James Bullard on monetary policy in 2008

James Bullard makes some good bullet points in a PowerPoint presentation:

Recessions are dated only long after the fact by an unofficial “NBER dating committee.”

The recession during this period was later dated as beginning in December 2007.

Incidentally, this is 10 months before Lehman-AIG.

However, during 2008 there was a debate as to whether the U.S. was in recession or not.

Based on the data at the time, the outcome of that debate was far from clear.

The real GDP data suggested that Fed easing had mitigated thecrisis up to mid-2008 and that the U.S. had perhaps avoided recession.

As of early August 2008, the growth picture for the U.S. economy according to available real-time data was relatively good.

In particular, estimates of real GDP growth were modest but positive for 2007 Q4, 2008 Q1, and 2008 Q2.
There was no recession according to the conventional definition of two consecutive quarters of negative GDP growth.

As of July 10, 2008, forecasts for the second half of 2008 were for continued modest growth.

According to today’s data, real GDP growth in the first quarter of 2008 was steeply negative, but this information was not available at that time.

There was a good case to be made that the “muddle through” scenario, which had apparently been correct for an entire year, would continue through the end of 2008.

That’s right.  As late as August 2008 both markets and economists thought we would avoid a recession. The collapse occurred in the second half of 2008. Then he makes a point I’ve been trying to get across for 5 years:

My argument is that the economy slowed substantially during the summer of 2008, greatly exacerbating the financial crisis and leading many financial firms to fail.

This slowing of the economy, however, was not readily apparent during the summer of 2008.

This is a crucial point.  Because the collapse of GDP was not known in real time, economists assume the post-Lehman intensification of the crisis was an exogenous shock, and causation went from financial crisis to subsequent recession.  Good to see a top Fed official recognize the reverse causation.

However I don’t like this:

But the rate cuts of early 2008 evidently did little to prevent the ï¬nancial panic, and may have exacerbated the situation to some degree … a point to which I will now turn.

I hate to see policymakers create ad hoc theories that are not consistent with market reactions.  The markets reacted very negatively to the smaller that expected rate cut of December 2007.  Bullard should stick with the tried and true—cutting the fed funds rate creates more AD than not cutting the fed funds rate target.  And more AD makes the crisis smaller.

Nick Rowe has a new post providing a partial defense on Mandel’s claim that a lack of goods innovation is the real AD problem.  I actually had thought of the same possibility as Nick, but didn’t discuss it.  The idea is that less innovation creates saving, which reduces velocity.  The reason I did not discuss this possibility is that if this is what Mandel had in mind, then he should not have responded to Matt Yglesias’s argument as he did.  Yglesias was essentially arguing that too little stimulus is the real problem.  If Nick Rowe’s interpretation of Mandel is correct, then too little policy stimulus is the real problem, as you’d want the Fed to offset any fall in velocity.  But I now regret not discussing that possibility.  It didn’t seem like the issue Mandel was getting at, but if Nick Rowe thought it was, then it is certainly possible I was wrong.

Tyler Cowen links to a blog post by Daniel Davies.  I disagreed with everything in the post, from beginning to end.  But it all seems to boil down to this:

  • So there is a structural shortage of domestic demand

There can’t be a structural shortage of demand, because demand is a nominal concept.  I wasn’t familiar with Mr Davies so I checked a few of his other posts. Yup, he’s a very smart guy, much smarter than me. So take this for what it’s worth.  Demand is fundamentally a nominal concept.  I don’t think any of the factors mentioned by Mandel, Davies, Tyler Cowen or anyone else matter for demand in a world where the inflation target is 5%.  Not even a tiny bit.

They matter a lot for secular RGDP growth stagnation, but not for demand.  And I still don’t think most people get that point.  The real problem is nominal.  Not China, not Germany, not lack of innovation, not income inequality, not debt, not housing, not banking, not deregulation, not current accounts, not fiscal policy, not serial bubbles, not wage stagnation, not anything real.  The real problem is not enough NOMINAL GDP, hence tight money.

(Oddly, I think Paul Krugman agrees with me on this point, but I can’t quite be sure.)

PS.  Don’t worry George; I don’t support a 5% inflation target.

PPS.  Of course there is also the “Great Stagnation.”  But that’s a different problem.

HT:  Vaidas


Tags:

 
 
 

41 Responses to “James Bullard on monetary policy in 2008”

  1. Gravatar of Geoff Geoff
    21. November 2013 at 20:30

    However I don’t like this:

    “But the rate cuts of early 2008 evidently did little to prevent the financial panic, and may have exacerbated the situation to some degree … a point to which I will now turn.

