Using institutional economics to predict effective government reforms | Mushtaq Khan (2021)

Core Theoretical Framework: Political Settlements and Organizational Capabilities

• Definition of Development: The primary distinction between developed and developing nations is the distribution of organizational capabilities (the ability to organize complex production, manage large numbers of non-familiar individuals, and sustain operations), not merely the existence of laws or markets.
• Institution-Organization Feedback Loop: Institutions (rules) and organizations (firms, parties, networks) interact dynamically; organizations create and distort rules, while rules shape organizational emergence.
• Rule of Law vs. Rule by Law:
  • Rule of Law: Exists when the penalty for rule violation is independent of who the violator is; requires a broad distribution of power where peers check each other.
  • Rule by Law: Exists when rules are enforced only on those unable to evade them, while powerful elites violate rules with impunity; typical in developing nations.
• Horizontal vs. Vertical Enforcement:
  • Vertical Enforcement: Top-down action by police, courts, or anti-corruption commissions; often ineffective in low-capability societies because enforcement agencies are themselves captured or lack credible power.
  • Horizontal Enforcement: Peer monitoring where organizations with equal power check each other; this is the primary mechanism for rule-following in practice (e.g., 90% of enforcement occurs this way).
• The Problem of "Low Capability": Corruption often stems not from moral failings but from a lack of organizational capability to compete globally; when organizations cannot become competitive, they resort to capturing political rents (subsidies) as a survival strategy rather than generating wealth through production.

Industrial Policy and Capability Development

• Necessity of Support for New Industries: Markets cannot spontaneously solve "late development" problems where productivity gaps are massive (e.g., 7:1 or higher); external support is needed to fund "learning by doing" and organizational experimentation.
• Failure of Simple Liberalization: Privatization and liberalization often fail in developing nations because the private sector often shares the same low organizational capabilities as the state, leading to private-sector collusion and capture rather than competition.
• Successful Case: Bangladesh Garments:
  • Success was not due to discovery or cheap wages but a specific policy design where a South Korean firm (Daewoo) received a transferable rent to train a local firm (Desh) in organizational management.
  • The rent was paid directly to the Korean firm for training, not captured by the local firm, ensuring the transfer of knowledge occurred.
  • Once 150 staff learned the system, they spun off into hundreds of new firms, creating a competitive industry.
• Successful Case: South Korea:
  • Industrial policy worked because Japanese colonialism decimated pre-existing horizontal patronage networks, preventing firms from colluding to capture subsidies.
  • The state could credibly threaten to withdraw subsidies from failing firms, forcing them to become globally competitive to survive.
  • Success relied on a "political settlement" where the military regime had the capacity to override powerful local business-political networks.
• Successful Case: India Automobiles (Maruti Suzuki):
  • Instead of upfront subsidies, the government offered a future market opportunity (protected domestic market) contingent on achieving high domestic content.
  • Foreign investor (Suzuki) had a self-interest in upgrading local suppliers to meet quality standards to protect their brand, effectively driving capability transfer without upfront state risk.
• Constraint on Policy Design: Industrial policy cannot simply copy the "South Korean model" in countries with entrenched horizontal networks; policies must be designed to align with the specific distribution of power and capabilities in that context.

Anti-Corruption Strategies: Evidence-Based Approaches

• Critique of Standard Anti-Corruption: Transparency, accountability, and vertical enforcement alone are insufficient because 70–80% of society is informal and the powerful do not follow rules regardless of prosecution.
• Strategy: Enhance Horizontal Checks: Identify sectors where some actors follow rules and others do not, analyze why (often due to market demands or peer pressure), and design policies to replicate those checks for the rule-violators.
  • Example (Skills Training): Corruption (falsifying job placement) occurred because low-capacity firms couldn't utilize skilled workers. Solution: Combine skills training with commercial loans to upgrade firm capabilities, forcing firms to demand real skills to maintain production efficiency.
  • Example (Climate Infrastructure): High corruption in embankments was reduced by designing projects with dual-use benefits (e.g., roads or schools), giving local "powerful" (but poor) communities an immediate stake in monitoring construction quality.
• Exit Strategies: In sectors where everyone is engaged in corruption due to a lack of alternatives (e.g., artisanal oil refining in the Niger Delta), enforcement is futile; the strategy must be to provide alternative livelihoods and infrastructure to break the "networked corruption."
• Development Precedes Rule of Law: While rule of law and development reinforce each other, the causal arrow is stronger from development to rule of law; as organizational capabilities spread, powerful groups demand rule of law to facilitate complex contracting.

Application to Developed Countries and Policy Analysis

• Roots of Populism in the West: The rise of populist movements is linked to the erosion of the "political compact" where the rich needed the working class for production; globalization moved production capabilities abroad, breaking the link that justified welfare states and leading to polarization.
• Addressing "Left-Behind" Regions: Policies focused solely on infrastructure (roads, rail) fail if they do not build local organizational capabilities; effective policy must support mid-level technology and local firm formation rather than attracting large, transient foreign investments.
• Political Economy Toolkit:
  • Context is Critical: General theories (neoclassical economics, "limited access orders") are useful heuristics but must be tested against local history and specific power distributions.
  • Key Questions for Policymakers:
    • Who are the "powerful" actors at the specific level of the policy being implemented?
    • Do any of these actors have an incentive to follow the rules and monitor peers?
    • If not, can the policy be redesigned to create that incentive, or is an "exit strategy" required?
• Methodology: Avoid relying on a single "textbook" model; instead, use a combination of frameworks (e.g., North/Wallis/Weingast, Khan's political settlements) and ground analysis in local empirical reality.
• Organizational Inefficiency is Universal: Even in developed nations, transaction costs are high; the difference is the degree of efficiency and the existence of incentives to reduce them, not the absence of friction.

Key Forward-Looking Statements and Disagreements

• Agreement with North/Wallis/Weingast: Development is a systemic equilibrium, not a pathology of "bad governance," but they overemphasize violence as the primary constraint and underemphasize productive organizational capabilities.
• Disagreement with "Limited Access Order" Theory: Rents are not only created by limiting access to markets; they are often created by opening new markets via policy (e.g., learning rents), and advanced nations also maintain persistent rents through rule-following influence (e.g., patents).
• Prediction for China: As Chinese firms (like Huawei) grow in complexity and number, they will increasingly demand a rule of law to facilitate global contracting, pushing China toward a rule-of-law system similar to South Korea's 2000s transition.
• Advice for Practitioners: Do not expect "big bang" revolutions or magic solutions; success requires incremental, sector-by-sector experimentation to build horizontal enforcement mechanisms that align with local interests.