J’accuse

Apologies to Emile Zola fans for the sophomoric title, but I wanted to get people’s attention as this is important.  In the past I had sort of given the Fed a pass on its behavior in the 3rd quarter of 2008, partly because we didn’t get a major stock market crash until the first 10 days of October, and partly because I hadn’t looked closely at the data.  A couple months back I looked at some graphs and realized that policy was already drifting off course in the third quarter.  I don’t know why it took me so long to look at daily data, but when I did so yesterday I was shocked by what I saw.  There are no excuses for the Fed’s behavior.  Last September they had all the information they needed to act decisively, and blew it.

At midyear the commodity boom was reaching a peak, and it was expected that headline CPI numbers would be quite high for the next few months.  Not surprisingly this near term inflation was priced into TIPS, with TIPS spreads near their cyclical peak:

Date       5 year TIPS yield    5 year T-note yield   5 year TIPS spread

7/1/08          0.67%                        3.33%                        2.66%

This was the period of maximum concern about inflation.  The ECB raised its target rate to 4.25% in early July, and I recall there was pressure on the Fed to do something about inflation.  They resisted, but the lingering effects of this pressure caused great mischief a few months later.  Then commodity prices began falling and inflation concerns began to wane:

Date       5 year TIPS yield    5 year T-note yield   5 year TIPS spread

8/1/08          1.06%                        3.23%                        2.17%

8/15/08        1.16%                        3.11%                        1.95%

9/2/08          1.28%                        3.00%                        1.72%

9/12/08        1.47%                        2.97%                        1.50%

Note how inflation expectations fell steadily as the US economy weakened dramatically in August and early September.  The first data announcement showing this weakness occurred on September 15:

The 1.1 percent decrease in production at factories, mines and utilities was more than forecast, Federal Reserve figures showed today. Car output slumped 12 percent, the most in a decade, and declines ranged from semiconductors to building supplies.

Today’s report indicates the domestic slump is pulling down a U.S. manufacturing industry that has been buttressed by record exports. The figures may stoke concern the economic downturn will deepen amid a housing recession, rising unemployment and a credit crunch that today sent Lehman Brothers Holdings Inc. into bankruptcy.

“We’re seeing more pervasive weakness in manufacturing, going beyond autos,” said Sal Guatieri, a senior economist at BMO Capital Markets in Toronto. “The economy is taking a lurch downward. This is another report that points to a recession.”

At the time the US economy had been in recession for 9 months, but the Fed did not yet know that.  The August numbers showed the recession spreading from housing to manufacturing, which until then had been supported by strong exports.  After the September 15th IP announcement, there can have been no doubt that recession risks were increasing sharply.  And as if that wasn’t enough, the bankruptcy of Lehman was announced on the previous day (Sunday.)  Stocks fell sharply on Monday the 15th.

With inflation expectations well below the Fed’s target, and with dramatically increasing risks of recession, the Fed had every reason to move aggressively on the 16th.  We clearly needed much higher NGDP growth expectations.  And any doubts about the need for stimulus were erased by the bond market’s behavior on September 15th, when 5-year inflation expectations plunged to the shockingly low level of 1.23%.  This is not the sort of number you would get if investors expected 5% NGDP growth over 5 years.

How did the Fed react to the rapidly intensifying recession, and the sharp drop in inflation expectations?  They issued a statement arguing that the risk of recession was roughly balanced by the risk of inflation.  Not a risk of excessively low inflation, but rather by risk of excessively high inflation.  Read it and weep:

Release Date: September 16, 2008
For immediate release

The Federal Open Market Committee decided today to keep its target for the federal funds rate at 2 percent.

Strains in financial markets have increased significantly and labor markets have weakened further. Economic growth appears to have slowed recently, partly reflecting a softening of household spending. Tight credit conditions, the ongoing housing contraction, and some slowing in export growth are likely to weigh on economic growth over the next few quarters. Over time, the substantial easing of monetary policy, combined with ongoing measures to foster market liquidity, should help to promote moderate economic growth.

Inflation has been high, spurred by the earlier increases in the prices of energy and some other commodities. The Committee expects inflation to moderate later this year and next year, but the inflation outlook remains highly uncertain.