    “I hate to see policymakers create ad hoc theories that are not consistent with market reactions. The markets reacted very negatively to the smaller that expected rate cut of December 2007. Bullard should stick with the tried and true””cutting the fed funds rate creates more AD than not cutting the fed funds rate target. And more AD makes the crisis smaller.”

    This is wrong. Markets “reacting positively” after the rate cut does not refute the argument that lowering interest rates exacerbated the problem. Think of a person who is drunk, but starting to sober up and on the verge of developing a headache/hangover. If he ingested even more alcohol, then his headache/hangover would be averted. Here, more of what causes headaches and hangovers appears to lead to a “positive reaction”.

    For the economy, cheapening borrowed money even further can make it seem like the good times are here again, but in reality are preventing needed corrections from taking place. These corrections can only take place if interest rates are allowed to rise back up.

    Problems caused by too lower interest rates can be exacerbated if the Fed lowers rates even further.

    —————————

    “Because the collapse of GDP was not known in real time, economists assume the post-Lehman intensification of the crisis was an exogenous shock, and causation went from financial crisis to subsequent recession. Good to see a top Fed official recognize the reverse causation.”

    Investment fell before even this. Why? NGDP wasn’t falling at the time. No, NGDP fell after investment fell. There is still a lack of understanding causation.

    —————————–

    “There can’t be a structural shortage of demand, because demand is a nominal concept.”

    True, but there is still a structure demand problem, namely, a problem with relative spending and relative prices, which non-market inflation exacerbates.

    ————————

    “The real problem is nominal. Not China, not Germany, not lack of innovation, not income inequality, not debt, not housing, not banking, not deregulation, not current accounts, not fiscal policy, not serial bubbles, not wage stagnation, not anything real. The real problem is not enough NOMINAL GDP.”

    No, the real problem is structural. Not aggregate demand, not NGDP, not aggregate spending. The real problem is not enough UNHAMPERED ECONOMIC CALCULATION.

  2. Gravatar of Mark A. Sadowski Mark A. Sadowski
    21. November 2013 at 20:38

    “I hate to see policymakers create ad hoc theories that are not consistent with market reactions.”

    I enjoy Bullard’s presentations more than almost any other FRB President. He has good analysis that is well supported by substantive evidence and research, and usually he says something that is highly quotable. But for some reason he always seems to go off the rails at some point. (Must have something to do with that St. Louis freshwater.)

    “Oddly, I think Paul Krugman agrees with me on this point, but I can’t quite be sure.”

    This seems to be a very common sentiment in the econoblogosphere.

  3. Gravatar of kebko kebko
    21. November 2013 at 21:36

    Emergency Unemployment Insurance was first passed on June 30, 2008, so apparently Congress and the President thought we were in a recession.

    The Davies piece made me think of this:

    http://zeezeescorner.tumblr.com/post/23606996850/simpsons-reverse-vampires

  4. Gravatar of jknarr jknarr
    21. November 2013 at 21:41

    I think Davies has one systematic point, one that can be nicely fitted into the Sumner AD framework: debt.

    Debt is like anti-aggregate-demand — can we treat it seriously as such? But, debt is nominal too, and not structural, and is encouraged by tight monetary policy. When money is tight — and interest rates trending lower — debt rushes in.

    All of these stagnation-structure-savings-innovation arguments are simple ad-hoc rationales for keeping the tight-money/high-debt status quo, and de facto pushing it to the next level of wealth confiscation, e.g. Cyprus and negative rates.

  5. Gravatar of Saturos Saturos
    21. November 2013 at 22:41

    But did you believe at the time that monetary policy was very contractionary even at the start of 2008?

  6. Gravatar of Jason Odegaard Jason Odegaard
    21. November 2013 at 23:00

    Hello Scott-
    Off-topic to this post, but do you have any thoughts or opinion on the shake-up happening at the Minneapolis Fed?

    From the local press:
    http://www.startribune.com/business/232695051.html
    Stephen Williamson was quoted by the Star Tribune, so I had to check out his blog:
    http://newmonetarism.blogspot.sg/2013/11/problems-in-great-white-north.html

  7. Gravatar of Erik Erik
    22. November 2013 at 01:23

    This post got me thinking about an old 90s song All Bout the money. Lyrics are very market monetarist 😉

    http://youtu.be/RbXUulY20qI

  8. Gravatar of J.V. Dubois J.V. Dubois
    22. November 2013 at 02:21

    “Oddly, I think Paul Krugman agrees with me on this point, but I can’t quite be sure.”

