Fixed Costs, Audit Production, and Audit Markets: Theory and Evidence

Abstract

We analyze the role of discretionary joint fixed costs in audit production. Given such costs, the investment decision and production of audit services must be analyzed over a client portfolio. We model this problem, and use monotone comparative statics (Milgrom and Shannon [1994]) to show the implications of variations in client-specific losses and the number of clients for the optimum level of fixed investment and audit assurance. We develop four hypotheses concerning the relations between audit quality and (1) the magnitude of potential client-specific losses; (2) average client losses in a portfolio; (3) the number of clients in a portfolio; and (4) the variability of losses in a portfolio. Using discretionary accruals as the audit quality proxy, we find evidence consistent with these hypotheses. Using PCAOB inspection results, and financial statement restatements as proxies for audit quality, we find weaker evidence consistent with hypotheses 2, 3 and 4.

What does an audit really cost? An exploratory study of auditor pricing of initial engagements in a fixed cost model

“…Technically, theSimunic (1980) model refers to the incremental costs of production, but almost all research since then has treated audit effort as the primary source of incremental costs.2 A recent working paper byGu et al (2019) models fixed costs within an audit firm and derives an optimal level of investment in fixed costs. Their related empirical results suggest that fixed costs can have a direct effect on audit quality, although this empirical observation is based on audit fees as is the other literature in this area, rather than direct costs of an engagement.3 The primary implication of theSimunic (1980) assumptions is that audit fees cannot reflect demand effects.…”

“…2 2 A recent working paper by Gu et al (2019) models fixed costs within an audit firm and derives an optimal level of investment in fixed costs. Their related empirical results suggest that fixed costs can have a direct effect on audit quality, although this empirical observation is based on audit fees as is the other literature in this area, rather than direct costs of an engagement.…”

“…16 16 While beyond the scope of this paper, a firm may have other fixed costs not related to direct labour, for example, office space, computers, technology development, system support, data acquisition and training. Some of these are discussed in Gu et al (2019). These costs also should be considered in calculating firm and engagement profitability since they represent the ‘overhead’ of the firm that must be ‘covered’ by contribution margin.…”