PurposeCanterbury, New Zealand, experienced two significant earthquakes in 2010 and 2011 with a devastating impact on both houses and land. Negative media attention to the potential financial risks of living near or on the new Technical Category 3 (TC3) land or on land in a flood zone has fuelled the perception of uncertainty over the negative property price impacts. This research aims to determine if residents’ perceptions of the risks associated with various types of land zones (e.g. TC1, TC2 and TC3) are reflected in property prices.Design/methodology/approachThis research analyses sale price patterns and the relationship between sale prices and house characteristics before and after both earthquakes. A three-step approach was taken by applying: an average trend analysis, Geographic Information Systems’ (GIS) hotspot analysis to identify possible spatial differentiations between the before and after-effects of the earthquakes and hedonic modelling to quantify the effect of house characteristics on sale price while controlling for and comparing three land zones (TC1 to TC3).FindingsThe data suggest that average sale prices increased after both quakes in TC1 and TC2 in contrast to TC3 zones, while close to 8,000 structures were demolished in red zones from 2010-2013 (supply was reduced). The econometric modelling suggests that higher sale prices are achieved by: newer houses across all land zones and more recent sale agreements only in TC1 and TC2 zones. Other observations include the effect of certain exterior façade materials on sale prices on the overall data set and in the individual TC1 and TC3 zones. In conclusion, the results suggest that although caution might exist for the TC3 zone, the quality of the house can override the stigma attached to the TC3 zones.Research limitations/implicationsA confounding factor in the research was that approximately 7,800 homes were rezoned red and/or demolished between 2010 and 2013 changing the supply and demand balance. Further, banks and other lenders updated their requirements for new lending on properties in the Canterbury region, requiring a number of reports from professionals such as structural engineers, geotechnical engineers and valuers before any new lending would be approved. Additionally, immediately after the September and February earthquakes, there was a 21-day stand-down period for earthquake-cover in Canterbury and without adequate insurance cover banks would not advance mortgage money, causing a short-term slowdown in the residential property market.Practical/implicationsThe outcomes of this research will be of interest to government agencies tasked with assessing compensation for affected property owners. For example, the Earthquake Commission (EQC) developed aDiminution of Value Methodology for Increased Flooding Vulnerabilitythat formed the basis of a High Court declaratory judgment decision in December 2014 that cleared the way for the EQC to start settling properties with increased flooding vulnerability. The EQC methodology was informed by the results of similar studies to this one, from around the world. Homeowners and rating valuers will also be interested in the results to understand how house prices have been affected by market perceptions towards earthquake damage, particularly in the worst-affected areas.Originality/valueThis study fills a research void regarding the price impacts of residents’ perceptions of the risks associated with various types of land zones that reflect the expected future liquefaction performance of the land.