Oddly, the original embargo was not lifted. The EU embargo, which is far less restrictive, was not lifted either (Canada never imposed an embargo; however, a large share of Canadian military hardware is of US origin, or else developed jointly with US firms, so the Commerce Department can exercise influence over re-sale of that technology to third countries. A case where this was used was in the late 1970's, when Ecuador, whose attempts to buy the F-16 were thwarted by the Carter arms embargo to Latin America, turned to the Israeli Kfir. The Commerce Department held up the sale on the grounds that the J79 used by the Kfir is manufactured by Gneral Electric—source. Soon after, GE and the Israeli foreign ministry prevailed on Washington to allow the sale to go ahead. Ecuador and Colombia are the two countries outside Israel that operate the Kfir as their frontline interceptor.) The reason for this was that London was collaborating with Washington on Chinese military proliferation and human rights.

This article by Colonel Daniel Smith of the US Army is of some interest to me on the complexities of the embargo. The frame of reference is after UK Foreign Secretary Jack Straw announced that his government, the lone EU holdout on lifting the ban, was withdrawing its longstanding opposition.

EU arms sales to China in 2002 were under $300 million and reached only $540 million in 2003. Russia and Israel are China’s top two arms suppliers [...]

[UK Foreign Secretary Jack] Straw assured the U.S. that Britain would pursue a “more effective arms control regime” for EU countries, presumably by narrowing the scope of sales permitted under the 1998 EU Code of Conduct for Arms Exports. Although this too, is a political and not a legal arrangement, the Code’s transparency provisions [...] are thought to have inhibited arms sales to China more than the 1989 embargo [did].

Looking at the three sections of the 1998 EU Code, where could Straw’s pledge be implemented? Not in the introduction, which describes the “state of play” in the world (the “whereas” section). In fact, after expressing a commitment to high common standards for restraining arms sales that could be used for repression or aggression, the basic enterprise is undercut by noting that EU countries wish “to maintain a defense industry as part of their industrial base as well as their defense effort.”

Not in the eight criteria governing arms trades with non-EU countries, since the Code has no penalties and no enforcement mechanism—not even an internal, mandatory process to resolve differences should EU member states reach opposite conclusions about an arms sale’s conformity to the criteria [...]

Perhaps in the Code’s twelve “operative provisions” that a prospective EU arms supplying country is to apply when considering a sale. However, only four operative provisions hold any real potential to preclude a transaction [...] Of these four, two are after-the-fact of an agreement if not the actual transfer. [These would no doubt be the ones guaranteeing line-item veto of arms transfers and the application of dual use restrictions on sales to police and miltary—a far easier restriction to bypass than the American one—JRM]

Objections to international weapons transfers are a topic that often unite liberals, leftist anti-imperialists, and paleoconservatives. Much of the early rancor suffered by the Clinton Administration was directed as his perceived "softness" towards Beijing—especially in view of his 1992 campaign accusations that the first Bush Administration was excessively tolerant of Beijing's human rights abuses. The "anti-globalists" also had the same rebuke for the Clintons. One alleged that easy arms sales worsened the national security landscape for the USA, while the other argued that the USA was abetting a massive human-rights violator.

Arms exports are a real problem for both reasons; in a way, they're a win-win situation for the producer, because if China's weapons were confined to Soviet designs it would be difficult to justify the vast sums spent on things like the F-22 or the F-35 Joint Strike Fighter. This is rather a trivial observation, however; despite the perverse incentives in the weapons industry, arms manufacturers are not necessarily evil people. They invariably find rational justifications for doing what serves their best interests, just like everyone else, and it is really up to legislatures or multilateral organizations to restrain them. The efforts of the latter to do precisely that are never adequate, and presumably never will be—even when the national government itself owns the defense industry, as is often the case (e.g., Russia, Argentina, France). I daresay the most plausible improvement would be a massive dose of public skepticism about the motives of any hawkish initiative.

Personally, I think it's amazing the EU arms embargo lasted as long as it did, in view of the lucrative potentials. It lasted in large measure, I think, because it was comparatively porous and had not interfered with any significant weapons transfer. Partly this is because so many European weapons manufacturers own subsidiaries in other countries that produce under license. The EU had its own reasons for not wishing to export the Rafale or Eurofighter Typhoon to all comers, but other items are produced under license in, say, Brazil or China itself. Moreover, I've heard allegations that US weapons sales to China surpass those of the EU, but am awaiting evidence or context. This GAO testimony references $312 million in US licensing agreements to China between 1991 and 1998, but more recent citations are welcome.