Economic growth or social expenditure: what is more effective in decreasing poverty and income inequality in the EU – a panel VAR approach

Velkovska, Ivana; Trenovski, Borce

doi:10.3326/pse.47.1.5.

2628
Views

314
Downloads

Economic growth or social expenditure: what is more effective in decreasing poverty and income inequality in the EU – a panel VAR approach

Ivana Velkovska*

Borce Trenovski*

Download citation

https://doi.org/10.3326/pse.47.1.5

FULL ARTICLE

FIGURES & DATA

REFERENCES

CROSSMARK POLICY

METRICS

LICENCING

PDF

Table 1
Literature review

Study	Period	States	Methodology	Results
Dafermos and Papatheodorou, 2010	1994-2007	14 EU members	Panel model	Social expenditure significantly decreases poverty and inequality
Nelson, 2013	1990-2008	28 EU members	Panel model	European social protection inadequate for redistribution
Cyrek, 2019	2007-2013	All EU members	Panel model	Effectiveness of social protection declined during euro zone crisis
Andrés-Sánchez, Belzunegui-Eraso and Valls-Fonayet, 2020	2011-2015	28 EU members	Panel model	The southern member states are less effective in tackling poverty and inequality
Molina-Morales, Amate-Fortes and Guarnido-Rueda, 2014	1996-2006	27 EU members	Panel model	The social protection expenditure is dependent on the political will of the country
Mansi et al, 2020	2009 -2018	EU and WB	Multiple regression analysis using the fixed effect model	Economic growth does have a significant impact on reducing poverty
Bosco and Poggi, 2019	2008-2011	26 EU countries	Dynamic three-level model	The risk of poverty is negatively related to the size of the structural social expenditure
Da Silva and Andrade, 2016	2003- 2013	EU-27	Nonparametric panel data model	Results show different effectiveness for different countries and for different levels of social transfers which above a saturation point result in inefficiency.
Sanchez and Perez-Corral, 2018	2005-2014	28 EU members	Dynamic panel models	The results show the existence of a negative correlation between public social expenditure as a whole and income inequality.
Hidalgo-Hidalgo and Iturbe-Ormaetxe, 2018	2005	17 EU members	Cross-section analysis	Public expenditure in primary education has a strong effect on raising individuals above the poverty line
Leventi, Sutherland and Tasseva, 2018	2013	7 EU members	Microsimulation models	Most cost-effectively in most countries are increasing child benefits and social assistance
Doina and Viorica, 2017	2015	27 EU members	Regression analysis	The expenditure made by the state have a significant influence on poverty reduction. The greatest influence is made by expenses for social protection and they are followed by the health care, business and education-related expenses
Cammeraat, 2020	1990-2015	22 EU and OECD members	2SLS regression models	Social expenditure reduces poverty and inequality without being harmful for GDP growth. Targeted schemes are most effective in reducing poverty, while social expenditure types with a universal character are more effective in reducing inequality.
Antonelli & De Bonis, 2017	2013	22 EU members	Cross section analysis	States that have the highest levels of social expenditure are also the ones with highest effectiveness of social expenditure
Caminada and Goudswaard, 2010	1990-2007	OECD and EU-15	Cross country analysis	If pensions are treated as transfers, we find a strong relationship between levels of social spending and antipoverty effects of social transfers and taxes. Social spending seems to be an important determinant of a country’s poverty outcome. Each percentage point of social expenditure alleviates poverty in both EU15 and non-EU15 countries by 0.7 percentage point on average
Caminada and Goudswaard, 2009	2005, 2006	OECD and EU-15	Cross country analysis	They do not find a strong relationship between levels of social Spending and antipoverty effects of social transfers and taxes. At the program level, family programs and child support alleviate poverty to a large extent.
Fonayet, Eraso and Sánchez, 2020	2007-2015	EU	Panel model	Correlation between social expenditure and the levels of poverty is not strong
Bayar & Sasmaz, 2018	2005-2014	Selected CE and EU countries	Causality analysis	There is no causal interaction between social expenditures and poverty in this sample

Table A1
pVAR results for model 1

Panel vector autoregression
GMM Estimation
Initial weight matrix: Identity
GMM weight matrix: Robust
No. of obs. = 180
No. of panels = 30
Ave. no. of T = 6.000
Variables	Coef.	Std. err.	z	P>\|z\|	[95% Conf. Interval]
dlogSPE
dlogSPE L1	-0.22	0.24	-0.93	0.35	-0.69	0.24
dlogSPE L2	-0.20	0.15	-1.29	0.19	-0.50	0.10
dPVR L1	0.00	0.00	1.97	0.04	0.00	0.01
dPVR L2	0.00	0.00	0.56	0.57	-0.00	0.00
dINQ L1	0.00	0.00	-1.21	0.22	-0.01	0.00
dINQ L2	0.00	0.00	-1.56	0.11	-0.01	0.00
dlogGDP L1	0.09	0.17	0.54	0.59	-0.24	0.42
dlogGDP L2	0.07	0.10	0.65	0.51	-0.14	0.28
dPVR
dlogSPE L1	-3.49	8.51	-0.41	0.68	-20.19	13.20
dlogSPE L2	-0.18	4.66	-0.04	0.96	-9.33	8.96
dPVR L1	0.05	0.19	0.29	0.77	-0.32	0.44
dPVR L2	0.02	0.09	0.22	0.82	-0.16	0.20
dINQ L1	0.01	0.24	0.07	0.94	-0.46	0.50
dINQ L2	0.16	0.15	1.09	0.27	-0.13	0.47
dlogGDP L1	-5.88	4.02	-1.46	0.14	-13.77	2.01
dlogGDP L2	-7.87	3.01	-2.61	0.00	-13.78	-1.97
dINQ
dlogSPE L1	-5.99	5.90	-1.01	0.31	-17.56	5.58
dlogSPE L2	-5.90	4.42	-1.33	0.18	-14.58	2.77
dPVR L1	0.19	0.14	1.38	0.16	-0.08	0.47
dPVR L2	0.12	0.07	1.70	0.08	-0.01	0.26
dINQ L1	-0.17	0.16	-1.07	0.28	-0.50	0.14
dINQ L2	-0.03	0.15	-0.24	0.81	-0.33	0.26
dlogGDP L1	-2.18	3.18	-0.68	0.49	-8.42	4.06
dlogGDP L2	8.04	2.57	3.12	0.00	2.99	13.09
dlogGDP
dlogSPE L1	0.01	0.26	0.07	0.94	-0.50	0.54
dlogSPE L2	-0.06	0.14	-0.46	0.64	-0.35	0.21
dPVR L1	0.00	0.00	0.45	0.65	-0.00	0.00
dPVR L2	0.00	0.00	1.34	0.18	-0.00	0.00
dINQ L1	-0.00	0.00	-0.33	0.74	-0.01	0.00
dINQ L2	-0.00	0.00	-1.08	0.28	-0.01	0.00
dlogGDP L1	0.32	0.22	1.42	0.15	-0.12	0.76
dlogGDP L2	0.01	0.08	0.22	0.82	-0.14	0.18

Source: Authors’ calculation using STATA 13.

Table A3
pVAR results for model 2

GMM Estimation
Initial weight matrix: Identity
GMM weight matrix: Robust
No. of obs. = 180
No. of panels= 30
Ave. no. of T = 6.000
Variables	Coef.	Std. Err.	z	P>\|z\|	[95% Conf. Interval]
dlogSPE
dlogSPE L1	-0.29	0.28	-1.06	0.29	-0.85	0.25
dlogSPE L2	-0.23	0.18	-1.28	0.20	-0.58	0.12
dPVR L1	0.00	0.00	1.76	0.07	0.00	0.01
dPVR L2	0.00	0.00	0.21	0.83	0.00	0.00
dlogGDP L1	0.13	0.17	0.74	0.46	-0.21	0.47
dlogGDP L2	0.12	0.10	1.14	0.25	-0.08	0.34
dPVR
dlogSPE L1	-2.59	9.71	-0.27	0.78	-21.62	6.43
dlogSPE L2	0.17	4.86	0.04	0.97	-9.35	9.69
dPVR L1	0.05	0.18	0.30	0.76	-.313	0.42
dPVR L2	0.03	0.08	0.46	0.64	-0.12	0.19
dlogGDP L1	-6.37	3.76	-1.69	0.09	-13.75	0.99
dlogGDP L2	-8.54	2.84	-3.00	0.00	-14.13	-2.96
dlogGDP
dlogSPE L1	0.00	0.28	-0.02	0.98	-0.57	0.55
dlogSPE L2	-0.07	0.15	-0.51	0.60	-0.37	0.21
dPVR L1	0.00	0.00	0.45	0.65	0.00	0.00
dPVR L2	0.00	0.00	1.22	0.22	0.00	0.00
dlogGDP L1	0.33	0.22	1.51	0.13	-0.09	0.77
dlogGDP L2	0.03	0.08	0.46	0.64	-0.12	0.20

