Pressure, Legitimacy and Innovative Behavior by Public ... · Pressure, Legitimacy and Innovative Behavior by Public Organizations* KOEN VERHOEST, BRAM VERSCHUERE, and GEERT BOUCKAERT

JOBNAME: No Job Name PAGE: 1 SESS: 54 OUTPUT: Thu Jun 21 11:00:03 2007 SUM: 70F337C0/v2451/blackwell/journals/gove_v20_i3/gove_367

Pressure, Legitimacy and Innovative Behavior byPublic Organizations*

KOEN VERHOEST, BRAM VERSCHUERE, and GEERT BOUCKAERT

According to New Public Management (NPM) doctrines, public organi-zations involved in service delivery and policy implementation will beinduced to innovative behavior if they have enough managerial autonomyand simultaneously are subjected to managerial pressure, such as resultcontrol by government or competition of other providers. This NPMpressure-response model is tested by using survey data on 84 Flemishpublic organizations. These tests provide evidence for the assumed effect ofNPM-like pressure on the innovative behavior of public organizations.However, the empirical model shows more complex relationships as isassumed by NPM doctrine. These complex relationships are corroboratedand explained by making reference to a multiple-case study of four Flemishpublic organizations. An expanded political/administrative pressure-response model, referring to legitimacy as a motivational force, is suggestedin order to explain innovative behavior by public organizations. This modelmay help to understand the preconditions for spontaneous adaptation ofpublic organizations.

New Public Management (NPM) and Innovative Behavior of PublicOrganizations

NPM, a set of administrative doctrines which emerged in the 1980s (Hood1991), was a reaction to the perceived lack of result- and customer-orientedness of public organizations that delivered services to the public orimplemented policy. One aspect was the perceived inertia of governmentbureaucracies and their lack of innovative behavior as to the adoption ofmanagerial techniques and the development of processes and products.According to NPM doctrines, these problems stemmed from “governmentfailures” referring to the inability of centralized, hierarchical organizedsystems to adapt to the fast-changing environment and to deliver servicesefficiently and effectively (Walsh 1995). The focus of NPM strategies waslargely aimed at establishing alternative forms of policy implementationand service delivery. These forms aimed at operational decentralizationwithin a context of strategic (re)centralization (cf. Hoggett’s 1996 notion of

*Catholic University Leuven

Governance: An International Journal of Policy, Administration, and Institutions, Vol. 20, No. 3,July 2007 (pp. 469–496).© 2007 The AuthorsJournal compilation © 2007 Blackwell Publishing, 350 Main St., Malden, MA 02148, USA,and 9600 Garsington Road, Oxford, OX4 2DQ, UK. ISSN 0952-1895

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

JOBNAME: No Job Name PAGE: 2 SESS: 54 OUTPUT: Thu Jun 21 11:00:03 2007 SUM: 4D500864/v2451/blackwell/journals/gove_v20_i3/gove_367

“centralized-decentralization”; Aucoin 1990; Campbell and Halligan 1992).In particular, the execution of tasks was supposed to be devolved tosingle-objective organizations with expanded managerial autonomy,which are controlled by the government on the basis of result norms (e.g.,“agencies” as defined by Pollitt et al. (2004). The practice of “let the man-agers manage” aimed at more managerial autonomy as a vehicle for betterperformance. Controlling public organizations on results has been intro-duced as an instrument for putting organizations under certain kinds of(internal) pressure, another vehicle for performance and innovation. More-over, this contract-based service delivery of goods and services had to beoptimized by the introduction of competition—as an instrument of externalpressure—by other public or private suppliers, whereever such competi-tive markets are possible. The ultimate aim of these practices—all gatheredunder the label NPM, although different in focus and scope—was to createmore innovative and, hence, better-performing public organizations by themeans of “disaggregation, competition and incentivization” (Dunleavy1997). Public management scholars observe indeed a shift in many Organi-sation for Economic Co-operation and Development (OECD) countriesto policy implementation and service delivery regimes with semi-autonomous public organizations figuring under increased levels of inter-nal and external pressures (OECD 2002; Pollitt et al. 2004). The migrationtrajectories differ, however, between countries (Bouckaert 2003; OECD1997; Verhoest and Bouckaert 2005).

The central question in this article is whether or not the combination ofmanagerial autonomy on the one hand, and managerial and market-likepressures, such as result control and competition on the other hand,induces public organizations to innovative behavior, as is assumed byNPM doctrines? Or, are other triggering factors into play?

Moreover, this article may help shed some light on the issue of spon-taneous adaptation by public organizations and its conditions. Innovativebehavior can be considered to be practices of spontaneous adaptation (seeCampbell in this Volume). As Campbell argues, spontaneous adaptation isfostered by latitude and autonomy of public managers. However, Camp-bell questioned the impact of an overemphasis on outputs (instead ofoutcomes) in the control by government, as well as the recentralization ofdiscretionary powers by recent Joined-up Government initiatives. In thisarticle, we study the impact of balancing autonomy and result control oninnovation and, hence, touch upon some of these issues raised by Camp-bell. We will provide evidence that organizational legitimacy is a crucialmotivator for spontaneous adaptation in the form of innovative behavior,and that autonomy and result control may be stimulating factors.

Figure 1 shows the basic argument of NPM as to the ways to stimulatepublic organizations for innovation.

The NPM argument thus boils down to two conditions for innovativebehavior, referring to the managerialist approach and the public choiceapproach (Aucoin 1990).

470 KOEN VERHOEST, BRAM VERSCHUERE, AND GEERT BOUCKAERT

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

42

43

44

45

46 22

JOBNAME: No Job Name PAGE: 3 SESS: 54 OUTPUT: Thu Jun 21 11:00:03 2007 SUM: 3ACC395E/v2451/blackwell/journals/gove_v20_i3/gove_367

First, “Let public managers innovate,” by giving them enough manage-rial autonomy to develop new ways of processing and handling theirbusiness. We define autonomy as the level of decision-making competen-cies (discretion) of an organization, whereas control focuses upon theconstraints, which the ministers/departments exert on the actual use ofthese decision-making competencies in order to influence the decisionsmade. Making a public organization more autonomous involves shiftingdecision-making competencies from external actors to the agency itself bydelegation, devolution, or decentralization. Managerial autonomy concernsdecision making as to the choice and use of financial, human and otherresources at strategic or operational level (for a discussion of the conceptof autonomy with respect to public organization, see Christensen 2001;Verhoest et al. 2004). Managerial autonomy enables and facilitates innova-tion, but does not induce public managers in itself to innovation. There-fore, pressure is needed.

Second, “Make public managers innovate” by putting managerial andmarket-like pressures on them (OECD 1997). This element was stressedquite strongly in NPM-like reforms in OECD countries. Pressure can becreated internally in the public sector by means of harsh performancestandards in performance contracts with government, to which the publicorganization is held accountable and to which sanctions or rewards arelinked. Sanctions and rewards function as drivers for performance and, asa condition for performance, for innovation because public organizationsand their managers are assumed to be self-interested actors (Boorsma andHalachmi 1998; Schick 1996). This process by which the political principalof a public organization sets the objectives and corresponding perfor-mance standards, monitors and evaluates the progress of the organization

FIGURE 1An NPM Pressure-Response Model with Respect to Innovations inPublic Management

Perceived problems as topublic organizations:

- inertia- lack of innovation- lack of customerorientedness

as elements of a broaderlack of performance

Enabling public organizations toinnovate by enlarging managerial

autonomy

Forcing public organizations toinnovate by:- internal pressures as resultcontrol by government- external pressures bycompetition of other providers

Public organizations are:

- more dynamic- more innovative- more customer oriented

as elements of a broaderimprovement ofperformance

PROBLEM NPM SOLUTIONS EFFECTS

PRESSURE, LEGITIMACY AND INNOVATIVE BEHAVIOUR 471

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

m100

1

JOBNAME: No Job Name PAGE: 4 SESS: 54 OUTPUT: Thu Jun 21 11:00:03 2007 SUM: 4E46AD57/v2451/blackwell/journals/gove_v20_i3/gove_367

with respect to these standards, and, if necessary, sanctions the organiza-tion or its managers will be referred to in this article as ‘result control.’

Governments can create external pressures from outside the publicsector by introducing competition by other suppliers in the form ofmarket-type mechanisms (Le Grand and Bartlett 1993; Walsh 1995). Thatmarket forces and, in particular competition, functions as a major spur forinnovation is an argument that is strongly developed within the literatureon innovation in private for-profit firms (Nelson 1993, 364).

Likewise, introducing market mechanisms and competition in thepublic sector is assumed to exert a downward pressure on costs of ser-vices; to increase the responsiveness of public agencies toward the serviceuser; and to stimulate action to ensure survival, such as innovative behav-ior (Common, Flynn, and Mellon 1992, 33). Not only present competitionby other suppliers, but also the threat of future competition, or potentialcompetition is considered to revitalize public organizations, for instancewhen introducing competitive tendering or market testing mechanisms(Common, Flynn, and Mellon 1992). Empirical evidence that introducingcompetition enhances performance, in particular efficiency of public orga-nizations is available, although not unequivocally (Fölster 1997; Hodge1996 v. Van Thiel 2002). However, research on the influence of competitionon innovation of public organization is rather scarce. It can be hypoth-esized that in competitive environments, public organizations are forcedto seek and adopt innovations, aiming for comparative advantages inrelation to (potential) competitors in order to achieve high-level perfor-mance, to assure the current organizational position and status.

