In recent weeks, U.S. Trade Representative Michael Froman has begun making outlandish claims about past U.S. trade agreements. These claims are not supported by the official U.S. government trade data. The Office of the U.S. Trade Representative’s (USTR) recent assertions that the North American Free Trade Agreement (NAFTA) has led to a U.S. trade surplus with Mexico and Canada and that the U.S.-Korea Free Trade Agreement (FTA) has increased U.S. manufacturing exports to Korea have been met with incredulity. These pacts’ recent anniversaries have spotlighted how the trade pact model on which the Trans-Pacific Partnership (TPP) is premised has led to massive trade deficits.
The premise that NAFTA would improve our trade balance was the basis for NAFTA proponents’ promises that the pact would create U.S. jobs. Many of the same government and industry sources made the same claims to sell the 2011 U.S.-Korea FTA. These pacts’ dismal outcomes – slow or even negative export growth, rising imports and burgeoning trade deficits – are intensifying congressional opposition to Fast Track authority for the TPP.
Rather than altering the trade agreement model to avoid repeating these outcomes, USTR appears intent on trying to change the data. To generate the outlandish claims about NAFTA and the Korea FTA, USTR employs a smorgasbord of data tricks to look out for in Froman’s testimony Thursday before the House Ways and Means Committee:
USTR’s Biggest Distortion: Counting Foreign-Made “Transshipped” Products as U.S. Exports
USTR’s primary data distortion is the decision not to use the official U.S. government trade data provided by the U.S. International Trade Commission (USITC).[i] Instead, USTR cites data that include what are called “re-exports.” These are goods made abroad that are simply shipped through the United States en route to a final destination. (The USTR figures would include as U.S. exports goods taken off a truck from Canada in California’s Port of Long Beach then shipped to their final destination in Korea, or goods shipped from China, unloaded in a California port and trucked to Mexico.) Each month, USITC removes re-exports, which do not support U.S. production jobs, from the raw data gathered by the Census Bureau.[ii] But USTR uses the uncorrected data, inflating the actual U.S. export figures.
- Using the official USITC data, U.S. export growth to countries with which we do not have FTAs has been 30 percent faster than to our FTA partners over the past decade.[iii]
- The USITC data show U.S. average monthly goods exports to Korea are down 11 percent, imports from Korea have increased and the U.S. average monthly trade deficit with Korea has swelled 47 percent since the enactment of the Korea FTA.[iv] The total U.S. trade deficit with Korea under the FTA’s second year is projected to be $8.6 billion higher than in the year before the deal.[v] Using the administration’s current export-to-job ratio, this drop in net exports represents the loss of more than 46,000 U.S. jobs.[vi] However since the FTA, foreign-made re-exports passing through the United States en route to Korea are up 13 percent on a monthly average basis.[vii] By counting these foreign goods as U.S. exports, USTR artificially diminishes the dramatic drop in actual U.S. exports to Korea, and errantly claims gains in some sectors.
- Using the USITC data, the 2013 U.S. goods trade balance with NAFTA nations was a deficit of $177 billion. The combined U.S. goods and services deficit with Mexico and Canada rose (in real, inflation-adjusted terms) from $9.7 billion in 1993 to $139.3 billion in 2012 (the year of comparison used by USTR).[viii] This NAFTA deficit increase of $129.5 billion, or 1,330 percent, represents hundreds of thousands of lost U.S. jobs.[ix] But adding re-exports has had an increasingly distortionary effect on the true NAFTA deficit, allowing NAFTA proponents to make the 2013 NAFTA goods deficit of $177 billion look less than half as large. By incorporating re-exports, USTR claims in recent press materials: “U.S. total goods and private services trade balance with Canada countries (sic) shifted from a deficit of $2.9 billion in 1993 to roughly balance in 2012 (surplus of $37 million).” But after removing re-exports and adjusting for inflation, the actual total U.S. goods and services trade deficit with Canada increased from $16.9 billion in 1993 to $49.1 billion in 2012. That’s a deficit increase of $32.2 billion, or 191 percent. Similarly, USTR claims: “U.S. total goods and private services trade balance with Mexico countries shifted from a surplus of $4.6 billion in 1993 to a deficit of $49.4 billion in 2012.” But after removing re-exports and adjusting for inflation, the actual total U.S. goods and services trade deficit with Mexico changed from a $7.2 billion surplus in 1993 to a $90.1 billion deficit in 2012. That’s a $97.3 billion decline in the U.S. goods and services trade balance with Mexico.
