This is the accessible text file for GAO report number GAO-06-231 
entitled 'U.S.-China Trade: Eliminating Nonmarket Economy Methodology 
Would Lower Antidumping Duties for Some Chinese Companies' which was 
released on January 10, 2006. 

This text file was formatted by the U.S. Government Accountability 
Office (GAO) to be accessible to users with visual impairments, as part 
of a longer term project to improve GAO products' accessibility. Every 
attempt has been made to maintain the structural and data integrity of 
the original printed product. Accessibility features, such as text 
descriptions of tables, consecutively numbered footnotes placed at the 
end of the file, and the text of agency comment letters, are provided 
but may not exactly duplicate the presentation or format of the printed 
version. The portable document format (PDF) file is an exact electronic 
replica of the printed version. We welcome your feedback. Please E-mail 
your comments regarding the contents or accessibility features of this 
document to Webmaster@gao.gov. 

This is a work of the U.S. government and is not subject to copyright 
protection in the United States. It may be reproduced and distributed 
in its entirety without further permission from GAO. Because this work 
may contain copyrighted images or other material, permission from the 
copyright holder may be necessary if you wish to reproduce this 
material separately. 

Report to Congressional Committees: 

January 2006: 

U.S.-China Trade: 

Eliminating Nonmarket Economy Methodology Would Lower Antidumping 
Duties for Some Chinese Companies: 

[Hyperlink, http://www.gao.gov/cgi-bin/getrpt?GAO-06-231]: 

GAO Highlights: 

Highlights of GAO-06-231, a report to congressional committees: 

Why GAO Did This Study: 

U.S. companies adversely affected by unfair imports may seek a number 
of relief measures, including antidumping (AD) duties. The Department 
of Commerce (Commerce) classifies China as a nonmarket economy (NME) 
and uses a special methodology that is commonly believed to produce AD 
duty rates that are higher than those applied to market economies. 
Commerce may stop applying its NME methodology if it finds that China 
warrants designation as a market economy. 

In light of increased concern about China’s trade practices, the 
conference report on fiscal year 2004 appropriations requested that GAO 
review efforts by U.S. government agencies responsible for ensuring 
free and fair trade with that country. In this report, the last in a 
series, GAO (1) explains the NME methodology, (2) analyzes AD duties 
applied to China and compares them with duties applied to market 
economies, and (3) explains circumstances in which the United States 
would stop applying its NME methodology to China and evaluates the 
potential impact of such a step. 

Commerce agreed with our findings, commenting that our report provides 
timely and helpful information on the NME methodology and its 
application to China. 

What GAO Found: 

Commerce’s methodology for calculating AD duties on nonmarket economy 
products differs from its market economy approach in that (1) since NME 
prices are unreliable, it uses price information from surrogate 
countries, like India, to construct the value of the imported products 
and (2) it limits eligibility for individual rates to companies that 
show their export activities are not subject to government control. 
Companies that do not meet the criteria or do not participate in 
Commerce investigations receive “country-wide” rates. 

China has been the most frequent target of U.S. AD actions. On 25 
occasions, Commerce has applied duties to the same product from both 
China and one or more market economy. China (NME) duties were over 20 
percentage points higher than those applied to market economies, on 
average. This is because average China country-wide rates were over 60 
points higher than comparable market economy rates. Individual China 
company rates were similar to those assigned to market economy 
companies, on average. 

Comparison of China and Market Economy Antidumping Rates for 25 
Products (1985-2004): 

[See PDF for image] 

[End of figure] 

Commerce can declare China a market economy if the country meets 
certain criteria, thus ending the use of surrogate price information 
and country-wide rates in China AD actions. These changes would have a 
mixed impact. Duties would likely decline for Chinese companies not 
assigned individual rates. Individual company rates would likely 
diverge, with those that do not cooperate with Commerce receiving rates 
that are substantially higher than those that do cooperate. In any 
case, it appears that the actual trade impact of the NME methodology 
will decline as the portion of total export trade conducted by Chinese 
companies assigned individual rates increases and as the country-wide 
rates that largely account for the comparatively high average rates 
applied to China decline in importance. 

What GAO Recommends: 

www.gao.gov/cgi-bin/getrpt?GAO-06-231. 

To view the full product, including the scope and methodology, click on 
the link above.  To view the database on antidumping case information 
against Chinese companies, click on the following link: 
http://www.gao.gov/cgi-bin/getrpt?GAO-06-652SP.  For more information, 
contact Loren Yager at (202) 512-4347 or yagerl@gao.gov.

[End of section] 

Contents: 

Letter: 

Results in Brief: 

Background: 

Commerce Employs a Special Methodology to Calculate China AD Duties: 

Commerce Has Applied AD Duties to China Frequently at Varied Rates: 

Ceasing to Apply NME Methodology Would Have Mixed Results: 

Concluding Observations: 

Agency Comments and Our Evaluation: 

Appendixes: 

Appendix I: Scope and Methodology: 

Appendix II: Additional Information on Duty Rates Applied to China and 
Market Economy Countries: 

Appendix III: Regression Analysis Results: 

Appendix IV: Comments from the Department of Commerce: 

GAO Comments: 

Appendix V: GAO Contact and Staff Acknowledgments: 

Tables: 

Table 1: Products Affected by AD Duty Orders against China, 1980-2004: 

Table 2: Comparison of China, Market Economy AD Duty Rates, 
Methodological Changes, and Potential Effects: 

Table 3: Summary Data on China AD Duty Orders, 1980-2004: 

Table 4: Average Duty Rates for 68 Orders against China, 1980-2004: 

Table 5: Average Duty Rates for 25 Products with Orders against both 
China and Market Economies, 1980-2004: 

Table 6: Results of Multivariate Regression Analysis of Duty Rates on 
Explanatory Variables for China, Country-Wide Rates, and Year: 

Table 7: Results of Multivariate Regression Analysis of Duty Rates on 
Explanatory Variables for China, Country-Wide Rates, and Individual 
Products: 

Table 8: Results of Multivariate Regression Analysis of Duty Rates on 
Explanatory Variables for China, Country-Wide Rates, Adverse 
Inferences, Other Facts Available, and Individual Products: 

Figures: 

Figure 1: Different Approaches to Determining AD Duty Rates: 

Figure 2: U.S. AD Orders in Place by Country as of December 31, 2004: 

Figure 3: Results of AD Petitions, 1980-2004: 

Figure 4: AD Duty Rates Applied against Imported Polyethylene Retail 
Carrier Bags (2004): 

Figure 5: Differences between Overall Average Duty Rates--China and 
Market Economies, 1980-2004: 

Figure 6: Differences between China Country-Wide and Market Economy 
All- Others Duty Rates, 1980-2004: 

Figure 7: Differences between Average Individual Company Duty Rates-- 
China and Market Economy Countries, 1980-2004: 

Figure 8: Average Number of Individual Rates per Case per Year, All 
Orders against China, 1980-2004: 

Figure 9: Average Country-Wide and Individual Rates, All Orders against 
China, 1980-2004: 

Abbreviations: 

AD: antidumping: 

HTS: Harmonized Tariff Schedule: 

ITC: U.S. International Trade Commission: 

NME: nonmarket economy: 

WTO: World Trade Organization: 

Letter January 10, 2006: 

The Honorable Richard C. Shelby: 
Chairman: 
The Honorable Barbara A. Mikulski: 
Ranking Minority Member: 
Subcommittee on Commerce, Justice, and Science: 
Committee on Appropriations: 
United States Senate: 

The Honorable Frank R. Wolf: 
Chairman: 
The Honorable Alan B. Mollohan: 
Ranking Minority Member: 
Subcommittee on Science, State, Justice, and 
Commerce and Related Agencies: 
Committee on Appropriations: 
House of Representatives: 

Imports from China have grown rapidly over the last decade, from a 
total value of about $42 billion in 1995 to over $196 billion in 
2004.[Footnote 1] While the prices of these Chinese goods are often 
lower than U.S. prices and, therefore, benefit consumers, this growth 
has presented a major challenge for U.S. producers that compete with 
Chinese products in the U.S. market. Some U.S. companies adversely 
affected by this growth have alleged that Chinese success in the U.S. 
market has come partly as a result of unfair trade practices. 

