August 18, 2007
Where's the risk?
Usually an economic downturn is associated in a big increase in the spread between corporate and Treasury yields. This spiked pretty dramatically last week, but still has a long way to go.
The above graph plots monthly averages of the corporate-Treasury yield spread through July. Going back to 1953, this spread averaged 170 basis points, and typically rose over a hundred basis points during an economic downturn. Insofar as there is a higher probability of default during an economic recession, even risk-neutral investors would require a higher yield on corporate debt when more businesses are failing.
This spread spiked substantially last week, as long-term Treasuries fell 20 basis points while the Baa yield showed little movement. Even so, the current spread of 206 basis points is not much above the long-term average or the short-term range we've seen over the last two years.
If the financial turmoil over the last few weeks reflects fears of a significant economic downturn and financial distress, I would have expected to see an even bigger spread at this point.
Posted by James Hamilton at August 18, 2007 04:08 AMdigg this | reddit
Listed below are links to weblogs that reference Where's the risk?:
» Risk in Financial Markets: Going Up? from Businomics Blog
Jim Hamilton over at Econbrowser has an interesting post entitled Where's the Risk? He looked at bond yield spreads and didn't see any higher risk. I wanted to look at daily data, in both long and short-term rates. I found [Read More]
Tracked on August 21, 2007 11:45 AM
good point Dr hamilton.
But today's world is vastly different. Many investors fearing market collapse, switched over to short term treasuries (very quickly nowadays due to internet)hence widening the spread a little bit.
this is where cutting rates will turn out to be a bad policy.
we have asset hyperinflation (US net worth has balooned to 70 trillion in a very short time but lots of air below it, expect either substantial damage to that or dollar losing its value by ~50%)
Bet what outcome is preferred by the wallstreet fat cats ,wannabe free marketeers, always lining up for govt dole as and when a need arise?
Posted by: andiron at August 18, 2007 05:25 AM
James, are you suggesting that the hubbub in the financial sector is, at least at this point, disconnected from the rest of the economy? I can't tell whether stuff like this is a reflection that's starting to change or rent-seeking. Both, I guess.
Posted by: Dave Schuler at August 18, 2007 07:26 AM
This raises a couple of questions.
So why would the rate on T-bonds drop? A government surplus with lower demand for borrowing? Cash-outs looking for buying opportunities elsewhere or to cover near-term cash needs?
Are there no time lags between these financial products?
Posted by: Joseph Somsel at August 18, 2007 08:08 AM
Great post, professor. Remarkable how a little common sense can put all this turmoil and all the current frenzy in adequate perspective. I would love to read your thoughts on the bid-ask spreads in the Treasury market and their predictive power. I would suspect that market events that eventually lead to recessions show up in the data in the form of sky-rocketing bid-ask spreads. We are simply not seeing this right now...
Posted by: Alberto at August 18, 2007 08:26 AM
What would happen to the curve above if the Fed did not step in and let the market worked its way out?
Posted by: Anonymous at August 18, 2007 09:09 AM
"the current spread of 206 basis points is not much above the long-term average"
The definition of "not much" is a matter of personal preference. When I look at the top chart, the only times I see a spread >200 basis points are during recessions, after recessions, just prior to recessions, and during times when there was a significant recession *somewhere* (for example, in 1986 there was a recession in the southwestern US, which is significant because it is part of the US; in 1998 there were recessions in Asia, which were significant because the crisis was so widespread).
Posted by: knzn at August 18, 2007 09:26 AM
knzn -- your point is confirmed by the fact that one of the tightest relationships in the economy is between the quality spread and manufacturing capacity utilization.
Typically the biggest widening of spreads occurs when treasury yields fall but rates for lower quality spreads do not follow suit.
Posted by: spencer at August 18, 2007 09:44 AM
I suspect that a stark widening of credit spreads is coming, professor, but maybe not right away.
This episode is different from past economic cycles in that mortgage spreads HAVE widened substantially, and the housing sector is in a recession and headed for something worse than that in 2008. Will an unusually severe housing recession in 2008 be enough to pull the US economy into a recession too? Considering that Consumer Spending constitutes about 2/3 of GDP and the linkage that's developed over the past 20 years between mortgage refi's and Consumer Spending, I'd have to answer: I think so.
