Causal inference of the nuke rumor effect on the tourism profit in the Fukushima area

Sorry for cross-posting.
Tokyo Electric Power Company announced that it would discount its indemnification for tourism in Fukushima and other three neighbor prefectures by 20% because that part was not attributed to its nuclear power station accident. Their evidence is that, in the earthquake in 1995, tourism in Akashi and Himeji cities lost their benefit by 20%.
http://www.tepco.co.jp/en/press/corp-com/release/110921
09-e.html

I don't think that Akashi and Himeji cities in 1995 is counterfactual of Fukushima and other three prefectures in 2011. They are different. In the first place, Akashi and Himeji "cities" were the most damaged areas in 1995, while some areas in Fukushima and other three "prefectures"
suffer from neither earthquake nor radiation but "rumor."

I have an idea of natural experiment. For example, Towada Lake, a tourist attraction, strides Iwate, Aomori, and Akita prefectures, among which only Iwate suffers from the rumor. Thus, the lakeside hotels in Aomori and Akita are counterfactual of those in Iwate; they share the same tourist attraction and are affected by earthquake and radiation, while only Iwate hotels are damaged by their prefecture name "brand."

If you have a good idea about how to estimate the nuke rumor effect on the tourism profit in the Fukushima area, please tell me. This might become professional contribution to the victims.

Thanks,

Kentaro

*******************************************************
*********
Kentaro FUKUMOTO, Ph. D.
Professor, Department of Political Science, Gakushuin University http://www-cc.gakushuin.ac.jp/~e982440/index_e.htm
*******************************************************
*********