February 2021

Sun Mon Tue Wed Thu Fri Sat
  1 2 3 4 5 6
7 8 9 10 11 12 13
14 15 16 17 18 19 20
21 22 23 24 25 26 27
Blog powered by Typepad

« Bob Higgs Gets It, Why Doesn't Everyone? | Main | Economic Freedom of the World: 2011 Annual Report »


Feed You can follow this conversation by subscribing to the comment feed for this post.

I am old enough to remember sending the TV to the repairman (or having him come over to fix it). I think that his services in inflation-adjusted terms cost more than a new TV today. To put things another way, we do not repair most small consumer durables because a new one is cheaper than the cost of repair!

I know someone who, after extensive research, bought a 20 year old GE waffle maker. With stamped aluminum body and heavy cast aluminum plates, it was and is the best available for its price. So there have been some real declines in quality.

Also, if there is a trend toward fewer hours worked to afford things, each new toaster is in fact incrementally cheaper as well.

Computers demonstrate the trend nicely. First computer, Osborne, cpm system, in 1982 cost $2,500, about 10% of gross salary. It had twin disk drives, one for the program and one to work on with 50k capacity. Last computer cost less in dollars, about 2% of gross salary and it has a hard drive with quite a lot of storage capacity.

Still the Osborne functioned as a word processor almost as well us a modern computer if you just wanted to generate pages of original text.


I just received a package in the mail. Parts for my grill that I ordered on the internet.

From a vulgar Keynesian perspective, planned obsolescence is just what we need.

But, with planned obsolescence, we cannot simply state that toasters and toaster ovens are 6 and 7 times cheaper today than they were in 1975. If they last half as long as the older versions, wouldn't that mean that they really are only 3 to 3.5 times cheaper?

This is potentially an argument in favor of one of "The Great Stagnation's" core theses: That we aren't as wealthy as we thought. At least it's an argument against naive cost comparisons; maybe the CPI factors in expected life span when calculating price changes as well as additional features, etc. (And not everything is getting a shorter life span: Cars last much longer than they did 30 years ago, and represent a much larger proportion of wealth than toasters.)

Captured in real wage and income data, why is this not double counting?

It seems to me that if CPI is composed with the view that "a car is a car and a toaster is a toaster" then that will lead to an underestimate of real income coming from the impact of cars getting better and an overestimate of it coming from toasters getting worse. We can't really tell which outweighs the other directly. Of course if the CPI is composed more carefully the question is which quality changes are included and which aren't.

Are there any statistics on what has happened to the profession of small-appliance repairman?

It seems from my observation that is an occupation that has pretty much disappeared. If so, I think it is because A) Appliances last so much longer now and B) New ones are so cheap. So most people just buy a new one when their old one breaks.

I know that when my television set conked out, it took me forever to find a repairman who was in a different city. And the price he quoted to fix it was just $100 less than what I could, and did, purchase a new larger one for.

I work in electronics, so I can comment on this from that point of view...

What Mike P says is true. Another reason is that modern electronics is much more difficult to repair. The process of reflow soldering used to attach components to PCBs is difficult to perform on a small scale.

Another problem is that repairing electronics is labour intensive and quite skilled, so as labour becomes more expensive it becomes more expensive.

I work in the silicon chip industry what's called "Design Evaluation". Designers create chip designs and have them made. Those prototypes are then passed to me. My colleagues and I evaluate them to see how well they perform, if they perform to the intended specs or not. That process takes many months, I report on it as it happens. Improvements are made to the design and new prototypes made. Finally, when a fully-functional and performing design is ready I do in-depth tests and draw up a data-sheet.

Repair would be very useful in what I do because I'm not involved in mass production. To test a particular model of chip I'll have a PCB designed with the chip on and it's associated circuits. Often these boards are very difficult to repair by hand because surface mount components and reflow soldering is used. Components sized "0402" are 1.0 × 0.51 mm so their solder pads are less than 0.5mm. That makes them very difficult to solder by hand, I can do that, but I can't solder the even smaller 0201 components.

What this means is that most PCB board based circuits are unrepairable unless they are designed very carefully to be repairable. We sometimes do that for in-house test boards, but mostly it's too much work and comes with too many other disadvantages in cost and circuit size.

Today it's usual that repair can only happen if whole boards can be replaced. In laptop computers for example (I used to work in that area) the hard drive, LCD screen, WiFi card, Bluetooth card, memory, keyboard and touch-pad can be replaced. That's because they're each separate boards and separate modules. Most of the reason for that isn't repair it's the other benefits of modularity such as having supplier competition, configurability and standard interfaces.

I wrote a long post here that the blog software didn't accept for some reason.

I work in electronics so I have some professional experience of these problems.

The reasons MikeP gives for difficulty of repair are correct. There are two other important reasons, firstly repair is labour intensive and labour has become more expensive. Secondly, the type of soldering we now use on circuit boards is very difficult to do by hand. That means boards can't really be repaired only replaced. That makes things like PCs which use many boards repairable, but many other types of consumer electronic aren't.

PC are designed in a modular way to facilitate competition between part suppliers, and to make standardisation and configurability possible. It just so happens that this also helps with repair, but that's not a big reason for designing that way.

I agree with your general point, but feel that the pile of dead toasters has been underpriced.

Steve's analysis is compelling. It illustrates the increasing modularized specialization of our lives. Modularized structures are a mixture of design and spontaneous emergence. The ease of repair absolutely is one component of the design in many cases, even if it an unintended result in others. Just as important, modularity allows for much more rapid adaptation and innovation. The internet itself is a result of this process. If the older more durable, intermittent repair model were "cheaper" for consumers (having greater value for enough of them), an entrepreneurial opportunity would exist to continue this. Its just hard to believe we are not much better off the way it now is.

The way you measure the cost of a toaster is a great illustration.

This would be a great way to illustrate the fallacy of the poverty line as well. It is easy for people to say that the poor keep getting poorer, but it is simply not true.

The poor of today can afford far more luxuries than the poor in our grandparents' time. The main problem with the poverty level, is that we have a faulty premise. We start with a line that is arbitrarily drawn, then come up with a conclusion based on the faulty premise.

Great job on the post!

The comments to this entry are closed.

Our Books