COLUMBIA, Mo. - Companies and shareholders depend on auditors to provide unbiased, independent analyses of financial statements, but new research from the University of Missouri shows auditors can be influenced and misled by information provided by management.
Nate Newton, assistant professor of accountancy in the Trulaske College of Business, and his co-authors asked practicing auditors to estimate expenses relating to a year-end fire at a fictional company. All auditors were given the same range of possible total costs for the incident and a set of company estimates, as is commonly done to demonstrate management's consideration of alternative assumptions. However, one group of auditors was given company estimates strategically grouped near management's preferred total expenses, and another group received a set of estimates more representative of the full range of potential expenses.
Newton and his co-authors found that the group given estimates supporting the company's preferred expenses were more than twice as likely to approve of management's preference compared to auditors who received a set of estimates more representative of the full range of potential expenses.
"We found that outside auditors were completely influenced by management's estimates, despite knowing in advance the full range of potential payouts," Newton said. "Industry standards tell auditors to evaluate management's consideration of alternative assumptions, so it is important for auditors to be aware that the process can influence their judgments."
Ideally, Newton said, auditors should ask for the information that led management to develop their estimates and create cost scenarios of their own. Companies often prefer to report lower costs; if auditors aren't critical of the assumptions made by a client, the result could be inaccurate financial statements.
In the year-end fire scenario, it is impossible to know the precise cost of the event because it involves uncertainties like whether lawsuits might be filed or the extent of physical damages to the fictional company. Still, working on cost scenarios independent of company estimates could result in a more comprehensive and accurate analysis, Newton said.
"Being aware of the potential bias created by the consideration of alternative assumptions can serve auditors, company management and investors because they can have a clearer portrait of a company's financial results," Newton said.
As part of the study, Newton and his co-authors also asked two groups of auditors to assess the risk that a company's financial statements would be misstated due to a lapse in internal controls. One group was given the option to rate a company as either low risk or high risk. The second group was asked to rate the company using one of three risk categories: low, moderate or high. When participants had two options, 37 percent rated the company as low risk. However, with three options, low risk was selected only 13 percent of the time.
Newton said the experiments illustrate that individuals who are making a determination amid uncertainty are highly influenced by the suggestion of benchmarks, despite knowing the full range of possibilities.
Newton's co-authors are Christopher Wolfe, accounting professor at Texas A&M University; and Brian Fitzgerald, assistant professor of accounting at Northeastern University. Their study, "The Effect of Partition Dependence on Assessing Accounting Estimates," will appear in an upcoming issue of Auditing: A Journal of Practice & Theory.