ISSN 1450-2275 Issue 96 December, 2017
© FRDN Incorporated
http://www.europeanjournalofeconomicsfinanceandadministrativesciences.com
Audit firms with Public Clients in Voluntary Audit-Market
Tatiana Mazza
Free University of Bozen-Bolzano, piazza Università 1, 39100 Bozen-Bolzano, Italy E-mail: tatiana.mazza@unibz.it
ORCID: 0000-0003-3516-9757 Stefano Azzali
University of Parma, via Kennedy, 6, 43100 Parma PR, Italy E-mail: stefano.azzali@unipr.it
ORCID: 0000-0002-2745-3347
Abstract
The aim of this study is to analyze benefits, in term of lower cost of debt, that non- Big4 audit firm with public clients help to obtain to their private clients in a voluntary audit market segment. Survey to partners of non-Big4 aiming to collect first evidences about the effects of auditing public clients have been performed. Multivariate regression on propensity score matched samples of non-listed companies audited by non-Big4 are run to test the hypotheses on the association between non-Big4 audit firm with public clients and cost of debt in their private clients. Results confirm expectations and reveal that the presence, the percentage and the number of public clients in a non-Big4 portfolio is negatively associated with cost of debt of the non-Big4’s private clients. Results show the externalities that the audit firms’ public client specialization has on private firms. Findings reveal that non-Big4 audit firm with public clients are able to transfer benefits associated with higher reputation, competences, litigation risks and incentives toward audit quality to private clients.
Keywords: Public firms, Non-Big4 audit firms, Cost of Debt, Voluntary audit market.
1. Introduction
The aim of this study is to analyze benefits that audit firm with public clients may provide to their private clients in voluntary audit market segment. In private firms, cost of debt (CoD) is one of the most important drivers of managers’ choices, given that debt is a significant resource that allow investment and that, in private companies, the main agency conflict is between lenders and managers/owners.
The research is motivated by a gap in the audit literature related to private companies and the non-Big4 segment of audit market. While literature investigated benefits (lower CoD) for private companies when audited by Big4 (Gul et al., 2013) and other benefits (higher reputation, competences, litigation risks, incentive to improve audit quality) associated with audit firms with public clients (Ittonen et al., 2015, Clatworthy and Peel, 2007, Hay et al., 2006), no research exists on the possibility to transfer previous benefits to private clients and non-Big4 segment of audit market.
Moreover, Italy is an interesting setting to investigate because: a) the non-BigN audit market share is significant in the private company segment (around 40 per cent); b) auditors are liable to third
parties, like lenders (Giudici, 2012).1 Since 2010 Italy has made obligatory a transparency report where audit firms list their public and other private public interest entities clients (such as financial institution), and this make it possible to classify audit firms by client characteristics. To the best of our knowledge, there is no prior literature on this topic.
We use a sample of non-listed Italian companies audited by non-Big4 in the period 2010-2014.
We firstly perform survey on non-Big4 that recently acquired their first public client to have a first overview on the effect of the work for these public clients and for the externalities on their other private clients. We secondly run multivariate regression analysis to test our hypotheses. Our regression analyses are run on propensity matched samples to assure comparability in term of size, agency costs toward lenders, profitability, complexity and financial risk between private clients audited by different groups of auditors.
In the matched sample of companies audited by non-Big4 with and without public clients (1138 observations), we find that presence of at least one public client in a non-Big4 portfolio is negatively associated with CoD of the non-Big4’s private clients. The benefit for private clients show the externalities that the audit firms’ public client specialization has on private firms. Focusing on the matched sample of companies audited by non-Big4 with public clients (718 observations), we find that the percentage of public clients of non-Big4 audit firm is negatively associated with CoD of their private clients. Finally, on the matched sample on companies audited by non-Big4 with public clients (594 observations), matched for the number of public clients, we find that the number of public clients of non-Big4 audit firm is negatively associated with CoD of their private clients.
Batteries of robustness tests run on different specifications for propensity score matched sample, alternative measures of agency costs between lenders and managers, a wider sample that include public companies, and different accounting standards, confirm our main results.
We contribute in several ways. Firstly, results show that non-Big4 audit firm with public clients are able to transfer benefits associated with higher reputation, competences, litigation risks and incentives toward audit quality to private clients. Secondly, non-Big4 audit firm have incentives to increase their percentage and number of public clients to increase their audit quality. Thirdly, we interpret our findings as suggestions for practitioners on the use of these auditors to mitigate the agency conflict between lenders and owners/managers in private firms.
The next session analyzes previous literature e reveals a gap in the audit literature related to private companies and non-Big4 audit market segment. Section 3 develops our hypothesis on the association between non-Big4 audit firm with public clients and CoD, specifying our expectation.
Section 4 presents our methodology that includes sample selection, surveys and multivariate regression models. Section 5 discusses the results and includes findings from the surveys, descriptive statistics, correlation matrix and regression models. Section 6 reports robustness tests and section 7 concludes the study.
2. Literature Review
Companies have incentive to increase audit quality in order to lower agency costs. Literature on agency conflict in private firms finds that as the demand for financial reporting and for external audits mainly arises from the need for debt contracting with banks and other private lenders (Lennox, 2005), principals are typically lenders (Peek et al., 2010; Power, 1997; Vander Bauwhede and Willekens, 2004). A bank may place more trust in client financial reporting and reduce the CoD when a high- quality auditor assures it. Previous studies show that banks tend to form different perceptions
1 One reason for this regulation is that Italy is a country where the main financing channel for companies is in the form of banks and trade creditors (third parties), and creditor protection is perceived to be more important than in Anglo-American jurisdictions. Moreover, Italian auditors were originally inside internal statutory audit committees. Once it was decided that directors and members of statutory audit committee were to be made liable for damages incurred by creditors, external auditors were put in the same position as members of statutory audit committee (Giudici, 2012).
according to the level of audit firm quality (Kelly and Mohrweis, 1989; Libby, 1979a; Libby, 1979b;
Strawser, 1994). Unlike public companies where internal corporate governance mechanism or surveillance of market authorities may mitigate agency costs, in private firms, audit quality may be the only available instrument to mitigate them (Cano-Rodríguez and Alegría, 2012). Moreover, Gul et al.
(2013), analyzing data from several countries in the period 1994 – 2006, find that CoD in firms with Big4 is lower than in firms with non-Big4 only in countries with stronger investor protection. Prior studies show that voluntary audited private firms compared to non-audited private firms have lower CoD, higher credit rating, easier access to external finance and lower EM (Minnis 2011; Melumad and Thoman 1990; Lennox and Pittman 2011; Hope et al. 2011; Kim et al. 2011; Dedman et al. 2014;
Dedman and Kausar 2012; Blackwell et al. 1998; Allee and Yohn 2009; Niemi et al., 2012; Collis, 2012).
Audit firm with public clients benefit in terms of: (1) reputation and competences acquired from auditing public clients, (2) incentives to improve audit quality stemming from high litigation risk associated with public clients, (3) incentive to improve audit quality stemming from stakeholder theory. First, audit firms with public clients have higher reputation and competences than audit firms with only private clients. Ittonen et al. (2015), analyzing a sample of public firms in Finland, find “that partners with greater public-client specialization provide higher quality auditing, since they have likely developed deep domain-specific knowledge and a keen sense of the reputational risks posed by public clients”. Clatworthy and Peel (2007) argue that the higher profile of public companies is likely to result in larger potential reputational losses for their audit firms. For this reason, public companies pay higher fees to their audit firm than private firms. Second, audit firms with public clients could be subject to higher litigation risk. Hay et al. (2006) conduct a meta-analysis comprising 12 studies, as well as other studies, that include in their audit fee model the difference between public and private status.
According to the 12 studies, a positive relation between dummy variables for public versus private companies and audit fees is ascertained. This is because public ownership is considered to increase the auditor's potential exposure to litigation that leads to higher audit fees. High litigation risk has been found to be one of the main incentives to perform a high-quality audit. Archival evidence suggests that audit fees do reflect variations in litigation (e.g. Palmrose, 1988; Pratt and Stice, 1994; Simunic and Stein, 1996; Seetharaman et al., 2002; Minutti-Meza, 2014). Thus, audit firms with public clients have higher AQ given the incentive from the high litigation risk. Third, public firms have to interact with more stakeholders and more sophisticated stakeholders (i.e. financial analysts). Public firms have more stringent accounting regulations and disclosure requirements, which contribute to a more developed culture of control and sensitiveness for the reliability of financial reporting. Kouaib and Jarboui (2014) studies the complementarities between ownership structure and the external audit quality. Mitra et al.
(2007) argue that to attract a large number of institutional investors, companies turn to auditors offering better services. Following stakeholder theory, audit firms that effectively audit public clients have higher quality. Their better environmental context in fact means they usually use more organized audit procedures for public clients, and their private firm clients can also benefit from these procedures.
Summarizing previous literature, no studies audit firms with public clients benefits in terms of reputation/competences, and incentives to improve audit quality. We cover this gap of the literature investigating the spillover effect on private firms with these high-quality auditors with public client.
3. Hypothesis Development
We hypothesis that audit quality benefits, in term of lower CoD, can be transferred to private clients of non-Big4 audit firms. Taking into account previous literature, it is an open question whether the higher reputation, higher competences, increased litigation risk and larger set of stakeholders (i.e. the effects of having a public client), apply to private clients when the audit firm is non-Big4. We expect that private clients will benefit from externalities that non-Big4 audit firms develop paying the cost to set up the system to audit public clients. No prior studies look at how externalities can be transferred to private firms, or whether the increase in reputation, competences, litigation risks and stakeholder
relationships affects private clients. The transfer of knowledge is even more important in the non-Big4 markets, where the number of public clients is lower and so having a public client makes more impact.
We hypothesis that reputation, competences, incentive from litigation and stakeholders’ relationships increases audit quality of non-Big4 audit firms with public clients. The quality control from the national oversight board to the audit of public clients increases this expectation of higher audit quality (Rija, 2017). We expect that private clients can benefit from externalities that non-Big4 audit firms have borne the cost of setting up for use with public clients.
Hp1: The audit of public clients in a non-Big4 portfolio is negatively associated with cost of debt of its private clients
With previous hypothesis, we compare non-Big4 with and without public clients. We also aim to contribute to literature focusing on companies audited by non-Big4 with public clients: just the presence of public clients cannot be sufficient to guaranty externalities to private clients. Then, we suggest testing the proportion of public clients out of an auditor’s total clients. Given that in the first hypothesis we expect that audit quality developed in working for public clients could be transferred to private clients, we next investigate if audit quality has different level depending on the magnitude of public clients work over the yearly work of an audit firm.
Hp2: The percentage of public clients over total clients of a non-Big4 audit firm is negatively associated with cost of debt of its private clients
Finally, the proportion of public clients over the total clients could not be the best variable to measure the benefits of lower CoD if the number of public clients is low. For example, when the non- Big4 has only one public client over a small total number of clients such as five clients, the high percentage (20%) of public clients out of total clients is coded to be better than the percentage (10%) non-Big4 with higher public clients, as for example 15 public clients out of 150 total clients. However, only one public client cannot be sufficient to create a base set for the change in audit procedures of the audit firms and to guaranty externalities to private clients. Supporting this view, prior literature investigates the auditor independence relation with the number of public clients. The willingness to resist client pressure is likely to increase with the number of public clients in the partner’s portfolio, because dependence on any one client diminishes, which should help to ensure audit quality. Ittonen et al. (2015) show that the negative association between abnormal accruals and public-client specialization can be attributed to auditors with three to six public clients. Based on this literature, we also use the number of public clients as independent variable and we expect different levels of CoD depending on the number of public client of non-Big4 audit firm.
Hp3: The number of public clients of non-Big4 audit firm is negatively associated with cost of debt of its private clients
4. Method
This paragraph includes methodologies used to select our sample of Italian companies and collect data (4.1. sample selection), to perform the research based on surveys to partners of non-Big4 audit firm with public clients (4.2. survey), and on statistical model (4.3. multivariate regression model).
4.1. Sample Selection
We start from 1149 Italian companies audited by non-Big4 audit firms (firms with two or more individual owners) with more than one client per year, appearing in Bureau Van Dick database (Table 1). We firstly drop public companies because they cannot choose among the different types of audit firms here analyzed. This database includes only the name of the last audit firm engaged and the year
of its engagement. Two downloads, one in 2012 and one in 2014, thus supplied the name of the firm that audited the list of clients in our sample at the end of 2012 and at the end of 2014. For each of the audit firms we have the starting year of the engagement. We include only the years for which we know that the audit firm was auditing a specific client, resulting in a sample period different for each firm (unbalanced sample). 2 All the firms in our sample voluntarily choose an external audit firm3. The problem of self-selection of the sample is lower than in prior studies because the comparison is not with firms that do not undergo audit, but between the types of audit firm that they engage. All the firms in the sample undergo audit.
Secondly, we compute the CoD and we drop observations with missing values for this variable.
The final sample used in the logistic consists of 1529 observations. PSM yields a sample of 1138 observations to be used in the main analysis.
Table 1: Sample selection
Description N.
Total number of Italian companies audited by a non-Big4 audit firm with at least 2 clients in the Bureau Van
Dick database in 2014 1149
Less public companies or companies subjected to mandatory audit in 2014 -254
Total number of firms 895
Total number of observations for the period 2010-2014 4435
Less observations with missing values necessary to compute variables related to cost of debt and independent
variables -2906
Total number of observations used in the first stage regression (untabulated) 1529 Less observations not matched in PSM and add observations replaced in PSM -391
Total number of observations in the matched sample 1138
4.2. Survey
We first run a survey to partners of non-Big4 to collect first anecdotal evidences about the effects of public clients. The survey was performed in February 2016. We addressed partners because we believe they are better able to answer questions, and are likely to have better knowledge of firm practices than lower ranked staff. We selected three audit firms which acquired their first public clients in 2007, 2008, and 2012, as shown in their Transparency report. In order to ensure external and internal validity of the questions, they were evaluated for clarity, completeness, and relevance by academic researchers not associated with this study. The questions aimed to evaluate the effects of the new public clients.
Thus, our first questions are about the effects of a specific public client acquired in a specific year on audit planning, scoping and risk assessment, testing, monitoring and remediation, on the composition of the audit team and relations with regulatory authorities, on audit fees and on insurance costs. Then, we let the second part of the survey open to receive any comments. See Appendix B for the text of the questions and the transcription of the comments received.
Further, as is customary in such studies, we informed participants that we would hold their responses in strict confidence, and report the results without revealing the name of the partner or audit firm. In addition, we explained that the interview is part of a research project under the auspices of a well-known university, which is widely recognized as trustworthy.
4.3. Multivariate Regression Model
Next, we run the following CoD model based on Equation (1):
2 In the period analysed in this research (2010-2014), Italian auditors used national auditing standards. These standards are similar to International Standards of Audit (ISA), and meanwhile Italy is moving towards their implementation. National standards are set by
“Consiglio Nazionale Dottori Commercialisti ed Esperti Contabili”. ISA have been mandatory in Italy since January 1st, 2015.
3 In Italy the audit of private firms can also be performed by an internal Board of Statutory Auditors or by one individual external auditor.
We exclude these audits from the sample. Our sample does not include firms not audited or subject to mandatory external audit.
CoDit = α + β1 public clientsit + β2 SIZEit + β3 LEVERAGEit + β4 QUICKit + β5 ROAit + β6
TANGIBLEit + β7 ALTMANit + β8 NEGATIVE EQUITYit + β9 LOAN MATURITYit + industry fixed
effect + year fixed effect+ e (1).
CoD is the average cost of financial debts for firm i and year t, which is the financial cost disclosed in the income statement following Generally Accepted Accounting Principles in Italy, scaled by the total amount of financial debts. The financial cost includes interest and commission. Following Francis et al. (2005), Karjalainen (2011); Cano-Rodríguez and Alegría (2012); Gul et al. (2013), we choose a measure that includes only interest-bearing debt. Li et al. (2010) support the use of CoD in analyzing the consequences of auditor choice for several reasons: the public debt market is significantly larger than the equity market in some contexts; CoD is relatively well defined with less mis-specification than cost of equity; CoD is not affected by the difference of more or less sophisticated investors given that the information environment in the debt market is characterized by numerous information intermediaries.
Public clients is respectively PUC, %PUC and #PUC for the three hypotheses. To compute these variables, we start from the list of audit firms that audit a Public Interest Entity prepared by the Italian oversight board.4 Next, we download Transparency Reports and we identify the number of public clients of each audit firm.
Independent variables were selected on the basis of numerous prior studies on CoD (Kim et al., 2011; Aobdia et al., 2015; Chin et al., 2014; Petersen and Rajan, 1994; Bharath et al., 2008;
Karjalainen, 2011; Graham et al., 2008; Lai, 2011; Pittman and Fortin, 2004). The literature on cross- sectional determinants of loan pricing, in general, finds that firm SIZE is inversely related to credit risk.
Agency theory predicts that the risk of agency conflicts, such as risk shifting and underinvestment, between a firm’s insider and outside lenders increases with financial leverage and leverage maturity structure. To control for this, we include LEVERAGE (Kim et al., 2011; Bharath et al., 2008; Graham et al., 2008; Aobdia et al., 2015; Karjalainen, 2011; Pittman and Fortin, 2004). QUICK or current ratios have been used in prior studies as a proxy of financial risk. Firms with a low value of this ratio may be suffering from liquidity problems, and they may be forced to use more expensive credit (Bharath et al., 2008; Aobdia et al., 2015). It is important to control for profitability through ROA;
banks and other private lenders are likely to charge lower interest rates to firms that are more profitable because such firms are better able to service their debt (Kim et al., 2011; Graham et al. 2008; Aobdia et al., 2015). We include TANGIBLE in order to have a measure of asset composition as determinant of CoD. The loan pricing literature suggests that owning tangible assets is inversely related to credit risk, given that they can work as collateral and, thus, the interest rate that lenders charge (Bharath et al., 2008; Aobdia et al., 2015; Graham et al. 2008; Kim et al., 2011; Karjalainen, 2011; Pittman and Fortin, 2004). We include the ALTMAN score of bankruptcy because debt holders may demand higher interest to cover this higher risk (Lai, 2011; Bharath et al., 2008; Graham et al., 2008; Aobdia et al., 2015). Lower values indicate more financial distress, so that a negative association is expected with CoD. Because about 2.8 percent of private Italian companies in our sample experienced negative equity during the sample period, we include the NEGATIVE EQUITY dummy variable as an additional control for credit risk. Firms with negative equity are riskier financially, and the debt holder may charge them higher interest as compensation (Kim et al., 2011; Karjalainen, 2011; Pittman and Fortin, 2004). We include LOAN MATURITY, because the lender requires a liquidity premium for longer-term debt and this liquidity premium translates into a higher loan spread (Bharath et al., 2008;
Aobdia et al., 2015; Graham et al., 2008; Lai, 2011; Karjalainen, 2011).5
We use propensity-score matching models, developed by Rosenbaum and Rubin (1983), to match a range of client characteristics to examine whether the auditor distinction can be attributed to
4Public Interest Entities are: entities governed by the law of a Member State and whose transferable securities are admitted to trading on a regulated market of any Member State; credit institutions in the EU; insurance undertakings in the EU; or designated by Member States as public-interest entities, for instance undertakings that are of significant public relevance because of the nature of their business, their size or the number of their employees (European Parliament, 2013).
5 Our analysis focuses on the CoD on the banks and other financial institutions. In our sample there are no public debts.
specific client characteristics6. Propensity-score matching models match observations based on the probability of undergoing a treatment, which in our case is the probability of selecting a PUC, %PUC below the median and #PUC<=3. Definitions of variables are shown in Appendix A. We use logit models, the most frequent approach (Guo and Fraser 2010, 135)7. We match with replacements with the following Equation (2) untabulated, within a maximum distance of 1 percent8:
Public clients dummyit = α + β1 SIZEit + β2 LEVERAGEit + β3 LOSSit + β4 ASSET_TURNOVERit + β5 QUICKit + β6 SIZE SQUAREit + industry fixed effect + year fixed effect+ e (2).
See Appendix A for the definition of Public clients dummy.
Independent variables are chosen on the basis of studies on audit firm choice.9 We include the most frequently used variables. We include SIZE, as included by all the studies analyzed, because large firms are expected to raise high quality of auditors, because they are better equipped to handle the audit efficiently (Chaney et al., 2004). We include LEVERAGE because high leveraged firms tend to choose higher quality auditors to reduce their higher agency costs (e.g., Chaney et al., 2004; Fortin and Pittman, 2007). We include LOSS to control for profitability.10 We include ASSET TURNOVER to control for transaction complexity because highly complex firms tend to choose high quality auditors equipped to handle the complexity (e.g., Chaney et al., 2004). We include QUICK ratio to control for financial risk as riskier firms tend to choose higher quality auditors with more experience and competences to audit risker clients more efficiently (e.g., Chaney et al., 2004; Fortin and Pittman, 2007). As suggested in DeFond et al. (2016) and as done by Fortin and Pittman (2007), we control for potential nonlinearities by including both firm size and its SQUARE. We choose to put the nonlinear term on size, because Lennox (2005) finds that the relation between auditor choice and size is not linear. We use some of the variables in both the choice model and in the CoD model (SIZE, LEVERAGE, QUICK). Following the suggestion of Lennox et al. (2012) of an exclusion restriction, we exclude asset turnover from the CoD model, given that it is not a significant determinant of interest rate. We also exclude LOSS from the CoD model, in the belief that the best CoD determinant is ROA.
We also exclude the nonlinear term of size. We also add specific control variables that influence CoD.
We next compute the goodness of the propensity score match using a Bias measure.11 Estimating Equations (1) and (2) we test the multivariate effect on CoD in the common support sample
6 See Lawrence et al. (2011), Lennox et al. (2012), DeFond et al. (2016) for an explanation of the difference between this method and Heckman (1979) model, and a description of matching models. Lennox et al. (2012) suggest that future research should make exclusion restriction, putting in the main model not the same variables used in the choice model and should explain why they decide to exclude the specific variables based on theory. They also suggest to report the independent variables used in all the models, and perform sensitivity analyses. Lawrence et al. (2011) do this sensitivity analysis reporting that results are robust using probit or logit, using matching with or without replacement, using bootstrapping, kernel weighting, and random subsamples, ordinary least square, Heckman self-selection model. They also in the main model include some new independent variables or excluding some of the variables used in the choice model and give the explanation for this different inclusion/exclusion. DeFond et al. (2016) argue that limitations of PSM are related to the research design, such as the number of control firms matched to each treatment firm (one-to-one or one-to-many matching), the closeness of the match (caliper distance), the non-linear terms included in the propensity score construction, and the replacement decision. They suggest remedies repeating the analysis varying all these research design choices. In this study, we repeat the analysis with different research design choices to address these issues and following the suggestion of Lennox et al. (2012).
7 All findings documented in this study are robust to using a probit model instead of a logit model to calculate propensity scores.
8 Results are the same whether we match with or without replacement, and changing the caliper distance at 0.5%. Moreover, results are the same if we switch from one-to-one to one-to-many matching. We repeat the analysis with coarsened exact matching and kernel weighting and results are also consistent with these methodologies.
9 We reviewed the following research to define the frequency of the variables used: Shipman et al. (2015); Kim et al. (2003); Weber and Willenborg (2003); Li, (2009); Chang et al. (2009); Behn et al. (2008); Guedhami and Pittman (2006); Louis (2005); Pittman and Fortin (2004); Mansi et al. (2004); Johnstone et al. (2004); Fortin and Pittman (2007); Choi et al., (2008); Choi and Wong, (2007); Francis et al.
(1999); Chaney et al. (2004); Campa (2013); Boone et al. (2010); Eshleman and Guo, (2014); Khurana and Raman, (2004); Lawrence et al., (2011); Lennox et al. (2012).
10Additional variables are defined here: LOSS=1 if net income is < 0 and 0 otherwise; ASSET TURNOVER= ratio of revenues to total assets at the beginning of the fiscal year (winsorized at the 1st and 99th percentiles); SIZE SQUARE = size*size.
11 Bias measures the similarity of the distributions of the first stage explanatory variables between the treatment group and the control group. It is calculated for each explanatory variable by dividing the difference in the means between the treatment and control groups by the square root of the average sample variances of the two groups (Rosenbaum and Rubin, 1985).
when the weight is generated.12 All the Equations are estimated with industry and year fixed-effects, in order to control for systematic differences across industries and years in the sample13. For the sake of brevity, industry and year indicator variables are not reported in the tables.
5. Results
5.1. Survey Results
The survey appears to show that it is not the fact that an audit firm is of high quality that helps it to acquire public clients, but rather that when it acquires public clients it needs to change its structure to be of higher quality. In general, the most significant effect of the acquisition of public clients appears to be on the change in the audit procedures. This is operated by the firm itself or at the request the Italian audit firm oversight board. There may also be a change in the composition of the audit team which needs to acquire higher competences. Some of the procedures involved in these changes are related to risk assessment, internal monitoring and safeguarding independence.
