8 REVIEW OF LITERATURE

8.1 INTRODUCTION

This review covers a wide range of issues related to tobacco in nine subject areas: demand, supply, and trade of tobacco leaves, demand for cigarettes, cigarette advertising, cigarette taxes, social costs of smoking, economic significance of tobacco industry, and farm issues associated with tobacco control. To select for the most relevant literature for the review, first, key words were used to search for articles related to the nine topics from three large databases: Arcola created by the National Agricultural library of the United States Department of Agriculture, National Smoking Data Base maintained by the Office on Smoking and Health of Centres for Disease Control and Prevention in the United States, and Ecolit (economic literature) available at the library of Kansas State University. The abstract and full text of each identified study were then reviewed, and results from the reviewed study summarized. The study period covered by this review is between 1985 and 2000, but a few important studies conducted before 1985 are also included.

8.2 DEMAND FOR TOBACCO LEAVES

Most of the tobacco produced in the world is not consumed as a final consumer product, but is used to produce tobacco products. Cigarette production accounts for most of the tobacco use in the world. In the United States, for example, cigarette production utilized 90 percent of the total tobacco consumed in 1996 (USDA, 1996). Other products include smokeless tobacco and cigars.

Demand for tobacco is primarily derived from consumer demand for cigarettes. Thus, factors influencing demand for cigarettes would also affect the demand for tobacco. These factors will be discussed in the later section on demand for cigarettes.

Cigarette manufacturers use tobacco and other inputs to produce cigarettes. An important question in determining the factors affecting demand for tobacco is whether there exists a substitutable relationship between raw tobacco and other production inputs. Sumner and Alston (1987) examined this question using a generalized Leontief cost function with data from United States cigarette production. Their results showed that utilization of United States-grown tobacco was not a fixed proportion of cigarette outputs. Both imported tobacco and other inputs were substitutes for United States produced tobacco. The estimated output-constant price elasticity of demand for United States tobacco was -2.3. The cross-price elasticity of demand for imported tobacco with respect to the price of United States tobacco was 2.0. Substituting tobacco for the other production inputs according to their study occurred in two different forms: increasing the proportion of additives to cigarettes and reducing tobacco wastage. These substitution processes were accomplished through developing new production technologies. The leaf-conserving technology had increased the filling capacity of tobacco. Other technology also permitted the use of the entire leaf. Because of this substitution, the amount of tobacco used to produce 1000 cigarettes had fallen from 2.70 pounds in 1950-54 to 1.70 pounds in 1987 (Sumner and Alston, 1987).

Tobacco grown in different countries or different regions in a country varies in type and quality. It is not an homogeneous product. The substitution between different types of tobacco such as flue-cured, burley, or oriental tobaccos was investigated by Beghin and Chang (1992). They estimated a share equation of four types of tobacco derived from a translog cost function using time series data of the United States cigarette manufacturing industry. The small elasticity of substitution estimated from their study implies a limited substitutability among the four types of tobacco used in the United States cigarette production. As with the results from the study by Sumner and Alston, they also found that domestic and foreign tobaccos were substitutes, but the degree of substitution in this later study was much smaller.

Rezitis, Brown and Foster (1998) used a dynamic model to improve the elasticity of demand for United States tobacco by United States cigarette manufactures over the two previous studies. Their result showed that demand for United States tobacco was more inelastic. For example, they estimated that the total own price elasticity for United States tobacco was -1.46, compared with a value of -2.5 estimated by Sumner and Alston. The own-price output-constant elasticity for United States flue-cured tobacco was estimated to be close to -1.0 by Beghin and Change and -2.0 by Sumner and Alston, and only -0.4 by Rezitis, Brown and Foster.

8.3 SUPPLY OF TOBACCO LEAVES

Tobacco is grown in more than 100 countries, including about 80 developing countries. Given its hardiness, tobacco can be grown in a variety of climatic conditions and topographies. Tobacco grows well in sandy soil with low water-holding capacity. It is tolerant to extreme weather conditions (Jacobs et al., 2000). World tobacco production is geographically concentrated. The four top countries, China, the United States, India and Brazil, produced about two-thirds and the top twenty countries produced more than 90 percent of the world production in 1997 (Jacobs et al., 2000).

World tobacco production has increased by nearly 60 percent from 1975 to1998. This increase, however, is not evenly distributed across production countries. Nearly all growth in production has come from developing countries. Between 1975 and 1998, production in the developed counties fell by 31 percent, while production in developing countries rose by 128 percent (Jacobs et al., 2000). The increase in tobacco production in developing countries can be attributed to several factors, including increased demand for cigarettes, a higher return for tobacco, and increased farm efficiency and quality of tobacco produced.

8.3.1 Factors affecting supply of tobacco leaves

In a competitive market, the supply of tobacco is determined by the price of tobacco and production costs of tobacco relative to its competing crops. Real prices of tobacco have fallen in most countries, but much less rapidly than prices of other agricultural crops and basic commodities. Tobacco is one of the most profitable crops in many countries. For example, in Zimbabwe, tobacco is roughly seven times more profitable than the next-best crop (Marvanyika, 1997). In the United States, the net return of about $2000 per acre from growing tobacco far exceeds the net return for most other crops. Tobacco is a cash crop for many farmers in developing countries. Thus, tobacco has been a crop that farmers prefer to grow in nearly all producing countries.

The market for tobacco does not resemble a free market (Coady, Pompelli and Grise, 1991). Governmental intervention influences the production and trade of tobacco in most of the tobacco producing countries. The types of domestic support and trade programs by countries are summarized in Table 8.1. More detail on polices of the intervention for the eight major tobacco trade nations and the European Community for 1982-87 is presented in Table 8.2.

Table 8.1: Types of domestic and trade programmes in producing and consuming countries

Source: Grise, V. N. 1990.

Table 8.2: Policies affecting tobacco production and trade, 1982-87

Source: Coady, Pompelli and Grise, 1991.

The level of governmental intervention varies considerably between nations. In some, governmental intervention is pervasive and weighs heavily in most production and trade decisions. High levels of intervention often involve a domestic subsidy, trade barriers, legal restriction, or agricultural production and export taxes. In other countries, governmental intervention is minimal and has little or no influence on tobacco production and trade. Measuring the overall intervention schemes, however, is difficult because of the diversity and complexity of these policies (Coady, Pompelli, and Grise, 1991).

Jacobs et al. (2000) summarized the differences in policies used in developing and developed countries. They found that developing countries tended to tax tobacco production since tobacco was an important source of foreign exchange earning and tax revenues. A few Sub-Saharan African countries are particularly dependent on tobacco and other primary commodities for foreign exchanges. Many low income countries rely on revenue from export industries excise taxes since income taxes are difficult to administer in less developed countries (Beghin, Foster and Kherallah, 1996, Pena and Norton, 1993). Argentina, Brazil, Turkey, and Zimbabwe all have export taxes on tobacco products. In some countries, centralized marketing boards or tobacco monopolies purchase tobacco at lower prices, implicitly taxing tobacco growers (Beghin, Foster, and Kherallah, 1996). Governments in some developing countries also subsidize tobacco farmers with credit, electricity, etc, which offsets the taxation. The extent of taxation, however, has been reduced in recent years due to the reduction of centralized monopoly purchasing as a result of GATT/WTO requirements, economic transition in former Easter-bloc countries and IMF mandates in several east Asian countries (Jacobs et al., 2000). Despite the taxation, tobacco is still more profitable than alternatives in most developing countries.

Governments in developed countries tend to subsidize tobacco farmers. Supply management is a widely used policy tool for tobacco production. The operation of the supply management differs between nations, however. In United States and Australia, basic components of the supply management are price support, production restriction through production quota, and import restriction through tariff and non-tariff measures (Zhang and Husten, 1998). In the European Union, the principal features for leaf tobacco are guaranteed prices to growers, buyers’ premium, export subsidies, and safeguards to protect the European market. Production control is also implemented when needed (Grise, 1990). Tobacco policy in developed countries has resulted in a higher and stable tobacco prices, and attracts more resources to tobacco production, except to the extent that production is restricted by quotas. The production quota has effectively reduced tobacco production in those countries. Eliminating production quotas would lead to an increase in supply of tobacco in those countries.

Estimating the elasticities of supply for tobacco is complicated by governmental interventions in tobacco production and trade. The marginal cost elasticity of tobacco production under the production quota system in the United States was estimated to be 0.25 (Goodwin and Sumner, 1992). The supply elasticity of tobacco without production quotas was estimated to be 7.0 (Fulginiti and Perrin, 1993). The supply elasticity for tobacco should be fairly large, particularly in the long run, since tobacco uses a small proportion of the arable land in the world as well as in any country and the net return from growing tobacco is several times that from growing the next best alternative crops in many countries.

8.4 TOBACCO LEAF TRADE

Tobacco is traded in the world market for three main reasons: (1) some countries do not produce tobacco, produce too little, or do not produce a particular kind of tobacco; (2) some countries do not produce high enough quality tobacco or enough of it to meet domestic demand; and (3) prices vary for given kinds and qualities between countries with different product demand mixes (Grise, 1990).

About one fifth of the global tobacco production was traded in the world market in 1997. The proportion of the export to production for individual countries varies from none to three-fourths (Jacobs et al., 2000). An individual country can be both exporter and importer of leaf tobacco since tobaccos produced in different countries are not homogenous. For example, Spain exported about half of its tobacco production and also imported 1.3 times as much as it produced in 1997.

8.4.1 Trade barriers

World trade of tobacco has been impeded by both tariff and non-tariff barriers. Many countries, including developed countries such as United States and the European Community and developing countries such as China, India, Zimbabwe, levy high tariffs on imported raw tobacco and tobacco products (Table 8.1 and Table 8.2). The non-tariff barriers that are used to limit imports include license requirements, restricted product lists, exchange control, mixing regulations (which govern the percentage domestic grown tobacco required in manufactured products), and quota restrictions. The pattern of trade is also distorted by export promotions through bilateral trading agreements, trade on concession terms, export subsidies and other government interventions in domestic production (Grise, 1990).

8.4.2 Impact of trade liberalization

Trade liberalization, defined as eliminating tariff and non-tariff barriers and domestic programs including supply control, price support programs and the other programs that alter the production decisions, would affect world tobacco production, consumption and trade. Grise (1990) analyzed the impact of trade liberalization on tobacco prices, production, consumption and trade.

Cross-country price comparisons of tobacco of the same type indicate that prices have been altered significantly by trade restrictions and domestic tobacco policy in major producing and consuming counties. Thus, trade liberalization would lead to a change in the price of tobacco in the world market. The United States has provided a price umbrella in the world market due to its relatively large share of that market. The price of United States-grown tobacco is significantly higher than from other major exporters in the world market. Removal of the tobacco program would lower the price of tobacco in the United States by 20-30 percent (Zhang, Husten and Giovino, 2000). Thus, the price of tobacco in both United States and world markets would be lower if trade were to be liberalized. But how much lower the world price would be is difficult to estimate. While an analysis of cross-country price differences could provide some clues, it does not take into account the fact that tobacco grown in different countries is not a homogenous product and that the transportation cost of shipping tobacco from different countries to the one destination would also differ. The wide variations in prices, however, would continue, reflecting different demands for various kinds and qualities of tobacco.

How much higher or lower would world consumption be in a liberalized policy environment? The higher prices which prevail under existing policies, combined with the working of support programs and import restrictions, limit the kinds and quality of tobacco available and discourages consumption. Thus, trade liberalisation would increase tobacco consumption. However, this effect would be small due to two main reasons. First, reducing the price of tobacco would have minimal impact on the price of cigarettes since tobacco accounts for only a small percentage of the retail value of cigarettes. Considering the monopolistic or oligopolistic structure of the tobacco industry in many countries, reducing the price of tobacco leaf may not have any major impact on the retail price of cigarettes and, given that demand for tobacco products is inelastic, any impact on cigarette consumption may be very small indeed.

Trade liberalization would be expected to have a substantial impact on trade patterns. It would reduce tobacco prices and increase consumption. Internal producer prices in importing and exporting countries with price support would fall when those policies are removed. Production in the importing countries would drop while their demand would rise in response to lower prices. Trade flows would expand, with the added production in the low-production-cost countries substituting for the lost production in the high-production-cost countries. The United States, Brazil, and Zimbabwe would be likely to increase their exports substantially while the EC countries, Australia, and Canada would import more tobacco. In addition, countries that have already been liberalizing their trade programs in East Asia, for example, would see a further increase in imports. The volume of trade would be expected to increase. With the current differences in production costs across countries, total trade liberalization would lead to an increase in trade surpassing 50 percent of the world production (Grise, 1990). Consumers would gain by traditional welfare measures. Increasing tobacco consumption, however, would also increase smoking related illnesses and deaths. Public health groups have expressed concerns about the possible impact of trade liberalization on global health, especially the impact in developing countries (Yach and Mackay, 1996), thus presenting a dilemma to policy makers. In the United States, for example, the question has arisen as to whether there is a policy conflict between the trade goal of promoting cigarette exports and the health goal of reducing domestic smoking and participating in the international anti-smoking movement (GAO, 1992).

No studies are found to quantitatively estimate the impact of trade liberalization on world production, consumption and trade. Available studies on the impact of trade liberalization have focused on a specific policy such as lowering import tariffs or imposing a domestic content requirement in an individual country such as United States (Beghin and Chang, 1992; Beghin and Hu, 1995). Multi-country and multi-commodity models which are used to estimate the impact of trade liberalization on trade flows in the United States have included tobacco, but these models implicitly assume that tobacco produced in different countries is a homogeneous product (Roningen and Dixit, 1989). These models ignore the supply management policy in United States tobacco production, two-way trade flows of tobacco, and the forward linkage of tobacco to cigarette production (Beghin and Chang, 1992). Quantitative studies which evaluate the impact of the trade liberalization on world tobacco production, consumption, trade are needed.

The ruling of the GATT council in the United States and Thailand dispute on cigarette trade has presented another example of health and trade policy conflict in the international context. To protect its monopoly position, the Thai government virtually banned cigarette imports. In addition, the government monopoly has completely stopped its own cigarette advertising and promotion since 1989. In response to petition from the United States Export Association, the United States Trade Representative submitted a complaint to GATT regarding the import ban of United States cigarettes. The GATT Council agreed that a ban on imports was a violation of international trade treaty. However, the council upheld the Tail government’s right to restrict over-all supply of cigarettes using a variety of policies including high cigarette taxes and bans on advertising. The GATT Council noted that the member nations could use various policies to protect health as long as they are applied evenly to domestic and foreign products. Furthermore, the council concluded that policies such as banning advertising, which made it more difficult for new foreign firms to compete with existing domestic firms, were justified under the GATT since advertising could increase the demand or cigarettes particularly among youth (Chaloupka and Laixuthai, 1996).

8.5 DEMAND FOR CIGARETTES

Modelling the demand for cigarettes has long been an interest of economists. Many economists once viewed cigarette smoking and other addictive behaviours as irrational and therefore not suitable for conventional economic analysis (Winston, 1980; Schelling, 1984). They believe that demand for cigarettes does not follow the basic law of economics including the downward-sloping demand curve. However, this view has changed as a large body of economic research demonstrates that demand for cigarettes clearly responds to changes in prices and other factors. Demand for cigarettes has been studied extensively since 1985.

8.5.1 Economic models and data used for estimating demand for cigarettes

Studies on demand for cigarettes have applied several types of economic models to different types of data with various estimation techniques. In general, two types of economic models are used: the conventional demand model and the addictive demand model. These models have been applied to two types of data: aggregate level data including time-series data for a single geographical unit and pooled cross-sectional time-series data, and individual level of survey data.

Conventional demand models which use aggregate data normally specify the demand equation in a such way that the quantity of cigarettes demanded is a function of cigarette prices, income, tobacco control policies and a variety of socioeconomic and demographic factors (Bishop and Yoo, 1985; Baltagi and Levit, 1986; Chaloupka and Saffer, 1992; Flewelling et al., 1992; Seldon and Doroodian, 1989; Seldon and Boyd, 1991; Showalter, 1991; Simonich, 1991; Tegene, 1991; Sung et al., 1994; Hu et al., 1994, 1995; Barnett et al., 1995; Tremblay and Tremblay, 1995; Yurekli and Zhang, 2000). But there are two exceptions (Baltagi and Goel, 1987; Peterson et al., 1992), in which a quasi-experiment approach was used to compare changes in cigarette consumption in states in the United States that have raised cigarette taxes to consumption in states where taxes have not changed. A small but growing number of studies have used data on individuals taken from large-scale surveys (Lewit et al., 1981; Lewit and Coate, 1982; Grossman et al., 1983; Jones, 1989; Wasserman et al., 1991; Blaylock and Blisard, 1992; Yen and Jones, 1996; Chaloupka and Grossman, 1996; Chaloupka and Wechsler, 1997; Chaloupka and Pacula, 1998a, 1998b; Farrelly et al., 1998). These studies differ from those using aggregated data, in that they normally estimate a two-part model, by estimating firstly the probability that an individual will smoke and, secondly, the level of consumption among smokers. The conventional demand model does not account for the addictive nature of cigarette smoking.

