Final Report Summary - HHPOLITICS (A Household Finance Theory of Political Attitudes and Political Behavior)
HHPOLITICS combines economic theories of choice and a combination of survey, experimental and administrative data to understand political expectations and attitude formation as well as political behavior. This includes research into the validity of common survey measures used in political economics, the sources and role of expectations in explaining political attitude formation, the links between preferences as measured by choice experiments and economic and political behavior, and the importance of distinguishing, in a fine-grained fashion, between different types of income and wealth when explaining individuals’ political attitude formation and political behavior.
For example, we show what “wealth” is not just wealth. Common measures of wealth, when employed in political science and political economy, conflate all types of wealth, including liquid wealth (e.g. bank deposits) and illiquid wealth (e.g. housing wealth). If hit by an economic shock, individuals can access the former, but not the latter, at least not in the short run. However, common theories of political support for social insurance and redistribution are based on composite measures of wealth that does not take into account its relative accessibility, or liquidity. We have developed survey instruments to measure such differences and have collected a large (+20,000) cross country survey that combines, for the first time, detailed measures of assets and liabilities, from economics, with detailed measures of political attitudes and preferences.
In addition, the project details the ways in which economic expectations and their fulfillment matter for political preferences. In particular, we show how information in one’s social network, measured from detailed population data through administrative registers, affect one’s expectations about unemployment and, in turn, how such revised expectations, resulting exclusively from observing economic outcomes in the social network, affect political preferences for social insurance. Similarly, we show how distinguishing between expected and unexpected economic outcomes, including unemployment and income developments, are key to understanding political preference formation. In particular, we show that experiencing unemployment does not alter political preferences for social and insurance and redistribution is such unemployment was expected; in that case, the expected unemployment was capitalized into current preferences and, as such, not altered subsequently. On the other hand, experiencing unanticipated unemployment and income shocks leads people to revise their political preferences and, in some case, leads to political polarization.
For example, we show what “wealth” is not just wealth. Common measures of wealth, when employed in political science and political economy, conflate all types of wealth, including liquid wealth (e.g. bank deposits) and illiquid wealth (e.g. housing wealth). If hit by an economic shock, individuals can access the former, but not the latter, at least not in the short run. However, common theories of political support for social insurance and redistribution are based on composite measures of wealth that does not take into account its relative accessibility, or liquidity. We have developed survey instruments to measure such differences and have collected a large (+20,000) cross country survey that combines, for the first time, detailed measures of assets and liabilities, from economics, with detailed measures of political attitudes and preferences.
In addition, the project details the ways in which economic expectations and their fulfillment matter for political preferences. In particular, we show how information in one’s social network, measured from detailed population data through administrative registers, affect one’s expectations about unemployment and, in turn, how such revised expectations, resulting exclusively from observing economic outcomes in the social network, affect political preferences for social insurance. Similarly, we show how distinguishing between expected and unexpected economic outcomes, including unemployment and income developments, are key to understanding political preference formation. In particular, we show that experiencing unemployment does not alter political preferences for social and insurance and redistribution is such unemployment was expected; in that case, the expected unemployment was capitalized into current preferences and, as such, not altered subsequently. On the other hand, experiencing unanticipated unemployment and income shocks leads people to revise their political preferences and, in some case, leads to political polarization.