What is cost disease? According to Wikipedia, it

is a phenomenon described by William J. Baumol and William G. Bowen in the 1960s. It involves a rise of salaries in jobs that have experienced no increase of labor productivity, in response to rising salaries in other jobs that have experienced the labor productivity growth. This pattern seemingly goes against the theory in classical economics for which real wage growth is closely tied to labor productivity changes.

The rise of wages in jobs without productivity gains is from the requirement to compete for employees with jobs that have experienced gains and so can naturally pay higher salaries, just as classical economics predicts. For instance, if the retail sector pays its managers 19th-century-style salaries, the managers may decide to quit to get a job at an automobile factory, where salaries are higher because of high labor productivity. Thus, managers’ salaries are increased not by labor productivity increases in the retail sector but by productivity and corresponding wage increases in other industries.

There’s a lot more to cost disease than the kind of salary bloat described by Baumol and Bowen. Tyler Cowen addresses it here. Scott Alexander picks up the ball and runs with it here and here. Arnold Kling, as usual, gets to the heart of the matter:

1. At any given time, you will have sectors where demand is growing faster than productivity (think of health care and education) and other sectors where productivity is growing faster than demand (think of manufacturing). In the sectors where demand is growing faster than productivity, you have rising relative prices, or “cost disease.”

2. In health care and education, you also have a lot of government intervention, and government intervention almost always takes the form of subsidizing demand while restricting supply. Of course, that is going to cause relative prices to be higher, thereby exacerbating “cost disease.”

3. I would argue that there are plenty of barriers to competition in the college market. Accreditation is one such barrier. But there are natural incumbent advantages as well. You may be able to enter the market for high school graduates who are in no way prepared for college. But trying to enter the market at the level of a top 100 college is nearly impossible.

4. There are plenty of barriers in health care, also. Clinics are a good innovation, but the real expenses in health care are in chronic illnesses, and clinics do not compete to treat diabetes, Alzheimer’s, and so on.

5. It is in the nature of organizations for middle managers to try to build empires, adding to cost without necessarily creating value. In for-profit businesses, the owners have an incentive to check this, because the owners want to maximize profits. In non-profits, the natural checks operate only when revenues are not rising to cover the cost of expansion. Non-profits only worry about the bottom line when it threatens to go negative.

In short, some “cost disease” is natural. At any given time, some industries will have demand growing faster than productivity. However, much of it is artificial, as government subsidizes demand and restricts supply. Finally, some of it results from the fact that non-profits are less efficient than for-profit firms.

As a former officer of a tax-funded, non-profit, professional services corporation (TNPSC), I know about cost disease in the quasi-government sector of the economy.

First of all Baumol and Bowen’s definition of cost disease as salary bloat, though incomplete, is correct. Because my company — call it XYZ Corp. — derived all of its funds from government sources, our salary policies required the approval of government contracting officers. How did we gain that approval? Every few years XYZ hired a consulting firm that had access to salary data for private-sector companies. The consulting firm would then undertake a “study” to compare private-sector salaries with those of XYZ. Lo and behold, by selecting the right set of private-sector companies and the right set of jobs in those companies, the consulting firm found that XYZ’s salaries lagged, and should be boosted by more than the usual annual rate to keep pace with XYZ’s private-sector “competitors.” XYZ’s above-market benefits package (approved by contracting officers) and below-market turnover rate were conveniently ignored.

A related trick was to set executive salaries so that they kept pace with the salaries of executives at other TNPSCs. And how did the larger TNPSCs justify the high salaries to which XYZ aspired? With “studies” showing that their executive salaries lagged those of their private-sector “competitors.”

It’s a joke to compare salaries paid by relatively stable TNPSCs — organizations that have cozy, long-term relationships with their government sponsors — and salaries paid by private-sector companies. In addition to cushy benefits packages, employees of TNPSCs are well-insulated from competition, unlike their private-sector counterparts. Thus employees of TNPSCs are compensated not only with handsome salaries and benefits, but they also enjoy a high degree of job security. Which is why turnover rates at TNPSCs are low relative to private-sector companies.

How does Kling’s list of reasons for cost disease apply to TNPSCs?

Demand vs. productivity. I’m unfamiliar with the current state of “demand” for (i.e., government spending on) TNPSCs. But over the long haul, since the inception of TNPSCs during World War II, government spending on them has risen by orders of magnitude. It’s probably safe to say that the productivity of TNPSCs has risen little. Advances in computation and data storage have enabled such firms to collect and analyze data pertaining to a broader range of subjects, and to do it more rapidly. But there’s been little real innovation in the tools of analysis, most of which were devised during World War II and the decades immediately following the war. And the basic approach to “solving” the problems of government agencies remains the same as it was in World War II: Define the problem, collect the relevant data, analyze the data to find a preferred solution to the problem, and report the results to the government client. It was and still is a labor-intensive process.

Government intervention and barriers to competition. TNPSCs are formally designated by the government. There are relatively few of them, and most of them have pedigrees that date back to the 1940s, 1950s, and 1960s.

The cozy relationships between TNPSCs and their various government sponsors changed somewhat in the 1990s when profit-seeking professional-services firms declared war on TNPSCs. Some TNPSCs suffered funding cuts as a result, but the cuts were far from fatal and TNPSCs compensated by finding a broader range of government sponsors to maintain them in the style to which they had become accustomed. Some of them spun off for-profit counterparts, with the aid of fees earned on government contracts. More stringent contracting procedures imposed as a result of the war on TNPSCs also forced them to emulate the task-by-task funding of for-profits. But that’s just a cosmetic change; it adds to the cost of running TNPSCs, which the government defrays, of course.

Empire-building. Kling’s analysis is spot-on. Here’s some personal testimony: From the mid-1980s to the mid-1990s, the component of XYZ that I managed grew significantly. Staffing probably doubled, and costs rose accordingly. It wasn’t until 1995, when XYZ suffered funding cuts resulting from the war on TNPSCs, that my empire shrank. I ran the support side of XYZ, which encompassed contracting, accounting, information services, publication services, facility operations, security, computer operations, computer programming, and personnel (called “human resources,” of course). I handed off the computer programming function to another manager, who could sell its services to new clients, and cut the staffing of the other functions by about 20 percent. I did it so that the managers of the research divisions — the ones that do the work for which clients pay — could take much smaller staffing reductions. Did the 20-percent cut in support services hinder the work of the research divisions? Not that I noticed.

Why, then, did I grow the support division? Because I could. That’s empire-building, and I was far from the only empire-builder in XYZ or other TNPSCs. Ambition abounds, and it leads to empire-building for as long as the money is there to support it.

Kling is right. Cost disease prevails where government subsidizes demand and restricts supply. TNPSCs are small potatoes compared with the health-care industry, which is the largest component of the quasi-government sector of the economy. The industry is government subsidized (e.g., through Medicare, Medicaid, and research funding) and sheltered from serious competition by a vast web of laws and regulations. Those laws and regulations also impose heavy cost burdens on health-care providers, their suppliers, drug companies, and insurance companies. But the burdens are defrayed to a large extent by government funding. It’s a vicious cycle that’s largely responsible for the high cost of health care in the United States.

It’s even worse in the official government sector, which includes the vast federal apparatus, all manner of State and local agencies, public schools and universities (and heavily endowed private ones) — and myriad contractors to all of the foregoing. Massive cost overruns, dismal performance, administrative bloat, pension-fund raids on the public treasury (i.e., taxpayers), open-ended “entitlement” programs, etc., etc., etc.  It’s the non-accountability swamp. And the only way to drain it is to say “no” — period, full stop, end of discussion.