An average of 45% of active equity risk taken by institutional investors can be regarded as uncompensated, according to a study by Northern Trust Asset Management. Investors that use active management have the highest share of uncompensated risk in their portfolios.
Northern Trust Asset Management surveyed 64 institutional, mainly US investors with a total of slightly over $200 billion in assets under management. Uncompensated risk refers to risk that has not been proven to generate alpha. Uncompensated risk is often an unintended by-product of compensated risk, i.e. a risk that produces alpha.
An important part of the uncompensated risk can be traced back to unintended overweights to certain sectors, which is often a by-product of factor allocations. An example is an overweight to low volatility stocks. ‘Investors are not necessarily realising they are also buying sector bias such as real estate or consumer staples,› says Michael Hunstad, head of quantitative investment at Northern Trust Asset Management. ‘Low-volatility stocks are often interest-rate sensitive too so investors in these stocks inadvertently pick up interest-rate volatility too. You saw this clearly during the market correction in 2016 when investors started to price in rate hikes. Those shares then suddenly became more volatile because of their exposure to interest rate risk.›
Technology stocks
Another, more recent example is the almost automatic exposure to technology stocks in a growth portfolio. ‘As an investor you think you’re getting exposure to growth stocks, but in the meantime you’re also investing heavily in cloud computing and internet technology.’
Of course, this has not necessarily been a problem in recent years, and there is a good case to make for overweighting technology stocks. ‘But there is no evidence that technology stocks by themselves have alpha potential. An unconscious allocation to a particular sector as a result of a factor allocation is an unintended risk. What you’re looking for as an investor is pure exposure to the growth factor,› says Hunstad. He therefore recommends ensuring that growth exposure is pursued across all sectors.
Active management
Uncompensated risks are mainly found among investors who make extensive use of active management. According to Hunstad, 50% of the equity portfolios of institutional investors are still being actively managed. ‘There’s a correlation between the number of [active] managers you have in your total portfolio and the cancellation of factors. One manager can be overweighting a certain stock while another underweights it. The more managers you have the higher the likelihood of seeing such cancellation effects.’
In practice, almost half of the active risk of investment portfolios disappears in this way (see figure). ‘The strongest example we saw was a portfolio of emerging market equities with six different managers. These managers turned out to have such opposite portfolios that they wiped out 90% of the active share of the total emerging market portfolio.›
So the more managers you appoint, the greater the potential problem of uncompensated risk. This suggests that the diversification advantage that you have by betting on several horses disappears at some point: if you diversify too much, you quickly end up with a portfolio that resembles the benchmark, but at a higher cost. ‘Nearly 50% of active risk is being lost this way. Capturing just 50% of active risk while paying 100% of active fees effectively translates in paying two times more for each realised basis point of active risk than originally thought,’ Hunstad notes.
This does not necessarily mean, however, that you should use just one manager per mandate. It is just a matter of defining those mandates very precisely, while continuously monitoring whether the managers you have appointed stick to them. Hunstad gives the example of a value portfolio. ‘One manager could use the concept of intrinsic value, and therefore buy Amazon, for example. But another manager may look mainly at the price/earnings ratio and not buy the stock. If you appoint both of these managers, you run the risk that their positions will neutralise each other.› If you make a strict definition in advance of what you understand by value, then something like this will not happen.
Multi-factor
The proportion of uncompensated risk in the portfolios examined remaining around 50% over the years (Northern Trust has been carrying out this study for 8 years now) suggests that it is a persistent problem that is difficult to address in practice. But it can be tackled.
‘The best solution to this is a consistent multi-factor approach in which you select stocks that combine pure factors such as momentum, value, quality and low volatility,’ says Hunstad. ‘In this way, the probability of different parts of the portfolio neutralising each other is lowest.›
According to a recent study by Invesco, 81% of institutional investors worldwide now use a multi-factor approach within their equity portfolio. For the (mostly American) investors in the Northern Trust study, this is not yet the case, perhaps because these tend to be smaller investors with only a few billion dollars in assets under management. ‘Most of the portfolios we analysed do not yet use a systematic factor approach,’ says Hunstad.
However, eventually also smaller investors will realise multi-factor investing is the most effective way to address the problem of uncompensated risk, Hunstad believes. ‘We have much better tools now to construct portfolios than even a decade ago. It’s like today we have much more information about calories and vitamins to get to an effective diet. It’s the same for portfolio construction really.’