Krugman on the limits of monetary policy

Here’s Paul Krugman, in his pessimistic mood:

You might think that it’s a fundamental insight that doubling the money supply will eventually double the price level, but what the models actually say is that doubling the current money supply and all future money supplies will double prices. If the short-term interest rate is currently zero, changing the current money supply without changing future supplies “” and hence raising expected inflation “” matters not at all.

And as a result, monetary traction is far from obvious. Central banks can change the monetary base now, but can they commit not to undo the expansion in the future, when inflation rises? Not obviously “” and certainly “credibly promising to be irresponsible”, to not undo expansion in the face of future inflation, is a much harder thing to achieve than simply acting when the economy is depressed.

Just to be clear, Krugman is not saying the central bank must promise a specific future money supply on a specific date, he’s saying the expected future money supply must be large enough to produce a specific expected future inflation rate (or price level.)

There are several possible solutions to the credibility problem.  One that both Michael Woodford and I have discussed is level targeting.  Suppose the central bank always falls short of its price level or NGDP target by 1%, due to a lack of credibility.  You might argue that this describes the current situation in Japan, for instance.  If the central bank has a level target, that shortfall only has significant macroeconomic effects in the very first year.  After that the level of the target variable continues to fall 1% below target, but the growth rate of the aggregate (which is what really matters) is always on target (after the first year.)

A second solution is to adjust the monetary base as needed to peg the price of CPI futures, or NGDP futures.  That will keep expected future growth in the nominal aggregate right on target.  If that seems “to good to be true,” it’s because it exposes the fact that worry about “liquidity traps” is actually worry about something else—the size of the central bank balance sheet.  But in the long run the central bank balance sheet will be smaller with a more expansionary policy, and perhaps (as the Swiss central bank showed a few years ago) even in the short run. If the central bank balance sheet is too big for comfort when you are hitting your target for expected future inflation, or expected future NGDP growth, then raise the Price level/NGDP target path or lower the rate of interest on reserves.

In my view this is where Krugman goes off course.  He assumes that if the central bank has done a lot, and has still fallen short, it would have had to do even more to succeed—bleeding into fiscal policy. In fact, just the opposite is true.  The central banks that succeed are those (like the Reserve Bank of Australia) that do the least. That’s because faster NGDP growth leads to a much lower desired ratio of base money to GDP, and hence smaller central bank balance sheets.

Krugman continues:

Just to be clear, I have supported QE in both Britain and the US, on the grounds that (a) central bank purchases of longer-term and riskier assets may help and can’t hurt, and (b) given political paralysis in the US and the dominance of bad macroeconomic thinking in the UK, it’s all we’ve got. But the view I used to hold before 1998 “” that central banks can always cause inflation if they really want to “” just doesn’t hold up, theoretically or empirically.

I seem to recall that Krugman was quite pessimistic about the ability of the BOJ to succeed in producing inflation, at least a few years ago.  But when they raised their inflation target to 2%, they did succeed in moving from deflation to mild inflation (albeit still short of the 2% goal.)  The good news is that we now know that the Japanese government can sharply depreciate the yen anytime it wishes to (something else that Krugman had originally doubted.)  That means the BOJ can inflate–it’s just a matter of how committed they are to make it happen.  In a sense this is also Krugman’s view (when he’s in the more optimistic mood), and he’d undoubtedly argue that the political push from Abe helped the BOJ.  In the past Krugman seemed to think fiscal stimulus was an easier sell politically.  That may be true in a few cases, but in Japan monetary stimulus has turned out to be the easiest part of the three-part reform project.  And monetary policy doesn’t have to worry about fiscal offset, in the way that fiscal policy must worry about monetary offset.

On the other hand, my recent Econlog post on the ECB shows an almost Krugmanian level of pessimism.  I just don’t see the institutional resolve at the ECB to raise inflation up to their 1.9% target. Europe badly needs to rethink everything they are doing, from the ground up.

HT: Ken Duda



57 Responses to “Krugman on the limits of monetary policy”

  1. Gravatar of Michael Byrnes Michael Byrnes
    16. December 2014 at 16:20

    But if Krugman had offered this as an explanation for the past six years (not as a conclusion about the effectiveness of monetary policy in general), you would agree with him, right? (Which is not to say that that justifies his broader claim).

  2. Gravatar of benjamin cole benjamin cole
    16. December 2014 at 16:20

    1. What atavistic impulses are behind the concern over the size of a central bank balance sheet? Why should the Fed have a balance sheet of $1 trillion or $5 trillion? How would various sizes impact real growth? Indeed, if a central bank can pay off national debt and nothing happens…

    2. It may be the Fed should explicitly commit to keeping a “large” balance sheet indefinitely, to signal resolve.

    3. If inflation and interest rates are dead, then is not QE a conventional tool?

    4. Given Fed sqeamishness regarding QE, and how quickly the US will hit ZLB in next recession, is it almost certain the Fed will be much too late in going to QE next time?

  3. Gravatar of benjamin cole benjamin cole
    16. December 2014 at 16:24

    PS on Europe: can you imagine being a citizen of the nation of Spain? It is agonizing to be a citizen of the United States, where one might feel there is some sort of lever as a voter to influence monetary policy, at least in the long run. The Spaniard?

  4. Gravatar of Morgan Warstler Morgan Warstler
    16. December 2014 at 16:26

    We get some good VISIBLE deflation (not the invisible digital stuff), I sure hope this sin’t causing NGDP to drop:

  5. Gravatar of Michael Byrnes Michael Byrnes
    16. December 2014 at 16:37


    Someone did a good blog post about oil prices:

  6. Gravatar of Ray Lopez Ray Lopez
    16. December 2014 at 17:50

    PK sez: “You might think that it’s a fundamental insight that doubling the money supply will eventually double the price level, but what the models actually say is that doubling the current money supply and all future money supplies will double prices. If the short-term interest rate is currently zero, changing the current money supply without changing future supplies “” and hence raising expected inflation “” matters not at all.

    It amazes me that PK is taken at his word–and Sumner seems to agree– when this sentence is clearly wrong. Let us do a thought experiment. Leaving aside money illusion, which would, if it exists, cause prices to rise (I know, I know, money illusion probably existed in the 1930s, it exists in small-scale college experiments involving students and small amounts of money, but it’s not clear in today’s more savvy age money illusion really exists, albeit sticky prices and wages caused by persistent central bank fiat money inflation does probably exist, but I digress): make the money supply 10 million times greater than today, but then tell people, Ben Bernanke style: “we promise, we won’t do it again”. Do you really think this will not raise prices? You cannot be serious if you do. Rate of change is important, as PK says, but absolute level of money, a jump from a previous level, is also important.

  7. Gravatar of bill woolsey bill woolsey
    16. December 2014 at 18:30


    We won’t do it again? The “problem” is when the Fed says that the quantity of money will return to its initial value in the near future.

  8. Gravatar of Major.Freedom Major.Freedom
    16. December 2014 at 18:42

    If history is a guide, then it is the case that central bankers take as most credible and most serious the intellectual environment and guidance from financial insiders and entrenched political elites, not academic economists.

    If a central bank appears to be acting against what “the average economist” expects, then it is because the economists don’t understand, or don’t want to accept, the central bank’s main influences.

    Economists by and large follow, and study, and react to what central bankers do. They need old published macro data before they are allowed to say or recommend anything that is taken seriously by the journal editing gatekeepers. By the time that happens, central bankers, financial insiders and political elite are already engaging in new plans and schemes.

    Once those schemes are made public, quite a number of self-important academic economists are helpless and tend to roll their eyes, tsk tsk, and finger wag as if their opinion ultimately mattered. Academic economist opinions are only made to appear to matter by central banks. They promote, finance, and reward those who toe the party line of financial insiders.

    Was the consensus of economists’ beliefs the motivator for why the NY Fed (secretly at the time) sent over $40 billion to Iraq to finance the “reconstruction”? Was the consensus responsible for at least $9 trillion bail out of Wall Street during the last crash? Was the consensus responsible for QE combined with IOR, which enabled the financial elite to be saved while Main Street crumbled?

    Oh how tragic a life can basement dwelling pencil necks live when they have both a PC and delusions of grandeur.

  9. Gravatar of Morgan Warstler Morgan Warstler
    16. December 2014 at 18:56

    Every once and a while, Scott goes astray…

    “However falling oil prices have no implications for global growth—it merely redistributes global wealth.”

    This is clearly wrong. Wrong,

    Low input costs for fleet of foot free market economies with a stronger penchant for entrepreneurial growth – like say tons of FREE DIGITAL SHIT….

    Is far better for global growth better than reduced supply rents paid to autocratic third world dictators who wouldn’t know what to do with wealth if God himself wrote is all down in an ancient manuscript for them to read.

  10. Gravatar of Adam Platt Adam Platt
    16. December 2014 at 19:02

    Title should read: Krugman on the limits of Krugman’s understanding of monetary policy.

    I think I’m beginning to grasp his misunderstanding here. He is only capable of thinking about monetary policy (or any policy for that matter) in terms of short term “stimulus”. Therefore he assumes that any money added now will simply be withdrawn once we’re done “priming the pump”.
    When he says monetary policy doesn’t work below the ZLB, what he actually means is that monetary policy doesn’t work below the ZLB ASSUMING that it doesn’t change long-term expectations! That basically means he misses the entire point of market monetarism. He seems to respect you Scott, you may want to reach out to him to explain this discrepancy.

  11. Gravatar of OhMy OhMy
    16. December 2014 at 19:52

    Hard to know whether to laugh or cry at this post. In a corporate world you would be fired for suggesting your first “solution”.

    “I have a brilliant idea, we will undershoot our sales targets by 1% every year, that’s how the growth rate will be correct”

    Wow, just wow. Please never apply for a job outside academia.

  12. Gravatar of Matt Waters Matt Waters
    16. December 2014 at 20:07

    As I commented in the last post, my only issue with these sorts of arguments is they often get causality backward. For example, “level targeting” is considered a solution to credibility. But whatever the target, for expectations to work, the Fed and ONLY the Fed has to be capable of hitting that target. The Fed’s actions have to change what everybody else does, even if nobody believes it will be successful.

    This is particularly important for the case where the US did not backstop the financial system. What if the most dovish Fed action possible was monetizing all of the debt and Agency MBS? If IOER is 0%, would the entire monetization merely become excess reserves? What if negative interest was possible on IOER and how would cash withdrawals be handled?

    The panics before the Fed all had sharp negative NGDP and RGDP declines. Solvent banks kept themselves from going under without a LOLR through suspending deposits. The panic in 2007-08 could have been even more severe in that, as Lehman showed, there was no way in our modern financial system to force depositors to renew their overnight repo. The market’s reaction to TARP failing is also indicative of the possible damage wrought by the banking panic.

