Armlovich Alex Armlovich
    6. September 2012 at 05:30

    I think Woodford’s new paper, in questioning the effectiveness of “Pure QE”, was questioning the permanence of a change in the money supply by large asset purchases:

    “[T]his argument concerns the effects of a policy of permanently increasing the monetary base. The “irrelevance results” of Krugman (1998) and Eggertsson and Woodford (2003) instead pertain to a policy that increases the monetary base only during a period over which the zero lower bound prevents the central bank from achieving its usual targets, while the central bank is expected to return to its usual (purely forward-looking) approach to policy once it is no longer constrained “” for example, by returning to the pursuit of its long-run inflation target, or by following a Taylor rule consistent with such a target.”

    Check out Figure 14 and Figure 15 in the paper; the monetary base explodes in Japan and the US but NGDP doesn’t even whimper, given the expectation that the increase is impermanent. That’s why asset purchases only work when accompanied by (credible) explicitly contingent guidance like an NGDP level target.

    In addressing asset purchases by themselves as a tacit form of guidance: (pg. 85)

    “But the kinds of large increases in the monetary base associated with “quantitative easing” in Japan or with the Fed’s recent programs do not suggest particular expectations about future policy in the same way: the expansions have been much too large for any plausible suggestion that they are intended to be permanent, nor is the size of the expansion tied in any obvious way to any aspect of the central bank’s future targets that one might be trying to signal.”

    If you don’t change the target, asset purchases will do little. People just assume the Fed will unwind the expansion (as it has promised)and return to the same old policy reaction function as soon as the lower bound is exited. And therefore the purchases end up having only small effects to the extent they make more credible the Fed’s promise to maintain low rates and/or narrowly push down yields if asset markets are segmented.(see pg. 66 regarding segmentation)

  3. Gravatar of Rajat Rajat
    6. September 2012 at 05:34

    Australian NGDP growth down to 3.3% y/y, the slowest in 20 years outside the financial crisis. The RBA is following the Fed down its dead end street.

  4. Gravatar of Bill Woolsey Bill Woolsey
    6. September 2012 at 05:46

    Perhaps you should explain better about the “buying up planet earth” raising nominal GDP. What are the concrete steps….

    Woodford argued that having the Fed by risky and long term assets will be ineffective in theory because taxes might have to be higher because the Fed might earn less money, and so provide less money to the Treasury. So, investors hold more “safe” Federal Reserve money and fewer risky securities.

    In my post on this, I said that there is an issue with “corner solutions.” What happens when the Fed is buying BAA debt, and everyone else sells their BAA debt (and holds Fed issued money) and the Fed is only buying from firms issuing the debt to buy capital goods?

    Who spends less on output? Is it Ricardian equivalence? People reduce their consumption so they can accumulate safe assets to pay the extra taxes they might have to pay if the Fed makes less money and transfers less to the Treasury?

    If the Fed spends an amount equal to the target for nominal GDP on currently produced output, then nominal GDP will be on target even if no one else spends anything.

    OK, it is hard to imagine that no one spends anything on consumer goods. Suuppose the Fed buy buys enough capital goods so that when that is added to consumption, spending it equal to the target. There is no direct purchases of capital goods by the private sector, only the Fed buys them. The Fed offers to rent them to private firms, which it may or may not do.

    I think some thought about “buying the world” is worthwhile, no matter how unrealistic.

    By the way, I completely agree that even if purchases of risky and long term assets is “less effective” that purchases of safe and short securities, that just means that more should be purchased.

  5. Gravatar of Alex Armlovich Alex Armlovich
    6. September 2012 at 05:49

    It’s funny: quantity theory policy (as presented in the Auerbach and Obstfeld paper Woodford mentions) and interest rate policy look functionally similar in the end. In meeting an NGDP target or a gap-adjusted price level target, permanent expansion of the monetary base requires keeping target interest rates low after exiting the zero bound (since hiking rates unwinds the bloated monetary base through OMOs), and keeping target interest rates low necessarily increases the permanent monetary base enough to meet the NGDP or price level target. Targeted expectations are the great unifier.

  6. Gravatar of johnleemk johnleemk
    6. September 2012 at 05:59

    This is one of the clearest, most lucid posts I’ve seen from you in a while, Scott. Great, succinct explanation of the market monetarist position, with many quotable soundbites (which I wager are more important than many writers give them credit for).

