International Conference on Financial Instability and Inequality in

International Conference on Financial Instability and Inequality in an Economically Integrated World Tsingua Sanya International Mathematics Forum (TSIMF) Thoughts on finance, commodities and inequality Esteban Pérez Caldentey esteban. perez@eclac. org

In the past decades the evolution of developed and developing economies is charactetized by three stilized facts • Rising inequality trends in terms of the indome distribution and functional distribution of income Ø The current levels of income inequality are the highest in over three decades and the most significant increase occurred during the ´Great Moderation. ´ OECD : GINI coefficient and ratio of the income of the highest to the lowest decile, 1985 -2013 10. 5 0. 33 10 0. 32 9. 5 9 0. 31 8. 5 8 0. 3 7. 5 7 0. 29 6. 5 6 1985 Income of the highest to the lowest decile (left axis) 1990 1995 2000 2005 2007 2008 Source : OECD (2014) GINI (Left Axis) 2010 2011 2012 0. 28

Inequality in most of the developing regions has also increased Gini coefficient (simple averages) 1990 -2010 Source: Alvaredo y Gasparini (2014). Handbook of Income Distribution 2

During the same period the wage share has declined for most developed economies EU 14 Others Europe North America Asia Developing countries Germany Austria Belgium Denmark Spain Finland Franve Ireland Italy Luxemburg Netherlands Portugal United Kingdom Sweden Island Norway Switzerland Canada United States Australia South Korea 1 Japan New Zealand México Average 1980 -89 … 61. 2 63. 0 58. 9 62. 3 62. 5 62. 7 61. 1 54. 3 65. 7 60. 8 60. 3 51. 6 … 53. 9 … 58. 4 61. 0 … 70. 1 … … Source: AMECO. Notes: 1/ 2012 last year available Wage share as percentage of GDP Average Avreage Average 1990 -99 2000 -07 2008 -09 59. 1 56. 8 55. 7 58. 8 54. 5 54. 6 61. 5 59. 7 60. 5 56. 1 54. 9 58. 0 60. 5 56. 8 58. 3 58. 2 52. 9 55. 3 57. 0 55. 8 56. 5 55. 3 46. 7 54. 0 55. 3 51. 7 53. 5 50. 6 51. 8 54. 1 62. 3 58. 5 59. 2 58. 9 57. 1 58. 3 59. 5 60. 3 49. 1 48. 1 49. 2 … 64. 1 57. 8 51. 4 46. 1 46. 9 66. 1 65. 2 64. 2 58. 3 55. 3 56. 0 60. 4 59. 8 58. 5 58. 3 55. 5 53. 5 69. 7 65. 1 62. 9 65. 8 61. 3 59. 8 46. 6 45. 5 49. 0 40. 7 41. 0 39. 1 Average 2010 -15 56. 4 55. 4 60. 5 56. 7 55. 6 56. 5 57. 5 49. 7 53. 5 53. 1 59. 7 54. 2 59. 2 49. 8 58. 6 47. 8 65. 5 55. 8 56. 8 53. 3 59. 9 59. 4 46. 1 36. 9

At the same time the financial sector has experienced significant growth • This is reflected by the growth of the financial sector relative to GDP on a national accounting basis which allows a comparative basis for different countries • The growth in the financial sector is observed for both developed and developing and emerging market economies. Rate of growth of the financial sector and GDP for selected regions/countries of the world 1980 -2005 9. 5 10 9 8. 0 8 7 6. 8 6 5 4 3 3. 6 2. 9 2. 7 2 1 0 United States Source: On the basis of Beck (2012) Financial Sector OECD GDP Developing countries

At the global level the expansion of finance can be illustrated by the change in the stock of financial assets Financialization. World GDP, stock of global financial assets, derivatives, 1980 -2014 800 711 692 700 648 604 598 586 600 636 601 500 418 400 299 300 242 211 262 285 294 178 200 108 95 100 0 222 243 272 51 11 12 1 1980 23 13 1990 33 47 51 GDP (US$ Billion) 2000 Worlrd 2005 2006 Source: On the basis of Deutsche Bank (2014) 57 63 60 65 Stock financial assets 2007 of global 2008 2009(US$ billion)) 2010 73 Derivatives 2011 74 2012 76 2013 78 2014

Studies analyzing the relationship between these three stylized facts tend to focus on the causality from income inequality to finance Income inequality (Personal income) Inequality (functional distribution of income) Finance • Four transmission mechanisms Ø Downward pressure on aggregate demand (demand is wage-led) Ø Financial deregulation losens the external constraint opening the way to debt-led growth Ø Higher inequality contributes to household debt Ø Rising inequality increases the propensity to speculate Source: Goda, et al. (2014), Rajan (2010), Stockhammer (2013), Perugi et al. (2016).

