Illuminating a Bit of Urban-Myth Economics

16 January 2021

On YouTube, I encountered a video selling something (I didn't get to whatever it was) based upon a controversial theory, and using economic prejudices to make his case. To make the economic part of his case, he told an old, true story, but left-out an important detail.

There are a few incandescent light-bulbs that have been in continual use for many decades. (He referred to one of these.) They haven't needed to be changed. But that doesn't somehow prove that the manufacturers of light-bulbs have formed a cartel that avoids selling us cost-effective light-bulbs that will last many decades.

The bulbs that last for decades run on DC (direct current) rather than AC (alternating current). AC displaced DC because of issues of generation and of transmission; you'd be paying more for electricity if your power company used direct current. And, that these bulbs are run continuously means that they are never turned-off, and thus have only been turned-on once. It is the strains on the filament from being turned-on repeatedly and from running alternating current that cause our incandescent bulbs to fail more quickly. If you really want your bulbs to last longer, then install rectifiers in your lamps, and either never turn the bulbs off or also install devices that gradually increase the current when the light is turned-on.

Or just accept that you're trading higher bulb costs for lower power costs and lower costs of lamps.

Tags: , ,

4 Responses to Illuminating a Bit of Urban-Myth Economics

  • Gaal Yahas says:

    There's evidence that the Phoebus cartel did intentionally design shorter lifespans into lightbulbs.

    https://spectrum.ieee.org/tech-history/dawn-of-electronics/the-great-lightbulb-conspiracy

    Possibly the most famous long-running lightbulb, the Centennial Light, is now running at 4 watts (originally 30 watts). I don't know whether it's AC or DC that powers it. Because it operates at such small load compared to its original spec (and certainly compared to the demands that more modern users put to lighting devices), I think of it more as a curiosity than data for anything.

    • Daniel says:

      It's very good to have a comment from you!

      Perhaps I should have noted the possibility of using bulbs at lower voltages than specified by manufacturers (with bulbs then having greater up-front costs given the rate at which photons were emitted). One could modify a lamp to use a tap of lower voltage or itself to lower voltage or to use bulbs rated for higher voltage.

      Firms often try to cartelize but, without state assistance, cartels crumble over time, as did the Phoebus cartel. The apparent reasoning of Mr Krajewski (the author of that piece) is problematic; it would be a surprise in any case had the consortium not been motivated by a desire to profit, but that isn't itself somehow evidence that they didn't improve things for the consumer. Nor, in judging them or their Japanese competitors, do we want uncritically to accept that summing the price of light bulbs over time gives a proper measure of cost; more specifically, one must account for the time-value of money to the consumer.

      I hadn't been keeping track of the bulb now called the Centennial Light, and didn't know that they were no longer running it at a original voltage. I understand the desire to preserve it, and yet this lowering of voltage indeed reduces it further to a curiosity.

      • Gaal Yahas says:

        I am glad you're still writing!

        I don't really know anything about the economics of this, and only know very little about the engineering as well. From what I understand, it is perfectly normal for an incandescent bulb to reduce in luminosity over its lifetime, as the filament gradually evaporates. Typically, thermal strains will eventually be the cause for a light going out as the material wears.

        The Wikipedia article on incandescent bulbs cites a materials engineering guide, bringing some interesting facts about the regime lightbulbs typically operate in:

        https://en.wikipedia.org/wiki/Incandescent_light_bulb#Light_output_and_lifetime

        So given an existing bulb, you can tune its lifetime vs. light output as tradeoffs of each other (as well as power and color temperature). These tradeoffs are not in general linear. This is not dissimilar to how some people overclock their computer CPUs to squeeze out more performance from them, at the expense of reliability or expensive cooling systems. Less commonly, some people intentionally underclock their systems because they do not need the full "sticker" performance, and instead want the machine to be more reliable.

        Computers are more sophisticated than light bulbs; and with the spread of laptops and other mobile computing devices, instantaneous power draw and dynamic power scaling become attractive features of the system.

        I don't know the exact physical reasons why pre-cartel lightbulbs used to last longer. It would be convenient to say that they just used thicker filament, so there was more to wear away until the yielding point. If that were true, one could argue that the manufacturers were purely greedy: by saving a trivial amount of critical material, which saved a tiny fraction of the cost of a bulb, they got to sell twice as many units. I don't think this is quite accurate from the physical side though. Thicker filament means other physical characteristics too, so it's possible there were other factors involved. And like I say, I don't really understand the economics of this. But the cartel was operating in secrecy, and when the secret was exposed GE was not permitted to sell the artificially limited bulbs. So buyers were not offered a concurrent choice between cheap 1000-hour bulbs and pricier 2500-hour bulbs, first because of the cartel and then because of regulation.

        Modern LEDs of course do operate in DC and much lower voltages. They are both more power efficient and last much longer than incandescent bulbs. I think the least reliable part of a household LED lamp, as well as the most expensive, is the circuitry to deal with the high voltage AC grid. It took profound technological advances to make both the light emitting diode and the power supply that fits in an Edison screw base and they are still being optimized for many parameters consumers care about: price, color, warmup time, longevity and so on. Given the choice it looks like markets did go with the up-front pricier device. Of course, regulators had a lot of say in this matter too. You mention time-value: I don't *think* that I'm brainwashed, though, in saying that I for one am kind of glad to be [mostly] done with the hassle of changing burned lights.

        • Daniel says:

          Actually, when it comes to lighting, in America the hand of the government has been pretty heavy. A popular wattage of incandescent bulb was outlawed at the Federal level, and in California regulation had the electrical producers heavily subsidizing fluorescent bulbs (so that for a time they were cheaper than incandescents) and offering to replace halogen lamps with lamps of a different sort.

          I bought a large stock of the fluorescents during the period of subsidy, and gave half to my parents (in Arizona!). My experience with these bulbs has been good except for disposal. There is no good alternative for me in my area.

          My experience with LED. home-lighting is much more limited, and mixed.

          (I remember when LED. home-lighting was first proposed. The idea seemed quite far-fetched, and it took a significant amount of time before it became practical.)

          My time-values are different from those of most people, and I take it that yours are as well. But of course that in-and-of-itself doesn't make us more rational, and certainly doesn't make us more economically rational.

Leave a Reply

Your email address will not be published.

This site uses Akismet to reduce spam. Learn how your comment data is processed.