U.S. companies that are adversely affected by unfair imports from China 
(or other countries) may avail themselves of a number of relief 
measures, including antidumping (AD) duties. The United States has 
classified China as a "nonmarket economy" (NME) country since 
1981[Footnote 2] and employs a special NME methodology to calculate AD 
duties on unfairly traded products from that country. This methodology 
is commonly believed to result in duty rates that are significantly 
higher than those applied to market economy countries. 

In light of increased concern about China's trade practices, the 
conference report on fiscal year 2004 appropriations legislation 
requested that GAO monitor the efforts of U.S. government agencies 
responsible for ensuring free and fair trade with that 
country.[Footnote 3] In subsequent discussions with staff from the 
House Appropriations Committee's Subcommittee on Science, State, 
Justice, and Commerce and Related Agencies, we agreed to provide a 
number of reports on import relief mechanisms and the manner in which 
these mechanisms have been applied to China. To date, we have issued 
three such reports, focusing on textile safeguards, safeguards 
applicable to other products, and countervailing duties.[Footnote 4] 

This fourth and final report on China import relief mechanisms focuses 
on AD duties. In this report, we: 

* explain the special methodology that the United States employs to 
calculate AD duties on products from China and other NME countries, 

* analyze the application of AD duties to China over the last 25 years 
and compare the duty rates applied to Chinese products with the duty 
rates applied to products from market economy countries, and: 

* explain the circumstances in which the United States would stop using 
its NME methodology to calculate AD duties on Chinese products and 
evaluate the potential impact of this step. 

To conduct our review, we examined applicable U.S. laws and regulations 
and World Trade Organization (WTO) agreements, including relevant 
portions of the agreement through which China acceded to WTO membership 
in 2001. We reviewed scholarly literature and consulted with trade and 
legal policy experts from the U.S. government, private sector trade 
associations, consulting and law firms, and academic institutions, as 
well as representatives of the WTO, the government of China, and other 
governments. In order to analyze U.S. application of AD duties to China 
and compare the duties applied to China with those applied to market 
economy countries, we used information from the Department of Commerce 
(Commerce) and the U.S. International Trade Commission (ITC), including 
notices of Commerce determinations appearing in the Federal Register, 
to construct a database on all U.S. antidumping investigations from 
1980 through 2004. We verified this database to the official sources. 
Our analyses focused on the 68 AD duty orders that Commerce issued 
against Chinese products during this period and especially on the 25 
cases in which Commerce also imposed duties on the same products from 
market economy countries.[Footnote 5] We performed additional 
(multivariate regression) analyses to determine the extent to which 
duty rate variations could be attributed to differences between China 
and these other countries, or to other factors, such as the type of 
product involved. Appendix I contains a detailed description of our 
scope and methodology. 

We conducted our work from June 2005 through December 2005 in 
accordance with generally accepted government auditing standards. 

Results in Brief: 

Commerce's special methodology for calculating the AD duties that it 
applies to China and other NME countries differs from its usual (market 
economy) approach in two key respects. First, since prices in NME 
countries do not reliably reflect the fair value of the merchandise, 
Commerce uses price information from surrogate countries (like India) 
to construct the value of NME products--and thus provide an appropriate 
basis for calculating AD duty rates--rather than relying entirely on 
information from the exporting country itself. Second, Commerce 
requires NME companies to demonstrate that their export activities are 
not subject to government control in order to be considered eligible 
for individually determined duty rates, rather than considering all 
companies eligible for such rates, as it does in market economy cases. 
NME companies that do not meet these criteria, or do not participate in 
Commerce's investigations receive "country-wide" rates. 

Over the last 25 years, the United States has applied AD duties against 
China more often than against any other country. On 25 occasions, 
Commerce applied duties to the same products from both China and at 
least one market economy country. The duty rates assigned in these 
cases varied greatly. On average, however, the rates applied to China 
were over 20 percentage points higher than those applied to market 
economy countries. This difference is attributable primarily to the 
comparatively high country-wide duty rates applied to Chinese companies 
not eligible for individual rates. These country-wide rates averaged 
about 98 percent--over 60 percentage points higher than the average 
duty rates assigned to market economy companies not receiving 
individual rates. In contrast, when Commerce calculated individual 
rates for Chinese companies, these rates were not substantially 
different, on average, from those assigned to individual market economy 
companies. 

Commerce has administrative authority to declare China a market 
economy, or find individual Chinese industries to be "market-oriented" 
in character--provided that China overall or individual Chinese 
industries meet certain criteria. Such a declaration would end 
application of the NME methodology to China, in whole or in part. This 
would (1) eliminate country-wide duty rates against China and (2) 
eliminate use of surrogate country information to calculate AD duty 
rates on Chinese products. These changes would have a mixed impact. 
Eliminating country-wide rates would likely reduce duty rates applied 
to companies not receiving individual rates. Individually determined 
rates would likely diverge into two distinct groups, with companies 
that do not cooperate in Commerce investigations receiving rates that 
are substantially higher than those that do cooperate. The impact of 
applying Chinese price information would likely vary by industry, and 
AD rates applied against China would continue to vary widely, both 
within and among cases. However, it appears that the significance of 
the NME country-wide rates is declining as more Chinese companies 
receive individual rates, although data that would permit 
quantification of the potential trade impact of these changes is not 
available. This suggests that the trade significance of the NME 
methodology now applied to China will likewise decline over time. 

Commerce provided written comments on a draft of this report, which are 
reprinted in appendix IV. Overall, Commerce agreed with our findings. 
In addition, Commerce, the Department of Homeland Security, and the 
Office of the U.S. Trade Representative, provided technical comments. 
We took these comments into consideration and made revisions throughout 
the report as appropriate to make it more accurate and clear. 

Background: 

Dumping refers to a type of international price discrimination wherein 
a foreign company sells merchandise in a given export market (for 
example, the United States) at prices that are lower than the prices 
that the company charges in its home market or other export markets. 
When this occurs, and when the imports have been found to materially 
injure, or threaten to materially injure, U.S. producers, U.S. law 
permits application of antidumping duties to offset the price advantage 
enjoyed by the imported product.[Footnote 6] 

Any domestic industry that believes it is suffering material injury, or 
is threatened with material injury, as a result of dumping by foreign 
companies may file a petition requesting imposition of AD duties. 
Interested domestic industries file petitions simultaneously with 
Commerce and ITC. If Commerce determines that the petitioning parties 
meet certain eligibility requirements,[Footnote 7] ITC determines 
whether the domestic industry has suffered material injury as a result 
of the alleged dumping (or is threatened with material injury). 
[Footnote 8] While ITC is completing its work, Commerce conducts an 
investigation to establish the duty rates, if any, that should be 
applied. 

To determine the duty rates to apply in an antidumping investigation, 
Commerce identifies (1) the foreign product's export price entering the 
U.S. market and (2) its "normal value."Commerce then compares these 
prices to determine whether--and by how much--the product's export 
price is less than its normal value. AD duty rates are based on these 
differences, which are called dumping margins.[Footnote 9] 

To establish a product's export price, Commerce generally refers to the 
prices charged in actual sales of that product to purchasers in the 
United States.[Footnote 10] To establish its normal value, Commerce 
generally refers to the prices charged for the product in the exporting 
company's home market. In the event that the product is not sold in the 
exporter's home market, Commerce may refer to prices charged for the 
product in another export market or construct a normal value based on 
costs of production in the exporting country, together with selling, 
general and administrative expenses, and profit.[Footnote 11] The two 
agencies make preliminary and, after additional investigation, final 
determinations as to whether injury has occurred (ITC) and the size of 
the duty, if any, that should be imposed (Commerce). When warranted, 
Commerce issues "duty orders" instructing Customs and Border Protection 
to apply duties against imported products from the countries under 
investigation. Both ITC and Commerce publish their decisions in the 
Federal Register. 

Since AD duties address unfair pricing practices, and pricing decisions 
are generally made by individual companies, Commerce generally 
calculates and assigns AD duty rates on an individual company basis. As 
a result, AD investigations generally produce a number of individually 
determined, company-specific rates, reflecting differences in the 
extent to which companies have dumped their products--that is, exported 
them at less: 

than their normal value.[Footnote 12] In addition, AD duty orders also 
generally specify a duty rate for other companies that have not been 
assigned an individually determined rate. 

In principle, Commerce bases its AD duty determinations on information 
obtained from interested parties--including foreign producers and 
exporters. Commerce obtains needed information from foreign companies 
by sending them questionnaires and following up with additional 
questions, as needed, and with on-site visits.[Footnote 13] 

However, both U.S. law and WTO rules recognize that, in some cases, 
officials charged with completing these investigations will be unable 
to obtain sufficient information. In such cases, Commerce officials 
apply facts available to complete their duty determinations.[Footnote 
14] This may include secondary information, subject to corroboration 
from independent sources. 

Moreover, if Commerce finds that an interested party, such as a foreign 
company under investigation, "has failed to cooperate by not acting to 
the best of its ability to comply with a request for information" then, 
in selecting among the facts available, Commerce may apply an inference 
that is adverse to the interests of that party. In applying adverse 
inferences, Commerce can use (among other things) information contained 
in the petition filed by the domestic industry seeking imposition of AD 
duties, the results of a prior review or determination in the case, or 
any other information placed on the record.[Footnote 15] 

This authority provides an incentive for foreign companies to provide 
the information that Commerce needs to complete its work. For example, 
in a 1993 case that involved two Brazilian companies, one company 
attempted to cooperate in the investigation but nonetheless was unable 
to provide the information that Commerce needed, while the other 
declined to provide any information at all. Commerce used facts 
available to determine that the first company should be subject to a 
duty rate of 42 percent. For the second company, Commerce selected 
adverse inferences from among the facts available and applied these to 
calculate a duty rate of 109 percent.[Footnote 16] 

Commerce Employs a Special Methodology to Calculate China AD Duties: 

The methodology that Commerce employs in NME cases differs from 
Commerce's usual (market economy) approach in two key ways. First, 
rather than rely entirely on information from the exporting country 
itself to establish a product's normal value, Commerce uses price 
information from surrogate countries to construct these values. Second, 
rather than consider all companies eligible for individually determined 
duty rates, Commerce requires NME companies to meet certain criteria to 
be considered eligible for such rates. Commerce generally employs 
different approaches to calculate duty rates for companies that do and 
do not meet these criteria. 

AD Calculations for NME Products Employ Third Country Information: 

In AD investigations involving products from NME countries, U.S. law 
requires Commerce to use a special methodology to calculate duty rates 
in view of the absence of meaningful home market prices and information 
on production costs. When a product from China or another NME country 
is the target of an AD investigation, Commerce officials use price 
information and financial data from an appropriate market economy 
country to construct a normal value for the product under 
investigation.[Footnote 17] India is the most commonly used surrogate 
for China. 

To apply this methodology, Commerce (1) identifies and quantifies the 
factors of production (e.g., various raw materials) used by the NME 
producers, (2) identifies market prices for each factor in a surrogate 
country; (3) multiplies volume times cost for each factor; and (4) adds 
the results, together with a reasonable margin for selling, general and 
administrative expenses, and profit (based on surrogate country 
financial data), to produce a constructed normal value.[Footnote 18] 
The dumping margin--and consequently the AD duty rate--is then 
determined by comparing this normal value with the NME company's export 
price to the United States. 

NME Companies Must Meet Certain Criteria to Be Considered Eligible for 
Individual Duty Rates: 

While all companies from market economy countries are eligible for 
individually determined or weighted average AD duty rates, companies 
from China and other NME countries must pass a separate rates test to 
be eligible for such rates. This test requires NME companies to meet 
two closely related criteria: they must demonstrate that their export 
activities are free from government control both in law and in 
fact.[Footnote 19] To provide a basis for deciding whether companies 
meet these criteria, Commerce requires these companies to submit 
information regarding: 

* whether there are restrictive stipulations associated with an 
individual exporter's business and export licenses, 

* any legislative enactments decentralizing control of companies, 

* any other formal measures decentralizing government control of 
companies, 

* whether export prices are set by or subject to approval by the 
government, 

* whether the company has authority to negotiate and sign contracts, 

* whether the company has autonomy in selecting its management, and: 

* whether the company retains the proceeds of its export sales and 
makes independent decisions regarding disposition of profits or 
financing of losses.[Footnote 20] 

Commerce Employs Different Approaches to Determine Duty Rates for 
Eligible and Ineligible NME Companies: 

As shown in figure 1, Commerce uses fundamentally different approaches 
to calculate duty rates to be applied against companies that do and do 
not pass the separate rates test. 

Figure 1: Different Approaches to Determining AD Duty Rates: 

[See PDF for image] 

[End of figure] 

Eligible NME Companies Receive Individually Determined or Weighted 
Average Duty Rates: 

As shown in figure 1, Commerce treats companies from China and other 
NME countries that pass Commerce's separate rates test like market 
economy countries when assigning duty rates. When practical, Commerce 
fully investigates and establishes individually determined duty rates 
for each eligible NME company, just as it does for each market economy 
company. To the extent that fully investigated NME companies cooperate 
with Commerce, they receive rates based on the information that they 
provide. As explained in the background section of this report, 
Commerce uses facts available, and may use adverse inferences, to 
calculate duty rates when the companies under investigation cannot or 
will not provide the information that Commerce needs. 

In both NME and market economy cases, Commerce may limit the number of 
companies it fully investigates when it is faced with a large number of 
companies. In such situations, Commerce generally calculates individual 
rates for the companies that account for the largest volume of the 
subject merchandise.[Footnote 21] In market economy cases, Commerce 
then calculates a weighted average of these rates and applies the 
resulting "all others" rate to companies that it has not fully 
investigated.[Footnote 22] Commerce does not routinely calculate 
weighted average duty rates in NME cases. However, when the number of 
NME companies eligible for individually determined rates exceeds the 
number that Commerce can fully investigate, Commerce calculates a 
weighted average rate and informs Customs of the companies entitled to 
this rate.[Footnote 23] 

Other NME Companies Receive Country-Wide Duty Rates: 

In cases involving China or other NME countries, Commerce calculates a 
country-wide duty rate for companies that could not (or did not attempt 
to) pass Commerce's separate rates test. In NME cases, Commerce assumes 
that all exporters and producers of a given product are subject to 
common government control and that all of these companies should, 
therefore, be subject to a single country-wide duty rate. Commerce 
begins its NME antidumping investigations by requesting information 
from the government of the country in question and from known producers 
and exporters. If Commerce cannot identify all relevant producers and 
exporters, or if one or more of the identified companies refuses to 
cooperate in the investigation, Commerce relies on adverse inferences 
to calculate a country-wide rate. Commerce then instructs Customs to 
apply the country-wide rate against shipments from any company other 
than those specifically listed as eligible for an individually 
determined or weighted average rate.[Footnote 24] 

Commerce Has Applied AD Duties to China Frequently at Varied Rates: 

Over the last 25 years, the United States has applied AD duties against 
Chinese products more often than against products from any other 
country. While AD duty rates have varied widely, on average the rates 
assigned to Chinese products have been higher than the rates assigned 
to the same products from market economy countries. We found that this 
is attributable primarily to the comparatively high country-wide rates 
applied to Chinese companies not eligible for individually determined 
or weighted average rates. When Commerce has calculated rates for 
individual Chinese companies, the average rates assigned to these 
companies have not been substantially different from those assigned to 
market economy companies. 

China Has Been the Most Frequent Target of U.S. Antidumping Orders: 

Over the last 25 years, Commerce has both considered and actually 
applied AD duties against China more often than against any other 
country. From 1980 through 2004, Commerce processed 1,046 AD petitions 
and issued 455 AD duty orders. One hundred and ten of these petitions 
(11 percent) and 68 of these orders (15 percent) focused on China--both 
are the largest number against any U.S. trade partner.[Footnote 25] 

The number of orders applied to China varied from year to year. For 
example, Commerce issued no AD duty orders against China in 1998 but 
issued 9 in 2003. Commerce had 272 orders in place as of December 31, 
2004. Fifty-five of these (20 percent) apply to China. As figure 2 
shows, this is also the highest percentage of any country. As shown in 
table 1, these duty orders have targeted a wide variety of products but 
have been concentrated in chemicals and plastics, metal products, and 
agricultural products. 

Figure 2: U.S. AD Orders in Place by Country as of December 31, 2004: 

[See PDF for image] 

Note: From 1980 through 2004 Commerce applied 68 orders against China. 
Thirteen of these were revoked, leaving 55 in place as of the end of 
2004. 

[End of figure] 

Table 1: Products Affected by AD Duty Orders against China, 1980-2004: 

Type of product: Chemicals, plastics, pharmaceuticals; 
Examples of affected products: Barium chloride; Polyethylene retail 
carrier bags; Bulk aspirin; 
Number of orders: 26. 

Type of product: Steel, other metals; 
Examples of affected products: Carbon steel butt-weld pipe fittings; 
Chrome-plated lug nuts; Pure magnesium; 
Number of orders: 20. 

Type of product: Agricultural products; 
Examples of affected products: Crawfish; Garlic; Honey; 
Number of orders: 5. 

Type of product: Other products; 
Examples of affected products: Brake rotors; Hand tools; Cotton shop 
towels; Automotive replacement glass windshields; Folding gift boxes; 
Number of orders: 17. 

Source: GAO AD database. 

Note: Product categories based on the Harmonized Tariff Schedule of the 
United States, Annotated. See appendix I for more information. 

[End of table] 

Moreover, petitions for AD duties against China have resulted in 
application of duties more often than those against other countries. As 
shown in figure 3, 62 percent of the petitions filed against China over 
the last 25 years resulted in duty orders, while the equivalent figure 
for all countries was about 43 percent.[Footnote 26] 

Figure 3: Results of AD Petitions, 1980-2004: 

[See PDF for image] 

Note: Petitions may not result in an AD order for several reasons. For 
example, if ITC does not find the domestic industry to be materially 
injured, or threatened with material injury, or if Commerce does not 
find that dumping has occurred, then the case is terminated. Also, 
Commerce may suspend a case if the United States reaches an agreement 
with the foreign government that would eliminate the impact of dumping 
(e.g., restrictions on exports). In addition, the petitioners may 
decide to withdraw their petition before ITC and Commerce have 
completed their work. 

[End of figure] 

China, Market Economy Rates Varied Widely: 

Over this 25 year period, Commerce issued duty orders against the same 
products from China and at least one market economy country on 25 
occasions.[Footnote 27] In 18 of these cases, Commerce calculated 
individual rates for companies from China and at least one market 
economy country. Fifteen of these cases involved more than one market 
economy country. In all, the orders applying to these 25 products 
contained a total of 243 individual, weighted average, and country-wide 
duty rates. Appendix II provides detailed information on the rates 
applied in each of these cases, as well as another 43 cases that we 
identified wherein Commerce applied duty rates to China but not to any 
market economy country. 

These rates varied a great deal--both among the orders applied to 
different products and within the orders applied to the same products. 
Overall, these duty rates varied from zero to 218 percent for China and 
from zero to about 244 percent for market economy countries. Figure 4 
shows the extent to which duty rates applied to a single product can 
vary.[Footnote 28] 

Figure 4: AD Duty Rates Applied against Imported Polyethylene Retail 
Carrier Bags (2004): 

[See PDF for image] 

Note: In addition to the five individual Chinese companies shown above, 
two Chinese companies received de minimus rates and were excluded from 
the order. 

[End of figure] 

Overall, China Rates Were Higher Than Market Economy Rates: 

The average AD duty rates imposed on Chinese (NME) exporters over the 
last 25 years have been significantly higher than those imposed on 
market economy exporters of the same products. Taking all rates into 
consideration (including those calculated for individual companies, 
weighted averages of these rates, and country-wide rates applied to 
China) the average rate applied to Chinese companies in the 25 cases we 
examined was about 67 percent--over 20 percentage points higher than 
the average rate of 44 percent applied to market economy companies. As 
figure 5 shows, the overall average rates applied against China were 
higher for 18 of the 25 products in which there were AD orders against 
both China and at least one market economy. 

Figure 5: Differences between Overall Average Duty Rates--China and 
Market Economies, 1980-2004: 

[See PDF for image] 

[End of figure] 

Difference in Average Rates Due Primarily to High China Country-Wide 
Rates: 

The difference between average China and average market economy duty 
rates was due primarily to the fact that the NME country-wide duty 
rates applied to China were substantially higher than the comparable 
all-others duty rates applied against market economy countries. In 
contrast, the individually determined duty rates assigned to Chinese 
companies in these cases were not substantially different, on average, 
from the individually determined rates assigned to market economy 
companies. 

Country-Wide Duty Rates Substantially Higher Than Market Economy All- 
Others Rates: 

On average, the country-wide rates applied to China in these 25 cases 
were substantially higher than the comparable all-others rates applied 
to market economy countries. The country-wide duty rates applied 
against China averaged about 98 percent--over 60 percentage points 
higher than the average 37 percent all-others duty rate applied to 
market economy exporters of the same products. Figure 6 shows that the 
China country-wide rate was higher than the market economy all-others 
rate in 21 of 25 cases. As explained below, this difference was due 
largely to the use of different methodologies to calculate country-wide 
and all-others rates. 

Figure 6: Differences between China Country-Wide and Market Economy All-
Others Duty Rates, 1980-2004: 

[See PDF for image] 

Note: In several cases, Commerce issued orders against several market 
economy countries, and we calculated an average all-others rate for all 
of the affected market economy countries. 

[End of figure] 

Country-wide rates were nearly always equal to or higher than the 
highest individually determined rate applied to any Chinese company, 
due to application of adverse inferences.[Footnote 29] According to 
Commerce, NME country governments themselves have never provided the 
information that Commerce needs to establish an appropriate country-
wide duty rate. In addition, Commerce officials stated that, in most 
cases, participating NME companies have accounted for only a portion of 
known exports to the U.S. market from their country, indicating that 
others had not come forward. In most cases, therefore, Commerce has 
used adverse inferences to determine country-wide rates. For example, 
in its investigation of carbazole violet pigment, Commerce assigned 
three fully investigated Chinese companies individually determined 
rates of about 6, 27, and 45 percent. However, since other known 
Chinese producers did not respond to Commerce's request for 
information, Commerce used adverse inferences to determine that all 
other Chinese producers should be subject to an NME country-wide rate 
of about 218 percent.[Footnote 30] 

In contrast, the comparable market economy all-others rates were lower 
than the highest individual company rates assigned in any given case 
(if more than one other individual rate was assigned).[Footnote 31] 
This is because, as discussed earlier, Commerce generally calculates 
all-others rates by averaging individually determined rates--excluding 
those derived entirely through application of facts available and those 
that are de minimis or zero. With regard to carbazole violet pigment, 
for example, Commerce investigated not only China but also India. 
Commerce assigned two fully investigated Indian companies rates of 
about 10 and 50 percent and weight-averaged these rates to determine 
that shipments from all other Indian producers should be subject to a 
duty rate of about 27 percent. 

