The consequences of a trade collapse: Economics and politics in Weimar Germany

What are the political consequences of de-globalization? We address this question in the context of Weimar Germany, which experienced a 67% decline in exports between 1928-1932. During this period, the Nazi party vote share increased from 3% to 37%. Using newly digitized data, we show that this surge was not driven by the direct effects of the export decline in manufacturing areas. At the same time, trade shock-induced declines in food prices spread economic hardship to rural hinterlands. We document that this indirect effect and the pro-agriculture policies put forward by the Nazis are instead key to explain their electoral success.

JEL Classifications: F14, F44, N14, N44

Keywords: Great Depression, Trade, Extremism, Nazi Party, Germany

The General Agreement on Tariffs and Trade-World Trade Organization system has successfully promoted trade integration in the post-World War II era. Over the past decade, this progress has been undermined by a new wave of protectionism that has gradually been embraced by a number of countries around the world. As geopolitical tensions continue to build, the risks that protectionism will become entrenched are very high.

As this process will play out over the next decades, many scholars have looked at the past to gain insights into what might be the future consequences of de-globalization. While much progress has been made in this area, we do not yet fully grasp whether and how it can impact the rise of extreme political parties. Understanding this is important, as these political forces have been gaining ground in many Western countries.

This paper tackles these issues focusing on Weimar Germany, a country that faced a 67% decline in demand for its exports between 1928-1932, equivalent to an 8% drop in GDP. Over the same period, democracy collapsed, with Hitler’s Nazi party growing from 2.6% of the suffrage in 1928 to 37.3% in 1932. To what extent was the rise in the Nazi party due to the trade collapse? What were the roles of sectoral linkages in explaining this?

To answer these questions, we combine newly digitized data on employment by industry and county from the 1925 Census and exports by product between 1928–32 from the German Trade Statistics. The concentrated and well-delineated economic geography of this export shock allows us to explore the linkages between economic and political outcomes across Germany at the time. An important empirical concern is that the export shock might be shaped by factors within Germany, which also affect other economic and political outcomes. To overcome this challenge we use newly digitized product-level trade data on US imports from Germany’s main competitors (France and the United Kingdom) to build a plausibly exogenous instrument.

Our analysis proceeds in three steps. First, we assess the effects of the shock on economic hardship across German cities. We find that greater exposure to this event led to a drop in local economic activity, as measured by an array of outcomes, including electricity consumption, local commuting as well as income-, corporate- and consumption-tax revenues. The export shock also caused an increase in unemployment in cities more exposed to it. In addition, we document that it had indirect repercussions across Germany. In particular, as cities’ local economies collapsed, the price of food items produced in their agricultural hinterlands declined.

Second, using detailed election data, we examine the political consequences of the trade collapse. We find that areas more exposed to the export decline experienced a smaller increase in support for Hitler’s party, as political forces representing the economic interests of middle class voters held their ground. Furthermore, we also document gains for the Communist party, especially outside its traditional strongholds. A plausible explanation for this result is that the economic policies proposed by the Nazi party – draconian cuts in unemployment benefits and their replacement with “work and bread” infrastructure-building programs – did not appeal to important sections of the electorate (Childers 2010). To further explore this idea, we construct occupation–specific measures of exposure to the shock, distinguishing among white–collar, blue–collar and self–employed workers. Our findings indicate that areas where the shock was especially severe for white–collar workers experienced the largest decline in Nazi party support. Areas in which instead the shock affected mainly blue–collar workers saw a modest increase in support for Hitler’s movement, whereas no clear pattern emerges for self–employed workers. These results are in line with the anecdotal evidence suggesting that white–collar manufacturing workers had a lot to lose from the economic policies proposed by the Nazis, as they would have undermined their higher economic standing, based on performing non-manual labour and enjoying better overall working conditions. At the same time, blue–collar workers were likely to find the proposed programs more attractive as they focused on providing unskilled employment opportunities. In additional results, we also document that the decline in Nazi support due to the shock was particularly strong in the presence of a high share of female workers – consistent with the idea that working women found the Nazi party’s rhetoric about their purely domestic role in society not particularly attractive.

Third, we explore the indirect consequences of the export decline. To this end, we develop gravity-based measures of exposure to locations that were directly impacted by the shock, based on their demand and trade costs. Building on earlier work by Wolf (2009), we construct these measures in two ways: First, we use effective geographic distance. Second, we exploit the effects of internal administrative borders. We uncover that indirect exposure to the export shock increased support for the Nazi party. Exploring the mechanisms behind this result, we show that the effect occurs through urban-rural linkages, i.e. the “indirect” rise in support for the Nazi party is solely driven by agricultural areas. Why did farmers exposed to economic hardship react so differently? Once again, the Nazi’s policy platform is a key factor. The economic promises of Hitler’s party to farmers and agricultural workers were particularly generous, including price controls, a moratorium on foreclosures, and state support.

An important lesson that emerges from the analysis is that de-globalization per se might not necessarily lead to an increase in support for radical parties: rather, the proposed policy response to deal with the aftermath of the shock plays a key role. As a result, understanding how economic shocks and policy proposals play out across different subgroups of society is of paramount importance. While we focused on a historical episode, our findings are relevant also for today’s policymakers, dealing with the consequences of the ongoing trade war.

Read Working Paper 010