I recently read an article by Jeremy Seigel, the Ph.D. economist who currently teaches at UPenn, about how Standard & Poors computes its coveted 500 Index’s Earnings Per Share. Believe it or not, S&P calculates EPS of the index on a simple sum basis. Meaning, if every company earned $1.00/share, the Index will have earned $500/share.
Dr. Seigel proposed that S&P switch how it computes its EPS to a market weighted summed, similar to how it calculates the price of the Index. I agree with this methodology and strongly disagree with S&P’s response to why it currently uses its current method.
To mirror Seigel’s example and to build on it, let’s assume we have an index of two stocks, AIG and ExxonMobil (XOM), and that they each make up 50% of this index. If you were to model this index, regardless of size, for every share of AIG stock you purchase you will also purchase an ExxonMobil share. Suppose AIG incurs heavy losses and posts huge negative EPS numbers, while XOM posts huge profits and large EPS numbers.
Due to the capital structure, equity places no value on negative EPS and large amounts of value on positive EPS. S&P’s methodology assumes that an equity holder will place a negative value on shares that incur an EPS loss. Would you, as a shareholder, be willing to lose more than your initial investment? One cannot lose more than 100% on a common equity position.
Under the way S&P calculates the 500 Index’s EPS, if AIG lost a great deal of money and the market corrected by decreasing the value of the shares, even though its share in the index has decreased, its loss is of equal magnitude to that of XOM’s gain.
Under market-weighted EPS, as AIG’s value in the index fell, so did its contribution to the overall EPS. The lower limit of this function is zero which is consistent with the lower limit of the value of the underlying equity.
In Dr. Seigel’s article, he outlines S&P reasoning for why they used the simple sum approach when calculating the Index’s EPS. S&P touted their amazing Indexing skills claiming that his methodology
failed the simple tests of both logic and index mathematics. A dollar earned or lost is the same, irrespective of whether it is earned or lost by a big index constituent or a smaller one. To use an analogy, we could hypothetically view the S&P 500 as a single company with 500 divisions, with each division having earnings and an implicit market value. The smallest of these divisions could have an outside loss that wipes out the combined earnings of the entire company. Claiming that these losses should be ignored or minimized because they came from a less valuable division is flawed.
What really failed here was S&P not actually understanding WHAT they were indexing. Their above analogy is true for a company, but the S&P 500 Index does not model the capital structure of firms. It models the common equity portion of the firms. Common equity holders are not LIABLE for the losses in the firms and are definitely not liable for the losses at other firms.
S&P also assumes that the equity positions are effected by each other’s results. Imagine an example where a company is 5% index weighted and because of a small float reports -$25.00/share while a 95% index weighted company, because of a much larger float relative to net income reports a huge profit, but amounts to a $2.00 per share profit. Would putting a value on this index using -$23.00/ EPS make sense? Or would a market-weighted EPS of $0.65 seem more reasonable for the index?
Hopefully, S&P can realize this fault and accordingly remedy their methodology in the calculation of their 500 Index’s EPS.
Disclosure: No positions