Path-Dependent Economic Growth, Demand Side
I have been thinking a bit about economic growth. This is a very rough outline of my thoughts so far. I hope to refine my thoughts, but who knows, I'm quite possibly on a wrong path.
The basic idea is that demand can play an underappreciated role in economic development. I am not talking about a Keynesian shortfall in aggregate demand. That's important, but it's not my current preoccupation. What I am more concerned with is the way in which the value of economic activity is context-dependent. To put it very briefly, it appears that there are two distinct ways to grow an economy: you can improve physical production (for instance, by designing a soda can that uses 20% less aluminum without any sacrifice in quality) or you can improve the dollar value of existing production (for instance, by running an ad campaign to convince people that your soda is delicious). In the latter case, there are no immediate changes that are visible "from 30,000 feet." It's just that where people were willing to spend $x per can of soda, now they are willing to spend $y > $x, and so the economic activity of producing soda is now more "productive" than it was previously. (I am not using the concept of productivity in its technical sense.)
Now it may sound as though I am skeptical about this second source of economic growth. But I'm not. It might be relatively unimportant in the very early stages of economic development, when demand is overwhelmingly for basic goods like food and shelter. But I don't see how a modern economy could grow without expanding its "sphere of demand" to encompass more and more products with highly contingent value.
I'll give a few examples to show what I mean. First, there are the classic examples of technologies that are worth nothing outside of a specific context. The first fax machine was entirely useless until the second fax machine came online, and then its value grew as more and more machines came into use. This isn't a new observation, of course. Everyone knows that the value of a fax machine is contingent on a specific background of technology and network effects.
But take something like craft beer. As far as I know, the technology for beer production hasn't dramatically improved in the last fifty years. (We know this on an intuitive level because the Belgians have been making excellent beer for all of that time.) Now it's true that hard-to-detect changes in things like transportation costs can have a big effect, but I think it's safe to say that it would have been perfectly feasible to brew and sell something like Smuttynose's Old Brown Dog Ale 50 years ago.
But what actually happened was that beer production was hugely consolidated after Prohibition, and good domestic beer only gradually started emerging in the last quarter of the 20th century. And when it emerged, it emerged slowly. It sort of "ratcheted up," as beer drinkers were exposed to more and more variety and started to demand higher and higher quality. In other words, even when it was abundantly obvious that you could profitably make and sell high-quality beer in this country, it took decades for the market to mature into today's embarrassment of riches. It seems highly unlikely that the "bottleneck" was on the production end. Consumers had to be "educated" to appreciate craft beer, and they had to be "conditioned" to paying a lot more per ounce of beer. And to some extent culture had to evolve to confer status on people who understand and appreciate good beer. (Very fancy restaurants now have beer menus—I don't think that was always the case.)
The same thing goes for all kinds of products. Frozen orange juice was a very important development at one point, but its value was contingent on relatively cheap and widespread refrigeration. But then as refrigeration got even cheaper and the nation got richer, frozen orange juice lost a lot of market share to "fresh" orange juice. Its value was contingent on a context that prevailed for a few decades and then faded away. (Although I think frozen orange juice still enjoys considerable market share.) Frozen orange juice is a good example because its trajectory reflects both classic technological factors and "demand-side" factors like national income and taste preferences.
A more demand-oriented example is canola oil. Canada is capable of producing a lot of rape plants, but for one reason or another consumers were not enthusiastic about buying rape or rapeseed oil. A simple rebranding to "canola" ("Canada" + "oil") allowed the market to flourish. (Note that unprocessed rapeseed oil is poisonous and is used only for industrial applications, so please don't imagine that the product currently labeled "rapeseed oil" can be substituted for canola. Canola, though, is one of the healthiest cooking oils.) The same thing happened with Chinese gooseberries, which were rebranded as "kiwi fruit." Or the Patagonian toothfish as "Chilean seabass."
It's discomfiting to realize that an economy's basic health can be so dependent on marketing and consumer perceptions. It's also a potential avenue for recessions to have strangely long-lasting effects. A given supply/demand equilibrium might be a fragile thing, and a sudden shift in perceptions might be hard or impossible to reverse. (Just by way of example, what if consumers had shifted massively back into the frozen orange juice market to save money? Then orange juice production would be disturbed for a while, and the big question would be where consumers would end up when the dust settled. In the meantime a lot of "productive" economic activity would suddenly be unprofitable, which is to say unproductive in an economic sense.) I think this sheds at least a little light on the old conundrum about an economy producing at an "unsustainable" level. Krugman typically points out that the previous level of production was obviously achievable, since the economy reached it without inflation. But was the previous pattern of demand sustainable? Maybe not. During the bubble there was a lot of "demand" for housing that obviously couldn't last.
As I said, these are just my rough thoughts. I'll have a lot more to say as I work through these issues.
The basic idea is that demand can play an underappreciated role in economic development. I am not talking about a Keynesian shortfall in aggregate demand. That's important, but it's not my current preoccupation. What I am more concerned with is the way in which the value of economic activity is context-dependent. To put it very briefly, it appears that there are two distinct ways to grow an economy: you can improve physical production (for instance, by designing a soda can that uses 20% less aluminum without any sacrifice in quality) or you can improve the dollar value of existing production (for instance, by running an ad campaign to convince people that your soda is delicious). In the latter case, there are no immediate changes that are visible "from 30,000 feet." It's just that where people were willing to spend $x per can of soda, now they are willing to spend $y > $x, and so the economic activity of producing soda is now more "productive" than it was previously. (I am not using the concept of productivity in its technical sense.)
Now it may sound as though I am skeptical about this second source of economic growth. But I'm not. It might be relatively unimportant in the very early stages of economic development, when demand is overwhelmingly for basic goods like food and shelter. But I don't see how a modern economy could grow without expanding its "sphere of demand" to encompass more and more products with highly contingent value.
I'll give a few examples to show what I mean. First, there are the classic examples of technologies that are worth nothing outside of a specific context. The first fax machine was entirely useless until the second fax machine came online, and then its value grew as more and more machines came into use. This isn't a new observation, of course. Everyone knows that the value of a fax machine is contingent on a specific background of technology and network effects.
But take something like craft beer. As far as I know, the technology for beer production hasn't dramatically improved in the last fifty years. (We know this on an intuitive level because the Belgians have been making excellent beer for all of that time.) Now it's true that hard-to-detect changes in things like transportation costs can have a big effect, but I think it's safe to say that it would have been perfectly feasible to brew and sell something like Smuttynose's Old Brown Dog Ale 50 years ago.
But what actually happened was that beer production was hugely consolidated after Prohibition, and good domestic beer only gradually started emerging in the last quarter of the 20th century. And when it emerged, it emerged slowly. It sort of "ratcheted up," as beer drinkers were exposed to more and more variety and started to demand higher and higher quality. In other words, even when it was abundantly obvious that you could profitably make and sell high-quality beer in this country, it took decades for the market to mature into today's embarrassment of riches. It seems highly unlikely that the "bottleneck" was on the production end. Consumers had to be "educated" to appreciate craft beer, and they had to be "conditioned" to paying a lot more per ounce of beer. And to some extent culture had to evolve to confer status on people who understand and appreciate good beer. (Very fancy restaurants now have beer menus—I don't think that was always the case.)
The same thing goes for all kinds of products. Frozen orange juice was a very important development at one point, but its value was contingent on relatively cheap and widespread refrigeration. But then as refrigeration got even cheaper and the nation got richer, frozen orange juice lost a lot of market share to "fresh" orange juice. Its value was contingent on a context that prevailed for a few decades and then faded away. (Although I think frozen orange juice still enjoys considerable market share.) Frozen orange juice is a good example because its trajectory reflects both classic technological factors and "demand-side" factors like national income and taste preferences.
A more demand-oriented example is canola oil. Canada is capable of producing a lot of rape plants, but for one reason or another consumers were not enthusiastic about buying rape or rapeseed oil. A simple rebranding to "canola" ("Canada" + "oil") allowed the market to flourish. (Note that unprocessed rapeseed oil is poisonous and is used only for industrial applications, so please don't imagine that the product currently labeled "rapeseed oil" can be substituted for canola. Canola, though, is one of the healthiest cooking oils.) The same thing happened with Chinese gooseberries, which were rebranded as "kiwi fruit." Or the Patagonian toothfish as "Chilean seabass."
It's discomfiting to realize that an economy's basic health can be so dependent on marketing and consumer perceptions. It's also a potential avenue for recessions to have strangely long-lasting effects. A given supply/demand equilibrium might be a fragile thing, and a sudden shift in perceptions might be hard or impossible to reverse. (Just by way of example, what if consumers had shifted massively back into the frozen orange juice market to save money? Then orange juice production would be disturbed for a while, and the big question would be where consumers would end up when the dust settled. In the meantime a lot of "productive" economic activity would suddenly be unprofitable, which is to say unproductive in an economic sense.) I think this sheds at least a little light on the old conundrum about an economy producing at an "unsustainable" level. Krugman typically points out that the previous level of production was obviously achievable, since the economy reached it without inflation. But was the previous pattern of demand sustainable? Maybe not. During the bubble there was a lot of "demand" for housing that obviously couldn't last.
As I said, these are just my rough thoughts. I'll have a lot more to say as I work through these issues.
posted by James | 9:36 AM
0 Comments:
Post a Comment
<< Home