September 07, 2009
Guest Blog: Financial Crisis and Reform Déjà Vu
By Simon van Norden
"Once you've seen one financial market crisis...you've seen one financial market crisis."
-- Attributed to Federal Reserve Board Governor Kevin Warsh by former US Treasury Assistant Secretary for Economic Policy Phillip Swagel in The Financial Crisis: an Inside View, March 2009, p. 4.
The financial crisis has set a lot of records so far; it's certainly the worst US banking crisis of my lifetime. Some, as suggested by the above quote, see such crises as unique events; each one is singular and there's not much to be learned about how to handle one from looking at past crises. For example, there's no precedent that I know of for a banking crisis involves the failure of the biggest counterparties for credit default swaps.
I think a much smaller number of people see the crisis differently; they think of it as another potato, a big one. No two potatoes are exactly alike in size and shape, but they all taste pretty much the same and you can use the same recipe for most of them. For that reason, it's interesting to see to what extent the current crisis behaves like other crises, even if it has some unique features.
I think there's some interesting parallels between the current crisis, the Savings and Loan (S&L) crisis of the 80s and 90s, and the Long-Term Capital Management (LTCM) Crisis of 1998. But before I talk about that, let me talk about what a "typical" banking crisis looks like.
The Basel View of "Typical" Banking Crises
If we set the way-back machine to 2004, a time long before terms like ARM, CDS, and AIG entered common conversation, we can see what people thought a typical bank crisis looked like. That's the year the guys in Basel who worry about such things published "Bank Failures in Mature Economies." They looked at the main banking crises in developed countries from 1980 to 2000 and asked themselves what they saw. To be sure, they saw some differences, but they also saw some patterns. Here's part of their main conclusions (note that "credit risk" is Banker for "bad loans").
Most of the widespread [banking] failures required some amount of public support, sometimes in very large amounts. All of the episodes that involved large amounts of public support were caused by credit risk problems. ...The widespread banking crises that involved credit risk were remarkably similar. A period of financial deregulation resulted in rapid growth in lending, particularly in real estate related lending. Rapidly rising real estate prices encouraged more lending, abetted by lax regulatory systems in many cases. When economic recessions occurred, inflated real estate prices collapsed, leading directly to the failures. (BIS, 2004, p.66)
That sounds a lot like what the US (and some other countries) experienced immediately afterwards. There had been some financial deregulation, which was followed by a period of very rapid growth in real-estate-related lending. Rapidly rising real estate values encouraged more lending. The biggest difference seems to be the last point; the recession did not cause real estate prices to collapse; they had peaked by 2006 and fell before the recession started. We could probably also argue about whether it was financial deregulation or "financial innovation that avoided regulations" that helped fuel the increase in real-estate lending. However, in this view the boom and bust cycle in real estate, the subsequent fallout for the banking sector, and the need for a major publicly-funded bailout is not remarkable; we've seen this kind of story before. In fact, Reinhart and Rogoff have gone so far as to tabulate what happens to government debt in the aftermath of a banking crisis. They find that real government debt increases by an average of 86% in the three years after the start of a crisis. So regardless of how you feel about the US government's spending during the crisis, it seems less remarkable when compared to what typically happens in a banking crisis.
Figure from Reinhart, Carmen M., and Kenneth S. Rogoff. 2009. "The Aftermath of Financial Crises." American Economic Review, 99(2): 466-72.
Three American Financial Market Crises
More support for the view that banking crises follow similar patterns can be found by comparing the last three US banking crises; the S&L crisis of the late 80s and early 90s, the collapse of LTCM and the most recent crisis. The S&L crisis closely followed the pattern described by the BIS report quoted above; financial deregulation, followed by a rapid growth in real estate lending, creation of local speculative bubbles in real estate prices, and the failure of institutions as bubbles burst (For descriptions of the S&L crisis, see BIS (2004) or the GAO 1996 report). The General Accounting Office put the cost of the S&L bailout to US taxpayers at $132.1 billion, or a bit under 2% of GDP (United States General Accounting Office (1996) "Financial Audit: Resolution Trust Corporation's 1994 and 1995 Financial Statements," Table 3 and author's calculations). That may seem small compared to the size of TARP or this year's projected federal deficit, but it was shocking at the time.
