In 2001, Europe’s state-owned or formerly state-owned telephone companies were in trouble. Having overspent in the late 1990s on projects such as third-generation mobile phone technology – condemned as a white elephant – and buying or building internet service providers – most no longer operational – the incumbents were encumbered with debt as the new millennium was dawning.
Despite, or indeed because of its parlous condition, the industry desperately needed capital. But at the same time bond investors felt the rating agencies were being slow to respond to the new reality of heavily debt-burdened telecoms and so were not overly eager to lend to an industry where downgrades were simply a matter of time. Ever resourceful, investment bankers dreamt up a way that seemed to offer a fair solution to issuers and investors alike. The step-up coupon bond was born.
Step-up coupons force issuers to raise their interest payments if and when their credit ratings fall. The idea was seized upon by borrowers and institutions and, between 2001 and 2002, telecom issuers were more or less obliged to use step language if they wanted to access the bond market.
More than €50 billion of step coupon bonds have been issued in that currency alone. In addition, bonds worth tens of billions of euros have been launched in US dollars and sterling. However a problem has emerged: while many thought it was difficult to place a value on the extent to which a step-up would mitigate the effects of a downgrade, it is now becoming obvious that it is nigh-on impossible to understand how a step-down in coupon will diminish the benefits of an upgrade.
“In 2001, step coupon bonds were perceived to be the only way that telecom companies could issue debt. Everyone thought it was a great idea at the time,” says Simon Surtees, a credit analyst at Gartmore Investment Management.
Times change however. Since 2001, telecom companies have made a dramatic recovery. Although there were rating downgrades in the early days, many are now experiencing rating upgrades. This means the coupon goes down, not up. “Two years ago, it was all about step-ups. Now the tide has turned,” says Roger Appleyard, telecoms credit analyst at ABN Amro.
The implications of this sea change are manifold. Investors are now having to cope with actual or potential coupon reductions on some of their highest-yielding paper. Certain issuers, such as KPN with its euro ’06 and sterling ’08 bonds, have already cut coupons. Others may be about to do so. “What was once an asset has become more of an economic liability,” says Stephan Michel, telecoms credit analyst at Barclays Capital.
The issue is complicated enough in itself, since nobody can accurately second-guess the rating agencies: step-up coupons are usually referenced to the ratings assigned by Moody’s and Standard & Poor’s, so nobody knows precisely what upgrades are going to occur and when they might take place. In addition, the step language is different in virtually every one of the 40-plus bonds issued between 2001 and 2003.
Some bonds, such as those of France Télécom, reduce the coupon by 25 basis points every time each of the rating agencies upgrades the company by one notch. Some, such as those of Deutsche Telekom – presently rated BBB+/Baa3 – reduce the coupon by 50 basis points if the credit moves to single-A. Some companies change their coupon as soon as a rating change occurs. Others only make changes after the next coupon payment, which may take place once or twice a year. And different language can apply to different issues launched by the same company.
France Télécom, for example, has a euro-denominated bond due in 2008 with an annual coupon, and sterling- and dollar-denominated bonds due in 2011 with semiannual coupons. The company was upgraded by Standard & Poor’s last May so the coupon on the sterling and dollar bonds has already been changed. The euro bond is still on the old coupon.
France Télécom’s situation is particularly intriguing since its next coupon date is in March and Moody’s has put it on review for possible upgrade. The agency makes much of its independence so naturally the coupon date will not influence the timing of its decision. Nonetheless, it is worth noting that KPN was upgraded the day before its coupon date last year, after having brought forward publication of its annual report specifically to give the rating agencies time to analyse all the information they deemed necessary.
Even for those who acknowledge that step coupons were attractive when they were first brought to market, there is a feeling that the myriad uncertainties surrounding bonds with step language make them almost impossible to price. “This is always the tricky question because, at this point, the pricing of step-down coupons is largely dependent on the timing, direction and degree of rating agency actions. The direction is fairly clear in most investors’ eyes. The timing and degree are two unknowns,” says Matthew James, telecoms credit analyst at Morgan Stanley.
Duncan Warwick-Champion, telecoms credit analyst at UBS, has built individual models for each step coupon bond. Overlaying possible outcomes and the probability of possible outcomes, he has come up with individual prices. “I think that most of the short-dated bonds are trading reasonably well according to my theory but it is harder to calculate the option value in situations where there are fewer points of comparison. Dollar bonds are also more difficult to price because there are other issues, such as the popularity of European issuers with dollar investors,” he says.
Warwick-Champion also acknowledges: “Everyone has a different methodology for calculating step-ups and market expectations have an impact too. The real world tends to affect the model theory.”
“The credit markets are imperfect,” adds Appleyard from ABN Amro. “They take time to digest information and adjust prices accordingly. The problem with steps is that a lot of investors don’t know how to value them and a lot of buy- and sell-side analysts don’t know how to either. There are varying degrees of sophistication among the valuing techniques and the credit rating agencies are difficult to predict.”
The situation is especially acute now because markets have changed and far greater attention is paid to minute alterations in yield and spread levels. Spreads tightened radically in 2003, affecting not only the pricing of individual bonds but also the difference in pricing depending on one credit rating or another. “Nobody used to worry much about a few basis points. Prices were changing by 15 to 30 basis points a day. Now five basis points really matter,” says UBS’s Warwick-Champion.
What this means is that the market’s inability to accurately price step bonds is becoming more frustrating as the dynamics of the sector encourage increased volatility. The more upgrades are viewed as a possibility, the more different outcomes need to be factored into individual attempts to price these bonds correctly.
“The experiment is over but a legacy of up to €100 billion remains and will be here for years to come,” says Andrei Gorodilov, telecoms credit analyst at CSFB.
Not only are step bonds likely to be here for the next 30 years, but many in the market believe they were fundamentally flawed from the beginning. “EMI offered a huge step-up of 150 basis points if it went sub-investment grade but when that happened, the price of the bond went from 97 to 86, a far more significant change than the coupon uptick,” says Warwick-Champion.
“People are being too heavily penalised as the ratings go up and they were insufficiently compensated on the way down,” says Gorodilov.
Step coupons were a product of their time. Popular at launch, they are now widely considered an irritation at best, but the market is confident that important lessons have been learnt. If similar conditions prevailed once again, uniformity of approach would almost certainly be one demand of the investment community. Fewer changes would be another. The market is unlikely ever again to see anything like France Télécom bonds, with their 25 basis point changes per rating notch per agency.
The week in Risk.net, May 19-25 2017Receive this by email