Policies to Promote Refillable Beverage Containers

Policies are what have driven the success and preservation of refilling in many Canadian and European beverage markets. The Canada and the Europe pages of this web site describe particular implementations of several types of policy instruments that promote or require refillable beverage containers. These implementations illustrate the many possible variations of each policy instrument. Furthermore, the experiences of some Canadian provinces and of some European nations illustrate the strengths and the weaknesses of the policy instruments that they have instituted and illustrate the characteristics of effective refilling laws. In a unifying context, this page discusses the types of policy instruments that can be used to promote or require refillable beverage containers, their strengths and weaknesses, and the ways in which they can be formulated. These include:

Refilling laws are usually formulated under the authorization of a broader legal framework. An act regarding waste management, environmental protection, product control, beverages, beverage containers, or packaging authorizes the government to pass legislation or to make regulations specifically for beverage containers. Policy instruments for refilling fall into three broad categories: economic, regulatory, and contractual. Economic instruments include deposit laws, eco-taxes, tradeable permits, and government subsidies. Regulatory instruments include bans, mandatory refilling laws, quotas, and mandatory stocking laws. A contractual policy instrument usually requires industry to enter into an agreement with the government.

Whether it is an economic, regulatory, or contractual instrument, formulating an effective refilling policy requires understanding how each actor in the packaging chain influences the packaging mix in an unregulated beverage market. Packaging manufacturers have no interest in refillable containers, but they have no influence on the beverage packaging mix [AGM, p. 69]. Consumers can influence the packaging mix in three ways. First, by refusing or neglecting to return empty bottles, consumers can deprive bottlers of the economic advantages of refilling and thus force bottlers to consider packaging more of their products in one-way containers. Consumers can also explicitly demand beverages in refillable containers, but only the most environmentally-conscious consumers likely will do so. Finally, many consumers in Latin America and in other parts of the world can afford to buy Coca-Cola and other soft drinks only in refillable bottles.

Bottlers who want to offer consumers beverages in refillable bottles must find a retailer, bar, restaurant, or other market that is willing to accommodate refillables and willing to take bottle returns. The market must also be close enough to the bottler to make the refilling system economically advantageous. On the other hand, bottlers who can control the marketing and distribution of their products may also be able to control their packaging choices [SAPH, p. 169]. In the United States, for example, some dairies own and operate chains of convenience stores. One such company, Stewart’s, packaged and distributed milk and soft drinks in refillable bottles to its stores in New York and Vermont [SAPH, pp. 282-300] but stopped using refillables during the 1990s. In Europe, many breweries own restaurants and pubs, and this arrangement makes refilling the ideal packaging option for beer [AGM, p. 60]. Bottlers can influence the packaging mix also by manufacturing a beverage product that is unique and popular in its market or a product that consumers identify as a local brand to which they are loyal [SAPH, p. 42]. Finally, just as Coke and Pepsi used advertising to glamorize one-way containers during the 1960s [CW, pp. 142, 324], they and other soft-drink giants can promote refillables in the same way. Beer companies can do likewise with their own vast advertising resources. Without proprietary or other special markets, however, the bottler’s control of the packaging mix is constrained by retailers’ preferences.

Of all of the actors in the packaging chain, indeed, retailers are the most powerful and the most adamant opponents of refilling [AGM, p. 89]. The retailer is the actor who determines the packaging mix by demanding that bottlers deliver beverages in one-way containers [AGM, p. 89] or by selling their private-label (store-brand) beverages in one-way containers at prices much lower than those of the national brands. Retailers can indirectly demand one-way containers, moreover, by refusing to take bottle returns or by otherwise interfering with the bottle return process.