    With a stable banking system and just a general deflationary malaise to the country, the market monetarism stuff all seems fine. QE in the US and inflation targeting in Japan have worked in part with the implicit backstops to those countries’ banks. It’s just hard to embrace market monetarism completely though seeing some possible boundary cases and not knowing where those boundary cases lie.

  13. Gravatar of Tom Tom
    17. December 2014 at 00:46

    Why not just admit the obvious, that when rates are zero, the monetary policy spigot is fully open, and there is nothing more the central bank can do to increase spending in the real economy, other than to spend in the real economy itself. Devaluation is not the kind of inflation anyone’s looking for. Import inflation is pure cost.

    Sure, it would be a shocking reversal, and might alienate your fans. But it’s so, well, blindingly obviously true.

    I know, I know, you can’t. Never mind. Carry on.

  14. Gravatar of Daniel Daniel
    17. December 2014 at 01:32


    when rates are zero, the monetary policy spigot is fully open

    You must be new around here

  15. Gravatar of dtoh dtoh
    17. December 2014 at 02:57

    For the millionth time. Expanding MB if you let ER go up by a comparable amount has no impact. Creating a book entry between the NY Fed and Chase is no different than creating a book entry between the NY Fed and the St. Louis Fed.

    You and PK should stop talking about the doubling the money supply. MB is a totally bogus and useless definition of the money supply if you don’t subtract out ER.

  16. Gravatar of Nick Nick
    17. December 2014 at 03:44

    Various fed members have argued for about two years now that one of the major downsides of current policy was potential (possibly quite lagged) ‘financial instability’ in the high yield market.
    This has put them in a position where they can tighten, invert the yield curve, wreck the high yield market, and blame the damage on ‘easy’ policy 18 months prior.
    I’m worried. To me we are entering ‘worse than a crime; a mistake’ territory again with the FOMC.

  17. Gravatar of Nick Nick
    17. December 2014 at 03:46

    Above comment inspired by this Tim Duy post:

  18. Gravatar of Michael Byrnes Michael Byrnes
    17. December 2014 at 04:09

    Matt Waters wrote:

    “As I commented in the last post, my only issue with these sorts of arguments is they often get causality backward. For example, “level targeting” is considered a solution to credibility. But whatever the target, for expectations to work, the Fed and ONLY the Fed has to be capable of hitting that target. The Fed’s actions have to change what everybody else does, even if nobody believes it will be successful.”

    I think actually hitting the target may be less important than the predictability of the Fed’s policy stance. During the past 6 years, we have had all sorts of debate about the risks of the high base, when and how the Fed will remove the excess reserves from the system, whether policy should be tightened in ordet to pop bubbles, asset prices falling on the release of good economic news because some people expect monetary contraction in response, etc. Under NGDPLT, all of that noise would be gone or greatly minimized. Think of it as a more credible and data-dependent announcement of when the first rate hike will happen.

  19. Gravatar of Daniel Daniel
    17. December 2014 at 06:35

    Ray Lopez

    sticky prices and wages caused by persistent central bank fiat money inflation does probably exist

    See, this is why people here think you’re an idiot – because you say dumb crap like that.

  20. Gravatar of ssumner ssumner
    17. December 2014 at 07:19

    Michael, I’d certainly agree with the part about temporary currency injections not being effective.

    Ben, Yes, balance sheets are a phony problem.

    Ray, you misunderstood his claim.

    Morgan, More oil, yes, but not lower prices due to less demand.

    OhMy, I don’t know whether to laugh or cry at commenters who tell me I’m crazy, but can’t find any specific flaws in anything I said.

    Matt, You miss the fact that the financial crisis was being caused falling NGDP expectations for 2009 and 2010. With level targeting asset prices hold up much better and the crisis is far milder (not that there wouldn’t be some financial problems.)

    Tom, Obviously you are not familiar with MM arguments. You might want to educate yourself before dismissing them. Interesting that my opponents (you, OhMy, etc) can’t muster a single argument against anything I said. That makes me even more confident that I’m on the right tract.

    And when was “the monetary spigot fully open?” other than in Zimbabwe?

    dtoh, The Fed controls the MB directly and both NGDP and the base minus ERs indirectly. You can talk about the base and NGDP, in which case adding MB-ER adds nothing to the model.

    Nick, My sense is that the biggest threat to the high yield market is a recession, or in specific cases like oil company bonds, falling oil prices. I don’t think rising rate s are a big threat, if those increases reflect stronger growth.

  21. Gravatar of Adam Platt Adam Platt
    17. December 2014 at 07:28


    Scott has addressed that point many times. It’s Bernanke you should be ranting against.

  22. Gravatar of Brian Donohue Brian Donohue
    17. December 2014 at 07:36

    Morgan, good comment on oil.

