Join our community of smart investors
OPINION

Broker Forecasts: so wrong, they’re right!

Broker Forecasts: so wrong, they’re right!
October 9, 2015
Broker Forecasts: so wrong, they’re right!

As Nobel-Prize winning behavioural economist Daniel Kahneman succinctly puts it in his book “Thinking, fast and slow” , “Errors of prediction are inevitable because the world is unpredictable.”

In fact, what is perhaps more surprising than the fact that brokers get it wrong, is how much faith the market continues to put in forecasts when the evidence is so clear that they should be taken as a matter of opinion rather than fact.

Contrarian US investor David Dreman in his book “Contrarian Investment Strategies: The Next Generation” cites his study of over 100,000 consensus earnings estimates compared with reported figures over the 24 years to 1996 which found that over two fifths of forecasts were more than 15 per cent out.

Mr Dreman also highlights the fact that these errors are ripe for exploitation when he writes: “A large number of studies in psychology indicate that human judgement is predictably incorrect.”

If the type of forecast errors brokers make are predictable, yet the market still sets store by the forecasts, a situation exists that clued-up investors can make money from.

This is how the opportunity is explained by Ardevora Asset Management, a fund management firm set up by Jeremy Lang and William Pattisson to exploit broker forecast errors along with other behavioural finance quirks: “[Brokers] are over-confident about their forecasting ability and are resistant to information that contradicts their views. We study the past pattern of forecast errors for signs of systemic upward skew.”

Indeed, while the Woodford research shows that brokers on aggregate tend to be over-optimistic, they can also often persistently be over-pessimistic when it comes to individual stocks.

The graph below shows how the stratospheric ascent of Ashtead's shares from the end of 2009 through to earlier this year was accompanied by persistent upgrades to 2015 consensus earnings forecasts.

For nearly four years, I’ve been attempting to exploit this forecast-error anomaly with my “Great Expectations” screen. I am extremely happy to say my strategy has done very well for readers to date. Since starting the screen in December 2011, it has generated a cumulative total return of 174 per cent (average portfolio size of 14 stocks) compared with 37.4 per cent from the FTSE 350 - the index the screen is conducted on.

The screen’s criteria are:

■ EPS forecasts for each of the next two financial years upgraded by at least 10 per cent over the preceding 12 months

■ EPS growth of 10 per cent or more forecast for each of the next two financial years

■ Share price momentum at least double that of the market over the last year and better than the market over the past six months, three months and one month

I will update the screen for subscribers this December, but for now, here are last year’s picks and the returns they generated to the end of September.

GREAT EXPECTATIONS TOTAL RETURN 2 DEC 2014 - 30 SEP 2015

NameTIDMTotal Return
Taylor WimpeyTW.53.6%
UniteUTG43.1%
Barratt DevelopmentsBDEV41.3%
PersimmonPSN37.7%
BellwayBLWY36.4%
CarnivalCCL29.2%
International Consolidated AirlinesIAG27.6%
Howden JoineryHWDN25.7%
Big YellowBYG24.9%
GraftonGFTU5.9%
Dixions CarphoneDC.1.0%
Go-AheadGOG-4.3%
AshteadAHT-8.0%
Average-24.2%
FTSE 350--3.8%

This blog has been adapted from a presentation made by the author at the Investors Chronicle’s Going for Growth Private Investor Seminar on 8 October 2015