Because retailers have the most influence on the packaging mix in any particular beverage market, a successful policy instrument must either compel or motivate retailers to sell beverages in refillable containers, through either direct or indirect means. Most importantly, the government must consider the current packaging mix when choosing a policy instrument and formulating its implementation as a refilling law. For example, economic instruments which are designed to give refillables a price advantage can fail in markets in which consumers have very few choices of beverage products in refillable containers or in markets in which an insignificant volume percentage of beverages is packaged in refillables. In markets in which one-way containers dominate, the most effective policy instrument is one which forces a complete or partial transition to refillables. In markets in which refillable containers dominate but one-way containers are gaining market share, the most effective policy instrument is one which is designed to preserve refilling. In such markets, the government has more policy options–economic, regulatory, or contractual–than it does in markets in which refilling is insignificant. All of the refilling laws in Europe and most of the laws in Canada were implemented while refillables still dominated the beverage markets in these places.

In order to work effectively, any refilling law must have a few essential provisions. One of the most important provisions is a set of criteria to define the refillable container(s) and refilling system. Such criteria may include any combination of the following.

  • The type of product with which the container must be refilled
  • A minimum number of refillings that the container must be able to withstand
  • The physical characteristics of the container
  • The amount of the deposit under which the container is sold
  • Minimum return rates, which gradually increase after the refilling system has been established
  • How effectively the labelling of the container indicates that it is refillable and that it is sold under a deposit
  • Other specifications for the deposit-return system

A successful refilling law must also enable clear and unambiguous detection of violations or have provisions for reporting compliance. The fines or any other economic sanctions must exceed the benefits of violating the law [AGM, p. 76]. Furthermore, governments must be prepared to administer and enforce the law by having the personnel, the resources, and the legal authority to do so. Other aspects of a refilling law help ensure its success in promoting or requiring the use of refillable beverage containers. Continued support for a refilling law from the public and from lawmakers may depend on how well it accomplishes its designated environmental, economic, or market objectives. Such objectives often include waste diversion or litter reduction. To ensure its success, finally, a refilling law must be effectively promoted and publicized [SAPH, pp. 245, 246]. For more discussion about policy instruments that require or promote refilling, see [AGM][OECD][ERC][CNWM].

Deposit Laws

A deposit law requires beverages to be sold under a deposit-return system, a system in which consumers pay a deposit for the containers when purchasing beverages and receive a refund of the deposit when they return the containers. Most of the deposit laws that were enacted in the U.S. and in Canada during the 1970s, 1980s, and 1990s were intended primarily to reduce litter and to recover and recycle one-way containers. When the first set of deposit laws became effective in the 1970s, refilling advocates hoped that the mandatory deposits would reverse the decline in return rates for refillable bottles and reverse the decline of refilling overall. However, deposit laws alone cannot put more refillable beverage containers on store shelves, but they do support refilling systems by giving consumers an incentive to return their empty containers.

Role in refilling. Deposit laws usually have an auxiliary role in laws that promote or require refilling. In this role, deposit laws may be separate laws or may be provisions of refilling laws, and they primarily regulate the deposit-return systems that bottlers and retailers operate. The regulations often specify minimum deposits, minimum refunds, or other rules that govern the return of empty containers. Minimum deposit amounts can prevent the use of deposit reductions as a competitive weapon, a tactic that bottlers used in the early days of deposits [BCNC]. By specifying minimum refund amounts which are less than the deposit amounts, deposit laws give retailers the option of keeping part of the deposit as a handling fee. Such a deposit scheme is used in the province of Prince Edward Island, Canada, but most retailers there give their customers the full refund [PEIS]. Rather than specify minimums, some deposit laws specify ranges or fixed amounts for the deposits. Other regulations that deposit laws often include pertain to the labelling of containers, the visibility of return stations, the transactions between the bottler and the retailer, the submission of reports to the government, and the licensing of the deposit-return system. Deposit laws may also specify deposit amounts for secondary packaging such as bottle crates.