We Still Have Big Deficits Without Fossil Fuels (And Corn Doesn’t Explain Korea Export Crash)
Despite USTR’s claim that our NAFTA deficit is all about fuel imports, the share of the U.S. NAFTA goods trade deficit that is comprised of petroleum, petroleum products and natural gas has declined under NAFTA, from 77 percent in 1993 to 53 percent in 2013, as we have faced a surge of imported manufactured and agricultural goods.[x] Even if one removes all of these “oil” categories from the balance, the remaining 2013 NAFTA goods trade deficit was $82.9 billion. The combined NAFTA goods and services deficit in 2012 minus oil was $38.3 billion.
Similarly, with respect to the Korea FTA, USTR claims“[O]ur trade balance has been affected by decreases in corn and fossil fuel exports, changes that are due to the U.S. drought in 2012 and change in Korea’s energy mix.”[xi] But even discounting both corn and fossil fuels, U.S. monthly exports to Korea still fell under the FTA, and the monthly trade deficit with Korea still ballooned.[xii] USTR claims that corn and fossil fuels explain the entirety of the export downfall largely by using an ill-suited 2011 versus 2013 timeframe that omits 10 months of available data and relies on a less relevant pre-FTA baseline. Usage of this less accurate timeframe produces a greater drop in corn and fossil fuel exports, and a smaller decline in exports of all other goods, than has actually occurred under the FTA when comparing the year immediately preceding the FTA with the full set of available post-FTA data. It is not surprising that the dismal FTA record remains without these products, given that of the 15 U.S. sectors that export the most to Korea, 11 of them have experienced export declines under the FTA.[xiii] No product-specific anomalies can explain away what has been a broad-based downfall of U.S. exports to Korea since the pact went into effect.
Not Adjusting for Inflation Counts Increased Prices as an Increase in U.S. Exports
USTR also inflates U.S. exports to Korea by failing to adjust for price inflation. For instance, in its recent Korea FTA news release, USTR claims: “In the two years that this landmark agreement has been in effect … exports of U.S. manufactured goods to Korea have increased … Made-in-America manufactured goods still grew their sales in Korea by 3 percent.”[xiv] Simply adjusting for inflation alone completely erases USTR’s claim of growth in exports of U.S. manufactured goods to Korea under the FTA. That is, even if one includes the distortion of re-exports and uses USTR’s timeframe, U.S. exports to Korea of manufactured goods fell slightly under the FTA after properly accounting for price increases.[xv] If one removes the re-exports (i.e., uses the official USITC data) and looks at the actual months that the FTA has been in effect, U.S. monthly exports to Korea of manufactured goods have fallen 5 percent on average relative to the year before the deal took effect. The United States has lost an average of more than $150 million each month in manufactured goods exports to Korea under the FTA. Manufacturing sectors that provide critical shares of U.S. exports to Korea, such as machinery and computers/electronics, have experienced steep export declines under the FTA (11 percent and 12 percent respectively). In contrast, of the four critical manufacturing sectors that have seen increases in average monthly exports to Korea under the FTA, none has experienced an increase of greater than 2 percent.[xvi]
Cherry-Picking Small-Dollar Winning Sectors, Omitting Major Losers to Distract from Net Losses
In its Korea FTA press release, USTR claims: “U.S. exports of a wide range of agricultural products have seen significant gains. … There were also dramatic increases in U.S. exports of key agricultural products that benefit from reduced tariffs under KORUS, including dairy, wine, beer, soybean oil, fruits and nuts, among many others.”[xvii] But the losses in U.S. meat exports to Korea under the pact alone nearly cancel out the combined export gains for all agricultural sectors that USTR touts as winners (a monthly average loss of $20.1 million in meat exports versus a combined $24.7 million monthly average gain in exports of dairy, wine, beer, soybean oil, fruits and nuts).[xviii]Average monthly exports of all U.S. agricultural products to Korea have fallen 41 percent under the FTA in comparison to the year before the deal. Ignoring this overall result, USTR singles out fruit as a winning agricultural sector under the FTA. But U.S. monthly average exports to Korea of all fruits have increased by just $312,120 under the FTA. This 1 percent increase could hardly be described as “dramatic.” USTR also highlights wine, but U.S. monthly average exports of wine to Korea have increased by just $370,378 under the FTA.[xix] The amount of wine sold in an average six minutes in the United States is worth more ($402,415) than the gain in U.S. wine exports to Korea in an average month under the Korea FTA.[xx]
