This morning, I blogged at I.I.I. about how earthquake insurance works in Japan. I want to add a bit to that, mainly to say residential insurance claims won’t hit the private market too hard, if at all.
But if losses get big enough, things could get dicey for the quasi-governmental Japan Earthquake Reinsurance.
As I noted, Japan Earthquake Reinsurance holds the vast majority of Japan’s residential exposure. Much of that is retroceded through treaties placed April 1. Based on JER’s 2010 annual report (pdf), here is the 2009 structure, spread over two treaties:
- JER retained the first ¥115 billion of risk.
- JER ceded 100% of losses in the layer of ¥1,007.6 billion excess ¥115 billion. The government and the private market each took 50%.
- JER ceded 50% of the ¥802.4 billion excess ¥1,122.6 billion to the government and retained the rest.
- JER ceded 100% of the ¥1,787 billion excess ¥1,925 billion. The government took 95%; private reinsurers took 5%.
- JER ceded 95% of the ¥1,788 billion excess ¥3,712 billion.
Converted at 81 yen per dollar:
- JER retained the first $1.42 billion of risk.
- JER ceded 100% of losses in the layer of $12.44 billion excess $1.4 billion. The government and the private market each took 50%.
- JER ceded 50% of the $9.91 billion excess $13.86 billion to the government and retained the rest.
- JER ceded 100% of the $22.06 billion excess $23.77 billion. The government took 95%; private reinsurers took 5%.
- JER ceded 95% of the $22.07 billion excess $45.83 billion.
JER summarized in this chart:
Note this is the structure at 4/1/09. I don’t have evidence the structure has changed much since – it hadn’t going back the past several years. Most changes involved stretching the upper limit.
JER also summarizes each entity’s exposure with the following table. I’ve translated the yen exposure to U.S. dollars:
The bulk of the private market’s exposure is in the $12.44 billion XS $1.42 billion layer, where it is on the hook for $6.22 billion. But much of that is already booked as loss reserves. The JER annual report describes the situation:
JER and non-life insurance companies save the risk premium of insurance premiums paid by policyholders as earthquake insurance risk reserves for the possible payment of earthquake insurance claims while the government saves government reserves in the earthquake insurance special account under law. In the event that an earthquake occurs and causes losses or damages, each of JER, non-life insurance companies and the government pays an insurance claim according to each liability as stipulated in the reinsurance scheme by withdrawing from reserves.
Here were the reserves as of July 2010:
All the parties probably have a bit more than that on hand nine months later. Even so, a little math shows the maximum payout beyond the 2010 cat reserves would be $850 million – not chicken feed, but a lot easier to swallow than $6.47 billion.
My back-of-the-envelope indicates residential insured losses would have to exceed ¥2.335 trillion ($28.8 billion) before the private market would start realizing a loss, at least in a U.S. accounting sense.
It’s anyone’s guess what residential losses will hit, but I note that most have estimated only about 20% of residences are insured.
On the other hand, this article indicates “replacement value for insured properties within 3km of the coastline in the four Japanese prefectures worst affected by the tsunami was $24B.” (h/t Robert Bear’s blog) It’s not immediately clear whether this amount is the sum of limits for standard losses or earthquake losses. Quake limits are capped at half the limit for standard losses.
A big hit could be a significant problem for the JER, though. It has $6 billion in cat reserves but only $20 million shareholders’ equity (Annual report, p. 32). At July 2010, its solvency ratio – using a calculation similar to RBC in the U.S. – was 161.6% (p. 38).
The cat reserves are the main number buoying the solvency ratio, as much as they can.
A solvency ratio below 200% normally indicates a need for “prompt corrective actions,” according to Japan’s Financial Services Agency. But JER drily notes that it enjoys special government protection that:
stipulates . . . that the government takes responsibility for support and for lending funds for the payment of insurance money. Because this is a form of special business, JER’s solvency-margin ratio is not usable as a figure to enable the administrative authorities to trigger an order for improvement.
That makes sense on paper, I suppose, but given the circumstances, I’d think the last thing the Japanese government will need is to recapitalize an insurer.