Here’s a good article on what happened to Passave’s (PSVE) IPO plans.
Based on this article, other information, as well as personal experience, here’s my take on what probably (READ: SPECULATION) happened. Anyone who has worked for a semiconductor company and developed silicon based on a new standard won’t find anything remarkable in my analysis. You’ve seen this movie before and you already know how it ends.
- Passave, as a startup, has an EPON device under construction. UT Starcom (UTSI) is in the broadband access business, and decides to partner with Passave to enter the Japanese FTTH market.
- UT Starcom builds a hardware launch schedule around Passave’s silicon schedule. They make commitments to Softbank/Yahoo! in Japan. Softbank/Yahoo wants the hardware pronto so they can strike first against NTT with FTTH service, and score a coup like they did with DSL.
- UT Starcom makes the assumption that they will be able to ship rev A silicon as production for Softbank/Yahoo subscribers. Any experienced hardware engineer knows this is pure suicide.
- The availability of quality Passave silicon slips beyond where UT Starcom expected it. This could be a lack of realistic expectation on the part of UT Starcom or a mis-representation of reality by Passave. It’s also possible Passave unknowingly gave a clean bill of health to silicon that still had bugs. Faulty product got shipped, and needed to be recalled.
- Someone didn’t complete a full characterization of the hardware at Passave, UT Starcom, and Softbank/Yahoo!. Softbank probably rushed the equipment into service without a full flight test. Rather than accept responsibility, they put a bullet in their relationship with UT Starcom.
- UT Starcom loses a key customer in a crucial new market. Winning new business outside of PHS is what they are supposed to be doing. Some hardware QA was sacrificed; maybe the customer (who drove the aggressive schedule) wasn’t well briefed about what was shipping. Heads roll in UT Starcom HQ in Hangzhou, someone has to pay.
- UT Starcom makes reasonable/unreasonable demands of their vendor to pay up. Passave refuses. UT Starcom decides to file a lawsuit, knowing the impact such a move will have on the IPO. Waits to file lawsuit until Passave is well into quiet period. Places themselves in very solid negotiating position.
The real issue here is the following: Did Passave provide UT Starcom with sufficient information, and UT Starcom chose to ignore this when it shipped equipment? Also – was there some sort of penalty/liability clause in their agreement with Passave? Passave has some smart folks, and smart folks in the semiconductor industry refuse to sign penalty clauses or exclusivity agreements.
We analyzed the Passave prospectus in detail a few months back. They have done an outstanding job of building a highly defensible business around EPON. They have great cash flow and can probably get by without an IPO if they focused on the EPON market only. NTT might even be willing to help them out.
The comm semi business is very tight knit. If Passave was upfront and honest with UT Starcom they’ll be OK. If they abused their partner then they are in big trouble with every other customer going forward. I am going to further speculate and make the following conclusions:
- It is suspicious that UT Starcom waited to file the lawsuit until just before the IPO. If Passave felt they were even remotely in the wrong they would have settled in advance to avoid any roadblocks.
- NTT has based their entire deployment on Passave. NTT is a cautious carrier, and won’t work with a supply chain that isn’t reliable, period. This would mean UT Starcom and Softbank/Yahoo! played too agressive and got burned. Looking at FTTH deployment in Japan, it appears Softbank never recovered as NTT has most of the subscribers.
If you know more, by all means, please post here.
[Update 16:00 14 DECEMBER 2005]
Several people have contacted me for copies of Passave’s financials and our research. A paper copy of the prospectus can be obtained from Merril Lynch, or you can get an electronic copy from multiple online sources. The link to the SEC HTML copy is here.