Quarterly Publication - October 2016
The Flaws in the Case for Passive Investing - October 2016
In recent years, passive investment products have gained significant market share across the world. In my view, John Bogle, long considered the godfather of passive investing, did the savings industry a great service, because there are many incontrovertibly good things that passive investing brings to the market:
- Passive products increase choice for the consumer − this is always a good thing.
- The case for passive products is premised on low fees, which puts pressure on active managers who charge inappropriately high fees (fees that are not justified by the value they have added in their funds over time).
- It threatens active managers who have not delivered outperformance or who do not produce truly active portfolios (that is, they construct portfolios that hug benchmarks).
- Passive strategies genuinely make sense for some investors. Examples include:
- Investors who have not done the due diligence themselves, or have not taken the advice needed, to select skilled active managers.
- Those who do not have the long time horizon needed to prosper in financial markets. (Unfortunately, these investors tend to churn out of the active manager who has recently underperformed in favour of the active manager who has recently outperformed. In the process, they end up chasing yesterday’s winner, buying high and selling low, and ultimately destroying lots of value.)
- Investors who have not done the due diligence themselves, or have not taken the advice needed, to select skilled active managers.
However, notwithstanding these positives, I think that many investors in passive products are seduced by the sales pitch without fully understanding some of the deep flaws intrinsic to the passive proposition.
This article outlines a number of these flaws. (Please note that these points do not need to be read in any particular order, but in our opinion are all worth considering.)
1. INACTIVE (PASSIVE) INVESTING ACTUALLY DOES NOT EXIST
The bad news is that all investment actions require an active decision. No matter how artfully the passive sales pitch is presented, all passive investments fundamentally require an active decision. This is something of a fly in the ointment, as it is at odds with the seminal idea of passive investing − that clients are unable to identify which managers will make the correct active decisions and should therefore select an alternative that requires no active decisions (and thereby get the return of the market).
There are countless examples that demonstrate this point. Equity funds are a good place to start. A market cap weighted benchmark is the only true passive benchmark because it is the only index that all investors can buy. Yet the proliferation of passive equity benchmarks in all major markets is bewildering. In the US there are more equity benchmarks than there are large-cap stocks. This crushes the very foundation on which the case for passive investing rests, because investors do not simply get the return of the market when they invest in passive equity products. Instead, they get the return of the equity benchmark they have selected after fees and other costs incurred. And the active decision taken in choosing a benchmark can result in a materially different outcome for investors over long periods.
The SA equity market today provides an instructive case study. The most widely used passive products in the retail market are domestic equity funds. Once a client decides to allocate capital to a passive SA equity product, he/she then needs to choose a specific fund (benchmark). The bad news is that there are many options, each of which yield a very different outcome over long periods of time. In the retail market, the FTSE/JSE Top 40 Index funds initially attracted the lion’s share of the SA equity money. However, over the last few years, SWIX 40 Index funds have outperformed the FTSE/JSE Top 40 Index funds. A significant part of this return differential has come from a lower weighting to commodity stocks in the SWIX 40 Index. In 2015, this resulted in a big net inflow (R1.4 billion) to SWIX 40 Index funds and a large net outflow (R1.4 billion) from the ALSI 40 Index funds. Clients in these products believed they were following a passive strategy and getting the return of the market. Yet, in having to make the seemingly simple choice between the SWIX 40 Index and the FTSE/JSE Top 40 Index, they were unwittingly putting themselves into the position of having to make the most difficult active investment decision in the SA market: how much to allocate to commodity stocks?
The numbers tell the story. Index funds are forced to track the market. Consequently, they owned lots of commodity stocks at the top of the cycle, when prices were high (by June 2008, the SWIX 40 Index funds had 51% invested in commodities while the FTSE/JSE Top 40 Index funds had 61% invested), and they owned very little at the bottom of the market when prices were low (by December 2015, the SWIX 40 Index funds had 8% invested in commodities while the FTSE/JSE Top 40 Index funds had 12% invested).
To give you a sense of the commodity conundrum faced by all active managers in SA today:
- Commodity markets are oversupplied and the outlook is bleak.
