Subject: Re: OT: S&P versus T-Bills?
From what I have looked at, it is better on average to invest now, rather than wait in T-Bills. As an example. Say you have an income stream of $1/day since the beginning of 1950. You can invest the $1 in the equivalent of SPY everyday, or you can invest in T-bills at 5% p.a. and when the market is a fixed amount below the maximum previous value, you can take all the money from T-bills and invest. Subsequent $1/day will be invested in the market if it is below your threshold, or held in T-bills until it again drops below the threshold.

You are right that, with all else ignored and only looking at the change in quotation (and with absolutely no regard with the amount of earnings you are receiving for each dollar invested), the 5 year return is on average higher (as well as the 1 year return) than average when starting from market highs. This can be considered to be a corollary of price momentum spanning multiple years (though rarely more than 15 years, as we have - until now - experienced.

On the other hand, the price/sales ratio and the CAPE ratio of the S&P500 is near an all-time high.Since 1880 the market has been at a lower valuation of 10 year real average past earnings more than 97% of the time.

So investing from this particular market high likely won't have an above average 5 year return.

You are observing an investment advantage for a particular strategy (in this case a variation of price momentum) as observed through all market conditions. You are then making projection that from this point, under average valuation conditions (making no distinction) you will have an investment advantage on average.

However we are far from having an average valuation condition. Valuations can be just about ignored when projecting 1-year returns, as the valuation has such a small effect, however the valuation really starts to matter for a 5 year return (and matters more so again for 10 year returns) and you can use that information in your favour, even combining with your observation that we are at an all time high.

For example, if you take all of the points in the past in which the market was at an all time high, and then sorted them into return deciles ranked by the CAPE, you will likely find that the 5-year returns (when starting from a market high) are very sensitive to the starting CAPE ratio. The 5-year returns starting from market highs, but the CAPE ratio in the upper 3% range will, will almost by definition, have resulted in significantly negative real returns.

- Manlobbi