The Early Years of RTIs

Motorisation advanced in three waves. Between ca. 1910 and 1950 cars were concentrated in the USA, between 1950 and 1975 they became widespread in Europe and from 1960 onwards in the rest of the world, especially in Asia. In the process, the number of cars, trucks and motorbikes exploded from roughly one million cars in 1910 to 50 million (1930), 100 million (1955), 500 million (1985) and 777 million in 1997. It surpassed one billion in 2010. In addition, average driving distances expanded, further increasing overall traffic exposure.Footnote ¹¹

The modern automobile, driven by a combustion engine, was developed and first used in France and Germany, but it was in the USA that it first became part of everyday life.Footnote ¹² The beginning hardly foreshadowed upcoming developments. Two deaths due to motor vehicles were registered in Great Britain in 1896 and one in the USA in 1899.Footnote ¹³ Indeed, initially it did not seem that cars would spread beyond a small group of eccentrics. Early drivers in the USA were wealthy sportsmen who used cars to demonstrate ‘conspicuous leisure’. Their attitude outraged many Americans, some of whom reacted with ‘intense anger, and even acts of violence – often tinged with class hostility’.Footnote ¹⁴ Part of this resentment reflected the frustration of the have-nots watching the haves flouting their riches, but another part resulted from the experience that speeding motorists were becoming a danger to other street users. The number of cars grew, and stories of children being run over by reckless drivers or ‘joyriding’ chauffeurs made frequent, sometimes exaggerated, headlines in the press.Footnote ¹⁵

Thus, RTIs formed part of discussions on automobiles from the beginning. They presented a more difficult problem than railway traffic, which had also caused new dangers to which travellers were unaccustomed, but where a strict separation of railway lines from the rest of the traffic, notably from pedestrians, had proved successful at reducing victims.Footnote ¹⁶ Such a separation was not practical for motorised cars and the question involved a renegotiation of space. The issue was less acute in the countryside where other road users were few and where the car was often the only means of rapid transportation and its advantages were obvious. It was here that the car, in the form of an affordable and unpretentious Model-T Ford, turned from a symbol of aggressive luxury to one of the social success of the hard-working man. Country doctors, who had the most need for fast and reliable transportation, often served as promoters of cars as respectable products.Footnote ¹⁷ Mass production brought the car within reach of most Americans, and between 1909 and 1920 the number of registered cars increased by 2750%.Footnote ¹⁸

In the cities, which often had efficient means of public transportation already, cars had fewer obvious benefits and were more of a nuisance. Throughout the 1910s and 1920s, an angry animosity greeted cars, as traditional street users resented drivers who were perceived as disturbing the public order and as endangering people’s lives. Most urban RTI victims were pedestrians, and most of these were children. In marked contrast to later times, early observers blamed motorists for their deaths, not children who had played on the streets or parents who had failed to watch them. Children were expected to play on the streets without needing surveillance.Footnote ¹⁹

Motorist interest groups fought their negative image. They portrayed drivers as a persecuted minority deserving protection and succeeded in redefining the issue away from one of injuries and death to one of freedom. Any restriction of the use of the car was constructed as inhibiting the people’s rights to choose their preferred means of transport and of street use.Footnote ²⁰ In the process, they effectively reconstructed a street from a place of public service to a ‘marketplace for transportation demands’, an expression of a more modern economic outlook.Footnote ²¹ In a deliberate campaign the car lobby ridiculed pedestrians as ‘jaywalkers’ and thereby as less legitimate users of street space.Footnote ²² By the 1930s, years before cars became majority traffic participants, most people had accepted that streets were primarily for them.Footnote ²³ In addition to determined lobby activities, the democratisation of car ownership changed attitudes. Increasingly, ‘people bought their first cars, not just because they were useful as well as fun, but because their self-respect demanded it’.Footnote ²⁴

But the spread of motorised individual transportation was more than the sum of private decisions. Everywhere, governmental planning and the construction of an extensive system of overland roads preceded mass ownership of cars which could make use of it. In the USA, President Wilson signed the first large-scale highway building programme into law in 1916 with the Federal Aid Road Act, at a time when less than 4% of the population owned any type of motorised vehicle.Footnote ²⁵ In Europe, widespread motorisation did not begin until the 1950s or even later but various measures spurred and anticipated future needs. In Germany, National Socialist policies of mass production of a ‘people’s car’ and highways largely served purposes of military preparedness.Footnote ²⁶ The Italian government similarly initiated a high-profile programme of highway building years before there were sizable numbers of drivers but discontinued the scheme during the economic depression.Footnote ²⁷ In Great Britain and France, car ownership was no longer restricted to the most privileged class by 1938.Footnote ²⁸

The increase in cars went hand in hand with an increase in the burden of road traffic accidents.Footnote ²⁹ In the USA the number of RTI deaths increased drastically, both in absolute and relative terms between the First World War and the 1930s. While rates declined slightly and remained relatively stable afterwards, the absolute number of RTI deaths decreased only during the years of war and then continued to climb until more than 50 000 people were killed in 1970 (Table 1). In Great Britain, RTI fatalities surpassed 7000 in 1930, 1933 and 1934, corresponding to a rate of approximately 15 per 100 000. This compared to a total of 9000 deaths caused by all types of traditional accidents (falls, drowning, burning etc.) in 1900. Unlike the USA, where RTI deaths receded during the Second World War, the British situation initially became more critical after 1939, when war-time blackouts exacerbated the pre-existing dangers of the road: 8000 people died on British roads in 1939, roughly 8,500 in 1940 and 9000 in 1941, approaching 18.8 per 100 000. The numbers subsided afterwards, due to a drastic reduction of car ownership and available fuel.Footnote ³⁰ Although these numbers were well below the rates experiences in the USA at the time, they were impressive when road casualties were compared to war casualties: 370 000 people were killed and wounded during the war, while 588 000 people were killed and injured on the roads during the same period.Footnote ³¹ The latter number never truly entered the collective memory.

Table 1: Motor vehicle deaths and death rates. Sources: US Bureau of Census, The Statistical History of the United States, Introduction by Ben J. Wattenberg (New York: Basic Books, 1976), Series Q 208–209, p. 719; US Bureau of Census, Table 1103, http://www.census.gov/compendia/statab/2012/tables/12s1103.pdf.

