The Pattern of Entry of Generic Drugs into the Brazilian Pharmaceutical Market *

Transcription

1 The Pattern of Entry of Generic Drugs into the Brazilian Pharmaceutical Market * Klenio Barbosa ** Everton Silva *** Abstract The presence of generic drugs in a market is of great importance to society as it increases competition between suppliers of these products, which can lead to a reduction in the price of already established drugs. In the late 1990s there were relevant regulatory changes in the Brazilian pharmaceutical sector, which allowed for a gradual increase in the share of generic drugs in the domestic market. This paper aims to document the entry pattern of generic drug suppliers in Brazil after the arrival of the first generic manufacturer in the domestic market. Using a sample of 1228 observations involving 148 different drugs (active ingredients), we investigated the relationship between the number of periods after the arrival of the first generic drug supplier in Brazil and the number of suppliers of genetic drugs with the same active ingredient. We find that there was a continuous entry of generic suppliers after the first generic in the market. However, this increase in suppliers occurs at decreasing rates. Keywords: Generic Drugs, Pattern of Entry, Pharmaceutical Industry. JEL Codes: K32; I18; L1; L65. * Submitted in May Revised in July ** Professor, Insper Instituto de Educação e Pesquisa, Brazil. kleniosb@insper.edu.br *** PanPharma, Brazil. everton.silva@panpharma.com.br Brazilian Review of Econometrics v. 37, n. 1, pp May 2017

2 Klenio Barbosa and Everton Silva 1. Introduction At the end of the 1990s, relevant regulatory changes occurred in the Brazilian pharmaceutical sector that allowed for a gradual increase in the share of generic drugs in the domestic market. Two big changes in Brazilian law during the 1990s reshaped the pharmaceutical industry in Brazil: ratification of the TRIPS agreement in 1996, including a provision for granting patents to inventions in the pipeline; and the Generics Act of 1999, which introduced the bioequivalence test and enabled the substitution of pioneering drugs by generics. The presence of generic drugs in a market is of great importance for society as it increases competition between suppliers of these products, which can lead to a reduction in the price of already established drugs. In addition, the entry of generics can increase the supply of drugs. Given the high share of spending on drugs in total health expenditure in Brazil, this process of increasing the number of medical drug suppliers via the entry of generics is of crucial relevance to Brazilian society. Thus, a better understanding of the way the entry of generics occurs is of the upmost interest for policy makers in Brazil, for the industry, and for academia in general. This paper aims to describe the pattern of entry into Brazil of generic drug suppliers after the arrival of the first generics manufacturer in the domestic market. Identifying, for example, whether there is a continuous entry of generic suppliers after the first generic in the market can provide empirical support for the identification strategy used in studies that analyze the effects of the entry of generic drugs (i.e., Caves et al., 1991, Arvate et al., 2013). In these papers, the number of days between patent expiry and the market operation date is used as an instrumental variable for the entry of generics into the market. Such studies implicitly assume that there is a continuous entry of generic drugs after their patent expiry, thus increasing the chance of generic suppliers entering into a public tender. However, no study prior to this article has sought to investigate whether this relationship suggested by Caves et al. (1991), and Arvate et al. (2013), is verified empirically. This article seeks to fill this gap in the literature. Based on a sample of 1228 observations involving 148 different drugs (active ingredients), we investigated the relationship between the number of periods after the entry into Brazil of the first generic supplier and the number of suppliers of generic drugs with the same active ingredient. We verified that there is a continuous entry of generic suppliers after the first generic in the market. However, this increase in suppliers occurs at decreasing rates. To the best of our knowledge, this is the first paper in the Brazilian literature involving the pharmaceutical industry which analyzes this relationship. If on one hand this article describes the pattern of entry of generic drug suppliers into Brazil after the arrival of the first generic manufacturer in the Brazilian market, thus supporting the identification strategies used by Caves et al. (1991), and Arvate et al. (2013), on the other hand it contains some limitations. First, 2 Brazilian Review of Econometrics 37(1) May 2017

