The first mistake: treating "resistance" and "incapacity" as the same thing
Reform teams tend to interpret low performance as one of two stories. The first story is moral: people resist reform because they are corrupt, conservative, or irrational. The second story is managerial: people would implement reform if they had enough capacity, training, and tools. Both stories are sometimes true. What damages strategy is when institutions commit to one story too early.
Merilee Grindle's work on reform emphasizes that reform is not simply about obstacles; it is about navigating them — understanding the political setting, choosing intelligently, and managing the process of implementation. Her "good enough governance" thinking pushes the same realism: reformers must reconcile what is technically desirable with what is politically feasible and institutionally doable, often through prioritization and sequencing rather than maximal reform packages.
The practical point: capacity constraints and political resistance are different problems. If the system lacks operational capability — staffing, data, coordination routines, operational budgets — then "anti-resistance" tactics can create compliance on paper and failure in practice. If the system is politically resisting because reform threatens rents or power, then "capacity building" can simply help opponents implement the reform in name while neutralizing it in substance.
"So the first political economy question is not 'Is the reform good?' It is: what is the binding constraint — resistance or incapacity — and how do we know?"
The second mistake: assuming formal rules define real power
A technically sound reform usually arrives as a formal design: a new law, a new organigram, a new procurement procedure, a new digital system. Formal change is visible and legible. It is also often the least decisive layer of institutional reality.
Helmke and Levitsky show that informal institutions — norms, patronage, clientelism, unwritten rules — can shape behaviour more strongly than formal rules. Reforms frequently "succeed" formally while failing functionally: new anti-corruption bodies exist but cannot prosecute powerful actors; civil service reforms exist but recruitment remains politicized; competitive procurement exists but bid-rigging persists through informal networks.
Mushtaq Khan's political settlements perspective pushes the same logic: institutions that work well in one context can perform poorly in another because the distribution of organizational power shapes enforcement, compliance, and the actual effects of rules. "Good institutions" are not simply imported; they are embedded in a settlement of power.
Once you see this, reform design changes. You stop asking only "What does the law say?" and start asking: who can block, distort, or selectively enforce the law — without being punished?
Incentives, veto players, and why "rational" reforms still lose
Even when reform is broadly popular in principle, specific reforms create losers. This is unavoidable because reform changes allocation: budgets, jobs, discretion, rents, access, authority. Political economy analysis treats this as the core reason "technical" reform becomes politically fragile.
The veto player framework provides a structured way to think about reform blockage. Tsebelis's work argues that policy change depends on the number and alignment of actors whose agreement is necessary; the more veto players and the more divergent their preferences, the harder major reform becomes.
This isn't only about parliaments or formal checks. In many environments, veto power is informal: security actors, business networks, party machines, unions, regional brokers, administrative elites. A reform can be technically perfect and still be impossible if it threatens a powerful coalition's core interests.
A useful rule: reforms fail less because people misunderstand them and more because people understand them very well. They understand who loses power, who loses rents, and who becomes accountable. That is not a reason to abandon reform. It is a reason to treat reform strategy as coalition-building and incentive engineering, not only as technical optimization.
Implementation is a political process, not a downstream technical stage
Pressman and Wildavsky's classic work on implementation showed why programs can fail even when they appear rational: implementation involves many decision points, many actors, and many dependencies — each one a potential failure node. Policy does not move as a single decision; it moves through a chain of agreements and actions, and each link is political.
That chain is shaped by discretion. Michael Lipsky's "street-level bureaucracy" perspective points out that frontline officials effectively become policymakers because they interpret vague rules under resource constraints and competing pressures. Even if the reform is not "resisted" at the centre, it can be reauthored at the front line through everyday coping strategies: rationing services, prioritising some cases, informal shortcuts, quiet non-compliance.
Why reform designs often fail even when politics is understood
Even when reformers do political economy analysis, reforms can still fail because the design implicitly assumes a state capacity and social legibility that does not exist. James Scott's Seeing Like a State captures a recurring pathology: states and reformers simplify complex local realities to make them administratively legible, often suppressing local knowledge and adaptive practices in the process. The result can be elegant systems that misfire because they are built for a simplified model of society rather than society itself.
Political economy realism therefore has two layers: the power layer (who wins/loses, who can block), and the cognition/legibility layer (what the reform assumes about how the state sees and controls reality).
What a political economy perspective actually changes
It changes the unit of analysis. Instead of treating reform as a technical package that must be implemented "correctly," you treat reform as a contested redistribution of authority and resources that will be reshaped by incentives, informal institutions, and discretionary implementation.
That implies different reform practice: you prioritize feasibility and sequencing, not maximal design; you assume formal rules will interact with informal rules; you map veto players and coalition constraints explicitly; you plan for implementation as a chain of political negotiations; you design for iteration and learning to avoid capability traps; and you treat diagnostics as problem-driven and operational, not academic.
"Technically sound reforms fail for reasons that are usually visible — if you look in the right place. They fail because reform is not a technical upgrade to a neutral system. It is a political act imposed on an institutional ecology of incentives, informal power, veto players, and discretionary implementers."