Executive Incentives, Works Councils and Firm Performance

Jirjahn, Uwe Article Executive Incentives, Works Councils and Firm Performance Schmollers Jahrbuch – Zeitschrift für Wirtschaftsund Sozialwissenschaften. Journal of Applied Social Science Studies Provided in Cooperation with: Duncker & Humblot, Berlin Suggested Citation: Jirjahn, Uwe (2003) : Executive Incentives, Works Councils and Firm Performance, Schmollers Jahrbuch – Zeitschrift für Wirtschaftsund Sozialwissenschaften. Journal of Applied Social Science Studies, ISSN 1865-5742, Duncker & Humblot, Berlin, Vol. 123, Iss. 3, pp. 397-421, https://doi.org/10.3790/schm.123.3.397 This Version is available at: https://hdl.handle.net/10419/292063 Standard-Nutzungsbedingungen: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may be saved and copied for your personal and scholarly purposes. You are not to copy documents for public or commercial purposes, to exhibit the documents publicly, to make them publicly available on the internet, or to distribute or otherwise use the documents in public. If the documents have been made available under an Open Content Licence (especially Creative Commons Licences), you may exercise further usage rights as specified in the indicated licence. https://creativecommons.org/licenses/by/4.0/ Schmollers Jahrbuch 123 (2003), 397-421 Duncker & Humblot, Berlin Executive Incentives, Works Councils and Firm Performance* By Uwe Jirjahn Abstract This paper investigates the interaction effect of works councils and managerial profit sharing on establishment performance. In the theoretical part, it is argued that this inter action effect depends on two relationships, namely the relationship between works councils and self-enforcing contracts and the relationship between agency and trust within establishments. The empirical analysis with data from German manufacturing establishments shows a negative interaction effect of managerial profit sharing and works councils on productivity. Works councils seem to be of particular importance for the economic success of establishments when no managerial profit sharing is in place. The theoretical interpretation of this finding is, however, complex. Moreover, it is im plicitly shown that the negative interaction effect is particularly prevalent in larger es tablishments. Separate estimates are performed with a subsample of establishments em ploying 21 to 100 workers. While the estimates show positive productivity effects of managerial profit sharing and works councils also for this subsample of smaller estab lishments, no statistically significant interaction effect is found. Zusammenfassung Der vorliegende Beitrag untersucht die Wechselwirkungen, die zwischen betrieb licher Mitbestimmung und einer Erfolgsbeteiligung für die Geschäftsleitung eines Be triebs bestehen. Im theoretischen Teil des Beitrags wird argumentiert, dass der Interak tionseffekt beider Variablen von zwei Beziehungen abhängt. Zum einen ist es von Be deutung, ob betriebliche Mitbestimmung und Reputationserwägungen des Manage ments substitutiv oder komplementär zueinander sind. Zum anderen spielt der Einfluss des Principal-Agent-Problems zwischen Eigentümern und Managern auf kooperative industrielle Beziehungen eine Rolle. Die empirische Analyse mit Daten des Hannovera ner Firmenpanels zeigt, dass Betriebsräte und eine Erfolgsbeteiligung für die Geschäfts leitung im Hinblick auf die Produktivität negativ miteinander interagieren. Betriebsräte scheinen somit dann von besonderer Bedeutung für den betrieblichen Erfolg zu sein, * I would like to thank two anonymous referees and the participants of the Workshop "Mitbestimmung und Mitarbeiterbeteiligung: Komplementär oder substitutiv?" in Wit ten for helpful comments. All remaining errors are my own. Schmollers Jahrbuch 123 (2003) 3 OPEN ACCESS | Licensed under CC BY 4.0 | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/schm.123.3.397 | Generated on 2023-04-04 12:32:19 398 Uwe Jirjahn wenn es keine Erfolgsbeteiligung für die Geschäftsleitung gibt. Die theoretische Inter pretation dieses Ergebnisses ist allerdings komplex. Die empirischen Resultate deuten zudem implizit darauf hin, dass der negative Interaktionseffekt insbesondere bei größe ren Betrieben vorzufinden ist. Werden die Schätzungen nur für Betriebe mit 21 bis 100 Beschäftigten durchgeführt, dann zeigen sich zwar weiterhin positive Produktivitäts effekte betrieblicher Mitbestimmung und einer Erfolgsbeteiligung des Managements, während sich hingegen kein statistisch gesicherter Interaktionseffekt beider Variablen finden lässt. lEL Classification: 124; 153; 133; D23; G32; L21 1. lntroduction FitzRoy and Kraft (1987) argue in their influential paper on works councils that the econornic effects of codetermination depend on the managerial envir onment in the firm. In spite of the merits of this argument, the interaction effect of executive compensation and works councils on firm performance has not been examined up to now. Drawing on recent principal-agent theory, this paper provides both a non-technical theoretical framework for analysing the interaction of works councils with manager incentives and an empirical test of the hypotheses. In the theoretical part of this paper, it is argued that the interaction of works councils and profit sharing for executive managers depends on two relation ships. First, works councils may help to overcome employer opportunism. However, codeterrnination is not the only solution to the employer's comrnit ment problem. Reputation concems may also induce an employer to behave honestly. This raises the question whether codetermination and self-enforcing contracts are substitutes or complements. Second, the principal-agent problem between the owners and the managers of the firm may exert two opposite ef fects on trustful employer-employee relations. On the one band, agency may have a commitment value, since managers are less likey to renege on implict contracts with the work force on behalf of short-term profits. On the other band, rent-seeking activities by managers may hinder trust and loyality within establishments. Combining the relationship between works councils and self enforcing contracts and the relationship between agency and trustful employ er-employee relations, four situations with different implications for the inter action of managerial profit sharing and works councils are identified. The empirical analysis uses pooled data from manufacturing establishments in the German federal state of Lower Saxony. While establishments with man agerial profit sharing are more likely to have a works council, the productivity estimates show a negative interaction of works councils and profit sharing for executive managers. This negative interaction effect seems to be particularly prevalent in larger establishments. Separate estimates are performed with a Schmollers Jahrbuch 123 (2003) 3 OPEN ACCESS | Licensed under CC BY 4.0 | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/schm.123.3.397 | Generated on 2023-04-04 12:32:19 Executive Incentives, Works Councils and Firm Performance 399 subsample of establishments employing 21 to 100 workers. While the esti mates show positive productivity effects of managerial profit sharing and works councils also for this subsample of smaller establishments, no statisti cally significant interaction effect can be found. The theoretical framework of this paper suggests two alternative explana tions for the negative interaction of managerial profit sharing and works coun cils. First, if managerial profit sharing decreases the commitment value of agency and works councils cannot foster trust and loyality without the man agers' cooperation, a negative interaction effect on establishment performance may result. Second, if managerial profit sharing induces managers to reduce their rent-seeking activities and works councils are not so important for build ing trust in establishments with less severe manager opportunism, a negative interaction effect is also likely to be found. The plan of the paper is as follows. Section 2 presents the theoretical frame work. In Section 3, the data set employed is reviewed and the results are pre sented. Section 4 concludes. 2. Theoretical Framework In this section, we argue that the interaction effect of works councils and managerial profit sharing depends on two relationships, namely the relation ship between codetermination and self-enforcing contracts and the relation ship between agency and trust within firms. In the first step, it is discussed whether works councils and self-enforcing contracts are substitutes or comple ments in building trustful employer-employee relation. In the next step, we analyse the implications of the agency problem between owners and executive managers for the employer-employee relations. The impact of agency on trust ful industrial relations is not clear-cut. On the one hand, agency may increase the range of feasible self-enforcing contracts, because managers are less likely to neglect an implicit promise than the profit maximising owners of the firm. On the other hand, rent seeking activities by managers may hinder trustful in dustrial relations. These two effects of agency have different implications for the role of managerial profit sharing and its interaction with works councils. Finally, we summarise the hypotheses for the empirical analysis. 2.1 Works Councils and Self-Enforcing Contracts The rights of the works councils are laid down in the Works Constitution Act (see Müller-Jentsch 1995 for a survey). Councils shall be elected by the whole work force of establishments with five or more permanent employees. However, their creation depends on the employees' initiative. Hence, works Schmollers Jahrbuch 123 (2003) 3 OPEN ACCESS | Licensed under CC BY 4.0 | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/schm.123.3.397 | Generated on 2023-04-04 12:32:19 400 Uwe Jirjahn councils are not present in all eligible establishments. The powers of the works councils can be quite broad. They have füll codetermination rights on a set of issues, including the introduction of new payment methods, the fixing of job and bonus rates, the allocation of working hours, the introduction and use of technical devices designed to monitor employee performance, and up and down grading. In these areas management cannot take action without the agreement of the works council. The councils have less strong consultation rights in matters such as changes in equipment and working methods that af fect job requirements. Their participation rights in financial and economic matters cover only information provision. The Works Constitution Act ties works councils to the general obligation to cooperate with management 'in a spirit of mutual trust ... for the good of the employees and of the establish ment.' There are two major explanations as to why codetermination may improve joint establishment surplus. First, councils may exert a collective voice role. Many working conditions are workplace public goods (Freeman 1976; Free man and Medoff 1979). Works councils communicate worker preferences to management helping to optimise the provision of those public goods and to reduce personnel turnover (Frick and Sadowski 1995; Frick 1996). However, the strong codetermination rights of the works councils seem to go far beyond the rights of a pure collective voice institution (FitzRoy and Kraft 1987). This observation brings us to the second explanation. Works councils with codeter mination rights may prevent employers from engaging in opportunistic beha viour (Smith 1991; Freeman and Lazear 1995). Employees will withhold ef fort and cooperation, when an employer cannot credible commit to take into account their interests. For example, employees fearing job loss due to techno logical change do not reveal information about potential productivity enhan cing innovations. Providing works councils with codetermination rights is one mechanism for protecting the interests of the work force. Hence, works coun cils may foster the workers' cooperation with the introduction of efficiency enhancing work practices, including pay for performance and profit sharing for employees (Heywood, Hübler and Jirjahn 1998; Heywood and Jirjahn 2002), employer provided further training (Gerlach and Jirjahn 2001), flexible working time arrangements (Dilger 2002; Hübler and Jirjahn 2003), human resource management systems (Jirjahn 2002), flexible work systems (Hübler and Jirjahn 2002) and investments in improving local environmental quality (Askildsen, Jirjahn and Smith 2002).1 1 The participatory rights of the works council may differ between the various work practices. However, the behaviour of works councils is not entirely predetermined by legislation. Works councils appear to play a very active role in decision-making that may go far beyond the rights laid down in the Works Constitution Act. First, managers often encourage works council participation in order to ensure the employees' coopera tion and effort. Second, works councils may use their codetermination rights in social Schmollers Jahrbuch 123 (2003) 3 OPEN ACCESS | Licensed under CC BY 4.0 | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/schm.123.3.397 | Generated on 2023-04-04 12:32:19 Executive Incentives, Works Councils and Firm Performance 401 Codetermination is not the only solution to the employer's commitment pro blem. Reputation concerns may induce an employer to behave honestly (Bull 1987). Therefore, self-enforcing contracts stand as an alternative in order to engender the trust, which is important for the workers' cooperation. This argu ment suggests that codetermination and self-enforcing contracts may be sub stitutive. Work councils should be of particular importance for building trust ful industrial relations, when self-enforcing contracts are not feasible. The re