Jaap van Dam, principal director of investment strategy at PGGM, one of the world’s largest asset owners known for its commitment to long-horizon investing, once asked what he called the million-dollar question: “Can we be reasonably certain that we will be rewarded for being a long-horizon investor? Because, if we’re not, then why bother?”
A sound answer to this question, as Jaap rightly put it, will determine whether long-horizon investing will really take off among asset owners.
We have been grappling with this issue within the Thinking Ahead Institute for some time now, and at the start of 2017 formed a working group of members who shared our passion to make more definitive progress on the matter. The first fruit to emerge from the efforts of the long-horizon investing working group is our recently-published paper: The search for a long-term premium. Our conclusion is that a sizeable net long-term premium does exist, and we have quantified its size as lying between 0.5% to 1.5% per annum (pa), depending on an investors’ size and governance arrangements.
Before going into the detail of how we arrived at our conclusion, I think a couple of higher-level observations might help. First, to the best of our knowledge this is the first time anyone has attempted to quantify the long-term premium and therefore this is potentially an important contribution. Second, our conclusion is not definitive proof but rather is the result of (a) deduction from disparate pieces of evidence and (b) our assumptions regarding the combination of the pieces. As a consequence we are exposing our analysis and our thinking to public scrutiny which carries a degree of personal risk. Should the paper survive its public testing, then I think it will become important—because at that point the default actions required from an investor subject to a fiduciary duty will change, and they will either have to work out how to harvest the long-term premium or explain why they are not doing so.
Hunting for evidence of the long-term premium is easier said than done. In an ideal world, we would run a regression of net investment returns against investors’ time horizons. Sadly the data to run this regression does not exist, as it is not possible to accurately measure the time horizon of investors.
As a result, we used an “indirect” approach, based on identifying return opportunities and ways to reduce drag on returns. We found eight building blocks of long-horizon value. Each is practical to implement, albeit with changes required to the investment process. Together, they provide evidence of a sizeable premium from long-horizon investing.
Long-horizon return opportunities
A study that examined over 2000 highly-intensive engagements with over 600 US public firms between 1999 and 2009 produced some revealing conclusions. The study showed that engagements with investee companies generate, on average, positive abnormal returns of 2.3% over the year following the initial engagement—clear evidence of the benefits of being active owners to encourage investee companies to take long-term approaches.
When investors are willing to pay for liquidity—in other words, sell assets below “fair value”—someone on the other side of the trade gets paid. One study suggested that long-horizon investors have the potential to earn additional returns of 1.0% pa at the expense of shorter-horizon investors by providing liquidity when it is most needed.
Another aspect of liquidity involves harvesting the illiquidity risk premium (IRP), which is well established as a source of return for long-horizon investors. When investors accept illiquidity, they accept greater uncertainty about the outcome because they are less able to liquidate the asset. The longer the capital is tied up, the more return investors expect by way of compensation. Academic studies point to a range of 0.5% to 2% pa for this particular premium—and even higher returns might be available to very long-horizon investors.
A fourth return opportunity for long-horizon investors comes from exploiting various mispricing effects via factors or smart betas. Decades of data suggest that this can add more than 1.5% pa relative to the cap-weighted index.
Investors have long been aware of thematic investing and belief in its ability to create value appears to be strong. However, the lack of consistency in implementation approach means we have been unable to find empirical evidence that categorically demonstrates the success of a thematic approach. This is therefore the most arguable of the eight building blocks.
Lower long-term costs / mitigating losses
A long-horizon mind-set can also usefully guide behaviours to reduce drags on investment returns. A study of over 400 US plan sponsor “round-trip” decisions (firing and replacing managers) compared post-hiring returns with the returns that would have been delivered by fired managers. It suggested that by replacing their investment managers, the plan sponsors on average gave up a cumulative 1.0% in the three years following the change—they “bought high” and “sold low” which is a drag that can be mitigated by a long-horizon mind-set.
Open-ended fund structures, despite the flexibility they provide, might not be fit-for-purpose for long-horizon investors who do not require nearly as much liquidity as other short-horizon shareholders. In such a structure, long-horizon shareholders effectively subsidise their short-horizon peers for their liquidity needs. One study found that liquidity-driven trading in response to flows (in particular redemptions) reduced returns in US open-ended mutual funds by 1.5% to 2.0% pa.
Last but not least, significant savings in transaction costs can be made by avoiding unnecessary turnover as a long-horizon investor.
It’s all about governance
Capturing the benefits of long-horizon investing will require a major shift of mind-set and significantly expanded skillsets by investors. In many cases, it entails incremental spending—e.g. expanding investment expertise in active ownership, or increasing the number of board meetings. In the paper we take two hypothetical funds to develop a reasonable estimate of the potential net long-term premium in practice.
The smaller fund (US$1bn) focuses on avoiding costs and mistakes. It reduces manager turnover, avoids chasing performance, and moves part of its passive exposure into smart beta strategies. The rationale is: if you don’t have the resources to win big, at least don’t lose. The net benefit of these efforts is potentially an increase in investment returns of about 0.5% pa.
The larger fund (US$100bn) has the governance and financial resources to consider all available options. It introduces long-horizon return-seeking strategies, while reducing its exposure to mistakes and costs. The net uplift to returns is potentially around 1.5% pa.
In the investment world where there are very few universal truths, it would be hubristic to conclude that we have proven the existence of the long-term premium. We are, however, “reasonably certain” that the costs of developing the mind-set and acquiring the skillsets to address long-horizon investing challenges are substantially outweighed by the potential return enhancements. Our work now turns to developing a range of practical solutions for investors to access the premium.