VoxEU Column Environment Industrial organisation

Innovation, emissions policy, and competitive advantage in the diffusion of European diesel automobiles

Diesel vehicles have never been popular in the US, but have dominated sales in Europe. This column presents new evidence explaining why this is the case. A change in preferences and the numerous competing suppliers benefited the diffusion of diesel cars. But more important was a European environmental policy that favoured CO2 reductions. As diesel vehicles are only produced by European manufacturers, this policy provided a competitive advantage for domestic producers equivalent to a 20% import duty.

Diesel vehicles in Europe

A week ago, on 18 September 2015, the US Environmental Protection Agency (EPA) accused Volkswagen (VW) of devising a sophisticated scheme to deceive environmental authorities when testing for NOx emissions.1 The notice of violation translated into a 20% drop in the value of VW shares three days later, the result of a possible recall involving up to half a million vehicles and $18 billion in fines. A likely outcome of this scandal may be the effective disappearance of diesel vehicles from the US market for a second time in two decades due to failure to meet emission standards. Diesel passenger vehicles have never been popular in the US but they have dominated sales in Europe for the last quarter of a century.

In a recent paper, Manuelli and Seshadri (2014) document the slow and lengthy diffusion of many major innovations. In order to explain the five decades it took for tractors to be widely adopted, they resort to the argument first put forward by Schumpeter (1950) that potential buyers delayed purchasing them in the hope of benefiting from additional incremental technology improvements. This must not be the case of diesel automobile engines after the introduction of the TDI engine in 1989. Within a decade, the market share of diesels in Europe grew from about 10% to well over 50%, and even exceeding 70% market penetration in the popular midsize saloon segment. This major transformation of the automobile industry is little known among non-Europeans and remains surprisingly ignored among innovation economists.

How was this fast diffusion possible? Why did diesels fail in other markets during the 1990s, most noticeably the US? In a recent paper (Miravete et al. 2015), we present a comprehensive analysis of this industry transformation in the 1990s using Spanish automobile registration data that distinguishes by make, model, and type of engine. Four elements help explain the fast diffusion of diesels in Europe:

  • Market initial conditions;
  • Evolution of driver preferences;
  • Technology imitation; and
  • A greenhouse-oriented emission policy which favoured diesel vehicles.

As diesel vehicles are generally only produced by European automakers, the emission policy employed by the EU provided European automakers with a significant competitive advantage over foreign imports in the domestic marketplace.

Initial conditions

It is commonly assumed that the reduced diesel fuel taxation is responsible for the large market penetration of diesel vehicles in Europe. This favourable tax treatment has its origin in the European Fuel Tax Directive of 1973 that aimed at harmonising energy taxation across Europe following the first oil crisis. Until recently, fuels were taxed by volume (easier to monitor) rather than by energy content, and diesel was favourably treated to limit economic harm to agriculture and road transportation.

Fifteen years later, by the time of the introduction of the TDI, this reduced taxation of diesel fuels was responsible only for a 10% market niche of large passenger cars and commercial vehicles, a far-cry from the market penetration of diesel vehicles during the 1990s. But this market niche allowed for other important developments. By 1990, there were plenty of mechanics trained in fixing diesel vehicles and every gas station across the continent included a diesel pump, thus reducing the cost of adoption for potential buyers.

Estimation – change of preferences

To explain the fast adoption of diesel vehicles during a period of historically low fuel prices we estimate an equilibrium model for the automobile industry in Spain – a country which exhibited diesel adoption rates representative of Europe as a whole. The fact that product offerings change rapidly during the decade invalidates the use of product characteristics as instruments for prices as it is common in Berry et al. (1995). We thus follow Petrin and Seo (2015) and allow automobile manufacturers to choose car attributes (e.g. horsepower) in order to maximise future expected profits.2

Figure 1. Automobile manufacturers unit costs (reference = Renault)

  • Our estimates reveal that drivers prefer smaller vehicles, favour models with high power/weight ratio, and dislike imported (non-European) cars.

Preferences are, however, quite heterogeneous regarding fuel cost of driving, performance, and the diesel technology. While diesels are not favoured when TDI was introduced, the perception of this technology improved during the 1990s, perhaps as a result of learning spillovers as it became more common to encounter diesel vehicles on the road.

Our estimates also confirm that Chrysler and Asian imports are substantially more competitive than European manufacturers. Taking market leader Renault as the reference, Figure 1 shows that only entry-level brands SEAT and Skoda can effectively compete against imports.


Bresnahan (2010) characterises general technologies as those that are easy to imitate and reuse by other manufacturers, thus limiting the profitability of innovations. As VW introduced the TDI engine, it should have anticipated that other automakers would reverse engineer those engines and offered their own alternatives within a short period of time. Will those spillovers turn this innovation unprofitable?

Figure 2. Volkswagen’s innovation rent capture across the decade

We use our model to simulate market outcomes under two alternative scenarios: a world without diesel vehicles; and one where VW is the single manufacturer of diesel engines. The difference between VW’s associated profits of these two scenarios defines the potential profits of the TDI innovation. A comparison with the actual profits reveals the share of those innovation rents captured by VW. Figure 2 shows that VW probably cannibalised its own sales of fuel-efficient gasoline models in the early stages of diffusion of the diesel technology, but later it was able to capture over 30% of the innovation rents, or about €250 million per year from the Spanish market alone.

  • Thus, imitation and competition is responsible for the dissipation of two-thirds of the innovation rents, which will benefit consumers in the form of lower prices and more products.

Emission policy and protection of the European automobile market

  • And finally, and perhaps most interestingly, a European emissions policy aimed at reducing greenhouse emissions proved to be a very effective tool to enable the popularity of diesel vehicles.