There are several versions of the addictive model that have been used for studying the demand for cigarettes: the imperfectly rational addiction model, myopic addiction model and rational addiction model (Chaloupka and Warner, 1999). The imperfectly rational addictive model assumes that consumers have stable but inconsistent short-run and long-run preferences. An individual chooses a future consumption path that maximizes current utility, but later in life changes this plan (Schelling, 1978). This model, however, has not been applied empirically to cigarette smoking. The myopic addiction model is similar in spirit to the habit formation or irreversible demand models. The basis of the empirical application of the model is that the current cigarette consumption depends on, besides the factors included in the conventional demand model, a “stock of habits” representing the depreciated sum of all past consumption (Mullahy, 1985; Baltagi and Levin, 1986). The rational addictive model is the most recent model used for modelling demand for cigarettes (Becker and Murphy, 1988; Becker et al., 1991; Pekurinen, 1991; Chaloupka, 1990, 1991, 1992; Keeler et al., 1993, Sung et al., 1994; Conniffe, 1995; Duffy, 1996; Cameron, 1997; Bardsley and Olekalns, 1998). The rationality here simply implies that individuals incorporate the interdependence between past, current, and future consumption into their utility maximization processes. This is in contrast to the assumption, implicit in myopic models of addictive behaviours, that future implications are ignored when making the current decision. In other words, myopic behaviours imply an infinite discount of the future, while rational behaviours imply that future implications are considered. Empirically, the demand equation is specified as the quantity of cigarettes demanded in the current period being a function of both past and future consumption as well as those other factors included in the conventional demand model.

Becker and Murphy (1988) and Becker et al. (1991) developed several hypotheses from the basic rational addiction model. First, the quantities of the addictive good consumed in different time periods are complementary. As a result, current consumption of an addictive good is inversely related to not only the current prices of the good, but also to the all past and future prices. Consequently, the long-term effect of a permanent change in prices will exceed the short-term effect. Moreover, the ratio of the long-term to short-term price effect increases as the degree of addition rises. In addition, the model predicts that the impact of an anticipated price change will be greater than that of a comparable un-anticipated price change, while a permanent price change will have a larger impact on demand than a temporary price change. Finally, price responsiveness varies with time preference: addicts with higher discount rates will be more responsive to changes in money prices that those with lower discount rates.

Data from different countries including the United States (Becker et al., 1991; Chaloupka, 1990, 1991, 1992, Keeler et al., 1993, and Sung et al., 1994), Finland (Pekurinen, 1991), Australia (Bardsley and Olekalns, 1998), the U.K. (Duffy, 1996), Greece (Cameron, 1997) and Ireland (Conniffe, 1995) have been used to test these hypotheses derived from the rational addictive model. Results from those studies are mixed. Studies using data from the United States, Finland, and Australia, in general, support the hypothesis derived from the rational addiction model, but those studies using data from the U.K., Greece, and Ireland found little support for the rational addiction model. These latter studies, however, are generally limited by the small number of observations available for the analysis and by the use of several highly corrected regressors.

Specific variables included in the demand model of each study vary, depending on the economic model used and the availability and type of the data. Important factors that have been evaluated include costs of cigarette smoking, consumers’ income, cigarette advertising and other promotion activities, and health information. The cost of cigarette smoking should be defined broadly, including not only the purchase price of cigarettes, but the time and other costs associated with smoking. Restrictions on smoking in public places and private work sites, for example, impose additional costs on smokers by forcing them outdoors to smoke, by increasing the time and discomfort associated with smoking, or by imposing fines for smoking in restricted areas. Similarly, limits on access to tobacco by youth may increase the time and potential legal costs associated with smoking. Findings on how each factor affects the demand for cigarettes from those studies are summarized below.

8.5.2 The effect of cigarette prices on demand for cigarettes

Cigarette consumption is found to be negatively related to price. The estimated price elasticity from those studies using aggregated data varies from -0.14 to -1.23, but most fall in the narrower range from -0.3 to -0.5, including the result from the two quasi-experimental studies (Baltagi and Goel, 1987; Peterson et al., 1992). The estimated price elasticities from the studies using individual-level data, in general, are comparable to those estimates from the studies using the aggregate data.

Nearly all of the studies of the price-demand relationship focus on the developed countries. Warner (1990) argued that price responsiveness in less developed countries is likely to be greater than in more developed countries, given the relatively low incomes and relatively lower level of cigarette consumption by smokers in poor countries. A few studies have evaluated the price-consumption relationship of cigarettes in developing countries. Findings from studies using data from Papua New Guinea (Chapman and Richardson, 1990), China (Mao, 1996; Xu, Hu and Keeler, 1998), South Africa (van der Merwe, 1998), Zimbabwe (Maranvanyika, 1998), and Taiwan Province of China (Hsieh and Hu, 1997) support this argument.

The question of whether youth are more or less responsive to prices than are adults has been examined in a number of studies using individual-level data (Lewit, et al., 1981; Lewit and Coate, 1982; Grossman et al., 1983; Wasserman et al., 1991; Chaloupka and Grossman, 1996; Farrelly, et al., 1998, and Tauras and Chaloupka, 1998). Findings from those studies are mixed. The earlier studies on this issue (Lewit, et al., 1981; Lewit and Coate, 1982; and Grossman et al., 1983) found that youth are more sensitive to prices than are adults. This result, however, was challenged by the study done by Wasserman et al. (1991), which found that the price responsiveness of youth was not significantly different from that of adults. Recent studies of youth and young adult smoking (Chaloupka and Grossman, 1996; Farrelly et al., 1998; Tauras and Chaloupka, 1998) generally supported the earlier results that the price sensitivity of cigarette demand was inversely related to age. Those recent studies estimated the price elasticity of demand for cigarettes by youth was between -1.1 and -1.3, very similar to -1.44 estimated Lewit et al. in1981.

The price responsiveness of sub-population groups by income levels has been investigated by a number of researchers (Townsend, 1987; Chaloupka, 1991; Townsend et al., 1994; Farrelly et al., 1998). Results from those studies indicate that cigarette demand is less price elastic for more educated or higher income individuals. For example, Farrelly et al. estimated individuals with family income below the sample median were 70 percent more responsive to prices than those with higher family income.

8.5.3 Income effect

Findings on how demand for cigarettes changes as consumers’ income increases are inconsistent. The estimated coefficient of the income variable in most studies of demand for cigarettes is significant and positive, implying that cigarettes are “normal” goods and that increasing income would have a positive effect on demand for cigarettes. However, a number of studies (e.g., Wasserman et al., 1991, Keeler et al., 1993, Yurekli and Zhang, 2000), particularly those using cross-sectional survey data also found that income has either an insignificant effect or negative effect on demand for cigarettes. A meta-analysis by Andrews and Franke (1991) who used results from 48 studies found that the weighted mean income elasticity is 0.36, which is significantly greater than zero. They also found that the income elasticity for cigarettes fell over time.

8.5.4 Smoking restrictions in public places and private work sites

As information on the health consequences of exposure to environmental tobacco smoking (ETS) has become more widespread, governments at all levels in many countries have adopted policies to limit smoking in public places and private work sites. A World Health Organization (1997) survey of tobacco control policies in 134 countries indicated that the vast majority of countries had some form of restrictions on smoking in public places. Although the restrictions are primarily intended to reduce non-smokers’ exposure to ETS, they can also affect the smokers since the restrictions reduce the smokers’ opportunities to smoke or otherwise raise the “cost” of smoking. Smoking restrictions may also alter the perceived norms related to smoking by changing attitudes concerning the social acceptability of smoking.

The impact which smoking restrictions have on cigarette demand has been evaluated in a number of studies (Wasserman et al., 1991; Chaloupka, 1992; Chaloupka and Saffer, 1992; Keeler et al., 1993; Chaloupka and Grossman, 1996; Evans et al., 1996; Chaloupka and Wechsler, 1997; Chaloupka and Pacula, 1998; Bardsley and Olekalns, 1998; Yurekli and Zhang, 2000). In general, smoking restrictions have been found to reduce both smoking prevalence and average daily cigarette consumption among smokers. For example, Yurekli and Zhang (2000) estimated that restrictions on smoking reduced cigarette consumption per capita by 4.5 percent in the United States in 1995.

