Trade Policy, Agreements and Taxation of Multinationals

Trade Policy, Agreements and Taxation of Multinationals Rising Wage Inequality and Trade Lecture 1 Meredith Crowley University of Cambridge July 2015 MC (University of Cambridge) Trade Policy, Agreements and Taxation of Multinationals July 2015 1 / 63

Lecture 1: Rising Wage Inequality and Trade In part 1 of this lecture, we will... review empirical evidence on increasing wage inequality around the world, review the shape of wage distribution, discuss how to measure inequality, examine different theories behind the rise in wage inequality. We will break to read and then discuss recent evidence that inequality is decreasing in Latin America. In part 2 of this lecture, we will... review the Ricardo-Viner Specific Factors Model, apply this model to the offshoring of jobs by Multinational Enterprises, review evidence on offshoring by US Multinational Enterprises. MC (University of Cambridge) Trade Policy, Agreements and Taxation of Multinationals July 2015 2 / 63

Income and wage inequality: rising around the world There have been dramatic increases in income inequality and wage inequality in many countries around the world for the past few decades. The share of a nation s income going to the top 10 % of a nation s population has increased in many countries. The difference in the wage earned by the 90th percentile worker and the 10th percentile worker has increased. Wages earned by more educated workers have increased more rapidly than those earned by less educated workers. the returns to a college education have increased Within demographic and within skill group wage inequality have increased substantially (residual inequality) some college graduates earn much more than others MC (University of Cambridge) Trade Policy, Agreements and Taxation of Multinationals July 2015 3 / 63

Income inequality: rising in the US and Europe Total income consists of wages and returns to capital. The y-axis measures the share of total income received by the top 10% of earners in the US and Europe. In 1970, the top 10% of Americans earned about 33% of US income. In 2010, the top 10% of Americans earned about 47% of US income. In both the US and Europe, income inequality has been rising for several decades. Income inequality in Europe and the United States, 1900 2010 Share of top income decile in total pretax income 50 percent 45 40 35 30 25 1900 1910 Top 10% income share: Europe 1920 1930 1940 1950 1960 Top 10% income share: U.S. 1970 1980 1990 2000 2010 Fig. 1. Income inequality in Europe and the United States, 1900 to 2010. The share of total income accruing to top decile income holders was higher in Europe than in the United States from 1900 to 1910; it was substantially higher in the United States than in Europe from 2000 to 2010. The series report decennial averages (1900 = 1900 to 1909, etc.) constructed using income tax returns and national accounts. See (24), chapter 9, Fig. 9.8. Series available online at piketty.pse.ens.fr/capital21c. MC (University of Cambridge) Trade Policy, Agreements and Taxation of Multinationals July 2015 4 / 63

Income inequality: rising in China from 1986-2003 There is less income inequality in China than in the US or Europe. In 2000, the top 10% of Chinese received about 25.5% of income, compared to 34% in Europe and 46% in the US. But income inequality in China has been rising. Similar increases in income inequality are observed in Japan and the United Kingdom. MC (University of Cambridge) Trade Policy, Agreements and Taxation of Multinationals July 2015 5 / 63

Income inequality: rising in the United Kingdom and Japan United Kingdom Japan MC (University of Cambridge) Trade Policy, Agreements and Taxation of Multinationals July 2015 6 / 63

Rising inequality between average workers in the US The previous figures showed that the top 10% of income earners are earning more over time. College/high school median annual earnings gap, 1979 2012 In constant 2012 dollars What about the average or median worker? 70,000 dollars 60,000 Household gap $30,298 to $58,249 In the US, the median income of a college-educated worker has increased more than the median income of a high school-educated worker over the last 30 years. This is evidence of increasing inequality between skilled and unskilled workers. What has caused increasing inequality? 50,000 40,000 30,000 20,000 10,000 Male gap $17,411 to $34,969 Female gap $12,887 to $23,280 0 1979 1982 1985 1988 1991 1994 1997 2000 2003 2006 2009 2012 Fig. 1. College/high school median annual earnings gap, 1979 2012. Figure is constructed using Census Bureau P-60 (1979 1991) and P-25 (1992 2012) tabulations of median earnings of full-time, full-year workers by educational level and converted to constant 2012 dollars (to account for inflation) using the CPI-U-RS price series. Prior to 1992, college-educated workers are defined as those with 16 or more years of completed schooling, and high school educated workers are those with exactly 12 years of completed schooling. After 1991, college-educated workers are those who report completing at least 4 years of college, and high school educated workers are those who report having completed a high school diploma or GED credential. MC (University of Cambridge) Trade Policy, Agreements and Taxation of Multinationals July 2015 7 / 63