Half-back deposit. One type of deposit law is designed to promote the sale of beverages in refillable containers. Under this type of law, often called a “half-back” deposit law, the consumer receives a full refund of the deposit for a refillable container but only half of the refund for a one-way container. One important concern about the half-back deposit is how to allocate the half of the deposit that the consumer does not receive for returning a one-way container. If that money is allocated to retailers or to bottlers, then they will have a disincentive for selling beverages in refillable containers. Consumers will have less of an incentive for buying beverages in refillables if the price of a beverage product in a refillable bottle is greater than the price plus half of the deposit of the same product in a one-way bottle of the same size. To promote refilling through a half-back deposit law, therefore, deposits not only must be high enough to motivate consumers to return containers but also must be high enough to give refillables a price advantage. The half-back deposit can be effective in markets in which refillables effectively compete with one-way containers, but bottlers or retailers may try undermine its effectiveness by selling beverages at prices that ensure an advantage for one-way containers [CNWM, p. C-4].

Consumers can buy beer in refillable bottles in the Canadian province of New Brunswick, which applies the half-back deposit to almost all beverage containers. Although the refillable bottle dominates the beer market in New Brunswick because Canadian brewers and beer drinkers prefer refillables, the effect of the half-back deposit on the beer packaging mix there may be a worthwhile investigation. The research for this web site focused on New Brunswick’s soft-drink market, where refillables had almost disappeared when the half-back deposit became effective in 1992. Two other Atlantic Canada provinces, Nova Scotia and Newfoundland also have the half-back deposit.

return to top


Like the half-back deposit law, eco-taxes are economic instruments which most often are designed to give beverages in refillable containers a price advantage. Usually, an eco-tax imposes a tax on all beverage containers except those that qualify as refillable. The research for this web site found that eco-taxes are the most prevalent and the most effective economic instrument for promoting refilling. Some European countries have successfully used eco-taxes since the 1970s for this purpose. Finland, Norway, Denmark, Belgium, and the Canadian province of Ontario levy eco-taxes on one-way beverage containers.

Types of eco-taxes. Most eco-taxes on beverage containers fall into one of three categories according to their intended function. One type of eco-tax is a cost-covering charge to finance environmental programs, including those that involve monitoring or controlling the use of natural resources. A cost-covering charge on one-way containers could be used to help finance waste management programs [AGM, p. 79], including recycling. An eco-tax whose revenues are used in this manner is an attempt to internalize the monetary costs associated with the environmental impacts of the use of one-way beverage containers. To internalize these costs means to force industry and perhaps consumers to pay for the environmental impacts of one-way containers. Because assigning monetary values to these environmental impacts is not yet an exact science, setting the amount of an eco-tax on this basis may spark controversy [ERC, pp. 18-19]. Another type of eco-tax is an incentive tax levied solely to correct environmentally destructive behavior. Taxes on one-way containers can provide incentives for consumers to buy beverages in refillables rather than one-way containers. A third type of eco-tax is one that is intended mainly to raise revenue, although it may be associated with some environmental goal. An eco-tax program may fall into more than one of these three categories [ERC, pp. 3-4].

Formulation. Formulating an eco-tax on one-way beverage containers involves deciding who will pay the tax, how to set the tax rates, and where the revenue will go. The effectiveness of an eco-tax on one-way containers can depend on which actor in the beverage packaging chain pays it. A policy of taxing packaging manufacturers for producing one-way containers would discriminate against makers of beverage cartons or aluminum cans because they cannot make refillable versions of their products. Because any bottler of any packaged beverage can use both refillable and non-refillable containers, taxing them seems to be the fairest and most effective way to use eco-taxes to influence the packaging mix. For many of the eco-taxes that this web site discusses, the bottler, importer, or distributor pays the tax upon shipment of beverages to the retailer or other point of sale. The recommendation that the bottler pay the tax instead upon purchase of a new container from the packaging manufacturer is proposed by the Organisation for Economic Co-operation and Development [OECD, p. 14]. Taxing the bottler seems to contravene the assertion that the retailer effectively controls the packaging mix. However, eco-taxes can compel retailers to stock beverages in refillable containers when the tax amounts are high enough to diminish their profits from beverages in one-way containers [AGM, p. 9]. When bottlers pay the tax, indeed, they pass the resulting price increase to the retailer, who then passes it to the consumer. Hence, the retailer and ultimately the consumer pay the tax indirectly. Directly taxing retailers or consumers rather than bottlers has its disadvantages. First, it complicates collection by increasing the number of establishments from which the tax must be collected. Second, taxing consumers directly for the purchase of one-way containers provides only a weak incentive for bottlers and retailers to offer beverages in refillables and requires consumers’ awareness of the tax.

