With funds and trusts, can investor choices be predicted?

Kepler probes the relationship between investment trusts’ discounts and open-ended funds’ inflows.

While there are many aspects of investment trusts that differentiate them from their open-ended peers – such as the presence of an independent board, the potential use of gearing, and their ability to smooth dividends – nothing is more fundamental than investors having to contend with the prospect of their investments trading at a discount to NAV. A wide discount can provide an attractive entry point for investors, offering the ability to enhance returns through a narrowing of the discount. We set out to discover whether the actions of investors in open-ended funds (i.e. the inflows and outflows of the products) have a correlation with the discounts of investment trusts, asking whether the demand for the two types of fund is driven by the same factors making it potentially exploitable by investors.

Correlations

Using the data provider Morningstar, we have been able to ascertain the monthly inflows and outflows of the Morningstar sectors of open-ended funds over the last five years (to 30/11/2021), in order to compare this data to the average monthly discounts of the funds’ respective sectors. Importantly, we only compare the major listed equity sectors that have more than five investment trusts in them. This is to allow meaningful comparisons, as open-ended strategies typically lack the capacity to hold unlisted investments due to their daily liquidity. We also need the investment trust peer group to have a sufficiently diverse sample, so as to reduce the skewing effects of individual trusts’ policies, such as aggressive discount control mechanisms.

INVESTMENT TRUST DISCOUNTS AGAINST OPEN-ENDED FLOWS

*Significance is determined at a 5% level. Source: Morningstar.

As can be seen in the above table, there is largely a weak but positive correlation across nearly all the major peer groups. This means that during periods of inflows into the funds in a given sector, investors can expect some degree of discount narrowing in the associated trusts. However, the most important takeaway here is that while closed- and open-ended investors show some signs of similar behaviours through buying to the same styles of strategies at the same time, we believe that the correlation between their actions is so small that it could not be utilised in any practical manner. And even where there are signs of positive correlation, almost all the sectors fail to pass the significance test, meaning that the positive correlation between the two data sets is too small to be reliable. The exception is the Asia Pacific sector, which also has the highest correlation. However, this is only 0.32.

One of the reasons why these correlations might be so low is because investment trusts don’t typically trade on high premiums, as boards will often issue shares to take advantage of this, suppressing the premium of a trust in the process. This means that in strong periods of inflows, the correlations may fall, as while inflows into open-ended funds can grow endlessly, the premiums of trusts will have a practical limit set on them.

This leads us to the next step in our analysis: testing the correlations between open-ended fund flows and discounts only during periods when trusts traded on an average discount. However, we found that only three sectors traded on an average premium during any of the 59 sampled months. Of those three, only the Morningstar Global Equity Income sector traded at an average premium for a meaningful amount of time (33 out of the 59 counted months). The Morningstar Japanese Smaller Companies and Global sectors traded on an average premium for three and six months respectively.

CORRELATIONS EXCLUDING POSITIVE PREMIUMS

*Significance is determined at a 5% level . Source: Morningstar

Given the few adjustments required when accounting for monthly premiums, their removal does little to impact our overall conclusion. However, for our limited sample size we can see that the removal of premiums does increase the overall correlation between the sectors, indicating that open-ended demand may have a slightly greater impact on shareholder behaviour than that inferred by our previous table. Nonetheless, the limited number of peer groups within this sample does limit the robustness of the data.

The Morningstar Global Equity Income sector’s anomalous average premium is in our view the result of one of the inherent advantages of investment trusts, in that trusts can offer income investors far more reliable income streams than their open-ended equivalents. For example, Murray International (LSE:MYI)has been able to offer investors at least a 4% yield for the majority of the last five years, a difficult ask given the effect of both rock-bottom interest rates and the impact that Covid-19 has had on global dividend payouts. While MYI trades on a historically wide discount today, over the last five years it has traded on an average discount of a mere 0.3%. The sector has also been a haven for those concerned about the impact of Brexit on UK companies’ dividends and about the weakness of sterling.

The above tables give us some indication of the nature of the relationship between discounts and open-ended investor demand, that being a very weak but positive correlation. The subsequent question that needs to be answered is why? Why, when at a fundamental level closed- and open-ended investment strategies seem so aligned, is the demand for one not fully reflected in the other? The answer, we believe, is because while their investment philosophies tend to be aligned, and thus will capitalise on the same market tailwinds and headwinds, it is the unique elements of investment trusts that lead to an asymmetry in investor demand.

DCMs

The most obvious reason as to why the correlation is so weak is that the discounts of investment trusts do not solely reflect the demand of investors for their shares. Investment trust boards can also intervene to manipulate the discounts, whereas the demand for open-ended fund units is solely at the behest of the market. Discount control mechanisms can allow a trust’s share price to be shielded during market drawdowns, giving investors more confidence about their ability to sell at a good price and thus perhaps reducing their inclination to jump the gun to avoid a widening discount.

