A look at parts of system that are unfit for purpose

Enough of the banks already. We know what they did, and roughly why they did it. Time to inspect those other parts of the financial system that have turned out unfit for purpose.

Last week I discussed how brokers' analysts are not in a position to do fundamental long-term research, since that would not suit the purposes of the investment banks that employ them. Now let us turn to the fact that even if they could produce such research, fund managers could not act on it.

To illustrate the problem, take residential mortgage-backed securities. In the depths of the market panic back in March, I pointed out that prime Dutch RMBSs were priced on the assumption of 75 per cent defaults - which was obvious nonsense.

Since then, the price of such securities has rocketed. But while some specialist funds have snapped them up, the big institutions seem to have missed the boat.

Before asking why, consider a fundamental point about risk. Of the three main types - liquidity risk, market risk and credit risk - the first two are essentially short-term in nature. So, as long-term investors, the institutions should be indifferent to them.

To repeat an analogy I have used before, suppose a bank is offering 2 per cent interest on short-term deposits and 3 per cent on five-year ones. If you know you don't need the cash for five years, the extra 1 per cent is free money.

But the institutions have debarred themselves from such thinking. The vast majority, in the UK anyway, have long since farmed out their investing to fund management firms, which they typically employ on three months' notice.

Both fund managers and institutions will deny that this makes them short-termist, but their behaviour says otherwise.

Mortgage-backed securities aside, take a more mainstream example. In the credit bubble, it was obvious that companies that succumbed to investor pressure to leverage up - through share buy-backs, for instance - were a long-term sell.

That is because the pricing of risk varies radically through the cycle. In the boom, investors looked only at the apparent rewards of leverage. But it was wholly predictable that come the bust, they would switch their attention to its dangers.

The snag was, of course, that no one could predict when the bust would come. So however much fund managers accepted the logic, the nature of their contracts meant they could not put it into practice.

How did it come to this? Back in the 1980s, pension funds in particular lived in a different and simpler world. They put their cash flow into domestic blue-chip stocks and Treasuries, and sat on it. The results were a private matter for the sponsoring employer, and in any case were smoothed over time to adjust for market cycles. But in the 1990s bull market, pension funds became a source of profit. And in the ensuing bear market, people began to notice the contrarian results being achieved by alternative investments. Even conventional investment came to be split into different styles.

So pension funds were farmed out to specialists. The question then arose of how the trustees - whose job it now was - should appraise their performance. And given the near-impossibility of distinguishing ex-ante between long-term strategic brilliance and short-term bungling, that could only go one way.

The final blow came from the accountants, with pension deficits being marked to market and brought on to the balance sheet. The arguments about that are for another day. But long-termist it isn't. All this results in a dysfunctional system and a correspondingly systemic problem. The extreme market volatility of the past decade has various causes. But one, clearly, is the fact that the stabilising effect of long-term holders has been largely removed.

And sooner or later, volatility is a risk which has to be paid for. One practical effect has been the stampede by employers from direct benefit to direct contribution pension schemes, whereby the cost of volatility is shifted to the employee.

How to fix all this? The accounting issue apart, there is in the end only one solution, in two forms.

First, the institutions need to decide who they can trust with their money, and give them long-term contracts accordingly. The agreement would specify fund managers' targets over perhaps five years, and the penalties for failure.

Second, there should ideally be a corresponding shift to the model of private equity rather than hedge funds. That is, investors should commit their money for a fixed term and abide by the consequences. In return for that risk, they should of course be charged much lower fees.

As to how feasible any of that is, I pass. But in a world of surtaxes on bankers' bonuses, more things are feasible than we once thought. And if not now, when?

tony.jackson@ft.com