Competing theories come to the fore as quantitative models put under the microscope
Quantitative investing is once more under the spotlight, writes Yuri Bender. Last year’s resignation of Axa Rosenberg founder Barr Rosenberg, known for his ‘Bionic Betas’ and early use of quant methodology, has created a new crossroads.
As a young and charismatic finance professor at Berkeley, California, in the 1970s, Mr Rosenberg used systematic thinking to explain the vagaries of previously unpredictable markets. His pioneering work involved taking the ‘beta’ measure of a stock’s volatility, devised by another Californian academic, William Sharpe, and mixing in data on companies’ liabilities, earnings and sector stats.
Like many academics turned investors, he packaged up and sold his computer-aided techniques to the institutional world, under the once invincible, Barr Rosenberg Associates (BARRA) risk management brand.
Ironically, it is the risk reporting that Professor Rosenberg championed, which eventually let him down. Under typical quant methodology, it is this data on risk which has an important influence on the construction of investment portfolios.
Axa Rosenberg had a leading position in the quant management world of the 1990s and as recently as 2009 was running $70bn (E52bn). But theirs was not the only game in town. One controversial heavyweight doing the rounds in Europe and beyond is Professor Robert Haugen, whose “minimum variance” methodology, used in his own consultancy and also as academic adviser to BNP Paribas subsidiary Alfred Berg, is particularly compelling.
His theories about market anomalies, going back to 1967, backed by empirical research suggest best returns are usually generated by lowest risk stocks. This turns on its head both the Efficient Market Hypothesis and Capital Asset Pricing Model, which power the curricula and income stream of Europe’s business schools.
The perceived academic wisdom, disproved by Mr Haugen, remains that high risk means potentially high returns in an efficient market place, where everybody apparently knows everything everybody else is doing. This often seems little more than a fantasy.
One of Mr Haugen’s key competitors is Fiona Frick, head of equities at Swiss private house Unigestion, who follows a similar process. She likens quant management to flying with an automatic pilot, while fundamental management uses a human pilot. In order to return safely, today’s investment journey needs both, she believes. Unlike Mr Haugen, Ms Frick occasionally uses volatile stocks for diversification.
Both the Unigestion and Alfred Berg processes tend to underperform in bull markets, which generally have thematic drivers, but outperform indices when markets are falling.
The breakdown in historical risk measures in 2008 has certainly powered the return of the quants, with some investors keener on the pragmatic rather than purist approach.
