You have surely heard the phrase economies of scale — meaning that when you make many copies of something each instance costs less than when you make only a few copies. Large companies are said to benefit from “economies of scale” — so there is pressure to become bigger. Every introductory economics textbook says something like this.
Here’s what none of them say: The more of Item X made by one company, the more “sterile” Item X becomes, meaning the less Item X is able to spark innovation. Call this sterilities of scale. You have never heard this phrase — I invented it. (I cannot find it anywhere on the Web.) But it is just as obviously true as the notion that when you make more of something you can make each one more cheaply. If 100 widgets are made by one company, there is going to be less innovation surrounding widgets than if 100 widgets are made by 10 different companies. Sterility of Scale 1: When ten different companies make something, more people are studying and thinking about and pursuing different ways of making it than if only one company makes it. Sterility of Scale 2: The more profitable a single item becomes (due to low cost of manufacture), the more pressure not to change anything — not to kill the goose that lays golden eggs. Sterility of Scale 3: The larger the company, the more employees who care only about preservation of their fiefdom (comparing 10 companies of 10 people each to 1 company of 100 people). See how obvious it is that sterilities of scale exist?
The two concepts — economies of scale and sterilities of scale — are equally elementary. But only one is taught. Study of innovation should be 50% of economics but in fact is close to 0%.
This is why Tyler Cowen’s The Great Stagnation is so important — because it begins to point to this great gap. Jane Jacobs did so, but had little or no impact. (At a Reed Alumni Gathering I was seated next to a professor of economics. “What do you think of the work of Jane Jacobs?” I asked her. “Who’s Jane Jacobs?” she replied.) I think human decorative preferences are so diverse (chacun a son gout, no accounting for taste) for exactly this reason, to avoid sterilities of scale. Diversity of preference makes it easier for many different manufacturers to thrive, which increases innovation. For example, diversity of furniture preference makes it easier for dozens of furniture companies to survive, thus increasing innovation surrounding furniture. Clayton Christensen’s The Innovator’s Dilemma describes many examples where large companies were much less innovative than smaller companies — so much so they often went bankrupt. Which suggests sterilities of scale can be fatal.
If there were more understanding that ten small things are going to be more innovative than one big thing, I like to think that scientists would better understand the value of very small research and grant sizes would go down. An illustration of the general cluelessness is someone who wrote to Andrew Gelman complaining that a sample size was only 30.
I started thinking about this after hearing Nassim Taleb discuss economies of scale (e.g., here).
> If 100 widgets are made by one company, there is going to be less innovation surrounding widgets than if 100 widgets are made by 10 different companies.
So, was there more or less ‘innovation surrounding’ computers when there were dozens or hundreds of companies selling computer processor chips for thousands and hundreds of thousands of dollars, as compared to the handful now selling for hundreds?
I question sterility of scale #2 in this sense: the more profitable a single item becomes, the more *potential competition* is drawn out of the woodwork. More people *outside* Apple are spending more money and effort developing tablets now as a result of iPad being so visibly profitable; if it had been less profitable it wouldn’t be seen as a valuable niche worth pursuing. All these competitors both provide innovation directly and incentivise Apple to keep innovating – concrete example: Apple *had* to add a camera in the latest iPad revision to keep in the running.
Sterility of scale #1 also seems problematic in that real standardization of one product within a product category tends to spawn massive innovation in the form of add-on products. When automobiles are all handcrafted, there’s not much of an aftermarket business opportunity making things like custom floor mats and spoilers and foglights. The iPhone and iPad may themselves be pretty “sterile” but the ecosystem they created in cases and apps and accessories is vibrant and would be much less so without a stable base to work on.
Sterility of scale #3 is much more plausible than the others. Though there are a couple of conflicting tendencies. One is that companies generally *get* really large by providing value to customers. To the degree that empire building interferes with profitability, it prevents company growth; the fact that one company won out suggests – doesn’t prove, but does suggest – that it’s the most efficient provider *despite* the tendency empire building.
Still, that one is a real phenomenon. The term you might be looking for is “diseconomies of scale”.
https://en.wikipedia.org/wiki/Diseconomy_of_scale
Interesting. We agree and are in the process of trying to integrate innovation into our curriculum. There are some recent books by Baumol that are looking at this. And, there’s Clayton Christensen’s work as well. But, it’s a challenge. In order for students to be able to “think outside the box,” they need to know what’s in the box.
