Why is systemic risk awareness so low?
The FT’s Roger Blitz and Peter Fitzgerald of Aviva Investors discuss global investors’ current low levels of systemic risk awareness. They also explain the optimal way to measure and hedge against the possibility of market collapse.
Produced by Alessia Giustiniano. Filmed by Petros Gioumpasis.
Transcript
You can enable subtitles (captions) in the video player
Donald Trump may be unsettling politicians, but there appears to be a lack of systemic risk in the market. So do investors understand systemic risk or are they simply ignoring it? With me to discuss this is Peter Fitzgerald, Head of Multi-Assets at Aviva Investors. Peter, what is your research showing us?
Yeah, so market risk is something that changes over time. And it's not necessarily something that is static. Now if one looks at a simple what I call the correlation matrix, which we would do on a heat map basis-- so that when something is color-coded red, it means that these assets are highly correlated together. And what you can see from a heat map, which takes a snapshot at the end of the year, is that there are some risk concentrations, but not actually that many. And there are opportunities to build diversified portfolios, because what you can see is that the blue and the lighter coloured yellow actually say that these assets are not necessarily moving together.
This is a snapshot from the end of the year. And what you can see is there are some flashes of red. Now if you were to look at this particular heat map around the period of the Brexit Referendum, you would have seen a lot more red, and therefore a lot more systemic risk in markets, and a much more difficult environment in which to build a diversified portfolio.
So at the moment, that would probably be quite less red and less significant for a systemic risk.
Correct. Correct. And what we say is important is to use a number of different factors. So correlation and you heat maps are important. However, you need to supplement those with some additional information.
What's the story about market turbulence since the crash?
So market turbulence, what it simply does is it measures how much of the current market is really being changed by one or two factors as they change over time. And what you can see from this particular chart, which gives you that measure of market turbulence, is today, actually, market turbulence is very low. Market participants are actually not too concerned with what's happening. And the big spikes that you can see relate predominantly to events around China and obviously the referendum in the UK.
So when we look for systemic risk, when are we going to know it's around? What's a way of measuring it?
So the challenge always that you have with systemic risk and with periods of market volatility is they tend to erupt almost without warning. And they generally follow a period where volatility and risks have been very, very low. So what we think is important is to measure the current market risks over a number of different timelines and windows to give you an insight into what may be changing beneath the surface in markets. But it is really difficult to pinpoint exactly when these regimes may or may not change.
Let's have a look at your final chart, Peter.
So the final chart looks at what we call the absorption ratio. And what this says is how much of the market conditions and market returns are being dominated by just one factor. And what you can clearly see here is a spike in that particular measure again around the middle of last year with the referendum in the UK. And you can see that today despite all the noise and concerns about Donald Trump, this ratio is actually relatively low, which suggests that the market movements and market behaviours are not necessarily being driven solely by Mr Trump.
OK so to conclude, how should investors play in systemic risk?
So the key thing to have is a portfolio that is well diversified-- one that you have done a range of stress tests on to see how would this portfolio have behaved in all of these known historical events. But you cannot simply use either one risk measure or simply rely automatically on historical relationships holding.
Peter Fitzgerald, thank you very much, indeed.