The oil price crash and global lockdowns have pushed the valuation ratio of international to U.S. stocks to levels not seen since 2000. Back then, the combination of a U.S. Tech bubble and Emerging Market crises resulted in 20+% annual international equity outperformance for the next decade. We estimate that our basket of diversified, developed and emerging international stocks should outperform by similar amounts over the next 10 years.
Investors are likely to look back in years to come and ask themselves why they were paying almost four times more for U.S. equity relative to the rest of the world at a time when the U.S. economy faced relative weakness and the dollar faced major policy-driven headwinds.
Fig. 1
Source: Bloomberg
The above chart shows our long and short portfolios. We are short an equal weighting of the SPX and the NDX and long an equal weighting of a number of international markets that trade at deeply discounted valuations and have undervalued currencies. Specifically, the U.K., Mexico, Chile, Colombia, Italy, Spain, Poland, Czech Republic, Turkey, Russia, Korea, Singapore, and Indonesia.
The following chart shows the average of the price-to-book ratio and the price-to-sales ratio for our long and short baskets on the left hand side, with the long basket now trading at its weakest level on record and the short basket still trading above its 2007 peak. On the right-hand side is the current trailing 12-month dividend yield of both indices which shows a similarly large divergence.
Fig. 2
Source: Bloomberg
As dividend payments tend to fluctuate more aggressively than sales or book values, we use the payout-adjusted dividend yield (see 'Global Equity Outlook: The U.S. Versus The Rest') as a way of adjusting the trailing dividend yield to a yield figure that gives a truer reflection of the sustainable level of dividend payments. In the U.S. for instance the SPX dividend payments as a share of sales, book values, and GDP are extremely high relative to historic high in part due to elevated profit margins which are highly likely to mean revert over time.
Fig. 3
Source: Bloomberg
The following chart shows how closely these valuation metrics have correlated with equity returns in the past. The last time the valuation divergence was as wide as it is currently, the long portfolio went on to outperform the short portfolio by ~20% annually for a decade. This excludes income returns from higher dividend yields in the rest of the world relative to the U.S.
Fig. 4
Source: Bloomberg, Author's calculations
Over a long-term time horizon the importance of valuations cannot be overstated. Many investors have argued that valuations no longer matter because we have seen expensive stocks get more expensive and cheap stocks get cheaper over recent years. However, this is simply not how valuations work. The following chart shows how the importance of valuations is a function of the investment horizon.
Fig. 5
Source: Bloomberg, Author's calculations
The longer the time horizon, the more important valuations become as a driver of returns. The chart below shows the correlation between the long/short valuation ratio and subsequent annualized returns over 1-5 years. Over the course of 1 year there is only a mildly positive correlation, but as the investment horizon increases, the correlation rises exponentially.
Fig. 6
Source: Bloomberg, Author's calculations
As the investment horizon rises past five years the correlation between valuations and subsequent returns becomes incredibly tight, to the point where at the 10-year horizon the R-squared between valuations and returns rises to a staggeringly high 0.97.
Fig. 7
Source: Bloomberg, Author's calculations
With this in mind, we could see international stocks continue to underperform in the near term as investors continue to chase large cap U.S. tech stocks to their 2000 extremes, but the odds of this happening are falling by the day. While the correlation between valuations and 1-year returns is low, the more extreme valuations become, the less likely it is that they continue to rise. As the charts above show, even on a one-year horizon we should still expect to see positive returns in our long basket relative to our short basket.
The dollar's strength over past few months has added insult to injury with regards to the performance of international stocks. As we argued here and here, the weakness in foreign currencies relative to the dollar has come despite a dramatic improvement in real yields relative to the U.S. dollar.
The chart below shows the spread of 10-year real bond yields for the countries in our long basket for which inflation-linked bond yields are available (the U.K., the eurozone, Chile, Mexico, Turkey, and Korea) relative to the U.S. Rising real bond yields should be supportive of international currencies but the panic has caused investors to completely overlook the fundamental picture.
Fig. 8
Source: Bloomberg, Author's calculations
From a long-term perspective relative high real yield spreads should help to see real exchange rates appreciate across our long basket, particularly given how cheap they have become. The average REER is now almost 20% below its long-term average which suggests we can expect roughly 2% annual gains in real terms over next decade. This expected tailwind is echoed in GMO's April White Paper.
Fig. 9
Source: Bloomberg, Author's calculations
The dollar's rise in the face of deteriorating fundamentals has triggered the U.S. Fed and Treasury into mounting an all-out monetary debasement drive. The dollar's reserve status has given U.S. policymakers greater leeway to pursue inflationary stimulus policies without leading to currency weakness. This dollar strength is a positive for the U.S. economy but conversely it is negative for domestic asset prices. We have no doubt that the priority of policymakers lies with supporting the financial sector at the expense of domestic purchasing power.
