MIRACLE

The first paper using MIRACLE data studies how pre-existing social institutions—measured from colonial-era household registries—shaped the local returns to Korea's national rural development programme, the Saemaul Movement. It shows that these institutions did not simply strengthen or weaken the effects of new infrastructure. They shaped which forms of infrastructure became economically valuable by influencing the development strategies communities pursued and the bottlenecks they faced.

Korea's Gyeongbu Expressway (1968–70) absorbed more than 20% of the national budget and reshaped the country's economic geography. MIRACLE tracks townships along the corridor before, during, and after construction to examine how pre-existing social institutions shaped whether new connectivity led to industrial growth, out-migration, or persistent stagnation.

Canonical dual-economy models, from Lewis (1954) through Harris and Todaro (1970), predict that agricultural productivity gains release labour from farming and push workers toward cities. Korea offers a counterexample: the nationwide rollout of high-yield Tongil rice varieties between 1971 and 1979 generated one of the fastest recorded increases in agricultural productivity, yet rural incomes converged toward urban levels during rapid industrialisation. MIRACLE links crop output, cultivated area, extension timing, and non-farm employment at the township level to distinguish among competing channels: labour reallocation into cities, local demand linkages within rural economies, and capital transfers from the state.

Korea restored 2.8 million hectares of forest between 1962 and 1987 while industrialising from extreme poverty, a trajectory that challenges the Environmental Kuznets Curve hypothesis that ecological recovery follows prosperity. Forests exhibit strong spatial complementarities: isolated patches of trees provide negligible ecosystem services, but the same trees within a contiguous canopy deliver watershed protection, soil stability, and microclimate regulation. Successful reforestation may therefore require spatially coordinated intervention (an environmental analogue to the "big push") rather than marginal incentives alone. MIRACLE links Korea Forest Service spatial archives to township-level economic data to trace how land-use restrictions, coal briquette distribution, and administrative coordination shaped both forest recovery and local economic adjustment. A STEG Small Research Grant is supporting the construction of a 400-township panel for this component.

Korea democratised in 1987 after barely two decades of industrial growth. This pace challenges gradualist accounts of democratic change, but aligns uncomfortably well with modernisation theory's prediction that development itself generates political transformation. MIRACLE provides the micro-data to examine three competing channels at the local level: the diffusion of Protestant missions, which established some of the country's earliest organisations based on broad participation; the Saemaul Movement's village councils, which may have built democratic capacity through collective decision-making under authoritarian rule; and structural transformation itself, through urbanisation, education, and the emergence of a middle class. Because each channel implies a different geography and timeline of political mobilisation, the panel allows them to be tested against one another rather than studied in isolation. Distinguishing among them reveals whether Korea's democratic transition was a rupture or the culmination of institutional changes already under way.