FOOL SCHOOL
Using Broker Forecasts & Recommendations

September 9, 2005

Pick up any paper and read through the financial section and you're likely to see a lot of comment on price movements being caused by broker recommendations. Typically, a broker will label a company either Buy, Sell or Hold and whenever one of them changes its view on a company, upgrading from a Hold to a Buy for example, it can cause a movement in the company's share price.

As far as long-term investors are concerned, however, broker recommendations are about as useful as a chocolate teapot. The main problem is conflict of interest. Brokers are often part of larger financial groups and they don't want to upset existing or potential clients. This means Buy recommendations tend to vastly outnumber Sells. Another problem is that many brokers produce somewhat obscure recommendations along the lines of Accumulate, Outperform and Weak Hold. Finally, brokers are often very short term in their outlook and place too much emphasis on overly simplistic relative valuations (e.g. company A in one sector is valued at 20 times earnings therefore Company B, also in the same sector, should be valued at a similar level).

Although the recommendations themselves are practically worthless, the forecasts produced alongside them, estimating items such as sales, profits and dividends for future periods, are much more useful. You do need to appreciate their limitations though.

The numbers of brokers covering a company will depend on its size. Some small companies have just one broker covering them (some of the tiniest firms have none at all). Larger companies, in the FTSE100 for example, may have a couple of dozen brokers following their fortunes. The average of all current forecasts for a company is referred to as the 'consensus forecast' and this the figure is displayed on most company information pages (here is ours for Vodafone, for example).

Most of these forecasts are compiled with fairly heavy guidance from the company itself. The house broker (who is retained by the company to advise them on stock market matters) will often get the first indication from the company of what they are expecting and other brokers will tend follow their lead. Typically, forecasts are produced for the next two financial years, although some in-depth reports might look a little further ahead.

Often forecasts for the current year will be fairly accurate, say within 10% or so of the eventual outcome. Forecasts for smaller companies, or those in fast-moving industries, tend to be less reliable than those for larger, more mature businesses. Forecasts for subsequent years tend to get less and less reliable with quite astonishing speed. Most brokers will update its forecasts every few months or so, meaning the consensus forecasts will shift throughout the year and are likely to become more accurate as you get closer to the end of the period they relate to.

If you want to place reliance on forecasts it pays to look at the detail behind the consensus figures, although this information can be hard to find. Some of the individual forecasts that make up the consensus figure might be months out of date and, especially if they predate any recent result or trading statements, they could be significantly wide of the mark. One way round this is to reconstruct your own consensus figure by looking at each individual forecast and discarding old figures. Although this takes more time, and some practice, it is usually worth the effort.