Thursday, September 13, 2007

A certain point of view

Yesterday's post set off a flurry of comments, most of which were telling me how wrong I was. But fortunately it was all very reasoned, and I'd like to thank every one who posted a comment for contributing to the blog.

I wanted to reiterate that I don't find anything wrong with being bearish. While I'm not terribly bearish right now, I do think the odds of recession are fair, and perhaps the stock market hasn't pricing that in properly. I honestly don't spend a lot of time thinking about whether the stock market is fairly valued or not. I do spend a lot of time on credit spreads, and those tend to correlate with stocks, but that's honestly about it. For what its worth, the money I've allocated to stocks stays invested at all times, and I rebalance regularly, regardless of my market view. I have some trading strategies I employ for my own money, involving all kinds of cute things like technicals and leverage and short selling, but its all bond related. I'm not a stock guy. I just know that stocks usually go up, and I like those odds.

There is a real difference between being bearish because some asset class is currently somewhat over valued and expecting a financial apocalypse. The former is looking for a correction, which are fairly common. The later is looking for something that has very rarely or possibly never happened before.

And here is the problem with forecasting something that hasn't happened before. You don't know when to conclude that it isn't going to happen. Take consumer debt loads. By some measures, consumer debt is extremely high. This leads some to conclude that a day of reckoning is forth coming. But the problem is that consumer debt has been high for a long time. By at least one measure consumer debt has outpaced GDP by a full 1% over the last 20 years. So if one held that consumer debt was too high, you could have held that opinion for the last decade. And yet until just now, there hadn't been any problems with consumer debt. If you had stayed out of the stock market from 1997 to 2007 you would have missed out on a 92% return. This despite a
historically bad bear market in 2000-2002. And even today, the jury is still out on how badly this will impact the stock market. So far its been very little.

As some commenters pointed out, bears also need to know when to get back in, which is very difficult. Say you had been bearish on stocks because you felt they were "irrationally exuberant" in December 1996 (Greenspan coined the phrase in a speech on 12/5/96). From 12/5/96 to 3/31/00 the S&P 500 advanced 111%, or about 25% annualized. But you stick with your theory, its just getting more irrational, you think. Then things start turning your way. From 3/31/00 to 9/30/02, stocks fall 44%. Do you re-enter the market? Well the S&P was at 744 on 12/5/96 when you decided to go bearish. It hit a low of 777 on 10/9/02. But still above the level where you determined things were "irrationally exuberant." Can anyone reading this honestly say they would have known to exit their short then? Of course, we know the rest of the story. Stocks are up 15% annualized in the nearly 5 years since 9/30/02.

Several commenters advanced something to the effect that periods of financial innovation end badly. Let me give you an example. In the 1980's part of the stock market boom was the burgeoning LBO market, which was made possible by the growing junk bond market. The junk bond market as we know it today was basically invented by Drexel Burnham and Michael Milken. Prior to this "innovation" junk bonds were almost exclusively fallen angels.

As readers undoubtedly know, in 1989, Milken was indicted on various securities law violations. Drexel would declare bankruptcy in 1990. Drexel had dominated the market making in high-yield debt, and therefore their exit ushered in a period of horrible liquidity in junk bonds. It would have been very easy to conclude that the popularity of junk bonds was just a fad, and that we would soon return to the previous norm of junk bonds being mostly fallen angels.

Meanwhile, junk performed very poorly. From September 1988 (when the SEC first sued Drexel) to February 1990 (bankruptcy) junk bond prices fell more than 12%, based on the Merrill Lynch High Yield Master. (Total return was still positive at 1.6%). The S&P 500 was up more than 26% during this period. For the rest of 1990, junk continued to fall, losing another 12% price wise, and falling about 1.7% in total return.

Whatever happened to the junk market anyway? Oh yeah, its still going strong. In calendar 1991 the high-yield market returned 39% and hasn't really looked back since. See, the innovation was a good one. Was Milken and Boesky and those guys all crooks? Sure. Did it end badly for them? Hell yeah. Would you have been wise to avoid stocks (or high-yield bonds) even had you predicted their downfall? Only if you timed it just right.

15 comments:

Anonymous said...

"flurry of comments, most of which were telling me how wrong I was"[tddg]

You're wrong! Most of the posts confirmed your premise that a rational approach is preferable to the absolutist ultra-bear ideology.

You must be counting posts that repeated the same assertion multiple times.

"I honestly don't spend a lot of time thinking about whether the stock market is fairly valued or not"[tddg]
For once, I can help you out. The stock market right now is by no means overvalued. But it does pay diddley and his brother squat for dividends, with respect to historical averages.

"the money I've allocated to stocks stays invested at all times"[tddg]
I would read any comments you have to make about deciding on allocation between stocks and your bond-related "cute things". Also any comments on your rebalancing criteria.

"Whatever happened to the junk market anyway? Oh yeah, its still going strong...the innovation was a good one"
*grudging compliment* this kind of insight provokes return visits to your blog. I've been wondering about the similarities between (Milken & junk bonds) with (rating agencies & structured mortgage backed securities) and hoped to hear a paid professional address this. I'd thought that Milken was originally correct. Junk bonds offered superior returns. Then, once everybody got in on it, junk bonds changed, and they no longer offered superior returns. After things settled down, the high-yield market was restored as a viable sector. Won't the structured mortgage backed securities turn out the same?

Accrued Interest said...

Maybe I exagerated a bit. I just want it to be clear that I don't mind criticism.

I have my personal money set up into different pots based on my liabilities. With my sort of long-term money, either for retirement or without a specific purpose, I keep a 50/50 split between bonds and stocks. However my bond money is actually highly speculative, and is basically my own personal hedge fund. So really I have 100% in risky investments in my long-term portfolio. I rebalance back to 50/50 each month or when I have cash flows. If I had a zillion dollars, I'd do the same thing.

