| CORPORATE FINANCE |
April 2001 |
E-BONDS: WILL THEY FLY?
Emily S Plishner looks at how Internet
bond auctions can help corporate issuers lower their cost
of capital.
By Emily S Plishner
Last August, when US-based Dow Chemical
became the first non-financial corporation to sell its bonds
directly to investors through a web-based Dutch
auction, it was hailed as a watershed event in the investment
banking world. The businesses of securities brokerage and
trading had already been revolutionized by on-line brokers
and electronic communication networks. Wall Street's most
lucrative franchise, the control of new-issue distribution,
was next. And corporate issuers could expect to reap the benefits
of lower transaction costs and better prices for their securities.
Why then, seven months later, is Dow still the only industrial
corporation to conduct an on-line bond auction? Two words:
bear market. Thanks to widening spreads on corporate bonds,
raising debt capital was a difficult proposition on-line or
off in the second half of last year. And corporate issuers
decided that in a bad market, they were better off with Wall
Street underwriters pricing the deal, rather than the open
market. "There is a fear among issuers that in a weak
market, the transparency of on-line pricing might work against
them," explains Ralph Cioffi, senior managing director
at Bear Stearns in New York.
Maybe so, but it's hard to argue with the success of Dow's
on-line auction. The company raised US$300 million in five-year
bonds using software developed by San Francisco-based investment
bank WR Hambrecht. The auction drew a far broader investor
base than usual, and every successful bidder walked away with
a full allocation. Dow is paying about the same interest rate
it would have paid had it taken the traditional syndicate
route, and it had to cough up less than half the typical underwriting
fee. "To me, it's a no-brainer," says Dow treasurer
Geoffery Merszei.
It did, however, take some courage on Dow's part. Wall Street
banks stand to lose a lot of power and revenue if they relinquish
control of new-issue distribution, and they're not above fighting
to save their bread and butter. Undoubtedly, plenty of corporate
issuers will decide that a few basis points aren't reason
enough to damage relationships with their investment bankers.
But in the long run, Internet bond auctions, particularly
for plain-vanilla investment-grade debt, could save corporate
issuers a significant amount of money.
E-Bonds and Dutch Auctions
The first corporate bond billed as an "Internet bond"
or "e-bond", was issued by Ford Motor Credit in
early 2000. The US$1 billion offering of three-year notes
worked much like a traditional bond underwriting, except that
the prospectus and other marketing materials were posted on
the web, and orders were taken by email. The issue, however,
was still priced by the lead manager, Lehman Brothers, not
by the investors that ultimately bought the bonds.
Over the next three quarters, a handful of other Internet
bond offerings followed, with issues by DaimlerChrysler, the
World Bank and BASF, and, later in the year, by Bear Stearns,
Deutsche Bank and, ultimately, Dow. While these offerings
were all billed as e-bonds, the last three were the only issues
priced and allocated through on-line Dutch auctions.
In a Dutch auction, investors bid for a particular amount
of a security at a specific price. The best bids are accepted
in the amounts requested until a clearing price is reached
for the entire issue. All successful bidders then get their
requested allocations at the clearing price. Off-line, it's
essentially the form that Treasury bond auctions take. But
the Internet brings a new dimension to the process, making
the model viable for securities with a smaller potential audience.
While most investment banks are fighting it every step of
the way, virtually everyone in the industry agrees that the
Internet is already changing the way securities are distributed.
And at least some players are preparing for the future. "It's
better to be on the train than under it," concludes Mark
Millender, managing director of debt markets at Bear Stearns.
Both Bear Stearns and Deutsche Bank have conducted their own
auctions using proprietary software accessible through their
websites.
Besides those from the two investment banks, the other e-bonds
were priced according to the usual mysterious ways of the
syndicate business. That process takes place in private among
investment bankers with a long history of back-scratching
and IOUs. The ultimate price of an offering has as much to
do with who owes what to whom on Wall Street as it does with
the level of demand in the marketplace. The Ford Motor Credit
issue, notwithstanding the administrative and marketing work
done on the web, was no different from any other underwriting
in this regard. "It's just glorified email," says
Millender.
Dow, in contrast, used four co-managers for its deal: Hambrecht;
Bear Stearns; HSBC; and Williams Capital Group. The two-hour
auction took place at Hambrecht's auction website, www.openbook.com,
and, according to Leland Crabbe, a portfolio manager at Credit
Suisse Asset Management in New York who participated in the
bidding: "It was the first [e-bond auction] that felt
real to the market."
What makes e-bonds revolutionary
is that the market, not Wall Street, sets the price. "That
is the whole point of using the Internet," says Merszei.
"Without question, the pricing will be more favorable
to borrowers." He expected Dow's issue to price between
98 and 102 basis points over Treasuries. It came in at 101,
with a clearing yield of 7.108 percent. And the underwriting
spread that Dow paid to its bankers was half the usual 60
basis points of a syndicate deal. "Even if we'd paid
102, the all-in price would have been very competitive,"
he says.
Help Wanted
Not surprisingly, when Merszei went shopping for a manager
for Dow's on-line auction, the A-list underwriters that usually
vie for the company's business showed little interest. Only
two banks were prepared to take the job: Hambrecht, which
had plenty of experience with on-line equity IPOs but next
to none with investment-grade debt, and Bear Stearns, a second-tier
player in corporate debt underwriting.
Most, if not all, of the leading fixed-income underwriters
have or are developing their own software to conduct on-line
auctions, but few are prepared to use it for fear of cannibalizing
their existing underwriting business. "We'd be kidding
ourselves if we didn't recognize that the business is changing,"
admits Jim Merli, New York-based managing director for Lehman
Brothers, noting that Lehman already does auctions of money-market
preferred securities. "If that's the way the market goes,
we are prepared to respond."
Like most investment bankers, however, Merli is in no hurry
to see it happen. Investors who participated in the Dow and
Bear Stearns auctions were reputedly discouraged by Wall Street
salespeople. There were predictions that Wall Street dealers
would refuse to trade the paper in the aftermarket, there
would be no research to support the issue, the auction technology
would fail, and not enough investors would participate.
Some suggested that without the moderating hand of the syndicate
desk to set allocations, investors might end up paying too
much for the bonds. It was also predicted that in a volatile
market investors would beat down prices. "They were telling
me it was bad for the issuer and bad for the investor,"
says Credit Suisse's Crabbe. "But it can't be bad for
everybody." Underlying all the arguments was the not-so-subtle
innuendo that auction participants could expect retaliation
when it came time to allocate the next traditional deal.
For the largest, most-favored investors,
that's a significant threat. But smaller investors generating
less commission for dealers and CFOs who manage their company's
pension fund in-house are already at the bottom of the pecking
order at the syndicate desks: their allocations in traditional
deals can't get much worse. In a Dutch auction, where the
best bid wins, smaller, less well-connected investors are
on an equal footing with the heavyweights, and are much more
likely to get their allocation than they are in a syndicate
underwriting.
At the end of the day, the intimidation
apparently had little effect. Plenty of investors logged on
for all three on-line auctions, pricing was good for the issuers,
and investors got their full allocations. Trading volume in
the aftermarket may have been light, but that was a positive
sign: successful investors were hanging on to their bonds.
And according to Merszei, Dow has not been blackballed by
other investment bankers. "Secretly, many of them - including
senior managers - say this is the wave of the future,"
he says.
The Merits for Issuers
The great promise of Internet bond auctions for corporate
issuers is a lower cost of capital. But one of the biggest
sources of expected savings - lower sales and marketing costs
- may not be as large as first imagined. All the parties involved
in the Dow deal insist that a successful Dutch auction requires
an active effort by salespeople, no matter how good the technology.
"Software doesn't answer questions," says Chris
Williams, founding partner of investment bank Williams Capital
Group. "It still calls for an educated salesperson to
explain the underlying credit and the relative value of the
issue." Williams has a dozen people in fixed-income sales,
and doesn't expect that to change.
While the four managers of the Dow issue
halved their commission on the deal (30 basis points versus
60), the discount may in part be a reward for Dow's willingness
to serve as guinea pig for the auction. "Everyone has
assumed that OpenBook is aimed at cutting fees, but it's not,"
says Robert Goldberg, co-head of debt markets at Hambrecht.
"For the issuer, it's a method that results in a better
price and more transparency."
Indeed, the greater transparency of Internet auctions is the
reason that issuers should realize better prices for their
bonds. In traditional syndicate deals, underwriters often
have an incentive to underprice the issue in order to attract
big institutions with which they would like to do more business.
Investors with large allocations can then make a quick profit
selling the underpriced bonds in the secondary market, because
of pent-up demand.
In Internet auctions, the book is built
on-line in real time, giving participants a much better sense
of the true demand for the securities in the marketplace.
Consequently, they're apt to make better bids. "Investors
will accept less yield if they know the bonds are likely to
go up in value," says Bear Stearns's Millender. And because
the price reflects true demand in the market, successful bidders
are also more likely to hold on to their investment. According
to Stephane Paquier, Dow's corporate finance director: "We
got long-term investors instead of traders. Most of our investors
are keeping our paper."
An added benefit of an open auction appears to be a broader
distribution pattern among investors - something issuers like
because it can reduce the volatility of the securities in
the secondary market. The Dow e-bond auction attracted 57
investors, about three times the usual number for a deal of
that size. Bear Stearns ended up with 129 investors, about
twice the normal number. "It may have been investor curiosity
as much as anything else," says Cioffi.
Whatever the reasons, the auctions clearly
generated a lot of interest in the investor community. "The
process unfolded just as our game theorists predicted: as
the auction progressed, the pace of bidding increased and
the spread came down," says Cioffi, who handled Bear
Stearns's auction of its own debt.
And the initial concerns about liquidity in the aftermarket
proved to be unfounded. Unlike stocks, bonds - even of large
issuers - are not terribly liquid, so the market-making support
of the investment community can be important. Although it
is difficult to measure liquidity, Dow's e-bond is trading
at prices and volumes related to the company's credit rating,
the size of the offering and prevailing interest rates - just
like any other bond issued by the company.
So what will kick-start the market for web-based bond auctions?
Falling interest rates could help. So too, could the successful
auction of US$5 billion in three-year notes on February 9
by residential mortgage agency Freddie Mac in the US. As Jon
Prince, manager of debt marketing for Freddie Mac, says: "An
auction is more interesting to a bull than a bear."
Yet many investors will likely sit on
the sidelines until one system is established as a de facto
standard. That hasn't happened yet, and in a shaky credit
environment with a recession threatening, it may not for a
while. Hambrecht, however, recently scored a huge endorsement
when Freddie Mac announced it would use the system for some
large tranches of its mortgage-backed securities.
Both proponents and opponents of
on-line Dutch auctions of corporate debt say one thing is
clear: the system works best for large, plain-vanilla issues
with good credit stories. That's why TradeWeb and Muni-Auction
are already up and running with agency issues. For corporates,
that means quality investment-grade debt, and maybe asset-backed
securities. Says Hambrecht's Goldberg: "It's always better
to introduce change with well-known credits that don't need
a lot of explanation."

Emily S Plishner is a freelance writer based
in New York. |