‘At least he built the Autobahn’. Many Germans remember this phrase from conversations with parents and grandparents pointing to how the Nazi regime could receive such widespread support. The regime’s overwhelming popularity at home was essential for its policies, from the aggressive pursuit of war abroad to genocide. The building of Germany’s motorway network has survived in popular memory as a palpable, unambiguously benign accomplishment of the National Socialist government; in retrospect, it serves as a ready explanation for the regime’s genuine popularity from 1933 onwards.
Can infrastructure spending really win ‘hearts and minds’? Are the ready explanations of elderly Germans merely an excuse – or do they capture an important reason for the Nazi regime’s early political success? Empirical studies typically find only weak support for the notion that infrastructure projects increase support for a government – spending is often targeted where it is most needed: in districts where electoral defeat threatens. Examples of papers that find some effect of pork-barrel spending include Levitt and Snyder (1997) and Manacorda et al. (2011). Berman et al. 2011 show that during the US occupation of Iraq, areas with large infrastructure spending saw reductions in violence. On the whole, however, economists and political scientists are skeptical whether pork-barrel spending works in practice (Stein and Bickers 1994, Feldman and Jondrow 1984).
Examining road-building in Nazi Germany is useful because the effectiveness of pork barrel spending under a dictatorship has not been examined systematically (Voigtländer and Voth 2014). Also, the regime’s rise in popularity is not well-explained. Even in March 1933, with the Hitler government already in office, ‘only’ 44% of Germans voted for the Nazi party. Nonetheless, within a few years, all the reports by opposition groups and by the regime’s own security service indicate very high approval rates (Evans 2006). How did the Nazi regime grow in popularity so quickly?
Building the Autobahn
Road-building received highest priority under the Hitler government. Within a few weeks of taking office, the regime introduced subsidies for car ownership and outlined an ambitious programme for the German motor industry. Within six months, a new company charged with building the world’s first nationwide highway system was set up; within nine months, the first stretches of motorway were under construction (Vahrenkamp 2010).
Despite claims to the contrary, military considerations never played much of a role. The Autobahn plans were partly based on plans drawn up by a think tank and a private company that had mapped out road trajectories in the 1920s (the Stufa and the Hafraba projects). The new roads used much of the earlier planning, but deviated in some parts.
Instead of gradually connecting metropolitan areas, road-building began on 17 sites simultaneously, traversing 131 out of the 901 electoral districts. Opening the first stretch of motorway therefore took longer; but its progress was visible in many more parts of Germany. Throughout the planning and building process, Nazi propaganda showcased the Autobahn as an important achievement. The first sods of earth were turned by Hitler himself in September 1933 – less than nine months after coming to office. For the November election, the event provided the main campaign image. Every opening of a new segment turned into mass events, with people lining the roads as the cars of party leaders drove by triumphantly. By 1935, the first stretches of motorway were open; by 1938, 3,500 kilometres of road were complete (Vahrenkamp 2010).
Figure 1 gives an overview of the 1934 state of the project. The initial plan connected major population centers all over the country. Eventually, almost half of all German electoral districts would be traversed by motorways (Phase III: construction planned). By 1934, however, construction was under way in only 15% of them (Phase I: under construction), with others designated as next in line (Phase II: approved).
Figure 1. Autobahn network in 1934
We ask if voters in districts where roads were being built supported the Hitler regime to a greater extent. But how can we know, given that free and fair elections were no longer being held? The regime did hold elections with some frequency to demonstrate its popularity. These were neither free nor fair. Storm troopers were present at the polling stations, and there was pressure to vote in public – voting in secret was optional, but frowned upon. Spoiled ballot papers were often counted as ‘yes’; the old and infirm were transported to the voting stations. In some places, voting papers were marked before being handed out so that voting behaviour could be tracked. Support was very high on average, but it would be absurd to claim that 90% or more of Germans from November 1933 onwards were all fully behind the regime; the true level of support cannot be established from these voting results.
Instead, we argue that we can analyse changes over time – that a significant decline in the level of opposition in one district, over and above general shifts in behaviour, can be taken as an indicator of changing support for the regime. The reason for this approach is that there was important variation at the local level. For example, in the 1934 plebiscite that vastly expanded Hitler’s powers as head of government, 24% of people in Aachen voted ‘no’; in Nuremberg, on the other hand, only 4.6% voted against the government proposition.1
To compare like with like, we examine the reduction in opposition to the regime between two elections – November 1933 and August 1934. In November 1933, new elections for the Reichstag were held, with the NSDAP the only party on the ballot paper. The 1934 plebiscite occurred after the death of President Hindenburg, and authorised combining the powers of Chancellor and of President in the hands of Adolf Hitler.
The impact of road-building
We compare electoral districts traversed by the Autobahn with those without road-building. In general, opposition to the regime declined slightly between November 1933 and August 1934.2 Where a district saw road-building, the shift was markedly larger (Figure 2). We plot the distribution of votes against the regimes; the dashed line shows the pattern for areas with motorway construction, and the solid line, for areas without. As is readily apparent, where the motorways were being built, the level of opposition was lower – the whole distribution is shifted to the left.
Figure 2. Change in votes against the Nazi Regime, November 1933- August 1934, conditional on Autobahn connection
We examine this result statistically, and find a significant decline in opposition. On average, opposition declined by 1.5% in areas where there was no road-building; where the Autobahn was being built, opposition declined by 2.4%, i.e. 60% faster.
Was the impact of road-building causal? Or did the regime simply build roads in areas where its popularity was surging anyway, perhaps in a bid to reward loyal local supporters? To deal with this question, we do two things:
- First, we correct for trends in voting results; our findings do not change.
- Second, we use information on roads planned before 1933. The concern that the Nazis built roads in areas with growing support can be examined empirically by comparing the road trajectories actually built with those planned in the 1920s. Where both coincide, it is not likely that attempts to chase voters influenced results; where the Nazis built a road, but earlier plans had envisioned none, it is possible that local preferences and conditions shaped plans.
In Figure 3, we compare the effect of road-building in different districts. The first two columns show average shifts in all districts – divided into those with and without road-building. It shows the additional ‘swing’ of almost one percentage point. The middle two columns repeat the analysis in the subset of districts where the 1920s plans called for road construction. The results are nearly identical. In other words, in the sample of districts where road-building was a good idea for technical or economic reasons, the effect is just as large as in the dataset as a whole. The final two columns show the treatment effect in areas where only the Nazis planned roads. If the ‘vote chasing’ hypothesis were correct (i.e. if the Nazis built motorways in areas where their popularity was rising anyway), we would expect bigger shifts in favour of the regime. This is not what we find; if anything, the additional reduction in opposition is smaller than in districts where the 1920s plans had already envisaged road building. In other words, where Nazi road-building deviated from pre-existing plans, it created less support for the regime than elsewhere. We therefore conclude that our results in the full sample are unlikely to be affected by strategic highway construction under the Nazis.
Figure 3. The effect of road-building on opposition to the Nazi regime
We also show that the reduction in opposition was largest in electoral districts closest to the motorway construction sites. The further away a district was from where the new roads ran, the smaller the ‘swing’ in favour of the regime.
Why it worked
The Nazi regime prioritised road-building as an economic stimulus measure. Original plans were for 600,000 workers to be employed; the actual maximum figure was 125,000. Recent analysis suggests that aggregate economic effects were modest (Ritschl 1998). The benefits in terms of transport were also minimal – Germany had one of the lowest rates of car ownership in Europe (Evans 2006).
Nonetheless, it is possible that local effects were much larger. Workers were initially housed in private homes in the villages and towns where the roads were being built; barracks were only built later. Those employed in building the road also spent money in inns and shops; construction crews organised film showings, and construction sites became minor local attractions – a popular destination for weekend trips (Eichner-Ramm 2008).
In modern democratic elections, ‘political budget cycles’ may be driven by politicians’ need to signal their competence. Similarly, the Autobahn served as a convincing proof of Nazi Germany’s ability to get things done – a project to showcase the ruthless energy and organizational capabilities of the new regime, as Hitler promised in his speech inaugurating the project. Sold as a key factor for economic revival, the rapid fall in unemployment after 1933 convinced many that road-building had ‘worked’. After the perceived incompetence and gridlock of Weimar politics, many Germans were undoubtedly impressed by the rapid progress in road-building. The propaganda machine took particular care to connect the roads in the public imagination with Adolf Hitler himself – the motorways were called ‘roads of the Führer’, piggybacking off the leader’s popularity and enhancing his image still further. While these effects would have affected voting in the country as a whole, it is plausible that the regime’s accomplishments in building the Autobahn were more salient for voters in districts where the new roads were taking shape (Gennaioli and Shleifer 2010).
Long before the Nazi regime committed its singular crimes, it had become remarkably popular in Germany (Evans 2006). Voting records from 1933 and 1934 reveal the effect of one factor that, according to many historians, boosted support for the regime – the building of the Autobahn. Using detailed information on the geography of road-building, we isolate the effect of construction on voting behaviour by analysing the ‘swing’ in favour of the regime over a nine-month period (November 1933 to August 1934). We find that opposition declined much faster where the new ‘roads of the Führer’ ran.
Direct economic benefits for residents in Autobahn districts may have played a role, but they were probably small. More importantly, the new roads provided concrete proof of the regime’s actions, delivering on its promise to get ‘Germany moving again’. Within a couple of months of taking power, a highly ambitious highway construction project was under way at 17 different locations all over the country, affecting more than 100 electoral districts. In other words, the visible progress of road construction made the regime’s ability to follow through on its promises salient for many Germans.
Combined with effective propaganda trumpeting the regime’s successes, the roads succeeded in winning the hearts and minds of many Germans. Nor were they the only ones to be impressed. When the US Army rolled into Germany at the end of World War II, one of the officers taken with the ease of transport on motorways was Dwight D. Eisenhower. When he became President of the United States, he lead the initiative to built the country’s interstate highway system.
Berman, Eli, Jacob N Shapiro, and Joseph H Felter (2011), “Can Hearts and Minds Be Bought? The Economics of Counterinsurgency in Iraq”, Journal of Political Economy 119 (4): 766–819.
Eichner-Ramm, Britta (2008), 70 Jahre Autobahn Göttingen - Kassel: Zeitzeugen berichten. Göttinger Tageblatt-Verlag.
Evans, Richard J. (2006), The Third Reich in Power. Penguin.
Feldman, Paul, and James Jondrow (1984), “Congressional Elections and Local Federal Spending”, American Journal of Political Science 28: 147–64.
Gennaioli, Nicola, and Andrei Shleifer (2010), “What Comes to Mind”, The Quarterly Journal of Economics 125 (4): 1399–1433.
Levitt, Steven D, and James M Snyder (1997), “The Impact of Federal Spending on House Election Outcomes”, The Journal of Political Economy 105 (1): 30–53.
Manacorda, Marco, Edward Andrew Miguel, and Andrea Vigorito (2011), “Government Transfers and Political Support”, American Economic Journal: Applied Economics 3 (3): 1–28.
Ritschl, Albrecht (1998), “Reparation Transfers, the Borchardt Hypothesis and the Great Depression in Germany, 1929–32: A Guided Tour for Hard-Headed Keynesians”, European Review of Economic History 2 (1): 49–72.
Stein, Robert, and Kenneth N. Bickers (1994), “Congressional Elections and the Pork Barrel”, Journal of Politics 56: 377–99.
Vahrenkamp, Richard (2010), The German Autobahn 1920-1945: Hafraba Visions and Mega Projects. BoD – Books on Demand.
Voigtländer, Nico, and Hans-Joachim Voth (2012), “Persecution Perpetuated: The Medieval Origins of Anti-Semitic Violence in Nazi Germany”, The Quarterly Journal of Economics 127 (3): 1339–92.
Voigtländer, Nico, and Hans-Joachim Voth (2014), “Highway to Hitler”, CEPR Discussion Paper No. 9983.
1 Both cities have sharply different historic traditions, reaching as far back as the Middle Ages – reflected in differences in anti-Semitism, amongst other factors (Voigtländer and Voth 2012).
2 Total votes in the country as a whole against the regime slightly increased. If we analyze results district by district, we find the opposite – the average district had slightly fewer votes against the regime, because we count all districts as equally-important units of observation. Weighted by population, we would also find a slight decline in regime support, driven by voting in a few large cities.