ATP: experiments with alpha and beta

“There is very little pure alpha” said Henrik Jepsen, chief investment officer of ATP, at the Fiduciary Investors Symposium in Amsterdam when reflecting on the giant Danish fund’s experiences with the return class.

The DKK 624-billion ($114-billion) ATP decided to merge the alpha and beta platforms of its investment portfolio earlier this year. This wound back a 2005 decision to create a designated separate alpha unit – in effect a hedge fund subsidiary.

Jepsen explains how ATP’s views on the alpha/beta divide shifted during its great efforts to generate alpha. He thinks any alpha generated after the 2005 split was always a form of smart beta. “We probably reduce alpha now, but by accident,” he says. “I’m not ruling out the existence of alpha but it’s very difficult to extract, particularly on a large scale.”

According to Jepsen, the alpha portfolio was a definite success in constantly generating positive returns with little correlation to the beta part of its $66-billion investment portfolio. “The problem was that the returns were simply too low on an absolute amount”, he says, compared to ATP’s total portfolio. Nonetheless, he reckons that ATP has “retained a hedge fund platform of international quality”, while lowering costs by now reuniting the alpha and beta platforms in a single investment portfolio unit.

“Back in 2005, we saw the world consisting of two types of returns: beta as a sort of reward for carrying market risk over time, and alpha that could give you returns by outsmarting the market,” Jepsen says. ATP’s thinking these days is that “alpha is much smaller than what we thought and what is considered to be alpha is very often a kind of beta – in being a form of exposure against a systematic risk factor”.

ATP’s non-benchmarked return-seeking investment portfolio will continue to be divided into five risk classes: rates at 20 per cent of risk budget, credit at 10 per cent, equities at 35 per cent, inflation at 25 per cent and commodities at 10 per cent. The risk allocations are designed to ensure “all classes are meaningful but there is not any one dominating completely”. The portfolio is vital is generating cash flow to meet benefit commitments and Jepsen says it is designed to perform “more or less all the time” due to capital considerations. Some 85 per cent of the portfolio is invested in house.

Outside of the investment portfolio, a similarly sized hedging portfolio is invested 50 per cent in bonds and 50 per cent in swaps (both purely in Denmark and the eurozone. The purpose of the hedging portfolio is “to generate the promised interest rate, which is the accrual of our guarantees” as well as hedging interest rate risk, according to Jepsen.

Liquidity focus

A new liquidity-risk management model is identified by Jepsen as another key recent adjustment to ATP’s investment activities. “Liquidity is only a problem when you need it and we have seen how debilitating liquidity crises can be, so we developed an extreme focus in making sure we never get caught with insufficient liquidity,” Jepsen states. The fund has initiated a system of stress testing its ability to generate liquidity and liquidity needs to daily, weekly and annual horizons.

“As we have a very large portfolio of interest rate swaps, our business model is largely dependent on a well functioning banking system, and managing risks like that is a focus,” says Jepsen. A new need to post “a very large sum” of collateral for derivatives in European central clearing rules has also had an important part to play in increasing ATP’s focus on liquidity.

Cockroach approach

ATP’s investment strategy naturally faces the same demands as its international peers in navigating a low-yield environment, which Jepsen also argues is prone to shock. “One of my concerns is that we have all these statistical models that generally underestimate the number of big shocks that we have every five years or so,” he says.

ATP is aiming for robustness in this environment. Jepsen cites a thought from veteran Wall Street risk manager Richard Bookstaber that investors can learn from the cockroach. The cockroach has a very good risk management approach due to the wind sensors in its hairs, he explains. “If you’re a pension fund, you want to survive in the long run and you can maybe focus less on specific statistical models,” says Jepsen.

Maintaining a balanced portfolio and extending diversification to protect against shocks are also outlined as guiding principles for ATP in the current environment. Jepsen urges his fellow investors to focus on their comparative advantages and adds that patience is needed, as “no investment strategy will work at all times”.

Jepsen says ATP has been able to buck a trend for risk parity portfolios to underperform this year as “we have been much higher with our equity allocation than is usually the case”. That shows the importance of being adaptable, adds Jepsen, despite stating that ATP retains its long-term faith in qualities such as diversification and balance.