Our current Keynes Trust-funded grant has laid the empirical foundations for an analysis of comparative coastal light provisions in the UK and France 1680-1911, which we will effect in the proposed project. The previous grant has successfully delivered on all data-set creation objectives, and produced extra resources, which now puts us in an ideal position to fully analyse the data we have produced and make a very significant contribution to the Cosian debate (which uses lighthouse economics as a litmus test of market failure in the provision of public goods).

This project will also set an ambitious research agenda for years to come with a key impact on multiple fields in economics and economic history (comparative Franco-British development, historical shipping, the role of publicly funded infrastructures in productivity growth, the role of the state in the protection of working conditions) using highly innovative methodologies from computational geospatial modelling and urban planning/economics.

For economists, including two Nobel laureates, lighthouses have become emblematic features of the argument on market failure in the provision of public goods. Despite this, rigorous empirical assessments have been hampered by the limited historical data available. All the datasets for England and Wales and most for France have now been created, and we are applying for follow-up funding to: i) complete the innovative methodological work started in our previous grant to provide the first rigorous quantitative comparison of France and Britain, ii) add comparable data for Scotland and Ireland as during the early nineteenth century more lighthouses were built in these parts of the U.K. than in England and Wales, and iii) publish our findings on the issue of market failure and light provision.

Published datasets

• Historical sailing routes (available as shapefiles) to and from 150 ports across Europe. The repository also contains all underlying data for the simulation (wind, waves, current, gusts, visibility cost surfaces). DOI: 10.5281/zenodo.8013873 (embargoed until January 2024).
• Historical wrecks database: a geodatabase containing the location of over 189,000 wrecks identified and geolocated. DOI: 10.5281/zenodo.8014019 This dataset has a restrictive licence since it uses material under copyright from several maritime agencies.

Unpublished datasets

• SCOT_REVS: light duty revenues (individual light duty income, and where available), 1680-1911, c.230 obs.
• SCOT_LIGHTS: extends our current data for Scotland back to 1680.
• IRL_REVS: light duty revenues (individual light duty income, where available), 1680-1911, c.230 obs.
• IRL_LIGHTS: c. 350 lights georeferenced with build dates and ranges.
• LOGBOOKS_XLS: a dataset containing all extracted logbook data.
• SCOT_WATERWAYS_SHP: a polyline dataset of all navigable waterways for Scotland 1680-present.
• SCOT_ ROADS_SHP: three topologically correct feature classes containing road vectors for all Scottish roads in 1680, 1830, and 1911.
• SCOT_ RAIL_SHP: a polyline dataset of the Scottish rail network from 1826 to the present.

Papers under review

• Litvine, A.D., Starzec, A., Lewis, J. ‘A new model for coastal routing based on probabilistic least cost-path analysis’. This article has now been submitted to Nature.

Working papers

• Litvine, Bogart, D., Shaw-Taylor, L., Dunn, O., Alvarez-Palau, E., ‘The coastal lighting revolution in England and Wales.’ We have completed the methodology, and built a model, but we recently discovered an issue in the underlying data, which we are currently fixing. We hope to have a first draft on the WP ready in the Autumn, which could be made available as a working paper in the Cambridge Working Papers in Economics Series.

Other outputs

• This project has spawned an honours thesis, and an 'Undergraduate research opportunity Project' at UC Irvine. Tae Young found some surprising results with some preliminary lighthouse and shipping data. With his permission, his poster could be added to our result page, if desirable.