In general, as Stephen B. Osborne (1998) states: “It [the literature oninnovative capacity of the public organizations and the voluntary sector]has thus been in terms of how to make them more like for-profit organi-zations, and consequently as more likely to be innovators” (Osborne 1998,159).

In the second part of this article, we will test this NPM assumption withdata of public sector organizations in Flanders (Belgium). The findings areexplained by referring to a multiple-case research (on a selection of casesfrom this survey) in the third part. The last part discusses this explanationwith reference to other theories and sets some future lines for research.

Testing the NPM Model of Autonomy, Pressure, and Innovation

Operationalization of the Model

Figure 2 models in a simple causal scheme the NPM argument as to theinfluence of managerial autonomy, result control, and (potential) compe-tition on the adoption of innovations or innovative behavior by the publicorganization. Managerial autonomy is an enabler and a facilitator and,hence, is to be considered as having a moderating influence on the effectof NPM-like pressures (i.e. result control and potential competition) oninnovative behavior by public organizations.


1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

42

43

44

45

JOBNAME: No Job Name PAGE: 5 SESS: 54 OUTPUT: Thu Jun 21 11:00:03 2007 SUM: 3F8E107F/v2451/blackwell/journals/gove_v20_i3/gove_367

This model is tested using data of 84 public sector organizations that arepart of the central government in the Flemish Community. In 2002 and 2003,a total of 152 organizations were subject to a survey of top managers of theseorganizations on their perception of autonomy, control, and related perfor-mance and innovative behavior, among several other items. The organiza-tions that responded to the survey are involved in policy implementationand service delivery. They belong to five different formal-legal types, whichare classified in an increasing distance from the core government:

Information about the methodology of data gathering and analysis is tobe found in the Appendix. As to the measurement of the level of manage-rial autonomy, our previous research (Verhoest et al. 2004) has extensivelyshown that there is no straightforward link between the degree ofautonomy and the types in Table 1. Hence, we do not use this typology asan operationalization. Rather, we operationalize managerial autonomy bymeasuring the level of operational managerial autonomy as to financialmanagement (OFA1 and OFA2) on the one hand and personnel manage-ment (OPA) on the other hand by several indicators.

Result control by government is operationalized quite narrowly by mea-suring the extent to which the public organizations are subject to sanctionsor rewards in case of bad, respectively good performance (RES). Thesesanctions or rewards may be oriented toward the organization as a wholeor to (some of) its individuals.1 The level of potential competition by otherproviders was measured by the presence of other suppliers with similarproducts and services on the market with which the public organization isactive (COM).

The extent to which the public organization adopts innovative practicesis measured by the presence and use within the public organization ofone specific kind of innovation. A general and exhaustive definition of

FIGURE 2An NPM Pressure-Response-Model of the Influence of ManagerialAutonomy, Result Control and (Potential) Competition on InnovativeBehavior by the Public Organizations

Managerialautonomy ofthe public organisation

Result control bygovernment

Competition byother suppliers

Innovativebehavior by thepublic organisation

+

+

+

moderating influence

+


1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

33


innovation is proposed by Osborne as “the introduction of newness into asystem usually but not always, in relative terms and by the application(and occasionally invention) of a new idea. This produces a process oftransformation which brings about a discontinuity in terms of the subjectitself (such as a product or service) and/or its environment (such as anorganization or society)” (Osborne 1998, 24; cf. Alter 2002).

In our survey, we asked top managements of public organizations towhat extent they used specific performance techniques. These techniquesrefer to specific instruments of performance management, quality manage-ment, and personnel management as well as to the development of newproducts and services. Possibly, the introduction of specific managementinstruments can be considered as an innovation when taking the classicalbureaucracy as default position and looking at the whole population ofpublic organizations in the Flemish context. However, such instrumentscan be long-standing practice within the public organization concernedand, hence, not to be labeled as an innovative practice for this organization.Therefore, we restrict our analyses to the performance techniques, whichrefer to pure innovation sensu stricto, that is ‘the development of new productsand services’ by the public organizations (NewSP). This performance tech-nique fits most neatly with the definition of innovation that we use. More-over, although the introduction of new managerial instruments mayameliorate performance of a public organization, in most cases, probablyonly the development of new products and services may cause sustainableperformance leaps. A last motivation to select only this specific perfor-mance technique out of the range we have surveyed, is that the presence ofother performance techniques in public organizations can be caused rela-tively more by imitating strategies, by government-wide trends or norms ofgood management, or is relatively more likely to be contingent upon factors

TABLE 1Five Types of Organizations and N in Sample

Type N

1: Core government organizations that are part of the legal person ofthe Flemish community, and that are in the budget of the coredepartment to which they belong

27

2: Internally autonomized agencies with an own budget (somemanagerial discretion), no legal personality.

9

3: Internally autonomized agencies with an own budget (somemanagerial discretion), legal personality.

11

4: Externally autonomized agencies with an own budget, public lawlegal personality and a governing board (indicative for decreasedministerial oversight).

25

5: Externally autonomized agencies with an own budget, private lawlegal personality and a governing board (indicative for decreasedministerial oversight).

12

Total 84


12

3456789

1011121314151617

1819

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

42

43

44

45

46

47

48

JOBNAME: No Job Name PAGE: 7 SESS: 54 OUTPUT: Thu Jun 21 11:00:03 2007 SUM: 3FC87064/v2451/blackwell/journals/gove_v20_i3/gove_367

such as the kind of product and the size of the organization. On the otherhand, however, there are also limitations to the use of this specific perfor-mance technique, that is, ‘the development of new products and services’ as anindicator for innovation by the public organization.An important issue hereis whether or not the development of innovative services and products isinduced from within the organization itself or by their political leaders(e.g., the Barzelay and Campbell [2003] case study on U.S. Air Force). In thelatter case, the issue of innovation as a form of spontaneous adaptation ismuted, and new products and services would only be developed whenthere is a pressure for adaptation from political leadership. The availablesurvey data do not allow us to distinguish between innovations stemmingfrom independent behavior of the public organization or from interventionof its political leaders. However, making this distinction is possible in thecase study research of four public organizations that will be discussedfurther in the succeeding discussions.

In order to test the model represented in Figure 3, we pursued a three-level strategy. First, we performed ordinal regression for the dependentvariable NewS&P, with OPA, OFA1, OFA2, COM and RES as independentvariables. The relevant relations were tested for spuriousness, moderatingand mediating effects by bivariate cross-tabulations, with and withoutcontrolling for third variables. This enables us to detect interaction effectsof independent variables on the relationship between other independentvariables and the dependent variable. In the Appendix, more details aboutthe survey, the operationalization of the concepts, the subsequent analyti-cal steps, and their outcomes are included.

Findings: Innovation as a Function of Autonomy, ResultControl and Competition

If we investigate the relationship between autonomy, result steering and(potential) competition on the one hand, and the development of newproducts and services on the other hand, some observations may be made(for details, see Appendix).

FIGURE 3The Influence of OPA, OFA1, OFA2, COM and RES on NewS&P

RES

COM

OPA

NewSP

A

B

C

E

D


1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

JOBNAME: No Job Name PAGE: 8 SESS: 54 OUTPUT: Thu Jun 21 11:00:03 2007 SUM: 4C431E34/v2451/blackwell/journals/gove_v20_i3/gove_367

At a first basic level, there is a positive and significant effect of havinghigh levels of OPA, the presence of sanctions and rewards (RES) andpotential competition (COM) on the development of innovative productsand services (NewSP) (A, B, and C on Figure 3. See also tables 2 and 3 inAppendix: correlation coefficients of respectively 0.30, 0.30, and 0.22,Regression estimates of respectively 1.2, 1.1, and 0.8). On a second level,interaction effects are found altering some of these relationships (table 4 inAppendix).

1. The relationship between potential competition and the develop-ment of innovative services and products (NewSP) is altered bycontrolling for the variable OPA: in the group of organizations thathave low levels of OPA a stronger effect of COM on NewSP is found,compared to the total sample (Figure 3D. See also table 4 in Appen-dix: correlation coefficient of 0.43 compared to total sample coeffi-cient of 0.22).

2. Next to that, the positive relationship between OPA and NewSP isaltered if it is controlled for potential competition. In the group oforganizations that face no other suppliers (no potential competition),the effect of OPA on NewSP is stronger than in the total sample(Figure 3. See also table 4 in Appendix: correlation coefficient of 0.45compared to total sample coefficient of 0.30).

When we confront the empirical model in Figure 3 with the theoreticalNPM model (see Figure 2), it is clear that the models differ on someimportant issues:

1. The influence of the variable “managerial autonomy,” measured byOFA1, OFA2, and OPA, is not restricted to one of a moderating kind.Managerial autonomy, in particular for operational decisions of per-sonnel management, seems to have a direct influence on the develop-ment of new services and products, independent of pressuretriggered by the presence of results control (RES) or potential com-petition (COM). In the NPM model, managerial autonomy wasassumed to be only an enabling factor for innovation, but not toenhance innovative behavior by the public organization in itself.However, other factors that are outside the model and that arerelated to OPA may induce the organization to innovate.

