Top management teams’ shared leadership and ambidexterity: the role of management control systems

Abstract

The study explores how top management teams’ shared leadership is related to organizational ambidexterity in public-sector organizations, theoretically and empirically considering how this relationship is contingent on the management control system. Using a sample of 85 Swedish municipal housing corporations, we find that shared leadership has a positive relationship with organizational ambidexterity in public-sector organizations. Moreover, increasing use of new public management control systems, based on combined reward and performance controls, positively moderates this relationship. The study also finds that traditional public management control systems, based on combined planning and administrative controls, do not moderate the relationship between top management teams’ shared leadership and organizational ambidexterity. Accordingly, this article contributes to the public and strategic management literature, as well as to managerial practice.

Points for practitioners

The article suggests that sharing leadership within top management teams can result in a balanced resource allocation in municipal corporations. To be more effective in achieving this balance, public sector managers might consider emphasizing new public management-inspired management control systems and de-emphasizing those of a more traditional type.

Keywords

ambidexterity management control system municipal corporations shared leadership top management team

Introduction

Organizational ambidexterity – a non-financial organizational outcome referring to organizational ability to engage in both explorative and exploitative activities – is a popular organizational outcome in strategic management and corporate governance (e.g. Lubatkin et al., 2006). Achieving organizational ambidexterity arguably relates to top managers’ characteristics, leadership styles and means of influencing organizational performance (Umans, 2013). Researchers generally agree that top management teams (TMTs) represent an important locus for resolving inherent tensions in balancing explorative and exploitative activities and in achieving organizational ambidexterity (e.g. Smith and Tushman, 2005).

Studies exploring the role of top managers in achieving organizational ambidexterity traditionally adopt a top-down approach, assuming that chief executive officer (CEO) leadership affects the TMT and other organizational members (e.g. Smith and Umans, 2015). CEO decisions are, in turn, reflected in managerial ability to balance the firm’s explorative and exploitative activities (Mihalache et al., 2014). Emerging research suggests an alternative perspective in which leadership behaviour is distributed among TMT members rather than concentrated in one individual (Mihalache et al., 2014). We posit that the latter perspective on leadership holds greater promise for understanding the emergence of organizational ambidexterity. The concept of TMT shared leadership embedded in that perspective represents a ‘team process where leadership is carried out by the team as a whole, rather than solely by a single designated individual’ (Ensley et al., 2006). It allows functionally diverse top managers to handle large amounts of information, create alternatives and better address conflicts and ambiguity (Umans, 2013). These aspects are shown to result in an ability to balance explorative and exploitative resource allocation (Mihalache et al., 2014).

Raisch and Birkinshaw (2008) argue that while considerable effort has been made to understand what triggers organizational ambidexterity, few studies explore the conditions under which ambidexterity emerges (for exceptions, see Jansen et al., 2009; Mihalache et al., 2014). According to Benner and Tushman (2003), organizational ability to achieve ambidexterity rests on the ability to address various strategies, structures and processes. Drawing on management accounting and control research, we propose that management control systems are closely associated with structures and processes, being essential in decisions concerning resource use and allocation (Merchant and Van der Stede, 2007). Management control systems should be regarded as important contingencies in the relationship between TMT shared leadership and organizational ambidexterity.

Organizational ambidexterity is recognized as a useful concept in understanding the outcomes of public-sector organizations (e.g. Smith and Umans, 2015) given that their ultimate focus is often not financial performance, but non-financial performance, specifically, the ability to efficiently allocate resources to satisfy stakeholder needs (Bozeman, 1987). Research into ambidexterity in the public sector has been limited (e.g. Aargard, 2011; Cannaerts et al., 2016; Choi and Chandler, 2015; Smith and Umans, 2015). Existing studies have commonly treated ambidexterity as an outcome (e.g. Smith and Umans, 2015), taken a leadership approach (e.g. Tuan, 2017) or studied ambidextrous employees (e.g. Aargard, 2011). Understanding how organizational ambidexterity is achieved in the public sector and the conditions under which it emerges would advance our understanding of how public-sector organizations create stakeholder value.

This study explores how TMT shared leadership relates to organizational ambidexterity in public-sector organizations and how this relationship depends on management control systems. We explore these relationships empirically using a survey distributed to the TMT members of Swedish municipal housing corporations, seeking to make several theoretical and practical contributions. First, we explore the link between TMT shared leadership and organizational ambidexterity, advancing our understanding of the role of leadership behaviour in achieving balance between explorative and exploitative organizational activities. Second, we explore the conditions under which TMT leadership facilitates ambidexterity, advancing our understanding of the internal organizational processes and structures enabling ambidexterity. Third, by exploring our model in a public-sector context, we contribute to the limited research into how public-sector organizations can more efficiently allocate resources, and into the triggers and contingencies that organizations should consider. In practical terms, this article helps public-sector organization managers to understand how a combination of leadership approach and management control system can give a more balanced resource allocation, ultimately improving organizational performance.

Theory development

Organizational ambidexterity

Organizational ambidexterity connotes the organizational ability to balance explorative and exploitative orientations when using resources (Duncan, 1976): exploration is characterized by search and discovery, while exploitation is characterized by disciplined problem-solving and refinement (March, 1991). The concept was introduced by Duncan (1976) and further explored by March (1991). Although organizational ambidexterity research was pioneered by strategy scholars, the concept has been investigated from various perspectives, including organizational learning, technological innovation, organizational adaptation, strategic management and organization design (Raisch and Birkinshaw, 2008). Early studies explored organizational ambidexterity in itself, whereas later studies paid greater attention to the mechanisms, that is, organizational structure, context and leadership (Tuan, 2017), supporting organizational ambidexterity. Another topic of research interest is organizational ambidexterity and its performance effects (e.g. Lubatkin et al., 2006).

