Less is more: the trusts that outperformed during the sell-off

Data shows the trend of investment trusts moving to more concentrated portfolios is a good thing for investors. 

William Sobczak is an analyst at Kepler Trust Intelligence.

Recipe for disaster?

There is some evidence that, in recent years, the rise of passive funds has caused investment managers to become increasingly concentrated, endeavouring to be considered active managers as opposed to ‘closet trackers’.

The most active trusts tend to generate the highest levels of alpha, as we have often noted, and on average managers have been decreasing their number of holdings in the hope of achieving this.

Academic support for taking this approach has been provided by Martin Cremers and Antti Petajisto, who have shown evidence that portfolios with higher 'active share' tend to beat their benchmark net of fees whereas the least active tend to match their benchmark index performance before fees, and under-perform after fees.

In this article we assess how the trusts that have been increasing their concentration have performed in the most recent market corrections.

We then assess the performance of the most concentrated trusts, including investors in both small and large-caps, through the COVID-19 crash, and compare how they have done relative to the broader investment trust universe and their peers. Our research shows that more concentrated funds are not necessarily more susceptible to underperformance in a falling market.

How does concentration impact your performance during a downturn?

The past three years have seen three corrections or near corrections in the MSCI ACWI index. The first was between 11/01/18 and 26/03/18, which saw a 9% fall across the MSCI ACWI.

Then in the periods 02/10/18 – 21/12/18 and 20/02/20 – 18/03/20 the same market saw greater than 10% falls. Over the past five years 15 trusts have, according to Morningstar data, decreased their number of holdings by at least 20 positions.

These can be seen in the table below, with the excess return over the benchmark they have generated in these periods of market drawdown.

As can be seen there is no clear pattern, in terms of performance, across the differing market corrections. Nine of the fifteen (60%) did worse on a relative basis in the most recent correction than they did in the January 2018 slump.

However, only seven of the fifteen (47%) did worse than in the Q4 2018 slump. On the other hand, only 6 managed to outperform their benchmark in the most recent correction compared to 10 in the first.

Scottish Investment Trust (LSE:SCIN) is the only trust of the sample to progressively improve its excess return through each correction.

2014 saw a shift to a ‘best ideas’ portfolio, with Alisdair Mckinnon taking over as manager. Subsequently to Alisdair taking the reins the board made a conscious decision to almost halve the size of its portfolio to the current level of 54 holdings.

Martin Currie Global Portfolio (LSE:MNP) is another trust to stand out in the most recent market correction, performing particularly strongly in the recent correction relative to the past two market falls. Again the trust saw a large shift in concentration after a manager change, with Zehrid Osmani taking over the portfolio in 2018. Over the past five years the portfolio has moved to having just 42 holdings, although it has gone as low as 30 in recent months.

However, looking across these trusts, only Manchester & London (LSE:MNL) would actually be generally considered to be highly concentrated in absolute terms with just 15 holdings. The large majority – although increasing concentration dramatically –are still reasonably diversified, with five of the 15 trusts having more than 100 holdings in their portfolios.

Therefore to look at the effect of concentration itself in the recent crash we think it is better to look at the most concentrated trusts across the wider universe.

In fact, if one looks at the most concentrated trusts in the investment trust universe, the performance during the most recent correction has, on average, been very strong.

The trusts with fewer than 30 holdings have generated an average excess return of 8.4% in the 2020 correction, relative to their respective benchmarks (see table below). In comparison the broader investment trust universe generated a negative excess return of -1.01% to their relevant benchmarks. On the other hand, the least concentrated trusts, those with over 100 holdings, had an average excess return of -2.01%.

It is worth noting, however, that many of the most concentrated trusts (in the table above) hold smaller companies.

These trusts generally show stronger outperformance of their benchmarks than large-cap portfolios in any case, with the markets offering greater alpha potential. Moreover some of the trusts in the table hold micro-cap companies.

These are relatively illiquid and are unlikely to be traded during a sharp decline, meaning NAVs may not really reflect what market prices would be and so performance in corrections might be flattered.

Some also hold unlisted companies which are only valued monthly or quarterly. We think these factors are more likely to be important than concentration in explaining the outperformance in the table above.

Consequently we have restricted our analysis further to look only at large cap trusts to see if there has been any link between concentration and performance in downturns.

On average the 24 large-cap trusts with fewer than 50 holdings have an excess return of -0.08% in the 2020 correction. However the large-cap trusts with over 50 holdings have a negative excess return of -8.44%, i.e have generated significant underperformance.

Thus, there would not appear in our view to be a causal link between a reduced number of holdings and greater susceptibility to drawdowns.

Furthermore, if we look at the large cap trusts in comparison to their specific sectors, the most concentrated trusts have performed considerably better.

Below we have created a table with the average returns for the trusts with less than 50 holdings vs those with more than 50 holdings in each large cap sector. All but the Global Equity Income sector, which has just one trust with fewer than 50 holdings, has seen the more concentrated trusts outperform.

Similar to the Global Equity Income sector, we would opine that the results for North America need to be taken with a pinch of salt as there is such a small sample of trusts to look at. On the other hand it is worth highlighting the UK Equity Income sector. In this sector the nine trusts with fewer than 50 holdings have generated significant excess returns than the 16 with more than 50 holdings in a period with awful returns for the UK stock market.

Within this sector, British & American (LSE:BAF), Finsbury Growth & Income (LSE:FGT) and Troy Income & Growth (LSE:TIGT) have been particularly strong performers over the COVID-19 correction. TIGIT has a little under 50 holdings, whereas FGT and BA are extremely concentrated with 14 and 17 holdings respectively.

Conclusion

We believe this data shows that concentrated portfolios can continue to outperform during bear markets and concentration does not always mean increased losses.

Clearly there is not enough evidence yet to draw statistically significant conclusions, but we believe the indications are that investors should not be concerned about the downside loss potential of an investment trust purely if it is more concentrated than its peers.

In our view the trend to concentrated portfolios is a good thing for investors and likely to reinforce the benefits of investing actively through a closed-ended fund. This is because some of the drawbacks of concentration are more problematic for open-ended funds. The latter have to sell when they are suffering outflows, which incentivises them to hold more stocks, either to counter-balance illiquid holdings or to guard against the risk of any one becoming illiquid in a crisis.