Unipension: Risk, adjusted
Martin Steward spoke with Niels Erik Petersen (pictured) and Søren Bang Andersen of Unipension, the consolidated administration service for Denmark's pension funds for architects, MAs, MScs and PhDs, agricultural academics and veterinary surgeons
The formation of Unipension in January 2008 brought a motley bunch of pension plan members together - 16,000 Danish architects (AP), agriculture academics and veterinary surgeons (PJD), and 68,000 diverse professionals with higher degrees (MP): the need for a wholesale re-alignment of risks, assets and liabilities was obvious.
Two big steps were taken right away: members re-elected all three schemes' boards, and in the process MP members were asked whether they would like to retain a guaranteed pension rate of up to 4.25%, or, like the members of AP, PJD and so many other Danish schemes, accept a new rate of 1.5% in return for the possibility of better returns from more investment risk. Ninety percent took the plunge.
"If you take a person in their early 30s who has a high guarantee, in a low interest rate environment that means that your ability to hold risky assets is very limited," notes CIO Niels Erik Petersen. "It doesn't make sense that a 30-year-old cannot hold risky assets. There is a high probability that his pension will be lower as a result. The members of MP had experienced years of unattractive returns and they wanted something more attractive. The high guarantees also meant that pension levels were not attractive to new members."
Moving from up to 4.25% to 1.5% reduced the interest rate sensitivity of the liabilities and improved the schemes' positions in relation to the Danish Financial Services Authority's traffic light scenario-based solvency requirements which, in turn, enabled money to be saved in the swaptions market by reducing the required liability hedge ratio, freeing-up capital for the return-seeking bonus-potential portfolio.
Which is not to say that this all happened immediately. The small matter of the worst financial crisis for 80 years intervened. "An alignment of the three plans on the liabilities side made sense at the end of 2008, but we felt that it wasn't the right time to make those administrative changes - we were focused on getting the big decisions right," says Petersen. "Alignment was taking place through 2009, and now the three portfolios are much more aligned in terms of managers."
Asset allocation is not set in stone for years to come - there is still a lot of thinking to be done about the potential impact of Solvency II - but the trend is clear, particularly for the biggest scheme, MP. Equity risk is on the rise: the strategic benchmark is up eight percentage points from where it was in September 2009, to one-third of the fund's assets.
But Unipension takes that equity risk in a smarter way than before. "We don't have any do-or-die bets," says head of risk management, Søren Bang Andersen. The in-house team, which was hired almost completely anew in December 2008, is focused on managing core/satellite beta portfolios - but this is far from simply buying index-trackers. For starters, in-house managers use futures to control the beta exposures introduced by external active managers, effectively in the form of a completion portfolio: "We prefer to get as much active risk from stockpicking as possible, so let's say that as a consequence of stockpicking a manager is underweight US," Petersen explains. "If we don't want to have that risk in our portfolio, not necessarily because we have a view on the US but just to maintain our risk exposure, we could take a long position in S&P500 index futures."
In addition, before the equity allocation had been adjusted upwards, the schemes made the most of their risk budget with some "very positive decisions", as Petersen puts it, from a top-down view: "We were long emerging markets, for example, so in spite of our low-risk portfolio we came out with a good result." Again, it was the move to 1.5% on the guarantee that set the context for these views: "Every decision was clearly inside our VaR limits," says Andersen, "because of the low constraints from our liabilities. The volatility horizon on the liabilities side acts as a big buffer for our solvency ratio, which meant that we were able to absorb the extra volatility in the asset portfolio."
The managers are able to make nimble dynamic allocation decisions because, as well as the cash budget they are given, they have a mandate to use futures to generate exposures - as long as they are balanced by cash holdings to avoid leveraged positions.
This investment style - which Petersen calls "opportunistic" and is designed to use the funds' long-term horizons and ability to hold illiquid or deep-value, high-carry positions - came into its own during 2008 and 2009. "There were a lot of different opportunities last year that you wouldn't normally expect to see in the textbook," Petersen observes. Andersen adds: "Participants in the market got hit in 2008 - they needed cash, and we had cash. We had a very liquid portfolio, and we took advantage."
For example, a cool DKK400m (€53m) was made for MP alone simply by depositing a portfolio of corporate bonds with a cash-strapped but government-backed Danish bank for repo. A portfolio of private equity secondaries with a total commitment value of DKK500m, DKK100m of which had already been drawn down, was another opportunity that came up for sale. "We literally got that portfolio for free," says Petersen.
