Q:  What is the history of this fund and how is your fund different from its peers?

The fund was launched in August 2006 and the fund is managed with a value investing strategy. We are unconstrained by the market cap of the company and invest in all companies from microcap to large cap, disregarding the weights in the benchmark indexes. This wasn’t the first benchmark unconstrained strategy that we launched. In 2004, my team launched a global value product in response to investor demand for higher returns, higher tracking error, and higher conviction strategies that have the potential to significantly outperform traditional benchmarks. The development of our benchmark unconstrained strategy was really done with the objective of trying to increase the return potential for investors and the way that we decided to do this is very different from the rest of our peers. In the early part of the last decade, when higher tracking error and higher conviction strategies were being developed by the industry, in the vast majority of cases the road taken by managers was to concentrate their portfolios. Therefore, 200-stock funds became 100-stock funds and managers went down the road of reducing the number of stocks in order to try and create products with the potential to significantly outperform the benchmark. That comes with an obvious higher stock-specific risk as well. So if you’ve got a 30-stock fund and get one of those calls wrong then that can have a really big impact on your portfolio’s performance.

Q:  How has your strategy evolved?

The road that we adopted was the complete opposite to the rest of the industry. Essentially, we went down the road of adding stocks, becoming more diversified in order to create a strategy that we felt could give investors the best chance to outperform the broader market. The way that we did that was to essentially throw away the cap-weighted index as a reference point in building the portfolio. Essentially we started with a blank sheet of paper and just said that if you want to create a strategy with potential for high returns, you need to incorporate a number of features into that strategy. First, we believe you need to invest alongside the favorable tail wind of value, since value over the long-run has been shown to be an outperforming style in equity markets. Secondly, we think you need to exploit breadth, so you need to expand your opportunities beyond the stocks defined in the typical benchmarks. Most of the benchmarks that are used today such as the MSCI World Index, MSCI EAFE Index or MSCI ACWI Index, are essentially focused on large cap stocks and they miss out on mid and small cap opportunities. What we wanted to do was to expand the opportunity set as widely as possible. Therefore, we invest from microcap all the way through to mega-cap stocks. We see a lot of return potential opportunities in small and mid caps which exist outside of the standard benchmarks. Finally, the strategy does not adopt cap weights. We identified over a decade ago a number of issues with cap weighting, which has increasingly been grabbing industry attention, leading to some alternative beta or smart beta approaches. Over ten years ago, we recognized that market cap based weighting was the exact opposite of what value investors are trying to do. Essentially, market cap weighting in an index is an anti-value strategy. By placing increasing weight on stocks that are becoming more expensive, you’re not able to rebalance out of those expensive stocks into cheap stocks, so we decided to throw away cap weighting. Our strategy starts off with more of a cost adjusted equal weight across all of the stocks in our universe. That was the genesis of our benchmark unconstrained product range. We started off with the global strategy in 2004 and launched the global fund ex-U.S. version in 2006, which is currently available here in the US as the Schroder International Multi Cap Value Fund.

Q:  What are the total assets under management in the strategy? 

We have about seven different strategies within our product lineup, manage over $40 billion in assets across all of those strategies, and cover most markets around the world from the developed to emerging markets but not frontier markets. At the moment, we prefer to go down the capitalization spectrum into small and mid caps. There are a lot of opportunities there without needing to take on the additional country risk associated with the frontier markets.

Q:  What kind of value are you searching for and what is your process of discovering investment value?

Essentially, our process involves a ranking of over 12,000 stocks in our global universe. We rank those stocks on a number of different value measures. The value measures are focused on dividend yield, earnings yield, cash flow yield, sales yield and book to price, so there are five broad categories of value measures which we use to define value but the weights that we apply to those will vary by sector. For instance, book-to-price, an asset-based measure has very little relevance for a company in the software industry, where its sales and earnings are generated not from hard assets but from the intellectual property of the employees. Therefore we have different weight schemes applied to those different industries but what we end up with is a global composite rank. We rank stocks in our universe from cheapest to most expensive and only focus on the cheapest third of that rank. When a stock falls out of the cheapest third of that rank, it’s a hard sell and then we reinvest in another opportunity in the cheapest third of the rank. In that way you could essentially describe it as a relative value approach where we’re always invested in the markets and the stocks essentially ranked on these value measures relative to one another. When they’re no longer in the cheapest third of this rank, they’re sold and the assets are reinvested in a new opportunity in the cheapest third.

Q:  What are the distinct components of the investment philosophy? 

