Nice analysis.

Dan - I just got an email from Salma Hayek. Seriously. From the One Campaign. Part of the "make poverty history" team. I thought perhaps you could give us some of your thoughts on the issue - or, if not, at least you know you can get an email from Salma via signing up at One.

ACTIVATING THE CHINA SEP FIELD

Daniel, you seem to have done an excellent job of erecting a SEP field around the problem of exporting manufacturing, engineering, and other jobs to China. Your loyal readers are overwhelmingly trying to tell you that they are feeling the effects, and many of us have constructive suggestions for what could be done about it. But you persist in telling us that there is no problem!

True, our experiences are "merely anecdotal" while you have statistics to back your argument. (Insert statistics joke here.) But really, what is wrong with recognizing that American workers are losing the competition in the global labor market and our government is doing nothing to help?

I'm not suggesting tariffs or embargoes. Just a recognition that China has unfair advantages (beyond wages) that can be ameliorated in taxes, exchange rates, labor laws, and environmental laws. Let's at least recognize that there is a problem. I don't personally know a single person outside of government or academia who isn't effected by offshoring. And yet it seems that everyone inside those protected areas is oblivious.

Actually, Dan does not have the statistics --much less , the empirical data -- on his side.

Dell Computer's 10-K annual reports shows that Dell had 23,500 US employees in 2000 and 13,000 foreign employees.

Five years later, on March 2005, Dell US employees has only increased by 1,100 to 24,600.

By contrast, Dell foreign employment increased by 27,600 employees!! to 30,600.

(Note also that Dell's US employment actually FELL below the 2000 level in intervening years due to large layoffs).
Ref: http://finance.yahoo.com/q/sec?s=DELL , select 10-K reports for 2000 and 2005 ,and search for"employees"

Plus there are the million plus foreign workers (not immigrants) that the Republican COngresses let into the USA for six years by raising the H1B visa limits in 1998-2000 -- in exchange for large donations from the Silicon Valley firms.

Hard to tell where those foreigners are because the Bush administration refuses to identify the corporations which are hiring/sponsoring them.

Of course, we now have a small cottage industry of pundits writing reports assuring the US public that nothing is wrong. Reports similar to the assurances that the vet gives to children as he leads the family dog away to be put to sleep.

WHat do you expect? The standard for intellectual integrity in academia today is "he whose bread I eat his song I sing".

Unlike our rich capitalists, America's workers can't hand out lucrative consulting contracts, lavish grants, or highly paid sinecures.

Larry says:

"Let's at least recognize that there is a problem. I don't personally know a single person outside of government or academia who isn't effected by offshoring. And yet it seems that everyone inside those protected areas is oblivious."

(Still) protected areas Larry. Don't worry. Their day is coming too.

Michigan released May jobs numbers today.

It may be 3 1/2 years after the recession ended, but Michigan is still sinking.

Manufacturing is sinking, of course.

High value white collar jobs are declining as well.

I think it was the WSJ (yesterday or today) with a piece something to the effect of "it still makes sense to go to engineering college." I wonder.

I'm dubious about this study. First, the survey was of hr managers; however, in well-managed companies, HR serves the line managers, not the other way around. The decision about outsourcing is not going to be made by HR.

Second, hiring is not necessarily done directly by the multinationals at all. In both manufacturing and services, it is often contracted out, and what is being contracted for is a task, not a specific set of employees.

I have to really question your comments that it is easier to retrain a college graduate then a
manufacturing employee.

Although a manufacturing employee has experience working in a specific environment, the job skills he will need in a new job or very similiar.

But a college graduate has years of specalized training in a field that are not directly transferable. As an example take petroleum engineering. I recently saw a stat that the average age of a petroleum engineer was 48. In the last oil boom we turned out too many petroleum engineers and for 20 years it was a backwater. Now we need them, but do not have young ones in the pipeline. Moreover, to retrain a laid off programmer with a MA to be a petroleum engineer would take something on the orders of 2-4 years.

A college graduate has years of very specialized embedded capital tht are not easily transferable to other fields. so to retrain them means throwing those years of education down the drain and restarting. But for a blue collar job the
shift to another field does not envolve losing so much embedded capital.

Spencer: most college graduates have degrees in academic subjects, not engineering. Most graduate jobs are desk-ish jobs, not engineering jobs. So while your specific examples work, your argument doesn't.

Don: you may want to observe the growing proportion of Dell's *revenues* that came from outside the US over the same period (in any case, an example of a single company is no more conclusive than an example of a single whiny ex-software-developer...)

Re the question about the retraining of college graduates: many jobs are more a function of attitude and implicit knowledge than they are of anything that can be implicitly trained for. Can you retrain an engineer as a salesman? Possibly..some engineers make great salesmen. But it's more about who they are than about any specific training.

Similarly, can you retrain a salesman as a graphics artist? Only if he already has some inherent sense for color and design, and the patience to sit at a desk doing highly-detailed work.

John B, I don't know what your point is. That some of Dell's revenue comes from outside the US, so it's OK to export jobs? That software developers are whiny and don't deserve retraining assistance or sympathy of any kind? The simple fact is that highly trained Americans are losing their jobs and are being forced to do "service economy" jobs at lower pay because neither the employer nor government will help retrain them. Also, practically no employer will hire a middle aged engineer starting in a new field, regardless of what retraining s/he has had. This is more than a loss to those individuals. It is a huge unused asset to the US.

Even though, on average, the US economy seems to be doing OK, we are mortgaging our future beyond our ability to pay. We are experiencing a "brain drain" for the first time in our history and most do not appreciate the effect that will have in the long term. And the steady loss of jobs that create anything of value means that we are turning ourselves into a 3rd world country.

Sorry to be so glum, but it annoys me that my 2 favorite whipping boys, Washington DC and academia, refuse to admit that there is a problem.

Re John B's comment: "in any case, an example of a single company is no more conclusive than an example of a single whiny ex-software-developer"
-------
Dell is a major computer company. The fact that its foreign employment rose by 27,600! employees in the past 5 years while it's US employment only rose 1,100 is a piece of empirical data that blows
the cited McKinsey report's vague, handwaving bullshit out of the water.

In my opinion, Drezner is borrowing from Glenn Reynold's tactics over at Instapundit -- i.e., if you want to float a particularly implausible line of bullshit, don't discredit yourself by stating it yourself -- just link to it and imply that it has credibility.

I would be happy to list other companies, but Hewlett Packard and IBM don't break their employment out by US vs foreign. Another example of the Bush SEC at work. Just as the Bush INS doesn't list who's hiring all those H1B foreign workers the Republican Congresses were bribed into admitting.

The "political science" question is why any US citizen would vote for the Republicans --- as opposed to viewing them with deep hatred and contempt.

I say that not as a liberal but as a long time member of the NRA, a former Republican, and a charter subscriber to Pat Buchanan's "American Conservative". I for one am fed up with FOx News and George Bush's fake patriotism. Whoring for the rich is whoring for the rich -- wrapping it in the flag doesn't disguise the fact that they're damaging this country.

My table is almost unreadable in the comments section, but it's what I've got, so I'll include it below for those really interested. It gives a snapshot of the rate of growth of IT employment for 'American' firms, domestic vs offshored. 'American' - hah, hah. We pay taxes, they abscond, offshore, etc. It's a real party, for those rich enough to be citizens of the global economy. I'll note here that Repugs don't give a hoot about the welfare of the American people - their whole agenda is just love songs to the corporatocracy, with no serious consideration to the problems of health insurance, re-training, and the like. Just last year the NSF did a study that found that 18,000 Americans died every year because they lacked health insurance. What we do to our own people is far worse than what any terrorists have done to us. What does that make us, a nation of suckers?

Outsourcing versus Headcount Reductions: The Case of India

(1) Current Manpower
(2) India Manpower
(3) Job Cuts Announced / Carried out,last 12 month
(4) Frozen hiring engineers globally while increasing India outsourcing

[Corporation] [CUR(1)] [IND(2)] [Plans,India Office] [Cuts]

Accenture 65000 3500 8000 Employees by 8/04 1000
Adobe Systems 3250 185 250 People in 6 months 260
Cadence 5000 315 Doubling team in 4 yrs 500
Cap Gemini 56500 800 2000 People by 12/03 1000
Cisco 34466 2300 NA (4)
Covansys 4556 2000 2800 People in 1 year 200
CSC 92000 1200 4800 People by 2004 607
EDS 138000 300 2400 People by 2005 8200
i2 2800 1000 Recruiting actively ~1800 people
IBM Glob. Serv. 150000 3100 10000 People In 3 years ~2000 people
Intel 79200 950 3000 People by 2005 4700
Keane 5819 623 2000 People by end 2003 607
Logica-CMG 24000 350 1000 People by end 2004 2650
Lucent 35000 570 NA 13800
Microsoft 55000 200 500 People in 3 years hiring
Oracle 40000 3159 6000 People in 1 yr 200
Sapient 1500 600 Growing India Center 863
SunMicro 36000 700 Growing India Center 5480
Syntel 2700 2000 650 NA
Texas Instr. 34400 900 1500 People by Mar '06 800
Xansa 5583 1200 6000 People in 2-3 yrs 502

Source: Morgan Stanley India

That some of Dell's revenue comes from outside the US, so it's OK to export jobs?

No. In fact, that's a complete non-sequitur. In fact, there was no statistic presented about "exporting jobs."

Even granting the strange concept of "exportign jobs" in some contexts, hiring a foreign worker does not mean an American job has been "exported." If I start selling my products in, say, India, then I want to open a New Delhi office of my company to service customers there. I am not "exporting" a job by hiring people to work in that office.

That's different than hiring people in India to supply goods or services to sell to Americans. That's what he means.

That software developers are whiny and don't deserve retraining assistance or sympathy of any kind?

Well, they're definitely whiny.

One theory is that economists think free trade is good because they all work in safe jobs, whereas software developers don't. Another, slightly more plausible one, is that economists understand economics and software developers don't.

Dell is a major computer company. The fact that its foreign employment rose by 27,600! employees in the past 5 years while it's US employment only rose 1,100 is a piece of empirical data that blows
the cited McKinsey report's vague, handwaving bullshit out of the water.

Or, rather, it shows that Dell hired thousands and thousands of foreigners and yet still increased their number of American employees. Which sounds win-win for everyone.

David, the only reason people disagree with you is because they don't understand economics? OK then. Let me try to get through to you one more time: I am not saying (nor is anyone, I think) that free trade is bad. I am saying that when highly trained people are out of work and we all can see that most of the products that used to be designed and made in the US are now made in China, it is not hard to connect the dots. That in itself is not bad. Trade is good. What is bad is that our govt is doing nothing to help US workers compete, or to help them train themselves, or to help them find work after they've been retrained. It's a loss to the whole country, not just the "whiners".

Re David's comment: "One theory is that economists think free trade is good because they all work in safe jobs, whereas software developers don't. Another, slightly more plausible one, is that economists understand economics and software developers don't."

Actually , a third theory is that economists' articles often have no real ties to reality because economists shape their argument depending upon which butt they wish to kiss. Which, obviously, is not that of US workers because US workers are not the ones handing out grants, sinecures, and consulting contracts. Check the encyclopedia under "Sophists"

To check out David's idea that the beggers of Delhi are buying massive numbers of Dell computers, I checked Dell's 2000 and 2005 10Ks for revenue.

After all, if Dell's foreign employment went up 27 times it's US employment, then Dell's foreign revenue should have increased 27 times more than it's US revenue. Right?

However, Dell's US revenue increased from $16.878B in 1999 to $30.507B in 2004, an increase of 181%.

Dell's foreign revenue increased from $8.387B in 1999 to $18.698B in 2004, an increase of only 223% --

NOT the increase of 27 x 181% or 4887% that we would expect if Dell's foreign employment was being driven by the need to provide field support to those beggers in Delhi.

