From diffusion to diffuse-ability: unpacking the textual drivers of global corporate sustainability policy spread

In the changing landscape of global corporate sustainability policy, understanding how policies spread across borders is crucial. Traditionally, diffusion research has focused on the role of actors—such as governments, international organizations, and other stakeholders—in spreading policies. However, our latest study, From Diffusion to Diffuse-ability: A Text-as-Data Approach to Explaining the Global Diffusion of Corporate Sustainability Policy, shifts the spotlight onto the attributes of the policies themselves. We argue that the textual characteristics of a policy significantly influence how much of its content is adopted by other entities.

The attributes that matter

Our research identifies three key textual attributes that determine the extent of policy diffusion: complexity, flexibility, and restrictiveness. Complexity refers to how easily a policy can be understood. High complexity, characterized by technical jargon and lengthy, intricate sentences, creates a cognitive burden for potential adopters, often leading to greater adaptation when the policy is implemented elsewhere. Flexibility, on the other hand, relates to the generality of the policy language. Policies that use broad, ambiguous language allow for greater adaptation to local contexts, leading to wider, but often modified, diffusion. Restrictiveness, defined by the degree to which a policy mandates specific actions, typically results in less direct copying of the text, as adopters may resist binding commitments.

Analysing 65 years of global data

To test our hypotheses, we compiled a novel dataset of 1,429 corporate sustainability policies (CSPs) issued by 100 countries, 20 international organizations, and 12 regional organizations over a span of 65 years. This extensive dataset allowed us to measure the degree of textual copying, or "diffuse-ability," between original policies and their adopted counterparts.

Using advanced natural language processing (NLP) techniques, we assessed how much text from a source policy was directly incorporated into an adopted policy. We employed three different methods—cosine similarity, substring matching, and a hybrid approach involving cosine similarity with five-gram word strings—to ensure robust findings.

Key findings

Our analysis yielded several important insights:

Complexity reduces direct diffusion

As expected, policies with higher complexity were adapted more when adopted by other entities. This was evident in the negative correlation between readability and textual diffusion—policies that were easier to read tended to be adopted with less alteration.

Ambiguity promotes adoption

Contrary to our initial hypothesis, we found that policies with more ambiguous language were more likely to be adopted without significant changes. This suggests that adopters might prefer to retain flexibility in how the policy is applied, rather than undertaking the cognitive effort to adapt the policy to their specific context.

Restrictiveness limits copying

Policies that included more restrictive language or legally binding mandates were less likely to be copied directly. Adopters seemed to resist the imposition of specific obligations, leading to greater adaptation of these policies during the diffusion process.

Implications for policy-makers

These findings have important implications for policymakers aiming to promote the global spread of corporate sustainability practices. By understanding which textual attributes facilitate or hinder policy diffusion, policymakers can craft their documents to maximize their impact. For instance, simplifying the language of policies and incorporating a degree of flexibility could enhance their adoption across different jurisdictions and sectors.

Conclusion

Our research highlights the importance of looking beyond the actors involved in policy diffusion and considering the intrinsic characteristics of the policies themselves. As the world continues to grapple with environmental and social challenges, ensuring the effective spread of corporate sustainability policies will be key to driving global change.

Read more about the author: Dr Adam William Chalmers

Read the full paper: From Diffusion to Diffuse-ability: A Text-as-Data Approach to Explaining the Global Diffusion of Corporate Sustainability Policy