- Supply is still increasing due to projects that were committed to at the top of the market.
- Demand is anaemic and depends heavily on China (which is at risk of a hard landing).
- As a result, commodity stocks trade at depressed levels. At the beginning of the year, commodity markets became so stressed that we estimate that many of these stocks were trading at a quarter of their underlying value. This explains why so many of them have doubled or tripled since their January lows.
- Is the 2016 rally a dead-cat bounce and are we in fact only halfway through a decade-long bear market? Or have we seen the bottom and are commodity stocks still cheap enough to buy?
There is an inherent irony in passive investing. Clients buy into the argument that they do not know which active manager will get the big calls right. In a flawed leap of logic, they are then seduced into thinking that active decisions are not required. In so doing, they unwittingly put themselves into the position of having to make some of the big active decisions themselves (for example, how much to allocate to commodity stocks, as noted above).
Given the fiduciary responsibilities that many advisers and boards of trustees have to the end investor, I question whether enough thought is given to the reality that active decisions cannot be removed from the investment process. This is the Achilles heel of passive strategies. Someone, somewhere is making an active decision. First, this needs to be acknowledged. Then the decision needs to be made by a skilled and experienced professional who will be held accountable for the call.
2. THE PASSIVE ASSET ALLOCATION PROCESS IS FLAWED
Asset allocation is generally accepted to be the most important investment decision that any allocator of capital makes. The gains or losses from selecting the right or wrong equity manager will typically be dwarfed by the gains or losses stemming from the right or wrong asset allocation decision (for example, allocating too much to bonds and not enough to stocks). Asset allocation is the big call and you need to get it right.
Unfortunately, once again, there is no such thing as a passive asset allocation decision. The conceivers of passive products understand this, which is why passive multi-asset class products are typically ‘hardwired’ to make rules-based asset allocation decisions using passive building blocks. While they may pitch this asset allocation process as being passive, in truth, investors are buying a fundamentally active asset allocation strategy.
As an example, many passive products use a fixed equity/ bond allocation that is rebalanced periodically. Typically, the optimal allocations are arrived at by analysing history and back-testing alternative allocations to find the ones that worked best (in the past). The rebalancing process is rules based − it typically happens either monthly, quarterly or on an annual basis (usually whatever has worked best in the past!).
Make no mistake, this is fundamentally a very active investment strategy. The investment decision is based on historical performance data and implicitly assumes that the future will look like the past. I question whether this will be the case. There are many reasons for this, but to name just a few:
- Over the last five decades the JSE has produced extraordinary, once-in-a-generation returns that are unlikely to be repeated in the future.
- The JSE itself looks nothing like it did ten years ago. Three of the six largest stocks listed on the JSE were not even listed on our market ten years ago.
- Central bankers responded to the global financial crisis with quantitative easing. Eight years later, interest rates in many countries are now negative. This is a grand experiment that poses significant risk to economies and to the savings industry worldwide.
- I believe that quantitative easing has created a bond bubble; one that has massively inflated historical bond returns and will result in massive losses for bond investors at some point in the future.
3. PASSIVE PRODUCTS ARE NOT APPROPRIATE IN A HIGHLY CONCENTRATED MARKET SUCH AS SA
One of Bogle’s strongest arguments in favour of passive investing is that investors in passive products remove stockspecific risk from their portfolios and simply get the return of the market. This is a compelling argument and it applies in many of the world’s more mature and deep markets. Investors in a passive Standard & Poor’s (S&P) 500 fund today have only 3% of their investment exposed to the single largest stock, while exposure to the ten largest stocks in their portfolio will amount to 18%.
Unfortunately, the SA equity market is highly concentrated. The largest stock in the SWIX 40 Index is Naspers, at an eye-watering 19%, while the top ten stocks in the index represent 47%.
Accordingly, one of the strongest arguments in favour of passive strategies does not apply in the SA market. Not only does it not apply, it is actually the reverse – there is an unmanaged risk latent in most passive SA equity products today. Investors in passive SA equity products do not avoid single-stock risk. Often they end up with much more single-stock risk, and they do so without a skilled and experienced investment professional being accountable for the appropriateness of that weighting.