putation mechanism is based on a repeated game between the employer and the employees. Such a repeated game requires long-term employment rela tionships and low discount rates. The reputation mechanism fails in case of short-term employment relationships and in case that the employer overly dis counts the future loss of trust and cooperation. In these situations a council with codetermination rights may protect the workers' interests. However, another way of looking at the relationship between codetermina tion and reputation effects suggests that both mechanisms may be comple ments. First, a works council facilitates communication and coordination among workers. To the extend coordinated actions of the work force result in a more severe punishment of employer opportunism, the employer's incentive to renege on an implicit agreement is reduced (Hogan 2001). Second, in an uncertain and complex world, it is necessary that workers have accurate infor mation about the firm's economic situation to verify whether the employer behaves honestly or not (Kreps 1990). A works council with comprehensive information rights may enforce implicit agreements through a better flow of information between the employer and the work force. Third, in case of multi ple equilibria, a council may help the employer and the employees to shift to an equilibrium with cooperative industrial relations. A firm, that is interested in building trustful industrial relations, needs to provide a credible signal in order to convince the employees of its intentions. Therefore, the employer may encourage the workers to elect a works council. Moreover, reputation concerns may increase the employer's willingness to cooperate with the coun cil. Tue findings by Askildsen, Jirjahn and Smith (2002) and Dilger (2002) confirm that the effects of works councils depend on a positive management attitude toward worker involvement in decision making. Up to this point, our discussion of employer-employee relations has been founded on the implicit assumption of a profit maximising employer. Now we turn to the principal-agent problem between owners and managers. Principal agent theory assumes that executive managers have interests different from those of profit maximising owners. When it is difficult to monitor the man agers' actions, managers have greater scope to pursue their own goals rather than the owners' interests. In what follows, we discuss the implications of the or personnel matters to obtain employer concessions in other fields, in which they have no legal powers. Schmollers Jahrbuch 123 (2003) 3 OPEN ACCESS | Licensed under CC BY 4.0 | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/schm.123.3.397 | Generated on 2023-04-04 12:32:19 402 Uwe Jirjahn agency problem for the employer-employee relations. Moreover, the princi pal-agent problem brings us to the role of profit sharing. Profit sharing for executive managers may help reduce agency problems by aligning the inter ests of managers and owners. However, the impact of managerial profit shar ing on trust and cooperation in firms is not clear-cut. 2.2 The Commitment Value of Agency In standard principal-agent models, managers lack incentives to perform their tasks well, when they have a rather small stake in the firm's profit. The managers' incentives to control costs and to improve productivity are low. However, the agency problem has not only an impact on the managers' effort but also on the feasibility of self-enforcing contracts. Tao and Zhu (2000) ar gue that the managers' incentive to renege on an implicit contract on behalf of short-term profits is lower than that of the owners. While the managers obtain only a small proportion of the benefit from opportunism on behalf of the own ers, they may entail a disproportionate personal cost from informal sanctions. Thus, agency has a commitment value for the owners of the firm in making self-enforcing contracts feasible. The commitment value of agency does not only apply to the relationship between firms analysed by Tao and Zhu. lt is also relevant for employer-em ployee relations within firms. To the extend the agency problem has a commit ment value, it may imply an increased range of self-enforcing contracts be tween the firm and the employees, particularly when no profit sharing plan for executive managers is in place. Profit sharing for executive managers reduces agency problems by aligning the interests of managers and owners. lt provides an incentive for managers to exert more effort. However, it also increases the managers' incentive to break implicit contracts with the employees on behalf of profit maximising owners.2 Thus, trustful employer-employee relations are less likely feasible, when managerial profit sharing is in place. The interaction effect of managerial profit sharing and works councils on firm performance depends crucially on the relationship between codetermina tion and self-enforcing contracts. If work councils and self-enforcing contracts are substitutes in building trustful and cooperative industrial relations, the im pact of councils on firm performance will be stronger in firms with less severe agency problems. In these firms, it is more likely that an implicit promise is 2 In light of this argument, profit sharing induces managers to renege on implicit contracts only when opportunistic behaviour is in the owners' interest. Of course, profit sharing provides no incentive to break implicit contracts when opportunism is not in the owners' interest. Without profit sharing the interests of managers and owners are only loosely aligned. Therefore, managers who receive no profit sharing have little incen tives to renege on implicit contracts regardless of the owners' interests. Schmollers Jahrbuch 123 (2003) 3 OPEN ACCESS | Licensed under CC BY 4.0 | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/schm.123.3.397 | Generated on 2023-04-04 12:32:19 Executive Incentives, Works Councils and Firm Performance 403 neglected for the short-term benefit of profit maximising owners. Since man agerial profit sharing reduces agency problems between owners and executive managers, a positive interaction effect of works councils and managerial profit sharing can be expected. If codetermination is complementary to self-enforcing contracts, works councils will be more effective in firms, where agency increases the range of feasible self-enforcing contracts. This is more likely, when no profit sharing for executive managers is in place. In this case, it can be expected that works councils and managerial profit sharing have a negative interaction effect on firm performance. 