These include long positions in higher-yielding US bonds and short positions in euro bonds, as well as hedged positions in high-dividend-yielding equities in Europe and dividend-growth stocks in Japan. In our Factor Fund portfolios, we emphasize relative growth (growth at a reasonable price, or GARP) in the US, but we’re more leveraged to value and income in Europe and Japan. At this point, we’re still defensively positioned in Asia ex Japan, as we wait for improvement in earnings revisions and profitability trends.
April’s capital-market performance was marked by sharp performance reversals (Performance, Displays 4–6). The US dollar dropped by 3%, reversing a fairly steady climb since mid-2014. At the same time, oil prices surged by about 20%, and previously underperforming Russian and Chinese equity markets posted similar gains.
It might be tempting to view this reversal as the reflation trade playing out, with markets reflecting stronger growth prospects. But we don’t see much support for that notion. Earnings revisions and economic surprises were still largely negative (Growth). Given this, we view the performance reversals as technicals or profit-taking—not evidence of an inflection point.
April’s trend reversal was most pronounced in equity markets. Stocks that had rallied since mid-2014 massively underperformed, while the worst-performing stocks rebounded. The same reversal occurred in terms of earnings revisions: stocks with positive relative revisions underperformed those with negative relative revisions. Further, a variety of growth and momentum strategies underperformed, with the biggest letdowns in North America and among the most crowded stocks and sectors (Risk and Correlation, Displays 5 and 6). Crowded trades in other asset classes experienced similar directional underperformance—including the US dollar.
Interestingly, we experienced a similar performance reversal away from crowded trades in March and April 2014. However, this year’s reversal hit harder. Many of our best-performing funds underperformed and laggards rebounded. At this point, we don’t see a need to adjust our investment positioning. But on the margin, the big market moves led us to trim our energy-related overweight and add to US small- and mid-cap exposures.
In the past few reports, we’ve looked at tactical considerations related to monetary policy, economic surprises and arbitrage opportunities in asset prices—with an actionable investment horizon of three to 12 months. It’s been nearly a year since we’ve discussed expected long-term returns by asset class (please see our July 2014 GMP). We concluded that risk premiums on traditional assets were low relative to history, and that it made sense to include alternative asset classes and leverage in strategic asset allocations.
This month we take a more granular look at long-term equity returns. As expectations of “depressed long-term equity returns” become more of a consensus view, we see a growing need to better understand the potential mechanism for structural equity underperformance—and how this view could turn out to be wrong (or premature).
Base Case: Modestly Lower Equity Returns
Equity indices have nearly tripled from the low point of March 2009, giving rise to concerns over the prospects of a near-term correction—and, more importantly, lower long-term returns.
On the surface, those concerns seem misplaced. Traditional valuation metrics seem more or less neutral for developed-market equities (Display 1). Of course, that’s a high-level assessment. Valuations vary substantially among regions; the US, in particular, is trading at the very high end of its historical range (Valuation [Equities/Bonds], Display 1). Still, there’s little evidence of a valuation bubble in equities.
Our Capital Markets Engine projects a range of potential returns for asset classes; it currently provides a median 10-year expected return of 5.8% for global large-cap equities (Display 2). That’s below a “normal” return level of 9%. The projected shortfall is partly tied to the assumption of lower inflation (and lower nominal earnings growth), but that accounts for only 90 basis points of the shortfall from “normal” returns.
The main driver of lower returns is the implicit assumption of a “normalization” in the relationship between nominal economic growth, return on equity (ROE) and interest rates. In other words, the base-case scenario for forward equity returns assumes that valuation multiples stay about the same, with earnings growth decelerating consistent with a lower cost of capital and lower nominal gross domestic product (GDP) growth.
Conceptually, it’s hard to argue against some of the normalization principles. For instance, assuming a 14% ROE with dividend and net-share buybacks at 50% of earnings, book value would grow at a 7% rate. That pace is simply too fast for nominal GDP growth of 5%. At those rates, corporate balance sheets and earnings would eventually dwarf all of GDP—and that’s mathematically impossible. From a historical perspective, when corporations overearn relative to the cost of capital, they inevitably end up taking on lower-ROE projects or creating a price umbrella that allows new entrants to come into the market. Either way, that profitability should fall back toward the cost of capital.
Our tools suggest that we should expect a relatively modest decline in expected returns, but we can conceive a much more pessimistic scenario. For example, with corporate net margins 30% above their historical average, mean reversion could put the brakes on earnings growth for years. We see this view as overly pessimistic, because it fails to account for much of the margin improvement being “structural”: coming from lower taxes, lower interest payments and reduced depreciation expense.
In fact, operating margins for global companies aren’t actually exorbitant—they’re only modestly above historical averages (Display 3). So, it’s critical to understand both the mechanism and timing for a normalization of ROE when assessing long-term expected equity returns. There’s also a trade-off with rates: if nominal growth and inflation accelerate, rates will also rise, which would depress valuation multiples.
Laying Out the Equity Market Scenarios
Could disciplined capital use by firms push back the onset of eroding profit margins and lengthen the bull market? It’s possible, but evidence of pricing power becomes the key to this scenario. To assess the possibility, we created a simple dividend-discount valuation model to look at the fair value of the global equity market under different assumptions (Display 4). Instead of trying to pin down a specific price target, we focused on gauging the market’s sensitivity to different variables.