  23. Gravatar of ssumner ssumner
    17. December 2014 at 07:36

    dtoh, I’d add one point. The minute you define “money” as the base minus ERs, Krugman will claim the Fed has no control over the base

  24. Gravatar of Nick Nick
    17. December 2014 at 07:43

    Professor Sumner,
    What are ‘rising rates’? The yield on the 10yr keeps flirting with 2. The only rates that are in danger of rising are the ones where we most directly observe the secondary effects of monetary policy. Growth isn’t causing ‘rates to rise’. It looks to me like growth is correlated with expectations of monetary tightening…
    Honestly, I hope I’m totally wrong about this. But you say the FOMC has no clear goal. I see statements from them making excuses for why the next recession won’t be their fault all the time. You say the risk to the junk market is recession … They say that if a down junk market causes a recession it won’t be their fault. I don’t think the story has to end badly but … Yikes …

  25. Gravatar of ssumner ssumner
    17. December 2014 at 07:52

    Nick, Yes, the Fed is not fulfilling its mandate, and that is very worrisome. My comment on interest rates was hypothetical. Thus if rates rose significantly in 2015 (which I view as unlikely) it would be harmful if due to tight money, but not if due to strong growth.

  26. Gravatar of maynardGkeynes maynardGkeynes
    17. December 2014 at 08:00

    I think someone (Brad?) at Equitable Growth thinks that PK is all wet on this.

  27. Gravatar of Charlie Jamieson Charlie Jamieson
    17. December 2014 at 11:53

    Another article ascribing superhuman powers to the Fed.
    The central bank does what the private banks want it to do. The private banks create the money and the central bank rubber stamps what they do. The only question going into the financial crisis was, ‘Would the central bank let private banks collapse, or would it buy their bad debt?’ When the Fed bought mortgage securities, that question was answered.

  28. Gravatar of TravisV TravisV
    17. December 2014 at 12:26

    Yglesias: “How the Federal Reserve boosted the economy with a single word “” “patient””

  29. Gravatar of Matt Waters Matt Waters
    17. December 2014 at 13:00

    Well, where the “blurred lines” existed in 2008 was not with the Fed’s stated goals. If we assume a rational market, then the stock market moved to signal an extreme long-term depression. TIPS spreads are tougher to use in 2008-09, because that market had a lot of liquidity issues. Treasuries should ideally all be “risk-free,” but on-the-run Treasuries have an irrational risk premium purely because market participants have agreed on making that those the most liquid securities. As LTCM found out, those liquidity premia go up in times of distress even though it’s completely irrational on a long-term, discounted-cash-flow basis.

    So, anyway, looking at the stock market, it forecasted extreme NGDP shortfalls for several years at the market’s trough in 2009. Nobody would have said that the Fed’s targets were so malleable that the range for their accepted inflation included severe deflation. Instead, the issue was with the Fed’s possible mechanics. If the mechanics permissible for the Fed only included the Fed funds rate, then the Fed could state an explicit 2% CPI level target or 5% NGDPLT all it wanted to. Bernanke could yell day and night that they wouldn’t raise the Fed funds rate until it got back up to that level target, and it wouldn’t make a difference.

    One possible issue is whether the market was honestly rational. QE1 was announced November 2008, but the S&P went from 1250 at beginning of 09/2008 to 900 in 10/2008. It stayed mostly flat and does not show a marked change in 11/08. It continued flat or declining until a bottom of 683 on 03/06/09. The only thing the bottom seems to correspond with is the Fed’s announcement of TALF, which was somewhat different than QE1 where the Fed expanded its allowed collateral.

    It’s hard to say that TALF marked some sort of dramatic shift. It’s even harder to say that the Fed’s policy goals in fact changed along with the smooth upward trajectory of the market in second half of 2009.

    Is this sort of accuracy too much to ask for of the markets and the Fed’s goals? Probably, I’m an engineer not an economist. I think David Beckworth made a post about expectations which he specifically addressed to a “mechanical engineer with an MBA,” which happened to describe me exactly. As a mechanical engineer, I can see how efficient markets could look quite turbulent and chaotic and still be efficient. You would expect the correlation between stocks to go up significantly in a situation where the Fed’s goals are unknown, as new information concerning the Fed’s actions is efficiently processed by the market to apply to all stocks. And the increase in correlation is exactly what we saw.

    But my literal, mechanical approach also brings me to somewhat different conclusions concerning expectations, monetary policy and (possibly) efficient markets. It would be easier if axiomatically markets were always efficient, but there’s too many cases where markets described an extremely unlikely or impossible future state of the world. Irrational bull markets are more likely than bear markets because it’s extremely tough to short stock at irrationally high levels.

    So the market is taken somewhat with a grain of salt, where sometimes the prices give an accurate expectation of Fed policy and sometimes they don’t. And again, Fed policy could be “5% NGDPLT” screamed from the rooftops but if the Fed only has the Fed Funds rate to work with, then the policy won’t be hit. They need to be able to get there with more market participants increasing NGDP, even if they have no idea who the Fed is or what their policy is, and just look at their own financial situation in isolation. Otherwise, even if the market acts rationally, they may rationally consider the Fed impotent as long as the possible mechanics don’t clearly meet the target without the market’s help.

  30. Gravatar of Daniel Daniel
    17. December 2014 at 13:03


    Another article ascribing superhuman powers to the Fed.

    Since they OWN the printing presses, I’d say they do have superhuman powers.

  31. Gravatar of Joel Aaron Freeman Joel Aaron Freeman
    17. December 2014 at 13:30

    Scott gives two solutions to Krugman’s concern: (1) Level Targeting and (2) Market-driven policy.

    These are perfectly reasonable solutions, but both are rules-based. It seems to me that New Keynesians are very averse to rules-based strategies. They always seem very unsure of themselves, and therefore keen on having an exit strategy given their lack of confidence.

