State Capacity as an Organisational Problem: Evidence from the Growth of the US State over 100 Years
| Researchers: |
Nicola Mastrorocco (University of Bologna and TIME) with Edoardo Teso (Northwestern Kellog and CEPR) |
| Partners: | TBA |
| Location: | US |
| Sample: |
about 1 million employee-year observation |
| Timeline: | 1817-1905 |
| Theme: |
Political Economy, History, Development |
We study how the organization of the state evolves over the process of development of a nation, using a new dataset on the internal organization of the U.S. federal bureaucracy over 1817- 1905. First, we show a series of novel facts, describing how the size of the state, its presence across the territory, and its key organizational features evolved over the nineteenth century. Second, exploiting the staggered expansion of the railroad and telegraph networks across space, we show that the ability of politicians to monitor state agents throughout the territory is an important driver of these facts: locations with lower transportation and communication costs with Washington DC have more state presence, are delegated more decision power, and have lower employee turnover. The results suggest that high monitoring costs are associated with small, personalistic state organizations based on networks of trust; technological innovations lowering monitoring costs facilitate the emergence of modern bureaucratic states.
Local News and Police Behaviour in the United States
| Researchers: | Nicola Mastrorocco (University of Bologna and TIME) with Arianna Ornaghi (Hertic School and CEPR) |
| Partners: | TBA |
| Location: | United States of America |
| Sample: | 8.5 Local TV News Stories in 300000 Newscast |
| Timeline: | 2017-2020 |
| Theme: | Political Economy, Urban and Migration |
Do U.S. municipal police departments respond to news coverage of local crime? We address this question exploiting an exogenous shock to local crime reporting induced by acquisitions of local TV stations by a large broadcast group, Sinclair. Using a unique dataset of 8.5 million news stories and a triple differences design, we document that Sinclair ownership decreases news coverage of local crime. This matters for policing: municipalities that experience the change in news coverage have lower violent crime clearance rates relative to municipalities that do not. The result is consistent with a decrease of crime salience in the public opinion.