Each day, terabytes of broker research clogs the email inboxes of money managers. But how reliable are these multitudes of forecasts? New research in the Australian and the US markets helps answer that question.
Ploughing through broker calls is a professional hazard for those money managers who base their value proposition on the notion that they can make consistent returns for their clients by exploiting perceived market mispricing.
The hope is that among the volumes of forecasts will be the gem that will earn investors big money. But it seems there is growing disillusionment among forecasting-based fund managers about the quality of the broker calls.
Australian financial markets research specialist East Coles regularly surveys fund managers about the research they receive. Results of its most recent Best Broker poll featured in The Melbourne Age.1
The overwhelming impression left by the findings was that managers’ reliance on analyst recommendations has diminished over the years, in part because of concern over inaccurate forecasts and in part due to the perception of potential conflicts of interest between analysts and the investment banking and brokerage businesses of their employers.
"We rely on it [research] a bit,” one fund manager is quoted as saying, “which is unfortunate because it has been wrong throughout 2009."
Fuelling the cynicism was an absence of warnings over a number of high profile corporate collapses in Australia during the financial crisis, including the failures of Allco Finance, Babcock & Brown and ABC Learning Centres.
For instance, in September 2007, an analyst at a major investment bank reaffirmed his ‘buy’ rating on ABC Learning Centres after the company announced plans for further global expansion and said it would lift fees.2
ABC Learning at that time was one of the world’s biggest listed operators of child care centres, with more than two thousand centres operating in Australia, New Zealand and the United States.
Within a few months of that ‘buy’ rating being issued, ABC’s shares had collapsed from above $5 to zero and the company went into administration by September 2007 under the weight of $1.5 billion in debt.
In January 2008, an analyst from a high-profile Australian investment bank issued a report with an “outperform” recommendation on the Queensland property development group MFS and a 12-month price target of $7.15 from the then price of just under $4. A week later, the shares had collapsed by 75 per cent to $1.3 MFS subsequently went into liquidation.
For sure, there are good broker calls as well. But the East Coles survey found these are few and far between, which leaves fund managers wondering whether they would be better off with a dart board.
Further evidence of persistent inaccuracy in broker forecasts came in another recent survey, this time on the US market and carried out by management consulting firm McKinsey, updating a similar survey almost a decade ago.4
Analysing earnings forecasts for S&P-500 companies over a quarter century to the end of 2009, McKinsey found that with few exceptions aggregate earnings forecasts tended to exceed realised earnings per share.
Actual growth surpassed forecasts only twice in 25 years, the consultancy found, and both times during the recovery following a recession.
Instead, the survey found the most accurate expectations were actually those built into prices on capital markets themselves.
“This pattern confirms our earlier findings that analysts typically lag behind events in revising their forecasts to reflect new economic conditions,” McKinsey said. “Executives…ought to base their strategic decisions on what they see happening in their industries rather than respond to the pressures of forecasts, since even the market doesn’t expect them to do so.”
Good advice – and yet another reason to see market pricing as the best unbiased estimate of expected returns on capital markets.