TCH: Insurance and the Great Kanto Earthquake

??????????? In the years before the Great Kanto Earthquake of 1923, fire insurance was gaining more widespread adoption in Japan. For many insurance companies, it was becoming their largest line of business. Unfortunately, the earthquake and the fires that followed challenged this growth. Many looked to the insurance companies to pay claims after the disaster but contract terms meant that they had no obligation to pay after an earthquake. In any case, they could not possibly afford to pay out all, or even most, of the potential claims from a calamity so large anyway. The result was a coordinated and state-supported settlement.

Insurers

??????????? Fire insurance in Japan began to develop in the late 19th century and, as in other countries, the insurance industry more generally grew rapidly after the turn of the century. Tokyo Fire Insurance, established in 1888, was the first dedicated fire insurance company in Japan. Already by the end of 1893, several other fire insurers were founded and some existing insurance companies branched into this line.

??????????? ?So much competition early in the history of the fire insurance business could have spelled trouble for the companies. As a result, the early history of fire insurance in Japan revolved around efforts to collude on pricing. The companies formed industry associations in order to limit competition. After some false starts, the ultimate end was the creation of a cartel, the Great Japan Alliance Fire Insurance Association, which brought together eighteen Japanese insurers and twenty-four foreign ones. A common agreement on prices was reached in 1917.

??????????? ?As the price collusion may imply, Japan’s government generally intervened little in business, though it eventually brought some light regulation to the insurance industry. The Insurance Business Act of 1900 strengthened reserve provisioning and disclosure requirements. This limited intervention was enough to prompt some speculative insurers to leave the market; so too did some foreign insurers who preferred to keep their reserves invested in their home market rather than set aside in Japan. Still, the industry grew. As in other countries around this time, new products were introduced, including fidelity bonds and insurance for boilers and engines, personal accidents, burglary and motor vehicles; all of these launched between 1904 and 1916.

??????????? In the 1910s and early 1920s, fire insurance became a large business line. This was true even for Tokio Marine which was already a very large marine insurance company. It branched out into fire insurance in 1914 and grew this business quickly starting from 1917; Tokio Marine came to have significant market share as it was by far the largest non-life insurance company in the country. For the industry as a whole, while marine insurance was still a bigger business than fire insurance in 1920, by 1923, the latter was nearly twice as large as the former.

Great Kanto Earthquake

??????????? The 7.9 magnitude Great Kanto Earthquake struck two minutes before noon on Saturday, September 1, 1923. It would be followed by nine hundred aftershocks over the next few days. It was most damaging to Tokyo, Yokohama, and the surrounding areas, the largest metropolitan area in the country. Thus, the earthquake struck the center of Japanese commercial, industrial, financial, and political life. The port at Yokohama, for example, was responsible for over 40% of the country’s trade in commodities. The epicenter was just south of the city. The Great Kanto Earthquake caused 105,000 deaths. Counting those who fled the Tokyo region, the metropolitan area’s population fell by 750,000.

??????????? Much of this damage was the result of more than one hundred and thirty fires that started within minutes of the earthquake; the disaster struck as lunch was being prepared by many households with small kitchen ranges. Making matters worse, a typhoon in Western Japan caused gale-force winds to spread the fires. In all, some 212,000 houses were destroyed by fire, more than the number destroyed by the earthquake directly. This is a fairly common outcome of earthquakes, the same being observed in San Francisco in 1906. In any case, counting heavily damaged buildings, the losses to property were immense; the Great Kanto Earthquake left large portions of the cities of Tokyo and Yokohama without recognizable buildings remaining.

Covering Damages

??????????? The city government estimated the losses at ￥5.5 billion, including ￥2.7 billion yen in damages to buildings and furniture and ￥2.4 billion in damages to factories and inventories. The latter destruction would have shut down large swaths of the area’s industrial capacity. Compared to these numbers, fire insurance coverage stood at ￥3.26 billion across nearly 660,000 policies in force. Of these, 291,157 policyholders were victims of the disaster.

??????????? However, insurance would not be a sizable source of funds for rebuilding. Insurance contracts excluded coverage for fires caused by earthquakes. The experience of insurers in the aftermath of the San Francisco Earthquake of 1906 led to new policy wording, adopted globally, that explicitly excluded such coverage; earlier wording varied and was often unclear. Nonetheless, insurance companies were under pressure by the government to provide some relief. Soon after the earthquake, Prime Minster Gonbei Yamamoto was making hopeful statements about the insurance companies paying out claims. His cabinet was also ignorant of the ability of insurers to pay. This disaster was simply far more catastrophic and the damages far more costly than anything the insurance companies could bear.

??????????? ?Unsurprisingly, the government was itself pressured by the victims of the earthquake; these included powerful people like industrialists who had lost their factories and bankers who had made loans secured by now ruined buildings. There were also demonstrations by masses of claimants. They all wanted insurers to make payment.

??????????? ?However, there was hardly anything the government could do. Some politicians called for extralegal actions against the insurers, a hopeless maneuver. Even if 100% of the reserves of all the insurance companies active in the region were confiscated for distribution to the earthquake victims, the amount would have remained woefully inadequate, a drop of the ocean compared to the losses. Needless to say, the loss of 100% of insurance reserves would have also prevented the writing of new policies and, at least for a fairly long period of time, the suspension of insurance underwriting in Japan.