2.3 Rent-Seeking Activities by Executive Managers The assumption of rent-seeking activities contrasts the notion that agency has a commitment value in enforcing implicit contracts between the employer and the employees. Managers may use their discretionary power to their own benefit. Career concerns, status and power are important goals of managers. Smith (1991) provides a detailed description of various types of manager op portunism destroying trust and cooperation within firms. One source for man ager opportunism results from their discretion over task assignment (Prender gast 1995). This discretion creates opportunities for the managers to hoard re sponsibility and authority. Another source of discretionary power are subjec tive performance evaluations by managers, which open the door to favoritism based on the managers' personal preferences toward subordinates (Prendergast and Topel 1996). Subjective performance evaluations allow managers to re ward and to promote their preferred subordinates. Moreover, they may use the discretion in performance appraisal to extract private services from the em ployees (Laffont 1990).3 This distorts the incentives for the employees to exert effort and reduces their loyality to the firm. Rent-seeking behaviour by man agers decreases the range of feasible self-enforcing contracts between the firm and the employees. Even when the owners are interested in their reputation, the agency problem may hinder trustful and cooperative industrial relations within the firm. Managerial profit sharing does not only provide an incentive for managers to exert more effort. Moreover, it generates an incentive to reduce rent-seeking activities by increasing the cost of these activities to the managers involved. In case of profit sharing, managers participate in the efficiency loss resulting 3 The class of possible transfers can be quite large. Managers may take credit for the employees' innovative ideas. They even may sexually harass subordinates. Laband and Lentz (1998) show that sexual harassment by superiors has a strong and negative im pact on the job satisfication of female lawyers in the US. Moreover, it increases their quit intentions. Schmollers Jahrbuch 123 (2003) 3 OPEN ACCESS | Licensed under CC BY 4.0 | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/schm.123.3.397 | Generated on 2023-04-04 12:32:19 404 Uwe Jirjahn from tbeir rent-seeking activities. Tbey may be ratber inclined to foster tbe cooperative industrial relations, whicb improve firm performance. Tbe interaction effect of profit sbaring for executive managers and works councils depends, again, on the relationsbip between codetermination and self-enforcing contracts. If works councils and self-enforcing contracts are substitutes in building trustful employer-employee relations, a negative inter action effect of works councils and profit sbaring for executive managers on firm performance can be expected. Since rent-seeking activities by executive managers are more severe in firms witbout managerial profit sbaring, works councils may play an important role in protecting workers against opportunis tic managers in those firms. Works councils may bave a less important role in increasing firm performance, wben profit sbaring provides an incentive for managers to establisb trustful communication witb the employees. However, profit sbaring for executive managers will interact positively witb works council presence, wben codetermination and self-enforcing contracts complement one anotber. Works councils will be more effective in firms, wbere managers bave a positive attitude toward trust and cooperation. A coop erative managerial environment may belp to avoid conflictual bargaining and excessive time sperrt in meetings and discussions. Managers, interested in long-term cooperation with the employees, may ask works councils to partici pate in a wide range of decisions in order to foster bigb-trust management employee relationsbips. Tbe managers' willingness to build cooperative indus trial relations may be induced by managerial profit sbaring. 2.4 Summary of Hypotheses Tbe interaction effect of managerial profit sbaring and works councils de pends on two relationships, namely the relationship between codetermination and self-enforcing contracts and the relationsbip between agency and trust within establisbments. Tbere are two different aspects of agency. On the one band, agency may bave a commitment value for tbe owners of tbe firm in mak ing self-enforcing contracts feasible. On tbe other band, managers bave discre tionary power, wbicb tbey may use to tbeir own benefit. Wbile botb aspects are likely to emerge simultaneously, tbe effect of profit sbaring on trustful employ er-employee relations depends on tbe dominating cbaracteristic of agency. If tbe commitment value of agency dominates, profit sbaring for executive man agers will reduce trustful industrial relations within tbe establisbment. If rent seeking bebaviour is the dominating cbaracteristic of agency, managerial profit sbaring will foster trustful employer-employee relations. In both cases, tbe in teraction of profit sbaring witb works councils depends on a substitutive or complementary relationsbip between codetermination and self-enforcing con tracts. Table 1 summarises tbe bypotbeses on tbe interaction effect. Schmollers Jahrbuch 123 (2003) 3 OPEN ACCESS | Licensed under CC BY 4.0 | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/schm.123.3.397 | Generated on 2023-04-04 12:32:19 Executive Incentives, Works Councils and Firm Performance 411 Most noteworthy, several innovative work practices are positively linked to the presence of works councils. The results clearly indicate that works coun cils may play a cruical role in establishing or maintaining these work prac tices. Employer provided further training is a statistically significant covariate of works council presence in both the estimates with the füll sample and the estimates with the subsample. This finding confirms the study by Gerlach and Jirjahn (2001). lt is consistent with the hypothesis that employee representa tion may alleviate market failures in human capital investments such as hold ups, poaching extemalities and coordination problems (see Askildsen and Ire land 1993; Smith 1991, 1994; Soskice 1994; Askildsen, Jirjahn and Smith 2002a for a theoretical discussion). Further, the use of piece rates is positively linked to works council presence in the estimates with the overall sample and also in the estimates with the subsample of establishments employing 21 -100 workers. This result is consistent with the hypothesis that works councils may help to overcome the ratchet effect.9 lt confirms the studies by Heywood, Hübler and Jirjahn (1998) and Heywood and Jirjahn (2002). These studies show that the link is particularly due to the positive association between per formance pay and works councils in establishments covered by collective bar gaining agreements. 10 Moreover, flexible working hours (Arbeitszeitkonten) are positively linked to works council presence in both the estimates with the füll sample and the estimates with the subsample. There is also a significant positive association between shift work and the incidence of works councils in the regression with the füll sample. These findings indicate that works coun cils may play a mediating role in the joint determination of working time sche dules by the employer and the employees. 11 9 Workers will withhold effort when they fear an increase in performance standards after a period of good performance. They are less likely to fear the ratchet effect and to withhold effort, when a works council prevents an inappropriate increase in perfor mance standards. 10 Using cross section data from the NIFA Panel, Dilger (2003) finds no robust asso ciation between works councils and performance pay. lt is argued that the works council effect may simply reflect a size effect. However, considering the results with the Hann over Panel, this interpretation lacks plausibility. Even when controlling for establish ment size, there is a positive link between performance pay and works councils. More over, the association between the use of piece rates and works council presence can also be found in the subsample of smaller establishments with 21 -100 employees. There may be other reasons for the results found by Dilger: First, separate estimates for cov ered and uncovered establishments are not performed. Therefore, it is not clear if the results are due to the uncovered establishments in the NIFA Panel. Second, compared to other international studies on the determinants of performance pay, only a thin set of control variables is used. Therefore, Dilger's study is likely to suffer from an omitted variable bias. Third, the NIFA Panel is restricted to the narrow sector of mechanical engineering. Further, the link between works councils and establishment size should be interpreted with caution. A related study by Hübler and Meyer (2001) indicates that establishment size may mask the "true effects" of works councils. This study clearly contradicts the interpretation that a works council effect is simply a size effect. Schmollers Jahrbuch 123 (2003) 3 OPEN ACCESS | Licensed under CC BY 4.0 | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/schm.123.3.397 | Generated on 2023-04-04 12:32:19 412 Uwe Jirjahn The negative link between the incidence of production teams and works council presence is not easy to interpret. Addison, Schnabel and Wagner (1997) argue that works councils and teams might be substitutive forms of participation in the workers' eyes. However, the relationship between these two forms of employee participation seems to be more complex than this inter pretation suggests. There may be direct and indirect links between teams and works councils, operating in opposite directions. On the one band, it might be more difficult for a works council to represent the interests of the entire work force, when individual work groups directly communicate with the manage ment. On the other band, works council presence is positively associated with training and performance pay. Several arguments suggest that these human resource management practices are complements to teams. Establishments do not choose a single work practice. They choose an entire system of more or less interrelated and internally consistent practices. Jirjahn (2002) uses more precise informations on employee participation, which are only available in the third wave of the Hannover Panel. lt is shown that the dual system of worker representation in Germany is not only associated with a traditonal sys tem of mass production but also with a flexible production system charac terised by group incentives, training, teams and regular meetings of managers and workers. This finding accords with a study by Hübler and Jirjahn (2002). Using the IAB Establishment Panel, Hübler and Jirjahn do not investigate the incidence but the introduction of teams. lt is shown that establishments with works councils are more likely to introduce teams. Moreover, taking into ac count the endogenity of teams, the study shows a positive interaction effect of works councils and the introduction of teams on firm performance. 12 Returning to the main focus of this paper, there is a positive and statistically significant link between profit sharing for executive managers and works council presence for the füll sample. No statistically significant association between managerial profit sharing and the incidence of works councils can be found for subsample of smaller establishments. These results with the pooled data for 1994 and 1996 confirm studies using the first wave of the Hannover Panel (Addison, Schnabel and Wagner 1997; Jirjahn 1998). The findings im plicitly mean that there is a positive association between managerial profit sharing and works councils in larger establishments. However, this positive association does not imply a particular interaction effect of managerial profit sharing and works councils on establishment performance. As the theoretical arguments in Section 2 show, the positive link between the two variables is 11 These findings accord with other international studies, which show that union pre sence is positively linked to the incidence of shift work (Duncan and Stafford 1980) and to cornpensating wage differentials for shift work (Kostiuk 1990). 12 The last finding accords with studies for the US. Cooke (1994) and also Black and Lynch (2000) obtain positive interaction effects of unions and alternative forms of ern ployee participation on firm performance. Schmollers Jahrbuch 123 (2003) 3 OPEN ACCESS | Licensed under CC BY 4.0 | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/schm.123.3.397 | Generated on 2023-04-04 12:32:19 Executive Incentives, Works Councils and Firm Performance 413 consistent with both a positive or a negative interaction effect. In case of a positive interaction effect, managers may encourage employees to elect a works council, when they are motivated by profit sharing. In case of a nega tive interaction effect, wokers may be more willing to elect a works council for their protection against too high managerial pressure or for a minimum protection against employer opportunism. This may be the case, even when profit sharing induces more managerial resistance to codetermination and the hostile industrial relations do not contribute to establishment performance. In order to examine the relationship between profit sharing for executive man agers and works councils in more detail, we have to investigate the interaction effect of the two variables on productivity in the next step. 3.3 Managerial Profit Sharing, Works Councils and Labour Productivity Table 3 presents the pooled productivity estimates with the füll establish ment sample. The coefficient estimates for the control variables are well deter mined and accord with theoretical standard expectations. Tue relationships be tween these variables and productivity are robust to the specification changes shown in Table 3. First, the structure of the work force plays an important role for productivity. The proportions of part-time