Diversification Strategy: Correlation Mistakes And Inflation Exposure

by: Veni Vidi Emi


Wealth managers and market pundits have praised diversification for years.

Many misunderstand the actual opportunities to diversify, while touting improper means of diversification.

We look at asset correlation and diversification alternatives using empirical studies, market experience, and general heuristics.

If you think emerging market exposure, real estate, carry trades, value factor exposure, or bonds make your portfolio bullet-proof diversified, read on.

Investors who are focused on minimizing risk, usually owners of retirement accounts or risk-averse individuals, often talk to wealth managers and have one word thrown at them repeatedly: diversification. Many financial pundits will claim that diversification is the only "free meal" in investing. Sometimes, the term "holy grail" is even thrown around. Below is a quick introduction to the common interpretation of a diversified portfolio.

Besides the rather blasphemous nature of the religious allegories, they are often used in the wrong context. In fact, 'diversification' is often used as a pretense for brokers to invest in illiquid assets with large bid-ask spreads to profit off risk-averse retail customers.

While your wealth-manager and broker should be compensated for providing you a service, you should make sure it's at least the right one.

In this article I will argue that:

  • Market-cap diversification isn't material when it matters;
  • Emerging market exposure is not diversification;
  • Bonds offer diversification, but not entirely;
  • Inflation-protection can be optimized intelligently for positive carry.

We also look at proposed solutions to attain actual diversification.

Diversification: Why'd You Want It Anyways?

Diversification is great because it reduces volatility.

Volatility hurts our potential access to leverage (unlikely if you're a retail investor), volatility means we risk having to sell assets at exactly the wrong time, and volatility ensures we don't get adequate sleep. All-in-all volatility, while not the end-all-be-all of risk measurements, is net negative.

In other words: given two similar assets, we would prefer the less volatile.

The regular spiel is that diversification is great at reducing volatility. Why? The law of large numbers means that the more return-streams we have, the more likely we are to experience returns closer to our long-term expected average.

Volatility reduction, source.

The volatility is reduced even further if we diversify with uncorrelated assets.

Diversification through uncorrelated assets. Source.

The solution seems simple, easily presentable at retail investor pitches, sounds smart, and looks great in chart-form: buy a diversified basket of uncorrelated assets to avoid volatility.

The method is simple. Test the correlation of many asset classes, factors, types of equity, size of equity, and alternative financial instruments to each other. Combine the assets with the least correlation.

Often you will end up with a correlation table that looks similar to the one presented below from Morningstar:

Great. If our core portfolio is U.S. mid- and large-cap growth companies, then we can diversify using small cap value stocks, foreign stocks, emerging markets, high-yield bonds, and real estate. We now have a nicely diversified portfolio that will protect us during downturns! Of course, we also add some investment grade bonds and cash, but both yield almost nothing! What good is diversification at inflationary levels of return? No, clearly it is superior to have high-yielding assets that offer diversification... right?

When The Film Breaks And The Fairy-tale Ends

If everything was this simple, this article wouldn't need to exist. Based on the research, I find that correlation and diversification are myths when markets are truly pressured. Who cares if their upside returns are diversified? As mentioned the value is in protecting the downside from undue volatility.

The problem is that many assets are uncorrelated on the positive side, but highly correlated when the markets turn. This leads to the illusion of correlation as a result of statistics, but the truth of the matter is that downside correlation through tumultuous markets approach 1 for many assets.

In "When Diversification Fails", a paper released on the 30th of May 2018 for the CFA journal for financial analysts, Sébastien Page, CFA, and Robert A. Panariello, CFA of T.Rowe Price discuss the downside correlation between asset classes. Below are selected charts and my commentary:

As it can be seen the correlation profiles during selloffs and rallies differ widely. For example, stock rallies and real-estate rallies look entirely uncorrelated (and thus strengthen diversification). In fact, investments in other industrialized countries (55-60% correlation in the table from Morningstar) are almost 90% correlated during U.S. stock selloffs.

The truth is that the economy is deeply intertwined. While a Parisian cafe might primarily serve local residents, the local resident income still depends on large corporations, who in turn depend on the industrial welfare of the United States.

The same holds true for high-yield bonds, stocks irrespective of capitalization, emerging market bonds, and mortgage-backed securities. The only two real holdouts are real-estate and corporate bonds.

What about portfolio strategies, such as implemented by hedge funds? These can either be simulated or invested in by accredited investors.

Source: When Diversification Fails

The takeaway is obvious. Most styles are highly correlated with stock selloffs. A few are less correlated (equity market neutral and merger arbitrage), but both have historically faced rapidly decreasing returns as the strategies became popular diversification tools (link 3,4 in the appendix).

The choice for hedge-fund styles, then, either promise little diversification or consistently low returns.

So, you can't diversify using traditional asset-class dispersion. What can you do?

