The Effects of Price Cap Regulation on Tobacco Market
Full text
The Effects of Price Cap Regulation on Tobacco Market by Diana Soraia Ferreira da Moita Master in Economics Dissertation Supervised by: Paula Sarmento, PhD 2014
i Biographic Note Diana Soraia Ferreira da Moita was born in the city of Viseu, Portugal, on August 19, 1991. In September 2009 she started the undergraduate studies in Economics from the Department of Economics, Management and Industrial Engineering of University of Aveiro, having finished it on July 2012. This dissertation is a component of the Master of Economics of Faculty of Economy of University of Porto which she is attending since September 2012.
ii Acknowledgments I would like to express my gratitude to all the people who helped me on the elaboration of this dissertation. I am especially grateful to my supervisor, Doctor Paula Sarmento, for her guidance, assistance and presence whenever it was necessary. Their advices and suggestions were indispensable to accomplish this work. For the motivation, patience and unconditional support I am also thankful to all my family and friends, particularly to Nelson, my parents, my sister and my brother.
iii Abstract The tobacco market has serious market failures that justify tobacco regulation. Besides being an addictive good with severe consequences for health, the worldwide tobacco industry is characterized by high concentration and consequently by the presence of market power exerted by tobacco manufacturers. The intention of this work is to study the economic effects of an innovative method to regulate the tobacco market: price cap regulation applied to tobacco producers. Price cap regulation has already been applied in many sectors around the world. However it was never related to tobacco market until 2010. In this work, it is developed a theoretical model with two periods where an upstream firm, subject to price cap constraint, sells to the downstream firms. Additionally the strategic behavior of the upstream firm and the asymmetric costs between downstream firms is also object of study. The model shows the impact of price regulation, the upstream behavior and the asymmetric costs on the market variables. The results indicate that price cap regulation reduces producer’s profits when this firm does not behave strategically and also that the consumers benefit from lower prices. However, if the tobacco retail price does not change, the government revenue will be increased and consumers only benefit from higher quantities. Regarding the cost asymmetry, a curious result is that it can be profitable for the upstream firm if there are asymmetric costs between his customers. JEL codes: C72, L51, L66 Keywords: Tobacco Industry, Economic Regulation, Price Cap
iv Resumo O mercado de tabaco tem sérias falhas de mercado que justificam a sua regulação. Para além de ser um bem aditivo, com consequências graves para a saúde, a indústria mundial do tabaco é caracterizada por ser altamente concentrada e, consequentemente, pela presença de poder de mercado exercido pelos produtores de tabaco. O objectivo deste trabalho é estudar os efeitos económicos de um método inovador para regular o mercado do tabaco: regulação price-cap aplicada aos produtores de tabaco. Este tipo de regulação já tem sido aplicado em diversos sectores em todo mundo mas nunca foi relacionado ao mercado do tabaco até 2010. Neste trabalho é desenvolvido um modelo teórico com dois períodos onde uma empresa a montante, sujeita a restrição de preço, vende para as empresas a jusante. Adicionalmente, o comportamento estratégico da empresa a montante e a existência de custos assimétricos entre as empresas a jusante é também objecto de estudo. O modelo mostra o impacto da regulação de preços, do comportamento a montante e dos custos assimétricos nas variáveis de mercado. Os resultados indicam que este regime regulatório reduz os lucros do produtor quando este não se comporta de forma estratégica e também que os consumidores beneficiam de preços mais baixos. No entanto, se o preço de retalho não se alterar a receita fiscal aumentará e os consumidores apenas beneficiarão de mais quantidades disponíveis. No que diz respeito à assimetria de custos, um resultado curioso é que pode ser lucrativo para a empresa a montante se existirem custos assimétricos entre os seus clientes. Códigos JEL: C72, L51, L66 Palavras-Chave: Indústria do Tabaco, Regulação Económica, Price Cap
v Table of Contents Biographic Note ............................................................................................................ i Acknowledgments ........................................................................................................ ii Abstract ....................................................................................................................... iii Resumo ........................................................................................................................ iv List of Tables .............................................................................................................. vii List of Figures ............................................................................................................ vii Chapter 1 Introduction .......................................................................... 1 Chapter 2 Literature Review............................................................ 4 2.1. Tobacco Industry .......................................................................................... 4 2.2. Price cap Regulation ..................................................................................... 9 Chapter 3 Portuguese Tobacco Industry – Market Structure Characterization ............................................................... 14 Chapter 4 Price Cap Regulation - The Model................. 17 4.1. Hypothesis and variables ............................................................................ 18 4.2. Solving the problem - PCR application ...................................................... 20 4.2.1. Scenario 1: Symmetric Costs between Downstream Firms and Upstream Myopic Monopolist .................................................................... 21 4.2.2. Scenario 2: Symmetric Costs between Downstream Firms and Upstream Strategic Monopolist .................................................................. 25 4.2.3. Scenario 3: Asymmetric Costs between Downstream Firms and Upstream Myopic Monopolist .................................................................... 27
vi 4.2.4. Scenario 4: Asymmetric Costs between Downstream Firms and Upstream Strategic Monopolist .................................................................. 31 4.3. Model’s results .............................................................................................. 33 4.4. Discussion ..................................................................................................... 39 Chapter 5 Conclusion ............................................................................ 42 Appendices ....................................................................................................... 45 Appendix A - Results .................................................................................................. 45 References ......................................................................................................... 46 Relevant Legislation ................................................................................. 50 Other Documentation ............................................................................. 50
vii List of Tables Table 1: Results Scenario 1 vs Scenario 2 ...................................................................... 34 Table 2: Results Scenario 1 vs Scenario 3 ...................................................................... 36 Table 3: Results Scenario 1 vs Scenario 4 ...................................................................... 36 List of Figures Figure 1: Market Structure ............................................................................................. 18
1 Chapter 1 Introduction Economic regulation is necessary when there are market failures that significantly damage the market efficiency as it is the case of the tobacco industry. This market has obvious failures not only caused by externalities and lack of information but also due to the market power exercised by the so-called "Big Four" (Philip Morris International, British American Tobacco, Japan Tobacco and Imperial Tobacco) of the tobacco industry which together represent about 70% of world market share (Bialous and Peeters 2012). According to the fourth edition of The Tobacco Atlas1, the profit’s combination of the mentioned companies was 35 billion dollars in 2010 which corresponds to the combined profits of the Microsoft, Coca-Cola and Mc Donald’s (Eriksen et al. 2012). The tobacco industry is currently subject to several forms of regulation which try to achieve quite different objectives. The World Health Organization (WHO) divided those measures according two different aims: the reduction of the demand for tobacco and the reduction of the tobacco supply. In addition there are other policies to protect the environment (Nunes 2014). In order to reduce the demand the following measures, among others, can be applied: Price and tax measures (price and tax policies, prohibiting or restricting sales and imports by international travelers of tax and duty-free tobacco products). Ban tobacco smoking in indoor workplaces and public places (including public transports). Disclosure, to the Government and the public, information about contents and emissions of tobacco products (by manufacturers or importers of tobacco products). 1 Document published in 2012 by the American Cancer Society and World Lung Foundation that includes all the statistical information about tobacco and tobacco industry.
8 youth smoking decisions and number of cigarettes smoked, but also the way they are acquired: with higher prices, teens are less likely to become regular smokers but have incentives to borrow cigarettes on “social market” (Katzman et al. 2007)5. Thus, the current tobacco control policies are ineffective in reaching this group of teens who obtain tobacco trough the social market (Powell et al. 2005) The features of tobacco market described and studied above justify tobacco regulation but the literature also investigates the impact and effectiveness of tobacco control policies. Nations across the world had hardly tried to “fight” tobacco consumption through various tobacco control policies (implemented for more than 20 years). Although the success of the measures and policies is mixed, the price controls seemed to be the most effective measure (Gruber, 2002; Goel and Nelson, 2006). The last 20 years are characterized by a sequence of mergers, acquisitions and privatizations that led to the creation of the largest transnational tobacco companies in terms of volume and market share, as it is now (Bialous and Peeters 2012). The new research published in the last few years focus on different policies that could be adopted on tobacco market, in particularly on the instruments that affect the producers’ profits. It is in this context that it is the major contribution of Gilmore et al. (2010) to the topic. These authors suggest the application of a price cap regulation imposed to the tobacco producers in the United Kingdom, without changing retail price. They argue that price cap would control the excessive margins of producers, increase government revenue (with the transfer of the excess profits through tax increases) and also promote benefits to public health (“(…)preventing downtrading to cheaper products and controlling unwanted industry practices such as cigarette smuggling, price fixing and marketing to the young”). Following the previous work, Branston and Gilmore (2014) show that the application of the price cap regulation imposed to tobacco producers is feasible. They determine the current profits of the tobacco companies in the United Kingdom, how much such profits could be reduced by the implementation of the price cap regulation and consequently 5 The authors used data from the Youth Risk Behavior Survey (USA). This survey distinguishes the youths who actually buy cigarettes of those who borrow.
9 the government revenue increase. The authors use the profits range6 evidenced by several European transnational companies operating in more competitive markets (e.g.: L’Oreal, Unilever, Danone, Nestle, Heineken, Carlsberg, etc.) and analyze two scenarios called “conservative scenario”, which allowed companies to make returns of 20%, and “optimistic scenario” allowing returns of 12%. The results show that “Applying a system of price-cap regulation in the UK would raise around £500 million per annum (US $ 750 million).” It is very important to understand the relation between the market growth (in value, not volume) and the profits to comprehend the industry behavior and consequently planning effective tobacco control policies. Global consumption is declining but industry’ profits continue to increase, which can largely be explained by the industry “(…) phenomenal pricing power which is now fundamental to its future.” (Gilmore 2012). There have been numerous efforts to align the public health objectives with tobacco companies’ actions. Callard and Collishaw (2013) summarize six proposals to do so including ”(..) proposals for new structures through which tobacco products would be supplied, profit controls on the industry and legislated market outcomes” (the profit control suggestion represents the work of Gilmore et al. (2010)). Additionally the different political points of view present on society can be seen as a barrier to the establishment of an optimal policy in the tobacco industry, especially supply-side interventions (Callard and Collishaw 2013). The works made by Gilmore (2012) and Callard and Collishaw (2013) reinforces the need to integrate new methods of regulation on tobacco market and both refer the price cap regulation suggested by Gilmore et al. (2010). 2.2. Price cap Regulation The literature discusses the application of price cap regulation since it was developed in the 1980’s by Stephen Littlechild, United Kingdom Treasury economist (Armstrong et al. 1994) and applied to British Telecom. Over the years several studies have been published about price cap regulation and this instrument has been largely applied by many countries in several sectors (energy, telecommunications, transports, etc.) 6 The profitability is measured using EBITA (Earnings before interest, taxation and amortization) margin.
10 Regulators should regulate firms with market power to avoid the abusive prices that could be set but at the same time the firm’s efficiency should also be promoted. The price cap regulation faces these problems (Cowan 2002). Price cap regulation involves fixing a maximum price during a regulatory period, under which the regulated firm can freely set prices. At the end of each regulatory period the regulator adjusts the maximum price considering inflation, efficiency gains and other exogenous costs. The factor representing the efficiency gains is deducted from the price charged in the previous period to regulation. Thus, the regulated firm has incentives to invest on cost reduction and the consumers also benefit from the productivity increases. Moreover, the higher regulatory periods, the higher costs saving incentives are but consumers only later benefit from the efficiency gains. Therefore, price cap regulation has been popular and seems to succeed in its main objective that is to create cost efficiency incentives (Cowan, 2002; Currier, 2007; Sappington and Weisman, 2010). Usually, the main problem frequently discussed is the difficult on setting the factor that represents the expected productivity increases, the so called X factor. The value of the X factor is usually defined taking into account the comparison between the rate of growth of the firm’s input prices and the other firm’s long-term productivity performance in the industry, the market competitive forces, negotiations between the firms and the regulator and the regulator’s general perspective on the firm efficiency. The definition of the X factor should also consider the consumer’s defense, the autofinancing ability and the technological dynamism (Pires and Piccinini, 1998; Currier and Jackson, 2008). Regarding the technological investments, there still remains the challenge to promote long term investment. Therefore price cap regulation “(…) may be more appropriate for industries without substantial investment requirements where there is excess capacity than for expanding industries with large investment plans.” (Cowan 2002). Also, sectors with greater technological dynamism (as telecommunications sector for instance) present higher X values than those who technological innovation is slower (Pires and Piccinini 1998). Additionally, if the X factor is set too high, the firm may not cover its costs. On the other hand, if it is set too low the firm could earn abnormal profits and prices still remain excessive. Since the productivity evolution is so important there are different methods to define the X factor such as historical analysis and benchmarking. However “If the regulator uses the past
