This document presents data and programs used to produce empirical evidence presented in the article and in the Online Appendix. Codes for this paper were run using Stata. The geographic information software used for Figure 2 is ArcGIS. Data folder The Data folder contains data for replication. The datasets used for the analysis contain historical Census data merged to present-day Census or HILDA survey data. 3 main datasets are used for the analysis in the paper: Census_Hist_final.dta: 2011 Population Census of Australia linked with historical Census. Hilda_Hist_all_final.dta: Merged waves 1, 2, 3, 4 of The Household, Income and Labour Dynamics in Australia (HILDA) linked with historical Census (for more information and HILDA source files: http://melbourneinstitute.unimelb.edu.au/hilda) Partner_Hist_final.dta: HILDA dataset linked with historical Census and augmented with respondent’s partner’s characteristics In addition the Data folder contains: • historical data.dta includes historical data for Panel A of Table 1 of the paper. • Historic_panel_reg.dta: The historical panel described in Table 1 of the Online Appendix and used for regressions in Table 2 of the paper. • HistoricalCrossSection_reg.dta: data for Table 3. • Historical Population Aggregate_ ABS.dta: This file is used for Figure 1. It contains the historical population aggregate from the ABS website (http://www.abs.gov.au/ausstats/abs@.nsf/mf/3105.0.65.001). • Census_Hist_1933.dta for Tables A7 and A8. Further detail on the datasets and a description of how the datasets are built are provided in the Online Appendix. The “Trans” subfolder in “Data” contains intermediate datafiles that are created during the analysis (in particular when running the Appendix Tables replication analysis). Dofiles folder The Dofiles folder includes programs necessary for replication of analyses in the paper and online Appendix. • Table1.do: Table 1 in paper. • Table2.do: Table 2 in paper. • Tables3to8.do: Tables 3 to 8 in paper • AppendixTables.do: Appendix tables • Figure1.do: Figure 1 Two additional dofiles are necessary to run the analysis in the paper and Appendix: cmgreg.do from Doug Miller to correct for small number of clusters (http://faculty.econ.ucdavis.edu/faculty/dlmiller/statafiles/) distance.do to compute distances to main ports (Tables A3 and A4). Out folder: This is the directory where all exported stata output (mostly excel tables) is stored.