  7. Gravatar of ssumner ssumner
    6. September 2012 at 06:04

    Saturos, Thanks, I hope to do a book soon. That should help put the argument all in one place. This one is part of a trilogy I have planned for today, two more are coming. All will address the Woodford/Krugman worldview.

    Elsewhere you asked why I like the EMH. I think my study of financial markets during the Great Depression was my biggest influence. I was continually amazed at how prescient the markets were during this period. How they saw one step ahead in the policy game. I have to admit I also like it for “bad reasons”:

    1. The elegance of the “wisdom of crowds” model.
    2. It’s contrarian on bubbles.
    3. It refutes loud-mouthed smarty-pants. “If you’re so smart . . . ”

    Rajat. Good point, but too soon to say:

    1. Isn’t Aussie population growth slowing?
    2. With China slowing, the commodity industries will slow. They are a less important part of NGDP. Recall that NGDP is actually a proxy for the true ideal policy, stable hourly wage growth. How are hourly wages doing?
    3. It’s all about level targeting; the key in the post-2009 period was making up for the shortfall.

    Having said all of that, you may well be right. A warning sign would be a deep drop in long term bond yields.

    Bill, I never got into the details because it seemed like a moot point. I can’t imagine the Fed ever getting NEEDING to go beyond T-securities. But in this recession they CHOSE to go beyond T-securities, and made huge profits.

    Thanks Johnleemk.

  8. Gravatar of Becky Hargrove Becky Hargrove
    6. September 2012 at 06:10

    And in a nutshell, Keynesianism equals….lack of confidence! Just because we’re all “dead in the long run” does not mean the battle has to be lost.

  9. Gravatar of dtoh dtoh
    6. September 2012 at 06:20

    Your critics have said OMO don’t work, and your response is OMO will work because you can do a lot of them! ???? Sorry, I don’t think that cuts it. So I’ll repeat my comment from an early post. You need to explain the channel/mechanism by which OMO work.

    If you explain it through the financial asset price channel, the problem and its resolution become elementary.

    Since you accept EMH, then an increase in financial asset prices has to cause market participants to exchange financial assets for real goods and services (spending on consumption and investment). This can happen either through the sale of financial assets currently held or through the issuance of new financial assets. No economist will disagree with this.

    It’s simply a question of correctly defining financial asset prices, and the correct definition is real (nominal minus inflation) risk adjusted expected annualized yields.

    Once you have the correct definition, then an increase in expected inflation (by definition) increases financial asset prices regardless of whether or not nominal interest rates are stuck at the ZLB.

    This is irrefutable and should be obvious.

  10. Gravatar of johnleemk johnleemk
    6. September 2012 at 06:26

    dtoh, you’re absolutely right that it should be obvious, which I would guess is why it didn’t even cross Scott’s mind to mention this. If someone has the power to buy up anything they want by creating as much money as they want, it’s blindingly obvious that they can drive up the prices of everything as much as they want. Yet for some reason every time it’s pointed out that the Fed could, in the blink of an eye, retire all of the US government’s debt, monetary policy skeptics maintain that they do not think this would materially affect inflation.

  11. Gravatar of dtoh dtoh
    6. September 2012 at 07:05


    No. That’s not what Scott thinks. He thinks the hot potato channel causes people to directly spend more on real goods and services because there is more MO floating around. This is precisely where Woodford, Cochrane, Fama, Krugman, etc. all disagree. They think extra M0 just pushes up velocity at the ZLB because people are indifferent to holding M0 versus holding Treasuries.

    Their argument is hard to refute as long as you cling to the hot potato channel. However, if you explain how monetary policy works through the financial asset price channel then their argument is trivial to refute.

  12. Gravatar of D.Gibson D.Gibson
    6. September 2012 at 07:14

    Is this only a problem, when there is no government debt left to buy? I can’t imagine our political system ever being deadlocked an unable to create debt. I am pretty sure that is what Congress does best. The last time we had a forecast for a surplus (far from zero debt), politicians went nuts with spending and tax cuts.

    Why does the Fed need to “profit” on what it buys? Will they run out of ink?

  13. Gravatar of Saturos Saturos
    6. September 2012 at 07:30

    Isn’t Aussie population growth slowing?

    Yes, it is, but only compared to the recent boom of the past seven years. The median projected rate is only 0.2 percentage points less than the historic trend.$File/41020_PopulationGrowth.pdf

  14. Gravatar of ssumner ssumner
    6. September 2012 at 07:47

    dtoh and johnleemk, The hot potato effect is the reason, and the only reason, why the steady state level of NGDP rises in the long run. The asset price chaneel is a factor in getting you from here to there, maybe the crucial factor.

  15. Gravatar of Doug M Doug M
    6. September 2012 at 07:55

    “Everyone (I think) agrees that if the Fed bought up all of planet Earth, then OMOs would be “effective,” i.e. they’d boost expected NGDP growth.”

    You would be wrong. I would expect NGDP to crater.

    “Indeed they often call the purchase of non-government securities “fiscal policy.””

    Central banks buy gold, and reserve currency, and that is called monitary policy and not fiscal policy. The Fed has recently purchased MBS — no one called that fiscal policy…

    “It seems unlikely to me that the demand for base money would ever exceed 10% of GDP”

    The Fed bought $1.3 Trillion in 2009 — about 10% of GDP, and NDGP was still negative.

  16. Gravatar of Patrick R. Sullivan Patrick R. Sullivan
    6. September 2012 at 08:06

    ‘For God’s sake the US government already does insane things like bail out auto companies and banks.’

    Et tu, Scott?

    I agree it was insane to bail out GM and Chrysler. If we’d let them go under, Ford or Toyota or someone else would have merely snapped up those assets at fire sale prices and continued building Chevrolets and Caddies (as long as there were customers who wanted them).

    But what alternative was there, in the crisis, to bailing out the banks (if we agree with Willie Sutton, and Friedman ans Schwartz, that that is where the money is). When we didn’t bail out the banks, in the 1930s it was catastrophic.

  17. Gravatar of Patrick R. Sullivan Patrick R. Sullivan
    6. September 2012 at 08:14

    ‘The Fed has recently purchased MBS “” no one called that fiscal policy…’

    Actually, John Cochrane just did. First he called it fiscal policy, then he switched to calling it combined fiscal-monetary policy.

  18. Gravatar of Major_Freedom Major_Freedom
    6. September 2012 at 08:40

    Indeed they often call the purchase of non-government securities “fiscal policy.” I think that’s wrong, the purchase of high quality corporate debt at going market prices by central banks is just as much “monetary policy” as the purchase of Treasury debt, or the lending of reserves to commercial banks.

    Can you say “cognitive dissonance”? What a crazy definition for “going market price.” This definition includes the prices that are formed in part by the demand that the non-market Fed presents!

    This is like saying the going market prices of securities would have been as high without the Fed standing as a purchaser.

    When central banks declare their intentions to purchase a particular security, by the law of supply and demand, they will not be paying the “going market price”. The price of that security will rise higher than the “going market price”. This is because the money the Fed presents as demand is money created out of nothing, and so the Fed will add a “demand” component to the supply and demand ratio that did not exist before, and hence the price will rise.

    Just because the market typically anticipates the Fed, and increase the price themselves prior to the Fed purchasing the security, it doesn’t mean the Fed is paying “the going market price.” The Fed will end up paying “the going market price PLUS a non-market Fed induced premium”.

    Empirically, notice that when the Fed declared and followed through on Operation Twist, long term yields decreased. The “going market price” for those treasuries no longer existed, and a “going market price PLUS a non-market Fed induced premium” took its place.


    Apparently in market monetarist land, if the Fed declared their intention to buy fake dog poop, solar powered flashlights, and Milli Vanilli CDs, and the prices of those things increased, then we’re supposed to believe that the Fed is just paying the “going market prices” for those things.

    Gee whiz.

  19. Gravatar of K K
    6. September 2012 at 10:35

    Scott: “Everyone (I think) agrees that if the Fed bought up all of planet Earth, then OMOs would be “effective”

    I’m not sure. Williamson consistently argues in favor of a strong version of Wallace irrelevance. Woodford also in his Jackson Hole paper (see also Curdia and Woodford 2011). I tend to agree with you. But to me, purchases of private assets reeks of crony corporatism and industrial policy.

    “I think that’s wrong, the purchase of high quality corporate debt at going market prices by central banks is just as much “monetary policy” as the purchase of Treasury debt”

    Why would we want to confuse things by calling that “monetary policy?” Any kind of investment is monetary policy by your preferred definition. If a private agent issues bonds to dig holes in the ground and then the CB buys the bonds, that’s not “monetary policy.” It’s Keynes by the back door. Of course that kind of “monetary policy” can escape the liquidity trap. I think you are sabotaging the debate by obfuscating the commonly accepted meanings of words. If you want to hand some kinds of fiscal policy to the CB, that’s fine. But it’s not neutral in terms of wealth transfers, and calling it “monetary” doesn’t make it so.