After 1950, mass motorisation took root in Europe, and widespread car ownership formed part of a series of massive socio-economic transformations, based on cheap fossil fuels. Within one generation, patterns of housing, shopping, work and leisure time all reflected a society that relied increasingly on cars. Rising RTI numbers were part of this development, and they changed shape in the process. While initially accidents had primarily involved one vehicle and fixed objects or pedestrians, the rising presence of cars on the streets meant that collisions increasingly engaged several vehicles. In the UK, crashes involving three or more vehicles represented a mere 1.5% in 1936–7 but 4.7% in 1953. In the USA, deaths resulting from collisions of vehicles with fixed objects increased by 80% from 720 in 1930 to 1300 in 1952 but during the same period deaths resulting from collisions between two or more vehicles increased by 140% from 5880 to 14.100.

In the USA, as the number of cars increased, death rates in proportion to vehicles decreased, as did the proportion of victims resulting from collisions with pedestrians first in relative and eventually in absolute terms. From 1950 onwards, more people died in collisions between two motorised vehicles than in accidents involving vehicles and pedestrians. Similarly, the number of pedestrian deaths per registered motor vehicle roughly fell by half in Switzerland, Sweden, the UK and Ireland between 1947 and 1953.Footnote ³²

These changes led to a shift of concern about RTIs from pedestrians to drivers but also to an increasing acceptance of cars and RTIs. Even in pre-war Great Britain, pro-car and pro-pedestrian associations had competed for dominance while government policies tended to favour the affluent, i.e. car drivers.Footnote ³³ After the Second World War, the widespread perception that the expanding car industry was instrumental to economic reconstruction as well as to social recovery provided automobiles with a positive connotation. This existing pro-car bias and the growing democratisation of car ownership complicated a perspective which focused on cars as the culprits of RTI mortality. Instead, discussions were muted and concentrated on ‘the human factor’. In Great Britain, the discourse focused on pedestrian behaviour and pedestrians accepted that it was primarily up to them to adapt to an increasingly motorised environment. This discourse was complemented by a new perspective of speedy cars no longer as the source of road dangers but as a proud sign of modernity, which required similarly modern road systems.Footnote ³⁴ Meanwhile, in the USA, attention still concentrated on drivers. Automobile associations and insurance companies organised driver education courses, designed to turn car users into ‘safe drivers’, while local administrations established and police enforced regulations which should ensure that driver behaviour was conducive to road safety.Footnote ³⁵ RTIs were perceived as a form of dysfunction, caused by individual traffic participants who were insufficiently adapted to the demands of modern life.

The Construction of a Public Health Issue

Remarkably, during the first half of the twentieth century RTIs never appeared as an international public health issue. The League of Nations, which pioneered data collection and assessments of the state of public health around the world, completely ignored RTIs.Footnote ³⁶ It was only in the 1950s, after the USA had recorded its millionth RTI death (in 1951) that the international community began to take note.Footnote ³⁷ The United Nations (UN) Economic Commission for Europe began publishing detailed annual road traffic statistics (the Statistics of Road Traffic Accidents in Europe) for eighteen European countries.Footnote ³⁸ In 1957, the First International Congress of Traffic Police, held in Eindhoven, discussed road safety measures. In line with prevailing attitudes, its recommendations focused on the weakest of potential victims, calling for better education of pedestrian road users, especially of children.Footnote ³⁹

Meanwhile, the World Health Organisation (WHO) discovered RTIs as a health issue and began defining it in those terms. As a start, it carried out a survey regarding motor vehicle accidents. Forty-seven member states returned questionnaires: between them they had recorded 102 552 deaths that year (79 810 of them males) out of a population of 650 million people, and numbers were rising. No records existed regarding the much larger number of people who were injured, often seriously. It became clear that RTIs affected predominantly young males and children where they took a staggering toll:

In the UK, RTIs killed fifteen times as many children as poliomyelitis in 1956, and twice as many as during the worst polio epidemic after 1945. In a complaint that was to be repeated many times during the following decades, the author of a study observed that these numbers aroused a fraction of the interest of that directed at other epidemics.Footnote ⁴¹ The WHO tried to raise the awareness of this invisible health issue, dedicating its 1961 World Health Day to the theme of ‘Accidents and Their Prevention’.Footnote ⁴² It also commissioned a study on Road Traffic Accidents.

This report, written by the Chief Medical Officer of the London Transport Executive, was published by the WHO in 1962 under the sub-title of ‘Epidemiology, Control, and Prevention’ and declared that RTIs constituted ‘a public health problem of the first magnitude’.Footnote ⁴³ But comparisons between countries, and, indeed, an assessment of the burden, were difficult since definitions varied widely. In Belgium, for instance, road traffic deaths described deaths which occurred at the site of the accident only, while in England, deaths occurring up to 30 days after the accident entered into the statistics.Footnote ⁴⁴ What was clear, however, was that there were gender- and age-specific differences regarding RTI risks. The most striking element then – as later - was the gender gap. In all participating countries, the ratio of male to female road traffic deaths ranged between three and five and, the author of the 1962 WHO study observed, resembled a ‘biological or sociological law’, so far little understood.Footnote ⁴⁵ The age distribution differed between traffic participants. Among pedestrians, children between roughly 1 and 10 were particularly at risk; prevalence then declined and rose again from age 65 onwards. The increase in old age resulted in part from the high probability of elderly people dying from injuries from which younger people recovered.Footnote ⁴⁶ However, within the age bracket of young adults, RTIs represented a major cause of death, causing ‘a serious economic loss to the community’.Footnote ⁴⁷

While RTIs, therefore, slowed down economic development in tangible, although unquantified ways, the essential role of the motorisation industry for economic growth was obvious. The report commented that evidently RTIs took an increasing part of national mortality as a country became ‘more highly developed and therefore more highly motorised’.Footnote ⁴⁸ The connection between development and motorisation appeared too obvious even to suggest a cost–benefit analysis. Despite his urgent calls to address RTIs as a public health threat, the author was ready to accept a certain level of RTIs as an inevitable side-effect of modernisation, and he merely suggested that ‘a balance between man and the new element in his environment, the motor vehicle, is being reached, and that a mortality of roundly 20 per 100 000 per annum is the price of introducing the motor vehicle on a large scale’.Footnote ⁴⁹ Thus, in the early 1960s, a cost of 20 lives lost for every 100 000 people seemed an acceptable price for the benefit of modernisation, a rate reached in the USA during the late 1930s and then briefly again in 1948, but which had consistently been surpassed since.

