Throughout history, wars and episodes of violent disruption have often been associated with reductions in economic inequality. However, the underlying dynamics and mechanisms behind the reduction in inequality during wars are less understood. This paper explores these dynamics and mechanisms in the context of Norway. We have four main findings. First, we document a large wage decline in income inequality in Norway that occurred during World War II and persisted until the mid-1960s. Second, we show that this decline in inequality was a result of convergence between occupations and across municipalities. Third, we use a newly digitized individual-level panel of official tax records, to estimate the effect of wartime local labor market shocks on employment, earnings, and inequality. Specifically, we test the effects of price shocks of food and forestry product on local labor markets and of wartime investment in industrial infrastructure, airports, and defense installations. Comparing regions that received investments with similar regions that did not, we find short run and long-lasting effects both on employment and income of investments in manufacturing, as well as strong effects on income in areas affected by the enormous wartime price increase of fish, agricultural food, and forest products.

By varying the intensity of compulsory schooling while keeping the number of compulsory school years constant, this paper generates new insights into the effect of human capital on fertility on the extensive and intensive margin. We leverage population-wide panel data for Norway in combination with a school reform in the 1930s changing the instruction time during the school year that allows us to isolate the effect of human capital investment on fertility behavior across the life-cycle from an incarceration effect. We present robust evidence of reduced total fertility driven mostly by an increase in the share of women with one or no children and an increase in the age at first birth by around four months.

High-quality administrative tax data has been instrumental in advancing economic research over the past several decades. However, the scarcity of data prior to the 1970s has posed a significant obstacle to the study of the distribution of economic outcomes across people and generations in historical settings. We introduce a dataset extending Norwegian tax records to 1900 by meticulously digitizing more than 270,000 pages. The dataset includes most Norwegian regions and data on name, occupation, place of residence, earnings, and wealth. We link observations across years to significantly enhance the usefulness of the data. We argue that the methodologies employed in this paper offer significant potential for application to a wider range of sources and countries.

We estimate the complete income distribution in Norway for 1892, 1906, 1913 and 1929. Compared to previous research, we benefit from better data and more advanced estimation techniques. Our thesis identifies several data weaknesses which have caused bias in previous studies. Much of the data previously used does not distinguish between individual taxpayers and impersonal entities such as stock companies and banks. Another weakness is that before 1921, dividends were not included in the income data. For 1929, the data allows us to create local-level estimates for each Norwegian municipality. We find that the pre-tax, pre-transfers Gini index is stable for the years we analyse, starting at 52 percent in 1892 and ending at 54 percent in 1929. The top 1% income share before taxes and transfers falls over time, declining from 19 percent in 1892 to 12 percent in 1929. We find that shocks to wealth might play a role in this development. Our results differ significantly from those of previous studies. First, we find a Gini index lower than Aaberge, Atkinson and Modalsli (2016). Second, we find lower top incomes and a different development over time than Aaberge, Atkinson and Modalsli (2013). Our results suggest that Norway was already among the most egalitarian countries in Western Europe between 1892 and 1929 in terms of income. However, our estimates are sensitive to total income and to the estimation of stock dividends. Historical estimates from other countries are likely to be sensitive too, leaving a considerable risk of error when comparing.