4.6 Human Resources Dedicated to Corporate Governance

The underlying concern of policymakers about institutional investors' low rate of deviation from their proxy advisors' voting recommendations generally is that investors fail to take part in the govemance of the companies in their portfolio (for a discussion of varieties of shareholder engagement, see Winter 2011). In this final section, we examine the human resources that funds devote to corporate govemance. As Table 6 shows, the funds in our sample have, on average, 12 staff devoted to corporate govemance. The median number of staff is markedly lower (8), due to one outlier.1 This suggests that the median fund, which holds stakes in 1,407 portfolio firms, has one staff member available to monitor about 188 portfolio firms. In fact, the actual number of firms per staff member is higher, for two reasons. First, we have only looked at US and European portfolio firms and a substantial part of each fund's assets is invested in emerging markets. Second, not all members of the fund's govemance team make voting decisions. The team typically also consists of members who specialize in engagement, focussing on a particular theme (e.g. environmental, social or govemance), asset class (e.g. equities or real estate) or region. Coupled with the fact that shareholder meetings are typically concentrated in the spring, we can see how the amount of time that staff members can devote to making a voting decision is limited. Thus, the need to prioritize becomes apparent. The previous section has shown some of the criteria that funds seem to be using in deciding on the amount of resources they dedicate to different types of voting decisions.

The members specialized in voting are in charge of giving the actual voting instruction, which they typically do through the logistic voting platform provided by the proxy advisor. Before giving a voting instruction, the voting specialist will have to decide how to vote. In doing so, the voting specialist may use the proxy advisor 's voting recommendation as input, which is typically shown on the screen next to management's recommendation. He or she may also use other inputs, such as information obtained from the meeting agenda, the company's website or annual accounts, analyst reports, colleagues within the governance team (such as the team leader in case of significant votes or colleagues who are engaging with the portfolio firm), or other institutional investors. In addition, the voting specialist may obtain information from the portfolio manager. Nathan & Metha (2010) are sceptical of the divide between voting specialists and portfolio managers, arguing that it results in voting decision-makers being overly focused on corporate governance standards and insufficiently on firm-specific information that portfolio managers consider relevant in making investment decisions. Most funds in our sample claim that there is frequent communication between their governance specialists and portfolio managers, particularly regarding proposals that have potentially significant value implications.

Finally, the voting specialist may obtain information from management of the portfolio firm. The funds claim that communication between management regarding items on the meeting agenda takes place occasionally, particularly when the fund is inclined to vote AGAINST management and seeks confirmation from management that it properly understands management's position.2 This sort of ad hoc communication can be distinguished from `structurar engagement with the portfolio firm, which is usually (but not always) conducted by other members of the governance team who are specialized in engagement. These engagements typically relate to one or more of the themes mentioned earlier: environmental, social or governance. These themes may or may not correspond with items on the meeting agenda. To the extent they do not, it suggests that the issues shareholders get to vote on are not necessarily the issues that they consider most relevant.