Individual Company Rates in China and Market Economy Countries Not 
Substantially Different on Average: 

On average, there was little difference between the individually 
determined rates applied to companies from China and those applied to 
market economy companies. The average individually determined rate 
applied to Chinese companies in these cases was 53 percent--a little 
less than the average rate of 55 percent applied to market economy 
companies.[Footnote 32] The median rate for Chinese companies was 42 
percent--the same as the median rate for market economy companies. 
Figure 7 displays the average individual company rates assigned to 
Chinese and market economy companies in the 18 cases in which Commerce 
assigned individual rates to both. As the figure shows, the rates 
assigned to Chinese companies were higher than the market economy rates 
in ten of these cases and lower in the other eight. 

Figure 7: Differences between Average Individual Company Duty Rates-- 
China and Market Economy Countries, 1980-2004: 

[See PDF for image] 

[End of figure] 

Our statistical analyses provided additional support for the importance 
of the country-wide rates in accounting for the overall difference 
between the duty rates applied to China and to market economy 
countries. Using multivariate regression analysis, we found that a 
number of variables, such as the type of product involved, accounted 
for some of the overall variation in duty rates. However, after 
accounting for the China country-wide rates there was no statistically 
significant difference between the duty rates applied to China and 
those applied to market economy countries. As explained in more detail 
in appendix III, we found essentially the same results when we expanded 
our analyses to include data on AD actions against NMEs other than 
China.[Footnote 33] 

Ceasing to Apply NME Methodology Would Have Mixed Results: 

In certain circumstances, Commerce may stop using its NME methodology 
in China cases--and thus begin applying its market economy methodology 
to determine AD duty rates against that country. Such a step would lead 
to important changes in the methods that Commerce employs to determine 
China AD duty rates and in the duty orders resulting from these 
proceedings. These changes would have mixed results. Eliminating 
country-wide duty rates would likely reduce duty levels for Chinese 
companies that are not assigned individually determined rates. 
Individually determined rates would likely diverge into two distinct 
groups, with companies that do not cooperate in Commerce investigations 
receiving rates that are substantially higher than those assigned to 
companies that do cooperate. The impact of applying Chinese price 
information to calculate the normal value of Chinese products would 
likely vary by industry. In any case, rates would continue to vary 
widely based on the circumstances of each case. While trade data that 
would permit analysis of the potential trade impact of these changes is 
not available, it appears that the trade significance of country-wide 
duty rates is declining. 

Commerce May Stop Applying NME Methodology to China in Certain 
Circumstances: 

Commerce has administrative authority to reclassify China and other NME 
countries as market economies or individual NME country industries as 
market-oriented in character. Such reclassifications would end 
Commerce's authority to apply its NME methodology to such countries or 
industries. Also, China's WTO accession agreement specifies that 
members may apply third-country information to calculate AD duty rates 
against that country, but this provision expires in 2016. 

Commerce Has Administrative Authority to Change China's NME Status: 

Commerce has the authority to reclassify China as a market economy 
country, in whole or in part. As we explained in more detail in a prior 
report,[Footnote 34] U.S. trade law authorizes Commerce to determine 
whether countries should be accorded NME or market economy status and 
specifies a number of criteria for Commerce to apply in making such 
determinations.[Footnote 35] Countries classified as NMEs may ask for a 
review of their status at any time.[Footnote 36] China has actively 
sought market economy status among its trading partners, and a number 
of them have designated China as a market economy. However, Commerce 
informed us that Chinese officials have not yet officially requested a 
determination as to whether their country merits reclassification under 
the criteria specified in U.S. law. In April 2004, the United States 
and China established a Structural Issues Working Group under the 
auspices of the U.S.-China Joint Commission on Commerce and Trade. This 
group is examining structural and operational issues related to China's 
economy that may give rise to bilateral trade frictions, including 
issues related to China's desire to be classified as a market 
economy.[Footnote 37] 

Commerce also has the authority to designate individual NME industries 
as market oriented in character, but has denied all such requests to 
date. Commerce determined in a 1992 case against China that, short of 
finding that an entire country merits designation as a market economy, 
it could find specific industries within such countries to be market 
oriented in character.[Footnote 38] Commerce officials noted that on 
several occasions Chinese industries responding to antidumping duty 
petitions have requested designation as market-oriented industries. To 
date, Commerce has denied such requests--primarily on the grounds that 
the Chinese companies in question submitted information that was 
insufficient or was provided too late in Commerce's process to allow an 
informed decision. 

China's WTO Commitment Allowing the Use of Third-Country Information 
Expires in 2016: 

When joining the WTO, China agreed that other WTO members could use 
third-country information to calculate normal values in antidumping 
actions against Chinese companies. Specifically, China's WTO accession 
agreement provides that in determining price comparability in 
antidumping investigations WTO members may use "a methodology that is 
not based on a strict comparison with domestic prices or costs in 
China."[Footnote 39] However, the accession agreement also specifies 
that this provision will expire 15 years after the date of the 
agreement--that is, by the end of 2016.[Footnote 40] 

After 2016, the ability of WTO members to continue using third-country 
information in AD calculations involving China would be governed by 
generally applicable WTO rules, according to officials at the Office of 
the U.S. Trade Representative. These rules recognize that when dumping 
investigations involve products from a country that "has a complete or 
substantially complete monopoly of its trade and where all domestic 
prices are fixed by the state," importing country authorities may have 
difficulty making the price comparisons through which AD duty rates are 
normally established. In such situations, importing countries may "find 
it necessary to take into account the possibility that a strict 
comparison with domestic prices in such a country may not always be 
appropriate."[Footnote 41] WTO rules do not provide any specific 
guidance about how this provision should be implemented; such decisions 
appear to be left up to individual members. 

Transition to Market Economy Methodology Would Bring Significant 
Procedural Changes: 

Ending application of the NME methodology to China would bring two 
significant procedural changes in AD duty investigations against 
Chinese products. First, such a step would eliminate NME country-wide 
duty rates from China AD orders. Commerce would instead assign an 
individually determined rate to every relevant Chinese producer or 
exporter. If the number of companies involved were too great to allow 
full investigation of all relevant companies, Commerce would apply an 
all-others rate--a weighted average of the individually determined 
rates to all other Chinese companies (excluding those rates based 
entirely on facts available or that are de minimis or zero). However, 
Commerce would retain its authority to use facts available to determine 
the rates that it applies to individual Chinese companies. Second, 
transition to the market economy methodology would end Commerce's use 
of surrogate country information to calculate the normal value of 
Chinese products. Application of the market economy methodology would 
generally require Commerce to set the normal value of Chinese products 
equal to their sales price in China. If the product were not sold in 
China, Commerce could refer to prices charged for the product in 
another export market or construct the product's normal value, or it 
could continue to construct the product's normal value--using factor 
prices from the Chinese companies under investigation rather than from 
surrogate countries. 

Eliminating Country-Wide Rates Would Likely Reduce Duty Levels for 
Companies Not Assigned Individually Determined Rates: 

The elimination of country-wide duty rates against China would likely 
reduce the duty rates applied to some Chinese companies. If Commerce 
applied its market economy approach to China, duty rates for companies 
not receiving individually determined rates would, in most cases, no 
longer be determined by applying facts available. Rather, Commerce 
would, for the most part, determine these rates by averaging the rates 
applied to fully investigated Chinese companies, with some exclusions. 
The default rate for uninvestigated Chinese companies would move, in 
most cases, from being the highest rate found to the average rate found 
among companies that cooperate in Commerce investigations. 

Though not predictive, available evidence suggests that the all-others 
rates that Commerce would apply to China under the market economy 
methodology would be significantly lower than the country-wide rates 
currently applied to that country. As already shown, China country-wide 
rates have generally been significantly higher than the all-others 
rates that Commerce has assigned to market economy sources of the same 
products. As shown in table 2, the average country-wide rate for the 25 
cases in which Commerce assigned duties to both China and one or more 
market economies was 98 percent, while the average market economy all- 
others rate was 37 percent. The average rate assigned to individual 
Chinese companies was 53 percent, and Commerce calculates all-others 
rates by weight averaging individually determined rates, excluding 
those that are derived entirely through application of facts available 
and those that are de minimis or zero. 