The downside risks to growth and the upside risks to inflation are both of significant concern to the Committee. The Committee will monitor economic and financial developments carefully and will act as needed to promote sustainable economic growth and price stability.

Voting for the FOMC monetary policy action were: Ben S. Bernanke, Chairman; Christine M. Cumming; Elizabeth A. Duke; Richard W. Fisher; Donald L. Kohn; Randall S. Kroszner; Sandra Pianalto; Charles I. Plosser; Gary H. Stern; and Kevin M. Warsh. Ms. Cumming voted as the alternate for Timothy F. Geithner.

People often attack my futures targeting ideas by pointing to market excesses.  “Ha! You want to turn monetary policy over to the markets, the same people who brought us the tech bubble and housing bubble.”  But is there any good evidence for this attitude?  If we could go back to September 16th and do it all over again, where should we set the target fed funds rate?  Should we rely on the Fed’s approach, which is basically like driving down the road while looking in the rear view mirror for signs that you might be drifting off course?  Or should we use forward-looking indicators like the TIPS spread?  Yes, this is just one example, but the stock and bond markets were consistently ahead of the Fed throughout this entire crisis, from December 2007 forward.

We don’t have NGDP futures (another inexcusable fact, as the government could create a NGDP futures market at very low cost—but that’s another post), so we will have to make do with these flawed TIPS spreads.  Yet even with all of their flaws, TIPS spreads were clearly indicating that the risks of recession were growing dramatically, and inflation risks were rapidly receding.  The Fed ignored those warnings.

The September 16, 2008 Fed statement is not an error of omission, it is simply an error.  They sent out a press release that suggested they were clueless about the state of the economy, even though they access to all the data presented here.  When you are steering a giant oil tanker there are no “errors of omission,” rather it is simply a question of whether you did your job well or poorly.  The Fed doesn’t have to issue a statement indicating that risks or recession and inflation are roughly equal.  That is an affirmative step on their part, completely disregarding the message from the markets.  And the markets weren’t pleased:

Date       5 year TIPS yield    5 year T-note yield   5 year TIPS spread

9/15/08          1.36%                        2.59%                        1.23%

9/16/08          1.61%                        2.64%                        1.03%

It doesn’t seem like it could get any worse, but in fact it got much worse before it got better.  The Fed continued to ignore the need for a more expansionary monetary policy, focusing instead on policies to bail out the banking system.  They had misdiagnosed the situation as one where banking weakness was causing economic weakness, whereas it was actually the other way around (just as in the early 1930s.)  If we hadn’t already been in recession for 9 months, Lehman probably wouldn’t have failed.

The S&P fell 200 points in the week of October 6, roughly 18%.  It is difficult to know what caused this crash, but it was probably related to perceptions of falling AD, as there were no major financial news stories, but lots of scary stories about AD plunging all over the world.  On Monday, October 6th, the Fed made an even bigger error than on September 16th, for the first time in their 95-year history they began paying interest on bank reserves.  Inflation expectations had been drifting lower before the October 6th announcement, but simply collapsed in the days afterward:

Date       5 year TIPS yield    5 year T-note yield   5 year TIPS spread

10/1/08          1.92%                        2.87%                        0.95%

10/2/08          1.75%                        2.68%                        0.93%

10/3/08          1.68%                        2.64%                        0.96%

10/6/08          1.70%                        2.45%                        0.75%

10/7/08          1.83%                        2.45%                        0.62%

10/8/08          2.27%                        2.70%                        0.43%

10/9/08          2.32%                        2.79%                        0.47%

10/10/08        2.57%                        2.77%                        0.20%

Normally markets will react immediately to news, but the interest on reserves policy was so bizarre, so far from anything traders were familiar with, that it may have taken a few days to sink in.  Even in late October one can find economists implying the decision should have an expansionary impact:

The rate reduction co-incides with a Fed rule change that allows it to pay interest on banks’ excess reserves – a change that experts say gives Fed Chairman Ben Bernanke the ability to provide almost unlimited liquidity to the US banking system. This October one-two combination will be a powerful lever for the economy, economists say, perhaps enough to finally begin to relieve some of the pressures on credit markets.

“The Fed is pushing on the accelerator as hard as it can,” says Mark Zandi, chief economist at Moody’s Economy.com. “By paying interest rates on reserves, the Fed is able to provide so much cash to financial institutions that there will be no reason for them not to lend to one another and, by extension, to business and households.”

