Swedroe: Don’t Go ‘Mad Money’

Data suggest Jim Cramer’s ‘Mad Money’ is a poor guide for investors.

Unlike many other personalities in financial entertainment, Jim Cramer, longtime host of the popular CNBC show “Mad Money,” manages a portfolio that invests in many of the stock recommendations he makes on TV.

Established in August 2001, with approximately $3 million, the Action Alerts PLUS (AAP) portfolio has been the centerpiece of Cramer’s media company, TheStreet, which sells his financial advice, giving a subscriber base that numbers in the millions access to each trade the portfolio makes ahead of time.

In March 2005, upon the launch of Cramer’s “Mad Money” show, the AAP portfolio was converted into a charitable trust, adopting the policy that any dividend or other cash distributions would be donated to charity. Cramer lists himself as a co-manager of the AAP portfolio.

Thanks to Jonathan Hartley and Matthew Olson, authors of the study “Jim Cramer’s Mad Money Charitable Trust Performance and Factor Attribution,” which appears in the Summer 2018 issue of The Journal of Retirement, today we can review an analysis of the AAP portfolio’s complete historical performance. We can also glean some idea of how Cramer’s TV stock picks may have worked for investors who listened to them.

Study In Madness

Hartley and Olson measured the performance of the AAP portfolio against various asset pricing models, including the single-factor CAPM (market beta), the Fama-French three-factor model (adding size, or SMB, and value, or HML), the Carhart four-factor model (adding momentum, or UMD), and five- and six-factor models that add the quality factor (QMJ) and betting against beta factor (BAB).

Their study covers the period Aug. 1, 2001, the AAP portfolio’s inception, to Dec. 31, 2017. Following is a summary of the authors’ findings:

• The AAP portfolio underperformed the S&P 500 Index’s total return both since inception and since the debut of “Mad Money.” From inception, the AAP portfolio provided a total return of 97% (4.1% on annualized basis), underperforming the S&P 500 Index’s return of 204% (6.8% on annualized basis).

• Over the full period, the AAP portfolio produced a Sharpe ratio of 0.16, half the size of the S&P 500 Index’s Sharpe ratio of 0.32.
• Relative to the CAPM, over its full history, the AAP portfolio produced an alpha of -2.92%, which was statistically significant at the 5% level.
• When analyzing the period January 2005 to December 2017, and adding the size, value and momentum factors to market beta, Cramer has a statistically significant negative alpha between 3-4%.
• The statistical significance of the negative alpha disappears when using the models that include the QMJ and BAB factors—the portfolio’s exposure to common factors explains its performance, not Cramer’s stock selections. The six-factor regression shows an alpha of 1.5%, though it was not statistically significant.
• The AAP portfolio has less exposure to the market beta factor than the market overall, contributing to its underperformance relative to the S&P 500 Index. This can be explained at least in part by the fund’s policy to not reinvest dividends—it holds cash to make annual charitable contributions (and thus should not be a reflection on Cramer’s stock-picking skills).
• The AAP portfolio shows evidence of exposure to small-cap stocks, growth stocks and stocks with low earnings quality (junk stocks).

The preceding findings on Cramer’s performance are consistent with those from prior studies.

Other Evidence

In their study “How Mad Is Mad Money?”, which appeared in the Summer 2012 issue of The Journal of Investing, Paul Bolster, Emery Trahan and Anand Venkateswaran examined Cramer’s buy and sell recommendations for the period July 28, 2005 through December 31, 2008. They also constructed a portfolio of his recommendations and compared it to a market index.

Their study covered 1,592 clear buy and 700 clear sell recommendations. They then assembled an equal-weighted portfolio. To try to capture prices available to the typical retail investor, they based their trades on closing prices on the following day. Stocks remained in the portfolio until there was a sell recommendation. They then adjusted the absolute returns for the portfolio’s exposure to the market beta, size, value and momentum factors.

Following is a summary of the authors’ findings:

• Investors are paying attention, as the stocks returned an abnormal and statistically significant 1.88% on the day following Cramer’s buy recommendation.
• The returns for recommended stocks were positive and significant for both the day of the show (0.38%) and the 30 days prior to the show (3.9%).
• However, the returns were negative and significant, at -0.33% and -2.1%, for days two through five and days two through 30 following the recommendation. After 30 days, the results are insignificant.

The bottom line is that, over this period, Cramer recommended stocks with momentum, both positive and negative. His recommendations affect price, with the impact reversing quickly, consistent with pricing pressure caused by viewers jumping on his buy recommendations.

Cramer’s sell recommendations also affect prices, though the impact doesn’t quickly reverse:

• The abnormal returns were negative (-0.73%) and statistically significant on the day his sell recommendations hit the market. They were also negative (-3.24%) and statistically significant in the 30 days prior. Again, this suggests Cramer picked stocks based on momentum.
• Unlike with his buy recommendations, returns remain negative and statistically significant for days one through five (-1.22%) and days one through 30 (-3.11%) following the sell recommendation.

