ACCOUNTING FOR COINCIDENT GROWTH COLLAPSES BRAZIL AND MEXICO

ACCOUNTING FOR COINCIDENT GROWTH COLLAPSES: BRAZIL AND MEXICO SINCE THE EARLY 1980 s Edmar Bacha and Regis Bonelli Conference in Honor of Albert Fishlow Casa das Garças, Rio de Janeiro 07/03/2014

SUMMARY • Growth collapses: a periodization • Capital accumulation and the growth collapses • Decomposing labor productivity growth • Brazil’s TFP’ and the terms of trade • Productivity growth and structural heterogeneity (by region, tradability, firm size, and formality)

GDP per capita relative to the US: Convergence to 1980 -81, divergence afterward (2014 US$, PPPs) 0. 50 BRA / US MEX / US 0. 45 0. 40 0. 35 0. 30 2014 2012 2010 2008 2006 2004 2002 2000 1998 1996 1994 1992 1990 1988 1986 1984 1982 1980 1978 1976 1974 1972 1970 1968 1966 1964 1962 1960 1958 1956 1954 1952 0. 20 1950 0. 25

0% 2014 2012 2010 2008 2006 MEXICO 2004 2002 2000 1998 1996 1994 1992 8% 1990 9% 1988 1986 1984 1982 1980 1978 1976 1974 1972 1970 1968 1966 1964 1962 1960 1958 1956 1954 1952 1950 10% Mexico and Brazil — 10 -year moving averages of GDP growth rates, 1950 -2014 (% a. a. ) BRAZIL 7% 6% 5% 4% 3% 2% 1%

Growth periodization, 1950 -2014 (% a. a. ) Periods Brazil Mexico Brazil's average Mexico's average GDP growth Post WW-II Golden Age 1950 -1980 1950 -1981 7. 4% 6. 8% Post-1980 Near Stagnation 1981 -2014 1982 -2014 2. 6% 2. 2% Lost Decade 1981 -1992 1982 -1993 1. 4% 1. 6% Reforms with Subpar Growth 1993 -2003 1994 -2001 2. 8% 3. 0% 2004 -2010 2002 -2010 4. 5% 1. 9% Day after the Great Recession 2011 -2014 2. 1% 2. 9% China Syndrome

0% -2% -4% -6% 2014 2012 2010 2008 2006 2004 K' 2002 2000 1998 1996 1994 1992 12% 1990 1988 1986 1984 1982 1980 1978 1976 1974 1972 1970 1968 1966 1964 1962 1960 1958 1956 1954 1952 1950 Brazil — Capital stock and GDP growth rates, 1950 -2014 (%) 14% Y' 10% 8% 6% 4% 2%

0% -2% -4% -6% 2014 2012 2010 2008 2006 2004 K' 2002 2000 1998 1996 1994 1992 10% 1990 1988 1986 1984 1982 1980 1978 1976 1974 1972 1970 1968 1966 1964 1962 1960 1958 1956 1954 1952 1950 Mexico — Capital stock and GDP growth rates, 1950 -2014 (%) 12% Y' 8% 6% 4% 2%

Decomposition of the capital stock growth rate • Start from: • • Pi I = S, Where: Pi is the implicit price deflator of gross capital formation, I is gross real investment, and S is total savings (domestic+foreign) in current prices. Inventory changes are assumed equal to zero. Divide both sides by Pi. K (where K is the capital stock): I/K = S/Pi. K Divide and multiply the right-hand side by Py. Y (where Py is the implicit price deflator of GDP and Y is real GDP): I/K = (S/Py. Y)(Py/Pi)(Y/K) Subtract the capital stock depreciation rate (δ) from both sides, and rearrange to obtain: K' = s(1/p)v – δ • Where: K’ = I/K – δ, s = S/Py. Y, and p = Pi/Py

Brazil: Decomposition of K’, 1950 -2014 BRAZIL Post WW-II Golden Age Lost Decade Reforms with Subpar Growth China Syndrome Day after the Great Recession Post-1980 Near Stagnation Years K' s v p δ* 1950 -1980 8. 8% 19. 4% 0. 506 0. 784 3. 6% 1981 -1992 3. 3% 20. 9% 0. 357 1. 009 4. 1% 1993 -2003 2. 1% 18. 3% 0. 352 1. 013 4. 2% 2004 -2010 2. 8% 18. 5% 0. 382 1. 024 4. 1% 2011 -2014 4. 0% 20. 3% 0. 383 0. 973 4. 0% 1981 -2014 2. 9% 19. 5% 0. 364 1. 009 4. 1%

