conflict between the principal and the agent may be due to the greater need to be responsive to a specific case, but being unable to go through the formal policy procedures, due to its presumed overly linear and one-dimensional relationship. To overcome this, collective discretionary room is then important in establishing this response, where policy design and implementation should be a result of long-winded negotiations and granted discretion that should instead be encouraging ‘creative compliance’ when need be.

Having critiqued the PA approach, this essay does not disregard the fact that the PAT has important and relevant points to make, such as how it can be used to rationalize individuals’ ‘coping mechanisms’ that would result in negative effects to fairness and overall policy implementation , and also how this rationalization is useful in understanding and overcoming policy situations of high-conflict and low-ambiguity, where top-down structuring crucial in determining policy and implementation success. This is due to the need of clarifying policy impacts and outcomes for specific policy areas determined by the decisions of the principal. However, this is only relevant if the policy networks and external bodies involved can be properly identified, hence low ambiguity. Unfortunately, most policy implementation cases are indefinitely intertwined and highly ambiguous. So although discretion may be minimised in situations of low ambiguity, policy discretion would still occur indefinitely across all sectors of policy, though the extent to which it is granted or practiced may be subject to its given circumstance.

For further alternative approaches and a deeper analysis to non-compliance and discretion, I refer to Leadbeater and Goss’ (1998) coined term of ‘civic entrepreneurship’, defined as a form of front-line work taking place in the spaces of local governance where traditional organisational structures are breaking down