This paper presents empirical analysis of the long-term impact of highly educated Jewish scientists who had fled Nazi Germany on scientific innovation in the United States (US).
In 1933 the Nazis introduced legislation requiring all non-Aryans to be removed from the civil service, leading to the summary dismissal of Jewish academics. After the annexation of Austria in 1938, dismissals were extended to Austrian universities. Approximately 2,400 German Jewish academics were among the roughly 133,000 German Jewish émigrés who, by 1944, had fled Nazi Germany and had sought refuge in the United States.
The analysis is based on datasets covering: (a) aggregate changes in US patents between 1920 and 1970 across research fields, which were differentially affected by the arrival of German Jewish chemists; and (b) changes in patenting for individual US inventors across research fields. The authors employ a difference-in-difference methodology, comparing changes in US patenting by US inventors in research fields of German Jewish émigrés with changes in US patenting by US inventors in fields of other (non émigré) German chemists. This approach controls for a potential increase in US invention in fields where German chemists (who dominated chemical research in the early twentieth century) were active inventors. They also control for variation in patenting at the level of research fields and years (e.g. due to scientific breakthroughs, variation in the speed of invention across the life cycle of a technology, and potential declines in innovation due to competing firms agreeing to pool their patents) as well as patenting class and year fixed effects.
Ordinary least squares (OLS) estimates imply that German Jewish chemists who sought refuge in the US led to an increase in US patenting by US inventors of at least 31 percent per annum after 1933. In technology classes that include at least one patent by a German Jewish émigré, US inventors produced at least 75 additional patents per year after 1933, compared with an average of 241 patents per class and year in technology classes with patents by other German chemists.
To investigate the timing of the increase in US invention, the authors estimate the difference-in-differences coefficient separately for each year, allowing it to be different from zero before 1933. Estimates of annual coefficients indicate that the observed increase in patenting cannot be explained by differential pre-trends. Annual coefficients are close to zero before 1933 and increase to the highest level in the 1950s and early 1960s. These results indicate that unobservable factors that preceded the arrival of the émigrés are unlikely to have been the driving force behind the increase in US patenting.
To examine whether OLS regressions reliably estimate the émigrés’ effects, the authors also apply an instrumental variables (IV) approach, using the pre-1933 fields of dismissed German Jewish chemists (determined before the Nazi rise to power) as an instrument for the fields of émigrés to the US. This controls for the possibility that émigrés selected to work in research fields in which US inventors would become more productive, or, more likely based on historical evidence, that selection into research fields was negatively affected by anti-Semitism in the US, which restricted access to the most promising fields.
Consistent with historical accounts of negative selection, instrumental variables estimates imply a 71 percent increase in patenting, which implies that OLS results underestimate, rather than overestimate, the true effects of émigrés on US invention. Compared with a mean of 241 patents per class and year between 1920 and 1970 in research fields of other German chemists, US inventors produce 170 additional patents per class and year in research fields of German Jewish émigrés.
The paper identifies three potential mechanisms for these results:
The arrival of German Jewish émigrés increased US invention by attracting a new group of domestic US inventors to the fields of émigrés, rather than by increasing the productivity of incumbent US scientists;
The émigrés’ effects on US innovation may have been amplified by networks of co-inventors who became active patentees in the fields of émigrés especially after 1940, and continued patenting through the 1950s; and
Co-inventors of co-inventors of the émigrés also increased inventive activity in émigré fields after 1933, and remained substantially more productive throughout the 1950s and 1960s.
These patterns suggest that a natural delay in the transmission of knowledge from German Jewish chemists to US inventors influenced the timing of the increase in US invention. Additionally, comparisons with patent data for a younger group of less prominent German Jewish scientists indicate that the fields of émigré professors may be a good proxy for the fields of a broader flow of German Jewish émigrés, which caused the observed increase in US invention.
The authors conclude that German Jewish émigré professors helped to increase US invention in the long run, by training a new group of younger US scientists, who then continued to train other scientists.