Though innovation certainly isn’t a center of most micro theory, I disagree that this question hasn’t been studied. There is a huge empirical literature on “the Schumpeter hypothesis,” which is roughly that large concerns are responsible for innovation. His conjecture is that large firms get large because they are doing something right, and, absence some significant barriers to entry, they are going to have to continue to remain innovative or they will get caught by their rivals. Wal-Mart might be a good instance of a tremendously innovative firm that had its profitability cut out from underneath it because its rivals mimicked its cost-saving measures. If you type “Schumpeterian hypothesis” into google scholar, you’ll get plenty of discussion of this very topic. Here’s where Schumpeter makes his case:
https://www.sp.uconn.edu/~langlois/Creative%20destruction.htm
Oh, and this,
https://www.schumptoberfest.com
Sterility #1 might be best rephrased as “market competition encourages innovation”, a concept that *is* taught.
Sterility #2 is one of a variety of pitfalls profitable companies can fall into; the chief check on this is…market competition.
Sterility #3 is one of a variety of pitfalls that can eventually limit the growth of large firms, aka “diseconomies of scale”. Again, the chief check on this is market competition.
So you could summarize all of these as “competition – including *potential* competition – encourages innovation.” That is taught as the conventional wisdom.
Economics tries to explain why the world look the way it does, so one learns that:
(1) Competition produces variety and innovation
(2) Economies of scale (including specialization of labor) produce large firms that reduce costs
(3) Diseconomies of scale ultimately limit the size and/or effectiveness of large firms.
To the degree that people don’t learn (3) or insufficiently appreciate (1), there are calls to forcibly break up large firms. (Many such calls are misguided because they under-appeciate (2) or don’t realize that government regulators themselves are also subject to (3).)
Good stuff. As soon as I started reading this, I thought of NNT. In the updated version of The Black Swan, he goes further into depth on the concepts of fragility/anti-fragility that you linked to. I love how he uses mother nature as a model that economics should strive to follow. The hono genus has survived for 2.5 million years not because there was a premium placed on efficiency, but because there was an emphasis on redundancy: 2 eyes, 2 lungs, 2 kidneys, etc. We could have traded some of these redundant features for a bigger brain, more muscle mass, etc, but at some point you need to make some trade-offs to create a robust system resistant to natural shocks. Modern economics could learn a lot from mother nature.
I see evidence of sterilities of scale in the local public school districts in California. The small school district in a nearby town with just a few schools frequently runs single class experiments (e.g. does a multi-grade class work better for students?), but the large, well-funded district that my kids attend is proud of their ability to have every teacher doing the same lesson in every classroom every day. I think that Sterility of Scale 3 is applicable in this case.
Gwern and Glen, yes, a tiny fraction of products benefit from standardization — sometimes these are called “producer’s goods” (e.g., nuts and bolts, chemicals for industrial processes), sometimes “platforms”, sometimes “two-sided markets”. If that’s what you mean, I agree. That is another basic point about innovation missing from introductory textbooks. I am not saying sterilities of scale are the only thing that differs between large and small companies. In the case of platform goods, what you say is also important.
Here’s the first sentence of the Wikipedia article you link to: “Diseconomies of scale are the forces that cause larger firms and governments to produce goods and services at increased per-unit costs.” Nothing about innovation. The whole article mentions innovation in only one section. Of course economists know that innovation exists; they simply pay very little attention to it. (Just as epidemiologists know that the immune system exists but pay very little attention to it.) The Wikipedia article is one more example.
Thanks for the Schumpeter link. If I were trying to teach about innovation to college economics students, I think I would tell stories about how a few products began. Then I would say: “You’re lucky. Economists haven’t studied innovation very much. They haven’t come up with basic rules about what makes innovation large or small. So you can. I want you to study several examples of how products began and from those examples come up with testable generalizations about innovation.” I might give the students links to stories of how products/services began or I might require them to find the stories themselves. This is a lot more exciting and educational than the usual stuff that is taught in intro economics and in this sense it is easier to teach about innovation than the more conventional stuff.
Have you read _The Wealth Of Nations_? Adam Smith had a very different model of innovation than you do and emphasizes sources of it that you might have overlooked. Consider Smith’s example of a pin factory. Suppose we have ten firms making pins from scratch, where each person does *all* the work involved in making each pin. You seem to imply that should be good for innovation because ten people will have ten different worldviews, ten different backgrounds. But now suppose one larger company applies assembly-line technology to making pins. One of the ways you get economies of scale is that at the new larger firm there’s likely to be one guy who *all he does* is sharpen, and he does this full time, hundreds or thousands of pins a day. Since all he thinks about is pin-sharpening and since he does so much of it, he *gets better* at it. He is likely to invent new innovative processes to make pin-sharpening more efficient. Ten companies making 1/10th as many pins – or 100 companies making 1/100th as many – are not in a position to do this. For a firm that makes a few pins a day, sharpening is such a small part of the overall task/job that realistically they’ll never make it much more efficient. Ten people take a lot longer to climb a learning curve than one person doing the same task ten times as often.
Market concentration thus can enable *more* technical innovation. Concentration of capital at industrial scale makes investments possible that have huge returns which weren’t possible for smaller-scale producers to achieve. So that is one story economists tell about how innovation occurs. Now, it’s not *quite* clear what you mean by “innovation” in this context – does accomplishing the exact same goal more cheaply count? – but in the way economists define it they do have a story to tell. In their story (in which cost savings *do* matter), economies of scale tend to matter a lot *more* than diseconomies. It’s not clear to me that they’re wrong to think this. If you think they are, you need to make more of a case.
“Learning by doing” is a related concept in economics. Firms learn how to produce things better and more cheaply as they produce more, but there are diminishing returns and we don’t learn as much from the 1000th unit as the 10th.
https://en.wikipedia.org/wiki/Learning-by-doing_%28economics%29
The idea was applied to Intel and microprocessors here, check out the graph on page 12: https://www.antitrustinstitute.org/sites/default/files/scherer%20commentary.pdf
Economies of scale are something you can measure or calculate. In Smith’s example, he ballpark-estimated that a combination of division of labor and learning-by-doing led to pins being hundreds or thousands of times cheaper that they otherwise would be. Now that we can calculate much more precisely how much innovation we get from scale, this helps economists explain questions like: “Why do firms exist at all? Why is there trade between nations? How do countries with relatively free trade prosper?”
Diseconomies of scale seem a lot smaller in magnitude than that, but do help to answer second-order questions such as: “Why don’t firms just keep growing and consolidating until one firm produces everything? Why don’t we just let the government produce all the goods and give them away for free?”
When you say “a tiny fraction of products benefit from standardization”, I can’t think of any products offhand that *aren’t* in that “tiny fraction”. Standardized cars are a platform based on standardized tires which are based on standardized rubber…standardized computer chips enabled innovative computers, standardized computers enabled innovative software, then standardized software itself became a platform… standardized food staples made food cheaper and safer, standardized electronics made fires less likely, standardized thread made standardized cloth made clothes in standardized sizes and shapes… So: what products are you thinking of that don’t benefit?
By “standardization” I meant being almost identical — as produced by one company. You are using the term more loosely. I don’t know what you mean by “standardized cars”. In any case, non-platform products include all sorts of food, all sorts of clothes, all sorts of cleaning materials, all sorts of decorative stuff, all sorts of art, all sorts of furniture. The innards of many products (“producer’s goods”) are standardized, of course, as I said — tires, rubber, and so on. If you simply mean that elements of standardization touch most products — sure, language will do that. One does not need large companies to produce that effect — to make loaves of bread vaguely the same size, and so on. Language is involved with the sale of most products, and it enforces a certain amount of uniformity — a “loaf” of bread is taken to be a certain size. Likewise, uniformity among producer’s goods (e.g., loaf pans) produces a certain uniformity among the final products. That will happen regardless of the size of bread companies. So long as bread companies don’t make loaf pans.
“Now that we have measured how much innovation…” — I see no measurement of innovation. I see a measurement of cost (“thousands of times cheaper”).
I find nothing about specialization improving/increasing innovation in Adam Smith. Perhaps that is true, but as far as I can tell Smith missed it completely. I have never heard any economist say that specialization improves/increases innovation.
Specialization of labor *is* itself an innovation; that was one of the core theses of WoN. (Though he didn’t use the exact word “innovation”. Rather he referred to “improvements”, including but not limited to “improvements in the productive power of labor“.) FInding a better production process that results in goods being thousands of times cheaper both *is* innovative and also *enables* additional innovation in all the downstream products.
On cars: From the Model T onward, standardization in how cars are constructed on a large scale enabled vast improvements in quality and maintainability compared to when car-building was a craft, and also enabled a huge aftermarket in things like better tires or speakers or replacement headlights. In short, cars became a platform.
Even art benefits from standardization of canvases and paint; furniture from standard fittings and cuts of wood and so on. (Human decorative preferences don’t actually strike me as particularly diverse. There are only so many common themes/styles. If they were truly diverse, Restoration Hardware and Ikea wouldn’t have been so successful)
The question that interests me is whether Smith said anything about what makes innovation more or less likely. As far as I know he did not.