Passive ownership has outgrown the governance model it inherited

The index fund is the most successful consumer financial product ever built. It took the cost of owning the equity market from something like 150 basis points a year to something like three, and it did this without a single advertising campaign that anyone remembers. I have spent thirty years watching financial products that promised to help investors and mostly helped their sponsors. This one delivered. That is worth saying before anything else, because what follows is a complaint.

The complaint is not about fees or performance or concentration in the index itself. It is about what happens to the votes. An index fund buys shares because a rule says to, and those shares come with the right to elect directors, approve compensation plans, and decide contested transactions. The fund did not want that right. It cannot decline it. And the volume of it has grown to a size that the people responsible for exercising it were never staffed to handle.

The arithmetic nobody planned

Consider what ownership of a typical large US listed company looks like now compared with a generation ago. The change was not engineered by anyone. It is the accumulated residue of millions of individually sensible decisions to stop paying for active management.

Twenty seven percent held by managers who are structurally indifferent to the fortunes of any single company. That is not a criticism of the managers. It is the definition of the product. An index fund cannot sell a company it disapproves of, because selling would be a tracking error, and tracking error is the one thing an index fund is genuinely forbidden to have. Its only instrument is the vote.

What a stewardship team can actually do

A large index manager's stewardship function might employ several dozen people. Those people are responsible for voting at thousands of companies, most of the votes falling within a proxy season that runs a few weeks. The maths of this is not in dispute and the managers do not really dispute it. They publish voting guidelines, apply them consistently, escalate the small number of genuinely contested situations, and hold a few hundred engagement meetings a year with the largest holdings.

That is a reasonable response to an impossible assignment. It is also, unavoidably, governance by policy rather than governance by judgment. A voting guideline says that a board should have a majority of independent directors, or that a compensation plan should meet certain structural criteria. It cannot say that this particular chief executive is drifting, or that this particular board has stopped asking hard questions, because knowing that requires the kind of attention that a concentrated active investor pays and a diversified passive one cannot.

We have replaced a system where a few hundred people knew their companies badly with one where a few dozen people know thousands of companies not at all. I would not claim the old system worked well. I would claim it failed in ways that were visible.

Owen Castellane, senior strategist at Bellweather Macro

The proxy adviser problem, restated

The usual next move in this argument is to blame the proxy advisory firms, and the usual next move after that is to point out that the largest index managers now vote their own guidelines rather than following a recommendation. Both moves are correct and neither settles anything. Guidelines written centrally and applied mechanically produce the same outcome as an outsourced recommendation applied mechanically. The location of the rulebook changed. The mechanism did not.

What has genuinely changed is that the guidelines are now a political object. Managers have been pulled into public arguments about whether their voting policies reflect one set of values or another, and the result is a defensive crouch in which the safest policy is the one that attracts the fewest letters. Governance conducted to minimise controversy is not governance.

What would actually help

1. 01Offer several genuinely different default voting policies within the same fund and let holders pick one at account opening, so that the concentration of a single house view dissolves without requiring anyone to vote on individual resolutions.
2. 02Require managers to disclose engagement outcomes rather than engagement counts. The number of meetings held tells you nothing. What changed as a result tells you everything.
3. 03Lengthen the proxy season, or stagger meeting dates across the year, so that the work of considering thousands of ballots is not compressed into six weeks by an accident of convention.
4. 04Stop treating the vote as the whole of ownership. Boards respond to sustained attention from holders who understand the business, and the vote is the crudest available expression of that attention.

I do not have a great deal of confidence that any of this happens soon. The index managers did not ask for the responsibility and gain nothing from expanding it, since stewardship is a cost centre in a business built on eliminating costs. Company boards are not lobbying for more engaged owners. And the investors who benefited most from the shift to indexing, which is very nearly all of us, have no mechanism to express a preference about the governance arrangements that came with it.

The awkward truth is that we bought a cheap, excellent product and got an ownership structure thrown in that nobody designed and nobody is accountable for. It has not blown up. It may never blow up. But when a governance failure eventually arrives at a large company, and one will, the question of who was supposed to be watching will have an unsatisfying answer, and everyone involved will be able to point truthfully at somebody else.