Multiple times, both on Seeking Alpha and other sites (e.g., Quora, other investment sites), I have come across two extreme viewpoints on emerging market investing. EM Bulls argue that because properly chosen emerging markets have sustained periods of GDP growth much higher than the US (and especially than developed markets such as the EU and Japan), one should overweight emerging market ETFs, especially now as EM valuation discounts relative to the US exceed 2-sigma levels. EM Bears point out that economic numbers are unreliable, accounting standards are different, the history of EM index returns shows that EM lags US returns over the last decade or longer, the rule of law is often weak, corporate governance can be poor and correlations are high enough that there is little diversification benefit.
In this article, I look at the link between GDP growth and stock market returns, relying on a very solid paper by an economist who looked at over 100 years of data across multiple markets. Then I show the critical piece of logic that both these types of arguments are missing and how to find above-average stock market returns in EM.
GDP Growth vs Economic Profits
In Figure 1, I show real GDP growth versus market returns from 1900-2002
Source: GDP vs Equity Returns Paper
for Belgium, Italy, Germany, France, Spain, Japan, Switzerland, Ireland, Denmark, the Netherlands, the UK, Canada, the US, South Africa, Sweden and Australia. At the beginning of this period, all would meet our current definition of emerging markets while only South Africa would still meet the EM definition today. The country with the highest real GDP growth over this long period was Japan, but it ranks 6th from the bottom in this group of 16 nations in terms of equity returns. The country with the lowest real GDP growth was South Africa (which explains why it is still an emerging market, with less than 1% real growth in a world where most countries enjoyed around 2% real growth for a century) but it had the 3rd highest real equity return! Contrary to much popular belief, the US is not number 1 in either category over this 103 year period.
In the next Figure, from the same paper, I also show a shorter period (1988 to 2002) for another set of 13 countries, with real returns followed by GDP growth in the last two columns. Again, GDP and stock returns don't really line up.
In fact, over large data sets, the computed correlation is actually slightly negative.
Why Does This Happen?
Over long periods, fast developing countries have high real GDP growth rates due in part to the following causes: the government invests in real (roads, communications, ports) and intangible infrastructure (R&D, education, the rule of law). Workers migrate from low productivity activities like subsistence farming to factory work, start small businesses, and generally see rising wages and savings. R&D leads to technology which finds its way into capital investment, making all workers more productive. National output goes up on all fronts.
At the start of the period, a certain set of public firms exists. Much of the growth, however, occurs due to government spending, the activities of unincorporated individuals (picture someone who was an agricultural laborer, moves to Shanghai and starts a restaurant) and private firms. A lot more firms come into existence during the development process, but again mostly private. Some of them eventually become public companies. GDP growth and wage growth is often highest in countries without a lot of monopoly or oligopoly power on the part of firms (technically called monopsony when it comes to labor); in these highly competitive environments, economic profits are lower, last for shorter time periods and as a result, public market returns are generally lower than in nations with more monopoly or oligopoly structures.
For readers interested in reading more of the economics and research behind this idea, I highly recommend The Full Paper . The two main exceptions mentioned in the paper I want to delve into here are low valuations and "monopoly situations." The author finds that all else equal, countries do well when starting from a low valuation, as measured for example by the Shiller PE, or any other earnings yield measure that is normalized for business cycles. Additionally, monopolies tend to do well as they capture economic profits and resist competition eroding those profits over time.
EM Bear Evidence
In the next figure, I compare the returns of the S&P 500, a broad-based EM index (which, for example, the ETF EEM tracks), and the MSCI Asia-Pacific index(MXAP), since the end of October 2006.
Including the most recent (Great) recession, we see that the S&P500 has returned 8.45% a year, while the broad EM index has returned 3.42% and the Asia-Pacific index, which focuses on the tiger economies of Asia, returned only 4.28%. We can also see that during the Great Recession, both EEM and MXAP had greater peak to trough declines and took longer to recover through their prior peaks.
Thus, on the surface, the EM bears pessimism appears to be plausible: why invest in regions that have lower long-term returns and more downside risk in recessions? These graphs seem to agree with the predictions one might make from reading the economics paper in the previous section.
EM Bull Evidence
The problem with ETFs is that they invest in companies regardless of economic profits. The term "monopoly situation", when combined with "cheaper valuations" should really be thought of as the practice of buying fairly-valued or undervalued companies that have some form of moat. While true monopolies are one such example, any firm with a natural network effect, high returns on capital protected by intellectual property or brands, companies with high switching costs, or operating in an industry with large returns to scale or a first-mover advantage might also be expected to earn significant economic profits that last for some period of time. How might an individual investor who doesn't have the time to find such companies on their own look for opportunities to invest in EM markets?
