Brief Holiday Update
By John Hussman
December 31, 2012
With Monday as the last opportunity for Congress to avoid some version of the “fiscal cliff,” there isn’t a great deal to say that will be relevant for long, but my impression remains that the most likely outcome is that fiscal policy will go over the “cliff” – if only for a week or so. The last Congressional Republican to vote for a tax increase was Pete Domenici in 1990. Whether one agrees with that position on taxes or not, it seems unlikely that House Republicans will violate their pledge to their own constituents and vote to raise taxes on December 31 – when precisely the same outcome can be obtained by voting a few days later to lower them immediately after they are automatically raised.
The House has already passed legislation that, if passed by the Senate, would avoid the fiscal cliff but wouldn’t raise taxes. That clearly isn’t acceptable to Democrats. A slightly modified proposal to raise taxes only on incomes above $1 million was pulled before a House vote, because even that tax increase was too objectionable to garner enough votes. So now, there’s some talk of the Senate passing a bill and sending it to the House, though this would seem to violate Article I Section 7 of the Constitution, which requires all bills for raising revenue to originate in the House. Then again, the Federal Reserve has repeatedly violated the Federal Reserve Act to encroach on the fiscal authority of the House in recent years (see The Case Against the Fed), so we shouldn’t rule out some novel interpretation of the law on this matter as well.
In any event, we shouldn’t be terribly surprised if we briefly go over the fiscal cliff. The real problem would be if there is significant difficulty in quickly passing legislation in early January, or if Congress merely passes an extension of existing law, which would result in a replay of the current situation – since some solution would still be required to avoid the need for the House to explicitly raise taxes.
It strikes me that we have more hurdles for the financial markets beyond the fiscal cliff, including unresolved risks out of Europe, a likely but as yet unrecognized recession in the U.S., and continuing weakness in our return/risk estimate for the stock market, driven by a broad syndrome of overvalued, overbought, overbullish conditions. On a positive note, last week’s decline did ease the near-term overbought conditions of the market, and while the employment component of the Chicago purchasing managers survey deteriorated, there was a pickup in the new orders component that was of modest benefit to the “coincident/leading” measure I discussed in How To Build a Time Machine. Of course, we also saw a pop in the Chicago new orders component in August 2008, rising from 48.7 to 55.3, before plunging to 29 in the next three months, so one month doesn’t make a trend. Still, a series of upticks in Chicago new orders component and the Philadelphia Fed Index (ideally large upward leaps in both) would be among the first places to look for evidence that recessionary evidence is easing. For now, our measures remain negative on the economic front, but we’re open to new data as conditions change.
Though our concerns still weigh heavily toward the defensive side, there are hints of progress toward the resolution of the lopsided market conditions we’ve seen. Our measures of return/risk have been hard-negative for a year, and dropped into the most negative 1% of the data in March (the S&P 500 was below March levels as of Friday’s close). Based on historical precedents, I don’t expect these measures to remain negative much longer, and while the terminal portion of a market cycle is hardly enjoyable for most investors, significant downturns can result in remarkably large shifts in our return/risk estimates. We remain very open to a more constructive investment view as the evidence changes.
On the valuation front, our estimates of prospective 10-year S&P 500 total returns have benefited from some amount of growth in fundamentals since March and a slight decline in the S&P 500 Index, resulting in an increase in our 10-year estimate to about 4.7% annually (nominal). We’re sometimes asked how interest rates factor into this analysis. On this front, it’s important to recognize that the prospective return on a security is a function of the cash flows that it can be expected to deliver into the hands of investors over time, compared with the price paid in the present. Interest rates don’t actually enter that calculation. Instead, you make the comparison with interest rates after the prospective return on a given security is estimated. Alternatively, you can couple the estimated future cash flows with a “required” rate of return that reflects whatever beliefs you have about how interest rates affect stock returns, and then discount back to a target price, but that method is often very sensitive to bad assumptions about future rates of return (see the book Dow 36,000 for a good example of this non-robustness). As a rule, analysts like to choose infinite-horizon return estimates that are very close to the infinite-horizon growth rates they estimate for fundamentals – a practice that can be used to justify arbitrarily high price targets. We prefer to estimate the expected rate of return taking the current price as a given.
So the question then becomes, given that the 10-year Treasury yield is just 1.7%, isn’t an estimated 4.7% 10-year total return on stocks good enough? My impression is that though stocks may outperform bonds over a decade, that decade is likely to involve dramatic market losses and recoveries, so there is significant risk in attempting to “lock in” a 10-year return on stocks. By doing so, one surrenders the significant potential opportunity to accept a higher intervening rate of return at some point where prices are depressed and valuations are more typical.
Moreover, as I’ve noted frequently over the years, there is not even remotely a one-to-one relationship between 10-year bond yields and 10-year prospective stock returns. Stocks are effectively instruments with a duration of close to 50 years, while a 10-year bond has a fraction of that duration. Regardless of whether we use our 10-year prospective return estimates, or instead use 10-year realized S&P 500 total returns, we find that regressing forward 10-year S&P 500 total returns on the 10-year Treasury yield results in a slope estimate of anywhere between 0.2 and 0.4, depending on the period used, and the correlation itself is less than 0.2, so it’s a very weak relationship. More importantly, the intercept of that regression is as high as 11% and never materially below 9%. In other words, there is little evidence that prospective 10-year S&P 500 total returns revert around a mean that is materially under 9-11% even at very low interest rates. That’s problematic here, because we estimate that a 10% prospective 10-year total return on the S&P 500 is consistent with a level – given present fundamentals such as earnings, dividends, book values, revenues and so forth – of about 900 on the S&P 500.
Keep in mind, however, that we don’t require the market to move to what we view as fair value or undervalue in order to accept a significant exposure to market fluctuations. In general, the best opportunities emerge when reasonable (if not depressed) valuations are coupled with an early strengthening in market internals following a significant market decline. That sort of opportunity has emerged regularly in market cycles over history. Though much larger stress-testing concerns about Depression-era data complicated the 2009-early 2010 instance, the present cycle is also quite long in the tooth, and I doubt that market cycles have been permanently repealed by monetary policy.
Among cyclical bull markets in secular bear periods, the current advance – at about 45 months in length – is second in length only to the 2002-2007 bull market advance (see Hanging Around, Hoping to Get Lucky). On average, these advances have been followed by market declines of about 39% (the average for all bear markets is closer to 32%). Cyclical bear markets in secular bear periods have typically wiped out not just half of the preceding bull market gain – which is average – but close to 80% of the preceding bull market gain. It’s not clear what portion will be surrendered in the present cycle, or necessarily when, but I have every expectation of observing the normal range of full-cycle opportunities as we move forward from here.
The foregoing comments represent the general investment analysis and economic views of the Advisor, and are provided solely for the purpose of information, instruction and discourse. Only comments in the Fund Notes section relate specifically to the Hussman Funds and the investment positions of the Funds.
Past performance does not ensure future results, and there is no assurance that the Hussman Funds will achieve their investment objectives. An investor's shares, when redeemed, may be worth more or less than their original cost. Current performance may be higher or lower than the performance quoted above. More current performance data through the most recent month-end is available at www.hussmanfunds.com. Investors should consider the investment objectives, risks, and charges and expenses of the Funds carefully before investing. For this and other information, please obtain a Prospectus and read it carefully.
(c) Hussman Funds