Liam Kennedy spoke with Angelien Kemna, chief investment officer of APG, the Netherlands’ largest pension asset manager with AUM of €278bn, about her policies of ‘minimum regret’ and ‘controlled simplification’
The Netherlands’ most powerful woman is keen to communicate her values, or at least those of her organisation. APG’s culture is different from that of a commercial asset manager, says Angelien Kemna, chief investment officer. “We realise that we manage a lot of money, also for a lot of people who do not get large amounts of pensions,” she says. “Individual managers here, top to bottom, realise this is captive individuals’ money.”
Owned by pension funds including ABP, the largest in the Netherlands with assets of €235bn as of September 2011, APG was created in March 2008 from ABP to comply with government regulation that required the separation of the pension fund from its executive operations, including administration and investment management. In September 2008 APG merged with the building sector pension fund’s Cordares, which at that time had assets of almost €30bn. Since then, APG has acquired other clients, including PensPlan in Italy, for which it manages a €100m mandate. The organisation had AUM of around €280bn as of end-2011.
Through ABP, founded in 1922, the Netherlands was an early adopter of funding for pension liabilities, although the institution changed course dramatically in terms of investments when it was privatised in 1996. It then became an independent entity and undertook a major programme of asset diversification under Jean Frijns as CIO, moving away from a portfolio primarily of Dutch government debt.
Kemna, named most powerful woman in the Netherlands by the feminist magazine Opzij in autumn 2011, has been CIO of APG since November 2009, succeeding Roderick Munsters. Indeed, alongside Ho Ching of Singapore’s Temasek, she is one of the few women globally to achieve a position of such seniority in institutional investment.
Since Kemna arrived at APG from academia - she is a quant by background and previously worked for ING - the CIO has introduced a policy of controlled simplification to the investment portfolio. This has meant re-evaluating the portfolio in terms of its overall goals but also from the perspective of the members and pensioners: “It is looking ultimately to our overall client portfolio that matters.”
First comes the choice of benchmark or ‘smart beta’ - non-standard benchmarks - which APG has embraced. For instance, €40bn of its €70bn developed equity portfolio is run using ‘smart beta’ benchmarks with a basis point fee and almost all internally “that is probably lower than anywhere in the market for an external mandate” as Kemna puts it.
Controlled simplification also means reducing complexity in the overall portfolio in terms of the strategies and instruments that are used. “What adds to performance?” the CIO asks. “Can we not move with bigger, simpler steps, back to basics to say we don’t need all these difficult instruments? We have already taken out a lot of derivative instruments that do not add value and only make things complex.”
Not everything is out though, and there is a commitment to hedge funds, for instance. Indeed, APG is the sole client of the fund of hedge fund manager New Holland Capital (NHC), which is based in New York and owned by its partners. APG will continue to use it, although with Kemna’s caveat: “You have to show to the outside world, if you add very expensive hedge funds through NHC, that they do add value to your client portfolio.”
And quant is certainly in - APG has one of the biggest quant portfolios globally with €30bn in developed market equities and what Kemna describes as “an extremely good team”.
The controlled simplification process means critical attention on external managers, which currently account for 24% of assets. In practical terms, this will not favour managers who think too much in terms of their own business, performance fees and “generating alpha by twiddling here and there”. APG has “defined much better what we really want from them and what we really do not like”, the CIO says.
“We think in terms of a certain strategy; within that we ask, does this external manager fit as an addition or does he do exactly the same as we do? We typically don’t want those anymore, they really have to add value. Also, on the illiquid side we see a lot more focus on valuation both from the regulatory authorities and from our clients, so we demand more transparency there as well. This year we are going to test how many external managers of illiquid assets in which we have large investments can bring us the valuation and transparency that we want. We will be very critical of our external managers.”
Managers of small accounts will also be terminated: “If I have external managers doing a brilliant job but only on €33m in a developed equity fund out of €70bn, then I really don’t think I need them. That won’t do either in the context of controlled simplicity.”
The majority of the controlled simplicity “homework” has been done, Kemna says. Now the task is to examine in a structural way how financial markets are developing. “Is diversification still working? Should we shy away much more from traditional benchmarks either to non-traditional ones or ones closely related to liabilities? Now we have settled most of the controlled simplicity programme and we’re down to execution, we’re also trying to rethink where we stand. Financial markets have changed a lot and given that we better understand our culture and what we do things for, how can we start adding new thinking that helps us with our goal?”
As one would expect, APG’s senior staff espouse a variety of views on the current state of capital markets and the wider economy. In common with other Dutch pension funds, APG sold out of Greek debt but still holds Italian government bonds - part of what Kemna terms a‘minimum-regret’ portfolio that aims to reduce downside risk in the event of a disorderly default. Some hedges are in place, but not to reduce the upside potential assuming a relief rally on capital markets, even if a sustained relief rally seems far off.
“If you have diversification of opinion, it helps you having a minimum-regret portfolio. It won’t urge you to do too many unnecessary actions. It has been my opinion for many years that if there is such a diversity of opinion you should not take either the one bet or the other, because that is not a sensible strategy for the longer term. Timing is the most difficult thing in asset management decisions as we all know.”
In the short term, Kemna says it is clear that the ECB needs to play a bigger role “because the nervousness is too high”. She adds: “For the medium term if you look at the sheer size of euro liabilities - and they will continue to grow - there is a huge demand for a safe, liquid euro bond market. I know you need extra measures on the fiscal side and on the guarantee side but if there is a clear-cut, well-defined market it will compete with the US and a lot of money that has flowed to the US will flow back to Europe.”
APG believes that its asset diversification has functioned effectively overall since the 2008 crisis, although Kemna says this is due, in large measure, to the illiquid part of the portfolio. “Our credit portfolio has done extremely well this year, as has some of the illiquid part, while we have seen that emerging market equity was the poorest performer over all,” she says.
“Currently, we continue to look at the portfolio as a whole, the role that all the asset categories have in building the client portfolio and the risk factors that are underlying them. In the last couple of months we have been giving a lot of thought to whether these high correlations are going to stay or not. If they are going to stay then we need to do some extra homework. We are not the only ones looking at these high correlations and it’s my guess that we will see many papers going forward to address these questions.”
Now APG plans to manage risk differently, although not in the same way as ATP, for example, which manages five pre-determined risk classes (rather than traditional asset classes) at the overall portfolio level.
“We always look at the liability profile, the assets we have and their role towards the liabilities. Given the high correlations for the various asset categories, we are thinking of taking things a step further and looking at whether there are other ways to create diversification. Or are the risk factors we are using sufficient, given the changes in the market we are seeing?
“Increased volatility and increased correlation between the various asset categories is something we certainly worry about. We are also rethinking the way we achieve diversification in long-term themes that are not necessarily correlated. We will look at changing benchmarks, moving more from beta to smart-beta, considering themes like healthcare for an ageing population, or scarcity of food and water and how we can play this theme for the next 10 years, at least, to make sure that in this part of our portfolio we look into some of the diversification. That will be done next year or the year after.
“Maybe it will encourage people who have more resources than we have to do some thought processes for us,” Kemna muses, thinking of ways to unleash academic firepower to the benefit of APG and others.
With the forthcoming introduction of the new national pensions agreement, the Dutch supplementary pension system is moving away from a system that focuses on nominal pension rights to one that focuses on real rights - in other words, recalibrating the financial assessment framework (FTK) to take into account the fact that the ambition of the system is to provide increasing pensions in line with inflation.
Kemna is aware of the law of unintended consequences, especially when it comes to such a significant financial player as the €800bn Dutch pensions industry. She warns that policies must be tested thoroughly for unwanted effects and that pension funds must not move in tandem into inflation-hedging assets in order to avoid future bubbles.
“If €800bn is going into all kinds of inflation assets, then that won’t work because we will create our own bubble. So I don’t think we should go all the way in that direction. We do seek inflation-like assets but I think that will be quite limited, particularly if inflation creeps up dramatically.
“We do look at what we call alternative inflation, where we use inflation linkers and project linkers, so that is an area we want to expand but it would not be sufficient. We continue to ask whether the implicit inflation instruments or more traditional asset categories have a higher return. It would be rather stupid to create your own bubble.”
With a long-term target of 6-7% per annum, this pretty much leaves conventional asset classes, in particular public equities, as the major inflation-generating instrument for a pension manager the size of APG. But non-standard, non-market-cap-weighted benchmarks will be at the core of the strategy.
“If you look at the straightforward equity market equities, they are relatively limited, so we have a wide diversification in various assets,” concludes Kemna. “And for equities, we have certainly started to move to smart beta strategies that are much more interesting for pension fund liabilities, such as the minimum volatility strategy, for example. In the traditional asset categories, we move away from traditional market indices because they are not, per se, ideal for pension liabilities.”
The FTK already placed a considerable cost on risk assets in terms of capital buffers, and the general trend in financial services regulation - be that Basel III or Solvency II - is certainly less tolerant of risk assets overall. There is also a strong debate, both in the Netherlands and in the wider EU, in the run-up to the proposed new pensions directive over the requirement for risk taking in capital markets to fund expected levels of pension versus the need for some level of security for those pension rights. Meeting the pension goals of the upcoming generation of Dutch teachers, civil servants and firefighters will not be any easier for Kemna or her successors at APG.
This article first appeared in the February issue of IPE magazine.