Over the past 13 years, and including the recent market advance, the S&P 500 has underperformed even the minuscule return on risk-free Treasury bills, while experiencing two market plunges in excess of 50%. I am concerned that we are about to continue this journey. At present, we estimate that the S&P 500 will likely underperform Treasury bills (essentially achieving zero total returns) over the coming 5 year period, with a probable intervening loss in the range of 30-40% peak-to-trough.
Why? First, with respect to 5-year prospective returns, it's important to recognize that returns at that horizon are primarily driven by valuations - not the "Fed Model" kind, but the normalized earnings and discounted cash flow kind. Stocks remain strenuously overvalued here, and only appear "fairly priced" relative to recent and near-term earnings estimates because corporate profit margins are more than 50% above their long-term norm. Meanwhile, corporate profits as a share of GDP are about 70% above the long-term average. As I detailed in Too Little To Lock In, these abnormally high margins are tightly related (via accounting identity) to massive fiscal deficits and depressed household savings rates, neither which are sustainable.
Our projection for 10-year S&P 500 total returns - nominal - is about 4.4% annually, which is far better than the 2000 peak, far inferior to the 2009 trough, and save for the period before the 1929 crash, worse than any prospective return observed prior to the late-1990's bubble - even in periods having similarly depressed interest rates.
Of course, rich valuations can persist for some time - predictably resulting in poor long-term returns, but often doing little to prevent short-run speculation and temporary gains. The issue is then to identify the point at which overvalued conditions are joined by sufficiently overextended conditions, and a sufficient loss of speculative drivers, to make rich valuations "bite" even in the shorter-term. This is where additional criteria come in, such as overbought technical conditions and extreme optimism in the form of low bearish sentiment, depressed mutual fund cash levels, and heavy insider selling. Presently, it doesn't help that T-bill yields and long-term bond yields remain higher than 6 months ago, and we have signs of oncoming recession. This is particularly evidenced by collapsing economic measures in Europe, softening economic performance in developing economies including China and India, and jointly weak year-over-year growth in key U.S. economic measures such as real personal income, real personal consumption, real final sales, and reliable leading indicators from the OECD and ECRI, as well as our own measures.
The combination of rich valuations, overbought conditions, overbullish sentiment, and deteriorating leading economic evidence can still unfortunately persist for months before being resolved. But once the hostile syndromes we've seen recently have emerged in the data, attempts at continued speculation have amounted to playing with fire. Similar conditions have repeatedly resulted in disastrous outcomes for investors. It would be nice to be able to "time" these outcomes better. We haven't found a reliable way to do so, and would still be concerned about robustness - sensitivity to small errors - even if we did. Yet even when unfortunate outcomes are not immediate, the fact that the S&P 500 has underperformed T-bills for 13 years is not very sympathetic to arguments that stock market risk has been worth taking overall, except in confined doses.
With regard to those doses, this would not be the time to swallow one. We presently identify market conditions as being in the most negative 1% of historical data based on the average expected return/risk characteristics associated with similar conditions, on a wide range of horizons ranging from 2 weeks to 18 months.
Continue reading.
Why? First, with respect to 5-year prospective returns, it's important to recognize that returns at that horizon are primarily driven by valuations - not the "Fed Model" kind, but the normalized earnings and discounted cash flow kind. Stocks remain strenuously overvalued here, and only appear "fairly priced" relative to recent and near-term earnings estimates because corporate profit margins are more than 50% above their long-term norm. Meanwhile, corporate profits as a share of GDP are about 70% above the long-term average. As I detailed in Too Little To Lock In, these abnormally high margins are tightly related (via accounting identity) to massive fiscal deficits and depressed household savings rates, neither which are sustainable.
Our projection for 10-year S&P 500 total returns - nominal - is about 4.4% annually, which is far better than the 2000 peak, far inferior to the 2009 trough, and save for the period before the 1929 crash, worse than any prospective return observed prior to the late-1990's bubble - even in periods having similarly depressed interest rates.
Of course, rich valuations can persist for some time - predictably resulting in poor long-term returns, but often doing little to prevent short-run speculation and temporary gains. The issue is then to identify the point at which overvalued conditions are joined by sufficiently overextended conditions, and a sufficient loss of speculative drivers, to make rich valuations "bite" even in the shorter-term. This is where additional criteria come in, such as overbought technical conditions and extreme optimism in the form of low bearish sentiment, depressed mutual fund cash levels, and heavy insider selling. Presently, it doesn't help that T-bill yields and long-term bond yields remain higher than 6 months ago, and we have signs of oncoming recession. This is particularly evidenced by collapsing economic measures in Europe, softening economic performance in developing economies including China and India, and jointly weak year-over-year growth in key U.S. economic measures such as real personal income, real personal consumption, real final sales, and reliable leading indicators from the OECD and ECRI, as well as our own measures.
The combination of rich valuations, overbought conditions, overbullish sentiment, and deteriorating leading economic evidence can still unfortunately persist for months before being resolved. But once the hostile syndromes we've seen recently have emerged in the data, attempts at continued speculation have amounted to playing with fire. Similar conditions have repeatedly resulted in disastrous outcomes for investors. It would be nice to be able to "time" these outcomes better. We haven't found a reliable way to do so, and would still be concerned about robustness - sensitivity to small errors - even if we did. Yet even when unfortunate outcomes are not immediate, the fact that the S&P 500 has underperformed T-bills for 13 years is not very sympathetic to arguments that stock market risk has been worth taking overall, except in confined doses.
With regard to those doses, this would not be the time to swallow one. We presently identify market conditions as being in the most negative 1% of historical data based on the average expected return/risk characteristics associated with similar conditions, on a wide range of horizons ranging from 2 weeks to 18 months.
Continue reading.