The base-case scenario (scenario 2), which appears to be consistent with current valuations, is that ROE and payout levels hold near their current levels, modest nominal economic growth is sustained, and interest rates rise slowly. A major downside scenario (scenario 1) would see ROE and payouts fall back toward normal levels while interest rates rise near nominal GDP growth rates. And in the upside scenarios (scenarios 3 and 4), real interest rates remain low by historical standards, but profitability stays at current levels and there’s some capital “misuse,” as payouts to investors remain high. Which outcome is most likely?
The Profitability Perspective
Let’s start by looking at the inputs, beginning with the prospects for sustained ROE, or profitability. From an overall standpoint, global ROEs don’t seem unusually high relative to history (Display 5). But these numbers mask two trends. First, the long-discussed buildup of excess corporate cash boosted book value and reduced ROE. Second, the profitability of the financials sector was depressed by regulatory pressures, while the energy and commodity sectors were depressed by overinvestment. If we adjust for those areas, profitability actually ends up close to historical peaks. It’s possible to quibble with the approach of excluding cash and certain sectors from ROE, but it does highlight that many sectors are actually overearning relative to their histories.
As we discussed in prior reports, we see little evidence of firms initiating low-ROI investments or growing their cost structure ahead of revenues. In fact, the growth of book value has largely tracked GDP growth since 2010 (Display 6). This is partly due to the increase in the payout ratio—with higher payouts reducing the funds available for reinvestment. In other words, companies are seeing few attractive growth opportunities (Display 7), and are increasingly choosing to return cash to shareholders.
The “sustainable” growth rate today (book-value growth after returning cash to shareholders) has declined near levels observed in the last three recessions (Display 8). Today, corporations in developed markets are paying cash to shareholders through dividends and share buybacks at a level of 64% of earnings. Will this payout ratio fall back to the more normal 50%, causing excess cash to build up and leading to unprofitable investments on corporate balance sheets?
In the US, the total payout ratio (Display 9) ranged from 45% to 50% for most of the post–World War II period, until rebounding after 2008. But if we look at the prewar era, corporations normally returned 65% to 80% of their earnings in dividends.
Can payouts stay at levels consistent with their long-term history? We think there’s a reasonable chance they will. Until there’s clear evidence of accelerating growth, corporations will invest conservatively; they’ll need less cash and can keep payouts higher. Barring an outright price war and/or a deflationary environment, we see a reasonable chance that ROE and payouts will remain at historically elevated levels.
The Interest-Rate Perspective
The level of interest rates is the second major consideration in assessing equity market valuations. Nominal short-term and long-term interest rates are extremely low (Display 10), even in the long-term historical context—punctuated by periods of deflation. Low real interest rates are somewhat less extreme in historical context, because there have been prolonged periods when rates were near zero (Display 11).
We interpret positive real interest rates as a premium investors demand for bearing the risk of an inflation shock. Historically, these shocks have typically been associated with wars, but more recently they’ve been linked to policy mistakes and commodity prices (Display 12). This could be an argument for a positive correlation between real interest rates and evidence of wage inflation. In other words, investors won’t demand a “risk premium” for inflation until there’s actual evidence of inflation.
Summing It Up: Where Is the Market Today?
What does this analysis say about the four scenarios in Display 4? We think scenario 1, the major downside scenario, is unlikely: interest rates could normalize, but a significant upturn in corporate hubris and investment would be needed to drive margin erosion. Likewise, scenario 4 seems far-fetched, because there’s little precedent for persistently negative real interest rates (which result in the average real rate of zero). So, we think the market is either fairly valued or undervalued by 10% to 20%, depending on the extent that central banks continue quantitative easing and driving extremely low interest rates.
Premium for Shareholder-Friendly Companies
Companies that engage in shareholder-friendly capital use are still trading at a valuation premium today. In this landscape, we think a focus on improving capital use may make the most sense.
The premium for high-yielding stocks is now historically high (Display 13). Following the 2000 tech bubble debacle, investors began to emphasize shareholder-friendly capital use above growth potential, which led to the expansion of valuation ratios. Disciplined capital use has become increasingly important in determining equity market valuations, due to high demand for stable income because of low rates and demographic trends.
Today, the price-to-forward-earnings ratios of companies with both high dividend yields and high share-repurchase yields are near historical highs. The valuations of stocks with the highest dividend yields alone are similarly extreme. There’s no evidence that the demand for yield will subside, so there are two ways for investors to reduce valuation risk in this segment. First, source high-dividend yields from Japan and possibly Europe, where premiums aren’t as high (Display 14). But there are drawbacks: very low relative yields in Japan and exposure to economic cyclicality in Europe.
An alternative approach would focus on dividend growth rather than dividend level. This may require more in-depth research, but it could also lead to a substantial payoff, because firms raising their payouts could be re-rated by the market (Display 15). This is hardly a novel observation; there’s substantial interest in “dividend growth” funds and a growing chorus calling for more capital discipline in Japan. But the valuation gap and opportunity are still present, because money flowing into the highest-yielding funds dominates, forcing a higher premium.
Our portfolios remain overweight risk (and reflation-oriented) assets under the assumption of improving nominal economic growth, strong liquidity and the persistence of accommodative monetary policy. Among risk assets, our biggest overweights are in equities and energy, with more modest exposure to high yield and an underweight in real estate investment trusts (REITs). Within equities, we favor cyclical sectors, non-US equities, and value strategies in Europe and Japan.
However, we believe exposure to income-generating investments is fairly important to reduce risk. In particular, bouts of deflation fears that caused a sell-off in the second half of 2014 could easily reemerge. That’s why we maintain exposure to a variety of carry trades across assets. For example, we’re long US, Singapore and Australia bonds while we’re short Bunds and Japanese government bonds. Recently, we’ve been adding to high-dividend-yield equities in Europe and dividend growers in Japan.