11 performance of the firm, such as profitability, as a guide to setting the X factor then management incentives will be undermined” (Treasury 1999). The plan of price cap regulation includes the definition of the time between the regulation will operate until it is reviewed. The gap between reviews is known as the regulatory lag. This issue is subject to some considerations. A short plan “can help to ensure that prices do not diverge too far from underlying production costs and that realized earnings do not depart too far from the target level (…)” and also could discouraged firms to innovate and reduce production costs. “In particular, if the X factor is re-set frequently to pass on to consumers in the form of lower prices any cost reductions that the regulated firm has achieved, then the firm’s incentive to secure these cost reductions will be limited” (Sappington and Weisman 2010). Additionally, on dynamic regulatory relationships can arise the ratchet effect. According to this effect the regulated firm has incentives to show less efficient in order to induce higher caps (Resende 1997). “In a price cap setting it might happen that cost savings (or at least a fraction of them) generated in excess of those anticipated by the regulator are clawed back to consumers through lower prices”. Therefore, the ratchet effect can origin cycles on cost cutting activity: “(...) firms’ cost cutting activity tends to increase in the early phase of the regulatory cycle, while it weakens as the price review approaches.” (Bottasso and Conti, 2009). Such effect can be the result of the regulator’s lack of commitment to a long term contract. “If the firm by its past performance reveals efficiency, the regulator adjusts the contract to reduce future costly rent given to the firm.” Thus, unless the regulator provides strong incentives to regulated firms, at the beginning of the contract, the efficient firms will tend to mimic the inefficient ones to ensure future rent (Dalen, 1995; Bottasso and Conti, 2009). However, the regulator can use efficiency-improving investment to prevent the ratchet effect: “if inefficient firms over time catch up with efficient firms due to efficiencyimproving investment, future information rent will decrease, thereby reducing efficient firms’ incentives to pool with inefficient firms” (Dalen, 1995) Another disadvantage is the lack of interest to produce with quality levels: “(…) it is not clear that regulators can appropriately trade-off price against quality”. Price cap regulation promotes the cost efficiency but this effort to reduce costs could have an
12 adverse effect on the quality of the good or service produced (Cowan 2002). Because ensure the quality may be expensive, the cost incentives under the price cap regulation may lead to a delivered service quality degradation (Currier and Jackson 2008). Limited by the regulator price, the firm will only invest in quality when the demand increases (and therefore revenue increases) exceed the costs of this investment (Ghirardi 2000). Thus, price caps should be somehow tied up to quality (Brennan 1989). However, there is little evidence that under price cap regulation exists quality degradation since the regulators frequently monitor service quality measures, public the results and consequently threaten with lower prices on next review (Banerjee, 2003; Cowan, 2002) As said before price cap regulation is usually applied in utilities. A very common scenario in utilities is the existence of partial separation. This separation is a vertical separation which means that the firm is separated in terms of activities practiced and some specific activities are subject to regulation and others are not. In some sectors, the regulated activity is downstream and the upstream activity is deregulated. Through a theoretical model, Reitzes (2007) studied the application of a price cap regulation in that case. The author shows that the imposition of downstream price cap jointly with an appropriate profit sharing rate (between the regulated downstream firm and consumers) can eliminate the market power exercised by the upstream firm, since it can be induced to marginalize the price. Thus, the regulator can indirectly constrain the market power exercised by the upstream affiliate through the regulation system applied to downstream affiliate. However, price cap regulation applied downstream is less effective if the upstream rivals also behave strategically. The comparison between price cap regulation and other forms of regulation is also discussed in microeconomic literature. Despite the problems and given the advantages of this method the authors argued that price cap regulation is superior to other types of regulation, namely rate of return regulation (Clemenz, 1991; Liston, 1993; Pires and Piccinini, 1998; Vogelsang, 2002; Currier and Jackson, 2008; Sappington and Weisman, 2010). Price cap regulation has been so popular since it provides incentives for cost reduction and promotes an efficient pricing (Vogelsang, 2002). Hence, price cap regulation is more effective than rate-of-return regulation regarding the creation of incentives to promote efficiency since under rate-of-return regulation the firms have no
13 incentives to cost efficiency (Currier and Jackson, 2008). Additionally, Clemenz (1991) sought to study the case where the regulator is not able to observe investment expenditures of a firm who wants to reduce its’ production costs and conclude that price cap method “(…)not only provide stronger incentives for investments in cost reduction than rate-of-return regulation, but that it is also capable of achieving a higher social welfare”.
14 Chapter 3 Portuguese Tobacco Industry – Market Structure Characterization This section describes the tobacco industry structure in Portugal mostly using the data from SABI7 (Iberian Balance Sheet Analysis System). SABI is a database that contains information about companies in Spain and Portugal that goes back to 12 years ago. On the worldwide stage there are four major transnational companies operating on all phases of the value chain of tobacco industry (excluding Chinese National Tobacco Corporation, a state-owned tobacco company, that owns about 98% of the Chinese market): the Phillip Morris International (PMI), the British American Tobacco, the Japan Tobacco International and Imperial Tobacco, in descending order by size on volume production (Gilmore et al. 2010). Moreover PMI is the market leader in cigarettes production and the most profitable company in the world, in this sector. Additionally, around the world there are more than 500 factories that jointly produce 6 trillion cigarettes. In Portugal the cigarette production was 25 billion pieces in 2010 (Eriksen et al. 2012). It is important to note that production and distribution are two distinct markets. Therefore, to find how many companies operate on each market the companies where search by CAE (Economic Activity Classification), which is different among these markets, in Portugal. The data found helped to describe the tobacco industry and will help to define some hypothesis for model’s construction. PMI is the only company of the “Big-Four” group that has a subsidiary in Portugal, called “A Tabaqueira”. “A Tabaqueira” is one of the largest companies in the country 7 SABI was accessed on February of 2014 to explore Portuguese tobacco industry specifically the number of firms, operational results and market shares. It was used the most recent data which refer to the year of 2012.