Mexico: Decomposition of K’, 1950 -2014 MEXICO Post WW-II Golden Age Lost Decade Reforms with Subpar Growth China Syndrome Day after the Great Recession Post-1980 Near Stagnation Years 1950 -1981 1982 -1993 1994 -2001 2002 -2010 2011 -2014 1982 -2014 K' s v p* δ ** 8. 0% 16. 4% 0. 656 0. 795 5. 6% 3. 3% 17. 3% 0. 470 0. 915 5. 6% 3. 7% 17. 5% 0. 443 0. 842 5. 5% 3. 5% 22. 2% 0. 410 0. 967 6. 0% 2. 8% 21. 5% 0. 391 0. 919 6. 3% 3. 4% 18. 5% 0. 438 0. 887 5. 8%

Brazil: Decomposition of output growth per worker y' = αk' + TFP’ Decomposition Brazil 1951 -80 1981 -92 1993 -03 2004 -10 2011 -14 1981 -14 y' 4. 2% -0. 8% 0. 3% 2. 2% 1. 1% 0. 4% L' alfa*k' TFP' 3. 1% 2. 5% 1. 7% 2. 2% 0. 7% -1. 4% 2. 5% -0. 2% 0. 4% 0. 5% 0. 2% 2. 0% 1. 3% -0. 2% 2. 2% 0. 4% 0. 0%

Mexico: Decomposition of output growth per worker y' = αk' + TFP’ y' L' 1951 -1981 3. 4% 3. 2% 2. 1% 1982 -1993 -1. 7% 3. 4% 0. 1% -1. 8% 1994 -2001 0. 7% 2. 3% 0. 6% 2002 -2010 0. 3% 1. 6% 0. 9% -0. 5% 2011 -2014 1. 0% 1. 9% 0. 4% 1982 -2014 -0. 2% 2. 5% 0. 5% -0. 7% Decomposition Mexico alfa*k' TFP' 1. 3% 0. 1% 0. 6%

Brazil — TFP’ and terms of trade (To. T) rate of change, 1981 -2014 6% 30% TFP per cent change To. T rate of change (right hand side axis) 4% 25% 20% 2% -4% 2014 2013 2012 2011 2010 2009 2008 2007 2006 2005 2004 2003 2002 2001 2000 1999 1998 1997 1996 1995 1994 1993 1992 1991 1990 1989 1988 1987 1986 1985 1984 1983 -2% 1982 0% 1981 15% 10% 5% 0% -5% -6% -10% -8% -10% -15% -20%

Brazil’s TFP’, terms of trade, and business cycles [IV* regression, 1981 -2014] Dep. variable: TFP' 34 obs. (1981 -2014) Adjusted R-squared = 0. 426 Coefficients St. Error Stat t P-value Intercept 2. 85 0. 92 3. 10 0. 004 Relative change To. T Capacity use gap (%)* 0. 18 0. 07 2. 81 0. 008 -0. 42 0. 15 -2. 72 0. 011

Regional dispersion of per capita incomes, Brazil and Mexico, 1989(90)-2011(13) (sigma=s. d. States’ per capita incomes/unweighted mean income)

Bra trade /nontrad Mex trade /nontrad 2009 2008 2007 2006 2005 2004 2003 2002 2001 2000 1999 1998 1997 1996 1995 1994 1993 1992 1991 1990 1. 8 1. 7 1. 6 1. 5 1. 4 1. 3 1. 2 1. 1 1. 0 0. 9 0. 8 0. 7 1989 Labor productivity ratios: Traded/non-traded relationship

Productivity growth by firm size (manufacturing, commerce and services) - Mexico, 1998 -2008 Firm size Mexico: Productivity growth 1998 -2008 (% a. a. ) 0 - 10 -6. 5% 11 -30 -2. 2% 30 - 100 0. 2% 101 - 250 2. 9% 251 - 500 2. 4% 501 + 5. 9% Total 2. 0%

Productivity growth by firm size (manufacturing) - Brazil, 1996 -2007 [incomplete] Firm size Productivity Productivty 1996 Brazil: Productivity 2007 1996 (2007 prices) growth 1996 -2007 0 - 10 3. 153 1182. 6 2967 0. 6% 10 - 29 168 33. 9 85 6. 4% 30 - 99 92 36. 5 92 0. 0% 100 - 249 156 55. 6 139 1. 0% 250 - 499 219 67. 6 170 2. 3% 500 + 325 100. 4 252 2. 3% Total 242 76. 6 192 2. 1%

50 40 Baja California Sur Nuevo León Chihuahua Coahuila de Zaragoza Baja California Sonora Distrito Federal Querétaro Aguascalientes Tamaulipas Quintana Roo Sinaloa Colima Jalisco Durango México 60 San Luis Potosí 70 Nacional Campeche Tabasco Guanajuato Yucatán Morelos Zacatecas Michoacán de Ocampo Veracruz de Ignacio de la Llave Nayarit Puebla Tlaxcala Hidalgo Chiapas 80 Guerrero 90 Oxaca Informality rates per State. Mexico, 2012 Taxa de emprego informal nas unidades federativas, 2012 - México. 80. 0 59. 6 42. 0 30 20 10 0

0 40 30 Distrito Federal São Paulo Santa Catarina Rio de Janeiro Paraná Mato Grosso do Sul Rio Grande do Sul Minas Gerais Mato Grosso Espírito Santo Goiás 50 Nacional Amapá Rondônia Pernambuco Roraima Rio Grande do Norte Alagoas Sergipe Tocantins Amazonas Acre Bahia Paraíba Ceará Piauí 80 Maranhão Informality rates per State. Brazil, 2012 Taxa de emprego informal por estado, 2012 - Brasil. 90 76. 5 70 60 47. 6 30. 0 20 10

Regression for States’ informality rates, Brazil and Mexico (2012) Dep. Var. : Informality Rate (%) R-square adjusted 0. 730 Standard error 6. 208 Observations 59 27 BRA, 32 MEX Coefficients Standard Error Stat t P-value Intercept 266. 8 17. 8 15. 0 0. 00% log (GDP pc PPP) -22. 8 1. 9 -12. 0 0. 00% Dummy MEX 10. 5 1. 7 6. 3 0. 00% Dummy Campeche 50. 0 7. 4 6. 7 0. 00%

Labor informality rates, Brazil and Mexico, Selected years (%)

Summary (macro) • Close association between GDP and K-growth collapses since the 1980 s • K-growth collapses not due to savings: culprits were lower outputcapital ratios and higher relative prices of investment • Near-secular stagnation of labor productivity is explained more or less equally by less capital deepening and TFP • Terms of trade and business cycle são important determinants of Brazil’s measured TFP

Summary (meso) • Significant “mesoeconomic” diferences: • States’ per capita incomes dispersion increased in Mexico but decreased in Brazil • Traded sector productivity is higher and increases much faster than non-traded sector productivity in Mexico. In Brazil traded sector productivity is lower but catchs up with non-traded sector productivity • Large firms’ productivity increases much faster than small and medium sized firms in Mexico. In Brazil, large firms’ productivity increases at the average rate for all firms • Informality is higher in Mexico than in Brazil, despite a higher per capita income. Informality remained constant in Mexico while in Brazil it declined substantially

Summary (overall) • Mexico opened up its economy to trade and succeeded in developing a firstclass industrial sector in the country’s richer Northern region. A similar domestic economic integration didn’t occur. The dynamism of the large exporting firms in the North did not feed back to the non-traded, informal, small and medium-sized firms in the country’s poorer Southern regions. The consequence was a very low labor productivity growth rate • Brazil succeeded in reducing economic polarization in several dimensions. The problem was that in contrast to Mexico’s her high-productivity largemanufacturing firms did not integrate into the world economy and thus saw their productivity growth slow down. This provided a weak lever to move the rest of the economy up, and the country lingered on in a low overall productivity path, except when the commodity lottery dictated otherwise. • Bottom line: integrate it both ways, domestically and internationally!