Bond traders had to figure this out for themselves, as they weren’t getting much help from economists.  And in fairness, it was hard to believe that the Fed would intentionally adopt a contractionary policy in the midst of an economic/financial hurricane.  I almost didn’t trust my own initial reaction (which was incredulity), until I saw it confirmed by people like James Hamilton and Robert Hall.

As I pointed out in an earlier post, it wasn’t just TIPS spreads, beginning in July or August lots of other indicators were also signaling tight money:

1.  Higher real interest rates.

2.  A strongly appreciating dollar.

3.  Sharply falling industrial production

4.  Housing price drops in formerly stable markets

5.  Severe stock price declines

6.  Severe commodity price declines

But the TIPS spreads are probably the most damning.  Admittedly the data isn’t perfect, I am not certain if the real interest rate figures were distorted by the way TIPS are indexed (if there is a lag, measured real rates will rise when falling commodity prices signal lower CPI numbers in the near future.) And the spread may reflect liquidity concerns.  But none of the issues people have raised would change the fundamental story.  If the spreads were reflecting liquidity issues, then that would also be a signal that monetary ease was needed.

At this point one might ask if I am just playing Monday morning quarterback.  Even if the Fed’s decisions of September 16 and October 6 look bad in retrospect; isn’t it unfair to criticize them harshly?  After all they did not know how bad things were going to get.  But all I am doing is holding them to their own standards.  In a December 2007 speech Ben Bernanke stated that Fed policy should be forward looking:

“The [macroeconomic] projections also function as a plan for policy””albeit as a rough and highly provisional one.  As I mentioned earlier, FOMC participants will continue to base their projections on the assumption of ‘appropriate’ monetary policy.  Consequently, the extended projections will provide a sense of the economic trajectory that Committee participants see as best fulfilling the Federal Reserve’s dual mandate, given the initial conditions and the constraints imposed by the structure of the economy.”  (Italics added.)

In other words, an “appropriate” monetary policy is one that equates the Fed’s long run forecast with its policy goals, and specifically the dual mandate given by Congress.  In the long run the Fed cannot influence the real growth rate of the economy; these are the “constraints imposed by the structure of the economy” that he refers to.  In the short run, however, some monetary stimulus may be appropriate when output is expected to be below potential.

The primary long run goal is low inflation, with 2% believed to be the implicit target.  Because the Fed can control inflation in the long run, they should also normally forecast a roughly 2% long run inflation rate.  According to the famous “Taylor rule,” the only time a below 2% inflation forecast would be appropriate is if output is expected to be well above potential.  In the fall of 2008 the real economy was clearly weakening, and yet the Fed also allowed long term inflation forecasts to fall far below 2%.  To use Bernanke’s terminology, monetary policy was “inappropriate.”  I would use more colorful language.

It’s not clear why Bernanke abandoned the Svenssonian idea of targeting the forecast.  Perhaps he thought that the financial crisis had to be dealt with first.  On October 15th Bernanke gave a speech where he presented a very bleak picture of the macroeconomic outlook.  It would have been nice if he had said that the Fed would do everything possible to get NGDP growth back up to 5% in the near future, in the hope that faster growth would help ease the financial crisis.  Instead he got things backward, suggesting that unless the financial crisis could be fixed, there was little prospect for higher growth.  In other words, he basically gave up on monetary stimulus, and argued that the financial problem had to be fixed first.

“Ultimately, the trajectory of economic activity beyond the next few quarters will depend greatly on the extent to which financial and credit markets return to more normal functioning.”

That was not what the markets wanted to hear.  Bernanke may not have been aware of this fact, but the markets viewed his forecasts as more than just forecasts, but also implied policy statements.  If as Fed chairman you have been talking about targeting the forecast, and your forecast is extremely bleak, don’t be surprised if the market thinks your policy target might be aimed a bit too low:

U.S. stock markets, which were already down when Bernanke began his midday speech, headed lower after the Fed chairman concluded his remarks.

“It’s not like he said anything we didn’t know. But the fact that he’s the Fed chairman makes it carry more weight,” said Robert Brusca of FAO Economics.

