Successfully reported this slideshow.
We use your LinkedIn profile and activity data to personalize ads and to show you more relevant ads. You can change your ad preferences anytime.

Wage Inequality and Structural Change

39 views

Published on

Wage Inequality and Structural Change

Published in: Economy & Finance
  • Be the first to comment

  • Be the first to like this

Wage Inequality and Structural Change

  1. 1. Wage Inequality and Structural Change Joanna Tyrowicz (GRAPE, IAAEU, UW and IZA ) Magdalena Smyk (GRAPE and WSE) Statistics Poland, Annual Congress, Warsaw, 2018 1
  2. 2. Motivation Income inequality is a major policy concern • Theory: inequality should ↑ due to globalization and skill biased technological change Feenstra and Hanson 1996; Acemoglu and Autor 2011 2
  3. 3. Motivation Income inequality is a major policy concern • Theory: inequality should ↑ due to globalization and skill biased technological change Feenstra and Hanson 1996; Acemoglu and Autor 2011 • Transition countries are an interesting case, because of great inequality ↑ Milanovic 1999; Brainerd 2000 2
  4. 4. Motivation Income inequality is a major policy concern • Theory: inequality should ↑ due to globalization and skill biased technological change Feenstra and Hanson 1996; Acemoglu and Autor 2011 • Transition countries are an interesting case, because of great inequality ↑ Milanovic 1999; Brainerd 2000 2
  5. 5. Motivation Income inequality is a major policy concern • Theory: inequality should ↑ due to globalization and skill biased technological change Feenstra and Hanson 1996; Acemoglu and Autor 2011 • Transition countries are an interesting case, because of great inequality ↑ Milanovic 1999; Brainerd 2000 But there are substantial methodlogical challenges • Majority of studies look at post-redistribution household (equivalized) income Milanovic and Ersado 2012; Aristei and Perugini 2012, 2014 2
  6. 6. Motivation Income inequality is a major policy concern • Theory: inequality should ↑ due to globalization and skill biased technological change Feenstra and Hanson 1996; Acemoglu and Autor 2011 • Transition countries are an interesting case, because of great inequality ↑ Milanovic 1999; Brainerd 2000 But there are substantial methodlogical challenges • Majority of studies look at post-redistribution household (equivalized) income Milanovic and Ersado 2012; Aristei and Perugini 2012, 2014 • Cannot account for individual characteristics 2
  7. 7. Motivation Income inequality is a major policy concern • Theory: inequality should ↑ due to globalization and skill biased technological change Feenstra and Hanson 1996; Acemoglu and Autor 2011 • Transition countries are an interesting case, because of great inequality ↑ Milanovic 1999; Brainerd 2000 But there are substantial methodlogical challenges • Majority of studies look at post-redistribution household (equivalized) income Milanovic and Ersado 2012; Aristei and Perugini 2012, 2014 • Cannot account for individual characteristics 2
  8. 8. Motivation Income inequality is a major policy concern • Theory: inequality should ↑ due to globalization and skill biased technological change Feenstra and Hanson 1996; Acemoglu and Autor 2011 • Transition countries are an interesting case, because of great inequality ↑ Milanovic 1999; Brainerd 2000 But there are substantial methodlogical challenges • Majority of studies look at post-redistribution household (equivalized) income Milanovic and Ersado 2012; Aristei and Perugini 2012, 2014 • Cannot account for individual characteristics −→ Develop measures of earned income inequality, adjusting for individual characteristics 2
  9. 9. Motivation Income inequality is a major policy concern • Theory: inequality should ↑ due to globalization and skill biased technological change Feenstra and Hanson 1996; Acemoglu and Autor 2011 • Transition countries are an interesting case, because of great inequality ↑ Milanovic 1999; Brainerd 2000 But there are substantial methodlogical challenges • Majority of studies look at post-redistribution household (equivalized) income Milanovic and Ersado 2012; Aristei and Perugini 2012, 2014 • Cannot account for individual characteristics −→ Develop measures of earned income inequality, adjusting for individual characteristics Question: is structural change conducive to growth in wage inequality? 2
