Columbia Business school publishes a biannual investing newsletter, in which MBA students compete for the best stock analysis and pitch.

The newsletter is called “Graham & Doddsville” — a reference to Warren Buffett’s repudiation of the efficient market theory on the basis that mere randomness couldn’t account for the tendency of Graham devotees (the inhabitants of the hypothetical town of Graham & Doddsville) to outperform the market.

It comes as a surprise then, that the winners of the newsletter’s security analysis competition hardly ever produce work that follows the central tenets of Graham’s investment theory.

Instead, the students relied on the very kind of forward projection of earnings and/or cash flows that Graham often cautioned against. Below I will briefly review some of the winning recommendations, and explain how even a cursory EPS trend analysis could have improved the results in most cases.

### Avoid Stocks That Are Cheap Because of a Weak Earnings History

In the Winter 2019 G&D Newsletter, the winning analysis recommended DXC Technology Company (DXC) at a price of $25 per share, and an attractive price to earnings (P/E) ratio of 5.6.

On the basis of the attractive P/E multiple and the expectation that the newly appointed CEO could further the “ongoing earnings stabilization,” the analyst set a 3-year target price of $70 for the company.

Looking at DXC’s average earnings per share (EPS) of $3.23 from 2013 to 2019, we get a P/E ratio of 7.73 at the time of the recommendation. We see zero earnings in 2015, and negative earnings in 2017. In essence, the low P/E multiple accurately reflects the company’s lack of earnings growth.

Today, it trades at $28.90 — a 16% increase — but its revenues of the last three years exhibit a downwards trend.

Therefore, the target price of $70 is almost certainly too optimistic. But at least the projection tended in the right direction.

### Use a Multi-Year Average EPS Figure to Calculate the P/E Ratio

Next up is Hanesbrands (HBI), which the winning analyst recommended as a buy at the price of $17, with earnings of $1.64 implying a decent P/E ratio of 10.

When we base the P/E ratio on the company’s average EPS of 1.07 from 2013 to 2019, we obtain a P/E ratio of 17. Such a high multiple is appropriate only where there is evidence of strong earnings growth, but in HBI’s case, the trend displayed by the data is rather irregular.

In 2020, the stock’s earnings collapsed and its leverage ballooned to unsustainable levels. Today, the stock is trading at only $5.90.

### Elevated P/E Ratios Must Correlate With a Strong Earnings Growth Trend

In Spring 2019, the winning stock was Aramark (ARMK) at a price of $31, and a P/E ratio of 17 based on its earnings per share of $1.81.

When we calculate ARMK’s seven-year average, we get a value of 1.2, implying that its actual P/E ratio is a lofty 25.8. But this multiple is somewhat justified by the strong growth trend exhibited by its EPS history.

Today, the company trades at $39 — a decent 26% return.

### Past Trends Are Stronger Predictors Than Future Expectations

The winter 2018 edition of the newsletter included a buy recommendation for Lion’s Gate Entertainment (LGF.A ) at $18 per share and an EPS of $2.15, for a P/E ratio of 7.

Averaging out the prior six years, we obtain a much lower EPS of $1.07, which implies an inflated P/E ratio of 17. As it turns out, the year in which the stock was recommended the stock reported abnormally high earnings, which of course resulted in an attractive P/E ratio.

The analyst set a price target at $26, implying a 44% upside. Today, the stock has lost more than half its market value, trading at $10.49.

Our last “winner” is Nordstrom (JWN), whose analyst predicted a target price of $94. At that time the stock was trading at $64, with EPS of $2.59 and a P/E ratio of 24. Even before COVID, the future prospects of large brick and mortar retailers was in question, yet the company’s earnings multiple implied strong future growth.

Since the company’s EPS trend was negative, taking the average of the prior six years actually improves the EPS ($3.13) and the PE ratio (20).

As we now know, the expected turnaround in growth never materialized for Nordstrom and the company now trades at a third of its value in 2018, at $19.80.

### Why the P/E Ratio Is Still Relevant

Many finance experts reject the use (or diminish the importance) of P/E ratios in serious equity analysis. But at Market Inference, we believe it is still an essential analytical tool, especially when based on a multi-year earnings per share average.

Out of the five recommended stocks that we reviewed, three performed dismally, while two others logged moderate positive performances.

The two stocks that performed well, DXC and ARMK, were the only ones whose P/E ratios accurately reflected their underlying historical earnings growth trend.

We believe that even a cursory price to earnings trend analysis proved to be a stronger indicator of future performance than the students’ qualitative and projection based analysis.

In their defense, the unexpected COVID-19 pandemic and subsequent market volatility and macro uncertainty certainly threw a wrench into their projections.

But this is exactly why Graham advocated for establishing a Margin of Safety based on past data.

The future is unpredictable, so the best we can do is buy shares in companies that are rationally priced to their historical valuations, so as to shield us from inevitable (negative) surprises in the future.