Stocks fell 9% that day, although as indicated they were already falling before the Bernanke speech.  After the rough week of October 6, TIPS spreads had leveled off for a few days.  But after the Bernanke speech they headed lower again, and by the 16th were actually negative!  Once again, liquidity concerns may have exaggerated this downward shift, but liquidity concerns also pointed to the need for monetary ease.

Date       5 year TIPS yield    5 year T-note yield   5 year TIPS spread

10/10/08        2.57%                        2.77%                        0.20%

10/14/08        2.77%                        3.01%                        0.24%

10/16/08        2.88%                        2.84%                       -0.04%

10/24/08        3.44%                        2.64%                       -0.80%

11/26/08        4.24%                        2.01%                       -2.23%

The economy continued to deteriorate in November.  By November 26 the 5-year TIPS spread peaked at an astounding negative 2.23%.  Real output was in free fall all over the world.  And yet the Fed still had not even exhausted its conventional monetary ammunition.  The fed funds target was still 100 basis points above zero.   Indeed the Fed did not make a single rate cut in the entire month of November.  What were they waiting for?  I have no idea.  I know of no mainstream macro model that didn’t suggest money was far too tight.   It wasn’t until mid-December that the Fed woke up to the scale of the crisis and began to move a bit more aggressively.  I’ll tell that story in a subsequent post.

To be continued . . .

[PS:  Although I frequently refer to ‘Bernanke,’ this isn’t personal.  I doubt any plausible Fed chairman would have done better.  The problem extends to the Fed as an institution, the ECB, BOJ, and indeed, entire field of monetary economics.  All are too backward-looking.  I plan a post on this problem in the near future.]


Tags:

 
 
 

35 Responses to “J’accuse”

  1. Gravatar of Marc Marc
    19. June 2009 at 04:14

    Your RSS has been hacked. I can post the RSS feed received for this post, probably a rule setting (Viagra mentioned many times)

  2. Gravatar of Alex Alex
    19. June 2009 at 04:30

    Scott,

    It is time to realize that the Fed never intended to stimulate the economy through monetary policy. Their only concern was to avoid a major run on the financial system. They injected all the liquidity banks needed to keep running but they ever had the intention for that money to leave the banks as loans to the private sector. This is consistent with the view that the Fed (no matter what the dual mandate says) is not here to actively smooth the cycle but as a lender of last resort. It is not about whether the Fed saw it coming or not, if their forecast was to high or to low, if they are looking at the right markets or prices, it is about what we want the Fed to do.

    Alex.

  3. Gravatar of ssumner ssumner
    19. June 2009 at 04:40

    Marc, Thanks, That’s been a recurring problem. I’ll look into it.

    Alex, The Fed did try to stabilize the economy from 1982 to 2007, so if you are right then it represents a sudden policy change. I don’t see any evidence that they don’t care about the business cycle.

    The role you describe was what the Fed did in the early 1930s—lender of last resort. It gave up on the real bills doctrine after the Great Depression.

  4. Gravatar of Daniel Daniel
    19. June 2009 at 05:55

    I think you have some bad data there. The 5-year TIPs real yield was nowhere near 4% in November 2008. Bloomberg has it at 2.30% on 11/26, not 4.24% as you have it.

  5. Gravatar of Thruth Thruth
    19. June 2009 at 06:31

    I’ve seen a few economists describe the events surrounding Lehman’s failure as a “liquidity event”, presumably to distinguish it from true deflationary money shocks. Is it really any different from a monetary shock? Looks like a duck, quacks like a duck and all… (I think this goes back to some of the previous discussions we’ve been having about the relevance of modeling the micro of banks and bank runs).

    Note also that Cleveland Fed stopped publishing its series on inflationary expectations about the time of the liquidity event.

  6. Gravatar of ssumner ssumner
    19. June 2009 at 06:42

    Daniel, I used St Louis Fred data (that’s not a typo, by the way).

    http://research.stlouisfed.org/fred2/series/DFII5?cid=82

    They have a lot of time series for various TIPS, and they all showed the high real rates in November. Can you provide a link to the Bloomberg data?

    I actual find your number much more plausible, as a -2.23% TIPS spread seemed very fishy to me, so I hope you are right. But I can’t find it. Does anyone know why Bloomberg and St. Louis differ so much?

