Flood cat bonds? Flood sidecars? How can the NFIP de-risk using ILS?

As the National Flood Insurance Program (NFIP) increasingly looks to risk capital, from traditional reinsurance and in the future capital markets, in order to de-risk itself and better protect the taxpayers who ultimately back it, it’s perhaps more likely we see ILS structures in the form of a sidecar than a flood catastrophe bond, S&P says.

The need for the NFIP to reduce the amount of risk it holds is clear, with the softened price of reinsurance capacity currently providing it with the opportunity to lay off risk with reinsurers and ILS funds through a traditionally placed reinsurance program.

This process began last year, when the NFIP visited the reinsurance markets for the first time, resulting in the Federal Emergency Management Administration (FEMA) making a $1 billion purchase of reinsurance protection for the NFIP at the January 2017 renewal, which was a major leap in progress.

That initial program was said not to have included any directly collateralized coverage from ILS funds, although there could be fronted ILS capital included, it’s difficult to say. But either way, the capital markets was not a focus for the NFIP’s first reinsurance placement.

With legislative efforts to reauthorise and reform the NFIP all seemingly mentioning reinsurance, risk transfer and in many cases the capital markets or insurance-linked securities (ILS) as opportunities to de-risk the flood insurance pool further, it seems inevitable that the ILS fund market and ILS investors will play a greater role at future renewals.

But flood catastrophe bonds may not be the first structure we see, beyond just collateralized reinsurance participation, according to rating agency Standard & Poor’s, with the agency highlighting the potential for innovative traditional re/insurers to bring capital markets capacity alongside their own in the form of flood reinsurance sidecars.

S&P notes in a recent report that there remain “significant challenges” for private capital to participate in the NFIP due to both a “regulatory and risk modeling perspective.”

The fact that the reinsurance market is clearly interested in taking on more flood risk from the NFIP is positive, S&P says, and could help to support greater private market flood insurance offerings coming to the market. But despite this the rating agency says it doesn’t expect a rapid uptick in this, at least in this calendar year.

That said, when the private market participation in U.S. flood risk does ramp up, S&P expects this will be from reinsurers in the main, with primary insurers participating to a lesser degree and likely on a well-reinsured basis.

“We believe reinsurers will lead the charge by providing the means for NFIP to manage its existing exposure as well as by extending their capabilities and balance-sheet capacity in partnerships with primary insurers,” S&P says.

The soft reinsurance market means that for reinsurers flood risk can be seen as an avenue of growth, but S&P expects them to be cautious, limiting their exposure to this risk.

S&P questions whether the NFIP may push for private capital participation purely by leveraging the reinsurance and capital markets for tail-risk protection, or whether it moves towards a full privatisation model of insurers at the front end and reinsurers at the back.

For reinsurers and ILS funds it would likely be preferable, in the long run, to have the primary insurance market taking an increasing amount of risk out of the NFIP, while they still reinsure the program and those insurers as well.

This model, which would be more akin to the way Florida Citizens has so successfully derisked itself, means the amount of risk seeping into private markets could grow more rapidly, as primary insurers would likely make heavy use of reinsurance capital, at least to begin with.

But it also means that both insurers and reinsurers would be working to better understand the risk, and the NFIP would likely become the risk carrier of last resort, which is perhaps what it was always supposed to be.

However, S&P believes the use of ILS to transfer the NFIP’s risk to the capital markets could be a way off, saying; “At this point, we don’t believe the prospect of using ILS in and of itself is a realistic proposition.”

“Usually ILS/CAT bonds are targeted for well-understood natural peril risk for which third-party models are fairly well developed, which we do not believe is the case today for flood risk,” the rating agency continues.

However, S&P says it is not dismissive of capital markets participation, as; “We envision there’s potential for sidecar deployment by a reinsurer to access third-party capital, as these reinsurers would have their own money at risk and can provide greater comfort to such investors.”

So S&P feels that reinsurance firms will lead the assumption of risk from the NFIP, with capital markets players and ILS funds participating behind them, in structures akin to sidecars.

The rating agency does note that ILS structures could be deployed at a later stage, but it clearly does not feel that the catastrophe bond market is ready to assume the NFIP’s risk.

Why would this be?

The lack of robust risk modelling has clearly been an issue, but in recent years the main modellers flood risk analytics offerings have improved dramatically and continue to do so.

Additionally, we have seen flood catastrophe bonds in the past. Triggers (parametric in nature, for flood and surge flooding) have been designed that ILS investors and fund managers have been able to get comfortable with. We’d imagine for the right layer of risk this could be repeated, with even more targeted triggers today, potentially parametric.

Index-triggers also present an option, with industry loss data for flood catastrophes already reported on by the main providers, it seems that a flood cat bond could be structured around, say, a PCS catastrophe loss trigger for a flood specific event.

So there are options and some of these options have been accepted in the past by ILS investors, so why not again?

There’s also the question of indemnity triggers and if an ILS fund would accept indemnity exposure to the NFIP through a collateralized reinsurance arrangement, which no doubt some would right now, why not through a cat bond structure?

ILS fund managers would happily participate in the traditional reinsurance renewal for the NFIP, alongside and on the same terms as the major reinsurers. So why not carve off a little of this traditional program, securitize it even privately and issue the first NFIP flood cat bond to a select group of the most experienced ILS fund managers? That could set a precedent for securitization becoming a major source of reinsurance capital as the NFIP de-risks.

Looking further to the future, the development of sensor technology and the interest in InsurTech could present a real opportunity for flood insurance triggers to be technology based, perhaps on river levels in flood prone areas, using sensors attached to buildings, satellite and remote sensing tech for water inundation (already used in weather-index coverage), and other innovations.

The transparency of such future, technology based, flood cat bond triggers could be an extremely efficient way for the NFIP to access risk capital and reinsurance capacity from the capital markets.

But right now, S&P is likely right that the majority of the capital markets and ILS funds participation in assuming the NFIP’s risk will likely be through the traditional reinsurance program, either as S&P says in a sidecar structure, or perhaps more likely as fully-collateralized traditional reinsurance arrangements.

But in time the catastrophe bond market definitely presents a real opportunity to FEMA as it looks to shift flood risk away from the NFIP and taxpayers. It really would be a good idea to integrate a small catastrophe bond issue in the next NFIP reinsurance renewal, as this will lay the groundwork for much greater cat bond investor participation in flood risk for the coming years.

Join Artemis in Singapore on July 13th 2017 for ILS Asia, tickets on sale here