At first glance, the LTCM crisis appears quite different; no bank failed (LTCM was a hedge fund), its failure was unrelated to real estate investment or credit risk, and the crisis was resolved at no cost to the taxpayer. However, the LTCM crisis showed that, as a result of deregulation, a systemic crisis could start outside the regulated banking system. Another GAO study noted:
The LTCM case illustrated that market discipline can break down and showed that potential systemic risk can be posed not only by a cascade of major firm failures, but also by leveraged trading positions. LTCM was able to establish leveraged trading positions of a size that posed potential systemic risk primarily because the banks and securities and futures firms that were its creditors and counterparties failed to enforce their own risk management standards. (US GAO (1999) p. 29)
The same report noted:
- Gaps in [US Government agencies'] regulatory authority limits their ability to identify and mitigate systemic risk (US GAO (1999) p. 24)
- Regulators did not identify weaknesses in firms' risk management practices until after the crisis (US GAO (1999) p. 16)
- Monitoring did not reveal the potential systemic threat posed by LTCM (US GAO (1999) p. 17)
and provided a variety of proposals (some of which are mentioned below) to reform the financial system by reducing systemic risks.
The success of those reforms can be judged by role of similar factors in the most recent US banking crisis. An important factor in the latter has been the role of trading in derivative securities, primarily mortgage-based securities and credit default swaps (CDS). Again, government oversight of this market was limited due to faith in the market's ability to manage its exposure to risk, and was further weakened by divided responsibilities between multiple agencies. Regulators and private lenders alike were again unaware of major firms' exposure to losses on derivative securities; this time even the heads of major financial institutions were not aware of the extent of their own exposures. Again, this was in part due to the lack of transparency, lack of clearing and high leverage afforded by trade in Over-the-Counter (OTC) derivatives (particularly those traded at Bear Stearns.) Again, weaknesses in firms' risk management practices became apparent only in hindsight. Again, major financial firms that were not regulated as traditional deposit-taking banks took on highly-leveraged positions and posed major systemic threats to the banking system. These included several investment banks (such as Bear Stearns, Goldman Sachs, Lehman Brothers, and CitiGroup) and the insurance company AIG.
Looking at recent events from this perspective, I still see the size of the losses as breathtaking, but the causes and dynamics seem much more familiar. What bothers me is that some of the suggested solutions sound pretty familiar too; make derivative trading more transparent, improve coordination among the regulators, give regulators more power to control systemic risk in new places, and so on. Despite that, not only was there another crisis, but it was much larger than the two previous crises combined.
This post written by Simon van Norden.
Posted by Menzie Chinn at September 7, 2009 05:01 PMdigg this | reddit
Very interesting - thanks for the guest post.
One question - do you accept the characterization that "The LTCM case illustrated that market discipline can break down"?
Or do you believe it is better characterized as a case of individually disciplined rational actors operating in a less than perfect market (along the lines of "The Limits of Arbitrage"?
Posted by: Robert Bell at September 8, 2009 04:48 AM
This crisis may be reflected as private casualties and irresponsabilities,but the primary facie responsible body is institutional.
Excessive money supply is a central bank power
Banks supervision is a central bank duty
Real CPI is a central bank mandate
BIS is responsible for the mandatory requirements on prudential ratios.Capital adequacy ratios were designed to indulge leverage with derivatives (see office of the comptroller report on derivatives outstanding) The assets pricing was far incommensurate with incomes and revenues (Case Schiller,equities markets)
The compression on interest rates is driven by central banks (primary dealers,derivatives)
The immune system of the economy had and has been suppressed with a benevolent laissez faire.
The rest is contemporary history.
Posted by: ppcm at September 8, 2009 06:44 AM
Thanks, great post.
By the way, LTCM's collapse was largely a result of real estate bubble collapses in Asia, especially in the sense that they triggered the collapse of the government debt bubble in Russia. Also, some estimates of the total cost of the S&L bailout topped $200 billion.