Note that managerial autonomy for financial management does not seemto have a direct effect on the development of new services and products,nor does it have a moderating effect. Probably, being able to make opera-tional decisions as to recruitment, evaluations, rewards, and promotion ofstaff is more important for innovating public managers than having mana-gerial flexibility with respect to financial management, that is, shiftingbudgets, taking loans, participations and setting tariffs. In public organi-


1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

42

43

44

45

JOBNAME: No Job Name PAGE: 9 SESS: 54 OUTPUT: Thu Jun 21 11:00:03 2007 SUM: 45893105/v2451/blackwell/journals/gove_v20_i3/gove_367

zations, personnel is typically the main source of production. Moreover, inorder to develop new services and products, one must be able to hire,motivate and promote the persons with the expertise needed.2

2. The mere existence of potential competition is sufficient for a directeffect on innovative behavior, which matches the theoretical NPMmodel.

3. Being subject to results control by the government, particularly sanc-tions or rewards in case of deviating performance, seems to stimu-late public organizations to innovate, as is assumed in the NPMmodel.

4. The NPM model assumes that a high level of managerial autonomywould enable or strengthen the effect of potential competition oninnovation. However, both the variables concerning managerialautonomy with respect to personnel management (high and lowlevels) and potential competition (present or absent) interact empiri-cally in an unexpected way. The effect of one factor on innovationincreases when the other factor is absent/low. However, the pres-ence of both variables does not necessarily increase the prevalence ofinnovation. The influence of each of both, independent of the othervariable, is quite similar in strength. Hence, the variables OPA andCOM seem to interact as potentially compensating leverages in theireffect on innovation by public organizations. Again, this points at adirect effect of managerial autonomy with respect to personnelautonomy on innovation, independent from the presence of poten-tial competitors. On the other hand, potential competition caninduce public organizations even when they have low levels ofmanagerial autonomy and, hence, flexibility with respect to person-nel management. This seems to imply that potential competition is astrong motivator for innovative behavior (certainly stronger thanresult control).

5. Moreover, the empirical model only explains part of the variance ofthe dependent variable, more exactly, about 25% (R2 = 0.25Nagelkerke, ordinal regression in Appendix). Therefore, the devel-opment of new products and services depend largely on otherfactors.

NPM does not provide an explanation for the direct and independentinfluence of managerial autonomy concerning personnel management oninnovation, nor for the unexpected interaction of potential competitionand managerial autonomy. Only the empirically observed effect of poten-tial competition and result control (as operationalized by being subject tosanctions and rewards or not) on innovation seems to fit the model. If themismatch between the theoretical model and the empirical findings isnot attributed to methodological deficiencies of the analysis, then the


1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

42

43

44


theoretical NPM model is not corroborated and the observed relationshipbetween the independent and the dependent variables should beexplained by other models or theories.

In the next sections, we first will discuss a multiple-case research thatsuggests a broader model of pressure, which seems to corroborate a morecomprehensive theoretical framework for the empirical findings than theNPM model.3 We then take a look at some broader theories in order toassess their potential explanatory power and their relationship to thetheoretical model that arises from the multiple-case research.

Explaining the Findings

Additional Multiple-Case Analysis

In an attempt to test some elements of NPM doctrines, one of the authorsanalyzed, in previous research (Verhoest 2002), the effect of managerialautonomy on the one hand, and of result control (including financialincentives) and competition as NPM-like pressure on the other hand, onthe performances. The case study research was on four public organizationswhich are included in the survey population, in particular agencies of type4, which are constituted by parliamentary decree and function under apublic law regime. They have a governing board that is consist of amixture of government and societal actors’ representatives. The powers ofthe Portfolio minister and the minister of Finance to intervene within thedecision making of these public organizations are strictly regulated by thestatutes of the organizations and corresponding legislation. Traditionally,the control system relied heavily on ex ante and input control, limiting tosome extent the managerial autonomy and flexibility of the agencies (VOI:Vlaamse Openbare Instellingen). In the period from 1992 to 2002, the Flemishgovernment changed its policy concerning the control of the VOIs from anex ante input orientation to an ex post result and market orientation(Verhoest 2005a; Verhoest et al. 2004). This was also the time scope of thefour case studies.

The research (Verhoest 2002) was a two-step multiple-case study, withan in-depth case study of two sections within one VOI as the first step. Thecausal model, which emerges from this first step, was then tested in threeadditional case studies of a more limited magnitude in order to find somefirst indications as to the wider external validity of this broadened causalmodel. Taking result control as a replication criterion, one additional casewas a literal replication of the in-depth case, whereas the other two caseswere theoretical replications (Yin 1994): Both the FES and VOI-A had aperformance contract with the Flemish Government; whereas the twoother VOIs under study (VOI-B and the Child Support Agency) had noperformance contract. The case study research was done by documentanalysis of legislation, documents related to control and market regulation(e.g., performance contracts, annual reports), historical analysis of deci-


1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

42

43

44

JOBNAME: No Job Name PAGE: 11 SESS: 54 OUTPUT: Thu Jun 21 11:00:03 2007 SUM: 46BA85ED/v2451/blackwell/journals/gove_v20_i3/gove_367

sions, data series analysis of performance data, and (semi-) structuredinterviews with senior managers of the VOIs, as well as with seniorofficials in the parent departments.

The independent variables “managerial autonomy,” “result control,”and “competition” were operationalized quite similarly to the analysis wedescribed earlier in this article. The independent variable under review inthis study was the performance of the public organization with respect tothe objectives as defined by the political principal (i.e., the spendingminister and the Flemish government). However, we analyzed, in addi-tion, the presence of the performance techniques as a precondition to goodperformance. Hence, we have descriptive data on the “development ofnew products and services” in the four public organizations.

Table 2 summarizes the findings for the four case studies. We willdiscuss shortly the main findings of this research by making reference tothe different case studies.

The in-depth case study analyzed the effects of managerial autonomyand NPM-like pressures on the performance within two divisions of theFlemish Employment Agency (FES), that is, the job brokerage division andthe vocational training division. The job brokerage division managed toachieve a considerable performance improvement during the period 1992to 1999. It developed new products and services, which were consideredas “revolutionary” and successful, even in international context. Theseinnovations were mainly based on online interactive technology and cor-responding changes in service delivery. The FES job brokerage initiatedthese innovations itself, independently from instructions of the spendingminister and the cabinet. The vocational training division only slightlyincreased its performance because it was only able to conduct some “incre-mental” and marginal changes to its product and service portfolio amongothers. The degree and success of “spontaneous adaptation” was clearlylarger for the job brokerage division compared to the vocational trainingdivision.

The in-depth case study showed that these different outcomes were tobe explained by three factors (see Verhoest 2005b for a detailed account ofthis case).

First, the divisions were induced to performance-enhancing and inno-vative behavior by means of managerial autonomy, in combination withmore result control, financial incentives and competition, but only if thesecontrol instruments are used in a well-balanced way and if certain condi-tions are taken into account. These conditions were mostly absent in thevocational training division.

Second, a more fundamental motivational force for performance-enhancing and innovative behavior was at stake in the FES divisions: theneed of the FES (divisions) for strengthening its legitimacy toward itscustomers and especially toward its political principals (in particular, min-isters, government, and parliament) in response to certain legitimacy-threatening factors or pressures. In the job brokerage division, this urge


1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

42

43

44

45

46

JOBNAME: No Job Name PAGE: 12 SESS: 54 OUTPUT: Thu Jun 21 11:00:03 2007 SUM: 4E0DE3C9/v2451/blackwell/journals/gove_v20_i3/gove_367

TAB

LE

2In

dep

end

ent

and

Dep

end

ent

Var

iab

les

inth

eIn

-Dep

thS

ingl

eC

ase

Stu

dy

and

inth

eT

hre

eA

dd

itio

nal

Cas

eS

tud

ies

VO

IM

anag

eria

lau

tono

my

Res

ult

cont

rol

Fina

ncia

lin

cent

ives

Com

peti

tion

Oth

erpr

essu

res

(on

legi

tim

acy)

Enh

anci

ngfa

ctor

sIn

nova

tion

Perf

orm

ance

impr

ovem

enta

FES-

Job

brok

.+

++

(+)

+(+

)+

+(+

)+

++

++

+(+

)+

+(e

xcep

tfo

req

uity

obje

ctiv

es)

FES-

voc.

trai

ning

++

(+)

++

++

++

(+)

–+

(+)

=/+

(exc

ept

for

equi

tyob

ject

ives

)FA

I-A

+(+

)+

(+)

+(+

)+

+(+

)+/

–+

+to

(+)

FAI-

B+

00

++

++

––

–+

+/–

Chi

ldSu

ppor

tA

genc

y+

(+)

00

++

++

(+)

++

++

+(+

)(e

xcep

tfo

req

uity

obje

ctiv

es)

Lege

nd:

Leg

itim

acy

thre

ats:

legi

tim

acy

thre

ats

evok

ing

stim

uli

for

beha

vior

alst

rate

gies

.In

nova

tion

:in

nova

tion

sco

ncer

ning

prod

ucts

and

proc

esse

s.E

nhan

cing

/in

hibi

ting

fact

ors:

fact

ors

whi

chen

hanc

esor

inhi

bits

the

rece

ptiv

enes

sfo

rst

imul

iof

perf

orm

ance

-enh

anci

ngbe

havi

or.

