IP trade and the 2015 Economic Report of the President

The 2015 Economic Report of the President came out earlier this week. While it contains a number of statistics and policy proposals (see summary), I want to highlight just one small section on intellectual property:

Box 7-1: Trade in Ideas

In 2013, U.S. companies paid $39 billion in royalties and licensing fees to foreign companies, and were paid $129 billion by foreign companies seeking access to intellectual property held in the United States. While this “trade in ideas” represents just 14.6 percent of all U.S. trade in services, it generates 40 percent of our $225 billion services trade surplus. Figure 7-i shows the level of imports and exports in 2013 for each of the four major categories of trade in intellectual property. Roughly two-thirds of this trade is intra-firm, with a greater share of this intra-company trade occurring in the trademark and franchise fees category (76 percent) than for industrial processes (69 percent), software (58 percent), or audio-visual materials (42 percent).

Trade in ideas is partly influenced by differences in countries’ intellectual property laws; as such, harmonizing the international treatment of intellectual property rights has become an important, and sometimes controversial, aspect of international trade negotiations. For example, the WTO Agreement on Trade Related Aspects of Intellectual Property Rights established minimum standards for various forms of intellectual property protection. Several economic studies, such as papers by Branstetter, Fisman, and Foley (2006) and Cockburn, Lanjouw, and Schankerman (2014), suggest that stronger patent protection in destination countries does promote outbound technology transfer, both within and between firms.

One reason that trade in intellectual property can be controversial is that ideas are non-rival goods that can be used by many parties at the same time, with little or no incremental cost per user. This feature of intellectual property also creates challenges for measuring international technology transfer because it implies that the location of an idea, which determines the direction of trade flows, is somewhat arbitrary. To compound that problem, there is no obvious market price for many intra-company transactions, so both the magnitude and direction of intra-company trade in ideas may reflect corporate tax and legal strategies, as much as they do business or economic realities.

All of these complications can produce some unusual outcomes in the trade statistics. For example, U.S. intellectual property exports to Bermuda were $3 billion in 2013, with 98 percent of that trade occurring between affiliated companies, a trade that largely occurs for tax reasons rather than economic reasons, as discussed in Chapter 5 of this Report. These intellectual property exports are about two-thirds the size of Bermuda’s $4.5 billion GDP. In the same year, U.S. intellectual property exports to France, whose GDP is 600 times larger than Bermuda’s, totaled $3.4 billion, with only 42 percent transpiring between related companies. Lipsey (2010) shows that foreign affiliates of U.S. multinationals located in a variety of low-tax countries report unusually high levels of intangible assets relative to both employees and physical capital.

While it is difficult to estimate the size of any measurement bias created by geographic reallocation of intellectual property within multinational firms, it is possible to say something about the likely impact on trade statistics. In particular, transfers of intellectual capital abroad at below-market rates and intra-company pricing that shifts income outside the United States will lead the official statistics to underestimate the true size of the U.S. services trade surplus–that is, what would be observed under competitive market prices or in a tax neutral environment. For example, the true value of intellectual property exports in Figure 7-i may be higher, and the value of imports lower, particularly for trade in ideas related to trademark and franchise fees, where the share of intra-company transactions is highest. This type of bias would also make U.S. companies that trade in intellectual property appear less productive, by artificially lowering their revenues and inflating their costs. The continued growth of intra-company cross-border trade within large multinationals suggests that these measurement challenges will only grow in importance for both tax authorities and government statisticians.

And, unfortunately, the measurement challenges are even greater than this description notes. This part of the Report only covers trade in formal intellectual property. That is far from a complete inventory of ideas or how ideas are traded. Nothing about non-IP protected ideas. Nothing about non-technological innovations. Nothing about other forms of intangible capital. Nothing about the value of ideas embedded in advanced technology goods and services. Nothing about non-monetized and non-market mechanisms of information/knowledge exchanges. In other words, the real story is much more complicated and thus more difficult to get our policy response right.