Simon's conclusion suggests that he thinks we should consider whether bailouts lead to larger crises. Another interesting angle for comparison would be the different types of bailouts that have been applied. It's one thing to bail out depositors, which was the focus of the S&L bailout, and quite another to bail out the owners of financial institutions' equity, debt and other obligations, which was the focus of the recent bailout.
Posted by: Tom at September 8, 2009 07:59 AM
Robert Bell: I agree with the first and am skeptical about the second. I suspect that LTCM's counterparties were myopic. For example, allowing LTCM to take large OTC derivative positions without posting margins is hard for me to reconcile with rationality. (I've heard explanations, but I'm not convinced by them.) Somewhat similarly, many large firms kept large exposures to real estate on their books through 2007 and into 2008 when it was very clear that a major real estate market correction was underway.
Posted by: Simon van Norden at September 8, 2009 08:18 AM
I concur. Very interesting and useful post.
So the Fed was warned no in unambiguous terms as early as 2004 and the weaknesses were spelled out in several documents.
So what was the failure here? Was it the result of macro factors (excess Chinese savings)? Was it analytical (not in the broad sense, but rather in the more day-to-day sense?)? Was it institutional (did the Fed lack risk-facing functions)? Was it political (did the Administration suppress analysis, reports, or fail to respond to warnings)? Was it one of mindset (people simply weren't prepared to accept negative interpretations of data)? Was it moral (why did no one at the Fed speak up? Who resigned in protest? Who has held himself accountable or been held to account)?
What needs to be changed to prevent a recurrence? Or is this simply a matter of human nature, like periodic forest fires, fated to recur when memory fades and tinder accumulates?
Posted by: Steve Kopits at September 8, 2009 08:38 AM
Steve Kopits: This ain't fate.
Reading the reports written in the aftermath of the LTCM collapse, it seems clear that policymakers encountered some similar problems last year. It's hard to believe that memories faded a lot since 1999.
On what needs to be changed...that's a separate post.
Posted by: Simon van Norden at September 8, 2009 09:34 AM
I think you're the first person I've heard suggest that the Russian debt default was caused by the collapse of a real estate bubble.
On the subject of different types of bailouts, there's nice analytical work done on that. I'm making my students read Honohan and Klingebiel (2004); Google-Scholar them and you'll find more.
Posted by: Simon van Norden at September 8, 2009 09:44 AM
There are 3 ways to to smooth out economic crisis potential amplitude:
1) increase the viscosity of the financial system = delay practical decision maker executive reaction to changes in money velocity everywhere, in all scales.
2) Reduce global /local money velocity as such
3) Reduce scale of global financial operations
Would any politician accept any of this? Or businessman?
Still it may not exclude some big one-off events.
The cost is : If depth of crisis (fluctuations of velocity of money) will be reduced, inovation will be delayed as well. There is a balance to be found.
Posted by: Ivars at September 8, 2009 10:12 AM
If you look at an historical chart of commercial paper it is pretty clear that it ramped up in an unprecendented manner at the start of the bubble ('04). Is there something similar in other crises? Any way to identify massive credit creation that could be a red flag.
Posted by: oops at September 8, 2009 10:40 AM
The one thing that seems to be an underlying, recurring theme, probably going back a few hundred years in banking, is excessive leverage and low capital.
This go around we made a quantum leap in crisis creation practices. A few that come to mind:
1) Much more securitization. Bankers are working on commissions and fees, then sell off the risk, or transfer it to a SIV. But then other banks buy the securitized products, and can use them towards capital requirements. And ratings agengies would hand out the neccesary, but dubious AAA ratings. How nuts is this?
2) We eliminated the "firewalls" between commercial banking and investment banking, and apparently insurance too.
3) We had the trade deficit induced savings flow from EMs to OECD countries. This helped make much more cash available.
4) We could buy "insurance" (CDS) on the ponzi scheme, if you wanted a warm feeling about your CDO position, then found out the insurance wasn't backed adequately.
5) OTC trading was not "transparent", nor on a regulated exchange, so there wasn't any risk assessment going on.
6) Moral hazard IS REAL, and with multi-million bonus payments...IT IS HERE NOW!