Meanwhile, the perception of RTIs as a public health issue remained concentrated in North America and Europe where it gradually changed its character. In the late 1950s, a diverse group of public health experts, politicians, lawyers and social activists began arguing that 50 000 RTI victims annually were unacceptable and that car designs, which did not prioritise safety, bore a large part of the responsibility for the number of casualties. They argued that getting drivers to act more responsibly was clearly not succeeding while making cars ‘crashworthy’ by supplying them with padded dashboards and stronger door locks would have instant effects. The political climate of social discussions and the civil rights movement of the time were propitious for the argument that people’s well-being went beyond the control of the individual. The public was introduced to the new arguments through congressional hearings and, above all, in Ralph Nader’s book Unsafe at any speed, published in 1965. The car industry, fearing expensive construction changes and liability claims, tried to portray the issue as one of personal freedom. This strategy remained unsuccessful in part, because insurance companies could be drawn over to the other side of the argument.Footnote ⁵⁰ One year later, Lyndon Johnson signed into law two bills raising safety standards in cars and roads, and established a new federal agency, the National Highway Safety Bureau (NHSB), in charge of RTI control.Footnote ⁵¹

These events marked an important shift in how the causes for RTIs were defined, with ambivalent effects. On the one hand, it weakened a perspective which blamed RTI victims for their fate. Thus, shifting attention to cars can be seen as a correction to a situation where the industry had largely been exempt from any accountability. On the other hand, a growing focus on vehicle safety spurred the expectation of technical fixes. This new reductionist view of the problem became obvious during the more recent controversy regarding airbags, which became mandatory in the USA and whose overall benefits remained unclear. This development has been blamed for a neglect of behavioural factors and, ultimately, for a relatively less positive development in the USA, compared to other industrialised countries.Footnote ⁵²

The change of perspective paved the way for a view of economic development not as a cause of RTIs, i.e. the problem, but as its solution. And eventually, the focus on vehicles established the view of drivers being the primary victim of RTIs deserving protection. No similar technical considerations were directed at protecting pedestrians, who, by that time, had been driven out of large part of public spaces. Implicitly, this was criticised by William Haddon, an American epidemiologist who developed a systemic approach to RTIs which integrated considerations of infrastructural factors with vehicles and users in the pre-crash, the crash and post-crash stages relevant to road traffic accidents.Footnote ⁵³ This ‘Haddon Matrix’ was criticised but spurred a shift towards a more comprehensive perspective of traffic and RTIs in the 1970s and would re-emerge in a transformed shape in the late 1990s.

As urban centres in industrialised countries experienced an increasing burden of traffic congestion with its economic, environmental and social costs as well as ever-increasing RTIs, many cities took measures designed to reduce the motorised traffic on their territory. A study of the Organisation for Economic Cooperation and Development (OECD) of twelve cities in as many countries revealed that, by the mid-1970s, eleven had restricted parking in their centres, eight had increased the frequency of public transportation services, ten had provided preferential treatment for public services (such as bus lanes etc.), ten had established pedestrian zones and five had made provisions for cyclists. These measures had reduced RTIs substantially, sometimes spectacularly. Thus, Nagoya reduced RTIs by 61% and RTI mortality by 59% in its central business district within a few years. Even Paris, which saw its overall RTI rate go up by 52%, enjoyed a reduction of RTI mortality by 24%. In Ottawa, where overall RTIs increased by 19%, the RTI rate decreased by 40% in those areas where measures to reduce transit traffic had been put in place.Footnote ⁵⁴ These changes could not gloss over the fact that private cars were still considered ‘the predominant mode of personal travel in North America’ and were gaining ground in many areas of Europe and Japan.Footnote ⁵⁵ Nevertheless, in a curiously contradictory development, the number of cars kept increasing while cars were no longer automatically considered the best or the most modern means of transportation everywhere in high-income countries, and pedestrians regained some of the urban ground from which they had been evicted some decades before.

At about that time, the WHO began addressing RTIs more seriously. In 1974, the World Health Assembly acknowledged the ‘extensive and serious individual and public health problems resulting from road traffic accidents’ and urged national health authorities to provide leadership in the issue. In its proposals of tangible measures the resolution remained well within the paradigm of the last decades, calling for ‘improved driver licensing standards and traffic safety education programmes’ and the application of ‘safety principles in the development of new types of vehicles’.Footnote ⁵⁶ The Regional Office for Europe was made responsible for the WHO programme on road traffic accidents.Footnote ⁵⁷ However, probably this office observed more than they caused the substantial RTI decline which was unfolding around them. Between 1975 and 1998, RTI mortality decreased drastically in virtually all high-income countries: by 63.4% in Canada, by 58.3% in Sweden and by 27.2% in the USA.Footnote ⁵⁸ In 1987, a public health expert stated that a 200-fold increase of cars had been accompanied by an only twenty-fold increase in RTI deaths and that the RTI death rate was lower in 1985 than at any time during the preceding sixty years except in 1948. Impressed, he commented: ‘The epidemic of road traffic deaths may be most remarkable for the way it has been controlled.’Footnote ⁵⁹

Unfortunately, this assessment overlooked the majority of global regions and people. In fact, RTIs were fast developing into a global issue with an immense public health burden concentrated in middle-income countries. Between 1975 and 1998, RTI deaths increased by 237.1% in Colombia, by 243% in China and by 383.8% in Botswana. In 1990, RTI assumed place nine in contributors to the global burden of disease. Globally, road traffic deaths increased from ca. 990 000 per year in 1990 to nearly 1.2 million in 2002.Footnote ⁶⁰

In the late 1980s, international organisations gradually became aware of the extent of the evolving problem and of the need to address it. Given that North America and Europe looked back on decades of experience and that RTIs appeared to show a positive downward trend, there was a natural tendency to apply lessons learned in the ‘developed’ world to problems encountered in ‘developing’ countries. But what exactly were they? Had RTI rates declined because of better drivers, more stringent regulations, safer vehicles, better protection for cyclists and pedestrians, more alternatives to travelling by car, medical progress or simply because the sheer number of cars provided the majority of traffic participants with a protective frame, so that cars served as both protection against and as threats to other cars and their passengers? Or any combination of those factors? And which of these findings were applicable to the rest of the world, to whose benefit and at what price? Discussions evolved around competing claims to historical analysis and their conclusions for development policies in those countries where car traffic was still low but rising.

In many ways, the situation in low-income countries at the end of the twentieth century resembled that of countries in Europe and North America at its beginning: RTI victims were primarily pedestrians and cyclists, vulnerable traffic participants who competed with an increasing number of motorised vehicles for road space. But there were also distinct differences. Researchers listed a fateful combination of reasons for the rising RTI burden, including a ‘traffic mix of incompatible users (pedestrians, cyclists, motorbikes, cars, and trucks) with, for example, communities living within the vicinity of roads or the lack of pavement along large urban streets’.Footnote ⁶¹ Passengers in Africa and Asia usually rode in multi-passenger vehicles such as overloaded mini-buses or taxis, so that individual accidents involved more people than the typical accident in a high-income country, which had usually affected the driver(s) of one or several cars involved. Comparing RTIs in different parts of the world in 2000, this difference was exacerbated by the lack of effective first aid and emergency medical care in low-income countries. According to a Harvard study 10 000 crashes resulted in 66 deaths in the US, but 1786 in Kenya and 3181 in Vietnam.Footnote ⁶²

Gradually, international institutions began to become aware of the issue. Given the complex nature of RTIs, they could not but become part of the tension between different international health agendas. By the 1980s, international health was no longer the domain of the WHO alone. A new major player was the World Bank, which had been alerted by WHO efforts at the conference of Alma-Ata in 1978 to connect public health policies to demands for increased global economic equality and regulations, while growing neoliberal tendencies among key members, notably the USA, called for further economic deregulations. During the 1980s and 1990s, the World Bank steadily increased its role in the international health scene. Through substantial investment in health projects and the generation of health-related data it integrated health into its overall programme of fostering a market-driven form of economic development.Footnote ⁶³ Part of this strategy was to identify poverty reduction through economic growth as the primary means of improving public health in low-income countries, and research programmes were designed accordingly.Footnote ⁶⁴ This situation gave rise to the Global Burden of Disease Project, which the World Bank launched in preparation of its 1993 World Development Report Investing in Health, which successfully established the work of the World Bank as a reference point for global health.Footnote ⁶⁵ The Global Burden of Disease Project introduced disability adjusted life years (DALYs) as new measurement unit, which combined mortality, morbidity and injury into a single number. The resulting data revealed RTIs as an unexpectedly important health burden. Road traffic accidents ranked ninth among the leading causes of DALYs, accounting for 2.4% of the total.Footnote ⁶⁶ Even more surprising – at least to the authors of the study – road traffic accidents were the second leading cause of DALYs for men between 15 and 44 year of age in high as well as in low income countries.Footnote ⁶⁷ These data prompted a growing number of research publications dedicated to RTIs in the following years. In line with the perspective begun in the 1962 WHO report, publications constructed RTIs in medical terms as a ‘neglected epidemic’,Footnote ⁶⁸ a ‘global epidemic’,Footnote ⁶⁹ a ‘global road trauma pandemic’Footnote ⁷⁰ or compared RTIs to Aids, insisting that one epidemic provided lessons for fighting the other.Footnote ⁷¹ Some publications invoked warfare, describing RTIs as ‘vehicular manslaughter’Footnote ⁷² or a ‘war on the roads’.Footnote ⁷³

However, this health discourse had a hard time competing against the entrenched view of roads and road traffic as essential economic infrastructure. The 1994 World Bank World Development Report on Infrastructure for Development was a case in point. It acknowledged that RTIs were a leading cause of death in low-income countries and accorded governmental ‘regulation to preserve safety standards in infrastructure service provision and delivery’ an important priority.Footnote ⁷⁴ Another paragraph observed the ‘very low rates of traffic accidents’ which the Brazilian city of Curitiba had achieved through ‘carefully designed public transport routes’.Footnote ⁷⁵ However, those were the only references to road crashes and could easily be overlooked in the midst of a report which, otherwise, overwhelmingly discussed the need for optimal cost-effectiveness of infrastructure services, ideally through privatisation. Ironically, this preference for market solutions for development challenges should have provided grounds for the World Bank to favour investments in railway lines and other forms of collective transport rather than roads. Railways, bus services, subways etc. could easily be supplied by private operators or under concessions, working under conditions of competition and, therefore, supposedly more efficiently and cost-effectively. The privatisation of roads was sometimes possible as toll roads, but in general roads would necessarily remain public spaces.Footnote ⁷⁶

But these theoretical considerations clearly had little effect on real lending decisions. Since the 1970s, World Bank commitments for transportation infrastructure had increasingly concentrated on highway building, which effectively dwarfed the sums spent on railway lines. Investment in urban transport were non-existent or a fraction of those for highways and also consisted largely of road constructions. This tendency would continue and even intensify in the beginning of the twenty-first century. The sums invested in rural and urban roads amounted to 80% of all infrastructure lending (Figure 1(a) and (b)). The World Bank was not the only source of global investments in transportation infrastructure but its distribution of funds was reflected in the development of road and railway lines in many countries, including El Salvador, Malawi, Brazil and India. (Figure 2(a) and (b)). Invariably, the length of paved roads increased substantially, sometimes dramatically, between 1960 and 2000 while that of railway lines stagnated. If any funds were spent on infrastructure supporting non-motorised transportation, such as buses, trams, subway or even secure sidewalks or bicycle lanes, they were considered too insignificant to be mentioned in World Bank publications. Clearly, although a large part of the population in low-income countries had no chance of ever owning cars, motorised road traffic had become the accepted model for transport development. Meanwhile, RTIs were becoming an increasingly pressing concern. A precise understanding of the situation was difficult since data were – and still are – often patchy or simply non-existent, but those that were available were impressive enough. Examples included El Salvador and Mauritius, both of whom experienced rapidly rising RTIs mortality rates even though the length of paved roads in Mauritius expanded only slightly from 1593 to 1699km between 1970 and 1990.Footnote ⁷⁷ (Figure 3).