The shape of the wage distribution Labour economists observe that the typical wage distribution is skewed, with a long right tail. A small number of people earn very high wages. The median wage is lower than the average wage. Why does the wage distribution have this shape? MC (University of Cambridge) Trade Policy, Agreements and Taxation of Multinationals July 2015 8 / 63

Why is the wage distribution skewed? Why do we usually observe that the wage distribution is skewed? Consider the remuneration of human capital and ability The most able people acquire more human capital and then are paid more (complementarity between human capital and ability). = the skewness is explained by the positive correlation between ability and human capital. MC (University of Cambridge) Trade Policy, Agreements and Taxation of Multinationals July 2015 9 / 63

Measuring Inequality There are many approaches to measuring inequality. We have looked at the share of earnings going to the top 10% of earners in several countries and the median incomes of skilled versus unskilled workers. Other measures used are: the Lorenz curve; the Gini coefficient; 90-10 wage gaps: the percent wage differential between the worker of the 90th percentile and the one in the 10th percentile; 50-10 wage gaps: it measures the inequality between the middle class and the low-income workers MC (University of Cambridge) Trade Policy, Agreements and Taxation of Multinationals July 2015 10 / 63

Measuring Inequality: the Lorenz curve To construct a Lorenz curve, we: rank the households according to their income level; divide the population in 5 groups (quintiles) of equal size; calculate the share of income received by each quintile. US 2010 (Bureau of the Census) Quintile Share of Cumulative Income share of income First 0.034 0.034 Second 0.086 0.120 Third 0.147 0.267 Fourth 0.233 0.500 Fifth 0.500 1.000 MC (University of Cambridge) Trade Policy, Agreements and Taxation of Multinationals July 2015 11 / 63

The Gini coefficient: A summary measure of the information in the Lorenz curve The area between perfect equality curve and actual Lorenz curve can be used to measure inequality Gini coeff. = area between perfect equality curve and actual Lorenz curve area under the perfect equality curve Gini = 0: perfect equality Gini = 1: perfect inequality This measure is not perfect. Two different distributions can give the same Gini coefficient. MC (University of Cambridge) Trade Policy, Agreements and Taxation of Multinationals July 2015 12 / 63

Urban wages in China: Lorenz curve and Gini coefficient 1.8.6 L(p).4 Figure 1. Lorenz Curves of Wages in Urban China 1988-2008 Year Gini coefficient 1988 0.237 1995 0.345 2002 0.348 2008 0.45.2 0 0.2.4.6.8 1 Percentiles (p) 45⁰ line wage88 wage95 wage02 wage08 Chinese wage inequality increased over 1988-2008. China acceded to the WTO in 2001. Source: Understanding Urban Wage Inequality in China 1988-2008: Evidence from Quantile Analysis by Simon Appleton, Lina Song, Qingjie Xia. February 2013. CHINA GROWTH CENTRE (CGC) Discussion Paper Series No. 18. MC (University of Cambridge) Trade Policy, Agreements and Taxation of Multinationals July 2015 13 / 63

Changes in US wage inequality: the empirical evidence Four facts about the US economy: 1 Since the 1970 s, the US economy has grown quite well more than doubled, in fact. Along with this, labor productivity grew. 2 Over the same period, wages started to lag behind productivity growth. 3 Inequality in wages also increased; median wages have been stagnant, and below-median male wages have fallen. 4 This coincides with the big wave of globalization. MC (University of Cambridge) Trade Policy, Agreements and Taxation of Multinationals July 2015 14 / 63

US productivity and wages: 1960-2006 MC (University of Cambridge) Trade Policy, Agreements and Taxation of Multinationals July 2015 15 / 63

Growing US wage dispersion: 1973-2007 MC (University of Cambridge) Trade Policy, Agreements and Taxation of Multinationals July 2015 16 / 63

