"Some of the Conclusions Drawn are Simply False"

Pushback against this:

Analysis!, Free Exchange: If you have been paying attention to the news, to financial experts, to economists ... then you may have heard that there have been some recent problems...

Alex Tabarrok ... has been pushing the argument that we may face recession, but that the financial crisis never threatened the real economy, and so the big government bail-outs were unnecessary. And now he has proof. Three economists from the research department of the Federal Reserve Bank of Minneapolis have produced a working paper purporting to debunk four myths about the financial crisis. Those myths are:

1. Bank lending to non-financial corporations and individuals has declined sharply.
2. Interbank lending is essentially nonexistent.
3. Commercial paper issuance by non-financial corporations has declined sharply and rates have risen to unprecedented levels.
4. Banks play a large role in channeling funds from savers to borrowers.

The authors of the paper next provide a damning analysis. In the best tradition of lazy undergraduates everywhere, they plot lines on graphs and draw wild conclusions. And on the basis of these conclusions, Mr Tabarrok writes his post, and credulous bloggers begin analogising the bail-out to the Bush administration's bogus claims about Iraq's weapons of mass destruction.

There are a few problems with all of this. First of all, some of the conclusions drawn are simply false. While rates on the highest quality non-financial commercial paper have behaved fairly well in recent weeks, rates for lower quality stuff have soared. The spread between the two, actually, is one of Calculated Risk's credit market indicators.

The failure to distinguish between the two types of paper is indicative of the broader, unwarranted credulity of the authors. For instance, many of the series they present actually show an unusual spike in bank lending during the crisis period. Are we to understand that for most banks, conditions actually improved, suddenly, sharply, and atypically while the rest of the financial world went to hell? Well, we might do that. Or we might suspect that the increase in bank lending was itself a product of tight credit conditions elsewhere—that borrowers were falling back onto lines of credit they normally wouldn't use thanks to the severity of lending conditions.

And of course, there is the inconvenient matter that the Federal Reserve and the Treasury went out and did all that stuff they did in order to prevent a massive breakdown in lending to the real economy. ... Now this does allow sceptics to say, "Well, how do we know things would have collapsed"? We don't, of course, but that doesn't change the fact that current lending takes into account massive government intervention to make sure that lending continued. The latter therefore can't be used to argue that the former wasn't necessary.

Maybe at some point we'll see some careful research that suggests that the threat the financial crisis posed to the real economy was drastically oversold. This, I'm afraid, isn't it.

Consider Figure 2B from their paper:

Here is their entire analysis of what this shows:

Figures 2A and 2B display analogous data for loans and leases made by U.S. commercial banks. Again, we see no evidence of any decline during the financial crisis.

Here's what I see. The bump in loans between September 10 and September 17 was probably this (9/14):

Federal Reserve Board announces several initiatives to provide additional support to financial markets, including enhancements to its existing liquidity facilities

• Press release
• Interim final rule -- Federal Register notice: 53 KB PDF | TEXT
• Primary Dealer Credit Facility
• Term Securities Lending Facility
  Related Statements by Other Agencies
• U.S. Department of the Treasury
• U.S. Securities and Exchange Commission
      September 15, 2008
      September 14, 2008

Then, we see the Lehman collapse, and this caused the Fed induced substitute lending to fall off from 9/17 to 9/24. Next, after the bailout plan is proposed lending takes off again (see the change between 9/24 and 10/1), but then it falls off again and turns negative after WaMu fails (see the change between 10/1 and 10/8). I don't see how you can look at this figure and come to the conclusion that there have been no disturbances in financial markets from the crisis generally, or from specific events.

And here's figure 2A which shows the same data since 2001:

Yep, no sign of any changes due to financial market problems and then recovery after Fed action in that series.

These three economists, V.V. Chari, Lawrence Christiano, and Patrick J. Kehoe, have done some excellent work in the past, and I expect better than this. Disclaimer or not, the first thing you see when you open the paper is "Federal Reserve Bank of Minneapolis Research Department," and this reflects poorly on the Minnesota Fed.

[Update: See also Credit Crunch: Did We Make It All Up?. Also: The Credit Crunch Isn't a Myth.]

Update: Evidence on the quantity of loans.

Comments

[rufus]

"In the best tradition of lazy undergraduates everywhere..."

Now that is funny!

[don]

I'm sorry. To me, it looks like the Fed acted to keep lending at unsustainable levels.
The Economist's "analysis" is on a par with much of the other analysis I see there. For example, they have never admitted the huge flaw in their 'Big Mac' index, which fails to properly account for value added taxes and so greatly misleads in ppp comparisons between the dollar and pound or euro.
Look at the second graph. I suppose if one accepts that the growth in borrowing prior to 2007 was appropriate ... but it looks to me like an unsustainable bubble. Nominal GDP certainly never grew anything like the growth in credit from 2001 to 2007. Maybe there is a cyclical element and 2001 is a cyclical low. But from the graphs shown I would not conclude the Fed's actions were necessary or desirable.

[MG]

According to a footnote to the October 10, 2008 FRB release H8, loans by large commercial banks increased in the week ending October 1 solely due to the acquisition of $259 billion in assets from nonbank institutions. This does not appear to be mentioned in the paper cited.

[Cynthia]

Sounds like Alex Tabarrok should be awarded the Ig Nobel Prize for economics.

[Jas Jain]

--
I am simply speechless. Conclusions drawn by economists are false? Hell, that makes them propagandists and NOT scientists, no Prof. Thoma?

It is not a revelation to me.

Jas

[John Booke]

Rate on Moody's Corporate Aaa paper has spiked higher over the past four weeks. Its more than a full percentage point from where it was in the middle of September. It is now 6.47%. The spread between Aaa and Baa has also climbed a full percentage point since mid September to 2.62%. Check out the H15 data on the Fed's website.

[phineas]

The article by the Minneapolis Fed folks just presents the data as the reality, and leaves the reader to think for himself about interpreting it. The economist.com blogger does the opposite, namely presents no interesting information about reality but goes on a rant about his personal synthetic interpretion.

Mark Thoma should've fed us the former and avoided the latter idiot. Luckily for me I came across the Minneapolis Fed material at another site.

[Patricia Shannon]

Jas Jain
In case you haven't noticed, the owner of this blog is an economist, and he is the one that posted this.

Can you direct me to a post in your own blog that disagrees with something you have said, or with others in your subject matter, whatever it is? Professor Thoma gives info about himself on this blog. All you say is that you have a PhD, not even in what subject.

You are really getting very boring, as well as being rude.

[Patricia Shannon]

Jas Jain, I also didn't see anyway on your blog for others to respond, much less insult you.

[me]

PLEASE enable "killfile" for this blog.

It's easy to do. It lets each user decide which userids to block. It's enormously helpful.

killfile – Userscripts.org
Provides a killfile for certain blogs. Covers livejournal, haloscan comments, most typepad blogs, most blogspot blogs, scienceblogs.com, and more as I add them ...
http://userscripts.org/scripts/show/4107

[tinbox]

"In the best tradition of lazy undergraduates everywhere, they plot lines on graphs and draw wild conclusions"

Sounds like Krugman's analysis of the oil market back in May. (It was very bullish.) And he's a great guy and a certified genius-and very willing to tell us what must be done now.

Among many other things, the Gov't is injecting equity on the easiest of terms into institutions that haven't failed...that doesn't seem over-hyped? You would have to be highly certain of the course of future events for that to be justified. There isn't any evidence that economists or Treasury officials have that sort of foresight.

[save_the_rustbelt]

I do love my economist friends, but they are incredibly divorced from the day-to-day activity of the "real economy," which seems to be synonymous with "Main Street."

Want tot know what is happening in the near term? Ask some CPAs.

[tinboxes are empty, like their rhetoric]

That's not what krugman was arguing. He was using a model with a failry sophisticated mathematical structure in the background, these guys are looking at lines on a chart - and not even doing that very well.

The question was whether there was a speculative bubble. Jime Hamilton destroyed the only evidence for that, just left it lying in a heap, and nobody has produced anything to replace it. There is, however, lots of evidence on the other side - real evidence. The speculation story is mostly bouncing around in an echo chamber on the internet and that is serving as validation to the believers as they keep hearing it from each other.

Since you've misrepresnted his point, here's what Krugman actually says. This comes right after he laughs at the guy that Hamilton destroyed:

"Is speculation playing a role in high oil prices? It’s not out of the question. Economists were right to scoff at Mr. Masters... but under some circumstances, speculation in the oil futures market can indirectly raise prices, encouraging producers and other players to hoard oil rather than making it available for use.

Whether that’s happening now is a subject of highly technical dispute. ... Suffice it to say that some economists, myself included, make much of the fact that the usual telltale signs of a speculative price boom are missing. But other economists argue, in effect, that absence of evidence isn’t solid evidence of absence. ...

In any case, one thing is clear: the hyperventilation over oil-market speculation is distracting us from the real issues. ...

Oil prices will fluctuate in the coming years... I wouldn’t be surprised if they slip for a while... but the long-term trend is surely up. ..."

So none of this - e.g. prices falling now - is inconsistent with his point.

(If the three Minnesota amigos are right, I don't think they are, but if so, then prices are in a downward bubble right now - the point is that it's dumb to say that anticipated weakness explains prices today, but anticipated world growth couldn't drive prices up then).

[John Mondragon]

Anyone notice the number of the Fed working paper?

Ominous...

[mmm_horsemeat]

"Or we might suspect that the increase in bank lending was itself a product of tight credit conditions elsewhere—that borrowers were falling back onto lines of credit they normally wouldn't use thanks to the severity of lending conditions."

Saying that credit availability is so 'severely' endangered that borrowers are forced to utilize credit from banks isn't the most persuasive argument. What next?

"Gasoline supplies had withered to the point that I was forced to fill up at Texaco instead of Chevron!"

"A calamitous shortage of Cherry Garcia at Albertson's forced me to my fall back supply available at Safeway, Kroger and/or Supervalu!"

Gosh, it makes one wonder how those rumors of trumped up hysteria ever got started...

[Uncle Billy Is Shocked]

The paper, intended for the MN Fed, was apparently done under the watchful and suspect eye of the NBER. Kind of silly to worry about the Fed's reputation until they accept a paper and agree with it or act on bad information. But this NBER...

[xfire]

"Want tot know what is happening in the near term? Ask some CPAs"

The same CPAs that certified Enron's books !!!

[xfire]

"Want tot know what is happening in the near term? Ask some CPAs"

The same CPAs that certified Enron's books !!!

[don]

This paper may not be stellar economic research at its scientific best, but it sure beats the heck out of the Hufbuaer et.at calculations about the gains from free trade, which were simply crazy and in the best tradition of the worst advocacy research. Yet BB cited the results in testimony before Congress. That showed either very poor fundamental economic understanding, or a tendency to use any lie that suits your purpose. Neither alternative reflects well on BB. I guess the idea was, don't attack the currency mercantilists (who provide a great deal of laons to U.S. borrowers), even though that is as far from free trade as the highest trade barriers we have seen globally in the past generation.
Yeah, just like the second graph - do anything to keep up unsustainable borrowing.
On the other hand, the data compiled by Walmart show purchases of baby formula spiking during paydays. That's the kind of thing gives me pause about accepting too much of a slowdown. But how about social support for the poor, including the working poor, rather than bailing out the high end of the financial sector?

[Ginger Yellow]

I think the most disturbing thing about the paper is the fact that they completely ignore asset backed commercial paper. Part of the reason that "total credit" has stayed relatively stable is that banks are now sitting on hundreds of billions of dollars of asset backed commercial paper from programmes they sponsor that used to be owned by investors. Now from the perspective of the "total credit" number, it doesn't matter if a bank owns ABCP as a willing investor or a sponsor, and in the short run it doesn't matter so much to banks' existing clients except for the increased cost, but it clearly has an impact on the availability of credit, especially since outstanding ABCP has plummeted. Similarly, even for the CP the paper does address, they don't do any of the analysis you would expect. The "analysis" of CP rates covers only one maturity (there's no attempt to see how the maturity profile has changed) and even less forgivably, it uses absolute rates, not spreads. So if the Fed cuts rates by 100bp and the CP rate falls by 50bp, say, then that's a sign that the CP market is healthy, according to the paper. That's just willfully wrong.

[brookside]

MG points out a footnote in the 10/10/08 FRB release H8 about the acquisition of $259 billion in assets from non-bank institutions which showed up in the 10/1/08 release. On 9/21 and 9/22 Goldman Sachs and Morgan Stanley opted into the Federal Reserve System and became commercial banks. Any bets their loan and asset data did not make it into the 10/24/08 H8 release but showed up in the 10/1/08 release.

[brookside]

MG points out a footnote in the 10/10/08 FRB release H8 about the acquisition of $259 billion in assets from non-bank institutions which showed up in the 10/1/08 release. On 9/21 and 9/22 Goldman Sachs and Morgan Stanley opted into the Federal Reserve System and became commercial banks. Any bets their loan and asset data did not make it into the 10/24/08 H8 release but showed up in the 10/1/08 release.

[Evan]

I thought the afore mentioned paper was of abysmal quality; it could never survive peer review. What are the credentials of the are the authors? Are they really economists? There is more you could have pointed out to further debunk the paper but perhaps you have given us enough to show it's sloppiness. For example, they refer to commercial paper issued while ignoring changes in total commercial paper. In fact, they don't even label the vertical axis in their graphs with units of money! A little knowledge is a dangerous thing.

[skeptonomist]

I am not a fan of Fed research and have not read the paper in question, but I noticed some of the same things plus some others and put the comment below on Calulated Risk on 10/15.
Figure 2B in the paper (reproduced by Mark) has more recent data on bank loans, and the strange jump has leveled off. MG and brookside may have the explanation for the jump - it has to be interbank transfers of some kind. Nevertheless the data on loans and leases still do not show any evidence of a freeze.

There is no question that there are big problems in credit markets, but I strongly suspect that there was a panic about a credit freeze, based on what may ultimately be rather inconsequential raises in interest rates, rather than looking at actual volume data. The good news is that the panic shows signs of subsiding.

Those who are not sloppy undergraduates and want to get to the bottom of this should not take the Fed's word, or mine or Mark's, but should look carefully at the actual data, including volume data, not just interest rates.
-------------------------------------------------------------Comment by skeptonomist at Calculated Risk 10/15/08:

I keep looking for evidence of a credit freeze in the real world (outside Wall Street) and so far have not found it. Take a look at the real data on commercial bank loans:

http://research.stlouisfed.org/fred2/categories/100

especially weekly data. Loans were tapering off earlier in this year, but since the middle of September (collapse of Lehman Bros) they have been shooting up, not down.

Anyway, you can see from past history on the graphs that loans decrease severely in a recession, so as we are entering a recession we should expect the curves to turn down - such a thing is not an indication of an extraordinary "freeze" due to special credit problems, which would presumably cause a sharp drop. Dean Baker has called the media on several anecdotes of supposed credit problems which are better explained as a result of normal recessionary contraction or other factors. These tales are not evidence of a credit freeze.

Much has been made of commercial paper and Paulson says he needs to buy it up, but the volume data

http://www.federalreserve.gov/releases/cp/volumestats.htm

are generally above normal. Rates for lower-quality paper are high, but not crippling - they were almost as high in the middle of last year, and they are still far below the average for the last 40+ years.

The TED spread has been comparable to what it was in the Bank and S&L crisis of the late 1980's, which was not considered to cause a major credit freeze. We can expect more bank failures as at that time, and continued flight from mortgage-backed securities to Treasuries, but so far I can see no evidence of a calamitous inability of business to get credit.

[Bill Woolsey]

I read the Fed paper. It didn't say that the financial crisis was a myth. Rather that there are four myths being spread about the crisis.

To claim that banks aren't lending is false. They are lending. Total lending by banks is expanding. That doesn't mean that all banks are lending more. Or that other sorts of lending haven't decreased. It just says that some banks are making loans to a degree that total lending has expanded.

Similarly, it is claimed that banks aren't lending to other banks. The data shows that banks are lending to other banks. That doesn't mean that every bank can borrow from other banks. Just that some banks are lending to others.

If you look at the volume statistics for newly issued asset backed and AB commercial paper, it dropped sharply in September. It didn't fall to zero. Credit in that market wasn't "frozen." It was just less and more expensive.

Hopefully, the low level analysis provided by these economists at the Fed will allow journalists and the like to tell a better story. Large money center banks can't borrow from other banks. Nonbank finance is much more expensive and shrinking for less credit worthy firms (like Ford Motors.) Many banks are expanding lending, but people who borrowed from large money center banks aren't getting loans like they had.

And, of course, perhaps they can explain why this is a problem.

Well, probably they can't. But perhaps Thoma can develop
some high level analysis to explain the actual nature of the problem. Some banks lending more, others less, less direct finance, higher rates for risky firms, lower rates for consumer borrowers... Explain why this is a problem. I can see how it might well be.

But the Federal Reserve paper didn't claim to show that there was no financial crisis or that it is no problem. Just that the crisis isn't one of banks not lending at all or no banks lending to other banks. That is, credit isn't "frozen."

[Jack]

I read their paper. The authors don't say that there were no disturbances in the financial markets. All they say is that there is no tangible effect yet on lending to non-financial sector. It may be that future lending would have dried up without these actions. However, there is certainly no evidence that non-financial firms were facing difficulty borrowing money. There were a lot of stories being bandied around in the media that even firms like McDonald were unable to borrow money. At a minimum, their data shows that this wasn't true. Also, the graphs clearly indicate that the volatility in lending one observed in September was not unusual relative to historical volatility. It is true that the authors do not produce formal statistical tests. But, I am quite convinced that formal tests would bear out that the volatility in lending has not significantly changed.

You advance the interesting hypothesis that the increase in lending might be due to firms tapping their predetermined lines of credit. Well, do you have any evidence in support of this?

Finally, you suggest that the very fact that the Fed and the Treasury took drastic actions is evidence that something was wrong. Your logic is circular here, isn't it? They need to share this evidence and justify what prompted them to take these drastic actions.

You seem to take at face value what our policy makers have been telling us. The conventional wisdom can often be wrong. I think the authors are raising legitimate questions. You shouldn't be so dismissive.