There are a couple of distinct components of our philosophy. One is that all of our strategies are based upon company fundamentals. We believe that what a company does and what you pay for it, the quality of the business and its valuation ultimately drive share price returns. So value and quality are the essential building blocks of all of our products. We believe there’s a premium to value investing as well as to quality investing. The premium to value investing is something that’s quite well researched. We like to incorporate value and quality into our strategies to varying degrees because they are also a natural hedge for each other, so quality tends to work when value struggles. That’s the first aspect, and a very important aspect of our philosophy is that we look at company fundamentals through our valuation and quality measures. Second, we use quantitative tools really as a very scalable way to invest across a broad universe, so we’re attracted to quant for the efficiency it provides the team to filter through a universe of over 12,000 companies. It also results in a very disciplined approach to rebalancing stocks within your fund where you don’t become emotionally attached to individual companies and you can be a lot more objective in your approach to rebalancing. The quant tools are just there as a means to an end. Another important part of our philosophy is that ultimately the risk management and portfolio construction role have to be performed by a portfolio manager. In our view, you can’t outsource that to an optimizer or a quant model because essentially those quant models are backward looking and the risks in portfolios are better assessed by a portfolio manager who is there trading every day and can actually see what sort of risks are building in a portfolio. In that way we believe that we can avoid some of the black box risk the traditional quant managers have experienced historically. That sums up the three different components. Our products are based on fundamentals of value and quality which we believe will outperform over time and that value and quality are best implemented across a broad universe using quantitative tools but with portfolio manager implementation.

Q:  What are the key steps in your investment process? 

There are three key steps to the process. The first step which I’ve alluded to already is the creation of a global value rank, ranking each of the stocks in our 12,000 stock universe from cheapest to most expensive and our target universe being the cheapest third of that rank. Now, if we stopped there, at that first stage, what we would end up with is a cheap portfolio but it would be very low quality. One of the issues when you create a value fund is that the undesirable characteristic you get is usually distress; companies that are cheap for a reason. They might be unprofitable, might have very volatile sales and earnings histories, and might have excessive debt and the inability to pay interest on that debt burden. Typically the problem with a more naive value strategy is that it’s low quality. Step two of our process, which is the stock selection aspect, focuses our attention on those stocks with higher quality within the cheapest third of our universe. What we want to do in a sense is to try and avoid those value traps and just focus the portfolio on the highest quality names within the cheap universe. Step three of the process of the portfolio construction is that we want to be as diversified as possible across region, sector, style and stocks. The portfolio is built bottom-up based on the value and quality opportunities and the role of the portfolio management team is to make sure that there are no disproportionate concentrations building in the fund. If a product is benchmark unconstrained, risk is no longer defined relative to an index, risk is about concentrated thematic bets. To give you an example, in 2008 when Lehman Brothers went bust, banks all became cheap at the same time and dominated the cheapest third of our value rank and all of a sudden a international equity fund could become a banking fund. That’s because the fund would be loading up on those cheap companies and that would essentially become a one sector, one bet portfolio. What we’re trying to avoid is those types of concentrations while remaining as diversified as possible. We then believe that we can extract the premium to value and quality investing on a much more consistent basis across time. Essentially, what we’re doing in step two is looking for characteristics that companies have had when they’ve outperformed their peer group historically; essentially what we use is what we describe as decision trees which basically look for those relevant combinations of factors associated with a high quality stock. I think the decision trees are a very interesting aspect of what we do because I think many of your more traditional quantitative approaches have been very linear in the way that they select stocks. Essentially, when they rank stocks they have got the value factors, growth factors, momentum factors and you fit linear regressions in order to try and predict share price outperformance. The issue with that is you can never really identify the exception to the rule. An example I find most people can understand quite well is if you were examining the impact of leverage on a company and whether high leverage was good or bad, the answer to that of course is it depends. Because, if you combine leverage with a company that is well run and has expanding margins and growing sales of course the use of debt in that situation is probably quite sound. In the reverse situation, if you combine leverage with a company with declining margins and declining sales, that’s terrible and so high leverage only make sense depending on what it is combined with. But how do you model leverage in the traditional quant framework which is so linear in its approach? You can’t model it effectively and in real life many of the terms which we examine on the financial statements of companies, valuation terms like dividend yield, only make sense depending on what they’re combined with. Is high dividend yield basically the result of a company which is very cyclical and has currently got very depressed share price? Or is it a high dividend yield from dividend aristocrat, a company that is very strong and dividend coverage ratios have been growing, dividends that are consistent for the last 25 years? It is a very different answer. For most of your traditional fundamental investors this type of thinking is quite ordinary. For quant managers, I think it’s quite foreign. But I think the decision trees which we use in step two are a very good way of emulating the way the fundamental investors think, and essentially we use these decision trees as a quantitative tool to scale up that checklist of items in order to apply it across every single stock across our 12,000 stock universe in order to look for all of those companies with those particular attributes. In a sense what we are looking for in our strategies is essentially these desirable stock attributes. So we don’t just want a traditional value fund, we want a value fund that is full of companies that have growing margins, growing sales, potentially high or low leverage depending on what it is combined with. So what we are looking for is this checklist of attributes and then we will buy two or three dozen stocks with those specific attributes. We don’t really care about the individual company name. In our strategies if two or three dozen companies all have the same desirable attributes in a sense we can regard them all as largely fungible for each other. We want to buy them all if they’ve got a very, say, defensive flavor of value and by buying two or three dozen of those opportunities we minimize the stock specific risk and we can become more confident that we’re getting exposure to that defensive flavor of value.

