Posted by: Nicholas Davis | March 17, 2011

On disaster economics

There has been a flurry of articles discussing the economic implications of the Japan earthquake and tsunami. While many worry about the direct impact on output and wealth, and some on the implications for global bond markets, a few have stated that the disaster could result in a net benefit to the Japanese economy, or at least some form of “silver lining”. See for instance here at the Independent, and here at the Huffington Post.

As I wrote in the comments to the Independent article, I appreciate that reconstruction could act as a short-term stimulus to economic activity and mean that people start spending a portion of their (relatively high levels of) savings. However, forcing people to spend their savings is not a measure of success on its own. Putting aside Sean O’Grady’s claim in the Independent that such savings represent “a classic case of deflationary psychology” (perhaps, under the circumstances, excess savings could now be seen as incredibly rational??), it matters how they spend that money. Struggling to replace basic amenities and productive assets destroyed in a disaster would only increase future growth if the resulting assets were more productive than previous versions, and meanwhile the value of the stored wealth is lost to replacement costs rather than being available for investment via the banking system or spending on other goods.

Perhaps part of the confusion is that people are not distinguishing between stocks and flows – the flow of output will rise, yes, but only because the stock of wealth has been significantly reduced. People who see disasters as producing some net benefit must either a) be ignoring the stocks altogether and only focusing on GDP flow as a measure of wealth, therefore being blind to the value of the installed capital that was destroyed (and argue that although GDP growth might take a hit from productive capacity taken off-line, the reconstructive efforts will more than compensate for this), or b) claim that the additional flow of economic activity spurred by the crisis will, in time, result in economic gains that will exceed the lost stock of value destroyed and take the economy to a higher level than was evident pre-crisis.

The first approach seems very short-sighted, while for the second to be true you need two things:

First, to assume a very heartless position in defining value: the GDP value of new flows of reconstructive activity will be qualitatively very different from the human, social and cultural stock of value lost. In accounting terms you may see a “gruesome calculus” that results in a net benefit – but I doubt that that represents any true improvement in the quality of life of the Japanese people.

Second, you need to assume a counterfactual where growth without the disaster is significantly less than growth with the disaster for a meaningful period of time. While Japanese growth has been trending down, continued flat or negative growth was not the only “non-disaster” scenario.

I like Jarrett Skorup’s analysis of this at the Mackinac Center:

The answer is found from French economist Frederic Bastiat in his 1850 essay, “What Is Seen and What Is Not Seen.” “There is only one difference between a bad economist and a good one: the bad economist confines himself to the visible effect; the good economist takes into account both the effect that can be seen and those effects that must be foreseen.

In other words, what one sees are the workers paid to rebuild these destroyed cities; but what is not seen is where that money would have been spent if not for the disaster. Instead of being used to construct buildings, it would buy food, shelter, clothing and other goods. With the disaster, society has a new building, but without it they have a building and food, shelter, clothing or whatever else this money could be spent on. A country is much richer without the disaster.

Such theory-wrangling aside, economists have actually looked at this problem empirically, as was reported in the Economist yesterday. So you need not get too involved in the “scarcity v abundance” debate as described at Cafe Hayek , and instead can simply quote from the 2010 paper  Catastrophic Natural Disasters and Economic Growth” by Eduardo Cavallo, Ilan Noy, Sebastian Galiani, and Juan Pantano:

“Even very large natural disasters, when not followed by disruptive political reforms that alter the economic system, including the system or property rights, do not display significant effects on economic growth. …

Thus, we conclude that unless a natural disaster triggers a radical political revolution; it is unlikely to affect economic growth. Of course, this conclusion does not neglect the direct cost of natural disasters such as the lives lost and the costs of reconstruction that often are quite large. (Emphasis added by W.W. at the Economist, original page numbers for the quotes are p27 and p28)

So, in my mind at least, to suggest that temporary increases in GDP flows or the ability to manage better the public budget compensates in some way for the costs of the disaster is both a failure of imagination in terms of the alternative scenarios for non-disaster economic growth, and worse than gruesome when you consider the losses that can’t be measured in accounting terms. Particularly when out best current evidence fails to show any long-lasting economic impact either way.



  1. It was widely thought that WWII ‘strategic bombing’ could be counter-productive, in that the citizens would go on a serious ‘war footing’, in addition to any action the authorities might take.

    There are some reports that suggest that the Japanese have been shaken, so they might conceivably start doing things differently. For example, they might take the green agenda to heart, in which case a few breakthroughs might well lead to improvements in the long-term, depending on how you value climate-change and risk. But I agree that optimistic message that we have been getting seems a tad simplistic.

    • Thanks Dave. I guess my concern is whether such a shake-up results in a permanent or only a temporary shift (both in terms of the stickiness of the shift’s structure and its impact on productivity), and whether the form and timing of the shift is significantly earlier or better than might be expected without the disaster.

