Tuesday, April 19, 2011

Why US Multinationals Expand Abroad

Mark Thoma points us to an article by David Wessel, who points out that new data from the BEA indicates that US-based multinational corporations (MNCs) decreased employment in the US while increasing employment outside the US:
Big U.S. Firms Shift Hiring Abroad
U.S. multinational corporations, the big brand-name companies that employ a fifth of all American workers, have been hiring abroad while cutting back at home, sharpening the debate over globalization's effect on the U.S. economy.
I would like to sound a note of extreme caution when interpreting such data. It's easy to jump to the conclusion that this data indicates that MNCs are shifting jobs overseas, and that foreign employment growth is coming at the expense of jobs in the US. However, that is probably not what's going on here.

The vast majority of employment and sales by the foreign affiliates of US-based MNCs are serving the local market. When GE, or Microsoft, or Coca-Cola, or American Express expand their operations overseas, it is almost always with the primary goal of satisfying local demand, rather than replacing workers in the US. Yes, of course some offshore outsourcing does happen (though much less over the past few years than happened in the early 2000s), but really that's not what's driving the dramatic difference in employment patterns of US MNCs within the US compared to outside the US.

The following table shows employment changes and sales growth between 1999 and 2008 by major region. (Source: BEA data.) As you can see, employment of US-based MNCs did indeed fall by about 8% in the US during the period, while employment by those same MNCs rose elsewhere in the world by about 30%. In fact, US-based MNCs increased employment in the Asia-Pacific region by almost 75% over the period.

However, this reflects the fact that sales grew much faster for MNCs in countries other than the US. For example, the US branches of US-based MNCs sold only about 46% more in the US market in 2008 than they did in 1999. Meanwhile, the affiliates of US MNCs located in Europe increased their sales (excluding to the US) by 122%, while affiliates in the Asia-Pacific region increased their sales to markets other than the US by 177%. In other words, US-based MNCs expanded their operations outside the US dramatically during the period, but not primarily in order to produce stuff to be shipped to the US; rather, that overseas expansion was to serve overseas markets, which were booming for them.

The next table provides another perspective on this phenomenon, looking at absolute values of sales. In 2008, for example, the US branches of US-based MNCs sold approximately $8.7 trillion to the US market, while selling about $550 billion to non-US markets (i.e. exports). Meanwhile, the affiliates of US-based MNCs in Canada sold about $108 bn to the US but about $485 bn to markets outside the US (presumably most of that within Canada itself). And affiliates in Asia shipped only about 5% of their production back to the US -- the other 95% of sales by those affiliates of US-based MNCs were to markets outside the US (again, probably mostly in the Asia-Pacific region itself).

The point is not to argue that offshoring never happens. It does. But the pattern of international trade, particularly when it comes to the activity of MNCs, is much more complex and nuanced than that. And the clearest implication of this data is that the primary motivation for MNCs to expand their operations outside the US is not to produce stuff more cheaply there to be sold to the US. Rather, MNCs expand overseas mainly to service overseas markets.

UPDATE: I fixed a calculation error in the tables and modified the text to match.


  1. Gpras8:29 AM

    <span>Kash - I would argue a slightly different point. If firms are initially exporting to foreign markets and then move those operations abroad to be closer to those markets, that activity directly substitutes foreign workers for domestic workers.   
    Compare that to offshoring where MNCs move abroad to take advantage of cheaper labor abroad. Again, there is a direct substitution against American workers but there is a positive productivity effect here as well.   
    There is some evidence that the former leads to more losses of American jobs than the latter. </span>

  2. coberly12:57 PM


    I don't see the logic of this. Are you saying that the fact I can't buy anything that is not made in China has nothing to do with the difficulty Americans are having in finding jobs?

    Are you saying that an 8% cut in American jobs is justified by selling more TO "china"?

    are you saying America can't make stuff here and sell it there?

    I mean, you could be right, but I don't see the logic.

  3. coberly - no, I'm not at all trying to argue that US MNCs never produce stuff overseas for shipment back to the US.  It happens to the tune of about $300+ bn per year, in fact, which is a pretty big number.  The point is simply that there's much more to the story than that.  In addition to making stuff overseas for sale in the US, US MNCs are also interested in serving local markets around the world.  And the data shows that this second piece is a much, much larger component of what MNCs do than the first piece, by a factor of more than 10.  So, given what a small fraction of the output of MNCs is actually shipped and sold internationally in general, it is not surprising that companies hire lots of workers outside the US when markets outside the US are booming.

    Gpras - I agree that to some degree MNCs produce stuff locally around the world when they could in theory have produced it in the US and exported it.  However, this is a relatively small effect, simply because the vast majority of the sales of MNCs in general are produced and sold locally.  When US exports by MNCs amount to about $500bn/year but sales of MNCs around the world rise by $3,000 bn/year (as they did between 1999 and 2008), it's only natural that the vast majority of that increase in demand is going to be met by local production rather than production in the US. (Though US exports certainly did grow considerably over the period as well.)

    As a general point, the data illustrates that over 90% of what MNCs do is serve local markets, and there's a reason for that - it's frequently easier, cheaper, and more convenient to be near your customers, particularly since most of the output of MNCs is actually services, not goods, and so can't really be produced in one country but sold in a different one even if MNCs wanted to.

  4. Kash I think you miss the point a little. Sure, foreign demand for goods made by US based companies has increased. That the products these foreign buyers purchase are made in the country of final demand does not mean that a US job wasn't off shored or, more likely, simply never created because production capacity was increased outside the US. 

    I would very much like to know the percentage of goods made by US based companies that used to be made in the US and exported to foreign buyers rather than domestically produced there as they are now. And I'm particularly interested in why. If goods are manufactured in countries in or near final demand because of import restrictions or tariffs by the home country, I have a real problem. I have less of an issue if legitimate cost factors are involved. However, either way, it is important to understand the cause in order to take corrective action here at home. If we simply shrug and say "there's nothing we can do to change things" then a long slow decline is inevitable.

  5. Anonymous3:47 AM

    There is definately a great deal to find out about this subject.
    I really like all of the points you have made.

    Feel free to visit my weblog - new cellulite treatment

  6. Anonymous3:47 AM

    Terrific work! This is the type of information that are supposed to be shared around the internet.
    Shame on Google for no longer positioning this submit higher!
    Come on over and consult with my web site . Thanks =)

    My site ... cellulite treatment reviews