0,no

tpre

sent

;(+)

,pre

sent

toa

very

littl

eex

tent

;+,p

rese

ntto

alim

ited

exte

nt;+

(+),

pres

entt

oa

mod

este

xten

t;+

+,pr

esen

tto

aco

nsid

erab

leex

tent

;++

(+),

pres

ent

toa

very

cons

ider

able

exte

nt;+

++,

pres

ent

toa

max

imal

exte

nt,=

,bei

nghe

ldco

nsta

nt.?

,no

enou

ghda

tato

mak

eco

nclu

sion

s.a Pe

rfor

man

ceim

prov

emen

t:ex

clud

esth

ein

flue

nce

ofex

tern

alfa

ctor

sas

muc

has

poss

ible

.


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21

JOBNAME: No Job Name PAGE: 13 SESS: 54 OUTPUT: Thu Jun 21 11:00:03 2007 SUM: 4B5BEB25/v2451/blackwell/journals/gove_v20_i3/gove_367

clearly stemmed from the perceived risks, caused by the gradual liberal-ization of the job brokerage market and the presence of other providers. Inthe perception of the FES management, this evolution increased the risk ofa reduction or a restriction of transactions with the Flemish government inthe future. And, more important, it increased the risk of abolition orreform of the FES (division). The other providers on the job brokeragemarket were more and more perceived as alternatives for the FES and theymobilized political pressure on the principals of the FES (i.e., interestgroups, spending minister, the cabinet and members of parliament). Assuch, the threats stemming from the mere presence of these real andpotential competitors and their appeal on the political principalsincreased the motivation of the job brokerage division. The legitimacy ofthe vocational training division was mainly threatened not only by therivalry with other actors for the allocation of governmental subsidies, butalso by the lack of political consensus about the role of the FES vocationaltraining in a free market and its increased political salience because ofsocietal evolutions.

Moreover, it proved that performance-enhancing behavior was onlyone of the possible behavioral strategies that the FES divisions followed inorder to retain/restore legitimacy. Competition-inhibiting behavior, com-munication and image-building, and symbolic responsive behaviortoward the political principals were elements of other strategies. Further-more, the risks of potential competition pressed the FES to emphasize itseconomic objectives, rather than its social objectives, which was contraryto the will of its political principals (i.e., goal conflict).

Third, the vocational training division was considered to be less recep-tive for external stimuli for performance-enhancing and innovative behav-ior, stemming from NPM-like pressures or more political pressures bylegitimacy threats. Or, it could not react adequately because of certaininternal features, which affects the receptiveness for, or the capability of, apublic agency to react to such stimuli negatively. Table 3 lists these features.

The three additional case studies provided supporting evidence onthese three factors.

The VOI-A4 was a medium-sized organization, active in the exploitationand the management of traffic infrastructure. The VOI-A did improve itsperformance to a large to modest extent in the period 1992–1999, althoughno development of major new products or services was to be noted. Atleast four reasons are into play. First, the economic evolution and theproblems on competing traffic infrastructures counted for this perfor-mance record to a considerable extent. Second, the extent to which theNPM-like pressures enhanced the performance seems rather modest tosmall because of rather low norms in the performance contract andbecause of a rather low level of competition in the perspective of thecontextual factors. Third, the quest for continuing legitimacy in the eyes ofits political principals and its customers was also, for the VOI-A, a moti-vating factor because its continued existence was, in the past, questioned


1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

42

43

44

45

46

JOBNAME: No Job Name PAGE: 14 SESS: 54 OUTPUT: Thu Jun 21 11:00:03 2007 SUM: 4C4460A0/v2451/blackwell/journals/gove_v20_i3/gove_367

TAB

LE

3Fe

atu

res

wh

ich

Infl

uen

ceth

eR

ecep

tive

nes

sof

Exe

cuti

veP

ub

lic

Org

aniz

atio

ns

for

Per

form

ance

-En

han

cin

gS

tim

uli

(Bas

edon

Ver

hoe

st20

05b

)

Feat

ures

enha

ncin

g/in

hibi

ting

the

rece

ptiv

enes

sof

the

orga

niza

tion

for

stim

ulit

oin

crea

sepe

rfor

man

ceIn

flue

nce

onre

cept

iven

ess

orre

acti

onca

pabi

litie

s(b

etw

een

brac

kets

are

the

case

sst

udie

d)

(a)

The

capa

bilit

ies

and

visi

onof

the

top

man

agem

ent

and

the

gove

rnin

gbo

ard,

like

–a

clea

rst

rate

gic

visi

onho

ldby

the

man

agem

ent

and

gove

rnin

gbo

ard

Rat

her

posi

tive

toth

eex

tent

this

stra

tegi

cvi

sion

alig

nw

ith

the

obje

ctiv

esan

dst

rate

gic

visi

onof

the

supe

rvis

ory

gove

rnm

ent

(FE

S;C

SV)

–th

eca

paci

tyw

ith

resp

ect

tole

ader

ship

and

inte

rnal

stee

ring

that

isco

ntin

gent

wit

hth

ein

tern

alst

ruct

ure

and

cultu

reof

the

orga

niza

tion

and

its

envi

ronm

ent

Rat

her

posi

tive

(+–

FES;

CSV

;–V

OI-

B)

–th

eex

tent

ofin

tern

algo

alin

cong

ruen

cean

dro

leco

nfus

ion

wit

hin

the

gove

rnin

gbo

ard

Rat

her

nega

tive

(–V

OI-

B;–

FES)

(b)

The

inte

rnal

cont

rols

yste

mbe

twee

nth

eto

pm

anag

emen

tan

dth

esu

buni

ts,l

ike

–th

eex

tent

ofpo

licy

auto

nom

yan

d/

orle

gala

uton

omy

ofth

esu

buni

tsR

athe

rne

gati

ve(–

FES

voca

tion

altr

aini

ng;–

VO

I-B

)

–th

eex

tent

tow

hich

the

inte

rnal

cont

rols

yste

mha

sex

peri

ence

wit

hre

sult

cont

rol

Rat

her

posi

tive

(+–

FES;

–V

OI-

A)

–th

eex

tent

tow

hich

ther

eis

ast

rong

need

for

stee

ring

onco

nten

tR

athe

rne

gati

ve(–

FES

voca

tion

altr

aini

ng;–

VO

I-B

)–

the

exte

ntto

whi

chth

ere

are

suffi

cien

tst

eeri

ngm

eans

tost

eer

and

cont

rolt

hesu

buni

tsR

athe

rpo

siti

ve(F

ES;

–V

OI-

B)

(c)

The

orga

niza

tion

alcu

lture

,–

the

exte

ntto

whi

chth

ere

isa

cultu

reor

ient

edto

war

ds

prof

essi

onal

valu

esR

ahte

rne

gati

ve(–

FES

voca

tion

altr

aini

ng;–

VO

I-B

)

–th

eex

tent

tow

hich

ther

eis

are

lati

vely

hom

ogen

ous

“cor

pora

tecu

lture

”R

athe

rpo

siti

ve(F

ES

job

med

iati

on;V

OI-

A;C

SV;–

VO

I-B

)

(d)

Kin

dof

acti

vity

and

polic

yd

omai

n,lik

e–

the

exte

ntto

whi

ch(o

utpu

tsof

)ac

tivi

tyan

dpo

licy

are

mea

sura

ble,

stab

lean

dro

utin

e-lik

eR

athe

rpo

siti

ve(F

ES

job

med

iati

on;V

OI-

A;C

SV)

–th

eex

tent

tow

hich

the

orga

niza

tion

hold

sth

ein

stru

men

tsto

stru

ctur

epo

licy

impl

emen

tati

onit

self

(e.g

.,se

rvic

ed

eliv

ery

byth

em

eans

ofow

npe

rson

nelr

athe

rth

ansu

bsid

izin

got

her

serv

ice

del

iver

ers)

Rat

her

posi

tive

(CSV

;–V

OI-

B)


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35

JOBNAME: No Job Name PAGE: 15 SESS: 54 OUTPUT: Thu Jun 21 11:00:03 2007 SUM: 483BC51E/v2451/blackwell/journals/gove_v20_i3/gove_367

and reintegration in the core administration or a merger with anotheragency emerged as alternatives. Rather than pursuing service innovationsas the FES did, the VOI-A did stress another behavioral strategy: influenc-ing the political principals by, for example, intense personal contacts andforms of political lobbying. Fourth, as the core activity of the VOI-A wasto keep the infrastructure available, there was only limited room toenhance performance and major innovations with respect to such routine-like and stable activities.

The VOI-B was a rather small and young organization, consisting of acore “headquarter” and highly autonomous regional units. Its core activi-ties were in the field of training and consulting. Although its human andfinancial resources increased considerably in the period 1992–1999, itsperformance record was very mixed, with some services improving andothers deteriorating, indicating a rather low ability to adjust to changingenvironmental circumstances. How is this performance record to beexplained? First, in the period 1992–1999, some of the NPM-like pressureswere absent as result control by the Flemish government was lacking andthe political principals even failed to formulate a clear policy strategy andobjectives. The existing financial incentives (creaming of budget surplusesby the government) were not stimulating efficient behavior. Moreover, theextent of actual managerial autonomy was too small to change internalstructures and personnel regimes fundamentally.

However, the story is not that simple. Competition by other providersas a pressure did increase strongly, evoking a strong urge to restore/retainlegitimacy vis-à-vis the customers and the political principals. Moreover,this need was intensified by a lack of political interest or consensus aboutthe role of the VOI-B, and by fierce political lobbying of the competitors.Despite several opportunities, the VOI-B did not succeed to work out aconsistent performance-enhancing or innovative strategy because ofseveral internal characteristics, inhibiting the receptiveness of the VOI-Bfor external stimuli and its capabilities to reorient its internal functioning(see Table 3).

The second VOI under examination without a performance contract isthe Child Support Agency (CSA, Kind en Gezin), giving young childrenand mothers pre- and postnatal care, and subsidizing private organiza-tions, providing child minding. The CSA is a rather large VOI as topersonnel and financial resources. Its performance record was improvingin a consistent way, particularly after 1996. An exception was its perfor-mance with respect to equity and justice objectives, which was rather low.The CSA introduced quite radical innovations with respect to manage-ment, products and processes. It was successful in attaining an image of aflexible performance-oriented organization, both with its customers andwith its political principals.

However, the innovative behavior of the CSA cannot be explained byNPM-like pressures because the control of the CSA by the Flemish gov-ernment did not change in the direction of more result control and more


1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

42

43

44

45

46

JOBNAME: No Job Name PAGE: 16 SESS: 54 OUTPUT: Thu Jun 21 11:00:03 2007 SUM: 45CCCF3B/v2451/blackwell/journals/gove_v20_i3/gove_367

financial incentives. Only the actual level of the managerial autonomy wasperceived to have increased. Nor did the direct competition by privatedoctors and by nonsubsidized “crèches” augment strongly in the nineties.However, the CSA was motivated for performance-enhancing behavior by,again, an urge to restore/retain its legitimacy, as it was questioned widelyby society and politicians in the beginnings of the 1990s. Private generalpractitioners were claiming its activities of pre- and postnatal childcare. Asto the subsidization of “crèches” were ideas emerging to reintegrate theseactivities into the core administration. As a response and because therewas no political strategy, the CSA and its governing board developed aradical strategic plan to reorient and reform the service delivery. Thisreform plan was carried out without formal authorization of its politicalprincipals. This plan has led to “spontaneous adaptation,” by the devel-opment of new ways of service delivery and an improvement of perfor-mance. The features of the CSA were favorable to a quick reaction tochanging environmental circumstances (see Table 3). Together with inno-vations and performance-enhancing behavior, the organization enhancedits legitimacy toward its political principals by other behavioral strategies,such as communication and image-building and responsive political-oriented behavior.

Findings: An Expanded Explanatory Model

The four case studies suggest a model with a broader framework ofpressures or motivations of public organizations to innovate (see Figure 4).

The numbers in Figure 4 refer to the relevant conclusions:

1. The managerial and market-like pressures as advocated by theNPM, that is, result control and competition, do motivate publicorganizations under specific conditions to innovate.

2. However, there are also other pressures of a more political nature,stemming from threats to the legitimacy of the public organization.Such pressures may be raised by a multitude of factors, which arenot taken into account in the NPM logic. “Political pressure” comesfrom the (threat of a) decline in societal or political support for thepublic organization, which may be caused by factors such as themere presence of other potential suppliers (and their appeal to poli-ticians), the lack of political consensus about the role and form ofpublic intervention, or the perceived political and societal expecta-tions stemming from an autonomous status. As to the last factor,autonomy in itself may be a cause for political pressure to innovate.The public managers of the VOIs under scrutiny did fear that a lossof legitimacy and support of society and political principals couldresult in a reduction of tasks, cutbacks of resources, restructurationor ultimately, abolishment of the public organization.


1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

42

43

44

JOBNAME: No Job Name PAGE: 17 SESS: 54 OUTPUT: Thu Jun 21 11:00:03 2007 SUM: 3565C375/v2451/blackwell/journals/gove_v20_i3/gove_367

3. Moreover, innovative behavior as an indication of spontaneousadaptation is only one behavioral strategy, which public organiza-tions may develop to both kinds of pressure. Elements of otherstrategies are, for example, communication and image-building,competition-inhibiting behavior, (sometimes symbolic) responsivebehavior toward the political principals, political coalition-buildingand lobbying, and goal-stretching. Therefore, also less-desirableresponses (from the viewpoint of the political principal and society)are possible.

4. Some internal features of the public organization, such as, forexample, the strategic vision and capacities of governing board andtop management, the internal control system and the organizationalculture, may enhance or inhibit the capacity to react to these pres-sures (see Table 3).

5. The extent to which the public organization succeeds in restoring orenlarging societal and/or political support may result in a decreaseof NPM-like or political pressure, or in an increase or at least apreservation of resources, tasks and autonomy.

Innovation as a Function of Managerial and Political Pressures

Does this case-based broader model of pressure-response help explain thefindings of the analysis of our survey data?

FIGURE 4A Political/Administrative Pressure-Response Model Based on FourCase Studies

Managerial andmarket-like pressure orthreats (NPM)

-Result control bygovernment

-Potential competition byother suppliers

Political pressure orthreats

- Mere presence of otherreal and potential suppliersand their appeal topoliticians

- Lack of politic consensusabout role and form ofpublic interventions

- Politic and socialexpectations stemmingfrom autonomization

- Political salience task

- Others

A broader concept of pressure

Societal and political supportfor the public organisation

Innovative behavior bythe organisation

Other pro-activebehavior like

communication andimage-building

Defensive behavior likecompetition inhibiting,

market dominating behavioror political lobbying

Internal features of thepublic organisation

enhancing or inhibitingpro-active behavior

1

2

3

4

5


1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

JOBNAME: No Job Name PAGE: 18 SESS: 54 OUTPUT: Thu Jun 21 11:00:03 2007 SUM: 49CA44A5/v2451/blackwell/journals/gove_v20_i3/gove_367

1. The observed direct influence of managerial autonomy, in particu-lar, concerning operational decisions of personnel management, onthe development of new services and products (and its interactioneffect on the relationship between potential competition andinnovation):

The model that emerges from the case studies suggests two pos-sible explanations. First, public organizations with managerialautonomy which are not subjected to NPM-like pressure may bemotivated to innovative behavior because they face other forms ofpressure from threatening legitimacy. Second, having more manage-rial autonomy in itself may raise “political” pressure for a publicorganization. Public organizations with relatively larger managerialautonomy may face additional expectations from their political prin-cipals and important actors in society (customers, interest groupssuch as labor unions): the expectations of being more flexible, toperform better and to be more innovative than their counterpartswith less managerial autonomy.5 This was echoed in the case studiesby interviewees referring to the greater visibility of their autono-mized organization.

2. The observed direct effect of potential competition on innovativebehavior (and its interaction effect on the relationship betweenmanagerial autonomy concerning personnel management andinnovation):

This empirical observation supported the theoretical NPM model.In addition, the perceived presence of other providers with similarproducts and services—our operationalization of potential competi-tion in the survey—may evoke also a political kind of pressure. Evenin the absence of direct competition of such other providers, themere presence of these providers and their appeal on the politicalprincipals of the public organization may create an uncertain andrisky political environment for the public organization, which moti-vates the public organization to restore or strengthen its legitimacy.Innovative behavior is one of the observed responses.

3. The observed direct effect of result control, operationalized as thepresence of sanctions or rewards, on innovative behavior:

This finding supported the NPM model, indicating that the pos-sibility of getting sanctions and rewards from government in case ofdeviating performance had a motivating effect on the developmentof new products and services to increase performance. However, wemay assume that being explicitly sanctioned or rewarded by politicalprincipals because of performance, affects reputation and legitimacyin the eyes of principals and customers. An organization that hasbeen sanctioned by principals in the past will probably face a biggerchance to lose financial means and competencies and be restruc-tured, downsized or abolished in the future. As such, the possibility


1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

42

43

44

45

JOBNAME: No Job Name PAGE: 19 SESS: 54 OUTPUT: Thu Jun 21 11:00:03 2007 SUM: 4A12F33F/v2451/blackwell/journals/gove_v20_i3/gove_367

of sanctions and rewards creates also a political pressure in additionto the managerial NPM-like pressure.

4. The empirical model in Figure 3 explains only a part of the varianceof the independent variable (R2 = 0.25 [Nagelkerke]), see regressionanalysis in Appendix):

It is clear from the case study research that public organizationmay be motivated to innovative behavior by any factor that endan-gers its legitimacy. These factors could be numerous and are notincluded in the NPM model. Moreover, some of the unexplainedvariance may be attributed to “innovation by default” in public orga-nizations (i.e., the intercept) or by other motivations of public man-agers, such as public spiritedness.6

Discussion

Legitimacy Threats as Drivers for Innovation andSpontaneous Adaptation

Empirical observations may be explained by reference to the expandedmodel, emerging from the case studies. If these explanations hold, then theconcept of pressure of NPM adherents is too narrow to explain innovationby public organizations. A broader concept, including political pressuresfrom legitimacy threats could be more useful. Moreover, the broadermodel suggests that some organizations will find it easier to respond tosuch pressures than others because of certain internal features, which arenot taken into account in the NPM model. However, if we take a look atother research and theories, is it possible to find support for the broaderconcept of pressure as being a trigger for innovation?

In his research on innovation in public services by voluntary and non-profit organizations (VNPO), Osborne (1998) tests several factors as totheir influence on innovative behavior of such organizations. Osbornefinds important support for his institutional hypothesis and assesses theinfluence of institutional factors, such as societal changes, central govern-ment perceptions, perceptions and expectations of funders and of othersimilar organizations, on innovation. He found that the major impetus forinnovation via these institutional factors is “the search for legitimacy. Thisis the benefit that innovation can bestow upon a VNPO—be it legitimacyin the eyes of their beneficiaries, their staff, their peers, or perhaps mostsignificantly, their funders” (Osborne 1998, 160, his emphasis). Whereasthe profit motive and direct competition are major factors for privatesector innovation, the legitimacy motive in the perspective of the institu-tional framework is central in voluntary and nonprofit sectors (Osborne1998; see also Verhoest 2002) for a similar conclusion for the public sector).This corresponds clearly with our notion of political pressures stemmingfrom legitimacy threats.


1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

42

43

JOBNAME: No Job Name PAGE: 20 SESS: 54 OUTPUT: Thu Jun 21 11:00:03 2007 SUM: 523F819D/v2451/blackwell/journals/gove_v20_i3/gove_367

Suchman (1995, 574) defines legitimacy as “a generalized perception orassumption that the action of an entity are desirable, proper or appropri-ate within some socially constructed system of norms, values, beliefs anddefinitions.” In his work on for-profit firms, he points out that organiza-tions seek legitimacy through “achievement strategies” that conform tothe external audience, manipulate the external audience, or informunaware audience members of the organization’s activities. Cashore (2002)adapts these concepts and strategies to a broader setting and givesexamples. Therefore, “Innovative activity is not the only way in which togain legitimacy. Providing a specialist service, being a campaigning orga-nization, or providing a key mainstream service could be equally valid.Indeed, some of the traditional organizations eschewed innovation quitepurposefully, in exchange for one of these other sources of legitimacy”7

(Osborne 1998, 160, cf. Carpenter 2001; Dove 2002).Similarly, our case studies showed that public organizations, in order to

strengthen their legitimacy, can also choose for defensive and reactivebehavioral strategies. Moreover, they even tried to influence the environ-ment in order to reduce the factors that cause legitimacy threats or toinfluence the criteria and frameworks of reference used by importantstakeholders to evaluate the public organization. A clear example of thelatter point is that the two VOIs with a performance contract in our casestudies sought to dominate the contract negotiations in order to insertperformance objectives in their contracts which were both closely alignedto their own interests and quite easy to achieve. Also, by the broad supportfor their innovations, which were mainly launched independently of theirpolitical principals, two VOIs (the FES and the CSA) influenced the policyframework in their field, and as such, the criteria on which they wereevaluated (see Carpenter 2001 for a rich description of similar processes).

This emphasis on legitimacy as a drive for organizations is not that new,and has been discussed thoroughly in the literature on sociological insti-tutionalism (Peters 1999; Scott 1995). In particular, the work of Powell andDiMaggio (1991) on organizational isomorphism8 and subsequent appli-cations to nonprofit sectors are of importance here. Isomorphism of anorganization to its institutional environment enhances legitimacy and byconsequence, results in “greater access to resources, which reduces mor-tality rates” (Singh and Lumsden 1990, 184 as quoted in Osborne 1998, 46).In the DiMaggio and Powell perspective, isomorphism induces pressuresto structural uniformity within an organizational field in coercive,mimetic or normative ways. Osborne (1998, 187) states that in order toexplain innovation in nonprofit sector, there is another pressure atwork—“a pressure to congruence with the prevailing expectations within theinstitutional field,” what he calls instrumental isomorphism. In our casestudies, there is evidence that the VOIs oriented themselves quite stronglytoward what they perceived as the expectations of their customers, interestgroups and, sometimes quite indirectly, their political principals (or somefactions of these principals) in order to enhance their legitimacy.9


1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

42

43

44

45

46

JOBNAME: No Job Name PAGE: 21 SESS: 54 OUTPUT: Thu Jun 21 11:00:03 2007 SUM: 47AF5401/v2451/blackwell/journals/gove_v20_i3/gove_367

The interviewed public managers in the case studies pointed at anotherimportant issue, namely, the benefits of a strengthened legitimacy. Whenpublic organizations successfully deploy behavioral strategies in order torestore or strengthen legitimacy, they get more and wider societal andpolitical support for an increase of resources, tasks and autonomy, adecrease of harshness of control and competition, or at least for a preser-vation of the level of all these aspects (see also Dove 2002). As such,innovative behavior as a manifestation of “spontaneous adaptation” is asmart practice in the sense of Bardach (1998, 35–36). Public organizationsmay behave innovatively to safeguard or to “forge their autonomy”(Behn’s [2001] notion of “earned” autonomy; Carpenter 2001). By doingso, they seem to exploit the opportunity (Bardach 1998) of a potentialwin-win situation: producing some result for the public (innovative ser-vices, . . .), while in the same run they try to enhance trust levels of publicand political principals—a condition sine qua non for their autonomy.

Conclusion

A statistical test provided evidence for the NPM-pressure-response modelon the innovative behavior of public organizations. However, the empiri-cal model showed more complex relationships as was assumed by NPMdoctrine, with managerial autonomy having an independent effect oninnovation. An explanation was suggested by a recent multiple-case studyof four Flemish public organizations. The case research indicated that anexpanded political–administrative pressure-response has more explana-tory power. Central in this broader model is the search for legitimacy bythe public organization in the eyes of its users, stakeholders and, ulti-mately, its political principals. Innovation was only one of the observedways to restore or strengthen legitimacy and public organizations differ intheir capacity to react adequately to these kinds of pressure. Additionalsupport for the importance of the search for legitimacy as motivatingfactor was found in other empirical research and the sociological institu-tionalism as a broader theoretical framework.

The research findings point at certain stimulating factors for innovationas a form of “spontaneous adaptation” by public organizations. Thesefactors are managerial autonomy, result control (in the form of sanctionsand rewards), competition, and more broadly, factors that threaten orga-nizational legitimacy. However, certain intraorganizational features mayenhance or inhibit the capacity of organizations to innovate, as is alsosuggested by other empirical research (Carpenter 2001; Osborne 1998).These features seem to interact in a complex way with the institutionalforces and on each other, and they function like moderating and interme-diate variables. Theories with relevance in this respect are, for example,network and resource dependency theory, theories about organizationalculture and contingency theory. This may suggest that the effect of thepressure for innovative behavior (be it NPM-like or political nature) is


1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

42

43

44

45

JOBNAME: No Job Name PAGE: 22 SESS: 54 OUTPUT: Thu Jun 21 11:00:03 2007 SUM: 5A9BAD99/v2451/blackwell/journals/gove_v20_i3/gove_367

contingent upon the presence of some of these additional factors (see theinternal features in the case study research, Table 3). Constructing andtesting a model, which encompasses such variables, could be a future lineof research.

Notes

1. The concept of “result control” contains several subitems (e.g., goalsetting,indicators, measuring, evaluating, sanctions and rewards), of which the lastsub-item “sanctions and rewards” has been chosen as an operationalization.Rewards and sanctions are closing the results control cycle and are the mosttangible element in this cycle. Moreover, tests of correlation between theresults control items and the innovation type showed that only “sanctions andrewards” correlates significantly (Tau-b = 0.30 [0.007], see Appendix).

2. The public managers of the organizations that were studied in a multiple-case research (Verhoest 2002—see below), stated that managerial autonomyconcerning personnel management was of greater importance than mana-gerial autonomy as to financial matters if they wanted to enhance organiza-tional performance.

3. Before we discuss some theories, we should ask ourselves if the relation-ships we draw in the empirical model can be the other way around. Then, thedevelopment of new products and services would enhance the rise of poten-tial competitors or the increase of managerial autonomy. Only the latterseems intuitively acceptable and is not to be excluded a priori. However, theinteraction effects would be hard to explain in such a turned-around model.

4. This case and the case of the FAI-B were made anonymous in order toprotect these organizations and their management from negative conse-quences of this research.

5. In that sense, managerial autonomy concerning personnel management ismore visible and has a more salient nature for political and societal actorsthan managerial autonomy concerning financial management (which ismore technical in nature and more difficult to understand). Public organi-zations, which have personnel regimes deviating from the standard govern-mental personnel statute, clearly are monitored quite closely by, for instance,labor unions and the associated political factions. Moreover, the case studyresearch indicated that public managers themselves valued managerial per-sonnel autonomy more than financial autonomy (Verhoest 2002).

6. A competitive theoretical framework with potential relevance is a combina-tion of open system theory and traditional administrative theory. In thisperspective, granting public organizations more managerial autonomy is initself a good thing to do because (1) public organizations with more mana-gerial autonomy can adapt more quickly and easily to changes in theirenvironments (open system theory), and (2) public managers having moremanagerial autonomy can be assumed to pursue a better performance (bythe means of e.g., innovation) because they are driven by a public sectorethos, a focus on the general interest, by loyalness to their political princi-pals. However, we lack the data in our survey to test this theory. Moreover,when assuming public sector ethos and loyalness as motivations for publicmanagers, one can assume that the introduction of result control and com-petition would have a negative effect on motivation, and thus, result in adecline of innovative behavior. However, this was not corroborated by thestatistical analysis.

7. VOI-A in the case study research found it more important to deliver a stableand continuous services, eschewing major disturbances. Moreover, its


1

2

3

4

5

6

789

1011121314151617181920212223242526272829303132333435363738394041424344454647484950515253

JOBNAME: No Job Name PAGE: 23 SESS: 54 OUTPUT: Thu Jun 21 11:00:03 2007 SUM: ABBF7430/v2451/blackwell/journals/gove_v20_i3/gove_367

activities were focused on maintaining and improving the availability ofinfrastructure, leaving little room for radical innovations.

8. Meyer and Rowan (1977), with their theory of rationalized myths suggestthat innovations with respect to organizational structures may be of a sym-bolic nature, with little effect on operations of the organization.

9. However, it is interesting to analyze to what extent the prevalence of theinnovation type that we have studied could be explained by coercive,mimetic or normative isomorphism. If that is the case, one would expectthat, for instance, the development of new products and services as aninnovation type is relatively much more present within the groups of orga-nizations residing under some ministerial departments than within thegroups under other departments. The minister and his department may playa major role in advocating such innovations for the organizations with whichthey are responsible. However, comparisons based on the departmentalaffiliation showed no clear links between the prevalence of innovation andthe mother departments. Also, the prevalence of the innovation type showedno clear link with the formal–legal types of organizations, indicating a lowlevel of uniformity or convergence within these types as to innovation.

References

Alter, Norbert, ed. 2002. Les Logiques de l’Innovation. Approche Pluridisciplinaire[Logic of Innovation. A Pluri-Disciplinary Approach]. Paris: La Découverte.

Aucoin, Peter. 1990. “Administrative Reform In Public Management: Paradigms,Principles, Paradoxes and Pendulums.” Governance: An International Journal ofPolicy, Administration and Institutions 2: 115–137.

Bardach, Eugene. 1998. Getting Agencies to Work Together: The Practice and Theory ofManagerial Craftmanship. Washington, DC: Brookings.

Barzelay, Michael, and Colin Campbell. 2003. Preparing for the Future: StrategicPlanning in the U.S. Air Force. Washington, DC: The Brookings Institution.

Behn, Robert. 2001. Rethinking Democratic Accountability. Washington, DC: Brook-ings Institution.

Boorsma, Peter, and Arie Halachmi. 1998. Inter and Intra Govenment Arrangementsfor Productivity: An Agency Approach. Boston, Dordrecht, and London: KluwerAcademic Publishers.

Bouckaert, Geert. 2003. “La Réforme de la Gestion Publique Change-t-Elle lesSystèmes Administratifs?” [Does the Reform of Public Management Changethe Administrative Systems?]. Revue Française d’Administration Publique [FrenchPublic Administration Review] 105–106: 39–54.

Bryman, Alan, and Duncan Cramer. 1999. Quantitative Data Analysis with SPSSRelease. 8. A Guide for Social Scientists. London and New York: Routledge.

Campbell, Colin, and John Halligan. 1992. Political Leadership and Age of Constraint:Bureaucratic Politics under Hawke and Keating. Sydney: Allen and Unwin; Pitts-burgh, PA: University of Pittsburgh Press.

Carpenter, Daniel. 2001. The Forging of Bureaucratic Autonomy. Reputations, Net-works and Policy Innovation in Executive Agencies, 1862–1928. Princeton andOxford: Oxford University Press.

Cashore, Benjamin. 2002. “Legitimacy and the Privatization of EnvironmentalGovernance: How Non-State Market-Driven (NSMD) Governance SystemsGain Rule-Making Authority.” Governance: An International Journal of Policy,Administration and Institutions 4: 503–529.

Christensen, Jorgen. 2001. “Bureaucratic Autonomy as a Political Asset.” In Politi-cians, Bureaucrats and Administrative Reform, ed. Guy Peters and Jon Pierre.London: Routledge.


123456789

101112131415161718

19

20

212223242526272829303132333435363738394041424344454647484950515253

JOBNAME: No Job Name PAGE: 24 SESS: 54 OUTPUT: Thu Jun 21 11:00:03 2007 SUM: DEB80263/v2451/blackwell/journals/gove_v20_i3/gove_367

Common, Richard, Norman Flynn, and Elizabeth Mellon. 1992. ManagingPublic Services. Competition and Decentralization. Oxford: Butterworth-Heinemann.

Dove, Suzanna. 2002. “Fragile Miracles: The Creation and Sustainability of Autono-mous Oversight Agencies in a Politicized Bureaucracy—The Case of Bolivia.”Tesi Doctoral: Universitat Autonoma de Barcelona.

Dunleavy, Patrick. 1997. “The Globalization of Public Services Production: CanGovernment Be ‘Best In the World?’ ” In Globalization and Marketization of Gov-ernment Services, ed. Andrew Massey. Basingstoke, UK: Macmillan.

Fölster, Stefan. 1997. “Ist der Systemwechsel in Schweden in Gefahr? Erfahrungenmit Privatisierung, Deregulierung und Dezentralisierung” (Is the systemchange in Sweden in danger? Experiences with privatization, deregulation anddecentralization). In Reformen des öffentlichen Sektors in Skandinavien. Einebestandsaufnahme, ed. Claudius H. Riegler and Frieder Naschold. Baden-Baden,Germany: Nomos Verlagsgesellschaft.

Healey, Joseph. 1996. Statistics. A Tool for Social Research. Belmont: WadsworthPublishing Company.

Hodge, Graeme Arthur. 1996. Contracting Out Government Services: A Review ofInternational Evidence. Melbourne: Monytech Pty.

Hoggett, Paul. 1996. “New Modes of Control in the Public Service.” Public Admin-istration 74 (Spring): 9–32.

Hood, Christopher A. 1991. “A Public Management for all Seasons?” Public Admin-istration 69: 3–19.

Le Grand, Julian, and Will Bartlett. 1993. Quasi-Markets and Social Policy. London:McMillan.

Meyer, John, and Brian Rowan. 1977. “Institutionalized Organizations: FormalStructure as Myth and Ceremony.” The American Journal of Sociology 83: 340–363.

Nelson, Richard 1993. National Innovation Systems. New York: Oxford UniversityPress.

OECD. 1997. In Search of Results: Performance Management Practices in Ten OECDCountries. Paris: Public Management Committee, OECD.

———. 2002. Distributed Public Governance—Agencies, Authorities and Other Autono-mous Bodies. Paris: OECD.

Osborne, Stephen. 1998. Voluntary Organizations and Innovation in Public Services.London and New York: Routledge.

Peters, Guy B. 1999. Institutional Theory in Political Science. The “New Institutional-ism.” London: Continuum.

Pollitt, Christopher, Colin Talbot, Janice Caulfield, and Amanda Smullen. 2004.Agencies: How Governments Do Things through Semi-Autonomous Organizations.Basingstoke, UK: Palgrave Macmillan.

Powell, Walter, and Paul DiMaggio (eds.). 1991. The New Institutionalism in Orga-nizational Analysis. Chicago and London: The University of Chicago Press.

Schick, Allen. 1996. The Spirit of Reform: Managing the New Zealand State Sector in aTime of Change. A report prepared for the State Services Commission and theTreasury. New Zealand.

Scott, W. Richard. 1995. Institutions and Organizations. London: Sage.Singh, Jitendra, and Charles Lumsden. 1990. “Theory and Research in Organiza-

tional Ecology.” Annual Review of Sociology 16: 161–195.Suchman, Mark. 1995. “Managing Legitimacy: Strategic and Institutional

Approaches.” Academy of Management Review 20: 571–610.Van Thiel, Sandra. 2002. Quangocratization: Trends, Causes and Consequences.

Utrecht, the Netherlands: Interuniversity Center for Social Science Theory andMethodology.


123456789

101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354

44


Verhoest, Koen, B. Guy Peters, Geert Bouckaert, and Bram Verschuere. 2004. “TheStudy of Organisational Autonomy: A Conceptual Review.” Public Administra-tion and Development 24: 101–118.

Verhoest, Koen, and Geert Bouckaert. 2005. “Machinery of Government and PolicyCapacity: The Effects of Specialisation and Coordination.” In Challenges to StatePolicy Capacity: Global Trends and Comparative Perspectives, ed. Martin Painterand Jon Pierre. Basingstoke, UK: Palgrave Macmillan.