workers, women, apprentices and blue-collar workers have negative productivity effects, while the share of university graduates exerts a positive impact on productivity. Second, there are also several relevant establishment characteristics. Rather interestingly, shift work has a strong positive impact on productivity.13 Moreover, establish ment size and a modern production technology increase productivity. Single establishments, which have no subsidiaries and are not themselves subsidi aries, are characterised by lower productivity. Further, the coefficient estimate for the time dummy for 1996 is positive and statistically significant. The coef ficient on the coverage by a collective agreement is negative but statistically insignificant. Finally, note that 13 of 19 broader defined industry dummies are contained in all estimates. The estimates shown in column (1) and (2) do not take into account the interaction of managerial profit sharing and works councils. Starting with the results in column (1), it can be seen that establishments with a works council have a significantly higher productivity. However, it is not controlled for shift work and managerial profit sharing. The regression in column (2) includes these variables. Both shift work and profit sharing for executive managers ex ert a strong impact on productivity. The positive influence of managerial profit sharing is consistent with the hypothesis that profit sharing may help to reduce agency problems. However, the coefficient on works council presence is now 13 This finding is consistent with the formal model by Mayshar and Halevy (1997). Schmollers Jahrbuch 123 (2003) 3 OPEN ACCESS | Licensed under CC BY 4.0 | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/schm.123.3.397 | Generated on 2023-04-04 12:32:19 414 Uwe Jirjahn Table 3 Pooled Productivity Estimates for 1994 and 1996 (All Establishments) Variable (1) (2) (3) Constant 158.70 158.14 148.53 (11.25)*** (11.18)*** (10.56)*** Time dummy 1996 9.409 7.847 8.748 (2.52)** (2.11)*** (2.35)** Establishment size 0.009 0.005 0.008 (number of employees) (2.40)** (2.43)** (2.34)** Production technology at the 20.031 16.149 19.29 newest level (4.80)*** (3.79)*** (4.59)*** Single establishment -11.42 -7.123 -10.422 (2.75)*** (1.69)* (2.49)** Proportion of blue collar -0.346 -0.442 -0.313 workers (%) (2.46)** (3.11)*** (2.23)** Proportion of part-time -0.726 -0.633 -0.663 workers (%) (3.27)*** (2.87)*** (2.99)*** Proportion of women (%) -0.566 -0.576 -0.579 (4.73)*** (4.89)*** (4.86)*** Proportion of apprentices (%) -1.924 -1.899 -1.867 (6.88)*** (6.85)*** (6.71)*** Proportion of university 0.967 0.933 0.982 graduates (%) (2.05)** (2.06)** (2.15)** Coverage by a collective -4.079 -5.286 -2.847 agreement (0.88) (1.16) (0.618) Works council 13.413 4.447 18.197 (2.79)*** (0.88) (3.35)*** Shift work 24.174 (5.30)*** Profit sharing for executive 10.707 22.550 managers (2.65)*** (3.26)*** Profit sharing for executive -17.261 manager * works council (2.09)** Industry dummies Included Included Included Number of observations 1057 1057 1057 R2 0.256 0.284 0.266 (4) 153.41 (10.90)*** 7.876 (2.13)** 0.005 (2.53)** 15.78 (3.71)*** -7.367 (1.76)* -0.434 (3.08)*** -0.610 (2.77)*** -0.579 (4.93)*** -1.873 (6.79)*** 0.966 (2.16)** -4.588 (1.01) 11.637 (2.15)** 24.393 (5.36)*** 22.129 (3.24)*** -18.186 (2.23)** Included 1057 0.287 Method: OLS. *, ** or *** denote significance at a = 0.10, 0.05 or 0.01. Heteroscedastic-con sistent t-statistics are in parentheses, White's (1980) method. Schmollers Jahrbuch 123 (2003) 3 OPEN ACCESS | Licensed under CC BY 4.0 | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/schm.123.3.397 | Generated on 2023-04-04 12:32:19 Executive Incentives, Works Councils and Firm Performance 415 statistically insignificant. There are two explanations for this result. First, as shown in Table 2, both shift work and managerial profit sharing are positive covariates of works council presence. If for example work councils have posi tive productivity effects through their role in establishing and maintaining shift work, the inclusion of a shift work variable will decrease the coefficient on works council presence. Second, the coefficient on works council presence will be overor underestimated, if it is not controlled for the interaction of works councils with other variables. The estimates in column (3) and (4) of Table 3 show that there is a strong and negative interaction effect of works councils and managerial profit sharing on productivity. Taking into account this interaction effect, both the coefficient on profit sharing for executive man agers and the coefficient on works council presence increase substantially. Moreover, the impact of works councils is statistically significant regardless of the inclusion of shift work in the regression. However, the coefficient on works council presence is clearly smaller when controlling for shift work. Interestingly, the negative interaction effect of managerial profit sharing and works councils contrasts with the positive interaction effect of profit shar ing for employees and works councils in the study by FitzRoy and Kraft (1995). This clearly indicates that works councils may play a different role for managerial profit sharing and for profit sharing applied to the work force as a whole. The negative interaction effect shown in Table 3 is consistent with the notion that works councils are of particular importance for the economic suc cess of establishments, when no managerial profit sharing is in place. How ever, the theoretical explanation for the interaction effect of profit sharing for executive managers and works councils is complex. The hypotheses sum marised in Table 1 show that there are at least two possible explanations. First, if profit sharing decreases the commitment value of agency and works coun cils cannot foster trust and loyality without the managers' cooperation, a nega tive interaction effect on establishment performance may result. Second, if managerial profit sharing induces managers to reduce their rent-seeking activ ities and works councils are not so important for building cooperation in estab lishments with less severe manager opportunism, a negative interaction effect is also likely to be found. Finally, we perform the estimates for the subsample of establishment with 21 to 100 employees. The results are shown in Table 4. With the exception of the insignificant coefficient estimates for establishment size and the time dummy, the results for the control variables are sirnilar to those shown in the estimates with the full sample. Managerial profit sharing exerts a positive influence on productivity also in the estimates with the subsample of smaller establishments.14 However, no statistically significant interaction effect of 14 Mueller and Spitz (2001) obtain a similar result for small and medium-sized firms. In their study, managerial ownership up to around 80 percent