    One possibility is setting a growth rate for base money. That is, set a base money growth rate at 5 percent, and then use discretion to adjust up or down to 6 percent or 4 percent as needed. This solves Krugman’s problem because a growth rate of 5 percent implies that base money will tend to increase at that rate for all future years. If they decrease to 4 percent, that implies that base money will tend to be lower than previously expected for all future years, and yet the monetary base never shrinks.

    I’m not advocating a first-best solution. I’m only suggesting something more palatable to people like Krugman and Yellen.

  32. Gravatar of dtoh dtoh
    17. December 2014 at 13:46

    The Fed has complete control over ER and therefore has complete control over the money supply (when defined as MB – ER). An increase in the base accompanied by an equal increase in reserves has zero impact on AD. Zero. By not being absolutely clear about this, you enable to people to make the annoyingly common argument that “doubling the money supply has had no impact on the economy.”

  33. Gravatar of Adam Platt Adam Platt
    17. December 2014 at 13:51

    Agree with you dtoh. Bernanke has done an incredible disservice to monetary policy by implementing QE without understanding what it’s supposed to do and how it’s supposed to work.

  34. Gravatar of Charlie Jamieson Charlie Jamieson
    17. December 2014 at 13:54

    Daniel, my point is that the banks have the printing presses, and not the Fed. Banks make loans and create the money.
    Even treasury bonds (a form of money, imo) are created by the government making fiscal policy.
    The Fed is a creature of the banking system. Stop analyzing what the Fed might do and understand what it is forced to do. Being powerless, it tries to act as if it has power, and we agree to pretend this fiction.

  35. Gravatar of Matt Waters Matt Waters
    17. December 2014 at 16:10

    The Fed doesn’t have complete control over the printing presses. That ultimately belongs to Congress and the President, who make the laws that limit the Fed’s actions. The Fed, for example, cannot directly fund the US government and instead can only buy and sell Treasury bonds on the open market.

    Bernanke used a clause in the original 1913 law for extraordinary circumstances to do a lot of the stuff he did in 2008-09. That clause had never been used before.

    I wouldn’t say the banks “have the printing presses” either. Let’s say somebody sells a million dollars in Treasury bonds to the Fed. The person selling it has to deposit that money in a bank right? Couldn’t the bank keep the entire million as excess reserves but then that person decide to buy something with his deposit? Let’s say he buys a new lakehouse in cash from a developer who uses the same bank, and let’s say all the contractors use the same bank. NGDP would go up in this case even though M2 only increased by the same amount as what the Fed printed.

    That’s a somewhat silly example, but it shows how we can put too much on to the banking system, M2, money multiplier, etc. There are many ways where NGDP increases without having to rely on bank loans. Banks are market participants like anyone else, who rely on expectations of future growth to make decisions like anyone else.

  36. Gravatar of James in London James in London
    18. December 2014 at 00:44

    FOMC last night made it clear they want to “normalise” monetary policy. What is normal monetary policy though? A happy 1950s time when the sun always shone, the economy grew strongly and America led the free world. In a way, I suppose it is.

    By normal they could just mean successful. In that case the policy is more or less normal now. So, they should wake up and realise there is a “new normal”.

    Perhaps, we should compulsorily retire central bankers at 40 so they don’t ossify and become stuck in their ways.

  37. Gravatar of Brian Donohue Brian Donohue
    18. December 2014 at 06:25

    @James in London,

    “By normal they could just mean successful. In that case the policy is more or less normal now. So, they should wake up and realise there is a “new normal”.”


  38. Gravatar of Charlie Jamieson Charlie Jamieson
    18. December 2014 at 12:35

    Matt, the banking system has the printing press because loans create deposits. Banks don’t lend you depositer money — they create new money.
    Bernanke did what he did to save the banks. He might have believed that saving the banks would save the economy — that is debatable.
    But he is a banker first, so he saved the banks.
    ‘Expectations of future growth’ is a nebulous concept. Collectively, market participants have a certain expectation. I think the Fed has marginal impact on these expectations.

  39. Gravatar of Matt Waters Matt Waters
    18. December 2014 at 13:06

    Part of the issue is the two sides are talking past each other. If you mean purely M2, then of course the banks have “printing presses” along with (NOT instead of) the Fed.

    But what matters is not M2. What matters is NGDP, full stop. I gave a case where banks do not make loans but NGDP increases. For a case in the real world, reserves actually went down 2000-07 as NGDP went up without required reserve ratio changing. A lot more activity and loans happened outside the formal banking system, where deposits were transferred but no new M2 was created.

  40. Gravatar of ssumner ssumner
    18. December 2014 at 17:53

    Matt, Whenever the Fed needs more powers the Congress salutes and immediately gives them the power they ask for. In any case, the Fed has never even come close to running out of assets to buy. That is definitely not the problem. The problem is the Fed has had the wrong target. There’s a reason they tapered; they were happy with the level of NGDP.

    And during 2008 they were not even at the zero bound.

    Joel, Krugman would hate a monetary base target; I’d guess that he’d much prefer either of my two suggestions.

    dtoh, It doesn’t matter what they control, it matters what Krugman thinks they control. And he doesn’t think they control MB-ER. Obviously you don’t have to convince me that the Fed can control any nominal aggregate they like, including the nominal size of the ice cream industry.

