Applying concepts from heat transfer to explain policy diffusion

Public policy is no longer just a tool for good governance. It is increasingly a competitive tool given the globalized economy in which competition occurs across national boundaries. Technology policy, like healthcare policy, accounting, and other specialized subject areas, is increasingly complex and requires involvement of subject area experts.

The policy literature highlights the need for a more sophisticated specification of the policy process to allow effective subject matter expert involvement. Most established approaches for explaining policy diffusion rely primarily on historical patterns of adoption to generate scores for predicting future adoption dates. This dissertation derives a measure of ease of policy diffusion based on the concept of heat transfer in conduction through a wall.

This approach performs a central engineering management function by identifying synthesis between physical and social systems. As the model is refined, the ability to apply laws governing heat transfer to the policy process will provide an increasingly accurate policy diffusion model based on an improved understanding of policy levers. The effect on policy design and subject matter expert involvement in policymaking mirrors that seen in heat transfer from centuries of empirical tests on the behavior of materials as temperature is altered.

Diffusivity, defined here as the rate of change in heat level, is used as a measure of the rate of policy change. In heat transfer, the diffusivity is a ratio of conductivity to volumetric heat capacity. Higher diffusivity implies easier thermal transfer and lower diffusivity, the reverse.

In heat transfer, conductivity measures the capacity to transfer heat across the barrier between two regions. In policy diffusion, it is equivalent to the resources available to overcome resistance to policy diffusion. Based on literature, institutional capacity is used as a proxy for conductivity.

Volumetric heat capacity is a measure of the capacity to resist heat transfer by absorbing heat flow through the barrier. It is the capacity to store thermal energy and resist heat transfer through the material. Based on the literature, culture is an ideal proxy for this measure.

The data on culture is, however, inconsistent and cross-sectional in comparison with other annually updated data sets. Affluence disparity, which is also discussed as a retardant to policy diffusion, is used for the policy diffusion model. The literature discusses the use of affluence because of the financial requirements for policy or institutional change.

The mapping from heat transfer to policy diffusion is conducted in two stages. The first is a conceptual map of heat transfer parameters to policy diffusion. This is followed by empirical tests – statistical tests and graphical comparisons of the variation of each of the parameters with temperature in heat transfer, and an approximation for temperature in policy diffusion. This approximation is made in compensation for the binary measure for the level of policy implementation – the ideal map for temperature.

Kendall’s tau tests of association are used to compare the diffusivity score with an innovativeness score generated from historical policy adoption rates among U.S. states. The factors influencing diffusion of the Bayh Dole act for technology commercialization and intellectual property protection are also discussed in light of the diffusivity scores.

From the rank association tests (Kendall’s tau), the adoption pattern expected from the diffusivity score ranks does not significantly match that expected from the innovativeness score ranks for the 34 US states considered. There is also no significant correlation between the diffusivity score and adoption year (tau = -0.17).

The negative correlation coefficient at a 0.085 level of significance (though not satisfactory at the 0.05 threshold) suggests the reverse of the pattern expected for adoption based on the diffusivity score – a finding of considerable interest for future study. There was a significant correlation (tau = 0.251) at the 0.05 level of significance, between the innovativeness score and the adoption year, implying that that higher innovativeness scores suggest earlier policy adoption.

While the sample size constrains the statistical power of conclusions from this test, the observed correlation suggests a need for further tests with refinements to the model to include additional levers, such as the relative size of policy jurisdictions, demographic differences, proportions of rural-to-urban populations, and the strength and support of issue networks.

Graphical analysis is then used to map the diffusivity relationship in heat transfer to the equivalent relationship in the policy diffusion process. Temperature dependence is the basis for such pattern mappings in heat transfer. In policy diffusion, the level of policy implementation (policy level) is the temperature equivalent. This measure is yet to be operationalized. As such, this study uses an approximation based on the heat flow rate equation in heat transfer.

This approximation introduces a limitation to the validity of the conductivity mappings. However, it is an essential first step in testing the similarities between heat transfer and policy diffusion levers. Of additional importance, this dissertation highlights measures, such as the policy level, degree of contact and distance between jurisdictions, and the policy flow rate, the operationalization of which are needed for a more sophisticated specification of policy lever interactions.

These initial mappings are based on the adoption of a Renewable Portfolio Standard in the United States. Limitations, such as a measure for policy level (the temperature equivalent), compromise the accuracy of mappings and the conclusive power of the model in its present form. However, this study has identified significant conceptual similarity between heat transfer in the physical sciences and policy diffusion in the social sciences. The identification of individual levers and explanation of lever interactions opens a new frontier in physical-to-social science research and begins a process through which the benefits of sophisticated specification from centuries of heat transfer and material science research can be exploited in policy diffusion.

Policy researchers, such as Berry and Berry (1992) and Mintrom (1997), stress the need for a more sophisticated specification of the policy process that is not solely reliant on historical events. This dissertation initiates work to address this need. In accordance with the definition of the engineering management profession, the research here is broadly focused with an aim to achieve synthesis between the physical and social concepts of diffusion.

The diffusivity model, in its current form, explains conceptual similarity between heat transfer and policy diffusion. Further work is required to support this conceptual synthesis empirically. One such avenue for improvement is the use of additional levers informed by the innovativeness score.