Chief Investment Strategist
U.S. equities have had a bang-up year so far, along with many other major equity markets around the world. This has raised the question in many investors minds about how this market advance stacks up compared with prior bull markets, and whether we have come too far, too fast. In this report we set out to study the historical record of bull market advances and investigate the drivers of stock market returns. Bull markets tend to be driven in the first half by a surge in earnings, and in the second half by an expansion in valuations and the current bull market cycle seems to fit the historic trend. We conclude that the advance in equity prices has been supported by the growth in realized profits. Current valuations have moved above historical medians and rest at the median level that prior bull markets have peaked. However, we think an unexpected change in central bank policy remains the most likely trigger to imperil the continuation of this bull market. As we expect the normalization of Fed policy over the next year to be moderate enough in scope to be digestible by the stock market, we remain positive on the outlook for global equity prices.
EXHIBIT 1: FAIRLY TYPICAL BULL MARKET
Source: Northern Trust, Bloomberg, Ned Davis Research, Inc., Standard & Poors. Data through 11/22/2013. Bull market time periods as defined by Ned Davis Research. Price-to-earnings (P/E) ratio calculated using trailing 4-quarters Standard & Poors as reported earnings per share before August 1990 and operating earnings per share after August 1990 due to data availability. 11/22/2013 earnings are Bloomberg Trailing 12-months per share figures.
We examined the 20 cyclical bull markets that occurred since World War II, with a bull market defined as an extended period of above-average stock price increases, interrupted only by relatively short declines. Bull markets since 1980 are shown in Exhibit 1, along with the median for all bull markets since 1945. Bull markets have ranged widely in length of time and price return. Although our current bull run doesnt look too long at 753 days, it stretches to 1,696 days if you merge the two bull markets since 2009 into one. While the median bull market has lasted 700 days, the 1990-1998 bull market lasted 2,830 days while the 2002-2007 bull market lasted 1,826 days. Price-to-earnings (P/E) ratios on trailing 12-month earnings generally rise throughout the bull market. The current bull market started with a P/E ratio in line with the median starting level. Interestingly, the current P/E ratio is in line with the median ending level (more on valuations later). Since 1949, the stock market has been in bull markets around 75% of the time and bear markets the remainder.
Historically, interest rates appear to have had little impact on bull markets. The median change in the 10-year U.S. Treasury over the bull markets we examined was an increase of 0.47%, ranging from a 3.35% increase (1980-81 bull market) to a 3.90% decrease (1984-87 bull market). This is consistent with the research in our recent Investment Strategy Commentary, Sensitivity Training, which demonstrated that global equities have little factor exposure to interest rates (especially during periods of low interest rates). What really matters is why interest rates are changing, and what it says for future growth. If rates are rising because the economy is strengthening, that may not prove problematic. If rates are rising due to inflationary worries, however, the market may begin to discount the impacts of tighter monetary policy on the growth outlook.
To understand the drivers of bull markets, we decomposed returns by attributing stock gains to either earnings growth or valuation change. By subtracting the growth in earnings during the bull market from the change in price, we are able to see the effects of valuation (multiple expansion) versus earnings growth on the market. Over the life of the bull market, earnings tend to be the driving force as stocks grow into valuations (with the 1990s being a key outlier, as valuations got to extreme levels during that decade).
EXHIBIT 2: DECOMPOSING THE BULL
Source: Northern Trust, Bloomberg, Standard & Poors. Data as of 11/22/2013. Earnings defined as trailing 1-year operating earnings per share.
As earnings gains mature and normalize in the latter half of the cycle, valuations show some evidence of becoming a larger piece of the equation. In the 1990s, the second half of the bull market witnessed 98% returns from valuation expansion versus 54% from earnings (versus 23% and 24%, respectively, in the first half). In the second half of the 2000s bull market, returns from valuations were 3% while returns from earnings were 28% (versus -25% and 59%, respectively, in the first half). Ned Davis Research chose to define separate bull markets from March 31, 2009 through April 30, 2011 and from October 31, 2011 to present. However, we think many investors view them as merely two parts of an ongoing bull market that began in March 2009. If viewed this way, we once again see earnings driving the first half and valuations becoming a more important contributor in the second half of the rally.
We can also see the earnings trend occurring in Exhibit 3 though these are 12-month-trailing figures, so the severity of the earnings decline and subsequent improvement is smoothed. We have also displayed profit margins, which are now back to pre-financial crisis levels and represent historical peaks. As can be seen in the chart, earnings growth has been impressive in this cycle increasing by 102% in the bull market from 2009-2011 (versus 91% in the 1990s bull market and 80% in the 2000s bull market). However, lately, earnings growth has slowed and has increased just 12% in the current part of the bull market. Starting from the beginning of the 2009 bull market, earnings have increased a cumulative 146%. Importantly, margin expansion has been an important contributor to recent gains while revenue growth has been less impressive causing some to question how much longer the growth in earnings can continue. We believe that profits do need to reaccelerate here through revenue growth, given our view that profit margins are unlikely to expand further. It is our expectation that this will occur in 2014, as global growth looks set to accelerate and the United States benefits from less fiscal drag.
EXHIBIT 3: BULL MARKET IN PROFITS
Source: Northern Trust, Bloomberg. Data through 11/22/2013.
What has led to the strength in profit margins and how sustainable might they be? On a long-term basis, we would cite falling input costs (e.g., outsourcing) and restrained wages as a percentage of revenue as major contributors. In the United States, employee compensation has fallen to 62% of national income, down from a 67.5% peak in the early 1980s and back to levels realized in the late 1960s. This has been a global phenomenon, as labors share of national income has fallen as well in other major economies such as Germany, Japan and China. With wages rising in recent years in China and other emerging markets, the easy gains from outsourcing and offshoring are likely behind us. But a decline in profit margins, absent a recession, would probably require a shift in the relative profit share between capital and labor. Some of the key catalysts for this shift to occur are clearly identifiable, but dont appear on the horizon. These include a relative increase in the cost of investment goods, and/or an increase in the bargaining power of labor. These both appear to be longer-term issues, and therefore we think they are unlikely to be disruptive to this cycles bull market. We do not see a catalyst over the near to intermediate term for labor to gain bargaining power unemployment remains high, productivity is solid and capital investment remains cheap. A key driver to investment goods becoming more expensive would be higher interest rates, which we do not see as a likely outcome over the next five years. Moderate growth, disinflation in the developed world, continued monetary policy accommodation and significant global savings all suggest a slow upward movement in rates over the intermediate term.
As we tackle the question of valuations, we see that while valuations have expanded steadily for two years, they do not appear to be overly stretched. The current 17.1 P/E compares to a long-term median (going back to 1954) of 16.4 and a median level of 17.1 at the end of post-World War II bull markets. We do not believe the much higher Shiller Cyclically Adjusted P/E (CAPE) ratio which currently shows a ratio of 25.4 versus a median of 15.9 is instructive here as it includes the dramatic earnings fall from the global financial crisis. Our research into valuations leads us to one major conclusion: valuations have little predictive value for returns over the next year, but are predictive toward returns over the next five years. So while we shaved our long-term (five-year capital market assumptions) equity return forecasts this past summer to reflect the valuation expansion, we do not believe current valuations will prohibit further market gains over the next year. This also highlights our thesis that a continued market advance will need to be supported by an acceleration of earnings growth, reducing the pressure on valuations.
EXHIBIT 4: EXPANSIVE VALUATIONS
Source: Northern Trust, Bloomberg. Data through 11/22/2013. Investment-grade bond yield represents the Barclays Investment Grade Credit Index.
We think valuations also need to be evaluated in the context of the alternatives available to investors; from this perspective, valuations on equities appear attractive relative to bonds. In Exhibit 4, we compare the earnings yield of the S&P 500 to the yield available from investment grade bonds. While some investors use the Fed Model (which uses the yield on the 10-year U.S. Treasury), we use the corporate bond index in this analysis to reflect credit risk. Currently, the S&P 500 sports an earnings yield of 5.8% (the inverse of the 17.1 P/E), while investment-grade corporates are providing a 3% yield. From 2004 to 2009, these metrics were pretty much comparable. A divide started in the aftermath of the financial crisis, as low government bond yields, along with investor risk aversion, pushed investors into corporate credit. We think this valuation differential supports the prospects of equities over the near to intermediate term.
So what could emerge to end this long-running bull market? While a recession isnt required for a bear market to ensue, we have typically seen double-digit declines in the most cyclical areas of the economy, like in housing and the purchasing manager indexes. In recent cycles, surges in credit creation have led to financial market turmoil, as has significant tightening of central bank policy. In the current cycle, we would expand that list to include the risk of structurally lower growth. In this risk case, global growth disappoints again despite the best efforts of central bankers and markets develop stimulus fatigue.
EXHIBIT 5: CAUSE OF THE SLAUGHTER?
Source: Northern Trust, Bloomberg, IMF World Economic Outlook. Credit to GDP data through 6/30/2013, central bank balance sheet data through 10/31/2013, global growth estimates as of 10/7/2013.
As shown in the first panel of Exhibit 5, overall credit outstanding in the United States has not been increasing, but has been declining since 2008 (despite the surge in government debt). The household sector has been deleveraging, with mortgage debt falling from 73% of GDP to 56%. Domestic financial debt, reflecting the deleveraging of financial institutions, has also fallen from around 118% of GDP to 84%. Both of these reductions have already reduced economic growth. Government debt held by the public (federal, state and local) has boomed from 54% to 90% since 2008, although the rate of growth has slowed of late. So the only real area of credit excess in recent years has been at the governmental level, which reduces the near-term economic risk of reversal. While there may be individual instances of corporate credit market excess, the aggregate behavior is not worrisome. We would cite the fact that high yield (non-investment grade) issuers this year continue to use the proceeds of their issuances in a credit-enhancing manner, not showing an increase in credit deteriorating uses such as stock buybacks or acquisitions.
Changes in central bank accommodation are another potential headwind to a continuation of this bull market. The rebounding stock market has been a desired outcome of Federal Reserve policy (based upon comments by both current Federal Reserve Chairman Bernanke and former Federal Reserve Chairman Greenspan), with the hope that it serves as a bridge to improved economic growth. Stock market studies do show that the market has performed better during the periods where quantitative easing (QE) has been active (Bianco Research calculates a 135% advance during periods of QE, with a 27% decline in the periods of no QE), so we must weigh the potential effects of its reversal. We believe the Federal Reserve is the central bank closest to starting the normalization process, but we expect this process to only occur concurrently with improving economic growth. With the outlook for inflation remaining benign, we do not believe the Fed will be forced to prematurely remove accommodation. We also see the likelihood of the labor market reaching the Feds objectives over the next year as fairly low (and it could be argued that this wouldnt be so bad for equities). So while the issue of the end of easy money remains our top risk case for the markets over the next year, we believe it is a manageable risk.
A final concern we would cite as a potential catalyst to end the bull market would be simple exhaustion. If the improvement in global growth expected in 2014 and 2015 disappoints once again, investors could reach stimulus fatigue and conclude that monetary policy is insufficient to restore the global economy to stronger health. Without stimulative fiscal and regulatory policy in the developed economies (with the notable exception of Japan), the global economy could face years of stagnation in this environment. The growth picture in 2014 should receive a boost from a decline in fiscal drag globally. In the United States, fiscal drag of 2.4% in 2013 is estimated to fall to 0.7% in 2014. So if private sector economic performance grows at the same pace in 2014 as it did in 2013, growth should be 1.7% faster as a result of the lesser fiscal drag. This gives us some confidence on the near-term growth outlook, but the intermediate to longer-term uncertainty remains.
The markets spectacular gains since the lows of 2009 certainly give reason to pause, as does the recent uptick in investor sentiment and valuation. Our study of historic bull market cycles shows that this one isnt all that unusual, save for the fact that it has been highly dependent on central bank largesse. Valuations have now moved modestly above long-term market averages, but still look attractive as compared with the alternatives of investment-grade bonds or cash. We think the next leg of this market advance will be supported by renewed profit growth, as the global economy accelerates in 2014. That leaves the most likely culprit to cause the end of this bull market as an unexpected withdrawal of monetary policy support, which has been so supportive of risk taking over the last five years. We will continue to debate and evaluate this issue as our primary risk case to equity markets over the next year.
Special thanks to Joe Welsch, Investment Analyst, for data research.