While public-sector research has increasingly explored concepts from strategic management research (e.g. Modell, 2012), ambidexterity has rarely been examined. For example, researchers have studied individual ambidexterity (Kobarg et al., 2017) and ambidextrous leadership (Tuan, 2017). Organizational ambidexterity and its antecedents have been examined in conference/working papers (e.g. Aargard, 2011) and a limited number of scientific articles (Cannaerts et al., 2016; Choi and Chandler, 2015; Palm and Lilja, 2017; Smith and Umans, 2015). Since the reforms of new public management (NPM) and new public governance, the sector has needed to address innovation (Choi and Chandler, 2015), an indicator of the relevance of further understanding public organizational ambidexterity.

TMT shared leadership and organizational ambidexterity in the public sector

Research argues that organizational ability to pursue and balance explorative and exploitative activities depends on top management ability to choreograph this complex process (Halevi et al., 2015). According to Simsek (2009), we need to understand the mechanisms through which this occurs.

Top managers are generally functionally diverse with varied priorities, resulting in an emphasis on explorative or exploitative activities (Mihalache et al., 2014) and possibly leading to balance. Since achieving balance between exploration and exploitation is important for improved organizational outcomes (Umans, 2013), TMTs need to use team processes to overcome personal biases and conflicts of interest (Smith and Tushman, 2005). This allows them to develop mental schemata for considering exploration and exploitation as mutually complementary rather than opposed (Carmeli and Halevi, 2009). Top managers’ ability to share responsibilities, such as strategic goal setting and execution, has been proposed as a dimension of TMT work reflected in ambidexterity as an organizational outcome (Mihalache et al., 2014). This TMT process has been called ‘shared leadership’ (Ensley et al., 2006). In a TMT context, it connotes the distribution of leadership tasks, such as strategic decision-making and execution, priority identification, and team maintenance, among TMT members, instead of the tasks being concentrated in the hands of the CEO (Mihalache et al., 2014). In other words, shared leadership represents the opposite of traditional vertical leadership, in which leadership is transmitted down to TMT members. Shared leadership instead emphasizes the lateral distribution of leadership and influence among all TMT members (Mihalache et al., 2014).

Public management scholars are increasingly aware of leadership theory (e.g. Nelissen et al., 2000), noting the importance of its concepts and, especially, of shared leadership in the public sector. Moreover, public managers arguably play important roles in influencing change and development in public organizations (Smith and Umans, 2015). According to Fernandez et al. (2010), understanding shared leadership in teams and on different hierarchical levels in public organizations helps in forming leadership teams and making these organizations more efficient. This appears especially important given empirical evidence that organizations with high levels of shared leadership tend to share strategic decision-making (Nelissen et al., 2000). Furthermore, Rainey and Steinbauer (1999) suggest that leadership is crucial for public organizations given their vague and multifarious nature, with numerous conflicting organizational goals.

In line with Mihalache et al. (2014), we argue that TMT shared leadership will lead to organizational ambidexterity given the TMT’s ongoing negotiations and roles, shared aspirations, and perceived team success (Manz and Sims, 1993). It has also been argued that successfully sharing leadership requires high team cohesion and high-quality mutual and collective interactions (Carmeli and Halevi, 2009). Team cohesiveness advances the evolution of shared vision and purpose, alleviating the negative effects of behavioural differences (Umans, 2013). Maintaining a diverse and complex behavioural repertoire (Carmeli and Halevi, 2009) while fostering a positive team environment helps TMTs better assess resource-allocation options, ultimately leading to organizational ambidexterity. This discussion leads to Hypothesis 1:

Hypothesis 1: TMT shared leadership is positively related to organizational ambidexterity in public-sector organizations.

Moderating role of management control systems in public-sector organizations

Management control systems

Researchers argue that the way forward in understanding how leadership affects the explorative/exploitative orientation of firms lies in understanding contingencies (Carmeli and Halevi, 2009). Jansen et al. (2009) found, for example, that an external contextual contingency significantly influenced a TMT’s transformational and transactional leadership, as well as explorative and exploitative orientations. Mihalache et al. (2014) found that connectedness serves as an important contingency factor in the relationship between TMT shared leadership, cooperative conflict and shared decision-making, which, in turn, are reflected in organizational ambidexterity. A more recent study of the relationship between TMT behavioural integration and organizational ambidexterity, contingent on environmental dynamism, found that the relationship is stronger when high environmental dynamism is present (Halevi et al., 2015).

To further explore relevant contingency factors, we turned to the management control literature, where contingency theory has been most significantly developed (e.g. Gerdin and Greve, 2004). Taking note of research into management control systems (e.g. Malmi and Brown, 2008) and applying it to public-sector organizations, we focused on two management control systems. One, which we call the ‘traditional public management control system’ (TPMCS), concentrates on input factors (Osborne and Gaebler, 1992), leading to a control system emphasizing input control (Spano, 2009) in which planning and administration are essential. We call the other system, influenced by the NPM reforms of the late 1970s, the ‘new public management control system’ (NPMCS). It concentrates on output control and performance measures (Speklé and Verbeteen, 2014).

