1# RECORD 13561959c7aabb108ebf10b83cf499e51f1695002# TITLE: Flood risk management and real-estate prices: between prevention and “crowding out” effect3# YEAR: 2025.0 VENUE: Frontiers in Environmental Economics DOI: 10.3389/frevc.2025.16037464# AUTHORS: Jeanne Dachary-Bernard5# PAGES: 10 CANDIDATE TABLE PAGES (1-based): [6]6# ABSTRACT: Adapting to climate change is a key challenge for metropolis, and it partly deals with population location and flood risk management. The objective of this article is to understand the way flood risk is actually taken into account in residential choices. To this end, we measure how flood risk influences property prices and in particular, we target two specific flood management public instruments: the flood risk prevention plan (PPRI) and the public insurance mechanism (CatNat). We applied the hedonic price method over the period 2011–2016 to the French metropolitan area of Bordeaux, extended to include its neighboring estuarine municipalities. We show that flood risk zoning produces an expected significant deflating effect on prices whereas the CatNat system's insurance mechanism would appear to have a perverse effect. These results highlight an interesting crowding-out effect on the residential market that we finally discuss.789===== PAGE 1 (score 9, numeric-paren 0, table-words 0, var-words 9) =====1011TYPE Original Research12PUBLISHED 23 May 202513DOI 10.3389/frevc.2025.160374614OPEN ACCESS15EDITED BY16Saul Nurick,17University of Cape Town, South Africa18REVIEWED BY19Carsten Lausberg,20University of Applied Sciences21Nürtingen-Geislingen, Germany22Douw Boshoff,23University of Pretoria, South Africa24*CORRESPONDENCE25Jeanne Dachary-Bernard26jeanne.dachary-bernard@inrae.fr27RECEIVED 31 March 202528ACCEPTED 30 April 202529PUBLISHED 23 May 202530CITATION31Dachary-Bernard J (2025) Flood risk32management and real-estate prices: between33prevention and “crowding out” effect.34Front. Environ. Econ. 4:1603746.35doi: 10.3389/frevc.2025.160374636COPYRIGHT37© 2025 Dachary-Bernard. This is an38open-access article distributed under the39terms of the Creative Commons Attribution40License (CC BY). The use, distribution or41reproduction in other forums is permitted,42provided the original author(s) and the43copyright owner(s) are credited and that the44original publication in this journal is cited, in45accordance with accepted academic practice.46No use, distribution or reproduction is47permitted which does not comply with these48terms.49Flood risk management and50real-estate prices: between51prevention and “crowding out”52effect53Jeanne Dachary-Bernard*54UR ETTIS, INRAE Nouvelle-Aquitaine Bordeaux, Bordeaux, France55Adapting to climate change is a key challenge for metropolis, and it partly deals56with population location and flood risk management. The objective of this article57is to understand the way flood risk is actually taken into account in residential58choices. To this end, we measure how flood risk influences property prices and59in particular, we target two specific flood management public instruments: the60flood risk prevention plan (PPRI) and the public insurance mechanism (CatNat).61We applied the hedonic price method over the period 2011–2016 to the French62metropolitan area of Bordeaux, extended to include its neighboring estuarine63municipalities. We show that flood risk zoning produces an expected significant64deflating effect on prices whereas the CatNat system’s insurance mechanism65would appear to have a perverse effect. These results highlight an interesting66crowding-out effect on the residential market that we finally discuss.67KEYWORDS68real-estate69prices,70flood71risk,72crowding73out74effect,75residential76choices,77flood78management791 Introduction80Climate change is creating an increasing risk for urban areas, due in particular to81more frequent and violent storms and flood phenomena (IPCC, 2022). The United Nations82estimate that by 2050 almost 65% of the world’s population will be living in urban zones,83thus implying a new allocation of additional resources and services (to the population)84(United Nations, 2019). The population in France is expected to reach 74 million in 2050,85essentially distributed around urban centers increasing their vulnerability to global change.86In such a context, and in this medium and long-term perspective, our cities have to deal87with new challenges in terms of governance and more particularly to respond to the88imperative of implementing measures to adapt to climate change.89French current flood risk management strategy involves a combination of devices,90alongside a number of classic flood defense infrastructures, the high maintenance costs91of which are causing local managers to consider more flexible adaptation measures. Since921982, plans to prevent natural risks (PPRN) (known as PPRI when related to flood risk)93are France’s main management and prevention tools. Through zoning and associated94easements, these plans govern urbanization rights in accordance with the exposure and95vulnerability of the zones at risk. In tandem with the “CatNat” compensation system, these96tools constitute an original national solidarity system (Barraqué, 2014). In 1982, France97created the “CatNat” system, a fund financed on the basis of national solidarity via home98and car insurance premiums. When there is flooding, the mayor of a municipality can ask99the State (via the Prefect) to declare the event as a natural disaster (“catastrophe naturelle”100in French), hence the terminology CatNat. If, following an expertise, the State makes such101Frontiers in Environmental Economics10201103frontiersin.org104105106===== PAGE 6 (score 63, numeric-paren 36, table-words 3, var-words 10) =====107108Dachary-Bernard10910.3389/frevc.2025.1603746110and u a vector of independent and identically distributed random111error terms. PPRI, CatNat and their interaction PPRI∗CatNat are112the variables of main interest, as previously defined.1133 Estimation results114The spatial autoregressive error model (SEM) is estimated via115maximum likelihood (ML), and estimations were performed under116R (package spdep). Results are set out in Table 2. The OLS base117hedonic model is an a-spatial model that serves as a benchmark118against which the SEM effects can be compared.119We will present and interpret the results in two stages. Before120focusing on the impacts of the main variable of interest (flood risk121variables), we comment first on the other determinants of housing122prices and underline some stylized facts.1233.1 Stylized facts124In line with the literature (Baumont and Legros, 2013), the125results relating to the intrinsic characteristics of property (variables126group Z) suggest that the greater its built area, the more land it127has, the more rooms, whether it has a terrace, a swimming pool,128TABLE 2 Results of the MCO and SEM for hedonic valuation of real estate prices (The test’s risk levels are noted: 1% “∗∗∗”, 5% “∗∗” and 10% “∗”.).129Variables groups130Variables131MCO132SEM133Constant (α)1349.3516 (<0.0000)∗∗∗1359.4305 (0.1485)136Z137surf _build1380.0048 (<0.0000)∗∗∗1390.0047 (0.0001)∗∗∗140surf_area1410.0450 (<0.0000)∗∗∗1420.0455 (0.0016)∗∗∗143Rooms1440.0115 (<0.0000)∗∗∗1450.0112 (0.0021)∗∗∗146Terrace1470.0770 (<0.0000)∗∗∗1480.0688 (0.0046)∗∗∗149Swim1500.1860 (<0.0000)∗∗∗1510.1777 (0.0068)∗∗∗152Bathroom1530.0763 (<0.0000)∗∗∗1540.0709 (0.005)∗∗∗155Garage1560.0287 (<0.0000)∗∗∗1570.0331 (0.0041)∗∗∗158Condition1590.0611 (<0.0000)∗∗∗1600.0579 (0.005)∗∗∗161E162dist_road1630.1278 (<0.0000)∗∗∗1640.1231 (0.0187)∗∗165dist_BX166≤0.000 (0.000)∗∗∗167≤0.000 (0.000)∗∗∗168dist_river1690.0472 (<0.0000)∗∗∗1700.0487 (0.0036)∗∗∗171Isolation1720.1041 (<0.0000)∗∗∗1730.0986 (0.0166)∗∗174dist_road∗isolation175−0.0129 (<0.0000)∗∗∗176−0.0123 (0.0021)∗∗∗177surf_nature1780.0125 (<0.0000)∗∗∗1790.0121 (0.0022)∗∗∗180surf_built_neigh181−0.0760 (<0.0000)∗∗∗182−0.0786 (0.0049)∗∗∗183height_built_neigh1840.1493 (<0.0000)∗∗∗1850.1485 (0.0053)∗∗∗186social_compo_G1187Ref188Ref189social_compo_G21900.0279 (0.0006)∗∗∗1910.0318 (0.0094)∗∗∗192social_compo_G31930.1560 (<0.0000)∗∗∗1940.1615 (0.0071)∗∗∗195BM1960.1946 (<0.0000)∗∗∗1970.1962 (0.0096)∗∗∗198V199CatNat2000.0748 (<0.0000)∗∗∗2010.075 (0.0066)∗∗∗202PPRI203−0.1143 (<0.0000)∗∗∗204−0.1153 (0.0158)∗∗205CatNat∗PPRI2060.0583 (0.002)∗∗∗2070.0532 (0.0178)∗∗208T209T2011210Ref211Ref212T20122130.0092 (0.1435)2140.0148 (0.0061)∗∗∗215T20132160.0160 (0.0103)∗∗2170.0223 (0.006)∗∗∗218T20142190.0136 (0.0353)∗∗2200.0152 (0.0062)∗∗∗221T20152220.0177 (0.0046)∗∗∗2230.0209 (0.006)∗∗∗224T20162250.0550 (<0.0000)∗∗∗2260.0577 (0.006)∗∗∗227Lambda (p)2280.1609 (<0.0000)∗∗∗229N23024,871231I moran on residuals (p-value)232−0.0069 (0.8047)233Frontiers in Environmental Economics23406235frontiersin.org236237238===== PAGE 7 (score 14, numeric-paren 0, table-words 0, var-words 14) =====239240Dachary-Bernard24110.3389/frevc.2025.1603746242more than one bathroom, a garage, and the better its condition,243the higher its price is. In addition to these structural variables,244temporal variables also prove to have a positive effect on prices245(taking 2011 as the year of reference), which relates to the general246price increase seen in the real estate market over that period in this247area (A’Urba, 2018).248Regarding the property’s environment (variables group E),249the distance to the Bordeaux urban center has a significantly250negative (but very low) effect on prices meaning that the prices251decrease with the distance to city center. Conversely, we notice252positive coefficients for the variables of distance from the river and253distance to the main road. The dichotomous BM variable is highly254significant in the model, which means that the metropolitan effect255is captured into prices. Several other variables that refer to the256environment surrounding the property affect prices. Thus, a large257green area, a small built area, and a significant average building258height within 200 m of the house, will increase its value. The market259therefore appears to valorize a green and dense neighborhood.260The social environment also affects house prices through positive261and significant effects relating to the social composition of the262neighborhood. Socio_compo_G2 and Socio_compo_G3 have highly263significant and positive coefficients. It illustrates the search for264neighborhoods that are socioeconomically homogenous and above265all without diversity (this mixed composition is represented by266the reference level for social composition). Finally, the isolation267index has a positive effect on prices. All other things being equal,268the market valorizes isolated properties but it should be noted269that the interaction term between this isolation index and the270distance from the road variable has an overall negative effect. It271means that people valorize isolated houses, but that a property272far from the main transport infrastructures will be less expensive,273all other things being equal. People look for isolated housing but274easily accessible.2753.2 Flood risk276As far as the results specifically relating to vulnerability to277flooding are concerned, the two variables, and their interaction,278reveal significant effects on house prices, albeit in different ways.279The three flood components of the hedonic equation are highly280significant. In the presence of the interaction term, the coefficients281γ and δ are marginal effects and must be interpreted as holding282the other variable equal to zero. For example, γ is the impact of283being located in the PPRI zone of a municipality that has not284experienced a high number of CatNat. Similarly, δ captures the285impact of being located in a municipality with a high number286of CatNat outside a PPRI zone. The interaction term θ informs287about the marginal effect of being located in the PPRI zone of288a municipality that has experienced a high number of CatNat289flood events.290PPRI variable has a negative and significant effect ( ˆγ =291−0.1143) on prices. This means that all other things being equal,292a property located in a flood area in a municipality that has not293experienced many CatNat events is less expensive than a property294located outside that zone. The discount due to a flood zone295location is ∼11% in municipalities that have not experienced many296CatNat events.297The CatNat variable, on the other hand, has a positive298effect on price (ˆδ = 0.0748), which means that all other things299being equal, the fact of being located in a municipality that300has experienced numerous floods recognized by decree as being301natural disasters leads to higher prices for houses that are not302located in flood zones. At first glance this is a counter-intuitive303result that requires an explanation. It needs to be interpreted in304relation to the compensative dimension that the variable CatNat305conveys. Indeed, recognition of a flood event as a CatNat leads306to financial compensation. The market would therefore appear307to value significant activity by a municipality in terms of post-308flood crisis management. According to the estimations, this CatNat309variable leads to an appreciation in property value of ∼7.8%.310Finally, cross-referencing the two flooding variables allows311us to capture their interaction effect. This interaction effect is312significantly positive ( ˆθ = 0.0583). It means that people positively313value being in the PPRI area of a municipality that has had314numerous CatNat events. This marginal effect is positive and may315be qualified as an “additional premium”. The impact of CatNat is316therefore almost twice that in a non-flooded zone, i.e., almost 12%317of the value of the property. In the same way, and symmetrically,318comparing with the PPRI variable, the discount is therefore reduced319to 5.6% of the value of the property: the drop in value due to being320located in a flood zone is lower when the municipality has shown321itself to be in a position to trigger aid.3224 Discussion/Conclusion323The results relating to stylized facts highlight the presence324of the traditional process of peri-urbanization driven by the325search for accessibility to the city center and to green amenities326in the immediate vicinity of the home. On the other hand, as327the literature has shown (Travers et al., 2009), proximity to the328main transport infrastructures appears to be mainly perceived as329nuisances which outweigh the advantages in terms of accessibility.330This point requires qualification regarding isolated housing for331which access to the road network is a sign of accessibility. The332phenomenon of socio-spatial segregation often highlighted in333contemporary urban forms would seem to be confirmed here,334with the search for homogeneous and non-mixed socio-economic335environments (Baumont and Legros, 2013). This result had already336been demonstrated on the smaller scale of the Bordeaux metropolis337alone (Décamps and Gaschet, 2013). Moreover, given the broader338metropolitan scale that we have chosen, our results also ask339questions about the emergence, parallel to this process of peri-340urbanization, of a process of urbanization. This is characterized341by a growing attractiveness for spaces far from the metropolis,342spaces that will (re-)become attractive in their own right (Bailly and343Bourdeau-Lepage, 2011).344However, alongside these more general processes, our main345results regarding the flood risk variables focus on two important346phenomena that must be taken into account when considering the347strategy for adapting cities to climate change.348The first phenomenon concerning the PPRI zoning, which is349standard in the literature, relates to people risk aversion. Indeed,350locating in a flood prone area is negatively valued, meaning that351people tend to move away from the risk and choose not to live in a352Frontiers in Environmental Economics35307354frontiersin.org355