How will today’s terrible earthquake and tsunami affect Japan’s economy, not to mention those of other global economies? After spending all day reviewing a bunch of academic research, the verdict remains unclear. For instance,  simulations run by the Inter-American Development Bank find only the biggest disasters have an  impact. Take disasters in the 99th percentile:

The 99th percentile cutoff is equivalent to a natural disaster that kills more than 233 people per million inhabitants. Although the number is large, many recent large events have exceeded this rate. For example, the 2004 Indian Ocean Tsunami killed 772 people per million inhabitants in Indonesia, and almost 2,000 per million inhabitants in Sri Lanka. Moreover, by the latest accounts, the 2010 earthquake in Haiti killed over 20,000 people per million inhabitants (see Cavallo, Powell and Becerra (2010))

For these types of disaster, the economic drag is persistent:

quake

But smaller, though still devastating disaster show little economic impact:

quake2

The study’s conclusion:

Contrary to previous work, we find that natural disasters, even when we
focus only on the effects of the largest events, do not have any significant effect on subsequent
economic growth. Indeed, the only two cases where we found that truly large natural disasters
were followed by an important decline in GDP per capita were cases where the natural disaster
was followed, though in one case not immediately, by radical political revolution, which severely
affected the institutional organization of society.

Contrary to previous work, we find that natural disasters, even when we focus only on the effects of the largest events, do not have any significant effect on subsequent economic growth. Indeed, the only two cases where we found that truly large natural disasters were followed by an important decline in GDP per capita were cases where the natural disaster was followed, though in one case not immediately, by radical political revolution, which severely affected the institutional organization of society.

But a highly regarded 2002 study that tracked disasters in 89 countries for 30 years came to a different conclusion, as summarized by the IADB:

Skidmore and Toya (2002) explain their somewhat counterintuitive finding by suggesting that disasters may be speeding up the Schumpeterian “creative destruction” process that is at the heart of the development of market economies. Cuaresma et al. (2008) attempt to investigate this creative destruction hypothesis empirically by closely examining the evolution of R&D from foreign origin and how it is affected by catastrophic risk. They conclude that the creative destruction dynamic most likely only occurs in countries with high per capita income.

This make some sense to me.  This is not the Broken Windows Fallacy where the repair of damage is thought to boost growth. That sort of thing is just redistributing economic activity and is at best is a wash. But if a disaster is a catalyst for better infrastructure or less regulation to promote rebuilding, S&T could be correct.  (Of course, it is worth noting that Japan already has great infrastructure after spending like crazy to boost economic output the past two decades. Plus there is the issue of taking on more debt by a highly indebted country.)