  41. Gravatar of dtoh dtoh
    18. December 2014 at 17:56

    If you roll over it certainly won’t convince PK. Build a robust definition of the money supply into your model and your arguments. When ER were typically zero, MB was a perfectly adequate definition. Now it’s not. Stop being stubborn.

  42. Gravatar of Matt Waters Matt Waters
    18. December 2014 at 19:17

    Well, the Fed did go to the zero-bound pretty quickly in 2008 after Lehman. I know the first FOMC meeting they didn’t, but when the severity of the recession became clear, they quickly went to zero.

    My argument boils down to this: Let’s say the Fed had none of its extraordinary support of the banking system in 2008. That means no Bear Stearns and AIG bailouts, no TALF facility, no discount window lending and no commercial paper/money market backing. Also Congress does not enact TARP or any sort of fiscal stimulus.

    Let’s say the Fed instead makes their policy clear: 5% NGDPLT where first the Fed funds rate is reduced and then Treasury and Agency MBS were bought until NGDP forecasts/futures meet that 5% target. Is there a guarantee that the 5% NGDPLT would be satisfied? It’s hard for me to say the Fed would meet the target WITHOUT the banking support.

    The empirical evidence of QE announcements affecting the stock markets occurred when the government also backstopped the banking system. Expectations work when the Fed is within the bounds of its power to fulfill those expectations without the markets’ help. If the Fed does not need the market, then the market will believe the Fed and front-run the Fed. The Fed then does less actual work. For QE and long-term rate forecasts, the market in fact did front-run the Fed where the higher asset prices had far more effect than the QE or low interest rates itself.

    If the Fed does need the market, then the market may or may not move NGDP growth towards the Fed’s target. If the Fed becomes completely stagnant or impotent, then the market becomes this chaotic game with multiple Nash equilibria. For example, if the monetary base was always constant, then the market could remain on a fixed NGDP growth forever. If even one irrational participant changes its action though, then everybody else would rationally follow that one irrational participant.

    The chaotic game of irrationality changing rational actions fairly clearly describes the market’s actions after Sep. 2008. Although the Fed kept the Fed funds rate at 2% on 9/16, the Fed reduced the rate on 10/8 by 0.5, 10/29 by 0.5 and 12/16 to zero percent. The market showed little reaction to these moves on these days. So, if the markets were efficient, they had already priced in these moves and markets nevertheless forecasted another Great Depression.

    My hypothetical above adds in the possibility of unlimited asset purchases, but of only Treasuries and Agency MBS. There are two question that come to mind:

    1. How would the Fed not run into the same issue after exhausting its asset purchases and still not moving economic growth to its desired level?

    2. Are there technical or financial stability issues with doing such large-scale asset purchases? I know that “financial stability” is somewhat nonsensical, but a lot of people do in fact respond with the financial stability issue. Also the fact that all these asset purchases have to be real buy orders executed at the New York Fed trading desk. Not everybody out there owning a Treasury is wanting to sell it right now and so buying every single one could be tough, to say the least. The premium to go all the way down the Ask list would probably make the Fed pay prices equivalent of negative yields.

    So you have all that, while fiscal stimulus is relatively simple in how it works and does not have these bounds or issues. It’s pretty easy to see how Krugman and many other Keynesians have been reluctant to toss aside fiscal stimulus.

  43. Gravatar of Matt Waters Matt Waters
    18. December 2014 at 19:25


    Defining monetary base is fine and easy. It’s just hard to see how much monetary base actually MATTERS. If the monetary base doubles, sure the price level will eventually double prices because people holding the ER will die and their heirs will spend it, or whatever.

    But in the long run we are all dead. People spending real money on real things in the present do care very much about near-term and medium-term economic growth. Economic growth, NGDP and prices can go up even with a decline in reserves (required and excess), as we saw 2003-08.

    Indeed, one tool the Fed has is the monetary base and it is a BIG tool. But they also have required reserves ratio and interest rate on excess reserves.

  44. Gravatar of ssumner ssumner
    18. December 2014 at 19:41

    dtoh, The base is a perfectly good definition of the money supply. Recall that I’m a MM, not a monetarist. I’m not claiming the quantity of base money is a useful policy indicator. It isn’t, and MB-ER also would not be a useful monetary indicator.

    Matt, You said:

    “Well, the Fed did go to the zero-bound pretty quickly in 2008 after Lehman. I know the first FOMC meeting they didn’t, but when the severity of the recession became clear, they quickly went to zero.”

    This is wrong, and in a very important way. If you look at monthly data, almost the entire crash in NGDP occurred in 2008, but the Fed didn’t reach the zero bound until mid-December. That means during almost the entire crash we can assume the Fed had monetary policy/AD/NGDP exactly where they wanted it. And that was the problem. They were driving the car by looking in the rear view mirror. I was almost pulling my hair out during this period.

    The Fed cannot perfectly stabilize NGDP, but they certainly can perfectly stabilize NGDP expectations. Had they done so, the financial crisis would have been very mild. Asset prices are forward looking.

    The rest of your comment misses the point. You talk as if I’m suggesting asset purchases are a good way of addressing a collapse in NGDP. Clearly they are not. They are an admission of failure. The choice is not between fiscal stimulus and QE (although even there QE is far better, far less wasteful) the real choice is between fiscal stimulus and a sensible monetary policy.

  45. Gravatar of Matt Waters Matt Waters
    18. December 2014 at 23:36

    What is a sensible monetary policy then? I take it you mean the Fed buying and selling NGDP futures until the future market meets the target.