8.5.5 Summary and discussion

A large number of studies have been conducted to estimate demand for cigarettes in many countries. The estimated price elasticities of demand for cigarettes from those studies vary. But most of the estimates tend to be less than one, i.e., the demand for cigarettes is price-inelastic. A meta-analysis of the determinants of cigarette consumption, which analysed results from 48 studies, yielded a weighted mean price elasticity of demand for cigarettes of -0.4 (Andrews and Franke, 1991). The short-term price elasticity recommended by an expert panel for policy analysis is -0.4 (National Cancer Institute, 1993). The long-term elasticity is about 1.5 times the short-term price elasticity. Youth and the poor tend to be more responsive to changes in cigarette prices. Income, in general, is positively related to demand for cigarettes. The estimated income elasticity is 0.36, but decreases with time. Restrictions on smoking in public places and private work sites have significantly reduced demand for cigarettes.

Studies on demand for cigarettes have applied different economic models to two different types of data, aggregated and individual level. Analysing each of two data-types has some advantages and disadvantages. The aggregate data are either time-series data or pooled cross-sectional and time-series data. High correlation among many of the key independent variables and prices can be a problem with time-series data. Consequently, estimates of the impact which prices and other factors have on demand can be sensitive to the inclusion and exclusion of the other variables. The problem with using the pooled data is the measurement of cigarette consumption. Using these data, smoking is normally measured by annual state-level tax-paid cigarette sales. Both cross-border shopping between the neighbour states and the long-distance smuggling from low-tax to high-tax states can occur due to differences in taxes on cigarettes. Failure to account for this will produce upward-biased estimates of the impact of price on cigarette demand. Finally, with aggregate data the demand and supply of cigarettes need to be modelled simultaneously since cigarette price, sale and consumption are simultaneously determined. In contrast, the use of individual-level data can ease some of the problems associated with aggregate data such as simultaneous biases resulting from the price and consumption, and multicollinearity between cigarette prices and other factors affecting the demand. In addition, using individual-level data can allow researchers to study the price responsiveness of different sub-population groups such as those based on income, education, and age. The problem with individual-level data is the accuracy with which consumption of cigarettes is measured. Self-reported consumption is typically under reported.

The use of a rational addiction model for modelling the demand for cigarettes has been controversial. Critics of the model argue that nobody would sit down at an initial period, survey future income, production technology, investment/addictive function, and consumption preference for a lifetime, maximize the discounted value of his expected utility and decide to become an alcoholic. Empirical work for testing the rational addictive behaviours has also yielded mixed results.

In spite of the large number of studies on demand for cigarettes, only a few studies have been conducted for developing countries, despite the increasing cigarette consumption.

8.6 CIGARETTE ADVERTISING AND PROMOTION

Cigarettes are one of the most heavily advertised and promoted products in the world. Cigarette advertising and promotion, however, have been controversial. Many countries restrict cigarette advertising and promotion, but levels of the restriction are very different. While some countries have few restrictions, others ban advertising and promotion completely.

8.6.1 Effect of advertising on cigarette consumption

There are two different views about the impact of cigarette advertising on cigarette consumption. The tobacco industry argues that tobacco is a mature industry, and thus advertising affects only the market share of advertised brands and has no impact on aggregate demand for cigarettes. In comparison, the public health community argues that advertising has a positive effect on demand for cigarettes. Advertising is particularly effective in recruiting young smokers.

The effect of cigarette advertising on cigarette consumption has been examined in three different ways: (1) examining the impact directly, using either annual or quarterly national aggregate expenditure over time or using cross sectional data; (2) investigating the impact of an advertising ban on demand for cigarettes; and (3) studying the effect of counter advertising on smoking (Saffer and Chaloupka, 1999). Studies that examined the impact of advertising on cigarette consumption and results from those studies are summarized in Table 8.3. Studies that used the aggregated data generally found at most a small effect of advertising on cigarette demand. Studies using cross sectional data concluded that advertising had a significant positive effect on consumption, which increased both the market share of the advertised brand and the market size of cigarettes in general. Studies on the effect of advertising bans yielded an inconclusive result and those on counter advertising found that counter advertising reduced cigarette consumption. Saffer and Chaloupka (1999) evaluated the impact of advertising bans using data between 1970 and 1992 for 22 OECD countries and concluded that a comprehensive set of tobacco advertising bans can reduce cigarette consumption and a limited set of tobacco advertising will have little or no effect. They estimated that cigarette consumption would fall by 6.3 percent if all 22 OECD countries had comprehensive bans.

Table 8.3: Studies evaluating the impact of advertising on cigarette consumption

Source: Saffer and Chaloupka (1999).

8.7 CIGARETTE TAXES

Tobacco and tobacco products have long been taxed, primarily because the relatively inelastic demand for these products make them an easy source of revenue. In recent decades, however, taxation has been applied to cigarettes and other tobacco products in order to promote health as well as raising revenue. Cigarettes and other tobacco products are taxed in a variety of ways. The most commonly used methods of taxation include specific taxes, value added and other ad valorem taxes, and import duties (Chaloupka et al., 2001). Specific taxes (or excise taxes) add a fixed amount to the price of cigarettes while ad valorem taxes are a percentage of the base prices. Cigarette taxes in some countries such as the United States have been imposed by various levels of government, national, state and local level. Governments in nearly every country impose taxes on cigarettes and other tobacco products. The levels of taxes imposed on cigarettes vary from county to county. In the high income countries, tax amounts to two-thirds or more of the retail price of a pack of cigarettes. In the lower-income countries, taxes generally amount to not more than half of the retail prices of cigarettes (Chaloupka, et al., 2001).

8.7.1 Impact of cigarette taxes on demand for cigarettes

A number of studies have examined the effect of cigarette taxes on retail prices (Sumner, 1981; Sumner and Wohlgenant, 1985; Sung et al., 1994; Barnett et al., 1995; and Keeler et al, 1996). When those studies conclude that increases in cigarette and other tobacco taxes would certainly result in a higher price for these products, differences exist in the estimated magnitude of the increase in retail prices for a give level of tax increases. Early studies (Sumner, 1981; Sumner and Wohlgenant, 1985) concluded that pricing behaviour of the cigarette industry was similar to that of firms in a competitive industry in spite of its oligopolistic structure, thus cigarette taxes were fully passed to the price of cigarettes at the retail level. Recent studies (Sung et al., 1994; Barnett et al., 1995; and Keeler et al, 1996), which accounted for the dynamic nature of an oligopolistic industry, or modelling the demand and supply for cigarettes simultaneously, however, found that cigarette prices were increased by more than amount of the tax increase. For example, Keeler et al. (1996) estimated that a one-cent increase in a state’s cigarette tax would raise retail prices in that state by 1.1 cents.

The levels of taxes vary across nations or states or provinces in a country. Increasing the level of taxes in a higher-tax nation or a higher-tax state in a country would lead to a larger price differential across countries or states and a higher incentive for cross-border shopping and cigarette smuggling. Thus, the impact which increasing cigarette taxes might have on consumption in an individual country or state depend partly on the changes in cross-border shopping and cigarette smuggling which result from the tax increase (see below section 8.7.3).

8.7.2 Cigarette tax increases and tax revenues

Results from studies on demand for cigarettes have shown that the price elasticity of demand for cigarettes is less than one (see section 8.5above on demand for cigarettes). Thus, cigarette tax increases would lead to an increase in total tax revenues. The issue, however, is complicated by an unintended effect, cigarette smuggling, which results from a cigarette tax increase. The increase in smuggling which might be stimulated by an increase in taxes can significantly dampen the increase in revenue which would otherwise be expected. The tobacco industry argues that increases in cigarette smuggling and other tax evasions from cigarette tax increases would actually lead to a reduction in tax revenues (British American Tobacco, 1994).

Studies on the impact of cigarette tax increases in the United States concluded that the tax evasion which is due to price differentials across states could be large in absolute dollar amounts but small as a share of the total tax revenue (ACIR, 1985 and Yurekli and Zhang, 2000). For example, Yurekli and Zhang (2000) estimated that the loss in tax revenues from tax evasion from cigarettes accounted for less than 6 percent of the total cigarette tax revenue between 1985 and 1995. There is no empirical evidence that a cigarette tax increase in a state has resulted in a decrease in the state’s total cigarette tax revenue in the United States. However, losses in tax revenues from a tax increase based on the Canadian experience were much bigger. Galbraith and Kaiserman (1997) estimated that each 1 percent increase in cigarette taxes in Canada would lead to a fall of 1 percent in taxed sales. Concern about loss in cigarette tax revenues contributed to the decision of cigarette-tax rollback in 1994 in Canada. One study in the United Kingdom estimated that the revenue elasticity for cigarette taxes during 1971 and 1993 was between 0.6 and 0.9 (Townsend, 1996).