We currently believe that Naspers is undervalued. For that reason, although it is a large weighting in our equity portfolios, it has been appropriately sized in accordance with our view of the risk-adjusted return that it offers. Fundamentally, however, it remains a risky stock. Most of its value comes from its Chinese internet holding, Tencent. The internet sits at the epicentre of creative destruction. Most of the world’s biggest internet companies today barely existed ten years ago. Will the winners of today dominate the internet ten years from now? In China, the risks are even greater because Chinese internet companies are not faced with meaningful competition from the global gorillas (Facebook, Google, etc.), all of which are not allowed to operate in China. Thus the incumbents implicitly depend on the support of their regulators to thrive. Tencent is the kind of stock that can easily become overvalued and decline precipitously at any time. It is not the kind of stock that should be close to 20% of a retirement portfolio, certainly not without an active decision supporting it and an investment professional accountable for the call.
4. PASSIVE BOND FUNDS ARE ALARMINGLY FLAWED
Bond funds are perhaps the most flawed of the passive products. The conundrum of setting an appropriate benchmark for a bond fund is even greater than that described for an equity fund. It is typically solved by adopting the well-known bond indices: the Citigroup World Government Bond Index (WGBI) for global bonds and the JSE All Bond Index for SA bonds.
The problem here is that the more indebted an entity, the more bonds it has in issue. And the more bonds it has in issue, the greater its weight in the index. This is a very perverse outcome. Investors in passive bond funds end up, unwittingly, in products with a systemic bias to more indebted (riskier) entities. All other things being equal, the more indebted an entity, the less creditworthy it is, and the higher its weighting in a passive bond fund.
The point is well illustrated by the WGBI today. Three countries stand out as having government debt levels that vary from worrying to terrifying: France, Italy and Japan. Their debt/GDP numbers are 97%, 133% and 248%, respectively. In the WGBI, Japan has a weighting of 23%, France a weighting of 8% and Italy a weighting of 7%. All three countries are at risk of a debt trap. Japan, in particular, continues to blithely rack up deficits with complete indifference to the country’s own insolvency. And yet, the bigger those deficits, the more bonds these countries will issue, and the more of their bonds passive bond funds will have to buy.
5. PASSIVE IS BECOMING DISCONCERTINGLY ACTIVE AS SMART BETA PRODUCTS GROW IN NUMBER
An interesting development in the passive industry is that, as passive has gained in acceptance and confidence, it has become more active. More and more active investment decisions are being designed into passive products (is the world not an amazing place?). The boundaries between active and passive are therefore becoming ever more blurred. All smart beta products are, in truth, semi-active products. Is this a bad thing? I think so:
- The risk in these products is that clients believe they are getting a passive product – one that will track the return of the market (albeit with a few tweaks here and there that happened to have worked out very well in the past). These tweaks are always ones that delivered excellent results in the past. The back-testing results are always compelling. However, financial markets are daunting places that humble the best. If the formula for success were as simple as repeating what worked in the past, we could all fill our investment teams with algorithms and get on with life …
- In many cases, clients do not realise that they are invested in products where far-reaching active decisions are in fact being made. This applies as much to the smart beta building block funds (bonds, equities, properties) as it does to the passive asset allocation funds. In most cases these important active decisions are not being made by a team with the skills, the experience, the extensive research process and the granular understanding of the underlying securities that ought to support any active decision-making process.
6. THE PASSIVE SALES PITCH IS PREMISED ON LOW FEES. THIS IS OFTEN FAR FROM THE TRUTH.
Although we do not have access to fee data in the institutional market, I assume that many large pension funds secure fees below 0.2% per annum (which I consider to be a fair fee for passive).
In the retail market, however, passive products are surprisingly expensive. In fact, many passive retail products seem to charge active-like fees for a passive service:
- The total investment charge (TIC) for the five largest equity tracker unit trusts in the retail market are still very high, at 0.78% per annum on average.
- The equivalent number for the largest equity exchange-traded fund (ETF) in the market is lower, but still high, at 0.46% per annum. (This was arrived at by doing a like-for-like comparison to a unit trust, which includes the brokerage costs incurred in buying and selling ETFs. In this calculation we used the cheapest brokerage deal we could find and watered down those brokerage costs over a 20-year holding period.)