Setting tax rates for one-way beverage containers involves three major decisions. The first is whether to use a per-liter rate or a per-unit rate. Under a per-liter rate, the tax increases with the volume capacity of the container, and so does the price difference between a refillable container and a one-way container of the same beverage: the price advantage of a 2-liter refillable PET bottle over its one-way counterpart will be greater than that of a 500-ml refillable PET bottle over its one-way counterpart. The second decision concerns whether to use a common rate for all packaging materials or to set different rates for different materials. For different packaging materials, the relative environmental impacts or the net costs of recycling may provide a basis for setting the tax rate for each material. The most important decision involves setting the tax rates high enough to diminish retailers’ profits from selling beverages in one-way containers. These rates may need to be raised in the future if new technology or cheaper labor lowers the costs of packaging beverages in one-way containers.

Having the tax rates and some beverage sales statistics enables the government to estimate the revenue from its eco-tax on one-way containers. Where to put that revenue is another major decision in formulating an eco-tax. For most of the eco-taxes that this web site discusses, the revenue goes to the national or provincial treasury. Using the revenues from beverage container taxes to finance the annual budgets of waste management or other programs can be difficult because the revenue depends on beverage sales and especially on sales of one-way containers. Revenues will fall if the tax has the desired effect–to decrease the market share of one-way containers. Advantages and disadvantages. Eco-taxes on one-way containers not only bring revenue but also let bottlers, retailers, and consumers have choices in beverage packaging. Eco-taxes motivate rather than force consumers to buy beverages in refillable containers. The allure of revenue from an eco-tax, however, is one of its disadvantages [GUZZ]. Because of its ability to generate revenue, a government may choose an eco-tax over policy instruments that are more effective or more appropriate in its jurisdiction for promoting refillable containers. Most importantly, eco-taxes and other economic instruments are effective only in markets where consumers can buy almost all brands of beverages in both refillable and one-way containers and where refillables hold a fairly significant percentage of the packaged volume of beverages.

return to top


Having covered two economic instruments, this discussion will now turn to regulatory instruments. One well-known regulatory instrument is the quota, which requires that an entire industry or individual companies package or sell a specified percentage of the packaged volume of their beverage products in refillable containers. The Ontario soft-drink market [SAPH, pp. 246-247] and the experiences of Germany and of Portugal have shown that quotas do very little to promote refilling. Quotas imposed as regulations have at least two disadvantages. First, quotas do not provide effective incentives for the consumer, retailer, or bottler to maintain a market for beverages in refillable containers [CNWM, p. C-1]. Second, quotas require costly and time-consuming collection of packaging data, and government and industry may have disputes over the accuracy of the percentages in packaging mix and over how they are calculated.

Quotas are more effective in stopping the decline of refilling than in increasing refilling. To stop the decline of refilling, a quota could be set at or below the current volume percentage of refillable containers in the packaging mix. Requiring each beverage company to meet the quota may be more effective than requiring the beverage industry collectively to meet it.

return to top


For a given type of beverage, a quota of 100 percent is a ban on all non-refillable containers for that beverage. Bans have proven themselves to be the simplest but most robust policy instrument for mandating refillable containers. When implementing a ban, the government needs few additional staff members to administer the policy [AGM, p. 88]. Enforcing a ban is easy also because of its inherent simplicity. Detecting a violation requires simply inspecting store shelves and finding a beverage product in cans or in one-way bottles. Indeed, anyone can detect a violation of a ban. The only disadvantage of a ban is that it denies consumers the option to buy beverages in cans. Consumers may occasionally want to buy canned beverages because of their portability. To provide consumers a choice of beverage packaging while promoting refillable containers, a ban could be applied only to non-refillable glass and plastic containers. Bans can be targeted also at specific markets, say, restaurants and bars.