Source: Authors’ calculation using STATA 13.

Table A5
pVAR results for model 3

GMM Estimation
Initial weight matrix: Identity
GMM weight matrix: Robust
No. of obs. = 180
No. of panels= 30
Ave. no. of T = 6.000
Variables	Coef.	Std. Err.	z	P>\|z\|	[95% Conf. Interval]
dlogSPE
dlogSPE L1	-0.19	0.22	-0.88	0.38	-0.63	0.24
dlogSPE L2	-0.18	0.15	-1.19	0.23	-0.47	0.11
dINQ L1	-0.00	0.00	-0.75	0.45	-0.01	0.00
dINQ L2	-0.00	0.00	-1.37	0.17	-0.01	0.00
dlogGDP L1	0.10	0.16	0.61	0.53	-0.22	0.42
dlogGDP L2	0.04	0.10	0.44	0.66	-0.15	0.24
dINQ
dlogSPE L1	-4.98	5.72	-0.87	0.38	-16.21	6.24
dlogSPE L2	-5.37	4.09	-1.31	0.18	-13.38	2.64
dINQ L1	-0.09	0.14	-0.67	0.50	-0.37	0.18
dINQ L2	0.02	0.12	0.20	0.84	-0.22	0.27
dlogGDP L1	-1.86	3.15	-0.59	0.55	-8.04	4.32
dlogGDP L2	7.25	2.68	2.70	0.00	1.99	12.51
dlogGDP
dlogSPE L1	0.03	0.26	0.12	0.90	-0.49	0.55
dlogSPE L2	-0.06	0.14	-0.44	0.66	-0.35	0.22
dINQ L1	-0.00	0.00	-0.20	0.84	0.00	0.00
dINQ L2	-0.00	0.00	-0.99	0.32	0.00	0.00
dlogGDP L1	0.32	0.22	1.44	0.15	-0.11	0.77
dlogGDP L2	0.01	0.08	0.13	0.89	-0.15	0.17

Source: Authors’ calculation using STATA 13.

Crossmark is a multi-publisher initiative from Crossref to provide a standard way for readers to locate the current version of a piece of content. By applying the Crossmark logo, the Institute of Public Finance is committing to maintaining the content it publishes and to alerting readers to changes if and when they occur.

Clicking on the Crossmark logo will tell you the current status of a document and may also give you additional publication record information about the document.

For more details see IPF Crossmark policy page.

March, 2023
I/2023

Tihana Škrinjarić
Introducing a composite indicator of cyclical systemic risk in Croatia: possibilities and limitations

Francisco Bastida
Political economics and citizens’ engagement in Croatia: a differential analysis

Petra Đurman, Anamarija Musa, Tereza Rogić Lugarić
The characteristics and effects of public participation in Croatian e-consultations in fiscal matters

Ivana Dobrotić, Teo Matković
Understanding territorial inequalities in decentralised welfare systems: early childhood education and care system expansion in Croatia

Ivana Velkovska, Borce Trenovski
Economic growth or social expenditure: what is more effective in decreasing poverty and income inequality in the EU – a panel VAR approach

Predrag Bejaković
Informality, Labour Mobility and Precariousness: Supplementing the State for the Invisible and the Vulnerable