Active managers fail to reduce the downside

Stockmarket-Stock-Market-FTSE-Performance-700x450.jpg

One of the advantages mooted by advocates of active management is that active fund managers can avoid the worst falls in the stock market, or at least limit their effect. This would be expected to result in the active funds displaying a higher skewness than the market index, as there would be fewer extreme negative returns pushing the skewness down.

A negatively-skewed distribution displays more extreme events below the mean, while a positively-skewed distribution has more events that are much higher than the mean.

Being more positively skewed could bring two theoretical benefits. First, if you could avoid the extreme negative values and just track the market on the way up you would outperform. Second, and it is this claim we are focusing on, there may be advantages to holding investments with a more positive skew than the market even if their returns are equal or lesser to those of the market.

ActiveUKequitygraph
corporatebondsgraph

Investors are loss averse on average, meaning that they value avoiding losses more than making gains of the same magnitude. This means any investment that suffers less extreme losses than the market in bad periods may have a premium attached to it that could justify lower returns. In other words, an investor might accept lower returns from their fund if the fund lost less in bad periods as this is more important to them than making more than the market, or tracking the market, in up periods.

However, our research suggests that active fund managers on average do not improve the skewness of returns.

sknewnessgraph

UK equity

We looked at actively managed funds in the IA UK All Companies sector with a 10 year track record that were benchmarked to the FTSE All Share. To ensure we captured truly active funds rather than closet trackers we excluded those with an annual tracking error below 4 per cent, resulting in 62 funds.

We then calculated the skewness of an equally–weighted portfolio of those funds and of the FTSE All Share over rolling one year periods, calculated at each month end starting at the end of October 2005.

In 57 per cent of the periods the skew of the active funds was lower than that of the FTSE All Share. This means in periods of negative skew the falls to the active funds were more pronounced and in periods of positive skew moves to the upside were less pronounced.

We can break this down to see how the active funds performed when the market’s returns displayed negative skew and when the market’s returns displayed positive skew.

We find that in 52 per cent of the periods that the market had a negative skew the skewness was more negative on the active funds than the market. This means that the funds did not provide the suggested benefit of reducing negative skew when markets were suffering sharper falls than gains.

negativeskewdata
Positiveskewdata

In positive skewness periods we find the results were very similar: skewness was more negative in 52 per cent of cases and the skewness was more positive or less negative in just 48 per cent of cases.

These results do not correspond with the claims that active funds improve on the skewness of returns by making it less negative.

In fact, it seems that the active funds displayed a higher skew principally in two periods: the 2009 to 2011 market rally and 2014, a strong year for small and mid caps. This may be explained by fund managers overweighting small and mid-sized companies, which saw sharper rises in market value.

UK corporate bonds

We found similar results for the UK corporate bond sector, when looking at the skewness of the returns of the IA Sterling Corporate Bond sector and the Iboxx UK Sterling Corporate All Maturities market benchmark.

We found that in 55 per cent of periods the skewness of the active funds was more negative than that of the market.

In this sector there was a sharp difference between the results in periods of negative and positive skewness for the market, but not in favour of the hypothesis being tested: when markets were negatively skewed the active funds were more negatively skewed 61 per cent of the time, while when markets were positively skewed the active funds were more positively skewed 53 per cent of the time.

Again, these results do not support the contention that active funds can reduce the negative skewness of returns, and in the case of corporate bonds the funds do particularly poorly when this quality would be most appreciated – when markets are more negatively skewed.

Survivorship bias

It is worthwhile comparing our bespoke portfolio of UK equity funds to the IA UK All Companies sector unfiltered for passivity or length of survivorship. The sector average includes the track record of funds that have closed during the period in question.

The IA sector shows an even more pronounced tendency for lower skewness: in 58 per cent of periods the sector has a more negative skew than the FTSE All Share.

This suggests that our focus on funds that are still available is actually flattering the performance of the average active manager – those funds that have been closed during the period have had an even stronger tendency to show lower skewness.

Skewpercentages

Asset-weighted average

However, as with all assessments of actively-managed funds it is important to recognise that no-one can buy the average fund, and the results the average investor gets could be very different from the results of the average fund. Some funds attract many more investors than others and subsequently grow to a large size.

There are some indications that the average investor may indeed get a fund with a less negative skew than the market.

A portfolio of the five largest funds in our UK equity list of 62, rebalanced annually, does have superior characteristics: in 57 per cent of periods it has a higher skewness than the market.

The funds were almost exactly as successful in periods of negative and positive skew for the market, delivering less negative skew in 57 per cent of negative periods compared to 56 per cent of positive periods.

However, there is no evidence that a fund with a better track record of limiting negative skew is more likely to outperform. In fact, our data shows there is no significant relationship between reducing the skew of the index and returns. The R-squared measure, showing deviation from the index, is 1.25 per cent.

We are strong believers in the benefits of active management, but we cannot find evidence that it reduces the number of sharp falls investors suffer, as shown by skewness. This means picking any old manager because you believe that active managers do better for skewness is just as wrong as picking any old manager because you believe the average active manager will outperform. There is no substitute for careful research and selection of individual managers.

Thomas McMahon is fund analyst at FE Invest.