    Thruth, Yes, I heard about the Cleveland Fed dropping the series. Does anyone know why? (Half jokingly I want to say because it conflicted with Fed policy, but I’m not that cynical.)

    And yes, it most definitely was a duck, as the macro data have since then clearly shown. I think the point you raised gets to the heart of the misdiagnosis that I keep referring to.

  7. Gravatar of D. Watson D. Watson
    19. June 2009 at 06:46

    Are you therefore changing your mind about your last post? I’m having trouble reconciling the two otherwise.

  8. Gravatar of Alex Alex
    19. June 2009 at 06:46

    Daniel,

    My guess is that the data is from the St. Louis Fed and that Scott is using the 5 year constant maturity http://research.stlouisfed.org/fred2/data/DFII5.txt
    If you are looking at the same series but from another source be careful since as you can see in the linked attached the yield drops from 4.24% in 11/26 to 2.03 in 12/01. This is because the Fed changed the composition of tips used to compute the yield. Newly issued tips protect you against inflation an deflation hence they have a lower yield than old tips. Maybe you are looking at a new tip in the Bloomberg screen.

  9. Gravatar of JimP JimP
    19. June 2009 at 07:02

    And all this is why I do blame Bernanke – and bitterly so – and continue to do so today. Millions of families destroyed – for no reason at all. It remains, to me, utterly stunning.

    And now we have Christina Romer in there working for Obama. To lift a quote about her from this article:

    http://www.forbes.com/2009/06/18/paul-krugman-new-york-times-liquidity-romer-opinions-contributors-alan-reynolds.html

    Start of quote: On the contrary, Christina Romer’s research clearly demonstrates that strong rebound of 1934-37 was “helped along” by a 42% increase in the money supply. She found, “monetary developments were very important and fiscal policy was of little consequence … Even in 1942, the year that the economy returned to its trend path, the effects of fiscal policy were small.” End of quote.

    Indeed – she KNOWS that monetary policy will work, and she KNOWS that fiscal policy will not work – yet there she is – on tv – shucking and grinning for the Obama fiscal run-away train.

    Romer and Bernanke both – it is gross and incomprehensible incompetence. And we are going to pay for it – with the stagnation Krugmam so devoutly wishes for. Unless we vote these idiots out.

  10. Gravatar of ssumner ssumner
    19. June 2009 at 07:04

    D. Watson. I know it seems that way, but the answer is no. In both cases I want the Fed to rely on consensus market forecasts, I don’t expect the Fed to see problems the markets don’t see. So there is no inconsistency.

    Alex, That might explain it, the Bloomberg numbers could have been for new TIPS. Because TIPS can’t fall below par, it is hard to get an accurate reading of deflation expectations. You need to use an older 10 or 20 year TIPS, with 5 years left to maturity. Those built up a lot of capital gains, and need high real interest rates in late 2008 to offset expected capital losses from near term deflation as commodity prices crashed. But the -2.23% figure still seems weird.

  11. Gravatar of Aaron Jackson Aaron Jackson
    19. June 2009 at 07:38

    Scott,
    Probably one of your top five posts (if not number one)! Nice job!

  12. Gravatar of ssumner ssumner
    19. June 2009 at 09:21

    JimP, I guess people are under pressure to toe the party line. I had a post a few months ago entitled “President Obama, Talk to Christy.”

    Aaron, Mucho gracias.

  13. Gravatar of Nick Nick
    19. June 2009 at 09:33

    I’d be careful extracting inflation data from TIPS during that time period (and I suspect the Fed was as well). The TIPS market isn’t very liquid from an institutional perspective. TIPS became almost like a credit product in the middle of the crisis. Investors needed safety and liquidity. TIPS only provided one of those. If you’ve got a big position in a government guaranteed 10 year bond that you can’t sell, it doesn’t do you much good. There were a whole lot of ‘money-good’ bonds trading at higher and higher yields during this time period. That being said, I think generally your point is correct, but the magnitude of how obvious it was is overstated.

