Fundamental S&P 500 Forward Price Projections

Here are fundamental valuation arguments for the S&P 500 reaching levels ranging from about 500 to about 1,500 by year-end 2010.

2009 Year-End Valuation on Trailing Operating Earnings:

Street Opinion: Barron’s Online April 2nd reported the Reuter’s “street” estimate of 2009 S&P 500 “operating earnings” (not including non-operating expenses, such as asset write-downs, plant closures, layoff costs and other “non-recurring” items) at $59.18. Those analysts who would put a 10 to 20 multiple on trailing operating earnings as a “normal” range would see a 2009 ending index level of 592 to 1184 (mid-point 796).

Fed Model: The formerly popular “Fed Model” that justified a P/E equal to the reciprocal of the 10-year Treasury rate, doesn’t make a lot of sense in this period of risk aversion-driven low interest rates. If someone were nutty enough to use that model and the street operating earnings estimates for 2009, they would see a trailing P/E of 34 as reasonable, which would predict an index price level of about 2012. That’s more than 25% higher than the highest value achieved in 2007 before the current troubles began, so we’ll discount that as plumb crazy. Take those guys away and put them in padded cells so they don’t hurt themselves.

Standard & Poor’s Projections: S&P projects 2009 index operating earnings at $49.02 on a top-down basis, and $62.36 on a bottom-up basis. A trailing P/E of 10 to 20 on those numbers would put the index at 490 to 1247 (mid-point 868) by year-end.

2009 Year-End Valuation on 2010 Forward Operating Earnings:

Street 2010 Estimates: Barron’s reports the Reuter’s 2010 street estimate of operating earnings at $75.32. A 10 to 20 forward multiple on that would put the S&P 500 index at 753 to 1506 (mid-point 1129) by year-end 2009.

Individual 2010 Estimates: Deutsche Bank projects 2010 operating earnings at $61.70. JP Morgan sees $65, and Standard and Poor’s sees $48.44. Using the 10 to 20 forward multiple approach, their numbers would produce:

• DB: 617 to 1234 (mid-point 935)
• JPM: 650 to 1300 (mid-point 975)
• S&P: 484 to 968 (mid-point 726).

2009 Year-End Valuation on Trailing and Forward “As Reported” Earnings:

“As Reported” earnings include everything — nothing left out — all the bad, as well as all the good — but not what the rules may allow to pass through the balance sheet without going through the income statement.

Standard & Poor’s top-down projections of “as reported” earnings as of 03/25/09 are:

• 2009 – $34.74
• 2010 – $41.49.

While we don’t know what the historical norms are for operating earnings over long periods (because the term is not well defined — sort of whatever you want to report and whatever you want to exclude); “as reported” earnings is well defined and there are long-term historical records of the trailing P/E on those earnings.

Over an 81 year period through 2007, and eliminating the 10% highest and 10% lowest multiples; the range of P/E’s is 10 to 20 and the median is 15 (see our study).

Using 10 to 20 times trailing multiples on S&P’s projected 2009 “as reported” earnings, we get a year-end 2009 index price range of 347 to 694 (mid-point 520).

Using 10 to 20 times forward multiples on S&P’s projected 2010 “as reported” earnings, we get a year-end 2009 index price range of 415 to 830 (mid-point 622).

If we bring the crazy Fed Model guy back out of his padded cell for a moment, and allow him to assume that interest rates will not rise from now through 2010, and let him work with forward “as reported” earnings at his lofty 34 times multiple, we see 2009 ending the year at 1411 (about 10% below the 2007 pre-crash high).

Historical Trailing “As Reported” Earnings and Index Price Levels:

You might want to make your own guesses through analogy to prior years. Here are the “as reported” earnings, ending index price levels, trailing P/E’s, and year-end 10-year Treasury interest rates for the past 10 years. No conclusion, just raw data for you to chew on.

• 2008: $14.97 / 903 / 60x / 2.25%
• 2007: $66.18 / 1468 / 22x / 4.04%
• 2006: $81.51 / 1418 / 17x / 4.71%
• 2005: $69.93 / 1248 / 18x / 4.36%
• 2004: $58.55 / 1212 / 21x / 3.94%
• 2003: $48.74 / 1112 / 23x / 3.77%
• 2002: $27.59 / 880 / 32x / 3.36%
• 2001: $24.69 / 1148 / 46x / 5.07%
• 2000: $50.00 / 1320 / 26x / 5.12%
• 1999: $48.17 / 1469 / 31x / 6.45%

Chart Analogy:

Our closest recent chart analogy to today might be 2002. Looking at the historical table above, they had a 32 trailing P/E multiple then. The banking and general economic context is worse now, however.

If we use the 32 multiple (oops! that sounds like the crazy Fed Model guy) on the $34.74 projected 2009 trailing “as reported” earnings, it might be possible to expect 1097 for the index by 12/31/09. Somehow that really doesn’t seem crazy at all, if all the government programs kick in and begin to work — just look out for the inflation and higher taxes that will be on the other side of the recovery.

Conclusion

Quantitative analysis can be comforting on the surface, but underneath earnings estimates and earnings multiples are highly subjective, widely dispersed, and not subject to strict definitions and methods.

In the end, quantitative and fundamental analysis is based in logic, but applied subjectively — and totally unable to predict shocks and/or arbitrary rule changes by government intervention. Remember that last week banks were failing due to mark-to-market losses, and this week they are doing much better, because new rules relax the mark-to-market requirement. No form of quantitative review could predict something like that.

Earnings estimates have been coming down, but they will probably overshoot, as they tend to do both up and down.

So in the end predicting price levels over short periods is just a bet — a better bet than at the casino, but still a bet.

The most important thing to do is to know what you want to own, why you want to own it, and then buy it when both its fundamentals and its price behavior are favorable to you.

After buying it, hold it regardless of its fundamentals, so long as its price behavior is favorable to you, using protective stops along the way.

When the price behavior becomes unfavorable (most likely when you are stopped out), put the security back on your shopping list and re-enter only when both the fundamentals and the price behavior become favorable once again.

If what you want to own is the S&P 500 index, the price behavior is still not fully favorable. The fundamentals surely don’t look good yet, and can only really look good when the banks, auto companies, and home values get straightened out.

The March index rise is the strongest since the 1930s, giving strong hopes to many observers and investors. However, the 1930s were also the beginning of several periods of multi-year index losses (see our study), so this rally does not necessarily portend a new upward world. We hope so, but we need more proof.

Clearly the current rally is powerful and appealing. You could say the price behavior is favorable, but the moving averages have not yet confirmed an upward direction. It could be the beginning of a bull, but we don’t know yet. Certainly, there are companies within the index that are surging, but the 500 in combination are still in an unclear price pattern.

Comments

[Dave Wrixon:]

Moving averages don't confirm anything. They are just a mathematical tool. Markets are driven by sentiment. Sentiments are ultimately molded by Earnings in the Medium term. Short-term sentiment is hostage to the mindless outpourings of the Media.

That is the big difference between now and the 1930s. The involvement of the masses in the financial markets was smaller, but the level of understanding amongst the masses was nearly non-existent. The one thing whose influence has changed but not really been evaluated is the Information Multiplier.

Apr 05 07:21 AM

[investor88:]

Fair enough trade with stop loss and some scepticism, control risk. Agree early days to call a new bull. That does not mean no trading because in the last month or two many stocks have doubled.

Apr 05 08:19 AM

[wise:]

The best managers use normalized long-term P/E averages, which would be the 15-16X, and that's effective to have been used in the past at market bottoms.

Apr 05 09:47 AM

[Lilguy:]

In this dour economic climate, any valuation method that uses "operating earnings" only, that is, "not including non-operating expenses, such as asset write-downs, plant closures, layoff costs and other “non-recurring” items," way overstates the value and future earnings prospects of companies.

These many "non-recurring" costs are now what's driving businesses ever closer to insolvency, starting with our major banks and auto industry. These deeply impaired assets are no longer able to produce the products and services that generate future corporate earnings. Moreover, they raise (and rightly so) the prospect of future writedowns of these and other assets, eroding confidence in earnings forecasts.