Changes in US wage inequality: the empirical evidence Source: Autor, Katz and Kearney (2008) FIGURE 2. THREE MEASURES OF WAGE INEQUALITY: COLLEGE/HIGH SCHOOL PREMIUM, MALE 90/10 OVERALL INEQUALITY, AND MALE 90/10 RESIDUAL INEQUALITY MC (University of Cambridge) Trade Policy, Agreements and Taxation of Multinationals July 2015 17 / 63

Changes in UK wage inequality: Source: Brewer, Sibieta and Wren-Lewis (2007) MC (University of Cambridge) Trade Policy, Agreements and Taxation of Multinationals July 2015 18 / 63

Inequality in developing countries: the empirical evidence Wage inequality is not only a feature of high-income countries. Middle-income and developing countries have seen a rise in wage inequality throughout the 1980s and 1990s. These increases in wage inequality in developing countries have coincided with reforms of domestic laws (labor market reforms), structural (macroeconomic) adjustment programs, better educational opportunities, and trade liberalizations. MC (University of Cambridge) Trade Policy, Agreements and Taxation of Multinationals July 2015 19 / 63

Changes in DC inequality: the empirical evidence Source: Goldberg and Pavcnik (2007) MC (University of Cambridge) Trade Policy, Agreements and Taxation of Multinationals July 2015 20 / 63

Changes in DC inequality: the empirical evidence Source: Goldberg and Pavcnik (2007) cont. MC (University of Cambridge) Trade Policy, Agreements and Taxation of Multinationals July 2015 21 / 63

Changes in DC inequality: the empirical evidence Source: Goldberg and Pavcnik (2007) cont. MC (University of Cambridge) Trade Policy, Agreements and Taxation of Multinationals July 2015 22 / 63

Changes in DC inequality: the empirical evidence Source: Goldberg and Pavcnik (2007) cont. MC (University of Cambridge) Trade Policy, Agreements and Taxation of Multinationals July 2015 23 / 63

Changes in wage inequality: why? There are several competing explanations why we have seen such a dramatic increase in inequality. From the empirical evidence we see that: there has been an increase in wage inequality between skilled and unskilled workers; in the US there has been an increase in the proportion of high skilled workers over low skilled workers. Go back to the basic demand and supply framework. MC (University of Cambridge) Trade Policy, Agreements and Taxation of Multinationals July 2015 24 / 63

Changes in wage inequality: why? Demand and Supply Shift The y-axis is the ratio of wages for high-skilled workers to the wage of low-skilled workers. The x-axis is the ratio of the number of high-skilled workers to the number of low-skilled workers. MC (University of Cambridge) Trade Policy, Agreements and Taxation of Multinationals July 2015 25 / 63

Changes in wage inequality: why? Supply Shift in the US Fast growth in supply of educated workers in the 1970s, slowed down afterwards (baby boomers) Immigration of low skilled workers But, the absolute supply of educated workers was increasing at the same time as the college wage premium was rising = IT MUST HAVE BEEN A DEMAND SHIFT AS WELL MC (University of Cambridge) Trade Policy, Agreements and Taxation of Multinationals July 2015 26 / 63

Changes in wage inequality: why? Demand Shift INTERNATIONAL TRADE: The Heckscher-Ohlin Model The North is abundant in high-skill workers. The South is abundant in low-skill workers. Countries export the goods which are intensive in the factor which they have in abundance. In the North: the factor used in the exporting industry (skilled labour) gains from trade while the factor used in importing industry (unskilled labour) losses from trade. Trade increases wage inequality in the North! But, this model would also predict declining inequality in the labor-abundant South which we do NOT observe. MC (University of Cambridge) Trade Policy, Agreements and Taxation of Multinationals July 2015 27 / 63

Changes in wage inequality: why? Demand Shift SKILL BIASED TECHNOLOGICAL CHANGE Introduction of technologies (computers) which are complementary to skilled labour and substitute for unskilled labour. This seems to be an important source of wage inequality within countries. Is the timing correct? Not exactly. But it depends on which countries you examine. MC (University of Cambridge) Trade Policy, Agreements and Taxation of Multinationals July 2015 28 / 63

Changes in wage inequality: why? Alternative explanations for rising inequality in high-income countries INSTITUTIONAL CHANGES IN LABOUR MARKETS Deunionization. Fall in the real minimum wage in the 1980s. High unemployment and low wage inequality (European experience): countries with strong minimum wage settings and unions experience a lower increase in wage inequality. MC (University of Cambridge) Trade Policy, Agreements and Taxation of Multinationals July 2015 29 / 63

