Nov 04, 2012 at 1:12 am
The damage from Sandy, both to peoples’ lives and the economy, keeps mounting as each day reveals how long it’s taking to get back to normal. By definition, this increases the economic costs of the crisis, which can be thought of as having two basic parts: costs that are sunk forever and costs that will be made up.
More on that in a moment, but first, here’s a chart of hurricane costs as a share of GDP covering the past few decades. The preliminary estimate of Sandy, by GS Research, is $30 billion, or about 0.2% of GDP, placing it in the top five. I think that’s low and so I added a bar of $50 billion, or 0.3% of GDP. That’s being discussed as an upper bound in the papers, but I suspect we’ll find in coming days that it’s a more realistic estimate.
Source: GS Research, *GS’s preliminary estimate of Sandy; **my guess.
Sunk costs are those that won’t be recovered by future economic activity. In the case of a storm, they’re vacations that won’t be taken, nights out for dinner and a movie lost forever, conferences that were cancelled and won’t be rescheduled.
Of course, other economic activities are simply postponed, and insurance typically covers about half the damage from these types of storms. So some of what you see in the economic indicators after these large storms are results that dip below recent trends for a few months and then go above recent trends, balancing out on average.
And then, of course, there’s the rebuilding, and that’s where net versus gross comes in. Rebuilding stuff that was damaged or destroyed adds to GDP, because of the “G”—it’s gross domestic product, and so it doesn’t net out either standard depreciation—capital consumption, meaning the wear and tear on machinery and other goods that make up our capital stock—or the destruction from hurricanes.
For that, you have to turn to net national product, which in some ways is a more revealing measure of the nation’s wealth than GDP, because it represents what’s actually available for consumption and investment. As the BEA itself says (they produce these data):
…net domestic product is a measure that indicates how much of the Nation’s output is available for consumption or for adding to the Nation’s wealth.
Of course, if they both grow together, it doesn’t much matter which one we look at. But the thing is, they’ve actually been diverging increasingly in recent decades, mostly due to the fact that an increasing share of output depreciates more quickly than the stuff it’s displacing (see figure). Durable goods, for example, comprised 16% of value added back in the 1950s, now they’re about 6%. Computers, software and IT investment depreciates far more quickly than steel bar, and this stuff has gone from being fractional in the national accounts to 5% of value-added today.
Source: BEA, NIPA accounts
The figure shows both trends in real GDP and NDP and well as the ratio of NDP/GDP (real, in both cases), which has been trending down for years. Note also that in recessions, net falls more than gross.
So, rebuilding from Sandy will add back some of what’s lost and that will show up as faster GDP growth than would otherwise occur, though it shouldn’t add to the net side of the accounts. Practically speaking, expect upcoming job and GDP reports to show losses related to the storm—I’ve seen estimates of up to half a point of growth shaved off of GDP in 2012Q4. Hopefully, as we replace and rebuild, we’ll get most of that back next year, but I suspect the upside will be more gradual than the downside.
And if we’re smart, we’ll go significantly further and build some infrastructure to help mitigate the impacts of these 100 years storms that now seem to hit us twice a year, projects like burying power lines or erecting sea walls. And if we’re really smart, we’ll take climate change serious as a factor in play here, and start taxing carbon. We could use the revenue from smart idea 2 to support smart idea 1.
But don’t hold your breath. American governance has not been a hotbed of smart ideas of late.
Thank you for joining the conversation. Comments are limited to 1,500 characters and are subject to approval and moderation. We reserve the right to remove comments that: