Even low import tariff rates have a significant compounding effect on the final retail price of medicines, which in turn impacts on affordability. While much of the “access to affordable medicines” debate is about intellectual property rights (IPRs) and business practices of pharmaceutical manufacturers, import duties and national protectionism are swept under the political rug. In this paper, we provide a synopsis of tariff barriers for exports of pharmaceutical products to the world’s major low and middle income countries (BRICS-MINT countries).
Studying the impact on final prices for consumers, we estimate that the annual compounded financial burden of import tariffs on pharmaceuticals and prevailing trade facilitation inefficiencies is highest for China (up to 6.2bn USD), Russia (up to 2.8bn USD), Brazil (up to 2.6bn USD) and India (737mn USD), followed by Mexico (663mn USD), Turkey (290mn USD), Indonesia (251mn USD), South Africa (177mn USD) and Nigeria (60,000 USD). For Brazil and India, tariffs on medicines increase their final price by up to 80 per cent of the original sales price ex factory. As most BRICS-MINT governments directly buy or settle patients’ invoices for a bulk of medicine products, the sum of all tariff-induced premiums on final prices for pharmaceuticals tends to exceed by far the tariff revenues initially collected by these governments’ customs authorities.
While import tariffs on medicines can cause substantial net losses for governments, taxpayers and patients, they effectively work as a subsidy for companies along national distribution chains. This may lead to a political economy, in which customs authorities and pharmaceutical distributors may have a common interest in maintaining (high) import tariffs. We call for all low and middle income countries to join the “zero for zero” pharmaceutical agreement, which would help to significantly cut the costs of medicines in general, reduce obscurity and absurdities in government spending and create better conditions for the access to medicines for low-income patients in these countries.
Research assistance by Julie Richert and Nicolas Botton is gratefully acknowledged.
Effective Rates of Import Protection in BRICS-MINT Countries
The nominal tariff charged by customs authorities only tells part of the story of the real burden imposed on final consumers. At the counter, the final price of a medicine paid for by a consumer is a combination of the manufacturer’s price, various mark-ups by importers, wholesalers and distributors, and retail pharmacies, doctors and hospitals respectively (see, e.g., IFC 2017; IMS 2014a). Survey data presented by the International Finance Corporation (IFC 2017; referring to Health Action International (HAI) survey data) show that numerous mark-ups along the medicine distribution chain can account for up to 90 per cent of the final price to the consumer, and often are in the 30-50 per cent range in countries with unregulated mark-ups. Specifically, according to the IFC (2017), mark-ups range from
- 25 to 30 per cent for importers,
- 25 to 50 per cent for wholesalers,
- 25 to 75 per cent for sub-wholesalers,
- and 50 to 80 per cent for retailers (for generics products).
At a first glance, tariffs on pharmaceutical products in BRICS-MINT countries may look as if they make up a small proportion of the total cost of medicines only, adding a negligible amount of money to the price of medicines paid exclusively by importers. Yet, because of multiple percentage mark-ups that are added to the base import price of medicines (i.e. the ex factory price of a drug), even low tariffs add significantly to the final price of a medicine product when several mark-ups are charged by distributors along the distribution chain (see, e.g., Goel 2015; Olcay and Laing 2005).
The compounding effect (CE) of import tariffs is reinforced by other measures, such as additional administrative costs due to inefficient border compliance procedures and documentation requirements.
In addition, in some countries, such as India, exclusive supply and distribution agreements, mandatory approvals and no-objection certificates issued (or not issued) by trade associations, limitations on the number of wholesalers, the abuse of market power and other forms of anti-competitive behavior may adversely affect competition in the market and final prices paid by consumers (patients or the government, i.e. taxpayers and contributors to health insurance programs; see UNCTAD India 2015). Importantly, in cases where anti-competitive behavior along the distribution chain of medicines gives power to incumbent companies to maintain (high) fixed percentage mark-ups, an import tariff will always work as a subsidy to distributors as it increases the upstream price on which mark-ups are based and absolute revenues of distributors respectively (see conclusion for additional considerations).
In the following, we estimate the compounding effect of imports tariffs on pharmaceutical imports to BRICS-MINT countries. We take into consideration average trade-weighted import tariffs for the HS 3004 product category. Since non-pecuniary border measures effectively contribute to the thickness of borders for imported commodities in general (see, e.g., Hornok and Koren 2011), we also take into consideration pre-shipment costs for exporters and importers due to current inefficiencies in border compliance procedures and documentation requirements. We do not take into consideration the tariff equivalents of non-tariff trade barriers (NTBs), which may arise due to complex and lengthy pharmaceutical testing and marketing approval procedures in the importing countries.
For exporters, import tariffs create two major types of distortions, which directly and indirectly feedback to prices at the counter: first, the nominal import tariff charged by customs authorities in the importing country, and second, administrative and trade facilitation costs related to import requirements, which increase the pre-shipment value of a commodity and its ex factory price respectively. As these two components add to the price of a medicine product at the very beginning of national distribution chains, the final distortions that they create for consumers in pharmacies, hospitals or doctors’ medical practices go well beyond the level of the actual tariff rate.
The following estimation illustrates how these distortions are passed on along the distribution chain of pharmaceutical imports in BRICS-MINT countries. We take into consideration a variety of components and their compounding effect on prices charged along the distribution chain of a pharmaceutical product, including:
- weighted average import tariffs for BRICS-MINT countries ),
- pre-shipment administrative costs related to country-specific inefficiencies in border compliance and import documentation procedures ),
- mark-ups charged by importers ),
- mark-ups charged by wholesalers),
- mark-ups charged by sub-wholesalers ),
- mark-ups charged by retailers ), and
- country-specific sales taxes on pharmaceutical products ),
It should be noted that the results presented below should not be taken by their precise face value. Due to data limitations and the use of proxy estimates for agents’ mark-ups across the distribution chain of medicines in low and middle income countries, we are not able to account for country-specific peculiarities in the logistics of medicines. In addition, we also did not account for preferential treatments of some imported medicines, e.g. special treatment resulting from tariff exemptions, tax exemptions or drug price controls and product-specific mark-up regulations, due to data limitations. However, the methodology chosen for this analysis stylizes the major determinants of final prices for medicines in a way that allows us to estimate the direction and the size of the financial distortions created by import tariffs and border facilitation inefficiencies in a common way across all countries under study. At the same time, all our assumptions are based on empirical observations in pharmaceutical markets. Table 4 provides an overview of those variables that are used in our estimations as well as the sources of this data.
Table 4: Description of data and data sources
The compounding effect of import tariffs on pharmaceutical products is calculated for low and high mark-ups along national distribution chains, according to the estimates provided by IFC (2017). We provide estimates for two hypothetical trading regimes: one regime that is characterized by zero import tariffs on pharmaceutical products (i.e. the elimination of all existing import tariffs on HS 3004 pharmaceutical products) and a second regime that is, in addition to zero tariffs, characterized by a 50 per cent reduction of costs resulting from prevailing inefficiencies in border compliance procedures and documentation requirements (based on the World Bank’ Doing Business indicators).
Figure 9 provides estimated numbers for the example of India. The Indian government imposes a lump sum tariff of 10 per cent on any kind of pharmaceutical product listed in the HS 3004 category. Assuming high mark-ups along the Indian distribution chain of a medicine product imported from abroad, the final sales price of a product sold at 5.00 USD ex factory (which corresponds to the pre-shipment value of a hypothetical drug) would be 36.48 USD under the current regime. This price is 3.97 USD higher compared to a price that would result from trade at zero-tariffs including a 50 per cent reduction of the costs resulting from prevailing trade facilitation inefficiencies (all other things equal). Accordingly, the overall compounding effect, i.e. the percentage mark-up on the pre-shipment import price of a drug, is 79.5 per cent of the sales price ex factory. For all countries under study, country-specific assumptions are outlined by Table 6 in the Appendix I.
Source: own calculations.
For all BRICS-MINT countries, a summary of country-specific estimates and the final country ranking is provided by Table 5. We provide estimates for (1) the compounding effect in per cent of the net import value of pharmaceutical products (see Figures 10 and 11 and Table 7 in Appendix I), (2) the absolute aggregate compounding effect based on 2016 import values (see Figure 12), (3) absolute price effects for a range of import prices ranging from 0.50 USD to 100.00 USD (see Appendix I, Figures 24 and 25), and estimates for the financial burden (potential savings) expressed (4) in per cent of total annual health expenditure (see Appendix I, Figure 26), (5) in per cent of total annual spending on medicines (see Appendix I, Figure 27) and (6) in per cent of total annual out of pocket spending on medicine (see Appendix I, Figure 28). Finally, estimates for the compounded financial burden are provided (7) on a per capita basis (see Appendix I, Table 8) and attributed to (8) the government (see Appendix I Table 9) and (9) individual private consumers and private households (see Appendix I, Tables 10 and 11). Country snapshots are provided in Appendix II.
Table 5: Summary of country-specific estimates and final ranking, high mark-ups scenario
Source: own calculations. Note: numbers represent absolute and relative estimates derived from the compounding effect () on the basis of high mark-ups in comparison to a scenario that is based on zero-tariff pharmaceuticals trade and a reduction of 50 per cent of costs related to general inefficiencies in border compliance procedures and documentation requirement. For Nigeria, estimates were calculated on the basis of 2014 import data (the latest data available).
For the high mark-ups scenario, a summary of indicators (1) to (9) as well as country-specific ranking is given by Table 5. When measured in per cent of the ex factory price of a medicine product (Figures 10 and 11), the compounding effect is highest in Brazil (up to 79.8 per cent) and India (79.5 per cent), followed by China (44.2 per cent), Indonesia (43.4 per cent), Russia (39.6 per cent), and Mexico (26.4 per cent). Even though Nigeria, Turkey and South Africa apply zero tariffs on imports of pharmaceuticals, the financial burden of inefficient trade facilitation measures amounts up to 16.2 per cent (Nigeria), 12.5 per cent (Turkey) and 11.1 per cent (South Africa) of the ex factory price.
