05: shaker-styles built on sand, or the trap of flood insurance

For many in coastal communities, storms are a way of life. We used to recognize that more explicitly in our architecture. Plank wall bungalows used to dot the coast, and face significant damage over the course of seasons, only to be rebuilt with little additional cost and little environmental footprint. One of the developers chronicled reported that their family "never quite felt safe in their first waterfront home", the windows rattled and the roof leaked. 

Over time, that need for security resulted in bigger houses, more hardscaping and infrastructure. The demand for coastal living of this kind meant that wetlands were filled as well. From 1940-1970, developers dreged 1/3rd of the sediment in Back Bay to build foundations for homes. All of this increased the vulnerability of the coast and the dollar value of damages that result from severe storms.

Broadly speaking, many stakeholders recognize that coastal building is problematic. Since the 1960's the federal and state governments have been advocating for more measured growth in coastal areas, or at least they've been paying lip service to that end. There is limited ability to create seamless policy from the federal level all the way down to the local, however, and every time the opportunity arises, it seems too politically costly to do so.

For example, in 1962, the Ash Wednesday storm struck the northeast. As a system, it was larger than Superstorm Sandy and Hurricane Katrina, both massive storms. There were five storm surges, inducing decades worth of erosion and coastal change overnight. At that time, the Governor of New Jersey proposed a building moratorium within 100 feet of the water UP AND DOWN THE ENTIRE EAST COAST. Additionally, the state greenlit $60M in funds to repurchase homes in flood prone areas. In my limited understanding, this is more money on a nominal basis than has ever been spent rebuying homes in US history. Politics intervened, however, and President Kennedy caved to popular demand (led by the mayors) and declared a federal emergency. 

Many attempts were made after this to direct local policy via state and federal incentives, but many of these efforts were circumvented (New Jersey attempted to take control of zoning for 25+ unit buildings, instead people built everything to 24 units or less) or poorly rolled out (federal flood insurance was never fully considered, rolled out with poor pricing and worse maps, and it's gotten worse from there).

Local authorities tend to favor increased development for the obvious reason that it flows through to their tax base and the less obvious reason that many coastal mayors, or those conveniently chosen for this study, have significant personal stakes in development projects.

Their inclination for capital investment has been heavily supported over the past century by the increase in mortgage accessibility and the pervasion of federal flood insurance. Both forms of leverage are heavily subsidized by the federal government. 

While mortgages are quite well known to be subsidized through government guarantees, flood insurance enjoys subsidies in three ways. First, it is mispriced due to poor mapping. 20% of flood damages come from map sectors showing no risk. Second, it is back stopped by treasury. The flood insurance program is currently $24B in debt to taxpayers. Third, it must accept all comers. The federal flood insurance program handles 70% of all outstanding flood policies in this country, and that percentage is drastically skewed toward high risk properties. For example, Louisiana has paid roughly $4B in premiums as of 2019, and received more than $19B in payouts.

Despite the high level stance that coastal footprint should be managed more prudently, the terms of access to this leverage have barely changed.

In the event of a natural disaster, flood insurance funds are tapped as well as state and local resources, but sizable funds are unlocked if a federal emergency is declared via presidential disaster distinction (PDD). The threshold for a distinction is typically $1 in damages per capita in a given population area (typically a state, but can be a county). This number is quite low – it's also unadjusted for inflation – and as a result, 1,300 PDDs have been declared from 2009-2019.

As mentioned in the previous post, federal funds are spent before local matches, essentially granting implicit leverage to coastal rebuilding efforts. It's also important to note that this distinction also creates perverse incentives against the purchase of flood insurance. Though required by law and by lenders, those who own homes or storefronts outright might avoid the purchase of insurance, knowing that any disaster that reaches their structure will likely bring with it a PDD.

This is a somewhat cynical take on disaster recovery as PDDs are critical in many communities to returning to a normal way of life. Gaul somewhat skillfully contains most of his inquiry to extremely affluent communities in the coastal northeast, revealing to the reader the subsidy granted at the expense of exploring where it is most productively used.

As with all forms of leverage though, it's most effective in the most liquid, high-value markets. From a market value perspective, second homes represent a bulk of total property at risk. In 1960, all of the property on Long Beach Island was valued at $100M. Today, it's worth more than $15B. From a wealth creation perspective, it seems quite logical to foist a lot of these problems onto the federal government.

In the wake of a disaster, paperwork is extreme. Those that have the capital to pre-fund repairs and hire lawyers or lobbyists seem to come out ahead. Cities commission armies of clerks to file for every program that might be a fit. New Jersey is exceptionally good at getting federal infusions for boardwalk renewal and shoreline restoration because they've been at it the longest.

As a result, something of a shoreline industrial complex has emerged. It's relatively simple to get funding to repair assets already in place, so developers build despite the risk, and then make the argument that "what's done is done, what's built is built." This works well inside a structure that thinks tactically, but not strategically. Despite generating 3% of New Jersey's jobs, billions of dollars are plowed into the shoreline economy because of the way that it flows through to construction contracts and real estate values.

This paradox applies doubly to beach rebuilding, a particular skill of the Army Corps of Engineers. Unfortunately, they're not in the business of building a "durable coastline", so much as they are "the one that existed before the last storm". Beach rebuilding is notoriously fragile to storm surge, tending to wash away. As a result, it requires extensive maintenance. On Long Beach Island alone (year round population 10,000) the Army Corps of Engineers expects to spend $500M in beach rebuilding cost, or $32M per beach mile.

A frequent refrain is that "there are no conservatives in a crisis." That's exceptionally true when it comes to natural disasters. In 2017, hurricane related losses totalled $300B and most of that money went through Texas Senator Ted Cruz, otherwise an advocate for the free market and against government spending.

As mentioned, federal flood insurance is currently $24B in debt to the US Treasury (including $16B in debt forgiveness from the Trump Administration). Congress sought to rectify that with the Biggert-Waters Act of 2012, which would redraw flood maps to accurately reflect risk. The act lasted two years before public outcry against the cost of (accurately priced) flood insurance led legislature to pass another act returning subsidies.

The political nature of these decisions is also reflected in what major stakeholders don't want to know. When asked how many of the homes insured by the National Flood Insurance Program were vacation homes, FEMA said they couldn't access that information as it would require a "costly computer search".

Most people struggle to understand the risk they're creating or allowing. One mayor only started to appreciate sea level rise when rungs on his dock started disappearing. Others admit that they're not ready to consider a different kind of life, let alone the types of damage their current lives might be doing.

Above all, we struggle to imagine the future of our climate. Scientists argue we've never really seen a black swan hurricane because all of the past storms could be interpolated from historical data. Houston planners argue they " don't think we're equipped as a society to make the decisions we're being asked to make about storms that we haven't seen."

Aggressive projections place $500B of property at risk of being below sea level by 2100. Our storm projections continue to be too conservative, but these home losses would be greater than those experienced in the GFC. Freddie Mac economists agree.

Five years ago, I would have said that this was a substantial problem, and then we printed trillions to deal with COVID. It will likely happen again. New Jersey is already expecting that it will take $1-2B to raise 12,800 vulnerable homes on beach islands. Retrofitting might be more expensive than rebuilding.

A recurring theme in these explorations is a high level advocacy from federal and independent authorities for managed retreat. Programs are put in place, and then proven impossible to use. Dauphin Island, a spit of land in the middle of the Gulf that houses exceptionally beautiful but vulnerable homes, attempted to sell some real estate back to FEMA. Two years after their application they were told that the properties didn't qualify. The average time for a federal repurchase is five years.

All in all, this reality creates a scenario in which people can't afford to stay if flood risk is priced appropriately, and they can't afford to leave. So they stay, but anxiously so. 

By 2100, it's estimated that 80M Americans may have to move due to rising sea levels.

I'm wondering how this all flows through reinsurance markets. In many cases, reinsurance affects wind risk more than flood risk, but at times, those can be one in the same. Interestingly, Florida (which has far more wind than flood damage) is one of the only states to have paid more in flood risk premiums than it has received in payouts.

Gaul frequently points to the fact that permits for new housing are being approved in tracts that also have funds available for voluntary acquisition. Wodnering if old houses are being bought, and new houses are more hardened against storm risk. The failure to change local zoning laws would indicate that's unlikely, but it would invalidate some of the concern about the growth rate in the liability.