Oh by the way, that Milken video -- of a panel moderated by Ilana Weinstein and featuring Cohen, Cliff Asness and Neil Chriss -- is long but worth watching both for some good insights on how to think about hedge funds but also, and especially, for Cohen's hilarious innocent hands-in-the-air disclaimers that he just runs a family office and has never even heard of hedge funds.

Elsewhere, Cohen's family office "is starting a venture capital unit," called Point72 Ventures, "to fund and help develop financial technology for asset managers." And: "Berkshire business model is simple and effective, yet rarely copied." And here is Ben Carlson on the scourge of investment consultants:

As a group, hedge fund performance has been abysmal over the past decade or so. Consultants and those picking the consultants don’t receive nearly enough blame for this. The hedge funds are, in most cases, just giving these large institutional investors what they ask for (usually a strategy to fight the last war). Consulting firms pitch their “access” to top performing funds, but that’s usually the funds that outperformed in the past, not to be duplicated in the future.

ROIC.

Look, any financial metric can be misleading in some circumstances -- "I could write a paper on perverse ways you could destroy your company by raising your ROIC," says Aswath Damodaran, tantalizingly, about return on invested capital, which the Wall Street Journal calls "all the rage" among financial metrics -- but the fundamental question is, should companies be thinking quantitatively about whether the projects they invest in are likely to return more than their cost of capital, or should they not be thinking about that? There are definitely cases where the answer is no! Starting Facebook in a dorm room: probably not a decision driven by expected-return spreadsheets. But for most regular public companies that mostly invest in actual business projects that are supposed to make money in the reasonably foreseeable future, it is weird not be judged in some quantitative way on how much money those projects make and how much you spent on them. That is just basic stuff. And yet:

The focus on ROIC helps explain why activists push decisions sometimes labeled short term.

Activists said they aren’t inherently opposed to investment projects, but that companies have to justify spending. If a plant or a new line of business falls short on expected returns, companies should find a different project or give the cash back to shareholders, they said.

One way to think about the short-termism debate is that many public-company investors really do want their companies to invest mostly in projects that are likely to return more than their cost of capital in the foreseeable future. This is a very reasonable demand for almost all projects at almost all companies. There are exceptions: Some companies are engaged in real moonshots that have only a low probability of working out, many years from now, but if they do work out they will be transformational. (These projects could have a high ROIC too, but that might be harder to prove.) Probably a lot of those companies shouldn't be public companies. But, again, there are exceptions; Google and Amazon seem to have earned the market's trust that they can both run core businesses economically and invest in long-term opaque projects. (And there are sometimes controversial moves the other way; e.g., the pharmaceutical business is mostly about collecting portfolios of moonshots, while Valeant's innovation was to get rid of the moonshots and focus on foreseeable return on capital.)

The question is often: Is this management team one of the exceptions? If a company wants to invest in something whose economic benefits it can't quite justify numerically, and shareholders object, does that mean that the shareholders are short-term idiots? Or does it mean that the managers are undisciplined empire-builders? I don't think there is a general answer to that question, though I tend to sympathize with the ROIC people: Most public companies probably should be trying to earn their cost of capital most of the time.

Bloodchain.