The Pol Antràs manuscript on *Global Production*

The subtitle of the book is Firms, Contracts, and Trade Structure.  Here is one interesting bit of many:

…in that same year [2011] the share of intrafirm trade reached a record 89.6 percent for U.S. imports from Western Sahara. Leaving aside communist dictatorships and disputed territories, and focusing on the 50 largest exporters to the U.S., Figure 1.3 illustrates that the share of intrafirm trade still varies significantly across countries, ranging from a mere 2.4 percent for Bangladesh to an astonishing 88.5 percent for Ireland.

In that paragraph you can think of “intrafirm trade” as basically standing in for multinationals.  One motivation for this book is that a lot of earlier theories focus on international trade across firms and governments, when in fact much of what goes in is intra-firm and thus requires some more subtle theories.  Chapter two of the book is an excellent introduction to how international trade theory has shifted from an emphasis on countries or sectors to individual business firms, a significant advance.  A Nobel Prize for Marc Melitz and Antràs is by no means out of the question someday.

The manuscript itself is here (pdf), some data files are here.  His home page is here.


This reminds me of Noam Chomsky's line that a lot of so called "free trade" is really just allocation within a small command control economy.

The Shame of MIT strikes again.

Sam's comments remind me that Bastiat is still relevant and that people still have trouble imagining the "things unseen."

Apparently, unless they see a flea market with their own eyes, some people can't imagine that competition exists and that it shapes the decisions of big companies to a large extent.

Well the Theory of the Firm says that resources are allocated within a firm not using prices (for supposed ease in transaction costs, though it's true some corporations attempt to assign prices, but they are artificial, typically done for tax purposes), that firms are typically overstaffed within ("three people to do one person's work", typically in the government or big Fortune 100 companies, the rationale being a company can fire and replace people in-house more efficiently than hiring outside people) and big companies are market makers not takers, meaning classic Smithsonian perfect competition does not hold.

Thus essentially, at its core, a market comprising firms is communistic, not true capitalistic. It's also true that society depends on advance from exogenous technological advances, typically from inventors not working for money (e.g., Nobel Prize scientists who are motivated by personal reasons or fame not money).

Such is the paradox of modern capitalism. I could write a book on it, model it with some computer code, join the economics profession and maybe win a Nobel Prize someday, but I have a life already.

I may be misreading, but Antras still considers ex-ante hold-up to be the problem that needs to be solved while working within the transaction cost approach. Hold-up is almost never seen in the real world though while it is ex-post negotiations that often prove to be substantially costly, if the parties to the trade did not anticipate these and work out an approximate solution.

Still, embracing a contractual approach and trying to give both formal and empirical content to it is a most excellent move.

The high % of intrafirm trade for Ireland is interesting. At some level it is driven by taxes, because US multinationals use Ireland as their tax haven of choice, but at another level it is driven by non tax considerations.

Their are a lot of tax havens, and in general the US structure avoids Irish tax. So why Ireland? US tax laws require a real presence, with real people doing real work. You can do this in Ireland, and not most of the "island" tax havens. Second, they speak English in Ireland, and it is a member of EU with lots of good treaties.

Why Ireland? Possibly Teddy Kennedy? I recall reading something I think back in the 90s about legislation and Microsoft or Apple. Plus Ireland overhauled their tax code in the late 80s or 90s, didn't they? Ireland has a lot of "troubles" back then.

If there weren't so many land mines everywhere, there would be incredible potential to exploit nearly 1000 km of amazing beaches.

I think the EU also complained because at one time they were a net receiver of monies, then the miracle happened and when it was time to pay their fair share back, or to help out the poorer countries, they were avoiding it.

I do wonder about double, triple, quadruple counting. If firms are moving raw materials and partially finished goods around the world, it will look like lots of "trade" but involve lots of double counting. Also, all that you're saying is that transaction costs have allowed firms to extend well beyond factory walls for greater efficiency. This is a form of gains from trade along with the beneficial trade in the final goods.

After that, Brown and Schwarzenegger will head over
the State Capitol's rotunda, exactloy where they'll celebrate the unveiling of Schwarzenegger's official gubernatorial portrait.

Simply wish to say your article is as surprising.
The clearness to your post is just cool and that i could suppose you're an expert on this subject.

Fine with your permission let me to seize your RSS
feed to keep up to date with coming near near post.
Thank you one million and please keep up the rewarding work.

Comments for this post are closed