POST OF THE DAY
Berkshire Hathaway
S&P Valuation and Hedging

Related Links
Discussion Boards

By mungofitch
January 15, 2010

Posts selected for this feature rarely stand alone. They are usually a part of an ongoing thread, and are out of context when presented here. The material should be read in that light. How are these posts selected? Click here to find out and nominate a post yourself!

Question 1: When last do you feel the S&P 500 was at or below fair value (ignore the very brief dip early last year)?

Give or take a bit of slop, the last few occasions would be Feb 2009, for only about 1 month Jan 1990, for only about 1 month Nov 1987, and pretty close for about a year afterwards Oct 1986, and pretty close for months on either side Mar 1986  ...and undervalued for many years before Mar 1986, back to about Nov 1973. Here's an attempt at an illustration. The vertical lines are decades. This approach has several subtle assumptions built in, mostly:

�(1) - the "green" line in my trend-earnings graph is right. This is definitely true in the past, but has been eyeballed a little bit for the most recent period. Each point is the average in the years before and after, and clearly we won't know for a few years the exact result of that for today. Note that the eyeballing has put the green line if anything a little bit high. This generosity in the earnings estimate means that the bearish case is weakened. So, if you think the corollaries of using the green line make a good bearish case, and you think the green line is too high, then the bearish case is even stronger. (2) - the "fair" value of the S&P is whatever level which would make the  multiple of trend earnings equal to its average since 1940 (13.36). This is just a pinch higher than its multiple since 1871 (12.87). For the record, you could do worse than remember the number "about 13". (3) - the result you get in the fair value graph depends a little on how much smoothing you do on the very long run cycles of earnings. I smoothed quite a bit, only 4 squiggles in 150 years. With less smoothing you get a less certain and squigglier line, but it is less likely to give you the conclusion of those very long periods of overvaluation and undervaluation.

Here is an interesting image on a related line or reasoning. This turns the logic around: since the current point of valuation is known but the trend line is not known for sure, where would the trend line have to be placed if the current price were "fair"? If you can not imagine the correct trend line passing through that point, then you are sure that the market is overvalued or undervalued as the case may be.

Question 2: Based on the fact that the index can remain overvalued for  extremely long periods of time, do you view all hedging costs (put  option premium, etc) completely like insurance - i.e. the chance of  recovering any of the premium is small, but you structure it so that if  you do get a payout, it is big.

I hedge. As you might expect, I mostly lose doing it. I can't really recommend it. The approach I have adopted recently is painful, but should reduce the long run cost of being hedged. Specifically, I have short index futures equal to a certain percentage of my portfolio, say 40%. I hold these futures all the time. Given that I expect the index to be no higher in several years' time, I expect this hedge to cost me net nothing over that time frame. It will be (has been) very painful during big bull markets, but will give me back almost the same amount of money in the next bear market. My overall portfolio value will be smoother, and I will be less scared using the small amount of leverage that I foolishly have in place. (Do as I say, not as I do!). I also have a few out-of-the-money index puts. I mostly try to buy them when they seem relatively cheap (bullish days of low VIX, high index levels, and complacency, like today). Generally these lose money, though occasionally pay off. My best month ever was Sept 2008.

One of the hard things with puts is that you are implicitly making a bet not only that the market will go down, but also when it will do so. Plus, you have to have a strategy of what to do if it does go down.  When do you sell your puts to realize the profit on them? Do you sell a corresponding amount of long positions to stay hedged? My current strategy is this: don't rely too much on puts. Keep a few lying around at a wide variety of strike prices like lottery tickets. Sell each one as it goes into the money (that's when its time premium is maximized), or if/when it doubles in value from your purchase price. Keep a put on the things most likely to go down based on what you know. (In 2000 it was better to have Nasdaq puts than Dow puts). Right now I have mine on financials, XLF, since I think some weakness might be coming.

Another approach is to pick a very basic (even crude) market timing system with extremely rare signals and behave slightly differently based on whether the system is bullish or bearish. A random example: boards.fool.com/Message.asp?mid=28220449. This one is based on the observation that the market tends to be a whole lot weaker if there hasn't been a new recent high lately. When a bear signal comes, either add hedging or sell stocks.

A much more rational approach is to pile up cash till you see things that are really cheap, then buy them, and forget about hedging.  This is by far the most difficult alternative in practice. Another very good approach is to delegate the problem of being patient and buying stuff cheaply to someone smarter than you are, like Berkshire. This is the easiest alternative.

Jim