THE INFLUENCE OF PROFESSIONAL QUALIFICATION ON CUSTOMER PERCEPTIONS OF ACCOUNTING SERVICES QUALITY AND RETENTION DECISIONS

The purpose of this paper is to examine whether professional qualification increases the quality of accounting services as perceived by the customers. We advance the measurement of service quality by industry-specific indicators and establish four dimensions of accounting service quality. We analyse the impact of professional qualification on these dimensions of quality and the impact of these dimensions on customers’ retention decisions. The analysis is based on survey data of 237 Slovenian small and micro firms that outsource accounting. We find that professional qualification is positively associated with only one of the perceived service quality dimensions – accounting competences – and only assurance, responsiveness and reliability and empathy are positively associated with customers’ retention decisions. Limitations of the study are attributed to the measurement of service quality as perceived by the customers. However, this is the only factor of choice that ultimately counts in the competitive market for accounting services.


Introduction
Due to the public interest dimension of accounting, the accounting profession is strictly regulated, but only where the benefits of safeguarding the public interest from the consequences of market failure offset its costs (O'Regan 2010). It is therefore not surprising that regulation of the accounting profession relates only to public interest companies. Furthermore, it is limited to auditors and does not relate to the provision of accounting services in the narrower sense (i.e. book-keeping and reporting). Currently, the EU regulations include no particular requirements regarding the professional qualifications of accountants. It seems anomalous that preparers of accounts are not subject to the same regulation as auditors yet it may well be that an accountant of an EU public company is not professionally qualified.
Because of lack of expertise and the related cost, small and micro companies tend to outsource accounting services to professionals that are specialised in the provision of accounting services to small businesses. In a non-regulated setting, providing accounting services to small and micro companies is open to anyone. Furthermore, the trend of regulating financial reporting of small and micro firms is moving away from tighter restrictions: the recently adopted EU Directive 2012/6/EU has introduced financial reporting simplifications for micro companies. Although simplification of financial reporting is desirable from the administrative cost perspective, it may not contribute to higher quality financial information for internal and external users. In these circumstances, service quality delivered by professional accountants is a highly relevant concern.
The purpose of our paper is to analyse whether professional qualification of accountants providing services to small and micro firms is positively associated with the quality of accounting services provided. Service quality is a complex construct, intensively studied in the marketing and quality management literature for the last 30 years (Parasuraman et al. 1985Cronin, Taylor 1992;Zeithaml et al. 1996;Rafiq et al. 2012). Existing definitions of service quality are to a large extent applicable to accounting services, but the specifics of this market and the importance of the public interest in particular must be considered and incorporated into this construct. To the best of our knowledge, only few studies have specifically addressed service quality in accounting: Scott and Walt (1995), Aga and Safakli (2007), and Walker et al. (2012). More specifically, the research questions of the study are: 1) to examine which dimensions of accounting service quality are positively associated with professional qualification and 2) to analyse which dimensions of service quality affect customers' retention decisions.
As the setting for our research, Slovenia was chosen as a representative of small Central and East European economies. Due to the relatively short accounting regulation history in these countries (other than in a planned economy) and the lack of strong and established professional organisations, these countries beg the question as to whether regulation of the accounting profession would increase the quality of accounting services.
Our study provides the following contributions to the literature. First, it provides empirical evidence on the effect of professional qualification on accounting service quality and is thus relevant for accounting regulation. Second, it extends prior research by elaborating the construct of accounting services quality. Third, it provides evidence on the dimensions of quality that affect costumers' retention decisions and thus informs accounting firms about factors that underpin long-term relationships with customers. Fourth, it discusses the regulatory implications of the low weight given by customers to professional qualification in the context of the public interest concern.