The TPMCS

The TPMCS includes elements of administrative and planning control. Administrative control is exercised through set processes for how work tasks should or should not be performed (Malmi and Brown, 2008). It usually distinguishes between administrative design and structure, governance structures, and procedures and policies. Administrative control can refer to, among other things, control through organizational structure (e.g. functional, multidivisional and matrix), lines of authority and responsibility (e.g. TMT and board structures), and operating procedures (Malmi and Brown, 2008). Clarity of command and clear procedures for performing work allow a leadership team to focus on leading the organization. Simultaneously, administrative control might be perceived as overly bureaucratic and narrow, demotivating personnel in their commitment to reach common goals. The hierarchy and vertical structures typifying such control might further reduce TMT members’ ability to participate in shared leadership activities, possibly because they need to get involved in top-down management and focus on governance throughout the hierarchy. This implies that even if TMT members practise shared leadership, its effect on organizational ambidexterity might be weakened by the bureaucratic structures associated with administrative control (Reimann, 1973). According to Uhl-Bien and Marion (2009), bureaucratic administrative structures are usually detrimental for organizations seeking to balance complexity, especially given these structures’ misalignment with leaders’ dynamic interactions, bonding and aggregation, which are integral to shared leadership. This suggests that administrative control has a negative moderating effect on the relationship between TMT shared leadership and ambidexterity in public-sector organizations.

Planning control is exercised by setting goals for different organizational units, with the idea that controlling through setting consistent goals allows organizations to attain these goals more effectively (Malmi and Brown, 2008). Strengthening awareness of these goals is central to such control given that individuals need to be made aware of the demands that management places on them. Planning control emphasizes both long- and short-term plans, with strategic goals representing the long-term and operational goals the short-term orientation. This duality of planning control matches the duality of organizational ambidexterity. Research has also shown that planning control characterizes environments in which innovative solutions are hindered because resources are allocated to improving existing ‘assets’ rather than creating new ones (cf. Jansen et al., 2006). This, in turn, could mean that planning control stimulates exploitation rather than exploration, disturbing the balance between the two that ambidexterity represents. TMTs that share leadership and are well equipped to balance exploration and exploitation (Mihalache et al., 2014) might be hindered by management control systems emphasizing exploitation. This structural mismatch would hinder the emergence of organizational ambidexterity.

Accordingly, we argue that the TPMCS is misaligned with TMT shared leadership in that it promotes a top-down view of organizational development, especially in decision-making (Wang et al., 2014). When combined with TMT shared leadership, a TPMCS is expected to have a negative moderating effect on organizational ambidexterity, leading to Hypothesis 2:

Hypothesis 2: Increasing the use of TPMCS has a negative moderating effect on the relationship between TMT shared leadership and organizational ambidexterity in public-sector organizations.

The NPMCS

The NPMCS includes performance and reward control. Performance control is usually exercised through budgets and financial and non-financial indicators of employee performance (Vázquez et al., 2006). In such control systems, managers on different organizational levels are accountable and responsible for their units’ performance and ability to adhere to the budget (Malmi and Brown, 2008). Performance control is among the most extensively used types of control in private- and public-sector organizations (Malmi and Brown, 2008). Setting performance-oriented goals and regularly following them up give staff freedom in how to achieve these goals. According to Lubatkin et al. (2006), this more rational and sustainable use of available firm resources ultimately improves organizational performance. Given this argument, we posit that flexibility in achieving goals in relation to TMT shared leadership positively affects organizational ambidexterity. This is especially the case given the alignment between TMT leadership flexibility and the flexibility emphasized by performance control (cf. Das, 1995), in which individuals are encouraged to attain individual goals aligned with organizational goals.

Reward control involves establishing various rewards for individuals or groups seeking to attain organizational goals (Bonner and Sprinkle, 2002). Reward systems are extrinsic or intrinsic: the former are represented by monetary incentives; the latter by non-monetary incentives, such as professional development opportunities or increasing independence in performing work tasks (Malmi and Brown, 2008). Reward systems, especially because of their monetary nature, are associated with increased competition between individuals and/or organizational units. Increasing use of reward controls in public organizations might be problematic given indications that they can create unhealthy and destructive competition (Kellough and Selden, 1997). However, recent studies of public-sector reward control have found that it results in more efficient organizational functioning when combined with clear leadership objectives (Kim, 2010). Reward control is associated with clear incentive schemata that individuals generally appreciate, especially in the public sector (cf. Karl and Sutton, 1998). Given that the latter has only recently seen the introduction of reward-related systems, management team members increasingly need to interpret and translate such systems and their functions (Hoch and Kozlowski, 2014). TMT ability to share this interpretative responsibility and leadership is instrumental in making sense of the new system, and the combination and alignment of the reward system with shared leadership potentially allow organizations to become more efficient, including in distributing and balancing resource use.

NPMCSs that incorporate aspects of performance and reward control are relatively new to public organizations. Such systems have been introduced to improve efficiency and reduce the complexity (cf. Boland and Fowler, 2000) usually associated with the functioning of public-sector organizations. NPMCSs do this by identifying performance goals and being clear about the associated rewards for success. Given the newness of the NPMCS, uncertainty remains as to goals and rewards being attained (cf. Pasquier and Villeneuve, 2007), an aspect requiring that more than just one individual play a leadership role (Morgeson et al., 2010). The approach’s novelty might further require that multiple management team members be involved in sharing information and perspectives (Hoch and Kozlowski, 2014). TMT shared leadership and the NPMCS are also aligned through interaction between the complex behavioural repertoire generated in TMT decision-making and the efficiency orientation of NPMCSs. Through this interaction, exploration and exploitation are balanced, leading to Hypothesis 3:

Hypothesis 3: Increasing the use of NPMCS has a positive moderating effect on the relationship between TMT shared leadership and organizational ambidexterity in public-sector organizations.