Figure 1: IBRD and IDA commitments: (a) millions of current US dollars. Source: World Bank, Infrastructure for Development, World Development Report 1994 (Oxford and New York: Oxford University Press, 1994), 149; (b) millions of US dollars. Source: World Bank Evaluation Group, A Decade of Action in Transport (Washington DC: World Bank, 2007), 23. ‘General transport’ corresponded to ‘urban transport’ and consisted predominantly of roads.

Figure 2: Railways tracks and paved roads in kilometres: (a) El Salvador and Malawi; (b) Brazil and India. Source: World Bank, Infrastructure for Development, World Development Report 1994 (Oxford and New York: Oxford University Press, 1994), 140–145; CIA World Factbook, https://www.cia.gov/library/publications/the-world-factbook/fields/2085.html#182. International Road Federation, World Road Statistics 2007, Data 2000–2005, Geneva 2007.

Figure 3: The crude death rate due to motor vehicle traffic accidents: El Salvador and Mauritius. Deaths per 100 000 population. Source: WHO Mortality Database.

In 1998, the International Federation of Red Cross and Red Crescent Societies presented warnings about the seriousness of RTIs as a ‘hidden global epidemic’ in its World Disaster Report.Footnote ⁷⁸ In response, the World Bank established the Global Road Safety Partnership (GRSP), a network of representatives of business, civil society and government organisations. Its membership read like the who’s who of the international car and oil industry, including General Motors, Honda, BP, DaimlerChrysler AG, Michelin, Renault, Shell, Total and Toyota. Several governments and international organisations, including the WHO, were also represented. The group was hosted by the International Red Cross, which provided contacts and credibility, and it was overseen by a Steering Committee.Footnote ⁷⁹ Within some years it supported practical road safety projects in several countries around the world. Activities ranged from targeting drunk-driving and speeding to promoting the use of helmets and seat-belts to separating motorised and non-motorised traffic or promoting public transportation.Footnote ⁸⁰ However, the main focus was on the behaviour of potential victims, portrayed as those most responsible for RTI deaths. Children, particularly, were described as lacking sufficient skills and knowledge to cope with complex road traffic situations. This line of argument had formed part of the conventional wisdom in high-income countries since approximately the 1940s, but it was quickly becoming outdated. A 2002 meta-study of controlled trials of pedestrian education programmes in high-income countries showed no evidence that such programmes reduced the risk of road accidents involving child pedestrians. No similar studies existed for low-income countries.Footnote ⁸¹

Including representatives of the automobile industry in this group brought technical expertise and assured close contact with those actors whose actions inevitably formed an important component of any road safety programme. But obviously their presence also precluded solution strategies which questioned motorisation in principle. In 2002, the secretariat for the Steering Committee was shifted to the Task Force for Global Health, an institution which had been founded to coordinate international activities of Primary Health Care after 1978, when the World Bank succeeded in shifting the Health for All approach of the Alma-Ata conference from a political strategy aimed at increasing economic equality to technical programmes of oral rehydration, breastfeeding and immunisation.Footnote ⁸² A similar strategy to define the issue as a technical problem appeared underway with regard to RTIs. However, even this met little response in the car industry, as the European Enhanced Vehicle Safety Committee (EEVC) realised. Its proposals regarding changes to the fronts of vehicles designed to make them less dangerous to pedestrians were not on the agenda of car manufacturers in 2002.Footnote ⁸³

The Competition of Development Concepts

At the end of the twentieth century, discussions on RTIs evolved into a more fundamental debate about global development. It was stimulated by the introduction of the Kuznets curve into RTI research. The Kuznets curve went back to a theory presented by Simon Kuznets in 1955. Drawing on historical data of industrialised countries he maintained that social inequality increased in the early phase of modernisation but decreased from a turning point onward as national income continued to grow. The relation between social inequality and economic growth evolved, therefore, along a curve the shape of an inverted U.Footnote ⁸⁴ In the 1990s, the model received a second life as a description of the relationship between economic growth and some pollutants, notably ${\mathrm{SO} }_{2} $ and smoke, and became known as the environmental Kuznets curve. This finding provided a welcome defence against accusations that the existing growth-oriented economic system was responsible for large-scale environmental destruction, and it was readily accepted by advocates of the contemporary economic system. In 1992, the World Bank incorporated it into its World Development Report on the environment.Footnote ⁸⁵ In the following years, further research challenged the curve: similar developments could not be reproduced for other forms of environmental burden. Besides, it was doubtful to what extent the effect resulted from the transfer of polluting industries to low-income countries, a strategy which could obviously not be imitated on a global scale.Footnote ⁸⁶ In 2000, van Beeck et al. appear to have been the first to observe a Kuznets curve of RTIs in relation to ‘prosperity levels’. Economic growth, they argued, was ‘not only associated with growing numbers of motor vehicles in the population, but also seems to stimulate adaptation mechanisms, such as improvements in the traffic infrastructure and trauma care’.Footnote ⁸⁷ This view suggested that economic growth would in itself lead to a reduction of RTIs, making a rise in gross domestic product (GDP) the best strategy to reduce the health burden of widespread motorisation. It also implied that RTIs as a health problem were largely solved in high-income countries.

This view was taken up by the World Bank. In 2003, the World Bank Development Research Group on Infrastructure and Environment issued a Policy Research Working Paper on Traffic Fatalities and Economic Growth.Footnote ⁸⁸ Analysing vehicles per person (V/P) and fatalities per vehicle (FN) data from eighty-eight countries for the period 1963–99, they found a confirmation of the Kuznets curve with a turning point at a per capita GNP of $8600 in 1985 international dollars. On the basis of these data and of prognoses of population and income growths, they projected that it would take many years for developing countries to achieve the low RTI fatality rate of existing high-income countries. RTIs in India, for instance, which had a per capita income of only $2900 in 2000, would only begin to decline in 2042 after a peak of at least twenty-four fatalities per 100 000 persons, or thirty-four when adjusted for estimated underreporting. Brazil would ‘already’ peak in 2032 and would experience an RTI mortality rate of twenty-six deaths per 100 000 persons as late as 2050, compared to a rate of around eleven enjoyed by high-income countries in 2000. Only on the last page did the text mention, almost in passing, that these projections were based on a continuity of ongoing policies, while measures such as mandatory helmet wearing or effective traffic separation might lower those numbers.Footnote ⁸⁹

