Page 8 - Working Paper (The Myths and Realities of Tax Performance Under Semi-Autonomous Revenue Authorities)
P. 8

DDTC Working Paper 0213
                                                                                                            8



                   SARA=0),  the  amount  of  net  FDI  inflows  (%  of   positive effect on tax revenue because those reflect
                   GDP), value-added of industrial sector (% of GDP),   the rising level of a country’s economy. Especially
                   population  growth  (%).  Variable  SARA    is  tested   for  the contribution of the industrial sector, it
                   in two different models, to find robustness of the   can be understood tax revenue will increase as a
                   variable.                                        movement towards a modern economy.

                      The results of the second stage of the analysis,   Of the four independent variables included in
                   can be seen in Table 3, 4, and 5. Institutional SARA   the model only age dependency ratio is negatively
                   model gives significant results at the 99 % level in   affecting the tax ratio. This can be understood as
                   both  models.  Coefficient  sign  indicates  a  positive   follows. An increase of the ratio  of numbers of
                   result (+) with a range between 3.0 to 5.1. That is,   non-productive  age  population  (aged  0-15  years
                   in a SARA type country, the tax ratio will increase   and ≥ 65 years) to the population of reproductive
                   by  3-5  %  of  GDP    SARA  model  positively  affects   age  (age  15-65  years)  leads  to  a  decrease  in  the
                   tax revenue because of its ability to adapt to rapid   tax  revenues. The burden of non-productive age
                   changes in the economic  landscape, forming  an   population to the population of productive age will
                   effective and efficient tax authorities, as well as the   actually inhibit the rate of  increase in economic
                   openness factor, and governance better.          activity (both consumption and investment).
                      In addition,  several other  control variables   Number of observations used for the analysis
                                                                                                          2
                   such as: the level of consumption, the contribution   is 519  for model 1 and 515  in model 2. R  value
                   of the industrial  sector,  economic openness,   for  model  1  (0.23)  is  higher  than  the  model  2
                   and  population  age  dependency ratio  of non-  (0.15). Although model 1 provides better results,
                   productive  to  productive  population  also has   this model can only explain  22-23  % variations
                   significant impact and gets the expected sign. Rate   of  all  observations.  Furthermore,  in  a  panel  data
                                                                                          2
                   of consumption, contribution of industrial sector,   regression  with  a  low  R  can be considered not
                   as  well  as  economic  openness  significantly  have   problematic.

                                                         Table 3 - Pooled OLS
                                                                                 Model 1           Model 2
                                            Indicator
                                                                            Coefficient  T-Stat  Coefficient  T-Stat
                    Institution (SARA or Non-SARA                              3.425   3.05***   3.002   2.89***
                    Consumption (% PDB)                                        0.062   2.24**
                    FDI Inflows                                                                  -0.002    -0.47
                    Value added of industrial sector                           0.220   6.07***    0.23   7.50***
                    Openness (international trade/GDP)                         0.014   2.63***
                    Population Growth                                                            -0.103    -0.50
                    Age dependency ratio                                       -0.145  -5.34***
                    Number of observation                                         519                515
                    R 2                                                           0.225             0.150
                   Dependent variable is tax revenue (% PDB). Symbol***, **, and * indicate that coefficient of independent variable is statistically significant at 99, 95, and
                   90 %.
                                                         Table 4 - Fixed Effect
                                                                                 Model 1           Model 2
                                            Indicator
                                                                            Coefficient  T-Stat  Coefficient  T-Stat
                    Institution (SARA or Non-SARA                              5.104   3.43***   4.407   3.16***
                    Consumption (% PDB)                                        0.064   2.71***
                    FDI Inflows                                                                  -0.002    -0.58
                    Value added of industrial sector                           0.245   7.15***   0.258   8.19***
                    Openness (international trade/GDP)                         0.015   2.67***
                    Population Growth                                                            -0.078    -0.45
                    Age dependency ratio                                       -0.138  -6.00***
                    Number of observation                                         519                515
                    R 2                                                           0.227             0.151
                   Dependent variable is tax revenue (% PDB). Symbol***, **, and * indicate that coefficient of independent variable is statistically significant at 99, 95, and
                   90 %.
   3   4   5   6   7   8   9   10   11   12