3 The Pattern of Entry of Generic Drugs into the Brazilian Pharmaceutical Market it does not allow for it to be identified how characteristics of the Brazilian market (such as income, population size, epidemiological and sanitary conditions, and other characteristics described in Morton (1997) and Kyle (2006)) affect the pattern of entry of generic drugs into Brazil. For this, a dynamic entry model would need to be estimated using parametric and non-parametric methods, such as in Reiffen and Ward (2005). However, the database described and used in this article is not rich enough for such methods to be employed. Second, the database used in this article enables the entry behavior of generic drugs in Brazil to be analyzed for a set of 148 active ingredients. Ideally, this analysis should be carried out for all of the active ingredients available and sold in the Brazilian pharmaceutical market. However, the inexistence of such a richly detailed database does not enable a more general investigation to be carried out. This article is organized in the following way. Section 2 describes the importance of drugs spending in the budget, the pharmaceutical market in Brazil and in the world, and the regulatory changes that occurred in the Brazilian pharmaceutical sector at the end of the 1990s. Section 3 describes the database used in this article and presents the methodology used to analyze the relationship between the number of periods after the arrival in Brazil of the first generic supplier and the number of suppliers of generic drugs with the same active ingredient. Section 4 presents the results and Section 5 concludes. Related Literature. This article is related to the literature on the entry of generic drugs into Brazil. As far as we know, few econometric estimations have been applied to the pharmaceutical sector in Brazil, with both attempting to measure the impact of the entry of generics on prices. Fiuza and Lisboa (2003) indicated that the prices of brand drugs would increase with an increase in the market power of similar ones, a similar result to that of Frank and Salkever (1997), but the database used was from prior to the Generics Act. Nishijima (2008) used a different sample (covering both the period prior to and the period after the Generics Act) and a different model from that of Fiuza and Lisboa (differences-in-differences) and obtained different results, in which the price of the reference drugs would fall in accordance with a greater number of generics. Lopes (2009) recently replicated the Fiuza and Lisboa model with more products in a more recent period and reached the same conclusions as them. Another important paper on the topic was developed by Fiuza and Caballero (2010). These authors estimate the entry of generic versions of drugs outside patent in the various therapeutic classes, using both count data and orderly multinomial models. The results from Fiuza and Cabalero indicate that a simple Poisson model has a worse adjustment, although the explanatory variables exhibit the same pattern of significance and signs. Most of the explanatory variables used are significant, especially the proxies for potential market size (lagged values for drug revenues), market concentration, and age of the drug of reference. Multiple brands (and not simply own generics) and evergreening strategies appear to be effective Brazilian Review of Econometrics 37(1) May

4 Klenio Barbosa and Everton Silva in deterring entries. 2. Institutional Environment 2.1 Spending on medicinal drugs in Brazil Spending on medicinal drugs represents a large part of total health expenditure in Brazil. Brazilian government health spending represented 7.2% of GDP in 2000, whereas a decade later it represented 9.0%, an increase of 1.8% according to the World Health Organization (World Health Organization (WHO), 2013,, 132). Silveira (2007) shows that family spending on medicinal drugs as a share of income is high and has increased over the last decades. Table 1, for example, shows that average monthly family expenditure per capita on medicinal drugs as a share of family income per capita rose from 10.26% at the end of the 1980s to 12.57% at the beginning of this century. Table 1 also shows that family spending on medicinal drugs in the first deciles is smaller than family spending on drugs in the last deciles. This shows that any policy that affects price and access to drugs will have a smaller impact on the budget of the first deciles of the population than the higher deciles. Table 1 Average monthly family spending per capita on medicinal drugs by deciles representing family income per capita (In R$ from January 2003) DECILE Medicinal Drugs st nd rd th th th th th th th Total Source: IBGE/POFs from various years, (Silveira, 2007, p. 151) 4 Brazilian Review of Econometrics 37(1) May 2017

5 The Pattern of Entry of Generic Drugs into the Brazilian Pharmaceutical Market 2.2 Pharmaceutical market in Brazil and the world The pharmaceutical market is highly relevant in various countries in the world. In the United States, for example, pharmaceutical industry turnover represents a little more than 3 times that of the second placed country, Japan, which in 2010 turned over more than 300 billion (IMS Health, IMS Market Prognosis, 2011, p. 24). From Table 2 it can be observed that a continuous rise is expected in the participation of emerging countries in the global market, such as India, which has shown significant growth, rising 3 positions between 2005 and 2010, and which is expected to rise by 4 more positions by Table 2 Global Ranking of the Pharmaceutical Market 2010 Ranking 2005 Ranking 2010 Expected Ranking USA 1 USA 1 USA 2 Japan 2 Japan 2 Japan 3 France 3 China 3 China 4 Germany 4 Germany 4 Germany 5 Italy 5 France 5 France 6 United Kingdom 6 Italy 6 Brazil 7 Spain 7 Brazil 7 Italy 8 Canada 8 Spain 8 India 9 China 9 Canada 9 Spain 10 Brazil 10 United Kingdom 10 Russia 11 Mexico 11 Russia 11 Canada 12 Australia 12 India 12 United Kingdom 13 South Korea 13 Australia 13 Venezuela 14 Turkey 14 Mexico 14 Turkey 15 India 15 South Korea 15 South Korea 16 Russia 16 Turkey 16 Australia 17 Netherlands 17 Poland 17 Mexico 18 Belgium 18 Netherlands 18 Argentina 19 Poland 19 Belgium 19 Poland 20 Greece 20 Greece 20 Belgium Source: IMS Market Prognosis, 2011, p. 23. The pharmaceutical market has undergone various mergers and acquisitions. Many companies have adopted this strategy with the aim of diversifying their business, serving more previously unexplored markets and taking advantage of the wave of growth in emerging countries and the rise in generic drugs. Brazil has increased its share in the world market, rising from 10th (tenth) in Brazilian Review of Econometrics 37(1) May

6 Klenio Barbosa and Everton Silva the ranking in 2005 to 7th (seventh) in It is expected to occupy 6th (sixth) position in This increase is mainly due to the rise in Brazilian income per capita. In the Brazilian market the 5 (five) biggest manufacturers correspond to approximately 28% of total sales. As can be observed in Table 3, one of these, Sanofi-Aventis, which bought Medley in 2009, represents 11.74% of the market. 2.3 Patents and the TRIPS agreement Patents are public concessions that aim to stimulate research, temporarily protecting competitor innovation and thus benefitting the patent holder. They are considered to be a reward for inventors intellectual efforts, serving as an incentive for them to bear the costs of research and continue discovering new inventions. Patent registrations are mainly concentrated in developed countries, with the United States, Japan, and Germany together representing more than half of world patent filings. Brazil occupies 24th position, representing 0.30% in global share. 1 In 1994, a World Trade Organization agreement brought great changes in terms of intellectual property rights. TRIPS (Trade-Related Aspects of Intellectual Property Rights) covers the possibility of patent owners (invention creators) being able to restrict access based on commercial actions, thus impeding third parties without authorization from producing, using, and selling an invention or utility model (Institute of Health, 2006, p. 14). The agreement establishes a global standard for patent protection, with rules on how to file and centralization at the WTO. TRIPS is the most comprehensive legal framework in the area of intellectual property law. It was created in 1995 and covers seven areas of intellectual property: Copyright, Trademarks (including brand services), Geographical Indications (products from a particular city or region), Industrial Design (protecting the shapes of certain goods), Patents, Integrated Circuit Design, and Trade Secrets (World Trade Organization (WTO), 2003). With regards to the pharmaceutical industry, the TRIPS agreement determines that a generic drug can be produced after the expiry or renouncement of patent protection, guaranteeing a maximum monopoly of 20 years, as long as the knowledge is shared and the formula and manufacturing procedure are made public. However, the TRIPS agreement also envisages mechanisms for governments to break patents by decreeing a compulsory license, making it possible for a government, company, or organization to use the patent, temporarily removing the patent holder s monopoly, as long as social and economic well-being are verified, with public health being considered a priority in the face of patent holder interests. As cited in article 31b of TRIPS: b) such use may only be permitted if the proposed user has made 1 Brazil is considered to be a country with a low level of innovation, with a share 10 (times) less than that of scientific research production. According to the UNESCO Report (2010), Brazil occupies 13th place in world scientific production, with 90% of production originating from public universities. 6 Brazilian Review of Econometrics 37(1) May 2017

7 The Pattern of Entry of Generic Drugs into the Brazilian Pharmaceutical Market Table 3 Biggest Pharmaceutical Manufacturers in Brazil 2011 (in US$ million) Rk Manufacturer Sales % Share Rk Manufacturer Sales % Share 1 EMS PHARMA 2, ASTRAZENECA BRASIL MEDLEY 1, NYCOMED PHARMA ACHE 1, TEUTO BRASILEIRO SANOFI-AVENTIS 1, MERCK SERONO EUROFARMA 1, BOEHRINGER ING NEO QUIMICA BIOLAB-SANUS FARMA NOVARTIS LEGRAND MSD SANDOZ DO BRASIL PFIZER ROCHE BAYER PHARMA D M IND.FTCA Source: IMS Health (Mat, December, 2011). Brazilian Review of Econometrics 37(1) May

8 Klenio Barbosa and Everton Silva Table 4 Global Ranking of Patent Filings Ranking Country % Share 1 United States 54,042 51,642 45,627 45, % 2 Japan 27,743 28,76 29,802 32, % 3 Germany 17,821 18,855 16,797 17, % 4 China 5,455 6,12 7,9 12, % 5 Republic of Korea 7,064 7,899 8,035 9, % 6 France 6,56 7,072 7,237 7, % 7 United Kingdom 5,542 5,467 5,044 4, % 8 Switzerland 3,833 3,799 3,672 3, % 9 Netherlands 4,433 4,363 4,462 4, % 10 Sweden 3,655 4,136 3,568 3, % 11 Canada 2,879 2,976 2,527 2, % 12 Italy 2,946 2,883 2,652 2, % 13 Finland 2,009 2,214 2,123 2, % 14 Australia 2,052 1,938 1,74 1, % 15 Spain 1,297 1,39 1,564 1, % 24 Brazil % Others 12,197 13,254 12,164 12, % Total 159, ,24 155, , % Source: WIPO (World Intellectual Property Organization, 2013, p. 28). efforts to obtain authorization from the right holder on reasonable commercial terms and conditions and that such efforts have not been successful within a reasonable period of time. This requirement may be waived by a Member in the case of a national emergency or other circumstances of extreme urgency or in cases of public noncommercial use. In the case of public non-commercial use, where the government or contractor, without making a patent search, knows or has demonstrable grounds to know that a valid patent is or will be used by or for the government, the right holder shall be informed promptly; (WTO, 2003). 2.4 Pharmaceutical market Generics in Brazil In a market with free competition and no barriers to entry, the expectation is that new competitors will enter, attracted by extraordinary market returns, and increase the supply of medicinal drugs, thus forcing a fall in prices until reaching marginal cost. Patent expiry makes the entry of competitors into the market possible, ensuring bioequivalence guarantees and enabling interchangeability between drugs and making them substitutes for reference drugs. In this context, the importance of generic drugs emerges. A drug is defined as generic if it is interchangeable with some reference drug. This possibility of exchanging one drug for another is guaranteed by the requirement of bioequiva- 8 Brazilian Review of Econometrics 37(1) May 2017

9 The Pattern of Entry of Generic Drugs into the Brazilian Pharmaceutical Market lence tests, which ensure that a particular drug has the same components and will produce the same effect as a brand drug. Generic drugs are of great relevance for the well-being of society as they make it possible to access treatments that were previously unviable due to the high price. Thus, part of the population with a lower purchasing power will gain access to the drugs. Berndt (2002) states that when there is 1 (one) competitor producing a generic drug, the price of the drug is 85% that of the reference one; when there are 2 (two) competitors, it decreases to 75% and so on until reaching cases of 10% of the price of the reference drug (BERNDT, 2002, p.63). One example of this dynamic are drugs for treating HIV/AIDS. The annual cost of treating HIV/AIDS in India, for example, fell from an average of almost US15thousandtoU S 150 per patient (Albuquerque, 2008,, p. 6484). The entry of generic drugs has also caused impacts on public health services, allowing for generic drugs to be bought if these are cheaper than the reference drugs and leading to government savings. These government costs are connected with purchasing drugs for their own hospitals and laboratories, which are used for free in public healthcare treatments or the prices of which are subsidized in public pharmacies. In 1999 Act 9,787 was passed in Brazil. This defines the concept of a generic drug as being a drug with the following characteristics: the patent of the reference drug has expired and standards having been established for its sale, such as packaging that includes an uppercase G and contains the inscription Medicamento Genà c rico and the name of the active ingredient in the formula (Agencia Nacional de Vigilancia Sanitaria (ANVISA), 2002,,p. 6). The first registrations of approved generics in Brazil took place in Figure 1 presents the evolution of generic drug registrations in the period between 2000 and Considerable growth can be perceived, in that in 2002 there were 511 registrations and after 10 years there were 3, 411 registrations. This represents more than a six-fold increase. After the commercialization of generics began in 2000, sales in comparison to the reference drugs have risen since, increasing their market share. In April 2009, for example, generic drugs presented a 14.29% share and in April 2013 they represented 22.97%, increasing their market share by 8.68% in only 5 (five) years. Table 5 shows that of the 20 (twenty) biggest generic manufacturers in Brazil in 2013, the 10 (ten) biggest represented 90.50% of the domestic market, with the first 5 (five), Medley, EMS Pharma, Neo Quimica, Germed Pharma, and Eurofarma, respectively, being especially important. It is worth highlighting that in developed countries, the market share of generics is much greater. For example, in December 2011, the market share of generics in units in the USA, Germany, and the United Kingdom, was 60%, 66%, and 60%, respectively (IMS Health apud Progenericos). In Brazil the share in units from Brazilian Review of Econometrics 37(1) May

10 Klenio Barbosa and Everton Silva Figure 1 Generic Drug Registrations Generic Drug Registrations Source: Progenericos, Table 5 Manufacturer Share April/2013 RANKING MANUFACTURER MARKET SHARE % 1 MEDLEY 23.94% 2 EMS PHARMA 21.77% 3 NEO QUIMICA 8.56% 4 GERMED PHARMA 6.90% 5 EUROFARMA 6.63% 6 SANDOZ DO BRASIL 6.21% 7 ACHE 5.48% 8 TEUTO BRASILEIRO 4.87% 9 LEGRAND 3.39% 10 PRATI DONADUZZI 2.76% 11 NOVA QUIMICA 1.39% 12 ZYDUS 1.10% 13 MERCK SERONO 1.10% 14 RANBAXY 0.98% 15 MULTILAB 0.63% 16 CIMED 0.62% 17 GEOLAB 0.51% 18 UNIAO QUIMICA 0.47% 19 VALEANT 0.41% 20 ARROW FARMACEUTICA 0.27% Source: IMS Health. 10 Brazilian Review of Econometrics 37(1) May

11 The Pattern of Entry of Generic Drugs into the Brazilian Pharmaceutical Market May 2011 until April 2012 is around 25%. 3. Database Some studies, such as Reiffen and Ward (2005), argue that the generics industry provides an interesting area for empirical studies for analyzing the competitiveness of a market. Reasons for this include: i) each active ingredient can be considered as an experiment; ii) it is possible to observe the initial opening of the market by taking patent expiry into account; and iii) the companies will bear the relevant costs of requesting authority approval before knowing when and how many rivals will enter into the market (REIFFEN E WARD, 2005, p. 37). The primary database in this paper contains information for various reference products-drugs in the Brazilian market, among which the following information should be noted: the date of entry into Brazil of the first generic drug for each active ingredient and its respective manufacturer, and the date of entry into Brazil of the other generic drugs by active ingredient and their respective manufacturer. Based on this information, the database used in this paper was constructed. Each observation in the sample contains information on the number of quarters after the arrival in Brazil of the first generic supplier for each active ingredient, and the number of generic drug suppliers in the Brazilian market for the same active ingredient. All in all, 1,228 observations were obtained and organized into 148 active ingredients, covering the period between 2000 and Table 6, for example, presents the active ingredient Acetylsalicylic Acid, for which there are 3 manufacturers producing it, besides the first generic entrant. Thus, it is possible to calculate the period difference between these manufacturers and the entrant and verify whether these periods were short or not. Figure 2 presents the number of active ingredients and the number of manufacturers. This figure shows that the most frequent active ingredients are those with 4 (four) manufacturers, presenting 18 (eighteen) active ingredients. Based on the 148 active ingredients and their 1228 entrant manufacturers, Figure 3 was elaborated, which relates the number of quarters after the arrival in Brazil of the first supplier of a generic and the number of suppliers of generic drugs with the same active ingredient. In order to elaborate this figure, the difference was calculated between the date of entry of the manufacturer and the first manufacturer producing the generic. This allowed for this difference to be grouped into periods and for it to be verified how the distribution of entry of the manufacturers occurs over time. Analyzing Figure 3, it is possible to note that as time passes (x axis Quarters) the tendency is for the number of manufacturers to increase, but at decreasing Brazilian Review of Econometrics 37(1) May

