VoxEU Column Economic history International trade Productivity and Innovation

Early 20th century American exceptionalism: Production, trade and diffusion of the automobile

As the current narrative goes, the loss of US manufacturing jobs is due to competition from China and one way to get the jobs back is by running tariffs up the proverbial flagpole.  This column argues that in the case of the automobile industry, history shows exactly the opposite occurred. In the early 20th century, the US achieved exceptionalism in innovation, production and trade in automobiles without domestic tariff protection, while foreign nations languished behind high tariff walls designed to protect their fledgling domestic automobile industries. 

The erosion of US exceptionalism in the automobile industry is obvious in Figure 1, reproduced from Goolsbee and Krueger (2015). The figure shows the fraction of US total vehicle sales accounted for by the big three: GM, Ford, and Chrysler. The total share has been in an uninterrupted free-fall since the 1960s. The share collapse from 90% to 60% is breath-taking and reflects a combination of increasing import penetration and a rising share of production by foreign multinational companies within the US.  When interpreting this figure in the context of job creation and destruction, it is essential to keep in mind that factories built by Ford in Michigan and by Nissan in Tennessee each employ US workers; foreign multinationals are a large part of the flipside of Figure 1. 

If economic exceptionalism means winning the gold medal in the automobile manufacturing Olympics, based on recent market shares of US domestic sales, Toyota would take the gold and Ford and Nissan the silver and bronze, respectively.1  Note, we have perhaps unfairly disqualified GM and Fiat-Chrysler for the use of performance-enhancing public bailouts during the Great Recession.2  The ranking could also turn out differently if we had taken input-output relationships and international supply-chains into account. As we shall see, judging exceptionalism at the turn of the 20th century is much simpler.

Figure 1 Market share of US automobile sales accounted for by GM, Ford, and Chrysler

Source: Goolsbee and Krueger (2015).

US exceptionalism in automobile adoption 

In a recent paper (Cheng et al. 2019), we examine the production, trade and diffusion of the passenger automobile from its humble origins at the turn of the 20th century to the brink of WWII.  At the dawn of the modern automobile industry, the US was unrivalled in automobile adoption. Figure 2 shows the timeline of registrations per 1,000 population in the US and an aggregate of 23 (mostly) industrialized countries, excluding the US.

Figure 2 Passenger Vehicle Registrations per 1,000 population in the US (blue line) and an aggregate of 23 foreign nations (red line) 

Note: The US y-axis (on the left) is 10 times the foreign y-axis (on the right). 
Source: Cheng et al. (2019).

The registration levels are highly correlated over time, with the exception of WWI when adoption levels stagnated abroad, but continued to grow in the US.   Adoption levels exploded during the roaring 1920s and collapsed during the international Great Depression. The most striking feature of the figure, however, is that the US adoption rate is systematically a factor of 10 higher than the rest of the world. In this period of economic history, there appears to be something uniquely American about the automobile.

US exceptionalism in production and trade

The definition of the automobile has some historical ambiguity, as there were electric, steam and eventually gasoline powered vehicles with the number of wheels between three and four even in the 19th century.  At the beginning of the 20th century, the automobile with a gasoline powered combustion engine, multiple cylinders, and four-wheels, established itself as the dominant prototype and entered into mass production. The US – led first by Ford and later by GM, along with Chrysler and much smaller companies – dominated the world market of vehicle production.

Figure 3 World market of vehicle production (number of vehicles)

Source: Cheng et al. (2019).

Figure 3 displays the world market shares of vehicle production by country. The cumulative share of US production over the 1922 to 1938 period was 80%. The nearest rival was the UK with a market share of 6%. We see a gradual erosion of the US market share of world production, but nothing approaching the movements implied by Figure 1. In fact, Figure 1, understates the global transformation of the automobile industry. It shows that substitution of production and sales within the US from Ford, GM and Chrysler to foreign multinationals such as Toyota, Nissan, and Honda, producing within the US.  As the adoption gap between the US and the rest of the world has narrowed dramatically, much of the growth in automobile production levels occurred outside of the US and production has become more globally balanced.

US production exceptionalism carried over into trade. Table 1 shows the share of automobile imports of 21 nations accounted for the US and the three largest foreign producers: the UK, France, and Canada. The upper section of the table isolates countries more dependent on US as a source. The average share of imports accounted for by the US is 91.3%, very close to the US world production share at the time. The remaining six countries are less dependent on the US for imports in part due to colonial ties.

Table 1 Share of automobile import value by source and importer

Notes: Values are averages over the years, 1923-1928. Group 1 are highly dependent on US as a source, Group 2 are less dependent on the US as a source.  The ‘*’ indicates a share is less than 0.1%.
Source: Cheng et al. (2019). 

Falling US automobile prices and foreign tariffs

Ford introduced the assembly line production of the Model T in 1914. This process innovation has been credited with reducing labour input requirements for chassis production from 12.5 to 1.5 (hours), a factor of more than eight (Baldwin et al. 1987). Over the decade from 1913 to 1923, the relative price of the automobile fell by a factor of 2.5.  

Figure 4  Blue solid line is the average US wholesale price

Notes: The red dashed line is average US export unit value. The red solid line marks up export unit value by median tariffs of 15% from 1913 to 1920 and 30% from 1921 to 1940. 
Source: Cheng et al. (2019).

Figure 4 shows the path of the average relative price of automobiles in the domestic US market, at the point of exportation from the US and the path of prices marked up by the median foreign tariff imposed by importers. Specifically, the figure marks up the export unit value by the cross-country average tariff of 15% (30%) from 1913 to 1920 (1921 to 1940).  The model in our paper predicts lower levels of adoption when relative prices are higher with the quantitative effect dependent the elasticity of substitution between automobiles and other goods. In placing tariffs on US imports, the effect is to lower the foreign adoption levels relative to those of the US. As there was considerable variation in tariff levels across nations (in 1921, the interquartile range is 15% to 40%), their adoption levels also varied widely.

We model the demand for automobiles and other goods as arising from CES preferences with non-unitary income elasticities. Using our model and the historical data that we collect, we ask whether the low adoption of US automobiles in foreign countries was due to the high price of US autos including the effects of trade frictions, such as tariffs, or due to lower levels of real income. Across the 23 countries in our study, we show that about 62% of the gap in diffusion levels between the US and foreign nations is due to price frictions such as markups, tariffs, and trade costs (with almost equal contributions of about 20%), while the remaining 38% is due to income effects. 


The source of early 20th century US exceptionalism in the automobile industry is evident in the rapid decline in the relative price of the US automobile. In large part, this decline was due to process and product innovation diffusing across domestic plants built by GM, Ford, and Chrysler. The price declines made passenger automobiles more affordable to Americans and rapid adoption followed naturally. Foreign governments reverted to highly protectionist tariffs, delaying adoption by their consumers and doing little to encourage foreign production beyond domestic demand, even in countries with established domestic brands. 

The general liberalisation of tariffs on manufactured goods under the GATT, combined with the liberalization of capital markets following the Bretton Woods Agreement led the US automobile industry to globalise in the same way that the liberalisation of immigration and migration of athletes has globalized the US professional sports industry.   The US auto industry now consists of both domestic and foreign multinationals, all of whom employ US workers. The industry is in many ways more exceptional than at its birth – it has matured to succeed in an environment with more international competitors than was true of the past.  This success, though not without setbacks, is evident by the fact that US manufacturers were not seeking protection from foreign competition during the interwar period and they were not lobbying for protection recently.

As the history of the automobile industry has shown, backing away from competition rarely serves the desires of the customer or the efficiency of the producer.

Authors’ note: The views expressed are solely the responsibility of the authors and should not be interpreted as reflecting the views of the Board of Governors of the Federal Reserve System or of any other person associated with the Federal Reserve System.


Baldwin, N, G N Georgano, M Sedgwick, and B Laban (1987), The World Guide to Automobile Manufacturers, Facts on File Publications.

Bond, E, M J Crucini, T Potter, and J Rodrigue, “Three tariff lessons from US interwar history”,, 27 September.

 Cheng, D, M J Crucini, H Oh, and H Yilmazkuday (2019), “Early 20th Century American Exceptionalism: Production, Trade and Diffusion of the Automobile,” NBER Working Paper No. 26121.

Comin, D A, D Lashkari, and M Mestieri (2019), "Structural Change with Long-run Income and Price Effects", NBER Working Paper No. 21595.

Goolsbee, A D and A B Krueger (2015), “A Retrospective Look at Rescuing and Restructuring General Motors and Chrysler”, Journal of Economic Perspectives 29(2): 3–24.


[1] . The 2018 year-to-date markets shares of General Motors, Toyota Motor Corp, Ford Motor Company, FCA (Fiat-Chrysler), Nissan Motor Co and Honda Motor Company were: 17.02, 14.63, 14.44, 12.98, 9.35, and 9.31.

[2] Fiat-Chrysler is doubly disqualified as a trans-national competitor, which appears to violate both the rules of Olympic competition and the definition of national exceptionalism.

[3] The preferences used in our study are adapted from Comin et al. (2019), who study the transition from agriculture, to manufacturing and services. They find a much larger role for income and a much smaller role for relative prices in that context.

[4] See Bond et al. (2018) for an historical perspective on US tariffs and their efficiency effects.

1,995 Reads