In a world of their own
Controlling risk while seeking out-performance is always a trade-off between what can be described as the actuarial and the entrepreneurial faces of investment management. Nowhere is this truer than in equity mandates that seek to encompass the global opportunity set.
Global equity mandates offer fund managers the most freedom to exercise their skills, but also leave institutional investors with the greatest complexity when it comes to structuring and allocating mandates to individual managers. While a global equity portfolio is the closest of any equity mandate to the ‘market portfolio’ beloved of followers of modern portfolio theory, does that mean it comes closest to justifying an indexed approach to investment?
A capitalisation weighted approach to global benchmarks would imply more than 50% of assets should be US based even at times when international investors, without a US home bias, may see the market as generally overvalued. Key issues for international investors seeking global mandates are therefore to what extent managers should be following a benchmark, and secondly, what exactly should that benchmark include.
Deviation from benchmark allocations can be made at both the country level, and also at the stock level within countries and across global sectors. Which has the greater effect will depend on whether the stocks in question have higher correlations to their country of listing or their global sector. An increasingly integrated global economy is increasing the importance of sector influences. As T Rowe Price point out: “A newspaper based in an American city does not compete with ones based in Sydney or Helsinki, and yet they all face the same competitive issue: the digital migration of data services, including classified advertisements, to on-line portals.”
Emerging market stocks however, are characterised by having higher correlations to their country of origin than to their global sector. While there is considerable ambiguity as to which countries should be included in the category of emerging markets, country selection within the subclass can create more outperformance than stock selection for some specialist emerging market managers. For global equity managers, the biggest issue is what if any allocation should be made to emerging markets.
The starting point of any strategy for global equities should be what are the prospects for the returns on global equity benchmarks? Dominic Rossi at Threadneedle argues that all other asset classes have been revalued to higher ratings leaving only equities generally despite the re-rating seen in specific segments such as emerging markets, giving a reasonable prospect of low double digit returns over the next few years. Steve Doherty of Aberdeen Asset Management is less sanguine: “History tells us that real returns have been around 6-7%” going on to add that “there is no reason to think that is not a reasonable starting point”. The issue for the institutional investor is should that be accepted as a reasonable return expectation for their own portfolio, perhaps enhanced somewhat through active management within tight tracking errors to the benchmark, or should they seek to gain much higher returns through finding managers willing to take large deviations away from the global benchmark weightings?
The choice of benchmark can of itself, make a significant difference since it can define the investable universe. In particular, whether the benchmark chosen includes emerging markets and how constrained the manager is to the benchmark weighting has been a critical factor in recent performance and may well continue to be so.
Aberdeen themselves recently reduced their weighting to emerging markets from 32% to 22% against an index weighting of 7%. Many managers running global mandates however exclude emerging markets altogether, preferring to run separate emerging market funds, which precludes them from taking an asset allocation bet. When it comes to comparisons of performance for global equities, it is clear that the recent years of exceptional emerging market returns will skew the results of any fund manager who excluded emerging markets for whatever reason.
Given the staggering differences in the performance of value stocks and growth stocks on a global basis over the last six years, with the MSCI World value Index outperforming the MSCI World Growth Index by over 50% to the end of 2005, the potential impact of style bias within an investment approach cannot be ignored.
Rossi sees a core strength of their firm as the thematic research they undertake covering style and sectors that enabled them to have a growth bias through the late 1990s, switching to value in the early 2000s and currently back into a slight growth focus. Understanding and timing style changes is not easy, but Virginia Maisonneuve at Schroders sees two other reasons for the value comeback in addition to the fall-out from the excessive growth valuations built up during the TMT bubble; value benefited from both the improvements in the liquidity environment globally following the 11 September attacks in New York, and its subsequent impact on the global economic cycle and value was also helped by the strong performance of cyclical stocks relative to defensives as cyclicals hold the major share of the value index with 86% of the MSCI World Value Index compared to 61% of the MSCI World Growth Index (as at the end of 2005).
Maisonneuve sees that “looking forward, shifts on the economic and monetary environments as well as relative valuations are pointing to the return of a favourable context for growth performance”, citing an expected slowdown in global economic growth, flat or inverted yield curves, which have often been a good signal for growth-style to move back into favour and thirdly, P/E ratios for growth stocks relative to value have now come back to attractive levels leading her to the view that growth stocks are looking cheap.