Predictably, the paper provoked different reactions among researchers. Its projections led Nitin Garg and Adnan Hyder to urge that countries like India should take active steps to curb RTIs along WHO recommendations well before the theoretical economic threshold.Footnote ⁹⁰ Other researchers accepted and confirmed a Kuznets curve behaviour of RTIs as the statistical representation of a development whereby increasing national income would allow investments in safer roads and vehicles.Footnote ⁹¹ In an analysis of data from forty-one countries from the years 1992–6 David Bishai et al. found that in low-income countries a ten per cent increase in GDP increased RTIs by 4.7% and RTI deaths by 3.1%. By contrast, GDP increases in high-income countries reduced the number of deaths, although not of crashes or injuries. The turning point appeared to be between $1500 and $8000 per capita income.Footnote ⁹² Other projects were more clearly tied to corporate interests. In a study that was financially supported by the automobile industry, Walter McManus of the University of Michigan Transportation Research Institute calculated the lives that would be saved by lowering either vehicles per capita or the fatalities by vehicle. Both would save lives but, he concluded: ‘Reducing motorisation (vehicles per capita) is unlikely to be used as a policy to reduce fatalities because it is inextricably linked to economic growth. Consequently, the focus should be on reducing fatalities per vehicle.’Footnote ⁹³ Clearly, this approach limited anti-RTI strategies to those not harmful to the large sector of the global economy which depended in some way on the construction or use of motorised vehicles. It also portrayed RTIs as a regrettable but temporary side-effect of modernisation.

However, new approaches to the problem emerged. One sign of changing attitudes appeared in a debate about terminology. In 1987, a group of intensive care specialists in New Zealand proposed to change ‘the discourse on road traffic injuries by rejecting the concept of “accidents…”’.Footnote ⁹⁴ In a widely publicised campaign they succeeded in changing the choice of words in the media. Soon, the debate spread to Europe. In 1993, the editor of the prestigious British Medical Journal (BMJ) adopted the same attitude, arguing that numerous injuries, including RTIs, were not accidental and, therefore, should not be called ‘accidents’. Instead, he proposed using ‘crash’.Footnote ⁹⁵ The change of terminology was slow to be accepted. Maybe the expression of ‘traffic accident’ was too deeply ingrained in the general vocabulary, maybe BMJ was insufficiently influential or maybe people were unconvinced by the idea that ‘accident’ should be reserved for those rare events which involved no human responsibility at all. Eight years later, ‘accident’ continued to be widely used, including in articles published by the BMJ. In reaction, BMJ, in its self-proclaimed position as ‘a leading communicator in medicine’ with a responsibility ‘to establish or follow standards in language’ banned the ‘inappropriate use of “accident”’ in its pages.Footnote ⁹⁶

The change in terminology remained controversial. When the psychologist Alan Steward (University of Georgia) and social worker Janice Lord, former National Director of Victim Services of Mothers Against Drunk Driving, argued against ‘accidents’ since crashes ‘caused by intoxicated, speeding, distracted, or careless drivers’ were no accidents and that such misnaming might create extra stress for crash victims and impede their recovery, several colleagues disagreed.Footnote ⁹⁷ Indeed, prior studies had suggested the opposite effect: patients who blamed themselves for car accidents recovered more rapidly than those who blamed others.Footnote ⁹⁸ Nevertheless, the new terminology was adopted by many researchers and was also in WHO publications. Some authors clearly considered this change an important move, which they took pains to explain.Footnote ⁹⁹ Meanwhile, the current classification of causes of death and disease, ICD-10 kept ‘accident’ and so did other researchers as late as 2012.Footnote ¹⁰⁰ In fact, the change of words in itself was of limited consequence. It did underscore that RTIs resulted from preventable causes rather than fate and thereby contradicted a view of RTIs as inevitable consequence of development. But the new wording offered no new perspective as long as it continued to refer to well-rehearsed factors such as child restraints or driver behaviour, which stayed well within the conventional discourse.

Other initiatives had more far-reaching consequences. In 1999, Kåre Rumar, professor at the Swedish Road and Transport Research Institute, insisted that, despite falling RTI rates in Europe, the problem was far from solved. To illustrate its continued significance he pointed out that in most European countries one out of three citizens would need hospital treatment after a road traffic accident sometime in their lives and that one in twenty people would be killed or injured in a road accident.Footnote ¹⁰¹ He urged a systemic approach. In what appeared like an expanded form of the Haddon Matrix he described a web of factors, including age, gender, traffic regulations, road maintenance, attitudes to safety policies, intelligent control systems, unclear distribution of responsibilities, etc. Overall, his explanations aimed at a change of perspective. Instead of accepting the need for motorised traffic as a given and focusing on ways to reduce its health price, intelligent strategies should begin with the underlying purpose of traffic (transportation and mobility) and the biological vulnerability of the human body to external shock, and then search for ways to combine highest benefits with least sacrifice on that basis. As a vastly underestimated measure he singled out reducing the exposure of traffic participants to the risks of motorised transportation:

This approach was taken up by the WHO, which was initiating a five-year-strategy.Footnote ¹⁰³ A first result was a study on the ‘Global road safety crisis’, issued in 2003, which incorporated input from various UN bodies. The report advocated integrating RTI consideration into a broader vision of urban development and transportation planning, which also included alternative modes of transport. A one-sided concentration on a car-based system of traffic was portrayed as aggravating social inequality since it invested ‘increasing resources in the building and maintenance of an infrastructure for private motorised transport, while overlooking the public transport needs of the larger part of the population’.Footnote ¹⁰⁴ It also contributed to further health problems since in ‘many high-income countries, increasing use of cars has led to a general decline in walking and an increase in sedentary lifestyles, which in turn has had adverse consequences in terms of increasing obesity and cardiovascular health problems’.Footnote ¹⁰⁵ Effective strategies to reduce RTIs were said to require a ‘systems approach’, aimed at identifying and addressing all relevant factors. Successful strategies of high-income countries could serve as orientation, but policies in low-income countries would have to be adapted or even newly created according to local circumstances. While these explanations promised a new approach, the list of relevant determinants (speeding, alcohol, helmets, safety devices, trauma care, road safety standards, traffic safety regulations, vehicle safety) appeared remarkably conventional and addressed, again, to a large extent the behaviour of road users.Footnote ¹⁰⁶ Nevertheless, the text appears to have been the first time that a high-level international report on RTIs welcomed – and by implication recommended – a cutback of the use of cars as one strategy to reduce the RTI burden.

