# RECORD 1d57857811e3f735e7c7dfe3dfb0c766f1aba868
# TITLE: Adaptive Reuse of Religious Heritage and Its Impact on House Prices
# YEAR: 2020.0  VENUE: Journal of real estate finance and economics  DOI: 10.1007/s11146-020-09798-x
# AUTHORS: Caixia Liu; Xiaolong Liu
# PAGES: 22  CANDIDATE TABLE PAGES (1-based): [13, 14, 15, 17, 19]
# ABSTRACT: Due to social demographic change and secularization, religious heritage sites in Europe are on the verge of losing their original functions. While the adaptive reuse seems to be a proactive strategy to preserve the historical and cultural value embedded in religious heritage sites, little is known concerning its external impact. This paper aims to fill this gap by investigating the external effect of reusing religious heritage on surrounding house prices. Employing both the parametric and non-parametric difference-in-differences hedonic model on a sample of 42 projects of reusing religious heritage and a rich dataset of housing transactions in the Netherlands, we find significant positive externality of reusing religious heritage on local house prices. The external effects are heterogeneous across differentiated project size and monumental status. Larger religious heritage reuse projects and those listed as national monuments exert greater influence on surrounding house prices.


===== PAGE 1 (score 5, numeric-paren 0, table-words 0, var-words 5) =====

Adaptive Reuse of Religious Heritage and Its Impact
on House Prices
Caixia Liu1 & Xiaolong Liu2
Accepted: 28 September 2020 /
# The Author(s) 2020
Abstract
Due to social demographic change and secularization, religious heritage sites in Europe
are on the verge of losing their original functions. While the adaptive reuse seems to be
a proactive strategy to preserve the historical and cultural value embedded in religious
heritage sites, little is known concerning its external impact. This paper aims to fill this
gap by investigating the external effect of reusing religious heritage on surrounding
house prices. Employing both the parametric and non-parametric difference-in-
differences hedonic model on a sample of 42 projects of reusing religious heritage
and a rich dataset of housing transactions in the Netherlands, we find significant
positive externality of reusing religious heritage on local house prices. The external
effects are heterogeneous across differentiated project size and monumental status.
Larger religious heritage reuse projects and those listed as national monuments exert
greater influence on surrounding house prices.
Keywords Religious heritage . Adaptive reuse . Urban revitalization . House prices
JEL-Classification C21 . D62 . D84 . O22 . R31
Introduction
Religious heritage constitutes an important element that shapes the urban historical and
cultural landscape in Western European countries. Over the last few decades, due to
https://doi.org/10.1007/s11146-020-09798-x
* Xiaolong Liu
xiaolong.liu@rug.nl
Caixia Liu
cx0619.liu@gmail.com
1
College of Economics, Jinan University, Guangzhou 510632, China
2
Real Estate Center, Department of Economic Geography, Faculty of Spatial Sciences, University
of Groningen, Groningen 9747AD, The Netherlands
Published online: 12 October 2020
J Real Estate Finan Econ (2022) 64:71–92


===== PAGE 13 (score 26, numeric-paren 0, table-words 3, var-words 16) =====

heritage reuse projects with size over 5000 square meters as a subsample. Column (3)
and (4) report the average treatment effect of large scale religious heritage reuse
projects. The coefficient presented in Column (3) indicates that religious heritage reuse
projects with size over 5000 square meters lead to an increase in local house prices by
7.15%, which is larger than the average treatment effect estimated using all religious
heritage reuse projects as shown in column (1). Similar finding is reported after
controlling for housing and neighborhood characteristics in Column (4). The results
Table 2 Descriptive Statistics of Full Sample
Mean
S.D.
Panel A: Housing characteristics
Transaction price (€)
221,776.97
144,487.2
Lift (yes = 1)
0.11
0.31
Size (m2)
102.07
42.7
Number of rooms
3.87
1.48
Parking (yes = 1)
0.2
0.4
Garden (yes = 1)
0.47
0.5
Balcony (yes = 1)
0.4
0.49
Roofterrace (yes = 1)
0.1
0.3
Building year <1945 (yes = 1)
0.5
0.5
Building year 1945–1959 (yes = 1)
0.1
0.3
Building year 1960–1970 (yes = 1)
0.13
0.34
Building year 1971–1980 (yes = 1)
0.06
0.25
Building year 1981–1990 (yes = 1)
0.06
0.25
Building year 1991–2000 (yes = 1)
0.08
0.28
Building year >2000 (yes = 1)
0.06
0.23
Housing type-Apartment (yes = 1)
0.57
0.5
Housing type-Semi-detached house (yes = 1)
0.32
0.47
Housing type-Corner house (yes = 1)
0.08
0.27
Housing type-Detached house (yes = 1)
0.03
0.18
Observations
342,439
Panel B: Neighborhood characteristics
Population density (per km2)
9536.01
6362.37
Average household size
1.99
0.49
Proportion of immigrants (%)
23.34
15.79
Population aged 0–14 (%)
13.55
4.98
Population aged 15–24 (%)
15.02
6.92
Population aged 25–44 (%)
35.87
8.86
Population aged 45–64 (%)
22.44
5.26
Population aged 65 and over (%)
13.12
7.38
Number of neighbourhoods
1623
83
Adaptive Reuse of Religious Heritage and Its Impact on House Prices


===== PAGE 14 (score 22, numeric-paren 12, table-words 3, var-words 3) =====

imply that large scale religious heritage reuse projects bring more positive impact on
local house prices than that of reuse projects with relatively smaller size.
Next to the scale difference among reuse projects, these projects also vary in terms
of their designated monumental status. In the Netherlands, historical buildings, includ-
ing religious heritage sites, are listed with various monumental status, such as national
monument, provincial monument, municipal monument or no monumental status at all,
based on their cultural and historical values. It is perceivable that religious heritage sites
designated with national monument status may bring reputation effects such that they
Table 3 Parametric Difference-in-Differences Estimates
Full sample
Project size≥5000m2
National monuments
DV: lnp
(1)
(2)
(3)
(4)
(5)
(6)
treat × post
0.0399***
0.0217***
0.0715***
0.0338***
0.0459***
0.0228***
(0.002)
(0.001)
(0.007)
(0.003)
(0.004)
(0.002)
Housing characteristics
no
yes
no
yes
no
yes
Neighborhood characteristics
no
yes
no
yes
no
yes
Year fixed effects
yes
yes
yes
yes
yes
yes
Neighborhood fixed effects
yes
yes
yes
yes
yes
yes
Observations
342,439
342,439
64,337
64,337
159,980
159,980
Adjusted R2
0.585
0.866
0.544
0.872
0.588
0.866
Note: *** , ** , * denotes the significance of coefficients at 1%, 5% and 10% respectively. The dependent
variable is the natural log of house transaction price. Robust standard errors are reported in parentheses
Fig. 4 Average Price Impact of 42 Religious Heritage Reuse Projects
84
C. Liu, X. Liu


===== PAGE 15 (score 36, numeric-paren 31, table-words 1, var-words 0) =====

Table 4 Non-parametric Difference-in-Differences Estimates: Full Sample
Distance (km)
0.1
0.2
0.3
0.4
0.5
0.6
Price impact
0.0348** (0.0224)
0.0346** (0.0178)
0.0345** (0.0160)
0.0324** (0.0155)
0.0316** (0.0158)
0.0312** (0.0162)
Distance (km)
0.7
0.8
0.9
1.0
1.1
1.2
Price impact
0.0318** (0.0161)
0.0332* (0.0173)
0.0349* (0.0173)
0.0359* (0.0210)
0.0353 (0.0192)
0.0334* (0.0175)
Distance (km)
1.3
1.4
1.5
1.6
1.7
1.8
Price impact
0.0321* (0.0191)
0.0310* (0.0162)
0.0295 (0.0179)
0.0270 (0.0164)
0.0217 (0.0145)
0.0119 (0.0128)
Note: *** , ** , * denotes the significance of coefficients at 1%, 5% and 10% respectively. Bootstrapped standard errors are reported in brackets
85
Adaptive Reuse of Religious Heritage and Its Impact on House Prices


