The Contagious Cost Disease

I found a notice in my letter box the other day. The notice informed me that a postal delivery service, run by the private company pin Mail, had not been able to deliver a parcel. I was asked to go and pick up my parcel at a supermarket in another part of the city. Using public transport, I figured, the whole journey would have taken about one hour and half. I called pin Mail. “No. Sorry. We cannot deliver the parcel a second time”.

So here we are. E-commerce dealers like Amazon enable us to go shopping without even leaving our desk. That’s technological progress. On the other hand, we observe severe cutbacks concerning the quality of labor intensive services like postal delivery. In the old days, there was postal delivery up to three times a day. That’s history. Today, we’ve arrived at DIY. “Please ride all across town and pick up your parcel yourself. Thanks! Your delivery team.”

What happened? The book (which happened to be in the parcel which I finally received because  Amazon sent me a second copy) contains an answer to that very question. It’s The Cost Disease: Why Computers Get Cheaper and Health Care Doesn’t by the economist William Baumol. Baumol is now in his nineties (and still a brilliant mind, as anyone who has witnessed him as a conference speaker will confirm). The book is the latest item in a series of writing on the cost disease which goes back to the 1960s. Basically, there is nothing new in Baumol’s recent book – besides the fact that data which our economy produces continue to render plenty of evidence for the existence of the cost disease.

Cost Disease: The Hypothesis

The cost disease-hypothesis claims that there are certain sections in the economy which are stagnant. They are not profiting from innovation and technological progress. They are not experiencing any efficiency gains. Baumol’s famous example for such kind of business is a string quartet, who delivers the same kind of music and performance today as it did two hundred years ago. Meanwhile, other sectors (like the car industry) did profit from efficiency gains. They are able to produce more value per hour with less labor involved. The result is, that either the respective products can be sold cheaper or that workers are being paid a higher salary. This is bad news for our musicians. In the first case, they will witness that the price of the service which they are offering is rising, relatively speaking – just because other things are becoming cheaper. As a consequence, less people might be willing to buy the musicians’ services. The natural reaction for our musicians would be to reduce the price of their services in order to increase sales. The only way for them to this this, though, is to reduce their own salary and to intensify their work (for instance by giving two instead of one gig every day). Now, consider the case that productivity gains in the car industry were handed over to employees in form of rising salaries. Again, this is bad news for our musicians, because now their income is reduced by that fact that everyone else besides them is enjoying a higher paycheck. To keep up, they will have to sell their services for a higher price. Again, this might lead to problems, because clients might not be willing to pay higher prices for music, when all other prices stay the same.


There have been lots of arguments about whether or not the cost disease actually exists. After all, it’s just a model. There is no chance to observe the cost disease out in the wild: For every instance that one could cite as evidence (such as rising costs in health care and education), other explanations are readily available. Some economists (most prominent: Tyler Cowen) also claim that such a thing as a stagnant sector just doesn’t exist, because every sector and every business profits from efficiency gains one way or the other.

The most clear cut example I can think of which makes plausible the existence of the cost disease is the rise in prices for scoops of ice cream sold by small vendors. Ice cream is labor intensive concerning the production in a small manufacture (as many small vendors do) and in regard to selling it. (The ice cream shop is open every day, all summer long. But not every day is a bright and hot Sunday.) In the early 1980s, the price for one scoop of ice cream was around 25 cents (then 50 Deutsche Pfennige). Prices in Germany generally jumped from 74,6 in 1980 to 105,7 in 2013 (assuming 1995=100). If the price for a scoop of ice cream would have evolved like the rest of the economy, then it would be around 40 cents by now. Actually, it’s 1 Euro. What is this if not cost disease caught in the act?

As I said, it is difficult to observe the cost disease in other areas with equal unambiguity. I once did a research on the economic history of print journalism. In one of my sources I found that within the typical German newspaper, absolute costs for editorial work had increased from 100 percent in 1979 to 195 percent in 1990. No other budget post within the production process of the ‘newspaper’ had similar growth rates. (To my great disappointment, statistics from 1990 onwards were not available to business outsiders.) These numbers support the assumption that the business model newspaper would have run into trouble soon or later even if private TV (which attracted many advertisement clients who formerly bought their advertisement space in the newspapers) and the internet would never had happened. Editorial work is just becoming more and more expensive. It seems that the cost disease is at least part of the gang who is killing the newspapers (see also John B. Judis’ recent piece on the New Republic on this issue).


Assuming that the cost disease does indeed exist: do we really have a problem with it? Baumol himself doesn’t think so. He states, over and over again, that the skyrocketing costs for labor intensive products and services must not really be troubling, after all: We can afford rising expenses for health care, education and ice cream from our local dealer, because the amount of euros that we earn per hour today is so much more than it was in the 1980s. Baumol just warns us that we should not make the mistake to expect that costs for education and health care (and ice cream) would stay the same. In other words: the solution to the cost disease which Baumol proposes is a change in expectations and attitudes. Like: “Let’s face the fact that the cost disease exists and adjust our budgets accordingly. We just have to spend a part of the additional money which we have in our wallets thanks to technological progress for higher prices for music, education and health care.” It sounds easy. Is it?

True: Whereas the average employee in Germany made €14.315 (after taxes) in 1980, he now makes €18.508 a year – inflation already counted out. There are actually 4.000 more Euros in our wallet every year today than there was in 1980. But does this mean that we have €4.000 more to spend on music, education and health care? Maybe we had. But today, the money is all swallowed by upgrades in our consumption habits. Statistical figures reveal that we are actually spending these additional €4.000 for transport, culture and leisure, and other things. There is just no way that we can decide, by some declaration of will, to cut these €4.000 out of our current budget in order to pay again for those labor intensive products and services which have gotten so expensive and which we have ceased to buy.

I think there are some straightforward practical conclusions one can draw from this. If you plan a career or think about starting a business in a labor intensive field, you better check out if the business that you are heading into belongs those kind of businesses that manage to hand over rising costs to customers and clients (as local ice cream vendors apparently do) or into the other category which has to react to rising production prices with mere work intensification and reduced salaries for employees. If you pursue already a career in the latter field: get out, better sooner than later.

If you are a wise king, concerned about the wellbeing of your people, you might want to consider one of Baumol’s further suggestions, addressing not only rising prices for labor intensive services but also environmental harm caused by those products which, due to productivity growth, have become more and more available. The proposal is:

“(…) reducing taxes or granting subsidies on beneficial products whose availability is threatened by increasing costs, and imposing higher taxes on less beneficial items (…)” (p. 68).

Defining and separating beneficial products from harmful ones might not be an easy task. It might even turn out to be impossible. But we should give it a try to at least think about it. If Baumol is right, productivity growth is not only of ecological concerns, but also a threat to our quality of life, which depends to a great extent on jobs and on affordable products and services in culture, education and healthcare. Shifting resources from innovative areas of the economies to stagnant ones might not be fancy (see Per’s recent report on how the head of the tax committee in the Swedish parliament reacted to a like-minded proposal). But it could well be the only option that we have.