During the eighties, most Dutch pension schemes invested primarily in domestic government bonds and local private placements. In the nineties, pension schemes began to increase their allocation to equities which, due to the strong equity performance in the second half of that decade, led to high coverage ratios. In most cases, the high coverage ratios were used for contribution holidays, lower pension contributions and improvements in pension benefits. However, the sharp fall in equities at the beginning of the decade pushed most pension schemes into a difficult position and triggered the Dutch regulator, PVK, to demand that pension schemes define clear steps on how to improve their solvency.
Although a recovery in share prices in 2003 helped most pension schemes to report better coverage ratios, the realisation that amongst other factors the substantial allocation to equities had jeopardised the pension system, resulted in the need for rethinking the investment strategy. No longer is an equity risk premium of more than 3-5% considered sustainable; no longer can we claim that the risk of short term insolvency can be neglected, nor should we overly test the intergenerational solidarity and the willingness of the sponsor to contribute at all times. With this in mind, an important asset class to consider is corporate bonds. Not only because they offer a better match with liabilities than equities, but also because they offer a risk premium by means of a credit spread, which helps to keep pension schemes affordable.
The European corporate bond market took off after the introduction of the Euro in 1999. Most market participants were required to become acquainted with this new market. Portfolio managers who found themselves responsible for managing credit portfolios, often had a background as duration and currency managers. For them, credit analysis was a new phenomenon. On the other hand, for the issuers, borrowers had to become accustomed with the pricing and other requirements of a public market, which are different than the characteristics they were used to in the private debt market. Initially, there were some doubts as to whether the European corporate bond market would take off at all, but now five years on the market has proven itself and is expected to develop further.
The specific risk associated with corporate bonds is default risk. The issuer must fulfil its obligation of payment of coupons and the principal payment of the loan, both in full and in a timely fashion. The reality is that the issuer could fail this obligation. The investor is paid a credit spread, which reflects the credit status of the borrower.
It is important to note that the return distribution is heavily skewed. In a marking-to-market context the upside of an improving credit status is limited, whilst the market loss of a deteriorating credit status is much larger. In the extreme case of default, the loss could be 100%. In practice this is less, but depends on the outcome of a work-out process. Bondholders can often make use of the market place to sell defaulted bonds to work-out specialists, which can avoid a lengthy and complex legal process.
Graph 2 illustrates a typical price movement caused by a credit event.
This skewed return distribution demands solid diversification. At the portfolio level, risk can be reduced dramatically with a much higher chance of realising the expected return, whilst maintaining a smaller chance of a large loss. Still, a single default in a portfolio with a hundred equally weighted and uncorrelated credits, would lead to a loss of 50 basis points assuming a recovery rate of 50%. This is a substantial loss in a corporate bond portfolio and it takes quite some time before this loss is offset by incremental spread returns. If individual positions are correlated the loss will be even larger. Therefore, the process of diversification should not be taken lightly. A distinction must be made between high and low quality credits. With regards to high quality credits, often specific credit events will trigger downgrades or even defaults, whilst general economic conditions have a larger impact on lower quality credits.
The current Dutch regulatory environment uses an actuarial rate of 4% to discount liabilities. When investors started to allocate funds to corporate bonds, the main attractiveness was the additional yield offered in a low interest rate environment to match the actuarial rate. The new regulatory framework for pension schemes in the Netherlands, Financieel Toetsings Kader (FTK) which is effective from 2006, removes this actuarial rate and instead requires that pension liabilities will be measured at fair value. The question here is what the appropriate discount rate should be. In a so-called liability benchmark portfolio approach, the cash flows of the liabilities are mirrored with the cash flows of risk-free nominal or index-linked bonds. The sum of the prices of the bonds used, determines the current value of the liabilities. Obviously, if risk-free bonds are used, the current value of the liabilities is relatively high. This approach might be too conservative. Yet using relatively high discount rates is not acceptable either, because it would take a rain check on future investment results. In the UK, discount rates based on AA-rated bonds is used, and it is possible that the Netherlands will follow this example. However, a valid option could be to use swap rates instead, as this market is more liquid and offers longer tenors. But, irrespective of what discount rate to use, it is important to realise that under the new regulatory framework, it is not just the additional return which makes corporate bonds interesting, but the risk characteristics too. The characteristics of pension liabilities and corporate bonds are very similar and therefore a much better match than equities. The core holding of a pension scheme could therefore consist of corporate bonds to limit a scheme’s shortfall risk. Obviously, it is very important to minimise the impact of defaults, because this could cause an undesirable deviation from the liabilities.
An often heard comment is that it is not the right time to invest in corporate bonds given the tight credit spreads. This is a valid argument from an asset only point of view. However, under the new regulatory regime, the value of the liabilities is impacted by the same risk factors as corporate bonds. More important is the duration gap between the assets and the liabilities, which is an interest rate risk rather than a credit risk. A similar suggestion is that the new regulatory framework will not be effective until 2006 and the “old” regime is still important. However, we expect the PVK to shortly announce guidelines for the interim period in order to be compliant in 2006.
As mentioned previously, many duration and currency managers were asked to manage credit portfolios despite having no experience in credit analysis. In order to buy time, index tracking was often the chosen strategy, although this involves more risk than often thought:
o A large and increasing percentage of a particular company in a credit index simply means that the corporate issues a lot of debt. An index tracking strategy would therefore mean that the portfolio is geared to companies with much debt.
o The index changes over time. A credit index is dependent on the borrowing requirements of companies. The charts above show how the duration and rating composition has changed over the last six years. As the charts show, this is far from a stable scenario. The index is a moving target with continually changing risk-return characteristics.
o One could argue that the risk of an index tracking strategy is not impacted by default risk, because the default would show up in the benchmark also. Although this is like the story of the three monkeys, it is not true in relative terms either. Strict rules determine when and at what price bonds are included and excluded from the benchmark. More likely than not, these terms are different to the opportunity set of the investor, as a result of which the investor will underperform their benchmark.
So a passive index tracking strategy does not work, but how can one instead implement an active strategy? Some investors have decided to use a top-down approach. The rationale behind this being that a relatively small investment team could generate extra returns by effective sector rotation, limit the need for credit analysis and reduce the risk of default by diversifying as much as possible within each sector. Chart 3 shows how the corporate bond sector composition has changed over the last six years.
Immediately, it becomes clear that sector rotation in practice is hard to achieve. Not only is liquidity an issue, but more importantly, some sectors dominate the universe whilst others consist of just a few issuers which may differ substantially in size and business profile. In many cases a sector position is in fact a single credit holding and does not match the requirements of diversification. Besides, as previously mentioned, lower quality credits are more impacted by general economic conditions, which dilutes the effect of diversification and emphasises the need for credit analysis.
So should we forget about a top-down approach altogether? A top-down approach can generate value, primarily as a result of the tactical asset allocation decisions made between government bonds and corporate bonds. Yet, the most important conclusion is that a bottom-up approach and thorough credit analysis cannot be avoided in managing credit portfolios effectively. One problem is that the real value of a bottom-up approach only becomes clear when it matters. As long as the conditions of the credit markets are stable, most strategies will work. Yet as soon as credit events occur and credit quality deteriorates, it is extremely important to know and monitor all individual credit positions sufficiently. It is at these moments that the significant added value of avoiding credit ‘blow-ups’ becomes apparent.
A diversified credit portfolio consists of many individual positions and it is not uncommon to have 150 names in a portfolio. This is still a small subset of the total market, which indicates that a substantial number of credit analysts is required. It is not just the number of analysts that is important, but the profile of the analysts. Every analyst can be told how to carry out number crunching and to look at specific ratios. This is certainly part of the job, but it is much more important that the analysts have an in-depth understanding of the sectors and companies they follow and understand corporate finance. A vital part of the analysis is the evaluation of bond documentation. Analysts must question which covenants are in place to protect the bondholder and where the bondholder ranks when a credit event takes place. It is important to realise that different bonds of the same issuer can have different covenants. A credit decision is thus not only about the financials, but also about which specific bond issue to invest in and at what price. In addition, experience and the ability to have in-depth discussion with the financial directors of issuers, is an absolute must. Credit analysis should therefore not be seen as a starting position at the beginning of career in investments. Good analysts are career analysts.
The focus of this article is credit risk, but obviously optimising corporate bond portfolios can only be done when all risk factors are taken into account. This is the job of the portfolio manager who relies on credit analysts for the evaluation of credit risk, and on credit dealers to stay in close contact with the market. The portfolio manager’s focus is on diversification, relative value, duration, risk budgeting and other restrictions determined by his clients.
All this illustrates, that managing corporate bond portfolios is very labour intensive and requires a large and experienced team. The costs of such a team are substantial and can only be afforded by large credit investors. However, size also has other important aspects:
o effective diversification can best be realised in a large credit portfolio;
o large investors with career analysts have a better relationship with the financial directors of issuers;
o investment banks contact large investors in the early stages of bringing a new issue to the market to discuss the terms of the bond;
o large investors have the ability to employ fulltime work-out specialists who not only become involved in times of negative credit events, but who also support analysts in investigating suspicious behaviour of issuers;
o large investors are better positioned to take advantage of new developments and to broaden the investment universe to create the most efficient portfolios.
So can anyone do this job as a natural extension of the fixed income portfolio? Yes to an extent, provided we are in a ‘credit friendly’ environment with stable or tightening credit spreads and no credit events. No, if the decision to invest in corporate bonds is a strategic one. In light of the new regulatory framework in the Netherlands, corporate bonds should be considered very seriously as a core investment to match liabilities. However, in investing in corporate bonds there is no shortcut. In order to protect a portfolio from the damaging impact of defaults and achieve proper diversification, a large team of experienced credit analysts, portfolio managers and dealers is a prerequisite. If this is too costly, hiring an external manager with the appropriate skill set is highly recommended. Do not be fooled, a default will happen and what will be the impact on your performance then?
Eduard van Gelderen is Director of Fixed Income at Prudential M&G, Weena 290, 3012 NJ Rotterdam, the Netherlands
Tel: +31 10 2821282