The need for more research was the reason usually trotted out by Harry Duynhoven, the previous Government's Transport Safety Minister, to excuse inaction over cellphone use by drivers.

It was a nonsensical defence, given the avalanche of research that showed the danger the practice posed to others on the road. The current Transport Minister has at least acted swiftly to ban cellphones in cars. Yet now Steven Joyce has reached for the same excuse to justify the Government's decision not to lower the blood-alcohol level for drivers from 80mg of alcohol per 100ml of blood to 50mg.

It must be a galling exercise for the minister because he surely knows there is more than enough evidence to take this important road-safety step.

A mountain of research is available from overseas, including Australia, which long ago moved to a 50mg limit. This is the point, according to the College of Physicians, at which most people's brains start to fuzz over.

It reckons a driver reading 50mg of alcohol is twice as likely to crash as one with no alcohol. One with 80mg is seven times more likely to crash. However, Mr Joyce says New Zealand-specific research on the level of risk posed by drivers with a blood-alcohol limit between 50mg and 80mg is needed, and this will be done over the next two years.

He does not explain why drivers here would react differently to alcohol and somehow set themselves apart from international findings.

The research has been enough to convince New Zealand health experts, and the likes of the Alcohol Advisory Council, the Drug Foundation, the National Addiction Centre and Alcohol Healthwatch, to strongly advocate lowering the limit. The Law Commission and the Ministry of Transport are, similarly, adamant.

The ministry estimates a 50mg limit would save between 15 and 33 lives, prevent up to 686 injuries and save between $111 million and $238 million every year. In situations like this, where those with expertise are virtually unanimous, governments should provide a principled lead.

That demand, in itself, negates Mr Joyce's other defence, that the Government must carry the public with it. Governments would soon make themselves redundant if they allowed themselves to be ruled by opinion polls.

His argument becomes even more irrelevant when the results of polls this year are considered. About 70 per cent of respondents supported lowering the limit.

Earlier polls found something closer to a 50:50 split. That suggests the public is becoming more concerned about the role of alcohol in a road toll that has remained stubbornly high.

The popular backing would surely escalate even further with greater awareness of what a 50mg limit means. Australian guidelines suggest that at this level the limit for men is two standard drinks in the first hour and one per hour thereafter. For women, the limit is one standard drink an hour.

Environmental Science and Research guidelines are even more generous. These levels are consistent with a social drink. They are not the stuff of a radical crackdown that would prompt outrage about the Government acting as a killjoy.

The Automobile Association, hardly an expert in medical matters, is one of the few supporters of the Government's decision. It has always wanted the emphasis to be placed on repeat drink-drivers.

Its appeal has been answered, with a zero limit for recidivists and under-20 drivers, as well as the introduction of car interlocks for repeat offenders. These initiatives are welcome, as would be sterner enforcement.

But the Government had no good reason for not making a 50mg limit the centrepiece of this road-safety package - only the fear of losing the votes of the unreasonable few who would see this as another nanny-state intrusion.