Bihar's Alcohol Ban: What the Data Actually Shows

Why Crime Data Cannot Settle the Prohibition Debate

You are handed two numbers. One says crimes against women in Bihar kept rising after the 2016 alcohol ban, so the ban clearly failed. The other, from a peer-reviewed Lancet study using quasi-experimental methods, says the ban prevented roughly 21 lakh cases of intimate partner violence. Both numbers are real. Only one of them is measuring what it claims to measure.

The Data Problem Nobody Names

NCRB crime data records what gets reported to police. Severe, stranger-perpetrated violence usually is. Violence inside a household, by a spouse or family member, usually is not - not because it doesn't happen, but because reporting it carries social cost the victim often cannot afford.

When reported crimes against women rise over a decade, that can mean more crime or it can mean more reporting, better legal aid and less stigma. The report recommending Bihar lift its ban treats a rise in reported numbers as a rise in actual harm.

That is the error at the centre of this story.

What the Better Evidence Shows

Quasi-experimental studies are built specifically to isolate a policy's causal effect from these confounds. Two such studies - one in The Lancet Regional Health (2024), one a peer-reviewed evaluation published in 2024 - found substantial reductions in intimate partner violence after the ban and separately, a decline of roughly 18 lakh cases of overweight and obesity among men.

This is not proof the ban is costless. Lost excise revenue is real, estimated at 14% of state revenue and it is real money the state cannot spend on schools or hospitals. But a 2019 study in the International Journal of Drug Policy estimated alcohol use costs the Indian economy 1.45% of GDP annually after adjusting for the very tax receipts the report treats as pure loss.

Where This Leaves the Policy Question

The honest framing is not "ban or no ban." It is: does the state have an obligation to weigh a hard-to-monetise social good - women not being beaten, men not developing preventable disease - against an easy-to-monetise fiscal loss?

Article 47 answers that question constitutionally, directing the state toward prohibition of substances injurious to health. But a constitutional mandate does not exempt policymakers from asking whether prohibition is the only or the best instrument or whether rationing, access controls and stronger de-addiction infrastructure could deliver similar gains without the enforcement costs, black markets and hooch deaths that come with an outright ban.

The report that sparked this debate was not dishonest. It was incomplete - it measured the visible cost of prohibition and treated the invisible benefit as if it did not exist. That is the trap public health policy keeps falling into: what shows up in a ledger gets counted, what shows up in a hospital chart or a woman's decision to finally leave does not.