But it is also important to consider how finance can also ´cause´inequality • The chain of causality from finance to inequality is relevant to both developed and developing economies. • In the case of developing economies this relationship is a central obstacle to development. • Three transmission mechanisms can be distinguished Ø Impact of financial crisis ü Crisis tend to have permanent effects on the economy and the distribution of income. Ø Labor market (wage disparity) ü In both developed and developing economies financial sector employees are concentrated at the top of the income distribution and their earnings exceed those of other employees with simila rprofiles (age, gender or education) (OECD, 2015). ü Induces the financial sector to act as a rent seeker. Ø The production structure

Selected empirical analyses on the relationship between finance and income distribution Authors Sample Findings Beck et al. (2007); Clarke et al. Developed, emerging and Faster levels of financial (2006) developing economies development are accompanied by faster reductions in income inequality and poverty Jauch and Watzka (2012); 49 advanced and middle Faster levels of financial Gimet and Lagoarde-Segot income countries development are not (2011) accompanied by faster reductions in income inequality and poverty Fournier and Koske (2012) Micro data set of OECD countries Bahmani-Oskooee and Zhang (2015) OECD (2015) Greater share of financial activities in GDP translate into greater earnings dispersion The positive impact of finance on the distribution of income is transitory OECD countries 1974 -2011 Higher levels of credit intermediation and stock markets are related with a more unequal distribution of income

Since the 1980´s LAC´s rate of growth has remained stuck at 2%. Rate of GDP per capita growth for Latin America and the Caribbean and East Asia and the Pacific (1962 -2014) (In percentages) 7 5. 9 6 4. 9 5 4 3 2 6. 1 5. 4 East Asia and the Pacific 4. 3 3. 4 3. 7 3. 0 2. 5 2. 1 2. 0 Latin America and the Caribbean 1 0 Asia del Este y Pacífico América Latina y el Caribe 1962 -69 1962 -79 1962 -89 1962 -99 1962 -09 1962 -14

Some crises have permanent effects on output and the distribution of income Distribution of the rate of growth of real GDP for Latin America and the Caribbean before and after the Debt Crisis (1980 -1983) Two-sample Kolmogorov-Smirnov test for equality of distribution functions Smaller group 1 2 Combined K-S Source: On the basis of World Bank development Indicators (2015) D 0. 0 -0. 87 P-value 1. 0 0. 0

Inequality in Latin America has declined since 2002 Gini coefficient for Latin America (18 countries), 1980 -2002 Source: Cornia y Martonano (2012); Amarante (2014)

Given the structure of financial markets increases in the rate of pro-cyclical policies that follow crises tend to increase inequality Interest Rate Notional credit demand Actual credit demand C D A Commercial bank interest rate Central bank interest rate Illiquidity B and risk Loans Source: Wolfson (1996)

United States Wages in the financial sector relative to non -farm pr. Ivate sector, 1929 -2006 Securities, commodities, venture capital, private equity, hedge funds, investment banks OECD Share of financial sector employees among the top 10%, 1% and 0. 1% earners, 2010 Source: Philippon and Reshef, 2010; OECD (2015)

The financial sector is an important extractor of rents • Mainstream macro-financial models include explicitly or implicitly a financial intermediation equation • Intermediation costs are supposed to reflect the compensation financial intermediaries receive for pooling funds, share risks, transfer resources, produce information and provide incentives. • Selected evidence for the US shows that the intermediation cost has increased. Ø The increase does not reflect better risk sharing or production information Ø Improvements in information technology should have reduced the intermediation costs. Ø Why is the non-financial sector transfering increasing income to the financial sector? (Philippon, 2012). • Other wide ranging studies for Europe show that the increase in the profits accounts for the bulk of the rise in the value added in the financial sector. Ø Between 1970 and 2010, the profit share of the income generated by the financial sector increased from 42% to 47%.

The financial sector is an important extractor of rents • Mainstream macro-financial models include explicitly or implicitly a financial intermediation equation • Intermediation costs are supposed to reflect the compensation financial intermediaries receive for pooling funds, share risks, transfer resources, produce information and provide incentives. • ”How is it possible for today´s financial industry not to be significantly more efficient Selected evidence for the US shows that the intermediation cost has increased. Ø The increase does not reflect better risk sharing or production information than the finance industry of John Pierpont Ø Improvements in information technology should have reduced the intermediation costs. Morgan a century ago? ” Ø Why is the non-financial sector transfering increasing income to the financial sector? (Philippon, 2012). • Other wide ranging studies for Europe show that the increase in the profits accounts for the bulk of the rise in the value added in the financial sector. Ø Between 1970 and 2010, the profit share of the income generated by the financial sector increased from 42% to 47%.

Finance can also ćauseínequality through its impact on the production structure Finance Inequality (functional distribution of income ) Inequality (personal distribution of income) • The chain of causality is the reverse the íncome inequality to finance approach. ´ • The finance to inequality view can be illustrated with the case of commodities. Ø Provides a sense of the complexity of the interrelationship between real and monetary factors. Ø Shows how financial factors can dominate real factors. Ø Shows how finance can reproduce both social and productive structures.

