Inequality and Globalization Judging the Data by yurtgc548

VIEWS: 3 PAGES: 40

									     Inequality and Globalization:
          Judging the Data
A comparison of the UTIP data set on world pay inequalities
    with the Deininger-Squire data set on world income
                       inequalities


                   A Presentation at
                   The World Bank
                    June 18, 2002
                  by
          James K. Galbraith
                 and
            Hyunsub Kum




The University of Texas Inequality Project

    http://utip.gov.utexas.edu
                Two Data Sets
•   Deininger & Squire      •   UTIP-UNIDO
•   Income inequality       •   Pay inequality
•   Household surveys       •   Establishment surveys
•   Comprehensive           •   Narrow
•   Official & Unofficial   •   Official Data Only
•   Bibliographic           •   Calculated “in house”
•   Gini coefficient        •   Theil’s T statistic
              Key Questions

•   Is the coverage sufficient?
•   Are the numbers accurate?
•   Are the data right for the research question?
•   Can we do better?
The Deininger-Squire data set suffers from major
deficiencies of coverage...
         Number of Observations Per Country,
                     1950-1997




                                                                      DK Observations
                                                                          1 - 10
                                                                          11 - 20
                                                                          21 - 30
                                                                          31 - 40
                                                                          41 - 50
                                                                                  N


                                                                             W        E
 7000              0                 7000               14000 Miles
                                                                                  S



        Version of D&S used by Dollar and Kraay, “Growth is good for the poor.”
The UTIP-UNIDO Data Set has fewer gaps ….

   Number of Observations per Country,
               1963-1999




                                                        UTIP Observations
                                                             1 - 10
    Note: Observation count for Russia includes USSR         11 - 20
    1963-1991; China and Brazil blended from multiple        21 - 30
    editions of UNIDO ISIC; all others based on 2001
    edition only.                                            31 - 40
                                                             41 - 50
                     Comparing Coverage

Coverage of Inequality Measures: UNIDO-UTIP / D&S Gini              (2834/ 653)

                       Before 1966 - 1971 -  1976 -  1981 -   1986 -   1991 -   1996 -
Continent               1965 1970    1975    1980    1985     1990     1995     1999
Africa                 28/ 3   91/ 3  111/ 3  127/ 6   123/ 6   87/ 14   96/ 25   40/ 3
Central & North
America                 24/ 19    48/ 12      62/ 16      58/ 20      67/ 16       55/ 27       49/ 14       20/ 0
Asia                    33/ 26    73/ 26      87/ 23      99/ 29     104/ 28      100/ 36       86/ 17       33/ 0
Europe                  52/ 15    99/ 15     105/ 25     110/ 38     115/ 47      122/ 49      106/ 26       48/ 0
Oceania                  9/ 0     17/ 1       20/ 2       20/ 7       24/ 5        24/ 7        16/ 0        5/ 0
South America           11/ 2     21/ 3       27/ 8       35/ 10      41/ 6        46/ 10       43/ 5        17/ 0


UTIP coverage count for this table is based on UNIDO ISIC 2001 edition only, where matching data for GDP per capita also
available; total UTIP coverage is about 3200 observations.
         Judging Accuracy
       Some Useful Indicators:

•   Consistency over time
•   Consistency across space
•   Correspondence to known events
•   Consistency with “common knowledge”
               Consistency across time…


                       Non-OECD                    OECD                                    Non-OECD                    OECD

    -2.51034                                                              55.1357
log(Theil)




                                                                       Gini
    -4.38301                                                              26.0357
                1963   1968   1973   1978   1983    1988   1993 1998                1963   1968   1973   1978   1983    1988   1993 1998
                         Non-OECD vs. OECD                                                   Non-OECD vs. OECD




                              UTIP-UNIDO                                                    D&S


             Bars indicate standard deviation around average value for each year
Consistency across space…

           Global Inequality
                  UTIP Rankings




                                  1963-1999 Averages
                                  <= 0.0178
                                  0.0178 - 0.03556
                                  0.03556 - 0.05158
                                  0.05158 - 0.07439
                                  0.07439 - 0.09872
                                  0.09872 - 0.8926
                                   Global Inequality
                                      UTIP Rankings




 196 3- 1999 Av e ra ges
    < = 0 .0 1 7 8
    0 .0 1 7 8 - 0 .0 3 5 5 6
    0 .0 3 5 5 6 - 0 . 0 5 1 5 8
    0 .0 5 1 5 8 - 0 . 0 7 4 3 9
    0 .0 7 4 3 9 - 0 . 0 9 8 7 2
    0 .0 9 8 7 2 - 0 . 8 9 2 6




Note: Data for Balkans, Czech Republic, Slovakia and post-Soviet states are
post-1991 only. Earlier data for prior boundaries are available from UTIP.
                                  Global Inequality
                                     UTIP Rankings




