How to Cure Baumol’s Cost Disease: Turn Services into Goods

Baumol’s Cost Disease is a phenomenon that was formally observed by William J. Baumol and William G. Bowen (history snubs poor Mr. Bowen) in their 1965 paper, On the Performing Arts: The Anatomy of their Economic Problems. Baumol and Bowen point out that it takes the same number of musicians the same amount of time to perform a Mozart string quartet today as it did when Mozart was alive. This implies that the productivity of classical musicians, or how much output (music) they produce per hour of labor, has remained constant over time. Meanwhile, productivity in sectors that employ more technology, like manufacturing, have seen productivity rise significantly over the years. This is where Baumol and Bowen’s work gets really interesting.

In addition to their observation that productivity has not increased in some sectors of the economy, like the robust Mozart string quartet industry, they also noted that real wages, meaning wages adjusted for inflation, have still increased in these sectors over time. This second observation is in direct contradiction with one of the basic tenets of classical economics: wage equals the marginal productivity of labor. In other words, at the margin, employers pay their employees based on how much they produce. But if the productivity of musicians hasn’t gone up, then why have their wages? The answer, say Baumol and Bowen, is that wages rise in stagnant sectors of the economy because producers must compete for labor by paying a competitive wage rate. If the real wage of a classical musician were the same as it was in Mozart’s time, no one would choose to be a classical musician. The opportunity costs, in this case the wages of other available jobs, would simply be too high.

This phenomenon is known as a “cost disease” because it has dire implications for the economy at large. As real wages increase in sectors of the economy that do not experience productivity increases, a larger share of the economy will be devoted to producing these services. And since the government disproportionately supplies many of them – e.g. health care, education, law enforcement – Baumol’s cost disease also predicts the government’s share of the economy will rise as well.

The string quartet example can be loosely extrapolated to the whole service sector of the economy because the defining characteristics of services describe a string quartet as well. There are generally considered to be five such characteristics: intangibility (you can’t hold it in your hand or store it in a warehouse), perishability (they cannot be reused once delivered), inseparability (they are dependent on the person providing the service), simultaneity (they are delivered and consumed at the same time),  and variability (every service is unique and cannot be replicated exactly). For the purposes of this essay, you can think of these five characteristics as barriers to increasing productivity. Before we talk about overcoming these barriers, let’s contrast services with goods. Goods are tangible, relatively nonperishable, separable from the provider, and consumable after delivery. For these reasons, the provider of a good is less critical to how and when it is consumed. Furthermore, goods can be mass-produced in factories, which are amenable to technological advancements in production processes. Services tend to resist technological innovation because the person providing a service must be present while it is consumed and he is the main determinant of its quality. Much of what we call technological innovation involves removing the human from the production process, e.g. robots on an assembly line. Since most service work is non-routine, it is difficult to take out the person delivering the service.

If services are resistant to increases in productivity and goods are not, I argue that we can cure Baumol’s cost disease by transforming services into goods. I will use the education industry as a prima facie case for how to achieve this type of transformation. There is a revolution going on right now in higher ed regarding online education. Companies like Coursera, MRUniversity, and Udacity have created online platforms for delivering college level classes. Their approach differs from other experiments in online education, like MIT OpenCourseWare and Edx, in that these companies are not merely videotaping professors giving their normal classroom lectures. They are creating classes from the ground up that are optimized for the online experience. For example, many of the classes emulate the format used by Khan Academy, which combines notes, pictures, and diagrams with a disembodied voice narrating the lesson. They claim this allows students to focus on the material and follow along more comfortably. In any event, the key technology all of these online platforms exploit is the same: video.

Though video technology has been with us since 1951, education innovators are just now using it to effectively disrupt higher ed. It is the perfect technology for this task because it flips the five characteristics of education that make it a service. Video enables lectures to be “reused” after they have already been delivered (perishability); it allows them to be delivered without the provider present (separability); it permits viewing to occur after a lecture is delivered (simultaneity); and it makes the service homogenous (variability). Once the college class has finished the transformation from a service to a good, it can be mass-produced (at a marginal cost of zero) and sold to the public. According to the statistics on the Khan Academy website, Sal Khan, the company’s founder, has delivered more than 200,000,000 lessons to students across the globe. Talk about productivity.

P.S. I realize that online video does not directly address tangibility. Though this characteristic is irrelevant to increasing productivity in higher ed, if it is an important feature to you, I suggest you download the videos, burn them to a DVD, and clutch it tightly.