The reduction of the drink-drive limit in Scotland in 2014 has not led to a fall in road traffic accidents. With a new Road Safety Strategy in England and Wales, Jonathan James (Department of Economics, University of Bath) and Marco Francesconi (Department of Economics, University of Essex) discuss why the policy change may have had no effect and the importance of enforcement-led and targeted approaches.
The government’s new Road Safety Strategy sets out plans to lower the legal drink-drive limit in England and Wales, bringing it into line with Scotland and with most European countries. Scotland reduced its limit from 80 to 50 milligrams per 100 millilitres of blood (0.05% blood alcohol concentration, or BAC) on 5 December 2014, while the limit elsewhere in Great Britain has remained unchanged since 1967. Alongside the proposed change to the legal threshold, the strategy places greater emphasis on enforcement and targeted preventative measures, including the potential use of alcohol interlock devices for high-risk offenders.
Even though road safety has improved substantially over recent decades, road traffic collisions continue to impose a large burden on public health. In the year ending June 2025, more than 125,000 casualties from road traffic collisions were reported in Britain. In 2023, drink-drive collisions in Great Britain were linked to around 260 deaths and over 6,000 casualties overall, and although absolute numbers have fallen since 2022, the proportion of road casualties involving alcohol has increased over the past ten years. Policies aimed at reducing drink-driving therefore remain a central part of efforts to improve road safety.
Limiting BAC for drivers has become a key policy tool used by governments across the world in their attempt to save lives on the road. In 2001, the European Commission recommended that all member states should adopt a legal maximum limit of 0.05 BAC or lower. In the UK, the government-commissioned North Review in 2010 on the legal framework of drink- and drug-driving reached the same conclusion, recommending a reduction from 0.08% to 0.05%. The inclusion of a lower limit in the new Road Safety Strategy reflects this established consensus.
While lowering the legal limit reflects international best practice and long-standing recommendations, evidence suggests that its impact on road safety depends on how it operates alongside other measures. Enforcement, transport alternatives and policies targeted at high-risk drivers are likely to be central in determining whether behaviour changes and collisions fall. Scotland’s 2014 reform provides a useful case study for addressing these questions.
Our approach
To examine the impact of the law change we use detailed administrative data for all motor vehicle collisions reported to the police and recorded using the STATS19 accident reporting form from the Department for Transport. We use all records from 2009 to the end of 2016 on over 1.2 million accidents. Each record contains details about the accident and the individuals involved, including their age and sex, the exact time and location of the accident, and its severity (distinguishing between fatal, serious and slight). We also have information on alcohol involvement.
We take three approaches to estimate the effect of the law. The first is to compare accident rates in local authorities in Scotland before and after the introduction of the reform, with accident rates in local authorities in England and Wales before and after. The comparison in accident rates with English and Welsh local authorities is needed as rates across the whole of Great Britain have been falling.
Local authorities in England and Wales might not be a good comparison group to those in Scotland. Therefore, in our second approach, we repeat the same comparison but for those local authorities that are close to the Scottish-English border (within 50, 100 and 150km).
The third approach is to use a statistical technique that picks local authorities in England and Wales that closely follow the trend in Scotland before the reform was enacted. This also takes other characteristics into account, including weather conditions, road congestion, socioeconomic status of the population and alcohol availability in each local authority.
What did we find?
Across all three approaches, there is no evidence that the lower limit introduced in Scotland has reduced road traffic collisions. This holds for fatal crashes, serious injuries, slight injuries and drink-drive accidents specifically. The results apply across different population groups, different times (nights and weekends), and across rural and urban areas.
Nevertheless, the 2014 reform was associated with a rise in anti-drink-driving sentiment among the public. But this change in attitudes was not sufficient to produce measurable reductions in alcohol consumption or collision rates in Scotland.
Did the reform have unintended consequences?
Despite the lack of a measurable effect on road traffic collisions, the Scottish reform did not have any knock-on negative effects. There weren’t observable increases in alcohol consumption, vehicle use or criminal activity unrelated to drink-driving. Nor is there evidence the hospitality sector, including employment in pubs, suffered as a result.
These findings are important for the current policy debate. They suggest that lowering the drink-drive limit did not impose hidden economic or behavioural costs, even if it did not, on its own, deliver large road safety gains. From a policy perspective, this implies that concerns about unintended consequences should not be a central obstacle to lowering the legal drink-drive threshold in the rest of the UK.
Why was the law not effective?
There are three main factors that can help to explain these findings.
The first relates to alternatives to driving under the influence. The lowering of the drink-drive limit did not mean that that people suddenly had better or cheaper ways of getting home after drinking. The cost and availability of other transport options, such as taxis and buses, were unchanged.
The second relates to deterrence and enforcement. Our study indicates that the reform was not accompanied by a sustained increase in enforcement or changes in policing activity, such as higher police numbers, roadside breath testing (including tests not linked to collisions), or drink-drive arrests and convictions.
A third factor concerns the composition of drink-driving offences themselves. A significant share of drink-drive collisions involve drivers with blood-alcohol concentrations well above both the old (0.08) and new (0.05) limits. For these high-risk drivers, marginal changes to the legal threshold are unlikely to be binding. This helps explain why lowering the limit, in the absence of additional measures, did not lead to large reductions in collisions.
These mechanisms help explain why changes to legal thresholds are unlikely to be effective in isolation, and why enforcement-led and targeted approaches feature prominently in the new Road Safety Strategy.
What does this mean for the new Road Safety Strategy?
The evidence from Scotland suggests that lowering the legal drink-drive limit is unlikely to deliver large reductions in road accidents unless it is embedded within a wider package of enforcement and targeted measures. International evidence points in a similar direction: when drink-drive laws become stricter, their effectiveness depends critically on how they are implemented as part of a wider policy package.
If the lower limit included in the new Road Safety Strategy is to have an impact, it will benefit from complementary changes, such as improvements in late-night transport options and more visible and targeted enforcement. This could include higher levels of roadside testing and greater use of random breath testing.
It is also important to note that a large share of drink-drive accidents involve drivers well above the legal limit. Policies that directly target high-risk and repeat offenders may therefore be needed, including reforms to the High-Risk Offenders Scheme and the use of preventative technologies, such as alcohol interlock devices that will only allow a vehicle to start with a breath sample under a set limit. A greater focus on this high-risk group is consistent with evidence on where drink-drive harm is most concentrated. It is also reflected in the Road Safety Strategy’s emphasis on enforcement and prevention, rather than thresholds alone.
This blog is based on ‘None for the Road? Stricter Drink Driving Laws and Road Accidents’. Read the paper in full.
All articles posted on this blog give the views of the author(s), and not the position of the IPR, nor of the University of Bath.
