FTAdviser.com

Advertising

Allan Conway sees himself as "policeman" to the £277m Schroder Global Emerging Markets fund rather than as a "traditional" fund manager. As head of emerging market equities at Schroders, he runs the fund but has a hands-off approach to stock selection. Instead, he manages the risks associated with stock selection undertaken by his team of sub-managers.

“I oversee the entire process to make sure it’s working how it’s supposed to. I am the policeman of the system,” he says.

The system is based on the belief that country selection is as important as stock selection. “Historical performance in emerging markets shows quite clearly that country selection has contributed 50-80 per cent of returns, and you ignore it at your peril,” he explains.

The weightings of the various emerging economies in the index are driven by a proprietary quantitative model with inputs ranging from macroeconomic indicators to market and stock data. Mr Conway chairs a monthly meeting to discuss the asset allocation template, but says he has limited scope to override the model.

Russia is one of the model’s more dramatic asset allocation bets. Over 14 per cent of the portfolio is invested in Russian stock, compared with only 11 per cent for the MSCI Emerging Markets index. India, by contrast, is given a weighting of just 1.1 per cent, significantly lower than its 5.7 per cent share of the index.

Stock selection, on the other hand, is qualitative. “We have yet to find a pure quant approach we’re happy with, so we do good old-fashioned fundamental research,” he says.

The stock-to-analyst ratio in Schroders’ emerging markets team is roughly 15 to 1, according to Mr Conway. “We have analysts in Asia, in Latin America, in the Middle East and in London. The fact that we are on the ground means we can get to know the companies very well,” he says, adding that his team relies entirely on in-house research rather than broker notes.

Assisted by this global network of analysts, the managers in London pick stocks from the countries for which they are responsible. They are, however, subject to an unusually strict risk control system which links their freedom to stray from the benchmark to their track record. Mr Conway refers to this as “alpha-adjusted tracking error”.

Tracking error – the volatility of a fund’s returns relative to the benchmark – is Mr Conway’s preferred measure of risk. He sets a “risk budget” for the fund, which he then breaks down by country. If a given manager is delivering excess returns, he will be allowed to take on more risk by investing further away from the benchmark. If, on the other hand, a manager has hit a period of under performance, he or she will be asked to reduce the tracking error.

The managers are also expected to observe a stop-loss discipline. If one of their stocks falls 15 per cent, they are usually obliged to sell. According to Mr Conway, this is contrary to most managers’ instincts.

“If you’ve got a stock with a price-earnings ratio of 20 and it falls to 15, the standard reaction would be to buy more, because a good stock looks even more attractive at that price,” he says. “At a lot of investment houses you’d be accused of lack of conviction if you sold. Here it’s the opposite. You have to have a very good reason to stick with it.”

As proof that these controls add value, he offers a statistic: over the past year, the stocks that were sold because of the 15 per cent stop-loss went on to fall a further 32 per cent. “That is pain we avoided by accepting we’re not right all the time,” observes Mr Conway.

Both of these control mechanisms are based on the philosophy that the manager is fallible. According to Mr Conway, the very best managers are only right 55-60 per cent of the time. By constructing an elaborate system whereby the manager risk is spread and the downside tightly regulated, he hopes to benefit from positive alpha while neutralising negative alpha.

“In normal circumstances, returns are normally distributed,” he says – referring to the classic symmetrical bell-curve associated with data like exam results or height. “The purpose of our risk controls is to skew returns to the upside. We aim to build up out-performance over time, incrementally. Consistent second-quartile performance will turn into first-quartile performance over the long term.”

This keenly rational approach was introduced in January 2005 after a decade of disastrous underperformance which still drags on the long-term track record. The fund is now in the second quartile over both one and three years to 18 July, according to Morningstar.

“We have outperformed the index in 11 of the 14 quarters since we initiated this approach,” says Mr Conway. But he usually prefers the information ratio – returns divided by tracking error – as a measure of performance. “Investors focus too much on straight returns,” he complains, claiming that in terms of information ratio, his fund ranks in the first quartile over three years.

One of the potential problems with such a decentralised approach to stock selection could be inconsistencies between the various components of the portfolio. To iron these out, or at least to be aware of them, the managers hold regular sector meetings.

“We look at the portfolio from a sector view and see whether it makes sense,” explains Mr Conway. “For example if we’re overweight Russian oil but underweight Brazilian oil we try to find a reason.” He stresses, however, that these meetings do not drive the portfolio, but act as a further safety check.