  10. 10. Data
  11. 11. Collecting source data • standardized data: (EU)SES, ECHP 3
  12. 12. Collecting source data • standardized data: (EU)SES, ECHP • standard methodology data: LSMS, census, ISSP 3
  13. 13. Collecting source data • standardized data: (EU)SES, ECHP • standard methodology data: LSMS, census, ISSP • LFS and HBS 3
  14. 14. Collecting source data • standardized data: (EU)SES, ECHP • standard methodology data: LSMS, census, ISSP • LFS and HBS • ULMS, RLMS, GSOEP 3
  15. 15. Collecting source data • standardized data: (EU)SES, ECHP • standard methodology data: LSMS, census, ISSP • LFS and HBS • ULMS, RLMS, GSOEP 3
  16. 16. Collecting source data • standardized data: (EU)SES, ECHP • standard methodology data: LSMS, census, ISSP • LFS and HBS • ULMS, RLMS, GSOEP −→ Overall 1650+ datasets, 800+ with wage data 3
  17. 17. Match between the OECD and our data 4
  18. 18. The measures • Entropy measures: Gini coefficient and mean log deviation 5
  19. 19. The measures • Entropy measures: Gini coefficient and mean log deviation • Positional measures: 9th to 1st decile ratio + top (9th:5th) and bottom (1st:5th) −→ raw inequality measures from individual wage data 5
  20. 20. The measures • Entropy measures: Gini coefficient and mean log deviation • Positional measures: 9th to 1st decile ratio + top (9th:5th) and bottom (1st:5th) −→ raw inequality measures from individual wage data • Inequality may change because 5
  21. 21. The measures • Entropy measures: Gini coefficient and mean log deviation • Positional measures: 9th to 1st decile ratio + top (9th:5th) and bottom (1st:5th) −→ raw inequality measures from individual wage data • Inequality may change because • individuals with the same characteristics get paid differently 5
  22. 22. The measures • Entropy measures: Gini coefficient and mean log deviation • Positional measures: 9th to 1st decile ratio + top (9th:5th) and bottom (1st:5th) −→ raw inequality measures from individual wage data • Inequality may change because • individuals with the same characteristics get paid differently • individual characteristics change (services, educational boom, etc.) 5
  23. 23. The measures • Entropy measures: Gini coefficient and mean log deviation • Positional measures: 9th to 1st decile ratio + top (9th:5th) and bottom (1st:5th) −→ raw inequality measures from individual wage data • Inequality may change because • individuals with the same characteristics get paid differently • individual characteristics change (services, educational boom, etc.) • Counterfactual distributions 5
  24. 24. The measures • Entropy measures: Gini coefficient and mean log deviation • Positional measures: 9th to 1st decile ratio + top (9th:5th) and bottom (1st:5th) −→ raw inequality measures from individual wage data • Inequality may change because • individuals with the same characteristics get paid differently • individual characteristics change (services, educational boom, etc.) • Counterfactual distributions • Parametric (linear regression): works well at the mean 5
  25. 25. The measures • Entropy measures: Gini coefficient and mean log deviation • Positional measures: 9th to 1st decile ratio + top (9th:5th) and bottom (1st:5th) −→ raw inequality measures from individual wage data • Inequality may change because • individuals with the same characteristics get paid differently • individual characteristics change (services, educational boom, etc.) • Counterfactual distributions • Parametric (linear regression): works well at the mean • Semi-parametric (DiNardo, Fortin and Lemieux, 1996): works well at tails 5
  26. 26. The measures • Entropy measures: Gini coefficient and mean log deviation • Positional measures: 9th to 1st decile ratio + top (9th:5th) and bottom (1st:5th) −→ raw inequality measures from individual wage data • Inequality may change because • individuals with the same characteristics get paid differently • individual characteristics change (services, educational boom, etc.) • Counterfactual distributions • Parametric (linear regression): works well at the mean • Semi-parametric (DiNardo, Fortin and Lemieux, 1996): works well at tails • Benchmark: the US economy 5