increases firm perfor mance. Schmollers Jahrbuch 123 (2003) 3 OPEN ACCESS | Licensed under CC BY 4.0 | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/schm.123.3.397 | Generated on 2023-04-04 12:32:19 416 Uwe Jirjahn Table 4 Pooled Productivity Estimates for 1994 and 1996 (Establishments with 21 -100 Employees) Variable (1) (2) (3) Constant 165.59 167.39 157.60 (8.06)*** (7.95)*** (7.66)*** Time dummy 1996 -2.496 -3.204 -2.973 (0.49) (0.64) (0.59) Establishment size 0.010 0.017 0.074 (number of employees) (0.83) (0.14) (0.62) Production technology at the 22.052 19.097 21.455 newest level (3.40)*** (2.88)*** (3.30)*** Single establishment -16.12 -14.71 -15.031 (2.52)** (2.29)** (2.32)** Proportion of blue collar -0.628 -0.700 -0.581 workers (%) (2.95)*** (3.30)*** (2.71)*** Proportion of part-time -0.872 -0.851 -0.865 workers (%) (2.43)** (2.40)** (2.41)** Proportion of women (%) -0.409 -0.392 -0.410 (2.60)*** (2.49)** (2.63)*** Proportion of apprentices (%) -2.026 -2.050 -1.954 (4.95)*** (4.98)*** (4.69)*** Proportion of university 2.096 1.945 2.053 graduates (%) (2.11)** (1.98)** (2.10)** Coverage by a collective -8.861 -8.79 -8.360 agreement (1.53) (1.53) (1.44) Works council 15.989 14.532 18.180 (2.32)** (2.14)** (2.60)*** Shift work 18.456 (2.82)*** Profit sharing for executive 12.375 15.652 managers (2.65)*** (2.01)** Profit sharing for executive -6.354 manager * works council (0.60) Industry dummies Included Included Included Number of observations 438 438 438 R2 0.323 0.343 0.331 (4) 165.38 (8.05)*** -3.277 (0.65) 0.021 (0.17) 18.952 (2.85)*** -14.668 (2.29)** -0.700 (3.31)*** -0.861 (2.39)** -0.393 (2.50)** -2.061 (5.00)*** 1.942 (2.00)** -8.739 (1.52) 18.066 (2.56)** 18.674 (2.84)*** 16.740 (2.20)** -7.789 (0.74) Included 438 0.344 Method: OLS. *, ** or *** denote significance at a = 0.10, 0.05 or 0.01. Heteroscedastic-con sistent t-statistics are in parentheses, White's (1980) method. Schmollers Jahrbuch 123 (2003) 3 OPEN ACCESS | Licensed under CC BY 4.0 | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/schm.123.3.397 | Generated on 2023-04-04 12:32:19 Executive Incentives, Works Councils and Firm Performance 417 managerial profit sharing and works council on productivity can be found. This result indicates implicitly that the negative interaction effect, shown in Table 3, is particularly prevalent in larger establishments. Most noteworthy, the positive coefficient on works council presence is statistically significant, even when controlling for managerial profit sharing and shift work. The find ing accords with the missing interaction effect and with the missing associa tion of works council presence with managerial profit sharing and shift work in this subsample of smaller establishments (see Table 2). An earlier study by Smith (1994) with data provided by the Institut für Ar beit und Technik shows also positive productivity effects of works councils for smaller establishments. Addison, Schnabel and Wagner (2001) fail to find a statistically significant productivity effect of works councils for smaller es tablishments. However, they restrict their analysis to productivity estimates for 1993 with the first wave of the Hannover Panel.15 In contrast, the results in Table 4 show that it is reasonable to use the füll information provided in a data set. The coefficient estimates for the various variables in Table 4 are generally better determined compared to those provided by Addison, Schnabel and Wagner. This casts some doubts on the specification used by Addison, Schna bel and Wagner and on their restriction of the analysis to the first wave. Final ly, Jirjahn (2003a) examines the productivity effects for establishments with 21 to 100 employees in more detail. Confirming the theoretical models in Freeman and Lazear (1995) and Hübler and Jirjahn (2003), the study by Jir jahn (2003a) shows that this positive productivity effect of works councils can be only found for smaller establishments covered by collective agreements but not for smaller establishments without collective agreements. 4. Concluding Remarks For our understanding of works councils it is crucial to take into account the factors that influence their economic effects. This study confirms the notion that the managerial environment plays an important role. The theoretical fra mework provided in this paper identifies two relevant relationships for the in teraction of executive compensation and works councils, namely the relation ship between codetermination and self-enforcing contracts and the relation ship between agency problems and trustful employer-employee relations. The empirical analysis shows a negative interaction effect of profit sharing for ex1s Addison, Schnabel and Wagner (2001) justify their restriction with the sample at trition over the waves of the Hannover panel. However, they do not provide any evi dence that panel attrition causes serious problems on using the entire data set. The ad hoc restriction used by Addison, Schnabel and Wagner allows no use of the various advantages a panel data analysis provides (e.g. less collinearity, identification of out liers, controlling for establishment-specific heterogenity). Schmollers Jahrbuch 123 (2003) 3 OPEN ACCESS | Licensed under CC BY 4.0 | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/schm.123.3.397 | Generated on 2023-04-04 12:32:19 418 Uwe Jirjahn ecutive managers and works councils on productivity. This negative interac tion effect seems to be particularly prevalent in larger establishments. The results of this paper clearly call for further research. First, the negative interaction effect of managerial profit sharing and works councils is consistent with at least two explanations. In order to test these two explanations, we need more direct information on the managers' willingness to cooperate with the works councils. One way would be to examine the impact of managerial profit sharing on the managers' attitudes toward codetermination. If managerial profit sharing reduces the commitment value of agency, a negative impact of profit sharing on the managers' attitudes toward works councils can be ex pected. In contrast, if profit sharing reduces rent-seeking activities of man agers, a positive impact of profit sharing on the managers' willingness to co operate with works councils can be expected. Second, the relationship be tween codetermination and the form of profit sharing deserves greater scru tiny. Third, further research should not only take into account the managers' compensation but also variables, which indicate the severity of information asymmetries between owners and managers. Fourth, further insights could ob tained by combining the analysis of this paper with an analysis of additional interactions such as the interplay of works councils and collective bargaining. References Addison, J. T. / Bellmann, L. / Schnabel, C. / Wagner, J. (2002), The Long Awaited Re form of the German Works