  14. Gravatar of ssumner ssumner
    19. June 2009 at 16:32

    Nick, Good point. This paper discusses the issue:

    http://www.econ.jhu.edu/People/Wright/tips_brookings_wright_discussion.pdf

    Wright says that transactions costs are pretty low, but the market may be somewhat segmented, for reasons that are a bit unclear. So you might be right. I am a bit confused by one conclusion in the paper, and perhaps someone can fill me in. Didn’t the CPI rise about 3% between April and August 2008? And if so, then why couldn’t the 5 year TIPS due April 2013 have priced in some risk of deflation in the fall of 2008. Isn’t Wright essentially assuming that those TIPS would have no risk of any decline? I.e., he’s treating them like they were issued in August. The actual anomaly he finds is pretty small, the yield only rose about 1/2% above nominal bonds, and is now back below. So I don’t see how that is an obvious anomaly, although it certainly might be.

    I should say that one reason I have confidence in this conclusion is that I am not just relying on the TIPS, but also a lot of other information. By November Fed policy was “inappropriate” by every single indicator I could find. All the markets (not just TIPS) were suggesting that the forecast was below the target. Every economist I talked to felt that way. The consensus forecast was way below the target, and so on.

    Still, I would like to know how much liquidity affects the TIPS spread. If it is a big problem, then it’s one more reason why we need a NGDP market, which probably would not be biased by liquidity concerns, as the Fed would buy and sell unlimited NGDP futures at the target price.

    One last point, with the TIPS spread down to 1.23%, and commodity prices in steep decline for 2 months, isn’t it a bit odd that the Fed was more worried about high inflation than low inflation on September 16th? Even if the TIPS spread isn’t perfect, what reason did they have to disregard it?

    BTW, Like Wright, I am perplexed by the market reaction to the March QE announcement. I don’t see why long term nominal rates fell. But then they rose later, so maybe the markets simply made a mistake.

  15. Gravatar of The fatal conceit of 100% reserves « Entitled to an Opinion The fatal conceit of 100% reserves « Entitled to an Opinion
    19. June 2009 at 18:25

    […] a weird coincidence: What are the odds that AidWatch and The Money Illusion would both have posts with the same French title at the top of their front page (not anymore, but a […]

  16. Gravatar of Current Current
    20. June 2009 at 04:18

    Why does it take all this elaborate work to target NGDP?

    Let’s say MV=PQ and Q is GDP. This is an accounting tautology.

    The central bank knows M, it can find that from the commercial banks who lodge their accounts with the central bank very regularly. Can’t V be estimated similarly by data from the commercial banks?

    Then once you have MV you can add a bit if you like.

  17. Gravatar of ssumner ssumner
    20. June 2009 at 09:46

    Current, No, V cannot be estimated from the commercial banks. It can only be estimated by dividing NGDP by M. But NGDP is observed with a lag of several months, so we never know the current level of velocity. That’s one reason we need NGDP futures markets, to “predict the present.”

  18. Gravatar of Current Current
    22. June 2009 at 03:23

    I don’t really see why you need to work that way. What it is really necessary to know is not V but rather the demand to hold money. It is the increases in that which cause the problem of secondary recession not the change in V it brings about. It seems to me that this can be found by looking at the books of the commercial banks.

  19. Gravatar of ssumner ssumner
    22. June 2009 at 06:10

    Current, What matters is the demand for money in real terms, or as a percentage of NGDP (velocity). In either case you need macroeconomic data. By the way, most base money is normally held by the public, as cash. The Fed knows exactly how much base money is in circulation, and exactly how much is held by banks. Thus they know cash. But they need to know the real, not the nominal amount of cash in circulation. If the nominal amount is unchanged and the real demand doubles, you get a depression. The only way you can see this problem develop is if you notice the price level falling. The money supply may not change at all.

  20. Gravatar of Current Current
    22. June 2009 at 08:14

    The aim here is to offset a secondary recession. It is to prevent what Knut Wicksell and Steve Horowitz call “Cumulative rot”. The situation where people wish to hold more money as a hedge against uncertainty, but the only way they can do it is by cutting their spending. Those cuts in spending cause recession to spread elsewhere.

    I agree that “If the nominal amount (of cash) is unchanged and the real demand doubles, you get a depression.” It is the increase in the demand for money that needs to be satisfied.

    By aggregating the problem you are looking at it in a way that is more complicated than necessary. A fall in the price level is one sign that there is a problem. It may be that people are not buying because they wish to hold more money, causing a drop in prices. However, why can’t this be observed more directly. If people wish to hold more money then in the short-run they will succeed by cutting expenditure. There will then be a decline in what Selgin calls the turn-over rate of bank money. Why not just observe this?

    I suppose it might be quite difficult to see for cash, but I think it could be done for bank accounts.

  21. Gravatar of 123 123
    22. June 2009 at 12:46

    Great post. It is 100% true that decisive action was requred in September right after Lehman.

  22. Gravatar of ssumner ssumner
    23. June 2009 at 06:09

    Current, The monetary base is nearly 50% cash today, as is M1. In addition, velocity is not the rate at which money turns over, velocity is PY/M. A lot of money turnover is not to buy goods and services. So measuring the rate at which money turns over doesn’t help you know when NGDP is falling, which is what you want to know. There is no shortcut to measuring NGDP itself. I see what you are getting at, but it won’t work in practice.

    Thanks 123.

  23. Gravatar of 123 123
    23. June 2009 at 12:46

    Scott, have you done a similiar analysis about other important turning points? For example in 2000 the policy was arguably too tight – it would be interesting to check if TIPS market indicated that.

  24. Gravatar of ssumner ssumner
    24. June 2009 at 04:42

    123, I don’t know what the spreads were showing. BTW, I don’t think money was too tight in 2000. I think it was slightly too tight in 2001, but I’m not sure if that would show up in TIPS spreads. I wish we had a NGDP futures markets.

    I do know that the stock market rallied strongly on the January 2001 larger than expected Fed rate cut. This suggests the market was (correctly) worried about recession. So that’s an interesting tidbit. If I could go back in time I might not do much different until the spring of 2001, when I would have eased a bit more aggressively.

  25. Gravatar of Current Current
    24. June 2009 at 08:57

    Scott: “The monetary base is nearly 50% cash today, as is M1”

    In the US it is. I think it’s ~15% in other developed countries. The monetary base is so large in the US because dollar notes are used widely outside the US. In the drug trade for example. Is it necessary to worry much about the base money held abroad?

    Scott: “In addition, velocity is not the rate at which money turns over, velocity is PY/M. A lot of money turnover is not to buy goods and services. So measuring the rate at which money turns over doesn’t help you know when NGDP is falling, which is what you want to know. There is no shortcut to measuring NGDP itself.”

    What sort of money turnover are you thinking of that is not buying goods and services? Gifts?

  26. Gravatar of ssumner ssumner
    25. June 2009 at 05:13

    Current, Money can turn over to by goods at a garage sale, or to make mortgage payments, or to buy stocks, or bonds, or existing homes. Those aren’t NGDP.

    No, the 50% is unusually low. The base is normally 90% cash. It is lower today because of the high levels of ERs. The US cash to GDP ratio is not unusually large (compared to most other countries), despite the drug trade.

  27. Gravatar of Current Current
    25. June 2009 at 08:19

    Scott: “Money can turn over to by goods at a garage sale, or to make mortgage payments, or to buy stocks, or bonds, or existing homes. Those aren’t NGDP.”

    Why do you want to take those things out though?

    (Actually I’m curious about this and I asked at Austrian Economists the opposite question. They’re debating something very similar).

    Scott: “No, the 50% is unusually low. The base is normally 90% cash. It is lower today because of the high levels of ERs. The US cash to GDP ratio is not unusually large (compared to most other countries), despite the drug trade.”

    Look at the graphs on the Wikipedia page on Money Supply.
    http://en.wikipedia.org/wiki/Money_supply

  28. Gravatar of ssumner ssumner
    26. June 2009 at 06:15

    Current, I take them out because I am interested in explaining NGDP. I am not interested in transactions. Why should I care about that?

    Nothing in those graphs conflicts with my point about the money supply.

  29. Gravatar of Current Current
    26. June 2009 at 06:49

    Scott: “I take them out because I am interested in explaining NGDP. I am not interested in transactions. Why should I care about that?”

    I don’t understand that line of reasoning. Isn’t the question we often discuss “what monetary policy is best”? In that case aren’t transactions important as well as GDP?

    Besides, how can the relative prices of existing capital and portions of GDP be dealt with differently from the money side?

    Even in a barter economy it would be possible for the price of GDP goods to fall and the price of existing capital goods to rise.

    Scott: “Nothing in those graphs conflicts with my point about the money supply.”

    We were discussing the base amount of cash. Look at the size of the monetary base versus M1 for the European Union.

  30. Gravatar of Lorenzo (from downunder) Lorenzo (from downunder)
    26. June 2009 at 21:26

    At the time the US economy had been in recession for 9 months, but the Fed did not yet know that.
    Is there are a more fundamental question. Can anyone do monetary policy well enough as an act of official discretion?

    A monetary economist might say “yes, if they follow my preferred policy”. But there is clearly no consensus, not even, as you have pointed out, about the language in which to analyse matters to find the current policy.

    Having looked at how official discretions operate elsewhere in the economy, I am deeply sceptical about them. Why is monetary policy different?

  31. Gravatar of Lorenzo (from downunder) Lorenzo (from downunder)
    26. June 2009 at 21:27

    That should say “to find the correct policy”

  32. Gravatar of ssumner ssumner
    27. June 2009 at 10:07

    Current, Yes, now I see your point. cash is a small share of M1 in Europe. My point was that in developed countries cash is a pretty big shar of GDP, around 5%. In that area the US is not unusual. It looks like M1 is much bigger i nEurope than the US.

    On transactions: Suppose NGDP falls from 14 trillion to 10 trillion, and forex transactions rise from 20 trillion to 30 trillion. I say we are in a depression, but if you looked at total transactions they’d be way up. That’s why I think NGDP is more informative,

    Lorenzo, Like you I am increasingly skeptical of discretion. I now favor having the market set the monetary base, as I don’t trust the Fed anymore. That’s why I favor NGDP futures targeting.

  33. Gravatar of Current Current
    29. June 2009 at 09:49

    Scott: “Yes, now I see your point. cash is a small share of M1 in Europe.”

    The New Zealand and Australia graphs though are similar to the Euro graph in terms of the base:M1 ratio. I see what you mean about the US though.

    Scott: “Suppose NGDP falls from 14 trillion to 10 trillion, and forex transactions rise from 20 trillion to 30 trillion. I say we are in a depression, but if you looked at total transactions they’d be way up. That’s why I think NGDP is more informative”

    But, can monetary policy really do anything about that case? How could an increase in money supply change what people want to do with that money in the “right” direction?

    When I mentioned this on the Austrian economics blog George Selgin wrote:
    “Demand for money is demand to hold money, not to spend it on this or that. (When money is spent, the demand for it falls, while the demand for the stuff it’s being spent on rises.) This distinction between the demand for money balances to hold on to (genuine demand for money) and the demand for money for immediate spending (pseudo demand) is a crucial aspect of monetary economics.”
    http://austrianeconomists.typepad.com/weblog/2009/06/what-monetary-policy-cannot-do/comments/page/2/#comments

    That seems to make sense to me. The fall in NGDP you describe above could happen in a barter economy.

    Do you have a link to a paper explaining your own version of monetary equilibrium theory?

  34. Gravatar of ssumner ssumner
    1. July 2009 at 05:15

    Current, Unfortunately I am not well organized, but I am pretty sure my post entitled “A short Course in Monetary Economics” is in the March or April archives (if not, then May.) I think you could find it in a few minutes.

    I’ll just take issue with one thing you said:

    “But, can monetary policy really do anything about that case? How could an increase in money supply change what people want to do with that money in the “right” direction?”

    Thus confuses relative and absolute price changes. Monetary policy cannot control relative prices, nor should it try. Suppose you expanded the money supply 40% to reflate the economy, and suppose NGDP went up 40%. Then GDP would rise from 10 back up to 14, and forex transactions would rise from 30 to 42. The NGDP rise is exactly what you want, and you don’t care about forex transactions. To summarize, you target the aggregate that best provides macro stability, and let other variables go where the market takes them.

  35. Gravatar of Current Current
    1. July 2009 at 09:14

    I read your post “A Short Course in Monetary Economics”. And I read your reply about GDP and other transactions.

    I don’t really agree with them. However, it seems like the world and his wife are arguing with you on this blog right now. So, rather than arguing with you now I’ll postpone it to another time when things are less busy.

Leave a Reply