Commodities have taken on an increasing role as financial assets in the sense that prices respond to changes in expectations about future demand conditions • Some of the manifestations of the growing role of commodities as financial assets include: Ø The growth in activity in commodity future markets including commodity derivatives. ü Commodity futures trading as well as the derivative industry expanded significantly around the time of the burst of the bubble dot com and the ensuing stock market crash in the early 2000 s. ü Highly liquid-low interest environment that prevailed before the Global Financial Crisis combined with the high rate of return of commodity futures relative to equities provided an incentive to invest in commodities. o Between January 31, 2004 and June, 2008, commodity futures’ rate of return (19. 5%) more than doubled that of equity (6. 0%) (Bhardwaj, 2010). ü Low rate interest environment after the Global Financial Crisis, Ø The strengthening of the co-movement among different commodity prices and between commodities and stock markets. Ø The use of commodities as collaterals for loans and credit.

Percentage of statistically significant correlations between returns and volatilities of commodity prices indices and stock markets Source: On the basis of Bloomberg (2015)

Another recent illustration of the consideration of commodities as a financial asset is the use of commodities as collaterals in financing deals to raise and invest in liquidity Scatter plot between rate of return on copper and inventories (July 1997 -September 2014) Source: On the basis of bloomberg (2015)

Another recent illustration of the consideration of commodities as a financial asset is the use of commodities as collaterals in financing deals to raise and invest in liquidity Scatter plot between rate of return crude oil and inventories (January 1990 and September 2014) Source: On the basis of bloomberg (2015)

United States Senate Permanent Subcommittee on Investigations state in its report on Wall Street Bank involvement with Physical Commodities (November, 2014, p. 3): • “Until recently, Morgan Stanley controlled over 55 million barrels of oil storage capacity, 100 oil tankers, and 6, 000 miles of pipeline. JPMorgan built a copper inventory that peaked at $2. 7 billion, and, at one point, included at least 213, 000 metric tons of copper, comprising nearly 60% of the available physical copper on the world’s premier copper trading exchange, the LME. In 2012, Goldman owned 1. 5 million metric tons of aluminum worth $3 billion, about 25% of the entire U. S. annual consumption. Goldman also owned warehouses which, in 2014, controlled 85% of the LME aluminum storage business in the United States. Those large holdings illustrate the significant increase in participation and power of the financial holding companies active in physical commodity markets. ”

The commodity spot-future price disconnect Bloomberg commodity spot and (3 month) future index 1991 -2016 (Daily) 600 Index 1991 -100 500 400 300 BCOM SPOT BCOM 3 MF 200 100 Pre-Global Financial Crisis Post-Global Financial Crisis 93 19 1/ 94 2/ 19 1/ 95 2/ 19 1/ 96 2/ 19 1/ 97 2/ 19 1/ 98 2/ 19 1/ 99 2/ 20 1/ 00 2/ 20 1/ 01 2/ 20 1/ 02 2/ 20 1/ 03 2/ 20 1/ 04 2/ 20 1/ 05 2/ 20 1/ 06 2/ 20 1/ 07 2/ 20 1/ 08 2/ 20 1/ 09 2/ 20 1/ 10 2/ 20 1/ 11 2/ 20 1/ 12 2/ 20 1/ 13 2/ 20 1/ 14 2/ 20 1/ 15 2/ 20 16 1/ 2/ 19 92 1/ 2/ 19 2/ 1/ 1/ 2/ 19 91 0 Source: On the basis of bloomberg (2015); Choski (1984).

3. 5 Chile: profit share and natural resource rents as % of GDP in logarithms (1985 -2011) 4. 1 4. 05 3 4 2. 5 3. 95 2 1. 5 Pinochet Aylwin 1 3. 9 Lagos Frei Bachelet 0. 5 3. 8 3. 75 3. 7 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 0 log natural resource (left axis) log profit share (right axis) Commodity boom 6. 5 4. 1 6 4. 05 5. 5 4 5 3. 95 4. 5 3. 9 4 3. 85 3. 5 3 3. 8 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 Chile: profit share as % of GDP and copper prices in logarithms (1985 -2011) log B. M. L. refined copper price nominal log profit share (right axis)

Salarios/PIB 60 55 50 45 40 35 30 25 20 Salarios/PIB Colombia 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 60 55 50 45 40 35 30 25 20 Chile 60 55 50 45 40 35 30 25 20 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 60 55 50 45 40 35 30 25 20 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 Salarios/PIB The share of wages has also declined in other natural resources based economies Bolivia Perú

From the functional distribution of income to personal income inequality Investment financed by retained earnings Capital income Wage income

From the functional distribution of income to personal income inequality Finance Productive structure Investment financed by retained earnings Capital income Personal income distribution Wage income Reproduces de social structure