Kissing Assets Goodbye

No, thanks for asking but the power is not back on, and not likely to be coming back on for some time. But one finds a way – after all, this is the whole point of a “disaster recovery” plan. I won’t be penning many articles in the next few days, but given the circumstances I thought it relevant to comment on disasters and economic growth.

After a hurricane or other natural disaster, there is always a significant confusion among economists about whether the disaster will hurt U.S. GDP, because many consumers and businesses are unable to consume for a period, or help GDP because of reconstruction expenditures.

This is a crazy debate, and it underscores a key shortcoming of economic statistics. The usual economic result is a short-term (a couple of weeks, perhaps, or a month) of softness in private expenditures, followed by an increase in GDP because of rebuilding. Disasters, measured by GDP, are usually additive: that is, growth in the quarter immediately after the event is higher than it would otherwise be, because more money is spent from savings and government expenditures rise (because of explicit relief payments but also because of increases in automatic expenditures such as unemployment claims and other such things).

But that’s clearly nonsense, to the dispassionate observer. The citizen’s welfare, his standard of living, is clearly lower than it was before the disaster; if it was not, we could regularly ignite the economy by destroying buildings and homes and rebuilding them.

Yet, the numbers are not wrong, per se. By definition, GDP=C + I + G + (X-M), and total expenditures clearly rise as savings (public and private) decline in the aftermath of a disaster. The problem isn’t that the numbers, such as they are, are wrong but that there is a deeper philosophical problem. Most economic data measures flows, not levels (the proper term is “stocks,” but I didn’t want to confuse readers by sounding like I was talking about equities – I am not). There is no economic “asset” and “liability” account for the nation. If the disaster occurred to a company instead of to a nation, the company would record an expense for the impairment of an asset (a destroyed building, equipment, etc) as one transaction and then separately record the purchase of a replacement asset (if it was a durable asset, this second transaction would merely exchange one asset, cash, for a durable asset). The net result would clearly be a decline in the net value of the company. But there is no national asset or liability accounts to credit for the destruction of national assets.

And there should be. If policymakers had to focus not on increasing the expenditures of the nation, but on building its “net worth,” I suspect we would see more sensible national policies.

However, that’s not the way it works – but we all know that disasters hurt our economy, whatever effect they have on GDP.