The average deal we've done is 50% institution, 50% individual. Make no mistake about it, the right institutions want to play. The right institution is the guy who has a dedicated small-company fund. But in the current process, those institutions also have been sifted out.
We are targeting the guy who really looks for special situations, who is part of a smaller fund that may be part of a bigger fund family. The whole point is not to exclude institutions, but to see that large institutions don't have a built-in advantage.
When did the lightbulb go off? When did you decide that a new approach to IPOs was needed?
I seriously started considering it after the Hambrecht & Quist IPO, in 1996. When we went on the road show, we suddenly hit a difficult market. There were some institutions that weren't going to buy our stock. But eventually, we hit a few who liked the story and gave big orders for the stock. The deal came together -- the market improved a little the week we were doing it. All of a sudden, all these institutions that didn't like the stock came in with orders. There was no way the other underwriters or H&Q could say no to them. We ended up distributing a lot of stock into the hands of people I knew didn't want it, who were buying it purely for the flip. So when our stock went up a couple of points, they flipped it. The guys who really wanted the stock ended up buying it in the aftermarket.
That's when I started thinking, "Why shouldn't we find people who really want to own the stock? Why should we give the stock to people who are going to flip it and only pay us back in commissions?"
Are there any experiences that reinforced your belief in the need for an alternative?
The Boston Beer IPO. My impression was that Jim Koch, the founder, really didn't want to go public. When he finally agreed to sell, he made it a condition that some of the stock had to be offered to his customers. Goldman Sachs and H&Q managed the deal, but neither knew how to do it. So Jim researched it and found out that you could advertise a new issue. He put coupons about the stock issue in his six-packs, and within a month or so we had 120,000 checks for a total of $55 million.
We didn't know how to distribute the stock efficiently. So we brought in Advest and a couple of brokers, rented a fulfillment program from Dreyfus, and eventually pulled it all together. But we gave only $15 million of the offering to Boston Beer's customers. We sent $40 million back. It bothered me. There was all this demand that was left unfulfilled.
What struck me also was the enormous retention rate for people who bought that $15 million worth of stock back in 1995. I made a deal with Al Rossow, who was then the chief financial officer, about tracking those shareholders. About a year ago, when Al left, about half of those people still held the stock.
It became clear to me what was happening. Your most loyal shareholder -- the shareholder you can bring into your IPO -- is either your customer or somebody who for whatever reason respects your company.
What was it about that high retention rate that you found so interesting?
That this was hopefully a way of building a brokerage business without having to advertise, especially with the advent of electronic brokerage and the unfulfilled demand that was out there.
I was convinced that the IPO stock was being distributed to the wrong investors. More accurately, the stock was not going to the right people, not to the people who really wanted to own the stock.
It was when I really started to look at our distribution at H&Q -- and started looking at other distributions -- that I realized the stock was systematically being placed in the hands of the big-commission players, who were usually not long-term investors. It's so much more profitable for the big investment banks to give the stock to institutions. The underwriters don't pay the brokers who sell to institutions very much, so the underwriters' gross margin on the commissions is much higher on institutional sales than on retail sales. Probably even more important, the underwriters have a reciprocal flow of orders from the institutions, and this means more commissions.
"Companies aren't interested in getting the absolute maximum dollar amount for their stock. They want happy shareholders."
I became convinced that as long as you've negotiated IPO prices that are below what the market is offering, you are, in effect, creating guaranteed profits for a group of investors. But you're also giving away something of real value, taking something away from the balance sheet of the company that's going public.
My reaction was to set up an underwriting firm geared to electronic brokerage that could place the stock more efficiently in the hands of long-term investors. At first, it hadn't occurred to me not to use a conventional underwriting form. But then I realized that the only way you could change the whole structure was through an auction.
Can you give any real-life examples of how Dutch auctions might work?
Take Instinet Group, which is a securities broker that offers institutional investors an electronic system for stock trading. Instinet is a believer in electronic markets. So when it was going public, it wanted to position itself as a new-age investment firm; it had thousands of trading terminals out in the institutional market. But it was afraid that under a normal negotiated deal, only a couple hundred of those customers would get Instinet stock, and the company would then have a very angry customer base. So the question for Instinet was, How do you satisfy that demand?
When Instinet decided to go public, its lead bankers agreed that we at WR Hambrecht would handle approximately a fifth of the offering. That allowed Instinet to tell its customers that if they wanted stock, they could go to our Web site and make a bid.
It was a fascinating experience. The Instinet IPO this past May was a 32-million-share offering, a $464-million deal. Our share was approximately 6 million shares. We could have done the entire deal ourselves. We ended up with bids for 74 million shares. Credit Suisse First Boston, the lead bankers, wanted to price the Instinet offering at $12 to $13. But based on our bid prices, CSFB ended up having to price it at $14.50. The stock opened at $18.50 and has traded around $17 to $20.
To me this is a perfect example of the Dutch auction's finding the right clearing price. It found, almost to the dollar, what the offering price should be.
Please e-mail your comments to editors@inc.com.
