Analysts getting less accurate with earnings estimates

After scandals such as Enron and the dot-com debacle, a host of new measures were put in place, most notably 2002's Sarbanes-Oxley Act, which took a hack at reforming company accounts, and 2000's Regulation Fair Disclosure, forcing companies to release material information to all at the same time.

Though earlier studies showed an improvement in the wake of the measures, a new report looking at 1993 to 2013 makes depressing reading.

"Analyst forecast error and dispersion significantly increased over the long-term post-regulation period. Therefore, even if we assume that the regulations caused the improvement in the observed analyst forecast properties in the short run, they did not have a lasting effect," write Hassan Espahbodi and Pouran Espahbodi of the University of Texas Rio Grande Valley and Reza Espahbodi of Washburn University.

Their study, published in the current edition of the CFA Institute's Financial Analysts Journal, shows that although accuracy and dispersion improved in 2003-05, after the new regs were in place, they have since deteriorated significantly.

The regulations were intended to end widespread abuses such as selective disclosure of important information to analysts, as well as practices that gave stock analysts an incentive to put the well-being of their firms' investment banking operations over those of their supposed buy-side clients.

The abuses may have been curbed, but analysts' product did not improve.

"We conclude, therefore, that these regulations did not collectively improve the information environment in the long term, despite the reduction in analyst conflicts of interest. The continued problem with the information environment, therefore, seems to be due to the quality of financial reports," the authors write.

It seems possible, too, that analysts are less accurate because they are being steered less well by management. Given the inherent corruption in that system, that may be an improvement, though not if you want to be able to stop thinking and just trade in reaction to analyst reports.

It may however, be a classic case of "garbage in, garbage out" as the study indicates that it is not so much that the analysis is worse but that company reports are less reliable.

While Sarbanes-Oxley was intended to stamp out fraud, fraud in company accounting ultimately comes not from poor controls but poor ethics on the part of agents — company executives — who can benefit from such fraud.

A survey of nearly 400 chief financial officers recently found that 20 percent believe that firms intentionally distort earnings. The economic magnitude of the distortion was big — 10 percent — and a significant proportion of the fibbing was underselling earnings.

That survey didn't have a pre-Sarbanes-Oxley benchmark, but given the growth in executive compensation tied to share options there is certainly probable cause for motivation.

Remember, company executives, especially the top ones with the power to distort, are paid largely in options, which rise in value with increased volatility. It could therefore be lucrative, if dishonest, for an executive to engineer an earnings miss around the time an option is to be priced followed by a strong showing above expectations later down the line.

Share analysts could have been the instrument, wittingly or not, of a strategy like this before Sarbox. Now that they must rely more on company accounts with less "color" from insiders the same aims may be achieved through different means.

More highly dispersed and inaccurate analysts' earnings reports would certainly be useful for self-dealing insiders. Those may simply be the natural outgrowth of a deterioration in company information. Or perhaps analysts were just poor all along at interpreting company accounts, something now unmasked as they cannot be spoon-fed.

This is not to say that the reforms were a failure. In many ways they succeeded, especially in eliminating some forms of conflict of interest.