Since late-February, our estimates of the market's prospective return/risk tradeoff (over a set of horizons from 2 weeks to 18 months) have persistently held in the worst 0.5% of all historical observations. It's always important to emphasize that we try to align ourselves with the average return/risk profile that has historically accompanied the particular set of investment conditions we observe at each point in time, but that the outcome in any specific instance may not reflect the average return, and may even fall outside of what we view as the likely range of outcomes. That said, the awful behavior of the market in recent weeks is very run-of-the-mill in terms of how similarly unfavorable conditions have usually been resolved historically, and there is no evidence that this awful prospective course has changed much. The chart I included three weeks ago in Dancing at the Edge of a Cliff presents similar periods for historical perspective.
It's probably needless to say that last week's decline improved valuations modestly - we presently estimate prospective 10-year total returns (nominal) for the S&P 500 about 5.5% annually, based on our standard methodology. Most bear markets have historically ended only after prospective returns moved above 10% (including bear markets in periods of very low interest rates, and also including 2009). Moreover, regardless of whether interest rates have been high or low, extended secular bear markets have ended - and secular bull market advances have begun - only when prospective 10-year returns have reached about 20% annually (see Too Little To Lock In for a chart on this). So it won't come as a surprise that we don't view a 5.5% annual prospective total return as having much investment merit. You don't "lock in" prospective stock market returns - you ride them out, and holding on for the expectation of a 5.5% prospective annual return is likely to involve a very bumpy 10-year ride.
Investors with most of their assets already invested and unhedged should hope that prospective market returns move no higher than about 8% through the completion of the present cycle, since even touching a prospective return of 10% in the interim would require an S&P 500 in the mid-800's. Though I think it's plausible that we'll establish prospective returns consistent with the start of a secular bull market at some point in the next few years, actually quoting the associated level for the S&P 500 would only strain credibility here. Investors have forgotten so much after just 3 years time that it seems fruitless to talk about secular lows that only occur every 30-35 years (even if the last secular low was all the way back in 1982).
At this point, the S&P 500 has achieved a cumulative total return of less than 10% since April 2010. Meanwhile, of course, there remains a great deal of faith in the "Bernanke put," because even though it's fairly obvious that QE has done nothing durable for the economy or the financial markets over the last couple of years, a hit of QE might at least be good for a few months of "risk on" delirium. If the American public can't get thoughtful economic leadership, at least Wall Street's speculative junkies can hope for a little taste of Q from Sugar Daddy.
One of the problems with QE here, however, is that it would essentially represent fiscal policy for the benefit of speculators, at taxpayer expense. To see this, note that the 10-year Treasury yield is now down to less than 1.5%. One wonders how Bernanke would be able to argue, with a straight face, that this is not low enough. Nevertheless, a 10-year bond has a duration of 8 years - meaning that each 100 basis point fluctuation in interest rates is associated with a change of about 8% in the price of the bond. So if you buy the bond and hold it for a full year, an interest rate change of of 1.5/8 = .1875, or less than 20 basis points, is enough to wipe out the annual interest and leave you with a negative total return.
So at this point, if the Fed buys Treasury bonds, it will predictably lose money - after interest - unless interest rates rise less than 20 basis points a year during the period that the Fed holds those bonds. Over the past year, the standard deviation of week-to-week changes in the 10-year Treasury yield has been about 13 basis points, so 20 bips over the course of a full year is nothing. Whether or not a speculator is willing to take a bet on lower yields, it's highly unlikely that the Fed could buy Treasury bonds here at a yield of 1.5% and ever expect to unload its portfolio later at even lower yields, because yields would shoot higher merely on the anticipation of Fed liquidation.
As a result, Treasury debt purchased by the Fed here would almost certainly result in capital losses, at taxpayer expense, and those capital losses would be an implicit subsidy to speculators who sold those bonds to the Fed at elevated prices. Of course, "sterilized QE" - where the Fed would buy bonds, and then pay banks 0.25% interest to keep the balances on reserve - would involve an even larger subsidy, and would then require only a 15 basis point move to put the Fed into loss mode.
"QE3 - subsidizing banks and bond speculators at taxpayer expense" - there's a pithy slogan. That doesn't mean the Fed will refrain from more of its recklessness (which will be nearly impossible to reverse when it becomes necessary to do so), but does anyone actually believe by now that QE would improve the economy, durably elevate risky assets beyond a few months, or materially relieve global debt strains?
Despite the uncertainties, our game plan remains fairly straightforward. As I noted two weeks ago in Liquidation Syndrome, "there may be latitude to take a more constructive stance between the point that any new monetary intervention produces an improvement in our measures of market internals, and the point where we re-establish an overvalued, overbought, overbullish syndrome. Without a material improvement in valuations or market action here, we remain defensive. Undoubtedly, the best outcome would be a strong improvement in valuations, followed by signs of improvement in our measures of market action, which is the typical sequence of events that complete a market cycle and can launch a very favorable investment environment.
"In the meantime, however, a bad week is unlikely to change our assessment unless that bad week includes a market crash. Last week was not a crash, though a free-fall appears increasingly possible, as the reality of emerging recession (and all that it implies for fresh credit risks, sovereign defaults, fiscal imbalances, banking strains and other problems) will likely smash against the consensus view of economic expansion in next few months. I continue to view the U.S. economy as most probably entering a recession that will ultimately be marked as beginning in May or June of 2012."