      It seems from history (at least in terms of the stultification of political structures) that we’d expect a brief period of reform but then a tailing off, in which case the issue is how long the higher level of productivity persists.

      I do think that some good things do emerge from such terrible situations – a greater sense of community, for example. But I just don’t think that the positive elements in any sent compensate for the negative, which is the sense I’m getting from some of the economic analyses that suggest that the disaster will boost Japan’s economy in the long-run.

  2. Nicholas,

    Selecting politicians every few years disadvantages those who take more of a long-term view than most, e.g. in avoiding or coping with disasters. Firms are also incentivised on the short term (not just banks). From this perspective maybe the best we can do, pragmatically, is to grow as fast as we can in the short term and cope with set-backs as they arise. The Japanese were not growing fast, perhaps due to cultural reasons (not appreciating the need for growth) but have still had a set-back. The shock may change their attitudes and hence growth, even within a fixed political system. You may be right, in that after a while they will return to their own ways.

    I see the current systems as encouraging short-term thinking. I think that we have very many situations in which such short-term thinking is inadequate, but tend to over-look them, in the interests of pragmatism. I conjecture that if some disaster resulted in use abandoning our narrow pragmatism for broader thinking then we would all do better. The key would be to keep our reasoning skills exercised by applying them to lesser problems. We might then be able to retain them indefinitely. I am blogging on this at

    • Dave,

      I agree with you and appreciate the point re combining pragmatism with being open to triggers that take any opportunity to embed broader thinking. However I disagree with the suggestion that a pragmatic solution to the realities of short-term incentives is to focus on short-term growth (were you advocating that as a strategy for everyone to take in the current circumstances, or simply saying that that is a logical strategy only when trapped within a short-term (but flawed) paradigm?).

      My sense is that we and our leaders have made an error of attribution in linking happiness and well-being so closely to measures of economic output. Hence why I disagree with the proposal that a renewed interest in and stimulus of growth is a silver lining of the Japan disaster. Focusing on growth as a risk-mitigation strategy (generating resources to buffer a subsequent disasters) – I’m fairly sure you would agree – creates a focus on short-term efficiency and growth that actually exacerbates the broader risks.

      What I feel we need is pragmatism in the long-term, including a redefinition of efficiency that incorporates a wider range of scenarios and risks (and appreciation of uncertainty) over longer time horizons. Given the realities of political terms etc, I suggest that the risk community focus on the truly long-term risks facing us (i.e. generational health financing and climate issues) and try to develop innovative mechanisms to bind present decision-making to future outcomes as much as possible. If we can solve for some set of risks that everyone agrees are truly long-term, perhaps it will be easier to then transfer those kind of strategies over to the more hidden risks that require equally long-term commitments. Actually, I see the rise of sovereign wealth funds and the mainstreaming of strategies around long-term investing as being a fruitful source of new ideas and long-term thinking that can be built on in other areas of policymaking.

  3. Nicholas,

    Sorry. I should have emphasised that by ‘pragmatic’ I meant the following of a pragma that has been found to work in the short-run, e.g. is ‘evidence-based’. For example, in the short-run we may set up rules and targets, and then change them periodically. In sufficiently simple cases one can anticipate long-term problems and give their avoidance a short-term value. Thus one adjusts the established pragma to take account of the longer-term issues.

    This seems to be what you are suggesting, and would help in some circumstances, but maybe not all. I would like to see us adopting strategies that are less reliant on having a complete list of all the things that could go wrong. The two aspects are complementary. I do not see the pragmatic strategies as transferring, but if done in an appriate spirit it should be that many findings and insights can transfer, and that the two aspects are ‘synergistic’. This might be seen as a more reasonable form of pragmatism, that used to be commonplace, before we were all ‘improved’.

    • Dave, thanks for the clarification. I agree that the two aspects are needed and are complementary.

      I think of it as follows – on the one hand, wherever we can lengthen or intensify the “shadow of the future” we should. Most of these situations will be (relatively) simple cases which are nevertheless subject to bias or structures that create barriers to longer-term strategies. On the other hand, we also need to accept that in a complex system we cannot know anything like the full set of future states that are possible (and to attempt to do so would be dangerous), therefore we also adopt two other strategies: 1) flexibility embedded in activities and operations, so as to able to adjust tactics to a changing environment as required, and b) flexibility embedded in outlook, such that we are able to shift higher-level strategies and even entire perspectives (i.e employ broader thinking) so as to be able to enact 1) in as timely a manner as possible (perhaps, in some cases, slightly ahead of the herd).

  4. Agreed. Hopefully it wont take too many disasters before these things are valued.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s


%d bloggers like this: