Friday 8 April 2011

Insurance bailouts

The New Zealand Government has provided a backstop guarantee to Christchurch-based insurer AMI. Nolan at TVHE hits the main issues. Too much of AMI's book was Christchurch-based for the amount of re-insurance cover it had for a sequence of major Christchurch events. Yes, two huge earthquakes in Christchurch was a low probability event. But it's relatively cheap to insure against low probability events.

Nolan rightly points out:Here's what I told the National Business Review yesterday, now in today's print edition:
University of Canterbury economist Eric Crampton said his biggest worry about the AMI bailout “is the signal it sends to New Zealand insurers. Why bother carrying reinsurance over and above the statutory minimum and why bother ensuring you have a balanced client base if the taxpayer will bail you out if anything really bad happens?”

He said the government was right to worry about systemic risks to reconstruction posed by an AMI collapse. “But I’m not convinced a bailout was the best option. AMI policyholders, myself included, enjoyed lower rates because AMI had underinsured against the risk of local disasters. And, as it was a mutual, we don’t really have anybody else to blame.

“News reporting suggests AMI had sufficient capital and reinsurance to cover the costs of the Christchurch disaster but then would have had nothing left against other adverse events.

“Under those circumstances, I could have expected a haircut to Christchurch claimants or a surcharge on AMI members to help in recapitalisation and to fund the purchase of additional reinsurance against other large events.

“It isn’t as though there would be no payouts to Christchurch claimants absent a bailout; rather, the company would likely have been wound down subsequent to paying out on claims.”
NBR left out my last line, which I think is an important one (culled due to space constraints I'd expect):
Transitional assistance during that wind down could have been enough to avoid systemic problems while keeping sharp incentives for other insurers, and their clients, going forward.
The NBR piece continues:
University of Auckland Business School BNZ chairman of finance David Mayes was skeptical about possible “moral hazard” from the bailout. “I’m very cagey on moral hazard – very few companies are running on the hope they will be bailed out.”

He said AMI’s mutual structure presented problems both in terms of raising capital and selling the company. Getting approval from policyholders was “more difficult than doing it through a shareholders’ meeting,” he said.

Reserve Bank spokesman Mike Hannah said the government bailout of AMI shouldn’t be seen as setting a precedent due to the unique situation. “This is not how you would deal with a distressed insurance company in normal circumstances. However, this is a very unusual set of circumstances.”
David is certainly right that few companies run on the hope of being bailed out. But they may underinsure in reinsurance markets against low probability events that would draw political sympathy. Here's the Sunday Star Times, HT Hickey:
Sunday Star-Times calculations show AMI invested 3.7% of premium income in reinsurance. Other insurers exceeded 10%, with Lumley General reporting 16.3%.
And the NBR reports elsewhere (subscription):
Rival insurers are privately fuming at the government’s decision to provide a support package to AMI policyholders in case the insurer’s reserves run out.

An insurance industry source told the NBR this morning that, when discussing with colleagues the possibility of an AMI bailout, there were two main responses.

“The first one was, the government is going to get into even more debt, and the second one was, what about moral hazard?”

However, the source said AMI’s rivals would be unlikely to speak out against the bailout because “they don’t want to annoy the government in case they need a bailout themselves one day.”

The source said AMI was understood to have had a lower level of reinsurance than other comparable insurers, meaning it was able to charge lower premiums than it would otherwise due to the reduced reinsurance cost.

AMI customers benefited from those lower premiums and, now that they were being backstopped by the government, other insurers with more prudent levels of reinsurance were being "punished."
What I most want to know is why AM Best gave AMI an A+ rating, about the same as the other folks insuring NZ property, when AMI had this kind of risk profile. I'm insured with AMI and have a claim sitting with them for damage to our swimming pool and sidewalks consequent to the February quake; EQC is likely covering the rest.

Here's what went through my head when we bought insurance after moving to New Zealand:
AMI has better rates than the other folks and the same credit rating. They seem to be mainly locally based, so they might be subject to problems if there's a big Christchurch event. But that kind of risk has to be cheap to lay off through reinsurance and, besides, if they haven't, there's no way the government would fail to bail out in the kind of local event that would take out the insurer. I'll go with the low rates.
I don't always like to have my expectations fulfilled.

I don't think that moral hazard problems generated by this would really hit the big insurers: anybody internationally based would have international resources to draw down in the event of a disaster and so wouldn't be bailed out. Rather, the industry as a whole winds up with too many locally-based mutuals that can lay off risk on the government.


  1. Like yourself I am an AMI customer. However I'm not really anticipating a payout. We lodged a claim for a crack in our concrete path after the September quake, but have yet to see an assessor, and I now have no expectation that we ever will. If we do, all good, but if not, no worries. It is a minor crack, and there are plenty of folk in Canterbury with serious concerns; I'd rather see their claims dealt with.

    As for the bail-out, I wonder if a better option might not have been for govt to simply offer a guarantee that they would pick up any excess claims once AMI resources had been depleted. That way the folk who bought insurance in good faith are covered, but AMI's slightly reckless management strategy isn't rewarded by a full bail-out. Thoughts?

  2. "That way the folk who bought insurance in good faith are covered, but AMI's slightly reckless management strategy isn't rewarded by a full bail-out. Thoughts? "

    Although that also signals to people that they should just go for the cheapest insurer, rather than thinking if the insurer is actually operating properly. This in itself leads to a moral hazard issue.

    If society wants to implicitly insure people in the case of low probability high cost events then so be it - but do it in a way that is transparent. If government wants to do it - it needs to make sure that, ex-ante, that is what society desires.

  3. @Lats: The folks who bought insurance in good faith are also the folks who didn't exercise due diligence in their roles as mutual partners (me included). But you're right that this does make the difference between bailing out AMI and bailing out non-insured: at least most of the former were trying to buy insurance. And what Matt said.

  4. @Matt & Eric Good points. I can see the similarity between investors in failed finance companies and folk who bought insurance from AMI, although I don't think there are too many people who would have predicted AMI faltering. Their premiums are still a bit cheaper than any other big insurer though, so I guess had I been more on to it that may have been an indication that the risk was somewhat higher.

  5. “Reserve Bank spokesman Mike Hannah said the government bailout of AMI shouldn’t be seen as setting a precedent due to the unique situation. “This is not how you would deal with a distressed insurance company in normal circumstances. However, this is a very unusual set of circumstances.””

    What is this man talking about? Surely very unusual sets of circumstances are, in the world of insurance, exactly the type of circumstances that normally lead to an underinsured insurance company becoming distressed.

    I see you have weeded out Peter Quixote’s enigmatic and largely incomprehensible comments – I quite enjoyed them though – they were like bright daisies set against the asphalt grey of economics.

  6. Personally I hope AMI front foots the issue by releasing publicly how it derived the levels of reinsurance coverage it found itself with.

    I'm sceptical of Vero's comments in the media, particularly given it is owned by SunCorp, which itself had to tap international markets with the govt. guarantee and has had to seek additional capital itself.

    As far as Suncorp goes, the various acquisitions to me look suspiciously like trying to become TBTF, given it sits below the size of the Australian big four which does anyone really doubt are TBTF?

    As far as the insurance industry goes the opacity of the industry makes it very difficult for the general consumer to make a truly informed decision. Once again private rating agencies show they are boxtickers, as does auditor EY, the same auditor that ticked off Lehman Brothers before it woke up one Monday morning in a $100 billion hole.

    I guess an analogy would be if you are buying a TV, you shouldn't expect the purchaser to need an electrical engineering degree.
    That is why we have laws/regulation to set the market boundaries, otherwise the capitalist society would breakdown very quickly.

  7. Also remember AMI is a mutual, it can not go to its shareholders to raise more capital - it doesn't have any. So where does it get more capital if needed? Unfortunately we now know the answer.

  8. Gentlemen, what is your opinion on the possible hypothesis that the government assurance is more to reduce the likelihood of panic among insured and insurers? (as opposed to actual default)

  9. Eric,

    What are the incentives that would be changed if the Government were to refuse a bail out here?

    (1) Is it that those running companies like AMI would be forced to ensure there is adequate reinsurance cover/properly spread risk, which would eliminate a business model of offering cheap because inadequately covered policies? If they don't do so, then they risk losing the company when catastrophe strikes. But, why isn't nationalisation the same as receivership from the perspective of those making decisions for the company - in either case, they "lose" it? And in the case of a mutual like AMI, isn't it likely that the incentives for managers, etc are geared towards increasing market share/number of customers - the bigger the company gets, the better the managers do (in the absence of profits to be distributed to shareholders). Plus you've got a very broadly spread set of "owners", with minimal oversight of the management - there's no-one with a big enough ownership stake to closely monitor the managers preferred strategies. So, all-in-all, how would changing from a "bailout likely" to a "bailout unlikely" model amend how management would choose to act?

    (2) Is it that those purchasing insurance would be forced to properly evaluate the ability of the provider to cover risk, shying away from companies like AMI on the basis that they are less likely to meet their obligations in the event of catastrophe? Of course, this relies upon enough individual consumers not only assessing the risks they need to insure against, but assessing the risks involved in their choice of insurance of risk versus the price for that insurance ... and we're not all so economically rational! But even so, as you point out, the ratings agencies all thought AMI was peachy - A.M. Best gave it an A+ "superior" rating for financial strength. So why would a rational consumer think they can do a better analysis job than the rating agencies and choose to pay more for insurance from another company with the same rating (i.e. Tower or State, etc).

    Point being, I don't think the "moral hazard" argument is as strong here as it was for (say) South Canterbury Finance ... where you could invest for 9.5% with a Government guarantee of repayment! However, there probably is an argument for tighter regulation about the amount of reinsurance cover companies must take out for such "unexpected" events.

  10. Where bailout is unlikely, folks are less likely to go with small mutuals that have no other source of capital on which to draw. Instead, they'd look to larger international brands. Which of course aren't failsafe. But are less likely to collapse as result of anything here - if your insurance company goes belly up when nothing in particular is going on here, you suck up the loss of premium and buy a new policy.

    The moral hazard problem here is smaller, and lots of AMI customers did do their due diligence by checking A.M. Best's rating. I won't trust A.M. Best's rating on anything now.

    Agreed that tighter regulation on mutuals is a likely outcome. Possibly even desirable.

  11. Bailouts plus no additional regs on mutuals probably yields an industry with too many small mutuals.