The motivational role of consultative participation in a multi-period target setting: An experimental study

ABSTRACT

Consultative participation is a process where employees provide inputs but managers retain control over the final targets. According to the latest research in accounting, it has no influence on employee performance, and can even decrease employee motivation. This paper distinguishes between the different types of consultative participation, depending on whether managers take into account employees’ proposals (real consultative) or ignore them (pseudo-participative). We propose that it is necessary to distinguish among the types of motivation in order to understand how participation impacts on employee motivation. We present a mediating model where autonomous employee motivation is the mediating variable. An experiment was conducted to test this model. Our findings show that real consultative participation increases autonomous motivation, while pseudo-participation decreases it. Autonomous motivation also mediates between participation and employee performance. No effects were found on controlled motivation.

RESUMEN

La participación consultiva es un proceso en el que los empleados proporcionan opiniones y sugerencias sobre los objetivos a fijar, sin embrago, son los directivos o superiores los que finalmente deciden sobre estos objetivos. De acuerdo con las últimas investigaciones en Contabilidad, este proceso no produce ningún efecto sobre el rendimiento de los empleados, e incluso puede llegar a disminuir su motivación. Este trabajo diferencia entre tipos de participación consultiva, atendiendo a si los superiores tienen en cuenta las sugerencias de los empleados (consultiva real), o éstas son ignoradas (pseudo-participación) Proponemos que es necesario diferenciar entre tipos de motivación, para comprender cómo la participación consultiva puede influir en la motivación de los empleados. Se presenta un modelo de mediación, donde la motivación autónoma es la variable mediadora, y se contrasta a través de un experimento. Los resultados muestran que la participación consultiva real aumenta la motivación autónoma, mientras que la pseudo-participación la disminuye. Los resultados también revelan que la motivación autónoma media entre la participación consultiva y el rendimiento de los empleados. Finalmente no se han encontrado efectos de la participación consultiva sobre la motivación controlada.

KEYWORDS:

• Consultative participation
• target setting
• autonomous motivation
• performance
• experiment

Acknowledgements

We acknowledge the many comments from the participants of the 20th Workshop on Accounting and Management Control ‘Memorial Raymond Konopka’ (IE University, Segovia, 2015), and participants of the 21th Workshop on Accounting and Management Control ‘Memorial Raymond Konopka’ (Universidad de Valencia, Valencia, 2016). We also acknowledge the comments from the faculties of the III Research Forum in Management Accounting and Control: Prof. M. Abernethy and L. Van Lent.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1. Consultative participation is also called voice (Libby, 1999).

2. Following Libby (1999) and Lindquist (1995), pseudo-participative is a type of consultative process where managers ignore the input of employees.

3. Accounting literature differentiates three roles of participation: motivational role, cognitive role and value attainment role (Chong et al., 2006).

4. The work by Brownell and McInnes (1986) analysed the relationship between budgetary participation, motivation and performance, but it did not differentiate between consultative and substantive participation. Moreover, results showed that the relationship between participation and motivation is insignificant.

5. The work by Wong-On-Wing et al. (2010) introduced types of motivation as moderator variables, between budgetary participation and performance, but does not differentiate between consultative and substantive.

6. The study by Chong et al. (2005a) analysed two dimensions of budgetary participation: involvement (employees and managers exchange information so as to facilitate decision-making) and influence (the degree to which a person is able to affect his own actions, targets, goals, budgets).

7. Libby (1999) did not find the consultative participation and/or explanation to exert any influence on employee performance, but the interaction between them did.

8. Attainable/Unattainable targets are a control variable explained in the Manipulation and Measurement section.

9. The superior was not an individual who participated in the experiment, but the computer played this role.

10. We follow Lindquist (1995) to manipulate high and low consultative participation. Based on a pilot experiment, the close-target employee proposal was set by increasing 20% over original participant proposal each (low consultative). We follow Libby (1999) and Byrne and Damon (2008) to manipulate pseudo-participation. The distant was determined after computing the average performance achieved in the pilot experiment. The target was the same for all participants, but was changed for every trial.

11. The items no. 4, 5 and 6 from Milani’s (1975) instrument were used to measure the influence on target setting.

12. This analysis allowed us to check if the manipulation of the independent variable – consultative participation – was done in a proper way.

13. To further examine the results, t-tests were used to compare means across high and low consultative conditions, for autonomous motivation. The results show no difference (t = 1.324; p > .10).

14. At this point, we studied the effect on autonomous motivation solely, given that differences have been found only in this type of motivation.

15. When we introduced total unattainable variable as a factor (subsample attainable vs. subsample unattainable), the results of the GLM were similar. There is no statistically significant effect of total unattainability on autonomous motivation (F = .227; p > .10), but the effect of consultative participation (real vs. pseudo) is significant in the model (F = 5.997; p < .05).

16. According to Hair, Sarstedt, Ringle, and Mena (2011), bootstrapping (5000 resamples) is used to generate standard errors and t-statistics, which allow the evaluating of the statistical significance of the path coefficients.

Additional information

Funding

This work was supported by the Consejería de Economía, Innovación, Ciencia y Empleo, Junta de Andalucía [SEJ 2395];