The survey supports our argument that the effect of acquisition of new public clients are transferred also to private clients audited by the same audit firm. We consider this first evidence as support for our hypothesis in the empirical archival analysis.
5.2. Descriptive Statistics and Correlation Matrix
Table 2 shows the descriptive statistics of CoD and its control variables. The mean CoD for financial debts (7.3 percent) and for bank debts (untabulated) are similar: this confirms that banks are the main channel providing financial resources to private Italian companies. The mean CoD is consistent with the literature (e.g. Minnis, 2011).
The client size has a mean of about 10 (about €45 million euro). The financial leverage of the companies is relatively high, liabilities are in average 69.7 percent of total assets, which is consistent with our expectation that debt financing is important in privately held firms. The Altman score shows the level of the bankruptcy problem, which lies about 1.475 in average, consistent with the literature (median in Reichelt and Wang, 2010). Firms have a low profitability (ROA of about 1 percent) given that in the period analyzed companies had not recovered yet from the crisis. Our sample firms have a relatively low level of tangible assets (in average 25.8 percent of total assets). On average, about 2.4 percent of private companies in our sample have negative equity during the sample period. Finally, the loan maturity shows that short-term debts are on average 78.8 percent of long-term debts. This means that short-term debts are lower than long-term debts. Within the long-term debts, loans account for the majority (untabulated). In Italy there are more bank loans than financing from bonds and other forms than in United States. Mansi et al. (2004) discuss that in the United States public debt securities represent a significant portion of the typical corporation’s value.
The purpose of PSM is to identify very similar companies, with the sole difference being the auditor chosen, for the purpose of comparison. There are no statistically significant differences among subsamples for the following variables: size, leverage, loss, asset turnover or quick ratio (untabulated).
This come as the result of the application of PSM. The test for mean difference shows that also other independent variables are very similar for our subsamples. This shows that our sample of private firms is balanced for each group.
12The software Stata creates a weight variable automatically. For observations in the treated group, _weight is 1. For observations in the control group it is the number of observations from the treated group for which the observation is a match. If the observation is not a match, weight is missing.
13 To run audit firm fixed effect, the independent variables must change across time for some substantial portion of the individuals. This is not the case in this study, because we know only the current audit firm for each client and the number of years of tenure since its engagement started, but we do not have information on the past audit firm.
Table 2 Descriptive statistics
Note. *,**,*** is respectively 0.1, 0.05, 0.001 the p-value of the t-test of the difference in the mean between PUC=1 and PUC=0 or between %PUC>=median and %PUC<median or between #PUC<=3 and #PUC>3 .
The correlation matrix (Table 3) does not show substantial problems of multicollinearity. The mean variance inflation factor is under 4. In this univariate analysis, PUC, %PUC, #PUC are negatively correlated with CoD, suggesting that they have higher audit quality, which is consistent with our expectation. CoD is also correlated with higher quick ratio, loan maturity and lower size, tangible as its determinants.
Table 3: Correlation matrix
1 2 3 4 5 6 7 8 9 10 11 12
1 CoD 1.000
2 PUC -0.041 1.000
3 %PUC -0.062 - 1.000
4 #PUC -0.073 - - 1.000
5 Size -0.056 -0.024 0.013 0.008 1.000
6 Leverage -0.007 0.043 0.030 0.022 -0.095 1.000
7 Quick 0.057 -0.016 -0.016 0.020 -0.067 -0.379 1.000
8 ROA -0.031 -0.030 0.039 -0.018 0.123 -0.208 0.132 1.000
9 Tangible -0.064 -0.041 -0.055 -0.035 0.074 -0.273 -0.049 -0.012 1.000
10 Altman 0.045 -0.004 -0.011 -0.087 -0.065 -0.332 0.281 0.433 -0.193 1.000 11 Negative
Equity 0.012 0.064 0.042 0.053 -0.103 0.283 -0.104 -0.403 -0.050 -0.242 1.000 12
Loan
Maturity 0.101 -0.036 0.042 -0.001 -0.115 0.136 -0.190 0.061 -0.437 0.272 0.103 1.000
Note. Pearson correlation coefficient. Refer to Appendix A for variable definitions. The correlation matrix is calculated for the sample of 1138 observations, except columns and rows 2 and 3 that is calculated for the sample of 718, 514 observations. Significant coefficient at 0.10 are in bold.
5.3. Regression Results
In Table 4, results of multivariate regression analysis reveal that coefficients of PUC (first column),
%PUC (third column) and #PUC (fifth column) are negative and statistical significant. Coherently with our first hypothesis, we find that the presence of public clients in a non-Big4 portfolio is negatively associated with CoD of its private clients (coefficient -0.008). Confirming our expectation, we find that the higher developed knowledge, sense of the reputational risks posed by public clients (Ittonen et al., 2015), higher profile of public companies linked to larger potential reputational losses (Clatworthy and Peel, 2007) higher litigation risk (Hay et al., 2006) and a larger set of stakeholders are factors leading to improve audit quality of audit firms with public clients. We support the argument that higher audit quality from competences, reputation, litigation incentives and stakeholders’ relationships can be
Full sample (N=1138)
PUC=1 (N=569)
PUC=0 (N=569)
%PUC
<
median (N=359)
%PUC
>=
median (N=359)
#PUC
<=3 (N=297)
#PUC
>3 (N=297)
Mean SD 25th p. Median 75th p. Mean Mean Mean Mean Mean Mean
Dependent variables
CoD 0.073 0.079 0.035 0.050 0.076 0.069 0.076 0.064 0.068 0.068 0.062
Independent variables
Size 10.078 1.337 9.341 10.162 10.935 10.047 10.110 10.483 10.364 10.482 10.455
Leverage 0.697 0.210 0.588 0.732 0.850 0.706 0.688 0.689 0.669 0.703 0.715
Quick 0.974 0.738 0.554 0.806 1.162 0.962 0.985 1.142 1.261 1.014 1.062
ROA 0.019 0.081 0.001 0.021 0.048 0.017 0.022 0.011 0.003 0.014 0.015
Tangible 0.258 0.244 0.050 0.194 0.387 0.248 0.268 0.258** 0.306 0.279** 0.235
Altman 1.475 1.337 0.710 1.292 2.031 1.470 1.480 1.294 1.278 1.368 1.348
Negative
Equity 0.024 0.152 0.000 0.000 0.000 0.033** 0.014 0.050 0.028 0.017* 0.044
Loan
Maturity 0.788 0.227 0.687 0.860 0.966 0.780 0.796 0.753 0.726 0.762 0.784
transferred from public clients to private clients and these high-quality auditors help to reduce the CoD of private firms.
Moreover, focusing on a matched sample of companies audited by non-Big4 with public clients, we find that the percentage of public clients out of total clients of a non-Big4 audit firm is negatively associated (coefficient -0.150) with CoD of private firms (second hypothesis). In other words, we find that positive effect on CoD could be improved by a high percentage of public clients out of the total clients.
Finally, on a matched sample of companies audited by non-Big4 with public companies, we find that the number of public clients (third hypothesis) is negatively associated (coefficient of -0.007) with CoD of private firms. Confirming our third hypothesis, we find that another significant variable useful to improve benefits coming from public clients is their absolute number. However, looking at the magnitude of the coefficient, the highest impact is given by the proportion of public client out of total clients.
Results related to control variables in the multivariate regression analysis (Table 4) show a negative relation between size, Altman score and CoD, and a positive relation between quick ratio, loan maturity and CoD. Size and Altman are inversely related to bankruptcy because debt holders demand higher interest to cover this higher risk (Lai, 2011; Bharath et al., 2008; Graham et al., 2008;
Aobdia et al., 2015); the lender requires a liquidity premium for longer-term debt, and this liquidity premium translates into higher loan spread (Bharath et al., 2008; Aobdia et al., 2015; Graham et al., 2008; Lai, 2011; Karjalainen, 2011). On the other hand, the quick ratio does not drive the choice of more expensive credit.
Table 4: Multivariate analysis of Cost of Financial Debt
Non-Big4 audit firms with Public Clients compared with Non-Big4 audit firm without Public Clients
Percentage of Public clients over total clients and Number of Public Clients
Estimate P-value Estimate P-value Estimate P-value
PUC -0.008 0.093
%PUC -0.150 0.052
#PUC -0.007 0.033
Size 0.001 0.694 -0.005 0.054 -0.004 0.251
Leverage 0.004 0.833 0.013 0.530 0.014 0.409
Quick 0.007 0.061 0.005 0.273 0.014 0.083
ROA -0.035 0.363 0.040 0.287 0.062 0.242
Tangible 0.000 0.978 -0.021 0.140 -0.013 0.473
Altman 0.001 0.817 0.001 0.673 -0.009 0.018
Negative Equity 0.000 0.986 0.017 0.461 0.016 0.668
Loan Maturity 0.049 0.001 0.031 0.053 0.040 0.059
Constant -0.020 0.667 0.087 0.019 0.056 0.369
Adjusted R-Squared 0.051 0.073 0.068
Year and Industry
Fixed Effect included included included
Observations 1138 718 594
Mean bias 4.7 5.4
Median bias 4.8 4.4
P-value 0.859 0.471
Note. Coefficient p-values are two-tailed, based on asymptotic t-statistics using White (1980) standard errors. Refer to Appendix A for variable definitions. PSM is run with replacement, at caliper (0.01), common, logit.
6. Robustness
6.1. Propensity Score Matched Sample
PSM can be performed with many specifications. We repeat the analysis with kernel matching, in which all treated units are matched with a weighted average of all control units with weights that are inversely proportional to the distance between the propensity scores of treated units and control units.
Calculation of weighting depends on the specific kernel function adopted. We repeat the analysis without replacement, changing the caliper distance at 0.5% and switching from one-to-one to one-to-
many matching. We follow D’Attoma and Pacei (2014) in presenting the results for different methods of PSM.
Table 5 reports that after matching, the mean bias for all explanatory variables is reduced to acceptable levels (Harder et al., 2010). It falls from above 8 before matching to below 5 after matching for PUC, from above 12 before matching to below 4 after matching for %PUC, from above 4 before matching to below 3 after matching for #PUC14. Table 5 also reports that, after matching, the p-values of the joint significance of the explanatory variables are not significantly different between the treatment group and the control group. In short, these test statistics suggest that the matching method is appropriate. We report the coefficients on PUC, %PUC, #PUC with the corresponding p-value and the number of observations in these regressions. Results are confirmed.
To investigate whether a high-quality auditor reduces CoD, Coarsened Exact Matching (CEM) is also used. CEM overcomes some of the limitations inherent in PSM (King et al., 2011; Iacus et al., 2012). CEM is a more robust matching technique that is not subject to random matching, because it directly matches on a coarsened range of covariates and does not rely on a first-stage propensity score model. DeFond et al. (2016) encourage research to explore the use of CEM in complementing regression analysis for the purpose of providing robust inferences. We use the same variables used in the first stage propensity score to perform the match. CEM shows the same results as PSM. We can therefore conclude that results are not driven by endogeneity.
Table 5: Alternative estimation of Propensity Score Matching
PS matching
Mean (Median bias) p-value
Estimate (N)
Mean (Median bias) p-value
Estimate (N)
Mean (Median bias) p-value
Estimate (N) Before
matching
After
matching PUC Before matching
After
matching %PUC Before matching
After
matching #PUC Kernel
(normal;
bandwidth = 0.06)
10.8 (7.8) 0.000
1.6 (1.4) 1.000
-0.008**
(1516)
12.4 (13.6) 0.000
3.1 (1.8) 0.983
-0.103 (923)
5.9 (4.3) 0.000
2.0 (1.2) 1.000
-0.002 (941)
Without replacement
2.6 (2.5) 0.996
-0.010**
(1021)
2.4 (1.7) 1.000
-0.143*
(626)
2.9 (2.2) 0.993
-0.005*
(573) Caliper
(0.005)
4.3 (4.3) 0.712
-0.007 (1122)
3.9 (2.8) 0.969
-0.156**
(698)
5.7 (4.3) 0.402
-0.007**
(579) One-to-
many (many=3)
3.2 (2.6) 0.984
-0.009**
(1276)
3.9 (3.3)
0.916) -0.125*
(809)
3.0 (2.1) 0.999
-0.004 (737)
CEM -0.011**
(935)
-0.091 (535)
-0.007**
(518) Note. Coefficient p-values are one-tailed, based on asymptotic t-statistics using White (1980) standard errors. Pseudo R2
for PSM p-values are two-tailed. See Appendix A for variable definitions. Results for Kernel (normal) and Kernel (Epanechnikov) are very similar.
6.2. Alternative Measure of Agency Costs between Lenders and Managers
We also repeat the analysis using the credit default risk rating provided by modeFinance. This company provides the Multi Objective Rating Evaluation (MORE) in order to assess the level of distress of industrial companies. It provides a creditworthiness opinion of risk class on a ten-point scale. See Table 6 for variable definition. The rating can be used for access to loans in negotiations with banks. This type of rating has been used in prior literature to develop measurement of CoD in unlisted firms (Intrisano et al., 2016). We requested the data on this rating for the matched sample used in the CoD analysis. Results show that clients of %PUC are associated with higher ratings than firms with a lower default risk (positive coefficient of 6.854 in Table 6, first column).
We repeat the analysis using a different proxy of the dependent variable CoD. We compare the financial costs to different values of the debt, changing the denominator of the variables. We use a
14 With the exception of caliper (0.005).
more restricted Cost of interest-bearing Debt, including only the Cost of Bank Debt15. This is an interesting measure in Italy where private companies are mainly financed by banks and not by bonds, as shown by the descriptive statistics. The results are confirmed (negative coefficient of -0.160 in Table 6, third column).
6.3. Public versus Private Clients
The regressions are next run using an enlarged sample including public firms. Nearly 206 public non- financial firms are audited every year, and the number of observations for the CoD analysis rises from 718 to 731. We find that the effect of audit firm reputation is more important for public firms, given the significant negative coefficient on the interaction %PUC*PUBLIC (-0.742 in Table 6, fifth column). This suggest that higher reputation, competences and litigation risk from the audit of a public client affects another public clients’ audit. The externalities borne for setting up a public client audit are higher for other public clients than for other private clients.
Table 6: Other robustness tests
Credit default Rating
class Cost of bank debt Sample of public
and private firms Interaction with IFRS Cost of Debt Estimate P-value Estimate P-value Estimate P-value Estimate P-value
%PUC 6.854 0.081 -0.160 0.090 -0.226 0.024
%PUC*PUBLIC -0.742 0.097
%PUC*IFRS -0.807 0.060
Control variables … … … … … … … …
Adjusted R-
Squared 0.261 0.078 0.094 0.080
Year / Industry
Fixed Effect included included included included
Observations 270 689 731 497
Note. Coefficient p-values are one-tailed, based on asymptotic t-statistics using White (1980) standard errors. See Appendix A for variable definitions. All the regressions presented are run on the propensity score matched sample.
Data availability for Credit default Rating class is only for audit firms with public clients. Additional variables are defined here:
CREDIT DEFAULT RATING CLASS= ordered variable that assumes value 10 if the rating is AAA (extremely strong), 9 if the rating is AA (strong), 8 if the rating is A (high solvency), 7 if the rating is BBB (adequate), 6 if the rating is BB (adequate in the country-industry), 5 if the rating is B (vulnerable), 4 if the rating is CCC (dangerous), 3 if the rating is CC (high vulnerable), 2 if the rating is C (pathological situations), 1 if the rating is D (no capacity to meet financial commitments)
PUBLIC= 1 if the client firm-year observation is listed; 0 if the client firm-year is unlisted.
IFRS= 1 if the unlisted firm is voluntary using International Financial Reporting Standards (IFRS) and 0 if the unlisted frim is using the national general accepted accounting principles.
IFRS and PUBLIC are used as independent variable also in the first stage logistic model to match samples.
15 Cost of capital in the audit literature (Khurana and Raman, 2004; Iatridis, 2012; Azizkhani et al., 2013; Cassell et al., 2013; Lawrence et al., 2011; Guedhami et al., 2014; Choi and Lee, 2014) has been measured by ex-ante cost of equity capital (for example with the models of Gebhardt et al., 2001; Claus and Thomas, 2001; Ohlson and Juettner-Nauroth, 2005; Easton, 2004; Gode and Mohanram, 2003). These models imply the use of financial analyst earnings forecasts and stock prices that are not available for private firms.
Other studies (Mansi et al., 2004; Fortin and Pittman, 2007; Li et al., 2010) measure the cost of capital with the marginal cost of debt (the yield to maturity at the issuance date for the largest bond the firm issued in year t+1, minus the Treasury bond yield with similar maturity) and the Standard & Poor's senior debt rating in year t. Standard & Poor's rates a firm's debt from AAA (indicating a strong capacity to pay interest and repay principal) to D (indicating actual default). Bond rates are less well-fitted in this context, given that the main source of financing is from banks and not from bondholders. In private firms, bonds are often similar to stock option and they may represent a supplement to shareholder remuneration.
The cost of total debt, measured using as denominator the amount of total debts (Pittman and Fortin, 2004; Kim et al., 2011; Lai, 2011;
Minnis, 2011; Causholli and Knechel, 2012) has a mean value of about 2% with a standard deviation of 2%, similar to other countries like Korea (about 2% in Kim et al., 2011), lower than United States (about 7% in Minnis, 2011). In Italy, cost of total debt is much lower because it includes non-interest-bearing debt. This proxy is therefore excluded from the analysis.
6.4. IFRS versus Italian GAAP
Effects would be higher if private clients use the same set of standards as public clients. In general, private firms adopt Italian General Accepted Accounting Principles (GAAP) and some of them voluntarily adopt International Financial Reporting Standards (IFRS). Bozkurt et al. (2013) find that accounting and auditing professionals think that in order to increase application of the standards at national level, increase in collaboration and information flow among institutions are necessary. In Italy this collaboration is high and the application of IFRS from voluntary adopters is high. We repeat the regression adding an interaction between %PUC and IFRS. Results show significant negative coefficients for the interaction %PUC*IFRS (-0.226 in Table 6, seventh column). The externalities are higher when the client adopts the same standards as the public clients that the firm also audits.
7. Conclusion
Based on previous literature (Gul et al., 2013, Ittonen et al., 2015, Clatworthy and Peel, 2007, Hay et al., 2006), this research aims to test the negative association between the presence, the percentage and the number of public clients in a non-Big4 audit firm’s portfolio and the CoD of its private clients.
We first run survey to partners of non-Big4 about the effects of public clients: they confirm that competences and higher audit quality that non-Big4 audit firm get from the audit of public clients are mostly transferred to their private clients. Using data from a sample of Italian private companies audited by non-Big4 audit firms with and without public companies, we next perform multivariate regression models on propensity score matched samples. Confirming our expectation, we find that: 1) the presence of public clients in a non-Big4 audit firm’s portfolio is negatively associated with CoD; 2) the percentage of public clients out of total clients of a non-Big4 audit firm is negatively associated with CoD of its private clients; 3) the number of public clients of non-Big audit firm is negatively associated with CoD of its private clients. Benefits associated with public clients, their number and percentage (higher competences, reputation, profile, litigation risks) can be transferred from public to private clients, and one of these benefits is the reduction of CoD of private firms.
We contribute to literature in several ways. Firstly, we improve the strand of audit literature that studies the voluntary audit market and the non-Big4 segment. Covering a gap in the audit literature, we show that non-Big4 audit firms gain a significant competitive advantage when they accept the challenge coming from public clients. Secondly, we contribute showing that other significant variables that could improve previous benefits are the percentage of public clients over total clients and the number of public clients. Thirdly, our result could be useful for practitioners in the selection of audit firm, suggesting them to choose non-Big4 with high number and percentage of public clients to mitigate the agency conflict between lenders and manager in private firms.
References
[1] Allee, K. D. and Yohn, T. L. (2009) ‘The Demand for Financial Statements in an Unregulated Environment: An Examination of the Production and Use of Financial Statements by Privately Held Small Businesses’, The Accounting Review, 84(1), pp. 1–25. doi:
10.2308/aecr.2009.84.1.1.
[2] Altman, E. I. (1983) Corporate Financial Distress: A Complete Guide to Predicting, Avoiding, and Dealing with Bankruptcy. New York: John Wiley and Sons. Available at:
https://books.google.it/books/about/Corporate_Financial_Distress.html?id=GbS- AAAAIAAJ&pgis=1 (Accessed: 12 June 2015).
[3] Aobdia, D., Lin, C. and Petacchi, R. (2015) ‘Capital Market Consequences of Audit Partner Quality’, The Accounting Review, 90(6), pp. 2143–2176. doi: 10.2308/accr-51054.
[4] Azizkhani, M., Monroe, G. S. and Shailer, G. (2013) ‘Audit partner tenure and cost of equity capital’, Auditing: A Journal of Practice and Theory, 32(1), pp. 183–202. doi: 10.2308/ajpt- 50308.
[5] Vander Bauwhede, H. and Willekens, M. (2004) ‘Evidence on (the lack of) audit-quality differentiation in the private client segment of the belgian audit market’, European Accounting Review, 13(3), pp. 501–522. doi: 10.1080/0963818042000237106.
[6] Behn, B. K., Choi, J. and Kang, T. (2008) ‘Audit Quality and Properties of Analyst Earnings Forecasts’, The Accounting Review, 83(2), pp. 327–349. doi: Article.
[7] Bharath, S. T., Sunder, J. and Sunder, S. V (2008) ‘Accounting Quality and Debt Contracting’, The Accounting Review, 83(1), pp. 1–28.
[8] Blackwell, D. W., Noland, T. R. and Winters, D. B. (1998) ‘The Value of Auditor Assurance:
Evidence from Loan Pricing’, Journal of Accounting Research, 36(1), pp. 57–70. doi:
10.2307/2491320.
[9] Boone, J. P., Khurana, I. K. and Raman, K. K. (2010) ‘Do the Big 4 and the Second-tier firms provide audits of similar quality?’, Journal of Accounting and Public Policy, 29(4), pp. 330–
352. doi: 10.1016/j.jaccpubpol.2010.06.007.
[10] Bozkurt, O., Islamoglu, M. and Öz, Y. (2013) ‘Perceptions of professionals interested in accounting and auditing about acceptance and adaptation of global financial reporting standards’, Journal of Economics, Finance and Administrative Science, 18(34), pp. 16–23.
[11] Campa, D. (2013) ‘"Big 4 fee premium” and audit quality: latest evidence from UK listed companies’, Managerial Auditing Journal, 28(8), pp. 680–707.
[12] Cano-Rodríguez, M. and Alegría, S. S. (2012) ‘The value of audit quality in public and private companies: evidence from Spain’, Journal of Management & Governance, 16(4), pp. 683–706.
doi: 10.1007/s10997-011-9183-4.
[13] Cassell, C. a., Giroux, G., Myers, L. a. and Omer, T. C. (2013) ‘The Emergence of Second-Tier Auditors in the US: Evidence from Investor Perceptions of Financial Reporting Credibility’, Journal of Business Finance and Accounting, 40(3–4), pp. 350–372. doi: 10.1111/jbfa.12016.
[14] Causholli, M. and Knechel, W. R. (2012) ‘Lending relationships, auditor quality and debt costs’, Managerial Auditing Journal, 27(6), pp. 550–572. doi: 10.1108/02686901211236391.
[15] Chaney, P. K., Jeter, D. C. and Shivakumar, L. (2004) ‘Self-Selection of Auditors and Audit Pricing in Private Firms’, The Accounting Review, 79(1), pp. 51–72.
[16] Chang, X., Dasgupta, S. and Hilary, G. (2009) ‘The Effect of Auditor Quality on Financing Decisions’, The Accounting Review, 84(4), pp. 1085–1117.
[17] Chin, C., Yao, W. and Liu, P. (2014) ‘Industry Audit Experts and Ownership Structure in the Syndicated Loan Market: At the Firm and Partner Levels’, Accounting Horizons, 28(4), pp.
749–768. doi: 10.2308/acch-50825.
[18] Choi, J. H. and Lee, W. J. (2014) ‘Association between Big 4 auditor choice and cost of equity capital for multiple-segment firms’, Accounting and Finance, 54(1), pp. 135–163. doi:
10.1111/acfi.12011.
[19] Choi, J. and Wong, T. J. (2007) ‘Auditors ’ Governance Functions and Legal Environments : An International Investigation’, Contemporary Accounting Research, 24(1), pp. 13–46.
[20] Clatworthy, M. and Peel, M. J. (2007) ‘The effect of corporate status on external audit fees:
Evidence from the UK’, Journal of Business Finance and Accounting, 34(1–2), pp. 169–201.
doi: 10.1111/j.1468-5957.2006.00658.x.
[21] Claus, J. and Thomas, J. (2001) ‘Equity Premia as Low as Three Percent? Evidence from Analysts’ Earnings Forecasts for Domestic and International Stock Markets’, Journal of Finance, 56(5), pp. 1629–1666. doi: 10.1111/0022-1082.00384.
[22] Collis, J. (2012) ‘Determinants of voluntary audit and voluntary full accounts in micro- and non-micro small companies in the UK’, Accounting and Business Research, 42(4), pp. 441–
468. doi: 10.1080/00014788.2012.667969.
