Abstract

Institutional investors are often criticized for their insufficient “engagement” with listed companies. Actually, information and other transaction costs limit their capacity to actively monitor investee firms, and to engage with their management. A partial solution is offered by Proxy Advisors (PAs), providing proxy voting services on a subscription basis. We use Say-On-Pay (SOP) in Italian listed firms to investigate PA activity and (institutional) shareholder voting behavior. Making use of a unique dataset, including information on how each shareholder voted in 2012, we analyze the behavior of different classes of shareholders and their relation with PA recommendations. Our main results can be summarized as follows: i) while shareholder dissent on SOP is low, in line with what happens in other developed countries, dissent by institutional investors is surprisingly high; ii) the voting behavior of institutional investors is strongly correlated with PA rec-ommendations. The influence of PAs is higher on nonblockholders (mostly internationally diversified pension and mutual funds) than on blockholders; iii) preliminary estimates show that the influence of PAs in Italy is at least as strong as (and probably stronger than) that observed in the US. This is coherent with the weight of non-domestic institutions, which can be hardly expected to independently analyze thousands of firms; and it is especially so in Italy, where most listed firms may be classified as small/medium cap firms on a comparative basis; iv) institutional investors, however, do not take PA recommendations at face value, but focus on the specific reasons of concern underlined in the reports (in particular, on the structure and long-term value creation of the remuneration policy). Severance pay has, by far, the largest impact.

We thank International Shareholder Services (ISS) and Glass, Lewis & Co. (GL) for providing us with 2012 proxy season reports for Italian listed firms. We also thank Giovanni Siciliano and Nadia Linciano for useful comments and Stefano Libera for research assistance. Any mistake remains, of course, our sole responsibility. Opinions expressed in this paper are exclusively the authors’ and do not necessarily reflect those of Consob.