European High Yield - Which Way Now for Default Rates?

1. What has prompted all the negative news flow in the European high yield market during May 2004?

The volatility experienced by the European high yield market in May 2004 could hardly be in greater contrast with the benign conditions and performance that characterised the early part of the year. Total returns, while unsurprisingly not maintaining the stellar momentum of the c. 30% level achieved in 2003, had continued on a positive course, achieving c. 4% YTD up to early May, while the new issue primary pipeline remained buoyant, with c. EUR10.3bn of supply pricing in the four months to April 2004, compared with EUR22.8bn for the whole of 2003.

However, after early May 2004 the tide turned, with secondary spreads widening dramatically (the DJ TRAC-X European HY index widened 40bp during the week to 15/05/04), and new issues either being pulled (eg:- Corus and Brenntag) or needing to be restructured or priced on more favourable terms for investors (eg:- Debenhams and ProSiebenSat.1 (rated ‘BB+’)). So what exactly has happened to transform a recently hot asset class into a relative pariah from the investor’s perspective? Have the underlying credit fundamentals of these issuers changed materially during such a short space of time? The answer, of course, is “no”, although there are factors, at a macroeconomic and geopolitical level, that help to explain the apparent sudden risk aversion of the investor base. While concerns over a range of macro issues have influenced the flight to quality, just as important are the numerous technical factors that have conspired to change the demand/supply dynamic in such a dramatic way.

The principal factors that have prompted the high yield sell-off, representing a combination of macro concerns and technical factors, are as follows:

• Threat of Rising US Interest Rates (Technical/Macro) – With the Fed clearly signalling an imminent upward movement in rates, investors that have borrowed cheaply to invest in higher yielding securities (so-called “carry trades”) have been rapidly unwinding their positions. While rising interest rates will usually be supportive of high yield relative to, say, Treasuries, because of the former’s lower sensitivity to rates, the technical issue of rising funding costs outweighs this theoretical benefit. From a macro perspective, rising rates that reflect an accelerating economy should benefit high yield issuers given the fixed coupon nature of their debt, enabling them to take advantage of economic growth and a potentially lower default environment while being protected from higher borrowing costs. However, on this occasion, the negative technical issue has held sway over the positive macro dynamic.
• Evolving HY Investor Base (Technical) – Another technical factor that has contributed to market volatility has been the presence and activities of new groups of investors in the European market during recent months. New classes of investor, such as hedge funds, traditional investment grade/crossover players and mutual funds, had all sought to take advantage of the market’s 18-month bull run during the latter part of 2003, but many of these players did not have the necessary skills or experience to maintain a long-term investment strategy in the high yield product. Hedge funds, in particular, have been unwinding the leveraged positions that they had accumulated via the carry trades mentioned above. Additionally, US investors in the European market, who had been taking on large positions to benefit both from spread tightening and from currency gains as the euro and sterling strengthened against the US dollar, have also withdrawn funds in recent weeks. Unsurprisingly, this has coincided with some rebalancing of the euro/US dollar exchange rates as the dollar has regained lost ground.
• Rising Oil Prices (Macro) – With world oil prices currently hovering at around USD40 a barrel, having risen from the USD30 average price seen in 2003, this clearly has the capacity to exert a material negative impact on the macroeconomic outlook for global growth prospects. On a simplistic level, a prolonged period of high oil prices has the potential to increase raw material costs for certain high yield issuers, as well as impacting the ability of their key customers to invest at previous or existing levels, thereby potentially impairing growth and squeezing cash flows. However, on balance, although the impact of rising oil prices can be significant for individual issuers, in isolation it does not appear to justify the level of outflows from the European high yield market in recent weeks. Rather, it forms part of a range of concerns that have prompted the flight to quality that has occurred.
• Geopolitical Risk (Macro) – Although a somewhat softer factor than the oil price issue, nonetheless the uncertainty prompted by the ongoing tension in Iraq and concerns over potential future terrorist threats have contributed to the general trend towards risk aversion by investors across the credit spectrum. The influence of this issue is extremely difficult to quantify, although any specific unforeseen event is likely to have a significant negative effect, at least in the short term, on the equity and non- Treasury credit markets.
• Supply Pipeline/Size of the Market (Technical) – Notwithstanding strong recent growth, the European high yield market remains a relative minnow compared with its much larger US cousin. At the end of 2003 Fitch’s European high yield universe stood at EUR74.1bn compared with the EUR563.0bn of outstanding issuance in the US market. Given the European market’s much smaller size, the sudden glut of new issues seen during the early months of this year has resulted in investors being resource-constrained when trying to analyse deals. Historically, this hasn’t been an issue for European investors, given the typically measured timing of new issues, which has traditionally allowed investors to assess deals at a manageable pace. With total new European primary supply of c. EUR6.5bn planned, albeit not executed, for May 2004 alone, however, investors have needed to make decisions about which deals to review and, significantly, have experienced the unique phenomenon of having their orders filled rather than scaled back on allocation. This has been particularly relevant for individual sectors such as cable (see comments below). Inevitably, this temporary change in the supply/demand dynamic has contributed to the spread widening seen in Q2 04.
• Other Factors (Macro/Technical) – These include factors such as the measures being taken by the Chinese authorities to slow down the growth in their rapidly expanding economy and the reduction in Japan’s US dollar currency intervention, both of which should take liquidity and demand out of the world economy, thereby further dampening the hitherto rosy macroeconomic picture.

