The Solow paradox, public goods, and the replicator economy.
by John MacBeath Watkins
Robert Solow, a Nobel-prize-winning economist, remarked way back in 1987 that "what everyone feels to have been a technological revolution...has been accompanied everywhere...by a slowdown in productivity growth.”
This has become known as the Solow paradox.
The golden age of productivity growth in the U.S. was between 1939 and 2000, with a slowdown in the 1980s, an increase in the Clinton Administration, and a slowdown again since.
What happened in 1939? Well, we began preparing for war. We didn't just build tanks, guns, ships, and aircraft, we also built roads and airports, and we dredge harbors and improved port facilities. Prior to World War II, flying boats were popular for serving areas that didn't have airports. After the war, there were plenty of airports.
The infrastructure binge continued after the war, and Dwight Eisenhower thought his greatest accomplishment was the Interstate Highway Act, which knit the country together with ribbons of road. Eisenhower understood logistics. He also understood that training was important if you wished to mobilize a large enterprise, and he elevated education to a cabinet-level office.
The federal investment in roads and education set loose the potential of the people and the land. And what have we done with this legacy of supply-side investment in public goods?
We've disinvested. Our public goods are getting old, and we've pushed onto students the cost of financing their education, so that someone can come out of college very easily in $100,000 debt. Higher education keeps getting cut while more is spent on other things, like prisons and welfare. Yet providing better education is one way we should be able to spend less on prisons and welfare.
Our bridges are getting old, some of our roads are getting rough.
But why didn't our technology give us the added productivity our disinvestment in public goods was taking away?
Maybe it did. Or maybe, sometimes technology is not necessarily useful for increasing measured productivity.
You measure productivity by seeing how many widgets are produced over a period of time by a given number of people. For example, in the cottage industry of music that existed before recorded music came along, you had to either make your own or hire a musician to make the music for you. Every song required a person making music to happen.
When recorded music cam along, you no longer had to have a musician present to have a song. This meant fewer people would be employed as musicians, but also that people at the top of the profession could provide music for a larger number of people. A musician could sing a song once, and millions of people could buy that song and play it repeatedly. There was more music in our lives, it was made by the best musicians, and the cost was lower. Productivity increased.
But we don't know how much, because we weren't calculating the productivity of musicians. A few musicians at the top were more productive, but once a record had been sold, it could be played many times. Those repeat performances were taken out of the economic sphere, and not counted as performances in any accounting sense. The metric became the sale of the record, rather than the performance of the song.
But what happened with the digital revolution in music? Well, this:
http://www.theatlantic.com/business/archive/2013/02/think-artists-dont-make-anything-off-music-sales-these-graphs-prove-you-wrong/273571/
Unless there was a dramatic decrease in the number of musicians, this represents a huge decrease in productivity. Far fewer songs are being sold, and if the number of musicians remains constant, their productivity, measured by the usual economic methods, has decreased dramatically.
But we know that this has not been accompanied by an increase in the cost of a song. What has happened instead is that much of the music produced has been taken out of the economic sphere altogether. People are pirating the songs, and getting music for free. There is a cost to this; it's not really as easy to steal a song as to buy it, but those who wish to sell a song are competing with the free copy that can be pirated by acquiring some skill and jettisoning some scruples.
In the realm of classified ads, most of those are free on Craigslist. Until recently, most newspapers have made their digital product free. As a result, whole swaths of the economy have come out of the economic sphere. When you produce something for a lower price, you increase productivity. When you produce it for free, in economic terms you aren't producing anything.
Thus, we have a different paradox, that of the replicator economy. On Star Trek, replicators can make anything you want for free. But if everything you need is free, how does anyone get paid? Musicians are already facing the replicator economy. Writers may face it soon.
This shows that not all technology produces increases in economic productivity, because some of it takes things out of the economic sphere.
In addition, highly-skilled artisans who were more productive than the average person found it impossible to keep making money at their craft. Take the example of the weavers and what William Black called the "dark, satanic mills" that replaced them.
They increased the number of yards of fabric per worker, and reduced the level of skill required by the worker. Weavers, who had made a good living because they were more productive than average, were put out of work. Some became Luddites, smashing the machinery that was eclipsing their way of life, but in the end, they lost.
They were replaced by low-skilled, low-paid workers, including in many cases children. The price of fabric went down, but the way of life of the people working to make the fabric became worse. And while productivity was increased in the making of fabric, the skilled artisans found their skill no longer required.
A skilled artisan who ends up working as a laborer or a waiter is going to become less productive. And every disruptive technology must have the effect of obsoleting some skills. It takes time for people to adjust, and some never will. Society as a whole may benefit, but in the disrupted industry, there is some immiseration, and among the displaced workers, there will be a decline in productivity. In fact, the immiseration of the obsolete workers removes the incentive for other industries to become more productive, because it drives down the price of labor.
So, what does increase productivity?
Full employment. I know, I know, productivity actually climbs in a recession because you lay off your least productive workers, but in the long run, only a shortage of workers convinces companies to make capital investments to reduce the number of workers needed. If you have to bid up the price of workers to attract employees, it makes sense to increase productivity.
Right now, we have the spectacle of cash-rich companies buying back their own stock, which is great for managers who have stock options, but not great for productivity.
Disinvestment in infrastructure has been bad for productivity, and we could kill two birds with one stone by catching up on that, which would increase employment, and build improvements that would unleash some productivity. Investment in public capital goods could increase employment enough to stimulate investment in private capital goods.
But what are the chances of that? We have an entire political party dedicated to the proposition that government spending can't produce jobs.Until we get better lawmakers, we won't have better policy.
Robert Solow, a Nobel-prize-winning economist, remarked way back in 1987 that "what everyone feels to have been a technological revolution...has been accompanied everywhere...by a slowdown in productivity growth.”
This has become known as the Solow paradox.
The golden age of productivity growth in the U.S. was between 1939 and 2000, with a slowdown in the 1980s, an increase in the Clinton Administration, and a slowdown again since.
What happened in 1939? Well, we began preparing for war. We didn't just build tanks, guns, ships, and aircraft, we also built roads and airports, and we dredge harbors and improved port facilities. Prior to World War II, flying boats were popular for serving areas that didn't have airports. After the war, there were plenty of airports.
The infrastructure binge continued after the war, and Dwight Eisenhower thought his greatest accomplishment was the Interstate Highway Act, which knit the country together with ribbons of road. Eisenhower understood logistics. He also understood that training was important if you wished to mobilize a large enterprise, and he elevated education to a cabinet-level office.
The federal investment in roads and education set loose the potential of the people and the land. And what have we done with this legacy of supply-side investment in public goods?
We've disinvested. Our public goods are getting old, and we've pushed onto students the cost of financing their education, so that someone can come out of college very easily in $100,000 debt. Higher education keeps getting cut while more is spent on other things, like prisons and welfare. Yet providing better education is one way we should be able to spend less on prisons and welfare.
Our bridges are getting old, some of our roads are getting rough.
But why didn't our technology give us the added productivity our disinvestment in public goods was taking away?
Maybe it did. Or maybe, sometimes technology is not necessarily useful for increasing measured productivity.
You measure productivity by seeing how many widgets are produced over a period of time by a given number of people. For example, in the cottage industry of music that existed before recorded music came along, you had to either make your own or hire a musician to make the music for you. Every song required a person making music to happen.
When recorded music cam along, you no longer had to have a musician present to have a song. This meant fewer people would be employed as musicians, but also that people at the top of the profession could provide music for a larger number of people. A musician could sing a song once, and millions of people could buy that song and play it repeatedly. There was more music in our lives, it was made by the best musicians, and the cost was lower. Productivity increased.
But we don't know how much, because we weren't calculating the productivity of musicians. A few musicians at the top were more productive, but once a record had been sold, it could be played many times. Those repeat performances were taken out of the economic sphere, and not counted as performances in any accounting sense. The metric became the sale of the record, rather than the performance of the song.
But what happened with the digital revolution in music? Well, this:
Unless there was a dramatic decrease in the number of musicians, this represents a huge decrease in productivity. Far fewer songs are being sold, and if the number of musicians remains constant, their productivity, measured by the usual economic methods, has decreased dramatically.
But we know that this has not been accompanied by an increase in the cost of a song. What has happened instead is that much of the music produced has been taken out of the economic sphere altogether. People are pirating the songs, and getting music for free. There is a cost to this; it's not really as easy to steal a song as to buy it, but those who wish to sell a song are competing with the free copy that can be pirated by acquiring some skill and jettisoning some scruples.
In the realm of classified ads, most of those are free on Craigslist. Until recently, most newspapers have made their digital product free. As a result, whole swaths of the economy have come out of the economic sphere. When you produce something for a lower price, you increase productivity. When you produce it for free, in economic terms you aren't producing anything.
Thus, we have a different paradox, that of the replicator economy. On Star Trek, replicators can make anything you want for free. But if everything you need is free, how does anyone get paid? Musicians are already facing the replicator economy. Writers may face it soon.
This shows that not all technology produces increases in economic productivity, because some of it takes things out of the economic sphere.
In addition, highly-skilled artisans who were more productive than the average person found it impossible to keep making money at their craft. Take the example of the weavers and what William Black called the "dark, satanic mills" that replaced them.
They increased the number of yards of fabric per worker, and reduced the level of skill required by the worker. Weavers, who had made a good living because they were more productive than average, were put out of work. Some became Luddites, smashing the machinery that was eclipsing their way of life, but in the end, they lost.
They were replaced by low-skilled, low-paid workers, including in many cases children. The price of fabric went down, but the way of life of the people working to make the fabric became worse. And while productivity was increased in the making of fabric, the skilled artisans found their skill no longer required.
A skilled artisan who ends up working as a laborer or a waiter is going to become less productive. And every disruptive technology must have the effect of obsoleting some skills. It takes time for people to adjust, and some never will. Society as a whole may benefit, but in the disrupted industry, there is some immiseration, and among the displaced workers, there will be a decline in productivity. In fact, the immiseration of the obsolete workers removes the incentive for other industries to become more productive, because it drives down the price of labor.
So, what does increase productivity?
Full employment. I know, I know, productivity actually climbs in a recession because you lay off your least productive workers, but in the long run, only a shortage of workers convinces companies to make capital investments to reduce the number of workers needed. If you have to bid up the price of workers to attract employees, it makes sense to increase productivity.
Right now, we have the spectacle of cash-rich companies buying back their own stock, which is great for managers who have stock options, but not great for productivity.
Disinvestment in infrastructure has been bad for productivity, and we could kill two birds with one stone by catching up on that, which would increase employment, and build improvements that would unleash some productivity. Investment in public capital goods could increase employment enough to stimulate investment in private capital goods.
But what are the chances of that? We have an entire political party dedicated to the proposition that government spending can't produce jobs.Until we get better lawmakers, we won't have better policy.
Comments
Post a Comment