The Long-Term Benefits of School Spending a Hundred Years Ago and a Surprising Contributing Factor

Investing in public education has long been viewed as an important policy tool for improving future labor market outcomes, especially for disadvantaged youth. In the early 1900s, cities were primarily responsible for this investment.¹ At the same time, many cities experienced rapid population growth fueled by European immigration. As a result, city schools increasingly served immigrant children, including German children.

During World War I (1914–1918), with Germany as the Allies' (and eventually America's) primary enemy, the status of ethnic Germans living in U.S. cities declined rapidly.² In response to rising anti-German sentiment, U.S. city governments increased investments in their public schools. Their goal was to rapidly assimilate German immigrant children by teaching them English and civic duty.³

For their paper, “The Impact of Early Investments in Urban School Systems in the United States,” Allison Shertzer of the Philadelphia Fed and Ethan Schmick of Marquette University examined the rise in public education spending by city school districts after World War I and tracked the outcomes for affected student cohorts decades later. By considering the war-related anti-German sentiment in their empirical approach, they sought to understand how early 1900s education investments impacted students of different backgrounds later in life.

The authors created a new digitized, city-level data set of public school resources for 340 of the largest U.S. cities from 1900 to 1930. To measure outcomes decades later, they matched individual school-aged children from the 1910 and 1920 censuses to their records in the 1940 census.⁴ This allowed them to study how returns to education spending varied by childhood socioeconomic status. And to capture the exogenous variation relating to anti-German sentiment, they developed a statistical instrument based on how the size of the German population differed by city before the war and on how education spending differed by city after the war.

The growth of education spending by cities after World War I, they found, “marked a significant departure from nineteenth century levels.” Between 1920 and 1924, average real expenditures rose from $79 to $143 per pupil, an 81 percent increase.⁵ Spending increased across all three main categories: teachers and supervisors, operations, and capital. But capital and operations expenditures grew the fastest, thanks to a burst of new school construction.⁶ This rapid rise in school spending did not occur again until the 1960s.

Their “German share” instrument was predictive of future increases in spending: Cities with a larger share of German immigrants before the war spent proportionately more on public schools after the war. No such predictive relationship was found, however, when they looked at the overall immigrant population, indicating the prominent influence of anti-German sentiment on school spending.

Using this instrument, they show that increased education spending significantly improved outcomes in terms of both educational attainment and wages later in life.⁷ A 10 percent increase in expenditures per pupil during their school-age years increased their schooling by about one month and raised their wages in adulthood by 1.6 percent. Overall, investments in education account for about 40 percent of the rise in educational attainment for cohorts born between 1895 and 1913. These investments benefited both native-born and immigrant children. The authors largely attribute these benefits to the wide appeal of the newly constructed schools, which motivated children to stay in school longer.

The investments in school spending were especially beneficial for children from lower socioeconomic backgrounds, regardless of their place of birth or family origin. For children of blue-collar fathers, the effect on primary school completion was three times larger and the effect on wages four times larger than for children of white-collar fathers. Thus, these investments “may have played an important role in the midcentury decline in inequality in the United States.”

Shertzer and Schmick’s paper provides evidence that World War I set the stage for a substantial increase in U.S. cities’ public school spending. It also shows that public investment in urban education driven by an exogenous shock — specifically, anti-German sentiment — had a meaningful impact on educational attainment and future earnings, particularly for disadvantaged students.