Daniel Thornton on QE

Here is Daniel Thornton of the St. Louis Fed:

The analysis presented here suggests that QE had little or no effect in reducing long-term yields relative to what they would have otherwise been.2 If QE did not significantly reduce long-term yields relative to what they would have otherwise been, it cannot have increased output or employment either.

I’m old enough to recall when the St Louis Fed had monetarist leanings.  A monetarist would immediately reply that interest rates are a lousy indicator of the stance of monetary policy.  But this statement isn’t even consistent with New Keynesian models.  After all, QE could easily raise the Wicksellian equilibrium interest rate in a NK model, and hence boost AD even if actual interest rates did not change.  Thus the term ‘cannot’ is way too strong.  What if the Fed had done Zimbabwe-style QE.  Would you expect interest rates to fall?  Would NGDP growth rise?

I’m also puzzled as to why he looks at time series data and not market reactions to policy announcements.

Elsewhere Thornton acknowledges that there are other possible mechanisms:

Another possibility is that other countries experienced a greater output decline relative to that of the United States, which caused their yields to decline compared with the United States. The second chart shows the gross domestic product (GDP) growth rates of Canada, France, Germany, the United States, and the United Kingdom since 2005. The GDP growth patterns of four of the five countries have been similar since the fourth quarter of 2008; the sole exception is France, whose growth declined more. Hence, it appears unlikely that divergent growth rates could account for the lack of support for QE.

Lots of puzzles here.  Surely US growth has exceeded British growth by a wide margin.  In any case, Britain also did lots of QE, so why make this comparison? Germany and France don’t even have their own monetary policies; they are part of the eurozone. And growth in the eurozone has been far below US levels, presumably due to the ECB’s unwillingness to be as expansionary as the Fed. Indeed they raised interest rates twice in 2011.

PS.  Off topic, but notice all the chatter about how unemployment is not a useful policy guide for the Fed.  Who argued the Evan’s rule should have been based on levels of NGDP.  I hate love to say I told you so . . . .

HT:  Jeff Livingston


Tags:

 
 
 

76 Responses to “Daniel Thornton on QE”

  1. Gravatar of WC WC
    11. February 2014 at 06:56

    It makes me cringe seeing Fed officials reason from a price change. I really don’t understand how this prevailing wisdom survives.

  2. Gravatar of TallDave TallDave
    11. February 2014 at 07:11

    And cue the chorus, once again:

    “Expectations uber alles!”

  3. Gravatar of Kevin Erdmann Kevin Erdmann
    11. February 2014 at 07:32

    The presumptions piled on top of presumptions on these topics amaze me. There is nothing false about the recent drop in the unemployment rate. Come July, when we’re below 6% with continued increases in wages, it will be interesting to see if everyone is still insisting on disbelief.
    There are so many moving parts in the monthly labor reports, it seems that no matter what your theory says should have happened, you can find something in there to support it.

    QE creates employment through lower rates, but rates increased, therefore, of all the labor indicators that are reported, the ones that show employment strength are wrong, and the months where all the indicators are strong are outliers.

  4. Gravatar of Felipe Felipe
    11. February 2014 at 07:41

    Off topic: Scott, have you seen the new Bhattarai, Eggertsson, Schoenle paper?

    In a simple DSGE model, we show analytically that more flexible prices always amplify output volatility for supply shocks and also amplify output volatility for demand shocks if monetary policy does not respond strongly to inflation.

    http://www.nber.org/papers/w19886

    I didn’t really understand the model, but under a taylor rule, it seems flexible prices do not improve the situation (they tested a few versions of the taylor rule). Unfortunately, they didn’t test an NGDP target. But most importantly, they didn’t test any level target. That is probably important.

  5. Gravatar of ssumner ssumner
    11. February 2014 at 09:54

    Kevin, It will be a very interesting year for the unemployment rate. It’s down 1.3% points in the past 12 months, and 0.3% the 12 months before that. We may be at the natural rate much sooner than most expect.

    Filipe. Those arguments never made any sense to me. For any given decrease in NGDP, a larger fall in P means a smaller fall in Y. I suppose they argue that more price flexibility creates a bigger fall in NGDP, which I think is utter nonsense.

  6. Gravatar of Kevin Erdmann Kevin Erdmann
    11. February 2014 at 10:26

    Considering the uptick in wage growth, I wonder if we have been getting pretty close to full employment, which was north of 6% when EUI was 75 weeks and over and minimum wage hikes were still in the near past, and it now will move back toward 5% over the next year. (speculative)
    http://research.stlouisfed.org/fred2/graph/?g=rZo

  7. Gravatar of Mark A. Sadowski Mark A. Sadowski
    11. February 2014 at 12:58

    In “An Injection of Base Money at Zero Interest Rates: Empirical Evidence from the Japanese Experience 2001-2006” by Yuzo Honda, Yoshihiro Kuroki, and Minoru Tachibana (March 2007):

    http://www2.econ.osaka-u.ac.jp/library/global/dp/0708.pdf

    Honda et al estimate four variable VARs (industrial production, core CPI, reserve balances and various financial variables) and find that the effect of QE on the yields of 5, 7 and 10 year Japanese bonds is significantly *positive*.

    I have VAR Granger causality tests that show that since December 2008 the US monetary base Granger causes nominal 10-year T-Note yields at the 10% significance level and that the impulse response is *positive*.

    I also find that since December 2008 the US monetary base Granger causes 5-year inflation expectations as measured by TIPS at the 1% significance level, and that the since April 2009 the UK monetary base Granger causes 5-year inflation expectations as measured by inflation-linked gilts at the 1% significance level.

    The effect on inflation expectations is as expected in each case, that is, positive. And since theoretically the effect of QE on inflation expectations should be positive, why anyone would expect QE to “significantly reduce long-term yields” is beyond me.

    P.S. Daniel Thornton and Stephen Williamson are evidence that St. Louis Macro is totally upside down from the rest of the planet’s.

    P.P.S.

    http://1.bp.blogspot.com/-BAnvd5XkIkc/UnKFj667RyI/AAAAAAAAAyM/-xbQBsjqcm0/s1600/QE+Rates.png

  8. Gravatar of Mark A. Sadowski Mark A. Sadowski
    11. February 2014 at 13:57

    Correction.

    “I have VAR Granger causality tests that show that since December 2008 the US monetary base Granger causes nominal 10-year T-Note yields at the 10% significance level and that the impulse response is *positive*.”

    should read

    “I have VAR Granger causality tests that show that since December 2008 the US monetary base Granger causes nominal 10-year T-Note yields at the 5% significance level and that the impulse response is *positive*.”

  9. Gravatar of Tommy Dorsett Tommy Dorsett
    11. February 2014 at 15:11

    Thornton utterly failed in this endeavor, as Japan would have had the most stimulative monetary policy in history from 1995-2011, by his definition.

    Of course it doesn’t help that the entire FOMC (including Yellen today) argues QE works by ‘lowering long rates.’

    No, an expected NGDP path lower than would otherwise be the case lowers long rates and the converse. That explains Japan and the positive relationship between US QE on periods and long rates.

  10. Gravatar of mpowell mpowell
    11. February 2014 at 15:22

    That chart is hilarious. Did you check it? Min is -20%, max is 10% with 5% ticks. And it runs from 2005 to present. Everyone bounces noisily around the 0-5% range. It’s impossible to discern any difference because there’s no resolution and no integrated growth measure. The 1-2% difference we might hope to find is buried in the noise. Someone throws a chart like that at you and you know they’re not interested in conveying information.

  11. Gravatar of flow5 flow5
    11. February 2014 at 16:48

    Thornton’s absolutely “on point”. Interest rates are determined by the supply of and demand for loan-funds (not the supply of and demand for money).

  12. Gravatar of lxdr1f7 lxdr1f7
    11. February 2014 at 16:52

    “If QE did not significantly reduce long-term yields relative to what they would have otherwise been, it cannot have increased output or employment either.”

    I agree with Scott “shouldnt” instead of “cannot” have increased output and employment is more accurate IMO.

  13. Gravatar of Major_Freedom Major_Freedom
    11. February 2014 at 17:21

    “I’m old enough to recall when the St Louis Fed had monetarist leanings. A monetarist would immediately reply that interest rates are a lousy indicator of the stance of monetary policy.”

    The term “lousy indicator” of course means that interest rate changes and/or levels do not tell us about NGDP. Orthodox monetarists did not, and do not, believe that interest rates are a “lousy indicator” because they don’t tell us about NGDP. No, they held that interest rates do not tell us what is happening to the price level, or the money supply, which is what orthodox monetarists are concerned about.

    NGDP is not a “good indicator” of monetary policy, if we do what MMs are doing, which is comparing existing policy with a standard external to the standard being critiqued, then we can argue this: In the case of MM, the standard being critiqued is price levels and the money supply, and the external standard (i.e. MM standard) is NGDP. We can critique the MM standard by comparing the MM standard of NGDP to an external standard.

    Now I recognize that this is challenging to do for those wedded to MM. With so many flawed convictions, from believing that unemployment cannot ever rise due to market forces dominating the Fed in determining money holding and spending patterns, to ignoring why prices are prevented from adjusting according to market forces, to ignoring capital and labor market distortions caused by inflation, with that bag of tools at one’s disposal, competing theories are understandably perceived as insane. I get it. Thinking outside the box is what pragmatists have incredible difficulty with. Pragmatists fear radicalism.

    If one dives into studying pragmatism, and philosophy in general (which is a rarity in MM circles), it is rather easily grasped that practising pragmatism doesn’t actually accomplish what it says can be done if it is practised. Why is this? Because pragmatism is mired inextricably in self-contradiction. In other words, pragmatism is wrong on its own terms.

    First, contrary to the content of pragmatism, it does not work. And since pragmatists believe that the only truth is what works, pragmatism isn’t true. Not only that, but there are major problems with the meaning of “work”. What does it actually mean? Work for what exactly?

    Second, pragmatism is itself an ideology, which means according to pragmatism’s hostility towards and rejection of all ideology, as well as its call to dismiss ideologues, means we have to dismiss pragmatism on its own terms.

    Third, how would pragmatic libertarians and pragmatic socialists like Sumner reconcile their mutually incompatible beliefs? If pragmatic libertarians believe economic freedom “works” in the long run, whereas pragmatic socialists believe central banking “works” in the short run, then there is nothing in pragmatism that can settle the dispute. For why, pragmatically, should everyone be concerned more with the short term as opposed to the long run? Or vice versa? How would a long term oriented pragmatist settle a dispute with a short term oriented one?

    Indeed, the short term pragmatist can assert that if the Fed sent checks to everyone, then standards of living would rise for everyone in the short term. He could dismiss all long term pragmatists as ideologues. Or vice versa.

    If the average person cannot be sold on radicalism, then a fortiori the criminal ruling class, those who might well lose in the long-run, certainly won’t be convinced either. At best, the anti-radicalist pragmatist might concede: “OK, I will agree that abolishing central banking sounds like a good idea. But pragmatically, to ease the transition and minimize the costs and disruption, let’s move toward that ideal very, very gradually.” And so we are again back, haphazardly, to the Republican or Democrat Parties, and MMs it seems, who are the self-professed overlords of gradualism.

    It is no accident that Sumner preaches gradualism. Sure, he believes in freedom, free markets, peace, and all the rest, but these have to be approached slowly, push and pull, piecemeal, by the democratic consensus. Nicely snug in the warm bosom of the status quo.

    But there is a problem.

