 Mark Hulbert is the editor of The Hulbert Financial Digest (HFD), a report on investment advisory newsletters. In April 2002, the HFD became a service of Marketwatch.com. Constructing hypothetical, model portfolios according to the advice contained in those advisories, the HFD provided the first objective means with which to compare the predictive ability of investment newsletters. It is the standard source for comparative performance data, drawing upon nearly 30 years of historical data. Mr. Hulbert is a regular columnist for the Sunday edition of The New York Times business section and is on the editorial board for the Journal of the American Association of Individual Investors.
We spoke with Mark Hulbert on March 27, 2008.
As someone with a highly analytical long term perspective on the markets, is the sub-prime crisis fundamentally different and significantly more severe than other recent crises in the US markets (S&L, LTCM, dot-com crash)?
I’ve been doing this for 28 years, and every crisis seems different from the one before it. But there is always similarity between them. In 1980, when I started out, we were worried about solvency because of third world debt that many said would not get repaid. There was a strong sentiment that we could avoid going bankrupt. The same thing happened with the Savings and Loan crisis, but somehow we muddled through. Martin Zweig, who used to edit one of the newsletters we follow, is credited with the adage “don’t fight the Fed.” I believe he is right. The Fed has more tools in their arsenal than we probably know about. They may not be right on every decision. But the intelligent bet is that the Fed will figure out a way to pull the rabbit out of the hat.
The data shows that the newsletters we track, on balance, are more often wrong than right. They are a contrarian indicator, and right now they are bearish, signaling the market is at its bottom. To bet the market will go down further says those who are more often wrong will get it right this time. In January they did not think we were at the bottom. This was when the scandal surfaced about the Societe Generale rouge trader, and that didn’t scare newsletter editors. So they were wrong then, reinforcing their value as a contrarian indicator. The editors are not as bearish now as they were last fall. At that time, they said it did not look like a bottom, and they were right. So they are not a perfect contrarian indicator.
Instead of taking the contrarian consensus view, another perspective is to look at those that have the best and worst records, in terms of their predictive abilities. Right now, those with the best records are significantly more bullish than the bottom performers. The qualifier is that they were also more bullish last fall (and were wrong). This is just another tea leaf, and should not be used for making market timing decisions.
What are editors saying about whether a continued decline in the dollar would be healthy for the economy and the markets, or would it be too inflationary?
Interestingly enough, there is a lot of parochialism among newsletter editors. Most never comment amount about the dollar or, if they do, they don’t cite it as a source of concern. Editors will talk about the effect on individual companies, but generally do not get into macro analysis. That has been true for a long time. I expect that attitude will gradually change, as the dollar becomes less of a reserve currency. There has not been a lot of incentive to worry, because the Fed could make the dollar go down and the rest of the world swallowed it. That will be less true in the future.
How have newsletter editors reacted to the credit crisis, especially the liquidity issues facing the financial sector?
No newsletter editor has changed his or her mind because of overall credit or liquidity issues in the market. The typical editors’ forecast is based on other market indicators. If an editor is inclined to be bullish, it is based on an undervaluation of assets, and vice versa if they are bearish. Editors will address writedowns as buying or selling opportunities, but do not provide analysis at a causal level.
In your columns, you report on a number of indicators that can be predictive of market direction. Recent ones have included corporate insider trading activity, trends in corporate financing, and of course sentiment from the newsletters you survey. Which of these indicators do you believe are the most reliable, and what are they saying now, with respect to the direction of the market over the remainder of this year?
We have not done the statistical work to test everything. The important question to ask is the time frame over which these indicators tend to work. The newsletter sentiment indicator – the contrarian analysis – is a short term indicator, predictive at a three month time horizon. The bottom they are now signaling is relative to the next three months. Insiders are predictive at the 12 month time horizon, because they act early to avoid the appearance of impropriety. The best and worst performing newsletters provide indicators that are predictive at a similar 12 month time horizon.