    Yes I know that feeling. Like when Krugman says how Central Banks should do more and “credibly promise to be irresponsible”, only to wake up and find another article how fiscal austerity in UK caused the recession. Or see him posting good stuff like this one: http://krugman.blogs.nytimes.com/page/2/ where he shows how monetary policy not offseting fiscal austerity is really to blame. But then he produces some incomprehensible post about debt deleveragin or something or other to muddle things up more. So frustrating.

    It sort of reminds me discussion with some libertarians. “Hayek supported gold standard see his quote here. No! He supported NGDP targeting see his quote here!”.

  9. Gravatar of Benjamin Cole Benjamin Cole
    22. November 2013 at 04:46

    If what Bullard says is true (I suspect a wee bit of revisionism), then it calls for much better real-time info in the hands of the Federal Reserve.

    Surely, with so much information tracked continuously and online, the Fed could get a better handle on the economy. I mean, jeez, I read that 7-11 knows how many packets of chewing gum by type it sells by location continuously.

    Real estate services such as Redfin, and Trulia provide amazing real estate information, very timely. Costar is another for the commercial side.

    Train and truck traffic is tracked pretty closely. I assume there is dead on intro on airline flights.

    The second problem I see is that the FOMC is unwieldily, meeting every six weeks, and then in secret. That means if it gets bum info for just a six-week period (that is, bad info at one meeting and the same bad info at the very the next meeting) we see a period of 12 weeks where the Fed flies blind. That is three months.

    Better a Fed Chief who is responsible for policy and reports directly to the President, and is fed lots of data continuously. I wonder has the Fed has taken steps to improve access to economic information? Or is that beneath its purview, sort of getting grubby fingernails?

    There might be valuable proprietary information available also, with a promise of confidentiality. Sales at major retailers etc.

    On the Mandel angle I am amazed at the weakness of the concept. Okay, forget that there has been a ton of new products–ipads and iPhones and PC 3-D games and gadgets for your car.

    Have any of these guys ever been married? The wife always wants a nicer house (you can spend money until you are dead duding up a house) more clothes, hairdos. Second homes, vacations.

    I never met a women yet who did not love going out to restaurants, the more expensive the better.

    Guys like $500 seats at Lakers games, and “girlfriends” who eat money.

    Believe me, Americans will spend money, even if nothing is invented.

    That said, The Fed probably has to goose demand through QE, as people are saving for reasons unrelated to the interest rate. And we are still nearly in ZLB-slo-go land.

    I am beginning to wonder if permanent QE is necessary, and the Fed rolls over on its portfolio. Don’t tell anyone but I am talking about monetizing the debt. Well, if it works, even resort to cardinal sins is not forbidden.

  10. Gravatar of Dan W. Dan W.
    22. November 2013 at 05:27

    I politely ask all readers to visit gasbuddy.com and view the 8-year “quick chart” history of gas prices. In January 2008 gas prices were below $3 a gallon. By July they were over $4 a gallon (a nominal price that has never been higher). Then the price collapsed and by December 2008 gas prices had fallen to below $2 a gallon.

    1) Politically, could the Fed justify any action (ie monetary easing) that could be blamed for pushing gas prices any higher? If you think yes then you work in the corner office of an ivory tower.

    2) What caused the volatility in gas prices in 2008? Was it speculation, manipulation or simply market changes in supply & demand?

    3) What does the non-linearity (ie chaotic) pricing of gasoline in 2008 say about the predictability and manageability of prices throughout the economy? What ability do Central Bankers have to target NGDP if the price of a major commodity in universal demand throughout the economy can vary in price by over 100% in any given year?

    Gasoline was not the only commodity to experience rapid price increases in 2007 – 2008. Grains and metals also spiked in price, only to collapse in late 2008. I submit that the elevated prices of commodities the first half of 2008 made it politically impossible for Central Banks to push the monetary accelerator. This is the real world. Economic models that do not recognize that citizens have limits to how much inflation they will tolerate are toys and should be treated as such.

  11. Gravatar of Matt McOsker Matt McOsker
    22. November 2013 at 05:32

    How did they not know when the Leading Index was dropping all the way along?

    The less hindsight graph
    http://research.stlouisfed.org/fredgraph.png?g=oKp

    Hindsight graph
    http://research.stlouisfed.org/fredgraph.png?g=oKo

  12. Gravatar of ssumner ssumner
    22. November 2013 at 05:50

    Mark, I agree.

    Kebko, The more I look back at the Bush administration, the stranger it seems. They did fiscal stimulus in the spring of 2008. Why? And then the extended benefits in June. Why?

    jknarr, I doubt he’d agree that tight money leads to debt.

    Saturos, I thought it was mildly contractionary until late 2008, when it seemed to have become very contractionary.

