Research Brief

Separating Auditing from Consulting: More Complex than it Seems

Market concentration, price and quality drive choice of firms

The business of auditing companies has a built-in conflict: The company pays the auditor’s fees, which might dampen the auditor’s enthusiasm for delivering unwelcome news to company management. Especially, say, news that could upset investors and disrupt the company’s activities (and diminish the value of stock that makes up most of management’s net worth).

Now add to that the audit firm doing myriad consulting work for the company, work that brings in more fees than the audit assignment, which might make it even harder to displease top executives.

Exhibit A in why auditors shouldn’t do consulting work for clients is, of course, Arthur Andersen, the now-defunct firm whose audit and considerable consulting work occurred at the scene of three spectacular accounting scandals roughly 20 years ago: Enron, Waste Management and WorldCom.

Opt In to the Review Monthly Email Update.

The solution, contained in 2002’s Sarbanes-Oxley Act, mostly banned firms from providing nonaudit consulting services to their audit clients. What could be simpler?

Separation of Services Is Not Simple

A paper by UCLA Anderson’s Henry L. Friedman and Tilburg University’s Lucas Mahieux, published in the Journal of Accounting Research, suggests that the separation brings with it significant complications for the audit and consulting industries — and notably, also for their clients, in terms of price and quality of services, especially as the audit industry has consolidated.

The authors develop a model that accounts for these factors within a competitive landscape of auditors and pure-play consulting firms, which they use to demonstrate possible outcomes of different regulatory scenarios. The model has important implications for regulators concerned with general corporate financial integrity (as measured by average audit quality) and with the efficient operation of the audit, consulting and general corporate worlds.

Consolidation among auditors is a big factor. When there was a “Big Eight” among accounting firms, an operating company had seven others to choose from for consulting services, minus perhaps one or two closely linked to a competing company. Today, with a Big Four (Deloitte, PwC, KPMG and Ernst & Young), the choice is very limited. These four are responsible for auditing somewhere around 98% of the companies listed on the S&P 500 and the U.K.’s FTSE 350 index.

The management consultancy market is somewhat less consolidated: The top six pure-play consulting firms (led by the “Big Three” of McKinsey & Co., Boston Consulting Group and Bain) together accounted for a little under 20% of the market in 2018. But the consulting arms of the Big Four audit firms accounted for nearly twice as much market share, 37.4% of the consulting market.

The paper’s model is even more streamlined than the real-life audit market. It uses three client firms with different risk profiles (high, medium and low), two auditors (a high-quality more expensive one and a lower quality cheaper one) and a single pure-play consultant.

The authors assume for the model that client firms engage auditors to closely examine their financial books and weed out potentially unprofitable business ventures (no seeking a rubber stamp here). The more high risk a firm’s profile, the more it will benefit from a high-quality audit, and the more it will be willing to pay for it.

The auditor in turn will invest more in inputs (such as training and recruitment of skilled staff) that boost audit quality. At the same time, a low-risk client firm may feel that a low-cost audit is perfectly adequate. An intermediate-risk client firm may be targeted by either a high-quality or low-quality auditor, depending on the cost of the audit.

These same clients may or may not want consulting services. The key factor influencing how auditors and consultants will price and invest in their services is the degree to which client firms’ demand for these two services — audit and consulting — correlate.

Researchers’ Model Examines Regulatory Setups

The baseline model assumes that the correlation between audit and consulting demand is positive: a client who needs auditing services also wants consulting.

The authors then make predictions for three scenarios:

Auditors face no restrictions on offering consulting, and the markets for both kinds of services do not interact in a way that benefits or disadvantages either type of provider. The authors use this scenario to highlight some basic internal dynamics of the audit market: A high-quality auditor can either charge a higher price and only serve the high-risk client or target the high- and medium-risk client at a lower price. The medium-risk client will choose based on whether its expected gain from using a high-quality auditor exceeds the higher cost. The auditing firms thus face a trade-off between serving more audit clients at a lower price or charging more to a single client for a high-quality audit. The strategy they choose affects the average audit quality in the market. If the high-quality auditor’s costs per client are low enough, it can offer its services to both the high- and medium-risk firm, thereby maximizing social welfare.

Audit firms are banned from offering consulting to their audit clients. This limits competition in the consulting market for those clients, allowing any consulting competitor to charge higher prices. The high-quality auditor can then maximize its revenues by offering its high-quality services only to the high-risk client, while selling consulting — now priced higher thanks to reduced competition — to the two firms it doesn’t audit. The medium-risk company switches to the low-quality auditor, which could be either good for social welfare (the firm is not overpaying for costly auditing services it doesn’t need) or bad (the firm is not getting the high-quality audit it needs given its level of business risk).

Auditors cannot provide consulting to any client. Once again, the two markets don’t interact at all. The consulting competitor now has to price its services at a level that can attract all three client firms, and the two auditors face similar incentives as in the first scenario. The effect on social welfare will once again depend on how the high-risk auditor prices its audit services and on the medium-risk company’s actual business risk.

The Consulting Ban Isn’t the Only Obstacle

The model offers regulators a nuanced way to think about how bans on consulting for audit clients can affect auditor behavior by considering features of these markets such as pure-play consulting competitors, price and quality differentials, and how these interact with the varying business risk of client firms across time.

Two decades of a partial ban have come amid other problems with the audit market. These include the brighter growth prospects in general for consulting services; the oligopolistic structure of the audit market; the dominance of auditors in the consulting market; the revolving door between the auditors and their client firms (executives at audit firms often go into senior roles with the biggest companies in the world); and the simple fact that auditors are tasked with being tough on the same entities that write their paychecks.

This latter issue exists regardless of any ban on auditor consulting. Given the moonshot bets that venture capitalists place on startups, it isn’t clear that investors believe it’s in their best interest to inflict a rigorous audit on a growing firm when growth rate is precisely what determines its market value.

Early investors were not complaining when Ernst & Young’s signed off on WeWork’s financial statements ahead of its failed 2019 initial public offering; it was only when potential investors questioned its $47 billion valuation and began scrutinizing its business model that the IPO fell apart.

Cross-Selling Services Produces a Better Audit?

The ban on consulting work for audit clients is far from absolute. Audit firms are permitted to do both kinds of work for non-U.S. clients and to provide consulting services to U.S. companies who are not also audit clients. The less-legal option of characterizing nonaudit services as part of an auditing engagement can mean big trouble and big fines. PricewaterhouseCoopers (now PwC) allegedly did this for 15 clients between 2013 and 2016. which led to a $7.9 million settlement with the Securities and Exchange Commission.

But Sarbanes-Oxley took away the most convenient cross-selling opportunities — auditors are in a perfect position to offer consulting help. One analysis found that in 2000, Waste Management had the highest auditing bill of any of the more than 300 major firms studied, shelling out $48 million to Andersen. The same year, its nonaudit bill came to $31 million to Andersen for other services, including $22 million to revamp the firm’s accounting and financial-control systems. Andersen earned slightly more from providing nonaudit services to Enron than from auditing it — $27 million to $25 million — in the year before the energy firm’s collapse.

Consulting accounts made up 70% of total revenues at the Big Four in 2017, up from 40% in the wake of Sarbanes-Oxley.

Those inside the audit industry maintain that consulting revenue does not affect their independence; they say the additional work across other areas of a client firm allows them better visibility into its operations and risk potential, which can  increase audit quality.

Friedman and Mahieux offer a more nuanced way of thinking about the audit market and its interaction with the market for consulting. They conclude that, given the complexity of the audit and nonaudit services, or NAS, markets, the outcome of a ban isn’t so easy to predict. “General bans on auditor provision of NAS can, via similar channels, increase or decrease audit quality and social welfare,” the authors write.

Featured Faculty

About the Research

Friedman, H. L. and Mahieux, L. (2021). How Is the Audit Market Affected by Characteristics of the Nonaudit Services Market? Journal of Accounting Research.

Related Articles

A bubble refracts light and the image of a window Research Brief / Markets

Market Bubbles Aren’t Entirely Irrational Exuberance

A model examines the relationships between innovation, speculation and market values

Migrant workers in the field. Research Brief / Supply Chain

How to Reduce the Risk of Forced Labor in Agricultural Supply Chains

Improving the search for contractors that don’t use coercion

Jay Powell Research Brief / Markets

Corporate Bond Market Meltdown Averted after Fed Action

Decade-old bank-risk limits may have exacerbated liquidity problems

Illustration of a man chasing money floating away while chained to a ball Feature / Compensation

Noncompete Agreements Reduce Worker Pay — and Overall Economic Activity

Researchers refute earlier claims