However, as Mark Burgess of Credit Suisse Asset Management points out, the danger for value investors is that “if things are no longer cheap, they think they are expensive” and his own firm has both growth and value related funds.
Any global investment process, whether purely quantitative, purely qualitative, or somewhere in between requires a number of discrete steps.
First comes the identification and updating of the universe of possibilities according to some criterion. This may be the complete universe of stocks within an index; it may also include stocks outside the index; or it could represent the sum of regional lists of recommended stocks by teams of analysts. The second step of stock selection requires an assessment of forecasted returns. These need to be combined with a methodology of assessing risk both at the stock level and at the portfolio level which can then be combined to give an idealised optimised portfolio, which then needs to be translated into an actual portfolio through trading.
The different approaches seen in the marketplace can be differentiated by the approach taken to each step in the process, and the relative emphasis given to benchmark and absolute risk. At one extreme, lies a completely indexed approach, which seeks to generate the global portfolio at a minimum cost, but leaving investors with no scope for out-performance.
At the other extreme of highly concentrated portfolios of 50 or so stocks that seek absolute returns with little if any direct influence by indices. In-between, there is also scope for approaches that seek specific objectives such as CSAM’s 60 stock global-high income fund, which has a 4.2% historical yield. There is always the danger that the yield could get cut, but “volatility in dividend yields is low” according to Burgess.
The use of quantitative screening in some form is present in most investment processes, but recent years have seen purely quantitative approaches for global equities gaining much more recognition from institutional investors.
While all such approaches benefit from the ability to analyse thousands of stocks in a comparable manner and even on a daily basis, the underlying philosophies can be very different. GMO for example, divide their global portfolios into a small number of discrete purely value and purely momentum sub-portfolios based on individual valuation models.
Invesco by contrast, in their global quantitative stock selection process, combine four concepts; earnings momentum, looking at the IBES data for earnings revisions; price and volume, examining the long-term and medium term performance relative to other stocks; management action such as the issuance and buying-back of stocks and debt and the market reaction to this; and relative value which looks at mispricing relative to the universe. These factors are combined into one measure, which is used in the portfolio optimisation process. Invesco’s Michael Fraikin sees this approach as distinctive to other quantitative houses which “tend to be small cap biased or value managers or both”.
AXA Rosenberg’s approach is based upon combining the fair value of a stock with its predicted year-ahead earnings to invest in undervalued companies with superior earnings. The fair value is determined using current fundamental data together with industry breakdown information to determine whether a company is over or undervalued relative to other companies in the same sector.
AXA Rosenberg likens this to the break-up analysis conducted by investment bankers in valuing companies. Earnings are estimated from a model that combines fundamental data, analysts’ earnings estimates and buy/sell recommendations with stock price movements. Again, as in the other quant approaches, information on relative rankings of stocks is put into an optimisation process that seeks to maximise return potential within the agreed risk parameters and at an acceptable level of transaction costs.
Optimisation approaches need not necessarily be based on out-performing a global equity index. Invesco for example, have a product optimised against cash where, as Fraikin describes, they “effectively maximise the equity Sharpe ratio while being completely index agnostic. This way, you are getting a much less risky portfolio than with a traditional equity portfolio but at the same time a higher exposure to the stocks we like.”
The global opportunity set may offer a cornucopia of riches, but there are not that many managers who are sufficiently well resourced to consistently differentiate the real gems from the fakes.
As Burgess points out: “People fall down with global by having a weak US team or a weak non-US. The issue in managing global money is how do you get consistency and no systematic bias. The winners are often there because they had a bias, for example, Japan or emerging markets.”
What constitutes a bias and what constitutes superior analysis is however, a question of perceptions. High conviction small stock portfolios, by their nature, can have very large deviations away from benchmark weightings. As well as their stance on emerging markets, Aberdeen in their 50-60 stock portfolio, also have a 16.5% weighting to the US against a benchmark weighting of over 50%; T Rowe Price’s 90 or so stock portfolio has an almost 20% exposure to emerging markets obtained through a bottom-up driven process.
