Survey skewing is simple if you ask the right questions
Despite survey showing disapproval of listed companies with dual share structures, most fund managers are happy to buy these types of stocks
Now gather round, children, for today's lesson on how to get the results you want from phrasing your survey questions just the right way.
Leading questions that you cannot answer with any credit to yourself are old hat and respondents recognise them. It is no good when flogging toothpaste, for instance, to ask, "Are you still so stupid as to use Wet Cement's gritty product?" You just annoy people when you do it that way.
Far better is to set the mood you want with the first few questions. Thus in toothpaste you first ask: "Are you in favour of good dental health?"
Everyone will tick the Yes box, of course, which makes it all the easier to get the answer you want when you then ask: "Don't you think you should buy lots and lots of our toothpaste?"
The evidence is plain … that big fund managers … are quite happy to sup with the devil
Similarly when doing a survey of fund managers on whether Hong Kong Exchanges and Clearing should allow differential voting rights by class of share, your best option is first to ask: "How would you rate the importance of fair treatment of all shareholders in a market?"
They will all tick the Very Important box, of course, and then you have them where you want them. Dare they approve later in the survey of anything that might not be deemed fair treatment? Dare they?
Just a tip on how you do it.
But the one thing you must not do in this survey is start by asking the question: "Does your firm buy any companies that have dual classes of shares with controlling shareholders having more votes than other shareholders?" Make very sure that you do not ask this question as you will then not get the 98 per cent No vote that the Asian Corporate Governance Association got when asking its respondents whether they were in favour of listed companies having such dual share structures.
The reason you will then not get an overwhelming rejection of this nefarious practice is that most of your respondents are, in fact, happy to buy such stocks. You cannot make hypocrites of them if you wish them to vote with the angels.
The evidence is plain, nonetheless, that big fund managers, the kinds purportedly surveyed by the association, are quite happy to sup with the devil.
For instance, the tech giant Google has three classes of shares. They are "A" shares with one vote each, "B" shares with 10 votes each and now "C" shares with no votes at all.
This structure was adopted so that the two nerds who control Google can issue unlimited numbers of "C" shares for acquisitions while still controlling the company absolutely with their "B" shares.
And the register shows that the biggest holders of "C" shares, aside from the two nerds, are Fidelity, Blackrock, Capital Group, Vanguard, State Street, T. Rowe Price and JP Morgan Chase. Big institutional investors are not at all deterred.
Likewise, the largest shareholders of the "A" shares of Facebook, which have lesser voting rights than the controlling nerds' "B" shares, are Fidelity, Vanguard, State Street, Blackrock, Invesco, T. Rowe Price, etc., same mob.
They buy these dual structure shares because they think the disadvantage is sufficiently offset by a share price discount of usually about 4 to 5 per cent, not the average of 19 per cent that the association says its respondents would ask of a dual-share Alibaba.
Mind you, that 19 per cent may be the right discount for Alibaba. I personally sympathise with the respondent who wanted a 50 per cent discount.
Oh yes, one more tip on survey skewing. Make the particular the general in the summation of survey results that may favour your, so to speak, not unnecessarily overt agenda.
In this case, point out that HKEx chief executive Charles Li Xiaojia has suggested that it may be a good idea to give innovative companies special concessions. Then tar the entire exchange with it as if the listing committee had not turned him down on Alibaba.
So easy, just like toothpaste.
