This column was first published in Business Day. 

The government, guided by the ANC, does not have a culture of evidence-based policymaking. This is unfortunate. There have been several determined efforts to shift it. Yet we are going into a phase where economic policy needs to shift dramatically but the tool to do it — evidence — is blunter than ever.

Things have been getting worse. The spectacle of saving the doomed SAA, purely because the ANC decided it should be before the Covid-19 crisis, has done serious damage to the government’s fiscal outlook. Not because the R10.5bn payout to resuscitate the airline is that material in the R2-trillion to be spent this year, but because it damages the government’s position on the public service wage bill, which is material.

The government is arguing in court that its current wage agreement must be broken because the country faces a fiscal crisis. It is appealing to regulations under the Public Service Act that says a collective agreement can only be entered into if the costs can be covered.

It must argue this while somehow ignoring the embarrassing fact that it is spending billions on the airline despite the fiscal crisis. There is no evidence that spending on the airline will deliver the government’s own policy priorities, particularly those set out in its own 2017 National Civil Aviation Policy white paper (itself an example of good evidence-based policy thinking).

Perhaps the more damaging case of ignoring evidence has been in the response to Covid-19. By and large, the government has attempted to react quickly in line with the science, at least on the medical front. But there have been glaring errors.

Last week the high court in Cape Town declared the ban on the sale of tobacco products to be one of those. You will remember, especially if you’re a smoker, that sale of cancer sticks was banned during the first lockdown. This, of course, did not stop consumption of them, it just meant smokers had to buy from car guards instead.

The result was a dramatic flourishing of the illicit economy, which is untaxed and finances untold other horrible illegal activity. This diversion of legitimate economic activity was highlighted through research at the time. The court, however, focused on the scientific connection between the ban and the battle against Covid-19 finding that there was none.

The lack of evidence-based thinking has also been prevalent in the economic response to the pandemic. It was hard to understand the initial ban on e-commerce during the lockdown, particularly as so many other countries had promoted e-commerce as a safe alternative to normal face-to-face trading.

There was also the spectacle of the list of “essential” clothing items that shops would be allowed to sell when they reopened published by the department of trade, industry and competition, that banned the sale of swimwear in May.

This is nothing new. One notorious earlier example was the 2014 visa rules that forced foreign tourists with families through a bureaucratic nightmare of carrying birth certificates to enter the country. This was damaging to the tourism industry and did nothing to confront the alleged problem being targeted of the trafficking of children.

Now, let me say that the concept of evidence-based policymaking can be abused. The whole point is to escape ideology about what policies should consist of. Fundamental objectives, such as eliminating poverty and inequality, are political choices.

Evidence-based policymaking gathers evidence and tests interventions to see whether they work to deliver on those objectives. Ideological shibboleths such as free markets or state-ownership can be tested to see if they deliver the desired outcomes. Successful policies are adopted and unsuccessful ones cast aside, no matter the ideological baggage.

Ultimately, the constitutional imperative that policy be rational demands that it be evidence based. This idea was baked into the National Development Plan, which was tasked with developing the research and evidence for recommendations to the government.

It was also driven by the regulatory impact assessment process introduced by the Mbeki government in 2007. This set down steps policymakers must take, including assessing alternative policy options and determining the impact on the economy, society and the environment. Any new draft law should have an impact assessment report along with it to be presented to cabinet.

That policy, however, largely gathered dust. It was an uncomfortable experience for the few political heads that tried it. Those undertaken were cursory and after the fact. They were dropped in 2013, and then replaced with the “socioeconomic impact assessment system” in 2015.

This set out rules for new policy documents that were sent to the cabinet to have detailed assessment reports to accompany them. It does not seem to have been more successful than the 2007 version. It is notionally still a requirement in policy development, but I have yet to see any such assessment documents.

Ultimately, evidence-based policy must be a culture more than a set of rules. Being attuned to and respectful of the evidence, even when it conflicts with deeply held beliefs, is a sign of intellectual honesty. It is a value system that should be fundamental to our policymakers.

Yet, as we go into a year in which policy will have to draw the line between financial collapse of the state or a robust economic recovery, its absence is being keenly felt.

• Theobald is chair of Intellidex