15 and the largest tobacco company in Portugal. British American Tobacco is represented as one of the shareholders of the smallest tobacco company (“SUTSociedade Unificada de Tabacos Lda.”) operating on production market so has almost no impact in Portugal. On tobacco production8 the market is highly concentrated with only five companies operating on this market: “Tabaqueira - Empresa Industrial de Tabacos, S.A.” “Empresa Madeirense de Tabacos S.A.” “Fábrica de Tabaco Micaelense S.A.” “STG Portugal, S.A.” “SUTSociedade Unificada de Tabacos Lda.” Of this five only two have higher significance: “Tabaqueira - Empresa Industrial de Tabacos, S.A.” and “Empresa Madeirense de Tabacos S.A.” It is noteworthy that it is the PMI subsidiary - “TabaqueiraEmpresa Industrial de Tabacos, S.A”- that have the greatest market share holding about 45% of the production market. Additionally, “Fábrica de Tabaco Micaelense S.A.” holds about 23% of production market but is also classified with another CAE9. Thus, this company operates not only on production market but also on retail market. Therefore 23% is not an exact value for tobacco production market share. The distribution market10 is considered as the wholesale market and not the final place where tobacco is sold (retailers on specialized stores11). The PMI subsidiary (which holds the largest market share on production market) has a “sister company” - “Tabaqueira II, S.A” that operates on the distribution market. On distribution market it is the PMI subsidiary that has greater expression in relation to other companies, holding about 11% of the market. The remaining firms have all market shares below 5% and more than 50% of the remaining firms have market shares below 1%. 8 Production market is classified with CAE 12000. 9 That company is also classified with CAE 47260, which refers to tobacco retailers on specialized stores. 10 The distribution market is classified with CAE 46350. 11 The retail market is classified with CAE 47260.
16 The main difference between Portuguese tobacco production market and Portuguese tobacco distribution market is the number of firms. The production market is very concentrated while on distribution market there are over two hundred companies. In addition, many of these distributors have more than one CAE which means that they operate and compete on other markets. Thus, the total revenue does not come from only tobacco distribution activity. This distinction allows underline even more the position of the PMI subsidiary. Regarding the tobacco price, it is know that tobacco is subject to taxation. Therefore, the final price includes the taxes imposed by the Portuguese government. Thus, at the packing process and distribution preparation stages, the final price is already established. In Portugal, about 83% of the retail price is composed by taxes (included the value-added-tax (VAT) and other tobacco taxes) (Nunes 2014), having generated, in 2013, a tax revenue around 1312, 9 € million (DGAIEC 2014). The “Imposto sobre o Tabaco (IT)” is part of the excise taxes group and is applied to the selling price of the cigarette packs.12 According to the “Autoridade Tributária e Aduaneira (AT) the IT rates currently in force for cigarettes are 17% for the Ad Valorem element and 87,33€ (per 1000 cigarettes) for the specific element.13 Additionally cigarettes are subject to a minimum excise duty that corresponds to 104% of the tax resultant from the application of the tax rate charged to the most popular price category (“Código dos Impostos Especiais de Consumo (Decreto-Lei n.º 73/2010, de 21/06 - redacção dada pela Lei n.º 83-C/2013, de 31/12)) ”. 12 Other types of tobacco as the cigars, cigarillos, fine cut tobacco etc. are also subject to taxation and there have been an increase on taxation levels among those types. 13 The specific element corresponds to a fixed value per number of pieces (in tobacco case per 1000 cigarettes) while the Ad-valorem element corresponds to a fraction of the value taxed, expressed as a percentage.
17 Chapter 4 Price Cap Regulation - The Model This chapter describes the model constructed to study the effects of price cap regulation imposed to an upstream firm that sells the product to two downstream firms Furthermore, the analysis of price cap regulation is also developed considering other aspects of the market. Then, the strategic behavior of the upstream firm and the asymmetric costs between downstream firms is also studied. The structure of the market considered in the model is the following. At the upstream level (the production market) there is a monopolist that sells the product to two downstream firms (the distribution market).14 Both distributors have to buy from the monopolist. Hence, we assume that there is no alternative supplier of the product.15 The distributors sell the product to the final consumers.16 The upstream firm is subject to price cap regulation. This means that at each period the regulator sets the maximum price allowed for this firm. We consider that initially the regulator announces that it will imposed a price cap on the second period which depends on the price observed in the first period and also on the expected rate of inflation and efficiency gains, as explained in Chapter 2. Hence the model is developed as a sequential game in order to have a better understanding of the application of the price cap mechanism. At the first stage, firms decide the quantities (as a monopolist in the case of the upstream firm and as Cournot followers in the case of the downstream firms), without any price cap constraint. Then, prices and profits of the first stage are set. At the second stage, firms compete again, and 14 As described in Chapter 3, the Portuguese tobacco industry is characterized by a highly concentrated production market that we represented by a monopoly. At the distribution market there is competition that we represented by the duopoly in which one of the firms is vertically integrated with the producer (“sisters companies”). 15 We do not consider the possibility of imports, for instance. 16 In the circuit there are also the retailers between the wholesale distributors and the final consumers. However the retailers are not relevant to the model.
24 )r-(a 3b 1 =q=q=q 11 1B1A (14) The total quantity is given by: )r-(a 3b 2 =Q 1 1 (15) And the price r1 expression is: 11 bQ 2 3 -a=r (16) 4th Step - M chooses Q1 At the fourth step the monopolist cost function is 11M Qc*F+ = TC and the profit expression is ). -(F+cQ*Q = rπ1111M The inverse demand function faced by the monopolist was determined in the previous step and is given by equation (16). Note that here the profit maximization problem of the monopolist does not have any price constraint. Then, the first order condition of this problem leads to a total quantity produced by M of: c)(ab3 1 =Q1 (17) In order to obtain the equilibrium values for all the other quantities and prices we substitute the value of Q1 into the previous expressions. The equilibrium values are the following: First period Total quantity: )ca( b3 1 Q1 Individual quantities: )ca( b6 1 qq B1A1 Prices: c 2 1 a 2 1 r1 and c 3 1 a 3 2 P1
