QE works in practice because QE works in theory

Well at least we are seeing progress.  The highly visible market response to rumors of QE tapering has led many economists to begin wondering why QE has such a strong impact. Ditto for the powerful reaction to monetary stimulus in Japan. There’s a sort of “it works in practice but not in theory” meme circulating.  I find this really frustrating, but I guess I should be happy that we no longer have to worry about the far more absurd notion that QE doesn’t work at all.

It’s long been known that temporary currency injections are not effective at the zero bound.  Less well known is that fact that temporary currency injections are not very effective when not at the zero bound. And that has led to some confusion about the applicability of the “Quantity Theory of Money” to the zero bound situation.  People might say “Yes, doubling the money supply will normally have a big impact on the price level, but not at the zero bound.”  I’ve probably said similar things.  But that’s actually slightly misleading; the real problem is the temporary nature of the injections, not the zero bound.  Consider:

1.  Interest rates are 7.8% on T-bills.  The Fed suddenly doubles the monetary base, and simultaneously announces the money will be withdrawn from circulation two weeks later.

2.  Tomorrow Janet Yellen announces that the monetary base will be doubled, IOR will be eliminated, and the Fed will maintain the enlarged base even after it exits the liquidity trap in a few years.

In case one the price level doesn’t change.  In case two the price level doubles (or more if you include the QE already done.)

And yet in case one the action was done while rates were positive (although they’d immediately go to zero and stay there for two weeks.)  In case two the action was done when rates were zero, but the impact depends crucially on the base being expected to stay enlarged after rates rise.

It’s clear from these two examples that the real issue is not zero rates; it’s the difference between temporary and permanent currency injections.

Miles Kimball has a new post on this topic, which cites Richard Serlin citing Brad DeLong citing Ben Bernanke:

Brad DeLong: Richard points to this from Brad DeLong as some of the best intuition for Wallace Neutrality that he had found up to that point:

Long ago, Bernanke (2000) argued that monetary policy retains enormous power to boost production, demand, and employment even at the zero nominal lower bound to interest rates:

The general argument that the monetary authorities can increase aggregate demand and prices, even if the nominal interest rate is zero, is as follows: Money, unlike other forms of government debt, pays zero interest and has infinite maturity. The monetary authorities can issue as much money as they like. Hence, if the price level were truly independent of money issuance, then the monetary authorities could use the money they create to acquire indefinite quantities of goods and assets. This is manifestly impossible in equilibrium. Therefore money issuance must ultimately raise the price level, even if nominal interest rates are bounded at zero. This is an elementary argument, but, as we will see, it is quite corrosive of claims of monetary impotence…

His argument, however, seems subject to a powerful critique: The central bank expandeth the money stock, the central bank taketh away the money stock, blessed be the name of the central bank. In order for monetary policy to be effective at the zero nominal lower bound, expectations must be that the increases in the money stock via quantitative easing undertaken will not be unwound in the future after the economy exits from its liquidity trap. If expectations are that they will be unwound, then there is potentially money to be made by taking the other side of the transaction: sell bonds to the central bank now when their prices are high, hold onto the cash until the economy exits from the liquidity trap, and then buy the bonds back from the central bank in the future when it is trying to unwind its quantitative easing policies. A Modigliani Miller-like result applies.

This argument seems to impress a lot of people who are much smarter than me, but the sad truth is that it’s wrong.  They only think it’s right because they aren’t paying close enough attention to the real world. To a casual observer it looks like there are central banks that are trying to inflate but failing, due to credibility problems.  In fact, in all of world history no fiat money central bank has ever tried to inflate and failed. Elite economists glance at the WSJ or NYT, see something about the BOJ doing QE and having trouble getting out of deflation, and murmur to themselves; “someone really needs to model that.”

In the 1980s we had lots of popular “credibility” papers explaining why central banks had so much trouble controlling inflation.  So the “credibility” explanation also seemed like a logical solution to the Japanese liquidity trap. There is just one problem. We now know those credibility papers are wrong.  It’s not difficult for central banks control inflation. It’s easy. And it’s not difficult for central banks to create inflation either. Elite economists were simply not paying attention to the concrete steps for the Bank of Japan took to tighten monetary policy despite being far short of their (ostensible) policy goals.  Or the concrete steps that the European Central Bank took in 2011 to tighten monetary policy. Or the concrete step the Fed would have taken last month if not for concern about the government shutdown.

People who follow central banks closely know that the slow pace of nominal GDP growth does not reflect policy impotence but rather policy hawkishness. Watch what they do, not what they say.

So QE works for very simple reasons. Permanent monetary injections are effective even at the zero bound. QE programs are a signal that central banks would prefer at least slightly faster nominal GDP growth. Slightly faster nominal GDP growth requires that at least a small portion of the currency injection be permanent. So by signaling a preference for slightly faster nominal GDP growth, central banks are implicitly signaling a preference to have at least a small portion of the QE program be permanent (for any given IOR rate). Markets believe the central banks (and why shouldn’t they?) And hence asset prices react to the QE program.

No new theory is needed. Permanent monetary injections are inflationary, and always have been. Contrary to elite opinion, central banks have no difficulty convincing markets that they wish to boost nominal GDP growth, when they actually do wish to boost nominal GDP growth. Rather the problem is internal. The problem is that people like Janet Yellen have difficulty convincing her colleagues within the central bank that faster nominal GDP growth is desirable.

I think these issues would be much easier to understand economists would look at monetary policy in the reverse direction. Stop thinking about QE as some sort of lever to move nominal GDP in the desired direction. Instead think about where you central banks want to end up, and how much base money the public demands when nominal GDP growth expectations are on target. This can be done most effectively by the thought experiment of a nominal GDP futures market. Have the central bank peg the price of nominal GDP futures at a level equal to the policy target, and the actual level of base money consistent with that equilibrium becomes endogenous. It might be either higher or lower than the demand for base money when the economy falls short of its nominal GDP target. In other words it’s not even obvious whether positive or negative QE  is needed to hit the target. Start with the target and then work back to the size of the monetary base.