    “The EMH says they should not lose money on average”

    If CB purchases have an effect on asset prices, then that’s almost certainly not true (the EMH won’t be satisfied for the CB).

    “Do or do not, there is no try.”

    Therefore Ben Bernanke can lift space ships with his mind?

    “How much stuff do we have to buy?”

    Given the definition of monetary policy that you’ve accepted for the purposes of this post (treasuries only), that isn’t going to work. If treasuries are negative beta (as Glasner has pointed out), buying
    treasuries will make things worse. Investors *need* those treasuries as portfolio insurance. That is the meaning of “shortage of safe assets.”

    Here’s the problem: the models that say that the ZLB is a real problem actually satisfy ratex/emh in a multiperiod sticky price monopolistic competitive economy. If you want to claim the Lucas Critique moral high ground you can’t do by arguing from a static model with assumptions about
    aggregate behavior (AS/AD). You need to demonstrate that the claimed behavior is actually consistent with the choices of intertemporally optimizing agents. You can do infinitely lived or OLG or whatever. But ratex is tough, and to the extent that it’s been properly done in macro, it doesn’t say what you (or Friedman) would have wanted it to say. It says the ZLB is real and fiscal stimulus is sometimes necessary to keep output at the intertemporal optimum. As Matt Rognlie said, the onus is on you to demonstrate the flaws in those models and show how fixing them leads to your preferred conclusions. Or cede the Lucas moral high ground.

    “Or at the very least, convince the Fed and ECB and BOJ to shoot for a more rapid trend rate of NGDP growth (as in Australia), so that nominal rates stay above zero.”

    Excellent idea. Or revoke paper money (or keep it in fixed supply) so we can take rates negative. The pain of doing that would be vastly smaller than the pain of Japan style permanent stagnation. And *everyone* would agree that sufficiently negative spot rate policy would do the trick.

  20. Gravatar of Rademaker Rademaker
    6. September 2012 at 14:24

    Aren’t there also political constraints on how much the Fed can do? Didn’t Bernanke get called clueless the last time he engaged in QE? What adjectives will the next rounds of QE be described in? Romney is already threatening to remove Ben Bernanke and replace him with (presumably) a more hawkish character. At what point does this grow into a dominant political sentiment?

  21. Gravatar of Rademaker Rademaker
    6. September 2012 at 14:26

    My point is: if it turns out that monetary action is required on an entirely different scale than was initially communicated, suspicion begins to arise among both the public and politicians, and the chance that these will throw a monkey wrench into the policy making process will increase.

  22. Gravatar of Max Max
    6. September 2012 at 15:12

    The meaning of the zero bound is that there’s no such thing as “base demand” when rates are pegged at the CB’s floor. The demand is whatever the CB supplies.

    Quantity of money and the bank rate are not two independent policy instruments.

  23. Gravatar of Bill Ellis Bill Ellis
    6. September 2012 at 15:54


    Yep, It seems to me that Scott, Yglesias and the rest, believe the Fed is actually independent of politics… Or that markets don’t take into consideration political realties…

    Funny about Yglesias …Just a few weeks ago he was explaining why Ben Bernanke does not take Ben Bernanke’s advice in terms of political pressure.
    Now he is saying that Ben can credibly make promises and ignores the possibility of political pressure in making his case.

    Scott has been cagey as to why even though Bernanke keeps saying he can do more…he doesn’t do more…all while all calling Bernanke’s excuses for not doing more nonsensical and lame.

    Scott has pointed out that a weakness in counting on Fiscal stim as a means of getting the economy going is that the FED’s suborn 2% or less inflation target will blunt the effects of fiscal stim. But that Inflation target is ultimately set by POLITICAL concerns.

    Scott does not explain how those same political concerns suddenly disappear and lose their power just because the Fed switches from Inflation targeting to NGDP targeting.

    ( I do think that NGDP is a far better target than Inflation, but I don’t see how it transcends political realities. I would hate to see it sold as a magic bullet and then discarded as discredited.)

  24. Gravatar of dtoh dtoh
    6. September 2012 at 16:30

    You said, “The hot potato effect is the reason, and the only reason, why the steady state level of NGDP rises in the long run. The asset price chaneel is a factor in getting you from here to there, maybe the crucial factor.

    When the Fed engages in OMO, there are two sides to the transaction. One creates excess supply of M0, the other creates excess of demand for financial assets. The only way to get back to equilibrium is through higher NGDP, which both supplies the higher demand for financial assets and absorbs the increased supply of money. So I agree that there is an absolute and necessary correlation.

