Tax mimicking in Spanish municipalities: expenditure spillovers, yardstick competition, or tax competition?

Bastida, Francisco; Benito, Bernardino; Guillamón, María-Dolores; Ríos, Ana-María

doi:10.3326/pse.43.2.1.

3698
Views

1208
Downloads

Tax mimicking in Spanish municipalities: expenditure spillovers, yardstick competition, or tax competition?

Francisco Bastida*

Bernardino Benito*

María-Dolores Guillamón*

Ana-María Ríos*

Download citation

https://doi.org/10.3326/pse.43.2.1

FULL ARTICLE

FIGURES & DATA

REFERENCES

CROSSMARK POLICY

METRICS

LICENCING

PDF

Table 1
Information on variables and descriptive statistics (continued)

Variable	Calculation	Literature	Mean	Std. dev.	Min	Max
uproptaxrate	Tax rate to be multiplied by tax base (urban property value)	Solé-Ollé (2003), Lyytikäinen (2012), Baskaran (2014), Allers and Elhorst (2005), Isen (2014), etc.	.62	.17	.2	1.23
neig_uproptaxrate	Average neighboring uproptaxrate		.62	.13	.35	1.14
cartaxaveragerate	Vehicle tax liability of municipality i, year t /Minimum vehicle tax liability according to law 2/2004, year t	Besley and Case (1995), Solé-Ollé (2003), Delgado, Lago‐Peñas and Mayor (2015)	1.35	.29	1	2
neig_cartaxaveragerate	Average neighboring cartaxaveragerate		1.36	.23	1	2.00
r_revenue1pc	Municipal direct and indirect taxes per capita (real 2002 €)	Solé-Ollé (2003)	251.76	199.26	23.37	5633.85
r_transfpc	Transfers received from upper-level governments per capita (real 2002 €)	Revelli (2001), Bordignon, Cerniglia and Revelli (2003), Bucovetsky and Smart (2006), Lyytikäinen (2012), Delgado, Lago‐Peñas and Mayor (2015), Edmark and Ågren (2008), Baskaran (2014)	377.75	203.29	45.24	4228.01
income	Per capita income of the municipality (real 2002 income levels). Ranging from 1 (lowest) to 10 (highest).	Besley and Case (1995), Brett and Pinkse (2000), Bordignon, Cerniglia and Revelli (2003), Gérard, Jayet and Paty (2010), Edmark and Ågren (2008), Lyytikäinen (2012), Cassette, Di Porto and Foremny (2012)	5.15	2.30	1	10
unemploy	Unemployment rate (%)	Besley and Case (1995), Revelli (2001), Bordignon, Cerniglia and Revelli (2003), Gérard, Jayet and Paty (2010), Lyytikäinen (2012), Cassette, Di Porto and Foremny (2012), Edmark and Ågren (2008), Delgado, Lago‐Peñas and Mayor (2015)	8.56	4.97	.5	36.97
lnpopul	natural logarithm of municipality population	Bordignon, Cerniglia and Revelli (2003), Heyndels and Vuchelen (1998), Delgado, Lago‐Peñas and Mayor (2015), Brett and Pinkse (2000), Edmark and Ågren (2008)	8.63	1.21	6.90	15.00
r_housevalue	Tax base (real estate value) according to municipal government’s urban property value (real 2002 thousand €)	Oates (1969)	30.43	23.11	2.65	196.83
MCideology	Municipal Council political sign (0 left; 1 right)	Bordignon, Cerniglia and Revelli (2003), Edmark and Ågren (2008), Delgado, Lago‐Peñas and Mayor (2015)	.49	.49	0	1

Variable	Calculation	Literature	Mean
majority	Majority of one party in municipal council=1, 0 otherwise	Solé-Ollé (2003), (2016), Delgado, Lago‐Peñas and Mayor (2015), Fiva and Rattsø (2007), Roubini and Sachs (1989)	.66
munelectionyear	Dummy election year (1 election year; 0 no election year)	Besley and Case (1995), Bordignon, Cerniglia and Revelli (2003), Edmark and Ågren (2008), Gérard, Jayet and Paty (2010), Isen (2014)	.25
munpreelection	Dummy pre-election year (1 pre-election year; 0 no pre-election year)		.26
munpostelection	Dummy post-election year (1 post-election year; 0 no post-election year)		.23
dumm_yearvaluerev	Property values have been updated by the municipality i in year t=1, 0 otherwise	Revelli (2002), Solé-Olle (2003)	.03
bipartisan	Municipal ruling party belongs to the two main national parties (1 belongs to one of the two main political parties; 0 doesn’t belong)	Revelli (2002)	.74
nation	National government alignment with municipal government. If both are conservative or both are progressive, dummy takes value 1. Value 0 otherwise	Brett and Pinkse (2000)	.52
region	Regional government alignment with municipal government. If both are conservative or both are progressive, dummy takes value 1. Value 0 otherwise	Brett and Pinkse (2000)	.62
cgov_1	Takes 1 if there was one change in municipal government stemming from elections in 2002-2013. Takes 0 otherwise.	Edmark and Ågren (2008)	.28
cgov_2	Takes 1 if there were two changes in municipal government stemming from elections in 2002-2013. Takes 0 otherwise.		.29
cgov_3	Takes 1 if there were three changes in municipal government stemming from elections in 2002-2013. Takes 0 otherwise.		.13
autcom	Autonomous community (region) where the municipality is located	Besley and Case (1995), Isen (2014)	Non reported

Variable income is only available as a discrete variable in levels 1-10 for the time window. It was constructed by the Klein Institute (Autonomous University of Madrid).

Descriptive statistics of variable autcom are not reported because they are not relevant for the analysis. They are N-1 dummy variables representing the region where the municipality is located. Thus, 16 more rows would make the table more complex, without adding meaningful information for the reader.

Table 2A
Property tax regressions

Dependent variable

uproptaxrate

Estimation method

GMM

Sample

Whole sample

Sub-samples

Whole sample

Sub-samples

Basic model

Interaction

Majority = 0

Majority = 1

Basic model

Interaction

Majority = 0

Majority = 1

neig_uproptaxrate

*** .2576

8.90

*** .2466

9.65

*** .1772

3.78

*** .2366

7.10

*** .0196

3.29

*** .0211

3.41

.0239

0.37

** .1207

2.50

uproptaxrate _(t-1)

*** .9543

39.83

*** .9292

41.92

*** .9125

32.30

*** .9327

31.53

*** .9411

198.27

*** .9407

199.42

*** .7594

24.09

*** .6557

18.25

r_revenue1pc _(t-1)

-.0000

-0.98

-.0000

-0.33

.0000

0.60

.0000

0.14

.0001

1.14

.0001

0.95

* -.0001

-1.95

*** -.0001

-4.28

r_transfpc _(t-1)

*** -.0000

-2.83

*** -.0000

-3.62

-.0000

-1.52

*** -.0000

-3.65

.0000

0.51

.0000

0.44

.0000

0.30

** .0001

2.33

income _(t-1)

*** -.0027

-5.88

*** -.0027

-6.36

*** -.0037

-4.03

*** -.0034

-5.87

*** -.0263

-2.79

*** -.0265

-2.81

** -.0261

-2.39

.0007

0.14

unemploy _(t-1)

* .0003

1.77

*** .0006

4.09

** .0008

2.51

*** .0006

3.36

*** -.0213

-3.18

*** -.0213

-3.16

.0056

0.38

** -.0224

-2.25

lnpopul _(t-1)

* .0022

1.95

*** .0026

2.69

*** .0047

3.41

* .0020

1.66

-.0014

-1.00

-.0014

-0.97

-.1580

-1.63

-.0186

-0.29

r_housevalue _(t-1)

*** -.0004

-6.49

*** -.0004

-7.83

*** -.0006

-7.44

*** -.0003

-5.00

*** -.0005

-7.92

*** -.0005

-8.10

*** -.0005

-3.80

-.0004

-1.45

MCideology _(t-1)

*** .0298

4.76

*** .0175

3.58

* .0131

1.82

.0012

0.16

-.0011

-0.78

-.0014

-0.94

.0015

0.30

-.0072

-1.09

majority _(t-1)

-.0023

-0.31

-.0059

-1.12

.0016

1.36

.0012

1.06

munelectionyear

*** -.0083

-9.19

*** -.0097

-11.50

*** -.0147

-8.35

*** -.0075

-7.14

** .0373

2.57

*** .0377

2.60

-.0184

-0.98

-.0094

-0.81

munpreelection

*** -.0082

-8.98

*** -.0092

-10.83

*** -.0107

-6.26

*** -.0071

-7.16

*** .0397

2.78

*** .0398

2.78

.0104

0.44

** .0401

2.05

munpostelection

.0003

0.37

-.0001

-0.15

*** -.0050

-2.79

** .0026

2.23

*** .0351

3.88

*** .0354

3.92

-.0045

-0.35

-.0212

-1.33

dumm_yearvaluerev _(t-1)

** -.0105

-2.13

-.0073

-1.45

-.0085

-1.14

-.0044

-0.74

*** -.0078

-2.81

*** -.0078

-2.81

.0044

0.55

.0061

0.90

bipartisan _(t-1)

** .0150

2.36

** .0103

2.41

.0058

0.93

** .0220

2.49

.0006

0.42

.0003

0.21

-.0046

-0.66

.0004

0.05

nation _(t-1)

*** .0163

4.89

*** .0102

4.28

.0040

0.77

*** .0121

4.28

*** -.0040

-2.63

*** -.0041

-2.67

** -.0103

-2.10

.0014

0.44

region _(t-1)

.0030

0.74

.0016

0.51

* -.0089

-1.88

.0042

1.41

-.0006

-0.56

-.0007

-0.67

.0031

0.58

-.0003

-0.07

cgov_1xneig_uptaxrate

-.0049

-0.81

-.0024

-1.18

cgov_2xneig_uptaxrate

-.0060

-0.99

-.0016

-0.75

cgov_3xneig_uptaxrate

-.0025

-0.32

-.0032

-1.21

m(2) test

z = 1.89
Pr = .059

z = 1.81
Pr = .070

z = 0.68
Pr = .500

z = 1.15
Pr = .251

Hansen test

chi2= 808
Prob = .000

chi2= 1016
Prob = .000

chi2= 454
Prob = .002

chi2= 595
Prob = .000

Hausman

chi2 = 20
Prob = 0.467
RE

chi2 = 22
Prob = 0.455
RE

chi2 = 877
Prob = 0.000
FE

chi2 = 474
Prob = 0.000
FE

R-sq overall

0.8933

0.8937

0.2313

0.4531

All models include:

A constant, which is not shown.
Dummy variables for Spanish regions, which are not shown. This variable adds to the income variable (income) to control for differences in economic development among Spanish regions (autonomous communities).

Below each coefficient, z value is reported. Significance: * 10%, ** 5%, *** 1%.

RE= random effects IV regression. FE= fixed effects IV regression.

IV regression: Instrumented variables= neig_uproptaxrate r_revenue1pc r_transfpc income unemploy. Instruments= lnpopul, munelectionyear, munpreelection, munpostelection, r_debtpc cartaxaveragerate, propimmigrants, dumm_yearvaluerev

Table 2B
Car tax regressions

Dependent variable

cartaxaveragerate

Estimation method

GMM

Sample

Whole sample

Sub-samples

Whole sample

Sub-samples

Basic model

Interaction

Majority = 0

Majority = 1

Basic model

Interaction

Majority = 0

Majority = 1

neig_cartaxaveragerate

*** .0571

4.47

*** .0520

4.41

* .0362

1.80

** .0346

2.35

*** .7573

2.93

*** .1813

4.89

*** .2608

4.83

*** .4249

8.20

cartaxaveragerate _(t-1)

*** .8943

48.48

*** .9026

52.87

*** .9431

43.05

*** .9165

44.93

*** .4317

9.60

*** 1.1178

28.39

*** .6133

31.85

*** .6093

45.62

r_revenue1pc _(t-1)

** -.0000

-2.51

* -.0000

-1.86

*** -.0000

* -.0000

-1.82

* .0005

1.66

*** -.0004

-4.96

.0082

1.23

** -.0109

-2.56

r_transfpc _(t-1)

-.0000

-1.60

*** -.0000

-2.64

.0000

0.18

** -.0000

-2.36

*** -.0002

-2.92

* .0003

1.71

.0015

0.43

** -.0059

-2.38

income _(t-1)

*** -.0014

-3.44

** -.0010

-2.50

** -.0024

-2.44

*** -.0016

-2.84

-.0353

-1.60

*** .0987

4.70

.0004

0.37

* .0016

1.77

unemploy _(t-1)

* -.0004

-1.87

*** -.0005

-2.69

*** -.0018

-4.80

-.0003

-1.22

-.0047

-1.15

*** -.0298

-5.49

-.0000

-0.05

-.0002

-0.41

lnpopul _(t-1)

*** .0153

7.95

*** .0137

7.79

*** .0116

4.63

*** .0139

7.07

.1141

1.06

-.0236

-1.29

** .0485

2.34

-.0173

-1.19

r_housevalue _(t-1)

*** -.0001

-4.75

*** -.0002

-5.57

*** -.0002

-4.00

*** -.0001

-2.78

-.0002

-1.05

** -.0004

-2.33

*** -.0085

-2.80

*** -.0090

-4.04

MCideology _(t-1)

*** -.0195

-3.74

*** -.0170

-3.78

.0023

0.31

*** -.0225

-2.99

.0015

0.22

-.0078

-1.23

-.0037

-1.44

*** -.0093

-3.07

majority _(t-1)