When policy is viewed in this framework all worries about “Wallace Neutrality” vanish into thin air. Instead the real problem is highlighted—if there is a real problem. That is, does the central bank need to buy “unconventional assets” in order to meet the demand for base money when expected nominal GDP growth is on target? In most cases the answer is no.

PS.  I highly recommend this post by Ryan Avent and this post by Marcus Nunes.

 


Tags:

 
 
 

61 Responses to “QE works in practice because QE works in theory”

  1. Gravatar of Michael Michael
    11. October 2013 at 08:48

    “People who follow central banks closely know that the slow pace of nominal GDP growth does not reflect policy impotence but rather policy hawkishness. Watch what they do, not what they say.”

    The problem, then, is that they are trying to thread a fine gauge needle. More inflation/NGDP growth RIGHT NOW, but still a 2% inflation ceiling now and forever.

  2. Gravatar of youko youko
    11. October 2013 at 08:58

    I have a stupid question: I was talking to one of my buddies last night trying to explain the basics of this post to him. He asked me a question that stumped me though: why would a bank be both willing to sell assets for cash one day, and then willing to buy assets for cash the next? (Or the reverse?). Especially if they know the initialtransaction is going to he unwound the next day? It seems like they should only be willing to take one side of that deal at a time.

    I sort of waved my hand and alluded to a portfolio rebalancing effect, so that the bank that makes the initial transaction offloads its new assets/cash onto someone else, so its not necessarily the same banks buying back their own stuff… But that answer was really unsatisfying. Am I missing something obvious?

  3. Gravatar of Saturos Saturos
    11. October 2013 at 09:25

    Michael, not so fine, since it is understood that when you target x% inflation you go over a little bit in some periods, which will be counterbalanced by going under the same amount in other periods, like our RBA does. The expectation of your forecast should hit your target – whereas the Fed consistently expects its mean performance to be beneath their own target on inflation. Hence they need to invoke third mandates, etc. to explain their incompetence. And now that the media is cottoning on to said incompetence, they are rationalizing their own broken expectations of technocratic efficiency from the Fed (it had to be the greedy private bankers who broke the economy) by proffering such excuses as well.

  4. Gravatar of Tom Stringham Tom Stringham
    11. October 2013 at 09:32

    I’m not an economist, so posts like this really help clear things up for me. I’ve been trying to wrap my head around the zero-bound question for a while.

    Tell me if I understand this right: does the idea that monetary policy fails at the zero bound really rest on the assumption that no one will want to lend at near-zero interest? And if that’s the case, isn’t the zero-bound an empirical question that depends on lenders’ deflation expectations over the period of the debt anyway? So isn’t it the case that if we expect deflation, real interest rates can still be higher than near-zero? (And if we don’t expect it, we win because we have inflation.)

    But ignoring the markets, I don’t see how this can possibly be a problem if the Fed can simply do what Keynesians wouldn’t predict a rational lender to do, which is to buy bonds regardless (and in spite of) whatever the interest rates are. And if they are buying from the US Treasury, as opposed to banks, then you can be sure the money won’t be hoarded by banks, or whatever the mechanism of the zero-bound process is.

    So doesn’t the fact that there is a central bank that can buy bonds very simply circumvent the whole intuition of the zero bound?

  5. Gravatar of dtoh dtoh
    11. October 2013 at 09:45

    Clearly, there is no Wallace Neutrality in the short term. If there was, a demand shock (drop in financial asset prices) would immediately result in a drop in wages. And…Japanese monetary policy would not have had any effect.

    The fact that Wallace Neutrality does not exist in the short term (wages are sticky but financial asset prices are not sticky) is the very reason that monetary policy is both effective and necessary.

  6. Gravatar of Alexander Alexander
    11. October 2013 at 09:54

    Scott: I am probably just slow, but I’m not sure I understand your objection to the DeLong quote.

    You presented two cases to illustrate how a permanent increase in the money supply leads to inflation while a transitory increase does not.

    DeLong wrote, “In order for monetary policy to be effective … expectations must be that the increases in the money stock … will not be unwound in the future after the economy exits from its liquidity trap.”

    You said that this argument is wrong. But isn’t DeLong’s point the same as the point you made earlier?

    In any event, Wallace Neutrality is an assumption about fiscal policy. Wallace showed that given an equilibrium real allocation and price level sequence, there are multiple configurations of government liabilities, taxes and government asset holdings such that are all consistent with the same equilibrium allocation and price level sequence. In models that feature Wallace Neutrality, the modeler has assumed that fiscal policy will always adjust to exactly offset the effects of government asset purchases. It’s not some deep property of financial markets.

  7. Gravatar of jknarr jknarr
    11. October 2013 at 10:08

    Scott, do you think that where the monetary base is held matters? One half is outside the US now.

    Half of currency is abroad, and half of reserves are now held at US branches of foreign banks (which do essentially no lending in the US). US prices compete with prices abroad, clearly, so a US guy with a $100 bill now has to compete with someone with a $100 bill in southeast Asia for the similar $100 product, but could the monetary transmission be different?

    I expect not, as the USD is a closed system, but the pallets of $100 bills going abroad appear to have less to do with US activity than generating imported inflation, which might be a different beast.

  8. Gravatar of Steve Steve
    11. October 2013 at 10:30

    Quantitative Easing Derangement Syndrome

    QED.

  9. Gravatar of Ralph Musgrave Ralph Musgrave
    11. October 2013 at 10:37

    Scott says the price level would double after Janet Yellan doubles the base. Why would it? When Yellan prints $Xbn and takes $Xbn of Treasuries off the private sector, the private sector is little better off. Moreover cash and Treasuries are similar in nature (particularly Treasuries near maturity). So the total value of private sector financial assets remains much the same, plus the NATURE OF those assets is not HUGELY altered. So on that basis I’d predict that the private sector would not go on on inflation boosting spending spree. And that’s exactly what happened in 2008 when the base was doubled: there was little effect. Certainly not much effect on inflation.

  10. Gravatar of Adam Adam
    11. October 2013 at 10:57

    Thanks for this. Miles’s post was long, detailed and interesting, but frustrating because the differences between the assumptions embedded in the MM result seemed so obviously inapplicable to the real world. I spent a few minutes trying to think of how to concisely state that, but you (no surprise) have done better than I.

    The good news is that after reading Miles I nope have an idea what Steve Williamson has been on about, but still think the MM assertion is self-evidently silly. The ability of the Fed to confound expectations alone should be enough to reject strong assertions of QE’s impotence.

  11. Gravatar of John John
    11. October 2013 at 11:04

    Scott,

    You said, “although they’d [interest rates] immediately go to zero and stay there for two weeks.” Wouldn’t interest rates not go to zero for the same reason, expectations, that the price level wouldn’t double? Where’s the incentive to make a loan that yields 0% interest when you can wait a week or two and get a yield?

  12. Gravatar of Mark A. Sadowski Mark A. Sadowski
    11. October 2013 at 11:06

    “Elite economists were simply not paying attention to the concrete steps for the Bank of Japan took to tighten monetary policy despite being far short of their (ostensible) policy goals. Or the concrete steps that the European Central Bank took in 2011 to tighten monetary policy.”

    1. BOJ

    A) The BOJ reduced the monetary base by 24.4% from January to November 2006 and economic weakness followed within months. Japan was one of the first major economies to have negative RGDP growth when it fell in 2007Q3. RGDP fell 4.7% at an annual rate in 2008Q2, and 4.0% at an annual rate in 2008Q3, causing Japan to suffer serious consecutive quarterly declines in RGDP before the U.S. did the same. RGDP proceeded to fall 12.4% at an annual rate in 2008Q4 and 15.1% at an annual rate in 2009Q1. All told RGDP fell 9.2% from peak to trough.

    In short Japanese RGDP fell sooner, faster and further than every other major country at the beginning of this recession. It’s hard not to connect the dots between this result and the BOJ’s sudden and sharp withdrawal of QE, given one literally followed upon the other within months.

    B) The only time between 2008 and April 2013 it was evident to people watching the BOJ’s monetary base that QE was actually taking place in Japan (despite their highly visible “open mouth operations” since the Great Recession), was between February and April 2011 when the BOJ’s monetary base expanded by 20.9 trillion yen, or by 20.7%, largely in response to the Tohoku earthquake and tsunami. But the monetary base was promptly shrunk by 8.4 trillion yen over the course of the next two months. It wasn’t until July 2012 that the monetary base set a new record.

    2. ECB

    The ECB raised the MRO rate in July 2008 and again in April and July 2011, and in each case financial crisis followed soon thereafter. The second financial crisis started with Portuguese acting Prime Minister Socrates’ bailout request, was marked by the peaks in the government bond spreads for all of the GIIPS with respect to Germany, and only concluded (paused?) after Spanish Prime Minister Rajoy requested a bailout in July 2012.

    P.S. Due to reductions in the ECB’s “enhanced credit support programmes” (the ECB has never done QE and has no plans to do any) the ECB’s monetary base shrank by 28.4% between July 2012 and August 2013. But there’s been nary a peep about this by elite economists, has there?

  13. Gravatar of ssumner ssumner
    11. October 2013 at 11:07

    Michael, You said;

    “The problem, then, is that they are trying to thread a fine gauge needle. More inflation/NGDP growth RIGHT NOW, but still a 2% inflation ceiling now and forever.”

    No, they believe inflation will stay below target for as far out as they can forecast. They’ve simply chosen to ignore their mandate. They aren’t even close to the eye of the needle.

    youko, Presumably the bank would expect to make a small profit on the exchange.

    Tom, No, those who worry about the zero bound feel that money injected in any form would simply end up in the banks as excess reserves. Even if spent on Treasuries from the government.

    Alexander, This is the part of DeLong’s statement that was wrong:

    “His argument, however, seems subject to a powerful critique:”

    DeLong’s hypothetical is not a powerful critique to Bernanke’s claim that monetary policy is always effective. It’s an extremely weak critique.

    Yes, I understand the assumptions of WN, but they don’t describe how monetary and fiscal policy operate in the real world. That’s my point. Theories need to use assumptions consistent with the real world.

    jknarr, No, it doesn’t matter where the currency is held. It just matters what the total demand is.

    Ralph, I’m simply saying money is neutral in the long run, which is not controversial, at least when interest rates are positive. So if interest rates are expected to be positive in the future, then the money will be expected to lead to much higher prices in the future. And then intertemporal commodity price arbitrage will cause prices to rise today. When rates are positive cash and bonds aren’t even close to being similar assets.

    The 2008 currency increases were temporary, according to the Fed. So there was no reason for prices to rise significantly.

  14. Gravatar of ssumner ssumner
    11. October 2013 at 11:13

    Adam, At first I thought when you said MM you meant market monetarism.

    John, It would be hard to get the public to hold an extra trillion in base money for two weeks, at any interest rate above zero. Imagine trying to get a positive rate in the fed funds market—all the banks would say “we’re swimming in cash already.”

    Mark, Very good points.

  15. Gravatar of JP Koning JP Koning
    11. October 2013 at 11:38

    “In case two the action was done when rates were zero, but the impact depends crucially on the base being expected to stay enlarged after rates rise. It’s clear from these two examples that the real issue is not zero rates; it’s the difference between temporary and permanent currency injections.”

    If interest rates are at zero across all maturities — say markets already expect the fed funds rate to be at 0% in 2017, 2021, 2025 etc — then even a permanent currency injection will have little effect. I don’t think that’s a controversial statement. Sure, it’s unlikely that the curve would actually lie at 0% for all maturities, but it serves as a boundary case. What do you think?

    “Tomorrow Janet Yellen announces that the monetary base will be doubled, IOR will be eliminated, and the Fed will maintain the enlarged base even after it exits the liquidity trap in a few years.”

    How does your thinking change if Yellen does all those things but doesn’t eliminate IOR?

  16. Gravatar of TallDave TallDave
    11. October 2013 at 12:25

    Yep, expectations uber alles.

  17. Gravatar of TallDave TallDave
    11. October 2013 at 12:28

    Ironic, isn’t it, that Yellen may save Obama’s Presidency despite being his consolation pick?

  18. Gravatar of Lorenzo from Oz Lorenzo from Oz
    11. October 2013 at 12:49

    On the Ryan Avent post, two quotes come to mind:
    Kurt Lewin — “there is nothing so practical as a good theory”
    Winston Churchill — “the further back you look, the further forward you see”

    In the C19th England was on the gold standard. In the C19th, England experienced dramatic episodes of asset price instability (“bubbles”). Does anyone think that the Bank of England’s monetary policy caused the asset price instability? Or was it the combination of technological change and “savings pushing on investment”, as Andy Harless would put it? Relatively high demand for assets when it was literally uncertain what the technological payoffs will be, so expectations were very “fragile” (i.e. susceptible to new information changing framings).

    A stable macroeconomic environment with steady positive growth expectations likely does push up the demand for assets, but why is that a bad thing?

    The Fed Governors do not seem to have a useful understanding of what they are there to do. Perhaps that is the difference between them and the RBA Board–which does have a clear understanding of what they are there to do. Australia also has a separate prudential regulation authority, APRA, so that probably helps the RBA focus.
    http://www.apra.gov.au/Pages/default.aspx

  19. Gravatar of Lorenzo from Oz Lorenzo from Oz
    11. October 2013 at 12:51

    BTW, if anyone ever raises tulipmania, just point them to this delicious article by Earl Thompson:
    http://www.econ.ucla.edu/thompson/Document97.pdf

  20. Gravatar of dtoh dtoh
    11. October 2013 at 13:22

    For QE (monetary stimulus) to work, it is both necessary and sufficient for it to cause a marginal increase in the exchange of financial assets for real goods and services by the private non-banking sector. If it simply causes an increase in ER, it has no effect on AD.

    Everything else including, whether an injection is temporary or permanent and the impact of the injection on the future expected path of interest rates, is irrelevant.

  21. Gravatar of Saturos Saturos
    11. October 2013 at 14:22

    The Problem With Forward Guidance: http://www.project-syndicate.org/commentary/the-problem-with-central-banks–communication-strategies-by-marcel-fratzscher

  22. Gravatar of Lawrence D’Anna Lawrence D'Anna
    11. October 2013 at 15:05

    I have some questions about the details of “Have the central bank peg the price of nominal GDP futures at a level equal to the policy target”.

    If I understand correctly, a NGDP future is a thingy that pays a set fraction of the nominal GDP of a particular year, at some point in the future, after the actual NGDP that actually happened has been measured. Presumably the fed pegs the price of NGDP futures by writing new ones if the price is too high, and buying existing ones if the price is too low.

    So my questions are:

    What happens if the fed buys all the NGDP futures? Escalating QE until somebody somebody who isn’t the fed wants to own a NGDP future?

    Doesn’t this mean that the NGDP futures could end up being a large fraction of the fed’s liabilities? Is this a problem?

    Doesn’t it also mean that privately-written NGDP futures could end up being a large fraction of the fed’s assets? What happens if they don’t pay up? Is this a problem?

    Once expectations take hold that NGDPLT is able to hit its targets, don’t treasuries become close substitutes for NGDP futures? Is it a problem if congress runs around writing zillions of almost-NGDP-futures while the fed is trying to control their price?

    What happens if NGDP futures for one year are on-target but NGDP futures for another year are off? How can the monetary policy of today effect tomorrow’s NGDP without also affecting next week?

    How does the fed know how to price a NGDP future? Don’t they have to choose a discount rate to know what money paid when the future pays off should be worth?

  23. Gravatar of Gordon Gordon
    11. October 2013 at 15:35

    Scott, would you say that those who dismiss QE as being ineffective are overlooking Christina Romer’s paper that central banks are far more timid than aggressive or do they disagree with her analysis? As a non-economist, I go back to my Simpson’s party analogy with Marge in the role of the Federal Reserve to understand the past credibility gap. Marge has a past history with the party goers as being so fearful of drunken guests that when she started pouring the drinks, they considered it a good possibility that she’d stop pouring and put all the bottles of alcohol away too soon. (Homer in the role of the U.S. government has the opposite credibility problem – poor self-control and a desire to please his friends.) Now that we’re well into QE3, there is finally some credibility that Marge will keep going until it’s clear that the party will be self-sustaining.

  24. Gravatar of maynardGkeynes maynardGkeynes
    11. October 2013 at 15:43

    Given the distributional effects, using unwound inflation to benefit asset holders at the expense bond holders is quite troubling to me. I don’t believe the Fisher effect applies here , because it seems pretty clear that the basis of QE is unexpected inflation.

  25. Gravatar of TravisV TravisV
    11. October 2013 at 16:59

    Everyone,

    Y’all should check out Lorenzo From Oz’s commentary at Skepticlawyer.com:

    http://skepticlawyer.com.au/author/lorenzo

    Awesome stuff!!!