25 Firm M profits: 6Fb)-c²2ac-²a( b6 1 1M Firms DA and DB profits: ²)ca( b36 1 B1A1 Second period Total quantity: )2ca( b3 1 Q2 Individual quantities: )2ca( b6 1 qq B2A2 Prices: c 2 1 a 2 1 r2 and 3 2 c 3 1 a 3 2 P2 Firm M profits: 6Fb)-²4-c²2ac-²a( b6 1 2M Firms DA and DB profits: ²)2ca( b36 1 B2A2 4.2.2. Scenario 2: Symmetric Costs between Downstream Firms and Upstream Strategic Monopolist On scenario 2 it is the monopolist behavior that differs from the first scenario, which means that the monopolist chooses the quantities of period 1 and 2 that maximize the sum of the profits in period 1 and 2, given by: )-(F+cQ*Q) + r-(F+cQ*Q = rΠ222111M (18) As on previous scenario, firms DA and DB have identical costs. Thus, at the first step firms DA and DB choose q2A and q2B that maximize their individual profits. Then, we reach the same results as in scenario 1 given by the equations (7) and (8). At the second step, firm M chooses Q2 maximizing M subject to the price cap constraint and considering the inverse demand function derived on previous step expressed by equation (9). Then, the firm M maximization problem is the following: Max Q2 )cQ(F+-Q*r + )cQ(F+-Q*r = 222111M s.t. -r r 12
26 The Lagrange function is: L = )rbQ 2 3 a()cQF()bQ 2 3 a(Q)cQF(Q*r 1222*2111 From the first order conditions we obtained the Q2 that maximizes firm’s M profit that is also the same that on first scenario and given by the equation (10). Then, at the third step firms DA and DB choose the first period quantities, in the same way as in the first scenario. The individual quantities that maximize the profits are equal for the firms and given by equation (14). Hence the total quantity is expressed by equation (15). At the fourth step firm M chooses Q1 in order to maximize M . On this maximization problem M have to consider the inverse demand functions derived on the previous steps and the Q2 found on second step through the Lagrangean first order conditions, (expression that depends on r1 which in turns depends on Q1) Then, the firm M problem on first period is the following: Max Q1 )cQ(F+-Q*r + )cQ(F+-Q*r = 222111M with 11 bQ 2 3 -a=r , 22 bQ 2 3 ar and )r22a2( b3 1 Q12 Thus, the equilibrium quantity is δ)c-(ab3 1 =Q1 . Notice that the parameter δ, which belong to the price constraint applied on second period, has an impact on the equilibrium quantity of the first period. The equilibrium quantities and prices are now obtained by backward substitution and are given by:
27 First Period Total quantity: )ca( b3 1 Q1 Individual quantities: )ca( b6 1 qq B1A1 Prices: 2 1 c 2 1 a 2 1 r1 and 3 1 c 3 1 a 3 2 P1 Firm M profits: 6Fb)-²-c²2ac-²a( b3 1 M Firms DA and DB profits: ²)ac( b36 1 B1A1 Second Period Total quantity: )ca( b3 1 Q2 , Individual quantities: )ca( b6 1 qq B2A2 Prices: 2 1 c 2 1 a 2 1 r2 and 3 1 c 3 1 a 3 2 P2 Firm M profits: 6Fb)-²-c²2ac-²a( b3 1 M Firms DA and DB profits: ²)ca( b36 1 B2A2 4.2.3. Scenario 3: Asymmetric Costs between Downstream Firms and Upstream Myopic Monopolist Here it is studied the effects of asymmetric costs between DA and DB. Without loss of generality it is assumed that DA is the more efficient company only bearing the wholesale price as unitary cost. Differently, besides the wholesale price, firm DB has an additional unitary cost of e, with 0e . Then DB’s cost function is given by 2B2B q*e)+(r = TC . This function indicates that the marginal cost of producing one unit of output is higher, on the value e, for DB than for DA. Thus, the profit function of DB is
28 B222B2 q*)erP( (19) Considering again the Cournot behavior for each firm the DA’s best response function is again given by equation (5). Contrary to what happens to firm DA, the best response function of firm DB is different from the previous scenarios. By maximizing π2B in order to q2B it is found the DB’s best response function: . 2b e-rbq-a = q 22A 2B (20) Solving the system with both best response functions we obtain the individual quantities produced by DA and DB: )r-e(a 3b 1 = q 22A (21) )r-2e-(a 3b 1 = q 22B (22) Thus, the total quantity is: )2r-e-(2a 3b 1 = q+q = Q 22B2A2 (23) And the derived inverse demand faced by M is now expressed by: 22 bQ 2 3 -e 2 1 -a = r (24)
29 Following the second step of first scenario the maximization problem of firm M is: Max Q2 )cQ+(F-Q*r = 222 M2π s.t. -r r 12 The Lagrangean function is now given by: L= )rbQ 2 3 e 2 1 a()cQF()bQ 2 3 e 2 1 a(Q 1222*2 From the first order conditions it is found the total quantity produced M on second period: e).-2r-2+(2a 3b 1 =Q 12 (25) Notice that, in this scenario, Q2 is influenced not only by the first period wholesale price (r1) and the parameter , that denotes the price cap action (as on the other scenarios), but also by the constant e that reflects the cost asymmetry between downstream firms. Considering now the first period and using the same reasoning as in the first step, in third step is obtained the best response functions of firms DA and DB that are given by: 2b rbq-a = q 11B 1A (26) 2b e-rbq-a = q 11A 1B (27) Solving the system with the best response functions, the quantities that maximize the individual profits of DA and DB are:
30 )r-e(a 3b 1 = q 11A (28) ).r-2e-(a 3b 1 = q 11B (29) Thus, the total quantity produced on first scenario is given by: )2r-e-(2a 3b 1 = q+q = Q 11B1A1 (30) And the price r1 expression is: 11 bQ 2 3 -e 2 1 -a = r (31) In the fourth step firm M maximize πM1 in order to Q1, considering the price r1 defined by equation (31), leading to the choice of e)-2c-(2a 6b 1 =Q1 . The equilibrium values are obtained by backward substitution and are the following: First Period Total quantity: )ec2a2( b6 1 Q1 Individual quantities: )e5c2a2( b12 1 qA1 and )e7c2a2( b12 1 qB1 Prices: e 4 1 c 2 1 a 2 1 r1 and e 6 1 c 3 1 a 3 2 P1 Firm M profits: 24Fb)-e²4ec4c²4ea-8ac-²a4( b3 1 1M Firms DA and DB profits: ²)e5c2a2( b144 1 A1 and ²)e7a2c2( b144 1 B1
31 Second Period Total quantity: )e4c2a2( b6 1 Q2 Individual quantities: )e54c2a2( b12 1 qA2 and )e74c2a2( b12 1 qB2 Prices: e 4 1 c 2 1 a 2 1 r2 and e 6 1 3 2 c 3 1 a 3 2 P2 Firm M profits: 24Fb)-e²²16-4ecc²4ea48ac-²a4( b24 1 2M Firms DA and DB profits: ²)e54c2a2( b144 1 A2 and ²)e74c2a2( b144 1 B2 4.2.4. Scenario 4: Asymmetric Costs between Downstream Firms and Upstream Strategic Monopolist In this scenario it is studied the simultaneous existence of asymmetric costs between wholesalers and the monopolist strategic behavior. As on third scenario, firms DA and DB have asymmetric costs. Thus, at the first step firms DA and DB choose the second period quantities that maximize their individual profits. Then, we reach the same results as in scenario 3 given by equations (21) and (22) for individual quantities and by equation (23) for the total quantity. Following the second scenario, M chooses the quantities Q1 and Q2 that maximize the sum of the profits in period 1 and 2 ( M ), expressed by equation (18). Thus, at the second step, firm M chooses Q2 maximizing M subject to the price cap constraint and considering the inverse demand function derived on previous step equal to equation (24). Then, the firm M problem is the following: Max Q2 )cQ(F+-Q*r + )cQ(F+-Q*r = Π222111M s.t δ-r r 12 ≤