One other key difference on this go-round has been the recycling of oil profits from OPEC members back into the capital markets. A couple of great reads on the issue include:
And in Randall Forsyth's "UP and DOWN Wall Street" column in Barron's, a hedge fund manager is quoted saying that foreign banks recycling trade surpluses (Asia) and recyling petrodollars (European) practically had standing orders for their trading desks to buy any high yielding US debt rated Triple-A . . . . . through the magic of CDOs, lots of lousy US credit was packaged and resold to the yield hungry recyclers as high grade investments.
I think (and hope) that the CDO mania is over, and if it is, one of the stranger sources for demand for junk debt may dry up entirely. When, if and as that happens, high yield yields should rise, maybe even fast and hard, depending on how many leveraged relative value fixed income hedge funds are left to have to sell into the decline.
Posted by: Anarchus at August 18, 2007 12:58 PM
Does the Phrase " the housing ATM is now closed"
mean anything to you?
Posted by: DH at August 18, 2007 01:00 PM
Stock market always leads corporate bonds. Corporate Bond investors are the dumbest of the dumb money. Long Live Fleckenstein. "Lord of the Bear Markets."
Posted by: pepto at August 18, 2007 01:15 PM
A question I would have is whether spreads are really comparable like this over long periods of time. Eg, that spike in spread around 1980 was occurring at a time when both inflation and interest rates were running around 15%. Is 3.5 (nominal) % spread on top of 15% nominal interest rate and real rates close to zero really comparable to 3.5% on top of 5% (ish) nominal interest rates and 3% (ish) inflation today. I'd say no - with inflation so high (and therefore necessarily uncertain), one might need a larger spread just to be sure one was going to get a real return at all. Thus this measure isn't isolating the "higher probability of default during an economic recession", but includes multiple factors.
Posted by: Stuart Staniford at August 18, 2007 01:34 PM
This rally caught my attention...
Ex ante why would anyone be interested in this chart?
Do you think it represents a "tradeable" assest of some sort? If you knew where it was "going to go" so to speak... is there anything anyone could do to take advance of it.. without taking on a host of other risks?
Comparing yield to maturity of two assest is comparing to worthless measures to each other and hoping that their failings wash out?
You know that the total return on a ten year year treasury is almost entirely driven by the reinvestment of the coupon.. i.e one's current view on the term structure or the forward path of rates..
The corporate even if such a thing could be traded with any size at all.. would and does depend no only on that forward term structure too.. and even more as it has a higher coupon to reinvest.. but on the cross correlation between the probability of default and the term structure of rates..
If you're worried bout the corporate's probability of default you're essentially implying that its likely maturity is much less than the 10 year term.. just like in mortgages (this might be a clue)....
One has to price in the correlation between the probability of default and forward rates..
Your thoughts on this would be welcome.. but lets assume that like me you haven't a clue.. so you have a huge variance term to worry about..
Now of course in the real world no one looks at this.. but everyone spends an inordinate amount ot time stripping out zero coupon models with varialbe rates of default and oh yes recovery..
Believe me I too wish there was some information in a nice pretty chart that one can get for free..
But alas.. the world is harder..
At best what we get is what we knew going in..
Things look riskier
By the way the critical driver is often the inability to go short a corporate even if you are sure its going to zero..
One easy test.. run this on a foward basis.
see what it cost you today to take on the bet..
being long a treasury verus being short a corporate.. if one sells a corporate today..
literally one will be able to invest the proceeds at less than 2% i.e the T bill rate for a month..
so one has to pay that huge coupon and receive perhaps 700 basis point less..
What happens if things get so bad there is a takeover.. and the Debt is made whole..
Big risk will then make the corporate defacto risk free..
One doesn't have to be Fisher Black to see that somehow equity volatility enters in to this...
Oh yes.. by the way did I forget taxes too??
Suggest you look at Elton and Grubers nice piece of almost a decade ago that looked at these things in a serious fashion..
Posted by: stan at August 18, 2007 01:42 PM
I couldn't easily find a description of the methodology in the computation of the Baa Seasoned Corporate index used. I suspect that there is a term-spread included in the graph - is this the case?
Posted by: James I. Hymas at August 18, 2007 01:45 PM
The relatively low corporate spreads reflect a combination of two factors:
2) Investment-grade corporate bonds have benefited from quality drift: a much higher percentage of total issuance than in the past is now junk.
3) Corporations have restrained capex spending (relative to out-sized profits) over the past five years, and have strong balance sheets to show for it. Years of peak profits make these corporations less vulnerable to financial shocks.
Looking at corporate spreads as a measure of recession risk, if you'll pardon the analogy, is like Wayne Gretzky skating towards where the puck just was. Much better to look at the locus of leverage accumulation, which obviously is consumer borrowing in the form of mortgages, auto loans, and credit cards.
Posted by: David Pearson at August 18, 2007 01:59 PM
The current unemployment numbers is severely understated.. the farming/housing sector does employ a high proportion of immigrants. Of late, there are news that they are remitting far less money back to their home country.. increasing protectionist trend emerging as well..
Signs of tougher times to come?
Posted by: Mike at August 18, 2007 06:47 PM
By this account, a parachute drop is risk free because we've successfully left the plane, and we haven't yet (1) discovered a failed parachute or (2) hit the ground.
So where's the risk in a parachute jump, by analogy?
We never know where and how much risk there is, until after we've reached ground. Even then, risk is an accounting of hard versus soft landings after the fact.
So it seems you implicitly think everyone has reached ground already, and therefore an accounting of risk can be made and conclusions drawn.
Given that there have already been a few hard landings in the last few weeks, I think the title question is jumping the gun by a long shot.
Posted by: KnotRP at August 19, 2007 01:11 PM
KnotRP, all I do is raise the question, why isn't the perceived risk priced in the corporate-Treasury spread? Do you have an answer to that question?
Posted by: JDH at August 19, 2007 03:47 PM
This may be part of the answer.
The increase in demand for Treasuries -- as we usually see in a liquidity event -- should have an effect on yields that depends on the current supply of Treasuries. Government debt levels are relatively high now (say compared to 1999), so that the demand increase will cause a smaller price effect today.
The importance of supply for the corporate-Treasury spread is documented in:
We show that low frequency movements in corporate-Treasury yields spreads can be quite well explained by the US Debt/GDP ratio.
Posted by: Arvind Krishnamurthy at August 19, 2007 03:48 PM
The markets heard your question JDH, and here's their answer:
Money market investors staged a dramatic flight to safety, knocking down yields on short-term US government debt in early US trading on Monday, as top Treasury and Federal Reserve officials continued behind-the-scenes efforts to shore up confidence in the credit market.
The yield on the one-month Treasury bill fell 160 basis points to 1.34 per cent in early trading. The yield on three-month Treasury bills tumbled to 2.51 per cent, 123 basis points below Friday’s close – a sharper fall than during the October 1987 stock market crash.
The scramble to obtain government paper at any price is a sign of extreme risk aversion, and suggests the Fed’s actions have yet to stabilise sentiment in the credit markets.
From the Financial Times.
Posted by: Stuart Staniford at August 20, 2007 11:48 AM
Actually, what interests me recently (and today) is the relative flatline of 90 day eurodollars v the "off the cliff" chart of the 90 day bill.
Yes yes yes this is primarily a good old American mess.
It was government polity to run the residential housing insanity flat out without any restraint whatsoever right up to the moment that they actually ran out of warm bodies to do the borrowing. And then they attempted to find more warm bodies through the immigration "grand compromise" but failed. And the game ended. Full stop. Commenting on this the head of a Swiss bank used the word "insanity" during the interview. How refreshing.
Housing (construction and valuation) is so integral to the proper operation of the American economy that the greatest probability is that government policy will shortly shift back to the promotion of Alt-A products and even subprime products, perhaps renamed in some bogus manner. Oh, and FF rates are headed back to into the 1-2 range needed to relight the fire.
Sounds insane...and it is. But it is the only trick we have...being the one-trick-pony that we are.
Posted by: esb at August 20, 2007 03:05 PM
esb, it's most definitely NOT just an American only mess . . . . . European Banks (especially the supposedly prudent Germans) have bought a ton of American-made CMOs/CDOs/CLOs in off-balance sheet conduits and are as vulnerable as lots of US institutions. In fact, Germany's already had TWO bank/S&L failures from "conduits gone wild". And apparently the residential real estate crisis looming in Ireland may ultimately have similar dimensions (relative to their economy) to ours.
This firestorm is getting really serious and even if the Fed started easing aggressively TONIGHT, I doubt if a recession in 2008 could be avoided (but I'd like to see them try) . . . . .
Posted by: Anarchus at August 20, 2007 04:07 PM
Stan's comment that this spread is not predictive seems right, but it still has some use. Though not predictive, it is a good co-incident indicator of perceived risk and often of recession. So co-incident is better than a lagging indicator, like the NBER's declaration that a recession has started - or ended. It would seem that some sophisticated math applied to this spread (and maybe others, like the spread with emerging country debt) would "declare" a recession earlier than the NBER - and maybe do it better than the circular analysis of historical GDP growth rates. Something to think about. For those of us who lack powerful regression tools, a quick monthly review of the spread numbers can inform one's gut about the likelihood that a recession is already underway. Look at the behavior in early 2000 in the detail chart above to get a sense of what to look for prior to the recession of 2001 and the market decline of 2001, 2002.
I do agree with James' comment that the recent financial distress could have been expected to created a larger spread. The fact that it didn't seems to say that the recent market distress is focused on the liquid trading of mortgage securities, and the companies who deal heavily in mortgage securities, and is not (yet) a market concern about manufacturing and non-financial services. These concerns may (I think will) come forward in the future, but we are not there now.
Posted by: Mike Laird at August 20, 2007 04:47 PM
> Do you have an answer to that question?
It's entirely possible to have a low spread between two highly risky investments. There is an implicit assumption in the conclusion that Treasuries are always and everywhere fixed at low risk, mostly because we're used to it being that way, and that if the spread is low, it implies the corporate component is low by association. Treasury risk is rising *up* to meet Corporate risk as we head into a recession. What this means to me is that there is simply no good place to preserve wealth at the moment.
Posted by: KnotRP at August 21, 2007 02:09 AM
Did subprime become less risky (low spread to prime) or did prime (ARM/IO/PO) become more risky, in the last few years....the spread between the closed, but I'm not sure I'd draw the conclusion that we've had a lot of low risk mortgage lending.
Posted by: KnotRP at August 21, 2007 02:14 AM
Jim, good subject to raise.
I put together a chart comparing commercial paper to T-bills at http://businomics.typepad.com/businomics_blog/2007/08/risk-in-financi.html
Posted by: Bill Conerly at August 21, 2007 11:49 AM
James Hamilton, regarding your analysis of GDP data to predict the onset of recession. Here are 4 measures that have predicted every recession in the past 40 years, when they all hit together. They have done so without any false positives. If it works, why not use it? This is from John Hussman, a former econ professor who now manages the Hussman Funds. The punch line? The probability of recession is high, but only 3 of 4 measures have hit, so the recession has not yet started.
Here are the 4 measures, again, thanks to John Hussman.
1) The "credit spread" between corporate securities and default-free Treasury securities becomes wider than it was 6 months earlier. This spread is measured by the difference between 10-year corporate bond yields and 10-year U.S. Treasury bond yields (or alternatively, by 6-month commercial paper minus 6- month U.S. Treasury bill yields). This spread is primarily an indication of market perceptions regarding earnings risk and default risk, which generally rises during recessions.
2) The "maturity spread" between long-term and short-term interest rates falls to less than 2.5%, as measured by the difference between the 10-year Treasury bond yield and the 3-month Treasury bill yield. A narrow difference between these interest rates indicates that the financial markets expect slower economic growth ahead. If the other indicators are unfavorable, anything less than a very wide maturity spread indicates serious trouble, regardless of unemployment, inflation, or other data.
3) The stock market falls below where it was 6 months earlier, as measured by the S&P 500 Index. Stock prices are another important indicator of market perceptions toward credit risk and earnings expectations. While the economy does not always slow after a market decline, major economic downturns have tended to follow on the heels of a market drop. Stock markets tend to reach their highs when the economy "cannot get any better" -- unemployment is low and factories are operating at full capacity. The problem is that when things cannot get any better, they may be about to get worse.
4) The ISM Purchasing Managers Index declines below 50, indicating a contraction in manufacturing activity. This index is strongly related to GDP growth, and when combined with the previous three indicators, has signaled every recession in the past 40 years.
Presently, all of these conditions are in place except the fourth.
Posted by: Mike Laird at August 21, 2007 06:25 PM