2. So, in view of the huge sell-off seen in the cable sector, have the prospects for that industry really changed materially?

Not really. Depending on their personal viewpoint, analysts will argue about the outlook for the cable sector, but most would agree that the fundamentals have barely changed in recent weeks. A further crucial consideration here is that cable should not be analysed as a single industry, given that the dynamics can vary greatly from country to country, depending on cultural and other legacy factors. For example, in certain European countries such as Germany, the population is accustomed to accessing television via a cable network, with the relatively low monthly bill for the basic Cable Analogue TV (CATV) service being seen as a utility payment, similar to the way that gas or electricity services are viewed in other countries. In the UK, however, the landscape is somewhat different, with pay TV services being less well established and viewed more as a premium product with an accompanying higher price point. Clearly, these two markets exhibit differing characteristics that influence the business models employed and the competitive environment in which the cable companies operate.

However, whichever model prevails in individual markets, it appears difficult to make a convincing link between any of the macro factors outlined above and the poor performance of recent cable high yield transactions both in the primary and secondary markets. Following the clamour among high yield accounts that resulted in significant oversubscriptions for the NTL (c. 2.5x) and Cablecom (c. 5x) deals in primary in early April 2004, the market’s reception for the TeleColumbus and ONO transactions in late April/early May 2004 could hardly have been more different, with both deals trading down to the low-mid 90s immediately after launch. At first glance, this mixed reception in primary could be viewed as somewhat perverse, given the troubled history of Cablecom and the fact that ONO (ONO Finance plc Senior Notes rated ‘B-‘ by Fitch) is one of the very few cable companies in Europe not to have defaulted or gone through a distressed restructuring in recent years.

That said, any suspicion that the contrasting market receptions for these deals were based primarily on credit fundamentals can be discounted, given the impact of the new supply on the earlier successful NTL and Cablecom deals. Following the TeleColumbus and ONO launches, the NTL and Cablecom bonds followed the sector down to trade in the mid 90s, indicating clearly that the market had become overweight in cable paper; again, a technical rather than a fundamental issue. Despite its 100bp coupon premium, the £375m NTL sterling tranche traded 1 point lower than the EUR225m euros in early May 2004, again reflecting the technical issue of demand/supply for sterling paper relative to the more liquid euro.

A gradual stabilisation in the sector during May 2004, evidenced by some modest spread tightening, did not come close to reversing the losses of that May, reflecting the sector’s misfortune in being exposed to a combination of general high yield market technical factors as well as its own oversupply issue at the same time.

3. But I thought default rates were on the decline?

Yes, generally they are. The table below highlights the improvement seen in the European high yield default rate in 2003, relative to the nadir of 2002, which saw the record high of 25.1%, reflecting the low point of the telecom/cable credit cycle when so many of those issuers defaulted. Despite a definite improvement in the level of diversification within the market, the telecom and cable sectors still represented the majority of default volume in 2003, although at 60.4% this was down materially on the 88.7% seen in 2002.

The principal contributor to European default volume in 2003 was Marconi, the UK telecommunications equipment manufacturer, which finally defaulted after lengthy restructuring discussions with banks and bondholders. The Marconi default represented 50.5% of total European default volume in 2003 and was clearly the reason that telecom and cable continued to represent such a large proportion of the defaulted universe.

Excluding Marconi and the default by cable operator Ekabel Hessen GmbH, the underlying default rate for non-telecom/cable issuers was just 3.4%, down from 5.6% in 2002. The other significant contributor to default volume in 2003 was the utilities sector, with EUR1.1bn, or 25.2% of the total, representing defaults by AES Drax and British Energy. The other sectors to experience defaults in 2003 were transportation and chemicals; full details of the individual defaulted issuers can be found in the table attached as Appendix 1 to this report. Weighted Average Recovery Rates (“WARRs”) continued their year-on-year improvement, achieving 24% for 2003, largely thanks to two principal factors. Firstly, the 36% average recovery rate achieved on the asset-rich utilities defaulters was a strong supporting factor, but more importantly, the 23% recovery achieved for bondholders on the dominant Marconi default reflected their status as pari passu lenders alongside the bankers, a legacy of Marconi’s history as a fallen angel with no senior secured debt in its capital structure. This 23% recovery rate for the telecom section of the defaulted universe compared extremely well with the WARRs achieved on telecom defaulted assets in 2001 and 2002, as highlighted in the following table:

Despite the continuing, albeit reduced, concentration of the telecom and cable sectors in the default statistics, the European high yield market continues to diversify away from these industries, illustrated by the fact that they comprised just 19.2% of outstanding issuance volume in 2003, down from 28.7% in 2002 and 47.5% in 2001. This can only be good for the long term health of the market, improving investors’ risk diversification and reducing the impact of technical sector concentration issues such as those that have affected the cable sector in recent weeks.

4. So is the market oversold?

That depends. From the perspective of a hedge fund investor, or any party not committed to a long term investment strategy in European high yield, the ability to lock-in or protect gains or to minimise losses over a short time horizon is of paramount importance. At times of extreme market volatility that they, ironically, have helped to create, such investors will tend to scramble for the exit in the hope that they are able to crystallise gains before the more traditional buy-and-hold players, and then await future arbitrage opportunities in this or other markets.

Despite the macro concerns highlighted in the answer to question 1 above, the credit fundamentals of many sectors and individual issuers do not appear to have altered so significantly as to warrant the sell-off that we have seen in recent weeks. However, in increasingly sophisticated financial markets, comprising an ever more diversified investor base, all market participants need to recognise that short-term, arbitrage-driven investment strategies will inevitably increase, bringing with them volatility and the potential for exceptional gains and losses for all players. The hiatus that is likely to follow the temporary withdrawal from the European high yield market of some of the shorter term value investors should serve to bring a greater level of stability and certainty to the market in the next few months. Overall, however, on the question of whether the market is oversold, this is only likely to be answered in the affirmative if spreads tighten and the market rebounds, and that shows little sign of happening at the moment.

5. And what about the picture for the future?

Up to May 2004, the picture was relatively rosy. As at the end of April 2004, there had been only a single default recorded in Fitch’s index, and that was more of a technical default, rather than having occurred as a consequence of operational or financial distress, albeit bondholders did suffer a modest loss of principal. The US parent of Midlands Electricity had put the UK business up for sale last year to raise cash and, following a bidding process, E.ON (rated ‘AA- ‘/’F1’), through its UK-based subsidiary Powergen plc (rated ‘A’/’F1’) was successful in acquiring the assets. The £360m and USD500m tranches of bonds issued by intermediate holding company Avon Energy Partners Holding (“AEPH”) received recoveries of 95.8% of par value from the Powergen consideration. This single default equates to a YTD default rate of just 0.6%, albeit this increases to 1.8% on an annualised basis.

