Wednesday, June 17, 2009

How many jobs are onshorable? Re-interpreting the Blinder numbers in the light of new trade theory

How many jobs are onshorable? Re-interpreting the Blinder numbers in the light of new trade theory. By Richard Baldwin
VoxEU, Jun 15, 2009

According to Alan Blinder, constant improvements in global communications will bring much more offshoring of “impersonal services’’, with an estimated 30 million to 40 million US jobs potentially offshorable. This column warns against taking these numbers at face value and recalls that the US is actually a net insourcer. With the advance of communication technologies, the US should see lots more service jobs “offshored” and lots more “onshored”.

Before the global crisis hit, offshoring was one of the scarcest things on rich nations’ economic radar screens – especially the offshoring of “good” service sector jobs. Alan Blinder was one of the first to point out the threat in his 2006 Foreign Affairs article “Offshoring: The Next Industrial Revolution?” He wrote: “constant improvements in technology and global communications virtually guarantee that the future will bring much more offshoring of ‘impersonal services’’— that is, services that can be delivered electronically over long distances with little or no degradation in quality.”

Blinder has more recently produced some estimates of the size of the revolution. And they make it look like “the big one”. Blinder (2009): "I estimated that 30 million to 40 million US jobs are potentially offshorable."

This sort of media-friendly statement is part of what I consider to be very confused thinking by non-specialists – not that the economists involved are necessarily confused, but tacitly or not, they are allowing the media to misinterpret the numbers.

Let me start off by saying that I consider Alan Blinder to be one of the world’s leading macroeconomic policy specialists. Moreover, I greatly appreciate the way he uses his knowledge of economics to make this a better world (rather than focusing entirely on impressing the other inhabitants of academe). This time, however, I’m not sure it has worked out right.

I don’t wish to take issue with his numbers or methods. I wish to question the implications of those numbers. The trouble is that his numbers are being interpreted in the light of the “old paradigm” of globalisation – the world of trade theory that existed before Paul Krugman, Elhanan Helpman, and others led the “new trade theory” revolution in the 1980s.

The new trade theory: Micro, not macro, determinants of comparative advantage

Krugman’s contribution, which was rewarded with a Nobel Prize in 2008, was to crystallise the profession’s thinking on two-way trade in similar goods.1 This was a revolution since the pre-Krugman received wisdom assumed away such trade or misunderstood its importance. In 1968, for example, Harvard economist Richard Cooper noted the rapid rise in two-way trade among similar nations and blamed it for the difficulty of maintaining fixed exchange rates. Using the prevailing trade theory orthodoxy, he asserted that this sort of trade could not be welfare-enhancing. And since it wasn’t helping, he suggested that it should be taxed to make it easier to maintain the world’s fixed exchange rate system – a goal that he considered to be the really important thing from a welfare and policy perspective (Cooper, 1968).

Trade economists back then took it as an article of faith that trade flows are caused by macro-level differences between nations – for example, national differences between the cost of capital versus labour. Nations that had relatively low labour costs exported relatively labour intensive goods to nations where labour was relatively expensive.

This is the traditional view that Blinder seems to be embracing.

What Krugman (especially Krugman 1979, 1980) showed was that one does not need macro-level differences to generate trade. Firm-level differences will do.

In a world of differentiated products (and services are a good example of this), scale economies can create firm-specific competitiveness, even between nations with identical macro-level determinants of comparative advantage. Krugman, a pure theorist at the time, assumed that nation’s were identical in every aspect in order focus on the novel element in his theory (and to shock the “trade is caused by national differences” traditionalists). His insight, however, extends effortlessly to nations that also have macro-level differences, like the US and India.

This brings us to interpreting Blinder’s 30 to 40 million offshorable jobs.

Blinder’s calculations

Blinder’s approach is easy to explain – a fact that accounts for much of its allure as well as its shortcomings.
  • Step 1 is to note that Indian wages are a fraction of US wages.
  • Step 1a is to implicitly assume that Indians’ productivity-adjusted wages are also below those of US service sector workers, at least in tradable services.
  • Step 2, and this is where Blinder focused his efforts, is to note that advancing information and communication technology makes many more services tradable. The key characteristic, Blinder claims, is the ease with which the service can be delivered to the end-user electronically over long distances.
  • Step 3 (the critical unstated assumption, if not by Blinder, at least by the media reporting his results) is that the new trade in services will obey the pre-Krugman trade paradigm – it will largely be one-way trade. Nations with relatively low labour costs (read: India) will export relatively labour-intensive goods (read: tradable services) to nations where labour is relatively expensive (read: the US).

