Is Gates planning a share stake shuffle ahead of an MS split?
Some of the company's requests suggest that very thing
MS on Trial Could it be that Bill Gates has a plan to reduce his shareholding to less than five percent, so that after the divestiture, he could still have shares in both companies and be recognised as the person running both of them? Gates would not relish being identified as a pariah, and to him the most distasteful of the DoJ's proposals must surely be his being prevented from being seen to control both companies after the divestiture.
The shareholding restriction follows from his being defined as a "covered shareholder" by virtue of holding more than five per cent of the shares. There has already been considerable speculation as to which company he would support, and we feel sure he would find such speculation most unwelcome. Here's our reasoning for thinking there may be a cunning plan.
A glance at Microsoft's last proxy statement shows that as of 10 September 1999, only Gates and Paul Allen held more 5 pe rcent of the outstanding common shares. In Allen's case it was 5.1 per cent, and there is no question that he has reduced his shareholding considerably since then to less than 5 per cent, and so will escape being labelled a covered shareholder. The DoJ agreed to raise the threshold to 5 per cent from 3 per cent following remarks by Judge Jackson on 24 May. It turns out that Steve Ballmer will also escape, since he had only 4.7 per cent last September, and although it is possible he has received some stock options not yet announced, it is unlikely since neither he nor Gates received any during fiscal 1999 - and there were no outstanding options as of last September. Gates has disposed of a considerable number of shares since the proxy statement, so his then 15.5 per cent holding must now be down a few per cent.
With that background, let's turn to Microsoft's proposed changes to the Final Judgement, and the DoJ's filing yesterday. Microsoft tried to get the paragraph containing the definition of a "covered shareholder" changed, so that the date of determination of whether Gates had more than 5 per cent was based on when the separated business had been set up, rather than the date of entry of the Final Judgement. We think that there is considerable significance in this, and float the following hypothesis: Gates wishes to ensure he has a less than a five per cent holding, so he might transfer just under five per cent of his shares to his wife, and the remainder to his children.
There are two factors supporting the hypothesis. One is the extra time being sought - months rather than days - before the transfer of ownership. Regulatory problems would almost certainly make this impossible before the Final Judgement is entered.
The other factor is that in the paragraph defining "non-covered shareholders", Microsoft introduced a complicated argument premised on the principle that the implementation of divestiture should be stayed indefinitely pending appeal. The argument is made more complicated by Microsoft's claim that, based on "recognised practice", "a .Non-Covered Shareholder' be a shareholder of record on the .record date' set forth in the documents that describe the issuance of shares to effectuate the divestiture".
Whether the DoJ suspected that there might be some such plan is uncertain, although there is an enigmatic comment that "Proper effectuation of the purposes of the Final Judgment requires determination of Covered Shareholder status as of the date of entry of the Final Judgment." The DoJ did offer an alternative text to give comfort that there would be no problem over the complicating point that Microsoft had raised. This was done as an accommodation, and the DoJ did not say that there was any substance in what Microsoft had raised.
We may get some further clues as to whether Gates is hell bent on owning shares in both WinCo and AppCo by seeing how fiercely Microsoft contests this point when it files its response on Wednesday. Of course, all this is hypothesis, and as with any good theory there needs to be an alternative hypothesis. In this case, it would be that we are wrong. ®
Sponsored: Benefits from the lessons learned in HPC