Complexity in institutional reform

• Institutional reforms’ effects are long-run, multi-dimensional, and hence unpredictable. Why are they so prevalent in the world? • Because politicians design reforms not to reach stated goals, but around unstated, often orthogonal or even contradictory, private goals. • High instrumental mismatch leads to incongruous institutions ill-suited to stated goals. Outcomes are good for reformers but not society. • Low mismatch leads to mostly congruous institutions. Stated goals are always achieved. Outcomes are good for society but bad for reformers. • Medium mismatch produces incongruous institutions. Stated and private goals are sometimes achieved. Outcomes for society are generally poor. Why is there so much institutional reform in the world? If institutions are the deep rules of the game that determine how societies are governed, collective decisions taken, and resources mobilized for public purposes, then changing them is bound to have effects that are long-run and multidimensional across politics, the economy and society. Such effects will be unpredictable. Politicians with short time horizons should flee such initiatives, but instead embrace them the world over. Why? Because politicians design reform processes around often unstated private goals that may be orthogonal, or even directly opposed, to a reform’s stated, public goals. We characterize instrumental mismatch as the gap between stated goals and the specific reform instruments politicians deploy. Such reforms lead to incongruous institutions ill-suited to their core purpose, and hence to outcomes that are bad for society. Through 13 case studies from Latin America, India, Rwanda and the UK, we test and refine the theory. High instrumental mismatch leads to incongruous institutions in all our cases. Stated goals are never achieved, but private goals are. This is by design. Incongruous institutions sometimes produce good outcomes for reformers, but not for society. Low mismatch leads to institutions that are mostly congruous. Stated goals are always achieved. This makes sense: politicians’ private goals do not conflict with reform’s stated goals, increasing reform coherence. Ex-post, such reforms are good for society but bad for reformers. Medium mismatch also produces incongruous institutions. Stated goals and private goals are sometimes achieved. Outcomes for society are generally poor – not surprising given incomplete reform design and implementation. A twelfth article mines this evidence to propose three game-theoretic models of institutional change from a complex systems perspective. Taken together, we call this the complexity approach to institutional reform .