The Commonwealth Bank is Australia’s biggest financial conglomerate. It is also the most widely owned of the big four banks, with 780,000 shareholders.
Computershare data from 2013 shows that only about 5% of shareholders bothered to vote at ASX 50 AGMs. On average, about 10% of these appoint the Australian Shareholders’ Association as their proxy representative.
For instance, at the recent CommBank AGM in Melbourne, the ASA represented 2463 shareholders who collectively owned 6.41 million shares worth about $530 million.
This sounds big, but ASA’s voting power is still swamped by the big institutional shareholders who now routinely vote their stock on all major resolutions at big AGMs.
For instance, ASA chose to abstain on my recent run at the CBA board given our recent association, but if they’d voted undirected proxies in favour, the overall “for” vote would have increased from 3.15% to about 4%. In other words, an ASA switch from “against” to “for” would have shifted the dial by about 1.7% at the CBA AGM.
Unfortunately, a lot of Australian retail shareholders don’t bother to vote because they feel swamped by the big end of town.
However, that would all change if companies started to follow this recently adopted ASA policy (page 11) on voting disclosure:
“A useful additional disclosure to gauge retail shareholder sentiment would be for companies to disclose, both at the AGM and later to the ASX, how many shareholders have voted ‘for’, ‘against’ and ‘open’ on a particular resolution.”
Some smaller companies already voluntarily do this, as can be seen in this recent example by NSX Ltd. The ASX was told that the remuneration report was passed by 99.4% of voted shares, but NSX also revealed that this comprised just 11 shareholders voting in favour and eight voting against.
The disconnect between retail and institutional shareholders was graphically demonstrated at the recent Fairfax Media AGM, where a majority of the voting shareholders supported my election to the board, but this only translated to a pathetic 0.92% of voted shares. Thank you Gina, assorted institutions and all three proxy advisers for the support!
Fairfax was good enough to voluntarily send through the full data set (see here), which ranked the four board candidates as follows in terms of retail shareholder popularity:
Todd Sampson: 76.92% support (1050 for, 315 against)
Peter Young: 73.53% support (1003 for, 361 against)
Stephen Mayne: 64.57% support (902 for, 495 against)
Roger Corbett: 60.46% support (852 for, 556 against)
However, 556 against voters at Fairfax didn’t stop chairman Roger Corbett being re-elected with 94% of directed proxies in favour.
Seeing as this sort of disclosure was now ASA policy and Fairfax had set a good precedent, I tried the same request with the Commonwealth Bank but received a one-line reply from company secretary Margaret Taylor earlier this week which read as follows: “The information which we make available on the voting is that released to the ASX.”
Inside the defensive CBA bunker, I suspect the data shows that I was more popular with retail shareholders than the least popular incumbent director Launa Inman, who struggled home with 98.63% of the voted shares in favour.
City of Melbourne councillors recently unanimously endorsed this open data policy, which commits to voluntary disclosure of as much useful contextual data as possible.
It’s a shame our biggest bank doesn’t share the desire to let the sun shine in.
CBA will no doubt eventually cave in September next year when they get another external board nomination that includes the following paragraph:
“If CBA releases the voter data related to the 2014 AGM, as was first requested in November 2014, I will withdraw this nomination and thereby not require the bank to publish a platform related to the virtues of open data in its 2015 notice of meeting.”
A similar request for the voter data has been made to Woolworths, but the supermarket giant is still thinking about it. Logic would suggest this is something that the ASX Listing Rules or the Corporations Act could very easily accommodate with a minor amendment.
After all, if voter sentiment must be captured for all votes mergers done by way of on schemes of arrangement (see how Westfield did it in May this year), why not release the same data on all AGM resolutions?
At the very least, it would lift the current pathetically low participation rates because punters could see their voting intentions reflected in the data and companies would respond to media coverage by more actively seeking the support of retail investors on key resolutions.