Since diesel vehicles were (and are) predominately produced by European automakers, the emission policy provided these automakers with a competitive advantage (diesels) to protect the domestic automobile market from foreign competitors who had not invested in this technology.

We are agnostic about the right targets for CO2 and NOx emissions in order to properly address different environmental concerns, or whether protecting the US or European automobile industry was ever considered when designing such environmental policies. Figure 3 simply documents how diametrically opposed the US and European emission standards were at a moment in time, in the year 2000.

Figure 3. US and EU emission standards in the year 2000

Source: www.dieselforum.org

Could these different policies be behind the drastically different fuel mix in the US and European automobile markets? What other implications could we learn from these environmental policies?

The 1990 US Clean Air Act Amendments directed the EPA to limit vehicle emissions responsible for acid rain, among them NOx. Three-way catalytic converters used to reduce emission in gasoline engines cannot cope with the high concentrations of NOx generated by diesel engines. Rather than investing in redesigning their diesel engines to meet these standards, VW and Mercedes stopped selling diesel vehicles in the US market altogether. Would the same happen in Europe with more stringent NOx emission standards?

Again we use our model to simulate market outcomes under increasingly higher retrofitting costs for diesel engines to meet these standards. Figure 4 shows the market share by type of fuel and geographic origin and marks the range of 1994 euro equivalence of the 2010 retrofitting costs using the inexpensive Lean NOx Catalyst and the expensive Selective Catalytic Reduction System. With the latter, the share of diesels falls below the 10% market share prior to the introduction of the TDI.

Although both the US and Europe have set increasingly demanding standards, the US authorities have always been more concerned about NOx than CO2. If the current EPA accusations are true, VW would have chosen again not to invest in meeting today’s standards as its ‘deception device’ only reduced NOx emissions when vehicles were tested. Otherwise, VW cars were emitting 10 to 40 times more NOx than the emission standards. This recent scandal validates in some unexpected way our conclusion that the stringent US NOx emission standards after 1994 are responsible for the failure of diesel vehicles in the US market. After years of trying to reintroduce diesels in America, VW’s behaviour will probably mean a second exit of diesels from the US automobile market.

Figure 4. Market shares by geographic origin and retrofitting cost

Figure 4 also shows another important result. The goals of CO2-focused environmental authorities and automobile manufacturers are aligned. European environmental authorities want to favour the production of diesel vehicles because it reduces greenhouse emissions. And European manufacturers like selling diesel vehicles as they are increasingly accepted by European drivers who favour them over gasoline-efficient imports. As retrofitting costs of diesels increase and their market presence vanishes, gasoline imports grow their market penetration. Thus, the lenient NOx European emission standards effectively protect domestic manufacturers against (mostly) Asian imports. While in 2000 the combined market share of imports was 11.8%, in a world without diesels our counterfactual analysis indicate they would reach a 19% market penetration.

The substitutability between domestic policy and import tariffs is known since at least the work of Bhagwati and Ramaswami (1963), but it has rarely been measured. We use our equilibrium model to evaluate what import tariff would reduce the market penetration of non-European manufacturers from 19% to 11.8% in a scenario without diesel vehicles. We conclude that European environmental regulations are equivalent to a 19.6% import tariff, about double the nominal import tariff set at that time.3


The quick adoption of diesel automobiles in Europe is a fascinating case study that involves preference changes and imitation of a general technology, perhaps unintended consequences of emissions standards, industrial and trade protection policies. Our work shows how the use of structural equilibrium oligopoly models can be used to evaluate the implicit cost of regulations, and that domestic environmental policies might be a powerful instrument to protect domestic manufacturers in a world where trade agreements limit the extent of tariffs. As in the case of tariffs, implementing environmental policies can lead to important inefficiencies. In our case, it allows for excessive market power for domestic automakers and, depending on the shadow cost of each type of emission, might induce drivers to purchase the least desirable technology.


Berry, S, J Levinsohn, and A Pakes (1995) “Automobile Prices in Market Equilibrium,” Econometrica, Vol. 63, pp. 841–890.

Bhagwati, J and V K Ramaswami (1963) “Domestic Distortions, Tariffs, and the Theory of Optimum Subsidy,” Journal of Political Economy, Vol. 71, pp. 44–50.

Bresnahan, T F (2010) “General Purpose Technology,” in B. H. Hall and N. Rosemberg eds, Handbook of the Economics of Innovation, Vol. II, New York, NY: North-Holland.

Manuelli, R and A Seshadri (2014) “Frictionless Technology Diffusion: The Case of Tractors,” American Economic Review, Vol. 104, No. 4, pp. 1368–1391.

Miravete, E J, M J Moral, and J Thurk (2015) “Innovation, Emissions Policy, and Competitive Advantage in the Diffusion of European Diesel Automobiles,” CEPR Discussion Paper No. 10783.

Petrin, A and B Seo (2015) “Identification and Estimation of Discrete Choice Models when Observed and Unobserved Product Characteristics are Correlated,” Mimeo, University of Minnesota.

Schumpeter, J A (1950), Capitalism, Socialism, and Democracy, New York, NY: Harper & Row, 3rd edition.


2 Automakers also internalise the effect of their choice of product attribute on price; however, sudden shifts in demand only lead to price adjustments and not to product relocation.

3 Given that the market penetration of imports in Spain is similar to most other European countries, this 19.6% import tax could be considered representative of the equilibrium protective value of European environmental regulations across the continent. France is the most important outlier among large European automobile markets, with Asian imports barely exceeding 5% share in year 2000. For the French case, the tariff equivalence of the European environmental policy is as high as a 35.3% import duty.

7,349 Reads