Method

The study’s empirical object was Swedish municipal housing corporations, an industry characterized by a need to be aware of changing demographic conditions, technological advances and overall societal developmental trends (Lind, 2001), suggesting good opportunities to capture elements of both explorative and exploitative behaviour.

In April 2015 (data-collection point), there were 262 municipal housing corporations in Sweden (Planken and Andersson, 2015). We sent a questionnaire by email to the Chief Executive Officer (CEO), Chief Financial Officer (CFO) and one other member of the TMT (usually the one responsible for building and production) of each of these corporations. We were able to send questionnaires to 557 of the 786 potential respondents as not all TMT members’ email addresses appeared on the corporations’ websites. In such cases, we sent the email to the information address with a request to forward it to the respondent specified. We sent the questionnaire to multiple respondents per corporation to minimize respondent bias. Unfortunately, this did not happen because only a few organizations yielded multiple respondents; in most cases, only one person responded per organization. Of 786 questionnaires sent, we received 119 (a 21% response rate); of these, 34 were incomplete (up to 70% incomplete) and were omitted from further analysis. The final sample comprised 85 questionnaires, with 49 completed by CEOs and 36 by CFOs. Non-response analysis identified no major bias among respondents and non-respondents with regard to municipal classification based on structural parameters. We used Harman’s single-factor test to check for common method bias (Podsakoff et al., 2003), including all independent and dependent variables in one unrelated factor analysis. The results indicated that the data had a multiple factor structure and limited common method bias. Respondents were asked to indicate their agreement with statements about their TMT (the independent variable) and organization (serving as the dependent and moderating variables) on a seven-point Likert scale ranging from (1) strongly disagree to (7) strongly agree.

Dependent variable: organizational ambidexterity

We used three of six original statements for explorative innovation and three of six original statements for exploitative innovation from Jansen et al. (2006), adapting them to the public-sector context. The selection and adaptation of the measures were based on findings of a case study involving seven interviews with top managers in two municipal organizations in southern Sweden. We used three of six original statements measuring explorative innovation and three of six original statements measuring exploitative innovation. Principal component analysis (PCA), used to validate the measures, revealed a two-factor structure, with the exploration and exploitation questions in line with predictions. To calculate organizational ambidexterity, in line with Lubatkin et al. (2006) and after checking for construct validity, we summed the six questions measuring exploration (α = 0.803) and exploitation (α = 0.765).

Independent variable

The measure of TMT shared leadership was adapted from Mihalache et al. (2014). We used four of the original eight statements, basing the selection on findings of the aforementioned case study. After checking for construct reliability, TMT shared leadership (α = 0.965) was measured as the sum of the four questions posed.

Moderating variables

Our management control system measures were inspired by Malmi and Brown (2008) and the case study findings. We developed eight questions, four to capture TPMCS and four to capture NPMCS. We performed PCA to explore the control system instrument developed for this study given the potential uncertainty concerning new instruments and their reliability. We checked for correlations between the variables (which exceeded 0.3); as a measure of sampling adequacy (which was high), we used the Kaiser-Meyer-Olkin test (above 0.7). The eigenvalues and screen plot indicated that both components should be used (all communality values exceeded the preferred 0.4). One factor comprised all questions about TPMCS and the other factor all questions about NPMCS; no cross-loading was observed.

The TPMCS scales comprised four questions about the planning and administrative control system dimensions, representing the input factors. These questions were inspired by Osborne and Gaebler (1992) and Malmi and Brown (2008). The summative scores of the four items achieved acceptable validity (α = 0.867).

The NPMCS scales comprised four questions about reward and performance control system dimensions, representing the output factors. These questions were inspired by Hood (1991) and Malmi and Brown (2008). The summative scores of the four items achieved acceptable validity (α = 0.787).

Control variables

We used six control variables. In line with previous studies, we controlled for team size, given its importance for shared leadership, and firm size (i.e. self-reported number of employees), as larger firms might have more available resources yet lack flexibility (Mihalache et al., 2014). We also controlled for respondent position (i.e. CEO/CFO) and respondent demographics (i.e. gender, age and higher education).

Analysis and results

Descriptive statistics (see Table 1) indicate that the sampled corporations were moderately ambidextrous (mean 25.21, max 42). The respondents perceived TMT shared leadership in their team to be relatively high (mean 20.12, max 28) and that TPMCS (mean 19.44, max 28) was used more than NPMCS (mean 13.68, max 28) in their organizations. These findings suggest not only that municipal corporations in our sample are employing shared leadership, but also that TPMCS, rather than NPMCS, remains the preferred method of control. In our sample, the average team had 4.58 members; 42% of respondents were CFOs and 58% were CEOs. The respondents’ average age was 49.53 years; 76% of them had higher education (i.e. bachelor’s degree or higher) and 32% were women. The sampled organizations had, on average, 46.34 employees. Our findings provide empirical evidence of relatively large, gender-diverse TMTs with high human capital (education) that closely resemble those of stock-listed corporations (Umans, 2013). This could indicate that municipal corporations in our sample might be facing complexity and dynamism comparable with that of private firms that require larger, diverse top teams with high human capital to address the aforementioned contextual specificities.

Table 1.

Means, standard deviations (SDs) and correlations.

	Variables	Mean	SD	1	2	3	4	5	6	7	8	9
1	Ambidexterity	25.21	6.41
2	TMT shared leadership	20.12	7.90	.28**
3	TPMCS	19.44	5.86	.50**	.63**
4	NPMCS	13.68	5.95	.21^†	–.18	–.20^†
5	TMT size	4.58	2.20	–.06	–.01	.14	–.09
6	Firm size	46.34	58.46	.12	.09	.26*	.01	.39**
7	Position	0.42	0.5	–.16	.05	.05	–.07	.21^†	.12
8	Gender	0.32	0.47	–.04	.03	.11	.03	.19^†	.16	.44**
9	Age	49.53	8.97	–.15	.10	.00	.01	–.03	–.03	–.23*	–.23*
10	Education	0.76	0.42	–.17	–.01	–.04	.02	.01	–.16	–.08	–.02	–.35**

Notes: **p < .01; *p < .05; ^†p < .10.

The correlation matrix indicates that TMT shared leadership and TPMCS are significantly positively correlated with ambidexterity, and that NPMCS is weakly significantly correlated with ambidexterity. This suggests that TPMCS in isolation (given the bivariate nature of correlation analysis) has a stronger relation to ambidexterity than NPMCS does, the latter being a relatively new phenomenon in public-sector organizations. TMT shared leadership also appears to be significantly positively correlated with TPMCS, which might be explained by the similarity of the input-related facets that both concepts possess. There is also a weakly significant negative correlation between TPMCS and NPMCS, indicating that when one of the systems is in use, our sample organizations adjust the use of the other system reciprocally. Finally, there is an indication of significant positive correlation between corporation size and TPMCS use, suggesting that larger firms in our sample might be more traditional in their application of control.

To test our hypotheses, we conducted multiple hierarchical linear regression analyses (see Table 2). There are indications of some multicollinearity in the three regression models, but at 1.283–2.715, it is acceptably below the recommended value of 4 (Pallant, 2013).

Table 2.

Regression analyses: dependent variable ambidexterity.

	Variables	Model 1		Model 2		Model 3
	Variables	Std. B	Std. error	Std. B	Std. error	Std. B	Std. error
	Constant^a	24.348**	5.421	14.745*	5.868	24.772**	5.555
Step 1	TMT size	–.132	.356	–.128	.326	–.174	.340
	Firm size	.185	.013	.056	.012	.114	.012
	Position	–.176	1.545	–.118	1.482	–.231*	1.462
	Gender	.017	1.635	–.032	1.505	–.049	1.535
	Age	.011	.085	.040	.080	–.084	.081
	Education	–.178	1.754	–.099	1.626	–.172	1.686
	R²	.098		.099		.114
	F for R²	1.356		1.350		1.544
Step 2	TMT shared leadership	.275*	.088	–.079	–.105	.251*	.086
	TPMCS			.623**	.178
	NPMCS					.356**	.115
	ΔR²	.074		.225		.146
	F for ΔR²	6.564*		11.963**		6.881**
Step 3	TMT sh. lead. × TPMCS			.124	.727
	TMT sh. lead. × NPMCS					.297*	.697
	ΔR²			.008		.067
	F for ΔR²			.883		6.910*
	Overall R² (Overall adj. R²)	.171(.093)		.332(.247)		.327(.239)
	Overall F for R²	2.186*		3.917**		3.725**
	VIF value, highest	1.283		2.715		1.310

Notes: ^aUnstandardized coefficient. **p < .01; *p < .05; ^†p < .10.

In Model 1, we tested Hypothesis 1, inserting control variables in step 1 and the independent variable (TMT shared leadership) in step 2. In line with Hypothesis 1, the results indicate a significant positive relationship between TMT shared leadership and ambidexterity (Std. β = .257, p < .05).

In Model 2, we tested Hypothesis 2, inserting control variables in step 1 and the independent variable (TMT shared leadership) and the moderating variable (TPMCS) in step 2. We inserted the multiplication effect between the standardized scores (to minimize the multicollinearity common in moderation modelling) of the independent and moderating variables in step 3. The results indicate that TPMCS does not moderate the relationship between TMT shared leadership and ambidexterity, so we found no support for Hypothesis 2.

In Model 3, we tested Hypothesis 3, inserting the control variables in step 1 and the independent variable (TMT shared leadership) and the moderating variable (NPMCS) in step 2. We inserted the multiplication effect between the standardized scores of independent and moderating variables in step 3. The results indicate, in line with Hypothesis 3, that NPMCS has a significant positive moderating effect on the relationship between TMT shared leadership and ambidexterity. The plot of the interaction can be found in Figure 1.

Figure 1.

Interaction plot – NPMCS.

Both regressions in Models 2 and 3 (see step 2 in Table 2) indicate that TPMCS and NPMCS have a strong positive direct relationship with ambidexterity, indicating that while representing different logics, these control systems might encourage the fine balance between exploration and exploitation. Furthermore, Models 1 and 3 (see step 2 in Table2) suggest that TMT shared leadership is related to ambidexterity, independently and in combination with NPMCS, indicating the importance of TMT shared leadership when new systems such as NPMCS are being implemented. Yet, in Model 2, only TPMCS (step 2 in Table 2) explains variation in ambidexterity – TMT shared leadership is not significant. The latter might indicate that TPMCS makes management less important and ‘drives’ ambidexterity on its own. In summary, our analysis indicates support for Hypothesis 1 and for Hypothesis 3, but no support for Hypothesis 2.

Discussion

Our study examined how TMT shared leadership affects organizational ambidexterity in public-sector organizations and sought to understand the moderating role of these organizations’ management control systems. Our study demonstrated that TMT shared leadership is positively related to organizational ambidexterity in public-sector organizations, in line with recent findings in private-sector organizations (e.g. Mihalache et al., 2014). The study also found that team efforts matter in public contexts and that the tensions between exploratory and explicatory orientations (e.g. Carmeli and Halevi, 2009) can be resolved because these efforts are embedded in the joint ability to exert influence (cf. Ensley et al., 2006). This study contributes to emerging research into ambidexterity and is applicable in public-sector organizations (e.g. Kobarg et al., 2017), where ambidexterity is said to represent an alternative way of exploring and operationalizing organizational outcomes (Cannaerts et al., 2016).

This study builds our understanding of the conditions under which shared leadership in public-sector organizations effectively achieves ambidexterity. It answers calls to explore the circumstances (Raisch and Birkinshaw, 2008) under which management antecedents matter for ambidexterity. While we found no support for the proposed negative moderating role of TPMCS, we did find support for the positive moderating effect of NPMCS on the relationship between TMT shared leadership and ambidexterity. The type of organization we studied is apparently guided by multiple purposes (Rainey and Steinbauer, 1999), and NPMCSs are a positive force in such settings given municipal corporations’ openness to such systems. One explanation might lie in the higher risk-taking behaviour demonstrated by management in municipal corporations, compared to that of ‘purely’ public entities (Smith and Umans, 2015). That propensity for risk might positively interact with performance rewards based on NPM-oriented systems, resulting in higher organizational effectiveness in resource allocation (Kim, 2010). Our findings are thus aligned with Hoch and Kozlowski’s (2014) idea that efficiency and uncertainty, being properties of NPMCS in public-sector organizations, require TMT shared leadership to translate them into an effective use of resources. Our findings also suggest that shared rather than individual leadership in combination with newly introduced systems (Morgeson et al., 2010) in the public sector might be the way forward in creating ambidextrous public entities. The negative moderating role of TPMCS found indicates a possible reinforcing, rather than a misaligned (Wang et al., 2014), relationship between TPMCS and TMT shared leadership. However, our analysis indicates that TPMCS directly and positively affects ambidexterity, indicating not only reinforcement, but also that TPMCS may be ingrained in public-sector organizations, constituting a force in itself.

The present results make several theoretical and practical contributions to the literature. Our study contributes to the strategic leadership literature by providing theoretical insight and empirical evidence that TMT shared leadership is important for public organizations, positioning shared leadership as a TMT-level construct. Our findings thus add to emergent research exploring the role of managers in public organizations by borrowing from strategic management literature (Modell, 2012). We further contribute to the field by illustrating how TMT shared leadership interacts with management control systems in facilitating organizational ambidexterity. Specifically, we do this by empirically illustrating how NPM-inspired management control systems amplify the positive effect of TMT shared leadership on organizational ambidexterity, thus contributing to the research exploring how public organization structures interact with management characteristics (cf. Karl and Sutton, 1998; Kim, 2010). The study complements the limited research into public-sector organizational ambidexterity (e.g. Aargard, 2011; Smith and Umans, 2015) by theoretically and empirically deriving ambidexterity facilitators. Finally, the study identifies two management control systems in public-sector organizations, contributing a new, yet theoretically derived (cf. Osborne and Gaebler, 1992; Speklé and Verbeteen, 2014), typology that facilitates understanding of organizational functioning. The results also inform practice because of the indications that balancing exploration and exploitation is a management-team function and that interactions between TMT members, especially in sharing leadership, are decisive. The study further indicates that introducing reward- and performance-oriented management control systems could enhance leaders’ joint ability to create ambidexterity, contributing to research exploring the emergence of specific leadership structures in organizations (e.g. Morgeson et al., 2010).

Conclusions

Our study offers two overarching conclusions. First, it suggests that the contingency perspective remains highly relevant for public management literature, yet the concepts available within the field might need to be borrowed from adjustment fields or re-conceptualized in order to gain new insights. This study has followed Chandler’s (1990) idea of fit between strategy-related concepts (TMT shared leadership) and structure-related concepts (public-sector-specific management control systems) in relation to organizational outcome – organizational ambidexterity. By doing so, this study not only illuminates and confirms some theoretically hypothesized relations, but also offers new insight into the functioning of public organizations and how this can be explored.

Second, we can conclude that municipal corporations represent a fruitful context for testing the application of new constructs developed and/or borrowed from other fields (cf. Smith et al., 2018). Possessing properties of both private and public organizations (Smith and Umans, 2015), these entities might represent an ideal setting where the transferability of concepts from private to public contexts, and vice versa, can be explored. Inspired by previous research on ambidexterity in private corporations, this study highlights its applicability in the public sector and uncovers its stimuli. Similarly, our findings on contingencies, represented by dual management control systems, can potentially be transferred to research on private corporations, where one could re-conceptualize traditional and new public management control systems into conceptually close shareholder- and stakeholder-oriented management control systems.

