nrpn: There have been a lot of changes at AP3 over the past few years – the alpha-beta separation, the introduction of a new strategies portfolio and an increased focus on dynamic asset allocation. How does it all tie in together?

Primarily, there were three things that were important to us; diversification, alpha-beta separation and the focus on alternative assets. That’s three features that quite noticeably have characterised our portfolio. About two years ago, we made the explicit decision to include investments in so-called new strategies. We felt that alternative assets provided diversification and good returns, but came to the logical conclusion that there are also other alternatives aside from the more traditional ones. We also realised that we needed to act more opportunistically in the market. During the summer of 2008, you already began noticing certain dislocations in the market. We needed to take advantage of that so we initiated the work with the new strategies portfolio. We had to create a structure so that we didn’t miss out on anything just because of a too rigid and inflexible structure.

nrpn: Are the portfolio of new strategies and the increased focus on dynamic asset allocation connected as means to better take advantage of current market situations?

Absolutely. Two years ago, we hadn’t reached the allocation model that we have today, but we started to realise that we were operating in a structure that was a bit too rigid. It was of course the events of 2008 that forced us to rethink how to do asset allocation as a best practice. We also hired a new head of asset allocation, Mårten Lindeborg, who provided some fresh ideas and directed the process towards where we are today. But the starting point was very much lessons learnt from the crisis. We realised that the time horizon for investments has to be shortened to take advantage of time varying risk premiums and that we had to make more mid-term allocation changes than before. We gave a lot of thought about ways to look at the portfolio and concluded that the best way was to talk about risk classes, rather than asset classes. That’s probably one of the major changes, which will become more visible next year.

nrpn: That sounds a bit like the model used by Danish ATP.

It’s an ATP type of model that is becoming more and more common. It’s nothing that ATP has invented but an accepted way of thinking about risk, certainly in the hedge fund world. We are of course not a hedge fund having long-only constraints, but it works in a similar way. A portfolio manager at a good hedge fund allocates risk between trades, has a risk budget and values assets at risk.

nrpn: What does it mean, more specifically, to start dividing the assets into risk classes?

The assets in the portfolio are put together differently and perhaps it makes you look a bit more at the underlying economical driving forces for different asset classes. We have seven risk classes; equities, fixed income, credit, inflation, foreign exchange, absolute return strategies and other exposure. Alpha becomes a separate risk class under absolute return strategies and all of a sudden, the alpha-beta separation has fallen into place. However, there is no mention of alternative asset classes as each risk class consists of different types of strategies. Equity risk includes both listed and private equity as well as the life science portfolio and a few other mandates. Real estate investments fall under inflation risk, together with forestry and agricultural land.

nrpn: Is the division into risk classes a way of keeping a closer eye on the risks of different investments or does it work more as a platform for changing the allocation?

It’s a bit of both. A very important consequence of this type of model is that it’s easier to change the risk more dynamically and to find connections between different types of investment and work with structural relations between different boxes. It’s not as much silo thinking as with a more hierarchic portfolio structure, where good investments may come along but they don’t fit anywhere in the portfolio. Here, that problem doesn’t exist and if nothing else, we always have the box for other exposure.

In a way, it connects everything quite nicely. The need to diversify and to find new sources of return, that was the background for alternatives, the need to separate alpha from beta to increase focus in the portfolio, and then finally the need to make the portfolio a bit more flexible, which was the continuing development that led to the dynamic asset allocation.

Printer friendly version

Email this article to a friend