    The question is what is the causal factor driving the increased demand. Is money pushing increased nominal spending or are higher financial asset prices doing it. IMHO, money is like gas…. the amount people drive determines the amount of gas they have in their tank not the other way around. If you look at real world causality, it’s higher financial asset prices which drives higher spending (NGDP). This in turn absorbs the increased supply of money….not the other way around.

  25. Gravatar of Rajat Rajat
    6. September 2012 at 19:01

    Scott, Saturos, nominal wages have also slowed from 4-5% a couple of years ago to sub-3% now.

    Blogger ricardian ambivalence also has this neat chart showing stagnation in hours worked.

  26. Gravatar of Rajat Rajat
    6. September 2012 at 20:15

    BTW, Australian 10yr government bond yields have fallen from 5%+ up until mid-2011 to 3% now. To date, most commentators have attributed this to ‘portfolio effects’, meaning international investors’ search for highly-rated bonds with good yields. To the extent that economists recognise slowing nominal growth, this is put down to the ending of the commodity/terms-of-trade boom, rather than the RBA’s stubborn resistance to promote growth at trend rates.

  27. Gravatar of Saturos Saturos
    6. September 2012 at 21:21

    I don’t understand Ricardian Ambivalence’s argument. Is he explaining the gap between employment and hours worked structurally or cyclically (NGDPically)?

  28. Gravatar of Saturos Saturos
    6. September 2012 at 21:30

    dtoh, once again, supply and demand for money stocks is what determines NGDP, even in an economy without any financial assets except money. Even in an economy without any real capital assets. MV = PY, V being constrained by people’s money holding requirements.

  29. Gravatar of Rajat Rajat
    6. September 2012 at 21:34

    Saturos, I thought he was explaining it cyclically, but also saying that the nature of the current cycle (20 years no recession) is such that the slowing growth in hours worked has not showed up (yet) in the UnN rate.

  30. Gravatar of Saturos Saturos
    6. September 2012 at 21:38

    But yes, in an economy where eg. the stockmarket does exist, stocks etc. will have to go up before spending does.

    Actually that’s not quite right either. As Nick Rowe would say there is “artsie non-linearity”; these things tend to be co-determined simultaneously. The economy jumps to the new equilibrium where eg. money has lost half its value, once everyone expects everyone else to behave as though it has. Expectations need to be about something, you say? Let them be about the central bank, and its infinite ability to arbitrarily set the available quantity of unit of account-ing and medium of exchange-ing material.

    No fiat money central bank can fail to get NGDP exactly where it wants. Of course, I think in places like Japan, the central bankers are not quite aware that what they want is in fact so perverse – they don’t know their own power, even as they exercise it.

  31. Gravatar of Saturos Saturos
    6. September 2012 at 21:52

    Rajat, has Australian NGDP growth really slowed so much? Marcus Nunes’s charts show it doing fine up to last year And for this year the RBA says that real growth has offset declining inflation:

  32. Gravatar of "When Keynesians worry about ineffective monetary policy, they really contemplate a situation where the central bank has no intention to do whatever it takes to succeed. The Fed should say, ‘We will do whatever it takes to ensure our forecasts "When Keynesians worry about ineffective monetary policy, they really contemplate a situation where the central bank has no intention to do whatever it takes to succeed. The Fed should say, ‘We will do whatever it takes to ensure our forecasts
    6. September 2012 at 23:02

    […] Source […]

  33. Gravatar of dtoh dtoh
    7. September 2012 at 01:50

    Saturos, I have to respectfully disagree. At the ZLB, money becomes dual purpose, it is both a medium of exchange and a store of wealth (financial asset). Thus V goes out the window and becomes almost entirely function of the transaction costs of converting Treasuries into reserves at the Fed.

    As for the no financial asset economy, I agree, but that’s a special case. You can also hypothesize about a no-money (credit only economy), where NGDP is purely a function of the price and availability of credit (i.e financial assets).