There are many examples of funds with effective discount control mechanisms, such as Martin Currie Global Portfolio (LSE:MNP), which operates with a zero-discount policy, buying and selling shares to keep MNP’s share price close to NAV. Likewise, the board of its global equity peer, Mid Wynd International (LSE:MWY), also operates an aggressive discount control mechanism, whereby the board aims to keep the trust trading within a 2% discount or premium. Both MNP and MWY are excellent examples of why there is a disconnect between discounts and open-ended demand. Both MNP and MWY are managed in a very similar manner to an open-ended equivalent, yet their discounts tend to remain at a consistent level, even when their open-ended equivalents see outflows. By coincidence both MNP and MWY are high-quality, growth-orientated global equity strategies which have been able to generate enviable returns over the long term while also offering attractive characteristics beyond their performance, be that MWY’s strong downside protection or MNP’s best-in-class ESG credentials.

Chasing discounts

While discounts do reflect excess sell pressure on a trust, they also represent a potential investment opportunity, as what goes down should, sometimes, come up. When it comes to buying trusts, we believe that investors do not merely account for potential NAV growth, but also for the additional benefit a narrowing discount can have on share price returns. This means that there may be a ‘rubber band’ effect, whereby trusts on particularly large discounts see a rapid narrowing as investors flock to the trust to capitalise on an attractive entry point. This acceleration in demand may also contribute to a detachment in correlation as investors buy into deeply discounted trusts when they see a value opportunity, which may not be the case when inflows into open-ended funds are being seen. A recent example of this occurred with Brunner (LSE:BUT), which traded on a circa 18% discount during October 2020 due to aggressive selling pressure after one of its major shareholders liquidated its position. Yet BUT now trades on a circa 8% discount due to the natural demand for its shares thanks to its top-decile performance over the last 12 months, and probably also due to investors capitalising on its anomalous discount. BUT’s recent NAV outperformance is thanks to its lead manager Matthew Tillett, who follows a balanced but concentrated approach to global equity investing which has allowed the trust to avoid many of the stylistic headwinds that 2021 presented.

Idiosyncratic strategies

While the NAV returns of open- and closed-ended strategies are at the mercy of the same market forces, investment trusts provide their managers with far greater flexibility as to how they approach investing. Freed from the confines of open-ended liquidity requirements and the occasionally onerous restrictions of UCITS regulations, investment trust managers can invest more freely than their open-ended peers. This means there are plenty of trusts with idiosyncratic strategies which often have tailwinds behind their share price demand, given they offer attractive sources of diversification or alpha. One such example is AVI Japan Opportunity (LSE:AJOT), a Japanese small-cap strategy which takes an unconventional, highly active approach to investing. The team aim to unlock shareholder value through a series of active engagement campaigns implemented across the majority of their holdings. The nature of their approach requires them to take large, long-term positions in their companies in order to have sufficient voting share in each company to enact changes. Such an approach is best suited to closed-ended. This strategy has garnered AJOT a loyal investor following, with the trust trading at an average premium of c. 2% since its launch in October 2018.

As we alluded to in our analysis, the demand for the resilient income streams that investment trusts can provide has helped narrow some discounts, be that through the use of gearing to enhance yields or the use of revenue reserves to smooth payments, or through tapping into capital accounts to support dividend payouts. It is the ability of investment trusts to provide greater, or less volatile, yields than their open-ended peers that has helped support the demand for their shares. This fact has been made apparent by the impact COVID-19 had on global dividends, where in the case of the UK the market saw an aggregate dividend cut of 44% over 2020. Even more recently, strategies like European Assets (LSE:EAT) saw renewed demand for shares over the last month, despite the difficult period that global equity markets have seen. This is likely due in no small part to EAT’s policy of paying out 6% of its NAV as a dividend each year, making it one of the only ways income investors can access high-quality European small caps without sacrificing yield, as EAT currently has an underlying portfolio yield of only 1.2%. For similar reasons JPMorgan Global Growth & Income (LSE:JGGI)has been able to trade at a largely consistent premium over the last three years. By paying out 4% of its NAV as a dividend each year, the managers of JGGI are freed from the restrictions of a formal income objective, and can invest in a portfolio of high-quality growth opportunities. While such companies are not typically associated with income strategies, JGGI’s yield is able to remain competitive with that of global equity income peers, while also generating one of the highest long-term NAV returns.

Conclusion

Investors are hopefully rational beasts, and their investment behaviour should in theory be the same when dealing with open- or closed-ended structures, as they are both susceptible to the same market headwinds and tailwinds; or at least, one would assume so at first glance. We believe it is not too great an assumption to expect investment trusts’ discounts and open-ended funds’ inflows to have some relationship, as they are both proxies for investor demand. Yet as we have shown here, the relationship between these two factors, though slightly positive, is weak. It is certainly too weak to infer any sort of reliable relationship between the two factors.

When pondering why the hypothesis doesn’t hold, we find that the nature of investment trusts leads to there being powerful external factors that influence the demand for investment trusts’ shares, such as their discount: factors not shared with their open-ended equivalents. Whether it be the strict discount control mechanisms of MNP and MWY, the unique offerings of AJOT, or the obvious income advantages that investment trusts offer, it is clear that the demand for investment trusts is as nuanced, and at times as resilient, as the advantages they provide.