The concept of a ‘systems approach’ was further elaborated in a World Report on Road Traffic Injury Prevention, a WHO report, published in 2004 in collaboration with the World Bank, which presented a wealth of data and analysis. Despite a reality of more than 1.2 million RTI deaths, twenty to fifty million people injured and rising numbers the report insisted that RTIs had to ‘be considered alongside heart disease, cancer and stroke as a preventable public health problem’ which responded ‘well to targeted interventions’.Footnote ¹⁰⁷ The report recommended and discussed in considerable detail a series of well-known practical measures such as speed control, the use of seat-belts and adequate child restraints as well as helmets for two-wheeler users, the enforcement of alcohol limits, good road design, improved vehicle standards and efficient post-crash medical care. These were only presented, however, as pieces of a much larger puzzle. Other, less frequently cited but relevant strategies included reducing the deprivation of underprivileged social groups, the separation of different types of traffic in clearly marked separate roads with different speed limits, mandatory daytime running lights, intelligent seat-belt reminders, traffic calming measures such roundabouts, road narrowings, chicanes or road humps, or crash-protective roadsides.Footnote ¹⁰⁸ And even these points were merely tactical elements in a far more comprehensive approach, designed not primarily to make motorised traffic safer but to provide living conditions which would satisfy human needs for food, household items, work and leisure activities in intelligent ways.

This task might entail measures to reduce the need for travel by ‘[l]and-use planning practices and “smart growth” land-use policies – development of high-density, compact buildings with easily accessible services and amenities’ or the ‘creation of clustered, mixed-use community services’ or encouraging the use of electronic mail for communication.Footnote ¹⁰⁹ If moving from one place to another could not be totally prevented, it could be organised in safer ways than through private cars. Calculations for the European Union regarding the risk of death in relation to distances travelled listed motorised two-wheeler users as running twenty times the risk of car occupants, who were seven to nine times safer than cyclists or pedestrians but ten times less safe than bus and train occupants. Thus, travelling by public transportation was by far the safest means of transportation, creating good reason to encourage it.Footnote ¹¹⁰ Reducing the need to travel in general and to travel by car in particular was also considered a very positive step since, as studies from high-income countries indicated that ‘under certain conditions, for each 1% reduction in motor vehicle distance travelled’ there was ‘a corresponding 1.4–1.8% reduction in the incidence of crashes’.Footnote ¹¹¹ Besides, such measures could have tangible health benefits in addition to reducing RTIs, such as increasing healthier life-styles through more walking and cycling, and reducing noise and air pollution.Footnote ¹¹²

However, as the report acknowledged, many of the road safety measures would not be applicable in developing countries, where the need to act was greatest. Since the large majority of global RTI victims were pedestrians and cyclists in low- and middle-income countries and since in the foreseeable future most people in those countries would continue to be walking, cycling and using public transportation, their needs would have to be the priority concern.Footnote ¹¹³ In the absence of facilities for pedestrian and cyclists many people were forced to travel with privately run services. These services, often in obsolete overloaded vehicles, driven by overworked drivers and owned by businessmen, who bribed traffic enforcement authorities, created substantial risks to occupants as well as other traffic participants. Among other measures, strategies had to be found that would simultaneously address the ‘safety of road users, the labour rights of drivers and the economic interests of the vehicle owners’. Footnote ¹¹⁴ Such a traffic pattern with its mix of dangerous motorised vehicles sharing road space with large numbers of vulnerable road users had never been experienced by high-income countries and therefore lessons from North America and Europe were of only limited use in the rest of the world.Footnote ¹¹⁵ In addition, the report showed a more complex connection between RTIs and development by citing estimated global costs of the RTI burden of US$518 billion per annum and $100 billion in developing countries ‘twice the annual amount of development assistance to developing countries’.Footnote ¹¹⁶ In this perspective, RTIs did not appear as a temporary price paid for a generally beneficial economic development but as a powerful impediment to development.

This WHO report was the most detailed study on RTIs as a global health problem to that date, and was meant to become the central reference point for further research. Its wealth of information made it a publication which was impossible to ignore. Notably its numbers regarding present and estimated future RTI victims became the standard components of virtually all following studies. The system’s approach included near-revolutionary elements. Questioning the connection between motorised transport and modern economic development by decentralising residential areas or by making use of online communication means challenged the logic of the traditional view of development, more so than the various traffic reduction measures begun in OECD urban areas during the 1970s. While pedestrian zones restricted traffic locally, adopting a reduction of the need to travel as a goal of intelligent housing and work plans could potentially change the concept of development as people knew it.

The structure of recommendations underlined the comprehensive nature of the report, listing what governments, policy, legislation and enforcement, the public, vehicle manufacturers, donors, communities, civil society groups and individuals could do. Clearly, preventing or reducing RTIs was considered everybody’s responsibility. However, the very comprehensiveness of the report obscured its radical components in the midst of a multitude of more conventional ideas. Indeed, the large number of recommended measures allowed adopting a broad approach that nevertheless could leave out individual onerous ideas. In the summary, only two recommendations out of forty-seven called for the establishment of public transportation and none explicitly mentioned the inclusion of traffic-reduction objectives in land use plans.Footnote ¹¹⁷

The UN instituted the United Nations Road Safety Collaboration, designed to implement the recommendations of the World Report. The group consisted of regional and global international organisations, including the World Bank and UNICEF, and a variety of other national and international bodies (governments, non-governmental organisations, donors, research agencies and the private sector) where the WHO held a coordinating role. This strand has also led the UN General Assembly in March 2010 to proclaim the period 2011–20 as the Decade of Action for Road Safety.Footnote ¹¹⁸ Thus, there has been a real effort on the part of the UN to construct RTIs as a global issue and to assume responsibility for it.