In short, valuing a corporation without considering all the factors that determine its current and future value is grossly misleading and can lead to massively bad investment decisions. It is rather like accepting the business headlines that essentially say the latest news on the economy (or company, or whatever) is looking good if you do not consider all the bad news. True, but so what!

Apr 05 10:39 AM

[pslater:]

All this talk of P/E ratios ignores the fact that P/E ratios ARE a function of interest rates. I agree that the Fed model is useless right now if you use the 10 year Treasury as the benchmark but the concept is sound. If you substitute the BBB interest rate (the average credit rating of the non financial S&P 500 companies is around BBB. For reference, see the iShares investment grade corporate bond fund - ticker LQD. Average credit rating is BBB+) for the 10 year Treasury you get a much more accurate picture of where P/E ratios should be. With the current BBB rate of around 8.5%, a P/E of about 11.8 is implied. (100 / 8.5 = 11.8)

An analysis of the last 21 years of actual year end index numbers compared to 'fair value' estimates using the BAA interest rate to imply P/E ratios shows that in recessions (1991, 1992, 2001, 2002, and 2008), the S&P tends to have about a 3.0 multiple to the implied P/E ratio (this is skewed by 2008 where the multiple is 4.8!)whereas in economic expansions (the rest of the years) the average is 1.6 (which is skewed upward by the 1998, 1999, and 2000 P/E spike).

If we use the 'as reported' trailing four quarter number of $14.97 we get a fair value for the S&P 500 of 529 ($14.97 X 11.8 P/E * 3.0 recession adjustment).

If we use S&P's 'as reported' EPS estimate for the next four quarters of $34.74 (a very bad idea!) and the economic expansion adjustment of 1.6, we get an index fair value of 655 (11.8 X $34.74 X 1.6 expansion adjustment).

Finally, if we use the 2008 recession multiple of 4.8 rather than the historical multiple of 3.0 and the 'as reported' EPS of $14.97 we get a 'fair value' of 847 ($14.97 X 11.8 X 4.8) which is less than 1% from where the index closed Friday!

In summary, this market is no bargain and the Fed model is useful if you apply the appropriate interest rate.

Apr 05 11:38 AM

[Scott's Inv...:]

I think one factor not given enough attention in the media and in earnings projections is the relationship between the recent surge in unemployment and scheduled resets on Alt-A and Option ARM mortages that are set to begin on a larger scale in the middle of this year. I think we will see a second wave of foreclosures starting later this year and going well into 2010, which will certainly impact the earnings of financials and homebuilders and can make relying on any projections dubious when attempting to come to fair valuations (then again, by modifying mark to market perhaps nobody will even notice!).

I think the safest way to play the market at this point is to watch price behavior and use stop losses and be willing to work with shorter time horizons then long term investors may have been previously accustomed to.

Apr 05 11:53 AM

[prudentinve...:]

Despite the uplifting talk in the media, I cannot yet see any fundamental basis for a sustainable long-term recovery. The government has, temporarily I hope, seized control of a large portion of the economy, and is picking its winners and losers and its allocation of labor, resources and rewards. Given the historic failure of centrally planned economies, I doubt that these actions are setting the stage for anything beyond a short-term consumption boost engendered by massive, inefficient, and inflationary spending.

What this means for stocks is beyond my ability to predict, but I doubt that historic analogies or models that have worked in the past are relevant in the present environment. Taking some profits in rallies and buying the dips, to continue holding some equities, seems prudent at this point in time.

Apr 05 12:10 PM

[benevolent ...:]

In 2002 the basis of earnings reports was changed to "operating earnings" rather than "total company earnings" so that the P/E ratio will be about twice of what it really is-thats unhealthy. Operating earnings omit several items, the main one being interest payments for debt service. From 2002 to 2008 debt of all publicly traded companies almost trippled. This according to the SEC.

Now there are two entities making claim to the same assets; share holders and bond holders and/or lenders. The latter has first claim and that makes share holders owning nothing as an aggregate. How can we trust these accounting methods when the prime directive of most corps is for the CEO's, who appear to be in collusion with BoD members (other corp CEO's), is to max their annual bonuses? Enron on a massive scale. CEO's collateralizing the corps assets (now liabilities) and showing this as profit. But not before the liabilities are set into a shell corp and stuctured as a lease. May be I am wrong-lets make sure.

