Governance is one of the most used and least defined words in development. Before we can ask whether it causes growth or how to improve it, we need to know what it is and how it is measured — and to be honest about how shaky those measurements are. This module builds the empirical foundation for the course and inoculates you against the most common errors in reading a governance number.
What governance means
A working definition
Governance is the set of traditions and institutions by which authority in a country is exercised — how governments are selected, monitored, and replaced; the capacity of the state to formulate and implement sound policies; and the respect of citizens and the state for the institutions that govern their interactions (Kaufmann-Kraay-Mastruzzi). Note it spans three things: how power is acquired (accountability), what the state can do with it (capacity), and the rules that constrain its use (rule of law). Corruption is one symptom within this larger system, not the whole of it.
The Worldwide Governance Indicators
The most-cited measure is the World Bank's Worldwide Governance Indicators (WGI), which aggregate hundreds of underlying variables from dozens of sources into six dimensions:
- Voice and accountability — can citizens participate in selecting their government; freedom of expression and association
- Political stability and absence of violence — likelihood of destabilisation by unconstitutional or violent means
- Government effectiveness — quality of public services, the civil service, and policy implementation
- Regulatory quality — ability to formulate and implement sound rules that permit private-sector development
- Rule of law — confidence in and obedience to the rules: contract enforcement, property rights, the courts, crime
- Control of corruption — extent to which public power is exercised for private gain
The measurement problem
These are perceptions, aggregated
The WGI and Transparency International's Corruption Perceptions Index are built largely from the perceptions of experts, firms, and households — not direct observation of governance. That creates several hazards: perceptions lag reality and are shaped by news and reputation (a halo or horns effect); sources are correlated (everyone reads the same headlines), so apparent agreement overstates precision; the indices have wide confidence intervals that rankings hide (the difference between the 45th and 55th percentile is usually statistical noise); and comparability over time is weak because the source mix changes. Treat a governance score as a fuzzy band, never a precise rank — and never report a one-place move as if it meant something.
De jure versus de facto
The deepest trap is mistaking the rules on paper for the rules in practice. A country can have an exemplary anti-corruption law, an independent-on-paper electoral commission, and a constitutional bill of rights (strong de jure institutions) while corruption flourishes, elections are managed, and rights are ignored (weak de facto institutions). Acemoglu and Robinson stress that de jure and de facto power can diverge sharply, and that real outcomes track the latter. Reform that changes the paper without changing the practice — passing the law, creating the agency — is the most common form of governance theatre, and the rest of this course is about telling the two apart.
Good enough governance
Merilee Grindle's corrective (Good Enough Governance, 2004) targets a practical error in the governance agenda: the tendency to hand developing countries an enormous checklist of 'good governance' reforms — dozens of dimensions to fix simultaneously — which is both impossible and unprioritised. Grindle argues for sequencing and selectivity: identify the few governance constraints that bind growth and welfare most in a particular country at a particular time, and fix those first. Not all governance deficits matter equally or can be addressed at once; pretending otherwise produces long reform wish-lists that overwhelm limited capacity and achieve nothing.
Exercise
A country's Control of Corruption percentile rises from the 32nd to the 38th over two years, and a minister claims this 'proves the anti-corruption drive is working'. (1) Give two reasons from the measurement critique to be cautious about this claim. (2) The same country has just passed a strong access-to-information law but agencies routinely ignore information requests. Frame this using de jure vs de facto and explain what you'd need to observe to judge real progress. (3) The government asks you to design a governance-reform programme and lists 25 priorities. Apply Grindle's 'good enough governance' to respond. (4) Propose one objective (non-perception) indicator that would more credibly track the anti-corruption drive than the WGI percentile.