That private equity programme is brand new for Unipension, but the "very small" in-house team working with advice from its fund of funds providers has already committed DKK2.5bn since the beginning of 2009.
"Some of the portfolio is funds of funds and some direct GP relationships that arose due to the opportunities that came up last year," Petersen explains. "We are long-term, but we try to exploit the short-term possibilities as well. The special situation in private equity made us change our strategy slightly - going more into mezzanine, for example, and some secondaries - but those strategies are also useful to mitigate the j-curve of such a young programme. We will probably not invest in emerging markets or venture capital, and do very little in mega-cap."
If private equity introduces an illiquidity risk that Unipension will have to consider in the context of Solvency II, they are as nothing compared with the complexities inherent in the schemes' real estate portfolios. The holdings are mostly residential rather than commercial property, and these residences, managed by Dan-Ejendomme, are offered for rent to scheme members at market rates. For this historical reason, the responsibility over real estate lies directly with the schemes' boards rather than with Unipension's investment department. Moreover, as befits a pension scheme for architects, perhaps, AP's portfolio is heavily skewed towards holdings with special architectural value. While these are valued quarterly, those valuations are based on virtually non-existent volumes.
"It's difficult for us to calculate the risk profile of those assets as turnover has been very limited," says Petersen. "Some of them have been held for longer than anyone can remember!"
Discussions are not restricted to illiquid asset classes - the solvency debate and the move to a lower pension guarantee introduces new questions about how to optimise between nominal and real returns, duration-based and real assets.
"All three schemes have nominal guarantees, and therefore aim to have the highest nominal returns possible," says Petersen. "Without the old, high guarantees we are given some extra freedom in relation to our focus on real or nominal returns. That's a discussion that is on the agenda for the coming years. So we view our real estate - together with infrastructure, inflation-indexed bonds, forestry, and so on - as part of our real investments, and we still haven't finalised where we are going with that."
Divesting from hedge funds
One area where it has made a decision, however, is hedge funds. It has recently completed divestment from its fund of fund investments, citing a lack of transparency and their failure to provide real diversification during the financial crisis.
Making its initial hedge fund allocation in the expectation that the investments would not be correlated with its traditional allocations to bonds, equity and property, it invested in two funds of funds - one a broad, diversified product and the other more opportunistic and concentrated. "The crisis of 2008 revealed that they did not bring the diversification effect that we'd hoped for when we most needed it," says Petersen.
Unipension was also disappointed with the kinds of strategy that their underlying managers began to pursue in the months leading up to the credit crisis, particularly in very illiquid instruments. "We hadn't expected them to invest in very illiquid structures, which, after all, it makes much more sense for us, as a pension fund, to do ourselves, directly," says Petersen. "We expected our hedge funds to be more market-opportunistic." When Unipension engaged its providers on these issues, "the answers we got did not change our decision to get out of these structures", he adds.
Asked if Unipension might reconsider funds of hedge funds if they could get a tilt towards the strategies that did perform well through the crisis - directional trading strategies like global macro or managed futures, which are also arguably best suited to managed accounts that provide more transparency - Petersen remains circumspect. "We don't know yet what our approach will be," he says. "But if we are going to go back to hedge funds, we are going to require much greater visibility so that we can calculate our risk exposures all the way through and be certain that the funds offer real diversification - including during financial crises."
In many ways the investment strategy at Unipension is still a work in progress, as the three separate schemes continue to be integrated, as manager rosters are rationalised and consolidated, as the right balance between beta and alpha is struck for each fund and for the group as a whole, as the full consequences and possibilities of the lower pension guarantee is realised, and as the longer-term impact of Solvency II is factored into its overall risk budgeting. But the outlines are already clear: Unipension wants to continue to develop its mission as an opportunistic long-term investor, an intelligent taker of illiquidity and carry risk. It appears to have demonstrated its capacity to apply that opportunistic approach in distressed markets, but it was its conservative positioning in liquid instruments that provided whatever downside protection it enjoyed during 2008, rather than the strategic allocations it made for that purpose. Downside risk management is crucial if it wants to make its deep-value opportunistic positions compound successfully for the long-term without threatening solvency limits in distressed markets: it remains to be seen whether it will choose to develop that in-house, as part of its increasingly sophisticated beta management process, or return to the alternative investment industry with a better sense of its requirements in future.