Q:  How do decision trees help in stock selection? 

We do focus on the cheapest third of the universe and so essentially within that we’re looking for the companies which we think have the best fundamental characteristics. We’re not going to find the highest quality names because oftentimes they will actually be quite expensive. There will be a lot of stocks, say in consumer staples or healthcare that essentially won’t fall within the cheapest third of our rank and so would not be considered for inclusion in our value strategy. In focusing on the cheapest third, the decision trees are basically the second cut, saying what are the stocks with the strongest fundamentals within that cheap universe and so it may not necessarily be that on an absolute basis their margins are the strongest in the industry or that the sales growth has been the strongest in that particular industry but just within the cheapest third they would be.

Q:  Why do you like to hold such a large number of stocks? 

We typically hold more than 500 stocks across all of our benchmark unconstrained strategies and so diversification is actually one of the important characteristics of what we do, not just from risk management. But actually holding a large number of stocks is an important aspect in how we try to generate returns in our strategies. And this is something that I find for most people is quite hard to understand because it really goes against the conventional wisdom that you need to concentrate your portfolio to generate strong outperformance. If anything, the International Multi Cap Value Fund has demonstrated the exception to the rule that you can be very diversified and still generate strong and consistent performance, and we’re outperforming our peer group, nearly all of whom are concentrated. So essentially what we’ve demonstrated with the fund is that number of holdings is not necessarily a good indicator of the return potential from a strategy. What we are more focused is on building a fund with high active share, which is essentially a measure of the active bets a manager has relative to an index, and most of the traditional stock pickers would create funds with high active share with just a few concentrated bets, whereas we prefer to build up a high active share from hundreds and hundreds of very small bets. In a sense, the analogy that I draw is one to the gaming industry. How does a casino make money consistently over the long-term? It’s not from the high roller who comes in and bets the house on red or black on the roulette table; it is actually through tens of thousands of individual patrons who are betting five dollar chips on the roulette wheel. That’s the way that the casino makes money; by tilting the odds slightly in their favor, they make more consistent returns the more bets that are placed not less, and essentially we’re trying to do the same thing. Using our decision trees in step two, the stock selection, we believe that we can tilt the odds in our favor in terms of picking stocks that will outperform their peers. Then, the road to more consistent returns is essentially to add stocks, not to reduce the numbers. This goes against the grain for many people and we end up with a portfolio with many holdings and 1,200 stocks. Obviously the fund is no longer about stock specific risk so if you get one call right or wrong on a company in isolation that individual call will have a very small impact on the portfolio. Essentially what we’re trying to do is build brick by brick, with many hundreds of bricks, in the hope that the overall fund then ends up becoming cheaper and higher quality than the index. When you look at the value characteristics and the quality characteristics, you’ll find that our 1,200 stock portfolio is cheaper and higher quality on almost every measure, and so what we end up with what, in our view, is an equity strategy that represents a superior equity beta for clients to hold which is more diversified than the index.

Q:  Can you describe the organization of your research team? 

Within the research team, we’re team of 29 people, based in London, Sydney and New York. The head of the team is Justin Abercrombie. He joined Schroders in 1996 and he has now built up a team where there are four senior portfolio managers, including himself, and another ten in the research department and three other individuals in the portfolio implementation. They are focused on creating a set of financials for all companies within our universe. So we create a pro forma or a standardized template, earnings statement, cash flow statement and balance sheet for all the companies within our universe, more than 12,000 stocks, in order to allow us to calculate those financial ratios, the value ratios and those quality indicators which go into our investment process. That represents one aspect of the research, making sure that the accounting treatment or the treatment of the underlying data is consistent between countries and between industries. The second part of the research process is essentially looking at refining the decision trees, so coming up with those check lists of attributes that we believe that high-quality stocks have to meet and a lot of research goes into that. Then, about 50% of the research agenda is just driven by what the senior portfolio managers see evolving in markets and our portfolios day-to-day. For example I mentioned the 2008 example when Lehman went bust, we started a year before that when sub-prime was blowing up that essentially kicked off a research project looking at banks and our exposure to banks and trying to become more discriminating within the banking sector. More recently, a year ago with the prospects of rising interest rates that kicked off another project, basically what Justin was worrying about and the senior portfolio managers was the impact of rising rates on interest rate sensitive areas of the market. For example life insurers with very long-tail liabilities are especially sensitive to rising interest rates, and so that sparked a research project looking at how we could become more discriminating among life insurance companies.