Strategy 1: Hedge Better Using An Equity-Bond Portfolio

The Equity-bond mix portfolio is the traditional response to high downside correlation among other asset classes. This portfolio construction has empiric backing.

In fact, a treasury-bond + risk equity portfolio (along the line of Nassim Taleb's barbell strategy) has a well-balanced risk profile when it comes to business cycle risk.

I italicized the last 3 words for good reason. The conditional correlation outlined above showcases the 'flight-to-safety' during business cycle downturns, but bonds and stocks can also depreciate rapidly at the same time for various reasons.

In the Journal of investment strategies 2014, 3-18 edition, Nic Johnson looks at stock-bond correlations across a wide range of macroeconomic situations and concludes that correlations between bonds and equities range due to a wide variety of factors. Often you will see negative correlation during business cycle downturns due to a "flight-to-safety" expectation, but inflation can reverse the correlation and cause both assets to decline materially at the same time.

The end result is that a mix-match of equity & bonds will lead to reduced volatility from business cycle changes, but won't protect you from heavy inflation/monetary risk. The risk is even more prevalent in tail events:

Which means that risk relating to the federal reserve/quantitative easing and inflation is likely to persist across bond-diversification, especially during stress events.

Strategy 1b: Removing Inflation Risk

Seeing as our correlated risk assets (real-estate, large-cap, small-cap, value stocks, growth stocks, high-yield bonds, emerging and foreign market equities & debt) are likely to depreciate materially in tandem during business cycle downturns, we diversify using high-grade government bonds.

The core risk factor that remains is heavy inflation. There are two interesting ways to counter inflation-risk.

The traditional method is to hedge using a specific set of commodities, some preferable to others. We will look at the literature for this.

The more modern method is to buy equities that remain stable (or appreciate) during inflationary periods. This method earns positive carry. I refer to these as inflation beneficiaries.

One can also engineer their portfolio through the paradigm of inversion and simply avoid equities in industries that are hurt by inflation.

When looking at commodities, we take our clue from "The Effectiveness of Asset Classes in Hedging Risk" published in the Journal of Portfolio Management, spring edition 2012. In this article two points of interest appear:

Diversification using gold, other precious materials, livestock, and agriculture is surprisingly effective, with correlations around -0.1 to 0.1 for most asset classes with both bonds and stocks during extreme movements and regular markets.

Source: The Effectiveness of Asset Classes in Hedging Risk

The second takeaway is that one shouldn't hedge using industrial metals during both inflationary and non-inflationary periods given the higher correlation with equity returns.

When it comes to buying inflation beneficiaries there exist many types, we will look at three types.

One type has access to inflation-beneficent commodities through operations that generate positive carry. Think of a miner. The issue with miners and other industrial beneficiaries of inflation lies in their capital intensity. Warren Buffett covered this in his essay from 1977, titled "How inflation swindles the equity investor". The point is that asset-heavy businesses requires reinvestment at peak prices in relation to the dollars already received, leading to yet higher capital intensity.

The key is to avoid asset-heavy operators. The easiest way is to acquire royalty companies that sponsor miners. The best explanation of investing in these is the one given by Murray Stahl in his Q4-2017 commentary on the pages 7..9. I highly recommend reading the letter if you're interested in a royalty company pitch through the paradigm of an inflation hedge.

Another method of investing in inflation beneficiary are through contractually secured cash flows. I personally know of two situation-types that are often structured in this manner.

The first are merger arbitrage situations with a rolling interest rate compensation. These are common in international markets with high-inflation currencies, but can be seen in U.S. cases. Read my article on the Fibria merger as an example of the Brazilian CDI adjustment in play. The cash payments at close of the merger are adjusted on a rolling basis for daily inflation expectations through interbank rates.

The second are from municipal or government contracts for recurring work across a long-span of time. The waste industry has a contract structure that benefits greatly from inflation (read my article on waste industry contract structure here).

Personally, my preference is for government contracts, given that inflationary pressures can impact the ability to merge/acquire the target company, but I believe both are better than commodities given the inherent positive carry.

Conclusion And Summary

We've covered why most asset-class "diversification" results in less volatile upside returns, but highly correlated downturn risk.

We've looked at the empirical data on correlations across asset classes and hedge fund strategies and found most lacking. The best alternative was the traditional bond-stock portfolio, but the portfolio was extremely susceptible to inflation-induced downturns.

We then looked at investor abilities to invest in commodities or inflation beneficiaries. While not recommended as an inflation hedge, we looked at which commodities to avoid. We also looked at positive carry equity trades that hedge inflation pressures.

Hopefully, this 101 introductory course on proper diversification from a qualitative level was useful.

1. The Effectiveness of Asset Classes in Hedging Risk

2. When Diversification Fails

3. The Shrinking Merger Arbitrage Spread

4. IQ Market Neutral Hedge Fund Proxy Index.

Disclosure: I/we have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours.

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.