Passive investing by selecting and tracking a market index (instead of selecting individual stocks or credits) has become enormously popular in the last decade. Both retail investors (through so-called ETFs) and institutional investors increasingly apply passive investing to large parts of their investment portfolios, including emerging markets. The rationale for passive investing is that markets are efficient and all relevant information is reflected in asset prices. Why pay an active manager (one that deviates from a market index or “benchmark”) high fees if there is, in the longer-term, limited scope for him or her to outperform a market index (because markets are efficient)? A case in point is the performance of hedge funds, the ultimate stock pickers of this world. Over 5 years until 31 August 2017, an investment in hedge funds (using the HFR Equity Hedge Fund index) would have made you 18% (3.4% per annum), whereas tracking the S&P500 over that period generated a return of 76% (11.9% per annum or even 14.3% if dividends were reinvested in the S&P500)

Getting what you are looking for…?

The problem of passive investing is that it only can exist when there are enough active managers operating in the market to ensure that asset prices reflect fair value (otherwise the market efficiency thesis no longer holds). So, there is a cap on how big passive investing can grow before markets run the risk of getting mispriced. Another issue is that companies with a large market capitalization attract more investments than smaller companies, irrespective of business performance, opening the possibility for these assets getting overvalued (until, eventually, reality sets in). As most trackers use relatively liquid indices or benchmarks, there is a tendency for assets that are captured in the benchmark universe to be more expensive than those that are “off-benchmark” (so a long-term investor would be better off buying off-benchmark assets). Finally, in order to achieve lowest cost (the main attraction of the investment style), passive managers must reach for scale. This might result in concentrated ownership and lax governance. Trackers like BlackRock, State Street and Vanguard already dominate U.S. equity markets, together being the largest shareholders in about 40% of U.S. listed companies. For these reasons, we are not a big fan of benchmark-hugging, although thanks to distorting central bank policies (going by the name of QE) over the last couple of years, blindly following benchmarks has not been detrimental to performance as correlations between assets have been very high due to an indiscriminate search for yield. And, to be fair, sometimes it is the cheapest and quickest way to get exposure to a particular market. For now, apparently, there are still enough active managers around as we do not see many cases of irrational pricing in the most common equity indices (other than that stocks generally seem expensive to us). But that may change if more investors adopt passive investing styles.

In any case, passive investing or benchmarking is not necessarily advisable for every asset class. For example, strictly following credit benchmarks does not make much sense as large debt issuers have a large weight in widely used (market capitalization) indices. Having a lot of debt is not necessarily a good thing (which is why many index providers have indices that cap exposure to a corporate credit to 2% or 3%, mitigating this problem somewhat). For example, China has a lower weight (4.13%) in the much-followed J.P. Morgan EMBI external sovereign debt index than Indonesia (4.37%), whereas Poland and Ukraine have equal weights (2.68%). Credit quality or other potentially relevant metrics, like size and diversity of the economy, don’t play any role in constructing the index. Countries that do not have monetary credibility and, therefore, can not issue debt in their own currency, are overrepresented in the index. Investors should ensure they understand how benchmarks are constructed.

The most curious benchmark we have come across is J.P. Morgan’s GBI EM Global Diversified index. This benchmark is used by nearly all institutional investors that invest in Emerging Markets sovereign debt in local currency (rouble, real, peso, etc.). We assume that investors want exposure to this asset class to capture relevant risk premiums.

The first risk premium is carry. Forward foreign exchange rates are not a good predictor of future spot rates. Actual currency depreciation is often less than forward rates price in and this difference gets larger for countries with high nominal interest rates (typically riskier countries). Thus, the idea is to invest in the currency of a risky country with a high interest rate, for example Brazil. The currency will depreciate but net-net this is more than compensated for by the interest earned on the cash balance (in reais). A similar carry exists for bonds, where a so-called “term premium” can be earned by holding long-duration bonds as opposed to short-duration bonds. The forward swap curve tends to price in a bigger rise in interest rates than those that actually materialize and this is more prevalent when the curve is steep. The second risk premium is value. You want to be long “cheap” currencies: you want to buy goods in a country with a cheap currency at a lower price (taking into account exchange rates) than in your home country. You would buy stuff and ship it home until the moment that the exchange rate adjusts so that prices in both countries are equal. This is referred to as the law of one price or purchasing power parity (PPP). The Economist, a newspaper, uses this concept and publishes the Big Mac index, which shows you where you can get cheap beef (the best beef is to be had in Argentina, of course). There are many reasons why PPP may not hold, certainly in the short-term. For example, goods may not be tradable, CPIs may have different underlying baskets, transaction costs may apply, consumption preferences may be country-dependent, etc.

In any case, you would expect that J.P. Morgan devises a benchmark that captures these risk premiums. However, this is not the case: benchmark weights are based on the amount of debt a country has (capped at 10%) and not on carry potential or value considerations. So, Poland, whose local currency sovereign bonds yield only 2.8% on average, has a 9.1% weight in the benchmark. In April of this year, Czech Republic re-entered the benchmark (it dropped out in 2008) despite yielding a measly 0.6% (and having a negative real interest rate!). Countries that do have carry potential, like Egypt or India, are not included, sometimes because of dubious reasons (e.g.  based on J.P. Morgan’s assessment of “liquidity”). Investing in this asset class using J.P. Morgan’s benchmark, especially when a passive or low tracking-error approach is adopted, doesn’t make any sense at all. Amending the benchmark, for example by including only countries that have a credit rating of BBB or lower (as a proxy for where carry can be found), or, even better, ditching the benchmark altogether and directly focusing on carry and value metrics would be a better approach, in our view.

Will passive or benchmark investing lead to disastrous results? Probably not, but for some asset classes you can do better by focusing on value drivers instead of following the herd.