12 Klenio Barbosa and Everton Silva Table 6 Example Database Active Ingredient, Commercialization Date ACTIVE INGREDIENT MANUFACTURER DATE OF INITIAL MAN ENTRY DATE QUARTER DIFFERENCE COMMERCIALIZATION DEXAMETASONE AC MG UNIAO QUIMICA F N 09/01/ /01/ DEXAMETASONE AC MG EMS PHARMA 09/01/ /01/ DEXAMETASONE AC MG GERMED PHARMA 09/01/ /01/ DEXAMETASONE AC MG GEOLAB 10/01/ /01/ DEXAMETASONE AC MG TEUTO BRASILEIRO 07/01/ /01/ DEXAMETASONE AC MG CIMED 02/01/ /01/ DEXAMETASONE AC MG SANVAL 08/01/ /01/ ACETYLSALICYLIC AC MG GREEN PHARMA 07/01/ /01/ ACETYLSALICYLIC AC MG CIMED 04/01/ /01/ ACETYLSALICYLIC AC MG BALM LABOR I FARMA 09/01/ /01/ ACEBROPHYLLINE EMS PHARMA 05/01/ /01/ ACEBROPHYLLINE GERMED PHARMA 08/01/ /01/ ACEBROPHYLLINE ACHE 08/01/ /01/ ACEBROPHYLLINE LEGRAND 11/01/ /01/ ACEBROPHYLLINE MEDLEY 12/01/ /01/ ACEBROPHYLLINE TEUTO BRASILEIRO 06/01/ /01/ ACEBROPHYLLINE NEO QUIMICA 10/01/ /01/ ACEBROPHYLLINE UNIAO QUIMICA F N 12/01/ /01/ ACEBROPHYLLINE FARMASA 07/01/ /01/ ACEBROPHYLLINE NOVA QUIMICA 11/01/ /01/ ACEBROPHYLLINE PRATI DONADUZZI 05/01/ /01/ ACEBROPHYLLINE GEOLAB 09/01/ /01/ ACEBROPHYLLINE CIMED 08/01/ /01/ ACECLOFENAC EMS PHARMA 05/01/ /01/ ACECLOFENAC GERMED PHARMA 07/01/ /01/ ACECLOFENAC MULTILAB 04/01/ /01/ ACECLOFENAC LEGRAND 03/01/ /01/ ACECLOFENAC NOVA QUIMICA 08/01/ /01/ ACETYLCYSTEINE MG PRATI DONADUZZI 03/01/ /01/ ACETYLCYSTEINE MG EMS PHARMA 05/01/ /01/ ACETYLCYSTEINE MG LEGRAND 04/01/ /01/ ACETYLCYSTEINE MG UNIAO QUIMICA F N 07/01/ /01/ ACETYLCYSTEINE MG GERMED PHARMA 10/01/ /01/ ACETYLCYSTEINE MG FARMASA 10/01/ /01/ Source: Elaborated by the authors using data provided by IMS. 12 Brazilian Review of Econometrics 37(1) May 2017

13 Acum. No. of Mans % Number of Active Ingredients The Pattern of Entry of Generic Drugs into the Brazilian Pharmaceutical Market Figure 2 Relationship between No. of Active Ingredients and No. of Manufacturers Number of Active Ingredients and Number of Manufacturers Number of Manufacturers Source: Source: Elaborated by the authors using data provided by IMS. Figure 3 Relationship between the Number of Quarters after the Entry of the First Generic and the Accumulated Number of Manufacturers (%) 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% Quarter After the Entry of the First Generic and the Accumulated Number of Manufacturers (%) Number of Quarters after the Entry of the First Generic into the Brazilian Market Source: Source: Elaborated by the authors using data provided by IMS. 15 Brazilian Review of Econometrics 37(1) May

