Can quant approach recover after performance slump?
The performance of the Jyske Invest Aggressive Strategy has not been up to scratch in the past two years. Philip Haddon assesses whether Jakob Greisen is set for a return to form after adjusting his processes.
As a former professional golfer, Jyske Invest fund manager Jakob Greisen (above) would admit his fund’s performance has not been up to par in the past two years. After three years of strong performance, his Jyske Invest Aggressive Strategy fund slumped well below the index.
As the graph below shows, from the end of October 2007 to the end of June 2009 his fund lost 50.4%, while the FTSE World index fell 37.3%. This saw him fall down the three-year sector fund manager rankings from near the top at the end of October 2007, when he was ranked 11/580, to near the bottom at the end of June 2009, when he was ranked 539/596.
The formerly AAA-rated Dane thinks the bear market revealed the flaws in the ‘VAMOS’ part of his quantitative process and he has acted to improve it. VAMOS stands for valuation, momentum and strength. The system scores stocks on the basis of their valuation and earnings momentum to create a shortlist on which Greisen will then perform due diligence to form his concentrated 40-stock portfolio. .
He says the process aims to seek out the sort of stocks that everyone craves: ‘cheap companies performing well’. But he accepts the process was slightly flawed, as he says many of his bets were too closely correlated and he was drawn into a commodity trap that hit performance.
‘We had a large exposure to commodities and commodity-related companies, as they had positive earnings momentum when momentum was falling all over the place. But when the financial crisis really hit and commodities were hammered, we were hard hit because we have a quant model guiding us.’
The Jyske Invest Aggressive Strategy fund also suffered as a result of its large exposure to Russia, especially to steel company Mechel, which became embroiled in a tax fraud scandal. Greisen sold the stock on the day that news of the accusations broke, but not soon enough to avoid a plummet in the share price.
Many fund selectors have criticised quant strategies’ performance during the crisis, and Greisen accepts the criticism as he underperformed on the way down and failed to keep up in the rallies of 2009.
‘When the market turns very fast, it is not the ideal market for a quant-driven manager,’ he says. ‘There was a fast bounce from the bottom in March, and at that time we were positioned for somewhat more defensive names. We were left a little on the sidelines there. For a quant-driven manager, we saw the worst two outcomes in 12 months. First there were sharp falls, then, once we had stabilisation in earnings estimates, we had another big shift in the market into early cyclicals, which do not score well in a quant model.’
To avoid getting caught out in future, Jyske Invest changed the quant part of its process to put a greater focus on risk and diversification.
‘We have not changed our screening model, but we have added a risk model to look at two factors. Firstly at the stability of a company – we have five factors to measure how volatile it is – and we also now have a diversification factor, to see how high a stock scores for diversifying the portfolio. We will still make lots of bets as it is an aggressive strategy, and it is still a stockpicking fund and never a benchmark fund. But we want to make sure that in future our bets are not too correlated.’
One result of the changes is that Greisen has recently been buying into US banks such as Goldman Sachs, Wells Fargo and, ‘a little more controversially’, Bank of America, on the back of strong short-term earnings momentum. ‘If you look at a one to two-year period, some of these banks could double in price,’ he says.
So, after a tough year and a re-think of his strategy, is Greisen’s Aggressive Strategy an attractive option for the next cycle of the market? He emphasises that his strategy is not purely quant based, saying ‘we don’t believe in just the quant model, we believe in our unique combination of quant and due diligence.’
He is also hopeful that markets will return to ‘normal’ over the next 12 months. ‘We hope that for the next 6-12 months, there will be a period where cheap companies with earnings momentum will outperform. Our backtesting shows that our process has worked for 16 out of the last 18 years, so we are confident the quant model will come back.’
But despite Greisen’s confidence, selectors seem likely to remain wary of investing in funds that are reliant on momentum and quant factors.
Greisen is relying on markets returning to ‘normal’, but after the bursting of the credit bubble and the events of the past two years, it is difficult to tell if such normality will return. Indeed, Pimco founder Bill Gross recently said that investors must adjust to a ‘new normal’, and I do not yet know if such quant approaches can do that.