Wednesday, July 30, 2014

Deflation: The Case for Worrying Less

One of the arguments as to why monetary policy should continue to be loose, both for the Federal
Reserve in the United States and for other central banks around the world, is to avoid the risk of a bout of deflation. But is that fear overstated? The Bank of International Settlements offers a discussion "The costs of deflation: what does the historical record say?" in its 84th Annual Report published last month. (For those not familiar with BIS, it's a Swiss-based international organization that has been around since 1930, and serves as a main forum for consultation and cooperation between central banks.

When looking at historical episodes of deflation, the BIS offers a simple comparison. Look at the five years before and after an episode of deflation started for a range of countries, and see what happened to growth of real GDP during that time. As a starting point, consider pre-World War I episodes of deflation from 1860-1901 in ten countries:  Belgium, Canada, France, Germany, Italy, Japan, Netherlands,  Switzerland, United Kingdom, United States. Some of these countries had multiple episodes of deflation during this time: for example, the U.S. economy had episodes of deflation starting in 1866, 1881, and 1891. When you match up the five years before and after the starting point of an episode of deflation with the path of GDP growth, here's what you get. The red line shows the price level (which peaks at time zero, because that's how the figure is constructed), and blue line shows the path of GDP relative to that time zero. For this time period, the onset of deflation on average doesn't seem to have any effect on economic growth.

How about during the early interwar period, meaning the 1920s and early 1930s? Here, the path of GDP growth across the 10 countries in this sample is definitely faster before the start of price deflations rather than after--but it remains positive. BIS writes: "In the early interwar period (mainly in the 1920s), the number of somewhat more costly (“bad”) deflations increased: output still rose, but much more slowly – the average rates in the pre- and post-peak periods were 2.3% and 1.2%, respectively. (Perceptions of truly severe deflations during the interwar period are dominated by the exceptional experience of the Great Depression, when prices in the G10 economies fell cumulatively up to roughly 20% and output contracted by about 10%.)"


Finally, what about more recent deflations from 1990 to 2013? For this time period, the sample now includes episodes of deflation in 13 places: Australia, China;l the euro area, Hong Kong SAR, Japan, New Zealand, Norway, Singapore, South Africa, Sweden,  Switzerland, and the United States (in the third quarter of 2008). The recent episodes of deflation have been much shorter: thus, the red line showing the price level dips slightly, but then starts rising again after about a year. The path of GDP growth looks much the same in the five years before and after the deflation. As BIS writes: "The deflation episodes during the past two and a half decades have, on average, been much more akin to the
good types experienced during the pre-World War I period than to those of the early interwar period ..."

BIS suggests four lessons that can be taken from this exercise.

"First, the record is replete with examples of “good”, or at least “benign”, deflations in the sense that they coincided with output either rising along trend or undergoing only a modest and temporary setback. ...
"The second important feature of deflation dynamics revealed by the historical record is the general absence of an inherent deflation spiral risk – only the Great Depression episode featured a deflation spiral in the form of a strong and persistent decline in the price level; the other episodes did not. ...   The evidence, especially in recent decades, argues against the notion that deflations lead to vicious deflation spirals. In addition, the fact that wages are less flexible today than they were in the distant past reduces the likelihood of a self-reinforcing downward spiral of wages and prices. ..."
"Third, it is asset price deflations rather than general deflations that have consistently and significantly harmed macroeconomic performance. Indeed, both the Great Depression in the United States and the Japanese deflation of the 1990s were preceded by a major collapse in equity prices and, especially, property prices. These observations suggest that the chain of causality runs primarily from asset price deflation to real economic downturn, and then to deflation, rather than from general deflation to economic activity. ... 
"Fourth, recent deflation episodes have often gone hand in hand with rising asset prices, credit expansion and strong output performance. Examples include episodes in the 1990s and 2000s in countries as distinct as China and Norway. There is a risk that easy monetary policy in response to good deflations, aiming to bring inflation closer to target, could inadvertently accommodate the build-up of financial imbalances. Such resistance to “good” deflations can, over time, lead to “bad” deflations if the imbalances eventually unwind in a disruptive manner."
In short, the grim experience of the 1930s and its combination of deflation and Great Depression looks like a special case. The typical deflation is not accompanied by a steep recession. Nothing here argues that deflation is desirable or worth seeking to encourage. But the greater danger of economic recessio or Depression seems to arise not out of price deflation, but instead when asset prices bubbles overinflate, and then burst.