    The problem is that never before in the history of mankind has any meaningful revolutionary change in human life been carried by the torch of pragmatism. For who in world would join a revolutionary movement, and commit himself for life to the fringe, for the sake of 10% more computers, or 5% more hamburgers? Who will man the barricades, either physically or spiritually, for more M&Ms or sports cars? For the middle east revolutions away from theocratic dictatorship, who in their right mind would believe pragmatism is responsible? No, they are being moved by deep moral convictions and radical thought, exactly the opposite of pragmatism.

    Nobody can claim that their values of liberty, free markets, and justice, are being accomplished or are capable of being accomplished through pragmatist ideology. Washington is dominated by pragmatists. Is Washington working?

    Why not let the crazy ideologues worry about the movement towards liberty? The pragmatists, as usual, will just take what comes, and if liberty does come, it will be because of the radicals. You’re welcome.

  14. Gravatar of benjamin cole benjamin cole
    11. February 2014 at 17:36

    The St Loo Fed also recently released paper to the effect that QE would have to quadruple in size to boost GDP…
    OT but worse: in her recent testimony Yellen straitjacketed herself into eliminating QE this year, come what may…
    The Fed will copy Bank of Japan 1992-2012? I think so…

  15. Gravatar of Tom Brown Tom Brown
    11. February 2014 at 19:10

    O/T for anybody: today I asked Nick Rowe to summarize the goal of good monetary policy. His response:

    “…the goal of good monetary policy is to try to make Say’s Law true.”

    Well I’m certainly no expert on Say’s Law, but I shared that w/ an Austrian-leaning friend and he responded:

    “If you outlaw fractional reserve banking (bank must sell 10 year bonds to get money for 10 year loans) and use gold and silver coins, Say’s law would be true.”

    His statement seems far from true to me. How would you respond? I launched into a simple Nick Rowe style story demonstrating how moving from a barter economy to a monetary one with gold as the MOA could cause a problem (in the face of a shock), which I’m imagining would push my example far from the Say’s Law ideal… (should the CB not respond to the shock appropriately) but since I don’t really know what I’m talking about I thought I’d toss that out to you folks and see how you might have responded. Thanks.

  16. Gravatar of ssumner ssumner
    11. February 2014 at 19:38

    Kevin, Maybe, but I’d focus on nominal wage growth as an indicator of overheating. Real wages require the use of . . . the price level.

    Which of course doesn’t really exist. 🙂

    Mark, Thanks, and the more recent Japanese QE seems to have boosted all sorts of aggregates in 2013.

    mpowell, I wondered about that chart as well. I knew already that the US had grown considerably faster than Britain, and the chart seems impossible to read.

    Tom, I agree with Nick, indeed I made a similar argument in a recent post at Econlog.

  17. Gravatar of Tom Brown Tom Brown
    11. February 2014 at 19:56

    Scott, thanks, I thought you’d agree w/ Nick, but off the top of your head, what is your response to my Austrian friend’s statement.

  18. Gravatar of Kevin Erdmann Kevin Erdmann
    11. February 2014 at 20:02

    Scott, I’m using it as an indicator of tightness in the labor market. Nominal wages would indicate monetary policy. (Or, rising price-adjusted wages coupled with rising unemployment would also indicate monetary policy.)

    Incidentally, it looks like job openings per unemployed worker is beginning to revert strongly back to the pre-2008 Beveridge Curve. I found a Boston Fed paper on it, where they disaggregated the shift in the ratio, and, oddly, they found that almost all of the excess unemployment was among:
    1) 20-34 year old job leavers, new entrants, and re-entrants
    2) 45+ year old job losers

    Other categories tended to be near the pre-2008 trends. The good news is that both categories appear to be pulling back to the old trend. In January looks like the trend continues with the 45+ year olds. Such a coincidence that it’s happening right when EUI ends.

    With all of these workers losing EUI, which I hear has a great big multiplier, I expect 1st quarter GDP projections must be very negative. Unemployment rate should be shooting up since we’re losing all that stimulus….

    http://idiosyncraticwhisk.blogspot.com/2014/02/december-jolts.html

  19. Gravatar of Mike Sax Mike Sax
    11. February 2014 at 20:45

    Tom did I not tell you he agrees with Nick?

  20. Gravatar of Tom Brown Tom Brown
    11. February 2014 at 21:18

    Mike, yes: but I already suspected that. I was more interested in how these nice MM folks would respond to that statement by Vincent.

  21. Gravatar of Tom Brown Tom Brown
    11. February 2014 at 21:23

    …and BTW, Nick thinks he got that statement from Brad DeLong!

  22. Gravatar of Benjamin Cole Benjamin Cole
    12. February 2014 at 03:47

    BTW, a Yi Wen, also of the St Louis Fed, produced a recent study to the effect that QE would need to be quadrupled to bring GDP back to trend, and that QE is deflationary.

    I realize that freedom of speech and diversity are wonderful things. But what does it mean when”serious” economists at the St. Louis Fed are producing conflicting studies, and studies that appear to say we can QE to the moon, pay off the national debt, and have little or no influence or positive influence on real economic output or inflation?

    Policymakers are supposed to believe what? The public is supposed to believe what?

    See http://thefaintofheart.wordpress.com/2013/10/25/st-louis-fed-u-s-national-debt-can-be-wiped-out-in-a-few-years-by-qe-and-inflation-reduced/

  23. Gravatar of ssumner ssumner
    12. February 2014 at 06:40

    Tom, Your friend is wrong, as you’d still have NGDP shocks and sticky wages.

    Kevin, I don’t know about Q1 GDP, as inventory swings can affect a single quarter. But I do agree the unemployment rate will fall faster with the elimination of extended benefits.

    I still have doubts about real wages. That’s partly because I don’t trust inflation numbers. The variation in your series in recent years is driven by variation in inflation that seems bogus to me.

    In the short run nominal wages may be a good indicator of tightness, as nominal wages are sticky. Any significant breakout from the recent 2% nominal wage gains track would reflect tightening of the labor market.

    Ben, The problem with those kind of claims is that they operate in a vacuum. You need to consider the signaling that accompanies the QE.

  24. Gravatar of ssumner ssumner
    12. February 2014 at 06:52

    Kevin, Interesting post. Is there any possibility that the drop in December reflected expectations that EUI would end? Was the media reporting the phaseout of EUI in December? Were recipients told their EUI might run out at the end of the year?

  25. Gravatar of Salemicus Salemicus
    12. February 2014 at 08:11

    Scott,

    What do you make of the new forward guidance issued by the BoE today? In the British press it is being portrayed as an embarrassing climbdown. For myself, I am much more concerned that they have still not moved to a “targeting the forecast” model.

  26. Gravatar of Kevin Erdmann Kevin Erdmann
    12. February 2014 at 08:32

    The unemployment drop in December did not come from 45+ year olds, so it presumably didn’t come from 45+ year old job losers that populated the EUI.
    The drop in January came entirely from the 45+ year olds. I don’t have January job openings data yet, but the Beveridge Curve for this group will push left again in January, back toward the pre-2008 trend.
    I expect to see the 45+ year old Beveridge Curve continue to shift left due to the end of EUI, as unemployment drops from this group. This unemployment decline will be related to a reshifting of the aggregate Beveridge Curve.
    Then, complimentary effects from the re-employment of this group should create broad employment improvements, including higher job openings, higher quits, etc., which will be associated with further drops in unemployment as we climb up the Beveridge Curve from there.
    I don’t have data that fine from North Carolina, but my speculation is that this sort of two act effect is what happened there, and that is why the unusual drops in unemployment there were at first associated with some exits from the labor force, but after EUI ended were increasingly associated with employment gains with each passing month.

  27. Gravatar of TallDave TallDave
    12. February 2014 at 09:05

    Mark A. Sadowski 11. February 2014 at 12:58

    Really, everyone commenting on economics should read this.

  28. Gravatar of flow5 flow5
    12. February 2014 at 09:55

    “what does it mean when”serious” economists at the St. Louis Fed are producing conflicting studies”

    See: “In the 1960s, the St. Louis Fed garnered a reputation as a maverick in the Federal Reserve System because of its espousal of monetarism”

    Inflation analysis cannot be limited to the volume of wages and salaries spent. To the Keynesians (& Friedman), aggregate monetary demand is nominal GDP, the demand for services (human) and final goods. This concept excludes the common sense conclusion that the inflation process begins at the beginning (with raw material prices and processing costs at all stages of production) and continues through to the end. Roc’s in MVt = roc’s in all transactions (roc’s in nominal-gDp is just a proxy).

  29. Gravatar of flow5 flow5
    12. February 2014 at 10:29

    We’re already seeing how bad their forward guidance is (for short-term interest rate projections):

    http://1.usa.gov/oLC2C9

  30. Gravatar of Tom Brown Tom Brown
    12. February 2014 at 13:07

    Mark, is there evidence for what happens to yields when QE is unwound? What would be your guess? Yields down? Depends?

  31. Gravatar of Mark A. Sadowski Mark A. Sadowski
    12. February 2014 at 15:35

    Tom Brown,
    Is it true that Mike Sax thinks that Keynes wanted to “drown the government in the bathtub and dance” on his own grave?

    http://2.bp.blogspot.com/_5JJarCb6DPo/Sn9mKH2wI8I/AAAAAAAAAdc/vK2OHcZXJUk/s1600-h/keynsie+dansingjpg

  32. Gravatar of Mark A. Sadowski Mark A. Sadowski
    12. February 2014 at 16:52

    Tom Brown,
    “Mark, is there evidence for what happens to yields when QE is unwound? What would be your guess? Yields down? Depends?”

    My interest in VAR studies on QE, and my Granger causality tests on QE, grew out of my studies of Post Keynesian empirical research on endogenous money. Most of these research papers do Granger causality tests on monetary aggregates and they usually find (among other things) that bank loans Granger cause the monetary base. This is taken as a confirmation of the endogeneity of money.

    But if a central bank is targeting the unsecured overnight interbank lending rate it would be very surprising indeed if bank loans didn’t Granger cause the monetary base, right? So it occured to me if the unsecured overnight interbank lending rate is fixed, and the central bank is increasing the monetary base at ad hoc rates, as in QE, then we might find that the monetary base Granger causes other variables under those conditions. And in fact that has turned out to be a very fruitful line of inquiry.

    So, in general, we would probably expect VAR studies, and Granger causality tests in particular, to only show that the monetary base causes other variables at the zero lower bound in interest rates.

    What do the historical incidents show us about the unwinding of QE?

    There are only a handful of nations which have done QE at the zero lower bound in interest rates for which we have easily accessible detailed data. In terms of the time they were at the zero lower bound in interest rates, they are: 1) the US from 1933 to 1937, and 1937 to 1941, 2) Japan from 2001 to 2006, 3) the US from 2008 to present, 4) Japan from 2008 to present, 5) the UK from 2009 to present, 6) Switzerland from 2009 to present, and 7) Sweden from 2009 to 2010. (I haven’t looked at cases #4, #6 and #7 owing to the short period of time that the monetary base was expanded.)

    In the case of the US Great Depression, the Fed left the zero lower bound without ever really decreasing the monetary base. Growth in the monetary base was paused for several months in 1936-37 and this was soon followed by the 1937 recession.

    When the US entered WW II the Fed left the zero lower bound and continued to increase the monetary base. Moreover there was never any need to exit from the more than six-fold increase in the monetary base that took place in 1932-48. The most that the monetary base decreased was from $48.413 billion in December 1948 to $42.960 billion in April 1950, or by 11.3%. And even that decrease probably had only minimal negative consequences because of the expected decline in real output following World War II. And keep in mind the entire decrease occurred away from the zero lower bound in interest rates.

    The BOJ reduced the monetary base by 24.4% from January to November 2006, with the call rate being raised significantly above the zero lower bound in June. So the call rate was raised in the middle of the period of time that QE was unwound.

    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 this recession. It’s hard not to connect the dots between this result and the BOJ’s sudden and sharp withdrawal of QE, and raising of the call rate, given one literally followed upon the other within months.

    The other cases are ongoing.

    My guess would be that unsecured overnight interbank lending rate should not be raised too quickly, and that QE should not ever be unwound (yields would probably go down).

  33. Gravatar of Mark A. Sadowski Mark A. Sadowski
    12. February 2014 at 17:00

    One more thing. Sweden is not ongoing. Sweden’s economy was doing quite well in 2009 to 2010 and has done much worse since it unwound QE, and bond rates have generally fallen.

  34. Gravatar of Mark A. Sadowski Mark A. Sadowski
    12. February 2014 at 17:08

    For more on Sweden see this post:

    http://www.themoneyillusion.com/?p=25716

  35. Gravatar of Mark A. Sadowski Mark A. Sadowski
    12. February 2014 at 18:15

    To be more specific, here’s the proportions that the monetary base fell from peak on a monthly basis before leaving the zero lower bound in previous QE incidents:

    1) US 1933-37
    1.1% from December 1936 to February 1937
    2) US 1937-41
    None
    3) Japan 2001-06
    17.5% from January to May 2006
    4)Sweden 2009-10
    11.9% from November 2009 to June 2010

  36. Gravatar of David Beckworth David Beckworth
    12. February 2014 at 18:21

    Mark A. Sadowski,

    That is a great insight on the monetary base and the ZLB. Have you written up your results in a paper? Would love to see it.

  37. Gravatar of Mark A. Sadowski Mark A. Sadowski
    12. February 2014 at 18:28

    David,
    Thanks. I’m planning on doing something with all of this at some point. I’m not sure what form it should take though given the dearth of related published research.

  38. Gravatar of Tom Brown Tom Brown
    12. February 2014 at 18:35

    Mark, first of all, thank you again for another impromptu analysis. I feel like I should be paying you (Scott I’d pay you too, but If I paid by the word, I’d be paying Mark a lot more). I’m always seeing people opine about what would happen should QE be unwound, but this is the first time (I’m aware of) that I’ve seen evidence. Just to be clear, the call rate is the overnight rate, correct?… so yield curves might flatten or be inverted then?

    Re: Mike’s comment: yes, I saw your comment… you one upped me with Keynes, but I (partially) beat you to razzing Mike about that:

    http://worthwhile.typepad.com/worthwhile_canadian_initi/2014/02/tiff-macklem-retail-competition-flexible-it-vs-ngdplt.html?cid=6a00d83451688169e201a73d75d333970d#comment-6a00d83451688169e201a73d75d333970d

    … Keynes himself dancing though was a nice touch!

    Mike, sorry to pick on you man, but you left yourself open for that one, you’ve got to admit! 😀

  39. Gravatar of Mark A. Sadowski Mark A. Sadowski
    12. February 2014 at 18:42

    Tom Brown,
    “Just to be clear, the call rate is the overnight rate, correct?… so yield curves might flatten or be inverted then?”

    Yes and yes.

  40. Gravatar of Tom Brown Tom Brown
    12. February 2014 at 19:35

    Mark or Scott: I was thinking that if the Fed, for some reason, wanted to use OMOs to set the FFR above the IOR rate (IORR) without first unwinding its balance sheet, it could do so by upping the reserve requirement (RR) to 167% = $2.5T/$1.5T (the ratio of reserves to checkable deposits). Is there some macro-economically important effect that might result from upping the RR to 167%?

  41. Gravatar of Benjamin Cole Benjamin Cole
    12. February 2014 at 20:31

    Mark and Tom:

    Fascinating. This supports my gut sense that the Fed needs to condition the public that QE is not exotic or unconventional or short-term. It is another tool to be used to keep the economy growing. And the Fed balance sheet may be unchanged for decades. I add on that the Fed needs to refer to the existing balance sheet as “average-sized’ or something like that, so people stop calling it a “large” balance sheet.

    However, the Fed has not done this; even Ben Bernanke did public handwringing on the difficulty of managing a large balance sheet. And now Yellen keep pointing to the day we stop QE.

    The Fed, like the Bank of Japan, has a squeamish aversion to QE and balance sheets.

  42. Gravatar of Erik Trygger Erik Trygger
    13. February 2014 at 02:50

    Scott, OT from this post but very relevant for the ZLB debate in the eurozone. According to Lars Svensson:

    ” In the Eurozone there can be a substantial difference between the official policy rate, the Main Refinancing Rate (MFR), and the interest rate that matters to the economy, the Eonia rate (the overnight rate in the Eurozone). Since the ECB controls the Eonia rate by controlling the floor of the interest-rate corridor around the MFR, one may see the Eonia rate as the actual policy rate in the Eurozone. Therefore the figure shows the Eonia rate instead of the MFR.”

    Judging by the Eonia rate the Eurozone probably ZLB for years.

    http://larseosvensson.se/2013/11/18/ekonomistas-a-comparison-of-monetary-policy-in-sweden-with-that-in-the-eurozone-the-uk-and-the-us-english-translation/

  43. Gravatar of Tom Brown Tom Brown
    13. February 2014 at 03:59

    Mark, in my comment above:

    http://www.themoneyillusion.com/?p=26159#comment-318452

    I was assuming that with excess reserves (ER) > 0, that the call rate (I was calling it the FFR) is driven down to close to the IORR, at least if the IORR > 0. Is that a correct assumption? After re-reading your comments, I’m not sure I have that correct.

  44. Gravatar of Negation of Ideology Negation of Ideology
    13. February 2014 at 05:34

    Mark, Tom and Ben –

    I agree this is fascinating. The attack on QE (well, one attack) seems to be that the Fed is going to have to dump all these long term bonds on the market at a loss. I’ve always been skeptical of that criticism. It seemed to me more likely that most of the increase in the base will turn out to be permanent, allowing the Fed to simply slow the rate of increase when necessary. If that’s not enough it can reduce the base somewhat by reducing bonds as they come due. Then it can raise IOR. Then it could sell some of the gold hoard.

    But if I understand Mark’s data correctly, those steps are probably unnecessary, most of those bonds will be held until maturity.

    Also, I think it would be wise if the Fed always held a mix of maturities of bonds. There’s no reason for the Fed to limit itself to short term bonds even in normal times. They should just hold enough so they can reduce the base by whatever amount is likely over that time. The base will never have to be reduced to zero.

  45. Gravatar of Mike sax Mike sax
    13. February 2014 at 05:56

    I don’t know about that Tom. I got a new post where I expand on this more.

  46. Gravatar of Tom Brown Tom Brown
    13. February 2014 at 06:14

    Mike, I just checked: I only see the post up on my “Scintillating Analysis” 😀

    Plus Nick & I were just discussing your comment … you might want to check it out. I think he makes a good point:

    http://worthwhile.typepad.com/worthwhile_canadian_initi/2014/02/tiff-macklem-retail-competition-flexible-it-vs-ngdplt.html?cid=6a00d83451688169e201a73d76f0ad970d#comment-6a00d83451688169e201a73d76f0ad970d

    … plus I engage in more scintillating analysis regarding what you might be thinking, but I may have it all wrong, so you may need to set me straight.

  47. Gravatar of Mark A. Sadowski Mark A. Sadowski
    13. February 2014 at 06:22

    Tom Brown,
    The BOJ calls their overnight rate the “call rate”. Different central banks use different names for the overnight rate.

    Interest on reserves was only instituted by both the Fed and the BOJ in October 2008. The Japanese call it the “complementary deposit facility rate”.

    The Riksbank is one of nine central banks by my count (Australia, Canada, New Zealand, UK, Denmark, Norway, Sweden, Mexico, Timor Leste) which have no reserve requirements at all.

    Raising reserve requirements increases the demand for the monetary base and thus it is contractionary. The Fed doubled reserve requirements at member banks from 13.0% in July 1936 to 26.0% in May 1937. Although most economists today think that the decision to sterilize gold inflows in December 1936 (thus effectively freezing the size of the monetary base) was the main cause of the 1937 recession, the increase in reserve requirements obviously also played a role.

    Usually interest on reserves acts as a floor for the overnight rate. This has not been the case in the US because some institutions which have deposits in account at the Fed (e.g. Fannie Mae and Freddie Mac) don’t get credited with any interest if they hold those balances overnight. This creates an arbitrage opportunity and results in the fed funds rate being lower than interest on reserves.

    As a practical matter the fed funds rate is no longer of much importance. Joseph Gagnon and Brian Sacks have argued that the Fed should change its policy interest rate to the soon to be instituted reverse repo rate:

    http://www.piie.com/publications/interstitial.cfm?ResearchID=2558

  48. Gravatar of Tom Brown Tom Brown
    13. February 2014 at 06:36

    Mark, Thanks.

    re: Gagnon & Sacks: JKH discussed that here:

    http://pragcap.com/a-new-operating-framework-for-the-federal-reserve

    I have not yet read their proposal.

  49. Gravatar of Tom Brown Tom Brown
    13. February 2014 at 06:49

    Mark, if raising reserve requirements is contractionary, then could that serve as an effective tool for putting the brakes on inflation should that ever start to become a problem? It seems like a good alternative since it wouldn’t involve any OMOs or any payment of IOR.

  50. Gravatar of Edward Edward
    13. February 2014 at 07:01

    This is truly
    annoying and awful
    http://www.bankrate.com/finance/federal-reserve/federal-reserve-policy-hurts-retirees-1.aspx

  51. Gravatar of ssumner ssumner
    13. February 2014 at 07:32

    Salemicus, I don’t have anything to comment on, as the press doesn’t seem willing to explain exactly what the forward guidance was previous to the change.

    Kevin, Thanks for that info.

    Mark, Interesting data. In 1937 the big issue was a rise in the demand for base money.

    Tom, A higher RR probably wouldn’t have a big effect right now, but the risk is certainly there that it could be contractionary. On the other hand the expectations channel suggests it might be immediately contractionary.

    Erik, The green line is the EZ rate, which appears to have reached the zero bound in late 2012. However I believe other rates still matter, at least if you look at market responses to ECB rate cuts.

  52. Gravatar of Max Max
    13. February 2014 at 07:33

    “It seems like a good alternative since it wouldn’t involve any OMOs or any payment of IOR.”

    If interest rates rise above IOR, then reserve requirements are effectively a tax on bank accounts. But under current conditions, they could massively increase reserve requirements without ill effect (at least, if they could convince people that it was not intended to be contractionary).

  53. Gravatar of Tom Brown Tom Brown
    13. February 2014 at 07:35

    Scott, do you think that raising the RR could be an effective brake on inflation then?

  54. Gravatar of Mark A. Sadowski Mark A. Sadowski
    13. February 2014 at 07:41

    Tom Brown,
    I tend to view interest on reserves as a more flexible (and more modern) alternative to reserve requirements. There is an enormous intersection between between those central banks that have instituted interest on reserves and which do not have reserve requirements. The only central banks which have both, to my knowledge, are the Fed, the ECB and the BOJ. Two of those only just instituted interest on reserves since the Great Recession and the third is, well, the ECB (a very odd duck).

    In short, having both IOR and RR is like wearing both a belt and suspenders. If it makes you feel more secure then great, but I suspect your pants will stay up just as well if you wear only one or the other.

  55. Gravatar of Tom Brown Tom Brown
    13. February 2014 at 07:48

    Max, I was thinking to raise the RR just enough to eliminate excess reserves. I see what you mean by a tax though… not all banks will have enough reserves to start out and would need to borrow from other banks (but those other banks would profit!)… and certainly if credit expands w/o an accompanying expansion of reserves through OMPs… again it’s a tax (presumably they’d need to borrow from the Fed at a new higher rate). It’s something I don’t see brought up very often, so I was curious what people thought.

  56. Gravatar of Tom Brown Tom Brown
    13. February 2014 at 07:51

    Mark, you don’t think RR’s are in some sense cheaper to implement since the CB doesn’t need to pay anything?

  57. Gravatar of Mark A. Sadowski Mark A. Sadowski
    13. February 2014 at 08:09

    Tom Brown,
    Why would a central bank need to worry about having enough money?

  58. Gravatar of Tom Brown Tom Brown
    13. February 2014 at 08:22

    Mark, good point: but they do need to keep their BS in balance right? Not that I imagine paying IOR will be a huge burden. But it is an outflow (and they aren’t getting any assets in return for that outflow)… plus the fact that it is an outflow slightly tends to counteract the contractionary purpose (i.e. it does put more base money out there overall). Plus, maybe RR is politically easier to sell to a public which may be dismayed to hear that not only has the CB expanded reserves, but is now paying interest on those reserves.

    I realize those are minor points… and perhaps the overall case for IOR (rather than RR) may well be stronger.

  59. Gravatar of Mark A. Sadowski Mark A. Sadowski
    13. February 2014 at 08:48

    Tom Brown,
    “…but they do need to keep their BS in balance right?”

    A shortage of assets would only be a problem to a central bank if it wanted to sell assets in order to reduce the monetary base. If such a situation ever arose it could simply draw on the treasury. (The ECB is obviously a more complicated case, but let’s ignore that for the moment.)

    Interest on reserves is currently in practice in several currency areas that comprise over half of global GDP. To the best of my knowledge, Canada has had it since 1992, Norway since 1993, Sweden since 1994, Australia since 1997, the Euro Area since 1998, New Zealand since 1999 and the UK since 2001. If this were a problem surely it would have been discovered before now. Moreover the trend is away from RR towards IOR, not the other way around.

  60. Gravatar of Max Max
    13. February 2014 at 08:54

    Tom, it’s a technically perfectly viable approach, but it sort of goes against the trend in central banking toward relaxing (or eliminating entirely) reserve requirements. The other option which doesn’t involve paying interest is to sell the long term bonds that the Fed bought.

  61. Gravatar of TravisV TravisV
    13. February 2014 at 09:26

    Dear Commenters,

    Why is the market pricing bonds like this?

    Jeff Gundlach: Most Bond Categories Are Greatly Overvalued Relative to Treasuries

    http://blogs.barrons.com/incomeinvesting/2014/02/12/gundlach-sees-treasury-rally-extending-10-yr-yield-falling-to-2-5/?mod=BOLBlog

  62. Gravatar of Tom Brown Tom Brown
    13. February 2014 at 09:31

    Max,

    “The other option which doesn’t involve paying interest is to sell the long term bonds that the Fed bought.”

    True, but they do risk taking a loss on them when they sell. As Mark points out, ultimately Tsy will bail them out if they accumulate negative equity, but that sounds politically dicey.

    Mark & Max: I don’t doubt that the trend it towards IOR and away from RR. I’m just not entirely sure why. I also don’t doubt that IOR has been proven to work just fine.

    I think that we should keep RR around as a back up measure at least… so I guess I like the belt and suspenders approach.

  63. Gravatar of Edward Edward
    13. February 2014 at 10:10

    TravisV

    Because people are stupid about bonds. They dont realize that bonds are just about the most unsafe investment out there

  64. Gravatar of Max Max
    13. February 2014 at 10:42

    Tom, “True, but they do risk taking a loss on them when they sell.”

    They can also lose money if they hold to maturity (if interest paid out exceeds interest earned).

    “ultimately Tsy will bail them out if they accumulate negative equity, but that sounds politically dicey.”

    They won’t need a bail-out, they’ll just withhold future profits. An asset will be conjured on the balance sheet to reflect this. The Fed has a very lucrative franchise (printing $100 bills), so its solvency is not in doubt.

  65. Gravatar of Tom Brown Tom Brown
    13. February 2014 at 11:37

    Max, you write

    “They can also lose money if they hold to maturity (if interest paid out exceeds interest earned).”

    I’m not getting what you mean there.

    “…so its solvency is not in doubt.”

    No argument there. But w/ RR it’s hard for me to imagine any problem for the Fed no matter how quickly they acted. Should the Fed want to unwind overnight, however, has got to be messier: e.g. it’s at least conceivable to me that they might be in the hole for a while afterwards… and then some yahoos from congress might get up in arms and demand an audit. 😀

  66. Gravatar of Max Max
    13. February 2014 at 12:15

    Tom, “”They can also lose money if they hold to maturity (if interest paid out exceeds interest earned).”

    I’m not getting what you mean there.”

    If the central bank buys a 3% bond and IOR averages over 3% for the life of the bond, then it loses money.

    If it imposes required reserves and doesn’t raise IOR, then it wouldn’t lose money, but this is a tax on bank deposits. Banks wouldn’t like that!

  67. Gravatar of ssumner ssumner
    13. February 2014 at 15:04

    Tom, Yes, but it’s a very clumsy instrument. I’d prefer open market sales.

  68. Gravatar of flow5 flow5
    14. February 2014 at 16:14

    “If this were a problem surely it would have been discovered before now”

    LOL. No one at the Fed can forecast either.

  69. Gravatar of flow5 flow5
    14. February 2014 at 16:22

    “I see what you mean by a tax though…”

    LOL. A brief “run down” will indicate just how costless, indeed how profitable – to the participants, is the creation of new money. If the Fed puts through buy orders in the open market, the Federal Reserve Banks acquire earning assets by creating new inter-bank demand deposits. The U.S. Treasury recaptures about 98% of the net income from these assets. The commercial banks acquire “free” legal reserves, yet the bankers complained that they didn’t earn any interest on their balances in the Federal Reserve Banks.

    On the basis of these newly acquired free reserves, the commercial banks created a multiple volume of credit & money. And, through this money, they acquired a concomitant volume of additional earnings assets. How much was this multiple expansion of money, credit, & bank earning assets? Thanks to fractional reserve banking (an essential characteristic of commercial banking) for every dollar of legal reserves pumped into the member banks by the Fed, the banking system acquired about 93 (c. 2006), dollars in earning assets through credit creation.

  70. Gravatar of Mark A. Sadowski Mark A. Sadowski
    14. February 2014 at 16:39

    flow5,
    “No one at the Fed can forecast either.”

    But we’re not talking about forecasting, are we?

  71. Gravatar of Tom Brown Tom Brown
    14. February 2014 at 16:42

    flow5, I interpreted Max’s tax comment to mean that raising the RR to eliminate excess reserves would add some cost to any further expansion of credit requiring borrowing of reserves by the banks, especially if the FFR were raised above the IOR.

    In a relative sense this would end up costing the banks more than if it had not been done (assuming the Fed didn’t provide the required reserves for further credit expansion by more asset purchases).

    I’m not sure why you bring up 2006. But if the RR had been raised then, say to 100%, and there were no LSATs by the Fed to create the required reserves, but instead the reserves had to be borrowed, then banks holding deposits would feel an extra burden (loss of spread on their BSs) compared to an RR level of 10%.

    I guess the overall point is that higher RRs are likely to result in thinner spreads for banks.

  72. Gravatar of Max Max
    14. February 2014 at 23:15

    Tom, “I interpreted Max’s tax comment to mean that raising the RR to eliminate excess reserves would add some cost to any further expansion of credit requiring borrowing of reserves by the banks, especially if the FFR were raised above the IOR.”

    Replace “especially” with “only”. If FFR=IOR, then RR isn’t a deposit tax because there would be no cash-equivalent asset with a higher return than reserves. The Fed would be supplying the required reserves at no cost.

    Incidentally, the current problem of rates being lower than IOR could be remedied by RR, but a simpler method would be for the Fed to borrow overnight from non-banks (effectively allowing non-banks to deposit money at the Fed).

  73. Gravatar of flow5 flow5
    15. February 2014 at 13:04

    But we’re not talking about forecasting, are we?

    No, the subject’s much clearer. The payment of interest on excess reserve balances demonstrates exceptional delusion. It’s for high level, conceptual thinkers, not bean counters.

    “I guess the overall point is that higher RRs are likely to result in thinner spreads for banks”

    Do the CB’s required reserve expenses outweigh the Fed’s “manna from heaven”? Not according to studies. The CBs pay for what they already own (interest on the savings deposits they accept). Interest is the CB’s (& the system’s), largest expense item. You just reverse Reg Q legislation gradually & get the CBs out of the savings business. The CBs would then become more profitable.

  74. Gravatar of flow5 flow5
    15. February 2014 at 14:53

    “The Fed would be supplying the required reserves at no cost”

    CBs create new money when the lend/invest. They do not loan out existing deposits (saved or otherwise). They could continue to lend even if the non-bank public ceased to save altogether. The lending capacity of the member banks is determined by monetary policy (not the savings practices of the public).

    Any device or policy that attracts and impounds CB held savings displaces its rate of utilization (eliminates its outlet). Unspent, or un-invested savings, represents an heretofore unrecognized leakage in Keynesian National Income Accounting. Unused savings exert a deflationary or net contractionary effect (unless offset by an easier FOMC policy).

    Both the elimination of Reg Q ceilings & paying interest on excess reserve balances allows the commercial bankers to outbid the non-banks for loan-funds. Such a cessation in the circuit income & the transactions velocity of funds will reduce real-output and cause a reorientation in monetary policy – ultimately leading to stagflation. That’s a pyrrhic victory.

  75. Gravatar of Tom Brown Tom Brown
    15. February 2014 at 15:34

    Max: I agree: “only” was the word to use.
    flow5: I agree that CB’s do no loan out existing deposits. However, I seriously doubt anyone here thinks they do!

    You both lose me w/ the rest (especially you flow5), but that’s OK… I’m not really looking for any further clarification at this point!… I’m happy to move on. 😀

    (However, I’ll think a bit more about what you said Max… maybe the light will click on at some point)

  76. Gravatar of flow5 flow5
    16. February 2014 at 11:39

    “I seriously doubt anyone here thinks they do!”

    What they don’t do is think (& all of the Fed’s chairman have said that the CBs loan out savings – that’s why Reg Q ceilings were eliminated – by mistake, i.e., the ABA’s politics – see Alton Gilbert’s delusion: “Requiem for Regulation Q: What It. Did and Why It Passed Away” ).

    Perform the reconciliation. Take reserve bank credit (the flip side of the FOMC’s 1966-69 bank credit proxy proviso), since Sept 2008: & subtract the DFIs loans + investments (i.e., CB credit + S&L credit + MSB credit + CU credit), & subtract changes to bank capital. Then the residual gives you roughly QE’s impact (zippo).

    And if you can’t forecast you have no baseline for any economic analysis.

    This is remedial. I called this historic inflection point too (July 2012).

Leave a Reply