Over the full period, a portfolio of Cramer’s picks lost 7.32%, slightly better than the S&P 500 Index’s loss of 8.72%. However, once returns were adjusted for exposure to the four factors, the results were all indifferent from zero. In addition, 98% of the returns to the portfolio of Cramer’s stock picks were explained by those factors. In other words, there is no evidence of any stock-picking skill—his picks are neither good nor bad. In the end, it’s just entertainment.

More Madness

The results of this study basically replicate the findings of a May 2005 study, “Is the Market Mad? Evidence from Mad Money” by Joseph Engelberg, Caroline Sasseville and Jared Williams, then three Ph.D. students at Northwestern’s Kellogg School of Management, including the fact that Cramer’s recommendations were momentum stocks.

They found that volume soars when Cramer recommends a stock. For example, the researchers found that, on the smallest quartile of stocks, volume is almost nine times more than normal on the day after his recommendation (and stays above normal for about four days, with the effect decreasing over time). Increased demand led to an overnight rise in prices of about 5% for the smallest stocks (where it can have the greatest impact), and about 2% for the authors’ entire sample of 246 unconditional recommendations made between July 28, 2005 and Oct. 14, 2005.

Unfortunately, the profits turn out to be as illusory as the tooth fairy—the run-up in price completely reverses within 12 trading days. The original gains turn into nothing more than market impact costs. In other words, after costs, Cramer’s picks have negative value to investors who act on his buy recommendations. However, the market is so efficient that there may be people who benefit from Cramer’s picks.

While demand for Cramer’s stock picks increases, short-selling volume (bets that the stock will fall) also increases. In the opening minutes of the day following a recommendation, short sales increase to almost seven times their normal levels, and they remain elevated for three days.

Who are these short-sellers? Likely candidates are hedge funds, who are exploiting the naivete of individual investors. Through their actions, short-sellers are helping keep the market efficient.

Consider A Shorting Strategy

There is one other study, “Shorting Cramer,” to review. In a 2007 column, Barron’s reported that it studied 1,300 recommendations and found, over the prior two years, viewers holding Cramer’s stock picks would have been up 12%, while the Dow Jones industrial average rose 22% and the S&P 500 Index rose 16%.

The research makes it clear that being highly intelligent (and entertaining, in Cramer’s case) is not a sufficient condition to outperform the market. The reason is simple: There are many other highly intelligent money managers whose price discovery actions work to keep the market highly efficient (meaning market prices are the best estimate we have of the right price). That makes it unlikely any active money manager will outperform on a risk-adjusted basis.

Cramer himself provides a fitting conclusion: In a 2007 issue of New York magazine, he stated: “God knows why, but there seems to be a market for this kind of idiocy.”

Post Script

While the focus in this article is on Cramer’s stock-picking skills, thanks to research by CXO Advisory Group, we also have insight into his market-timing skills. CXO set out to determine if stock market experts, whether self-proclaimed or endorsed by others (such as in the financial media), reliably provide stock market timing guidance.

To find the answer, from 2005 through 2012, they collected and investigated 6,584 forecasts for the U.S. stock market offered publicly by 68 experts (including Cramer), employing technical, fundamental and sentiment indicators. Their collection included forecasts, all of which were publicly available on the internet, that went back as far as the end of 1998. They selected experts, both bulls and bears, based on web searches for public archives with enough forecasts spanning enough market conditions to gauge accuracy.

The researchers’ methodology was to compare forecasts for the U.S. stock market to the return of the S&P 500 Index over the future interval(s) most relevant to the forecast horizon. They excluded forecasts that were too vague as well as forecasts that included conditions requiring consideration of data other than stock market returns.

They matched the frequency of a guru’s commentaries (such as weekly or monthly) to the forecast horizon, unless the forecast specifies some other timing. Importantly, they took into account the long-run empirical behavior of the S&P 500 Index. For example, if a guru said investors should be bullish on U.S. stocks over the current year, and the S&P 500 Index went up by just a few percentage points, they judged the call incorrect, because the long-term average annual return has been much higher. Finally, they graded complex forecasts with elements proving both correct and incorrect as both right and wrong (not half right and half wrong).

Following is a summary of CXO’s findings:

• Across all forecasts, accuracy was worse than the proverbial flip of a coin—just under 47%.
• The average guru, and Cramer specifically, also had a forecasting accuracy rate of about 47%. Cramer finished 39th out of the 68 experts.
• The distribution of forecasting accuracy by the gurus looks very much like a bell curve—what you would expect from random outcomes. That makes it difficult to tell if there is any skill present.

Of course, there were a few with fairly good records, which is what you would randomly expect. But only five of the 68 gurus posted scores above 60% (the highest score was 68%), while 12 had scores below 40% (the lowest score was 22%). Remember as well that strategies based on forecasts have no costs, but implementing them does.

The research shows that whether it involves predicting economic growth, interest rates, currencies or the stock market, or even picking individual stocks, gurus’ only value is to make weather forecasters look good. Keep this in mind the next time you find yourself paying attention to some guru’s latest forecast. You’re best served by ignoring it.

Larry Swedroe is the director of research for The BAM Alliance, a community of more than 140 independent registered investment advisors throughout the country.