2 Thoughts On The Evolution Of S&P500 Earnings

Includes: SPY
by: Braun Research

A rather special long term graph describing the evolution of the S&P500 earnings since 1935 (source) stirred up two ideas that may prove useful: one is coming from my former engineering soul while the other is related to my current incarnation as a practician in the investment world. Should you care to bear with me, I thought that you might be intrigued [click images to enlarge]:

Click to enlarge

1. To an engineer's eye, the chart on the left is clear evidence that the S&P 500 earnings behave lately very similarly to the way any automated feedback system would with a cyclical stimulus being applied to it and with not enough damping (friction), much like the one described in the graph on the right (Source: Forced Vibrations). The engineers call this phenomenon "resonance." In terms of design, if nothing is done, the vibrations will reach a point at which the structure will be torn apart by the massive amplitude. This was the cause of the collapse of the Tacoma Narrows Bridge. (Click here for a video clip of this catastrophe). The frequency of the wind blasts matched the natural swinging frequency of the bridge. Another example was when the space shuttle's turbo pump was accidentally designed with a resonance structure. Fortunately, the engineers noticed the problem before take-off. The error cost NASA millions of dollars. For information on the space shuttle example go here. For those interested, this link is incidentally also a good primer on the practical significance and the possibility of diverse interpretations of probabilities and risks in high value physical systems, with usefully transferable concepts to the investment world, beyond and above the deafening noise being currently heard there on the subject of risk.

The economic system described in this earnings graph is certainly more complex than a bridge or a shuttle pump. But, just as a poorly designed engineering product, it appears ready to oscillate well beyond the acceptable limits set by whatever (or whoever) attempts to regulate it in an endeavor to bring it back to stability. It is not unlike the WWI story of the army units ordered to get out of lockstep when marching over a bridge, lest the rhythm of their march happens to match the natural frequency of the bridge and make it collapse. Similar examples are numerous and questions can obviously be raised by these - possibly fortuitous but still compelling - analogies: What is the economic "stimulus" at work? What "frictions" are there and what other frictions may be needed? Are we anywhere close to the failure point of the system? Are earnings, particularly as measured by the S&P500, a true reflection of the economy? etc. You can dismiss the analogy but would it be wise to at least consider…

2. The practical investment value which may be to point out that, before our economic system demonstrating this potentially self-destructive behavior is either mended or, God forbid, it comes crashing down, some kind of objective analysis could be a very useful tool to help sense the impending tops and bottoms of the oscillations, in time for corrective action to be taken by wise, clear-sighted PMs who, despite being unable to influence the entire system, can and should protect those who entrusted that PM with the management of their hard earned money. When looking for such an analysis one should be able to find a macro-economic approach that has been - provably - able to sense in a timely way, both the Dot-com bubble and the Credit bubble and has been bullish on the current up-leg that started some time around March 2009; albeit a pin-point precision cannot and should not be expected. Such an approach should also have provided, over a reasonably long time (say 8-12 years), demonstrable help in focusing on those companies (stocks) who were not in lockstep with the foolish, uniformly marching crowd (proven good stock picking). There are such research providers and managers but it takes an intense, pro-active effort to find them as they tend to be smaller organizations with resources too limited for repetitive promotion. You should require a proof of outperforming a benchmark over certain longer intervals of say 2-3 years within say an 8-12 years history of the portfolio. Such performance should cover making the initial investment from practically any starting point rather than the usual 1, 3, 5 or even 10 calendar year performance graphs that are being typically offered and which, in this author's very humble opinion, are not much worth, simply because few investors start their investment on (or even close) to Jan. 1. Those who deal frequently with performance measurement issues, are well aware that even a small shift of any measurement period can and does inflict major variations in the reported performance. It is further important to have strong assurances that the investment philosophy and practice underlying the portfolio did not change over the period covered. Otherwise, although past performance never guarantees future results, the likelihood of any performance to continue in the future becomes even more questionable than it normally is.

Tying all this back to the earnings graph that caught the engineering part of my soul, the point is that the red growing forecasted portion of the curve (though short and a little hard to notice) will have to hit a peak, like all other cycles did. That peak seems now to be about 5-7 years away according to this graph and additional independent research does confirm this time frame. It is very useful to have someone with a provable track record (A) follow the developments for you and warn you in good time when a peak is approaching, and (B) help you in the meanwhile select stocks which are most likely to take advantage of the next few years of this growing period. It is unlikely to be a generalized growth: it is a stock-picker's market and the "weak sisters" will be weeded out - mercilessly.

An actual portfolio that has been run for 11 1/2 years on these principles proves that such an approach is feasible. It uses a set of highly disciplined analyst-behavior criteria combined with a DCF analysis for the buy decisions and a set of equally disciplined selling criteria. It produced in this period an average annual alpha of 6.5% after all brokerage and management fees were paid and it exceeded the S&P500 in all three year periods starting in any month during the almost 12 years. The portfolio was long only, with no leverage and the management approach did not change.

Reactions will be much valued on both points.

Disclosure: I have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours.