VoxEU Column Frontiers of economic research

Car thieves: Not too bright, please!

How economically minded are car thieves? This column presents evidence from the Netherlands suggesting that car thieves stay away from cars in unpopular colours because of their relatively low resale value. It argues that driving a car in a bright, uncommon colour such as yellow is a highly effective deterrent against car theft – about as effective as an expensive security device.

Students in the Netherlands have been found painting their bicycle pink or yellow. The bright colours are thought to reduce the risk of theft as it is easier to spot the thief and may make the bike less easy to sell. As only a few students are willing to take the effort and want to be seen riding a yellow or pink bicycle, these paint jobs have been stable practice for at least a couple of decades. How much of a deterrence effect it has is yet unknown.

Similarly, a car thief interested in making money can be expected to go for cars with the highest resale value, i.e. cars in the most popular exterior colours. Colour is key in the car market. A car in silver or yellow goes for the same price at the dealership, but the resale value greatly differs between the two. The National Auto Auction Association estimates that on average a used car in a popular colour sells for $1,000 more than the same car with a less desirable colour (NAAA 2010). The preference for common colours may be strengthened by the fact that a stolen car with a common colour is less easy to spot by police officers that have been alerted a car theft has occurred. That may be particularly important for those using a stolen car to commit another crime such as a robbery.

In other words, economically inspired models of criminal behaviour predict relatively low theft rates for cars in uncommon, bright colours (Becker 1968, Felson and Boba 2010). Currently, most car thieves are in it for the money. Joyriding and other temporary theft have become almost extinct with the rise of the highly effective electronic engine immobiliser that prevents hotwiring, a popular technique of stealing a car (Vollaard et al. 2010).

Conservative taste

Buyers of new cars in the Netherlands have a strong preference for three colours: black, blue, and silver/grey. Together these colours account for almost 80% of cars leaving the dealership over the last few years. As Figure 1 shows, car buyers have become more and more conservative in their taste. Red and green have been out of fashion for a couple of years now, particularly silver/grey has become very popular.

Tastes in other countries are also relevant as a non-negligible part of all stolen cars are meant for export – to eastern Europe in particular. Although there is some variation across countries which colours are most popular (with some variance in the popularity of white), there is great similarity in taste when it comes to unpopular colours. Data from DuPont, a major manufacturer of car paint, show that preferences in the Netherlands are roughly in line with those in other European countries (DuPont Automotive 2009). Taking a worldwide view, black, blue, white, and silver/grey constituted some 86% of new car sales in 2009. The other 14% is left for colours such as green, yellow, red, and orange.

Figure 1. New car sales by colour, the Netherlands, 1990-2009

Source: CBS/RDW

Thieves go for black, blue, and silver/grey

To put the economic perspective on car theft to the test, we combine data on the passenger car fleet and all stolen cars in the Netherlands for 2004-2008. The data were provided by the Department of Motor Vehicles (RDW), police, and Statistics Netherlands (CBS). The data relate to the fifteen most popular makes: Alfa Romeo, Audi, BMW, Citroen, Fiat, Ford, Honda, Mazda, Mercedes-Benz, Opel, Peugeot, Renault, Toyota, Volkswagen and Volvo. We focus on cars aged up to three years, as these new cars are most likely to be stolen for their resale value (85% of all car thefts are in this category).

Figure 2 shows the average risk of car theft by car colour during 2004 to 2008. In line with the above prediction, the three most common colours – black, blue, and silver/grey – are stolen more frequently than the uncommon colours. Pink is the perfect deterrent: none of the 109 pink cars (aged up to three years) have been stolen over the last few years. Noticeable is the rate of theft for black cars, which is higher than the rate for the most popular colour, silver/grey. There may well be something else than colour that makes black cars relatively attractive to car thieves. Black is more popular for luxury makes. However, if we exclude the three luxury makes Audi, BMW and Mercedes-Benz, we find the same pattern – albeit at a slightly lower overall rate of theft (right bars in Figure 2). It could be that the most expensive models (independent of make) are more likely to be black, but that is not clear and not easy to test.

If we exclude cars in black because of this possible ‘luxury model bias’, the average rate of theft of cars in common colours blue and silver/grey is almost 40% higher than for cars in other less common colours. That puts the deterrent effect of an uncommon car colour in about the same range as the effect of a good car security device such as the Lojack track-and trace system (Gonzalez-Navarro 2008).

Yet, while some people undoubtedly get pleasure out of driving a yellow or orange car, an uncommon colour is not without its costs. Anecdotal evidence from used car dealers that is in line with NAAA (2010) suggests the resale value penalty to be some 5% to 10%, which can be as much as €1,000. This “implicit price” for the deterrent effect of an uncommon colour needs to be reduced by the value of the joy of driving such a car. Actually, if the theft reduction effect of an offbeat colour was hitherto unknown to car buyers, then the resale penalty must be equal or lower than the utility derived from the colour: otherwise, the colour would not have been chosen in the first place. In other words: with a security device like Lojack going for some €500, deterrence through a bright car colour may be at least as good a deal as deterrence through security.

Figure 2. Theft risk by colour (%), cars up to three years old, 2004-2008, the Netherlands

Note: Left bars relate to 15 most popular makes in the Netherlands (Alfa Romeo, Audi, BMW, Citroen, Fiat, Ford, Honda, Mazda, Mercedes-Benz, Opel, Peugeot, Renault, Toyota, Volkswagen, Volvo). Right bars exclude the three luxury makes (Audi, BMW, Mercedes-Benz).Source: CBS/RDW

Car colour has become more important

When we look at how strongly theft risk and exterior colour are related, we find that car colour is becoming more important. It is three times more important now than it was 15 years ago. There are two reasons why this should be so. First, as car buyers have become more conservative in their taste, the difference in resale value between the three most popular car colours and the other colours has become more pronounced. Second, as cars have become hard to steal (thanks to engine immobilisers) most car thieves are now professionals rather than young people looking for a joyride. Clearly, professional thieves have a stronger preference for common colours.

It’s resale value, not the fear of getting caught

Is it only resale value that drives the preference of thieves for cars in common exterior colours or does the probability of apprehension also play a role? The recent history of car theft gives us some idea. Red is obviously a bright colour that attracts attention – including that of the police. Red is also a colour that has fallen out of fashion since the turn of the century (Figure 1). In the beginning of the 1990s around 25% of all new cars were red, now the number is close to 5%. The decline of red doesn’t only go for the Netherlands, but is a worldwide trend according data from DuPont. If thieves are primarily interested in resale value and do not care much about being spotted in a bright coloured car, then we should see higher rates of theft for red cars in the 1990s. That is exactly what we find. Figure 3 shows that, just with the colour silver/grey, the popularity of red in new car sales is tightly linked with the prevalence of red among stolen cars. This suggests that car thieves do not seem to be particularly worried about being picked out from traffic by police.

Figure 3. Popularity of colour in new car sales vs. theft risk by colour, the Netherlands

Source: CBS/RDW


Differences in theft rates between cars in common and uncommon colour suggest that resale value is on the mind of car thieves. We find evidence that it is indeed the resale value rather than the fear of getting caught that is driving this difference. If the aversion to driving a car in an offbeat colour is not too high – or if someone actually enjoys it – then buying deterrence through an uncommon car colour may be at least as good a deal as buying deterrence through an expensive car security device.

The author would like to thank Rene van Stralen for excellent research assistance.


Becker, Gary S (1968), “Crime and punishment: an economic approach”, Journal of Political Economy, 76:169-217.

DuPont Automotive (2009), “Color popularity report”, Wilmington (DE).

Felson, Marcus, Rachel Boba (2010), Crime and everyday life, 4th edition, Sage, Los Angeles.

Gonzalez-Navarro, Marco (2008), “Deterrence and geographical externalities in auto theft”, mimeo, Princeton University.

National Auto Auction Association, 2010, Evaluating a vehicle.

Vollaard, Ben, Jan van Ours, Stefan Toonen (2010), “Costs and benefits of government intervention in victim precaution. Evidence from a natural experiment in car security”, mimeo, Tilburg University.

8,609 Reads