Verhoest, Koen. 2002. “Resultaatsgericht verzelfstandigen. Een analyse vanuit eenverruimd principal—agent perspectief. Doctoraatsverhandeling (Result drivenautonomization. An analysis based on an extended principal agent perspective).Leuven: K.U. Leuven—Faculteit Sociale Wetenschappen.

———. 2005a. “The Impact of Contractualization on Control and Accountability inGovernment-Agency Relations: The Case of Flanders (Belgium).” In Contracts,Performance and Accountability, ed. Gavin Drewry, Carsten Greve, and ThierryTanquerel. ••: EGPA/IOS.

———. 2005b. “Effects of Autonomy, Performance Contracting and Competitionon the Performance of a Public Agency: A Case Study.” Policy Studies Journal 33:235–258.

Walsh, Kieron. 1995. Public Services and Market Mechanisms: Competition, Contract-ing and the New Public Management. Basingstoke, UK: Macmillan.

Yin, Robert K. 1994. Case Study Reseach. Designs and Methods. Thousand Oaks, CAand London: Sage.

Appendix: The Statistical Analysis of the Survey-Data

Operationalisation of the Variables in the Statistical Analysis

Independent Variables: Autonomy Competition & Result Steering

Operational Personnel Managerial Autonomy (OPA)

Low levels (0): organization agrees on zero or one of following statements: a.organization can issue extra-legal advantages to personnel member, b.promote personnel member, c. evaluate personnel member, d. appointpersonnel member.

High levels (1): organization agrees on three or all of abovementionedstatements.

Operational Financial Managerial Autonomy 1 (OFA1)

Low levels (0): organization agrees on zero or one of following statements:within the regulative framework set by oversight authorities; a. organiza-tion can engage in loans, b. organization can set tariffs for services andproducts delivered, c. organization can participate in private law legalpersons.

High levels (1): organization agrees on two or all of the abovementionedstatements.

Operational Financial Managerial Autonomy 1 (OFA2)

Low levels (0): organization cannot shift budgets


123456789

10111213141516171819202122

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

42

43

44

45

55

JOBNAME: No Job Name PAGE: 26 SESS: 54 OUTPUT: Thu Jun 21 11:00:03 2007 SUM: 5860C205/v2451/blackwell/journals/gove_v20_i3/gove_367

High levels (1): organization can shift budgets on one, two or three offollowing cases: a. between personnel and running costs, b. between costsand investments, c. over different years.

Competition (COM)

Low levels (0): organization faces no other suppliers on market for itsservices and products.

High levels (1): organization faces other suppliers (few or a lot) on marketfor its services and products.

Result Steering (RES)

Low levels (0): organization is no subject to sanctions (in case of badresults) nor rewards (in case of good results).

High levels (1): organization is subject to sanctions (in case of bad results),rewards (in case of good results) or both sanctions and rewards.

Dependent Variable: Innovative Technique

Development of Innovative Services and Products (NewSP)

Low levels (1): organization never develops innovative services and prod-ucts.

Intermediate levels (2): organization develops innovative services andproducts to a small extent.

High levels (3): organization develops innovative services and products toa great extent.

Sample Size (N) and Distribution

TABLE 1N of cases per variable and distribution per value

Value // IV COM OPAB OFA1 OFA2 RES

‘0’ 40 40 51 39 37‘1’ 43 38 28 37 40

Val // DV NewSP

‘1’ 21‘2’ 29‘3’ 25

Compared to the total number of cases in the sample (table 1, pg 5), we havea smaller number of cases in the statistical analysis, due to the fact that notall organizations responded to every survey questions (missing data).


1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

2728

2930313233

343536373839

40

41

42

JOBNAME: No Job Name PAGE: 27 SESS: 54 OUTPUT: Thu Jun 21 11:00:03 2007 SUM: 8B087515/v2451/blackwell/journals/gove_v20_i3/gove_367

Statistical Analysis of the Model

Step 1: Correlations (Tau-b) between Independent Variables andDependent Variable

TABLE 2Correlations between independent and dependent variables

COM OPA OFA1 OFA2 RES NewSP

COM Corr Coef 1,000 -,038 -,073 ,120 ,142 ,219Sig. (2-tailed) , ,737 ,524 ,304 ,214 ,046N 83 77 78 75 77 75

OPA Corr Coef -,038 1,000 ,423 ,563 ,207 ,295Sig. (2-tailed) ,737 , ,000 ,000 ,079 ,008N 77 78 76 73 73 73

OFA1 Corr Coef -,073 ,423 1,000 ,503 ,162 ,149Sig. (2-tailed) ,524 ,000 , ,000 ,167 ,181N 78 76 79 75 74 73

OFA2 Corr Coef ,120 ,563 ,503 1,000 ,240 ,171Sig. (2-tailed) ,304 ,000 ,000 , ,045 ,131N 75 73 75 76 71 70

RES Corr Coef ,142 ,207 ,162 ,240 1,000 ,300Sig. (2-tailed) ,214 ,079 ,167 ,045 , ,007N 77 73 74 71 77 72

NewSP Corr Coef ,219 ,295 ,149 ,171 ,300 1,000Sig. (2-tailed) ,046 ,008 ,181 ,131 ,007 ,N 75 73 73 70 72 75

There seems to be strong relationships between COM, OPA and RES onthe one hand (DV) and NewSP (IV) on the other hand. Next to that strongand significiant relationships between independent variables (OFA1, OPAand OFA2) also emerge, indicating some potential collinearity betweenindependent variables.

Step 2: Ordinal Regression

All independent variables are entered in the model for NewSP (depen-dent).

TABLE 3aOrdinal regression (all independent variables in model)

IV Estimate Sig

OPA 1.3 .03RES 1.1 .04COM 1.0 .05OFA1 .28 .64OFA2 -.34 .58

R2 = .25 (Nagelkerke)


1

2

3

45

6789

1011121314151617181920212223242526

27

28

29

30

31

32

33

34

35

3637

3839404142434445

46

66


Manual backward elimination leads to final model. In two steps, OFA1and OFA2 are eliminated, because they show no significant effect inregression. Moreover, no relationship with the dependent variable wasobserved in the correlation analysis. Finally, also chances to collinearity inregression decreases, as both variables strongly correlate with each otherand OPA. The model we become is then:

TABLE 3bOrdinal Regression (after manual backward elimination)

IV Estimate Sig

OPA 1.2 .01RES 1.1 .03COM .8 .07

This model explains 24% (.24 Nagelkerke R2) of the variance in dependentvariable NewSP. The model fit significance is .001.

Step 3: Crosstabulation

The three variables in the model are put in a crosstabs-analysis, in order tocontrol their effect on NewSP for the other variables (Healey 1996, 407 andfurther, Bryman and Cramer 1999, 234 and further).

TABLE 4Crosstabulation results

Initialrelation

Tau-b(Sig.)

Controlledfor

Tau-b(Sig.) for: Effect

OPA– NewSP .30 (<.01) RES For RES = 1: .36(.009)

No considerablemediatingeffect

OPA– NewSP COM For COM = 0: .45(.001)

Relation OPA –NewSPaffected bylevel of COM

RES – NewSP .30 (<.01) OPA X2 test not sign –RES – NewSP COM X2 test not sign –COM – NewSP .22 (.04) OPA For OPA = 0: .43

(.001)Relation COM–

NewSPaffected bylevel of OPA

COM – NewSP RES X2 test not sign –

Relationships of variables in model (OPA, COM and RES) on NewSP,controlled for OFA1 and OFA2 (variables removed out of model) did notshow a significant mediating effect.


1

2

3

4

5

6

78

91011121314

15

16

17

18

19

20

21

2223

242526272829303132333435363738394041

42

43

44

JOBNAME: No Job Name PAGE: 29 SESS: 54 OUTPUT: Thu Jun 21 11:00:03 2007 SUM: 2789717C/v2451/blackwell/journals/gove_v20_i3/gove_367

AUTHOR QUERY FORM

Dear Author,During the preparation of your manuscript for publication, the ques-

tions listed below have arisen. Please attend to these matters andreturn this form with your proof.

Many thanks for your assistance.

QueryReferences

Query Remark

q1 Au: Is the short title running head ok? If not, pleaseprovide one.

q2 Au: Acoin 1990 has been changed to Aucoin 1990so that this citation matches the reference list.Please confirm if this is correct.

q3 Au: “Appendix 1” has been changed to “Appendix”as there is only one appendix found in the article. Isthat okay?

q4 Au: Healey 1996: Please provide the state/countryfor Belmont.

q5 Au: Verhoest 2005a: Please provide the place ofpublication.

q6 Au: Please check whether ‘,’ in the date should bechanged to decimal points.

SNP Best-set Typesetter Ltd.Journal Code: GOVE Proofreader: EmilyArticle No: 367 Delivery date: 21 June 2007Page Extent: 28 Copyeditor: Rhys

Documents

Pressure, Legitimacy and Innovative Behavior by Public ... · Pressure, Legitimacy and Innovative Behavior by Public Organizations* KOEN VERHOEST, BRAM VERSCHUERE, and GEERT BOUCKAERT