Posted by: Cedric Regula at September 8, 2009 10:45 AM
The immediate cause of the collapse of LTCM was the failure of Russian bonds but the real cause was the arrogance of the LTCM management that created the algorithms that led them to move from 90 to 1 to over 200:1 leverage at the height of folly. When levered up that much, any slight breeze brings down the house of cards. When Genius Failed is a very readable account of the collapse; reading it now you can draw a straight line from LTCM to the mortgage bundlers and their government non-regulators who together brought on the current mess.
Posted by: akpundit at September 8, 2009 11:33 AM
Two things come to mind:
1 - The Basel accords have gone a long way to assigning strict defined limits to risk positions, a sure way to increase real risk. Anyone who has ever worked in a trading environment knows well that when risk limits are well defined with hard numbers, positions will be taken right up to the actual limit, the limits will be 'gamed' with all maner of approaches, and often approvals will be sought to even slightly exceed limits. In contrast, with only a vague definition of limits, risk is held lower to avoid inadvertantly and unpredictably being called out. The later case, vague limits, seems more perilous to an outsider, but it is orders of magnitude safer in actuality: a fact the BIS seems painfully unaware of.
2 - I still fail to understand why no one saw the situation at AIG developing. Reading their annual and quarterly reports and filings (in hindsight), their rapidly growing ABS super senior insurance business should have thrown up all sorts of red flags (especially given the language they used to explain the business), and there are lots of people on Wall Street who have a history of finding these things and betting (heavily) against them. Thus I am bewildered why no Ackman, etc came to take a large short position in AIG, and why no one started beating a drum about this until it was way too late. Their must simply have been too many lower risk (long) ways to make money....
Posted by: Dr. D at September 8, 2009 11:36 AM
I am not really sure of Norden's point but the following sentence was interesting.
"Again, weaknesses in firms' risk management practices became apparent only in hindsight.
My economics friends often point out to me that most financial innovations such as exotic derivatives come into existence to skirt prohibitive regulations. As Norden reveals in the quote above regulators only recognize the risk in hindsight. Then, by the time a new regulation is written and passed into law or enacted into policy, traders are already finding new innovative ways to skirt the new controls.
Even honest regulations and regulators can never keep up with the innovations of the millions of traders in the market. New regulations usually hinder traders create problems for traders that do not need to be regulated and so simply introduce friction into the economy. The new regulations are too late in the process and obsolete.
What Norden seems to be saying is that we need to allow companies to fail when they get caught on the wrong side of the risk. If they know they will pay the price they will be more prudent in engaging in risky activities (moral hazard anyone?)
The next time you review one of JDH's great charts on the FED balance sheet just consider you are watching moral hazard in the making. The market will regulate and discipline better than any bureaucratic regulator.
Posted by: DickF at September 8, 2009 12:14 PM
The central banks manipulation of interest rates is the primary cause in my mind, followed by the securitization of assets. The drop in the prime rate from 9 to 5 between 2001 and 2002 led to a historic movement of people into housing, ostensibly by using adjustable rate mortgages. Then the just-as-quick reversal of the rate from 5 to 8 from 2005 to 2006 set up the default, bankruptcy, lower demand, unemployment death spiral.
Posted by: Gridlock at September 8, 2009 12:43 PM
It is not surprising to me that we get bubbles. People are herd animals for the most part. While many are questioning whether we have enough regulation, I wonder about the moral hazards created by bailing them out. What about the role of Fannie and Freddie? One can hardly begin to sort out the role of individual firms from the government and quasi government players involved in the market. Did firms really ignore risk or where they encouraged or even pressured to do so?
Posted by: Hitchhiker at September 8, 2009 12:56 PM
There's a lovely Michael Lewis article "The End of Wall Street's Boom" from Nov. 2008 on Portfolio.com which recounts the adventures of some of those who recognized the problem early and bet their investments accordingly.
Posted by: Simon van Norden at September 8, 2009 01:04 PM
1) I'm struck by the apparent contrast between your conclusion
The market will regulate and discipline better than any bureaucratic regulator.
and the GAO quote in the article
The LTCM case illustrated that market discipline can break down.
I'm not sure how you square those.