- The TICs for smart beta products are significantly higher than the pure equity trackers; in many cases these are close to those that genuinely active funds charge.
7. SOME PASSIVE PRODUCTS UNDERPERFORM THEIR BENCHMARKS BY A LOT MORE THAN THEIR EXPENSE RATIOS
The passive sales pitch leaves one with the impression that a passive product will give its client the returns of the benchmark after fees. However, an analysis of the historical returns delivered by passive retail products/ETFs demonstrates that this is not always the case.
Passive products underperform their benchmarks to the extent that they do not perfectly mirror their benchmarks, as well as due to the trading costs they incur. As more money flows into passive products, I think this underperformance will become more pronounced. Why?
- Flows into passive products result in an increased supply of scrip lending in the market. Passive products earn a fee income from scrip lending, but as supply increases, that fee income will decline.
- When indices are rebalanced (as stocks fall away or are added to the index), passive products will increasingly struggle to mirror their benchmarks as more and more money competes to do exactly the same trade.
An analysis of the retail market does not reveal a uniform experience across the different product providers. Some products have not suffered any performance drag at all, whereas for others it is as high as 0.5% per annum (this needs to be added to the fund's TIC to calculate total underperformance).
In the end, a thoughtful analysis of the passive sales pitch reveals many flaws that are glossed over by its proponents. As is so often the case in life, the theory is frequently very much at odds with the reality. Although passive undoubtedly has its place in the market, we observe that it comes with as many negatives as it does positives.
FINALLY, WHAT DOES THE GROWTH IN PASSIVE ASSETS MEAN FOR ACTIVE MANAGERS?
In my opinion, true active managers have nothing to fear. Passive investing leverages off active investing, because active managers make markets more efficient than they would otherwise be. The two strategies are, for this reason, complementary. Markets function best when there is a broad universe of investors with different strategies and time horizons. The growth in passive strategies actually increases the opportunity set for the genuinely active manager. It does this by increasing liquidity in the market. It also makes markets less efficient because it fundamentally biases the investment process towards buying high and selling low. It systematically gives higher weights to overvalued stocks and lower weights to undervalued stocks.
A good practical example would be index rebalancing days (these happen once a quarter and are my favourite days in the office because of the opportunity they provide to buy cheap stocks and to sell expensive stocks in size). On these days, passive products are forced to sell stocks that have performed poorly enough to fall out of their respective benchmarks and to buy those stocks that have performed well enough to move up into their respective benchmarks.
By definition, active managers cannot deliver outperformance if markets are efficient. They endeavour to buy low and sell high. In order to do so, they need someone on the other side of the trade. Passive money is here to stay. It no doubt adds to the stress levels of rational long-term managers (by definition the inefficient pricing of assets has to cause short-term underperformance in their funds). But ultimately it creates opportunity.
Many of our clients ask us to critique the passive proposition. Although this article was penned in answer to that request, I think that as an active manager we ought to heed the wise words of Jeff Bezos, the founder of Amazon. He notes that most businesses spend too much time watching their competitors. Amazon has succeeded because of its relentless focus on its clients, not its competitors. The active manager that cuts out all the noise and delivers compelling results for clients over long periods of time (and charges a fair fee for that service) will prosper regardless:
- Over Coronation’s 23-year history, our institutional (pension fund) SA equity portfolios have outperformed their benchmarks by 3% per annum before fees. R100 million invested on the day we opened for business in 1993 would be worth R3.8 billion today, after all fees and costs. That same amount invested in passive alternatives, at a fee rate as low as 0.2% per annum, would be worth R2.8 billion.
- Similarly, R100 000 invested in our SA equity unit trust on the day it launched in 1996 would be worth R3.1 million today, after all fees and costs. The same amount invested in a passive All Share Index-tracking unit trust fund at a TIC of 0.78% per annum (approximately what the two largest retail index tracker funds charge) would be worth only R1.3 million, and worth just R1.4 million if it had been invested in an All Share ETF (with an all-in cost of 0.46% per annum).