To provide consumers a choice between domestic and imported beverages, imports could be exempted from the ban. Such an exemption would be reasonable also because refilling may be impractical for beverages imported from overseas bottlers. Denmark’s policy allows non-refillable glass and plastic containers for imported beer, soft-drinks, and carbonated mineral waters but requires these imports to be sold under a deposit-return system. For these three types of beverages, in addition, Denmark bans all non-refillable containers for products made and sold in Denmark and bans all cans. The only other ban on non-refillable containers is that of the Canadian province of Prince Edward Island, which bans all non-refillable containers for all beer and soft drinks. Unlike Denmark, Prince Edward Island must impose its ban on all beer and soft drinks regardless of where they are produced. Allowing one-way containers for beverages from other provinces would certainly bring cans and probably bring one-way PET bottles to the island.

The ease of imposing a ban depends on how prevalent refillable containers are when the ban becomes effective. If refillables command nearly 100 percent of the market, then beverage companies would need to invest little money and effort to comply. If the market share is much less than 100 percent but nevertheless significant, then industry will need to invest some time and effort to achieve full compliance. Where refillable beverage containers still hold a significant market share, however, the cost of making the transition completely to refilling will be much less than the cost of maintaining the packaging status quo [AGM, pp. 88-89]. If refillable containers hold an insignificant share of the beverage market, then a ban should give beverage companies plenty of time to make the transition from one-way containers to refillables. return to top


Although quotas have yet to work effectively as regulations, quotas have proven effective in agreements between government and industry. The agreement between Quebec brewers, the provincial government, and Recyc-Quebec requires that no more than 37.5 percent of the total number of containers in a brewer’s sales be non-refillable. Another agreement, the Dutch Packaging Covenant II, stipulates that beverage companies and importers cannot substitute refillable with one-way beverage containers unless they can demonstrate that the environmental impact of their one-way containers is less than or equal to the impact of their refillable containers. Agreements such as these take a contractual rather than a regulatory or economic approach to managing beverage containers. This approach involves enforcing a contract rather than enforcing laws. Other advantages of an agreement include the flexibility that they give industry [CNWM, p. C-8] and the choice in beverage packaging that they ultimately give consumers. One inherent disadvantage of agreements is the lack of any provisions that force industry to enter into them. To compel industry groups to enter into an agreement, the agreement can be an alternative to a more restrictive policy instrument. In Quebec, hazardous materials regulations govern the containers of brewers and distributors who choose to not sign the agreement or to not sell beer only in refillables.

return to top


For more information about some of these sources, go to the annotated bibliography (B) or to the links.


  • [AGM] Golding, Andreas. Reuse of Primary Packaging. (Main Report) Brussels: European Commission, 1999. (B)
  • [BCNC] Woodroof, Jasper Guy, and G. Frank Phillips. Beverages: Carbonated and Noncarbonated. Westport, Connecticut: Avi Publishing Company, Inc., 1981. (B)
  • [PEIS] Morawski, Clarissa. “P.E.I.’s Beverage Container Program.” Solid Waste & Recycling 4.5 (1999): 16. (B)
  • [CNWM] Citizens’ Network on Waste Management (CNWM). A Strategy to Promote Refillables and Reuse in Ontario. Kitchener, Ontario: CNWM, 1997. (B)
  • [CW] Louis, J. C., and Harvey Yazijian. The Cola Wars. New York: Everest House, 1980. 372 pages. (B)
  • [GUZZ] Macdonald, Doug. “Beer Cans, Gas Guzzlers and Green Taxes: How Using Tax Instead of Law May Affect Environmental Policy.” Alternatives 22.3 (1996): 12. (B)
  • [ERC] ECOTEC Research and Consulting, et. al. Study on the Economic and Environmental Implications of the Use of Environmental Taxes and Charges in the European Union and its Member States. Brussels: European Commission, 2001. (B)
  • [OECD] Organisation for Economic Co-operation and Development (OECD). Beverage Containers: Reuse or Recycling. Paris: OECD, 1978. (B)
  • [SAPH] Saphire, David. Case Reopened: Reassessing Refillable Bottles. New York: INFORM, Inc., 1994. (B)