??????????? Still, insurers were not all hostile to any sort of contribution to the victims. Many supported a settlement that would still see the draining of most of their reserves. In light of the situation, such a compromise seemed generous. However, doing nothing would have damaged the industry’s credibility. A holdout insurer could be perceived as financially weak or stingy and either is a threat to future success in the market. Stronger insurers also wanted to move on from the calamity by making payment. If no assistance was provided at all, then people and companies might stop buying fire insurance and the market would disappear. Already, prospective customers were holding back on buying insurance in the period of uncertainty.

??????????? After months, the insurers agreed to provide some relief. The government wanted insurers to make payments at 10% of the value of the policy. Some insurers essentially refunded one year’s premium instead. Even a payment of 10% of the policy limit towards relief for the victims, in the context of a disaster as large as this earthquake, was beyond the ability of many companies to pay. For example, Yokohama Fire, a relatively large firm, had a reserve fund of approximately ￥7.9 million but even the payment of one-tenth on claims would have been ￥12.5 million for this insurer.

??????????? In all, insurers in Kanto and Kansai paid out ￥123 million. Even this ‘smaller’ sum was not small and paying it strained the industry. To allow such a large portion of insurer’s capital to be paid out, the government provided a 4% interest loan that allowed companies to pay. This was crucial to the effort. Only few insurance firms were able to make the payments without any government support.

Other Financial Effects

??????????? Ikeuchi Yukichika, in his book Shinsai Keizai Shigan (Personal View of the Earthquake Disaster Economy), published just three months after the disaster, argued the earthquake set Japan’s economy back years. He was an economic advisor to several firms and had a long academic training; his opinion certainly carried weight. Certainly, whatever the long run impact, the disruption in the short run is clearly difficult to overstate. Some goods were in short supply, like construction goods for rebuilding, but others were accumulating in huge surpluses, like coal for which demand by factories collapsed. The economy was in a very disorderly state.

??????????? In light of the scale of the destruction, it is remarkable that, Ikeuchi Yukichika’s prediction notwithstanding, the economy generally recovered quickly. That said, financial problems persisted and unlike the loss of buildings to fire, these were not localized. Japan was wholly dependent on Tokyo-based financial institutions. Then as now, financial and commercial activity in Japan was relatively centralized. As a result of the damage sustained in Tokyo, many bank records were lost. Further, many businesses ceased to exist and sizable investments in property or other assets in the capital region were lost. This was very problematic for banks. The result of all these problems was that credit was withdrawn and cash transactions were replacing trade credit, even in other parts of Japan far from the epicenter of the earthquake; local chambers of commerce and Ikeuchi Yukichika reported this concerning development.

??????????? The government took large actions to mitigate the financial consequences of the earthquake. It initially implemented a one-month moratorium on the settlement of commercial bills and debts. The Bank of Japan was also required to advance money against bills presented by banks, providing liquidity. As is often a hazard in such endeavors, the assistance provided by the government and the Bank of Japan propped up weak banks just as it had good ones. This helped cause a later banking crisis in 1927. In the end though, compared to the Great Depression of the early 1930s, the lasting economic impact of the earthquake was comparatively mild. However sharply reduced compared to the hopes of many, the payouts of insurance companies must have counted for something in making this so.

Lesson

??????????? ?Decisions on individual insurance claims are often straightforward at least as far as interests are concerned. An insurer has an incentive to scrutinize a claim in light of the insurance contract, its inclusions and exclusions. However, in the context of a high-profile claim or multiple such claims, particularly where there are large public interests at stake, the calculus is more complicated. The question also goes beyond a single insurer. These sorts of situations are collective action problems. Stingy insurance companies can ruin the public’s impression of an entire industry and the contract terms lose significance. The result often is a collective solution not necessarily set out in an insurance contract and often requiring some external support and pressure alike, this time coming from the government.

More from the Tontine Coffee-House

???????????Read?about how the 1906 San Francisco Earthquake changed the nature of insurance against earthquakes. Also, learn how the?Great Fire of Hamburg?and the Glarus Fire of 1861 were formative events in the reinsurance industry. Consider subscribing to this blog’s?newsletter?or checking out?book recommendations, which include many of the sources often referenced in my posts.?

Further Reading

1.????? A History of Insurance in Japan, Swiss Reinsurance Company Ltd, 2017.

2.????? Hunter, Janet. “Thinking about the Economic Consequences of the Great Kantō Earthquake.” The Asia-Pacific Journal: Japan Focus, 1 Sept. 2023.

3.????? Hunter, Janet. “‘Extreme confusion and disorder’? the Japanese economy in the Great Kantō Earthquake of 1923.” The Journal of Asian Studies, vol. 73, no. 3, Aug. 2014, pp. 753–773.

4.????? Takau Yoneyama. “The Great Kanto Earthquake and the response of insurance companies: a historical lesson on the impact of a major disaster.” Hitotsubashi Journal of Commerce and Management, vol. 43, no. 1, Oct. 2009, pp. 11–26.