Other organisations reacted with their own reports. Each referred to the 2004 World Report, and provided its own interpretation in line with its own outlook. In subtle ways, they carried out a competition of concepts: one designed to safeguard motorised transport by making it safer through technical, legal and administrative modifications, another designed to modify the entire system of transport by questioning the prioritisation of its motorised form. A report by WHO Europe, which also came out in 2004, was among the latter. It emphasised the synergistic value of various anti-RTI measures and the environmental component of overall anti-RTI policies. In the process, it portrayed finding ways to reduce RTIs as part of a strategy for sustainable development:

The same objective of ‘sustainable development’ was echoed in the title of Make Roads Safe: A New Priority for Sustainable Development, published by the Commission for Global Road Safety in 2006. The Commission was created by the FIA Foundation for the Automobile and Society in 2005 with former NATO Secretary-General Lord Robertson as chairman. Its advisory board included prominent people connected with the automobile sector, but also members of the WHO, the World Bank and the OECD.Footnote ¹²⁰ The campaign ‘Make roads safe’ has been supported by a number of celebrities and world leaders such as British Prime Minister Tony Blair and Archbishop Desmond Tutu.Footnote ¹²¹ The Commission also supported the GRSP created by the World Bank in 1999. Cooperation and overlap of personnel and website space with the GRSP was such that it was questionable to what extent these were distinct bodies. The GRSP website included a link from which the 2004 World Report on Road Traffic Injury Prevention could be downloaded and it issued guidelines for the implementation of its recommendations.Footnote ¹²²

However, below the surface of agreement to behavioural, technological and administrative changes, underlying differences about where transport policies in particular and socio-economic development in general should be heading were played out. Make Roads Safe: A New Priority for Sustainable Development focused attention on road improvements while referring to developmental challenges.Footnote ¹²³ The report argued that low-income countries could and should learn from the experiences of high-income countries but without having to imitate every step. Instead of repeating the Kuznets curve relation between economic growth and RTIs, found in Europe and North America in the 1960s, low-income countries should use additional measures as a form of shortcut to lower RTIs.Footnote ¹²⁴ Rather than aiming at a ‘system’s approach’, like the 2004 World Report, the Commission proposed a ‘safety system’s approach’, which recalled the Haddon Matrix of 1968. This combination of pre-crash, crash and post-crash measures for people, vehicle and environment made the approach appear comprehensive, while accepting motorised transport as a given and narrowing ‘environment’ to the technical qualities of a road: its design, markings, maintenance, protection, pedestrians’ crossing and rescue facilities.Footnote ¹²⁵ Improving the safety of vehicles held little promise in low-income countries, the report argued, since few people could afford modern cars and the variety of vehicles used prevented a quick effect. Ruling out an emphasis on people as a way of ‘blaming the victim’ the report focused on roads as central elements. This strategy was not limited to low-income countries since, ‘[a]mongst the best performing industrialised nations, improved road infrastructure remains the major source of expected future contributions to casualty reduction targets.’Footnote ¹²⁶ Nevertheless, the development of low-income countries, formed a central argument using the Millennium Development Goals (MDGs) as points of reference. The report criticised the MDGs for ignoring RTIs and insisted on the central importance of roads for achieving these goals in a long-term, ‘sustainable’ manner.Footnote ¹²⁷ Thus, according to the report, the aim could not be development without roads but development with lots of roads of good quality. A central recommendations, therefore, was that ‘at a minimum 10% of all road infrastructure projects should be committed to road safety and that this principle should be rigorously and consistently applied by all bilateral and multilateral donors’.Footnote ¹²⁸ Thereby, the report reconfigured RTIs from a development to a development-assistance-programme issue.

Three years later, the WHO issued a Global Status Report on Road Safety, which sought to provide an overview over the state of RTI related conditions in all countries worldwide. Based mainly on information gathered in a questionnaire, the report addressed a broad range of factors: institutional settings, the quality of data, vehicle and infrastructure standards, legislation on some of the main behavioural risk factors, medical care and exposure to risk. While the behavioural factors (speeding, drink-driving, use of motorcycle helmets, use of seat-belts and child restraints) received relatively most attention in terms of pages, the exposure to the risk, defined as ‘the existence of policies to encourage non-motorised modes of transport and public transport and strategies to achieve these, and levels of motorization’ were also addressed and related questions were included in the questionnaire.Footnote ¹²⁹ The results reflected years of prioritising transport by car: forty-four per cent of all countries worldwide had no policy, national or local, that encouraged public transport, and sixty-eight per cent had no policy that encouraged walking or cycling as an alternative to motorised transport. In an obvious attempt to produce an inspiring model, the report presented positive examples from Bogotá, Sweden, Delhi and Lagos, where policies of this type had tangibly reduced RTI numbers.Footnote ¹³⁰ In addition, the need for such policies was addressed in the first of five central recommendations, although it was framed as an issue of ‘road design and infrastructure, land use planning and transport services’, understating its systemic nature.Footnote ¹³¹

The World Bank had cooperated with both the Commission for Global Road Safety and with the WHO, and in many ways its position has been the most interesting and possibly the most important, given its financial clout and its influential voice in the development debate. A 2007 report, issued by the in-house Independent Evaluation Group of the World Bank, painted the picture of an institution in search of a position regarding a rapidly evolving issue. The report spelled out the stakes involved: an expected ‘huge expansion’ of the global automobile market, based on the motorisation of China and India, which predicted that ‘over the next 20 years, more cars may be built than in the 110-year history of the industry’.Footnote ¹³² Clearly, decisions in this issue involved lots of money, and a huge private market, in which the Bank and its investors could hardly remain disinterested, especially since those two countries had absorbed almost half of all World Bank lending in the transport sector since 2000.Footnote ¹³³ However, the Evaluation Group also recognised that the pressure of a three per cent annual growth rate of cars on global roads increasingly created environmental and health problems – and thereby eventually economic burdens.Footnote ¹³⁴ Times were changing and the Bank risked becoming outdated unless it changed, too. The report regarded the dominance of road building in the present World Bank transport portfolio as a threat to its long-term relevance. While highway construction would remain important, other ‘transport modes and themes’ were also becoming significant and it was ‘essential to see transport opportunities with a multimodal setting of integrated urban and rural concerns’.Footnote ¹³⁵ Given the importance of the ‘poor in urban and rural areas of developing countries’ and the bad condition of non-motorised transport facilities open to them, which had been neglected, including by World Bank activities, the report strongly recommended that the Bank begin funding related projects despite ‘the lack of or very low revenue-generation nature of such projects’.Footnote ¹³⁶ After all, holistic road safety approaches were increasingly being pursued ‘in all regions’.Footnote ¹³⁷ Presumably, the World Bank would be well advised to recognise these tendencies, their demands and opportunities at an early stage. In other words: