Evan Soltas
Apr 29, 2012

A Disservice to Recovery

Surprisingly, weakness in services consumption is the recovery's worst dampener

How is this recovery so weak? I'm not looking for a "why" -- I think we have a good sense already of the destabilization of nominal spending and aggregate demand -- but rather a more technical and more proximate cause for why GDP growth has been so weak. (Indeed, some have called this a GDP-less recovery.)

What I found was that perhaps the most significant "dis-contribution" to GDP growth has been in services consumption. And although there are other factors which we will put aside for this post, that realization surprised me, because so much of our conversation has focused on other components of GDP. In the past, I've talked about the level of investment spending as the largest contributor to the level of overall GDP is this recession. But in terms of affecting -- i.e. slowing -- the growth rate of GDP in this recovery, one source is far more significant than the others: services consumption.The slowing growth rate of services consumption alone has taken a full percentage point or more off annual GDP growth. Stop and think about that: a full percentage point. The only other component of GDP which compares is real government expenditures, whose slowed growth is taking roughly half a percentage point off annual GDP growth. (Paul Krugman has recently focused on the latter, calling it fairly enough the "secret of our non-success," but the slowed growth of services consumption is, as I've written, more significant.)

Economists now that consumption composes by far the largest fraction of GDP, although it is often stable enough that smaller but more volatile components -- largely investment -- outweigh it from time to time in terms of how they swing the level of GDP in recessions. But in the long run, growth rates matter. And since consumption composes 70 percent of output, and services roughly 64 percent of all personal consumption expenditures, it follows that services consumption -- at 45 percent of all output -- is more important than I ever thought it before writing this post.

For this reason, services consumption is dragging down the growth of the broader consumption component, and through it GDP. This graph shows the year-over-year percent annual change in services and total consumption. Down go services, and with it, consumption and the strength of our recovery from recession.As we can see, real consumption growth has been on a secular decline since the late 1990s. That is a problem for the American growth model, which is, as we've seen, heavily dependent on services consumption.

The weakness in services is not only reducing the growth rate of GDP in this recovery, but it's also dragging down employment growth. Although much attention has focused on the contraction in goods-producing industries, this has been going on for decades, and goods production is a much smaller sector of our economy. Moreover, after the depths of contraction in goods-producing jobs, that sector is now seeing a historically rare employment recovery. And yet services employment growth remains far off trend -- it is generating approximately 1 million fewer jobs a year than it did in the 1990s, and several hundred thousand fewer jobs a year than it did in the 2000s. (Not coincidentally, the slow services employment growth correlates extremely well with the reduction in services consumption growth.)

We can look one level deeper, too: what types of services are seeing slowed consumption growth? This is an annual change, in billions of 2005 dollars, of the various components of services consumption.A second graph, because I fear that one may be hard to read; this is a graph of the quarterly level of spending in billions of 2005 dollars in each service sector.My interpretation of these data is that the slow growth is services is broadly felt and is not concentrated in any single "problem sector." In the 1990s, services consumption was supported by strong growth in several categories -- mainly, as we can see, housing and utilities, health care, financial, and other. Recreation, transportation, and food and accommodations played smaller but positive roles. Contrast that to today. Spending growth on housing and utilities services is muted, financial services spending just came out of contraction a year ago, a few odds and ends -- spending growth in transportation services and other services has dropped.

Services consumption deserves far more attention than I've seen it receive. It is one of the major weak points of our recovery: services consumption growth alone is taking a full percentage point off annual GDP growth and reducing employment growth by up to 1 million jobs a year. The slowdown in services, lastly, is not concentrated in any particular sector. Weak services consumption, you might say, does a tremendous disservice to the recovery.

A hat tip for the inspiration for this post goes to Naufal Sanaullah