In taking the decision to pull what had ended up as a truncated $1.5 billion 10-year bond offering,Pacific Century CyberWorks-Cable Wireless HKT (PCCW-HTK) managed to salvage a tiny piece of dignity from a colourful new chapter in its short, but sorry history.
At a launch spread of what had widened out to 325bp over Treasuries, PCCW-HKT abandoned a deal that had become uneconomic in terms of the price it was prepared to pay in order to extend its maturity profile. The move was applauded by the market.
As one head of fixed income research says, "It sends out a positive signal because it shows that PCCW wasn't prepared to raise money at all costs and is, therefore, not in liquidity bind. It was a prudent move. There comes a point when the incremental cost to lengthen the maturity over the average cost of bank debt (115bp over Libor) simply isn't worth it."
In some respects, the company was extremely unlucky. The possibility that Argentina might have to reschedule a portion of its $130 billion outstanding debt sent all emerging market spreads wider, and made the pricing of any primary markets deal extremely difficult as roadshows came to an end.
So too, going into the deal, it would have been entirely natural to imagine that international investors would have bought a credit that had secured a higher than expected rating of Baa1/BBB. Until that point, the Asian bond markets had been enjoying a bull run all year; PCCW-HKT would have hoped to ride on the back of this same strong demand and re-finance its debt at a fortuitous point in the interest rate cycle. And while the telecom sector remains out of favour, PCCW-HKT is an incumbent fixed-line operator that has maintained its dominance in the face of competition, retaining an 88% market share for business lines, and 97% share for residential lines. As a result, it enjoys some of the strongest cash flow margins of any company in its sector worldwide.
Companies regularly pull bond deals and/or restructure them to fit altered market conditions. But very few receive the kind of lambasting that PCCW has been subject to.
Only five months ago, Hutchison Whampoa also abandoned plans for a 30-year tranche on a $1.5 billion bond deal after investors failed to deliver the price it wanted. The crucial difference between Hutchison Whampoa and PCCW-HKT, however, is that while the former made sure that it looked like it was acting from a position of strength, the latter has appeared paranoid, uncommunicative and defensive from the very moment news of its impending bond deal became known.
Many observers believe that the company has mismanaged virtually every aspect of the transaction and in the process, lost any leverage it might have had with investors. Global co-ordinator Morgan Stanley has also been roundly criticized by every single Asian debt professional for putting together a strategy that sought to re-finance PCCW-HKT's syndicated loan in a single shot. The three joint lead managers bought in at a later date Barclays, HSBC and JPMorgan have not escaped criticism either. Competitors believe that too much time was spent making sure they kept their syndicate positions rather than questioning a strategy that had raised concerns from the outset.
Following Malaysia's bad example
In particular, a number of commentators point to the example of Malaysia, whose equally unsuccessful sovereign bond issue of July 1999 shared many of the same mistakes. Neither entity followed the textbook example of how to complete a successful deal. Firstly, start small, build momentum, then tighten the pricing or increase the size as the book starts to overflow. Do not tell the market upfront that the intention is to raise billions of dollars and then spook investors further by stating that the size will be increased again if demand is strong enough. Secondly, do not launch an ambitious deal right on top of Christmas, or the summer holidays. Thirdly, if market conditions deteriorate, complete the roadshows then tell investors that the deal will be launched when spreads stabilize. And fourthly, try to think of the client's needs, not of the fees likely to be generated.
For many Asian debt experts, the failure to follow these cardinal rules says as much about the psychology of the Malaysian government and PCCW management as it does about bad advice. As one of long-standing fixed investor concludes, "We've been waiting a decade for Hong Kong Telecom to do a bond deal and this is what we end up with. Is this the way that a well-managed blue chip Hong Kong company should approach the debt markets? The conservative and very British mindset of the old Hong Kong Telecom would never have led to a deal like this. It's so typical of the brash, suit-less, tie-less approach of the new PCCW, which starts at the top and permeates all the way down. It's all, 'I want it big and I want it now.'"
Why do a bond deal anyway?
With the benefit of hindsight, many observers argue that PCCW-HKT should never have attempted a bond deal in the first place. Having analyzed the company's capital structure, most fixed income analysts come to the same conclusion. In essence, they argue that the key issue for the company is to make sure that it can keep upstreaming dividends from the cash cow (PCCW-HKT) to the parent company (PCCW), which has little incoming income, but a sizeable debt load and capital expenditure needs.
