Sweden’s AP2 backs own dynamic bets

swedenA committed ‘return seeker’, Sweden’s Andra AP Fonden (AP2) exploited the repricing of risk during the financial crisis by investing decisively in convertible bonds and credit, says Tomas Franzen, chief investment strategist at the SEK204.3 billion ($28.5 billion) fund. Now it is looking at real assets and emerging Asia to further diversify its sources of return. Simon Mumme reports.

Tomas Franzen

AP2, one of the five ‘buffer funds’ in the Swedish pension system, has a strong appetite for risk. As a “return seeker, not a liability hedger,” the fund targets a 5 per cent return above inflation, says Franzen. “To achieve this goal, we need take on various risk premia at a rather large extent.”

This is evident in its overall 58 per cent allocation to equities, which saw the fund’s performance sink by -24 per cent during 2008, then boom 20.6 per cent in 2009- “the best result since the fund’s start,” its website declared – as markets rallied.

Before this rebound, however, AP2 responded to “huge dislocations in various markets” to buy risk in the dark days of late 2008 and early 2009, Franzen says. “Specifically, we saw that we could be more diversified in corporate risk by owning not only equities in companies, but also corporate bonds and convertibles and private equity.”

AP2 watched the convertible bond market freeze as hedge fund managers were forced to return capital to flighty investors or those in need of liquidity. It then appointed some of the same managers, many of which specialised in convertible arbitrage strategies, to run long-only portfolios of convertible bonds.

“We saw, before the crisis, that 70 per cent to 80 per cent of the assets in the [convertibles] market were held by arbitrageurs. Then it wasn’t possible for them to finance their positions because original investors in these strategies took their money out, and credit, for use as leverage, dried up,” Franzen says.

“What we saw was a flood of convertibles sold into the markets with sharp increases in yields and pricing optionality. So it was quite natural to pursue managers of these assets but on a pure beta, long-only basis rather than an alpha-arbitrage kind of way.”

These new exposures, which now account for 3 per cent of AP2’s strategic portfolio, were funded from its global equities portfolio. Meanwhile, the fund also sold down 4 per cent of its sovereign debt holdings to capture the “pure risk premia'”on offer in corporate credit markets, Franzen says. As 2009 progressed, AP2 was heavily exposed to the great beta recoveries in debt and equity markets.

“Last year, corporate credit and convertibles performed better than what most people expected, so a lot of that repricing has been done over the past year. But some beta returns still seem to be available.”

The choice to invest in credit and convertibles were made within a new decision-making framework introduced after an unsuccessful application of tactical asset allocation. Seen as three- to five-year exposures, the convertibles and credit investments indicate a more dynamic approach to long-term asset allocation at AP2.   

“It’s part of the strategic portfolio, and more of an attitude to strategic asset allocation, you could say. It’s not changing the portfolio so that it matches a pre-specified risk budget or return goal, but when risk is repriced we would like to be a bit more pro-active.

Another manifestation of this strategy, which AP2 implements without input from a consultant, was the fund’s decision to increase its currency hedge as the Swedish kroner was hammered, Franzen says.

Get real

Some investors describe 2009 as the year for beta. AP2 clearly benefited from the substantial rally in risk assets, and now pursues investments in timberland and agriculture “as a more pure diversification play” and not an opportunistic exploitation of risk premia, Franzen says.

“As far as returns are concerned, the dynamics of these returns should be very dissimilar to other assets [because] many of these investments are illiquid.”

Another investment theme for these assets is the growing demand for farmland to feed the emerging middle classes of Asia and Latin America.

AP2 will invest in the US, Australian and New Zealand timberland market before allocating to the less mature and riskier timber and agriculture markets of Latin America and Eastern Europe. Over time, it aims to “build a diversified global portfolio in these areas” but has not set aside a specific amount of capital, preferring to “handle this like we handle private equity” and allocate money when opportunities to invest with good managers arise, Franzen says.

He is confident the capital lock-ups involved in these investments will not prevent the fund from meeting its payment obligations to the Swedish pension system in the next 10 to 20 years. Currently, the fund’s investment income sustains a payment rate of about 1.5 per cent to 2 per cent each year. But in the next five years or so, this rate will rise to about 4.5 per cent, and the fund will be forced to liquidate part of its portfolio to meet this need, “which will have some consequences for taking illiquidity risk”. Still, AP2’s illiquid investments are well below its threshold of15 per cent.

The fund has allocated to emerging market equities and debt since 2002, “but as global markets have drifted more and more into emerging markets, we find ourselves in a neutral position in the strategic portfolio,” Franzen says. Now the fund will assess a strategy to lift the exposures “to establish a global overweight to these markets, in which we will also consider taking a regional approach to favour the Asian region in the emerging markets world”.

Drag tracing

Fully 80 per cent of AP2’s assets are managed internally. In 2009, the fund created a new unit, the strategic exposure and trading team, to ensure that its portfolio was implemented with optimal efficiency. “It is not very much alpha-driven,” Franzen says. “It’s very much about cost-efficiency and handling our currency hedges.”

The team got off to a bumpy start. “Last year, it was quite tricky to keep strategic weights and not have return drag. But coming into this year we’ve seen a huge improvement in the return drag, and that’s a result of our increased focus and attention on this.”

In 2008, the team generated a 50 basis point loss, or ‘return drag’ on the portfolio. But this implementation shortfall has since been reversed: in the first three months of 2009, the team recorded no loss. However, similar to the way in which the performance of active managers is influenced by market conditions, these results are not fully attributable to the team’s skill.

“It’s not only to do with what we can achieve ourselves, it’s also due to what kind of market environment you’re in. Lately, the dynamics of the market have been somewhat kinder to us.”

AP2’s other internal teams, which run portfolios of Swedish equities, fixed income and quantitative, enhanced-index strategies, have been in place for a number of years. Many of its passive funds run fundamental indexation strategies instead of conventional market capitalisation-weighted indices, which it regards as more inefficient.

The fund was an early adopter of non-market capitalisation weighting, using fundamental-, equal- and GDP-weighted indices as investment vehicles to reduce the long-term return drag from market capitalisation weighted equity portfolios. Last year, these non-market capitalisation weighted portfolios helped boost AP2’s equity returns following the market dislocations of 2008.

But fundamental indexation can make rebalancing the fund’s strategic portfolio problematic: “one positive of a market capitalisation portfolio is that it has liquidity,” Franzen says.

AP2 does not plan to increase the amount of money it manages internally in the foreseeable future. External managers are appointed to oversee exposures in which the team does not have adequate expertise, such as global equities, private equity and ‘”pure alpha” tactical asset allocation strategies, Franzen says.