For what its worth, I have a large theoretical value in my business, but I ignore that when thinking about asset allocation.

I completely agree with your last comment. Milken may have been a slimeball, but a viable junk market has been a net positive for the economy. I think CDOs are a net positive for the economy. I view all this as growing pains.

Anonymous said...

"" I think CDOs are a net positive for the economy. I view all this as growing pains. ""

In 1992-93 my mom bought a bunch of CMO's. I think they were somewhat new back then. Her nephew was her broker. They promptly fell 20% in value in 1994. She hung on, collected the coupon and they all eventually got called at par a few years later.

Accrued Interest said...

Yeah CMO's kind of went through the same thing, although it wasn't a credit problem it was a speed problem.

Anonymous said...

""Yeah CMO's kind of went through the same thing, although it wasn't a credit problem it was a speed problem.""

The value of the CMO went down and the supposed life of the bond was greatly extended, IIRC. Is that what you mean by speed?

knzn said...

I don't think anyone can realistically expect to increase their total return by timing the (unlevered) aggregate stock market. The stock market just has this amazing underlying tendency to go up, and that tendency seems to randomly reassert itself at unpredictable times. But maybe one can reduce the volatility of return by staying out when the risk of a big drop is particularly high. (For example, a reasonable person might have gotten out in 1996, when the S&P500 P/E went above 20, and then maybe come back in in 2005, when it dropped below 20 again.) If you can reduce volatility, then you can increase your total return by using leverage that otherwise would have been too dangerous (or, e.g., by staying 100% in stocks when your usual risk tolerance would have called for some allocation to high-grade bonds).

(For the example I gave, it looks like volatility-adjusted returns were better for Treasury bonds during Nov96-Feb05 than they were for stocks, so a levered T-bond position would have outperformed an equally volatile stock position.)

Accrued Interest said...

knzn: I think that's a lot more reasonable than to say you can actually time the market.

Dave M.: See http://accruedint.blogspot.com/2007/07/how-does-cmo-work.html.

CMO's are designed to dampen prepayment risk, but when prepayments are extremely high or low, the "modeling" breaks down and prepayment risk actually becomes greater in the CMO (or at least as bad) as the underlying MBS.

Accrued Interest said...

You know, my bottom line is that I believe in the long run, people get paid to take risk. If you can figure out ways to systematicaly take more risk, you are likely to make more money. Not all risks fit the definition of a systematic risk, mind you, but that's my philosophy.

Anonymous said...

"Whatever happened to the junk market anyway? Oh yeah, its still going strong..."

lol -- i don't disagree. but one might anticipate that junk bonds and CDOs alike are really all part of a longer-term trend of increasing the gearing of the capital markets. the function both and many other innovations (from fiat currency to SIVs) essentially perform is to increase effective money supply by multiplicative leverage.

the core questions being posed by many a permabear are these: could/does that process (which has been in place in this incarnation at least since the 1970s) have a secular endpoint, and can run in reverse?

i think one would have to answer, regardless of disposition, obviously yes. of course, inflection points in such secular trends are (as you note) very rare, and it cannot pay to sit short in wait of it year after year. but that does not make the permabear argument irrelevant, and insuring against the unlikely but devastating is not unwise imo.

Accrued Interest said...

I guess my view is that investing is a game of probabilities. (see http://accruedint.blogspot.com/2006/09/mail-time.html)

Its not a game of possibilities. So I never would say that a secular reversal of the long leveraging of the economy is impossible. But I prefer to look at debt levels are they are now and consider whether its possible to service this debt. How debt levels compare to the past isn't all that relevant if, as a whole economy, we can service our debts.

Its not unlike looking at Cisco Systems. In 1999 they had no debt. Now they have a fair amount. Should I be worried about this? Well, not too much, because they have a level of debt they can easily service.

Anonymous said...

anon said.."but that does not make the permabear argument irrelevant, and insuring against the unlikely but devastating is not unwise imo."

Yes, but in a devastating event, the chances are good that the counterparty insuring you will not stay solvent long enough to pay you off.

And, for that matter, in the case of the permabull's case of financial armageddon, does it really matter if you have your money in a money market fund (or in your mattress) instead of stocks? We'll all be on the barter system and your dollars will have no value anyway.

Anonymous said...

Dear TDDG,

I wrote an essay about something similar to this titled Financial Aphasia (http://bluecollardollar.com/financial_aphasia.html)


But mostly, I wanted to compliment you on your project. Keep up the good work!

Best,

Paul

Paul Petillo
Managing Editor/BlueCollarDollar.com
http://bluecollardollar.com
http://retiringwithaplan.blogspot.com

heydood said...

The only problem that I have with your analysis is that most of your examples are from the last 30 years or so. This is amazingly common in the media and financial blogs. It is logical, as most of the writers/analysts tend to focus on events in their memory. That is why most "financial armageddon" events only occur every 60-80 years or so, as those who lived through them have died off and those who didn't make the same mistakes. Great blog.

Accrued Interest said...

Dan:

Thanks for the kind words and for reading and commenting.

There is a legitimate question as to what time periods in history remain relevant and what is no longer relevant. For example, is anything under the gold standard relevant any more? Or pre-Volcker, when the Fed was less focused on inflation first?

I think the answer is its relevant in some ways, but it most ways its not. By this I mean, a 1970's type oil shock could happen today. But the Fed's reaction today would be very different. So it wouldn't be correct to assume that the economy's reaction to another oil shock would be comparable to the 1970's.

Accrued Interest said...

Paul: Thanks for the link. Keep commenting!