Professional qualification, accounting service quality and customers' retention decisions
In the EU, the vast majority of companies (92.2 percent) are micro firms with fewer than ten employees (Annual… 2012). Therefore, small and micro firms have a significant collective impact on national economies.
Although not subject to as much oversight and regulation as financial statements of public companies, financial statements of small and micro companies are important to various groups of users. The question of whether public interest is respected in the nonregulated segment of the accounting services market is therefore of high importance. In the light of the financial crisis, the demand for high quality accounting services is continuously increasing. Although accounting services in the narrow sense (bookkeeping and reporting) are non-regulated, professional organisations organise professional education programmes leading to professional qualifications. Given the scope, sophistication and prices of these programmes they are less suitable for the accountants that provide services to small and micro firms.
Regarding the professional qualification of accountants, abundant literature concentrates on the presentation and comparison of different professional certification training and the relevance of professional certification in accounting education (Strouhal, Zemánek 2008;Boyd et al. 2009;Cory, Huttenhoff 2011;Wheeler, Low and Samkin 2014). However, none of these studies analyses the impact of professional qualification on the quality of services provided.
In the marketing and management quality literature service quality is defined as conformance to customer specifications (Berry et al. 1988: 35;Zeithaml, Bitner 2003). Research suggests that customers perceive service quality as a combination of multiple factors, relevant in a given context. Berry et al. (1988) identify the following five service quality dimensions: reliability, assurance, tangibles, empathy and responsiveness. Different authors suggest diverse types of dimensions for different sectors and there is no clear consensus regarding the dimensions of service quality (Scott, Walt 1995).
Dimensions of service quality in accounting were investigated by Scott and Walt (1995), Armstrong and Davison (1995), Aga and Safakli (2007), and Walker et al. (2012). Scott and Walt (1995) analysed quality dimensions important for the selection of an audit firm. Armstrong and Davison (1995), Aga and Safakli (2007) explored service quality of professional accounting firms operating in Australia and North Cyprus respectively. Walker et al. (2012) investigated management accounting service quality.
In a discussion of the dimensions of accounting services quality, the role of public interest in accounting has to be considered. In line with the Public Interest Framework (recently presented by the International Federation of Accountants), securing the public interest in accounting practice relates primarily to preparation of high quality accounting information, application of high standards of ethical behaviour and sound professional judgment. Ethical behaviour and professional competence are also stressed by the International Ethics Standards Board of Accountants (IESBA) Code of Ethics for Professional Accountants (IESBA 2012). The Code requires that accounting professionals comply with five fundamental principles, namely integrity, objectivity, professional competence and due care, confidentiality and professional behaviour. Professional institutions that provide professional qualification programmes appear to broadly promote the notion of public interest and the respect of the Code of Ethics.
Given that professional qualifications are designed to increase the quality of accounting services, our first research question is whether professional qualification of accountants is positively associated with the quality of accounting services as perceived by the customers. We expect that among service quality dimensions professional qualification has the strongest influence on accountants' professional competences, described as the duty to maintain professional knowledge to ensure that clients receive a competent professional service. As professional qualification programmes promote other principles of the Code of Ethics (integrity, objectivity, confidentiality) which to a certain extent resemble the assurance, empathy, responsiveness & reliability dimensions of service quality, we assume that professional qualification has beneficial effect also on these other dimensions [of service quality]. Therefore, beyond its influence on accountant's professional competences, we expect professional qualification to positively affect assurance, responsiveness & reliability and empathy.
The association between professional qualification and accounting service quality is also important for accounting professionals when deciding whether it is worth gaining qualifications in terms of competitive advantage. Due to the positive effect of customer retention on the company's profit (Tatikonda 2013) an important question is whether accountants' professional qualification is associated with their customers' retention decisions mediated by the service quality. There is little benefit in holding professional qualifications if customers [who are free to choose] do not distinguish between qualified and unqualified accountants.
In the marketing and quality management literature, service quality is viewed as the central construct that drives customer loyalty (Cronin, Taylor 1992;Zeithaml et al. 1996;Fullerton 2005), which, in turn, is a central determinant of customer retention (Gerpott et al. 2001). The results of prior research related to accounting are somewhat inconclusive: Scott and Walt (1995) report that quality dimensions associated with customers' selection of an audit firm are not significantly associated with actual changes of audit firms.
Beyond increasing competences, a large emphasis in professional qualification is placed on the development of ethical behaviour and accountants' awareness of the public interest. This may not coincide with the interest of the clients. We therefore expect that perceived competences may not be as strongly related to customers' retention decisions as other dimensions of quality because accountants with higher competences may be less subservient to clients as regards reporting of accounting figures. Our second research question is, therefore, which of the quality dimensions are positively associated with customers' retention of their accountant.
Overall, we hypothesise that: i) professional qualification positively affects all dimensions of accounting service quality with the strongest impact being on professional competences and ii) customers' retention decisions are positively affected by all dimensions of accounting service quality with assurance, responsiveness & reliability and empathy having the strongest impact. We present these relationships in the following conceptual mode:

Methodology
Sample A random sample of 2,000 small and micro companies was drawn from the Slovenian register of companies Bizi.com. The selection of companies was based on the criteria for small and micro entities as defined in the EU Fourth Directive. The choice of this size of firm is based on the idea that most small and micro firms outsource the accounting function to an external provider, i.e. a professional accounting firm. Hence, the respondents represent the users of accounting services. We sent out e-mail questionnaires to executives in the period August to October 2012. We obtained 382 responses (a 19.1% response rate). Of these, 280 companies outsource financial reporting to accounting professionals (of these, 43 respondents indicated that they were not aware of whether their accountant has a professional qualification or not) and 102 carry out the financial reporting tasks themselves. We used the responses of the 237 firms that outsource the accounting services and are aware of the qualification of their accountant. The mean education of the respondents is at college degree level. 75% of the sample firms have fewer than 10 employees.

Professional qualification
The respondents were asked to indicate whether their provider of accounting services has completed one of the professional qualification programmes (QUALIF) organised by the Slovenian professional associations. In Slovenia, professional qualification programmes are provided by two professional organisations: The Slovenian Institute of Auditors and the Chamber of Accounting Service Providers. Only candidates with a college or university degree can enrol and the curricula of both programmes focus on accounting, financial reporting, accounting-related regulation, ethics, taxation and corporate, labour and other relevant legislation. The professional qualification of the accountant providing a service to the respondent is measured as a dummy variable.

Accounting service quality
The instrument to measure service quality in accounting was adapted from the SERV-QUAL instrument . The appropriateness of an unmodified measurement instrument for accounting services is severely limited (Rao, Kelkar et al. 1997) as industry-specific measures of service quality are needed to appropriately measure the construct. The fundamental difference between accounting services and other types of services relates to whom the service provider ultimately serves. Unlike other services, the provision of financial accounting information serves the public interest. From that perspective some quality dimensions may collide; it would, for example, not be appropriate for quality assessment to be dominated by an overly empathic attitude of the accounting firm towards the client.
Similar to many prior studies (Aldhizer et al. 2002;Waldmann, Raghavan 2002;Hong, Goo 2003;Saxby et al. 2004;Aga, Safakli 2007) we follow the original 10-dimensional and the truncated five-dimensional SERVQUAL instrument (Parasuraman et al. 1985; in the development of our instrument. The wording of indicators adapted to the accounting context is based on the Saxby's et al. (2004) study. Given the specific focus of our study, we complemented the scale with additional indicators and omitted one dimension. Additions to the scale were based on in-depth interviews with three accounting experts (practitioners) and our own judgment. It made sense to eliminate the visible dimension of accounting services, i.e. the tangibles. The full list of 29 initial items and their references is available from the authors upon request. The respondents were asked to indicate their personal judgements regarding specific features of accounting services on a 5-point Likert scale.
We explored the dimensionality of researched indicators by performing exploratory factor analysis (principal axis method, rotation varimax). The results are presented in Table 1. A four-factor solution is proposed: 1) competence (COMPET), 2) responsiveness and reliability (RESPREL), 3) assurance (ASSUR) and 4) empathy (EMPAT). The total variance extracted for all constructs is above 60%. Based on the factor analysis and without a loss in terms of substance of the constructs we trimmed the original list of 29 indicators to 21 to increase the discriminant and convergent validity of the constructs in the final structural model. Eight indicators were eliminated in the analysis because of their cross-loadings on different factors or the lack of significant loadings on any factor.

0.826
Q22 The behaviour of the accounting firm employees is in line with their profession.

0.739
Q20 Employees of the accounting firm are kind and polite.

0.714
Q13 Employees of the accounting firm are always willing to help the customer.

0.690
Q17 The accounting firm provides the services according to agreed deadlines.

0.686
Q14 Employees of my accounting firm respond to customers' questions and requests promptly.

0.629
Q12 Accounting firm's employees provide the service promptly.

0.592
Q18 The accounting firm provides the services as was agreed with the customer.

Retention decision
We assess the customer's retention decision (RETENT) by way of two indicators measuring the inclination to keep the same accounting firm for one and for five years. The variable is measured on a scale ranging from 1 to 5, whereby 1 represents improbability and 5 represents certainty that the accountant will be retained [in the future].

Model estimation method
SmartPLS 2.00 software (Ringle et al. 2005) utilising a partial least squares (PLS) approach to structural equation model (SEM) estimation (Wold 1985;Lohmöller 1989) is used to test the conceptual model summarised in Figure 1. PLS is a components-based structural equation modelling technique that focuses on maximising the explained variance of endogenous target constructs in the model. It is a prediction-oriented variancebased approach. It simultaneously estimates the structural paths and measurement paths in the model using an iterative ordinary least squares (OLS) regression-like procedure.
The PLS procedure has been attracting interest among accounting and management researchers in recent years (see Lee et al. 2011;Hair et al. 2012 for an overview); it has an advantage over the variance-covariance based structural equation modelling technique in that it allows the use of both formative and reflective variables in the model (Chin 1998) and is less restrictive regarding the distribution assumptions.

Measurement model
As shown in Table 2, the factor loadings for the separate dimensions of quality exceed the 0.7072 threshold, providing evidence of satisfactory individual item reliability. Additionally, we report cross-loadings for all indicators to present the discriminant validity of constructs. To assess the internal consistency of the constructs Table 3 presents Cronbach's α values ranging from 0.826 to 0.941. Corroborating the Cronbach's α analysis, Dillon-Goldstein's ρ (composite reliability) values are also at around 0.9 for each construct, providing strong evidence of unidimensionality.
Convergent and discriminant validity of the two composite constructs is assessed by the average variance extracted (AVE), which represents the average variance shared between a construct and its indicators. AVE greater than 0.50 indicates satisfactory convergent validity, whereas discriminant validity is assessed by comparisons of AVE to the variance shared between any two constructs. As shown in Table 3, the AVE values for all quality dimensions are greater than the 0.5 cut-off point, indicating an acceptable level of convergent validity. In Table 4, the square roots of the AVE are presented on the diagonal of the correlation matrix, and are all greater than the corresponding correlations. This provides evidence of sufficient discriminant validity of the reflective constructs despite relatively high cross-loadings, showing that each indicator loads higher on its theoretical construct than on any other factors.   Off-diagonal elements are the correlations among constructs. For discriminant validity, diagonal elements should be larger than off-diagonal elements. N.a. is non-applicable and relates to the construct measured with one item. ** p < 0.01, *** p < 0.001 (two-tailed test). Table 5 summarises the structural analysis and Figure 2 provides a diagrammatic representation of the significant structural paths in the proposed model. The predictive validity of the parameter estimates is estimated via a cross-validated redundancy index or so-called Stone-Geisser Q 2 test (Geisser 1974;Stone 1974) and R 2 . As shown in Table 3, the redundancy for all latent variables is satisfactory, with all Q 2 values being greater than 0, providing additional support for the model's predictive relevance.

