By Theodore Benjamin Kogan, ShareAction supporter

22 September 2015

When the results of the Crimea referendum showed that 97% of those who voted were in favor joining Russia, we all knew why. You just don’t win by that much in popular votes without fraud.

But in corporate governance, despite a ‘clean’ voting process, these kinds of figures are normal. In fact, for all 2014 votes of the FTSE 100, the average support for management proposals was also 97%.

There’s a variety of well-explored reasons for this. Individual shareholders don’t really bother with voting; institutional shareholders have no real incentive to actively use their votes, and so on. All of this contributes to so-called passive voting, whereby shareholders side with management by default.

The mounting body of evidence that suggests that voting – corporate governance’s central accountability mechanism – is malfunctioning is worrying. It stresses the importance of understanding what differentiates those shareholders who effectively practice true corporate stewardship from those who don’t. That’s why I decided to focus on this for my Master’s thesis for the MSc in Public Policy at UCL.

In the course of my research, I came across ShareAction’s report on the different disclosure levels of various UK asset managers. The Stewardship Code’s ‘comply or explain’ practice meant institutional investors were in three different tiers with respect to voting disclosure: those who chose to explain (i.e. no disclosure), those who chose to comply (i.e. disclosure to the minimum recommended standard, which is to disclose the voting record), and those who chose to go beyond minimum disclosure requirements (i.e. disclosure that includes not just the voting record, but also voting rationales).

These tiers in voting practices seemed worth investigating. On a strictly theoretical level, it seemed intuitive that institutional investors who actively put effort into greater disclosure than is necessary would also put more effort into voting itself, and that institutional investors who put effort into minimum disclosure would put more effort into voting than those who did not put effort into disclosing anything.

I called the five full-disclosure investors Tier 1, the sixteen partial-disclosure investors Tier 2, and the remaining non-disclosure investors Tier 3, and began collecting data on remuneration-related proposals. I focused on remuneration because it receives the least consistent and the weakest support from shareholders – though at 91% on average, the support is still overwhelming. I identified 206 such proposals by the FTSE 100 in 2014.

The results showed that Tier 1 is indeed comprised of much more ‘active’ voters on the whole (while the difference between Tier 2 and Tier 3 is marginal).


One implication of this is that 97% support for management should not be normal if investors put real effort into corporate monitoring. If Tier 1 opposes management 20% of the time, and abstains 8% of the time, simply because the institutions that constitute it care more about being proactive and responsible, then perhaps others should aspire to that standard – it’s not an unreasonable one to set.

Tier 1’s concerns don’t seem to be particularly frivolous. Among given rationales for opposition to management are: “we do not consider continued employment to be an adequate performance condition for awards upwards of 200% of salary” and “a vote against this proposal is warranted [because of] severance payments to […] Pier Luigi Foschi who had oversight over health and safety at the company during the Costa Concordia disaster”. To verify that concerns systematically weren’t frivolous, I looked at the rationales for dissenting votes, categorized objections to remuneration policies, and counted the number of times each type of rationale was evoked by Tier 1 investors.


Across all remuneration-related resolutions, the most prominent reasons for opposition to management were a poor pay-to-performance linkage, inappropriate (usually excessive) levels of pay, inappropriate discretionary payments, poor disclosure, and inappropriate (usually excessive) pension arrangements. Those rationales with conceivably less impact on shareholder value, such as overly complex remuneration schemes, an excessive focus on the short term, and lack of independence of the remuneration committee, are all invoked far less. And when, for each resolution, I tally up the maximum number of times a justification is given by the five Tier 1 asset managers, I find that it is the most reliable predictor of Tier 2 and Tier 3 voting behavior. This means that the more Tier 1 managers agree that a resolution is problematic, the likelier Tiers 2 and 3 are to vote against – though at a lower overall rate of opposition.

So how do we get Tier 2 and Tier 3 to emulate Tier 1? I don’t know. The Times’ Investor Democracy campaign, encouraging institutional investors to listen to their clients on matters of voting, could be part of the answer. Organizations like ShareAction will also be instrumental in bringing it about. But I trust that, one way or another, it will happen. Perhaps someday in the not-too-distant future we will regard 97% average support for management with the same disbelief as we regard the 97% support for the Crimea-Russia unification referendum.

Thanks Ben! If you have any questions for Ben or would like to see his Master’s thesis, you can get in touch at