We measure the economic effects of urban growth boundaries or greenbelts that prohibit new construction beyond a predefined urban fringe. We focus on England, where 13% of the land area is designated as greenbelt land. Using spatial differencing, we show that the external effects of these regulations are substantial (about 15-20%) but very local. In contrast to the previous literature, we find no evidence for internal or 'own-lot' effects. We further show that supply effects are important: greenbelt policy reduces housing construction in greenbelts by about 80%, thereby increasing prices throughout the housing market by about 4%. We show that greenbelt policy implies a negative welfare cost of about £ 7.5 billion a year (0.5% of England's GDP). We further find evidence that greenbelts are no popular recreational destinations (proxied by geocoded pictures), and do not imply longer commutes or more housing CO2 emissions.