Constitution Act, IZA Discussion Paper No. 422, Bonn. Addison, J. T. / Schnabel, C. / Wagner, J. (2001), Works Councils in Germany: Their Ef fects on Firm Performance, Oxford Economic Papers 53, 659 -694. -(1997), On the Determinants of Mandatory Works Councils in Germany, Industrial Relations 36, 419 -445. Askildsen, J. EI Ireland, N. (1993), Human Capital, Property Rights and Labour Mana ged Firms, Oxford Economic Papers, 45, 229-242. Askildsen, J. E. / Jirjahn, U. / Smith, S. C. (2002a), Works Councils and Investments in Skills, Working Paper. -(2002b), Works Councils and Environmental Investment: Theory and Evidence from German Panel Data, CESifo Working Paper No. 785, Munich. Bellmann, L. / Kohaut, S. / Schnabel, C. (1999), Flächentarifverträge im Zeichen von Abwanderung und Widerspruch: Geltungsbereich, Einflussfaktoren und Öffnungsten denzen, in: L. Bellmann/V. Steiner (eds.), Panelanalysen zur Lohnstruktur, Qualifi kation und Beschäftigungsdynamik, BeitrAB: Nürnberg, 11 -40. Black, S. E. / Lynch, L. M. (2000), What's Driving the New Economy: The Benefits of Workplace Innovation, NBER Working Paper No. 7479. Schmollers Jahrbuch 123 (2003) 3 OPEN ACCESS | Licensed under CC BY 4.0 | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/schm.123.3.397 | Generated on 2023-04-04 12:32:19 Executive Incentives, Works Councils and Firm Performance 419 Brand, R. ! Carstensen, V. / Gerlach, K. / Klodt, T. (1996), Tue Hannover Panel, For schungsstelle Firmenpanel, Discussion Paper No. 2, Hannover. Bull, C. (1987), The Existence of Self-Enforcing Implicit Contracts, Quarterly Journal of Econornics 52, 147-159. Cooke, W. N. (1994), Employee Participation Programs, Group-Based Incentives, and Company Performance: A Union-Nonunion Comparison, Industrial and Labor Rela tions Review 47, 594-609. Dilger, A. (2003), Payment Schemes, Returns and Works Councils, Schmollers Jahr buch, this issue. -(2002), Ökonomik betrieblicher Mitbestimmung: Die wirtschaftlichen Folgen von Betriebsräten, Mering: Rainer Hampp Verlag. Duncan, G. J. / Staffo rd F. P. (1980), Do Union Members Receive Compensating Wage Differentials? American Econornic Review 70, 354-371. FitlR.oy, F. R. / Kraft, K. (1995), On the Choice of Incentives in Firms, Journal of Eco nomic Behavior and Organization 26, 145 -160. -(1987), Efficiency and Internal Organization: Works Councils in West German Firms, Economica 54, 493 -504. Freeman, R. B. (1976), Individual Mobility and Union Voice in the Labor Market, American Econornic Review 66, 361 -368. Freeman, R. B. / Lazear, E. P. (1995), An Econornic Analysis of Works Councils, in: J. Rogers/W. Streeck (eds.), Works Councils -Consultation, Representation and Co operation in Industrial Relations, Chicago: University of Chicago Press, 27 -52. Freeman, R. B. / Medojf, J. L. (1979), The Two Faces of Unionism, The Public Interest 57, 69 -93. Frick, B. (1996), Co-deterrnination and Personnel Turnover: The German Experience, Labour 10, 407-430. Frick, B. ! Sadowski, D. (1995), Works Councils, Unions and Firm Performance: The Impact of Workers' Participation in Germany, in: F. Buttler / W. Franz/ R. Schettkat/ D. Soskice (eds.), Institutional Frameworks and Labor Market Performance: Com parative Views on the U.S. and German Economies, New York: Routledge, 46-81. Gerlach, K. ! Jirjahn, U. (2001), Employer Provided Further Training: Evidence from German Establishment Data, Schmollers Jahrbuch 121, 139-164. Hall, B. J. / Liebman, J. B. (1998), Are CEOs Really Paid Like Bureaucrats? Quarterly Journal of Economics 111, 653-691. Heywood, J. S. / Hübler, 0. / Jirjahn, U. (1998), Variable Payment Schemes and Indus trial Relations: Evidence from Germany, Kyklos 51, 237 -257. Heywood, J. S. / Jirjahn, U. (2002), Payment Schemes, Gender and Industrial Relations in Germany, Industrial and Labor Relations Review 56, 44-64. Hogan, C. (2001), Enforcement of Implicit Employment Contracts through Unioniza tion, Journal of Labor Economics, Vol. 19, 171 -195. Schmollers Jahrbuch 123 (2003) 3 OPEN ACCESS | Licensed under CC BY 4.0 | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/schm.123.3.397 | Generated on 2023-04-04 12:32:19 420 Uwe Jirjahn Hübler, 0. / Jirjahn, U. (2003), Works Councils and Collective Bargaining in Germany: The Impact on Productivity and Wages, Scottish Journal of Political Economy 50, 1-21. -(2002), Arbeitsproduktivität, Reorganisationsmaßnahmen und Betriebsräte, in: L. Bellmann/ A. Kölling (eds.), Betriebliche Reorganisation und Fachkräftebedarf, Nürnberg: Beiträge zur Arbeitsmarktund Berufsforschung, 1-45. Hübler, 0. / Meyer, W. (2001), Industrial Relations and the Wage Differentials within Firms, Schmollers Jahrbuch 121, 285 -312. Jirjahn, U. (2003a), Produktivitätswirkungen betrieblicher Mitbestimmung -Welchen Einfluss haben Betriebsgröße und Tarifbindung? Zeitschrift für Betriebswirtschaft, Ergänzungsheft 4, 63-85. -(2003b), X-Ineffizienz, Manageranreize und Produktmarktwettbewerb, Bern et al.: Peter Lang Verlag. -(2002), The German Experience with Performance Pay, in Paying for Performance: An International Comparison, edited by M. Brown and J. S. Heywood, New York: M. E. Sharpe, 148 -178. -(1998), Effizienzwirkungen von Erfolgsbeteiligung und Partizipation -Eine rnikroö konomische Analyse, Campus: Frankfurt. Kostiuk, P. F. (1990), Compensating Differentials for Shift Work, Journal of Political Economy, 95, 1055 -1075. Kraft, K. / Niederprüm, A. (1999a), Determinants of Management Compensation with Risk-Averse Agents and Dispersed Ownership of the Firm, Journal of Econornic Be havior and Organization, Vol. 40, 17 -27. -(1999b), Ist die Vergütung von Managern im Zeitablauf flexibler geworden? Zeit schrift für betriebswirtschaftliche Forschung 51, 787 -804. Kreps, D. M. (1990), Corporate Culture and Economic Theory, in: J. E. Alt/ J. A. Shepsle (eds.), Perspectives on Positive Political Economy, Cambridge: Cambridge University Press, 90 -143. Laband, D. N. / Lentz, B. F. (1998), The Effects of Sexual Harassment on Job Satisfac tion, Eamings, and Tumover among Female Lawyers, lndustrial and Labor Relations Review, Vol. 51, 594-607. Laffont, J. J. (1990), Analysis of Hidden Gaming in a Three-Level Hierarchy, Journal of Law, Economics and Organization 6, 301 -324. Mayshar, J. / Halevy, Y. (1997), Shiftwork, Journal of Labor Economics, Vol. 15, 198 - 222. Mueller, E. / Spitz, A. (2001), Managerial Ownership and Firm Performance in German Small and Medium-Sized Enterprises, Centre for European Econornic Research, Dis cussion Paper No. 01 -72. Müller-Jentsch, W. (1995), Germany: From Collective Voice to Co-management, in: J. Rogers/W. Streeck (eds.), Works Councils -Consultation, Representation and Co operation in lndustrial Relations, Chicago: University of Chicago Press, 53 -78. Schmollers Jahrbuch 123 (2003) 3 OPEN ACCESS | Licensed under CC BY 4.0 | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/schm.123.3.397 | Generated on 2023-04-04 12:32:19