Advisor Perspectives welcomes guest contributions. The views presented here do not necessarily represent those of Advisor Perspectives. The views of this article represent those of the author and do not necessarily represent the views of Capital Advisors, Inc.
Might today’s historically low interest rates in the U.S. persist for years to come? The latest Quarterly Review and Outlook by Van Hoisington and Lacy Hunt of Texas-based Hoisington Investment Management forces readers to consider that possibility, refuting the reversion-to-the-mean mindset that causes many people to expect higher interest rates in the not-too-distant future.
If the Hoisington model for the economy turns out to be right, the implications for the stock market are unfavorable. That may seem counterintuitive; lower interest rates, after all, generally increase market valuations. But in this case they portend a slowly growing economy, which will depress prices.
The key variable in Hoisington’s forecast of persistently low – if not even lower – interest rates is an expectation of sluggish economic growth for the next decade or so. Hoisington and Hunt cite three real-world examples and three academic studies to support the idea that the structural growth rate of a nation’s economy declines by at least one percentage point whenever countries reach a certain threshold of indebtedness. One percentage point may not sound like much, but for a mature economy like the U.S.’s that historically grows about 3.0% annually, net of inflation, it cuts growth by a third.
Citing the research of Reinhart and Rogoff, Hoisington and Hunt suggest that developed economies slip into a prolonged period of slower growth whenever government debt approaches 90% of the value of the overall economy, as measured by gross domestic product (GDP). The current government debt-to-GDP ratio in the U.S. is nearly 100%.
Hoisington and Hunt make the case that 2008 marked the beginning of a fourth episode of suppressed economic growth in the wake of over-indebtedness, following the three prior examples they cite in their report, two of which occurred in the U.S. and the other in Japan. The first was triggered by a massive debt buildup to finance the railroads in the U.S. in the 1860s-70s. The second was the period of over-investment and rampant asset speculation in the 1920s that led to the Great Depression. The third example was the debt-financed expansion in Japan that peaked in 1989. In the words of Hoisington and Hunt:
“The relevant point to take from this analysis is that U.S. economic conditions beginning in 2008 were caused by the same conditions that existed in these above-mentioned panic years. Therefore, history suggests that over-indebtedness and its resultant slowing of economic activity supports the proposition that a prolonged move to very depressed levels of long-term government yields is probable.”
The inability of the U.S. economy to reach escape velocity three years into a feeble recovery from the last recession supports the Hoisington model. If this worldview gains wider acceptance over time, it is likely to influence stock prices negatively in the decade to come. Specifically, stocks will likely be priced at a lower average multiple of earnings if investors internalize a slower expected growth rate for the economy. When combined with today’s below-average starting point for the dividend yield from stocks, a shrinking price-to-earnings ratio (P/E) would make it virtually impossible for the U.S. stock market to deliver double-digit returns over the next seven to 10 years.
This conclusion runs counter to conventional wisdom about stocks, which holds that lower interest rates should lead to a higher P/E ratio in the stock market. Conventional wisdom makes sense as long as the long-term outlook for the economy and corporate profits remains stable. But if economic growth experiences a secular shift lower, as Hoisington and Hunt expect, the reduction in earnings growth will overwhelm the benefit of low interest rates in the valuation calculation of the stock market.
Ed Easterling of Crestmont Holdings, LLC has conducted extensive research into the variables that determine the P/E ratio in the stock market. One is the expected growth rate of corporate earnings. Since long-term earnings growth is bound by the pace of economic activity in aggregate, a secular decline in the growth rate for GDP should translate into a similar reduction in the long-term growth rate of corporate earnings.
Easterling’s research suggests that a 1% decrease in the long-term earnings growth rate implies a two- to seven-point reduction in the average P/E ratio for the stock market. He bases his conclusion on a range of analytical approaches, including a long-term dividend discount model, academic formulae, and a deductive process using the components of return in the stock market. The conclusion from each approach is consistent directionally, even if the range is broad.
Investors should expect a lower range for the P/E ratio in the stock market in the years to come if the U.S. economy continues to struggle under the weight of an excessive debt-to-GDP ratio.
The table below provides a matrix of possible outcomes for the S&P 500 Index over the next 10 years, based on different of assumptions for the pace of future earnings growth and the trend in the P/E ratio. The table uses the Shiller method for normalizing the earnings per share and P/E ratio of the S&P 500 Index.
10-Year Target for the S&P 500 Index
Real Earnings
Growth |
+2
|
Unchanged
|
Change in P/E
-2.0 |
-4.0
|
-6.0
|
1.0% |
1,657.6 |
1,518.5 |
1,379.3 |
1,240.1 |
1,100.9 |
2.0% |
1,829.4 |
1,675.7 |
1,522.2 |
1,368.6 |
1,215.0 |
3.0% |
2,016.8 |
1,847.4 |
1,678.2 |
1,508.8 |
1,339.5 |
4.0% |
2,221.4 |
2,034.8 |
1,848.4 |
1,661.9 |
1,475.4 |
5.0% |
2,444.4 |
2,239.2 |
2,033.9 |
1,828.7 |
1,623.4 |
Source: Capital Advisors, Inc.; Robert J. Shiller - www.econ.yale.edu/~shiller/; Standard & Poor’s
For context, here is a summary of recent conditions in the stock market, as of July 2012:
S&P 500 Index Price |
1,369.8 |
Normalized Earnings |
$ 62.98 |
Shiller P/E (July 2012) |
21.8 |
Shiller P/E (100-Yr. Avg) |
16.3 |
Real Earnings Growth1 (1991-2010) |
2.5% |
1. Real Earnings Growth represents the 20-year compound annual growth of the normalized earnings per share of the S&P 500 Index, as calculated by the Shiller method, between 1991 and 2010.
It is possible to devise a scenario for decent long-term returns in the stock market from the inputs listed above. For instance, if real earnings growth compounds at 5.0% per annum over the next 10 years and the normalized P/E ratio increases by two percentage points (this combination appears as the lower left box in the table), then the S&P 500 would grow around 6.0% annually before dividends, and maybe 9.0-9.5% including dividends. But note that 5.0% earnings growth is double the pace of the past 20 years, when America’s debt-to-GDP ratio was not dragging the economy downward, as it will over the next decade.
A more sobering forecast would see future earnings growth around 2.0% or lower, as implied by the Hoisington model of the economy, and a decline in the normalized P/E ratio of at least a couple percentage points, as indicated by Easterling’s research. In this scenario, U.S. stocks could very conceivably zigzag for another decade without making any lasting progress to the upside.
Other long-term forecasting approaches support low expectations for stocks in the coming decade. One approach tracks the correlation between the beginning valuation of the stock market, as measured by the Shiller P/E ratio, and subsequent 10-year returns. Over the past 140 years, there have been 177 10-year holding periods (measured monthly) that began with a Shiller P/E between 20 and 24, a range that captures the recent Shiller P/E of 21.8. The average 10-year return among all these periods was 6.4% annualized, including reinvested dividends.
Another approach, popularized by GMO’s Jeremy Grantham, assumes corporate profit margins and P/E ratios will revert toward their long-term mean over rolling seven-year forecast horizons. Grantham’s latest seven-year forecast for domestic large-cap stocks projects an annualized return of less than 4.0% from today’s starting point.
A strong case can be made for a secular decrease in economic activity in the U.S. until the nation’s debt-to-GDP ratio returns to a sustainable level well below 90%. This process might take a decade or more. Even though expected returns in the bond market look terrible, bonds still have an important role to play in most investors’ portfolios because interest rates may remain suppressed for many years to come, and they may even go lower if deflation takes root in the global economy.
For the stock market, multiple analytical approaches suggest long-term returns may be well below average for U.S. stocks from today’s starting point. Stock dividends take on added significance when expected returns are low because a dividend yield of 3.0% or better may represent more than half of the total return from stocks over the next seven to 10 years. High-quality, dividend-paying stocks are a smart bet today.
Finally, there is nothing wrong with a little extra cash. It pays nothing, but it gives a portfolio options. When the menu of investment opportunities is unattractive, hold some cash and wait for the markets to deliver better choices at some point in the future. It always does, eventually.
References
Hoisington, Van, and Hunt, Lacy, 2012, Quarterly Review and Outlook – Hoisington Investment Management
Easterling, Ed, 2010, Probable Outcomes – Secular Stock Market Insights, (Cypress House, Fort Bragg, CA)
Easterling, Ed, 2012, Crestmont Research, LLC
Keith Goddard is the CEO and Chief Investment Officer for Capital Advisors, Inc., a Tulsa, Oklahoma based asset manager. Keith is also the co-manager of the Capital Advisors Growth Fund (CIAOX). As of June 30, 2012, Capital Advisors served as manager and advisor to approximately $1.1 billion in client assets.
Read more articles by Keith C. Goddard, CFA