Structural model
The estimated path coefficients (see Table 5 and Fig. 2) indicate the strength and the direction of the relationships among the specified variables. The statistical significance of the parameter estimates is assessed using a bootstrap procedure with 1,000 replacements. The analysis shows support for our conjecture that professional qualification has the strongest effect on competences of the accountants. The structural coefficient is positive (b = 0.192, p < 0.01), while there is no significant association with other dimensions of quality. As argued previously, the association between professional qualification and quality dimensions is more indirect. Competences seem to be indeed an antecedent of other quality dimensions and strongly influence them (COMP → ASSUR b = 0.264, p < 0.01; COMP → EMPAT b = 0.284, p < 0.01; COMP → RESPREL b = 0.647, p < 0.01). Not surprisingly, there is also a very high association with assurance, empathy and responsiveness & reliability.
With respect to the effect of service quality, our study provides support for the association between the standard SERVQUAL dimensions and customers' retention decisions.
The structural coefficients between assurance, empathy, responsiveness & reliability and retention are 0.274, 0.230 and 0.379, p < 0.01 respectively. We expected the weakest relationship between accountant competences and retention and we find no evidence at all of a direct relationship between the two variables. All in all, it seems that the association between professional qualification and retention by the customer is not direct: it is mediated by perceived competences of the accountant and by their impact on other quality dimensions.  The overall model explains 66.4% of the variance in customers' retention decisions by the proposed dimensions of quality. Professional qualification only modestly explains the variation in perceived competences (3.7%). Variance explained in relations between competences and other dimensions exceeds 40% (see Table 3). These findings sug-gest that the professional qualification of accountants only weakly explains customer perceptions of accountant competences, but that perceived competences create strong confidence in the accountant that has a beneficial effect on all other dimensions of service quality -assurance, responsiveness & reliability and empathy. Only these latter dimensions, in turn, affect the customers' retention decisions. Therefore, competences are not of direct importance for a customer in the decision to maintain a certain professional accountant, but are rather an antecedent of other quality dimensions.

Conclusions and discussion
The purpose of the paper is to raise the question as to whether financial reporting in small and micro firms performed by professional accountants should be (to some extent) regulated in the form of required qualification. We address this question by analysing whether professional qualification positively affects the perceived quality of accounting services provided to small and micro firms, which is expected, in turn, to affect the customer's retention decision. Although the trend in accounting regulation seems to be going in the reverse direction we believe that the issue of regulating the professional preparers of accounts is somewhat different to the lessening of administrative burden of firms. The introduction of minimum required professional qualifications would oblige professional accountancy services providers to comply with certain education, training and ethical standards.
We are not aware of any prior studies investigating the relation between professional qualification, accounting service quality and customers' retention decisions in small and micro firms. One of the contributions of this study is the elaboration of the service quality measurement, appropriate for accounting services. We complemented the existing measurement model with indicators that are industry specific and put an increased emphasis on accountant competences. We identified four dimensions of service quality: competences, assurance, responsiveness & reliability and empathy. The latter three dimensions coincide with the dimensions of the general service quality model , in which competences are only a subcomponent of assurance. In our model, on the other hand, competences are a self-standing dimension, which may be particular to the profession under analysis.
We find that professional qualification has a positive impact on customer perceptions of accountants' competences. The explanatory power is, however, low. We further find that competences are positively associated with other aspects of service quality, namely assurance, responsiveness & reliability and empathy.
Ultimately, our study finds that service quality dimensions impacting on customers' retention decisions are the same as those found in non-accounting studies. In other words, the market will choose those accountants that are perceived as more empathetic, reliable and responsive to individual needs of the customers, those who assure privacy and invoke trust. The competences are of secondary importance. This finding is significant and is in line with the finding of Armstrong and Davison (1995), Aga and Safakli (2007) who also find empathy to be one of the most important dimensions of quality as perceived by customers. The finding suggests that the market will not demand formal qualification of accountants, because customers prefer an accountant that is loyal. These characteristics of a personal relationship between a client and an accountant may raise serious concerns regarding an accountant's independence.
The study has some important regulation-related implications. The finding that professional qualification significantly affects professional competences of accountants shows that the market is aware of the benefits of professional qualification. However, the finding that professional competences are not highly valued by the customers in their retention decision suggests that, from the viewpoint of the market, regulation of the accounting profession is not desired. What matters most is the allegiance to the customer. Such an attitude leaves the users of accounting information rather vulnerable and the assurance of public interest remains unsolved. Regulation of the accounting profession is therefore warranted from the viewpoint of the public interest in all those countries in which the market for accounting services is not sufficiently self-regulated in the sense that a professional accountant's affiliation to professional organisations and acquisition of professional competences would be required in order to be accepted by the customers and by the users of accounting information.