Sugar tax: IRR cannot be both umpire and player

Thomas Johnson says paid-for research will always be controversial

Sugar tax research: IRR cannot be both umpire and player

In a recent article on Politicsweb, Enough with the pseudo-experts, John-Kane Berman, a fellow at the Institute of Race Relations (IRR), sided with climate change denialists. He's as entitled to his opinion as climate change celebrities with whom he took issue. But his excessive, emotional language ("Holocaust denialists") made me wonder why he's so upset, and about IRR's mission of "promoting political and economic freedom".

My questions about IRR deepened with IRR’s CEO Frans Cronje’s response to media criticism of its Coca Cola-sponsored sugar tax study.

Consultants who conduct studies for clients – like IRR's Coca Cola sugar study – may feel obliged to tell clients what they want to hear and make favourable findings supporting the clients’ position. In a way it's what they paid to do. Grant Thornton's unsubstantiated rosy World Cup and Olympic economic predictions are examples of SABC "good story", emperor- without-clothes style of reporting.

Before Arthur Andersen’s ignominious collapse following Enron and other scandals, far too many “independent auditors” failed in their duty to fairly and factually report on clients’ financial position so as not to displease them and favourably present clients’ statements to end-users. (Note auditors are legally liable in the event they misreport when they know a third party, e.g., creditors, shareholders, etc, will rely on their report).

Even after Andersen’s and Enron’s fraud changed accounting forever with governments introducing more stringent legal requirements, auditors still skirt what’s ethically permissible. I worked as an accountant, and it was common knowledge and from direct experience this practice occurred, including among so-called reputable firms. (Needless to say, it was not one we condoned.)

Unlike auditing there’s no specific legal restriction on consultants (including audit firms conducting non-audit consulting work), excluding the usual contractual conditions, directing how they perform investigations or research for clients. But professional consideration applies: for the credibility of their work and reputations it’s incumbent they apply their minds to the facts impartially and clinically – they should not “change the facts to support the conclusion they want”.

Paid-for research, even when conducted by university-based researchers, is controversial because often the independence and, therefore, credibility of researchers are questioned. Examples are GMO companies – Monsanto appears to be the most cited – paying for research that exonerates GMO foods from adverse health and environmental effects. Scientific American's complaint about industry-sponsored GMO studies, one that applies to similar studies on climate change, sugar, etc, goes to the heart of the problem:

“Unfortunately, it is impossible to verify that genetically modified crops perform as advertised. That is because agritech companies have given themselves veto power over the work of independent [emphasis added] researchers. Only studies the seed companies have approved ever see the light of a peer-reviewed journal. In a number of cases, experiments that had the implicit go-ahead from the seed company were later blocked from publication because the results were not flattering.”

Of course, it’s unlikely clients would give researchers free rein to investigate the topic wherever it leads. They impose specific, often stringent, terms of reference that limits the investigation and ultimate credibility of the findings – manipulating facts to suit the conclusion desired (complaints about the Seriti Arms Deal Commission come to mind).

This happened with a case I recently wrote about – the City of Cape Town’s business plan study for the Cape Town Stadium. There consultants were only permitted to investigate the five business plan models the city gave them to reach the conclusion the city desired the stadium shall be financial viable, even when objective facts state otherwise.

In cases like these the motives of clients limiting the scope of the investigation or only publishing favourable conclusions – conclusions that supports its agenda – is dubious and dishonest and creates distrust. How far the “independent” researcher is complicit in this deception depends if he was aware his report would be misused.

In Frans Cronje’s article responding to media criticism of its Coca Cola sugar study he said:

“The Financial Mail alleged the IRR engaged in the SSB [sugar sweetened beverages] tax work knowing what the outcome of its research would be, and approached funders on those grounds. Of course we did. We have extensive experience of secondary taxes and know how they harm economies and delay the structural reforms that are necessary to secure real economic turn-arounds. That is why we approached the sugar industry in the first place.”

Let me play devil’s advocate in this dispute because IRR’s affiliations bear scrutiny.

Research is conducted to learn something new or confirm a theory, in other words, it’s supposed to expand knowledge. But IRR’s bid for the Coca Cola study when they a priori “[knew] what the outcome would be; off course we did” is suspect. And I’m surprised Cronje, who appears unable to see past his indignation, does not think so too.


It would be a different matter if IRR had already completed an independent, possibly but not necessarily unfunded, on spec research project and attempted to sell the output. Then its integrity and credibility could not be questioned, at least not by people who don’t have an axe to grind.

I also don’t have a problem IRR used its experience and expertise in a field – secondary taxes on the economy – because this is expected and common – you don’t hire an organisation that has no experience (except, of course, on BEE/government contracts).

But to claim the findings of a bespoke research study is completely credible when the a priori conclusion (the economy would be harmed) is not inevitable or obvious is disingenuous.

The main problem I have, though, is the IRR is not unbiased, as proven by Cronje’s quote above. While we, or most of us, have biases, tendencies and pre-conceived ideas, isn’t it the job of an independent researcher to put that aside and look at things squarely and unemotionally?

Cronje claims there’s “an impenetrable Chinese wall between our sales and fundraising staff, and our policy writers and researchers”. But how much, if any, of this operational independence, particularly as it relates to SSB and Coca Cola research, is negatively influenced by IRR’s staunchly pro-economic freedom and pro-private sector position? This is what he says:

“One of the reasons the South African economy is in such trouble is that the private sector is regularly denounced, insulted, disparaged and assailed when it warns against hostile and counter-productive policy. When it tries to explain its side it is silenced in exactly the manner the saga set out above attests to so well. At the IRR we want to give investors a voice so that their concerns may be heard and responded to.”

From this statement we may conclude Cronje and IRR devoutly believe the private sector are victims and are being subjected to unfair and vindictive policies and taxes. The sugar tax is one such measure. Is it not fair then to conclude they could not, as a matter of principle, reach a different conclusion, i.e., a sugar tax is beneficial to health, industry and the economy?

(Incidentally, I strongly disagree with Cronje industry is an innocent bystander to the economic circumstances the country finds itself in. I wrote before that by its controlled, monopolistic/oligarchic and innately uncompetitive nature and reaping exceptionally high profits, it must share blame with government.)

I don’t know what IRR means by “economic freedom”. Do they mean laissez faire, and little to no taxes and policy guiding the private sector? This is unrealistic.

But the (unintended) consequences of its entrenched and biased pro-private sector position means its Coca Cola report will be questioned because it’s not independent, it has a conflict of interest due to its connection to industry, and they had already reached an a priori conclusion the sugar tax would “harmful”. Coca Cola’s and the sugar industry’s cause would have been better served if they had commissioned fully independent, non-conflicted researchers without limitations of scope.

Putting aside the spat with media groups, which are not innocent and have their agendas, IRR/Cronje know any research is open to criticism and they should have expected it, particularly because in this instance it’s valid. Therefore, his high-dudgeon is curious. I suspect it’s more an attempt to defend the credibility of a study that probably cost Coca Cola a lot of money and already is being dismissed as biased soon after it left the printers.


I read IRR’s report “Against the Sugar Tax” and others in the media about it. For my penny’s worth (and for reasons that do not necessarily coincide with IRR), I agree the tax will not, and government knows it will not, meet its stated objectives. Therefore, it’s a revenue generating exercise in disguise.

According to Wikipedia, countries that implemented a sugar tax or are thinking about it are: France (2012), Mexico (2013), Norway, SA (2017), UK (2018), St Helena (2014) and US. Interestingly, Denmark had a “soft drink tax” since the 1930s but repealed it in 2014 and their “fat tax” in 2013. Wikipedia does not mention Hungary, which has had a “hamburger/crisps/fat tax” since 2011.

Quoting from the Journal of Adolescent Health and Contemporary Economic Policy (both 2009), Wikipedia says “taxes would need to be raised substantially to detect significant associations between taxes and adolescent weight", and “a percentage point change in a soft drink tax would affect body mass index (BMI) by a very small amount—about 0.003 points”.

It also states the long-term effects of the tax on obesity in Mexico has not been determined, and a general criticism, “There is limited evidence the sugar tax reduces consumption of heavily sugared products”.

Advocates of the tax should say if the three and four years since it was introduced in Mexico and France respectively is sufficient time for it to achieve its goals to a medical certainty. I doubt they would stake their professional reputation on it.

Tax on alcohol and cigarettes have had little effect on changing habits. South Africa’s alcohol consumption has climbed and SA is now one of the top 20 drinking nations, double WHO’s African region average. Regarding smoking, an article by P Reddy (2015) et al in the South African Medical Journal (vol. 105 no.8) states:

“Tobacco interventions used have included legislation such as the Tobacco Products Control Amendment Act, hikes in excise duty on cigarettes, and health promotion interventions to educate and improve individuals' health knowledge. Despite initial successes, there have been recent increases in tobacco use between 2008 and 2011 among SA youth, particularly girls.”

And we know bad habits frequently start in youth.

As with the plastic shopping bag tax (levy), the idealistic objectives were not met and it ended up being another source of revenue from an already wrung-dry taxpayer. As a lay economist I think the sugar tax is another empty, hand-wringing gesture from government who’s clearly out of ideas on how to change people’s bad habits. And they should not have to take all the responsibility. But please, stop trying to nanny people. Either allow them to have it, or ban it.

If they’re serious, an easily implementable proposal is to have legislated limits on the amount of sugar, fat, etc permissible in drinks and food – like limiting brine in frozen chicken – and have hefty fines on manufacturers for exceeding those limits. It’s easier controlling at source – the manufacturer.

But a sugar tax, like similar ones, is simply a revenue collection exercise with the knowledge it will not go where it belongs – to the health budget and to promote a healthy lifestyle among citizens.