2) In quoting me, you state that I reveal that regulators only recognize the risk in hindsight. I didn't know that; there's no reference to regulators in the quote that you use and I did not intend one. For an example of how firms are harmed by the breakdown of their own risk-management practices, why not read Michael Lewis's August 2009 article on AIG in Vanity Fair?
3) My name is not Norden. It is van Norden.
Posted by: Simon van Norden at September 8, 2009 01:23 PM
Simon, thanks, but I've read the various articles by Lewis (and while fun reads, they are generally overhyped, IMHO of course).
That the AIG debacle effectively took place in full public view should bother people a lot, especially those calling for more regulation.
Posted by: Dr. D at September 8, 2009 01:53 PM
I am very sorry for calling you "Norden" rather than "van Norden." In the past I have always used both but a few years ago I was chastised for writing "von Mises" rather than "Mises" by someone who knew him. Since that time I have dropped von, van, de, du, and others. I promise you I did not intend any disrespect and hope you will accept my apology.
Posted by: DickF at September 8, 2009 02:22 PM
No worries on that score! Just wanted to nip the problem in the bud...
Posted by: Simon van Norden at September 8, 2009 02:45 PM
van Norden wrote:
1) I'm struck by the apparent contrast between your conclusion
The market will regulate and discipline better than any bureaucratic regulator.
and the GAO quote in the article
The LTCM case illustrated that market discipline can break down.
I'm not sure how you square those.
To answer your question let's look at the entire GAO quote.
The LTCM case illustrated that market discipline can break down and showed that potential systemic risk can be posed not only by a cascade of major firm failures, but also by leveraged trading positions. LTCM was able to establish leveraged trading positions of a size that posed potential systemic risk primarily because the banks and securities and futures firms that were its creditors and counterparties failed to enforce their own risk management standards.
When risk management standards are not enforced why would the GAO say "market discipline can break down?" Wasn't this rather a violation of market based lending standards? This was just another case of moral hazard. The "banks and securities and futures firms" felt that if a problem occured either LTCM would be bailed out or they would be bailed out or both. And they were essentially correct because - I will quote from wikipedia because I am lazy -
"Goldman Sachs, AIG and Berkshire Hathaway offered then to buy out the fund's partners for $250 million, to inject $3.75 billion and to operate LTCM within Goldman's own trading division. The offer was rejected and the same day the Federal Reserve Bank of New York organized a bailout of $3.625 billion by the major creditors...."
In truth the LTCM incident proves that the market does discipline BECAUSE LTCM failed. Even though they had the "greatest minds" these "greatest minds" failed. Their model was not sound. Once again to quote Wikipedia (for the same reason)
LTCM's strategies were compared ... to "picking up nickels in front of a bulldozer" — a likely small gain balanced against a small chance of a large loss, like the payouts from selling an out-of-the-money option.
But it also proved that our markets are not free markets. The NY FED over-rode a market offer to purchase LTCM by twisting the arms of its creditors. It was noted at the time that the involvement of the NYFED would simply create more moral hazard in the future. Oh how the past two years have proven that true!!
So the simple answer: I don't agree with the GAO.
Posted by: DickF at September 8, 2009 03:00 PM
I am sorry but I don't think the real problem is the failure of regulation (though moral hazards like agency issues certainly exacerbated the fallout). Let's not forget that after the dot com bubble the central banks were desperate to drive up leverage to prevent deflation. The leveraged financial firms were simply doing the central bankers' bidding. In such a case you couldn't expect them to behave any other way. It was completely rational to expect real estate prices to go up or at least stay put (or to at least have faith that the central bankers could prevent a nation-wide collapse, which none other than THE central banker himself implied). It is interesting that though the European banks are more leveraged the problems started in the US and Europeans bore about half of losses from the subprime fallout in the US. Europeans certainly seem comfortable with the level of leverage US was getting up to. However the distribution of income in the US does not support the leverage that the Europeans are accustomed to in high tax welfare states.
Posted by: HZ at September 8, 2009 04:40 PM
I guess you reminded me of a few more things. Like gross incompetence played a big role in this. Krugman just published a new paper stating that macro as it was understood, is wrong, and we need something new. He suggests it will be a messy new theory, but probably should include something like watching out for Minsky Moments. So the Nobel Prize winner hath spoken.
