Figures & data
Table 1. Descriptive statistics and mean comparison tests by urban vs. rural Panel A of Table 1 describes all the variables that are used in our analyses and defined in Appendix A Table A1. Panel B provides mean comparison tests with unequal variances by urban and rural firms. The semi-annual investor data are obtained from the security register center of euroclear in Sweden between 2006 and 2010, and data with market prices and firm-level information are obtained from datastream
Table 2. Capital structure, locally biased ownership, and abnormal returns. The table presents the results from capital structure (CS) and abnormal return (AR) regressions. Our main variables of interest are i) firm-level overall local ownership, ii) local ownership in firms by locally biased owners (Local-local-bias), and iii) local ownership in firms by individuals who have made a local investment but hold an unbiased portfolio (Local-nonlocal bias). The urban dummy indicates whether the firm is headquartered in an urban region and interacts with the overall local ownership measure. Intercept, year and firm fixed effects are included, standard errors are clustered on firms, and t-values are shown within parentheses. The coefficients are marked with ***, **, * for 1%, 5%, and 10% significance levels, respectively
Table 3. Results from instrumental variable approach. The panels revise the previous analyses by using an instrumental variable approach in a 2SLS estimation. The analyses consider the time varying preferences of investors and use internal instruments, one period and two period lagged values, for local ownership variables. Panel A revises capital structure analyses to test hypothesis 1. Panel B tests hypotheses 2a and 2b by examining the urban vs. rural dimension, and Panel C revises abnormal returns regressions to test hypothesis 3. In the models, time fixed effects are included, and standard errors are clustered in both firm and firm districts. The t-values are shown within parentheses, and the coefficients are marked with ***, **, * for 1%, 5%, and 10% significance levels, respectively
Table 4. Capital structure, locally biased ownership, and abnormal returns: Inflation-adjusted values. This table revises the analyses in by using inflation-adjusted variables. Continuous variables in absolute terms such as total assets, fixed assets, etc., are deflated by the consumer price index (CPI in decimal form) by using the first period in our data as the base year (period). The relative measures such as return, ownership, etc., are deflated by the inflation rate that is calculated based on our CPI measure
Table A1. Definitions of the variables
Table A2. Descriptive statistics of the inflation-adjusted variables. Continuous variables in absolute terms such as total assets, fixed assets, etc., are deflated by the consumer price index (CPI in decimal form) by using the first period in our data as the base year (period). The relative measures such as return, ownership, etc., are deflated by the inflation rate that is calculated based on our CPI measure. The table shows descriptive statistics of these deflated variables. Panel A describes all the variables and Panel B provides mean comparison tests with unequal variances by urban and rural firms using inflation adjusted values
Table A3. Results from instrumental variable approach using inflation adjusted variable. The panels revise the previous analyses of 2SLS estimation in by using inflation adjusted variables. The analyses consider the time varying preferences of investors and use internal instruments, one period and two period lagged values, for local ownership variables. Panel A revises capital structure analyses to test hypothesis 1. Panel B tests hypotheses 2a and 2b by examining the urban vs. rural dimension, and Panel C revises abnormal returns regressions to test hypothesis 3. In the models, time fixed effects are included, and standard errors are clustered in both firm and firm districts. The t-values are shown within parentheses, and the coefficients are marked with ***, **, * for 1%, 5%, and 10% significance levels, respectively
Data availability statement
Due to the nature of this research, participants of this study did not agree for their data to be shared publicly, so supporting data is not available.