Weekly Wrap-up: Valuation Models
The following article was originally published at The Agile Trader on November 20, 2005.
Our Dynamic Trading System remains long SPY. The System has no position in QQQQ. While it is likely that the System will remain long the SPYs through December, trading signals, both long and short, may very well begin arriving on the Qs by the middle of next week and throughout most of December.
In one sense this is a great time of year for the System as strong seasonal tendencies make the System's trades extremely profitable on a historical basis. On the other hand, it can be boring at times as the System has fewer signals than normal. Over the long-term the System averages about 2 positions per month on the SPYs and about 4 positions per month on the Qs. In aggregate the average is about 70 trades (round trips) per year. But in the November time frame the System may sit on its long positions for quite a while longer than it does during the rest of the year.
That said, ultimately our goal with the System is not ENTERTAINMENT but MAKING MONEY. And the System has been doing a good job of that lately with 10 of its past 11 trades in the "green" for better than 19% in profits in the past 3 months (including our open SPY position, which is more than 4%). Moreover, our auto-execute traders in our Futures Service have 267% in trading gains in the same time frame, exhibiting the upside of the System's efficient application in the e-mini SPX and e-mini NDX contracts.
Of course past performance is not a guarantee if future returns. But, if you would like more information about our Futures Service, its risk, rewards, subscription fees, and how to open an auto-trade account (which automatically executes our signals in the futures markets while you go about having a life, rather than staring at the computer monitor all day) you can email us at email@example.com.
**** **** ****
I'd like to revisit a market valuation model that we last looked at in this space on January 23, 2005. Why? Because over the weekend I read a surprising number of articles written by apparently disgruntled bears who were at their most vitriolic on the subject of the economy, the stock market, and the bubblicious credit markets. And ringing cries for a return to the Secular Bear rang out across the land. (Not surprising given the stock market's recent breakouts to new cycle highs on a number of its benchmark indices - discussed more below).
The following 3 paragraphs reiterate what I wrote 10 months ago, but I've updated the numbers to those that are presently accurate.
If you watch CNBC or read any number articles in the financial press you'll hear pundits articulating their personal views. "The market is wildly over-valued!" "Bear markets don't end until market has a PE of 5!" "The market's fairly valued!" "The market's cheap!" And while everyone's got an opinion (dogma?) very few among those who express a view will tell you just how it is derived.
In that context, I'd like to take a look at a (crude) valuation model at which I've been gazing intermittently over the past 8 years; one that makes a series of very conservative assumptions about what the SPX is worth, and one which yields a result very close to the actual SPX price surprisingly often. This is an opportune moment to do just this because the SPX's trailing PE is within 0.6 of its 46-yr median (now 16.5, with a median of 17.1), and just how people reckon that the index is historically expensive remains a mystery to me--one which I wish media pundits would solve for me by revealing their methodologies.
So, here are our (conservative) assumptions: The SPX is worth the current value of future earnings over the next 30 years and will have no "enterprise value" beyond those earnings (i.e., that the value of the SPX 30 years hence will be exactly the amount of its undistributed earnings over those 30 years). Earnings per share will grow at 6.9% over the next 5 years (as they have done over the past 46 years) and then will CEASE GROWING AT ALL. We will use the current yield on BAA corporate bonds (6.31%) to discount the current value of those future earnings. (This is much higher--more conservative--than is used by the Fed's Fair Value model, which discounts the SPX's earnings yield against the 10-yr Treasury Note, currently at 4.5%.)
Here's one graphic representation of what we're talking about.
This model projects earnings of $80.77 for the coming 52 weeks, which is 4.3% below the current consensus. The model further projects that 5 years from now annual EPS will top out at $105.48 and remain flat for the following 25 years, with those earnings discounted at 6.31% per year so that EPS 30 years from now are currently worth just $16.83. And earnings beyond that are worth ZERO.
The sum of the next 30 years' discounted earnings (purple columns) is now $1,349, which is what this model says is Fair Value for the SPX, about +8% from the SPX's current value.
This result is not terribly out of line with what we see when we examine the relative yields on the SPX, BAA Corporate Bonds, and Treasuries.
The SPX has an earnings yield that is higher than both BAA and Treasury Bonds. That is, investors are paying more for $1 in earnings on the SPX than they are paying for $1 in yield on the SPX or on BAA Bonds. That makes the stock market cheap on a relative basis.
On average over the past 46 years investors have been willing to pay about 15 times the yield on the 10-Yr Note (average 6.7% yield, now 4.5% yield) and about 16 times the yield on the SPX (average 6.3% yield, now 6.8%).
Looking at the chart of the relative PEs we see that the SPX PE on F52W EPS is now 14.8 while the Price/Dividend Ratio of the 10-Yr Treasury is at 22.1.
So, despite the stock-market rally to new SPX 4-year highs, the PE on the index remains quite low on both an absolute basis (average 16, now 14.8) and on a relative basis (about 33% below the P/D on the 10-Yr Note).
The Fed's Fair Value calculation (Forward 52-Wk EPS divided by the yield on the 10-Yr Treasury) gives us a target of 1876 ($84.44/.045)=1876.
That result is probably a bit insane at the moment as it does not take into account the riskier post-9/11 world in which we live (huge debt overhangs, high energy prices, geopolitical uncertainty, etc.). We try to quantify the riskier world with our Risk Premium calculation. Risk Premium is defined in this space as the difference between the SPX F52W EPS yield and the 10-Yr Treasury Yield. At this point that calculation is: 6.8% - 4.5% = 2.3%. And that's quite high.
Our Risk Adjusted Fair Value(RAFV) target is reckoned by taking the F52W EPS consensus ($84.44) and dividing that by the sum of the 10-Yr Treasury Yield and the Median Post-9/11 Risk Premium (now 1.92%). So, our RAFV target is $84.44/(.045+.0192)= 1315.
The blue line on this chart is the SPX price. The red line is RAFV calculation. We continue to expect that the year-end rally will take the SPX up very close to its RAFV target, much as did the rally in late '03.
The impediment to further upside remains the flattening yield curve.
This chart plots the SPX Trailing PE (pink) and the Yield Curve (blue). We define the Yield Curve in this space as the spread between the 10-Yr Treasury Yield and the Effective Fed Funds Rate. With the SPX now yielding about 4.5% and the Fed Funds Rate at 4% the Yield Curve extends over just +0.5%. And as you can see on the chart, when the Yield Curve is narrowing (blue line falling) the PE is generally falling (the bubble of the latter '90s excepted). So, the market's PE will have trouble expanding unless/until the Fed takes its foot off the brake (stops raising rates and flattening the curve).
If the Yield Curve inverts, then the economic growth is likely to slow to a crawl or even go negative. But if/when the Fed takes its foot off the brake, we could see PE multiples expand again.
**** **** ****
The SPX is currently testing the 1253 level (61.8% retracement of the '00-'02 bear market). A break of 1253 has us looking for 1291 (66.6% retracement). And if 1291 breaks to the upside, then we're looking for 1368 (76.4% retracement).
If we head for those higher targets this year, then we would expect that the trip down into the 4-year Cycle Low in the fall of '06 will be of the relatively more mild variety, as we have studied over the past couple of weeks. If the market fails its test of 1253 then the risks of a nastier bear market in '06 increase.
Best regards and good trading!