The Framework Convention Alliance for Tobacco Control

Industry’s dire warnings deconstructed


by Grieve Chelwa, PhD Candidate, University of Cape Town

In late October 2012, a study conducted by NKC Independent Economists was released that attempts to measure the primary elements of the tobacco value chain in 15 African countries that are part of the COME SA/SADC /SACU * regional blocs. The study was commissioned by the Tobacco Institute of Southern Africa (TISA), a body representing the tobacco industry in the region.

The study’s release coincides with the Fifth Conference of Parties (COP 5) in South Korea and is most likely meant to influence deliberations around Articles 17 and 18 of the Framework Convention on Tobacco Control (FCTC). Articles 17 and 18 speak to issues of alternative livelihoods and the protection of the health of workers engaged in the growing and processing of tobacco.

The aim of this short write-up is to critically engage with some of the findings of the TISA study, as well as to question the methods used by the study.

 Some of the study’s main findings are:

  • The tobacco industry (farming, processing, and selling) provides employment to a total of 4.4 million people who support a total of 24 million dependents in the 15 countries covered.
  • Taxes collected (VAT and Excise taxes) totalled of US$5.6 billion in 2011 for the 15 countries.
  • Total value of trade (imports and exports) was about US$3 billion for the 15 countries.
  • Total value of the tobacco value chain (raw tobacco production, exports, imports, taxes) was estimated at US$10 billion (see section 2.4).

In what follows, I interrogate these findings and the methods in closer detail.

2.0. General Comments

2.1. Data sources. Apart from consulting reputable data sources, such as the IMF, World Bank and FAO, for macro variables (which they mostly get right, see 2.5 below), other variables more pertinent to answering the research question seem to have come from organizations with an interest in the continued survival of the tobacco industry.

It is good practise to reference, in full, a particular data source (and to provide internet hyperlinks where possible) so that others may interrogate the methods and assumptions underlying a particular estimate. This is not done in this study, and it therefore casts doubt on the validity of the figures quoted in the report.

2.2. The study reports a total of 296,890 establishments (wholesale, formal retail, informal retail) selling finished tobacco products in the 15 countries covered. This number is reported without any further commentary but the implied conclusion is that these establishments rely on selling tobacco products and failure to do so would jeopardize their financial position. But these establishments sell other fast moving consumer goods (bread, milk, eggs, etc…) alongside tobacco products, and it is a stretch to imply that 296,000 entities would cease to exist if prevented from selling tobacco.

2.3. The study reports that there are 1.2 million small-scale tobacco farmers, employing 3.9 million people in the region. 18 million people are in turn dependent on the farmers and their employees for their livelihoods. Further, the study reports that the value of raw tobacco produced in 2011 was US$1.2 billion. This data comes from only nine out of 15 countries:  Angola did not report any data while Egypt, Bostwana, Lesotho, Namibia, and Swaziland had a “not applicable” entry against their names.

Do these numbers represent a significant proportion of each country’s total agricultural employment and is the income for each person reliant on tobacco agriculture significant? By analysing objective data from 160 sources alongside the TISA study, I have estimated the numbers of small-scale tobacco farmers and associated employment levels.

On average, tobacco employment constitutes 2.5 percent of total agricultural employment in the nine countries reporting. Excluding the two outliers (Malawi and Zimbabwe) would bring the average contribution of tobacco employment down to 0.93%.

As far as incomes are concerned, each person (farmer, employee and dependent) reliant on tobacco farming obtained an income of US$56 on average in 2011. This represents only 10% of the region’s average GDP per capita.

Lastly on this point, the study’s presumption is that 1.2 million small-scale farmers (alongside their employees and dependents) would be rendered destitute if they could not sell raw tobacco anymore.

This is unlikely to be the case. As Warner (2000) has argued, money not spent on buying tobacco (both raw and manufactured) would be spent on something else, either in agriculture or elsewhere.

This would create additional employment for those genuinely at risk of losing their livelihoods.

2.4. The study commits a couple of methodological errors in arriving at what it calls the Tobacco Value Chain Total Value (TVCTV ) for the 15 countries considered. As stated earlier, the value of this amount is estimated at US$10 billion. A footnote defines this number as the summation of the value of raw tobacco production, value of total exports, value of total imports and taxes paid. In adding all these numbers together, the authors are either guilty of double counting or adding “apples” to “oranges.”

Further, tobacco tax revenues are not at risk of declining (or disappearing) as countries try to reduce tobacco consumption through tax measures. The consensus in the economics literature is that tobacco is price inelastic. Therefore, excise tax hikes can only serve to increase the revenue collected by governments from tobacco.

2.5. I t seems that the study has correctly reported the usual macro variables (i.e. percent of Agricultural GDP due to tobacco, area under cultivation, production in tonnes, value of exports, imports etc…), if only because these statistics are easier to cross-check against reputable online databases. But the fallacy is in the way these figures are added together in quantifying the value chain (see point 2.4 above).

3.0. Conclusion

This short write-up has highlighted some of the methodological shortcomings in the TISA study released at the end of October 2012. Further, using the TISA study’s own-estimates, I have shown that the tobacco farming sector is not as important, from both an agricultural employment and income perspective, as the study makes it out to be. In addition, the value ascribed to the tobacco value chain is likely to have been overstated if one accounts for double counting.

This story was taken from Day five Bulletin issue 121, released at COP5 on 16 November 2012.

Share This