hier auf Englisch (Investopedia), etwas fundierter als der deutsche Beitrag in P. 35
Quelle: http://www.investopedia.com/articles/03/112703.asp
The Fed Model And Stock Valuation: What It Does And Does Not Tell Us November 27, 2003 By John Summa, CTA, OptionsNerd.com
When the stock market was reaching record new highs in 1999 and 2000, many stock valuation models began sounding the alarm - flagging what in hindsight proved to be an extremely overvalued stock market. However, most investors unfortunately chose to ignore this available information, believing instead that we had entered a "new economy", immune to past problems like long, painful bear markets.
The major market averages remain well below the levels of those days of "irrational exuberance", despite having rallied considerably off the bear market lows of October 2002.
Given the extended rally of 2003, therefore, I thought it would be a good time to look at stock valuation using the so-called Fed Model to see where we stand today. Are we in a mini bubble? Or is this still a buyer's market? As you'll see, the answer is not that simple.
Let's begin with a quick review of the basic valuation model popularized by economist Ed Yardeni who, in his own words, has dubbed it the "Fed's stock valuation model, though no one at the Fed ever officially endorsed it." This model compares two key benchmarks: the interest rate yield of the 10-year U.S. Treasury note and the forward operating earnings per share of the S&P 500 stock index.
As you'll see, these variables can be tinkered with (i.e., substituted with others) to suit one's own subjective preferences regarding what works best for relative comparisons. (For example, some may substitute reported earnings or trailing earnings for the S&P 500 side of the ratio, or use a different interest rate yield in the numerator). That said, let's see what the Fed Model is telling us now and what it has indicated in the past.
The Fed model as of November 14, 2003, despite a major rally off February 2003 lows, had the S&P 500 still well below fair value. You can seen in Figure 1 that the Fed Model shows the S&P fair value at 1,426.30, despite a rally of close to 30% from its 2003 lows to the latest level of 1050.35. In other words, the S&P 500 is still 26% below fair value, or undervalued by 376 points, as indicted in Figure 2.
At the historic market top of March 2000, it is worth noting that the model showed a fair value of 954.91 versus the actual closing high of 1,527.46. It's sort of like the cartoon character Wily Coyote, who remains suspended in midair for a few seconds before realizing he has run off the edge of a cliff. Oops!
After the major bear market plunges of 2000, 2001 and 2002, the model finally began to show the S&P 500 as undervalued. This occurred during the big swoon lower that started the summer of 2002 and continued into the month of October, as can be seen in Figure 2. But the rally of 2003 has made the S&P 500 considerably less undervalued than was the case in March this year (when the Fed Model showed the S&P 500 over 700 points undervalued). The S&P 500 remains, therefore, from a relative value perspective still a buy, according to the Fed Model. However, there are some underlying dynamics at work in the model's variables that lead to some interesting paradoxes.
Figure 1 - S&P 500 Fair Value According to the Fed Model's Valuation. (im Anhang unten, A.L.)
For example, note that the S&P 500 became more than 800 points undervalued in June of this year, despite the major rally off the lows of March. At that point, S&P 500 fair value was over 1800, according to the Fed Model, as can be seen in Figure 2. How could the S&P get so undervalued after it had rallied 25%? The answer can be found in an examination of the wild swings in the Treasury markets in the spring of 2003, with first a lower yield, then a reverse to a higher yield.
When it became clear to bond traders that an economic recovery was likely, long-term rates moved sharply higher, such as the Treasury 10-year note yield, which is one of the two variables in the Fed Model. Recall that the 10-year yield, however, first fell to 3.1% in June of this year, which is what caused the spike higher in S&P 500 fair value as seen in Figure 1, despite the rally in the stock market. The yield promptly reversed course, though, and is just above 4.2% after getting as high as 4.6% in August. While the S&P hovered in a trading range through most of the summer, the reversal lower again in the 10-year yield from the August highs drove fair value higher again to 1,500, with the S&P 500 getting 500 points undervalued despite climbing higher in September. This dynamic may explain why the bulls have had an easier ride, since many asset allocation models are driven by Fed Model valuations or some variation of it, and that would indicate a shifting to stocks from bonds.
But is the model useful for the average investor? Is the market really still a buy? I would argue that taken on its own, the model could be problematic. Remember, all the Fed Model is telling us is that when the forward earnings yield on the S&P 500 (how much you would earn per dollar paid) is less than the 10-year bond yield (how much you make from holding a 10-year note), then stocks have gotten too expensive. In other words, it does not pay to hold stocks because you can earn more in a Treasury bond with less risk. Conversely, if the earnings yield on the S&P 500 is higher relative to the 10-year note yield, stocks are said to be undervalued--that is, investors are at least being compensated for taking on the greater perceived risk of stocks.
So stocks are still "cheap." But when we take a look at absolute value comparisons, such as the earnings yields of the S&P 500 relative to past levels, we get a very different story.
Take a look at Figure 3 below, which contains the historical reported earnings to price ratio of the S&P 500. This series paints a rather gloomy long-term outlook for stocks.
Figure 3 - S&P 500 Earnings-to-Price Ratio with Mean and Standard Deviations Source: Pinnacledata.com. Chart generated using MetaStock Professional. (liefer ich als Anhang im nächsten Posting nach - A.L.)
Big secular bull markets begin when the earnings yield is substantially above its long-term average, preferably 2 standard deviations above. But as you can see from Figure 3, the reported earnings yield is considerably below its average (currently on a reported basis at 3.29), not only a level below its long-term average, but one that preceded the long bear market that began in the late 1930s and continued into the early 1940s.
I am not suggesting that we will have a repeat of this pattern, but this historical valuation measure does suggest that it may be difficult to sustain a new bull market similar to the one we saw during the 1980s and 1990s or the 1950s and 1960s, at least until absolute valuation returns to reasonable levels.
Bullish investors, meanwhile, may be in for another painful lesson about markets.
By John Summa, CTA
Die Grafik zeigt, dass das Fed-Modell zu stark schwankenden Aktien-Bewertungen kommt - die die realen Firmen-Werte, die sich fundamental errechnen lassen, sicherlich nicht in dieser Stärke aufweisen. Allein von 2002 bis 2003 errechnete sich laut Fed-Modell ein "fairer" SP500-Wert zwischen 1000 und 1800! |