Source: own calculations. Note: estimates do not include a reduction of (pre-shipment) trade costs due to inefficient border compliance procedures and documentation requirements.
Source: own calculations. Note: estimates include sales taxes and a reduction of 50 per cent in (pre-shipment) trade costs due to inefficient border compliance and documentation inefficiencies. For Nigeria, estimates were calculated on the basis of 2014 import data (the latest data available).
The compounding effect per capita is highest for Russia (19.08 USD), Brazil (12.62 USD), Mexico (5.20 USD) and China (4.52 USD), followed by Turkey (3.66 USD), South Africa (3.16 USD), Indonesia (0.96 USD), and India (0.56). Due to low import volumes as well as zero import tariffs, the estimated numbers are only marginal for Nigeria.
When measured in per cent of annual out of pocket spending on medicine, the financial burden “directly” imposed on patients is highest in South Africa (36.2 per cent), Russia (24.0 per cent), Turkey (15.7 per cent), Brazil (13.5 per cent) and China (11.5 per cent), followed by Mexico (7.7 per cent), Indonesia (5.8 per cent), and India (4.5 per cent). The fact that total government spending on healthcare relative to total spending on healthcare is comparatively high in China (63 per cent), Russia (82 per cent) and Brazil (64 per cent) explains the relatively low direct impact on patients out of pocket spending in these countries. However, as the citizens of these countries have to pay taxes and make financial (health insurance) contributions to government-run healthcare programs, the citizens of these countries have to bear the financial burden imposed by import tariffs and inefficient trade facilitation measures and, accordingly, leaves them with less disposable income. Due to low import volumes as well as zero import tariffs, the estimated numbers are only marginal for Nigeria.
The results also indicate that the aggregated adverse impact of import tariffs on pharmaceuticals and cross-border trade facilitation inefficiencies (see Figure 12), is highest in China (up to 6.2bn USD), Russia (up to 2.8bn USD), Brazil (up to 2.6bn USD) and India (737mn USD; high tariffs, low import volume), followed by Mexico (663mn USD), Turkey (290mn USD; zero import tariffs), Indonesia (251mn USD), South Africa (177mn USD; zero import tariffs) and Nigeria (60tsd USD; zero import tariffs).
Taking into account annual government tariff revenues reveals that the financial burden of import tariffs and trade facilitation inefficiencies that can be attributed to government spending on medicines exceeds tariff revenues by 3.36bn USD in China, 1.97bn USD in Russia and 1.35bn USD in Brazil, followed by 360mn USD in Mexico, 171mn USD in India, and 15mn USD in Indonesia. In other words, due to the compounding effect of import tariffs, governments alone tend to finally pay (or reimburse) between two and six times the amount they collect as tariff revenues at the border, while they would save that amount if trade would take place at zero tariffs.
The aggregate average ranking indicates that the adverse financial impact of import tariffs on pharmaceuticals and prevailing cross-border trade facilitation inefficiencies is highest for Russia, Brazil and China, followed by Mexico, Turkey and Indonesia. Although import tariffs for pharmaceutical products are comparatively high for imports to India, the overall compounding effect for India is contained by India’s comparatively low volume of pharmaceutical imports. The fact that South Africa and Nigeria do not impose tariffs on pharmaceutical imports reflects these countries’ good overall score in the ranking of BRICS-MINT countries.
Source: own calculations.
 In their study, Olcay and Laing (2005) report that the compounded effect on final prices may add 20 per cent to the price of a medicine product. They conclude that “there are NO good reasons why those countries should retain tariffs. Tariffs on medicines target the sick which cannot be good public policy” (see page 36).
 Prices for individual medicines can significantly vary among countries. In a survey of 60 countries, a study by Health Action International (HAI 2010) found that the price a patient would have to pay for10ml soluble human insulin in 2010 varies significantly. Although Indonesia ($16.61-$51.15) and South Africa ($32.89-$40.47) had high prices, Nigeria ($18.65-$23.65), Brazil ($20.23-$25.46), India ($3.35-$7.89), and Turkey’s ($16.48-$16.80) prices were very low. None of these countries had higher prices than the US ($51.95-$62.39) and Austria ($76.69).
 For Nigeria, estimates were calculated on the basis of 2014 import data (the latest data available).
 Due to data limitations we did not account for preferential treatments of some imported medicines, e.g. special treatment resulting from tariff exemptions, tax exemptions or drug price controls and product-specific mark-up regulations. As several exemptions apply for sales taxes and government(-mediated) purchases of pharmaceutical products, government revenues from sales taxes have not been taken into consideration. Although sales taxes impact on the revenues and expenditures of several strands of government, the net impact on total government revenues would be neutral. For final consumers, i.e. patients, applicable sales taxes substantially increase the price of medicines that are purchased in hospitals, doctors’ practices and pharmacies.