One approach would be to find EM specialists who "talk the talk" as far as writing about value-based approaches and looking for companies with sustainable competitive advantages, and who also have some performance history so they look like they are probably "walking the walk" as well. For example, here is the investment philosophy on the website of Matthews Asia Funds, which has specialized in the Asia-Pacific region since the early 1990s:
We believe many of the region's widely used indices are backward looking and are not representative of the industries and companies that will be successful in the future. We look to invest in companies whose potential has yet to be fully recognized by the market.
Long-term focus on Asia
We believe a long-term approach is the most effective way to capitalize on Asia's evolution. Matthews has been investing in Asia since 1991 and we draw on our experience to identify companies that stand to benefit from the growth and development of markets throughout the region.
We employ a fundamental, bottom-up investment process that seeks to identify companies with sustainable long-term growth prospects, strong business models, quality management teams and reasonable valuations. This research process involves more than 2,500 company meetings each year.
So they seem to talk the talk. What about their results?
In the next figure, I compare three funds that have been around since 2006 from the Matthews fund family (I do not know of specialists in other regions that have as long of a history) against the S&P500 (which won handily when compared against the indices). The funds are a China-focused fund (MCHFX), a broader fund focused on the "Tiger Economies" within Asia, which includes China (MAPTX), and the broadest fund of all(MAPIX) which is a dividend-growth focused Asia-Pacific fund which includes some exposure to Emerging Asia such as Vietnam, Thailand, Cambodia, etc.
With the benefit of active management, it appears that the gap with the S&P 500 mostly disappears. Indeed, the China-focused fund has to date, slightly beaten the S&P 500, whereas the other two funds have been nudged out by the S&P 500 (8.45% vs 8.18% and 7.91% after expenses). During the Great Recession, it appears that the broader funds were considerably less volatile than the China-focused funds, which had a pretty scary peak to trough drop. Additionally, all of the active funds recovered through their prior peaks earlier (2010) than did the S&P 500 (2012) following that recession.
I don't know about you, but if I had seen the evidence presented in this article so far, all I would have concluded at this point is that if I want to invest in EM, it certainly shouldn't be through ETFs or index products, and I should probably stick to the Asia-Pacific region. However, I would not be convinced that there was any intrinsic reason to invest in the Asia-Pacific region over just continuing to have a home bias in favor of staying within the S&P 500.
This is where my final figure comes in. It uses a model which assumes that relative valuations between the US and other markets, measured on the basis of normalized, cyclically adjusted PE, revert to normal over the next 7-8 years. You may have read that right now, people are paying more for US equities relative to other markets (per unit of earnings) than they have 95% of the time in historical data.
It also takes into account current dividend rates, assumes no growth in dividends (which underestimates EM returns, since dividend rates tend to grow in line with GDP), but assumes that sales grow in line with recent GDP (which probably overestimates EM returns, since future GDP may slow relative to the last 5 years, but this offsets the dividend assumption in large part).
Granted, this is a simple model, it only applies over longish time horizons of 7-10 years, and is subject to some model error due to the three assumptions above. However, if we think about how much in annual percent terms each of the three assumptions noted above might be off by, relative to the size of the expected performance difference between the US and most of the rest of the world, we see that a very rough conservative confidence interval accounting for these errors might be for 3%-8% out-performance for well-selected EM equities. I don't know if that final piece of information swayed you, but it swayed me as I began to dig into it deeper. For readers interested in looking into this topic further, I also recommend looking at the relative valuation of the dollar, as it is an important component of EM returns as well.
The thought process engaged in by EM bulls who advocate the use of very broad ETFs to participate in the stock market growth that "must arise" in higher GDP growth nations is no different from the faulty thought process of people investing in high revenue growth, low or no profit companies such as Uber (UBER), Tesla (TSLA) and especially Beyond Meat (BYND). However, in dismissing the entire EM sector as a result, we are likely to be just like investors who failed to invest in companies like Microsoft (MSFT). The key is to look for moats and reasonable price if not outright fire sales, and of course be diversified. If we cannot do the work required ourselves, the next best thing is to pay someone who can do it well to do it for us.
Disclosure: I am/we are long MAPIX. 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.
Additional disclosure: I am also short UBER.