  27. 27. Results
  28. 28. Overall time trends: Gini coefficient 6
  29. 29. Overall time trends: Gini coefficient 7
  30. 30. Overall time trends: 9th to 1st decile ratio 8
  31. 31. Overall time trends: Gini coefficient Monthly wage - Hourly wage - Hourly wage - original Parametric ACS2000 Parametric ACS2010 OLS RE FE RE FE RE FE 9th-to-1st Transition 0.243 0.006 -0.183*** -0.188*** (0.167) (0.111) (0.048) (0.055) Time 0.005 0.004 0.005 -0.006 -0.012 -0.004 -0.008 (0.030) (0.023) (0.023) (0.014) (0.015) (0.016) (0.017) Transition#Time -0.065 -0.052* -0.052* 0.075*** 0.078*** 0.071*** 0.071*** (0.042) (0.031) (0.031) (0.020) (0.020) (0.022) (0.023) Gini Index Transition 0.130*** 0.065*** -0.033*** -0.033*** (0.040) (0.017) (0.008) (0.008) Time 0.015*** 0.014*** 0.012** -0.0002 -0.0008 -0.001 -0.001 (0.005) (0.005) (0.005) (0.002) (0.003) (0.002) (0.003) Transition#Time -0.0242*** -0.0254*** -0.0242*** 0.012*** 0.012*** 0.012*** 0.012*** (0.007) (0.006) (0.006) (0.004) (0.004) (0.003) (0.004) Mean Log Devation Transition 0.078*** 0.077*** -0.029*** -0-.024*** (0.0201) (0.022) (0.008) (0.007) Time 0.027*** 0.018*** 0.014** 0.0003 0.001 0.0003 0.001 (0.008) (0.006) (0.006) (0.002) (0.003) (0.002) (0.002) Transition#Time -0.04*** -0.037*** -0.034*** 0.004 0.003 0.004 0.002 (0.011) (0.009) (0.009) (0.003) (0.003) (0.003) (0.003) Obs. 548 548 548 418 418 418 418 Countries 31 31 31 30 30 30 30 9
  32. 32. Time trends: summary • Essentially, immediate adjustment in wage inequality in transition countries −→ decompression of wages srtong despite institutional arrangements 10
  33. 33. Time trends: summary • Essentially, immediate adjustment in wage inequality in transition countries • Growth in inequality mostly due to decompressing lower half of the earnings distribution −→ decompression of wages srtong despite institutional arrangements 10
  34. 34. Time trends: summary • Essentially, immediate adjustment in wage inequality in transition countries • Growth in inequality mostly due to decompressing lower half of the earnings distribution • Very slow negative trend towards lower earnings dispersion −→ decompression of wages srtong despite institutional arrangements 10
  35. 35. Time trends: summary • Essentially, immediate adjustment in wage inequality in transition countries • Growth in inequality mostly due to decompressing lower half of the earnings distribution • Very slow negative trend towards lower earnings dispersion • Counterfactual distributions more compressed in transition countries −→ decompression of wages srtong despite institutional arrangements 10
  36. 36. Time trends: summary • Essentially, immediate adjustment in wage inequality in transition countries • Growth in inequality mostly due to decompressing lower half of the earnings distribution • Very slow negative trend towards lower earnings dispersion • Counterfactual distributions more compressed in transition countries • Convergence towards similar characteristics −→ decompression of wages srtong despite institutional arrangements 10
  37. 37. Time trends: summary • Essentially, immediate adjustment in wage inequality in transition countries • Growth in inequality mostly due to decompressing lower half of the earnings distribution • Very slow negative trend towards lower earnings dispersion • Counterfactual distributions more compressed in transition countries • Convergence towards similar characteristics • No convergence towards similar wages −→ decompression of wages srtong despite institutional arrangements 10
  38. 38. Structural change Table 1: Wage compression and the indicators of structural change 5th-to-1st 9th-to-5th Raw Parametric DFL Raw Parametric DFL High-skilled Neg 0 Neg Neg Neg Neg #Transition 0 0 Neg Neg Neg 0 Obs. 330 289 289 289 330 289 Countries 26 25 25 25 26 25 11
  39. 39. Structural change Table 1: Wage compression and the indicators of structural change 5th-to-1st 9th-to-5th Raw Parametric DFL Raw Parametric DFL High-skilled Neg 0 Neg Neg Neg Neg #Transition 0 0 Neg Neg Neg 0 Obs. 330 289 289 289 330 289 Countries 26 25 25 25 26 25 R&D Neg 0 Neg Neg Neg 0 #Transition Neg 0 Neg Neg 0 0 Obs. 387 344 344 344 387 344 Countries 31 30 30 30 31 30 11
  40. 40. Structural change Table 1: Wage compression and the indicators of structural change 5th-to-1st 9th-to-5th Raw Parametric DFL Raw Parametric DFL High-skilled Neg 0 Neg Neg Neg Neg #Transition 0 0 Neg Neg Neg 0 Obs. 330 289 289 289 330 289 Countries 26 25 25 25 26 25 R&D Neg 0 Neg Neg Neg 0 #Transition Neg 0 Neg Neg 0 0 Obs. 387 344 344 344 387 344 Countries 31 30 30 30 31 30 Trade Neg - - Neg 0 0 #Transition 0 0 Neg 0 Pos 0 Obs. 488 416 416 416 488 416 Countries 31 30 30 30 31 30 11
  41. 41. Structural change Table 1: Wage compression and the indicators of structural change 5th-to-1st 9th-to-5th Raw Parametric DFL Raw Parametric DFL High-skilled Neg 0 Neg Neg Neg Neg #Transition 0 0 Neg Neg Neg 0 Obs. 330 289 289 289 330 289 Countries 26 25 25 25 26 25 R&D Neg 0 Neg Neg Neg 0 #Transition Neg 0 Neg Neg 0 0 Obs. 387 344 344 344 387 344 Countries 31 30 30 30 31 30 Trade Neg - - Neg 0 0 #Transition 0 0 Neg 0 Pos 0 Obs. 488 416 416 416 488 416 Countries 31 30 30 30 31 30 Employment Neg Pos Neg Neg 0 Neg #Transition 0 Pos Neg Neg 0 Pos Obs. 470 403 403 403 470 403 Countries 31 30 30 30 31 30 11
  42. 42. Structural change Table 1: Wage compression and the indicators of structural change 5th-to-1st 9th-to-5th Raw Parametric DFL Raw Parametric DFL High-skilled Neg 0 Neg Neg Neg Neg #Transition 0 0 Neg Neg Neg 0 Obs. 330 289 289 289 330 289 Countries 26 25 25 25 26 25 R&D Neg 0 Neg Neg Neg 0 #Transition Neg 0 Neg Neg 0 0 Obs. 387 344 344 344 387 344 Countries 31 30 30 30 31 30 Trade Neg - - Neg 0 0 #Transition 0 0 Neg 0 Pos 0 Obs. 488 416 416 416 488 416 Countries 31 30 30 30 31 30 Employment Neg Pos Neg Neg 0 Neg #Transition 0 Pos Neg Neg 0 Pos Obs. 470 403 403 403 470 403 Countries 31 30 30 30 31 30 High-tech export 0 0 Pos 0 Neg 0 #Transition Neg 0 Neg 0 Pos 0 Obs. 458 399 399 399 458 399 Countries 31 30 30 30 31 30 11
  43. 43. Conclusions
  44. 44. Conclusions We provide consistent estimates of unconditional and conditional measures of earnings inequality • Wage dispersion increased rapidly in early transition, mainly through prices 12
  45. 45. Conclusions We provide consistent estimates of unconditional and conditional measures of earnings inequality • Wage dispersion increased rapidly in early transition, mainly through prices • Initially level was lower, now is higher than Western Europe 12
  46. 46. Conclusions We provide consistent estimates of unconditional and conditional measures of earnings inequality • Wage dispersion increased rapidly in early transition, mainly through prices • Initially level was lower, now is higher than Western Europe • Decline, but very slow 12
  47. 47. Conclusions We provide consistent estimates of unconditional and conditional measures of earnings inequality • Wage dispersion increased rapidly in early transition, mainly through prices • Initially level was lower, now is higher than Western Europe • Decline, but very slow • Conditional estimates reveal that workers still more similar in transition countries than in Western Europe 12
  48. 48. Conclusions We provide consistent estimates of unconditional and conditional measures of earnings inequality • Wage dispersion increased rapidly in early transition, mainly through prices • Initially level was lower, now is higher than Western Europe • Decline, but very slow • Conditional estimates reveal that workers still more similar in transition countries than in Western Europe 12
  49. 49. Conclusions We provide consistent estimates of unconditional and conditional measures of earnings inequality • Wage dispersion increased rapidly in early transition, mainly through prices • Initially level was lower, now is higher than Western Europe • Decline, but very slow • Conditional estimates reveal that workers still more similar in transition countries than in Western Europe Structural change per se is not a big force • Skill biased technological change is correlated with lower earnings inequality • No stable results for globalization Inequality form below more relevant than that of above. 12
  50. 50. Questions or suggestions? Thank you! w: grape.org.pl t: grape org f: grape.org e: j.tyrowicz@grape.org.pl 13

×