Limitations and future research

This study has several limitations. The small sample and having only one respondent from each organization represent major limitations. We tried to alleviate this by sending the survey to all TMT members in the sampled companies and by sending several reminders. Our study might further suffer from respondent bias, which was partly addressed by performing statistical bias testing, though we cannot fully disregard this threat to validity. Another limitation is the lack of information on TMT human capital: it was impossible to collect relevant data because only limited information is available on TMT members in public-sector organizations. Finally, the constructs used came from the literature on private, not public-sector, firms; this limitation especially applies to the construct measuring ambidexterity, where explorative and exploitative innovation sub-constructs were used. By adjusting the instruments and terminology to better fit the public-sector context, we addressed this limitation.

Despite its limitations, our article serves as a starting point for exploring the relationship between public-sector organizational characteristics and the emerging phenomenon of shared leadership, specifically, its effect on various organizational outcomes. For example, future studies could consider the impact of TMT composition on ambidexterity in public-sector organizations. Further studies could elaborate on the roles of management control systems, for example, developing a more detailed instrument for measuring process, performance and public value (Spano, 2009). Finally, future studies could deconstruct ambidexterity by exploring the aspects of TMT leadership that matter in explorative versus exploitative orientations given that some organizations might pursue one or the other as their main strategy.

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship and/or publication of this article: This research was partly funded by the Torsten Söderberg Foundation.

Timurs Umans was an Associate Professor in Corporate Governance at Kristianstad and Linnaeus Universities, Sweden. He is now Senior Associate Professor in Accounting and Management Control, Jönköping International Business School, Jönköping University (Sweden). His research interests circle around top management team functioning in both public- and private-sector organizations.

Elin Smith is an Associate Professor in Corporate Governance at Kristianstad University, Sweden. Her research interests circle around the governance of public- and private-sector organizations.

William Andersson is a former master’s student in the Accounting and Auditing programme at Kristianstad University, Sweden.

William Planken is a former master’s student in the Accounting and Auditing programme at Kristianstad University, Sweden.

References

Aargard

P (

2011) Organizational ambidexterity: How to be both innovative and efficient in the public sector. Working paper No. 5/2011, Roskile University.

Benner

Tushman

ML (

2003) Exploitation, exploration, and process management: The productivity dilemma revisited. Academy of Management Review 28: 238–256.

Boland

Fowler

A (

2000) A systems perspective of performance management in public sector organisations. International Journal of Public Sector Management 13(5): 417–446.

Bonner

Sprinkle

G (

2002) The effects of monetary incentives on effort and task performance: Theories, evidence, and a framework for research. Accounting Organizations and Society 27(1): 303–345.

Bozeman

B (

1987) All Organizations Are Public: Bridging Public and Private Organizational Theories. San Francisco, CA: Jossey-Bass.

Cannaerts

Segers

Henderickx

E (

2016) Ambidextrous design and public organizations: A comparative case study. International Journal of Public Sector Management 29(7): 708–724.

Carmeli

Halevi

MY (

2009) How top management team behavioral integration and behavioral complexity enable organizational ambidexterity: The moderating role of contextual ambidexterity. The Leadership Quarterly 20: 207–218.

Chandler

AD (

1990) Strategy and Structure: Chapters in the History of the Industrial Enterprise ( vol. 120). Cambridge, MA: MIT Press.

Choi

Chandler

SM (

2015) Exploration, exploitation, and public sector innovation: An organizational learning perspective for the public sector. Human Service Organizations: Management, Leadership & Governance 39(2): 139–151.

10.

Das

TK (

1995) Managing strategic flexibility: Key to effective performance. Journal of General Management 20(3): 60–75.

11.

Duncan

(1976) The ambidextrous organization: Designing dual structures for innovation. In: Kilmann

Pondy

Slecin

(eds) The Management of Organization ( vol. 1). New York, NY: North-Holland, pp. 167–188.

12.

Ensley

Hmieleski

Pearce

CL (

2006) The importance of vertical and shared leadership within new venture top management teams: Implications for the performance of startups. The Leadership Quarterly 17(3): 217–231.

13.

Fernandez

Cho

Perry

JL (

2010) Exploring the link between integrated leadership and public sector performance. The Leadership Quarterly 21(2): 308–323.

14.

Gerdin

Greve

J (

2004) Forms of contingency fit in management accounting research – a critical review. Accounting, Organizations and Society 29(3): 303–326.

15.

Halevi

Carmeli

Brueller

NN (

2015) Ambidexterity in SBUs: TMT behavioral integration and environmental dynamism. Human Resource Management 54(S1): 223–238.

16.

Hoch

Kozlowski

SW (

2014) Leading virtual teams: Hierarchical leadership, structural supports, and shared team leadership. Journal of Applied Psychology 99(3): 390.

17.

Hood

C (

1991) A public management for all seasons. Public Administration 69(1): 3–19.

18.

Jansen

Van den Bosch

Volberda

H (

2006) Exploratory innovation, exploitative innovation, and performance: Effects of organizational antecedents and environmental moderators. Management Science 52(11): 1661–1674.

19.

Jansen

Vera

Crossan

M (

2009) Strategic leadership for exploration and exploitation: The moderating role of environmental dynamism. The Leadership Quarterly 20(1): 5–18.

20.

Karl

Sutton

CL (

1998) Job values in today’s workforce: A comparison of public and private sector employees. Public Personnel Management 27(4): 515–527.

21.

Kellough

Selden

SC (

1997) Pay-for-performance systems in state government perceptions of state agency personnel managers. Review of Public Personnel Administration 17(1): 5–21.

22.

Kim

J (

2010) Strategic human resource practices: Introducing alternatives for organizational performance improvement in the public sector. Public Administration Review 70(1): 38–49.

23.

Kobarg

Wollersheim

Welpe

, et al. (2017) Individual ambidexterity and performance in the public sector: A multilevel analysis. International Public Management Journal 20(2): 226–260.

24.

Lind

H (

2001) Den nya bostadspolitiken (Uppl. 1) [The New Housing Policy (1st edn)]. Umeå: Boréa bokförlag [Boréa Publishing].

25.

Lubatkin

Simsek

Ling

, et al. (2006) Ambidexterity and performance in small- to medium-sized firms: The pivotal role of top management team behavioral integration. Journal of Management 32(5): 646–672.

26.

Malmi

Brown

D (

2008) Management control systems as a package – opportunities, challenges and research directions. Management Accounting Research 19(1): 287–300.

27.

Manz

and

Sims

Jr (1993) Business Without Bosses. New York, NY: Wiley.

28.

March

JG (

1991) Exploration and exploitation in organizational learning. Organization Science 2: 71–87.

29.

Merchant

Van der Stede

WA (

2007) Management Control Systems: Performance Measurement, Evaluation and Incentives. Harlow: Pearson Education.

30.

Mihalache

Jansen

Van den Bosch

, et al. (2014) Top management team shared leadership and organizational ambidexterity: A moderated mediation framework. Strategic Entrepreneurship Journal 8(2): 128–148.

31.

Modell

S (

2012) Strategy, political regulation and management control in the public sector: Institutional and critical perspective. Management Accounting Research 23: 278–295.

32.

Morgeson

DeRue

Karam

EP (

2010) Leadership in teams: A functional approach to understanding leadership structures and processes. Journal of Management 36(1): 5–39.

33.

Nelissen

Denhardt

Lako

CJ (

2000) The pursuit of significance: A survey among top managers of public administration in The Netherlands. Public Management: An International Journal of Research and Theory 2(2): 219–238.

34.

Osborne

Gaebler

T (

1992) Reinventing Government: How the Entrepreneurial Spirit is Transforming the Public Sector. Reading, MA: Addison-Wesley Publishing Company.

35.

Pallant

(2013) SPSS Survival Manual: A Step by Step Guide to Data Analysis Using IBM SPSS (5th edn). Berkshire: McGraw-Hill Education.

36.

Palm

Lilja

J (

2017) Key enabling factors for organizational ambidexterity in the public sector. International Journal of Quality and Service Sciences 9(1): 2–20.

37.

Pasquier

Villeneuve

JP (

2007) Organizational barriers to transparency: A typology and analysis of organizational behaviour tending to prevent or restrict access to information. International Review of Administrative Sciences 73(1): 147–162.

38.

Planken

Andersson

W (

2015) Top management teams’ shared leadership and control systems: Ambidexterity in municipal housing corporations. Master’s thesis, Kristianstad University.

39.

Podsakoff

Scott

MacKinzie

, et al. (2003) Common method biases in behavioral research: A critical review of the literature and recommended remedies. Journal of Applied Psychology 88: 879–903.

40.

Rainey

Steinbauer

P (

1999) Galloping elephants: Developing elements of a theory of effective government organizations. Journal of Public Administration Research and Theory 9(1): 1–32.

41.

Raisch

S and

Birkinshaw

J (

2008) Organizational ambidexterity: Antecedents, outcomes, and moderators. Journal of Management 34(3): 375–409.

42.

Reimann

BC (

1973) On the dimensions of bureaucratic structure: An empirical reappraisal. Administrative Science Quarterly 18(4): 462–476.

43.

Simsek

(2009) Organizational ambidexterity: Towards a multilevel understanding. Journal of Management Studies 46: 597–624.

44.

Smith

Umans

T (

2015) Organizational ambidexterity at the local government level: The effects of managerial focus. Public Management Review 17(6): 812–833.

45.

Smith

Umans

Thomasson

A (

2018) Stages of PPP and principal–agent conflicts: The Swedish water and sewerage sector. Public Performance & Management Review 41(1): 100–129.

46.

Smith

Tushman

ML (

2005) Managing strategic contradictions: A top management model for managing innovation streams. Organization Science 16: 522–536.

47.

Spano

A (

2009) Public value creation and management control systems. International Journal of Public Administration 32(3/4): 328–348.

48.

Speklé

Verbeteen

FHM (

2014) The use of performance measurement systems in the public sector: Effects on performance. Management Accounting Research 25: 131–146.

49.

Tuan

TL (

2017) Reform in public organizations: The roles of ambidextrous leadership and moderating mechanisms. Public Management Review 19(4): 518–541.

50.

Uhl-Bien

Marion

R (

2009) Complexity leadership in bureaucratic forms of organizing: A meso model. The Leadership Quarterly 20(4): 631–650.

51.

Umans

T (

2013) Top management team’s cultural diversity and firm performance: The mediating role of ambidextrous orientation. Corporate Ownership & Control 11(1): 882–891.

52.

Vázquez

JMS

Elorza

MLV

Jerónimo

JMR (

2006) Control problems in distribution channels: Empirical evidence of management control systems contributions. Revista de Contabilidad – Spanish Accounting Review 9(18): 11–29.

53.

Wang

Waldman

Zhang

Z (

2014) A meta-analysis of shared leadership and team effectiveness. Journal of Applied Psychology 99(2): 181–198.