Revising the beloved Fed Model -(so soon?)
Valuation is a tricky topic. If you think back to the second half of the 1990s, it was a parameter that was largely ignored altogether.(Timely information) In essence, most top- down investors ditched valuation after Chairman Greenspan uttered the now- famous adage of “irrational exuberance.” If valuation didn’t matter then, why should we focus on it now? Could it be that valuation mattered all along but that the so-called Greenspan Model is the wrong way to go?(Gasp - could it be?!)
Remember that it viewed the equity market as fairly valued as early as 2001 and that it now states that the S& P 500 is approximately 40% undervalued. Does that make sense? Is it still useful?(Still?) Is the stock market really 40% undervalued?
We have examined the theoretical underpinnings of the Greenspan Model and have identified two avenues for improvement. Let’s begin with the denominator, or the discounting factor — long-term interest rates. Many individuals will refer to S& P 500 fair value as being the inverse of the ten- year Treasury note yield. This is essentially a crude version of the Greenspan Model. In today’s world, the inverse of a 4% yield would imply a fairvalue P/ E multiple of about 25x. The problem with this approach is that it uses a risk-free rate (Treasuries) as a denominator when equities are not a risk-free asset class. (that is not what my broker told me in 99)An alternative would be to look at corporate bond yields instead and to find an index that would mimic or proxy the composition of the S&P 500.
The second issue with the Greenspan Model is the use of earnings as an appropriate numerator. What are earnings? Operating, core, or reported? Trailing or 12-month forward? Who knows these days? One alternate method that we feel circumvents the entire debate is to use revenues (or page views - how about click throughs?) instead of earnings. Of course, this substitution also carries limitations, since it is difficult to find forward estimates of revenues — leaving trailing as the only index-encompassing alternative. Nonetheless, it avoids the major issue of balance sheet massaging, since revenues are typically what they are (barring unfortunate exceptions like WorldCom and Enron).
Keeping in mind that valuation models are not ideal for market timing (they are informative, not predictive), we have compiled two alternative methods of valuation for the stock market, both using corporate bond yields instead of Treasury yields, with one coupling them with revenues rather than earnings. The conclusion that can be drawn from these two unorthodox valuation methods is that the stock market is likely in fair-value territory — a far cry from the 40% undervaluation that the Greenspan Model, with its abnormally low Treasury yield input, would imply. (You can't make this stuff up)
Our corporate bond proxy uses a blend of the Moody’s Aaa and Baa corporate bond indices and is designed to mimic the composition of the S& P 500 as closely as possible. Using corporate bond ratings from companies that account for 83% of the total S& P 500 market cap, we created suitable weights to apply to the Aaa and Baa bond yields. A valuation model comparing the forward P/ E of the stock market to the fair value P/ E implied by this blended corporate bond yields suggest that equities are indeed trading at attractive multiples. In fact, on this basis, the stock market has not been this cheap since 1989.
Valuation = Blended Moody's Aaa & Baa Corporate Bond Yield Less 12- Month Forward Estimated S& P 500 Earnings Yield Divided by the 12MF S& P 500 Earnings Yield
A similar approach using price to revenues yields a slightly different valuation conclusion. Indeed, discounting the current price-to-revenues ratio with our corporate S&P 500 bond proxy suggests that the equity market is now in fair-value territory for the first time since 1994. It is not as compelling as the forward earnings method, but at least addresses the accusation that the stock market is cheap solely because earnings are artificially inflated. Revenues are what they are, and the market is now in line with its pre-bubble historical range. (Comforting)
February 12, 2003 Bear, Stearns & Co. Inc |
