Taxation and Regulation of Plastic Shopping Bags in Botswana and South Africa

While many wealthy countries have tried to reduce the use of plastic shopping bags, Bostswana and South Africa recently tried to do the same by charging for bags that were previously issued for free. This commentary describes recent research on these two countries that yielded some surprising results: the long-run effect of the price on plastic bag use was smaller than that in the short run, and the effect did not vary much across income groups. The smaller long-run effect suggests that it may be more difficult to permanently reduce plastic bag use in these countries.

Plastic disposable shopping bags have proved themselves valuable social assets to households throughout the third world. Despite their light weight, strength, and durability, there are downsides to such bags: once ripped or torn they lose their value, and when too readily and cheaply available, the incentive to re-use them drops sharply. Disposable shopping bags have become a feature of life in Botswana since 1975, when they were first widely issued. In South Africa, they have been issued to shoppers gratis at most major supermarkets for even longer. This long-term, widespread use has resulted in a negative visual externality: no one is responsible for the used bags littering urban streets and catching on fences in rural areas.

A point of debate is whether this externality has a more physical dimension. That plastic bags and related ties, cordage, and netting can be unsightly is clear. However, documented cases of wild and domestic hoofed animals, such as goats and pigs, dying as the result of ingesting plastic bag remnants are surprisingly rare, given the public profile of this issue. (As this was written, a Google search turned up 821,000 hits for sites with the words “plastic bags kill gut mammals.”) From a pure cost-benefit perspective, it is far from clear that these physiological externalities would warrant intervention.

Despite the poor scientific support for intervention, there has been a growing public conviction that disposable plastic packaging, and in particular shopping bags, impose environmental externalities. In the European Union, society has long been educated regarding the need to separate wastes, minimize use of oil-based products and so on and unsurprisingly, charges on plastic bags were widely supported and proved effective. The United States is slowly following suit; the first state-level plastic-bag legislation was recently enacted in Washington, DC, although several cities and counties have had them in place since 2007.

Interestingly, this has not been confined to affluent societies. Following NGO and public pressure, Botswana intervened in 2007 to cut the use of the lightweight disposable plastic shopping bags that were formerly distributed gratis at supermarkets. Four years earlier, in May 2003, South Africa had introduced similar legislation regulating the gauge of plastic used in plastic bags, and requiring open payment for them at checkouts. Active lobbying by plastics manufacturers saw substantial easing of the regulations initially proposed. But in smaller economies such as Botswana’s, bags are typically imported, and lobbying by industrial interests is less likely.

What interventions took place and how successful have they been? In an attempt to answer these questions, we sought details regarding the number of bags issued monthly, the sizes of the bags, and the prices charged from the four major retail chains in South Africa and Botswana. Because each of the four chains target a different income group, the data gave some sense of the income elasticity of demand for plastic shopping bags, as well as the price elasticity.

In both countries, the legislation had begun by trebling the minimum gauge of plastic used. The resulting bags could be loaded more heavily without bursting, and could withstand far more re-use. In Botswana particularly, the focus of policy was not merely on curbing the number of plastic bags issued, but on encouraging their recycling, reducing plastic litter, and ensuring the safe disposal of residual plastic waste. The private sector mutually agreed to aid the process by no longer distributing bags without charging, though the actual prices levied for them varied across the major retailers.

Our key finding was that charging for bags had a far greater impact in the short run than in the long run. Indeed the actual price charged was less important than the act of charging, a finding that held true over the entire range of prices being levied. Moreover, this effect seemed relatively insensitive to the incomes of consumers. The initial impact of levies was also far greater than that of subsequent price increases. Conventional utility theory, which presumes that people act rationally, gives little help in explaining this pattern of behavior. More useful might be prospect theory—or at least the endowment effect, the notion that people attach an irrationally high value to things they see as “theirs,” especially when these are threatened. The public may have seen free bags as a “right” and therefore reacted against paying, irrespective of the price. Once they had become accustomed to paying for plastic bags, the price became relatively insignificant as, for most of them, the price of the packet was low relative to the cost of the purchases made and carried in it.

In order to estimate consumer responsiveness to these levies, it was necessary to normalize the use of bags for the volume of purchases and for the volume of the bags used. The result can then be expressed as the real value of purchases per bag. Any increase in this figure indicates success in the policy.

The impacts of the charges have varied across counties and across time. In Botswana, the charge appears to have been more successful than in South Africa, albeit the charge has been in effect a far shorter time. Simply showing the number of bags issued could misrepresent the impact of the charges over time, since the volume of retail sales also drives the demand for bags. Accordingly, the effect of the 2007 legislation and charges is represented using an index comprising the number of bags issued by each of the four major retail chains, divided by the real value of their sales.

Firm 1 focuses on high-income consumers, while firm 4 targets a lower-income market-segment. It is clear that the success of the policy has been greatest among consumers from these two extremes.

In South Africa, the impacts of the charge have been far less clear cut. Surveys of consumer behavior indicate that relatively few are reusing the bags to carry shopping; instead, they are being used to carry household waste. Recycling waste-contaminated bags is economically unfeasible. In this respect, the new legislation has not been successful. Moreover, data from three of the four retail chains indicated that, despite the charges, bag use increased once consumers had become accustomed to paying for them.

Our findings identified two interesting aspects of the problem. First, when levying charges for plastic bags, consumer sensitivity to the charge declined over time; that is the sensitivity of plastic bag demand to price was less in the long run than in the short run. This is the antithesis of conventional textbook economics. Second, both the very poor and very rich have revealed a willingness to pay for good heavy duty plastic bags. This is shown in the diagram below. Again, the index shows the number of bags issued by each firm corrected for any changes in their total real sales volumes.

Firm 1 targets high-income consumers, while firm 4 targets low-income consumers exclusively. In comparing figure 2 to figure 1, note that South Africa’s plastic bag levy was implemented in 2003, four years ahead of Botswana’s.

It can be seen that in both Botswana and South Africa, shoppers in retail outlets that target low-income consumers demand just as many if not more bags per $100 worth of shopping. Two reasons have been postulated. One is that, not having cars, and typically living at a distance from the retail outlets, they do not shop in bulk, but buy small amounts as and when needed. Their bags are therefore less fully filled. Another is that, when buying similar products of similar volume, the poor are likely to buy the cheaper item. The results are clear: in both countries the plastic bag charge has been clearly regressive.

South Africa and Botswana’s experiences with regulation and taxation of plastic bags may offer useful insights for other LDCs (least developed countries). Certainly their experiences suggest that long-run reduction of plastic bag use may be more difficult to achieve in LDCs than it has been in first-world economies, such as Ireland or Denmark, where ecologically aware consumers pushed for the legislation and were already accustomed to waste sorting and recycling. In retrospect, it is unclear whether or not the threefold increase in the gauge of plastic being used was warranted. It is also clear that publically displaying the price of bags and stressing the payment may make the tax more effective.

The particular interest of Botswana’s experience lies in its voluntary roots. This was legislation that the public wanted. In South Africa, it had long been noted that it was stores targeting the very poor (and the very affluent) that issued heavy-gauge bags. Despite the regressivity of plastic bags charges, opposition to the regulation and pricing of plastic shopping bags did not come from the poor in either country. In the absence of price competitive biodegradable packaging, regulation and taxation of plastic bags may appear to offer a feasible short-term solution, even in LDCs, to the problem of plastics in the environment. The South African experience, however, suggests that consumers adjust to such taxes and charges, and in consequence these may provide only a limited solution in the long run.

​