    Jason, I tend to agree with the sentiments in Williamson post (diversity of opinion is important), with the proviso that I don’t know any of the facts, so my agreement on this specific case is conditional on Williamson being about about the facts.

    Thanks Erik.

    JV, I feel your pain.

    Ben, Another good reason for futures targeting.

    Dan, You said;

    “Politically, could the Fed justify any action (ie monetary easing) that could be blamed for pushing gas prices any higher?”

    The Fed will be blamed for everything under the sun whatever it does. Look at what people think QE has done! So it might as well do the right thing. If you recall the Fed’s policy of cutting gas prices from $4 to $2 was wildly unpopular, as the economy collapsed.

    Matt, The previous recessions were no different, economists almost never predict a recession until it is (in retrospect) well underway.

  13. Gravatar of Dan W. Dan W.
    22. November 2013 at 06:33

    Scott,

    I agree 100% with you about the knee-jerk blaming of the Fed – they are an ever convenient scapegoat. It is for this reason that we all know when QE will end and when the financial asset bubble will pop. It will be when increases in energy and food prices become a political liability.

    Put simply QE is brought to us by Fracking, technology driven productivity and an under-performing global economy. None of which the Federal Reserve can claim responsibility. But as they say, better to be lucky than good.

  14. Gravatar of Patrick R. Sullivan Patrick R. Sullivan
    22. November 2013 at 08:34

    I do know D-squared, having debated him a number of times. One thing is for certain, he isn’t as smart as HE thinks he is.

  15. Gravatar of Anders Anders
    22. November 2013 at 09:02

    Scott, what if you consider the various sectors of the economy in terms of their desire to run a surplus or deficit (ie net lend/borrow) as a function of their income?

    If the entire private sector tightens its stance – ie increases propensity to net lend at any level of NGDP – then NGDP will clear at a lower level, cet. par.

    It seems to me very likely that, since 2000, the private sector has indeed increased its propensity to net lend, even with loose monetary policy.

    The solution has to be that if the private sector insists on being a net lender, the government has to _embrace_ being a net borrower – ie accept a structural deficit.

    What don’t you like about this analysis? It’s certainly nominal…

  16. Gravatar of ssumner ssumner
    22. November 2013 at 12:10

    Dan, You may know, but I certainly don’t. I don’t even know if there is an asset bubble.

    Patrick, Given his writing style I’m not surprised.

    Anders, Not, that’s a real analysis, not a nominal analysis. To make it nominal you’d have to explain how monetary policy reacts to shifts in the private propensity to lend.

    And you have not explained why we need a public deficit. Why not simply target NGDP?

  17. Gravatar of Mark A. Sadowski Mark A. Sadowski
    22. November 2013 at 13:06

    Scott,
    On the subject of 2008, I finally got around to assembling all of the relevant data series involved in the Fed’s Credit and Liquidity Programs:

    http://research.stlouisfed.org/fred2/graph/?graph_id=147243&category_id=0

    The following is the sum of the amounts in the Fed’s Credit and Liquidity Programs, the monetary base (SBASENS), and the difference in billions of dollars.

    Date——–C&LP-Base-Diff.
    2007-11-01—-0–835–835
    2007-12-01—-4–836–832
    2008-01-01—68–832–764
    2008-02-01—77–830–753
    2008-03-01—70–834–764
    2008-04-01–276–830–554
    2008-05-01–318–837–519
    2008-06-01–324–839–516
    2008-07-01–350–849–500
    2008-08-01–366–849–483
    2008-09-01–420–913–493
    2008-10-01-1110-1113—-3
    2008-11-01-1545-1459-(-86)
    2008-12-01-1687-1670-(-17)
    2009-01-01-1520-1715–195
    2009-02-01-1331-1561–230
    2009-03-01-1293-1639–346
    2009-04-01-1168-1760–592
    2009-05-01–974-1776–802
    2009-06-01–784-1680–896
    2009-07-01–611-1674-1063
    2009-08-01–490-1706-1215
    2009-09-01–444-1798-1354
    2009-10-01–379-1942-1564
    2009-11-01–311-2025-1713
    2009-12-01–250-2024-1774
    2010-01-01–210-1997-1787
    2010-02-01–172-2119-1947
    2010-03-01–153-2076-1923
    2010-04-01–145-2016-1871
    2010-05-01–142-2011-1869
    2010-06-01–139-2000-1860
    2010-07-01–134-1996-1861
    2010-08-01–131-1992-1861
    2010-09-01–121-1960-1839
    2010-10-01–116-1965-1850
    2010-11-01–115-1971-1856
    2010-12-01–112-2015-1903
    2011-01-01–100-2044-1944
    2011-02-01—88-2211-2124
    2011-03-01—85-2396-2312
    2011-04-01—82-2502-2420
    2011-05-01—80-2567-2487
    2011-06-01—75-2646-2571
    2011-07-01—70-2685-2615
    2011-08-01—63-2658-2595
    2011-09-01—60-2635-2576
    2011-10-01—58-2642-2584
    2011-11-01—52-2604-2552
    2011-12-01–100-2619-2519
    2012-01-01–142-2638-2495
    2012-02-01–147-2700-2552
    2012-03-01–100-2652-2552
    2012-04-01—65-2644-2579
    2012-05-01—53-2612-2558
    2012-06-01—46-2620-2574
    2012-07-01—46-2651-2605
    2012-08-01—39-2648-2609
    2012-09-01—23-2595-2572
    2012-10-01—16-2613-2597
    2012-11-01—15-2647-2632
    2012-12-01—14-2673-2660
    2013-01-01—10-2740-2730
    2013-02-01—-7-2841-2834
    2013-03-01—10-2940-2930
    2013-04-01—10-3015-3005
    2013-05-01—-8-3116-3107
    2013-06-01—-3-3197-3194
    2013-07-01—-3-3293-3290
    2013-08-01—-3-3398-3395
    2013-09-01—-2-3488-3486
    2013-10-01—-2-3590-3588

