A leading researcher says there's an urgent need to understand New Zealand's "tragedy on the horizon" - climate-driven costs that could hit coastal properties and banks.
Tens of thousands of homes that currently sit within a kilometre of the coast are likely to be hit by a combination of sea level rise and more frequent and intense storms under climate change this century.
One report last month predicted thousands of homes in our main four centres could face soaring insurance premiums - or have some cover pulled altogether - within only the next 15 years.
In a major new study pulling together experts across geology, climatology, banking and real estate, Otago University's Associate Professor Ivan Diaz-Rainey will take a wider look at the crisis, exploring when and to what extent it will impact property values.
"The 'when' is a really critical question - that depends on what [climate change] scenarios play out, but right now, we want to get some bounds around how house prices are going to be affected," Diaz-Rainey said.
"When South Dunedin was flooded in 2015, for instance, at first there was a 15 per cent discount on those affected houses, before it recovered to around 4 per cent over the long term.
"But obviously, if these properties start flooding frequently, and there are no defences, then the effect is different."
Models already indicated that houses were expected to sell for discounts in areas with higher risks of damage from natural disaster, such as flooding.
That lower price compensated for the higher costs of insurance and repair.
One analysis suggested that, in areas with a one-in-100 year flood risk, the average steady-state sale price discount was about 4.6 per cent.
That of course changed as the return period for flooding narrowed eventually to the point where homes were being permanently inundated.
The current average holding period of a New Zealand house was about 7.4 years - but those homes imminently threatened with permanent flooding were likely to become stranded assets more quickly than that period.
Diaz-Rainey said that raised big implications for the banking sector - and his study would investigate the potential thresholds for state changes, such as banks withdrawing lending.
The state changes themselves could trigger a fire-sale of assets, as homeowners sold up before their assets were stranded, or simply walked away from homes with negative equity.
Given more than 60 per cent of the four largest banks' loan portfolios are made up of housing loans, even a relatively small proportion of stranded assets could threaten financial stability.
"The big central banks around the world are now taking this extremely seriously, as is the Reserve Bank of New Zealand who are a partner in our project," he said.
"In New Zealand some banks have 90 per cent of their assets in domestic real estate. But this is an evolving situation, and it really depends which pathway we go down."
That was where the other major component of the project, supported with an $869,000 Marsden Fund grant, would add more clarity.
A sweeping assessment, released by the Government last year, set out scenarios New Zealand faced based on emissions rising at the current rate, and resulting in a projected 67cm of sea level rise, and 3C of temperature increase, by 2090.
It found about 675,500 Kiwis lived in areas already prone to flooding, with a further 72,065 living in the firing line of where some of the most dramatic effects of sea level rise could hit.
Buildings, too, were at extreme risk – nearly 50,000 of them were currently exposed to coastal flooding, and at the highest range of warming scenarios, that could rise to nearly 120,000 this century.
One of its authors, Niwa coastal scientist Dr Rob Bell, said New Zealand likely only had a few decades to get proper plans in place to prepare for rising seas.
Diaz-Rainey said much of the modelling to date had drawn on what's called "bath tub" modelling, which assume a uniform effect of sea level rise rather than taking account of other dynamics like actual ground water levels.
"These models can both over-estimate and under-estimate the risk. What we've done in a pilot study is compare the bath tub model to a geometric model based on actual measurements of groundwater," he said.
"We found the areas that get flooded are totally different. The overlap between them is very low in moderate scenarios, but then in high-emission scenarios with runaway climate change, everything gets flooded. So, geology matters."
Based off that pilot study, which was focused on South Dunedin, Diaz-Rainey and his team will develop refined sea level rise and flooding models accounting for a wide range of geological factors, which they'd use to look at effects on property.
Another unique part of the three-year project was its use of mātauranga Māori, or Māori knowledge, with iwi eventually helping to identify threatened places that are also wāhi tūpuna, or culturally important.
Diaz-Rainey expected the climate threat would hit properties and banks with both quick shocks, like those caused by Canterbury's earthquakes, but also gradual impacts, as seen with the slow build-up to a leaky homes crisis that ultimately cost $47b.
"Cumulatively, we need to start to get a better sense of what the bounds are. From the conversations we've had with banks already, those big questions of when and to what extent are definitely of interest to them."
The Government was currently working on a national adaptation plan, setting out how to make communities more resilient to climate change.