Did Building the Autobahn Help Hitler Win Public Support?

This study investigates whether infrastructure spending can boost support for authoritarian regimes by examining Nazi Germany’s Autobahn construction between 1933 and 1934. Using newly digitized data on voting outcomes, road construction timing, and geographic proximity for over 3,000 cities, the authors find that support for Hitler increased significantly in areas near new highways. The persuasion effect ranged from 8.7% to 17.2% of previously opposing voters. Economic gains like reduced unemployment did not explain the effect; instead, propaganda—especially via radio—amplified the perceived competence of the regime. The impact was strongest in politically unstable regions of the former Weimar Republic.

Full Citation and Link to Article

Voigtländer, Nico, and Hans-Joachim Voth. 2024. “Highway to Hitler.” American Economic Review 114 (5): 147–83. https://doi.org/10.1257/aer.20210203

Extended Summary

Highway to Hitler: 1,000-Word Summary

Central Research Question

The central question posed by Voigtländer and Voth (2024) is whether infrastructure spending in autocracies can generate popular political support, thereby helping to entrench authoritarian regimes. Specifically, the authors examine whether the construction of the Autobahn in Nazi Germany during 1933–1934 played a causal role in increasing public support for Adolf Hitler’s regime. This inquiry contributes to broader concerns in political economy: can visible signs of economic competence, even in the absence of real economic improvements, influence public opinion and electoral behavior under authoritarian rule?

Previous Literature

The paper builds on several strands of existing literature. One is the role of elections in autocracies, where scholars such as Egorov and Sonin (2014) and Gandhi and Lust-Okar (2009) have explored how electoral processes can legitimize dictatorial regimes. Another is the literature on the economic determinants of political support, especially studies that explore how targeted government spending or propaganda affects voting behavior (e.g., Adena et al. 2015; DellaVigna and Gentzkow 2010). Related work has also examined the effect of infrastructure on economic outcomes, such as Donaldson (forthcoming) on railroads in India and Faber (2014) on highways in China. What distinguishes this paper is its focus on the political—not economic—returns to infrastructure investment, especially in the context of early Nazi Germany’s transformation into a “consensual dictatorship.”

Data

The authors assemble a new dataset combining multiple sources:

• Voting data from over 3,200 towns and cities for three key elections: the March 1933 semi-free election, the November 1933 Nazi-controlled election, and the August 1934 plebiscite on Hitler becoming both President and Chancellor.

• Historical maps showing locations of planned, approved, and under-construction segments of the Autobahn network as of mid-1934.

• Census data from 1925 and 1933 providing socio-economic variables like unemployment, urban population, religious composition, and occupational structure.

• Radio signal strength data (as a proxy for propaganda exposure) provided by Adena et al. (2015).

• Measures of political instability in Weimar Germany’s states, based on the index from Satyanath et al. (2017).

The authors use geographic information systems (GIS) to measure proximity of each city to highway construction and use these distances as a central explanatory variable.

Methods

The core empirical approach is a difference-in-differences design, measuring changes in pro-Nazi voting between November 1933 and August 1934 as a function of proximity to Autobahn construction. The main treatment variable is either continuous distance to the nearest construction site or a dummy for being within 20 km of a site. All vote shares are standardized to allow comparison across elections with different structures (e.g., referendums versus party-list votes).

To address potential endogeneity (e.g., if the Nazis chose highway routes for political reasons), the authors use instrumental variables (IV) based on “least-cost paths” between terminal cities, calculated from terrain data. These paths represent the most economically logical routes for highway construction, conditional on geography. The IV estimates confirm and strengthen the OLS results.

They also explore heterogeneity in effects using interaction terms: for example, between highway proximity and radio signal strength, and between highway proximity and political instability at the state level. Placebo tests are conducted using earlier election outcomes (1924–1933) to verify that results are not driven by pre-existing trends in Nazi support.

Findings/Size Effects

The authors find compelling evidence that proximity to Autobahn construction increased support for Hitler in the 1934 referendum:

• In the most basic OLS specification, moving from 100 km away to 1 km from a construction site increased pro-Hitler votes by 2.8 percentage points.

• IV estimates using least-cost paths produce even larger effects, suggesting that measurement error or political targeting does not drive the results.

• The estimated persuasion rate—the share of previously opposing voters who were converted due to the infrastructure project—ranges from 8.7% to 17.2%, comparable or higher than effects found in other voting studies.

The authors carefully explore alternative explanations. While Nazi propaganda portrayed the highways as solving unemployment, the data show that cities near construction sites did not experience greater reductions in unemployment between 1933 and 1935 than more distant cities. This suggests the economic impact of the highways was more symbolic than real.

Propaganda played a significant role. The impact of highway construction on support for the regime was much larger in areas with stronger radio signal strength. In low-radio-access areas, highway proximity had little effect. In areas with high signal strength, persuasion rates were around 30%—nearly triple the average effect. This supports a propaganda-competence mechanism: visible government action combined with persuasive messaging changed public perceptions.

The effect was also moderated by political context. Highway construction increased support most in states with high levels of political instability during the Weimar Republic. In more stable areas, such as Prussia or regions with long-standing governments, the effects were smaller. This suggests that visible action and propaganda were especially persuasive where citizens were disillusioned with democratic gridlock.

Finally, the authors show that roads already “approved for construction” but not yet underway had similar effects on voting as those already under construction. This highlights the power of expectations and visibility: even the promise of action, amplified by propaganda, was enough to win support.

Conclusion

Voigtländer and Voth provide novel evidence that public infrastructure projects can function as powerful tools for autocratic consolidation, even in the absence of substantial economic benefits. In the case of Nazi Germany, the construction of the Autobahn network helped to legitimize Hitler’s rule during a fragile transition period by creating the illusion of economic recovery and government competence. The effect was significantly amplified by propaganda through radio, and it was especially potent in politically unstable areas.

These findings contribute to several scholarly debates. First, they show how autocratic regimes can use visible infrastructure to project legitimacy, sidestepping the need for actual policy performance. Second, they highlight the importance of media in magnifying the symbolic power of government action. Third, the study shows how early, seemingly technocratic interventions (like road building) can have profound political consequences, helping to convert reluctant citizens into regime supporters.

While focused on a specific historical episode, the implications extend to modern contexts where authoritarian leaders use infrastructure as a performance signal. In an age where democratic backsliding is again a concern, the study is a cautionary tale: roads may not just lead to prosperity—they may also pave the way to dictatorship.