Neptune Investment Management manages £6.5bn (€7.9bn) in almost 30 different equity strategies, pursuing a sector-driven philosophy. Martin Steward discussed the process with Douglas McDowell, head of client investment strategies

MS: Why take sectors as your starting point?
DMcD: "Your portfolio is a collection of businesses, and understanding the dynamics of those businesses is the only logical way to go about deciding how to approach the real investable world. I struggle to think of examples of companies that have been among the 50 best in the world - irrespective of how well they have been managed - if the business they are involved in is a poor one. Increasingly, where a company is domiciled - and certainly where it is listed - tells you nothing about its business. A strategist can say, ‘I think the US is cheap, go and find some things to buy'; or a benchmark might have a set allocation to the US. But then you select stocks without any consideration as to whether those US businesses are cheap or expensive relative to the same kind of businesses elsewhere. While any global portfolio would be expected to have some sort of regional spread - and ours must have exposure to at least seven countries - we have no regional risk budgets: big regional overweights and underweights fall out exclusively from trying to create a distribution of stocks that we feel will make money, given our internal constraints on sector and stock limits and risk measures."

MS: Is there not a danger of missing the strong winners from weak sectors?
DMcD: "Although it is generally true that we wouldn't buy stocks in industries whose outlook is poor, there are always exceptions and we try very hard not to have any preconceived notions about what portfolios should look like. Our process doesn't say, ‘I know I don't want to have any businesses in this industry'. In fact you can't really do that, because you need to research businesses that do not conform easily to MSCI sub-sectors - particularly in industrials, where the business you want to buy often has other bits and pieces bolted onto it. Similarly, we offer a Russia fund, for example, and the sectoral biases in the MSCI Russia index do not reflect the real economy - it's heavily weighted to oil and gas and underweight consumer staples, where we expect a lot of growth. This not a problem in the US, Europe, UK or Japan - but you have to become more nuanced relative to your idealised sector matrix as you move away from those bigger markets.

"So our starting point is sector analysis, understanding the operational dynamics of the 30-plus MSCI sub-sectors over the next 3-5 years - but that analysis is based on bottom-up research into the small number of major players that dominate each industry group. This framework allows us to filter lots and lots of information without having hundreds of people doing commoditised research, most of which has zero value - important if you run concentrated portfolios of 30-50 stocks, as we do. Once we have identified our favoured industry groups, we apply a standard discounted cash flow analysis based on bull, bear and neutral scenarios to come up with a blended value for each stock in those groups. If we think there is at least 25% upside to current valuation a stock is selected to form part of the distribution of our global sector matrix, and also the buy list for the portfolio manager. Portfolio stock selection is down to the individual fund manager, and much of it is down to how it will fit into their particular portfolio, or how they feel about the investment case: if there are four pharma stocks on the buy list, you can choose to have your overweight in one of those stocks or all four. But nothing can be bought unless it's on the buy list, and unless there is a very compelling reason to position away from the global sector matrix, portfolio managers also have to follow that. We estimate that performance attribution is around 60% sector and 40% stock-specific."

MS: Is it possible to run a sector-driven strategy that is not in some way ‘thematic'?
DMcD: "While it's always easier to communicate what we do through thematic ideas we try very hard not to. We like emerging market consumption stories and that's undoubtedly in the back of our researchers' minds, but what we don't do is say, ‘The big growth story for the next few years is going to be emerging market consumption, go and find things to buy to exploit that'. In terms of style, valuation is very much the last stage of the process - there always has to be a business case for being overweight a sector and we don't buy companies or sectors simply because they are unloved and undervalued. And while the portfolio will exhibit pro-growth or defensive characteristics through the cycle, what we don't do is say, ‘We're all pretty bearish, let's find things to own to create a defensive portfolio'. Overall, the sector research tends to reinforce our innate top-down views."

MS: Sectors aren't always the main driver of risk. In Europe country risk has become much more important as a driver of risk relative to sectors. How do you manage that?
DMcD:: "There will be times - periods of excessive risk aversion or risk appetite, for example - when the fact that a company happens to be Chinese will be more important than the business it actually conducts. But unless we can convince ourselves that that dynamic is going to be a significant long-term shift in the way investors value companies, we will ride through it. We have worked against the idea of geographic silos for nine years, and over that time we believe that the development of international capital markets has moved the world in our favour rather than the other way around."