Financial Times; Oct 16, 1999

Just why are portfolios turned over so often?

Churn: Transaction costs arising from so many deals are costing investors plenty, says Ian Orton

"Inactivity strikes us as intelligent behaviour. Neither we nor most business managers would dream of feverishly trading highly-profitable subsidiaries because a small move in the Federal Reserve's discount rate was predicted or because some Wall Street pundit had reversed his view on the market. Why, then, should we behave differently with our minority positions in wonderful businesses?" - Warren Buffett, 1996.

A similar question could be asked of UK fund managers. If the stocks they pick for the portfolios they manage on behalf of investors are so wonderful, why do they turn over their portfolios so often?

Precise statistics covering the buying and selling activity of UK funds do not exist. But a casual inspection of the reports and accounts of a reasonable cross-section of unit and investment trusts shows that annual portfolio turnover rates of between 70 and 100 per cent are not untypical. In other words, fund managers are prepared to hold a stock for just over a year before trading it for something else.

This practice is not unique to the UK. According to John C. Bogle who founded Vanguard Group, a leading US fund manager, the annual portfolio turnover of the typical US general equity fund averaged about 17 per cent from the 1940s to the mid-1960s. In 1997, however, average turnover of US equity funds stood at 85 per cent, a fivefold increase.

This seems to be light years away from the "buy and hold" philosophy that has served Warren Buffett and the shareholders of his Berkshire Hathaway investment company so well. It is also at odds with the advice given to prospective investors by the UK mutual fund sector. Buy and hold seems to be a good idea for the investor when it comes to buying a financial product - but not for the fund manager working on that investor's behalf.

Fund managers are not short of reasons to justify a more active approach to management. The most obvious reason is that it helps to boost performance. A more active - or even hyperactive - approach to stockpicking enables managers to make effective use of the investment research produced by an army of in-house and external analysts.

This can, for instance, identify market pricing anomalies or highlight companies with as yet unrealised earnings potential that might boost their share price. Alternatively, it might identify companies whose prospects are less rosy and shares are obvious sell candidates.

"Some funds have an aggressive growth mandate and this invariably results in greater trading activity as fund managers attempt to exploit special situations," says Anne McMeehan at the Association of Unit Trusts and Investment Funds (Autif).

A less obvious reason is that a manager might be replaced by one with different ideas about the way a portfolio should be structured. This always produces an increase in turnover.

Of course, if all this increased activity actually benefits the fundholder, there should be few complaints. The problem, however, is that there is little conclusive evidence that it does.

With one or two exceptions, managers seem to be very reticent about sharing their wisdom with investors. This is unfortunate, because high portfolio turnover can be quite costly - and this, ultimately, drags down fund performance.

Studies suggest that fund transaction costs can vary between 0.5 and 2 per cent a year in the US, the precise figure depending upon the size of the fund and the level of turnover.

While there are no comparable figures for the UK, some experts estimate that a mainstream fund that turned over its portfolio by 100 per cent during the course of a year would incur transaction costs of about 1 per cent, excluding stamp duties. This figure could be higher for those that invested in foreign markets.

Given the good returns posted by most funds during the 1990s the impact of these costs on overall fund performance has been relatively small. But, if equity returns revert to their long-term trend, then the situation could be different.

According to an equity-gilt study produced by Barclays Capital, UK equities have generated annual real returns of about 8 per cent over the past 80 years. If the costs associated with a 100 per cent portfolio turnover rate for a UK fund do amount to 1 per cent, this means that investors could be losing up to 12.5 per cent of the available return in transaction costs alone.

In future, managers will have to justify these costs by showing that they really do add value.

Your rough guide to transaction costs 

Portfolio-related transaction costs are an invisible item as far as UK unit and investment trusts are concerned. Examine the annual report and accounts and you will find little mention, writes Ian Orton. Even so, it is possible to estimate the fund's portfolio turnover rate.

The annual reports and accounts of UK authorised unit trusts provide a summary of material portfolio changes during the year. These provide details of total fund purchases and sales. The turnover rate is simply the lesser of purchases or sales of portfolio investments as a percentage of average fund assets.

If, during a year, a £100m fund bought £100m of stocks and sold £100m, its turnover would be calculated as 100 per cent, although the fund executed £200m of purchases and sales in total.

It should be remembered that the figure produced by this calculation is only approximate. Remember that open-ended funds, such as Oeics and unit trusts, will probably be taking in new money over the year as fresh investors buy into the fund or existing ones increase their holdings. Alternatively, the fund will also be paying out money as investors redeem all or part of their holdings. This will have an impact on portfolio purchases and sales and - by extension - the portfolio turnover rate. This is very much a first step to estimating transaction costs.