Such paltry gains pale in comparison to the more than $20 million lost on average under each month of the FTA in U.S. exports to Korea of meat – one of the sectors that the administration promised would be among the biggest beneficiaries of the Korea deal.[xxi] Compared with the exports that would have been achieved at the pre-FTA average monthly level, U.S. meat producers have lost a combined $442 million in poultry, pork and beef exports to Korea in the first 22 months of the FTA.[xxii]Since the FTA, U.S. average monthly exports of poultry to Korea have fallen 39 percent below the pre-FTA monthly average. U.S. poultry exports to Korea have been lower than the pre-FTA monthly level in every single month since the FTA’s implementation. U.S. average monthly exports of pork to Korea since the FTA have fallen 34 percent below the pre-FTA monthly average, and U.S. average monthly exports of beef to Korea have fallen 6 percent below the pre-FTA monthly average.[xxiii]
Omissions and Data Tricks to Hide Massive Auto Sector Deficit Growth Under the Korea FTA
The USTR data on U.S. automotive trade with Korea under the FTA is based on a series of tricks. USTR claims: “Since the Korea agreement went into effect, U.S. exports to Korea are up for our manufactured goods, including autos … overall U.S. passenger vehicle exports to Korea increased 80 percent compared to 2011, and sales of ‘Detroit 3’ vehicles are up 40 percent.”[xxiv] In fact, exports to Korea of U.S.-produced Fords, Chryslers and General Motors vehicles increased by just 3,400 vehicles from 2011 to 2013.[xxv] But given that pre-FTA exports of “Detroit 3” vehicles was also tiny – 8,252 vehicles – USTR can express the small increase of 3,400 cars as a “40 percent” gain. Meanwhile, 125,000 more Korean-produced Hyundais and Kias were imported and sold in the United States in 2013 (after the FTA) than in 2011 (before the FTA), when Hyundai and Kia imports already topped 1.1 million vehicles.[xxvi]
And USTR’s claim of an “80 percent” rise in passenger vehicle exports, in addition to being inflated by increases in re-exports and prices, omits the export of auto parts, which constitute the majority of the value of U.S. automotive exports to Korea. U.S. average monthly exports of auto parts to Korea have fallen 12 percent under the FTA, offsetting much of the rise in passenger vehicle exports.[xxvii] After including auto parts, excluding foreign-made re-exports, using the more FTA-relevant timeframe and adjusting for inflation, U.S. average monthly automotive exports to Korea have increased by only 12 percent under the FTA, while average monthly automotive imports from Korea have risen by 19 percent.
The disparity is even starker in dollar terms: While U.S. average monthly automotive exports to Korea under the FTA have been $12 million higher than the pre-FTA monthly average, average monthly automotive imports from Korea have soared by $263 million under the deal. The tiny gains in U.S. exports have been swamped by a surge in auto imports from Korea that the administration promised would not occur because of its additional FTA auto sector measure negotiated in 2011. In January 2014, monthly automotive imports from Korea topped $2 billion for the first time on record. The post-FTA flood of automotive imports has provoked a 19 percent increase in the average monthly U.S. auto trade deficit with Korea.[xxviii]
Using a Selective Time Frame to Measure the Outcomes of the Korea FTA
Rather than compare the post-Korea-FTA period to the 12 months prior to the FTA’s implementation (i.e., April 2011 through March 2012), USTR uses calendar year 2011 as a baseline. This means that USTR omits data from the three months immediately prior to the FTA’s 2012 implementation (January through March 2012) and replaces it with data from the same three months in 2011. This difference matters, since U.S. exports to Korea in the first three months of 2011 were 9 percent lower than in the first three months of 2012, giving USTR a lower baseline of comparison that makes the downfall in U.S. exports look less severe than if using the three most recent pre-FTA months.[xxix] In addition, USTR uses only calendar year 2013 to assess the FTA’s record, omitting 10 months of available post-FTA data (April through December 2012 and January 2014). While a comparison between 2011 and 2013 could serve as a second-best approximation in the absence of more precise data, the more FTA-relevant monthly data is readily available.
[i] USITC data can be found at U.S. International Trade Commission, “Interactive Tariff and Trade DataWeb.” Available at: https://dataweb.usitc.gov/.