Tuesday, July 29, 2014

Universal Basic Income: A Thought Experiment

Pretty much all current public programs to assist households with low incomes suffer from the same seemingly unavoidable problem: As the household's income rises, the amount of assistance provided by the public program is phased out. For example, a person who earns an extra $100 might find that eligibility for public assistance has been reduced by $50. Economists call this reduction in benefits as income rises a "negative income tax," and it is not unusual for studies to find that certain working poor families face negative income tax rates in the range of 50% of the marginal dollar earned, 60%, or even higher (for examples, see here and here). These high negative income tax rates diminish work incentives for those with low incomes.

There's a way out called "universal basic income." Ed Dolan explores "The Pragmatic Case for a Universal Basic Income" in the Third Quarter 2014 issue of the Milken Institute Review (free registration may be required for access).

The idea of a universal basic income is that every U.S. citizen would be entitled to a chunk of money each year. This amount would not vary based on income level, or employment, or disability, or age, or any other reason.  Specifically, when a low-income person works and earns income, the universal basic income check would not be reduced in any way.  The "negative income tax" rate is zero percent.

The idea of a universal basic income raises obvious questions. How much would it be per person? How would it be financed? Are the politics of such a program conceivable? Let's tackle these in turn.

How much? Dolan suggests that we aim at having a universal basic income that reaches the poverty line. The current poverty line for a family of three is roughly $18,000. Thus, in round number terms, the goal might be to have a universal basic income of $6,000 per person, perhaps paid every other week. (One can easily imagine having a number that is a little higher for adults and a little lower for children, or keeping some of the money for children in trust and giving it to them over the first decade or so of adulthood, but let's not make this example too fancy.)

How would this amount be financed? The starting point is that this program would replace programs currently aimed at those with low incomes: for example, it would replace welfare, Food Stamps, the Earned Income Tax Credit and the Child Credit. For the sake of this thought experiment, Dolan suggests not including public policies that help low-income families with education or health (like Medicaid). The other programs for supporting low-income Americans spend roughly $500 billion.

But remember, the universal basic income goes to everyone, not just those with low incomes. Thus, Dolan suggests that a second source of revenue would be to eliminate a number of tax deductions and deferrals, including the mortgage-interest deduction, the deferral of income taxes on retirement, and the deduction of charitable contributions. He also proposed eliminating the personal exemption. (In keeping his argument separate from health care financing issues, Dolan does not propose touching the tax exemption for compensation paid in the form of employer-provided health insurance.) These changes would raise about $1.2 trillion. Most middle-income and upper-middle income families, who either do not itemize deductions at all or don't have that many deductions to itemize, would come out ahead with $6,000 per person rather than using these tax provisions, although those at the highest income levels who make substantial use of these provisions would end up paying more.

Finally, Dolan suggests that Social Security recipients can have a choice: they can either receive the Social Security payments they are already entitled to by law, or they could instead receive the universal basic income. Those receiving the lowest Social Security payments now would benefit from this choice, and the rest of the elderly would be unaffected by the plan.

Taking these steps together, Dolan estimates that the U.S. could fund a universal basic income of $5,800 per person. In addition, consider some of the side benefits. Work incentives for those with low incomes would be greatly improved. The need for government to set up and enforce complicated eligibility rules for those with low incomes would be eliminated. The tax code could be greatly simplified, and many more people could file very simple tax returns. At the most basic level, everyone in the United States would be guaranteed an amount close to the poverty level of income.

What about the politics of a universal basic income? It's no surprise that many who lean liberal like the idea of guaranteeing a basic income. However, the idea has a reasonable number of conservative and libertarian supporters, who like the idea of a program that addresses the basic concern over helping those with low incomes, but in a clean, clear way that involves much less interference of eligibility rules and phase-ins and phase-outs in people's lives. Dolan claims that there are lively debates over a universal basic income happening behind the scenes between those with very different political persuasions.

The idea of a universal basic income is appealing to me in theory, but I have a hard time believing that once enacted, the U.S. political process would be willing or able to leave it alone. One one side, those who favor higher tax rates for those with high incomes would immediate start trying to figure out ways to claw back payments to those with high incomes. On another side, there would be continual pressures to reinstate programs like Food Stamps, or targeted welfare payments for certain types of families, or favored tax provisions for home-buying or charitable contributions or retirement. There would be continual political pressure to alter the amount of a universal basic income, as well. The U.S. political system does not excel at replacing complexity with simplicity, and then leaving well enough alone.







Monday, July 28, 2014

The Decline of Milk

All those "Got Milk?" advertisements are trying to push back against the tide: U.S. milk consumption has been falling for several decades. Jeanine Bentley of the US Department of Agriculture lays out some "Trends in U.S. Per Capita Consumption of Dairy Products, 1970-2012" (June 2, 2014). Here's the decline in milk.


The decline seems to be part of a generational shift in what people choose to drink. Bentley explains:
"A 2013 ERS [Economic Research Service] study found that while Americans continue to drink about 8 ounces of fluid milk when they drink milk, they are consuming it less frequently than in the past. Americans are especially less apt to drink milk at lunchtime and with dinner. National food consumption surveys reveal that Americans born in the early 1960s drank milk 1.5 times a day as teenagers, 0.7 times a day as young adults, and 0.6 times a day in middle age. In contrast, Americans born in the early 1980s entered their teenage years drinking milk just 1.2 times a day and were drinking milk 0.5 times a day as young adults. Competition from other beverages—especially carbonated soft drinks, fruit juices, and bottled water—is likely contributing to the changes in frequency of fluid milk
consumption. In addition, substitutes for cow’s milk (including nut milks, coconut milk, and soy milk) have provided alternatives for consumers."
For the overall category of dairy products, the decline in milk consumption has been partially offset by a rise in cheese consumption: "Over the last four decades, Americans have increased their consumption of cheese, especially Italian varieties such as mozzarella, parmesan, and provolone. In 2012, cheese availability was 33.5 pounds per person, almost triple the amount in 1970 at 11.4 pounds. Availability of Italian cheeses increased to 14.9 pounds per person from 2.1 pounds in 1970. Since 2005, availability of American cheese has remained around 13 pounds per person. The inclusion of cheese in time-saving convenience foods and in commercially manufactured and prepared foods such as frozen pizza, macaroni and cheese, and pre-packaged cheese slices has increased consumption. The popularity of cheese-rich Italian and Tex-Mex cuisines has also contributed to increased cheese consumption."

Nonetheless, with the decline in milk consumption, dairy consumption has been falling over time. 



(In case you are wondering what the "availability" of milk means in the figure title, here is Bentley's explanation: "ERS’s [Economic Research Service's] food availability data calculate the annual supply of a commodity available for humans to eat by subtracting measurable nonfood use (farm inputs, exports, and ending stocks) from the sum of domestic supply (production, imports, and beginning stocks). Per capita estimates are determined by dividing the total annual supply of the commodity by the U.S. population for that year. Although these estimates do not directly measure actual quantities ingested, they serve as a proxy for Americans' food consumption over time.)

Friday, July 25, 2014

Beware Faraway Shareholder Meetings

An old trick in Washington policy circles is to obscure unwelcome news by releasing it late on Friday afternoon--and if it can be Friday afternoon of a three-day weekend, so much the better. Many companies seem to have another method of obscuring unwelcome news: they schedule their annual shareholder meeting to a more remote location. Yuanzhi Li and David Yermack present the evidence in "Evasive Shareholder Meetings." An overview of the paper is freely available in the NBER Digest for July 2014. The actual paper is National Bureau of Research Working Paper #19991 (March 2014), and while it is not freely available, many readers will have access through library subscriptions.

Li and Yermack look at data on 10,000 annual shareholder meetings held between 2006 and 2010. In one set of companies that in a certain year choose to hold their shareholder meeting at least 1,000 miles from the corporate headquarters. After such meetings, the company is more likely to announce unfavorable quarterly earnings, and also experiences an stock market return 3.7% less than a benchmark for the overall stock market over the following six months.

Similarly, when a company holds its shareholder meeting at least 50 miles from a major airport, its stock underperforms the market benchmarks in the following six months. When companies hold their meeting both more than 50 miles a major airport and more than 50 miles from corporate headquarters, the company's stock underperforms the market by 6.8% in the six months after the meeting.