    My issue is not with such a policy. Talking about level targeting, communication and expectations just makes your argument…unclear. For the reasons I outlined, I can’t see a model where level targeting works alone without the mechanics solved. Those mechanics could include asset purchases, interest on excess reserves or participating directly in an NGDP futures market.

    For the 2008 monetary policy, the monetary policy stance of the Fed was ultimately bimodal. You had a post saying that Australia had “far better monetary policy” than us or Europe. I don’t think Australia’s monetary policy was that different. Their economy’s natural rate just didn’t go below zero, just like our natural rate didn’t go below zero in 2001.

    The disagreement comes down to whether NGDP IS monetary policy, or if monetary policy is a range of different actions, given what happens in the economy outside of the Fed’s control. Those actions can include communication, such as rate forecasts. Given different conditions in the economy, the same exact governors on the FOMC can give vastly different NGDP outcomes. In 2008, based on 2001-03 rate movements, every informed market participant could have reasonably expected the Fed to move to zero rates relatively quickly if the market was correct in its dire NGDP forecast. Moving to and staying at zero was far better than what the ECB did, but it wasn’t enough to overcome the market agreeing on an equilibrium with a lower NGDP.

    So, if the mechanics are the issue with creating NGDP growth, then the technical issues with mechanics would need to be worked out. Hopefully the NGDP market will ultimately be the solution, but saying the Fed only needs to give a level target without participating in the market just doesn’t seem true.

  46. Gravatar of dtoh dtoh
    19. December 2014 at 00:38


    I agree. I’m not claiming that the money supply (no matter how you define it) is a good indicator of the stance of monetary policy. We both know that it is not, but any monetary policy relies on a transmission mechanism and even if you subscribe to the HPE theory, there is no transmission if increases in the MB are matched by increases in ER. YOUR policy prescriptions even if perfectly followed by the Fed would have zero effect if Fed action consisted only of exchanging ER for Treasuries or other assets. To be credible a policy must have both a target and a tool, and if your policy recommendations rely on an MB definition of the money supply and ignore ER, then the tool (and the policy) will not be credible. (And…. as you always say expectations have to be about something.)

  47. Gravatar of James in London James in London
    19. December 2014 at 03:50

    Matt, Matt, Matt
    You are almost there. If NGDP forecasts are on a nice steady growth path, then the policy is right. MM’ers state that the Fed, mostly, just needs to state the policy clearly for us to get there. This is something they have not done. Sometimes the Fed have come close, for a while, in an implicit way, used QE, Forward Guidance, etc etc. It is only 25% good enough. It is better than not doing it, but it could be waaaay better.

    If they announced tomorrow that they had discussed NGDP Forecast Targeting at an FOMC meeting or at Jackson Hole, bond yields would rise straight away, implied NGDP Futures would rise. Something like that happened when it was announced that Michael Woodford was at Jackson Hole (remember Scott’s blog that day). And, again, when Mark Carney raised the idea in Speech in December 2012 after he’d been appointed BoE Governor but before he arrrived. Expectations are that powerful. the “market” is ready for it.

  48. Gravatar of Brian Donohue Brian Donohue
    19. December 2014 at 06:19


    The Federal government deficit this year is $480 billion. Something like $10 trillion in new federal debt since the crisis.

    I remember when ‘fiscal stimulus’ and ‘deficit spending’ were not entirely unrelated concepts.

  49. Gravatar of Don Geddis Don Geddis
    19. December 2014 at 08:49

    @Matt Waters: “What is a sensible monetary policy then?

    It’s setting a clear, stable, target, e.g. NGDPLT @ 5% annually.

    I can’t see a model where level targeting works alone without the mechanics solved. Those mechanics could include asset purchases

    Of course. Expectations do 99% of the work … but they don’t do 100%. You seem to be a person of the concrete steppes. Almost all the action is in communicating the clear target. But sure, you need SOME kind of backup threat as well. Chuck Norris actually does need to be able to beat up individual people. For a central bank, the threat is the expansion of the money supply, via the very standard Open Market Operations, aka asset purchases of Treasuries.

    There are no concerns about zero rates, or liquidity traps, or unconventional policy, or concrete steps. Just state a clear target, and make the threat of unlimited (and perhaps exponentially growing) OMOs, until the target is reached.

  50. Gravatar of ssumner ssumner
    19. December 2014 at 14:16

    Matt, You are thinking about things in the wrong way. Your posts make no sense unless you believe that concrete action X has Y effect, regardless of whether the NGDP target is 5% or 2%. You talk as if the target path doesn’t matter, only the concrete actions matter. But it’s just the opposite, monetary policy is 98% target path and 2% concrete actions. Australia’s monetary policy was radically different from ours, because they had a much higher target path for NGDP. That’s why they did better, without any QE at all.

    dtoh, You said;

    “YOUR policy prescriptions even if perfectly followed by the Fed would have zero effect if Fed action consisted only of exchanging ER for Treasuries or other assets.”

    That would be good enough, if they were following my policy. Because if ER was all that changed, that means that’s all the public wanted to hold when NGDP expectations were on target. It’s the market that determines where the money ends up.

