The passive approach to bonds

Related Categories

In the mid-1990s, we were approached by a client and asked if our long experience of running passive global equity accounts could be applied to a fixed income portfolio. Underperformance by many active global and US bond managers in standard competitive universes was well documented and the prospect of paying considerably lower fees for a passive strategy provided a strong cost incentive.
Having spotted the water, the horse swiftly started to drink. Today, our passive bond business worldwide amounts to almost e26bn and covers an array of bonds from Treasury inflation-protected securities to developing debt.
One could be forgiven for thinking that running such portfolios is straightforward, but it has not been without its challenges for European fund managers in recent years. Economic and monetary union presented obvious tests for investors, not only in guessing the ‘ins’ and ‘outs’, conversion rates and approximate capitalisation weights of euro area or global indices, but largely in the sheer scale of the redenominalisation effort. For passive managers focused on minimising risk to the respective benchmark and likely owning a higher proportion of index securities than more active counterparts the work involved was multiplied.
As European currencies became more and more perfectly correlated and European yields converged, commentators soon pointed to the future prominence of credit sectors. The arguments were based initially on the lure of augmented yield and some opportunity for diversification. In the US, it was hoped that the growth of credit sectors might alleviate a perceived shortfall between demand for and supply of government bonds resulting from government fiscal rectitude.
The forecasters were correct and until recently, the size and interest in non-government markets within Europe and the US have grown at an impressive pace. Following the moves by many European plan sponsors to abandon national government benchmarks and adopt new euro-area ones, there has been a second-generation move to so-called ‘broad investment grade’ (BIG) or ‘aggregate’ indices. These indices incorporate much broader categories of debt in addition to domestic governments, such as corporate bonds, regional governments, agency paper supranational issuers, mortgage-backed securities and asset-backed securities. All are included with the proviso that the issuer is at least rated investment grade (Baa3 or better by Moodys or BBB– or better by Standard & Poors). Managing portfolios using such a diverse array of bonds has made the manager’s remit far trickier in several ways, even if the chosen strategy is a passive one.

The first problem is obvious. There are a far greater number of securities in the new indices. The Salomon Smith Barney EuroBIG currently covers 1,077 separate securities compared to 274 on the EMU Government Bond Index (EGBI). On the Lehman Euro Aggregate there are 3,911 in total ,of which only 339 make up the treasury component. The typically smaller size of many European funds compared with much larger US schemes makes full replication an inefficient method of passively managing portfolios against these benchmarks. However, the lower correlation between bonds in certain sectors of an aggregate index means the security selection process used to achieve index returns must be that much more considered than on a government-only portfolio. As a first step aggregate indices present the manager with many more index characteristics to target. No longer is it a simple question of replicating country weights, duration, cash flow distribution and yield. Additionally, there is a whole range of sector and rating quality criteria to meet too.
Optimisation software promises to make the job easier, but exclusive reliance on such systems is not without its pitfalls. For example, they often rely on historical data to estimate a bond’s risk based on its sensitivity to movements in the treasury curve and, for corporate bonds, an additional estimate of the sensitivity to general movements in ‘spreads’. That is, the movement in corporate yields relative to treasury yields. An average spread curve is estimated, sometimes using the swap curve as a proxy, and the bond’s rating is used to scale up if of lesser quality than the swap curve (technically AA-rated) the sensitivity to a general spread movement. Conversely, this is scaled down if the bond is more highly rated. The problem is that this estimate is only good on average, if at all. There is little in many of these models to capture the behaviour of individual issuers and even less to capture the future or potential behaviour of these bonds. Right now the Salomon EuroBIG carries an average rating of AAA–. With euro-area economic activity slowing and consolidation still prevalent, the probability that individual issuers can retain their high ratings is diminishing. Diversification of corporate names can limit damage but a larger exposure to one issuer relative to the benchmark can still lead to significant underperformance, even in investment grade territory. Experiences with Xerox late last year are a prime example. At the end of September, the bonds were trading at 95, but as the bond market learned the severity of the company’s cash-flow problem the bonds plummeted and now trade below 64. The situation was not helped by the patchy liquidity that has developed in global debt markets over the last couple of years. The Long-Term Credit Management (LTCM) affair, a general increase in spreads on corporate paper and consolidation in the banking sector have all played their part in reducing market liquidity in non-government bonds.
The Xerox situation was largely unforeseen by rating agencies and the price volatility was definitely not forecast by standard risk software sitting on most fund manager’s desks. As a result, naïvely optimised portfolios designed to track the index might easily have done anything but.

Where the deficiencies lie in existing tools is apparent. Managers in the US are already some way ahead of their European counterparts in researching and implementing greater sophistication to help overcome stories like that above. Newer corporate bond risk and pricing models have many more inputs and aim to capture the differential balance sheet risks between various companies and industries. There are, however, a couple of impediments preventing these models making the crossover to the European bond market. Firstly, European accountancy is not yet on a level playing field and thus it is difficult to develop models with factors to compare the cash flow and balance sheet health of European companies, when treatment can still vary across the region. Secondly, despite growth the European credit markets do not yet rival the US in terms of the numbers of issuers in all of the market sectors. With only a limited number of issuers in certain sectors, there are probably still too few datapoints to try and significantly increase the number of inputs into the models, for statistical reasons.
Absent any developments in existing optimisation software for European managers, the only real hope of avoiding pitfalls such as this is for them to combine stringent risk management with old-fashioned credit analysis to construct portfolios to track these new benchmarks and therein lies the irony. Those quantitative passive managers without credit analysis skills could well come unstuck in the current environment. It is only those managers who can combine experience and resources on the credit side, coupled with the more usual quantitative tools and analytics who are likely to be able to deliver the low level of tracking error expected on a passive fixed income portfolio going forward. But experienced credit analysts do not come cheap, especially in Europe. So, charges for passive management against these new aggregate benchmarks could also be under some upward pressure.
Rachel Roberts is a fixed income engineer with State Street Global Advisors in London

Have your say

You must sign in to make a comment


Your first step in manager selection...

IPE Quest is a manager search facility that connects institutional investors and asset managers.

  • QN-2444

    Asset class: Trade Finance.
    Asset region: Global.
    Size: USD 10m.
    Closing date: 2018-06-25.

  • QN-2452

    Asset class: Hard Currency.
    Asset region: Emerging markets.
    Size: USD 300 mil..
    Closing date: 2018-06-22.

Begin Your Search Here