Discussion topic: Wage inequality in Latin America See handout: Levy Yeyati and Pienknagura, Vox 10 June 2014 Please read the handout, Wage compression and the decline in inequality in Latin America: Good or bad? by Levy Yeyati and Pienknagura. In this article, the authors describe a decline in income inequality in Latin America starting in 2003. What do the authors claim is behind the decline in Latin American income inequality? Which arguments do you find most convincing and why? Do you regard the decline in inequality as a good or bad outcome for Latin America? Why? MC (University of Cambridge) Trade Policy, Agreements and Taxation of Multinationals July 2015 30 / 63

Discussion topic: Wage inequality in Latin America Figure 2. Supply of education across regions Notes: LAC-7 & Uruguay: Argentina, Brazil, Chile, Colombia, Mexico, Peru, Uruguay, and Venezuela. SEA MICs: Indonesia, Malaysia, Philippines, South Korea, and Thailand.EE MICs: Estonia, Hungary, Poland, and Slovakia. PCEs: Australia, Canada, New Zealand, Norway, and Sweden. AT MICs: South Korea, Singapore, and Thailand. Source: Barro-Lee (2010) and de la Torre, Levy Yeyati, and Pienknagura (2013) MC (University of Cambridge) Trade Policy, Agreements and Taxation of Multinationals July 2015 31 / 63

How has globalization impacted the distribution of wages? Globalization refers to many things : international trade; capital and labour mobility; the role of multinationals; outsourcing; the role of technological and information diffusion; fewer trade barriers. We will begin by studying the impact of trade liberalizations on the distribution of wages. MC (University of Cambridge) Trade Policy, Agreements and Taxation of Multinationals July 2015 32 / 63

The Specific Factors Model This is sometimes called the Ricardo-Viner model. The Specific Factors Model lets us look at implications of trade for the DISTRIBUTION of income within country. It helps us understand which productive factors gain and which lose from a trade liberalization in the short run. What is a specific factor? A factor of production that is unable to move into or out of an industry. A specific factor is not of any use in other industries in the short run. A factor is specific if time or resources are needed to move it from one industry to another. Examples: High skill and low skill workers. Land. Machinery. MC (University of Cambridge) Trade Policy, Agreements and Taxation of Multinationals July 2015 33 / 63

The Specific Factors Model: Autarky Outline of the model 2 goods: Manufactures (M) and Food (F). 2 specific factors: Capital (K) used to make Manufactured goods and Land (T) used to make Food. 1 mobile factor: Labor (L) used in production of both goods. Labor markets clear. All workers work in equilibrium. L M + L F = L. MC (University of Cambridge) Trade Policy, Agreements and Taxation of Multinationals July 2015 34 / 63

The Specific Factors Model: Autarky CRS Technology: Q M = Q M (K, L M ), Q F = Q F (T, L F ). Diminishing marginal product of labor MPL = Q M (K,L M ) L M. MC (University of Cambridge) Trade Policy, Agreements and Taxation of Multinationals July 2015 35 / 63

The Specific Factors Model: Autarky The marginal product of labor in manufacturing output and labor space exhibits diminishing returns. The marginal product of labor in food is similar. MC (University of Cambridge) Trade Policy, Agreements and Taxation of Multinationals July 2015 36 / 63

The Specific Factors Model: Autarkic Equilibrium Define FOOD as the numeraire good, that is, define all relative prices in units of food rather than in dollars. P F 1 and P M P M PF With competitive markets, the value of the marginal product of labor is equal to the wage. The value marginal product of labor is the price*mpl. So, w = VMPL M = P M MPL M and w = VMPL F = P F MPL F. MC (University of Cambridge) Trade Policy, Agreements and Taxation of Multinationals July 2015 37 / 63

The Specific Factors Model: Autarkic Equilibrium Draw the value marginal product of labor curves in each sector subject to the labor endowment constraint. This shows the equilibrium wage (in terms of units of food) and the labor allocation. MC (University of Cambridge) Trade Policy, Agreements and Taxation of Multinationals July 2015 38 / 63

