EPS Revisions for S&P 500 Companies

Here is a brief summary of data I collected from Yahoo Finance for companies in the S&P 500 Index.

EPS Estimates for the current fiscal year compared to estimates 90 days ago:
UP: 166 (33%)
DN: 306 (62%)
UNCH: 24 (5%)

In aggregate, current year estimates were revised down 4.7%.
The average upward revision was 4.2%
The average downward revision was 9.9%

Not much of a surprise as to which type of companies received the largest revisions: Energy & commodities-Up and Financials & consumer goods-Down.

Out of the 496 companies I could find estimates for (current & 90 days ago), 479 have positive EPS estimates and 17 negative, up from 11 negative estimates 90 days ago.

I eliminated the negative EPS companies when calculating the revisions to consensus estimates since percentage changes for negative values do not make sense. Therefore, the 9.9% figure for average down revision is slightly understated.

Back in February, when the market was trading at 13.7x estimates I theorized that investors were thinking estimates were too high and would be revised down, opposed to the market being relatively cheap. (see low multiples mean market cheap?) Currently, the S&P 500 is trading slightly higher than it was in February when I made those comments, yet according the data from the WSJ, the estimated P/E has risen to about 14.5x. It appears logical, the multiple has risen close to 6% and estimates have fallen close to 5%, and the market price level has increased a touch as well.

The question remains: “Are earnings estimates still too high?”

The S&P is still trading at low multiple considering that the 10-year currently yields 3.48%, and that multiples have generally been in the high teens for the past couple decades. So, does suggest more downward revisions? Well, if the investors were to think the economic slowdown will be brief and shallow, it would stand to reason that the S&P would be trading at a much higher multiple- reflecting the expectations of a strong rebound and higher future earnings.

Many do believe that the economy will strengthen considerably in the second half of this year and that earnings will return to double-digit growth rates. However, the S&P’s P/E multiple doesn’t convincingly confirm that sentiment. Perhaps another explanation exists.

A point I touched on in my February commentary was the increase to the equity risk premium, or ERP. Investors require a premium to hold risky equities over holding risk-free assets, such as Treasury Bills. The ERP over history has been 5.5-6.5%, but in recent times it has fallen to 2-4%, depending on which method used and which expert one asks.

In the last several months to almost a year, that risk premium has definitely increased. Using the price level of the S&P 500 index and expected earnings growth to calculate the ERP, the current implied ERP is somewhere around 5.5%. A higher ERP translates into lower price multiples. On the margin, when investors perceive increased risk to holding equities, they will pay less for $1 of EPS, hence a lower P/E multiple.

Investors have been less willing to pay up for equities. There has been an explosion of market volatility. Volatility is another way of saying uncertainty, which could be described as risk, actually it's the definition of risk.

Current conditions make the future impossible to predict. Times such as - when today was like yesterday, which was like the day before that, and so on, with static market and economic conditions, then it’s easier to assume that the future will be similar to the present. Volatility in the market is absent; investors perceive there to be less risk, hence P/E multiples are higher.

Currently, today is rarely anything like yesterday, and tomorrow is anyone’s guess. Will there be another big bank write-down? Which market will seize up next? We’ve had mortgages, auction-rate, muni’s, Libor, etc. Bear Sterns saw something like more than 10 billion in liquidity evaporate in the span of a day. It’s been an extremely uncertain and volatile period.

Getting back to the earnings estimates question- Does the market think estimates are not too high? That there will not be a rash of revisions? It’s hard to say. What isn’t hard to say is that even if investors think earnings forecasts are reasonable, they have a low degree of confidence (or certainty) that those estimates will prove accurate. Whether estimates are too high or not, or if a recession will be shallow or deep- actually may not be the question. Either way, investors are not willing to take that bet, instead they are paying low multiples for equities and buying low risk assets yielding negative returns (after factoring in inflation).

I have included tables for the 25 largest (pct) revisions- Up and Down

TOP 25 UPWARD REVISIONS (90 Days)


TOP 25 DOWNWARD REVISIONS (90 Days)

Do Low Multiples Mean the Market is Undervalued?

S&P 500 is trading at low price multiple to expected earnings, 13.7 according to WSJ. The historical forward P/E has been in the range of 14-16, depending on how far you look back. With interest rates incredibly low, 3.88% on the 10-year, should make the fair-value multiple even higher.

According to my calculations, the S&P 500’s mean P/E = 14.2 and median = 13.2. Thirteen companies were excluded due to negative earnings. The highest P/E was 90, and only ten firms had multiples greater than 30. The chart shows the frequency distribution of the individual firm’s P/Es constituting the index.

So is the market cheap? The low price multiples suggest that it is.




The Consensus estimates point to a strong recovery. According to Thompson, Analysts predict earnings to jump 15.3 percent this year.

In my opinion, that magnitude of growth is wildly optimistic. The Market isn’t buying it either. It’s not that the market is cheap, it’s that investors believe consensus estimates are too high. Why do I think that? Because if the market had full confidence in the forecasted numbers, I doubt the market would trade at these multiples. Hence, investors are pricing in lower earnings than the consensus forecasts thus making multiples higher.

The most probable outcome will be downward revisions to the earnings estimates. It’s possible that some of the consensus numbers are stale, meaning analysts have been slow to update them. This seems plausible given the current environment of uncertainty, and the inherent lack of visibility, may delay updates to estimates. In some cases, analysts may be waiting for a clearer picture going forward, or updated guidance from firms before making revisions to this and next years’ full year estimates.

Another possibly is that the required rate of return investors demand from equities rose, the “Equity Risk Premium.” This implies that the Market perceives increased risk inherent in the equity markets. This seems plausible since volatility has increased relative to years past. Higher perceived risk leads to higher demanded returns which compresses price-earnings multiples.

In my opinion, stocks are not as cheap as forward multiples suggest. Earnings estimates are too high and are likely to come down. In addition, the ERP has increased pinching multiples. However, if the economic slowdown begins to appear less severe as the Market is expecting, then stocks would be rather cheap. That would mean that analysts are not overestimating future earnings. However, given the turmoil in the housing market and the relationship to consumer spending, it’s likely the coming quarters will be weak.