Insiders were bullish in October of last year, but not as much as today. Going back 12 months, they were modestly bullish, and market was down 5%-6%. They were wrong, but not spectacularly wrong.
Interestingly, insider activity and newsletter (contrarian) sentiment are now saying similar things. But others might reach a bearish conclusion, and it is never the case that all indicators point the same way.
Yale professor Robert Shiller has shown that ten year normalized P/E ratios are predictive over a ten year horizon, although this indicator has been wrong most of this decade. [Ed. Note: see our article on this subject.] It is modestly predictive over a one year horizon. The problem is that, with 140 years of historical data, there are not that many non-overlapping ten year periods.
Shiller’s indicator was right in the 2001 to 2002 timeframe. The scary thing is, at the bottom of that bear market, valuations were still higher than at the high of other bull markets. This suggests the subsequent bull market in this decade was cyclical, and not a major bull market staring with a low valuation. [Ed. Note: see our article on this subject.]
In the newsletters you monitor, several appear to have offered consistently reliable guidance over long periods of time. Is there any common denominator among these newsletters, either in their analytical approach, the rigor of their analysis, the style of investing they advocate, or the people that write them?
I have given a lot of thought to this over the years. I expected a different answer. It turns out the most compelling common denominator is the top newsletters consistently follow a discipline. A disciplined adherence to a strategy works, as long as it is a decent strategy. A lot of times, the newsletters at the bottom jump ship as soon as their strategy is out of favor, and end up selling low and buying high. They are always being whipsawed in and out of the market.
I believe this is a profound observation. It is not the strategy itself that makes the difference. It is the discipline, especially when one is out of sync with the market. For every strategy, be it growth or value, large or small cap, there is a newsletter that has done well and one that has done poorly.
The academic literature says value and small cap trump growth and large cap (on a risk-adjusted basis). But there are a lot of issues with the academic work. Studies have shown that the “value effect” is heavily reliant on a small number of stocks without much liquidity. Plus, if you eliminate January from the Fama-French database, there is no difference between value and growth in their model.
You write regularly on topics that are at the forefront of academic research, some of which purport to identify ways to outperform passive index investing. Yet, you advocate (at least for individual investors) a passive buy-and-hold approach. Is there any academic research, particularly in the last couple of years, which has persuaded you that individuals can take a more aggressive approach?
This is a very complex question. The problem with “buy and hold” is not a statistical problem; it is a psychological problem. Most investors find it intolerable to buy and hold through a bear market. It might be statistically inferior to engage in active management, but they can live with it. Based on the track record of the newsletters we follow, it is better to have an inferior strategy and stick with it, than to overreact when the market turns against you.
No recent academic work has changed my fundamental attitude. Buy and hold is hard to beat.
I wrote about a recent study by Dartmouth Professor Kenneth French. He shows, as percentage of total market cap, the amount spent on active management has remained remarkably constant over a long period of time. This is despite significant changes that, for example, have narrowed bid-ask spreads. Another important phenomenon is retail investors, as a whole, are holding a smaller share of assets, while institutional ownership is growing.
Many believe hedge funds earn high fees by profiting from the mistakes of individual investors. In order for this to happen, given the portion of assets held by individuals, French shows that retail investors are sacrificing 1% in annual returns through active management just for hedge funds to earn their fees. The actual sacrifice is greater to the extent that hedge funds earn alpha.
In academia, it used to be the only way to get ahead was to say the markets were perfectly efficient and couldn’t be beaten. Now, that is not the case. Almost all the papers are finding anomalies that cannot be explained by the overall model. But the magnitude of these anomalies is modest and it leaves open the question of whether it is economically worth it.
Have you adjusted your own asset allocation or investment strategy as a result of the sub-prime crisis?
No. I am not an active investor. Most of my assets are in a 401(k) plan and are highly diversified through funds.
Display article as PDF for printing.
Would you like to send this article to a friend?
Remember, if you have a question or comment, send it to
. |