===== PAGE 16 (score 6, numeric-paren 0, table-words 1, var-words 4) =====

are more attractive to local residents, and there is a higher chance for them to be
qualified for financial support aiming at maintenance and renovation than those without
such a distinguished status (Koster and Rouwendal 2017). Therefore, it seems plausible
that such reputation effects would be reflected in the estimated treatment effect so that
projects of reusing religious heritage sites with national monumental status would
impose greater positive externality on local housing market than that of other projects.
To check if this is the case, we re-estimate the model using only those sites listed as
national monuments and results are displayed in Column (5) and (6). Comparing the
coefficient estimates of the treatment variable shown in Column (5) and (6) with their
respective counterparts listed in Column (1) and (2) that are estimated using the full
sample, it appears that the average treatment effect is indeed larger for projects of
reusing religious heritage listed as national monuments.
Non-parametric Difference-in-Differences Estimates
The parametric difference-in-differences specification offers a rather straightforward
way to measure the average treatment effect of reusing religious heritage sites, how-
ever, it fails to make distinction of treatment effect variation along both the spatial and
temporal dimensions. In contrast, with the non-parametric difference-in-differences
specification, we are able to recover the gradient of treatment effect of reusing religious
heritage sites across space and time. Fig. 4 is the wireframe mesh plot which illustrates
the average treatment effect of 42 religious heritage reuse projects over space and time.
The mesh grid color is determined by the value of house price in natural log, and
proportional to surface height. As shown in Fig. 4, the house price surface before the
commencement of religious heritage reuse projects is quite flat, which validates that
Fig. 5 Price Impact of Religious Heritage Reuse with Project Size larger than 5000 Square Meters
86
C. Liu, X. Liu


===== PAGE 17 (score 28, numeric-paren 23, table-words 1, var-words 0) =====

Table 5 Non-parametric Difference-in-Differences Estimates: Project Size≥5000m2
Distance (km)
0.1
0.2
0.3
0.4
0.5
0.6
Price impact
0.1471*** (0.0195)
0.1435*** (0.0115)
0.1276*** (0.0256)
0.1122*** (0.0396)
0.1007** (0.0513)
0.0845 (0.0561)
Distance (km)
0.7
0.8
0.9
1.0
1.1
1.2
Price impact
0.0683 (0.0564)
0.0566 (0.0531)
0.0505 (0.0482)
0.0510 (0.0448)
0.0526 (0.0426)
0.0522 (0.0412)
Distance (km)
1.3
1.4
1.5
1.6
1.7
1.8
Price impact
0.0488 (0.0398)
0.0417 (0.0379)
0.0312 (0.0343)
0.0198 (0.0288)
0.0099 (0.0226)
0.0016 (0.0160)
Note: *** , ** , * denotes the significance of coefficients at 1%, 5% and 10% respectively. Bootstrapped standard errors are reported in brackets
87
Adaptive Reuse of Religious Heritage and Its Impact on House Prices


===== PAGE 18 (score 8, numeric-paren 0, table-words 2, var-words 4) =====

house prices close to the religious heritage site and house prices slightly further away
from the site share similar trends before the treatment. After the religious heritage starts
the reuse process, the house price surface jumps up steeply, indicating that religious
heritage reuse indeed has positive impact on local house prices. 5 years after the start of
the projects, the religious heritage reuse triggers house price appreciation within its
close proximity, i.e. 0.1 km, by approximately 5.2%.
To illustrate the treatment effect of religious heritage reuse numerically, Table 4
summarizes those point estimates shown in Fig. 4 and the price impacts reported in
Table 4 are the average treatment effect estimated as the price difference between the
post-treatment years, i.e. 0 to 5 years, and pre-treatment years, i.e. -5 to −1 years, at
each 0.1-km distance to the religious heritage site. Between 0.1 and 0.7 km distance to
the religious heritage site, house price appreciation is 3.1 ~ 3.5% after the start of the
reuse project, and is statistically significant at the 5% level. Similar degree of price
appreciation is observed for houses located between 0.8 and 1.4 km distance to the
religious heritage site, but is only significant at the 10% level. For houses that are
located further than 1.4 km to the religious heritage site, the treatment effect becomes
smaller and statistically insignificant.
Next, we keep those religious heritage reuse projects with size more than 5000
square meters to investigate whether the treatment effect is heterogeneous based on the
project size. Fig. 5 presents the mesh plot of the price impacts of large-scale religious
heritage reuse projects. Similar to the non-parametric results using the full sample, the
house price surface remains low and flat up to the start of the religious heritage reuse
projects. When reuse projects are initiated, the house price surface is first elevated
sharply then flattens off gradually until year 5. For houses located 0.1 km away from
the religious heritage site, the price appreciation is about 20% after 5 years since the
start of reuse project.
Fig. 6 Price Impact of Reusing Religious Heritage Designated as National Monument
88
C. Liu, X. Liu


===== PAGE 19 (score 33, numeric-paren 28, table-words 1, var-words 0) =====

Table 6 Non-parametric Difference-in-Differences Estimates: National Monuments
Distance (km)
0.1
0.2
0.3
0.4
0.5
0.6
Price impact
0.0692** (0.0314)
0.0670** (0.0308)
0.0637** (0.0279)
0.0578** (0.0265)
0.0517** (0.0262)
0.0466* (0.0255)
Distance (km)
0.7
0.8
0.9
1.0
1.1
1.2
Price impact
0.0438* (0.0246)
0.0427* (0.0235)
0.0426* (0.0227)
0.0415* (0.0222)
0.0387 (0.0215)
0.0344 (0.0242)
Distance (km)
1.3
1.4
1.5
1.6
1.7
1.8
Price impact
0.0323 (0.0244)
0.0315 (0.0237)
0.0290 (0.0225)
0.0248 (0.0173)
0.0191 (0.0151)
0.0106 (0.0109)
Note: *** , ** , * denotes the significance of coefficients at 1%, 5% and 10% respectively. Bootstrapped standard errors are reported in brackets
89
Adaptive Reuse of Religious Heritage and Its Impact on House Prices


===== PAGE 20 (score 11, numeric-paren 0, table-words 4, var-words 3) =====

Table 5 reports the average treatment effect of large-scale projects across space.
Overall, the house price appreciation triggered by these large religious heritage reuse
projects exhibits spatial decay which is more pronounced than that reported in Table 4,
and the treatment effect of these projects are much greater in comparison with the non-
parametric results using the full sample. Specifically, large projects contribute to house
price increase by 14.7% at the 0.1-km distance, which is statistically significant at the
1% level. From 0.2-km to 0.5-km distance to the religious heritage site, the extent of
house price appreciation drops gradually from 14% to 10% and remains statistically
significant at least at the 5% level. Beyond 0.6 km, the average treatment effect keeps
decreasing over space but becomes statistically insignificant.
Further, we restrict the non-parametric estimation using religious heritage sites
designated with national monument status. Fig. 6 demonstrates the treatment effect of
these religious heritage reuse projects over space and time. The house price surface
exhibits similar pattern to those shown in Figs. 4 and 5. At 0.1-km distance to the
religious heritage site listed as national monument, local house price appreciation is
approximately 7% after 5 years since the initiation of reuse project. Table 6 reports the
spatial distribution of the average treatment effect of reusing the religious heritage sites
with national monument status. Overall, the house price impact decreases with the
distance to the religious heritage site. Between 0.1-km and 0.5-km distance to the
religious heritage site, reusing religious heritage listed as national monument pushes up
local house prices by 5 ~ 7%, and the treatment effect is statistically significant at the
5% level. In terms of the magnitude, the price impacts are larger than those reported in
Table 4. From 0.6-km to 1-km distance to the religious heritage sites, the average
treatment effect drops gradually from 4.7% to 4.2% which is significant at the 10%
level. These religious heritage reuse projects do not seem to have significant impact on
house prices further than 1.1 km distance to the sites.
Conclusion
In this paper, we examine the external effect of reusing religious heritage on local house
prices. We begin with employing the standard parametric difference-in-differences
approach to estimate the average treatment effect of religious heritage reuse by
comparing house prices within an inner ring, i.e. treatment area, with house prices in
the outer ring, i.e. control area, before and after the treatment. We then resort to a novel
non-parametric difference-in-differences method to estimate the spatial and temporal
variation of the treatment effect of reusing religious heritage sites.
Using 42 completed reuse projects that spread across the Netherlands and span the
period from 1995 to 2017, findings from both the parametric and non-parametric
estimation reveal the presence of positive external effect of reusing religious heritage
sites on surrounding house prices. We further show that the project size and monument
status are the driving factors in determining the magnitude of observed positive
externality associated with reusing the religious heritage. Overall, these results indicate
that the adaptive reuse contributes to not only the historical and cultural preservation in
place but also the welfare of local communities in terms of housing price capitalization.
This study could be a useful reference towards better understanding of the welfare
implications of historic preservation and urban revitalization policies. To a certain
90
C. Liu, X. Liu