32 The Lagrange function is: L = )rbQ 2 3 e 2 1 a()cQF()bQ 2 3 e 2 1 a(Q)cQF(Q*r 1222*2111 From the first order conditions is obtained the Q2 that maximizes firm’s M profits which is the same as in scenario 3 and given by equation (25). In third step firms DA and DB decide q1A and q1B, respectively. Then, we reach the same results as in scenario 3: equations (28) and (29). Hence the total quantity is given by equation (30). As on second scenario, at the fourth step firm M chooses Q1 in order to maximize MΠ considering the inverse demand functions and the second period quantity that depends on the wholesale price charged on first period. Then, the firm M maximization problem on first period is the following: Max Q1 )cQ(F+-Q*r + )cQ(F+-Q*r = Π222111M with 11 bQ 2 3 e 2 1 -a=r , 22 bQ 2 3 e 2 1 ar and )er22a2( b3 1 Q12 Thus, the equilibrium quantity is e)-2-2c-(2a 6b 1 =Q1 . Therefore, the equilibrium total quantity produced on first period depends not only on the δ parameter of the price constraint restriction but also on the constant e, which reflects the costs asymmetry on distribution market. By backward substitution the equilibrium values of the fourth scenario are the following: First Period: Total quantity: )e2c2a2( b6 1 Q1 Individual quantities: )e52c2a2( b12 1 qA1 and )e72c2a2( b12 1 qB1
33 Prices: e 4 1 2 1 c 2 1 a 2 1 r1 and e 6 1 3 1 c 3 1 a 3 2 P1 Firm M profits: 24Fb)-e²²4-4ecc²44ea-8ac-²a4( b12 1 M Firms DA and DB profits: ²)e52a2c2( b144 1 A1 and ²)e72a2c2( b144 1 B1 Second Period Total quantity: )eδ2c2a2( b6 1 Q2 Individual quantities: )e52c2a2( b12 1 qA2 and )e72c2a2( b12 1 qB2 Prices: e 4 1 2 1 c 2 1 a 2 1 r2 and e 6 1 3 1 c 3 1 a 3 2 P2 Firm M profits: 24Fb)-e²²4-4ecc²44ea-8ac-²a4( b12 1 M Firms DA and DB profits: ²)e52c2a2( b144 1 A2 and ²)e72c2a2( b144 1 B2 4.3. Model’s results In this section we present the results of each scenario and compare them in order to analyze the effects of price cap regulation, cost asymmetry and monopolist strategic behavior. The results are summarized in a single table that can be found in the Appendix A. From the comparison of the results we conclude that there is a result which is verified in all scenarios. It is observed that q2A, q2B and Q2 are always higher than q1A, q1B and Q1. Moreover, r2 and P2 are always lower than r1 and P1. These results mirror the action of price cap regulation on the second period. With a maximum price on the second period, that by definition is lower than the price observed on the previous period, quantities rise on the second period. In the following description of the results we take the simplest scenario, scenario 1, as the baseline scenario that will be compared with the others scenarios.
40 reduction on period’s 2 price due to the effect of the inflation, efficiency gains and possible exogenous costs. From the model we conclude that all the second period variables are influenced by the parameter , reflecting the action of price cap regulation. That action is also visible when the only firm that works on production market, market where it is applied the price regulation, maximize their profits strategically. In that situation, the first period variables are also influenced by . Moreover when it is studied the scenarios where there are cost asymmetry between distribution firms, all the variables are influenced by the constant e. That constant represents the difference among the production costs of the distribution companies. Hence, in a market with cost asymmetry the variables values mirror that difference. With the application of the price cap regime, the production costs and final prices (r and P) suffer a reduction and the market quantities increase (from first to second period). Thus, distribution firms can produce more at a lower production cost (they consequently earned higher profits) and consumers benefit from extra quantities at a lower price. However, as the price charged by production monopolist reduces, he faces a trade-off between to sell more at lower price or maximize his profits knowing that will be subject to regulation. Therefore, when the monopolist has strategic behavior (scenarios 2 and 4) he earned the same profits on both periods in order to reduce the effect of the price cap regulation on his profits. When he maximize his profits in each period separately (scenarios 1 and 3), i.e., when he does not take in account that he will be subject to regulation on second period, his profits are higher on first period comparing with the second period. Also, if the condition k4 defined in previous section is satisfied it is also satisfied that k2 . Hence, it is assured that the production monopolist earns the higher profits when he behaves strategically and his clients have cost differences (scenario 4). On the opposite case, i.e., when he maximizes his profits with no concerns about the regulation that will suffer on next period and his clients have identical costs (scenario 1), he receives the lowest profits. Thus, price cap regulation is more effective on reducing producer profits when he does not behave strategically. Moreover the monopolist not
41 only has incentives in behave strategically but also benefit when his clients have asymmetric costs. Regarding the cost asymmetry, when the firms operating on distribution market have the same production costs they produce exactly the same and equally shared the profits, on both periods. If exist cost asymmetry it is the most efficient firm who produces more and earns higher profits. Additionally, it may be profitable for firm M if his clients have differences on production costs. If the difference on production costs between distribution companies is too high, firm M earns higher profits. In that case, the monopolist is willing to reduce the price charged on first period for not lose the inefficient firms (what could happen as an outcome of the impossibility of those firms resist on market). Thus, the greater the cost asymmetry, the greater is the incentive of the monopolist to behave strategically. With concern to the price charged by the monopolist firm we note that this price is higher when the monopolist strategically maximizes his profits and his clients have identical costs. On the contrary, that price is the lowest when the monopolist clients have asymmetric costs. Thus, the distribution companies also benefit with the existence of asymmetric costs, specially the most efficient one, since it has no additional production costs. Furthermore, it is in the case when the distribution firms have identical costs and the monopolist working on production market does not behave strategically (scenario 1) that the final price is the lowest and the total amount available on the market is higher, on both periods. The consumers have more benefits on that case. On the contrary, when there is cost asymmetry and strategic behavior by the upstream monopolist (scenario 4) the total quantity available is the lowest and the final price charged to consumers is the biggest of all scenarios. Moreover, when the monopolist has a strategic maximization profits and assuming that his clients has identical costs (scenario 2) the final price paid by the consumers is higher. In these situations the consumers lose. Finally, we can conclude that although the price cap regulation is an effective measure there is no scenario where all the economic agents benefit.