Re David's comment: "Even granting the strange concept of "exportign jobs" in some contexts, hiring a foreign worker does not mean an American job has been "exported."
-------------
Jobs are exported when rich men layoff workers in this country so that they can move capital abroad, set up businesses, and hire foreign workers in order to take advantage of the tax breaks the Republican Congresses give for foreign imperialism ..er, investments.

Bush's tax cut for the rich --funded by stealing $Trillions from middle class
America's Social Security Trust Fund --
created jobs in China, not in the US.

From 2001 to 2004, US Direct Investment Abroad increased by $102.8 Billion/YEAR. By contrast, US Fixed Investment,NonResidential (i.e., business investment in the USA) increased by only $52.7 Billion/YEAR.

Note that roughly $1 Trillion has to be invested EVERY YEAR in US plant just to keep our current level of employment
--due to the rapid depreciation of modern manufacturing (five year useable life of computer systems today
vice the 30 plus year life cycle of steel mills and railroads 100 years
ago.) The diversion of needed capital abroad is why the ranks of permanently US unemployed have soared and number in millions

Of course, those foreign investments would never occur if they were not protected by the blood and taxes ($450B/year + defense budget) of US workers.
Can't have a foreign business empire if you can't count on the US Marines to maintain order and twist a few arms when need be.

David, the only reason people disagree with you is because they don't understand economics?

I would certainly say that's true in DTG's case. (Witness bizarre claims like this: "After all, if Dell's foreign employment went up 27 times it's US employment, then Dell's foreign revenue should have increased 27 times more than it's US revenue. Right?" No, not right, DTG. There's no reason why there should be a 1-1 correlation, particularly when the workers in the two areas are paid different amounts.)

What is bad is that our govt is doing nothing to help US workers compete, or to help them train themselves, or to help them find work after they've been retrained.

In principle, that's reasonable. In practice, what exactly do you think the government should be "doing," and why do you think the government would do a good job at it? Since when is government good at "helping people find work"? For that matter, why do you think government "is doing nothing"?

Sigh
Re David's comment:
"I would certainly say that's true in DTG's case. (Witness bizarre claims like this: "After all, if Dell's foreign employment went up 27 times it's US employment, then Dell's foreign revenue should have increased 27 times more than it's US revenue. Right?" No, not right, DTG. There's no reason why there should be a 1-1 correlation, particularly when the workers in the two areas are paid different amounts.)"
---------------
Recall that it was David above who argued that Dell's huge increase in foreign employment was the result of Dell selling computers overseas (in India, in David's example.)

It David who does not understand economics :

From 1999 to 2004, Dell US revenues and Dell foreign revenues increased by roughly the same rate:
181% (US) vs 223% (foreign).

However, the number of Dell US employees only increased by roughly 4% whereas the number of Dell's Foreign employees increased by 235% !!

Note David's misunderstanding. Wage differences and exchange rates do not come into it. Dell
has the same technology and processes overseas as in the US. If Dell's US operations and foreign operations were really separate systems , then
you would see similar percentage increases in employees to get similar percentage increases in revenues in each system.

I.e., if I have to increase my number of employees by 200% in the foreign system in order to increase my foreign revenues by 200%, then I would have to similarly increase my number of US employees by 200% in order to increase my US revenues by 200%.

Given that Dell's US revenues increased by roughly the same percentage as foreign revenues (181 vs 223),
The increase of 235% in number of Dell's foreign employees versus the increase of only 4% in Dell's US employees shows that Dell's increases in US sales were based on work done by foreign employees. I.e., that Dell is shipping AMerican work abroad on a massive basis.

It may be that I misjudge David, of course. I'm assuming that he's making an argument whereas he may really be giving a hilarious parody of Tucker Carlton -- the conservative advocate who excels at presenting the illusion of a rational argument. An illusion that disappears when you look for content and discover that there is no "there" there.

Generally, I agree with David that the govt is not necessarily the best place to look for solutions to any kind of problem. But in this case, it is govt policy that largely created the problem, so only the govt can "undo" the damage. For example:

1. A tax system that inherently penalizes domestic production and encourages imports.
2. An environmental policy that pretends that we are green while paying the Chinese to polute like made while producing the products that we used to make.
3. A wide variety of workers protections that I certainly would not argue with, but they're useless because we condone abhorent practices in China every time we buy something made there.
4. Trade agreements that are supposedly based on free trade with trade partners that control their currencies at artificially low levels.
5. Regarding retraining: Why is it that a mid-career professional can deduct the cost of classes to further a dead-end career, but not to train for a new career?

The list goes on. States within the US compete with each other to attract employers because they know that it's good for the local economy and builds their tax base. Why doesn't the US govt recognize that the same competition exists between countries?

If Dell's US operations and foreign operations were really separate systems , then you would see similar percentage increases in employees to get similar percentage increases in revenues in each system.
I.e., if I have to increase my number of employees by 200% in the foreign system in order to increase my foreign revenues by 200%, then I would have to similarly increase my number of US employees by 200% in order to increase my US revenues by 200%.

There is no rational reason why that would be true. A necessary but not sufficient condition for that to be true is that workers be equally productive in each country -- something which clearly isn't the case. (And keep in mind that "employees" can be anything from janitors to engineers; why DTG thinks that throwing out gross employment numbers is illuminating is beyond me.)

-------

Larry: 5. Regarding retraining: Why is it that a mid-career professional can deduct the cost of classes to further a dead-end career, but not to train for a new career?

Probably because
(1) the very definition of "dead-end career" would imply that classes would _not_ further it -- so those who benefit from this are people whose careers are not "dead end" at all.
(2) if we enacted the latter policy, it would be rather difficult to limit the class of beneficiaries to those "training for a better career." Now, if one simply wants to make education generally deductible, fine -- but otherwise, there's no real limiting principle.

But that having been said, there _are_ taxpayer-subsidized programs out there for retraining people for new careers. Every time a free trade bill is passed, they up the funding for these.

The list goes on. States within the US compete with each other to attract employers because they know that it's good for the local economy and builds their tax base. Why doesn't the US govt recognize that the same competition exists between countries?

I don't know why you think government doesn't recognize this. I realize that the prototypical "whiny software engineers" think this is a new phenomenon because it suddenly affects some of them for the first time, but autoworkers (to pick one example) have been dealing with these issues for decades.

I'm just not sure what you think government can do about it. It's easy to (say) slap a tariff on imported televisions or cars or beef or underwear. It's a little more difficult to do the same for imported snippets of code or for the imported results of the reading of X-rays.

David, I don't know why you have it in for software people. When US mfg jobs left the country, economists said, that's OK, we'll retrain them as computer techs, etc and we'll all be better off in the long run. But now it's some of the most highly trained workers who are impacted. What is a PhD in physics supposed to do at age 50? or a mechanical engineer? or even a computer programmer? I know many examples of such folks who are currently in retail sales. Clearly you do not work in one of the effected fields, and I'm happy for you. But there is a huge loss in human capital taking place, even if you have no sympathy for those effected.

Re David's comment:"There is no rational reason why that would be true. A necessary but not sufficient condition for that to be true is that workers be equally productive in each country -- something which clearly isn't the case"

GIven that we are talking about the same company --same technology, same processes -- any productivity INCREASEs should be similar. The productivity baseline of US vs foreign workers --revenue /employee -- is clearly evident in the 10Ks as a starting condition for 2000. It is not even in the ballpark of being an alternative explanation for the massive difference in Dell hiring of foreigner vs US employees. IF foreign employees were as unproductive as David's argument suggests, then Dell would not have hired 27,000 of them.

David's inability to understand plain English and simple concepts --when it refutes his opinion --reminds me of Rush Limbaugh.

Some excerpts from recent AP news story, located at
http://www.detnews.com/2005/technology/0506/21/0tech-220566.htm
------------------
....Mo's decision to reboot his nascent career reflects a subtle but potentially significant industry shift. As tens of thousands of engineering jobs migrate to developing countries, many new entrants into the U.S. work force see info tech jobs as monotonous, uncreative and easily farmed out -- the equivalent of 1980s manufacturing jobs.

The research firm Gartner Inc. predicts that up to 15 percent of tech workers will drop out of the profession by 2010, not including those who retire or die. Most will leave because they can't get jobs or can get more money or job satisfaction elsewhere. Within the same period, worldwide demand for technology developers -- a job category ranging from programmers people who maintain everything from mainframes to employee laptops -- is forecast to shrink by 30 percent.

Gartner researchers say most people affiliated with corporate information technology departments will assume "business-facing" roles, focused not so much on gadgets and algorithms but corporate strategy, personnel and financial analysis.

"If you're only interested in deep coding and you want to remain in your cubicle all day, there are a shrinking number of jobs for you," said Diane Morello, Gartner vice president of research. "Employers are starting to want versatilists -- people who have deep experience with enterprise-wide applications and can parlay it into some larger cross-company projects out there."

Career experts say the decline of traditional tech jobs for U.S. workers isn't likely to reverse anytime soon.

The U.S. software industry lost 16 percent of its jobs from March 2001 to March 2004, the Washington-based Economic Policy Institute found. The Bureau of Labor Statistics reported that information technology industries laid off more than 7,000 American workers in the first quarter of 2005. ...
...Thousands of U.S. companies have opened branches or hired contractors in India, China and Russia, transforming a cost-saving trick into a long-term business strategy. Offshoring may be a main factor in eroding enthusiasm for engineering careers among American students, creating a vast supply of low-wage labor in eastern Europe and Asia and driving down worldwide wages.

The average computer programmer in India costs roughly $20 per hour in wages and benefits, compared to $65 per hour for an American with a comparable degree and experience, according to the consulting firm Cap Gemini Ernst & Young.

According to the most recent data from the National Science Foundation, 1.2 million of the world's 2.8 million university degrees in science and engineering in 2000 were earned by Asian students in Asian universities, with only 400,000 granted in the United States.
------------
This pretty much refutes Drezner's articles and the McKInsey report he cited.

David, I don't know why you have it in for software people.

Larry, I don't "have it in for" software people. While I personally am not currently one -- I left that path a while ago -- the rest of my family (siblings, father, wife) are, so if I believed as you did I would be worried. I don't, so I'm not. But who I do "have it in for" is whiny people -- a category in which software people seem to be overrepresented at the moment. (Although, of course, most software people are not.) It's not just that some of them think they're so special that they should be immune from the same competition that other industries have been subject to for quite a while, but that whenever someone points out that, in fact, there's no data to back up their cries that the sky is falling, they attack either the data or the professions of the people gathering the data. Most people in software and other technical fields are not bagging groceries, even if you happen to know "many examples" of such.

-----

DTG: IF foreign employees were as unproductive as David's argument suggests, then Dell would not have hired 27,000 of them.

They would if they were a lot cheaper, you twit.

When I said
"DTG: IF foreign employees were as unproductive as David's argument suggests, then Dell would not have hired 27,000 of them."

David responded
"They would if they were a lot cheaper, you twit"

To which I respond:
NOT if the foreign workers were being hired SOLELY to generate foreign sales , viced being hired to generate sales in America.

The reason is that if salaries are lower in a foreign country, then revenues from selling a Dell computer IN THAT COUNTRY will also be lower--i.e., if that worker's output is only sold within that country.
Given that Dell's industrial process is the same in US and foreign countries, if you pay a US worker $100K to generate $400K in revenues, then you will pay an Indian worker 100K rupees to generate 400K rupees in sales. If anything, your Indian operation is likely to be less profitable, because you have to bring in expensive foreign manufacturing equipment from outside.

Only if you can shift a low paid foreign worker's product from his native country to sell in America do you gain significantly -- and have reason for hiring 27,000 foreign workers while having large layoffs in the USA (albeit with a net gain of 1000 US workers after 5 years).

WHich, in cause David's handwaving has muddied the subject, is the issue under discussion.