8.7.3 Cigarette tax increases and cigarette smuggling

Differences in cigarette prices among countries and among different taxing jurisdictions as a result of differences in cigarette taxes create an incentive for both casual and organized cigarette smuggling and other forms of tax evasion. This smuggling problem is exacerbated by a number of factors, including the relatively easy transportation of tobacco products, the high potential profits from this illegal activity, the presence of an informal distribution network in many countries, the availability of tax free and duty-free cigarettes, and nonexistent or relatively weak policies concerning cigarette smuggling and lack of enforcement (ACIR, 1985; Joosens and Raw, 1995, 1998; Joosens and van der Merwe, 1997).

There have been relatively few econometric analyses of the impact of price differences on organized and casual cigarette smuggling. Nearly all of these studies published are based on annual state-level cigarette sale data from the United States (ACIR, 1985; Baltagi and Levin, 1986; Chaloupka and Saffer, 1992; Becker et al., 1994; Yurekli and Zhang, 2000). Those studies have concluded that the casual and organized smuggling of cigarettes could account for a significant share of sales in those states, but at an aggregate level, smuggling accounted for less than 5 percent of cigarette consumption. A recent study estimated that the magnitude of worldwide cigarette smuggling accounts for 6 to 8 percent of global cigarette consumption (Merriman, Yurekli, and Chaloupka, 2001).

8.8 SOCIAL COSTS OF TOBACCO CONSUMPTION

8.8.1 Economic costs of smoking

The economic costs of smoking have been estimated in many countries, but mainly in the industrial countries. The high prevalence of smoking in developed countries could lead to higher economic costs in the future since most of the smoking-related illness would occur many years later after smoking.

Two general approaches have been used in estimating the economic costs of smoking. The prevalence approach values the present costs associated with existing cases of smoking-related diseases. In contrast, the incidence approach values all of the future costs associated with new cases of smoking-attributable diseases during the reference year. The former provides an estimate of the current economic burden of smoking, while the latter is more useful for evaluating interventions that might interrupt the development of smoking related illnesses. A special case of the incidence approach is to estimate the net social cost a smoker imposes on non-smokers in his life-time. This estimate is used for estimating the level of negative externality of smoking, thus finding the optimal level of excise taxes on cigarettes.

The cost-of-illness approach includes three components: (1) direct medical costs associated with smoking related illness; (2) indirect morbidity costs associated with lost earnings from work attributable to smoking; and (3) indirect mortality costs related to the loss of future earnings due to premature smoking-related deaths. Different methods have been used to estimate the various cost components, with attributable-risk methodology commonly used in estimating smoking-related disease incidence and medical costs, and the human capital approach employed in placing a value on lost years of life. The willingness-to-pay approach has also been used to value the smoking-related premature deaths. The economic costs of smoking have been estimated for a number of countries including the United States (Bartlet et al., 1994; United States Department of Health and Human services, 1989), Canada (Collishaw and Myers, 1984), and China (Jin et al., 1995). The global economic cost associated with smoking could be very large. For example, the annual economic cost estimated for the United States is approximately $100 billion dollars (Chaoluapka and Warner, 1999).

The net costs imposed on non-smokers by smokers have been studied by a number of researchers in the United States and Europe (Leu and Schaub, 1983; Manning et al. 1989, 1991; Hodgson, 1992; Viscusi, 1995; Barendregt et al., 1997; Warner et al., 1998). Diverse conclusions have been reached from those studies. Some evidence supports the conclusion that a smoker normally spends more than a non-smoker on medical costs, and thus imposes a net cost on non-smokers. When including the costs associated with other consequences of smoking such as absence from work resulting from smoking related illnesses, reduced income tax receipts and additional social security costs as a result of premature sickness and death, smokers impose positive costs on non-smokers but which are smaller than the level of cigarette taxes. One study, however, has estimated the cost that a smoker imposes on non-smokers can be as large as $4.80 for a pack of cigarettes (Hay, 1991).

The differences between estimations of medical costs associated with smoking can be attributable largely to the selection of diseases to be included. Diseases that are found to be associated with smoking have grown as research on smoking continues. Previous estimates of medical costs associated with smoking perhaps have underestimated the “true” cost since none of the studies has included the entire list of diseases that could now be attributable to smoking. In addition, nearly all of the studies excluded the medical costs associated with treating diseases that are possibly related to ETS (second hand smoking). Furthermore, the cost of smoking in earlier studies did not include the time and transportation costs associated with smokers obtaining medical care for smoking related illnesses and the costs associated with fire caused by smoking. Intangible costs associated with smoking-related diseases such as suffering and pain are also not included in estimates.

Using productivity loss to value the indirect costs associated with smoking is also challenged by many researchers. Applying this method implies that the value of a person is reflected by his or her earnings, with the life value for that person equal to the discounted stream of the future earning. Many researchers believe that the lost earning is a poor measure for the value people place on their health or on their lives. The willingness-to-pay approach should be a theoretically correct approach to value both morbidity and mortality.

Estimating the net cost a smoker imposes on non-smokers has been a difficult task both conceptually and empirically. The identification of costs to be considered as external or internal, and of what should be considered as transfers, have been controversial. In addition, empirical evidence on the number of diseases and the extent to which each disease can be attributable to smoking varies with different epidemiological studies. Using different assumptions can lead to different conclusions. The estimation of the net cost is also dependent on the system of health care and social security in a country. For example, in the United States, some studies suggest that, by dying early, smokers subsidize non-smokers’ social security payment. In the United Kingdom, by contrast, research indicates that sickness benefits paid to smokers and pensions paid to their dependents compensated for the lower direct pension benefits paid to smokers as a consequence of their earlier average age of death. In developing countries in which old-age expenses are largely a private matter, the social benefits of dying early would not exist and hence would not offset any negative externalities of smoking.

8.9 ECONOMIC SIGNIFICANCE OF TOBACCO

Successful efforts to reduce smoking would lead to a smaller tobacco industry. It has been argued that the economic vitality of many countries depends on a strong tobacco industry and that the adoption of anti-smoking policies would inflict economic damage. The impact which a reduction in smoking would have on employment in particular has been studied most extensively.

8.9.1 Employment and tobacco

The economic significance of the tobacco industry in a country or a region of a country has been evaluated by two different types of studies. Some studies have shown that tobacco industry has generated millions of jobs, implying that a reduction in tobacco consumption would lead to many job losses. Other studies, which assume a dynamic economy with adjustment taking place between industries, have shown that there would be little negative impact or even a gain in jobs if tobacco were to be eliminated from the economy.

Those studies which have estimated the gross employment associated with the tobacco industry using either a macroeconomic model or accounting techniques have estimated number of jobs both directly and indirectly related to the tobacco industry. Direct employment included those jobs from the core sectors of the tobacco industry: tobacco farming, cigarette manufacturing, and distribution and retailing of cigarettes. The indirect employment included jobs from two sources: jobs derived from the supplier sectors such as fertilizer and paper industry and jobs induced by the spending of the core and supplier sector employees and their families. The expenditure induced employment is a result of the ripple effect of the tobacco industry. The estimated number of jobs that are indirectly associated with the tobacco industry was normally estimated as many times the number of direct employment. The number of jobs that are dependent on the tobacco industry is generally overstated in these studies for two main reasons. First, the estimated jobs that are indirectly associated with the tobacco industry are not the actual jobs dependent on tobacco spending. Secondly, those studies assume that the reduction in tobacco spending from a lower tobacco use will disappear in the economy (or zero opportunity cost for the resources used) and will not generate new jobs from alternative spending activities. The results from several such studies are summarized in Table 8.4.

Table 8.4: Estimates of tobacco-related employment

In contrast, other studies, typically from academic institutions, have estimated the net contribution of the tobacco industry to employment, i.e., the change in employment after considering the redistribution of the same resources to alternative uses. Regional economic models and input-output models used in those studies in most cases found little or no negative effects from tobacco control policy on economic output and employment. Reducing smoking would lead to job losses in those sectors immediately associated with cigarette production such as tobacco manufacturing and farming, or in tobacco dependent regions. But these losses are generally outweighed by increases in employment in other industries or in non-tobacco dependent regions. The estimated net change in employment also depends on assumptions used as to how the resource or spending saved from the tobacco related activity would be reallocated to other sectors of the economy and how governments may react to possible loss in revenue from tobacco sales. The results of several such studies are summarized in Table 8.5.