Threadneedle’s Rossi does however, argue that the advantage of global equities is that it is possible to have diversified portfolios without sacrificing alpha through the ability to find uncorrelated themes throughout the world in a way that regional and country managers cannot and so favours 120 or more stocks in their high alpha product.
Using a more traditional qualitative approach does require extensive teams of analysts if the global opportunity set is to be explored in any detail and the use of quantitative screening often provides an essential first screening for many managers.
However, Aberdeen is an example of a firm that seeks to cover the global universe without the use of an initial quant screen. As Stephen Doherty explains: “We look at companies as open books, assessing them on the factors we think important. We say that the scientific part of our process is getting into the company, writing the company note, building on that knowledge base over that period of time. The artistic bit or the, ‘why you would pay us to be active’ is how we construct the portfolio, why we have some stocks at a 3% weighting and others at a 1% weighting.
“Our regional equity managers visit thousands of companies each year and their best picks make up a 300 stock global universe which my team works from. When we do a quality assessment the things we are looking at are; how the business looks; what the prospects are; how long have the management team been there and what motivates them; do we trust them. Then we look at the financials, the strength of the balance sheet, and the use of gearing. We are looking for companies we can understand and are who are running their business for the shareholders”. He adds: “Some have described our investment process as ‘GASP’, growth at a Scottish price.”
The temptation with global portfolios is to have them run as separate regional portfolios by essentially a committee of competing individuals, none of whom may be willing to admit mistakes to their erstwhile colleagues and potential rivals.
For firms such as Schroders, with regional teams spread in 26 countries around the globe, “the key is global perspective with local knowledge”, according to Maisonneuve. “We combine the regional approach with the global sector approach. The global sector specialists are all sitting here in London.
“While the vast majority of the stocks we buy are already covered by our team of analysts, we won’t ignore an opportunity if a stock isn’t one of them. Instead, we will bring our global and local team together to assess the stock. Crucially, our local guys are not incentivised to cover the market, but to find stocks which will outperform.”
The real strengths of such approaches may well be the ability to have a clear line of responsibility for the shape of the overall portfolio that can utilise the recommendations of stock and sector specialists in a completely objective and unbiased manner.
T Rowe Price is another example of a large firm leveraging off its investment in research infrastructure to manage global equities. Rob Gensler, as the portfolio manager, has ultimate responsibility for the portfolio, although it is based on the recommendations of the firms 110 or so globally based sector and regional researchers and fund managers, giving Gensler the ability to take strong bets based on a universe that has already been sifted thoroughly for attractiveness, with all stocks either buy-rated by the analysts or owned by another portfolio manager at the firm.
Another example of investment groups developing global products is Janus that has just launched a global research-based product, with 100 to 150 stocks from all industry sectors and companies, with a minimum market cap of $1bn.
“This will bring to three, our global equity offerings,” explains David Schofield, of Janus International in London. This is a research product based on “best ideas”, but it is not a concentrated focus approach, with a target alpha of around 3%.
It takes a fundamental approach, compared with the international launch in Februaray, when the group’s subsidiary INTECH, introduced a “core risk managed” global equity offering, with its structured quant approach and up to 700 stocks, with alpha of 2.5% plus.
These two different approaches as being very much in parallel as core-type products, he says.
The group’s longer established global product has been the concentrated portfolio, with just 30 to 40 stocks. “This has along-only absolute return emphasis, fitting into the unconstrained bucket, and more of a satellite-type holding.”
The market may be polarising between the two ends of the spectrum, the core and the more aggressive products, Schofield adds. “There could be a squeezing out of the managers in the middle. We want to be ready for this development in the market.”
Threadneedle’s global equity process launched in 2004, replacing its old balanced management approach based on regional sub-portfolios, is another example of a firm seeking to produce global portfolios that are not stock heavy through stapling together a number of regional portfolios. Rossi and his three colleagues running the portfolio see themselves as leveraging off the 120 or so investment professionals all located in London.
While a global equity mandate is in many senses an intuively appealing approach of subcontracting equity asset allocation decisions to fund managers, choosing an enhanced index approach with full coverage of the global opportunity set and a few hundred stocks will give startingly different results to the highly focused portfolios of a hundred or less. Combining managers with different styles, and approaches may well be the most attractive option, but unlike the case of US equities, it is not so obvious how that should be done.