  26. Gravatar of Andrew_M_Garland Andrew_M_Garland
    11. October 2013 at 21:50

    Do I understand correctly? Permanent “injections” of money increase inflation. Inflation increases nominal GDP.

    I thought we are interested in increased real production, that is real GDP. What good is nominal GDP? What if there is no increased real production? I thought inflation upsets markets and delivers value to those at the beginning of the inflation chain. There are trickle-up benefits as the new money delivers real value to the first spenders (the government and its contractors), and steals value from the most vulnerable savers (cash savings accounts of the old).

    “Injections” of new, permanent, inflationary money are done when the Fed buys government bonds (buying directly or indirectly). The government spends the money (yay!), and the public pays a silent tax in the decreased value of its cash holdings (boo!)

    We certainly have real inflation. The real value of all cash balances and financial instruments payable in cash (bonds) decreases. Why, exactly, does real production magically increase? And, does real production increase by more than the implicit tax on cash and financial paper?

    What are the measured real benefits and costs? I see why the government and its contractors like this “money injection” policy. Why would anyone else like this policy?

    EasyOpinions

  27. Gravatar of Benjamin Cole Benjamin Cole
    11. October 2013 at 23:09

    So…a central bank prints (digitizes) a lot of money, and buys a lot of assets…the new money enters circulation…and stimulated growth until capacity is reached, after which it tends to be inflationary.

    Is this really debatable?

  28. Gravatar of Benjamin Cole Benjamin Cole
    11. October 2013 at 23:45

    Open question for anybody:

    It seems taken for granted that QE resulted in an enlargement of bank reserves.

    I have pondered this, as a “rational man” would not sell government bonds just to bank the cash. The bond holder is already liquid and has no risk, and even earns a modest yield. So why sell a government bond just to bank the money, and lose the interest yield?

    Could it be that swelling bank reserves reflect something else? Flight capital? People leaving the stock market (in the early days after 2008)? People banking money until real estate stabilized? Banks sitting on assets as they earn 0.25 percent?

    Due to the rational man theory, I do not see many selling government bonds just to bank the money…that is not a compelling story line…

  29. Gravatar of Mike Mike
    11. October 2013 at 23:57

    ssumner

    “The highly visible market response to rumors of QE tapering has led many economists to begin wondering why QE has such a strong impact.”

    Strong impact on asset prices, but a weak impact on inflation and unemployment.

    “There’s a sort of “it works in practice but not in theory” meme circulating. ”

    Surely unemployment is too high and inflation is too low. Maybe the term “monetary policy” should be redefined to “asset price policy”

  30. Gravatar of Mike Mike
    12. October 2013 at 00:14

    “It’s not difficult for central banks control inflation. It’s easy. And it’s not difficult for central banks to create inflation either.”

    Central banks don’t control inflation they influence it and can set an upper bound to it. They target the cost of fed funds (and required reserves) which influences banks funding costs which affect their lending and investing prices and quantities. Therefore the banks control inflation as much as the central bank because the deposits they create affect prices the same way currency does. The central bank doesnt control or dictate the amount if deposit growth. Also deposits outnumber currency so banks probably influence inflation more than the central bank.

    So I’m guessing you think the fed would attain its UE and inflation targets if it did higher amounts of QE? Hitting those targets probably necessitates the dow at 25000 or similar. The increase in asset prices has been disproportional to the improvement in the economy would you agree? QE is imbalanced.

    “It’s long been known that temporary currency injections are not effective at the zero bound. Less well known is that fact that temporary currency injections are not very effective when not at the zero bound.”

    How does the CB inject currency though? Its out of the CB’s hands. All the CB can do is create reserves. The lending markets have to be expanding deposits for the demand for currency to pick up (assuming stable C/D ratio). The CB cant just expand currency if deposits dont grow. Banks and people will demand a certain proportion of the broader money as currency. But deposits need to expand, therefore the banking system determines the amount of currency through demand, currency is not controlled through supply. The fed supplies reserves which may be converted into currency if demanded.

  31. Gravatar of dtoh dtoh
    12. October 2013 at 00:43

    Benjamin Cole, you said;

    “Due to the rational man theory, I do not see many selling government bonds just to bank the money…that is not a compelling story line…”

    This is exactly right. The non-banking private sector (drug dealers/tax evaders excepted) holds only the amount of money needed for transactions. The only reason they sell bonds is because they intend to exchange bonds (financial assets) for real goods and services. They money is merely an intermediate instrument which is needed to effect the exchange of financial assets for real goods and services.

    Monetary stimulus works through expectations and because it raises the real price of financial assets causing a marginal increase in the exchange of financial assets for real goods and services.

  32. Gravatar of Philippe Philippe
    12. October 2013 at 02:48

    “The non-banking private sector (drug dealers/tax evaders excepted) holds only the amount of money needed for transactions.”

    What about savings deposits, time deposits, etc?

  33. Gravatar of dtoh dtoh
    12. October 2013 at 04:32

    “What about savings deposits, time deposits, etc?”

    Talking about money used in OMO, i.e. MB.

  34. Gravatar of Philippe Philippe
    12. October 2013 at 04:51

    The central bank doesn’t give non-banks MB when it conducts OMOs. Non-banks receive credits to their bank accounts. The non-banks could withdraw these bank deposits as cash if they wanted to, but they could just as easily transfer the deposits to savings or time deposits which earn interest.

  35. Gravatar of John John
    12. October 2013 at 05:42

    Scott,

    The point I was trying to make is that some people would have to borrow during those two weeks but everyone would have an incentive not too. The same situation applies with the extra money. Some people would go out and spend it even though they know that the money will be yanked away in two weeks. It just seems very inconsistent to say that expectations will mean that people won’t go out and spend the extra money to bid up prices but they will make loans. If you know that the money is going away, you have no incentive to make any loans or spend any money. It seems like you are missing the point that you’ve made so well in the past about the role of expectations.

  36. Gravatar of ssumner ssumner
    12. October 2013 at 06:01

    JP, Yes, in that case bonds would be cash, and an OMO would simple swap cash for cash—no effect.

    As long as the IOR is kept at the current level, my claim would be no different. Obviously in theory there is a IOR increase large enough to negate any monetary injection.

    TallDave, Yes, ironic.

    Lorenzo, Very good comment.

    dtoh, You are saying for QE to work it must work, everything else is irrelevant. But those things are relevant to whether it works in the first place. Changes in financial prices are evidence of it working. They are an EFFECT.

    Lawrence, I have a paper on NGDP futures targeting in the right margin, which answers all your questions.

    Gordon, Yes, I think a lot of people do overlook Romer’s arguments.

    MaynardKeynes, If the Fed uses QE to target NGDP, then it will also minimize unfair redistribution between lenders and borrowers. Of course actual Fed policy (low NGDP) has massively redistributed resources toward government bondholders.

    Andrew, Check out my short course on money in the right margin.

    Ben, They don’t typically sell bonds to hold money, they almost always turn around and immediately invest the money in some other asset, like stocks.

    Mike, Miles Kimball obviously knows that if QE is having a big impact on asset markets it is also having a significant impact on the economy. Lots of people don’t understand that but elite economists do, which is why they are finally realizing that QE works in practice.

    QE is not my preferred tool for monetary policy (I prefer NGDPLT) but if the Fed insists on using QE I favor doing enough to get expected NGDP growth up to 5%. I have no idea where stocks would go in that case, and really don’t care.

    As far as banks controlling inflation, I very much doubt it. Consider two hypotheses:

    1. In the 1980s the nation’s commercial banks got together and decided to gradually bring inflation down to 2%, and then keep it there.

    2. In the 1980s the Fed decided to gradually bring inflation down to 2%, and then keep it there.

    I can’t even imagine anyone claiming theory 1 is more plausible than theory two.

    The Fed creates base money, the banks and public determine how it is split between reserves and cash. During normal times, about 95% of new base money becomes cash with a few days.

  37. Gravatar of ssumner ssumner
    12. October 2013 at 06:03

    John, Longer term loans would still pay interest, it’s merely the overnight loan rate that falls to zero. So waiting two weeks to make a loan would not help at all. Still not sure that answers your question, but it’s the best I can do.

  38. Gravatar of Geoff Geoff
    12. October 2013 at 06:13

    You can garner the intellectual bankruptcy of market monetarism by noticing that its standard of “success” and “goal achievement” is more inflation, and not sustainable higher living standards.

  39. Gravatar of Mike Mike
    12. October 2013 at 06:32

    dtoh

    “Monetary stimulus works through expectations and because it raises the real price of financial assets causing a marginal increase in the exchange of financial assets for real goods and services.”

    The increase in demand for goods and services as a result of increased asset prices is too small. This is because many people don’t have assets or have few assets so there is no increase in demand from them. Also people with many assets demand for goods and services wont change much if their wealth increases.

    Plus the credit channel. Only the people with assets experience improved balance sheets and accessibility to credit.

    Dealing directly with the broad public by expanding money on the other hand directly increases demand for goods and services because people have a higher propensity to spend rather than reallocate portfolio. Directly receiving money means balance sheets improve increasing accessibility to credit. Higher demand for goods and services also means higher employment and higher accessibility to credit because of employment.

  40. Gravatar of Mike Mike
    12. October 2013 at 06:51

    “As far as banks controlling inflation, I very much doubt it. Consider two hypotheses:

    1. In the 1980s the nation’s commercial banks got together and decided to gradually bring inflation down to 2%, and then keep it there.

    2. In the 1980s the Fed decided to gradually bring inflation down to 2%, and then keep it there.

    I can’t even imagine anyone claiming theory 1 is more plausible than theory two.”

    No one “controls” inflation. Both the banks and the CB influence it.

    “The Fed creates base money, the banks and public determine how it is split between reserves and cash. During normal times, about 95% of new base money becomes cash with a few days.”

    Correct and that is determined by how many deposits are created by banks. The more deposits the more currency and reserves needed. As a result of deposits created by banks a demand for required reserves and currency will come about.

    When the fed creates reserves it pushes down the rate on reserves which affects but does not control the rate of deposit creation. The amount of deposits created determine the demand for required reserves and currency. Therefore both the banks and the fed affect inflation but no one controls it. Many factors affect the amount of deposits created by the banking system, the cost of fed funds is one factor.

  41. Gravatar of libertaer libertaer
    12. October 2013 at 08:14

    Mike, you said:

    “Dealing directly with the broad public by expanding money on the other hand directly increases demand for goods and services because people have a higher propensity to spend rather than reallocate portfolio…”

    I’m an advocate for NGDP targeting done through helicopter drops and a VAT for vacuum cleaner operations. (My reasons have nothing to do with the effectiveness of various transmission channels, I’m just against government spending by indebting future taxpayers, “no taxation without representation” and all that. Now, without government bonds -and since you don’t want central banks to buy private assets either- you need a monetary policy without the buying of assets at all. Hence, heli drops and VATs.)

    Now imagine such a monetary policy would be enacted. What would happen? By only announcing helicopter drops NGDP expectations would rise instantly, firms would invest, people expecting higher incomes would start to consume more and stocks would go up… No difference to NGDP targeting done through QE.

    So yes, money should be spend into the economy through helicopter drops, but no, it wouldn’t change anything. Chances are you wouldn’t need any actual heli drops. Monetary policy works always through expectations, the tools you use don’t matter much… only the target counts.

  42. Gravatar of libertaer libertaer
    12. October 2013 at 08:31

    Scott,

    “Ralph, I’m simply saying money is neutral in the long run, which is not controversial, at least when interest rates are positive. So if interest rates are expected to be positive in the future, then the money will be expected to lead to much higher prices in the future. And then intertemporal commodity price arbitrage will cause prices to rise today. When rates are positive cash and bonds aren’t even close to being similar assets.”

    This is another reason why NGDPLT through heli drops and VATs is superior to monetary policy by buying assets. Buying assets is like doing a helidrop and a vacuum cleaner operation at once, giving and taxing away at the same time. This only works if what you take away behaves differently in the long run than what you give. NGDPLT through heli drops would work even if we had negative interest rates forever.

  43. Gravatar of dtoh dtoh
    12. October 2013 at 08:32

    Mike,
    You’re interpreting the definition of financial assets incorrectly and too narrowly and overlooking the fact that you can exchange an asset by going short. The exchange of financial assets includes not just selling what’s in your portfolio, it also includes the issuance of new notes, CP, increased draw down on a line of credit, home improvement loans, increased balance on your credit card, etc.

    Credit is just one type of financial asset. The issue is not the availability of credit, the issue is the price of financial assets generally (including credit). The most basic and irrefutable economic theory tells you that when the real price of financial assets rises relative to the price of goods and services, then you will see an increase in the exchange of financial assets for real goods and services, i.e. an increase in AD.

    Also you talk about “directly receiving money.” This is not the way OMO works. It’s an exchange. If it were a helicopter drop the mechanism would be different, but we don’t have helicopter drops.

  44. Gravatar of JP Koning JP Koning
    12. October 2013 at 09:03

    “Therefore both the banks and the fed affect inflation but no one controls it.”

    Mike, there is nothing the banks can do to affect the price level that the Fed cannot counterbalance. But it is not the case that Fed influence over the price level can be counterbalanced by banks. There is an asymmetry.

    (You sound like someone called lxdr1f7, who I’m talking with at my blog about these same issues)

  45. Gravatar of dtoh dtoh
    12. October 2013 at 09:25

    Scott, you said;

    “You are saying for QE to work it must work, everything else is irrelevant. But those things are relevant to whether it works in the first place. Changes in financial prices are evidence of it working. They are an EFFECT.”

    Your description of my argument would be correct if I had said QE works because it raises AD. What I said is that QE WORKs because of an increased exchange of financial assets for real goods and services.

    I overstated my point by saying the other things are irrelevant, but what I will say is that they are secondary and that they are not deterministic. Expected future OMO, the future path of interest rates, and ER are only relevant to the extent that they amplify or attenuate the change to the real price of financial assets caused by Fed purchases.

    The rise in real financial asset prices is not an effect. It is the primary and direct result of Fed purchases of financial assets and is the mechanism by which monetary policy works.

    I think the reason for the lack of clarity in your descriptions of how QE works is because you think of OMP primarily as the issuance of money rather than as the purchase of financial assets. I believe you do this because of your reliance on the analogy of an exogenous increase in apples (or gold or whatever) working through an HPE mechanism. This is not the way monetary policy works. There are no helicopter drops, and there is no HPE.

    Monetary works because the Fed pushes up the real price of financial assets. You would get exactly the same effect if commercial banks all decided to spontaneously drop interest rates.

  46. Gravatar of Philippe Philippe
    12. October 2013 at 09:30

    “I’m just against government spending by indebting future taxpayers, “no taxation without representation” and all that.”

    Strictly speaking only the government is indebted. They can pay the bill by creating money. If there is too much money creation in the future there could be lots of inflation, which is a slightly different issue to being indebted.

  47. Gravatar of paul Einzig paul Einzig
    12. October 2013 at 11:39

    Excess reserve levels have gone from the 1 billion dollar range in the 1990’s to the 2.2 trillion dollar range today. Is this not proof that in fact QE is ineffective, that the lions share of new liquidity never even sees the light of day?

  48. Gravatar of ssumner ssumner
    12. October 2013 at 16:00

    Saturos, That’s just an awful piece. First of all, he doesn’t seem to understand the Fed’s policy, which calls for raising interest rates if inflation rises above 2.5%. And second, there is no proposed alternative policy target. What should the Fed be targeting, and how should they target it? You can’t beat something with nothing.

  49. Gravatar of ssumner ssumner
    12. October 2013 at 16:07

    Paul, For the millionth time, temporary currency injections have little effect, so the evidence you cite is meaningless.

    Mike, Since the Fed doesn’t control inflation, who causes a 2% inflation target to be achieved?

    Libertaer, Helicopter drops are extremely wasteful as the money must later be vacuumed up with distortionary taxes.

    dtoh, But you haven’t explained what causes asset prices to change. It can’t be purchases, because purchases by themselves don’t cause asset prices to change. Otherwise how could purchases of bonds by the Fed make bond prices fall?

  50. Gravatar of dtoh dtoh
    12. October 2013 at 16:16

    Paul, you wrote;

    “Excess reserve levels have gone from the 1 billion dollar range in the 1990″²s to the 2.2 trillion dollar range today. Is this not proof that in fact QE is ineffective, that the lions share of new liquidity never even sees the light of day?”

    If the amount of increase in ER exactly matches the amount of assets purchased by the Fed in OMP, then OMP will have no effect.

    However, the amount of ER is entirely at the discretion of the Fed. So yes, “QE” as implemented by the Fed has not been fully effective, but that’s simply a result of poor execution by the Fed.

  51. Gravatar of Mike Mike
    12. October 2013 at 18:59

    JP Koning

    “there is nothing the banks can do to affect the price level that the Fed cannot counterbalance.”

    Control seems to imply that its up to their discretion what the rate of inflation will be. Influence is a better description IMO, sometimes the fed has a strong influence on inflation and other times like now it doesn’t. Otherwise the inflation target wouldn’t be called a target it would be called a setting maybe.

