How does the economy affect a religious phenomenon? A panel approach to international pilgrimages to the Shrine of Fátima

Matheus Belucio; José Alberto Fuinhas; Carlos Vieira

doi:10.2478/ejthr-2023-0010

.blurhash-client-img { display: none !important; }

How does the economy affect a religious phenomenon? A panel approach to international pilgrimages to the Shrine of Fátima

European Journal of Tourism, Hospitality and Recreation

Volume 13 (2023): Issue 1 (December 2023)

By: Matheus Belucio , José Alberto Fuinhas and Carlos Vieira

Open Access

|Nov 2023

Figures & Tables

Portuguese international tourism
Source: World Bank (2020)

Official pilgrims
Source: Belucio et al. (2020)

Table 1:

Statistical description of the data

Variable	Obs.	Mean	Stand. dev.	Minimum	Maximum
pil	290	4307.65	9111.46	10	69540
gdppc	290	1927727	6393161	12323.84	37500000
depa	290	22400000	25500000	261000	143000000
une	290	8.12	5.14	0.29	27.33
exc	290	97.61	8.42	69.42	129.96
lpil	290	6.963	1.706	2.303	11.150
lgdppc	290	11.348	1.989	9.419	17.439
ldepa	290	16.315	1.230	12.472	18.779
dlpil	261	0.145	0.727	−1.901	3.203
dlgdppc	261	0.014	0.035	−0.155	0.215
dldepa	261	842058	2981744	−17200000	18400000
dune	261	0.068	1.146	−3.072	6.603
dexc	261	−0.798	5.086	−21.251	12.405

[i] Notes: ‘l’ in front of the variables means that the value was transformed into a natural logarithm; Stand. Dev. denotes standard deviation; rates cannot be transformed into a natural logarithm due to statistical restrictions.

Table 2:

Panel unit root tests

Maddala & Wu (1999) Panel Unit Root test
Specification without trend				Specification with trend
Variables	lags	Chi-Square	p-value	Variables	lags	Chi-Square	p-value
lper	0	72.878	0.090	lper	0	86.045	0.010
dlper	0	303.272	0.000	dlper	0	277.706	0.000
lgdppc	0	55.546	0.5670	lgdppc	0	455.452	0.000
dlgdppc	0	865.527	0.000	dlgdppc	0	681.756	0.000
une	0	101.813	0.000	une	0	370.916	0.000
dune	0	402.546	0.000	dune	0	273.373	0.000
exc	0	66.322	0.212	exc	0	124.349	0.000
dexc	0	318.098	0.000	dexc	0	225.599	0.002
ldepa	0	32.529	0.997	ldepa	0	112.73	0.000
dldepa	0	213.454	0.000	dldepa	0	239.402	0.000

Pesaran (2007) Panel Unit Root test (CIPS)
Specification without trend				Specification with trend
Variables	lags	Zt-bar	p-value	Variables	lags	Zt-bar	p-value
lper	0	−1.436	0.075	lper	0	−2.196	0.014
dlper	0	−6.965	0.000	dlper	0	4.618	0.000
lgdppc	0	0.107	0.542	lgdppc	0	0.807	0.210
dlgdppc	0	−3.656	0.000	dlgdppc	0	1.535	0.062
une	0	−1.495	0.067	une	0	−2.037	0.021
dune	0	−5.126	0.000	dune	0	0.509	0.305
exc	0	−2.179	0.015	exc	0	0.323	0.627
dexc	0	−4.223	0.000	dexc	0	−2.978	0.001
ldepa	0	0.882	0.811	ldepa	0	2.059	0.980
dldepa	0	−1.653	0.049	dldepa	0	−1.559	0.060

[i] Notes: ^*, ^*** denote statistical significance at the levels of 10% and 1%, respectively; Maddala & Wu (1999) Panel Unit Root test assumes the cross-sectional independence and H0: series in l(1). Pesaran (2007) Panel Unit Root Test (CIPS) assumes that cross-sectional dependence is in the form of a single unobserved common factor and H0: series is I(1)

Table 3:

Variance inflation factor statistics

Variable	VIF
Lgdppc	1.10
Exc	1.09	Mean VIF	1.07
Une	1.06
Ldepa	1.04
Dlgdppc	1.60
Dune	1.48	Mean VIF	1.33
Dldepa	1.17
Dexc	1.08

[i] Note: Dependent variables are the number of pilgrims in the natural logarithm and the first difference, respectively.

Table 4:

Correlation matrix and cross-sectional dependence (CD)

	Correlation matrix					Cross-sectional dependence
	lpil	lgdppc	Une	exc	ldepa	CD-test	p value	corr	abs(corr)
lpil	1					29.27	0.000	0.459	0.500
lgdppc	−0.2629^*	1				35.94	0.000	0.564	0.664
une	0.4568^*	−0.1734^*	1			11.35	0.000	0.178	0.498
exc	−0.0152	−0.2436^*	−0.0496	1		13.83	0.000	0.217	0.575
ldepa	0.3892^*	−0.0782	0.1281^*	0.1287^*	1	31.87	0.000	0.500	0.566
	dlpil	dlgdppc	Dune	dexc	dldepa
dlpil	1					23.32	0.000	0.386	0.473
dlgdppc	0.2009^*	1				28.16	0.000	0.466	0.541
dune	−0.2057^*	−0.5664^*	1			21.14	0.000	0.350	0.517
dexc	0.1472^*	0.1586^*	−0.1373^*	1		5.60	0.000	0.093	0.403
dldepa	0.1574^*	0.3093^*	−0.1486^*	0.2563^*	1	8.19	0.000	0.135	0.305

Note: The CD test has N(0,1) distribution under H0: cross-sectional independence.

* denotes statistical significance at 5%.

Table 5:

Hausman test

FE vs. RE	FE vs. RE
Chi2(8) = 151.74^***	99.99^***
Default option by Stata	Sigmamore option by Stata

Notes:

*** denotes significance at the 1% level.

Table 6:

Specification tests

	Statistics	Probability
Modified Wald test	275.61	0.0000
Pesaran's test	19.945	0.0000
Wooldridge test	33.975	0.0000

[i] Notes: H0 of Modified Wald test: homoscedasticity; H0 of Pesaran’s test: residuals are not correlated and follow a normal distribution; and H0 of Wooldridge test: no first-order autocorrelation.

Table 7:

Estimation results (corrected shocks)

Dependent Variable: DLPIL	Coefficients	FE	Coefficients	FE-DK
Constant	−41.51392	^***	−41.51392	^***
jap2016	−2.82544	^***	−2.82544	^***
ndl2017	2.047733	^***	2.047733	^***
Dldepa	2.35E-08		2.35E-08	^**
Dlgdppc	2.760323	^**	2.760323	^**
Dexc	0.013365	^*	0.013365	^*
lpil(−1) (ECM)	−0.9630423	^***	−0.9630423	^***
Diagnostic statistics
N	261		261
R²	0.5233		0.5233
F stat	F(9, 223) = 27.21		F(9, 8) = 4874.67
Prob	^***		^***

[i] Notes: ^***, ^**, ^* denote statistical significance at 1%, 5% or 10% levels, respectively; the Driscoll & Kraay (1998) standard errors were used to estimate the models.

Residues of estimation
Source: Authors’ elaboration

Table 8:

Impacts and speed of adjustment

Dependent Variable: DLPIL	Coefficients	FE	Coefficients	FE-DK
Short-term impacts
Dldepa	2.35E-08		2.35E-08	^**
Dlgdppc	2.760323	^**	2.760323	^**
Dexc	0.013365	^*	0.013365	^*
Long-term impacts
ldepa(−l)	1.133986	^***	1.133986	^***
lgdppc(−l)	2.8l2038	^***	2.812038	^***
une(−l)	−0.0381511		−0.0381511	^***
Speed of adjustment
ECM	−0.9630423	^***	−0.9630423	^***

[i] Notes:

[ii] ^***, ^**, and ^* denote statistical significance at 1%, 5% and 10% levels, respectively; the ECM denotes the coefficient of the variable LPIL lagged once.

References

Authors

Metrics

Articles in this issue

DOI: https://doi.org/10.2478/ejthr-2023-0010 | Journal eISSN: 2182-4924 | Journal ISSN: 2182-4916

Journal RSS Feed

Language: English

Page range: 110 - 124

Submitted on: Aug 2, 2022

Accepted on: Jul 27, 2023

Published on: Nov 18, 2023

Published by: Polytechnic Institute of Leiria

In partnership with: Paradigm Publishing Services

Publication frequency: 2 issues per year

Keywords:

Pilgrimage,

Economic determinants,

Panel data analysis,

Economics of Religion,

Religious tourism,

Shrine of Fátima

Related subjects:

Business and economics,

Business management,

Business management, other,

Industries,

Tourism, hospitality, travel,

Event industry, leisure industry,

Sports and recreation,

Sports and recreation, other

© 2023 Matheus Belucio, José Alberto Fuinhas, Carlos Vieira, published by Polytechnic Institute of Leiria
This work is licensed under the Creative Commons Attribution 4.0 License.

Volume 13 (2023): Issue 1 (December 2023)

How does the economy affect a religious phenomenon? A panel approach to international pilgrimages to the Shrine of Fátima

Figures & Tables

Figure 1:

Figure 2:

Table 1:

Table 2:

Table 3:

Table 4:

Table 5:

Table 6:

Table 7:

Figure 3:

Table 8:

Figure 4:

Paradigm

My account