.0025

0.41

-.0048

-1.01

-.0038

-0.64

-.0076

-1.39

munelectionyear

*** -.0126

-14.54

*** -.0115

-13.59

*** -.0148

-7.62

*** -.0107

-10.05

.0032

0.27

*** .0633

3.49

*** -.0134

-4.68

*** -.0088

-3.85

munpreelection

*** -.0086

-9.05

*** -.0079

-8.56

*** -.0091

-4.21

*** -.0073

-6.45

.0171

1.45

** .0237

2.15

*** -.0085

-3.18

*** -.0062

-3.02

munpostelection

*** .0044

3.80

*** .0053

5.27

.0000

0.02

*** .0067

4.80

.0079

1.53

*** .1072

5.62

.0007

0.27

*** .0071

3.93

dumm_yearvaluerev _(t-1)

** .0079

2.52

*** .0108

3.42

.0054

0.91

.0033

1.01

.0044

0.89

-.0108

-1.12

** .0103

2.44

.0052

1.58

bipartisan _(t-1)

*** -.0200

-3.73

-.0037

-0.79

-.0080

-1.37

-.0050

-0.59

.0115

1.58

* .0113

1.86

-.0048

-1.70

-.0065

-1.71

nation _(t-1)

-.0019

-0.76

-.0033

-1.59

-.0047

-0.93

-.0038

-1.60

-.0002

-0.05

*** -.0159

-3.28

.0006

0.31

** -.0034

-2.27

region _(t-1)

-.0033

-0.91

.0019

0.64

.0052

1.04

** .0058

2.00

-.0003

-0.06

-.0011

-0.17

* .0044

1.90

-.0012

-0.53

cgov_1xneig_cartaxrate

-.0009

-0.50

-.0022

-0.61

cgov_2xneig_cartaxrate

-.0004

-0.20

.0013

0.29

cgov_3xneig_cartaxrate

-.0026

-0.94

-.0052

-1.14

m(2) test

z = -0.02
Pr = 0.983

z = -0.04
Pr = 0.966

z = -1.00
Pr = 0.318

z = 0.99
Pr = 0.321

Hansen test

chi2= 607
Prob = .000

chi2= 840
Prob = .000

chi2= 440
Prob = .006

chi2= 464
Prob = .001

Hausman

chi2 = 7584
Prob = .000
FE

chi2 = 11
Prob = .885
RE

chi2 = 1843
Prob = .000
FE

chi2 = 665
Prob = .000
FE

R-sq overall

0.6487

0.6693

0.9134

0.8221

All models include:

A constant, which is not shown.
Dummy variables for Spanish regions, which are not shown. This variable adds to the income variable to control for differences in economic development among Spanish regions (autonomous communities).

Below each coefficient, z value is reported. Significance: * 10%, ** 5%, *** 1%.

IV regression: Instrumented variables= neig_uproptaxrate r_revenue1pc r_transfpc income unemploy. Instruments= lnpopul, munelectionyear, munpreelection, munpostelection, r_debtpc cartaxaveragerate, propimmigrants, dumm_yearvaluerev.

Table 3
Tax competition (Tiebout) regression

		Variable description
Dependent variable	movetoneigh	Neighbors’ population change / municipality_i population change. Population change= population_t / population_t-1
uproptax_rel _(t-1)	.0061 0.92	Average of neighbors’ urban property tax rate / municipality_i urban property tax rate
cartax_rel _(t-1)	-.0003 -0.48	Average of neighbors’ car tax rate / municipality_i car tax rate
income_rel _(t-1)	.0013 0.63	Average of neighbors’ income level / municipality_i income level
unemploy_rel _(t-1)	.0034 1.25	Average of neighbors’ unemployment rate / municipality_i unemployment rate
r_housevalue_rel _(t-1)	.0000 0.24	Average of neighbors’ real house value / municipality_i real house value
r_transfpc _(t-1)	-.0000 -1.22	See table 1
MCideology _(t-1)	.0056 1.18
majority _(t-1)	-.0021 -0.49
munelectionyear	*** .0020 2.78
munpreelection	*** .0022 2.84
munpostelection	*** .0017 2.83
dumm_yearvaluerev _(t-1)	-.0048 -1.29
bipartisan _(t-1)	*** .0182 2.62
nation _(t-1)	.0049 1.63
región _(t-1)	.0006 0.18
propimmigrants_(t-1)	.0142 0.66	Municipality_i immigrant population / Municipality_i total population
m(2) test	z = 0.45 Pr = 0.656
Hansen test	chi2= 82.65 Prob = .338