    The fed mainly controls the supply of reserves and reserves requirements. It doesnt directly determine the broader money supply, or the demand for money. It does influence demand for money and broader money. It can react to changes in demand for money and expectations in order to reach inflation targets. To say it controls inflation is too far. It would need to be omnipotent.

    It may need to become more and interventionist and adopt more and more extraodinary measures to do so.

    I am lxdr1f7.

    ssumner

    “Mike, Since the Fed doesn’t control inflation, who causes a 2% inflation target to be achieved?”

    Its not being achieved. It may not achieve it either unless it resorts to absurdities such as charging negative rates or buying up large amounts of safe assets and forcing people to take on more risk.

    Even if it achieves its inflation target and brings down UE through QE how much higher would all the asset markets have to be? If the markets need to appreciate so much to reach inflation target markets may become too volatile and gyrate too wildly to changes in economic variables. Therefore the fed may achieve target inflation only in the short term.

    Changing the mechanism so the fed deals directly with the public would make the fed more effective at targeting inflation but “control” is too strong a word IMO.

  52. Gravatar of libertaer libertaer
    13. October 2013 at 02:56

    Philippe, you said:
    “They can pay the bill by creating money. If there is too much money creation in the future there could be lots of inflation, which is a slightly different issue to being indebted.”

    As long as the central bank is independent (submitting to an inflation or NGDP target), they can’t monetize the debt.

    Scott, you said:
    “libertaer, Helicopter drops are extremely wasteful as the money must later be vacuumed up with distortionary taxes.”

    An alternative to taxes as vacuum cleaners would be paying interest on reserves. And a Chuck Norris who is clear about NGDP level targeting and able to do helicopter drops would have to do even less than a Chuck Norris who can only do QE.

    But do you agree with my point that in a world with interest rates forever negative monetary policy by buying government bonds wouldn’t work?

  53. Gravatar of John John
    13. October 2013 at 07:12

    Scott,

    Sorry if I was unclear and thanks for attempting to answer my convoluted question. It just seems to me that private lenders, given that they aren’t coerced, would still demand interest given that everyone knows that the extra liquidity is going to be withdrawn and the path of interest rates hasn’t changed over the long term. Long term expectations affect decisions right now. I thought that was why you were saying that temporary increases in money won’t affect prices in a meaningful way.

  54. Gravatar of Lorenzo from Oz Lorenzo from Oz
    13. October 2013 at 14:35

    Travis V : thanks! 🙂

  55. Gravatar of Philippe Philippe
    13. October 2013 at 17:18

    Libertaer, that doesn’t make any difference.

  56. Gravatar of Ralph Musgrave Ralph Musgrave
    13. October 2013 at 22:06

    Scott,

    You claim that for a monetary base increase to have a stimulatory effect, the central bank has to announce the increase as being permanent. Problem there is that any announcement by a government or central bank to the effect that a counter cyclical measure is permanent is a contradiction in terms. It won’t be taken seriously.

    A counter cyclical stimulatory measure is one which by definition may be reversed if and when the private sector gets too confident. It’s a bit like the recent announcement by the Bank of England that interest rates will stay low for two years: no one believes it.

    Of course the same problem applies to fiscal stimulus: i.e. the effect of ANY stimulatory measure, monetary or fiscal, is always reduced by income smoothing.

  57. Gravatar of Saturos Saturos
    14. October 2013 at 03:07

    Ralph, there is no time inconsistency in NGDP level targeting, or at least hitting your own inflation target. Announcing that the base will be wherever it needs to be to hit that target, or a weighted average of it and the unemployment target consistently over time, instead of implicitly targeting much lower levels of aggregate demand, has no reason not to be taken seriously.

  58. Gravatar of Geoff Geoff
    14. October 2013 at 15:41

    “In case one the price level doesn’t change.”

    If any of that additional money comes into existence through purchasing anything, then it is wrong to claim that prices won’t rise.

    If by “price levels” it is meant to refer to finished goods, then there might be an increase in those prices.

  59. Gravatar of libertaer libertaer
    15. October 2013 at 01:30

    Philippe,

    “Libertaer, that doesn’t make any difference.”

    Why do you think that? If the central bank follows an inflation or NGDP target, they can’t print money as they wish if this creates inflation.

    And even if they ignore the target and create a lot of inflation, this would be just like a tax. So the debt is paid by future generations, be it through taxes or inflation.

    By the way, I’m not a debt paranoid. My problem with people like Krugman is not that we can’t go deeper into debt. We can, but we shouldn’t. On top of it all, it would be useless. Japan did it and it didn’t work out. Almost 20 years of stagnation.

    Monetary policy is the only game in town.

  60. Gravatar of ssumner ssumner
    16. October 2013 at 10:12

    libertaer, You said;

    “But do you agree with my point that in a world with interest rates forever negative monetary policy by buying government bonds wouldn’t work?”

    I don’t know what “interest rates forever negative” means. Can you be more precise?

    Mike, The Fed can simply peg TIPS spreads or CPI futures if its serious about inflation targeting. Fortunately it’s not serious about inflation targeting.

    Ralph, You said;

    “You claim that for a monetary base increase to have a stimulatory effect, the central bank has to announce the increase as being permanent. Problem there is that any announcement by a government or central bank to the effect that a counter cyclical measure is permanent is a contradiction in terms. It won’t be taken seriously.”

    No, by “permanent” people mean enough has to be permanent to hit their long run inflation or NGDP target. They don’t mean it all has to be permanent.

  61. Gravatar of Sumner, Krugman and ZLB Denial | Last Men and OverMen Sumner, Krugman and ZLB Denial | Last Men and OverMen
    18. February 2017 at 08:44

    […] in Kimball’s proposal.)  I won’t discuss why this is wrong here; I’ve done a zillion other posts explaining why the zero bound is not a problem.”   […]

Leave a Reply