Productivity paradox (aka Solow's paradox)

Computers, Internet, mobile technology, social media, and other digital technologies are so ubiquitous that we cannot imagine our lines without them. Many have even referred to the digital revolution as the third Industrial Revolution. It would seem obvious that all this digital technology has transformed our lives and has led to higher productivity and GDP growth. Yet, the data indicates that productivity growth in the U.S. has not been substantially higher during this digital age. Even as digitization took off in the U.S. starting in the 1980s, productivity growth in the decades since has been lower than productivity growth during the period of 1947-1973. This is known as the productivity paradox or Solow's paradox after Robert Solow remarked back in 1987: “you can see the computer age everywhere except the productivity statistics.” How is this possible? Could it be that this digital age is not as transformative as we think?

I think there are four possible explanations: 

I.  1947-1973 was the anomaly: The period following WWII was unique in American history for several reasons. First, the US economy had suffered through the Great Depression and anemic economic growth for a decade and was likely ripe for a productivity rebound. Second, numerous innovations developed for war efforts later were commercialized and applied in the economy. The unique nature of this period set the stage for enhanced productivity growth for a few decades and makes it incomparable to other periods.

II.    Solow was premature; productivity growth was high during 1995-2005; future gains will take time: The computing hardware advances that began in the 1950s and 1960s and the software advances that began in the 1970s led to the development of database management systems and enterprise software packages. There is some evidence that this wave of digitization caused the significant productivity boost that the US economy witnessed from 1995-2005 (see exhibit).

Just as it took decades for the productivity gains from computing advances to manifest, it will take time for the Internet, mobile technology, big data analytics, and other digital technology to affect the economy. Technology takes time to permeate throughout the economy and society as it becomes more affordable and as use cases are developed. It also takes time for firms to employ the technologies in ways that enhance productivity. This argument is supported by evidence from past GPTs. The mechanical revolution of the First Industrial revolution started in the mid-1700s and productivity gains took several decades to take hold. A similar story can be seen following electrification in the Second Industrial evolution.

III.  The paradox can be explained by the services sector: Over the last few decades, the U.S. economy has been transitioning from a mostly manufacturing-based economy to one dominated by the services sector. This transition could be dampening productivity growth rates (independent of digitization) for two reasons. The first is a composition effect: services generally experience slower growing productivity gains then manufacturing. Education and healthcare are the prime examples of this phenomenon (see Baumoil cost disease) and hence as these sectors comprise a larger share of the economy, productivity gains will mechanically also slow. Second, GDP (and hence productivity) of the services economy are notoriously more difficult to measure compared to the manufacturing sectors. Again, with the rising share of the service economy, the mis-measurement of GDP and productivity could be worsening.

IV.  Digitization exposes the gap between GDP and societal welfare: Imagine a world where robots are not scarce and can provide all the products and services one would desire or need. GDP may in fact be low or even zero but human welfare could be quite high. We are nowhere near that utopian fantasy but even if we have traversed a very small part of the road to that world, one can easily imagine that GDP and social welfare have begun diverging recently and that divergence is leading to a misleading interpretation of low productivity gains. Social media hints at this divergence: access to social media (e.g. Facebook, Twitter, Snapchat) is very cheap or even free and likely below the average consumer’s willingness to pay. Hence, social welfare has risen more than GDP would predict for social media. There are other similar examples in news (e.g. online news, blogs), communication (e.g. email, messaging, peer-to-peer) , entertainment (e.g. fantasy sports), and education (e.g. MOOCs) where the price of the product/service is significantly less than consumers’ willingness to pay. 

Why is Paris 4-5 times as dense as DC?

Cities are awesome!  Some people have an immediate negative view of big cities - congestion, pollution, traffic, expensive, etc. But those features do not necessarily follow from cities. Rather, they are a result of poor urban policy. What does seem to follow from cities is high productivity and economic growth. While causal evidence is hard to prove, there is much suggestive evidence that urbanization drives a host of positive economic factors of which productivity and economic growth are two of the more important ones. See here and here.

Density defines a city. And when it comes to density, cities in the United States are leaving significant economic growth on the table because of our low-density cities.

Paris is five times as dense as Washington, DC despite the fact that both cities are of comparable land size, have significant height restrictions, and restrict large pieces of land for national park and building / monument space.

My wife and I visited Paris recently and since we live in DC, noticed two key things that could explain the significant density difference.

1) American cities quickly becomes sparse as you move from the inner core because of the housing types. Paris has miles and miles of 6-7 story apartment buildings. DC has a few taller apartment buildings in the core (Dupont, Logan Circle, West End, etc) but very quickly, you find brownstones and single family homes.