The catch

This last step is factually incorrect, as recent work by Mary Amiti and Shang-Jin Wei (2005) has shown. They note: “Like trade in goods, trade in services is a two-way street. Most countries receive outsourcing of services from other countries as well as outsource to other countries.”

[graph: US insourcing and outsourcing of jobs]
Source: Author’s manipulation of data from Amiti and Wei (2005), originally from IMF sources on trade in services.

The US, as it turns out, is a net “insourcer”. That is, the world sends more service sector jobs to the US than the US sends to the world, where the jobs under discussion involve trade in services of computing (which includes computer software designs) and other business services (which include accounting and other back-office operations).

The chart shows the facts for the 1980 to 2003 period. We see that Blinder is right in that the US importing an ever-growing range of commercial services – or as he would say, the third industrial revolution has resulted in the offshoring of ever more service sector jobs. However, the US is also “insourcing” an ever-growing number of service sector jobs via its growing service exports. The startling fact is that not only is the trade not a one-way ticket to job destruction, the US is actually running a surplus.


None of this should be unexpected. The post-war liberalisation of global trade in manufactures created new opportunities and new challenges. To apply Blinder’s logic to, say, the European car industry in the early 1960s, one would have had to claim that since the German car industry (at the time) faced much lower productivity-adjusted wages, freer trade would make most French auto jobs “lose-able” to import competition. Of course, many jobs were lost when trade did open up, but many more were created. As it turned out, micro-level factors allowed some French firms to thrive while others floundered, and the same happened in Germany. Surely the same sort of thing will happen in services, as trade barriers in that sector fall with advancing information and communication technologies.

In short, what Blinders’ numbers tell us is that a great deal of trade will be created in services. Since services are highly differentiated products, and indivisibilities limit head-to-head competition, my guess is that we shall see a continuation of the trends in the chart. Lots more service jobs “offshored” and lots more “onshored”. What governments should be doing is helping their service exporters to compete, not wringing their hands about one-way competition from low-wage nations.


1 Full disclosure: Krugman was my PhD thesis supervisor and we have coauthored a half-dozen articles since 1986.


Amiti, M. and S.J. Wei (2005), “Fear of Service Outsourcing: Is it Justified?”, Economic Policy, 20, pp. 308-348.
Blinder, Alan (2006). “Offshoring: The Next Industrial Revolution?” Foreign Affairs, Volume 85, Number 2.
Blinder, Alan (2009). “How Many U.S. Jobs Might Be Offshorable,” World Economy, 2009, forthcoming.
Cooper, R. (1968). The Economics of Interdependence. New York: McGraw-Hill.
Grossman, G. and E. Rossi-Hansberg (2006a). “The Rise of Offshoring: It’s Not Wine for Cloth Anymore,” July 2006. Paper presented at Kansas Fed’s Jackson Hole conference for Central Bankers.
Krugman, Paul (1979). "Increasing returns, monopolistic competition, and international trade," Journal of International Economics, Elsevier, vol. 9(4), pages 469-479,
Krugman, Paul (1980). "Scale Economies, Product Differentiation, and the Pattern of Trade," American Economic Review, vol. 70(5), pages 950-59, December.
Krugman, Paul (1991), “Increasing Returns and Economic Geography”, Journal of Political Economy 99, 483-499.

The U.S. has been bringing soldiers home as soon as they get any experience

General McChrystal's New Way of War. By Max Boot
The U.S. has been bringing soldiers home as soon as they get any experience.
The Wall Street Journal, Jun 17, 2009, page A13

Gen. Stanley McChrystal was appointed commander in Afghanistan to shake up a troubled war effort. But one of his first initiatives could wind up changing how the entire military does business.

Gen. McChrystal's decision to set up a Pakistan Afghanistan Coordination Cell means creating a corps of roughly 400 officers who will spend years focused on Afghanistan, shuttling in and out of the country and working on those issues even while they are stateside.

