Research on John Carver’s Policy Governance Model Reveals Its Implementation and Effectiveness
By Patricia Dautel Nobbie (pnobbie@dhr.state.ga.us)

New research on John Carver’s Policy Governance Model is shedding some light on the implementation and effectiveness of the model in nonprofit boards of directors. Patricia Dautel Nobbie’s doctoral dissertation, “Testing the Implementation, Board Performance, and Organizational Effectiveness of the Policy Governance Model in Nonprofit Boards of Directors” is the first comprehensive, empirical examination of the model in use. The dissertation was recognized for the Gabriel G. Rudney Award for Outstanding Research in Nonprofit and Voluntary Action.

In the John Carver’s Policy Governance Model, the board drives the organization through policies developed in four areas: ends (mission); executive limitations (acceptable practices for the chief administrative officer); board-staff link (power and accountability relationships); and, governance processes (board responsibilities and relationship to those served). The board grants discretion to the CEO to advance the “ends” policies or mission of the organization without violating the “means” incorporated in the three remaining policy areas. Estimates are as high as several thousand organizations in the U.S. and Canada that claim to be operating according to the Policy Governance model. The research examined relationships between implementation of the Policy Governance model, performance of the board of directors, and the effectiveness of the organization. Several contextual variables previously explored in the literature were included in the framework, such as size of board, age of the organization, hours of board training, use of a consultant, and so forth.

Data was collected through surveys constructed with items derived from the literature on board governance, board performance and five models of organizational effectiveness: goal achievement, resource acquisition, internal processes, CEO job satisfaction and CEO performance. Data for investigating model implementation and board performance were gathered from chief executive officers, chairpersons and board members of 32 organizations that have implemented or are in the process of implementing the model.

Data to examine comparative organizational effectiveness were drawn from two control groups. The first consisted of 290 CEOs from a randomly selected set of 1,500 nonprofit organizations that filed 990 tax forms in 1999. This sample was stratified to match the proportions of organization types in the National Taxonomy of Exempt Entities (NTEE) classification system to each category in the Policy Governance sample. The second control sample was comprised of CEOs of nonprofit organizations that participated in board development activities from the National Center for Nonprofit Boards (NCNB) (N=26). In these samples, four dimensions of organizational effectiveness were assessed: goal achievement, resource acquisition, internal processes and CEO job satisfaction. The use of control groups allowed examination of whether measures of effectiveness from Policy Governance organizations varied significantly in these four areas from either randomly selected organizations whose governance practices are undetermined, or from organizations that had participated in governance development. Organizations in the control groups that indicated they had attempted board change using Policy Governance were removed from the samples to avoid contamination.

Findings and Implications: Implementation
Prior to this study, little was known about the extent to which an organization actually implemented the policies and practices prescribed by the model.

In this research, data from 234 board members of 32 Policy Governance organizations revealed that board members consistently, and to a high degree, practiced Policy Governance behaviors more than traditional board behaviors. Board members appeared to make the “paradigm shift” that Carver refers to, and there is little evidence that they blended traditional and Policy Governance board behaviors. The degree of implementation of the model was significantly related to the number of hours of training received in the model and the length of time board members stated they had operated under the model. However, partial correlations of 12 distinct Policy Governance behaviors (controlling for time after implementation) revealed that board members did not practice all behaviors to the same degree, an indication that the model is not practiced as a whole as Carver recommends.

Board Performance
The second area of examination in the study was the relationship between extent of implementation of the model and board member’s perceptions of enhanced board performance. Board members, CEOs, and to a lesser extent, board chairpersons perceived significant improvement in the board’s performance after adoption of the Policy Governance model. The relationship between the mean level of implementation of Policy Governance behaviors and board members’ perceptions of improvement in board performance was positive and significant at the individual and organization levels.

Organizational Effectiveness
In the Policy Governance organizations, there appears to be a relationship between the extent of implementation of the model, and board members’ and CEOs’ impressions of enhanced organizational effectiveness. This was true for board members in the area of goal achievement and for CEOs in the areas of internal processes and job satisfaction.

In comparing the Policy Governance sample and the control groups, Policy Governance CEOs’ measures of effectiveness varied significantly from the random sample in goal achievement and job satisfaction. However, there were no significant differences in perceptions of effectiveness between Policy Governance CEOs and the CEOs from the NCNB sample. The evidence suggests that CEOs of Policy Governance organizations perceive their organizations have greater performance in the terms of goal achievement and job satisfaction than do CEOs of organizations in a random sample whose governance processes are unknown. However, there is no evidence here to support that the performance of Policy Governance organizations is superior to that of organizations that have adopted an alternative method of board governance.

The results would seem to indicate that boards benefit from adoption of, and adherence to a model of governance, but there is little to support that Policy Governance model yields superior results in terms of board performance and organizational effectiveness. Boards in this sample that faithfully adhered to the guidelines of Policy Governance reported good outcomes, with some exceptions. However, results varied between different types of organizations, and varied according to training and how long an organization stayed with the model. The research opens several doors to increased scrutiny of the Policy Governance model, and aspects of the relationships between implementation of a model, board performance and resulting effectiveness.