The Specific Factors Model: Autarkic Equilibrium In the Autarkic Equilirbrium, the relative price of manufactures to food is equal to the ratio of marginal products of labor. P M MPL M = P F MPL F P M = MPL F P F MPL M price ratio = slope of PPF. MC (University of Cambridge) Trade Policy, Agreements and Taxation of Multinationals July 2015 39 / 63

The Specific Factors Model: Autarkic Equilibrium The relative price (p a = P M PF ) is equal to the slope of the PPF! MC (University of Cambridge) Trade Policy, Agreements and Taxation of Multinationals July 2015 40 / 63

The Specific Factors Model: Free Trade Equilibrium Next, suppose there are two countries: Home and Foreign. The countries are identical in terms of preferences and technologies. The countries differ in that Home has a larger capital endowment: K > K. What does this imply about the PPF s of each country and autarkic prices? MC (University of Cambridge) Trade Policy, Agreements and Taxation of Multinationals July 2015 41 / 63

The Specific Factors Model: Free Trade Equilibrium Home has more capacity to produce manufactures. With identical preferences, the autarkic relative price of manufactures is greater in Foreign than in Home. p a > p a. (Abundant manufactures are cheap in Home). MC (University of Cambridge) Trade Policy, Agreements and Taxation of Multinationals July 2015 42 / 63

The Specific Factors Model: Free Trade Equilibrium Let p w be the relative price of manufactures to food in the world, i.e. in the free trade equilibrium. Note that p a < p w < p a, so the relative price of manufactures will increase in Home under free trade. What happens to aggregate welfare? MC (University of Cambridge) Trade Policy, Agreements and Taxation of Multinationals July 2015 43 / 63

The Specific Factors Model: Free Trade Equilibrium In the aggregate, there are gains from trade. The economy is consuming on a higher indifference curve. MC (University of Cambridge) Trade Policy, Agreements and Taxation of Multinationals July 2015 44 / 63

The Specific Factors Model: Free Trade Equilibrium For labor, the model s predictions for welfare are ambiguous. MC (University of Cambridge) Trade Policy, Agreements and Taxation of Multinationals July 2015 45 / 63

The Specific Factors Model: Free Trade Equilibrium Summarizing the predictions for workers: The wage rises in terms of food, but the increase in the (relative) wage is less that the increase in the relative price of manufactures. w (but by < P M ). Labor moves into manufacturing and out of food production: L M, L F. The amount of manufactures produced goes up and the amount of food produced falls: Q M, Q F. (Note: technology doesn t change when the economy opens to trade). The real wage declines in terms of manufactures and rises in terms of food. w P M, w P F Because of the ambiguous effects on real wages, it is unclear if labor is better or worse off. If we knew the utility function of workers, and how much they value consumption of manufactures versus food, we could make a definitive statement about the welfare of workers. MC (University of Cambridge) Trade Policy, Agreements and Taxation of Multinationals July 2015 46 / 63

The Specific Factors Model: Free Trade Equilibrium But what about the capitalists? Are they happier? Yes. MC (University of Cambridge) Trade Policy, Agreements and Taxation of Multinationals July 2015 47 / 63

The Specific Factors Model: Free Trade Equilibrium What do land owners think? Are they happier? No. MC (University of Cambridge) Trade Policy, Agreements and Taxation of Multinationals July 2015 48 / 63

The Specific Factors Model: Free Trade Equilibrium Decomposing the gains from trade: gains from exchange and gains from specialization. MC (University of Cambridge) Trade Policy, Agreements and Taxation of Multinationals July 2015 49 / 63

The Specific Factors Model: Free Trade Equilibrium Returning to the discussion from part 1 of this lecture, what insights into changes in the income distribution do we gain from the specific factors model? Under a trade liberalization, in the short run, the returns to fixed factors of production will differ. The fixed factor of production that was relatively abundant under autarky (but scarce globally) will experience an increase in its return. The fixed factor of production that was relatively scarce under autarky (but abundant globally) will experience a decrease in its return. For the mobile factor (labor), the real gains will depend critically on the utility function of workers. In the long run, we would expect investment in capital and a decline in the return to capital. Similarly, in the long run, land would lie fallow or be used to house factories so that the return to land would increase. MC (University of Cambridge) Trade Policy, Agreements and Taxation of Multinationals July 2015 50 / 63