  51. Gravatar of TravisV TravisV
    19. December 2014 at 15:24

    Krugman today:

    “Now suppose that we’re in a New Keynesian world in which prices are temporarily sticky; so P is given. And suppose we’re at the zero lower bound, so r=0. Then there’s only one moving part here: the expected future price level. Anything you do “” monetary or fiscal “” affects current consumption to the extent, and only to the extent, that it moves the expected future price level. Full stop, end of story.

    An immediate implication is that the current money supply doesn’t matter. The future money supply matters, because it can affect the future price level, so a permanent increase in M can affect the economy “” but that effect works entirely through expectations. What you do now matters only to the extent that people take it as an indication of what you will do in the future.”

  52. Gravatar of Matt Waters Matt Waters
    19. December 2014 at 15:50

    Total debt is not the number that matters. The debt held by the public didn’t go up quite as much as $10 trillion, but did go up a lot. Around 40 percentage points of GDP or 7 percent a year. WW2 spending was 80 percentage points of GDP in a lesser amount of time. Also, much of the debt held by the public financed spending which was already happening before, whereas WW2 was new spending.

    I can see how, from the tenor of the last paragraph and my comments, that I it seems I agree whole-heartedly with Krugman about monetary policy’s ineffectiveness. Krugman and other Keynesians just have not been able to explain empirically how NGDP has continued to expand despite government austerity. Ezra Klein posited that it was because state governments compensated for federal austerity, but state governments are essentially big private firms. Their boat goes up and down with all the other boats. They don’t affect the level of the water like either the Fed or the fiscal budget.

    But all that said, yes I am certainly a card-carrying member of the people of the concrete steppes. At the end of the day, I want A leads to B leads to C. If you look at every dollar of NGDP, that is a dollar that somebody out there spent for a reason. That is the “concrete steppes” for me. Not the Fed mechanisms for monetary policy per se, but the concrete actions by real market participants. Those actions can include the effects of guidance or communication from the Fed.

    I could put it this way: how would a pure zero multiplier of deficit spending and efficient markets describe how the market reacted to TARP? That is a fiscal action after all. So if the multiplier is zero, then NGDP would have been unchanged which means the market was irrational.

    To sum up my views, I see a laser-focus from Krugman on interest rates and fiscal deficits while I see a laser-focus from MM on the central bank’s target and policy. I’m trying to piece together all the raw observations and figure out what could explain themselves. From what I can tell, some is pure mechanics, some is rational expectations and some is pure irrational animal spirits.

  53. Gravatar of dtoh dtoh
    20. December 2014 at 17:11

    You said, “That would be good enough, if they were following my policy. Because if ER was all that changed, that means that’s all the public wanted to hold when NGDP expectations were on target. It’s the market that determines where the money ends up.”

    I think you’re skirting the question. Expectations have to be about something and that means at some point the Fed has to have a tool which they can use which causes the non-banking sector to hold more (or less) money.

    The primary non-expectational tool that the Fed uses regardless of what target they choose (rates, inflation, unemployment, or NGDP) is asset purchases/sales. If the market and the economics profession is in the habit of gauging the level and efficacy of the Fed’s use of their policy tool by changes in MB alone (without regard to changes in ER), then they will draw the wrong conclusions.

    (And… when the Fed is setting the IOR rate, it’s the Fed not the public that determines where the money ends up.)

  54. Gravatar of ssumner ssumner
    21. December 2014 at 11:40

    Matt, My views on fiscal are more nuanced than you might assume. For instance, cuts in MTRs help for supply-side reasons. Employer-side payroll tax cuts (or VAT cuts) can help if the central bank is targeting inflation. TARP may have helped for non-fiscal reasons, stabilizing the banking system. TARP did not add to the national debt in the long run, whereas traditional fiscal stimulus does. That’s my objection to traditional demand-side stimulus.

    Under NGDPLT I’d oppose TARP, under the actual regime I’m sort of agnostic.

    dtoh, The profession should certainly not be looking at the MB to gauge Fed policy. In 2007 that was universally viewed as idiotic. Not sure how and why it changed.

  55. Gravatar of dtoh dtoh
    21. December 2014 at 16:00

    I didn’t say the profession should be looking at MB to gauge Fed “policy.” Policy consists of both a target (or targets) and a tool (or tools). I said the Fed’s primary “non-expectational” tool was asset purchases/sales and that the profession was gauging the efficacy of that policy “tool” based on changes in MB. (How many times have we heard arguments along the lines of: “The CB has doubled the money supply without effect.” Just look at the PK quote you provide at the start of this post and the one at the start of your most current post.)

    1. I agree that you need to look at the target to judge efficacy of the overall policy.

    2. I agree that with a credible policy, expectations will do almost all of the work.

    3. However, expectations have to be about something, and credibility depends on having an effective non-expectational tool.

    If you were to poll the profession and ask on a scale of 1 to 10 how effective have Fed asset purchases (e.g. QE) been at raising AD, what do you think the response would be.

    Alternatively, if you were ask the profession to quantify how much action the Fed has taken, what percentage of the respondents would cite changes in MB as a measure of Fed action?

  56. Gravatar of ssumner ssumner
    22. December 2014 at 06:55

    dtoh, The MB is a policy tool. MB-ER is a possible policy target. There is a big difference. You cannot simply substitute one for the other.

  57. Gravatar of dtoh dtoh
    22. December 2014 at 07:14

    Let’s go back to the beginning. If the Fed increases MB and ER by an equal amount is there in your view an HPE transmission mechanism by which this translates into higher AD?

Leave a Reply