Income and Road Safety in OECD Countries

This article, titled “The relationship between income level and road traffic deaths: an empirical analysis for 22 OECD countries“, delves into the significant global challenge of road accidents and their profound socio-economic implications. Road accidents are unequivocally recognized as a major socio-economic problem worldwide, contributing to thousands of deaths, injuries, and disabilities on a daily basis. Globally, approximately 1.3 million people die annually due to road traffic accidents, with a striking 93% of these fatalities occurring in low- and middle-income countries. The economic burden imposed by these accidents is immense, often consuming an estimated 1–3% of the Gross National Product (GNP) in many nations. Furthermore, it is projected that fatal and non-fatal road injuries will collectively cost the global economy approximately $1.8 trillion between 2015 and 2030.

Authored by Yuksel Bayraktar, Serdar Aydin, Mehmet Firat Olgun, Ayfer Ozyilmaz, and Metin Toprak, the study sets out to comprehensively investigate the complex interplay between road accidents and income levels across 22 OECD countries over a substantial period from 1975 to 2018. Beyond this, it critically examines the convergence of traffic accident deaths for 34 OECD countries between 1994 and 2020. The research addresses a notable gap in the existing literature by employing advanced econometric techniques. For analyzing the relationship between income and road accident deaths, it utilizes Poisson Regression, Negative Binomial, and Quantile Regression Fixed Effect models. For the convergence analysis, a sophisticated fractional frequency unit root test with structural breaks is employed. These specific methods were chosen to provide robust estimations, especially crucial in scenarios where count data (like accident deaths) might not follow a normal distribution or when variables exhibit extreme values or unobservable individual heterogeneity. The study specifically focuses on GDP per capita as an indicator of income level, given its direct relevance to the income-road safety relationship.

The Methodology involved careful consideration of data characteristics. Since the number of deaths due to road accidents is count data, the Negative Binomial Regression method was preferred over Poisson Regression when overdispersion was present, as Poisson regression can lead to biased and inconsistent estimates under such conditions. The Negative Binomial Fixed Effects model explicitly accounts for heterogeneity and relaxes the assumption of equal mean and variance. For situations where the dependent variable, though not strictly count data, might have extreme values, the Quantile Regression Fixed Effects estimator was used. This method is less sensitive to extreme values and accounts for unobservable individual heterogeneity, which is particularly relevant when accident deaths may vary significantly across countries. For the convergence analysis, the stochastic convergence method was utilized, transforming the data to examine if individual country death rates converged to the OECD average. The Bozoklu et al. structural break fractional frequency unit root test was applied to test for stationarity, which indicates the presence of convergence, taking into account potential shocks from country dynamics. The variables used in the models included Fatality (road accident deaths), lnGDP (GDP per capita), lnHS (Health spending), and lnAC (Alcohol consumption), all taken as natural logarithms in the models.

The key findings of the study are multifaceted and insightful. Firstly, a prominent inverted U-shaped nonlinear relationship between road accident deaths and per capita income was discovered. This finding is consistent with the Kuznets Curve hypothesis, suggesting that as per capita income increases, road accident deaths initially rise up to a certain threshold, after which they begin to decline. The underlying mechanisms for this often involve an initial increase in vehicle ownership and travel demand in the early stages of economic development, coupled with less investment in road safety. However, as income levels further rise, societies demand and governments implement more robust road safety policies, invest in infrastructure, and individuals may opt for safer vehicles, leading to a reduction in fatalities. The calculated turning point, where deaths begin to decrease, varied slightly depending on the estimation method used. It was calculated as $15,578 using the negative binomial fixed effects method. For countries with low accidental deaths (20th quantile), the turning point was $14,738, while for those with high accidental deaths (80th quantile), it was $14,845 using quantile regression fixed effect estimation. This indicates that countries with lower accident rates reach the turning point at a slightly lower income level.

Secondly, the study found that increased health expenditures significantly reduce road accident deaths. This effect was negative and statistically significant at the 1% level across various models. Specifically, quantile regression results indicated that a 1% increase in health expenditures reduces accidental deaths by 0.833% in countries where accident-related deaths are low. This aligns with previous research highlighting the positive impact of a stronger health system and access to quality medical care on road safety outcomes.

Thirdly, higher alcohol consumption leads to an increase in these fatalities, showing a positive and statistically significant effect on road accident deaths at the 1% significance level. Quantile regression estimation results demonstrated that a 1% increase in alcohol consumption increases accidental deaths by 0.52% in countries with high accident-related deaths. This finding reinforces the well-established detrimental impact of alcohol on driver abilities and road safety.

Finally, regarding convergence, the analysis demonstrated that 21 out of 34 OECD countries are converging to the OECD average in road accident deaths. This suggests that policies aimed at prevention and improvements in road safety are effective in these nations. The concept of convergence is often linked to “technology transfer,” where countries with higher accident rates adopt road safety technologies and policies from those with lower rates, as well as harmonization efforts in policies, particularly relevant for groups like EU countries. However, 13 countries did not show convergence.

The authors conclude by emphasizing the critical role of policymakers in implementing and reinforcing effective road safety measures. Key policy recommendations include:

• Robust infrastructure investments, such as improving traffic lights, curves, and widening highways.
• Implementing deterrent penalties for risky behavior by pedestrians or drivers.
• Launching public awareness campaigns to educate citizens on traffic rules and life safety.
• Encouraging the use of public transport in densely populated areas to minimize traffic congestion.
• Developing regulatory policies and penalties for motorcycles, including mandatory helmet use, given their significant share in road accidents.
• Ensuring adequate inspection, high safety standards, and improved quality of regulatory institutions.
• Measures to prevent drug and alcohol consumption while driving.
• Enhancing the speed of access to accidents and the quality of medical care.

The study’s findings provide crucial insights for policymakers, especially by highlighting the income threshold for the decline in road accident deaths and the importance of targeted interventions like health spending and alcohol regulation to improve road safety outcomes in OECD countries.

Reference: Bayraktar, Y., Aydin, S., Olgun, M. F., Ozyilmaz, A., & Toprak, M. (2025). The relationship between income level and road traffic deaths: an empirical analysis for 22 OECD countries. BMC Public Health, 25, 2809. https://doi.org/10.1186/s12889-025-23726-9