One of the ideas that has bubbled up as a result of the financial crisis - and the apparent failure of shareholder oversight within it - is to go back to the drawing board in terms of portfolio construction. The argument runs something like this, because institutions have a wide range of equity holdings (in order to diversify risk) there is realistically no way that they can act like an 'owner' of all of them. They also arguably don't really have much of an interest in doing so precisely because they appear to spread their risk, hence they can even just about handle BP tanking because their overall exposure is low. Therefore in a world of widespread portfolios, shareholder oversight is undermined.
So what's emerging instead? Well some argue, and I admit my technical knowledge here is non-existent, that once you get past a certain number of stocks the benefits of diversification (in terms of risk reduction) tail off pretty sharply. Therefore holding several hundred, or even thousand, lines of stock might not do you any good anyway. As as a result I've heard a few people argue recently that a good ownership-oriented portfolio would basically be small in number of holdings, but with big weightings in each stock, and with the plan being to hold for the long term. The institution would make particular efforts to get to know the management and forge decent relationships on the basis of a shared understanding that this was a long-term commitment.
All well and good, but what does that remind you of? To me it looks similar to the private equity governance model, just diluted a bit. Diluted, that is, by the fact that you don't own the company outright, just a sizeable slug of it, and don't have a load of debt focusing your interest. Or to provide a more accurate comparison, isn't this the sort of portfolio an asset manager might have run 15 or 20 years ago?