[iii] U.S. International Trade Commission, “Interactive Tariff and Trade DataWeb,” accessed February 11, 2014. Available at: https://dataweb.usitc.gov/. The statistic is a comparison of the average annual growth rate of the combined inflation-adjusted exports of all non-FTA partner countries versus that of all FTA partner countries from 2004 through 2013 (adjustments have been made to account for the changes in these two categories as non-FTA partners have become FTA partners). All data in this memo is inflation-adjusted according to the CPI-U-RS index of the U.S. Bureau of Labor Statistics (which provides indices up through 2012) and the online inflation calculator of the U.S. Bureau of Labor of Statistics (which provides an approximate index for 2013). U.S. Bureau of Labor Statistics, “Consumer Price Index Research Series Using Current Methods (CPI-U-RS),” U.S. Department of Labor, updated March 29, 2013. Available at: https://www.bls.gov/cpi/cpiursai1978_2012.pdf. U.S. Bureau of Labor Statistics, “CPI Inflation Calculator,” U.S. Department of Labor, accessed March 10, 2014. Available at: https://www.bls.gov/data/inflation_calculator.htm.
[iv] In this paragraph and throughout, figures concerning average monthly trade levels with Korea compare data from the year before the FTA’s implementation and from the 22 post-implementation months for which data are available. U.S. International Trade Commission, “Interactive Tariff and Trade DataWeb,” accessed March 10, 2014. Available at: https://dataweb.usitc.gov/.
[v] The projection for export losses under the FTA’s first two years assumes that trends during the FTA’s first 22 months continue for the remaining two months for which data are not yet available. U.S. International Trade Commission, “Interactive Tariff and Trade DataWeb,” accessed March 10, 2014. Available at: https://dataweb.usitc.gov/.
[x] Trade in petroleum, petroleum products and natural gas is defined as NAICS 2111 and 3241 for data since 1997 – when NAICS replaced the SIC classification system – and SIC 131, 291, 295, and 299 for data before 1997.
[xii] Corn is defined as NAICS 111150 and fossil fuels are defined as NAICS 211111, 211112, 212112 and 212113. U.S. International Trade Commission, “Interactive Tariff and Trade DataWeb,” accessed March 10, 2014. Available at: https://dataweb.usitc.gov/.
[xviii] “Meat” includes beef (defined as SITC 011), pork (defined as SITC 0122, 0161 and 0175) and poultry (defined as SITC 0123 and 0174). Dairy is defined as NAICS 2111511, 311512, 311513, 311514 and 311520. Wine is defined as NAICS 312130. Beer is defined as NAICS 312120. Soybean oil is defined as NAICS 311222 and 311224. Fruits are defined as NAICS 11310, 11320, 111331, 111332, 111333, 111334 and 111339. Nuts are defined as NAICS 111335 and 111992. U.S. International Trade Commission, “Interactive Tariff and Trade DataWeb,” accessed March 21, 2014. Available at: https://dataweb.usitc.gov/.
[xix] Fruits are defined as NAICS 11310, 11320, 111331, 111332, 111333, 111334 and 111339. Wine is defined as NAICS 312130. U.S. International Trade Commission, “Interactive Tariff and Trade DataWeb,” accessed March 10, 2014. Available at: https://dataweb.usitc.gov/.
[xxi] “Meat” includes beef (defined as SITC 011), pork (defined as SITC 0122, 0161 and 0175) and poultry (defined as SITC 0123 and 0174). Dairy is defined as NAICS 2111511, 311512, 311513, 311514 and 311520. Wine is defined as NAICS 312130. Beer is defined as NAICS 312120. Soybean oil is defined as NAICS 311222 and 311224. Fruits are defined as NAICS 11310, 11320, 111331, 111332, 111333, 111334 and 111339. Nuts are defined as NAICS 111335 and 111992. U.S. International Trade Commission, “Interactive Tariff and Trade DataWeb,” accessed March 21, 2014. Available at: https://dataweb.usitc.gov/.
[xxvii] Passenger vehicles are defined as code 300 and 301 in the one-digit End Use classification system, while auto parts are defined as 302. U.S. International Trade Commission, “Interactive Tariff and Trade DataWeb,” accessed March 10, 2014. Available at: https://dataweb.usitc.gov/.
[xxviii] Total automotive exports and imports are defined as code 3 in the one-digit End Use classification system. U.S. International Trade Commission, “Interactive Tariff and Trade DataWeb,” accessed March 10, 2014. Available at: https://dataweb.usitc.gov/.