Portugal created its pension reserve fund, the FEFSS, to fill the gap between social security contributions and pension payments that was expected to appear around 2015. The aim was to provide a buffer fund to make good the shortfall.
Since then the fund has grown substantially. Last year, IGFCSS, the state-run investment management operation that runs the fund, became the largest pensions asset manager in Portugal. Assets under management at 31 December 2001 totalled e3.8bn, putting it ahead of Pensoesgere, a leading pension fund management company, with assets of e3.6bn.
Today, FEFSS has assets of e4.2bn, and is expected to grow to e7.5bn by 2005. Most of this growth has occurred in the past five years. Until 1996 it received only a trickle of cash from the social security ministry’s treasury. However, in 1997 the government decided to increase the inflow, and since then the fund has received more than e500m a year.
The fund has performed creditably. In 2000 and 2001, years of negative returns for Portugal’s pension funds, it achieved 4.1% and 3.3% nominal returns respectively.This is largely because the fund is run defensively, with 95% confidence that it will not produce a nominal negative return in any one year. Henrique Cruz, the board member of the Porto-based IGFCSS responsible for managing the fund, explains: “We have had one main objective for the fund, which is to preserve the capital in real terms. So we have a portfolio that does not put at risk the real return on a year-by-year basis”.
Now Cruz and his team are considering other, more adventurous, options. The fund has just completed an ALM study with the US investment bank Lehman Brothers which posits a number of options, from retaining the status quo, to trying to beat a benchmark. These options are currently being put to the IGFCSS advisory board and will eventually go to the minister for social security.
The ALM looks at options from now until 2016. This is the accumulation phase of the fund, when social security contributions are greater than pension payments. “Projections show that around 2016, the social security treasury surpluses will be gone, and then the fund will start to be used for
pensions,” says Cruz. “By 2034 the fund will have run out of money and there will be a projected deficit of 1.8% of GDP. But this deficit is projected to remain constant for the future. So now we have 34 years to think of a solution that is equivalent to the deficit.”
Cruz reckons the IGFCSS team can prolong the life of the fund for another three to five years if they are allowed to move away from the present conservative investment strategy to a riskier strategy of trying to beat a benchmark, with the possibility of nominal negative returns on a year-by-year basis.
“What we need to discuss with the advisoryboard and the government is the risk budget. The big question is whether the government still wants positive returns every year, or whether they are prepared to accept one or even two negative years and then a positive year to compensate?”
Cruz fully understands why the board and successive governments take a cautious approach to investment. “The fund is a very scrutinised fund. You have to keep it in mind that you are not managing a pension fund for 500 employees. You are managing the money from five million Portuguese people. It is a big responsibility.
“But I think that now there should be discussion on this because by putting this present constraint on the fund you are making people – potentially – poorer in the future. It’s a trade-off between the short-term risk and the long-term poverty”.
Statutory investment rules also limit the fund’s risk. By law, the fund must hold at least 50% of its portfolio in Portuguese government debt. There is also a 20% cap on equities, which must be euro-denominated. The fund currently holds about 9% of its portfolio in equities.
Cruz would like the rules revised to allow the fund to invest in OECD equities that are not euro-denominated. “Currently the fund can invest outside the Euro-zone but only in euro-denominated assets. When you go to outside this you can get higher returns for the same amount of risk because of diversification”.
Portuguese government debt is another issue. The current strategy requires Portuguese debt to converge to the amortised costs. This ensures a low risk and return, similar to a coupon. The ALM study shows that one way of getting a better return, within the existing one-year positive return strategy, is to limit the maturity of the debt to, for example, five years.
He would also like a maximum placed on Portuguese government debt allocation: “Right now you don’t have a maximum limit. So maybe what would be acceptable is a maximum of 60%, or 55%.”
“Even if you conclude that you shouldn’t have so big a percentage of government debt in your portfolio, you still have the problem of financing the Portuguese treasury. My suggestion is that, as the fund will continue to grow, you can have a strategy that will not affect the Portuguese financing. If you don’t buy more debt, you don’t invest in the same proportion for the future. So your 50% will fall to 40% and to 35% without having to sell”.
Cruz would also like to invest more in real estate, currently performing well in Portugal. “Real estate is very good for inflation hedging, because the fund pays pensions which will be actualised according
the Portuguese inflation. But the fund invests in financial assets that reflect Euro-zone inflation. So we don’t have a perfect hedge. The only way to have a perfect hedge is buying real estate in Portugal.”
The fund currently has only 1.5% of its portfolio invested in real estate, compared with a mean of 4% for Portuguese pension funds and with the ALM’s allocation of 20% to 25%.
This is for historical reasons, he says. “The fund does not have the infrastructure to do real estate investment. But we have started to recruit a team for that recently.”
He is also concerned about liquidity. “Pension funds have a different horizon than we have. They don’t have in mind that one-day they will have to sell real estate. We have to keep that in mind. Somewhere in the future we will have to start to sell and then we will have trouble with the liquidity of real estate assets. We have decided recently to go to 5% in the near future. But if we want to go to more than 5%, we have really to discuss this problem of liquidity.”
Once the IGFCSS has put various IT and fund administration solutions in place, it plans to outsource a small part of FEFSS assets – between 15% to 20% – to external managers.
Mandates will be specialist, says Cruz. “This is because we want to keep the asset allocation decisions with us. We expect those specific mandates to follow a benchmark with a maximum tracking error. This way we can ourselves compensate for any bad performance by the specialist managers and keep the portfolio within its volatility tunnel.”
Outsourcing a balanced mandate would not suit the fund, Cruz says. “There are some assets that are more weighted in our portfolio than they are in the international benchmark. They need more attention from the manager than they would get if you provided a balanced mandate. For example, we have 50% of Portuguese government debt. If you put that in a balanced mandate a manager will just look at it as a residual asset.”
The mandates will not be put to public tender
until IGFCSS has gained approval from the government. One change that could encourage FEFSS to adopt a more risky investment strategy would be a government guarantee that the funds would not be touched until 2016. Governments in other countries, such as Sweden and Ireland, have given such guarantees. “Other pension reserve funds can take more risks because of this,” says Cruz.
If the government were to lock the money into the FEFSS fund, Henrique Cruz would find he had a far larger risk budget at his disposal.

Topics