Today, units typically spend six to 12 months in a war zone, and officers typically spend only a couple years in command before getting a new assignment. This undermines the continuity needed to prevail in complex environments like Afghanistan or Iraq. Too often, just when soldiers figure out what's going on they are shipped back home and neophytes arrive to take their place. Units suffer a disproportionate share of casualties when they first arrive because they don't have a grip on local conditions.

There was a saying that we didn't fight in Vietnam for 10 years; we fought there for one year, 10 times. The North Vietnamese, on the other hand, continued fighting until they were killed or immobilized. That gave their forces a huge advantage.

In Vietnam, units already in the field would get individual replacements from home, thus making it hard to maintain unit cohesion. Sometimes new soldiers were killed before anyone even knew their names.

The policy now is unit rotation -- an entire battalion or brigade (or a higher-level staff) trains together, deploys together, and leaves together. That makes for better cohesion, but makes it even harder to maintain continuity because there is little overlap between units.

In a tribal society like Afghanistan's, the key to effectiveness is having personal relationships with tribal elders, which argues for keeping troops in place much longer than currently is the case. But there are limits to the stress that soldiers can endure -- effectiveness degrades severely for anyone who spends too long in combat. And in an all-volunteer military, there is always the danger that if troops are forced to be away from their families too long they might not sign up for another hitch.

The U.S. Special Operations Command (the military command for all special operations units) has responded by creating a deployment cycle whereby units spend roughly six months deployed in a war zone and six months at home, keeping tabs on their area of operations while they're away and returning to the same area time after time. This arrangement, which has been in use for several years, allows personal relationships to be cultivated and continued while still giving troops some downtime.

It's an intriguing approach, and one that Gen. McChrystal, a veteran of special operations, is now migrating to the conventional military world. The new Pakistan Afghanistan Coordination Cell is an attempt to strike a balance between personnel needs and war-fighting needs, and it is a move in the right direction.

I would argue for going even further by extending staff deployments (which aren't as stressful as combat jobs). Volunteers should be allowed to spend years at a time in places like Afghanistan -- not only soldiers but also diplomats and intelligence officers.

Who would volunteer to live in such an inhospitable environment? Well Sarah Chayes, a former NPR reporter, has been living and working in Kandahar since 2001. While in Afghanistan recently, I also met a former Special Forces soldier, now working as a State Department counter-narcotics contractor, who said he has been in Afghanistan for four years. Such people are invaluable for their knowledge of the local landscape.

The British, from whose glory days we can still learn many lessons, recognized this. Gertrude Bell, Richard Francis Burton, T.E. Lawrence and numerous others made an outsize contribution to their empire by "going native." They may have been sneered at by typical army officers, who were primarily interested in polo, whist and gin, but the knowledge they acquired proved invaluable.

Consider the case of Col. Sir Robert Warburton, a 19th century artillery officer who was the offspring of a marriage between a British officer and an Afghan princess. He spent nearly 30 years on the Northwest Frontier of India working as a political officer, negotiating with tribesmen who were (and are) suspicious of all outsiders.

"It took me years to get through this thick crust of mistrust, but what was the after-result?" he wrote in his memoirs. "For upwards of fifteen years I went unarmed amongst these people. My camp, wherever it happened to be pitched, was always guarded and protected by them. The deadliest enemies of the Khyber Range, with a long record of blood-feuds, dropped those feuds for the time being when in my camp."

Warburton retired in May 1897. Within months the frontier was aflame with a great uprising that took tens of thousands of troops to suppress. (You can read all about it in Winston Churchill's first book, "The Story of the Malakand Field Force," which contains eerie echoes of current fighting on both sides of the Afghanistan-Pakistan border.) Warburton, who had been known as the "King of the Khyber," was convinced that if he were still on the job, the contacts he had cultivated would have allowed him to prevent the uprising. He may well have been right.

What Gen. McChrystal realizes, in effect, is that we need to create our own Robert Warburtons. If his experiment succeeds, future commanders can build on the precedent to provide the kind of cultural and linguistic skills that we will need to win the long war against Islamic extremists.

Mr. Boot is a senior fellow in national security studies at the Council on Foreign Relations. He is currently writing a history of guerrilla warfare.

Health Reform and Competitiveness

Health Reform and Competitiveness. WSJ Editorial
WSJ, Jun 17, 2009

Democrats have spent years arguing that corporate tax rates don't matter to U.S. competitiveness. But all of a sudden one of their favorite arguments for government-run health care has become . . . U.S. corporate competitiveness. Political conversions on this scale could use a little scrutiny.

"Businesses now recognize that if we don't get a handle on this stuff then they are going to continue to be operating at a competitive disadvantage with other countries," President Obama recently remarked. "And so they anxiously seek serious reform."

Sure enough, many business leaders who should know better have picked up the White House theme. "You won't fundamentally solve the problems in business until you solve the problem of spiraling health-care costs, which is driving everybody crazy," said Google CEO Eric Schmidt the other day.

Messrs. Obama and Schmidt need to brush up on their economics. Employers may write the checks to the insurance companies, but workers still pay for the coverage they get from those employers. The total cost of an employee is what matters to businesses, and fringe benefits are as much a part of compensation as cash wages. When health costs rise, firms don't become less competitive, as if insurance were lopped out of profits. Instead, nonhealth compensation drops. Or wages rise more slowly than they otherwise would.

A recent study from none other than the White House Council of Economic Advisers notes exactly this point: If medical spending continues to accelerate, it expects take-home pay to stagnate. According to the New York Times, White House economic aide Larry Summers pressured CEA chairman Christine Romer to make the competitiveness argument, "adding that it was among the political advisers' favorite 'talking points.'" Ms. Romer pointedly retorted, "I'm not going to put schlocky arguments in there." How the schlock gets into Mr. Obama's speeches is a different question.

It's certainly true that the U.S. employer-based insurance system can dampen entrepreneurial spirits. There's the "job lock" phenomenon, in which employees fear leaving a less productive job because they're afraid to lose their health benefits. Another problem is that insurance costs more for small groups than the large risk pools that big corporations assemble, meaning that it's harder to form new businesses that can offer policies. But all this is really an argument for developing the individual health insurance market, where policies would follow workers, not jobs.

As for the competitiveness line, it's nonsense for most companies. The exceptions are heavily unionized businesses like auto makers that have locked themselves in to gold-plated coverage, especially for retirees. They have a harder time adjusting health costs and wages. Other companies might get a bit more running room in the short run if government assumed all health costs a la the single-payer systems of Western Europe. But over time the market would clear -- compensation being determined by the demand for and supply of labor -- and wages would rise. Or they might not rise at all if health-care costs are merely replaced by the tax increases necessary to finance Mr. Obama's new multi-trillion-dollar entitlement.

This is where the real competitiveness argument is precisely the opposite of the one pitched by Messrs. Obama and Schmidt. Consider the European welfare states, where costly entitlements and regulations make it extremely expensive to hire new workers. The nearby table lays out the tax wedge, the share of labor costs that never reaches employees but instead goes straight to government. In Germany, France and Italy, the tax wedge hovers around 50%, in part to pay for state-provided health care.

By contrast, the U.S. tax wedge was around 30% in 2008, according to the OECD. In other words, the costs of providing insurance would merely be converted into a larger wedge, which would itself eat into compensation. This is why Europe has tended to have higher unemployment and slower economic growth over the past 30 years.

If Democrats really want to increase U.S. competitiveness, they could look at the corporate income tax, which is the second highest in the industrialized world and a major impediment to U.S. job creation when global capital is so fluid. Or drop their proposals to raise personal income-tax rates, which affect thousands of small- and medium-size businesses that have fled the corporate tax regime as limited liability companies or Subchapter S corporations. Or cut capital gains rates, which deter risk taking and investment. Or rethink their plans to rig the rules in favor of organized labor by doing away with secret ballots in union elections.

On all these issues and more, Democrats want to increase, not reduce, the burdens on U.S. business. Their health-care line is, per Ms. Romer, "schlocky" political spin.

WSJ Editorial Page on Federal President's Position on Iran's Elections

Obama's Iran Abdication. WSJ Editorial
Democracy interferes with his nuclear diplomacy script.
The Wall Street Journal, Jun 17, 2009, page A12

The President yesterday denounced the "extent of the fraud" and the "shocking" and "brutal" response of the Iranian regime to public demonstrations in Tehran these past four days.