But I do remember Greenspan and his epiphany that we cannot have deflation "because the Fed has a printing press.". So now we see that Keynes wasn't kidding about that Liquidity Trap thing, monetary policy is "zero bound" and isn't working, and Ben is buying $1.2 Trillion in MBS trying to prop up the $11 Trillion mortgage market and housing prices. I consider monetary policy to be "regulation", and if Central Banks believe they can buy their way out of any mistakes in judgement they make, especialy huge ones that econ 101 students wouldn't even make, then we are in big trouble. But bottom line it's a credit economy, and monetary policy is very weakly linked to credit. So we get swings from bubbles to debt deflation busts from credit, not monetary policy directly. Consequently, the Great Moderation has been looking more and more manic depressive that past 10 years.
But loose lending standards and toxic loans had a lot to do with it. In 2006, 30% of all loans were variable teaser rate loans, and then there were the no-down liar loans. I think this just all got obfuscated by securitiaztion into CDOs and the upper tiers with AAA ratings just fooled a lot of people. We didn't really get CDOs dissected in the media until after the fact.
And the rest of the government was complicient with all the irresponsible behavior as well. It's just that the banks ran off with all the money in personal bonuses.
Of course now we have the Obama trioka; Summers, Geithner and Bernanke.
They all were overseeing the mess develop, and were even key players in getting lots of the problem de-regulation steps implemented.
And they can't really fix anything without it being "bad for the economy". Wonder were we go from here?
Personally I think we need to go back to a "utility" type of banking system which has depositors and the banks make personal, mortgage and biz loans. Sounds like I'm longing for S&Ls again, but maybe we could have some better regulated ones and sound lending standards. The other thing to note is Canada has zero reserve banking, and there have been no problems there. Could compensation (lack of excessive...) have something to do with that?
Things like trading of any kind should be allowed only by stand alone hedge funds, mutual funds, and private equity funds. This way traders and money managers are working for investors who knowingly have there money at risk. And investors get the reward. No more of this situation where a bank can get billions in near zero interest deposits from savings and checking accounts, make the funds available to to its trading division, who in turn uses it to drive up our cost of food and gasoline, and then "earn" up to $100 million for a trader, $500 million for the bank, and if anything goes wrong, the taxpayer gets the bill.
We've got to be completely crazy as a country to let that go on.
Posted by: Cedric Regula at September 8, 2009 07:04 PM
Maybe the idea of "utility" banking has merits and you only need to look to Canada for such a paradigm which has weathered the financial-banking crisis fairly well.
Posted by: Brian at September 8, 2009 07:40 PM
Looking back, the take-home from LTCM for me was that banking institutions could game the government, pulling in super returns via leverage and yet be bailed out when the inevitable happened.
Posted by: PeeDee at September 8, 2009 08:30 PM
Monetary policy works through driving asset prices and thus investment demands. Thus I would argue that toxic loans were what the central bankers were asking for at the end of the long cycle when interest rate was approaching zero. Indeed if every loan officer behaved like Warren Buffett the central bankers would have been completely emasculated and monetary policy could not be conducted. However in the end they run into the zero bound as you said and had to rely on fiscal rescues. If anything they have so far escaped the blame by successfully shifting the blames to others. That does not bode well for the future.
I agree the agency issues were running amok but that was not the cause of this credit collapse. The real harm caused by the agency issue is the large scale mis-allocation of resources. On the human side you have MIT graduates going after financial engineering zero sum games in stead of real world engineering problems. On the investment side we have the massive over-investment in US residential housings. I would also argue that the large volatility in the commodity market is only confusing not helping the real economy. Of course when you award people involved in zero sum games on a commission basis, you are asking them to create volatility -- the one sided compensation formula will ensure that some agents will gain immensely from volatility while no agents will have to pay for losses.
Posted by: HZ at September 8, 2009 09:07 PM
When risk management standards are not enforced why would the GAO say "market discipline can break down?" Wasn't this rather a violation of market based lending standards? This was just another case of moral hazard.