If the trend up to May 2004 had continued it would have extended the positive default rate trend, and there are a number of reasons to believe that this was not an unrealistic expectation. Firstly, as can be seen from table 4/graph 1 below, the rating mix within the European high yield universe has evolved in a positive manner in terms of credit quality during the past three years. From the point at June 2001 when over 75% of the universe by volume was rated ‘B+’ or lower, including 24.8% in the ‘C’, ‘CC’ and ‘CCC’ categories, the rating mix at March 2004 reveals a much more robust level of credit quality, with 46.2% in the ‘BB’ range and only 12.6% in the ‘C’/’CC’/’CCC’ range. This improvement partly reflects the nadir of 2002, when many of the lowly rated issues as at June 2001 defaulted. Additionally, however, it reflects the quality of some of the high yield issuance during 2003, when well established issuers, including some fallen angels, accessed the market, increasing its overall credit quality. Examples of such ‘BB’ rated issuers included Vivendi Universal (now rated ‘BBB-‘ by Fitch), Fresenius Finance BV, Valentia Telecommunications Ltd, Ericsson (rated ‘BB-‘) and Vendex. Table 3 suggests that the trend through the first quarter of 2004 appears similarly positive, with the percentage of ‘BB’ rated issuers increasing from the December 2003 level and the ‘C-CCC’ bucket increasing only slightly. This trend is supported by the positive information flowing from the upgrade/downgrade statistics in Q4 2003, when 10.3% of the European high yield universe was upgraded, compared with the 4.6% that was downgraded during the same period.

Interestingly, however, a review of the EUR4.9bn of new issuance that was completed in April 2004 reveals a somewhat different trend in terms of credit quality, with just 21% of new issuance rated in the ‘BB’ range. While only 6% of new deals during April were assigned ‘CCC’ ratings, the remaining 73% that fell within the ‘B’ range reflected the aggressive structuring of transactions that had become increasingly common in an over-heating market; of that 73% ‘B’ range universe, 9% of deals were rated ‘B’, while the other 64% were rated ‘B-‘.

This data supports Fitch’s own experience in the context of the agency’s rating activity of 2003 vintage LBO transactions, which reveals that after two consecutive years of reduction in average deal date leverage ratios, 2003 saw the inflection point, with Total Debt/EBITDA rebounding to 4.8x from the low point of 4.4x seen in 2002. Senior Debt/EBITDA followed suit, matching the 3.8x level seen in 2001 (see table 3). The high levels of liquidity in the private equity and senior debt markets have also contributed to the more aggressive structuring of transactions. In the LBO sector, this means being able to meet the expectations of financial sponsors regarding the optimal capital structure needed to maximise their return on equity. In turn, this translates into higher financial leverage and tests the appetite of bank credit committees to sanction such aggressive financial structuring from a senior debt perspective.

Given the above rating mix trend, and in view of the average time to default for the European high yield market (70%+ of 2003 defaults occurred between two and four years after issuance), Fitch believes that absent another large investment grade default on the scale of Worldcom, default rates for 2004 and 2005 should remain at a similar level to that seen in 2003. Indeed the 2004 rate may even be lower than the 6.4% recorded for 2003. However, in view of the recent trend towards more aggressive structuring as evidenced by the April 2004 rating mix and the trend towards higher leverage multiples, and depending on the evolution of the macroeconomic environment, the agency could envisage an increasing default rate in 2006. Prospects for recovery rates will clearly be dependent on the nature of the defaulted issuers, although the continuing diversification of the market away from telecom/cable would appear likely to support improving potential recovery rates for bondholders.

Graph 1: European Corporate Bond Market Speculative Grade Quarterly Rating Mix Trend

Source: Fitch Ratings

Appendix 1

Note: Rachel Hardee co-authored this report which was first published by Fitch on 27 May 2004.