Indices fail to measure up
Global bond indices have serious flaws including a low weighting in emerging markets
• Capitalisation-weighted bond indices do not represent minimum risk portfolios and are unsuitable for performance comparisons.
• Existing benchmarks created by investment banks reflect their own trading activities rather than an objective assessment of investment opportunities.
• Illiquidity in bond markets exacerbates the flaws arising from using them.
In the world of equities, the development of capitalisation-weighted indices has reduced the complexity of risk assessment and simplified portfolio management through index funds. At first sight, this approach could be applied to fixed interest assets. Indeed, there is a proliferation of bond indices competing for attention. There are, however, differences in both the underpinnings of bond indices as well as their implementation.
In the emerging market context, there are arguments that the current provision and usage of emerging market debt (EMD) indices represents market failure. This affects investors and issuers through misallocations of capital, inadequate diversification, increased financial instability and a misrepresentation of active investment decisions in passive benchmarks.
A well-structured bond portfolio should be diversified across a range of credits with no undue concentration. Index providers are faced with the choice of coverage versus liquidity. Liquid indices with a narrow coverage are attractive to index-providers as they can form the basis of lucrative derivative products. However, a bond fund manager needs to access the whole market and his performance needs to be measured against that objective and risks controlled relative to the total market. This is difficult to solve.
A fundamental difference between equity and debt indices is that the amount of an issuer’s debt represented in a bond index is not sensitive to the pricing of its debt. The credit spread forms only a small part of yield except in high yield or distressed debt portfolios. As a result, there is the perverse result that the weaker an entity becomes financially through the issuance of more debt, the more an index will weight that entity, even though its attractiveness is reduced. In contrast, the weightings of unsuccessful companies with decreasing equity prices will automatically reduce in a capitalisation-weighted equity index.
Global bond indices typically only have an allocation of 2%-4% of EMD despite the fact that they represent a larger part of the universe
Conversely, naively allocating capital to sovereign debt according to market capitalisation weights would increase financial instability. In the peculiar world of developed market debt markets driven by quantitative easing (QE) this is a problem. Investors are continuing to invest in low yielding developed market sovereign debt driven by financial repression while under-allocating to higher yielding EMD that is far less leveraged.
Global bond indices typically only have an allocation of 2-4% of EMD despite the fact that they represent a larger part of the universe. One recent development has been the launch of the JP Morgan GBI Aggregate Diversified index, which has a 20% allocation to emerging market sovereign debt with 94% of the index being investment grade. While this is an improvement, it only represents a benchmark for a small amount of capital.
Any market cap weighted bond index cannot be viewed as a minimum risk portfolio. In the case of emerging markets, there is an additional flaw in that only 9% of all bonds are included in the main benchmark indices and most of the excluded bonds are not un-investable, says Jan Dehn, head of research at Ashmore. A serious problem with existing EMD indices is that they are produced by investment banks whose business activities revolve around the developed world, and therefore miss out on smaller emerging markets such as Kenya. The provision of accurate global EMD indices represents a public good for which profit-maximising entities are ill-equipped to satisfy. As a result, to judge active managers by their performance relative to a bond index is misleading, since the EMD indices represent active stock selection decisions.
In the past, the use of flawed indices as benchmarks was not a problem as active managers could rectify flaws. Since the global financial crisis, however, liquidity in bond markets has become variable as banks have reduced the amount of capital allocated to market-making. Salman Ahmed, chief investment strategist at Lombard Odier Investment Managers (LOIM), argues that this exacerbates flaws in index construction for active managers who benchmark hug indices as exiting positions becomes costly.
For investors, indices have two purposes. First, they are tools to measure manager performance. Secondly, they can help understand where to invest through the construction of what the investment institution views as a strategic “minimum risk” portfolio.
The problem with benchmarks for performance comparisons is that define the universe in which the manager invests. This is the case when they are index-tracking. As a result, markets not included in indices are not subject to investment flows. The JP Morgan EMBI GD index has 65 countries within it compared with 30 a decade ago. But there are 165 EM countries and most of them have local markets. Investors are therefore investing in an asset class where there is far more diversification than they can access.
Investors should also consider what represents the best long-term strategic benchmark given the difficulties arising from capitalisation weighted benchmarks for bonds. Looking at economic fundamentals is one approach. LOIM has developed a smart-beta type benchmark based on fundamentals rather than market cap, which incorporates factors such as GDP and leverage. Purely price-based approaches are also possible although this relies on forecasts of returns. An alternative that can make sense is to consider a risk parity approach which sets risk targets and weights assets according to those targets.
There is no single optimal weighting of different sectors within EMD. Moreover, the marketplace itself is evolving, which, given the trend of further reductions in the market share of hard currency sovereign debt, suggests that current allocations should be discounted. While existing indices are poor at covering the universe and establishing minimum risk benchmarks, that is a market failure. It is the responsibility of governments and supranational institutions to address these failures. Dehn says that they have ignored the problems associated with EMD indices. For little cost, he argues, they could either produce indices themselves or subsidise other providers to provide coverage where they do not trade. Meanwhile, institutional investors should be wary about sticking too closely to any EMD index.