14 Klenio Barbosa and Everton Silva rates. This indicates some non-linearity in the relationship studied. In addition, it is possible to verify the % quantity of manufacturers in accordance with the number of quarters. For example, in the first 2 years, almost 30% of the manufacturers in the sample entered into the market. In order to identify the entry pattern of generics into the Brazilian market, linear regression models will be used to estimate the relationship between the number of periods after the entry into Brazil of the first supplier of the generic for an active ingredient, and the number of suppliers of generic drugs with the same active ingredient. The first model that will be estimated consists of a quadratic equation model, as set out in the following equation: Y i = a + bx i + cx 2 i + ε i, (1) in which Y i is the accumulated percentage of the number of manufacturers, X i is the difference in periods (in quarters) between the date of the first generics manufacturer in the Brazilian market and the entry of the entrant manufacturer, and ε i is the error term. The estimation of equation (1) enables the pattern of entry into Brazil of generics to be identified via parameters b and c. In particular, if c is negative, this means that the relationship is concave and therefore the number of suppliers increases at decreasing rates. If c is positive, this means that the relationship is convex and therefore the number of suppliers rises at increasing rates. In order to identify the elasticity effect of the number of periods after the 1st entry of a drug on the number of manufacturers, a log-log type model will be estimated. This model is represented by the following equation: ln (Y i ) = α + β ln (X i ) + υ i, (2) in which ln(y i ) is the natural log of the accumulated percentage of the number of manufacturers, ln(x i ) is the natural log of the difference in periods (in quarters) between the date of the first generic manufacturer in the Brazilian market and the entry of the entrant manufacturer, and υ i is the error term. In equation (2) β is the coefficient that indicates elasticity, and with this it is possible to relate the impact that a 1% variation in the quarter will cause in X% on the accumulated percentage of manufacturers. For the estimation of equations (1) and (2), the ordinary least squares method was used. The statistical significance of the parameters in equations (1) and (2) will be tested using the student t test and the associated p-value. 4. Results In this section the results from the estimations of equations (1) and (2) are presented. Table 7 presents the results from the estimations. 14 Brazilian Review of Econometrics 37(1) May 2017

15 The Pattern of Entry of Generic Drugs into the Brazilian Pharmaceutical Market Column (1) in Table 7 presents the estimations of equation (1). As this equation is a second degree polynomial, 3 parameters were estimated. All of the estimated parameters are significantly different from zero. This indicates that equation (1) adequately represents the pattern of entry of generic drugs after the arrival in Brazil of the first supplier of a generic with the same active ingredient. The estimation of equation (1) shows that the coefficient associated with the X i variable (number of quarters after the arrival of the first generic in Brazil) is positive, and the coefficient associated with the Xi 2 variable is negative. These parameters determine a concave function, a parabola with downwards concavity, for the relationship between the number of periods after the arrival in Brazil of the first supplier of a generic with the same active ingredient and the number of suppliers of generic drugs with the same active ingredient. This means that after the arrival in Brazil of the first generic drug there is a continuous entry of suppliers, but at decreasing rates. That is, the longer the time that has passed since the entry of the first supplier, the tendency is for the percentage of supplier entries to increase, but at decreasing rates. It can therefore be concluded that over time there is a decrease in entrants into the generics market, after the entry of the 1st manufacturer producing a particular active ingredient. Table 7 Results from the Regressions Dependent Variables Y ln(y ) (1) (2) Xsquared sd: t: p: X sd: t: p: ln(x) sd: t: p: Constant sd: sd: t: t: p: p: Notes: st=standard deviation, t=calculated tstatistic, p=p-value. Column (2) in Table 7 presents the estimations of equation (2). This equation assumes a concave function for the relationship between the number of periods after the entry into Brazil of the first supplier of a generic with the same active ingredient and the number of suppliers of generic drugs with the same active in- Brazilian Review of Econometrics 37(1) May

16 Klenio Barbosa and Everton Silva gredient. This means that in this estimation it is assumed that after the entry of the first generic into Brazil there is a continuous entry of suppliers, but at decreasing rates. Despite this, equation (2) has an advantage: its log-log formulation allows for the effect of elasticity on the variation in the number of periods to be directly estimated, as well as on the number of entrant manufacturers after the first manufacturer of a generic in Brazil. This parameter is the β coefficient. From Table 7 it can be noted that all of the estimated parameters for equation (2) are significantly different from zero. This indicates that equation (2) also adequately represents the pattern of entry of generic drugs after the arrival in Brazil of the first supplier of a generic with the same active ingredient. The estimation of equation (2) shows that the coefficient associated with the ln(x) i variable (natural log of the number of quarters after the entry of the first generic into Brazil) is positive. This means that after the arrival the first generic in Brazil there is a continuous entry of suppliers. Note that this relationship was also estimated via equation (1). Table 7 presents the estimated parameter for the β coefficient. This estimation is equal to This means that increasing the number of periods (quarters) after the arrival of the first generic into Brazil by 1% will result in a 0.84% increase in the accumulated percentage of manufacturers of generics with the same active ingredient in Brazil. 5. Conclusion This article identified the pattern of entry into Brazil of generic drug suppliers after the arrival of the first generic manufacturer in the domestic market. Based on a sample of 1228 observations involving 148 different drugs (active ingredients), the relationship was identified between the number of periods after the arrival in Brazil of the first generic supplier and the number of suppliers of generic drugs with the same active ingredient. It was verified that there is a continuous entry of generic suppliers after the first generic in the market. However, this increase in suppliers takes place at decreasing rates. With this it can be concluded that there is a greater concentration of generic suppliers entering in the first years, with these entries reducing over time. References Abbott, F. M. (2001). The trips agreement, access to medicines, and the WTO Doha ministerial conference. Working Paper 36, Public Law and Legal Theory. Agencia Nacional de Vigilancia Sanitaria (ANVISA) (2002). genéricos: Oriente-se. Brasília. Medicamentos Albuquerque, R. C. (2008). Patentes farmaceuticas e acesso a medicamentos. In XVII Congresso Nacional CONPEDI, Brasília. 16 Brazilian Review of Econometrics 37(1) May 2017