  34. Gravatar of Saturos Saturos
    7. September 2012 at 03:04

    dtoh, I think the ZLB is the special case. It only matters because we conduct monetary policy by exchanging newly-created money for Treasuries (and perhaps eventually other securities). If the real yield on capital, and the securities which allow claims on it to be traded, were forever zero (Keynes’ nightmare), then yes we’d have a problem. Then in the case of inflation expectations vanishing to zero we’d either have to use some mechanism to make money less attractive (such as Lars Svensson’s foolproof mechanism) to restore expectations, until people no longer regarded it as an adequate substitute for real saving. Or we should permanently switch to a better way of conducting monetary policy. We could create a “flexible gold standard”, as in Nick Rowe’s latest post, adjusting the money supply by changing the price of gold, which has no upper bound. Or we could have a moving peg of our currency to a basket of others. Or we could use NGDP futures contracts. Or we could do a currency reform every seventh Tuesday. We’re talking about a fiat-money central bank here.

    As for your no-money hypotherical economy, don’t you realize that the absence of money means either: no unit of account, in which case NGDP doesn’t exist (because standardized prices don’t exist); or no unit of exchange, which means NGDP doesn’t matter, as a drop in the flow of money is irrelevant/has no impact on the exchange of real goods and services? So I think you (and most economists) really need to better appreciate the conceptual centrality of money here. To quote Nick Rowe (again): “Recessions are always and everywhere a monetary phenomenon”.

    Scott, when they give you the Nobel I hope you insist that it be shared with Nick, at the very least.

  35. Gravatar of Dtoh Dtoh
    7. September 2012 at 04:57

    I totally agree with your first paragraph, except if you want to make money less attractive just go electronic and give it a negative interest rate. I also think it’s important to emphasize your point that monetary policy is an exchange. It’s not the mere issuance of money.

    As for a no money economy (credit only), there’s no reason you can’t have a unit of account. It can be anything as long as you can control the supply and everyone agrees on it.

  36. Gravatar of Rajat Rajat
    7. September 2012 at 20:14

    Saturos, Marcus’s chart is pretty small. I just calculated y/y NGDP from the national accounts and found much lower rates over the last couple of years than over the past two decades (excluding the GFC). But I accept that the real shortfall has been the last 18 months or so, since late 2010. Based on peak participation rates, Australia’s UnN rate is about 6%. I don’t see why that is necessary.

  37. Gravatar of ssumner ssumner
    8. September 2012 at 10:53

    Patrick, The 50% fall in NGDP was catastrophic. If we target NGDP at a 5% growth rate, then a few bank failures will be tolerable.

  38. Gravatar of Shining Raven Shining Raven
    26. September 2012 at 01:07

    Thank you very much for this post, now I finally understand what your NGDP-targeting is about. I could not understand the mechanism that you wanted to use, but now I do: You want to use an expansionary fiscal policy a la Keynes, but you want the Federal Reserve to do it instead of the “government” (Congress), and you want to call it “monetary policy”, since it is the Fed that is doing it. Got it.

    I think K at 6. September 2012 at 10:35 nails it: What you call “monetary policy” is nothing of the sort, at least if the normal use of the word has any meaning.

    Sure, if the Central Bank buys up “the world”, as you say, this would certainly raise NGDP. But why exactly would this be preferable to Congress doing the same thing? It certainly would be exactly as inflationary (which of course in this case is a good thing and expected, as I understand), but the Central bank would have to make an awful lot of investment decisions in where to put its money.

    Exactly why would it be better to have the Central Bank decide where to invest all this money instead of all of us (through our elected representatives)? So then the Central Bank would decide where to build a new hospital, or a toll road? Or in which company to invest (by buying their “high-quality assets”)? This would certainly open the door for a lot of cronyism.

    Why would it be better if “unelelected government bureaucrats” make these investment decisions? (And yes, by any normal measure the Fed is of course part of the govnerment.) Do you think the Fed would be immune for long from a terrible onslaught of lobbying? Now this goes on at the Congress, because this is where the public money is, but do you think people are not going to try to influence the Fed’s decision once it starts to invest in the private sector to an unprecedented degree and starts buying assets left and right? Yeah, sure.

    So I don’t really see the point of your idea. There are already mechanisms in place for doing fiscal policy. I suggest to use them, instead of coming up with a half-baked idea of a new path of doing this through the Fed, which will have the same problems (waste and cronyism), without having any safeguards (in the form of public accountability) in place yet.

  39. Gravatar of Translating Market Monetarists « azmytheconomics Translating Market Monetarists « azmytheconomics
    28. September 2012 at 09:43

    […] owner of 100% of all Treasury liabilities, lower IOR to 0%, and inflation would not change at all. Market monetarists don’t believe this could happen. 2. Inflation/NGDP would increase despite 0% interest rates. This may seem kind of Green Lanternish […]

Leave a Reply