Floods do not sink prices, historical memory does: How flood risk impacts the Italian housing market †

Anna Bellaver ⋆,, Lorenzo Costantini ⋆,, Ariadna Fosch ⋆,, Anna Monticelli, David Scala, Marco Pangallo ‡, · 2025

Read Floods do not sink prices, historical memory does: How flood risk impacts the Italian housing market † by Anna Bellaver ⋆,, Lorenzo Costantini ⋆,, Ariadna Fosch ⋆,, Anna Monticelli, David Scala, Marco Pangallo ‡, (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2502-12116.txt.

Introduction

Bellaver, Costantini, Fosch, Monticelli, Scala, and Pangallo ask in twenty twenty-five whether single floods cut house prices in Italy, or whether repeated flooding does. Italy is highly flood-prone but unevenly, which makes it a good test case. The team built a dataset of over half a million mortgage-backed sales over eight years and applied price models plus a before-after design.

Single floods did not lower prices in at-risk areas. Repeated exposure did, with drops of a few percent in the most flooded regions. Young buyers won no discount in risky areas, while seasoned buyers did. Buyers in risky areas also had lower incomes. The authors stress memory and institutions in how risk prices in.

Catalog record

CREATORAnna Bellaver ⋆,, Lorenzo Costantini ⋆,, Ariadna Fosch ⋆,, Anna Monticelli, David Scala, Marco Pangallo ‡,
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE113 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2502-12116

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Floods do not sink prices, historical memory does: How flood risk impacts the Italian housing market † Anna Bellaver ⋆,, Lorenzo Costantini ⋆,, Ariadna Fosch ⋆,, Anna Monticelli, David Scala, Marco Pangallo ‡, Abstract Do home prices incorporate flood risk in the immediate aftermath of specific flood events, or is it the repeated exposure over the years that plays a more significant role? We address this question through the first systematic study of the Italian housing market, which is an ideal case study because it is highly exposed to floods, though unevenly distributed across the national territory. Using a novel dataset containing about 550,000 mortgage-financed transactions between 2016 and 2024, as well as hedonic regressions and a difference-in-difference design, we find that: (i) specific floods do not significantly decrease home prices in areas at risk; (ii) the repeated exposure to floods in flood-prone areas leads to price reductions, up to 4% in the most frequently flooded regions; (iii) low-income buyers are more likely to settle in flood-prone areas. 2 2 footnotetext: We would like to thank Paolo Bajardi, Diana Bruni, Matteo Coronese, Paola Danelutti, Adriano di Fabio, Barbara Lenti, Jacopo Lenti, Laura Li Puma, Michela Manuelli, Carla Monferrato, Yamir Moreno, Silvia Ronchiadin, Simone Scarsi, and Fabio Verachi for their helpful comments at various stages of this work, as well as seminar participants at EAERE and IAERE, where this work was selected among the top 5 presentations by young researchers, to be included in a special session at the Italian Economic Society 66th Annual Conference. 1 1 footnotetext: Equal contribution. 3 3 footnotetext: Corresponding author: [email protected] JEL Codes: D83, Q51, Q54, R31 Keywords: Flood risk; housing markets; hedonic regressions; awareness 1 Introduction Research on the impact of flood risk on home prices, primarily based on U.S. data, typically finds small impacts, except in the years immediately following major flood events, when prices substantially decline in affected areas. For example, prior to Hurricane Sandy, flood risk had little influence on home prices in New York City. However, in the years after the storm, properties in flood zones –despite being undamaged– sold for up to 8% less than properties in safer areas ( Ortega and Taṣpınar, 2018 ) . Bin and Landry (2013) attribute this pattern to the availability heuristic ( Tversky and Kahneman, 1973 ) , suggesting that individuals must experience floods firsthand to perceive them as a significant threat and demand a price discount for homes at risk. In this paper, we consider an alternative yet complementary hypothesis, which we term historical memory . We suggest that, in places with storms not as destructive as hurricanes and different housing market institutions than in the U.S., isolated flood events may be insufficient to significantly shift market beliefs and prompt the pricing of flood risk properties. Instead, repeated flood exposure in the same region may have a more substantial influence. If this hypothesis is true, home prices in frequently flooded areas are likely to reflect flood risk more prominently than those in rarely flooded areas, even for properties that remain undamaged. We operationalize the concept of historical memory by constructing a measure of buyers’ awareness of flood risk. This measure rises following each flood event but gradually declines as memories of past floods fade. A single flood, even if severe, may not raise awareness sufficiently for flood risk to become reflected in home prices, as buyers might perceive it as an isolated incident. In contrast, repeated exposure to flooding heightens awareness, making it more likely that flood risk is capitalized into property values. Our measure offers a concrete implementation of the ideas proposed by Tobin and Newton (1986) and Pryce et al. (2011) , enabling empirical testing with real-world data. To test this hypothesis, we conduct the first systematic study of the Italian housing market, which is an ideal case study for several reasons. First, Italy faces significant flood risk, with over 10% of its population and buildings exposed to flood risk ( Trigila et al., 2021 ) , and approximately 40 major floods recorded since 2000 ( EM-DAT, 2024 ) . Second, Italy recently experienced a major flood in the Emilia-Romagna region on May 16–17, 2023, which displaced 45,000 people and caused damages estimated at around 10 billion euros ( EM-DAT, 2024 ) . Third, flood exposure varies across regions: while flood risk affects the entire country, Northern Italy experiences more frequent events than the South. Fourth, unlike markets in the U.S. or the U.K., Italy lacks institutional requirements for flood risk disclosure during property sales and neither mandates nor promotes flood insurance. Finally, floods in Italy typically result from storms with a less catastrophic impact than the hurricanes or tropical storms common in the U.S. Given this environmental and institutional context, we frame and test our historical memory hypothesis as follows. We argue that, without an institutional structure that encourages the sharing of information regarding flood risk, households rely on the historical memory of major floods to identify the most at-risk areas. Due to the relatively limited destructive potential of individual floods, a single event may not significantly shift market beliefs. However, repeated flood occurrences gradually reinforce risk awareness, eventually leading to its capitalization in home prices. The recent Emilia-Romagna flood provides an opportunity to test the first aspect of this hypothesis, while the variation in flood risk across the Italian territory enables us to examine the second. The main reason why there is so little research on how flood risk impacts the Italian housing market is that systematic data only started to exist in the last few years. 1 1 1 We are only aware of one study assessing the impact of flood risk on prices ( Rosato et al., 2017 ) , using 197 data points manually collected from real estate agencies in Venice. Only recently systematic data about the housing market became available from the housing advertisement website Immobiliare.it ( Pangallo and Loberto, 2018 ; Loberto et al., 2022 ) , but concerning our research question, these data have only been used to study the impact of a sea wall in Venice ( Benetton et al., 2023 ) or floods in Emilia-Romagna ( Gentili, 2025 ) (the study by Gentili (2025) is posterior to our work, which was first posted online in February 2025). In this paper, we take advantage of a novel database that contains the near-universe of mortgages issued by Intesa Sanpaolo, the largest Italian bank. Our final dataset contains about 550,000 mortgages issued from 2016 to 2024, which is about 15% of Italian mortgage-financed transactions, and 10% of total transactions. 2 2 2 The data are highly representative of the Italian housing market throughout the entire national territory, see Supplementary Section A.2 . For each transaction, we know the sale price, the date, the income of the buyer, a proxy for whether the buyer is under-36, and several physical characteristics of the transacted home, including its geographical coordinates. We use the coordinates to determine whether a home is inside a flood-prone area by performing spatial matching with risk maps provided by ISPRA, the Italian Institute for Environmental Protection and Research ( Trigila et al., 2021 ) . Using the same approach, we also assess whether

Read the complete text