42 Chapter 5 Conclusion Economic regulation is used when there are market failures which require the government intervention to increase the market competition, to reduce prices on essential goods or even to reduce profits higher than the normal ones. These excess profits can be the result of one of the market failures studied in economics: market power exercised by companies. On tobacco market, the market power exercised by tobacco producers, on a global scale, is one of the failures. The “Big-Four” group of tobacco industry holds around 70% of world market share. In Portugal, one of the companies that belong to that group has a subsidiary in the country. That company is the higher tobacco company and one of the largest companies in Portugal. The literature has intensely discussed the features of price cap regulation and tobacco market but has never related this type of regulation to tobacco market until 2010. Gilmore et al. (2010) were the first to suggest that the application of price cap regulation on tobacco market could raise government revenue and bring benefits to public health. The following work of Branston and Gilmore (2014) applied this idea to United Kingdom and using the profits range evidenced by some European transnational companies operating in more competitive markets the authors reach some interesting conclusions about tobacco market. This dissertation arises following these previous works and contributes to understand the flexibility of price cap regulation and how such instrument could change tobacco market. Thus, the main goals of this dissertation were to study the economic effects of price cap regulation imposed to tobacco producers and to answer the following research questions: “Does price cap regulation reduce producer’s profits?” and “How price cap regulation affects the government revenue and consumer welfare?”
43 To achieve these objectives, we built a theoretical model inspired on the characteristics of Portuguese tobacco industry. That model sought to study the effects of price cap regulation on a market with the characteristics defined on Chapter 4. The model allows the understanding of what happens if the monopolist operating on production market has a strategic thinking, anticipating the effects of regulation. In addition, the model included the study of asymmetric costs between distribution companies, which makes one of the firms more efficient than the others. From the analyses of the results we conclude that price cap regulation is an effective policy instrument since it reduces the price charged by producers to distribution companies, which is the immediate result of the price cap imposed to tobacco producers. Furthermore the final price is also reduced and the quantities produced are increased. One of the conclusions which enable to answer the research question is that producer’s profits are reduced by price cap regulation but only when the firm does not behave strategically. This is a very important conclusion, especially to regulators. It is expected that a large company quickly adopts a strategic behavior after the implementation of the price regulation. Thus, the regulator must take into account this conclusion. Another important conclusion is that price cap regulation reduces the final price charged to consumers. It would be a good signal if tobacco was a good with no health consequences, which is not the case. Nevertheless and following the proposal of Gilmore et al. (2010), the final prices do not suffer any changes, in order to not encourage tobacco consumption (especially between the youth) and consequently bring benefits to public health. Therefore, the difference would be reestablished by tax increasing. So, the government revenue would rise by 3 2 of the net effect of the inflation, expected efficiency gains and exogenous costs when the producer does not behave strategically and by 3 1 of that effect when he strategically maximizes his profits. Thus, the tax revenue that can be collected by the government is higher when there is no strategic behavior.
44 On a final note, the consumer only benefit of larger quantities available on the market. One of the model’s results is that after price cap regulation the prices are lower and the quantities are higher. However, if this measure is applied to tobacco market it must be ensured that final price is maintained (by applying the suggestion of Gilmore et al. (2010)). Therefore, the consumer welfare can be increased since they pay exactly the same price for more quantities. The theoretical model presented can be applied to any market with similar features as those considered. Moreover, for future research it would be important to study the case of upstream competition and the situation where the firms have an informational advantage against the regulator (with respect to production costs or to the demand faced by firms, for instance). Finally, if price cap regulation is implemented it would be interesting to analyze, through simulations with real data, if the results obtained in the model occur.
45 Appendices Appendix A - Results Results π1B π1A Q1 r1 πM1 Scenario 1 Scenario 2 π2B P2 P1 πM2 q2A Period 2 r2 q2B π2A Scenario 3 Period 1 Scenario 4 Q2 q1A q1B c 3 1 a 3 2 3 1 c 3 1 a 3 2 2 1 +c 2 1 +a 2 1 c 2 1 a 2 1 k b6 1 )( b3 1 ²)-(k 3b 1 b6 1 )( b6 1 ² b36 1 ²)( 36 1 b3 1 -c 2 1 +a 2 1 2 1 -c 2 1 +a 2 1 3 2 -c 3 1 +a 3 2 3 1 -c 3 1 +a 3 2 )( b3 1 ²)4k( b6 1 ²)-(k 3b 1 )2( b6 1 )( b6 1 ²)2( b36 1 ²)( b36 1 )2( b3 1 )e2( b6 1 e 4 1 c 2 1 a 2 1 b24 1 )e52( b12 1 )e72( b12 1 ²)e72( b144 1 e 6 1 c 3 1 a 3 2 )e42( b6 1 e 4 1 c 2 1 a 2 1 )e542( b12 1 )e742( b12 1 ²)e542( b144 1 ²)e742( b144 1 e 6 1 3 2 c 3 1 a 3 2 e 4 1 2 1 c 2 1 a 2 1 ²)4( b12 1 )e22( b6 1 )e522( b12 1 )e722( b12 1 ²)e52( b144 1 ²)e522( b144 1 ²)e722( b144 1 e 6 1 3 1 c 3 1 a 3 2 )e22( b6 1 e 4 1 2 1 c 2 1 a 2 1 ²)16( b24 1 ²)4( b24 1 )e522( b12 1 )e722( b12 1 ²)e522( 144 1 ²)e722( b144 1 e 6 1 3 1 c 3 1 a 3 2
46 References Armstrong, M., Cowan, S. and Vickers, J. (1994), Regulatory Reform: Economic Analysis and British Experience, MIT Press. Ashenfelter, O. and Sullivan, D. (1987), "Nonparametric Tests of Market Structure: An Application to the Cigarette Industry", The Journal of Industrial Economics, Vol. 35, Nº 4, pp. 483-498. Babcock, B. A. and Foster, W. E. (1992), "Economic Rents Under Supply Controls with Marketable Quota", American Journal of Agricultural Economics, Vol. 74, Nº 3, pp. 630-637. Banerjee, A. (2003), "Does incentive regulation ‘cause’ degradation of retail telephone service quality?", Information Economics and Policy, Vol. 15, Nº 2, pp. 243-269. Bialous, S. A. and Peeters, S. (2012), "A brief overview of the tobacco industry in the last 20 years", Tobacco Control, Vol. 21, Nº 2, pp. 92-94. Bottasso, A. and Conti, M. (2009), "Price cap regulation and the ratchet effect: a generalized index approach", Journal of Productivity Analysis, Vol. 32, Nº 3, pp. 191201. Branston, J. B. and Gilmore, A. B. (2014), "The case for Ofsmoke: The potential for price cap regulation of tobacco to raise £500 million per year in the UK", Tobacco Control, Vol. 23, Nº 1, pp. 45-50. Brennan, T. (1989), "Regulating by capping prices", Journal of Regulatory Economics, Vol. 1, Nº 2, pp. 133-147. Brown, A. B., Rucker, R. R. and Thurman, W. N. (2007), "The End of the Federal Tobacco Program: Economic Impacts of the Deregulation of U.S. Tobacco Production", Review of Agricultural Economics, Vol. 29, Nº 4, pp. 635-655. Cabral, L. M. B. (2000), Introduction to Industrial Organization, MIT Press.
