Custom Proxy Advice and the Architecture of Shareholder Voice

A review of “Custom Proxy Voting Advice” by Edwin Hu (University of Virginia), Nadya Malenko (Boston College, NBER, FTG, CEPR, and ECGI) and Jonathon Zytnick (Georgetown University). The prizewinning paper of the 2025 John L. Weinberg/IRRCi Research Paper Award. 

Here’s how proxy voting could work: Asset managers carefully consider thousands of proposals across thousands of companies, evaluate the merits of each, and cast their votes in a way that reflects both fiduciary duty and broader stewardship principles. The sheer scale of this task suggests that many asset managers may seek ways to simplify their decision-making process.

In practice, they rely on proxy advisors. These firms—like ISS or Glass Lewis—supply voting recommendations that streamline the voting process. Critics argue this makes institutional investors passive, outsourcing their influence to third-party advisors with their own views. But the paper Custom Proxy Voting Advice by Edwin Hu, Nadya Malenko, and Jonathon Zytnick, recently awarded the 2025 John L. Weinberg/IRRCi Research Paper Prize, shows that this common narrative is too simplistic.

As the authors discussed in a blog post on the Harvard Law School Forum, we now know that proxy advice today is not a one-size-fits-all service. Around 80% of funds in the sample receive customised proxy recommendations. These differ from benchmark advice in over 20% of ballots. The process is straightforward: before the proxy season, a fund communicates its ideological preferences—say, stricter views on ESG or a more lenient stance on director overboarding. The proxy advisor encodes those views into a policy engine. Then, as votes roll in, the fund can accept the customised recommendation or manually override it.

In an ECGI interview with Tom Gosling, Prof. Malenko elaborates on how this system works in practice. Some investors opt for bespoke custom policies; others subscribe to ready-made thematic options like “climate” or “governance-first.” The key point is that these aren't passive defaults—they are expressions of strategy. What’s more, the researchers track when and how funds vote differently from the advice they are provided by analysing voting timestamps. If a fund casts its vote on a different date than its auto-submission setting, that’s a manual intervention—a signal of attention.

The theoretical framework underpinning the paper is where things get particularly interesting. The authors model proxy voting as a two-stage process. First, a fund decides whether to invest in customisation. This requires upfront costs—time, money, and effort—but yields voting advice aligned with its worldview. Second, it receives recommendations and decides whether to investigate specific items further. This structure introduces two key mechanisms: substitution and complementarity.

In non-controversial cases, customisation acts as a substitute—removing the need for further scrutiny. But in high-stakes or contentious cases, it frees up cognitive and organisational bandwidth. This is where the complementarity kicks in: funds that customise are actually more likely to pay attention when it counts. They use proxy advisors not as crutches, but as intelligent filters.

This insight flips much of the regulatory narrative on its head. Rather than seeing auto-submission as disengagement, this paper shows that it may facilitate more effective stewardship. When done right, customisation doesn’t replace accountability—it enables it.

The empirical findings are robust. The majority of investors customise their policies while still devoting scrutiny to key votes. This undermines the stereotype that proxy advisors impose uniform ideologies and that funds are blindly following proxy advice. Instead, they function as vote engineers—building the infrastructure through which institutional investors can express diverse priorities at scale.

Among the paper’s most valuable contributions is its recognition of attention as a scarce resource. Stewardship isn’t just about conviction; it’s about capacity. And customisation, it turns out, is a rational way of allocating that capacity. Investors who customise are better able to prioritise high-salience decisions without drowning in routine ones.

For policymakers, the implications are clear. Banning pre-filled ballots or discouraging auto-submission may sound like a win for active oversight, but could actually backfire. These measures might erode the customisation infrastructure that allows investors to reflect their values. And when investors default back to generic benchmarks, true diversity in corporate governance outcomes may shrink, not grow.

Ultimately, Custom Proxy Voting Advice helps reframe how we think about proxy advisors—not as ideological monoliths, but as scalable enablers of institutional differentiation. It is this quiet shift—from advice to infrastructure, from standardisation to expression—that marks a deeper evolution in shareholder governance.

-------------------------------------------

The ECGI does not, consistent with its constitutional purpose, have a view or opinion. If you wish to respond to this article, you can submit a blog article or 'letter to the editor' by clicking here.