Legislation passed late in 2010 completely changed the FRR’s purpose and liability stream. Martin Steward spoke with its CIO about the consequent re-tooling of its investment strategy

It’s no fun when the referee changes the rules of the game in mid-play. Imagine how it would feel if he decided, one-quarter of the way through a 90-minute match, that he was going to blow the full-time whistle after 45 minutes instead.

You are already sympathising with Philippe Aurain, CIO of France’s €37bn Fonds de réserve pour les retraites (FRR), and his colleagues. Established in 1999 - taking income from the social tax on income from estates and investments, the surpluses from the Fonds de Solidarité Vieillesse (FSV) and the Caisse Nationale d’Assurance Vieillesse (CNAV), the sale of some state-owned assets and a number of endowments - the FRR was charged with investing that income until 2020. At that point, its assets were to be dispensed in support of the long-term survival of France’s pay-as-you-go (PAYG) pensions system (the general Caisse Nationale d’Assurance Vieillesse des Travailleurs Salariés, or CNAVTS, and the Régime social des independents, or RSI, for the self-employed), which was expected to come under pressure from demographic ageing as the cohort of retired baby boomers grew.

Unfortunately, the government decided that it had already come under enough pressure from the financial crisis. As part of its controversial social security reforms, parliament passed legislation in December 2010 that will see the FRR make 14 annual payments of €2.1bn per year to the Caisse d’Amortissement de la Dette Sociale (CADES), which is used to refinance the debt incurred to pay pensions. The first payment will be made this year and the last in 2024, when the fund will close down. Furthermore, the fund will no longer receive any income.

It is difficult to imagine a more radical change to the mission of a state reserve fund. Not surprisingly, match tactics have had to change radically, too.

“In a way, you could say that now we are talking about FRR 2.0 - it’s a totally different fund, we are starting from scratch,” says Aurain, who has led development of the new strategy since he took over as CIO from Nicolas Sobczak in April 2010. “And yet one could equally argue that nothing has changed: FRR was always here to complement the PAYG system, all that has changed is the timeframe. Before, we faced no drawdowns before 2020 and a 30-year time horizon. Now, drawdowns have been brought forward by 10 years.”

Indeed, Aurain is admirably sporting in response to the referee’s decision. He spins it positively: “We now have a precise liability, a precise time horizon and a precise objective.” He seems genuinely to have relished the challenge of drawing-up the new strategy.

“From my appointment in April 2010, I was involved at the beginning of this new project for the FRR,” he says. “It was always my view that some kind of LDI process was best suited to the objective of delivering a secure, well-defined cash flow. Before, we used simulated liabilities to calibrate our asset allocation. Now we really need to secure this liability. That means we are now thinking in terms of a financing ratio.”

That underlines how different the FRR is now, especially compared with its pre-crisis composition. In fact, the big changes began to be discussed just as the FRR was finalising a newly de-risked version of its strategic portfolio: going into the crisis with about 60% of its portfolio in equities, it decided not to rebalance but to let that allocation fall naturally to 45%.

“Our performance since inception is about 3% annualised,” Aurain notes. “We’d have been much happier with 10%, but in our disappointment we have to recognise that, given our long-term timeframe, the level of equity exposure was not unsuitable. Our steady approach - not selling in a panic - helped us to benefit from the subsequent recovery.”

That 60/40 pre-crisis allocation now looks more like 40/60. But, more importantly, the structure has changed from a standard balanced portfolio strategy to a formal liability-matching one. Initially, 85% of its new liability cash flows will be hedged with a portfolio 50% in matched French bonds, 25% in international sovereign bonds and 25% in global investment grade credit. The non-domestic half will be non-matching, but managed with an overlay, so that it reflects the duration of liabilities as closely as possible.

The FRR will then pursue an opportunistic, dynamic de-risking strategy that re-defines the role of the investment committee previously in charge of setting strategic asset allocation.

“Now they must re-assess the opportunity either to increase the hedge ratio or decrease it,” Aurain explains. “The bounds in which they make these decisions are defined by the financing ratio - if it rises to a certain level a meeting will be triggered to assess the opportunity for profit-taking from the performance portfolio, which would effectively increase the hedge ratio; and if it goes too low a meeting would decide whether it is now time to hedge 100% of the liabilities, or take more risk because of market opportunities.”

That option to go 100% cash flow-matched if things turn sour is important, as it indicates the considerable leeway the FRR has under its new obligations. As Aurain wryly observes, the fund’s new LDI structure makes it look similar to a lot of UK defined benefit pension schemes, with one crucial distinction - the FRR is 140% funded. Indeed, perhaps it’s not so bad that the referee shortens the game, after all, if you are already 3-0 up with a row of defenders waiting to come off the subs’ bench. That cushion means you can hold those reserves back and press forwards.

“The 140% funding level is why we are not 100% hedged today,” says Aurain. “But once you have largely hedged your liabilities you are free to think of the other parts of the portfolio in terms of independent risk budgets.”

So, while the fund has gone from 60/40 to 40/60, the result is a performance portfolio that is more innovative and thoughtful about its risk-taking than the old one. For a start, it has an absolute return target of 6% per annum, rather than a market benchmark.

“The performance portfolio reflects some lessons learned during the crisis,” says Aurain. “First, diversification is very useful, but it isn’t useful for managing extreme risks. Second, you can’t avoid extreme risks. The first conclusion is that you must hedge your liabilities. The second is that you must have more flexibility to manage cycles and risk. That’s why we didn’t want a benchmark that would tie us to a pre-set asset allocation.”

There is a strategic portfolio. It is innovative in its attempt to balance fundamental economic risks rather than simply asset classes (so 70% is marked out, not for ‘equities’, but for ‘entrepreneurial and business-linked assets’ - equities and high yield bonds) and in splitting the equity portion equally between European, non-European developed, and emerging markets.

“That allocation will be subject to a lot of flexibility,” says Aurain, “but in any case it is far away from the usual benchmarks, particularly in terms of exposure to emerging markets, and a long way from what we used to have. Our emerging markets portfolio will be innovative in other ways, too. Most of it will be traditional local exposure, but we will also try to select managers who have a good track record in investing in developed market companies that have significant business exposure to emerging markets.”

Equity exposure will be a mix of passive and active strategies. But the FRR is pushing boundaries here, too, combining traditional cap-weighted passive strategies with others tracking alternatively-weighted indices. “We want to test a few different methodologies,” Aurain explains. “Initially we are looking to test fundamental-weighted in the form of RAFI, which we already use; minimum variance; and probably something in the maximum Sharpe ratio family.”

Just as it does for active strategies, the FRR plans to analyse these alternative betas for underlying risk exposures and use those insights to over-or underweight them based on macroeconomic views. “This worked very well for RAFI in 2009, for example, which is clearly value-biased,” notes Aurain.

In addition to ‘entrepreneurial and business-linked assets’ the performance portfolio will have about 15% of emerging markets exposure, in the form of bonds, and 15% in real assets in the form of commodity futures and listed property.

“We will remain passive in commodities because we want to steer away from the debate about speculation and prices,” Aurain explains. “We will probably employ an optimisation methodology to minimise the negative roll yield from futures that are in contango, and we will only invest in energy and metals indices - not agricultural commodities.”

The fund had launched a tender process for unlisted property management, but abandoned it to maintain as liquid and flexible a portfolio as possible, consistent with its new, shorter, time horizon. For the same reason, distributions from €900m worth of investments made in 2006-07 to private equity funds managed by Pantheon Ventures, Access Capital Partners, Neuberger Investment Management, and a secondaries fund with Axa Private Equity Europe, will be re-allocated to the listed equity portfolio rather than re-committed to unlisted assets.

With the strategy in place, the FRR is now deep into the long process of re-defining existing mandates and tendering new ones. What has emerged is a strictly-defined obligation, but in many ways a more challenging and complex one: taking risk to de-risk and de-risking to take risk, one might say. Its new freedom, paradoxically, comes from this strict definition of goals - combined with the 40% excess funding. If all goes well, at the end of its life in 2024 the fund should be able to deliver a lump sum of perhaps €18bn to CADES, over and above the cash flows defined in the new legislation.

“Probably no-one would have expected such a change, to be honest,” Aurain muses when asked to reflect on the new shape of his role as CIO. “But it has been really fascinating to develop this proposal. My first job at FRR in 2003 was to come up with the FRR’s first strategic asset allocation, so in a way this last project has brought things full circle.”