The fact that the Fed has stooped to buying junk bonds following a relatively mild peak-to-trough asset price drawdown is all you need to know when it comes to their priorities. For now, there has been little reaction in the FX markets but the longer the dollar stays strong, the more likely it is that U.S. policymakers throw more caution to the wind with regards to monetary debasement ultimately leading to long-term dollar weakness. The dollar's reserve status ultimately gives policymakers more rope with which to hang themselves.
We see U.S. growth averaging roughly 0.5% over the next decade from the 2019 peak as headwinds in the form of high debt levels, low savings rates, extreme financial distortions, and weakening demographics make the economy ill prepared to deal with the ramifications of the coronavirus pandemic (see U.S. Growth: Flattening The Curve). This may well turn out to be superior to the outlook facing the U.K., Italy and Spain, which face similar economic distortions and boast less flexible labor markets. Meanwhile, the Russian and Colombian economies face serious economic hardship from the collapse in oil prices. However, the majority of countries in our long basket should see growth far exceed the U.S.
Singapore and South Korea face significant demographic related growth headwinds but their fewer economic distortions and considerably higher savings rates should allow growth to exceed the U.S. Meanwhile, real GDP growth in Mexico, Chile, Poland, Czech Republic, and Indonesia should far outstrip the U.S. Even Turkey which faces political and external debt challenges should find some support from the collapse in oil prices.
Fig. 10
Source: Bloomberg, Author's calculations
In fact, the drop in oil prices should benefit growth more significantly in the average international long basket relative to the U.S. given the latter's newfound net oil exporter status. The chart above shows how the median terms of trade of the international long basket of countries has improved significantly as commodity prices have crashed. Overall, we think it is safe to assume that real GDP growth will average at least 0.5pp higher in the long countries relative to the U.S.
The spike in real interest rates seen in the emerging market economies is causing near-term economic disruption but is actually a blessing in disguise. While central bankers in the developed world are intent on keeping real interest rates negative to prevent businesses from undergoing a painful rise in borrowing costs, higher real rates are exactly what economies need to foster recovery.
The spike in real yields across EMs will help the process of creative destruction take place as weak companies will go out of business and resources will be directed to more productive uses, ultimately benefiting long-term growth. The strong growth seen across Emerging Markets in the 2000s had its roots in the painful economic crises which occurred in the late-1990s including the Asian Financial Crisis and the Russian default.
U.S. Short Portfolio: We forecast our short U.S. portfolio to return an annualized 1.3% from dividend income reflecting the payout-adjusted dividend yield, plus 0.5% from dividend gains due to real GDP growth. This implies 1.8% annual gains in the absence of any valuation mean reversion. In reality, valuations are highly likely to decline from current extremes, and the extent of the mean reversion will ultimately depend on the rate of real return that investors require on U.S. stocks.
Historically this has been 6.5% annually but there are reasons to believe that the Fed's stimulus efforts could reduce this to as low as 4% as explained here. This puts the total real return outlook somewhere between -7.6 and -12.8%. This compares unfavorably to the returns seen from 2000 to 2010 due largely to the weaker real GDP growth outlook.
10-Year Real Total Return Forecast By Component, %
Payout-Adjusted Dividend Yield | Required Rate Of Real Return | Real GDP Growth | Valuation Mean Reversion | Currency Performance Contribution | Total Real Return | Returns From 2000-2010 | |
U.S. | 1.3 | 6.5 | 0.5 | -14.2 | 0.0 | -12.4 | -6.8 |
U.S. | 1.3 | 4.0 | 0.5 | -9.4 | 0.0 | -7.6 | -6.8 |
International | 4.9 | 6.5 | 1.0 | -1.1 | 2.0 | 6.8 | 15.9 |
International | 4.9 | 4.0 | 1.0 | 5.0 | 2.0 | 12.9 | 15.9 |
Source: Author's calculations
International Long Portfolio: We forecast our long international portfolio to return an annualized 4.9% from dividend income, 1.0% from real dividend growth, and some degree of gains from valuation mean reversion. The degree to which we will see upside or downside valuation mean reversion will again depend on the required rate of real return and our base case is for total returns to range from +6.8 to +12.9%. Again, this compares unfavorably to the returns seen from 2000 to 2010 due in part to the weaker real GDP growth outlook and the fact that valuations ended 2010 at historically elevated levels.
Overall, we expect annual real total return outperformance in international stocks to range between 14.4% and 20.5% depending on the ultimate level of valuation mean reversion. This outlook is in line with the price performance implied by relative valuations in Figure 4 and is broadly in line with GMO's most recent 7-Year Forecast.
This article was written by
Disclosure: I am/we are short SPY, NDX. I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.