Political Reforms and the Decline of Mortality: Evidence from Swiss Cantons (1870-2000)

欧洲历史上最引人注目的人口变化之一是死亡率的下降。德国人的预期寿命从1871年的37岁增加到1933年的61.3岁；在瑞士，寿命从1875年的40岁左右增加到1945年的63.4岁。在这个研究项目中，我们分析了政治改革在多大程度上促进了这种死亡率的下降：比例代表制的普及和选民通过直接选举、倡议和公投直接参与政治决策。我们的实证分析将集中在瑞士，这是唯一一个地方政府可以选择其选举制度的国家。最重要的是，我们看到随着时间的推移，政治制度发生了相当大的变化。例如，在1890年至1945年间，25个州中有17个改为比例代表制。我们在差异中的差异框架中利用州一级政治改革的时机来估计政治制度对短期和长期年龄和特定原因死亡率的因果影响。为了进行实证分析，我们将收集1870年至2000年间所有州的死亡率、公共财政和政治制度的档案记录并将其数字化。我们还将利用宪法改革和邻国改革等工具变量，作为直接民主和比例代表制的工具。然后，我们研究了政治制度如何影响死亡率的机制，特别是通过它们对公共卫生投资或受教育程度的影响。根据我们的估计，我们计算出死亡率下降的多少可以用比例代表制、行政直接选举和直接民主的转变来解释；以及公共卫生和教育投资的作用。作为我们分析的最后一步，我们使用1850年至1950年间12个欧洲国家的跨国数据集，为我们结果的外部有效性提供了启发性证据。综上所述，本研究项目将对文献做出三个重要贡献：(1)我们的研究首次分析了死亡率下降的根本政治原因；(2)确定政治改革影响人口健康的关键机制；(3)我们探讨了比例代表制和直接选民参与是否对社会福利有有益的影响——超出了文献中讨论的对公共财政的影响。我们的实证结果将为研究选举制度和直接选民参与的福利效应的理论模型提供见解。我们的研究结果还将为发展中国家提供重要的经验，指导他们如何设计政治制度，以改善教育等公共产品的有效提供。我们收集的数据将在项目完成后公开提供。

One of the most spectacular demographic changes in European history is the mortality decline. Life expectancy in Germany increased from 37 years in 1871 to 61.3 years in 1933; in Switzerland, it increased from around 40 years in 1875 to 63.4 years in 1945. In this research project, we analyze to what extent political reforms contributed to this mortality decline: the spread of proportional representation and direct voter participation in political decisions through direct elections, initiatives and referendums.Our empirical analysis will focus on Switzerland, the only country where subnational governments can choose their electoral system. Most importantly, we see considerable variation in political institutions over time. For example, 17 of the 25 cantons switched to proportional representation between 1890 and 1945. We exploit the timing of political reforms at the canton level in a difference-in-differences framework to estimate the causal effect of political institutions on age- and cause-specific mortality in the short- and long-run. For the empirical analysis, we will collect and digitize archival records on mortality, public finances and political institutions for all cantons between 1870 and 2000.We will also employ instrumental variables using constitutional reforms and reforms in neighboring states as an instrument for direct democracy and proportional representation. We then investigate the mechanisms how political institutions influence mortality, in particular through their effects on public health investments or educational attainment. Based on our estimates, we then calculate how much of the mortality decline can be explained by shifts to proportional representation, direct elections of the executive and direct democracy; and what role investments in public health and education play. As a final step of our analysis, we provide suggestive evidence on the external validity of our results, using a cross-country dataset for 12 European countries between 1850 and 1950. In sum, the proposed research project will make three important contributions to the literature: (1) Our study is the first to analyze the fundamental political causes of the mortality decline; (2) we identify crucial mechanisms through which political reforms affect population health; and (3) we explore whether proportional representation and direct voter participation have beneficial effects on social welfare - beyond their effect on public finances discussed in the literature. Our empirical results will provide insights for theoretical models studying the welfare effects of electoral systems and direct voter participation. Our findings will also provide important lessons for developing countries on how to design political institutions that improve the effective provision of public goods like education. The data we collect will be made publicly available once the project is completed.