Discussion of ‘Strategic pricing by Big 4 audit firms in private client segments’

1. Introduction

I am delighted to have the opportunity to comment on Dutillieux et al., (hereafter, DSW). These comments relate to the version of the paper presented at the Accounting and Finance Conference in April 2013 and may not be as directly related to the final published version of the paper. My comments are in four main areas, regarding the importance of the research question, the setting in which the study is located, the theoretical development of the study and some thoughts regarding both supply- and demand-side considerations around auditor change. I conclude with some suggestions as to where the study could be extended which would further the contribution of an already useful study.

2. Importance of the research question

DSW ask an important research question, namely what is the nature of competition for new clients in the private company market. Whilst we know a reasonable amount about competition between the Big 4 and non-Big 4 and within the Big 4 in the public company market in many countries, including for initial audit engagements, we know very little about the private audit company market, or about how competition plays out for this group of companies. DSW take a two-pronged consideration to the issue of competition by examining both price and quality. This is a useful approach because there may be trade-offs occurring between price and quality and if this is so, this is important that it be known.

3. National setting

In addition to the private company setting, DSW's study is conducted in Belgium. This setting is important and appropriate as private companies make up a significant proportion of the economy in many countries and combined they are an important part of the audit market. There are some opportunities presented by the regulations and structure of the audit environment in Belgium. The data used by DSW are proprietary so that whilst fees are disclosed to the regulator, auditors and clients are not able to easily compare fees to their competitors. In addition, auditors are awarded 3-year audit mandates (which may or may not be renewed). This enables DSW to explore audit fees in a multiperiod setting, which is a strength of the study. Consideration could be given to an alternate measurement of the dependent variable which might be to consider the fees over the term of a 3-year mandate rather than in a single year. Other interesting features of the Belgian market about which further information could be provided include a discussion on whether there are prohibitions on solicitation and restrictions on advertising of audit services and how new clients are obtained by auditors would be helpful as well as some explanation on the choice of 2004 as the year of interest and the likely representativeness of that particular year. It would be useful to consider whether there are any differences in the competitive landscape for audit services in the private company market as well as in the Belgian audit market as this would aid in an assessment of the generalizability of this study to other settings.

4. Theoretical development

Currently, the theoretical development of the study is fairly limited. Experience Good Theory (EGT) as proposed by Shapiro (1983) suggests that you need to have experienced the benefits of purchase in order to be able to assess its quality or value. EGT has been used to explain supplier behaviour across a broad range of settings – banking, advertising industries and trade union membership to name a few. EGT seems to be applicable to the audit environment and has been used by Craswell and Francis (1999) to explain pricing of initial audits by specialists versus nonspecialists. There are certainly other theories worth considering in this setting, for example, audit as a credence good might be pertinent for private companies voluntarily seeking audits.

5. Supply- and demand-side considerations around auditor change

Why do (private) clients change auditor? There are many reasons for clients changing auditor (see Stefaniak et al., 2009; for a comprehensive review of this literature as well as Fontaine et al., 2013; recent interview data primarily with private company CFOs). These include having a disagreement with the current auditor to seek a more understanding auditor, seeking a lower audit fee, for a growing firm needing a bigger auditor to meet needs, for a firm planning to list or raise equity or debt needing a high-reputation auditor. The assumption underpinning DSW's study is that clients initiate all auditor changes from a demand perspective. However, there is evidence that auditors dismiss clients and manage their clienteles from a supply perspective as well (for example, Johnstone, 2000) which should also be considered.

Related to this is how do auditors in the private company market attract clients? More particularly, how should auditors attract the best clients? This is an interesting question because it raises the issue of whether auditors are good at picking winners. The analysis employed by DSW uses future sales growth over the next three years (2004–2007), a fact which is not known at the time of the auditor change. It would be interesting to also establish the past sales growth history of these clients and whether specialist auditors are able to pick winners ahead of sales growth which not known at the time of the switch. DSW's study has the potential to contribute to the industry specialization literature by examining whether specialist auditors are better at selecting high-quality clients relative to other auditors.

DSW may also wish to consider alternate measures of desirable clients, book-to-market is one but there are other factors that make clients attractive to auditors, some of which may not be measurable using a data set like that employed by DSW but may be worth considering. From my experience in practice, desirable clients were those who were ready for the audit to commence, with competent staff, good support for external audit from Audit Committee or Board, were keen for the audit to be completed on time (and on budget), receptive to feedback, were profitable, growing and in a growth industry and were likely to be issued with an unqualified audit opinion. From a financial perspective, desirable clients were those that were not sensitive to fee increases, purchased NAS, paid on time and referred the firm to others and were keen to maintain a long-term relationship. This leads me to consider that from the perspective of auditors, are clients an experience good too?

6. Extensions to the current study and considerations for future research

So, where could study go further? I have identified four main areas where the authors could consider extending their analyses to aid in a more complete understanding of competition in the private sector audit market. I have a number of suggestions for the authors regarding the wording and subsequent testing of the hypotheses. In their current form the wording of the hypotheses is unwieldy and long. I would encourage the authors to focus on tightening and ultimately shortening the hypotheses which would enhance the readability of the paper as well as clarify the hypothesis testing to be undertaken.

It is important to identify the most appropriate sample for testing of hypotheses. Currently, DSW compare ongoing clients of the Big 4 with new clients of the Big 4. It may be worth considering whether this is the right comparison group. To grow their overall business, Big 4 need to cherry pick the best (or the least worst) of the non-Big 4′s clients and some focus might be usefully placed on this group. It is hard to justify dropping current clients of the non-Big N from your sample when they are an important part of a competitive audit market especially as you have motivated the location of the study in Belgium with the spread of clients between Big 4 and non-Big 4. Whilst the focus is on the Big 4, it is the behaviour of the Big 4 relative to their competition that is of interest. After focussing on increasing your sample size, I would question the inclusion of finance and real estate industries in your analysis. Generally, but not always, these industries are excluded from audit fee studies due to differences in their audit fee drivers.

In relation to the model and the variables employed, I would consider the R² of the models to be low for fee models compared with other countries (around 40%). In particular, the lack of importance of client size in its association with audit fees is unusual. I would also suggest that there may be other variables that might address concerns regarding the completeness of the model to be employed for hypothesis testing. Including variables for type of audit opinion, controls for the year of mandate and the number of successive mandates would seem to be useful as well as other variables appropriate to private company market, for example, ownership concentration and leverage. I like the inclusion of the whole sample in the calculation of the market share metrics as it provides the interested reader with a good overall picture of the audit market being investigated. Currently, the robustness tests are fairly minimal and the results of these could be more clearly spelled out.

I would encourage the authors to explore their findings in relation to multiple periods after the 2004–2007 period examined in the study. For example, if audit firms offer a discount now, what happens later? Do clients remain with them or do they leave? The role of NAS could be more fully explored as often there is greater variation in demand for NAS from private companies than would be usual for public companies. Given the lack of public disclosure of this data, analysing the role of NAS is likely to be of great interest. By way of conclusion, I found this to be an interesting study asking an important question and utilizing a unique and appropriate dataset to answer the research question posed. I hope these suggestions have proved helpful to the author team and I thank Steven Cahan for giving me the opportunity to provide these comments.