Causal inference, alcohol bans and Covid-19 in South Africa: a short comment

As in other countries, South Africa has used various forms of restrictions on societal activity in an attempt to slow or prevent the spread of SARS-CoV-2 (‘Covid’). One measure that is relatively unusual is the limit on alcohol sales, which has varied in severity from a complete ban on any sales or transporting of alcohol to less severe variations on that such as banning only sales for off-site consumption, or limiting such sales to particular days and hours.

Such measures have drawn some vehement criticism, not least from the alcohol industry itself. One large player in that industry, Distell, commissioned a piece of research which argued that there was no defensible basis for these measures. That in turn was widely cited in the media, and at least one editor claimed that it showed: “There’s no way the alcohol bans in SA have been based on credible science. They’re based on prejudice.”

In the same month (April 2021) I was contacted by a civil society organisation for an expert opinion on that report. I wrote a short assessment, which takes a dim view of the approach and claims of the report – with corresponding implications for associated assertions that use it as a ‘scientific’ basis for opposing alcohol restrictions.

Alcohol_trauma_SMM_FINAL

As I indicate in my comments, this is ultimately an empirical question on which I have no prior views. The claim that a reduction in access to alcohol does significantly reduce the demand for hospital resources that are needed for critical Covid-19 cases is plausible. Whether it is true remains to be seen. A number of papers have been published on the subject, see:

https://onlinelibrary.wiley.com/doi/full/10.1111/dar.13310

https://www.sciencedirect.com/science/article/pii/S2211419X20301464

https://journals.co.za/doi/abs/10.4102/phcfm.v12i1.2528

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7719204/

I leave thoughts on those, and others which are likely to be out soon, for later work.

On the issue of economic impact, which the alcohol industry emphasises, there are certainly also concerns. However, it is useful to remember that industry estimates of economic harm from limiting their activities are often exaggerations of the net economic impact. Reductions in consumption also have a significant negative impact on government revenue from excise duties, though these are arguably quite small when compared to the broader economic and fiscal harm of ‘lockdown’ measures.

Elsewhere I have outlined in detail my views on the balancing act required of decision-makers, especially for less wealthy countries, in dealing with the pandemic. I argued that contrary to the conventional wisdom in 2020, South Africa’s response was deeply flawed and caused social and economic harm without adequate benefits in terms of long-term health outcomes. That remains my view, but it does not follow that every decision is flawed: in my assessment, the restrictions on alcohol sales/consumption, even if unnecessary or ineffective, are amongst the least of the government’s failures.