The United States and Europe are in a high-tech dispute. Their conflict lies in determining the difference between a LCD computer monitor and a flat screen television, and a WTO decision may wind up turning on the meaning of “for.”

But first, some background. The 1996 Information Technology Agreement (ITA), a plurilateral agreement adopted under the auspices of the WTO by the world’s major IT-producing nations, lowered those members’ MFN tariffs on information technology products to zero by 2000. Since 1996, another 42 WTO members have joined the original 29 signatories.

Yesterday, the United States, Japan, and Taiwan filed a request for a WTO dispute settlement panel to review the European Union’s compliance with the ITA. In the USTR’s words:

The EU in the past several years has adopted a series of measures that resulted in new duties on imports of specific high-tech products – cable boxes that can access the internet, flat panel computer monitors, and certain computer printers that can also scan, fax and/or copy… These products were included in the ITA… However, the EU claims it can now charge duties on these products simply because they incorporate technologies or features that did not exist when the ITA was concluded.

Of course, the European Commission sees it differently:

The EU, as required by WTO law, bases its customs classification exclusively on the objective characteristics of the products. Where changes in technology have given a product multiple functions – for example, a digital photo camera that also records large amounts of high-quality video – then these products in many cases are objectively different products falling outside of the original product categories covered by the ITA and are classified as such by the EU and others. The US claims this is a violation of the ITA. But both the spirit and explicit provisions in the ITA make it clear that extension to new products to reflect technological change would not be automatic, but based on periodic review by signatories…

Is it a LCD monitor or a flat screen TV? The ITA gives duty-free treatment to computer monitors, not to monitors for consumer electronics such as TV or DVD players. What the US claims are LCD computer monitors are in fact screens equipped with a Digital Visual Interface to allow use with consumer electronics such as DVD players. They are therefore properly classified as video monitors and not covered by the ITA. Incidentally, the classification of such products by US customs is similar to EU practice.

It’s a little unusual to see a WTO dispute about product classifications – usually conflicts revolve around how to calculate duties, the eligibility of safeguard mechanisms’ application, etc. Why can’t we just match a Harmonized System code from the text of the ITA to the EU tariff schedule and make sure the the latter number lies below the former?

First, the Information Technology Agreement was negotiated under the HS1996 product classification scheme. High-tech products have certainly evolved by leaps and bounds since then (should we just say an iPod is a CD player for tariff purposes?) and the HS2007 revisions were dedicated to information technology and communication products. Unfortunately, it is difficult to translate HS1996 tariff agreements into HS2007 tariff schedules:

The WCO Members agreed as a primary goal of the third HS review to conduct an overhaul of the provisions in the technology area in the HS2007 amendment…

In order to assess the impact of HS2007, and to serve as guidelines in transposing the schedules of concessions, the ITA participants asked the Secretariat to prepare a model list in HS2007 through a technical transposition which, like the methodology used for schedules of concessions, maintains the actual product coverage of the new list strictly identical to the original one. However, it goes without saying that several of the above-mentioned divergences in classification would not be solved through this technical exercise.

If ITA participants decide to strictly adhere to the original product coverage, the list in HS2007 cannot take advantage of the improved HS structure on IT products. In many cases, the new HS2007 subheading cannot be directly included in ITA lists because these subheadings normally combine previous “ITA” with some “non-ITA” subheadings. In order to exclude these non-ITA parts, many ex-outs and complicated descriptions need to be introduced by the Secretariat in the HS2007 model list, even though those non-ITA parts sometimes consist of only a minor part of the subheading and represent a very small amount of trade.

But more importantly, the Information Technology Agreement didn’t even use HS1996 codes in many cases!

[P]roducts were specified in two attachments of the Ministerial Declaration: Attachment A and Attachment B. Attachment A consists of two lists of categories of products legally defined by their HS1996 codes. Attachment B is a list of legal product descriptions without reference to their HS codes; restrictions on these products shall be liberalized “wherever they are classified”. Although these lists have provided a good guidance in terms of product coverage, there are still some ambiguities due to the lack of clear HS classifications…

[F]or the products listed in Attachment B and a number of items in Section 2 of Attachment A, consensus was reached only on the textual description of the products, but not on the corresponding HS codes. The ITA participants need to designate national codes based on their own interpretations and classifications.

A WTO committee that was supposed to add new products to the original coverage never reached any agreements.

Flat panel displays fall into Attachment B, which also specifically excludes TVs:

Flat panel displays (including LCD, Electro Luminescence, Plasma and other technologies) for products falling within this agreement, and parts thereof… The agreement does not, therefore, cover televisions, including high definition televisions.

So the LCD monitor needs to be for an automatic data processing machine, as computers are known under the ITA. And the European Commission’s description of the US tariff schedule seems to be right. LCD computer monitors entering the US under tariff line 8528.61 are duty-free, as they are projectors “of a kind solely or principally used in an automatic data processing system of heading 8471.” But tariff line 8528.69.50, for “other projectors, color, with a flat panel screen, display diagonal exceeding 34.29cm” applies a tariff of 5%. Flat panel televisions (8528.72.72) also face a 5% duty.

How will the WTO dispute panel evaluate the difference between a LCD monitor and a flat panel TV? Will the panel have to think hard about the meaning of “for”? I leave any further analysis to the good folks at the International Economic Law and Policy Blog, who may actually be qualified to predict where this case is going.

It seems unavoidable that product innovations will outpace their regulatory classification, especially in trade agreements that must be negotiated between governments. Are trade conflicts resulting from that lag equally unavoidable?

Here’s an exercise for an undergraduate course in international economics: What’s wrong with this graph from a Fast Company special report on China in Africa?

Graphing each country’s foreign exchange reserves, the text says “China still has a huge war chest for African deals and, unlike the US, doesn’t make demands for transparency or human rights.”

[HT: Alex Gadzala]

Here’s an exercise for an undergraduate course in international economics: What’s wrong with this graph from a Fast Company special report on China in Africa?

Graphing each country’s foreign exchange reserves, the text says “China still has a huge war chest for African deals and, unlike the US, doesn’t make demands for transparency or human rights.”

[HT: Alex Gadzala]

Here’s an exercise for an undergraduate course in international economics: What’s wrong with this graph from a Fast Company special report on China in Africa?

Graphing each country’s foreign exchange reserves, the text says “China still has a huge war chest for African deals and, unlike the US, doesn’t make demands for transparency or human rights.”

[HT: Alex Gadzala]

I very much doubt that the empirical evidence supports Dan Mitchell on this:

Indeed, that [higher corporate taxation] is why American companies almost always become the subsidiary rather than the parent when there is a cross-border merger.

Almost always?!? UNCTAD’s stats for 2006 M&A say that US entities made $171.3b in purchases and $172.2b in sales.

GATT Article XXIV (which is supposed to discipline preferential trade agreements) 8(b):

A free-trade area shall be understood to mean a group of two or more customs territories in which the duties and other restrictive regulations of commerce (except, where necessary, those permitted under Articles XI, XII, XIII, XIV, XV and XX) are eliminated on substantially all the trade between the constituent territories in products originating in such territories.

In practice:

Two views of how to interpret the ‘substantially all trade’ provision have crystallised. The quantitative approach favours a statistical benchmark on the proportion of trade covered – for example, 90 percent of all existing trade between the parties. The qualitative approach argues that no sector (or at least no major sector) should be excluded from RTA trade liberalisation…

[I]n the Trade and Development Cooperation Agreement concluded between the EC and South Africa… ‘substantially all trade’ was interpreted to mean an average of 90 percent of all items currently traded between the two countries. The inclusion of the word ‘average’ permits the use of an asymmetrical interpretation… approximately 94 percent of South African exports were covered versus 86 percent of EC exports.

But of course, that’s a perverse method of calculation. It overestimates liberalisation in the same way that the trade-weighted average tariff measure underestimates protection. Measuring “substantially all” by existing trade volumes stacks the deck against meaningful liberalisation:

A basic dilemma facing EU negotiators of these FTAs is that, according to their negotiating mandate, they must not undermine the finely tuned border protection of the CAP and the Common Fisheries Policy. At the same time, they must ensure that the agreement is compatible with Article XXIV… The European Union seeks to resolve this dilemma by interpreting WTO rules as requiring free trade to be established on 90% of the total bilateral trade flows. Since EU tariffs on most industrial products are zero or very low (exceptions are, for example, clothing and motor vehicles) the European Union has little difficulty in liberalizing imports of all, or practically all, industrial products. Also, since imports of agricultural products and fisheries are limited by (sometimes prohibitive) border protection they account for only a small proportion of existing total imports from the partner country. As a result, the European Union is able to make a sufficient contribution to the fulfillment of the 90% criteria by fully liberalizing imports of manufactured goods but, as shown in Table 6, only around 60% of its imports of agricultural products. Similar calculations, it is argued by the European Union, also enables the partner country to protect sensitive industrial and agricultural sectors of its economy while remaining within the EU’s interpretation of requirements of Article XXIV.

Australia has been pushing for WTO members to agree to a definition of “substantially all trade” at the Doha round, though I suspect they’re not making much progress.

GATT Article XXIV (which is supposed to discipline preferential trade agreements) 8(b):

A free-trade area shall be understood to mean a group of two or more customs territories in which the duties and other restrictive regulations of commerce (except, where necessary, those permitted under Articles XI, XII, XIII, XIV, XV and XX) are eliminated on substantially all the trade between the constituent territories in products originating in such territories.

In practice:

Two views of how to interpret the ‘substantially all trade’ provision have crystallised. The quantitative approach favours a statistical benchmark on the proportion of trade covered – for example, 90 percent of all existing trade between the parties. The qualitative approach argues that no sector (or at least no major sector) should be excluded from RTA trade liberalisation…

[I]n the Trade and Development Cooperation Agreement concluded between the EC and South Africa… ‘substantially all trade’ was interpreted to mean an average of 90 percent of all items currently traded between the two countries. The inclusion of the word ‘average’ permits the use of an asymmetrical interpretation… approximately 94 percent of South African exports were covered versus 86 percent of EC exports.

But of course, that’s a perverse method of calculation. It overestimates liberalisation in the same way that the trade-weighted average tariff measure underestimates protection. Measuring “substantially all” by existing trade volumes stacks the deck against meaningful liberalisation:

A basic dilemma facing EU negotiators of these FTAs is that, according to their negotiating mandate, they must not undermine the finely tuned border protection of the CAP and the Common Fisheries Policy. At the same time, they must ensure that the agreement is compatible with Article XXIV… The European Union seeks to resolve this dilemma by interpreting WTO rules as requiring free trade to be established on 90% of the total bilateral trade flows. Since EU tariffs on most industrial products are zero or very low (exceptions are, for example, clothing and motor vehicles) the European Union has little difficulty in liberalizing imports of all, or practically all, industrial products. Also, since imports of agricultural products and fisheries are limited by (sometimes prohibitive) border protection they account for only a small proportion of existing total imports from the partner country. As a result, the European Union is able to make a sufficient contribution to the fulfillment of the 90% criteria by fully liberalizing imports of manufactured goods but, as shown in Table 6, only around 60% of its imports of agricultural products. Similar calculations, it is argued by the European Union, also enables the partner country to protect sensitive industrial and agricultural sectors of its economy while remaining within the EU’s interpretation of requirements of Article XXIV.

Australia has been pushing for WTO members to agree to a definition of “substantially all trade” at the Doha round, though I suspect they’re not making much progress.

Here’s a typical story on rice prices:

Experts say rice prices are rising because of a mix of irrational panic, weather problems – typhoons in the Philippines, a cyclone in Bangladesh, flooding in Indonesia and Vietnam – and an overall reduction in the amount of land dedicated to rice farming. There are also strong suspicions of hoarding, something that the Thai commerce minister recently encouraged before reversing himself.

And here’s a story from Steve H. Hanke and David Ranson:

The most recent rice price spike is partially the result of countries such as India and Egypt imposing restrictions and bans on exports, plus the desire of other governments including the Philippines, the world’s largest rice importer, to bulk up their stockpiles. But the blame for the long-term trend of higher prices should be placed upon those who’ve delivered a weak U.S. dollar…

But determining what’s behind the escalation in commodity prices involves a principle to which economists universally pay lip service but in practice often ignore or forget: the distinction between nominal prices and relative (real) prices. Nominal prices of commodities are determined by the value of the currency used to stipulate them, while relative prices of commodities are determined by supply and demand — scarcity, glut and other “real” causative factors…

Officials should stop wringing their hands over sky-high rice prices caused by alleged changes in rice’s supply-demand fundamentals, and politicians should refrain from pointing accusative fingers at speculators and hoarders. The rice-price problem is a weak dollar problem. Until the dollar strengthens, the nominal dollar prices of rice and other commodities will remain elevated.

The claim that the rice-price problem is a weak dollar problem implies that (1) buyers of rice using other currencies should be unaffected and (2) the price of rice relative to other imports should be unchanged for US consumers.

Here’s Thai rice, up 140% over the last year:

And here’s the dollar-baht exchange rate over the last year:

Looks like Thai rice exporters are enjoying an increase in real income (and Thai rice importers a real decline).

Moreover, domestic prices are up:

And it is not just the international market that is in crisis. From October, 2007, to March, 2008, domestic rice prices increased by 38 percent in Bangladesh, 18 percent in India, and more than
30 percent in the Philippines.

This isn’t just a nominal illusion.

Next, Hanke and Ranson offer some chart about gold prices that’s difficult to interpret:

Since none of my friends earn their income in gold, I have no idea why this matters. The nominal price of rice and gas are up over the last few months; the nominal prices of other imports (such as my holiday in Ibiza) are also a bit expensive, but their prices haven’t risen nearly as fast. When you divide one nominal price by another, you get a relative (real) price. And look, the real price of rice is up over the last year!

To see that this isn’t a dollar problem, look at the graph provided by Asia Times‘ anonymous columnist “Spengler,” who argues the opposite of Hanke and Ranson but is also wrong:

There are long-term reasons for food prices to rise, but the unprecedented spike in grain prices during the past year stems from the weakness of the American dollar…

The chart below shows the price of 100 pounds of rice against the euro’s parity against the US dollar during the past 12 months. The regression fit is 90%. There is an even tighter relationship between the price of rice and the price of oil, another store of value against dollar depreciation.

As the chart makes clear, the ascent of the cost of rice to $24 from $10 per hundredweight over the past year tracks the declining value of the American dollar. The link between the declining parity of the US unit and the rising price of commodities, including oil as well as rice and other wares, is indisputable.

Allow me to try my hand at disputing. Take a look at the left vertical axis. The dollar price of rice more than doubled over the last year. Check out the right vertical axis. The euro rose at most 20% against the dollar (and you inverted the axis label, oops). The price of rice in euros is up quite a bit, r-squared be damned. Find me a currency that has doubled against the dollar in the last year.

I suspect that the correlations mentioned by Hanke and Ranson are equally meaningless:

For example, during the long period of the dollar’s strength, from the end of 1979 to the end of 2001, gold suffered a cumulative decline of 40%, while rice experienced a cumulative decline of 52%. During the subsequent six years from the end of 2001 to the end of 2007, the price of gold rose 191% and the price of rice rose 127%.

Goldbugs can eat more rice then. But this doesn’t explain the (real) crisis in rice.

David Ricardo:

Under a system of perfectly free commerce, each country naturally devotes its capital and labour to such employments as are most beneficial to each. This pursuit of individual advantage is admirably connected with the universal good of the whole… It is this principle which determines that wine shall be made in France and Portugal, that corn shall be grown in America and Poland, and that hardware and other goods shall be manufactured in England…

If Portugal had no commercial connexion with other countries, instead of employing a great part of her capital and industry in the production of wines, with which she purchases for her own use the cloth and hardware of other countries, she would be obliged to devote a part of that capital to the manufacture of those commodities, which she would thus obtain probably inferior in quality as well as quantity.

The quantity of wine which she shall give in exchange for the cloth of England, is not determined by the respective quantities of labour devoted to the production of each, as it would be, if both commodities were manufactured in England, or both in Portugal.

England may be so circumstanced, that to produce the cloth may require the labour of 100 men for one year; and if she attempted to make the wine, it might require the labour of 120 men for the same time. England would therefore find it her interest to import wine, and to purchase it by the exportation of cloth.

John Nye:

Prior to the late 1600s, the British drank plenty of wine, mostly French, a little Spanish, but virtually nothing from Portugal. The wars of 1689-1713 gave the Portuguese allies the opportunity of ten lifetimes. Beginning in 1703 a treaty was signed granting Portugal access to British markets for their wines—generally of a much lower quality than those of France, and often needing to be fortified with brandy or spirits in order to keep from going bad. The Methuen Treaty (as it was known) promised that Portuguese tariffs would always be at least a third lower than those of other nations, most especially France.

Of course, most of the Portuguese wine trade was dominated by British ships, merchants, and even vintners working in Iberia. The end of hostilities between Britain and France was seen as a grave threat to all these British interests, and vigorous lobbying by brewers, distillers, and the Anglo-Portuguese merchants stopped attempts to return to the period of open trade with the French.

Trade theorists have learned their lesson: use Greek letters rather than real world examples!

Hat tip: EconTalk.

David Albouy recently posted a new (Feb ’08) version of his critique of Acemoglu, Johnson, and Robinson’s 2001 AER paper that used European settler mortality rates as an instrument for modern institutional quality.

I haven’t had time to read through the back-and-forth exchanges to form my own judgment. Has any third party summarised the dispute and commented? Albouy is revising and resubmitting at AER, so at least some of his claims must be plausible, huh?