Table 8.5: Estimated net changes in employment from the loss of tobacco consumption*

* The net impact of employment lost in the tobacco industry offset by new employment in other industries.

Source: Jacobs et al, 2000.

The research questions addressed by the two groups are different although they may appear to be the same. The first group of studies have estimated the number of jobs related to the tobacco industry, both directly and indirectly. In contrast, the second type of studies have addressed a scenario of how the economy would adjust if tobacco is eliminated from the economy. The questions of how many jobs are related to the tobacco industry and how many people would become unemployed without tobacco are two quite different questions. To assume that an economy would not adjust, and that all tobacco-related jobs would disappear, is not realistic., An economic system perhaps would adjust to any shock. However, the adjustment could occur slowly and the resources freed from eliminating one economic activity may not be fully used by other activities. Unemployment could be significant during the transition period.

8.10 IMPACT OF TOBACCO CONTROL ON FARMERS

The aim of tobacco control is to reduce tobacco consumption, and hence production would be reduced. Tobacco farmers would be affected during this transition. Studies have been conducted to examine a number economic issues related to tobacco farmers during this transition.

8.10.1 Tobacco control policies and farm revenues

The impact of federal cigarette taxes on farmers’ incomes have been examined in various studies in the United States, e.g. Chase Econometrics, 1985; Price Waterhouse, 1992; Sumner and Wohlgenant, 1985; Brown, 1995. Some studies use a macro-econometric model or accounting method to estimate the impact of cigarette tax increases on each sector of the tobacco industry including the farm sector. These studies have concluded that increased cigarette taxes would reduce demand for tobacco, and reduce the incomes of tobacco farmers. For example, the Price Waterhouse study (1992) estimated that increases in the federal tax on cigarettes from 24 to 48 cents per pack in the United States would lead to a loss of farm revenues by $50 millions and farm jobs by 8140. In contrast, Sumner and Wohlgenant (1985) and Brown (1995) claimed that the impact of tobacco measures such as increasing cigarette taxes on farmers’ revenues would also depend on the response by the federal government to the lower demand for tobacco due to the governmental intervention in tobacco farming. They used a partial equilibrium model to simulate the impact which increases in cigarette taxes would have on revenues for two groups of people: tobacco quota owners and tobacco farmers (those who rent the tobacco quotas). They have concluded that adoption of a policy to fix the tobacco quota while allowing the support price to fall would lead to a loss of revenue for quota owners and no change in revenue for the tobacco farmers who rent quota for growing tobacco. But, adopting the converse policy would lead to a revenue gain for the quota owner and a revenue loss for the quota renter.

8.10.2 Effectiveness of farm diversification efforts

Reducing cigarette consumption can create economic hardship for those whose livelihood depends on tobacco. Thus, in developed countries, there have been efforts to diversify the economic activities for tobacco farmers and reduce their dependence on tobacco farming.

There have been efforts in a few countries to encourage tobacco farmers to grow alternative crops. In the United States, the efforts have been scattered and farmers have not been offered any financial incentive to switch crops. The search for alternatives has included the labour-intensive specialty crops such as vegetables and value-added activities such as food processing. The high return from growing tobacco has generally limited the impact of efforts to encourage the production of alternatives.

In Canada, the Tobacco Diversification Plan provided tobacco farmers with incentives to stop growing tobacco and to develop alternatives in the 1980's (PAHO, 1992). Significant numbers of farmers have ceased tobacco production through this program, but many participants acknowledge that they would have quit tobacco farming without the program. The apparent success of the program is further qualified by the finding that 24 percent of the participants continue to work in tobacco farming as employees, rather than as entrepreneurs (PAHO, 1992).

Several studies have evaluated the possible alternative crops for tobacco in developing counties. Those crops include cassava in Brazil, sugar cane in Kenya and chillies, soybeans, cotton, and mustard in India (Jacobs et al., 2000). A study on potential crops in Bangladesh has found that a number of vegetable species and pulses could be more profitable than tobacco. Rose blooms have been identified as profitable alternatives to tobacco in Zimbabwe (Maravanyika, 1998). Yach (1996) reported that more than fifty alternative crops and land use have been identified.

Many tobacco growing households are already quite diversified, growing several different crops and with family members holding off-farm jobs. For example, in the United States, some flue-cured tobacco farms also grow soybean, corn, cotton and wheat. On smaller United States farms typical of burley tobacco, beef cattle are common (Jacobs et al., 2000).

From the experience of developed countries, changing tobacco farming to other crop production seems to be an irrational policy tool and has a very limited success. Tobacco farmers in developed countries have responded to the decline in demand for tobacco and diversified their operations to some extent. Tobacco farmers could not be expected to reduce their tobacco production as long as tobacco remains more profitable than the other crops. In most countries, there is no crop which can substitute for tobacco with the same level of profitability. Thus, other crops can only be supplementary for tobacco.

8.11 REFERENCES

Advisory Commissions on Intergovernmental Relations (ACIR). 1985. Cigarette tax evasion: A second look. Washington DC.

Andrews, R.L., & Franke, G.R. 1991. The determinants of cigarette consumption. Journal of Public Policy and Marketing 10(1):81-100.

Baltagi, B.H., & Goel, R.K. 1987. Quasi-experimental price elasticities of cigarette demand and the bootlegging effect. American Journal of Agricultural Economics 69(4):750-754.

Baltagi, B.H. & Levin, D. 1986. Estimating dynamic demand for cigarettes using panel data: The effects of bootlegging, taxation, and advertising reconsidered. The Review of Economics and Statistics 68(1):148-155.

Bardsley, P. & Olekalns, N. 1998. Cigarette and tobacco consumption: Have anti-smoking policies made a difference? Working Paper, Department of Economics, The University of Melbourne.

Barendregt, J.J., Bonneux, L. & van der Maas, P.J. 1997. The health care costs of smoking. New England Journal of Medicine 337(15):1052-1057.

Barnett, P.G., Keeler, E. & Hu, T.-W. 1995. Oligopoly structure and the incidence of cigarette excise taxes. Journal of Public Economics 57(3):457-470.

Bartlett, J.C., Miller, L.S., Rice, D.P. & Max, W.B. 1994. Medical care expenditures attributable to cigarette smoking - United States, 1993. MMWR 43:469-472.

Becker, G.S., Grossman, M. & Murphy, K.M. 1991. Rational addiction and the effect of price on consumption. American Economic Review 81:237-241.

Becker, G.S. & Murphy, K.M. 1998. A theory of rational addiction. Journal of Political Economy 96(4):675-700.

Beghin, J.C. & Chang, R.E. 1992. Differential products and supply controls in the analysis of agricultural policy reform: The case of tobacco. Agricultural Economics 7:301-315.

Beghin, J.C.,. Foster, W.E, & Kherallah, M. 1996. Institutions and market distortions: International evidence for tobacco. Journal of Agricultural Economics 47(3):355-365.

Beghin, J.C. & Hu, F. 1995. Declining U.S. tobacco exports to Australia: A derived demand approach to competitiveness. American Journal of Agricultural Economics 77(2):260-267.

Bishop, J.A. & Yoo, J.H. 1985. Health scare, excise taxes and advertising ban in the cigarette demand and supply. Southern Economic Journal 2:402-411.

Blaylock, J.R. & Blisard, W.N. 1992. U.S. Cigarette Consumption: The Case of Low-Income Women. American Journal of Agricultural Economics 74(3):668-705.

British American Tobacco. 1995. Tobacco taxation guide: A guide to alternative methods of taxing cigarettes and other tobacco products. Woking (England): Optichrome The Printing Group 1994.

Brown, A.B. 1995. Federal cigarette taxes and smoking restriction: Impacts and policy implications. American Journal of Agricultural Economics 77(4):946-951.

Cameron, S. 1997. Are Greek smokers rational addicts? Applied Economics Letters 4(7):401-402.

Chaloupka, F.J. 1992. Clean indoor air laws, addiction, and cigarette smoking. Applied Economics 24(2):193-205.

Chaloupka, F.J. 1990. Men, women, and addiction: The case of cigarette smoking. National Bureau of Economic Research, Working paper number 3267.

Chaloupka, F.J. 1991. Rational addictive behavior and cigarette smoking. Journal of Political Economy 99(4):722-742.