2) Americans occupy more square feet per person than the French. The hotel rooms, the apartments, and the homes are all bigger in Washington, DC.

So there is no doubt that supply side constraints (e.g. height restrictions, zoning laws) are hindering density and economic growth in the U.S. 

And while those supply side policies should be changed, it is unlikely that U.S. cities will still be as dense as those in other developed countries.

Technology does not replace workers overnight

The rise of the robots is captivating economists, journalists, scientists and others as computing power, algorithms, data, and sensors are coming together to revolutionize what machines can do. And while machines will definitely replace workers in certain professions, they wont do it overnight. Instead, I think there are three reasons that the substitution will happen quite gradually. Lets look at ATMs as an illustrative example.

Here is an analysis by James Bessen that shows the number of bank tellers over time. Bank tellers rose quickly in the 60s and 70s and hit about 500,000 bank tellers in 1980. There are about 515,000 bank tellers today as can be seen in data from the BLS. Between 1980 and today the number of bank tellers has essentially been flat.

Bessen's numbers above clearly indicate that bank tellers did not disappear overnight. Here are three possible and not mutually-exclusive explanations:

1) During the upswing part of business cycles, the vulnerable job will likely decrease slightly or remain close to flat and and wages could be stagnant. During the good times, firms prefer not to layoff people but rather work them out through attrition. Then, in a downturn, those jobs are cut and may not come back. You can see this in the first link above. As ATM exploded, you see the number of tellers basically flat through the 90s. Then, the recession in 2000 allowed a bunch of tellers to be cut. Then, again, a slight to gradual increase in tellers after that recession until 2007/2008. Then, the number of tellers has been on a downward trajectory since. In fact, there has been a drop of almost 90,000 (~15%) bank tellers from 2007 to 2014 (605K to 515K).

2) Another mitigating factor is that a new technology likely increases demand for the product/service (either because of lower prices or more convenience) and this causes the firm to expand that offering. When they expand that offering, depending on the product/service, they may need to hire a few humans and so this can attenuate the initial substitution effect of the technology

3) A third explanation is that the occupational demands of a bank teller likely have changed. Tellers have progressed from only depositing and withdrawing cash to also performing more advanced transactions, trouble shooting, and maybe even some sales. Hence, even as occupations replace workers, many workers in that occupation can and will improve their skills to remain viable. And in fact, other adjacent occupations could be negatively effected. For example, as bank tellers upskill to out-compete ATMs, they may replace some sales workers who are more expensive.

Batman? Echolocation in humans

An incredibly interesting podcast recently from This American Life introduces the concept of how echolocation can help blind people "see." Further, it touches on the power of high expectations and the age-old trade-off between protection / complacency versus pushing boundaries. 

Work versus leisure

Back in 1930, J.M. Keynes predicted that within a hundred years, living standards in "progressive countries" would be 4-8 times higher and this would leave people more time to enjoy the good things in life. He further predicted that the working week would be drastically cut to 15 hours a week, with people choosing to have far more leisure as their materials needs were satisfied.

The first part of Keynes prediction was essentially correct. The living standard in Europe and the US is 8x times higher than it was in 1930 and we have 15 more years for living standards to get even higher by 2030.

However, the second part of his statement has not been realized. Per person "working hours" have come down a touch in the last 50 years but household "working hours" have gone up due to women's entry into the labor force. But regardless of how we count it, no society is close to the 15 hour work weeks that Keynes predicted.

Many people in the developed world could work a few hours a week, spend time with our loved ones, and pursue our hobbies. I would guess that many people would find this more attractive than our current equilibrium. One of the standard regrets from old age is having worked too much during their life and having spent less time with friends, family, and hobbies. So most of us are working too hard and will likely regret it, anywhere from a little to a lot, at the end of our lives. Yet, we continue to do so. Why?

One reason is that Keynes and economists may have underestimated how much utility humans get from working. For many it is part of their identify and defines a significant purpose in their lives. This leads to a question of the nature of the standard upward sloping labor supply curve but that is a conversation for another day.

Instead, I would like to focus on a second reason. I posit that having more leisure is likely most beneficial if our friends, family, neighbors, and communities also are taking leisure. So there is a network effect here. However, 15 hour work weeks are an unstable equilibrium because humans are fundamentally competitive. Imagine we were in a world where we all only worked 15 hours a week and spend our weeks blissfully as described earlier. Then one of our neighbors or siblings, or friends starts working 20 hours a week and buys a slightly bigger house and car and clothes and receives a promotion at work. My prediction: the 15 hour work-week would quickly unravel to the equilibrium we have today. Not everyone would have to be competitive for the 15 hour work-week to unravel. Even if only a small number of competitive people start working more, the reduced network effect can reduce the incentive for the others to maintain their leisure. Hence, income and its cousin consumption are positional goods because humans are fundamentally competitive and so the 15 hour work-week would not be possible even if it led to a higher utility.

One way to address the phenomenon of positional goods is to tax or cap consumption of that good. In a sense, this is what Europe has done. With their high income taxes and caps on weekly hours worked (though these caps can probably be skirted, i think they seep into the culture over time), Europe is trying to slowly move towards a Keynesian world. It is difficult to know what level of work maximizes utility but it is not necessarily true that societal welfare is higher in the US despite our higher incomes compared to Germany and France for example. Which equilibrium would you prefer? One where everyone works 30 hours per week and earns 60 cents on the dollar? Or one where everyone works 50 hours a week and earns a full dollar? 

What is capital?

There is a debate on the economics blogosphere on whether human capital is really capital? 

Capital is one of the most ubiquitous terms in our modern economy. But what does it mean? My take is that capital should be defined very broadly; however, in doing so, it is crucial to be aware of the important differences among the various forms of capital. In that sense, I agree with Noah Smith's view that depending on the context, we can lump different forms of capital together or keep them separate. 

I would define capital broadly as: if by investing in something today, it rewards you with higher consumption / utility in the future, that something can be called capital. Capital captures the important trade-off between consumption today and potentially higher consumption in the future. For example, many goods /services we purchase (e.g. food, TV, clothes, entertainment) are not capital. In a capitalistic system, private property rights protect individuals' ownership of capital.

By this definition, there are many forms of capital: financial capital, physical capital, land capital. human capital, social capital, knowledge capital.  

However, there some very key differences among these different forms of capital. 

1) Both social and knowledge capital have positive externalities and are non-rivalous
2) Physical and land capital depreciate the more you use them; human and social capital appreciate the more you use them
3) Human and social capital cannot be bought and sold
4) To reap the returns from human capital requires effort, which is a dis-utility

This fourth point is important and has interesting and significant implications, which Branko Milanovic discusses in more detail. 

"This distinction has some direct implications regarding individual welfare. $10,000 of income from human capital means I have to work x hours of work; $10,000 of income from my financial [or physical or land] capital, imply no work. Since work, in turn, implies effort and creates disutility (for otherwise, how would we explain upward sloping supply curves of labor?), it is obvious, I think, that in utility terms, $10,000 from human capital will bring fewer utils than the $10,000 collected from owning other forms of capital." 

So I dont have much of an issue with calling labor as "human capital" but ignoring the differences across types of capital can be misleading or wrong depending on the context. 

Selma

Robert Lucas once said about the topic of economic growth: "The consequences of human welfare involved in questions like these are simply staggering; once one starts to think about them, it is hard to think of anything else."

I like this quote; it captures one of the reasons I went into economics. 

Interestingly, I think there is a parallel with regards to race in America. When one starts reading, learning, and thinking about the role of race in American history, it is hard to think of anything else as it pertains to this great country. 

My wife and I saw the movie Selma this Martin Luther King, Jr weekend. Here is a beautiful song from the movie: Glory

When you play not to lose, you lose (NFC Championship game edition)

Incredibly poor mindset by the Packers who, led by their coach, played not to lose instead of playing to win:  

1) 4th and goal from 1-2 feet in the first quarter. I can somewhat understand kicking the FG at 4th and goal fron the 2 on the next series, but from 1-2 feet, are you kidding me? You have to be aggressive to win the NFC championship game against the best team on the road, with the best D. When you have an opportunity like that, you have to go take it...

2) In the 3rd quarter, the Seahawks faced a 2nd and 31. They had a ten-yard run, but still faced a 3rd and 21. At this point in the game the Seahawks had been absolutely horrendous on offense and the score was 16-0 in favor of the Packers. The Packers D had been dominant. So what did the Packers do? They rushed 2, sat back and Wilson had about 7 seconds to complete a 30-yard pass that led to their first points a few plays later... 

3) The Packers completely avoided Richard Sherman, the all-pro CB for the 'Hawks in the first game of the season. Then on the 1st drive of the NFC Championship game, he intercepted Aaron Rodgers in the end zone. He is good. I get that. But then, he hyper-extended his elbow early in the 4th quarter. Essentially he played the 4th quarter with one arm. And not once did the Packers run a play in his direction in the 4th quarter to see if he could defend a pass or make a tackle...

4) With 5:13 left in the 4th quarter, R. Wilson threw his 4th interception. M. Burnett intercepted that pass with plenty of room to run, but instead went down. What??? That was unbelievable. He could have run the ball easily into Seahawks territory. Click here to see how much room he had to run...

5) After that interception, here are the next three plays by the Packers: 
(i) Run for loss of -4 yards
(ii) Run for loss of -2 yards
(iii) Run for 2 yards
Where kind of play calling is that to win the game? If you want to take time off, try a screen pass, a bootleg swing pass, something. This play-calling was beyond conservative and led to... 

6) When the 'Hawks got the ball next, they were still down 19-7. They had been dominated by the Packers D for 56+ minutes. So what the Packers do? They offered the 'Hawks a nice, cushy prevent D from the Packers. Not the D that had shut them down the entire game. The 'Hawks promptly went down the field and scored a TD in just over a minute... 

Look, if the Packers had stopped the 'Hawks on either the unbelievable 2-point conversion or the onside kick, the would have won the game. The players have to execute. I understand that. 

But coaches have the responsibility to put their players in positions to execute and win. And looking at the evidence, the person with most culpability for the Packers collapse yesterday is Mike McCarthy. Answer this question: would Belichick and the Pats done any of the 6 situations above? No. As much as I dislike the Patriots, they are champions because they go out and win games. They dont play to lose. Playing not to lose is an excellent recipe to lose...   

Bill Barnwell of Grantland writes, "...Mike McCarthy is one of the worst in-game managers in the league..." After what I saw yesterday, I can't argue with that assessment.    

Technology and the future of work

One of the biggest challenges that society will face going forward is how to handle the impact of technology on society. The challenge has a new twist but not because technology is replacing human labor. That has been happening for centuries. Think luddites of yesteryear. But the luddites have been wrong over the last few centuries as economies have been able to weather technological change via a mix of economic growth, expanding education, labor-based institutions, and re-distributive policies.

The uniqueness of today's situation is rather the rate at which labor-saving technological progress is potentially occurring. If an industry replaces human capital with technology over the course of 100 years (as was the case with agriculture in the US), that provides society with plenty of time to soften the adjustment and re-deploy human capital. When it happens over 10 years, that is a significant challenge. 

Note: Globalization has exacerbated the effects of technological change but by itself is not as worrying because economic theory tells us that the effects of globalization cannot go on forever, whereas technological disruption can and likely will.  

Re-distributive policies must remain a component of the solution going forward (to preserve the pareto efficient nature of technological change). But the key trick will be to identify clever solutions beyond re-distribution. One framework is to think about ways of increasing the number of people who are owners of capital ("owners of robots"). This allows people to diversity their income so that even if robots replace their labor income, they can maintain their capital income. And if robots do not replace their jobs, they maintain their labor income.  The problem is that capital ownership is risky and poorer people often do not have the means to manage risk. So how to circle that square?

One thing I have been thinking about borrows from finance theory. Even if capital ownership is risky, if it is negatively correlated with labor risk, the overall portfolio of income becomes less risky. More thoughts to come on that idea. 

Another idea, put forth by Dani Rodrik here, is that the state can play a role in managing some of the risk. 

Pick and choose your Federalism

One of the early discussions at the founding of our nation was the topic of Federalism: how much power should be given to the central government and how much should remain with the states? This is a very interesting, important, and fascinating philosophical, legal, political, and social question and highlights the intelligence and sophistication of our founders.

The pro-Federalist view was captured in the writings of Alexander Hamilton, James Madison, John Jay and others in the Federalist Papers as they put forth the intellectual arguments for a Constitution that would provide powers for a strong federal government. These debates led to the Constitution, the Bill of Rights, and other important documents.

Crudely, conservatives today are more aligned with the anti-Federalist view and progressives are more aligned with the Federalist view. I say this alignment is crude for two reasons. 

First, it is arguable that the federal government today is more powerful than even the pro-Federalist founders envisioned. So Hamilton, Madison, and Jay may have been Federalists in the 18th century when the United States was a loose collection of states but it is possible that they would view the power of the federal government today with weariness. Many legal scholars debate whether the Federalist papers apply only to the specific context of the founding of our country or represent a broader philosophical position. 

Second, the stance of conservatives today, while ostensibly anti-Federalism at first glance, reveals itself as a 'pick and choose' your Federalism on deeper examination. On many issues, conservatives argue that the Federal government is getting too powerful and that responsibilities should be pulled back to states. However, when it comes to a few significant issues (gay marriage, legalization of marijuana), conservatives have tried to use the courts and the power of the federal government to over-ride decisions made by states. Many conservatives have tried / wanted to pass a constitutional amendment against gay marriage. The House of Representatives is trying to over-ride DC's vote to legalize marijuana.  And attorney generals from Oklahoma and Nebraska are suing Colorado for legalizing marijuana. For conservatives, federalism often seems to be an opportunistic cover to oppose any policies they dislike.