Poor say-on-pay votes behind shareholder engagement in US

Analysis of a dataset specially prepared for RI by Equilar.

In a study of S&P 100 companies, research firm Equilar found that disclosure of shareholder engagement has gone from 2 in 2011, the first year of Say on Pay, to 55 in 2015. The number of companies jumped to almost a third of the sample in the second year of Say on Pay – the advisory vote to approve executive pay in the US.
There has not only been an increase in disclosing shareholder engagement. A dataset specially prepared for RI by Equilar shows that there is a strong correlation between engagement disclosure and poor say-on-pay votes. A poor say-on-pay vote is defined as anything from minority support to 75 per cent support. Almost all votes receive 90+ per cent support in the US. As significant is the finding that companies disclosing shareholder engagement received lower average levels of shareholder support of pay practices in every year since the votes began.
Matthew Goforth, Equilar’s Research and Content Specialist, concurred with the conclusions RI reached based on the data. While the first year of engagement disclosure did not necessarily follow on from a low level of support for Say on Pay in the prior year, the pattern was pretty clear.
In 2012, for example, this included Goldman Sachs which disclosed shareholder engagement following a 72.9 per cent support for Say on Pay in 2011, and Hewlett-Packard which engaged following a failing vote of 48.2 per cent. HP’s support has been climbing steadily since. It reached 90 per cent in 2014, and the company disclosed no engagement in 2015.The trend continues. In the latest year, Ford, WalMart, pharmacy chain CVS, and Citigroup all disclosed shareholder engagement following a dip in support in 2014. Others appear to have introduced it simply because it is fast becoming not just best, but standard practice, such as IBM and pharmaceutical group Abbvie.
Patterns are not simple, however. Johnson & Johnson saw a vote of 60 per cent in 2011, disclosed shareholder engagement in 2012, but received a lower vote in that year of 55 per cent, showing it is often not enough to deflect further criticism. It would be interesting to see a correlation between proxy advisory vote recommendations, say-on-pay votes and the relationship between pay and performance. Clearly, simply talking to shareholders is not enough. Action is needed in many cases.
For example, Occidental Petroleum began disclosing shareholder engagement in 2012 but received only 62.8 per cent support in 2013; another sign that engagement is not the panacea for shareholder unrest on pay. And Pfizer has disclosed shareholder engagement every year since Say on Pay was introduced. Even this did not help in 2011, when it received pay support from only just over half its shareholders.
Other companies – Simon Property Group is a good example – engage with shareholders for different reasons. Although Simon was facing heavy shareholder opposition from 2012 onwards, it was also facing a lawsuit challenging an excessive “retention” stock award to CEO David Simon, the son of the founder.

Pay fortunes go up and down. Telecommunications company Qualcomm saw only 68.5 per cent support in 2012, and disclosed engagement the following year. Support returned to top levels for a couple of years – with no engagement disclosed – but it fell to 55 per cent last year. It would appear more than likely that shareholder engagement will be disclosed this year.
One would expect the behaviour of controlled companies – companies where founders control a majority of the voting stock – to be similar, but that is not the case. For example, cable company Comcast and Facebook are both controlled companies and only have a Say on Pay vote – as is permitted – every three years. Both have also experienced high levels of support for executive pay, but Comcast has disclosed shareholder engagement since 2013, while Facebook has not engaged at all. This year may be different for Facebook which might seek support after losing a legal battle with a shareholder over non-executive director pay.
Goforth commented that while shareholder engagement was focused on pay in the early years of Say on Pay, it has expanded lately to include more diverse topics such as succession planning for the CEO and broad strategy issues.Of course, we know that disclosing the fact of shareholder engagement does not necessarily equate to quality engagement. Oracle provides a good example. The company disclosed engagement in its last two annual reports, but a majority of its shareholders voted against pay in the last four years running. It would appear that the company must either have met with shareholders who already supported its pay practices, or that these conversations did not go well.
Equilar’s Goforth added that it is sometimes difficult to understand what the particulars of shareholder engagement consist of, since disclosure is variable. While the annual report is the most common source for engagement disclosure, sometimes companies’ investor relations pages contain more detail. Engagement can consist of anything from a survey of major investors to board members meeting with a group of the company’s largest shareholders. It could be a letter writing campaign for support, or telephone calls (as we saw with Bank of England’s campaign for support for the recombination of its chairman and CEO positions). While the involvement of non-executive directors in shareholder engagement is increasing in the US, noted Goforth, their participation is still in the minority, with most engagement conducted by the corporate secretary or management or both.