If there has been one lasting casualty of the collapse of the dotcom bubble, it has been sell-side research — the securities analysis undertaken by investment banks. Regulators have restricted it and investors have ignored it. RCM, a fund-management group, estimates that research budgets fell by more than 35% between 2000 and 2005. This month Prudential, an American financial group, said it would shut its research operation.
The main problem is that the business rarely pays. Regulators have sought to separate research from the corporate-finance arms of investment banks. In the dotcom era, analysts acted as if their opinions were for hire; if a company was planning to float or mount a takeover, they were happy to recommend the stock—provided their bank managed the deal.
Nowadays research is supposed to be paid for by commission income. But that is a problem now that investors can trade without using a broker, thanks to programme trades or order-matching systems. A survey by Greenwich Associates, a consultancy, found that the amount of commission dollars devoted to research has fallen by half in the past five years.
Research has also become more difficult. Analysts used to be the conduit between company managers and investors. But new regulations prevent them gaining privileged access to information. When it comes to number-crunching on shares, investors should have the computers to do it for themselves.
Things might have been even grimmer for the research departments of investment banks had it not been for hedge funds. A study by Booz Allen Hamilton, a consulting firm, found that whereas only 10% of large fund-management groups thought sell-side research was very important, more than 60% of hedge funds did so. Hedge funds trade a lot and they need new ideas; they are happy to pay handsomely for the best ones.
This means investors have split into two camps. The biggest fund-management companies, such as Fidelity, have their own research departments. Hedge funds rely on sell-side research. So, too, do smaller fund managers, but they often receive ideas only after they have been pitched to hedge funds.
One might expect buy-side research to be better than sell-side. It should not be subject to the conflicts of interest that affect the sell-side; the analysts should be interested only in finding the best stocks. Furthermore, because they do not publish their recommendations, the buy-side analysts are unlikely to face another problem that afflicts their investment-banking counterparts—angry chief executives who do not like sell notes on their stocks.
However, a recent study found that buy-side analysts were much less accurate in forecasting earnings than sell-side analysts and their buy recommendations performed much worse (although, bizarrely, the sell recommendations of the sell-side performed better than the buys).
The study focused on only one fund-management group. But the academics also compared the results of fund-management firms that claimed only to use their own research with those that used a mixture of in-house and sell-side analysis. Statistically there was no difference between the performances.
Why aren’t buy-side analysts better? One problem may be pay; analysts can earn a lot more at an investment bank. (And the best ones are increasingly being poached by hedge funds, the latest example being Henry McVey, Morgan Stanley’s chief American investment strategist.) Another issue may be the status of analysts within the fund-management industry. Traditionally, fund managers did their own research. That still happens at the smallest groups; attending (say) Tesco’s results presentation may not be the best use of a star manager’s time. Those analysts who were hired tended to be junior staff who did the grunt work; compiling the data that the portfolio manager needed. Most analysts saw their job as a stepping stone to managing money.
Those buy-side firms that give a more important role to analysts run the risk that they dominate the process, creating portfolios that the fund manager is obliged to pick from. This in effect takes the creativity out of fund management. There can also be the problem of “groupthink” when all the analysis is done in-house.
So outside analysis will probably still find a role for itself. Some of this will happen at investment banks, even if the service is a loss leader for the rest of the operation. Independent research boutiques are a better bet. After all, they live or die on the quality — and price — of their work.