Management Accountants’ Occupational Prestige Within the Company: A Social Identity Theory Perspective

Abstract

Professional associations and researchers in management accounting have attempted to increase the prestige of management accountants. Although studies have suggested that occupational prestige is important, the prestige of management accountants within the employing organisation and its consequences are still not fully understood. Building on social identity theory, we investigate the effect of occupational prestige, as perceived by management accountants, on organisational–professional conflict. We suggest that prestige can mitigate conflict because management accountants with high prestige will identify more strongly with their organisation, as they see it as a source of self-esteem and might believe that they are taken more seriously by managers. At the same time, we hypothesise that prestige exerts an indirect conflict-increasing effect via professional identification. This is based on the idea that prestige may cause proud management accountants to start to identify more strongly with their profession and become hard-liners, unwilling to compromise the values associated with their profession in the interest of the firm. Results from a series of three surveys support the indirect conflict-increasing effect. Moreover, the results suggest a direct conflict-reducing effect of perceived prestige. The implications for research and practice are discussed.

Acknowledgements

We thank Laurence van Lent and an anonymous reviewer for their valuable input on earlier versions of this paper. Moreover, we like to thank Lukas Goretzki, Ludwig Voußem, Raphaela Erhart and Fabian Kleinschmit for their helpful comments and inspiring discussions.

Notes

¹Accounting research has applied social identity theory, for example, in the contexts of control in teamwork settings in terms of incentive systems (Towry, 2003), the design of incentive and information systems (Heinle et al., 2012) and the effects of relative performance evaluation (Krishnan et al., 2010).

²A competitive mediation occurs if independent variable X has multiple effects (e.g. one direct and one indirect effect) on the dependent variable Y and these effects have opposite signs. For example, the antecedent may have a positive indirect effect on Y via mediator M and a negative direct effect (Zhao et al., 2010).

³Henceforth, we will use the terms occupation and profession interchangeably.

⁴This might be particularly true for management accounting, as ‘management accounting work is more strongly characterized by the specifics of the organization than, for example, audit[ing work]’ (Ahrens and Chapman, 2000, p. 478).

⁵The ICV is a large professional body of controllers in German-speaking countries.

⁶If the outliers are kept in the sample, the signs of all significant effects from the main model remain the same as in our main test with the 188 observations. Likewise, all hypothesis tests that are significant in the main test remain significant (p < .05; two-tailed).

⁷The psychometric literature (e.g. Raykov, 2005; Schafer and Graham, 2002) suggests that full information maximum-likelihood estimation is preferred over list-wise deletion. The covariance coverage of our data ranged from 73% to 100% and thus exceeds the required minimum of 10% (Muthén and Muthén, 1998–2010).

⁸In an untabulated alternative specification, we also used indicator variables for hierarchical level. The signs of all effects were identical and all significance levels remained largely similar.

⁹We also calculated a number of models (not tabulated) with additional control variables, such as need for belongingness and need for self-definition. Since these variables are beyond the focus of this paper and did not influence the results, we dropped them from the main model to improve the ratio between the number of observations and the number of estimated parameters (Tabachnick and Fidell, 2007, p. 123).

¹⁰All p-values are based on two-tailed tests.

¹¹These variables can be considered relevant because previous research has shown that they have an impact on many other variables that are important for organisational performance (e.g. organisational identification has been shown to influence absenteeism, effort and the like; see Riketta, 2005).

¹²For more details, see the online supplement at the journal's Taylor and Francis website.