Chaloupka, F.J., & Grossman, M. 1996. Price, tobacco control policies and youth smoking. National Bureau of Economic Research, Working paper no. 5740.

Chaloupka, F.J., Hu, T-W., Warner, K.E. & van der Merwe, R. 2000. Taxation of tobacco products. In P. Jha & F.J. Chaloupka, eds. Tobacco Control Policies in Developing Countries, Oxford University Press, 2001.

Chaloupka, F.J., & Laixuthai, A. 1996. U.S. Trade policy and cigarette smoking in Asia. National Bureau of Economic Research, Working Paper No. 5543.

Chaloupka, F.J., & Pacula, R.L. 1998a. Limiting youth access to tobacco: The early impact of the Synar amendment on youth smoking. Working paper, Department of Economics, University of Illinois at Chicago.

Chaloupka, F.J., & Pacula, R.L. 1998b. An examination of gender and race differences in youth smoking responsiveness to price and tobacco control policies. National Bureau of Economic Research, Working Paper No. 6541.

Chaloupka, F.J., & Saffer, H. 1992. Clean indoor air laws and the demand for cigarettes. Contemporary Policy Issues 10(2):72-83.

Chaloupka, F.J. and Warner K.E. 1998. The economics of smoking. In A. Culyer, & J.Newhouse, eds. The Handbook of Health Economics, 2000.

Chaloupka, F.J., & Wechsler, H. 1997. Price, tobacco control policies and smoking among young adults. Journal of Health Economics 16(3):359-373.

Chapman, S. & Richardson, J. 1990. Tobacco excise and declining tobacco consumption: The case of Papua, New Guinea. American Journal of Public Health 80(5):537-540.

Chase Econometrics. 1985. The economic impact of the tobacco industry on the United States economy in 1983. Chase Econometrics, Bala Cynwyd, PA.

Coady, S., Pompelli, G. & Grise, V.N. 1991. Government policies and programs affecting tobacco production and trade in major tobacco trading nations. Tobacco Situation and Outlook, US Dept. Agriculture, TS-216 (September 1991):33-37.

Collishaw, N.E. & Myers, G. 1984. Dollar Estimates of the Consequence of Tobacco Use in Canada, 1979. Canadian Journal of Public Health 75(3):192-9.

Conniffe, D. 1995. Models of Irish tobacco consumption. Economic and Social Review 26(4):331-347.

Duffy, M. 1996a. An econometric study of advertising and cigarette demand in the United Kingdom. International Journal of Advertising 15:262-284.

Duffy, M. 1996b. Econometric studies of advertising, advertising restrictions, and cigarette demand: A survey. International Journal of Advertising 15:1-23.

Evans, W.N., Farrelly, M.C. & Montgomery, E. 1996. Do workplace smoking bans reduce smoking? National Bureau of Economic Research, Working paper number 5567.

Farrelly, M.C., & Bray, J.W. 1998. Response to increase in cigarette prices by race/ethnicity, income, and age groups - United States, 1976-1993. Office on Smoking and Health, Morbidity and Mortality Weekly Report 47(29):605-609.

Flewelling, R.L., Kenney, E., Elder, J.P., Pierce, J., Johnson, M. & Bal, D.G. 1992. First-year impact of the 1989 California cigarette tax increase on cigarette consumption. American Journal of Public Health 82(6):867-869.

Fulginiti LE, & Perrin R.K. 1993. The theory and measurement of producer response under quotas. Review of Economics and Statistics 75:97-106.

GAO. 1990. Trade and Health Issues: Dichotomy between U.S. Tobacco Export Policy and Antismoking Initiatives. United States General Accounting Office. GAO/NSIAD-90-190. Washington, D.C., Amy, 1990.

Galbraith, J.W. & Kaiserman, M. 1997. Smuggling and Demand for Cigarettes in Canada: Evidence from Time-series Data. Journal of Health Economics 16(3):287-301.

Goodwin, B.K. & Sumner, D.A. 1992. Estimation of market supply parameters under mandatory production quotas. Unpublished paper, Department of Agricultural and Resource Economics, North Carolina State University, December 1992.

Grise, V.N. 1990. The world tobacco market - government intervention and multilateral policy reform. United States Department of Agriculture, Economic Research Service, Statistics Report No. AGES 9014.

Grossman, M., Coate, D., Lewit, E.M. & Shakotko, R.A. 1983. Economic and other factors in youth smoking. Washington: National Science Foundation.

Hay, J.W. 1991. The harm they do to others: a primer on the external costs of drug abuse. In M.B Krauss & E.P. Lazear, eds. Searching for alternatives: drug-control policy in the United States. Stanford (CA): Hoover Institution Press, pp. 200-225.

Hodgson, T.A. 1992. Cigarette smoking and lifetime medical expenditures. The Milbank Quarterly 70(1):81-125.

Hsieh, C.-R. & Hu, T.-W. 1997. The demand for cigarettes in Taiwan: domestic versus Imported cigarettes. The Institute of Economics, Academic Sinica. Nankang, Taipei: Discussion Paper No. 9701.

Hu, T.-W., Bai, J., Keeler, T.E., Barnett, P.G. & Sung, H.-Y. 1994. The impact of California Proposition 99, a major anti-cigarette law, on cigarette consumption. Journal of Public Health Policy 15(1):26-36.

Hu, T.-W., Qui-Fang, R., Keeler, T.E. & Bartlett, J. 1995a. The demand for cigarettes in California and behavioral risk factors. Health Economics 4(1):7-14.

Hu, T.-W., Sung, H.-Y. & Keeler, T.E. 1995b. Reducing cigarette consumption in California: tobacco taxes vs. an anti-smoking media campaign. American Journal of Public Health 85(9):1218-1222.

Hu, T.-W., Sung, H.-Y. & Keeler, T.E. 1995c. The state anti-smoking campaign and the industry response: the effects of advertising on cigarette consumption in California. American Economic Review 85(2):85-90.

Jacobs, R., Gale, F., Capehart, T., Zhang, P. & Jha, P. 2000. The supply-side effects of tobacco control policies. In P Jha &. F.J. Chaloupka, eds. Tobacco control policies in developing countries, Oxford University Press, 2000.

Jin, S.G., Lu, B.Y., Yan, D.Y. et al. 1995. An evaluation of smoking-induced health costs in China 1988-1989. Biomedical and Environmental Sciences 8:342-349.

Jones, A.M. 1989. A systems approach to the demand for alcohol and tobacco. Bulletin of Economic Research 41:85-105.

Joossens, L. & Raw, M. 1995. Smuggling and cross border shopping of tobacco in Europe. British Medical Journal 310:1393-1397.

Joossens, L. & Raw, M. 1998. Cigarette smuggling in Europe: Who really benefits? Tobacco Control 7:66-71

Joossens, L. & van der Merwe, R. 1997. Cigarette trade and smuggling. Project update #7, The Economics of Tobacco Control Project, Cape Town, South Africa.

Keeler, T.E., Hu, T.-W., Barnett, P.G. & Manning, W.G. 1993. Taxation, regulation and addiction: A demand function for cigarettes based on time-series evidence. Journal of Health Economics 12(1):1-18.

Keeler, T.E., Hu, T.-W., Barnett, P.G, Manning, W.G., & Sung, H.Y. 1996. Do cigarette producers price-discriminate by state? An empirical analysis of local cigarette pricing and taxation. Journal of Health Economics 15:499-512.

Leu, R.E., & Schaub, T. 1983. Does smoking increase medical expenditures? Social Science & Medicine 17:1907-1914.

Lewit, E.M. & Coate, D. 1982. The potential for using excise taxes to reduce smoking. Journal of Health Economics 1(2):121-145.

Lewit, E.M., Coate, D. & Grossman, M. 1981. The effects of government regulation on teenage smoking. Journal of Law and Economics 24(3):545-569.

Manning, W.G., Keeler, E.B., Newhouse, J.P., Sloss, E.M., & Wasserman, J. 1989. The taxes of sin: Do smokers and drinkers pay their way? Journal of the American Medical Association 261(11):1604-1609.

Manning, W.G., Keeler, E.B., Newhouse, J.P., Sloss, E.M., & Wasserman, J. 1991. The costs of poor health habits. Cambridge (MA): Harvard University Press.

Mao, Z.Z. 1996. Demand for Cigarettes and price policy: a time-series analysis in Chinese. Working paper, School of Public Health: West China University of Medical Sciences.

Maravanyika, E. 1997. The economics of tobacco in Zimbabwe. Economics of Tobacco Control Project, Update No. 9: University of Cape Town.

