An alcohol licensing regime should have one big job: to ensure that licensed outlets operate responsibly, first by vetting applications and then by monitoring compliance.

Its measures should be proportionate to the risks being addressed, and cost-effective. Licensing should not be the primary tool for reducing every kind of harmful alcohol-related behaviour. It needs other policies targeted at specific harms. Without those, too much of the burden of harm reduction falls on local alcohol policies and licensing conditions.

The Sale and Supply of Alcohol Act does what it says on the tin. It regulates the sale and supply of alcohol, including through outlet licensing and the Local Alcohol Policy framework.

The Government’s proposed amendments make modest improvements. Very low-risk activities, like providing a small drink as part of a hairdressing service, will no longer require a licence. It will be easier for non-wineries to run tasting rooms. Licence applicants will have a right of reply to objections, improving procedural fairness.

But the bigger problem remains. Local alcohol policies and related licensing conditions are not particularly effective at targeting specific harms.

Some current processes also allow official reporting agencies, like Police and Medical Officers of Health, to impose “voluntary” conditions on applicants such as prohibitions on the sale of single cans, restricted hours, or one-way door policies for bars – as the price of avoiding a contested licensing hearing. That process must change.

Our submission suggested adding a proportionality and cost-effectiveness principle to the Act. Measures taken under the Act should be used only where they are a cost-effective way of addressing harms, and only where expected benefits exceed the costs, including costs imposed on lighter drinkers and their suppliers.

Applicants should also have a right of reply to official reports.

We also noted complementary measures, outside the Act, that are better targeted at substantial harms.

For example, South Dakota targets repeat alcohol-related offending through monitored no-alcohol conditions. Participants must avoid drinking and are tested frequently; violations bring swift but modest sanctions, typically a night or two in the cells. RAND’s evaluations found reductions in repeat drink-driving arrests and domestic-violence arrests.

New Zealand could trial a similar model and, if it works here, adopt it.

Abstinence is up and hazardous drinking is down. A licensing regime focused on responsible sale and supply, complemented by separate measures targeting specific harms, would work better than asking licensing to do everything.

Dr Eric Crampton's submission, Sale and Supply of Alcohol (Improving Alcohol Regulation) Amendment Bill, was lodged on 14 May 2026.