"These elections are an atrocity," he said. "If [Mahmoud] Ahmadinejad had made such progress since the last elections, if he won two-thirds of the vote, why such violence?" The statement named the regime as the cause of the outrage in Iran and, without meddling or picking favorites, stood up for Iranian democracy.

The President who spoke those words was France's Nicolas Sarkozy.

The French are hardly known for their idealistic foreign policy and moral fortitude. Then again many global roles are reversing in the era of Obama. The American President didn't have anything to say the first two days after polls closed in Iran on Friday and an improbable landslide victory for Mr. Ahmadinejad sparked the protests. "I have deep concerns about the election," he said yesterday at the White House, when he finally did find his voice. "When I see violence directed at peaceful protestors, when I see peaceful dissent being suppressed, wherever that takes place, it is of concern to me and it's of concern to the American people."

Spoken like a good lawyer. Mr. Obama didn't call the vote fraudulent, though he did allow that Ayatollah Ali Khamenei "understands the Iranian people have deep concerns about the election." This is a generous interpretation of the Supreme Leader's effort to defuse public rage by mooting a possible recount of select precincts. "How that plays out," Mr. Obama said, "is ultimately for the Iranian people to decide." Sort of like the 2000 Florida recount, no doubt.

From the start of this Iranian election, Administration officials said the U.S. should avoid becoming an issue in the campaign that the regime might exploit. Before votes were cast, this hands-off strategy made sense in that the election didn't present a real choice for Iranians. Whether President Ahmadinejad or his chief challenger, Mir Hossein Mousavi, won wouldn't change the mullahs' ultimate political control. Mr. Mousavi had been Ayatollah Khomeini's Prime Minister, hardly the resume of a revolutionary.

But Friday's vote and aftermath have changed those facts on the ground. Like other authoritarians -- Ferdinand Marcos in 1986 or Slobodan Milosevic in 2000 -- Tehran misjudged its own people. Having put a democratic veneer around their theocracy, they attempted to steal an election in such a blatant way that it has become a new and profound challenge to their legitimacy. Especially in the cities, Iranians are fed up with the corruption and incompetence rampant in the Islamic Republic. This dissatisfaction was galvanized by the regime's contempt for their votes and found an accidental leader in Mr. Mousavi. The movement has now taken on a life of its own, with consequences no one can predict.

The Obama Administration came into office with a realpolitik script to goad the mullahs into a "grand bargain" on its nuclear program. But Team Obama isn't proving to be good at the improv. His foreign policy gurus drew up an agenda defined mainly in opposition to the perceived Bush legacy: The U.S. will sit down with the likes of Iran, North Korea or Russia and hash out deals. In a Journal story on Monday, a senior U.S. official bordered on enthusiastic about confirming an Ahmadinejad victory as soon as possible. "Had there been a transition to a new government, a new president wouldn't have emerged until August. In some respects, this might allow Iran to engage the international community quicker." The popular uprising in Iran is so inconvenient to this agenda.

President Obama elaborates on this point with his now-frequent moral equivalance. Yesterday he invoked the CIA's role in the 1953 coup against Iranian leader Mohammad Mossadeq to explain his reticence. "Now, it's not productive, given the history of the U.S.-Iranian relations, to be seen as meddling -- the U.S. President meddling in Iranian elections," Mr. Obama said.

As far as we can tell, the CIA or other government agencies aren't directing the protests or bankrolling Mr. Mousavi. Beyond token Congressional support for civil society groups and the brave reporting of the Persian-language and U.S.-funded Radio Farda, America's role here is limited. Less than a fortnight ago, in Cairo, Mr. Obama touted his commitment to "governments that reflect the will of the people." Now the President who likes to say that "words matter" refuses to utter a word of support to Iran's people. By that measure, the U.S. should never have supported Soviet dissidents because it would have interfered with nuclear arms control.

The Iranian rebellion, though too soon to call a revolution, is turning out to be that 3 a.m. phone call for Mr. Obama. As a French President shows up the American on moral clarity, Hillary Clinton's point about his inexperience and instincts in a crisis is turning out to be prescient.