The title of this blog entry may seem a particularly dumb question. Coastal legislators, the simple answer proceeds, want the Texas Windstorm Insurance Association to keep rates low because it helps their constituents’ pocketbooks. But is that really true? Or, more precisely, under what circumstances would that be true?
The first thing to recognize is that low premiums have a somewhat different effect when they go to Allstate than when they go to TWIA. One can quite understand policyholders not wanting to pay Allstate very much money because its profits simply go for the benefit of Allstate’s shareholders. TWIA, on the other hand, has no shareholders. To the extent TWIA collects “too much” in premiums, that is extra money available to pay future claims of future coastal policyholders. Over collection by TWIA favors future coastal residents in a way that over collection by Allstate does not.
One can get at a better answer to this question by asking why TWIA insists that policyholders pay much or, indeed, anything in advance premiums at all. Since the risk of tropical cyclones is difficult to quantify, why not just have a post-disaster assessment scheme? It would work as follows: when we know how large the damages are from some hurricane relative to TWIA cash on hand (which might be zero if there were no premiums), TWIA simply “post-assesses” its policyholders to pay in that much money over some period of time. We wouldn’t need no stinkin’ models; we’d just see what happens. Take imaginary Hurricane Barry that creates claims (plus lost adjustment expenses) of $1 billion above any cash TWIA happens to have on hand. Because TWIA has about $70+ billion of property under its control, if TWIA could spread the payback period over 5 years, some back of the envelope math suggests that the consequence of this scheme would be that someone with $200,000 worth of coverage would have to pay roughly $571 per year to take care of the claims caused by Barry.
This proposed post-assessment alternative sounds pretty good. I bet many TWIA policyholders with $200,000 properties would be delighted to pay only $571 per year for tropical cyclone coverage.
Unfortunately, the proposed alternative won’t work as inexpensively as described. This baseline figure of $571 is a significant underestimate. There are two reasons: interest expenses and post-assessment shrinkage.
The interest issue is easy to understand. Few policyholders with demolished properties want to wait 5 years living in a tent or in a home decorated with blue tarp until all the assessments are collected before getting their claims paid. So, to pay for Barry, TWIA has to borrow the $1 billion. But now TWIA policyholders don’t have to just pay back $1 billion, they have to pay it back with interest. Interest expenses alone can raise the annual payback amount substantially.
There’s also a shrinkage problem. Once the price of being a TWIA policyholder jumps due to the assessment, TWIA will no longer be assessing the amount of property it did prior to the storm. Thus, to continue with our example, to compute the necessary assessment, TWIA won’t be dividing the amount of damages by the $70+ billion; the denominator will be smaller. And this means that the assessment, because it will be spread over less property, will have to be larger. But the larger assessment means that yet more people will drop out of TWIA, which means that the assessment will have to be yet larger. One can’t easily say whether and if this spiral will stop, but one can be confident that the size of the assessment will increase significantly due to both the interests costs on the loan and due to the reduced size of the TWIA pool.
Footnote here. There is another way to run an assessment scheme: don’t assess future policies; assess policies in place at the time the disaster struck. Such a method avoids shrinkage as I’ve described it, but it creates another problem: collection. Instead of just cancelling policies that don’t pay assessment, TWIA now has to actually track down over a period of many years former policyholders some of whom just will not have the money to pay what is due. That’s expensive and will not yield a 100% recovery rate. And, as I read the Texas statute, it’s not what’s contemplated in the event of a large storm. Anyway, whether done by assessments on future or in-place policies, post-assessment has problems.
I don’t think it’s unreasonable to believe that for something like Hurricane Barry, interest and shrinkage would raise the post-assessment by about 25% over its baseline cost. But no one knows this number with any certainty. And this means that there is a substantial risk that the scheme won’t work or, that conservative lenders, will worry that it won’t work and will accordingly charge very high interest rates. Indeed, TWIA appears to have discovered that, although the Texas statute will let it borrow $1 billion following a disaster to be repaid with policyholder assessments, the market appears willing only to lend $500,000. So, post-assessment schemes are risky.
There’s an additional problem, however. And, this one is a killer. I just provided an example with a $1 billion Hurricane Barry. If we start increasing the size of the storm to, say $2 billion (over cash on hand) hypothetical Hurricane Tanya (or we add another smaller hurricane in the year such as $1 billion hypothetical Chantal), the shrinkage problem starts to grow. Indeed, with a more serious storm such as Tanya (or Barry + Chantal), there is a significant risk of a death spiral in which there are not enough TWIA policyholders left to repay the assessment.
What this means is that a sensible insurance scheme has to balance at least two factors: you don’t want insureds paying too much because — well — that’s money out of the hands of insureds and doing so could needlessly depress the coastal economy. On the other hand, you don’t want insureds paying too little because that creates the risk of a storm large enough to plunge any assessment scheme into a death spiral.
So, what an intelligent discussion of this issue needs to do is to discuss a balance of these two problems with actuarial and climate science in mind. By focusing only on the over-collection issue, certain coastal legislators are sadly distorting and deterring an intelligent political debate. The consequence of this distortion could be quite serious for coastal residents. They may find that the assessment scheme they counted on to help TWIA pay for losses just does not work. They will then need to come begging to the state or Uncle Sam for a bailout. Indeed, that is precisely what the latest actuarial study engaged by TWIA/TDI predicts will occur a substantial percentage of the time if coastal premiums are not raised substantially.
Now, what I could imagine a coastal politician saying in response to this blog (I do hope they read it — and read it all the way to the exciting end), is that the post-assessment problem is overstated because the Texas statute limits the size of assessments that can be made. Moreover, they might say, TWIA does not depend exclusively on policyholder assessments to pay for tropical cyclones that cause damages in excess of cash on hand. And, to some extent, this rebuttal is true. But, if we break it down, what the politicians are really saying then is that small premiums aren’t a problem because they expect someone else to pick up the pieces if a big storm hits. That is, TWIA policyholders aren’t to be treated like ordinary policyholders who are in serious difficulty if their insurer (predictably) goes insolvent after charging too low premiums; TWIA policyholders stand in a privileged position relative to other Texans.
I have a lot of problems with this rebuttal. First, there is still a problem because, as I have pointed out at some length, the other sources (limited assessments on non-TWIA coastal policies; limited assessment on Texas insurers) will not yield enough cash to pay fully for losses incurred after serious storms in the densely populated parts of the Texas coast. Even the coastal politicians now appear to recognize this fact but poo-poo what they believe is a 1.5% annual risk of this occurring. But even if there were zero risk of a TWIA default, the rebuttal rests on the dubious premise that coastal residents are entitled to some special privilege that the rest of Texas does not enjoy.
And, here, we get to the core of the issue. After you put all the smokescreen rhetoric aside, what certain coastal politicians are really saying is that the coast is indeed special. But why? Yes, many find the coast lovely, but so too do many find the Texas Hill Country or even the stark beauty of the northern panhandle with its silos, vistas and hints of ancient vulcanism. Yet neither Kerrville nor Dumas insureds have the benefit of getting full insurance on their $200,000 property while paying the premiums sufficient to cover only a policy with a $50,000 limit. If the preference for the coast is based on aesthetics, then let us be explicit and subsidize coastal beauty. Yes, the coast contributes to the Texas economy. But so does Houston, so does Dallas, so does El Paso, so does Stephenville and so does Dalhart and Childress and Texarkana. Dollars are dollars. Should Galveston residents assume the risks posed by locating property in Dallas because Dallas contributes greatly to all of us here in Texas?
So, in my opinion, the “coast contributes” justification is completely bogus.
Are coastal residents disproportionately afflicted by some imperfection in the insurance market from which the rest of Texas is exempt? Might that justify the privilege? I see no evidence of any special, irrational animus towards the Texas coast on the part of the insurance industry. Greedy insurers would as much enjoy profiting from coastal consumers as they do profiting from those in Stephenville. Why, if they could do so at a rate they thought profitable, would insurers collectively sit back and let a government entity take over the coastal insurance business? The fact that many (and not all) companies won’t write on the coast is more easily explained by the low prices of its publicly subsidized competitor — TWIA — or by feared regulatory price constraints than by some vast conspiracy. The most likely reason prices are high on the coast is because risks there are particularly uncertain and particularly correlated and because stockpiling the requisite cash to pay for large claims is expensive.
Now, is it the “fault” of a Galveston policyholder that they live in a place subject to uncertain, correlated risk? No. There is no moral failing. Is it a “choice?” That depends. For some, such as the owner of a second home, quite probably. Moreover, even for a first home, people choose to have a million dollar home rather than a $500,000 home. For others, however, less so. The current scheme in Texas, however, treats TWIA insureds as special regardless of the degree to which their ownership of property is a choice. In general, it’s a good idea for insurance prices to reflect risk accurately because that, in the long run, encourages intelligent investment decisions. Societies that subsidize risky activities for long periods of time end up with more risky activities and become, as the pandering of some coastal politician suggests, addicted to them at great expense to the rest of society.
The better rebuttal — and the one I actually hope starts to gain prominence — is that giving TWIA lots of money has enabled it to waste money purchasing incredibly expensive reinsurance in a market whose competitiveness is open to question. Because reinsurance has cost TWIA something like six times the actuarial value of the risk it assumes, policyholders would do better, this argument proceeds, for TWIA to charge less in premiums or to charge the same in premiums, pray for a few good years, and to stockpile the extra funds. This argument, by the way, has some support in the recent report of TWIA’s actuarial consultants. But, if wasting money on reinsurance is the issue, the better solution is to restrict TWIA’s discretion in using its funds for this purpose. TWIA could, after all, buy reinsurance even if premiums were reduced. Now, how exactly to intelligently restrict TWIA’s discretion in reinsurance purchasing is the subject for another day. In the mean time, however, the purported advocates of coastal policyholders had best be very careful lest their concerted advocacy in support of what they wish for, lower premiums, succeeds and a large insurer has insufficient resources to pay claim as a result. Pity the coastal politician seeking another term after they deprived the insurer of residents in their districts the resources that insurer needed to pay claims fully.