Applying the Specific Factors Model: Offshoring From McLaren, Chapter 11 Outsourcing: The process whereby corporations procure inputs from several countries and allocate the tasks of production across several countries in order to minimize costs. AKA: Globalization of the production process or Fragmentation of the supply chain Example: the ipod American product, assembled in many countries of parts made in many countries. Offshoring (or sometimes outsourcing): a firm hires workers abroad for a task that used to be done domestically Examples: Maquiladoras - Mexican plants used by US firms; Indian call centers. What is the impact on workers in the country that outsources tasks? MC (University of Cambridge) Trade Policy, Agreements and Taxation of Multinationals July 2015 51 / 63

Applying the Specific Factors Model: Offshoring From McLaren, Chapter 11 Suppose we have two sectors, a Multinationals sector and a Non-multinationals sector. Each has sector-specific capital, but labor can move freely between them. This is the specific factors model we just learned. MC (University of Cambridge) Trade Policy, Agreements and Taxation of Multinationals July 2015 52 / 63

Applying the Specific Factors Model: Offshoring From McLaren, Chapter 11 MC (University of Cambridge) Trade Policy, Agreements and Taxation of Multinationals July 2015 53 / 63

Applying the Specific Factors Model: Offshoring From McLaren, Chapter 11 Assume that only multinationals are able to offshore tasks. What happens when it becomes easier to do offshoring? It depends on whether foreign labor is a substitute for domestic labor or Foreign labor is a complement with domestic labor. MC (University of Cambridge) Trade Policy, Agreements and Taxation of Multinationals July 2015 54 / 63

Applying the Specific Factors Model: Offshoring Foreign labor as a substitute for domestic labor MC (University of Cambridge) Trade Policy, Agreements and Taxation of Multinationals July 2015 55 / 63

Applying the Specific Factors Model: Offshoring Foreign labor as a substitute for domestic labor Multinationals domestic employment shrinks relative to non- multinational employment as offshoring occurs. Offshoring lowers the wages of Home workers. MC (University of Cambridge) Trade Policy, Agreements and Taxation of Multinationals July 2015 56 / 63

Applying the Specific Factors Model: Offshoring Foreign labor as a complement for domestic labor MC (University of Cambridge) Trade Policy, Agreements and Taxation of Multinationals July 2015 57 / 63

Applying the Specific Factors Model: Offshoring Foreign labor as a complement to domestic labor Multinationals domestic employment expands relative to non- multinational employment as offshoring occurs. Offshoring raises the wages of Home workers. MC (University of Cambridge) Trade Policy, Agreements and Taxation of Multinationals July 2015 58 / 63

Applying the Specific Factors Model: Offshoring Evidence on US Multinational Employment - Green is Parent Employment MC (University of Cambridge) Trade Policy, Agreements and Taxation of Multinationals July 2015 59 / 63

Applying the Specific Factors Model: Offshoring Evidence on US Multinational Employment - Green is Parent Employment MC (University of Cambridge) Trade Policy, Agreements and Taxation of Multinationals July 2015 60 / 63

Applying the Specific Factors Model: Offshoring Interpreting the evidence on US Multinational Employment Employment in US Multinational Parents trended up through the sample period. Employment in US overseas Affiliates of Multinationals also trended up. Over the sample period of 1988-2006, US Multinationals increase their domestic workforce by just as much as their foreign workforce, suggesting that the complements model is more appropriate for understanding offshoring. MC (University of Cambridge) Trade Policy, Agreements and Taxation of Multinationals July 2015 61 / 63

Summary of Lecture 1 In part 1 of this lecture, we... reviewed empirical evidence on increasing wage inequality around the world, reviewed the shape of wage distribution, discussed how to measure inequality, examined different theories behind the rise in wage inequality, and discussed recent evidence on inequality in Latin America. In part 2 of this lecture, we... reviewed the Ricardo-Viner Specific Factors Model, applied this model to the offshoring of jobs by Multinational Enterprises, reviewed evidence on offshoring by US Multinational Enterprises. MC (University of Cambridge) Trade Policy, Agreements and Taxation of Multinationals July 2015 62 / 63

Readings Levy Yeyati and Pienknagura, Wage compression and the decline in inequality in Latin America: Good or bad? at VoxEU.org, 10 June 2014. McLaren, International Trade, Chapter 11. MC (University of Cambridge) Trade Policy, Agreements and Taxation of Multinationals July 2015 63 / 63