Apr 05 03:53 PM

[jstratt:]

Thanks for your analysis of earnings!

On one hand using earnings from a period of frozen credit markets may not be a fair valuation of future earnings expectations.

On the other hand who is to say that we will not refreeze and continue with a destruction in commerce.

In the meantime I will feel better about the direction when we get though this earnings season. For the record I am estimating 1020 on the SP500 by year end with current data. Roughly I think growth rates cause 20% of the loss from highs and 20% is from longer term damage to business activity.

Over the longer term stocks are worth more than 53% of their peak value.

Apr 05 07:09 PM

[mac.barron:]

A few important points:

Mr. Shaw states: "The [2009] March index rise is the strongest since the 1930s, giving strong hopes to many observers and investors. However, the 1930s were also the beginning of several periods of multi-year index losses . . ."

The Federal Government has been pumping literally TRILLIONS of dollars into distressed financial institutions. Part of the effect of this has been to create a market rally. Although the Government can keep the printing presses (both the real ones and the digital ones) humming for a while they can have a significant effect raising the markets.

However, at some point the addition of capital will have to be curbed. As yet, no significant observations have been made of dramatic increases in the domestic manufacturing economy, nor do any seem to be in the offing. So we have these facts to work with:

The record of history shows that the potential for a serious bear trap.

When we ask: "Is this time different?" We can not point to any significant motive force that might spur the U.S. economy ahead after the spending party is drawn to a close.

Therefore, it seems prudent to take necessary hedging steps to provide for the potential of both sudden and sustained drop in the domestic equities markets and a loss of value of the U.S. Dollar.

Apr 05 09:48 PM

[Fighting Yoda:]

Most recent big bear market was – 1974 – 84. During this period the Average S&P PE was 9.5, low of 7.4 in ’74 & ’79, and a high of 12.4 in ’83. I think we are going to have a ’74-’84 kind of bear market – there are too many much bigger problems – printing money simply will not solve these problems, likely would create some. Spending binge is over, the economy has to delever and start form a new base.

The earnings forecasts for ’09 are almost consensus $50 and below. A 10 multiple would get you to 500. A highly optimistic/bullish S&P would be 600. The current rally is simply a sucker rally, with every piece of worst than expected bad news the market goes up. Friday’s job report was a major horror story – it emphatically stated – no end in sight. Wall Street pros are simply drawing lambs to the slaughter.

Apr 06 12:32 AM

[Fighting Yoda:]

There is some mention about low interest rates mean higher PEs, conceptually makes sense and some recent data would support it. However during the ‘30s interest rates were equally low, PEs stayed low, market kept losing. So instead of interest rates I would focus on inflation, during Depression we had Deflation, and we are headed in the same direction now, Fed’s attempts to create inflation would fail. We already have significant drops in commodity, home, auto prices – all pointing to deflation. I don’t think anybody predicts a rebound in prices – at best stabilization at some point.

If we have deflation, even disinflation – markets would continue to fall.

Apr 06 12:46 AM

[Stone Fox C...:]

This talk of using 08 or 09 reported earnings or operating earnings is absurd. To value a company, you'd always use a discounted cash flow of future earnings. In that case, an analyst would attempt to factor 10, 11, 12 and so on earnings. They'd never focus on trough earnings to derive a value. Using a normalized earnings approach may be a accurate method. If earnings peaked around $100, I'd have a difficult time thinking normalized would drop below say $80.

If Citigroup is expected to have normalized operating earnings around $30B, I'd like to see the analysis that uses that number and not the reported for 2008. It's absurd to value a company on the worst case scenario. Claiming the market is expensive now is just as bizarre. To get there an analyst has to take the worse case scenario and not factor any of the reflating and low interest rates going on. Crazy is you ask me. As usual when the market goes higher, all of a sudden the estimates for 2010 will rise dramatically and we'll start hearing how at S&P 1,000 the market is cheap.

Apr 06 12:55 AM