47 Callard, C. D. and Collishaw, N. E. (2013), "Supply-side options for an endgame for the tobacco industry", Tob Control, Vol. 22 Suppl 1, Nº, pp. i10-13. Clemenz, G. (1991), "Optimal Price-Cap Regulation", The Journal of Industrial Economics, Vol. 39, Nº 4, pp. 391-408. Cowan, S. (2002), "Price-cap regulation", Swedish Economic Policy Review, Vol. 9, Nº, pp. 167-188. Currier, K. M. (2007), "A practical approach to quality-adjusted price cap regulation", Telecommunications Policy, Vol. 31, Nº 8–9, pp. 493-501. Currier, K. M. and Jackson, B. K. (2008), "A pedagogical note on the superiority of price-cap regulation to rate-of-return regulation", Journal of Economic Education, Vol. 39, Nº 3, pp. 261-268. Dalen, D. M. (1995), "Efficiency-improving investment and the ratchet effect", European Economic Review, Vol. 39, Nº 8, pp. 1511-1522. Dohlman, E. (2010), Post-Buyout Experience: Peanut and Tobacco Sectors Adapt to Policy Reform, DIANE Publishing Company. Fulginiti, L. and Perrin, R. (1993), "The Theory and Measurement of Producer Response Under Quotas", The Review of Economics and Statistics, Vol. 75, Nº 1, pp. 97-106. Ghirardi, A. (2000), "Estratégias de Regulação e Qualidade dos Serviços Públicos", Available at https://www.bnb.gov.br/content/aplicacao/ETENE/Anais/docs/ren2000_v31_ne_a26.pd f, Accessed on July 2014. Gilmore, A. B., Branston, J. R. and Sweanor, D. (2010), "The case for OFSMOKE: how tobacco price regulation is needed to promote the health of markets, government revenue and the public", Tob Control, Vol. 19, Nº 5, pp. 423-430.
48 Gilmore, A. B. (2012), "Understanding the vector in order to plan effective tobacco control policies: an analysis of contemporary tobacco industry materials", Tob Control, Vol. 21, Nº 2, pp. 119-126. Goel, R. K. and Nelson, M. A. (2006), "The Effectiveness of Anti-Smoking Legislation: A Review", Journal of Economic Surveys, Vol. 20, Nº 3, pp. 325-355. Gruber, J. (2002), "The Economics Of Tobacco Regulation", Health Affairs, Vol. 21, Nº 2, pp. 146-162. Katzman, B., Markowitz, S. and McGeary, K. A. (2007), "An empirical investigation of the social market for cigarettes", Health Economics, Vol. 16, Nº 10, pp. 1025-1039. Keeler, T. E., Hu, T.-w., Barnett, P. G., Manning, W. G. and Sung, H.-Y. (1996), "Do cigarette producers price-discriminate by state? An empirical analysis of local cigarette pricing and taxation", Journal of Health Economics, Vol. 15, Nº 4, pp. 499-512. Laux, F. L. (2000), "Addiction as a market failure: using rational addiction results to justify tobacco regulation", Journal of Health Economics, Vol. 19, Nº 4, pp. 421-437. Liston, C. (1993), "Price-cap versus rate-of-return regulation", Journal of Regulatory Economics, Vol. 5, Nº 1, pp. 25-48. Pires, J. and Piccinini, M. (1998), "Modelos de Regulação Tarifária do Sector Eléctrico", Available at http://www.bndes.gov.br/SiteBNDES/export/sites/default/bndes_pt/Galerias/Arquivos/c onhecimento/revista/rev907.pdf, Accessed on July 2014. Powell, L. M., Tauras, J. A. and Ross, H. (2005), "The importance of peer effects, cigarette prices and tobacco control policies for youth smoking behavior", Journal of Health Economics, Vol. 24, Nº 5, pp. 950-968. Raper, K. C., Love, H. A. and Shumway, C. R. (2000), "Determining Market Power Exertion between Buyers and Sellers", Journal of Applied Econometrics, Vol. 15, Nº 3, pp. 225-252.
49 Raper, K. C. and Love, H. A. (2007), "Market power in tobacco: Measuring multiple markets", Agribusiness, Vol. 23, Nº 1, pp. 35-55. Reitzes, J. D. (2007), "Downstream price-cap regulation and upstream market power", Journal of Regulatory Economics, Vol. 33, Nº 2, pp. 179-200. Resende, M. (1997), "Regimes regulatórios: possibilidades e limites", Pesquisa e Planejamento Econômico, Vol. 27, Nº 3, pp. 641-664. Rucker, R. R., Thurman, W. N. and Sumner, D. A. (1995), "Restricting the Market for Quota: An Analysis of Tobacco Production Rights with Corroboration from Congressional Testimony", Journal of Political Economy, Vol. 103, Nº 1, pp. 142-175. Sappington, D. E. M. and Weisman, D. L. (2010), "Price cap regulation: what have we learned from 25 years of experience in the telecommunications industry?", Journal of Regulatory Economics, Vol. 38, Nº 3, pp. 227-257. Showalter, M. H. (1999), "Firm behavior in a market with addiction: the case of cigarettes", Journal of Health Economics, Vol. 18, Nº 4, pp. 409-427. Sullivan, D. (1985), "Testing Hypotheses about Firm Behavior in the Cigarette Industry", Journal of Political Economy, Vol. 93, Nº 3, pp. 586-598. Sumner, D. A. (1981), "Measurement of Monopoly Behavior: An Application to the Cigarette Industry", Journal of Political Economy, Vol. 89, Nº 5, pp. 1010-1019. Treasury, A. (1999), "Price regulation of utilities", Economic Roundup, Vol., Nº 1, pp. 57-69. Vogelsang, I. (2002), "Incentive Regulation and Competition in Public Utility Markets: A 20-Year Perspective", Journal of Regulatory Economics, Vol. 22, Nº 1, pp. 5-27.