Q:  How do you measure the effectiveness of the decision tree process? 

We can essentially measure the effectiveness of the trees by looking at other performance attributions of our funds; we can decompose that into the sector contributions. The year before last information technology was the detractor for our strategy and so that sparked a research project to look at the decision trees for the technologies sector and how we need to revise that. The performance attribution allows us to measure or assess the performance of the trees and whether they’re working or not and then that will lead to a potential revision or some project to refine those trees. We’ve been using the trees for a long time; I mentioned that our team started off with an enhanced index strategy in January 2000, we’ve now got a 14-year track record. In that enhanced index strategy it is very benchmark aware but focuses on similar attributes to our benchmark agnostic strategies. In our global index strategy the only driver of returns are the stock over- and under-weights which are determined from the trees. Over its 14-year track record we’ve only had two years of underperformance and a batting average of 67%, so in two out of three months the trees have essentially been resulting in a portfolio that outperformed the index. So it has been a very consistent stock selection methodology and we’re constantly reviewing it using our performance attribution. We will review trees where the batting averages have started to decline.

Q:  How do you adjust the allocation in the fund? 

Risk is about absolute concentrations, so typically we try to make sure that there are no large exposures building in the funds with upper limits imposed at the sector level. We have no more than 30% in one sector and no more than 75 basis points in one stock. But the exposures at the country and sector level are monitored every day and we find that we effectively make it harder for our traders to continue to load up in that particular theme long before we hit any of the upper limits. For instance, banks. When Lehman went bust in 2008, the temptation from a value strategy would be to load up on cheapness so to buy more and more bank stocks because they were also cheap. Our response in that environment was essentially to raise the quality hurdle so as the fund is becoming more concentrated to a theme, in this case banks, we essentially raised the quality criteria that a bank needed to meet in order to be included in the fund. By raising that quality hurdle, we make it more difficult for the fund to continue to load up on that cheapness within the market.

Q:  What is your official benchmark, if you have any? 

We report performance against half a dozen different benchmarks. Within the mutual fund we report both against MSCI ACWI ex-US as well as against the MSCI EAFE index. But bear in mind that our emerging market exposure in the strategy has averaged between 10% and 20% since inception, with the long-term average around 15%, so that would probably place us closer to an MSCI ACWI ex-US benchmark than EAFE index. Some of our clients also like to compare the fund against the investable market index versions of those benchmarks. We don’t really care which benchmark clients measure us against because the benchmark largely doesn’t come into consideration when we’re building our fund. The starting stock weights essentially are a cost adjusted equal weight and then we’re just looking to focus on the highest quality names within our value universe and maintain diversification. Our approach is largely agnostic towards benchmark, so we have clients who monitor us against half a dozen different benchmarks.

Q:  How do you define risk and how do you manage different kinds of risk? 

Risk is very difficult term to define. A lot of people just look at, for instance, volatility or tracking error in benchmark relative space. We don’t believe you can really boil risk down to just one number. Risk can come in many different guises and, for us, risk is more about concentration – avoiding a big bet on just one thing within the portfolios. The role of the portfolio management team is to understand where all those bets are being taken and making sure that they don’t constitute a large proportion of the fund. Every day we actually run a risk report about 20 pages long which essentially x-rays the portfolio along a number of different dimensions. We look at industry exposures, country exposures, currency exposures, our active share in different regions and sectors of the world, the contribution to the fund’s total active share, and style skyline. We look at how our portfolio tilts on various value, quality, growth, momentum, and other risk indicators relative to the broader market, and we also conduct a scenario analysis. We look at the risk measures of the portfolio - your volatility, tracking error, and beta of the fund - in multiple different environments. In a stressed environment when you know the correlations between stocks go to one, you can have a very different measure of risk than in a more benign environment. How would the portfolio perform in rising markets versus falling markets? Or in value versus growth environments? We come up with a range of different estimates for beta, volatility, and tracking error which downplays the emphasis on a single risk number which may not be appropriate if market conditions change. By conducting a scenario analysis we believe can get a good sense of when we’re likely to outperform or underperform. Then we try to maintain a strategy which we think will loosely fit across most environments rather than being overly specified or optimized for just one type of market. We want a product that will be more all-weather across a lot of different regimes. That’s the purpose of the scenario analysis as well, to make sure that there are no nasty unconstrained edges building in the fund. But ultimately the way the portfolios are shaped comes down to a subjective call so this is what I mentioned at the very beginning, the portfolio implementation is done by portfolio managers. We don’t believe you can outsource portfolio construction to an optimizer and risk model for the reasons I just outlined in that there are lots of different environments and risk models change between stressed and benign markets. So ultimately it comes down to the experience and intuition of your portfolio management team as to where you want to trade off risk and reward in the fund and that will become a subjective investment call.