17 The Pattern of Entry of Generic Drugs into the Brazilian Pharmaceutical Market Arvate, P., Barbosa, K., & Gambardella, D. (2013). Generic-branded drug competition and the price for pharmaceuticals in procurement auctions. Working paper, C-Micro. Bae, J. P. (1997). Drug patent expirations and the speed of generic entry. Helath Services Research, 32: Berndt, E. R. (2002). Pharmaceuticals in U. S. healh care: Determinants of quantity and price. Journal of Economic Perspectives, 16: Caves, R., Whinston, M., & Hurwitz, M. (1991). Patent expiration, entry, and competition in the U.S. pharmaceutical industry. Brookings Papers on Economic Activity, Microeconomics:1 48. Fiuza, E. P. S. & Caballero, B. (2010). Estimações de entrada de medicamentos genéricos no Brasil usando modelos de contagem modelos ordenados. Texto de discussão 1511, IPEA. Fiuza, E. P. S. & Lisboa, M. B. (2003). Credence goods and market power: an econometric study of the Brazilian pharmaceutical industry. Estudos Economicos, 33: Frank, R. G. & Salkever, D. S. (1997). Generic entry and the pricing of pharmaceuticals. Journal of Economics and Management Strategy, 6: Iizuka, T. (2008). Generic entry in a regulated pharmaceutical market. The Japanese Economic Review, 60: IMS Health, IMS Market Prognosis (2011). Available from: http: // 20for%20Healthcare%20Informatics/Documents/The_Global_Use_of_ Medicines_Report.pdf Data de Acesso: 24/06/2013. Kyle, M. (2006). The role of firm characteristics in pharmaceutical product launches. The RAND Journal of Economics, 37: Lopes, J. A. (2009). Estudo sobre a reação de preço dos medicamentos líderes de mercado à introdução de concorrentes genéricos e similares. Master s thesis, EESP/FGV. (Dissertação de Mestrado não publicada). Morton, F. S. (1997). The strategic response by pharmaceutical firms to the medicaid most- favored-customer. The RAND Journal of Economics, 28: Nishijima, M. (2008). Os preços dos medicamentos de referência após a entrada dos medicamentos genéricos no mercado farmacêutico brasileiro. Revista Brasileira de Economia, 62: Brazilian Review of Econometrics 37(1) May

18 Klenio Barbosa and Everton Silva Organização das Nações Unidas para a Educação, a Ciência e a Cultura (UNESCO) (2010). Relatório unesco sobre ciência. Available from: Accessed: 24/06/2013. Referência e Treinamento em DST/AIDS de São Paulo/Instituto de Saúde (2006). Propriedade intelectual, patentes & acesso universal a medicamentos. São Paulo. Reiffen, D. & Ward, M. (2005). Generic drug industry dynamics. The Review of Economics and Statistics, 87: Silveira, F. G. (2007). Gasto e consumo das famílias brasileiras contemporâneas, volume 2. Brasília. World Health Organization (WHO) (2013). World health statistics. Genebra. World Intellectual Property Organization (WIPO) (2013). Pct yearly review the international patent system. Available from: sites/www/freepublications/en/patents/901/wipo_pub_901_2013.pdf Accessed: 24/06/2013. World Trade Organization (WTO) (2003). Intellectual property: protection and enforcements. Available from: whatis_e/tif_e/agrm7_e.htm Date of Access: 06/05/ Brazilian Review of Econometrics 37(1) May 2017