No. A necessary condition for moral hazard is that agents are making a rational decision based on knowing the true risks that they face. The quants at LTCM were surprised when their models broke down; moral hazard does not explain that.
One says "market discipline can break down" in precisely such a case because it was extensively argued that firms with bad risk management would rationally expect to be punished by an efficient financial market, so they would avoid lax risk management. Instead, we saw bad risk management punished; equity investors at LTCM, AIG and other firms lost billions.
You're right to note that such incompetence was eventually punished by the market. The GAO noted this contradicted the argument that merely the credible threat of such punishment would be enough to enforce good risk management.
I'd also be interested to know the source of your claim that the NY FED over-rode a market offer to purchase LTCM. In particular, who at the NYFRB did so, when, and how did they compell LTCM to accept a lower offer?
Posted by: Simon van Norden at September 9, 2009 05:20 AM
The other thing to note is Canada has zero reserve banking, and there have been no problems there.
Actually, Canadian banks face higher capital requirements than US banks. They are also helped by the fact that (1) Canadian housing prices did not fall 30%, (2) they have large deposit bases, and (3) the Canadian mortgage market is much more regulated (for example, all mortgages for greater than 80% of property value must be insured.) The IMF has done some nice overviews of Canadian banking and mortgages this year; for example, check out their 2009 Article IV consultation with Canada.
Posted by: Simon van Norden at September 9, 2009 05:32 AM
A lot of basically different stuff is thematically similar enough that writers will roll it into the big hair ball called "The credit crisis of 2007-2009" or something like that. But if we want to think like engineers and see what needs fixing, it's better to focus on smaller sub-problems. Example: in retrospect it seems clear that without FDIC like guarantees on most money market type funds, the large size of those funds collectively makes the system vulnerable to a bank run. This Bloomberg piece gives some nice color on how that got started.
The Lehman failure wouldn't have triggered a run on money markets unless it was plausible that a bunch of other big commercial paper issuers might fail. That could conceivably have happened because of a big natural disaster or war, but in this past crisis it was only a sudden re-evaluation of the balance sheets and cash flow situation of many financial firms. The debate about how inefficient the market is will go on, but in this case it is clear that the market as a whole wasn't connected with the info in 2006 that a 25% decline in housing prices over two years would directly lead to insolvency at a lot of large investment banks.
I'm sympathetic to the proposed partial fix that IBs need higher capital ratios, but perhaps a lot can also be done by putting greater disclosure requirements on 1) the money market funds that are eligible for a Federal backstop, 2) the commercial paper issuers that are eligible to have their paper put into the money market funds that are eligible for the backstop. Stuff like that is certainly easier than than asking the Fed to prevent "bubbles" that might mistakenly look attractive to IBs.
Posted by: Josh Stern at September 9, 2009 07:12 AM
Yes, let me guess, things were like the eighties in the US.
(1) Canadian housing prices did not fall 30%
Because they did not go up 30%?
(2) they have large deposit bases
Because banks paid interest to depositors?
3) The Canadian mortgage market is much more regulated
So maybe this is why prices didn't skyrocket?
Also, we used to have the mortgage insurance requirement here too on any loans with less than 20% down.
So I think we are on to something here, although now we still have 3% down loans from the USG, temporary $8000 credit for first time buyers, and they can also save for a mortgage payment in their 401K, then withdraw it for a first time home purchase.
But I think you need a job to get a mortgage now, so we have improved a little.
Posted by: Cedric Regula at September 9, 2009 08:08 AM
You just reminded me of another thing. Lehman was funding 30:1 leverage with overnight repos in the money markets.
In line with not allowing any organization involved in financial trading of any sort to have access to nearly unlimited near zero cost funds, thereby making it more transparent to the "little people" what their cash is really being used for, I would nix this practice as well.
Most people think their money market pays .5% to 1% because it is low risk. Not because they are morons and it is being used by a hedge fund at 30:1 leverage to make zillions for hedge fund traders.
We need to simplify life a bit. Who wants to analyze .5% money markets for untold hours anyway?
Unfortunately, we are going the other direction here GS and MS applied to be commercial banks, and Ben said, "Sure!".
The good part is they get regulated like banks and can only go 12:1 leverage. Bad part is they will pay us nothing for deposits, and give the money to their traders to play with.
Posted by: Cedric Regula at September 9, 2009 09:12 AM
van Norden wrote:
One says "market discipline can break down" in precisely such a case because it was extensively argued that firms with bad risk management would rationally expect to be punished by an efficient financial market, so they would avoid lax risk management.
Who in his right mind would ever argue that any system, free market or otherwise, could totally "avoid lax risk management?" Are you seriously arguing that because the free market cannot totally control behavior we need a coercive government that can?
I will conceed that a totalitarian government can make people do things that free people would not do, but even the most draconian totalitarian government cannot "avoid lax risk management."
This is a strawman argument.
Posted by: Anonymous at September 9, 2009 09:42 AM
1) There was substantial appreciation in Canadian real estate from 2000-2007; the IMF says it was comparable to that in the US, but I'm not sure where their numbers come from.
2) My bank is paying 0.00% on chequing accounts and 0.35% on 90 day GICs (term deposits), which is typical right now.
Posted by: Simon van Norden at September 9, 2009 12:37 PM
Professor van Norden,
If one divides the problem into two parts it may be easier to see the cause:
1) Central bank management of aggregate demands: such management (monetary policies) would be ineffective if lax risk management is not pervasive esp when demand for credit is insufficient (as exhibited by the low interest rate -- esp. in real terms -- for riskless investments).
2) Bad actors (agent issue) determine how weak credits are distributed.
I would think the first factor is causal to a credit crisis -- it ensures that lax credit management exists widely since otherwise monetary policy would be ineffective. Factor two shapes the exact path of credit boom and bust and how resources get allocated or mis-allocated.
Posted by: HZ at September 9, 2009 02:00 PM
Memory serving, the GS/Berkshire offer was rejected by LTCM, not the FRBNY.
Which is a rational move in the "Gambler's Paradox": if you're likely to lose anyway, taking a sure loss (a small payout and assimilation into GS) is less traditionally considered less "rational" than Betting the House on a Fed bailout.
Posted by: Ken Houghton at September 9, 2009 05:00 PM
Simon: If you've read at least 1000 words on the Russian crisis of 1998, anywhere, you've read that it was triggered by the 1997 Asian crises. And if you've read at least 1000 words on the Asian crises of 1997, anywhere, you've read that they involved the collapses of real estate bubbles. This is so uncontroversial, I've never even heard of there being alternative views.
Posted by: Tom at September 9, 2009 05:06 PM
SvN: Excellent post! And thanks to Menzie Chinn for the invite.
Some opine that high oil prices were the catalyst of the 2008 US financial sector crisis. Yet, from my admittedly weak knowledge of recent financial crises, energy price shocks were not a factor in pre-2008 crises. Am I safe in assuming that there are no informative empirical regularities or stylized facts with respect to specific evens that catalyze financial crises?
Posted by: GNP at September 10, 2009 10:03 AM
I would love to see an analysis over much much more than 20 years!
Posted by: Adrian Scott at September 10, 2009 02:02 PM
Things have been going downhill ever since they killed off the Templars and the Templar style of 100% reserve banking back in the 1300s.
Posted by: Cedric Regula at September 10, 2009 04:08 PM
The only problem with your conclusion is that the LTCM crisis was contained to the hedge fund itself and a few of it's clearing partners. The economy barely felt a ripple.
Posted by: Ian Sassoon at September 11, 2009 10:03 AM
Your memory matches mine, but apparently not DickF's.
I'm glad we agree that the Russian collapse was not caused by a Russian real estate bubble.
So who exactly are you saying claims that the Russian crisis was caused by a real estate bubble? As for the Asia crisis, the use of the word "caused" is not optional; "involved" is not a close substitute. The concensus view on the cause of the 1997 Asia crisis focuses on international capital flows and emphasizes governance issues; real estate - not so much.
Yes, the LTCM crisis was nicely contained (at no cost to the taxpayer, as I noted.) So where's the problem with the argument?
Posted by: Simon van Norden at September 13, 2009 03:43 PM