As I am sure many Americans have done over the past few weeks, I have had many discussions with friends about the recent stock market gyrations and potential economic fallout from the financial crises. To get some perspective, I went to Yahoo! Finance and looked at a chart of the DJIA prices from 1929 to the present. I took a rough estimate of the high price before a precipitous fall and then tried to find the low before things started to get better. Below are a the results of this exercise, which are by no means scientific. You can find the spreadsheet here.
The first three columns are fairly obvious. In the third, I applied previous percentage loses to the relatively recent high fro the Dow Jones Industrials (~one year ago). This is not a prediction, but rather just fodder for the discussion.
20 Comments
October 7, 2008 at 6:15 am
I’m always amused at analysis of the stock market that attempts to organize predictions around what happened in the past. When it comes to the stock market, I’m a big believer in chaos theory — there are innumerable factors that can all have an impact on the result at any given moment.
I don’t think this downturn will match previous experiences. It’s possible that because most people do a similar analysis that once the market hits 8,459 that they will start buying and this analysis will be spot on. But I think that almost every aspect of this market, which is composed of millions of actors, is different than any market before it. Just as every market before it was as well.
A more interesting analysis would be to perform a mass analysis of every stock (that is analyzable) using estimates that reflect the uncertainty of the markets today (down, pick a #) and fundamental valuation benchmarks (6X EBITDA Est. steady state business, 13X EBITDA Est. growth — pick your own), calculate the necessary prices and reconstitute the index. That would give me a great deal of comfort of the bottom of this decline.
October 7, 2008 at 6:16 am
Why am I so not surprised YOU are the one who is running such an analysis!! This is very insightful! And, for some reasons, it reminds me of the jelly beans exercise;)
October 7, 2008 at 2:26 pm
Thanks Jake. As I noted, this isn’t a prediction, but rather an analysis that shows what previous downturns would have equated to in today’s terms.
Great to hear from you fffabulous! As a former engineer, I know that math is your friend
Also, note that I included the most recent events in the average calculation — sort of double counting. If I exclude the most recent events from the calculations, the adjusted mean is 58% with a DJIA price of 8160.
October 7, 2008 at 3:47 pm
Understood that this is just analysis, but then we have to get all “statistically” and question whether the first observation is an outlier, right? Maybe the structure of the market and participants actions are materially different than the last five and therefore be excluded. And then we have to have a conversation about whether an average is the appropriate calculation.
I only bring up the objection because of people’s fascination of using math to model aggregate human behaviors has recently led to pretty bad outcomes (i.e. ISV ratings). It’s as if we’ve gotten to lazy to do the qualitative analysis required so we resort to mathematical shortcuts. Only they’re not shortcuts because they don’t lead to any answers.
I’ve done a ton of modeling in my career as a banker and turn-around specialist, and the one thing I know is the more modeling I do, the more I’m convinced that understanding the motivations and incentives of the human actors within the scenario will lead to a much better understanding than the results of the models.
Our current crisis is a crisis of financial models, both from the ratings agencies side to the portfolio risk measurement side. Everyone relied on the models, but no took a step back and said, “Why are we doing this and is it sufficient?”. So I hope my comments are seen as encouragement to apply the same insight you apply to your other modeling exercises and come up with something more meaningful in this situation…
October 7, 2008 at 5:55 pm
Jake, so what is your best guess with respect to the local minimum of the DJIA during this downturn? For the record, mine is 8,000 (although it may take some before before we get there).
October 7, 2008 at 7:02 pm
I don’t have the resources to run the model that I outlined earlier. Whatever that came up with would be a good floor in my opinion. But then I would factor in that there are many equity investors that will be “seeking the bottom” and will want to be in the market for an eventual upturn and therefore may stop valuations from declining to absolute intrinsic valuations.
So, faced with the inability to model the entire market, I’ll resort to making intrinsic value estimates company-by-company and hopefully purchasing companies at a discount to those estimates. But I’m sure there are large hedge funds that do have market models that calculate intrinsic values based on the latest data for each stock and recompose the index to identify a theoretical minimum value. It would be interesting to see what their results look like.
Is that a good enough non-answer for you?
October 7, 2008 at 8:10 pm
I feel like Katie Couric having a conversation with Sarah Palin…. Why don’t you have John get back to me?
October 7, 2008 at 9:48 pm
Mike,
so the big question is are you going to sell?
October 7, 2008 at 9:58 pm
Ouch. Although I’m not sure I’d be much happier being compared to Katie either…
October 7, 2008 at 10:59 pm
Jake, touché.
Sean, nothing left to sell…
October 9, 2008 at 6:23 pm
In case you’re interested, a great overview of the problems with risk modeling from the former G.C. of LTCM.
http://www.washingtonpost.com/wp-dyn/content/article/2008/10/01/AR2008100101149.html
October 10, 2008 at 1:56 pm
We bounced solidly off of 8000 this morning, Mike. You may be vindicated yet!
October 11, 2008 at 4:05 am
I think the low will be 5500 - 6000 though it may take another 6 months or so to get there.
October 12, 2008 at 6:35 pm
It would be interesting to use the same approach to estimate how long it will take the economy to recover.
October 23, 2008 at 9:47 pm
I keep a copy of the Shiller PE10 spreadsheet with a few of my scenarios here:
http://spreadsheets.google.com/ccc?key=pLJJLZL5xN362xBFVaibEYA
The PE10 (sometimes called “Graham’s PE”) has been historically very high over the past decade. In other downturns, it’s gone down into the 6-8 range. Last time was in the early 80s. We’re currently ~15. S&P of ~400 is a reasonable worst-case scenario. That’d be DJIA of ~4000. You could argue that PEs should be growing slightly over time, but the 80s wasn’t all that long ago.
Shiller’s paper on valuation ratios:
http://www.econ.yale.edu/~shiller/data/peratio.html
October 24, 2008 at 4:23 am
John, great comment. I believe that Shiller is using current price for the “P” and historic earnings for the “E” in P/E. Some (academic economists, for example) would argue that stock prices are a prediction of future earnings potential and that for P/E to be meaningful one would need to compare today’s price against “expected” future earnings. While it’s easier to compare against past data, I wonder how the same analysis would look based on the consensus forward earnings expectation of the market.
Having said that, my back-of-the-envelope analysis is entirely based on past past data, as Jake pointed out…
Bottom line — if the DJIA hits 4000, I’m putting everything I own up for sale on Ebay so that I can invest in stocks
October 24, 2008 at 3:05 pm
John, the flaw that I see with assuming that we will retest those kinds of lows is that it ignores the advancement in corporate governance across the market. I would argue that multiples should float higher over time as the sophistication of management increases. Granted, you can argue that management is worse off given the general mood today, but I believe that management teams are much more effective at managing cash flow and performance than they were 20-50 years ago. Therefore, they have a somewhat lower risk which should equate to higher multiples.
Of course if banks lever up to 30X with inadequate risk measures, my argument looks kind of stupid.
November 3, 2008 at 2:17 pm
Thought you might appreciate this:
http://online.wsj.com/article/SB122538449722784635.html?mod=testMod
November 13, 2008 at 9:17 pm
Another hard bounce off of 8,000.
November 15, 2008 at 4:02 am
“I am a strong believer in the importance of models, which are to our minds what spear-throwers were to stone age arms: they greatly extend the power and range of our insight. In particular, I have no sympathy for those people who criticize the unrealistic simplifications of model-builders, and imagine that they achieve greater sophistication by avoiding stating their assumptions clearly. The point is to realize that economic models are metaphors, not truth.” -Paul Krugman
Leave a Reply