Maravanyika, E. 1998. The search for an optimal tobacco control policy in Zimbabwe. In: I. Abedian, R. van der Merwe, N. Wilins & P. Jha, eds. The Economics of tobacco control: towards an optimal policy mix. Cape Town (South Africa): Applied Fiscal Research Centre, University of Cape Town.

Maravanyika, E. 1998. Do financially viable alternatives to tobacco growing exist in Zimbabwe? Economics of tobacco control project, Update No. 12: University of Cape Town.

Merriman, D., Yurekli, A. & Chaloupka, F.J. 2000. How big is the world-wide cigarette smuggling problem? In F.J., Chaloupka, & K.E. Warner, eds. Tobacco control policies in developing countries, Oxford University Press, 2000 pp. 365-392

Mullahy, J. 1985. Cigarette smoking: habits, health concerns, and heterogeneous unobservables in a micro-econometric analysis of consumer demand. Dissertation, Charlottesville VA: University of Virginia.

National Cancer Institute. 1993. The Impact of Cigarette Excise Taxes on Smoking Among Children and Adults. Summary Report of a National Cancer Institute Expert Panel. Bethesda, MD, National Cancer Institute, 1993.

PAHO. 1992. Pan American Health Organization, Tobacco or Health: Status in the Americas. Scientific Publication number 536.

Pena, P.P. & Norton, G. 1993. The effects of sectorial and economic-wide policies on tobacco production in the Dominican Republic. Journal of Agricultural and Apllied Economics, 25:151-64.

Pekurinen, M. 1991. Economic aspects of smoking: is there a case for government intervention in Finland? Helsinki: Vapk-Publishing.

Peterson, D.E., Zeger, S.L., Remington, P.L. & Anderson, H.A. 1992. The effect of state cigarette tax increases on cigarette sales 1955-1988. American Journal of Public Health 82(1):94-96.

Price Waterhouse. 1992. The economic impact of the tobacco industry on the United States economy. Arlington VA: Price Waterhouse.

Rezitis, A.N., Brown, B.A. & Foster, W.E. 1998. Adjustment costs and dynamic factor demand for U.S. cigarette manufacturing. Agricultural Economics 18:217-231.

Roningen, V. & Dixit, P.M. 1989. Economic implications of agricultural policy reforms in industrial market economies. USDA-ERS Staff Rep. AGES 89-36.

Saffer, H. & Chaloupka, F.J. 1999. Tobacco Advertising: Economic Theory and International Evidence. National Bureau of Economic Research, Working Paper No. 6958.

Schelling, T.C. 1978. Egonomics, or the art of self-management. American Economic Review 68:290-294.

Schelling, T.C. 1984a. Choice and Consequence. Cambridge (MA): Harvard University Press.

Schelling, T.C. 1984b. Self-command in practice, in policy, and in a theory of rational choice. American Economic Review 74:1-11.

Seldon, B.J. & Boyd, R. 1991. The stability of cigarette demand. Applied Economics 23:319-326.

Seldon, B.J. & Doroodian, K. 1989. A simultaneous model of cigarette advertising: Effects on demand and industry response to public policy. Review of Economics and Statistics 71:673-677.

Showalter, M.H. 1991. Essays in applied econometrics. Essay III: Monopoly behavior with intertemporal demands. Dissertation, Cambridge (MA): Massachusetts Institute of Technology.

Simonich, W.L. 1991. Government anti-smoking policies. New York: Peter Lang Publishing.

Sumner, D.A., & Alston, J.M. 1985. Removal of price supports and supply controls for U.S. tobacco: An economic analysis of the impact. Washington, DC: National Planning Association.

Sumner D.A. & Alston, J.M. 1987. Substitutability for farm commodities: The demand for U.S. tobacco in cigarette manufacturing. American Journal of Agricultural Economics 69:258-265.

Sumner, D.A. & Ward, R. 1981. Tax changes and cigarette prices. Journal of Political Economy 89(6):1261-1265.

Sumner, D.A., &. Wohlgenant, M.K. 1985. Effects of an increase in the federal excise tax on cigarettes. American Journal of Agricultural Economics 67(2):235-242.

Sung, H.-Y., Hu.,T.-W. & Keeler, T.E. 1994. Cigarette taxation and demand: An empirical analysis/model. Contemporary Economic Policy 12(3):91-100.

Tauras, J.A. & Chaloupka, F.J. 1998. Price, clean indoor air laws, and cigarette smoking: Evidence from longitudinal data for young adults. Working paper, Department of Health Management and Policy, University of Michigan.

Tegene, A. 1991. Kalman filter and the demand for cigarettes. Applied Economics 23(7):1175-1182.

Townsend, J.L. 1987. Cigarette tax, economic welfare, and social class patterns of smoking. Applied Economics 19:355-365.

Townsend, J.L. 1996. Price and Consumption of Tobacco. British Medical Bulletin 52(1):132-42.

Townsend, J.L., Roderick, P. & Cooper, J. 1994. Cigarette smoking by socioeconomic group, sex, and age: effects of price, income, and health publicity. British Medical Journal 309(6959):923-926.

Tremblay, C.H., & Tremblay, V.J. 1995. The impact of cigarette advertising on consumer surplus, profit, and social welfare. Contemporary Economic Policy 13(1):113-124.

United States Department of Agriculture. 1996. Tobacco situation and outlook report. Washington, D.C.: US. Dept of Agriculture, Economic Research Service, 1996, TBS-236.

United States Department of Health and Human Services. 1989. Reducing the Health Consequences of Smoking: 25 Years of Progress. A Report of the Surgeon General. Atlanta: U.S. Department of Health and Human Services, Public Health Service, Centers for Disease Control, National Center for Chronic Disease Prevention and Health Promotion, Office of Smoking and Health. DHHS Publication No. (CDC). 89-8411.

Van der Merwe, R. 1998a. The Economics of Tobacco Control in South Africa. In I. Abedian, R. van der Merwe, N. Willins & P. Jha, eds. The economics of tobacco control: towards an optimal policy mix, Cape Town, South Africa: Applied Fiscal Research Centre, University of Cape Town.

Van der Merwe, R. 1998b. The implications of falling tobacco consumption for employment in South Africa. In: The economics of tobacco control in South Africa. Cape Town: School of Economics, University of Cape Town, June 1998:107-130.

Viscusi, W.K. 1995. Cigarette taxation and the social consequences of smoking. In: J.M. Poterba,, ed. Tax policy and the economy. Cambridge MA.: Massachusetts Institute of Technology Press, pp. 51-101.

Warner, K.E. 1990. Tobacco taxation as health policy in the Third World. American Journal of Public Health 80:529-531.

Warner, K.E., Hodgson, T.A. & Carroll, C.E. 1998. The medical costs of smoking in the United States: Estimates, their validity, and their implications. Ann Arbor MI.: Department of Health Management & Policy, University of Michigan.

Wasserman, J. Manning, W.G., Newhouse, J.P. & Winkler, J.D. 1991. The effects of excise taxes and regulations on cigarette smoking. Journal of Health Economics 10(1):43-64.

Winston, G.C. 1980. Addiction and backsliding: A theory of compulsive consumption. Journal of Economic Behavior and Organization 1(4):295-324.

World Health Organization. 1997. Tobacco or health: A global status report. Geneva: World Health Organization.

World Health Organization. 1999. Making a difference, World health report 1999. Geneva: World Health Organization.

Xu, X.,. Hu, T.-W., & Keeler, T.E. 1998. Optimal Cigarette Taxation: Theory and Estimation. Working paper, Department of Economics, University of California, Berkeley.

Yen, S.T. & Jones, A.A. 1996. Individual Cigarette consumption and addiction: a flexible limited dependent variable approach. Health Economics 5 (1):105-117.

Yurekli, A.A. & Zhang, P. 2000. The impact of clean indoor-air laws and cigarette smuggling on demand for cigarettes: An empirical model. Health Economics 9(2):159-170.

Yach, D. 1996. Tobacco in Africa. World Health Forum. 17, 29-36.

Yach, D. & Mackay, J. 1996. Tobacco control from a global and national perspective. South African Medical Journal 86(8):931-2.

Zhang, P. & Husten, C. 1998. The impact of the tobacco price support program on tobacco control in the United States. Tobacco Control: An International Journal 7:176-182.

Zhang, P., Husten, C. & Giovino, G. 2000. Effect of the tobacco price support program on cigarette consumption in the United States: An updated model. American Journal of Public Health. 90(5):746-750.