Just to be clear, this effect is not rooted in effects from companies that are already known to be performing poorly, or already facing controversy, before the shareholder meeting occurs. Li and Yermack summarize (citations omitted):

We find little evidence that meetings are moved to distant locations when a firm has had  a bad year, or when public information suggests that firms should expect confrontation; in fact, analysis of the agendas for the meetings in our sample suggests that companies are more likely to meet near headquarters when they expect hostile shareholder proposals or board elections that may be subject to protest voting. This may occur because the company is more comfortable arranging security, working with law enforcement, and controlling access to the meeting site in its own jurisdiction. Companies may also be relatively unconcerned with the publicity value of controversial agenda items, since these are known by everyone in advance and often have easily predictable outcomes.
Instead, we find that managers schedule long-distance meetings when the firm is experiencing adverse operating performance that is not already known to the market. Company stocks perform very poorly in the aftermath of remote meetings, and part of this result stems from disappointing quarterly earnings announcements following these meetings. By moving the meeting far away, the managers might forestall shareholder or news media questioning that could lead to the early disclosure of adverse news. ...
Scheduling a meeting far from headquarters provides a straightforward opportunity for managers to discourage attendance. Research shows that firms tend to have high ownership in their local communities and that local analysts tend to forecast stock performance better than distant analysts. A long-distance shareholder meeting would inevitably reduce participation from both of these cohorts as well as the local business press, who may be the most knowledgeable people about the company. ...
The poor performance of companies following long-distance meetings suggests that management knows adverse news when choosing the location of these meetings, and it may move them far from headquarters as part of a scheme to suppress negative news for as long as possible. While this motivation seems understandable, it is less obvious why shareholders fail to decode such an unambiguous signal at the time the meeting location is announced.
Of course, now that this research has been published, investors are going to be on the lookout. Companies that schedule shareholder meetings far from corporate headquarters or far from a major airport should at a minimum expect to come under greater scrutiny--and perhaps even see an immediate fall in their stock price, on the presumption that a faraway shareholder meeting is a negative signal about future earnings.



Thursday, July 24, 2014

Long-Term Budget Deficits

Much of the sound and fury about U.S. budget deficits involves what should happen in the next year or so. Of course, short-run decisions about red ink do matter, and have a way of bleeding over into long-term decisions. But this focus on the short-term also risks missing the longer-term context of how U.S. government deficits and debt have changed since the Great Recession started in 2007, and where they are headed in the next couple of decades. For this purpose, my go-to starting point is "The 2014
Long-Term Budget Outlook" published  this month by the Congressional Budget Office.

Here's an overview of the basic CBO budget forecast, which shows some of the costs of the current path, but as I will explain, is probably to optimistic. This "extended baseline" forecast is based on existing law, including any ways in which the law requires future changes: for example, if current law requires a tax to be changed a few years in the future, that future change is included in this forecast. This figure shows government spending and revenues as a share of GDP. The gap between them doesn't look large, but remember that with the US GDP now in the range of $17 trillion, a gap of 1 percentage point is a deficit of $170 billion.


I'd draw two lessons from this figure. First, you can see the large jump in accumulated federal debt during the Great Recession, from less than 40% of GDP in 2008 to 74% of GDP in 2014. Federal debt relative to GDP is now at the second-highest level in US history, trailing only the explosion of debt used to finance the fighting of World War II. Second, federal debt in the longer term is projected to rise more slowly in the next few years, but to keep rising to 106% of GDP by 2039--which would be equal to the debt/GDP level in 1946.

Here's a figure showing the government debt/GDP ratio throughout U.S. history. You can see the previous debt/GDP peaks at the Revolutionary War, the Civil War, World War I, the Great Depression, World War II, and the 1980s and early 1990s. According to the CBO, the U.S. government is currently on autopilot to set an unwelcome new record.

The discussion of the effects of large budget deficits often seems to be an argument over the possibility of catastrophic debt overload and the risk of Greek-style or Argentinian-style debt defaults. But arguing over catastrophe misses the real and near-term costs of the higher budget deficits. The CBO lists three of them.

First, the current high level of government debt, and the projections for the next 25 years, mean that the U.S. government lacks fiscal flexibility. Before previous debt spikes, we had started with a relatively low debt/GDP ratio, and so we had room to borrow as needed. But with the debt/GDP ratio already at 74% and rising, we now lack that flexibility. I sometimes say that the U.S. could afford to fight the Great Recession with large budget deficits. But having done so, it wouldn't be nearly as easy to enact a similar deficit boost in the future if economic or foreign policy considerations might seem to warrant it.

Second, the current spending patterns of the U.S. government are starting to crowd out everything except health care, Social Security, and interest payments. The bottom three lines on this graph show rising government spending on major health care programs (like Medicare and Medicaid), on Social Security, and on interest payments on past borrowing. The top dark green line shows spending on everything else the government does, falling steadily as a share of GDP over time. Again, this is the projection based on current law.

Third, large government borrowing means less funding is available for private investment. CBO writes: "Large federal budget deficits over the long term would reduce investment, resulting in lower national income and higher interest rates than would otherwise occur. Increased government borrowing would cause a larger share of the savings potentially available for investment to be used for purchasing government securities, such as Treasury bonds. Those purchases would crowd out investment in capital goods—factories and computers, for example—which makes workers more productive. Because wages are determined mainly by workers’ productivity, the reduction in investment would reduce
wages as well, lessening people’s incentive to work."

These long-run dangers don't mean that an abrupt large reduction in budget deficits should happen immediately, when the economy is still struggling to generate a respectable recovery. But it does mean that we should be thinking seriously about small changes for the near future that will phase into larger effects over the next couple of decades.

I said earlier that this scenario is optimistic. I don't mean that the CBO has biased its baseline estimates: indeed, the report goes through in some detail the underlying projections behind these numbers about productivity and economic growth, health care costs and life expectancy, and interest rates (which affect the costs of financing government borrowing). But the baseline estimate, by law and custom, focuses on what is specifically in the law. This seemingly sensible rule offers a temptation to politicians, who can enact spending cuts or tax increases that aren't scheduled to start for five or 10 years. These proposals make the projected deficits look better over the next five or ten years, and then the policies can be changed later.

Thus, CBO calculates an "alternative fiscal scenario," in which it sets aside some of these spending and tax changes that are scheduled to take effect in five years or ten years or never. Remember that the extended baseline scenario projected that the debt/GDP ratio would be 106% by 2039. In the alternative fiscal scenario, the debt-GDP ratio is projected to reach 183% of GDP by 2039. As the report notes: "CBO’s extended alternative fiscal scenario is based on the assumptions that certain policies that are now in place but are scheduled to change under current law will be continued and that some provisions of law that might be difficult to sustain for a long period will be modified. The scenario, therefore, captures what some analysts might consider to be current policies, as opposed to
current laws."

What changes are assumed in the alternative fiscal scenario? As one example, the category of "Other Non-Interest Spending" in the chart above does not plummet: "Federal noninterest spending apart from that for Social Security, the major health care programs (net of offsetting receipts), and certain refundable tax credits would rise after 2024 to its average as a percentage of GDP during the past two decades—rather than fall significantly below that level, as it does in the extended baseline."

On the tax side, the usual political trick is to have various tax deductions or credits scheduled too expire in the future, which makes the projected deficit appear lower, except that when the time comes for expiration the tax provisions are renewed again. Thus, the baseline revenue estimates rise from 17.6% of GDP in 2014 to 18.3% of GDP by 2024 and 19.4% of GDP by 2034 (and keep rising after that). In the alternative fiscal scenario," tax revenue rises to 18.1% of GDP, which is "slightly higher than the average of 17.4 percent over the past 40 years," notes the CBO--but then tax revenues don't continue to rise above that level.

My own judgement is that the path of future budget deficits in the next decade or so is likely to lean toward the alternative fiscal scenario. But long before we reach a debt/GDP ratio of 183%, something is going to give. I don't know what will change. But as an old-school economist named Herb Stein used to say, "If something can't go on, it won't."


Wednesday, July 23, 2014

Unemployment and Labor Force Participation: Revisiting the Puzzle

The US unemployment rate has been painfully high in the Great Depression and its aftermath, but the high unemployment rates of the early 1980s look even worse--at least at first glance. However, the 1970s and 1980s were a time when a rising share of the adult population, and especially women, were entering the (paid) labor force, while the last few years are a time when the share of the adult population is in the labor force is declining .  Indeed, it's been a standard concern in the last few years that the official unemployment rate is not capturing the true pain in the labor market, because the official unemployment rate only includes those who are looking for work--not those who have become discouraged and given up looking. What's is the interaction between the unemployment rate and the labor force participation rate telling us?

Here'a figure showing the post-World War II unemployment rate. The monthly peak of the unemployment rate at 10% in October 2009 was second-highest of any post World War II recession, behind the peak of 10.8% in November and December 1982. In addition, from December 1979 to August 1987, a period of almost eight years, the monthly unemployment rate exceeded 6%. More recently, the unemployment rate first rose above 6% in August 2008, and as of June 2014 was at 6.1%. Thus, it's plausible that that the current stretch of monthly unemployment above 6% will last a little more than six years--which is dismal, but still with some way to go before matching the high unemployment rates that prevailed during most of the 1980s.



However, this comparison doesn't feel quite fair. After all, during the 1970s and 1980s, the labor force participation rate (that is, the share of adults who either had jobs or were unemployed and looking for jobs) was rising from 60% to about 67%. Since the start of the Great Recession, the labor force participation rate has fallen from 66% down to about 63%. An unemployment rate falling back below 6% was more comforting in the 1980s, with a rising share of adults working, than it would be in 2014, with a falling share of adults working. Here's a figure showing the labor force participation rate in the post World War II period.


To investigate these issues, the Council of Economic Advisers has just published "The Labor Force Participation Rate Since 2007: Causes and Policy Implications." The rise in the labor force participation rate from about 1960 up through the mid-1990s was driven both by the baby boom generation reaching working age, and the dramatic entry of women into the (paid) labor force. The decline since about 2000 is largely because the rising proportion of women in the labor force levelled off, and the aging of the baby boom generation is raising the number of retirees. Here's a figure from the report showing the labor force participation rate for men and women separately.

The CEA also notes that during every recession, the labor force participation rate tends to fall a little, as some people give up looking for jobs and thus aren't counted as officially "unemployed." After presenting its own analysis, and showing that it fits fairly well with previous studies of the subject, the CEA notes: "Up until the beginning of 2012 the [labor force] participation rate was generally slightly higher than would have been predicted based on the aging trend and the standard business cycle effects. But in the last two years, the participation rate has continued to fall at about the same rate even though the unemployment rate has been declining rapidly."

In other words, the drop in the labor force participation rate up through about 2012 was explainable based on the aging of the population and the common patterns during a recession. (Here's a post from April 2012 that refers to a study making this point.) In this telling, the real puzzle is not why the labor force participation rate fell up to 2012, but why it has continued to fall so quickly since 2012. To explain what is different about the period since 2012, compared with the period after previous recessions, the CEA report focuses most heavily on how long-term unemployment has been different in the aftermath of the Great Recession. For example, here's one figure showing the share of the total unemployed who have been out of work more than 27 weeks. In past recessions, this share peaked at about 20% of the total unemployed. In the Great Recession, the share of long-term unemployed peaked at above 40% of all unemployed, and even now remains at historically high levels. (Here's a post from July 2013 on the legacy of long-term unemployment.)




Another measure of long-term unemployment is the average duration of unemployment. Again, remember that only those who are actually looking for a job are counted as officially unemployed, not those who have become discouraged and stopped looking. In the last few recessions, the average length of unemployment peaked at around 20 weeks. In the Great Recession, it peaked at about 40 weeks, and is still at a discomfitingly high 35 weeks.


The CEA report looks at some other potential reasons for why labor force participation has dropped off, like a rising share of the 16-24 year-old age bracket being in school, and other breakdowns by age, gender and disability status. At least to me, the most important bottom line is not to focus on the somewhat sterile argument of who should "really" be counted as unemployed. After all, the Bureau of Labor Statistics does also collect statistics on "discouraged" workers and "marginally attached" workers. Instead, the key message is that the lower unemployment rate includes a much larger than usual share of long-term unemployed. Waiting for the U.S. economy to re-absorb these workers has not been showing good results. 


Tuesday, July 22, 2014

The Next Wave of Technology?

Many discussions of "technology" and how it will affect jobs and the economy have a tendency to discuss technology as if it is one-dimensional, which is of course an extreme oversimplification. Erik Brynjolfsson, Andrew McAfee, and Michael Spence offer some informed speculation on how they see the course of technology evolving in "New World Order: Labor, Capital, and Ideas in the Power Law Economy," which appears in the July/August 2014 issue of Foreign Affairs (available free, although you may need to register).

Up until now, they argue, the main force of information and communications technology has been to tie the global economy together, so that production could be moved to where it was most cost-effective. As they write: "Technology has sped globalization forward, dramatically lowering communication and transaction costs and moving the world much closer to a single, large global market for labor, capital, and other inputs to production. Even though labor is not fully mobile, the other factors increasingly are. As a result, the various components of global supply chains can move to labor’s location with little friction or cost." 
But looking ahead, they argue that the next wave of technology will not be about relocating production around the globe, but changing the nature of production--and in particular, automating more and more of it. If the previous wave of technology made workers in high-income countries like the U.S. feel that their jobs were being outsourced to China, the next wave is going to make those low-skill workers in repetitive jobs--whether in China or anywhere else--feel that their jobs are being outsources to robots. Brynjolfsson, McAfee, and Spence write:
Even as the globalization story continues, however, an even bigger one is starting to unfold: the story of automation, including artificial intelligence, robotics, 3-D printing, and so on. And this second story is surpassing the first, with some of its greatest effects destined to hit relatively unskilled workers in developing nations.
Visit a factory in China’s Guangdong Province, for example, and you will see thousands of young people working day in and day out on routine, repetitive tasks, such as connecting two parts of a keyboard. Such jobs are rarely, if ever, seen anymore in the United States or the rest of the rich world. But they may not exist for long in China and the rest of the developing world either, for they involve exactly the type of tasks that are easy for robots to do. As intelligent machines become cheaper and more capable, they will increasingly replace human labor, especially in relatively structured environments such as factories and especially for the most routine and repetitive tasks. To put it another way, offshoring is often only a way station on the road to automation.
This will happen even where labor costs are low. Indeed, Foxconn, the Chinese company that assembles iPhones and iPads, employs more than a million low-income workers -- but now, it is supplementing and replacing them with a growing army of robots. So after many manufacturing jobs moved from the United States to China, they appear to be vanishing from China as well. (Reliable data on this transition are hard to come by. Official Chinese figures report a decline of 30 million manufacturing jobs since 1996, or 25 percent of the total, even as manufacturing output has soared by over 70 percent, but part of that drop may reflect revisions in the methods of gathering data.)
If this prediction holds true, what does this mean for the future of jobs and the economy?

1) Outsourcing would become much less common. After all, if most of the cost of production is embodied in capital like robots and 3D printers, then the advantage to cheap labor becomes minimal. Brynjolfsson, McAfee, and Spence write: "As work stops chasing cheap labor, moreover, it will gravitate toward wherever the final market is, since that will add value by shortening delivery times, reducing inventory costs, and the like."

2) For low-income and middle-income countries like China that have thrived on being the workshops and manufacturing centers of the global economy, their jobs and workforce would experience a dislocating wave of change.

3) Some kinds of physical capital are going to plummet in price, like robots, 3D printing, and artificial intelligence doing many more tasks in both manufacturing and services. Especially as robots become capable of building more robots, capital goods will be abundant in a way that will not generate high returns to capital.

4)  So if many workers are going to find their jobs disrupted and many makers of capital equipment are going to find themselves in a brutal competitive battle to reduce price and raise capabilities, who does well in this future economy? For high-income countries like the United States, Brynjolfsson, McAfee, and Spence emphasize that the greatest rewards will go to "people who create new ideas and innovations," in what they refer to as a wave of "superstar-based technical change." For the MBA students at MIT and NYU, where these authors are based, this probably qualifies as thrilling news. But for the typical worker, the largely unspoken implication seems fairly grim. If you aren't a superstar entrepreneur, then you are likely to be replaced by a robot, or a lower-paid work in another country, or you'll have to scramble against all the other non-superstars to find a job in the remainder of the economy.

This final forecast seems overly grim to me. While I can easily believe that the new waves of technology will continue to create superstar earners, it seems plausible to me that the spread and prevalence of many different new kinds of technology offers opportunities to the typical worker, too. After all, new ideas and innovations, and the process of bringing them to the market, are often the result of a team process--and even being a mid-level but contributing player on such teams, or a key supplier to such teams, can be well-rewarded in the market. More broadly, the question for the workplace of the future is to think about jobs where labor can be a powerful complement to new technologies, and then for the education and training system, employers, and employees to get the skills they need for such jobs. If you would like a little more speculation, one of my early posts on this blog, back on July 25, 2011, was a discussion of "Where Will America's Future Jobs Come From?"