  18. Gravatar of Jim Jim
    22. November 2013 at 15:11

    But the problem IS income inequality and income distribution.

    This leads directly to a demand shortfall, which leads to low nominal GDP.

    It is what it is.

    And on the eve of the 116th meeting, I have to say,

    Beat Cal!!!!!!!!!!!!!!!!!!!!1

  19. Gravatar of kebko kebko
    22. November 2013 at 16:28

    Jim, I have a question. If something happened that meant every household had an income equal to the current top quintile, would that also cause a demand shortfall? If not, what is the difference between that context and the current unequal context that prevents the shortfall?

  20. Gravatar of Jim Jim
    22. November 2013 at 16:42

    kebko, if fiscal policy provided for every household suddenly receiving a PERMANENT tax credit (financed via debt) equivalent to the income of the current top quintile, I’d imagine that (i) we’d have higher inflation, and (ii) we would resolve the demand shortfall.

    Why? Because I believe that the bottom four quintiles have a much higher propensity to consume than the top quintile.

    I’d also appreciate it if someone better versed in Economics than me would attempt to answer your question, kebko.

  21. Gravatar of kebko kebko
    22. November 2013 at 19:03

    To clarify my question, I don’t mean a transfer. I mean some hypothetical development that leads to everyone earning what was, at the start,a top quintile income. Does that lead to a demand shortfall? And, if not, why does a population only partially populated with high income people have a demand shortfall?

  22. Gravatar of Daniel Daniel
    23. November 2013 at 01:06

    income inequality […] leads directly to a demand shortfall

    Does not compute, bro.

    Situation 1 – one guy has 1000$ to spend and another has 100$.

    Situation 2 – one guy has 600$ and the other 500$.

    Are you actually saying that in situation #1 there is less aggregate demand ?

  23. Gravatar of Daniel Daniel
    23. November 2013 at 01:07

    Damn it

  24. Gravatar of ssumner ssumner
    23. November 2013 at 10:26

    Mark, Thanks for the data. Do you notice any interesting patterns?

  25. Gravatar of Secondary Sources: U.S. Optimism, Showrooming, 2008 Policy – Real Time Economics – WSJ Secondary Sources: U.S. Optimism, Showrooming, 2008 Policy - Real Time Economics - WSJ
    23. November 2013 at 11:05

    […] –2008 Policy: Scott Sumner looks at St. Louis Fed President James Bullard‘s explanation of the 2008 crisis. “Because the collapse of GDP was not known in real time, economists assume the post-Lehman intensification of the crisis was an exogenous shock, and causation went from financial crisis to subsequent recession. Good to see a top Fed official recognize the reverse causation. However I don’t like this: “But the rate cuts of early 2008 evidently did little to prevent the financial panic, and may have exacerbated the situation to some degree … a point to which I will now turn,” [Bullard said] I hate to see policymakers create ad hoc theories that are not consistent with market reactions. The markets reacted very negatively to the smaller that expected rate cut of December 2007. Bullard should stick with the tried and true””cutting the fed funds rate creates more AD than not cutting the fed funds rate target. And more AD makes the crisis smaller.” […]

  26. Gravatar of errorr errorr
    23. November 2013 at 16:48

    I don’t think that inequality is relevant to a shortfall inAD although I believe inequality may be important to how the failure of policy distorts the economy and harms future growth. I also believe that monetary policy does not effect all quintiles in some heterogeneous way.