Why take on additional interest expense, the argument runs, when this will neuter net income at the very time when the company needs to keep dividends high?
As Jason Carley, Merrill Lynch's head of Asian fixed income research says in a recent report, "We estimate that HKT will incur an extra 39% in cash interest costs in FY 2002 (from $227 million to $316 million) should the bond issue be $2.5 billion, and up to 58% higher (to $359 million) if the issue reaches the floated $3.8 billion."
It is a view echoed by UBS Warburg's head of Asian fixed income research, Stephen Cheng. "Given the dependence on HKT cash flows, we believe one of the group's objectives is to enhance the ability to upstream cash to the parent level," he says. "The ability to upstream dividends will grow in importance as the maturity date on PCCW's major debt draws nearer."
The covenants issue
Yet the very objective of the bond deal was to re-finance most of the $4.7 billion syndicated loan which constitutes virtually all of PCCW-HKT's outstanding debt. In doing so, restrictive covenants, which hinder the company's ability to upstream dividends to the parent, would be removed. Raising only $1.5 billion would have left PCCW with the worst of both worlds, however. It would have vastly increased the company's interest costs and at the same time would not have taken out enough of the loan to rid itself of the covenant problem.
But as a number of observers have pointed out, why did the leads think that international investors would ever want to the buy the bonds of a company from which cash would have been systematically drained in order to support a much weaker parent? Any investor of sound mind would have seen through the strategy straightaway and would want to be compensated as such.
As Carley puts it, "While investors do benefit from the presence of bank covenants and cross default, the fact those covenants were not replicated in the bond issue raises significant risk should the bank loans be prepaid, renegotiated or the covenants themselves be waived by the bank lenders."
And as one head of Asian DCM at a competitor bank adds, "As a bond investor the last thing you want to have to rely on is another lender's covenants. I wouldn't have bought this deal at 400bp over. It all comes down to whether the company has earned the market's trust. And with PCCW and Richard Li, the answer has to be a resounding no."
In structuring the deal, investors were partially compensated with the kind of step-up, step-down coupons that have become prevalent in European telecom-related deals. For every notch upgrade or downgrade from either of the rating agencies, PCCW-HKT's coupon payments would have moved by 25bp. However, this too has also been criticized.
"There is a very big difference between being paid 25bp if British Telecom is downgraded from single A to high triple B and being paid 25bp if PCCW-HKT is downgraded from triple B to double B," says one banker. "Investment grade investors don't even want to contemplate the prospect of a deal turning junk. It frightens them and once the idea registers in their brains, they want a lot more than 25bp per notch downgrade."
No Asian demand
One of the key aspects that has set PCCW-HKT apart from the year's run of successful Asian bond deals, has been its failure to generate local demand. Observers report that the final order book was heavily dominated by US accounts and sparsely populated with those from the company's home region. By contrast, Asian deals and particularly those from Hong Kong, have seen tight pricing driven by a combination of local banks awash with liquidity and retail investors moving from choppy equity markets to the safer havens of debt.
The leads have been criticized for either not realizing that Asia was unlikely to drive the deal, or underestimating the impact of relying on US investors to price such a large and complicated bond offering. The latter scenario is thought to be more likely.
Given that three of the lead managers on the bond offering are also bank lenders to the group, they would have been well aware that the sector is trying to reduce its exposure to the credit, not increase it. So too, HSBC has acted as lead manager on one of the parent's convertible bonds and would have known that retail investors, bruised and disillusioned with the company's performance in the equity markets, would be in no mood to listen to its supplications in the debt markets.
Competitors argue that relying on US investors was always likely to increase indicative pricing levels and particularly if they thought they smelt blood. It is believed, however, that the leads always knew that the bond issue would need to be priced in the mid to high 200bp level over Treasuries. It was only when market conditions deteriorated and momentum was lost, that spreads levels spiralled beyond 300bp and the deal completely unravelled.
"PCCW-HKT is not a Deutsche Telekom or an AT&T," notes one observer. "There are not many investors that would consider switching out of either of these two credits into an Asian telecom play. There is a limited pool of funds to buy any kind of Asian deal and to get such a large size away, PCCW-HKT would've had to come at a premium to international comparables."
European telecom credits with similar ratings have been trading in a 220bp to 250bp band for 10-year debt in recent weeks. Pricing PCCW-HKT at a 30bp to 40bp premium to these levels might have seemed a sensible objective. This would then have put the company in line with the secondary market trading levels of Telekom Malaysia.
Traditionally, Malaysian corporations of comparable credit quality to Hong Kong corporations would be expected to trade wider because of the "Mahathir" premium. The large size of PCCW-HKT's transaction would have partially counterpointed this premium.
On Friday, Baa2/BBB-rated Telekom Malaysia's outstanding 2010 bond was trading at a bid/offer spread of 295bp/280bp. Other similarly rated Hong Kong credits were trading much tighter. Baa1/BBB-rated Dao Heng Bank, for example, has a 2007 transaction bid at 200bp over Treasuries and Baa2/BBB-rated Bank of East Asia a 2011 bond bid at 232bp over.
Further down the ratings spectrum, BBB-/Baa2-rated Citic Pacific has a 2011 bond bid at 325bp over Treasuries. Making sure that it did not receive a rating in the low triple-B range was one of PCCW-HKT's key objectives earlier this year, since it knew that a low triple-B rating would entail a sizeable jump in spread. The point was reinforced by the performance of Citic Pacific, which came to the international bond markets in late May and had to stretch a lot further to source demand for a $450 million deal.
How weak is PCCW's financial profile?
The other comparable credit, which ultimately compromised PCCW-HKT, is PCCW itself. The parent's December 2005 convertible bond is currently trading at a spread of about 460bp over Treasuries on an implied double-B rating.
For most observers, the relationship between the operating company (PCCW-HKT) and parent company (PCCW) is the key to the whole transaction. A number of analysts believe that while there is no cash crunch looming at the parent level, this largely hinges on how it is able to finance a $2 billion commitment for the construction of Cyberport before its 2007 deadline.
To the financial year ending March 2001, PCCW-HKT paid an interim dividend of HK$4 billion ($514 million). However, because of changes to the covenants attached to its $4.7 billion syndicated loan this February, the amount will be severely restricted in the coming few years.
The loan covenants state that if PCCW's total debt to EBITDA exceeds 3.5 times, it can only dividend up 35% of net profits. If this level falls to between 2.5 times and 3.5 times, it can dividend up 75% and if it falls below 2 times, it can dividend up 100%.
Given that its net profit for FY 2001 totaled H$5.879 billion ($755 million), the company paid out a dividend representing 68% of the total. For the coming financial year, analysts estimate that pre-bond issue, the company would be able to pay out about $211 million to $260 million and post-bond issue, about $170 million. This is based on an assumption that PCCW-HKT will not be able to reduce its debt to EBITDA ratio from its current level of 4.05 times and will report a reduction in net profits during FY 2002.
The parent on the other hand, is estimated to have cash needs of about $640 million per annum. This figure is comprised of annual interest payments amounting to about $94 million, losses stemming from its Internet businesses, which are expected to fall from $180 million per annum to about $100 million over next two years and the funding of Cyberport, for which it needs to need to commit $300 million.
In its ratings release, Standard & Poor's said that the parent has approximately HK$10 billion ($1.3 billion) in cash, which analysts say will tide it over to the end of 2002. In its offering memorandum, PCCW-HKT also reported that it has cash balances of HK$1.58 billion ($203 million). At the parent level, the company has debt of $2.1 billion, of which $1.696 billion comes due in December 2005.
Where does PCCW stand now?
Those who believe that the company should not have attempted a bond deal argue that the company should stay on the sidelines for another year. "It's not critical for the company to raise debt at this stage since tranche A of the syndicated loan does not need to be re-financed until 2003," says one banker. "What it should do is wait until its new Internet strategy begins to bear some fruit and losses are reduced. When this happens, equity investors will come back to the stock and it will be a good platform to raise debt."
"Very few people have much faith in the management abilities of current PCCW executives," another concludes. "They should wait until a senior industry figure has been appointed to take up the day-to-day running from Richard Li."
The idea that the company may attempt to come back in September is a prospect that few relish, but on past experience, many may yet expect.