Why the Cost of Government Services Is Rising

A look at productivity growth shows why the amount of Americans working manufacturing is shrinking, and why this will cause the price of government programs to rise.

By + More

Chad Stone is chief economist at the Center on Budget and Policy Priorities.

Productivity growth—the growth in how much the economy can produce without workers necessarily working longer or harder—is the key to a rising material standard of living. It also plays a central role in explaining why a shrinking share of Americans work in manufacturing—and why we probably should get used to paying more to support government programs we value, like public education and Medicare.

Washington Post columnist Steven Pearlstein recently made these points by drawing on the insights of economist William Baumol and using one of my favorite pieces of economic pedagogy, the simple two-sector general equilibrium model.  If I haven't scared you away with that, here's Pearlstein's synopsis of Baumol and his collaborator William Bowen's insight (which they originally made about rising costs in the performing arts but subsequently applied more generally to the disparity between productivity in the goods versus the services sector):

No matter how innovative people were in coming up with new technology and new ways of organizing their work, Baumol and Bowen reasoned, it would still take a pianist the same 23 minutes to play a Mozart sonata, a barber 20 minutes to cut the hair of the average customer, and a first-grade teacher 12 minutes to read her class Green Eggs and Ham. Based on this observation, the duo predicted that the cost of education and health care would inevitably outstrip the price of almost everything else.

[ See a collection of political cartoons on the economy.]

Here's the simple story about how that happens in an economy in which productivity rises much faster in manufacturing than in services and government: If the division of labor between manufacturing and services remained the same in the face of rapidly rising manufacturing productivity, we would soon be overwhelmed by a flood of manufactured goods but stuck with the same level of services. In fact, we would almost surely prefer to take the gains from higher manufacturing productivity in the form of a more balanced mix of goods and services.

That's where supply and demand kick in. The price of services will rise over time relative to the price of manufacturing goods to prevent a glut of manufactured goods and a shortage of services. Service providers will have to offer higher wages to attract the workers they need out of manufacturing jobs and into service sector jobs.

[ See a Slideshow of the 10 best cities to find a job.]

This, of course, is a toy model; the real world is more complicated. It's a topic for another day—why wages generally have not kept up with productivity growth since the 1970s and what determines the level of pay in manufacturing versus other industries. Moreover, the transition to a more service-oriented economy is not a smooth process but rather proceeds in fits and starts, with disruptive downsizing and plant closings that throw people out of work with no guarantee they will find a new job quickly. Government policy has a role to play here, not by trying to block necessary structural adjustments, but by doing its best to maintain full employment and a healthy job market and helping workers who face permanent job losses make the transition to a new job.

Rising costs to provide education, healthcare, and other government services are the flip-side of falling costs to produce manufactured goods in the toy model. Obviously, the story is more complicated in the real world here as well. First, it's an overstatement to say there is no productivity growth in education, healthcare, or other government services, or that artificially high prices and wasteful procedures are not a factor in healthcare. And, yes, the development of new health technologies has been an important force driving healthcare spending up, not down. But that's because those don't reduce the cost of supplying existing services the way productivity improvements in manufacturing drive down costs; rather, they tend to increase demand for higher-cost medical services in today's healthcare market.

[ See a collection of political cartoons on healthcare.]

None of this detracts from the key Baumol-Bowen insight, however—that it's perfectly natural for an economy benefiting from strong productivity growth in its private goods-producing industries to spend an increasing share of its income on services, including government services. Trying to hold government spending to its average historical level as a share of national income or GDP would likely result in an unacceptable cut in the real level of government services over time.