    Inequality is only relevant when and if incomes become permanently stagnant. If Scott is wrong and all these people who want to blame structural issues are right my dislike of many redistributionist policies becomes morally repugnant.

    Of course I believe Scott is mostly correct.

  27. Gravatar of Secstag: All Things to All People? | askblog Secstag: All Things to All People? | askblog
    24. November 2013 at 16:36

    […] Oy. Scott Sumner comments, […]

  28. Gravatar of Mark A. Sadowski Mark A. Sadowski
    25. November 2013 at 13:24

    Scott,
    “Mark, Thanks for the data. Do you notice any interesting patterns?”

    Primarily that the size of the programs was even larger than I remember, and that I was not aware that there was a surge in dollar swaps in early 2012. The latter was probably in response to the Euro Area sovereign debt crisis.

    Until Lehmans filed for bankruptcy on 9/15/2008 the Fed was fully sterilizing its Credit and Liquidity Programs. I saw a recent post where someone was arguing that there was a correlation between the Fed’s consequently declining holdings of securities and declining stock market indicies from December 2007 through September 2008. It was interesting and prodded me into finally putting together the data.

    Note that the difference between the Credit and Liquidity Programs and the monetary base didn’t recover to the levels it had been in November 2007 until June 2009, just as the economy troughed. Coincidence? Probably, but it’s still very interesting, and worth looking into further. And now that the data is all put together that’s possible.

  29. Gravatar of Mark A. Sadowski Mark A. Sadowski
    25. November 2013 at 13:31

    Off Topic.

    Recall back in April and May David Beckworth did a couple of posts on the University of Michigan/Thompson Reuters Survey of Consumers where households are asked how much their dollar (i.e. nominal) family incomes are expected to change over the next 12 months. Well, I’ve been doing a lot of Granger causality tests on my own and in response to discussions in the econblogosphere and one topic that came up in the comment thread at Yichuan Wang’s most recent NoahPinion post was in fact this survey. As a result of that conversation I did some tests and I find that nominal income expectations Granger cause 10-year T-Note yields at the 1% (on what David calls the “without” series, because it excludes the most extreme responses to the survey) and 10% (the “with” series) significance level during the period December 2008 through September 2012. This is of course consistent with term structure theory.

  30. Gravatar of Mark A. Sadowski Mark A. Sadowski
    25. November 2013 at 13:50

    Off Topic.

    I also have some Granger causality test results over the period since December 2008 that I think may be of particular interest to David Glasner, because of his paper that found a correlation between inflation expectations and the S&P 500 index since 2008.

    I find that there is bidirectional Granger causality between the S&P 500 and 5-year inflation expectations as measured by TIPS at the 1% significance level, that the S&P 500 Granger causes the PCEPI at the 5% significance level, that inflation expectations Granger causes Nondefense Capital Goods Excluding Aircraft Industries (ANXAVS) spending at the 5% significance level, and finally that the monetary base Granger causes inflation expectations, the S&P 500 and the PCEPI at the 1%, 5% and 10% significance levels respectively.

  31. Gravatar of Mark A. Sadowski Mark A. Sadowski
    25. November 2013 at 16:21

    Off Topic.

    This past weekend I expanded my Granger causality tests to more thoroughly cover four zero lower bound episodes when the monetary base was significantly expanded (i.e. when there was QE): 1) US from May 1933 to February 1937, 2) US from July 1937 through November 1941, 3) Japan from March 2001 through May 2006 and 5) the UK from April 2009 through September 2013. Note that since Japan only started QE in earnest in April 2013 I have not included this episode yet because the period is too brief. The time periods were selected based on the level of interest rates. The US periods in the 1930s are when the 3-month T-Bill yield was 0.29% or less. The Japanese period was selected based on when the call rate was 0.1% or less. The UK period was selected based on when the bank rate was 0.5% or less. The most recent US episode is when the federal funds rate dropped below 0.25%.

    I shall review the test results for the US over December 2008 through September 2013 and then cover each of the other four periods.

    1) US December 2008 – September 2013

    The monetary base Granger causes loans and leases at commercial banks at the 5% significance level. The M1, M2 and MZM money multipliers each Granger cause loans and leases at commercial banks at the 5% significance levels. These results are the exact opposite of what Accomodative Endogeneity predicts.

    The monetary base Granger causes 5-year inflation expectations as measured by TIPS at the 1% significance level. The monetary base Granger causes the PCEPI, the real broad dollar index, the 10-year T-Note yield (the impulse response is significantly *positive*), the S&P 500 and the DJIA at the 5% significance level. In addition the 10-year T-Note Granger causes the monetary base at the 10% significance level.

    2) US May 1933 – February 1937

    For the US during the 1930s I used the Friedman and Schwartz M1, M2 and M3 money supply measures (M4 was not available).

    The monetary base Granger causes loans at Federal Reserve System member banks at the 5% significance level. The M2 and M3 money multipliers Granger cause loans at Federal Reserve System member banks at the 5% and 10% significance level respectively. This is the exact opposite of what Accomodative Endogeneity predicts.

    In addition the monetary base Granger causes commercial bank deposits at the 5% significance level.

    3) US July 1937 – November 1941

    M1, M2 and M3 Granger cause loans at Federal Reserve System member banks at a 5%, 10% and 5% significance level respectively. This represents a partial rejection of Accomodative Endogeneity.

    However, commercial bank deposits Granger cause loans at Federal Reserve System member banks at the 10% significance level which is the opposite of what one might expect if Accomodative Endogeneity were true (i.e. that “loans create deposits”).

    4) Japan March 2001 – May 2006

    Japan has two measures of lending counterparts to broad money: 1) “loans and discounts” and 2) “loans and discounted bills”. Japan has four measures of broad money: M1, M2, M3 and L.

    There is bidirectional Granger causality between the monetary base and lending (both measures) at the 1% significance level.

    M1 granger causes loans and discounts at the 5% significance level. There is bidirectional Granger causality between M2 and loans and discounts at the 1% significance level. There is bidirectional Granger causality between M3 and loans and discounts at the 1% significance level. Loans and discounts Granger cause the M1 money multiplier at the 5% significance level.

    Loans and discounted bills Granger causes M1 at the 1% significance level. M2 Granger causes loans and discounted bills at the 1% significance level and loans and discounted bills Granger causes M2 at the 5% significance level. There is bidirectional Granger causality between M3 and loans and discounted bills at the 10% significance level. There is bidirectional Granger causality between the M2 money multiplier and loans and discounted bills at the 10% significance level. The M3 money multiplier Granger causes loans and discounted bills at the 10% significance level and loans and discounted bills Granger causes M3 at the 5% significance level. There is bidirectional Granger causality between the L money multiplier and loans and discounted bills at the 10% significance level.

    All of this together is broadly supportive of Structural Endogeneity. However the sum of “deposit money” and “quasi money” (the deposit component of broad money with the sole exception of CDs, which was not available for the entire period) Granger causes both measures of lending at the 1% significance level which is the opposite of what one might expect if Accomodative Endogeneity were true.

    In addition, the monetary base Granger causes the Nikkei 225 at the 10% significance level and there is bidirectional Granger causality between the monetary base and the sum of deposit money and quasi money at the 5% significance level.

    5) UK April 2009 – September 2013

    The UK’s measure of broad money is denoted M4. In addition the BOE also estimates M1, M2 and M3 measures according to Euro Area standards. The lending counterpart of M4 is denoted M4Lx.

    M1 Granger causes M4Lx at the 10% significance level. This represents a partial rejection of Accomodative Endogeneity.

    The monetary base Granger causes 5-year inflation expectations as measured by inflation-linked gilts at the 1% significance level. The monetary base Granger causes the real effective exchange rate of the pound sterling and the industrial production index at the 5% significance level. The monetary base Granger causes the deposit component of the Euro Area measures of broad money at the 10% significance level. In addition the deposit component of M4 Granger causes the monetary base at the 10% significance level.

  32. Gravatar of Mark A. Sadowski Mark A. Sadowski
    25. November 2013 at 18:20

    “The M3 money multiplier Granger causes loans and discounted bills at the 10% significance level and loans and discounted bills Granger causes M3 at the 5% significance level.”

    should read

    “The M3 money multiplier Granger causes loans and discounted bills at the 10% significance level and loans and discounted bills Granger causes the M3 money multiplier at the 5% significance level.”

  33. Gravatar of W. Peden W. Peden
    25. November 2013 at 23:51

    Mark A. Sadowski,

    I also recommend looking at the Retail M4 series (similar to the old M2 in the US, i.e. cash + retail deposits) and the M4x series (which removes the financial holdings of money that have been the cause of much of the instability in the M4 series over the years).

    A similar series for the US to M4x can be obtained by adding large time deposits to M2 and subtracting financial deposits & currency-

    http://research.stlouisfed.org/fred2/graph/fredgraph.png?&id=M2SL_LTDACBM027SBOG_FBTCDTQ027S&scale=Left&range=Custom&cosd=1974-01-01&coed=2013-10-01&line_color=%230000ff&link_values=false&line_style=Solid&mark_type=NONE&mw=4&lw=1&ost=-99999&oet=99999&mma=0&fml=a%2Bb-%28c%2F1000%29&fq=Quarterly&fam=avg&fgst=pc1&transformation=lin_lin_lin&vintage_date=2013-11-26_2013-11-26_2013-11-26&revision_date=2013-11-26_2013-11-26_2013-11-26

    – and then adding financial commercial paper-

    http://research.stlouisfed.org/fred2/graph/?id=M2SL

    The Fed doesn’t measure retail deposits in the same way as the BoE, so comparing M2 with Retail M4 isn’t possible.

    To a large degree, the UK statistics in the post-war period were so disrupted by regulatory distortions (e.g. the infamous “Corset”, which was naturally great at disguising figures) that the best option is generally to just look at the base for any study that goes back past the 1990s.

  34. Gravatar of Mark A. Sadowski Mark A. Sadowski
    26. November 2013 at 13:16

    W. Peden,
    I was aware of the M4ex aggregate, and its lending counterpart M4Lxex, but I neglected to run tests on it in part because the downloadable version of both series only date to July 2009, and in part due to the fact no other central bank computes a similar broad money aggregate (consequently I’m a little skeptical of it).

    As for Retail M4, the BOE doesn’t really advertise it as a separate aggregate although fundamentally it is what the previous M2 aggregate became in 1992.

    However, in response to your comment I ran Granger causality tests on Retail M4, M4ex and M4Lxex. What I find is that Retail M4 Granger causes M4Lx at the 10% significance level, M4 Granger causes M4Lxex at the 1% significance level, M4Lxex Granger causes M4 at the 5% significance level and M4Lxex Granger causes M4ex at the 5% significance level.

    Given my skepticism of M4ex and M4Lxex I tend to discount those results. The Retail M4 result adds further evidence against Accomodative Endogeneity, although strictly speaking it is not an official aggregate.

  35. Gravatar of W. Peden W. Peden
    26. November 2013 at 15:23

    Mark A. Sadowski,

    Thanks! Interesting stuff nonetheless.

    So, aside from the trivialities of double-entry book-keeping, this “loans create deposits” stuff doesn’t have a solid evidential basis?

  36. Gravatar of Mark A. Sadowski Mark A. Sadowski
    26. November 2013 at 16:10

    W. Peden,
    “So, aside from the trivialities of double-entry book-keeping, this “loans create deposits” stuff doesn’t have a solid evidential basis?”

    Let me put it this way.

    Suppose the central bank is targeting the inflation rate (or some macroeconomic quantity) using the unsecured overnight interbank lending rate as its policy instrument. We probably would find that bank lending Granger causes deposits, broad money and the money multiplier, and that bank lending Granger causes the monetary base, as has been found in much of the empirical research on endogenous money.

    But if the unsecured overnight interbank lending rate is fixed at the zero lower bound, and the central bank is increasing the monetary base at ad hoc rates as in QE, then the monetary base effectively becomes the policy instrument, and we may find that these causal relationships would reverse.

    In short, the direction of causality very much depends on how the central bank chooses to conduct monetary policy. It is not a natural law as many endogenous money enthusiasts imagine it is.

  37. Gravatar of W. Peden W. Peden
    26. November 2013 at 16:23

    Mark A. Sadowski,

    Even more interesting. It would be an exaggeration to call the current situation “monetary base control”, but it’s certainly interesting how QE had shaken matters up.

  38. Gravatar of ssumner ssumner
    29. November 2013 at 08:44

    Mark, Doesn’t a finding that macro variables Granger cause financial asset prices imply the EMH does not hold?

  39. Gravatar of Mark A. Sadowski Mark A. Sadowski
    29. November 2013 at 10:43

    Scott,
    Yes, if markets knew in real time what the values of the macro variables were, and that they knew those variables were correlated with those outcomes in financial asset prices.

  40. Gravatar of ssumner ssumner
    29. November 2013 at 12:13

    Yes, I forget about the data lag. That makes sense.

  41. Gravatar of Mark A. Sadowski Mark A. Sadowski
    29. November 2013 at 12:25

    For what’s worth I found an econometric mistake with respect to the results derived in response to W. Peden’s comments.

    “M4 Granger causes M4Lxex at the 1% significance level, M4Lxex Granger causes M4 at the 5% significance level.”

    should be changed to

    “M4 Granger causes M4Lxex at the 10% significance level.”

Leave a Reply