196 3- 1999 Av e ra ges
   < = 0 .0 1 7 8
   0 .0 1 7 8 - 0 .0 3 5 5 6
   0 .0 3 5 5 6 - 0 . 0 5 1 5 8
   0 .0 5 1 5 8 - 0 . 0 7 4 3 9
   0 .0 7 4 3 9 - 0 . 0 9 8 7 2
   0 .0 9 8 7 2 - 0 . 8 9 2 6
                                  Global Inequality
                                     UTIP Rankings




196 3- 1999 Av e ra ges
   < = 0 .0 1 7 8
   0 .0 1 7 8 - 0 .0 3 5 5 6
   0 .0 3 5 5 6 - 0 . 0 5 1 5 8
   0 .0 5 1 5 8 - 0 . 0 7 4 3 9
   0 .0 7 4 3 9 - 0 . 0 9 8 7 2
   0 .0 9 8 7 2 - 0 . 8 9 2 6
                                  Global Inequality
                                     UTIP Rankings




196 3- 1999 Av e ra ges
   < = 0 .0 1 7 8
   0 .0 1 7 8 - 0 .0 3 5 5 6
   0 .0 3 5 5 6 - 0 . 0 5 1 5 8
   0 .0 5 1 5 8 - 0 . 0 7 4 3 9
   0 .0 7 4 3 9 - 0 . 0 9 8 7 2
   0 .0 9 8 7 2 - 0 . 8 9 2 6
Global Inequality
   UTIP Rankings



                    196 3- 1999 Av e ra ges
                       < = 0 .0 1 7 8
                       0 .0 1 7 8 - 0 .0 3 5 5 6
                       0 .0 3 5 5 6 - 0 . 0 5 1 5 8
                       0 .0 5 1 5 8 - 0 . 0 7 4 3 9
                       0 .0 7 4 3 9 - 0 . 0 9 8 7 2
                       0 .0 9 8 7 2 - 0 . 8 9 2 6
        World Bank Inequality
         D&S Gini Coefficients, 1950-1997


                                                                        <= 30.06
                                                                        30.06 - 34.66
                                                                        34.66 - 39
                                                                        39 - 44.2
                                                                        44.2 - 51.51
                                                                        51.51 - 62.3




Note the reported heterogeneity of North America and Europe, and the homogeneous
measurements for Asia, with low inequality comparable to northern Europe and Canada.
                                                                   Inequality (Gini)
                                                                    <= 30.06
                                                                    30.06 - 34.66
                                                                    34.66 - 39
                                                                    39 - 44.2
                                                                    44.2 - 51.51
                                                                    51.51 - 62.3




Elementary economics suggests these differences in inequality are implausible in an integrated
region. If inequality were really so much greater in France than in Germany, wouldn’t low-
skilled French workers migrate to Germany to sweep the streets?
                                                          Inequality (Gini)
                                                           <= 30.06
                                                           30.06 - 34.66
                                                           34.66 - 39
                                                           39 - 44.2
                                                           44.2 - 51.51
                                                           51.51 - 62.3




           Inequality (Gini)
           <= 30.06
           30.06 - 34.66
           34.66 - 39
           39 - 44.2
           44.2 - 51.51
           51.51 - 62.3




The UTIP data and the D&S data cannot both be right. If Indonesia or India has
highly unequal pay, how does it arrive at highly equal incomes – more equal than
Australia? Through a strongly redistributive welfare state? Ha! Alternatively, if
low Ginis in those countries reflect egalitarian but impoverished agriculture, then
why are Ginis so high in agrarian Africa?
Correspondence to known events…
                                                                                                Figure 7
                                                                                                                          Inequality in Iran and Iraq
                                    Inequality in China
                                              and Hong Kong
                                                                                                         160
                                                                                                         140
    300
                                                                                                         120
                                                                                                         100
    200                                                                                                                                                                    Revolution
                                                                                                         80
                                                                                                         60

    100                                                                                                  40
                                                                                                                                                                                       War
                                                                                                         20
                                                                 Tiananmen                                 0
                                                                                                                63         67              71         75              79          83            87         91
      0
            72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98                         65              69         73              77         81             85         89             93


                                 China                        Hong Kong
                                                                                                                                                           Iran            Iraq




                   Inequality in the Southern Cone                                                                             Inequality in North Am erica
          300
                                                                                                   200


          250

                                                                                                   150
          200
                                           Banking
          150                              Crisis                                                  100

                                                                                                                                                                                                GATT Entry
          100
                                                                                                    50
                                                  Falklands War
          50

                                       Military Coup                                                 0
           0                                                                                               71   72   73   74    75    76   77   78   79    80    81   82   83   84   85    86   87   88   89   90   91   92   93   94   95   96
                71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96

                                                                                                                 United States                                  Canada                                    Mexico
                                  Chile              Argentina         Brazil




Data for China drawn partly from State Statistical Yearbook
                           Inequality in Scandinavia
150




100




50




 0
        71 72 73 74     75 76 77 78 79 80 81 82 83 84 85 86 87               88 89 90 91 92 93       94 95 96


                      Finland                Sweden                Norway                 Denmark




                    Ine quality in Ce ntral Europe
      250




      200




      150




      100




      50




       0
            63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95


                        Czechoslovakia               Hungary                      Poland
Consistency with “common knowledge”…

          European Countries Ranked in Order of
        Inequality by the World Bank, Low to High
          1970               1992
          UK                 Spain
          Sweden             Finland
          Belgium            Belgium
          Netherlands        Netherlands
          Finland            Italy
          Germany            Germany
          Denmark            UK
          Greece             Sweden
          Spain              Denmark
          Norway             Norway
          Portugal           France
          Italy              Greece
          France             Portugal
                                       Source: Deininger and Squire
                                       Data are for nearest available year
European Countries Ranked in Order of Industrial
Earnings Inequality Using the Theil Statistic, Low to High


   1970                   1992                   Using the UTIP
   Norway                 Norway                 inequality
   Finland                Denmark                rankings, one
   Denmark                Finland                finds that
   Germany                Netherlands
                                                 countries in
   Netherlands            Sweden
   UK                     UK                     Europe that
   Belgium                Germany                have less
   Sweden                 Belgium                inequality also
   Greece                 Austria                have less
   France                 Greece
                                                 unemployment.
   Austria                Portugal
   Italy                  France
                          Spain
                          Italy               Source: OECD STAN
                                              and authors’ calculations
                                              Data are for year reported.
Consistency over time and space together…


   With the UTIP data, we can review changes in
   global inequality both across countries and
   through time. Nothing comparable can be
   done with the Deininger and Squire data set,
   for the measurements are too sparse and too
   inconsistent.
The Scale
Brown: Very large decreases in inequality;
more than 8 percent per year.
Red Moderate decreases in inequality.
Pink: Slight Decreases.
Light Blue: No Change or Slight increases
Medium Blue: Large Increases -- Greater
than 3 percent per year.
Dark Blue: Very Large Increases -- Greater
than 20 percent per year. h
1963 to 1969
1970 to 1976




 The oil boom: inequality declines in the producing states, but rises in the
 industrial oil-consuming countries, led by the United States.
1977 to 1983
1981 to 1987




    … the Age of Debt
1984 to 1990
 1988 to 1994




The age of globalization…
3D Surface Plot (Tngall4ax.STA 3v*5360c)
      z=0.05+0.001*x+-3.974e-6*y




               A regression of pay inequality on
               GDP per capita and time, 1963-1998.




                                                     0.008
                                                     0.016
                                                     0.025
                                                     0.033
                                                     0.041
                                                     0.049
                                                     0.057
                                                     0.065
                                                     0.074
                                                     0.082
                                                     above
                        Global Pay Inequality
                                      Time Effect, 1963-1997
                   0
T im e e ffe c t




               -0.1



               -0.2



               -0.3
                                                                                                        Time Effects
                                                                                                       Dollar & Kraay data set
                                                                                             0.5
               -0.4                                                                          0.4
                       63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97
                                                                                             0.3
                                                                      Year                   0.2

                                                                                             0.1

                                                                                               0
                                                                                               1950   1960    1970    1980   1990   2000
The time effect from a two-way fixed effects panel data analysis
of inequality on GDP per capita, with time and country effects.
Globalization and Inequality

        Overall, pay data reveal a strong upward trend in
inequality across countries, over time, with an inflection point
in the early 1980s.

        It is a reasonable inference that global macroeconomic
forces, notably rising real interest rates and the debt crisis,
followed by the implementation under pressure of neoliberal
policies, were responsible for this worldwide pattern of rising
inequality in pay structures.
   Are Pay Data Appropriate?
• There are some instances of selection bias:
  where industrial job losses affect mainly
  low-income workers, increasing inequality
  will be understated – especially in the UK.
• In very rich countries, trends in capital
  income can lead to large differences
  between the trend of pay inequality and of
  income inequality – especially in the U.S.
           But in General…
• Trade and technology do affect income
  mainly through pay.
• Manufacturing pay is a fair indicator of the
  movement of all pay.
• Pay is a large subset of all income.
• Most income dynamics are derived from
  pay dynamics.
           But in General…
• Trade and technology do affect income
  mainly through pay.         (Galor and Tsiddon 1997, Aghion and Howitt 1997.)


• Manufacturing pay is a fair indicator of the
  movement of all pay.          (Cf. Galbraith & Wang on China, 2001.)


• Pay is a large subset of all income.                            (Williamson 1982)


• Most income dynamics are derived from
  pay dynamics.  (Acemoglu 1997).
       Conclusion:

Used with care, good pay data are
  better than bad income data.
            Can we do better?
•   UNIDO industrial data has 28 sectors.
•   OECD’s STAN has 39 sectors.
•   Chinese State Statistics ~ 500 cells
•   Russian State Statistics ~ 900 cells
•   Brazil, Mexico: monthly observations
•   Will the World Bank take up the challenge
    of collecting national data in detail?
For more information:




    The University of Texas Inequality Project


        http://utip.gov.utexas.edu

								
To top