Table 2: Comparison of China, Market Economy AD Duty Rates, 
Methodological Changes, and Potential Effects: 

Average China (NME) rates; 
Overall [A]: 67%; 
Group [A]: 98%; (Country-wide); 
Individual rates[B]: All: 53%; 
Individual rates[B]: Cooperative: 51%; 
Individual rates[B]: Adverse inferences: (Rarely applied)[C]. 

Average market economy rates; 
Overall [A]: 44%; 
Group [A]: 37%; (All others); 
Individual rates[B]: All: 55%; 
Individual rates[B]: Cooperative: 16%; 
Individual rates[B]: Adverse inferences: 77%. 

Change from NME to market economy methodology for China companies; 
Overall [A]: Chinese price information replaces surrogate price 
information; 
Group [A]: Country-wide rates eliminated, uninvestigated companies 
receive all others rate; 
Individual rates[B]: All: Separate rates test eliminated. 

Potential effect on average China rates; 
Overall [A]: Effect unknown but likely to vary by industry; 
Group [A]: Rates likely to be significantly lower; 
Individual rates[B]: All: Rates likely to fall into two distinct 
groups; 
Individual rates[B]: Cooperative: Rates likely to be relatively low for 
cooperative companies; 
Individual rates[B]: Adverse inferences: Rates likely to be relatively 
high for uncooperative companies. 

Source: GAO analysis of Commerce data. 

[A] Averages based on 25 products with comparable China and market 
economy cases (1985-2004). 

[B] Averages based on subset of 18 products with comparable China and 
market economy cases in which individual rates were applied. However, 
averages are nearly identical for full group of 25 products. 

[C] Commerce applied adverse inferences only three times, for an 
average rate of 78 percent. 

[End of table] 

Individually Determined Rates Would Vary, Depending Upon Cooperation: 

A simple comparison of the average individually determined duty rates 
calculated under the NME and market economy methodologies suggests that 
a change in methodology would not result in any significant overall 
change in duty rates applied to individual Chinese companies. For the 
comparable cases, individual AD duty rates for Chinese companies 
averaged 53 percent and were not substantially different from 
individual market economy company rates, which averaged 55 percent. 

However, a more detailed examination of the data indicates that the 
impact of a change in methodology on individual Chinese company duty 
rates would depend on the extent to which Commerce applies adverse 
inferences to calculate these rates. The rates assigned to individual 
companies under the market economy methodology fell into two distinct 
groups, depending on whether the companies cooperated with Commerce 
investigations. In the 25 cases that we examined in detail, about half 
of the fully investigated market economy companies cooperated with 
Commerce. On average, Commerce assigned a duty rate of about 17 percent 
to these companies.[Footnote 42] Commerce found the other half of the 
fully investigated companies uncooperative and, therefore, applied 
adverse inferences to determine the duty rates to be applied to these 
companies. On average, Commerce assigned a duty rate of about 77 
percent to these uncooperative market economy companies.[Footnote 43] 

Though not predictive, this suggests that a change from the NME 
methodology for China would result in a significant number of 
(cooperative) companies receiving relatively low rates, while another 
significant group of (uncooperative) companies would receive relatively 
high rates.[Footnote 44] Our regression analysis confirmed the 
importance of adverse inferences as a determinant of variation in duty 
rates. As explained in appendix III, we found that application of 
adverse inferences tends to increase duty rates by a large margin. 

Impact of Applying Chinese Price Information Would Vary By Industry and 
Is Likely to Decline over Time: 

The impact of using Chinese price information on China AD duty rates 
would likely vary from one industry to another under the market economy 
methodology. Chinese prices are widely viewed as distorted to varying 
degrees. Where prices for key inputs are artificially low, relying on 
Chinese price information would produce an artificially low normal 
value. The result would be an AD duty that is lower than would be 
obtained by applying surrogate country input prices. Conversely, where 
Chinese prices are artificially high, AD duty rates may be higher if 
based on Chinese prices. To the extent that Chinese economic reforms 
bring Chinese prices more into line with world markets, the impact of 
abandoning the use of surrogate country information can be expected to 
decline. At any point in time, however, the probable effect of such a 
methodological change in an individual industry investigation would 
depend on the particular facts applying to that industry. The net 
impact of changing the source of price information on overall China 
duty rates cannot be estimated with confidence. 

Duty Rates Will Continue to Display Great Variation: 

Our multivariate regression analyses suggest that, regardless of 
changes in methodology, there will continue to be a great deal of 
variation among the AD duty rates applied to products from China and 
other countries. Our analyses showed that application of country-wide 
duty rates to China largely explained the difference between the 
overall average duty rates applied to China and to market economy 
countries. Eliminating these rates would likely have a substantial 
overall reducing effect on China rates. However, a number of other 
factors, such as the type of product involved, also helped to account 
for differences among rates overall, and these factors will continue to 
have an impact on duty rates, regardless of whether Commerce applies 
country-wide rates to China. Furthermore, even after taking these 
factors into account, our analyses still explained only about half of 
the total variation in duty rates.[Footnote 45] This means that about 
half of the: 

variation in duty rates is attributable either to idiosyncratic factors 
or to systematic factors that we did not capture in any of our 
variables.[Footnote 46] 

Trade Significance of Country-Wide Rates Appears to Be Declining: 

Available evidence suggests that the volume of trade affected by 
country-wide rates is declining and that, consequently, the trade 
impact of China duty orders will in the future depend increasingly on 
the magnitude of the individually determined rates. Commerce officials 
observed that in the early 1980s it was not unusual for China AD duty 
investigations to produce only a country-wide rate. However, as the 
Chinese economy has evolved, individual Chinese companies have become 
more likely to request--and receive--individually determined or 
weighted average rates. Since 1980, Commerce has applied country-wide 
rates alone in only 15 of 68 Chinese AD orders, and the last of these 
occasions was in 1995. The majority of all Chinese AD orders (about 78 
percent), and all such orders issued over the last 10 years, have 
included at least one individual company rate. 

Neither Commerce nor Customs and Border Protection maintain trade data 
that would permit analysis of changes over time in the relative volume 
or value of products imported into the United States under the country-
wide or various individual duty rates listed in AD duty orders. 
However, as figure 8 shows, the average number of Chinese companies 
assigned individually determined rates (or assigned a weighted average 
rate) has been growing, though there continues to be variation from 
year to year. For example, in 2004 Commerce placed duties on six 
Chinese products and in doing so assigned individually determined or 
weighted average rates to 53 Chinese companies. Anecdotal evidence 
suggests that along with this rise in company interest in obtaining 
individual rates has come an increase in the volume of trade covered by 
these rates. For example, in one recent case Commerce fully 
investigated and assigned individually determined rates to four 
companies accounting for more than 90 percent of Chinese exports to the 
U.S. market. Commerce then assigned a weighted average of these rates 
to 9 additional companies, leaving only a very small portion of all 
Chinese exports to be covered by the country-wide rate. [Footnote 47] 

Figure 8: Average Number of Individual Rates per Case per Year, All 
Orders against China, 1980-2004: 

[See PDF for image] 

Notes: This table includes only the companies assigned individually 
determined or weighted average rates in initial AD duty orders. Other 
companies may be assigned such rates in subsequent reviews of these 
orders. The annual figures are the number of individual companies 
granted such rates divided by the number of AD duty orders. Commerce 
did not issue any individual company duty orders against China in 1980, 
1981, 1982, 1985, 1987, 1988, 1989, or 1998. In the tapered roller 
bearings cases, Commerce originally put an order in place in 1987, but 
amended it in 1990. We use the information from the 1990 amendment in 
the above graphic. 

[End of figure] 

Concluding Observations: 

On average, Commerce's application of its NME methodology has produced 
AD duties on Chinese products that are substantially higher than those 
applied to the same products from market economy countries. Changing 
China's NME status--and thus eliminating the application of this 
methodology--would have a variety of impacts. The duty rates applied to 
companies that do not receive individual rates would likely decline. 
Chinese companies that cooperate in Commerce investigations may also 
receive comparatively low rates. However, the impact of these lower 
rates on overall China averages may be offset, to some extent, by 
application of adverse inferences to assign relatively high rates to 
individual Chinese companies that do not cooperate in Commerce 
investigations. 

The net effect of these changes cannot be predicted. Such a prediction 
would require knowledge of price distortions in diverse Chinese 
industries, changes in these distortions over time, pricing decisions 
by Chinese companies in reaction to these changes, and decisions by 
U.S. companies about whether they should seek relief. Nonetheless, 
while the NME methodology is applied, it appears that the actual trade 
impact of using this methodology will decline as the portion of total 
export trade conducted by Chinese companies assigned individual rates 
increases, and as the country-wide rates that largely account for the 
comparatively high average rates applied to China decline in 
importance. 

Agency Comments and Our Evaluation: 

The Department of Commerce provided written comments on a draft of this 
report. These comments are reprinted in appendix IV. Overall, Commerce 
agreed with our findings, observing that the report provided timely and 
helpful information on the NME methodology and its application to 
China. 

Commerce identified a small number of apparent errors in our database. 
We re-examined our data, making corrections when necessary, and updated 
our analyses; these corrections did not have any significant impact on 
our findings. Commerce also made a number of technical comments, 
focusing primarily on our description of its NME methodology. We took 
these comments into consideration and made changes throughout the 
report to insure its clarity and accuracy. We also made a number of 
technical corrections suggested by the Department of Homeland Security 
and the Office of the U.S. Trade Representative. 

We are sending copies of this report to the Secretaries of Commerce and 
Homeland Security, the International Trade Commission, the U. S. Trade 
Representative, appropriate congressional committees, and other 
interested parties. We also will make copies available to others upon 
request. In addition, the report will be available at no charge on the 
GAO Web site at [Hyperlink, http://www.gao.gov]. 

If you or any of your staff have any questions about this report, 
please contact me at 202-512-4347 or [Hyperlink, yagerl@gao.gov]. 
Contact points for our Offices of Congressional Relations and Public 
Affairs may be found on the last page of this report. GAO staff who 
made major contributions to this report are listed in appendix V. 

Signed by: 

Loren Yager: 
Director, International Affairs and Trade: 

[End of section] 

Appendixes: 

Appendix I: Scope and Methodology: 

To address our objectives, we examined and summarized applicable U.S. 
laws and regulations, as well as relevant World Trade Organization 
(WTO) agreements. These included the Agreement on Implementation of 
Article VI of the General Agreement on Tariffs and Trade 1994--commonly 
known as the "antidumping agreement"--and China's WTO accession 
agreement. We conducted a literature search and reviewed relevant 
scholarly and legal analyses and Department of Commerce (Commerce) 
determinations.[Footnote 48] In order to corroborate and broaden our 
understanding, we consulted with trade and legal policy experts from 
the U.S. government, private sector trade associations, consulting 
firms, academic institutions, law firms with broad experience in trade 
actions involving China, as well as representatives of the WTO, the 
government of China, and other governments concerned about Chinese 
trade practices, including the European Union and Canada. 

In order to analyze the application of antidumping (AD) duties to China 
and compare duty rates applied to China with those applied to market 
economy countries (our second objective) and to evaluate the potential 
impact of ceasing to apply the nonmarket economy (NME) methodology to 
China (our third objective), we collected information from the 
Department of Commerce and the International Trade Commission, 
including notices of Commerce determinations appearing in the Federal 
Register. We used this information to construct a database on all U.S. 
AD investigations from 1980 through 2004. In addition to information on 
the countries and products involved and the status of each 
investigation, our database included the duty rates applied upon 
completion of each new antidumping investigation against China during 
this period, as well as the duty rates applied against any producers of 
the same products from other countries. We verified this database to 
the official sources and found the data to be sufficiently reliable for 
the purposes of this report. This database is accessible on-line at 
www.gao.gov/cgi-bin/getrpt?GAO-06-652SP. We verified this database to 
the official sources and found the data to be sufficiently reliable for 
the purposes of this report.

Our analyses focused on the 68 cases during this time period wherein 
Commerce imposed AD duties on Chinese products, and especially on the 
subset of 25 cases in which Commerce imposed duties against a similar 
product from one or more market economy countries.[Footnote 49] 
Specifically, the 25 cases included all market economy cases that had 
the same product name and were initiated within 1 year of an AD 
investigation against China. In all, we assembled data on 303 company- 
specific, weighted average, and country-wide duty rate determinations 
on Chinese products, and an additional 168 duty rate determinations on 
market economy products. Appendix II provides additional analyses of 
this data. 

As part of our examination, we also performed multivariate regression 
analyses to determine the extent to which duty rate variations could be 
attributed to differences between China and these other countries, or 
to other factors, such as the type of product involved. Appendix III 
provides more information on these analyses and their results. 

In addition to comparing China and market economies, we also collected 
information on duty rates that Commerce applied to products from other 
NME countries at the same time as it applied them to similar products 
from either China or a market economy. Appendix III provides 
information on the results of our analyses of this data. 

We did not collect or analyze information on duty rates applied against 
market economy countries in cases where no parallel action was taken 
against China or any other NME country. Therefore, our analyses of 
market economy duty rates are specific to the sample of market economy 
orders in which a corresponding NME order was also in effect. Inclusion 
of other market economy product duty rates may have produced different 
results. However, we determined that the appropriate comparison between 
China and market economy countries was between the 25 similar products. 
We found through our regression analyses (discussed in app. III) that 
the product being investigated does help explain the variation among 
rates and it is, therefore, important to make an appropriate 
comparison. In addition, duty rates for the 43 remaining orders against 
China alone followed a similar pattern as those contained in the 25 
cases where we drew comparisons with market economy duty rates. The 
average country-wide rate for these 43 orders against China was higher 
than the country-wide rate for the 25 orders (118 percent compared to 
98 percent), and the average individual rate was lower (41 percent 
compared to 53 percent) for the 18 orders with individual rates. These 
results were consistent with our findings that the country-wide rates 
tend to be significantly higher than individual rates. 

In order to group specific products subject to AD orders into groups of 
similar products, we used the Harmonized Tariff Schedule (HTS) 
classifications for each product, as reported in the Federal Register 
announcement of the order. The HTS is the official U.S. classification 
of goods imported into the United States and includes 99 chapters 
covering all goods imports. In addition, the HTS chapters are grouped 
into larger sections that cover broad types of related products. The 
categories we used in this report are based on these HTS sections and 
chapters. Specifically, the category "Chemicals, plastics, 
pharmaceuticals" comprises HTS chapters 28 through 40 (which includes 
all chapters under the section "Chemical or allied industries"). The 
category "Steel, other metals" comprises HTS chapters 72 through 81 
(which includes most chapters under the section "Base metals and 
articles of base metals" except those chapters covering articles of 
base metals). The category "Agricultural products" comprises HTS 
chapters 1 through 24 (which includes all chapters under the sections 
"Live animals; animal products," "Vegetable products," "Animal or 
vegetable fats, etc.," and "Prepared foodstuffs, beverages, spirits, 
and vinegar; tobacco and manufactured tobacco substitutes"). The 
category "Other products" comprises all other HTS chapters. 

We conducted our work from June 2005 through December 2005 in 
accordance with generally accepted government auditing standards. 

[End of section] 

Appendix II: Additional Information on Duty Rates Applied to China and 
Market Economy Countries: 

This appendix provides additional information on the antidumping (AD) 
duty rate data that we assembled for this report and provides some 
additional analytical information, including brief discussions of 
variation in the duty rates applied to China over time, Department of 
Commerce (Commerce) determinations on whether Chinese companies should 
be considered eligible for individual rates, and duty rates applied to 
selected market economy countries. 

Duty Rates Imposed on China: 

The overall average duty rate for all 68 orders against China from 1980 
through 2004 was 65 percent. This was the result of 72 country-wide 
rates (on 68 products) with an average duty of 111 percent and 158 
individual company rates with an average duty of 44 percent.[Footnote 
50] These rates ranged from zero to about 384 percent (see table 3). In 
our analysis, we identified 25 orders against China in which there was 
also an order against a market economy country on the same product put 
in place within 1 year from the order against China. Table 3 shows 
overall average duty rates from the 25 orders against China that were 
matched to market economy cases and the 43 orders in which no market 
economy order was identified. Table 4 at the end of this appendix 
provides information on each of the 68 orders against China, and table 
5 provides comparative information for each of the 25 cases in which 
duties were also applied against market economy producers. 

Table 3: Summary Data on China AD Duty Orders, 1980-2004: 

25 orders matched to similar market economy orders; 
Type of rate: Individual company rates; 
Number of rates: 50; 
Mean (%): 52; 
Minimum (%): 0; 
Maximum (%): 162. 

Type of rate: Country-wide rate; 
Number of rates: 25; 
Mean (%): 98; 
Minimum (%): 3; 
Maximum (%): 218. 

Total; 
Number of rates: 75; 
Mean (%): 67; 
Minimum (%): 0; 
Maximum (%): 218. 

43 orders not matched to similar market economy orders; 
Type of rate: Individual company rates; 
Number of rates: 108; 
Mean (%): 41; 
Minimum (%): 0; 
Maximum (%): 292. 

Type of rate: Country-wide rate; 
Number of rates: 47; 
Mean (%): 118; 
Minimum (%): 1; 
Maximum (%): 384. 

Type of rate: Total; 
Number of rates: 155; 
Mean (%): 64; 
Minimum (%): 0; 
Maximum (%): 384. 

All 68 orders; 
Type of rate: Individual company rates; 
Number of rates: 158; 
Mean (%): 44; 
Minimum (%): 0; 
Maximum (%): 292. 

Type of rate: Country-wide rate; 
Number of rates: 72; 
Mean (%): 111; 
Minimum (%): 1; 
Maximum (%): 384. 

Type of rate: Grand total; 
Number of rates: 230; 
Mean (%): 65; 
Minimum (%): 0; 
Maximum (%): 384. 

Source: GAO AD database. 

Note: The overall average individual duty rates listed in this table 
for the "matching" orders are for the 25 products with both China and 
market economy orders. However, only 18 of these products had 
individual rates imposed on both China and at least one other market 
economy. We use these 18 products for our comparison of individual 
company duty rates imposed on China and market economies, rather than 
the 25 products, in the rest of the report. The average individual 
company duty rate imposed on China for these 18 products is 53 percent, 
rather than 52 percent. 

[End of table] 

About 78 percent (53 AD orders) of the 68 AD orders included not only 
country-wide rates but also individually calculated rates for at least 
one Chinese company. Of these, about 54 percent (37 orders) included 
company-specific rates that were lower than the country-wide rates 
imposed in the same cases. With regard to nonmalleable cast iron pipe 
fittings, for example, two Chinese companies submitted detailed 
information and met Commerce's criteria for assignment of individually 
determined rates. Other Chinese pipe fitting companies, however, did 
not provide any information. Commerce assigned the two cooperating 
companies duty rates of between 6 and 8 percent--a fraction of the 76 
percent country-wide duty rate applied in this case.[Footnote 51] 

Only 15 orders issued against China during this period included just a 
country-wide rate. Most of these orders date from the period before 
1991 when Commerce had not yet begun applying its separate rates test. 
However, from 1991 through 1995 Commerce issued six orders that 
contained only a country-wide rate. In most of these cases, Chinese 
companies failed to respond to Commerce requests for information. For 
example, in one case Commerce solicited information through both the 
Chinese government and the relevant Chinese industry association. 
However, the industry association responded that no Chinese producer or 
exporter wanted to participate in Commerce's investigation. Commerce, 
therefore, used facts available to establish a country-wide duty rate 
of about 156 percent.[Footnote 52] 

In 12 of the 68 orders, all the individual rates issued were equal to 
the country-wide rate. In some cases, Commerce specified an individual 
rate for one company and then used this rate as "facts available" to 
establish a country-wide duty rate at the same level.[Footnote 53] For 
example, in its investigation of refined brown aluminum oxide from 
China, Commerce requested information from the government of China and 
more than 20 Chinese companies. Only one of these companies responded. 
Commerce found that this company qualified for its own duty rate and 
determined that this rate should be about 135 percent. Commerce 
determined that the failure of the other companies to provide requested 
information justified application of an adverse inference to determine 
the country-wide rate. Since the rate established for the lone 
cooperating company was higher than any of the rates suggested in the 
petition requesting imposition of duties on this product, Commerce set 
the country-wide rate equal to the rate applied to the one cooperating 
company--135 percent.[Footnote 54] 

Duty Rates against China Have Fluctuated over Time: 

We found that there was a slight tendency for duty rates applied 
against Chinese products to rise over the period of our analysis, as 
well as to fluctuate over time. As figure 9 shows, individual company 
and country-wide duty rates tended to be larger from 1992-2004 than 
from 1980-1991. In addition, the individual company rates demonstrate a 
cyclical pattern over time. In our regression analysis, we found that 
there was a small positive trend in AD duty rates against China over 
time that was statistically significant. This result is consistent with 
research that has shown that overall U.S. AD margins have increased 
over time.[Footnote 55] 

Figure 9: Average Country-Wide and Individual Rates, All Orders against 
China, 1980-2004: 

[See PDF for image] 

Note: Commerce did not issue any new AD duty orders against China in 
1980, 1981, 1982, 1985, 1987, 1988, 1989, or 1998. In the tapered 
roller bearings cases, Commerce put an order in place in 1987 but 
amended it in 1990. We use 1990 information in the above graphic. 

[End of figure] 

Average Duty Rates on 68 Orders against China and Subset of 25 Orders 
Matched to Market Economy Orders: 

Table 4 shows the duty rates on the 68 orders imposed on China between 
1980 and 2004. Table 5 then shows the duty rates on the 25 orders 
imposed on China in which we also found matching orders imposed on 
market economies. 

Table 4: Average Duty Rates for 68 Orders against China, 1980-2004: 

Year and product: (1983) Cotton shop towels; 
Average (all rates) (%): 35; 
Average country-wide rate (%): 36; 
Individual company rates: Number of rates: 2; 
Individual company rates: Average (%): 34; 
Individual company rates: Minimum (%): 30; 
Individual company rates: Maximum (%): 37. 

Year and product: (1983) Greig polyester cotton print cloth; 
Average (all rates) (%): 22; 
Average country-wide rate (%): 22. 

Year and product: (1984) Barium chloride; 
Average (all rates) (%): 15; 
Average country-wide rate (%): 15; 
Individual company rates: Number of rates: 1; 
Individual company rates: Average (%): 15; 
Individual company rates: Minimum (%): 15; 
Individual company rates: Maximum (%): 15. 

Year and product: (1984) Chloropicrin; 
Average (all rates) (%): 58; 
Average country-wide rate (%): 58; 
Individual company rates: Number of rates: 1; 
Individual company rates: Average (%): 58; 
Individual company rates: Minimum (%): 58; 
Individual company rates: Maximum (%): 58. 

Year and product: (1984) Potassium permanganate; 
Average (all rates) (%): 40; 
Average country-wide rate (%): 40; 
Individual company rates: Number of rates: 1; 
Individual company rates: Average (%): 40; 
Individual company rates: Minimum (%): 40; 
Individual company rates: Maximum (%): 40. 

Year and product: (1986) Iron construction castings; 
Average (all rates) (%): 12; 
Average country-wide rate (%): 12. 

Year and product: (1986) Paint brushes; 
Average (all rates) (%): 127; 
Average country-wide rate (%): 127. 

Year and product: (1986) Porcelain-on-steel cookware; 
Average (all rates) (%): 67; 
Average country-wide rate (%): 67. 

Year and product: (1986) Steel wire nails; 
Average (all rates) (%): 6; 
Average country-wide rate (%): 6. 

Year and product: (1986) Wax candles; 
Average (all rates) (%): 54; 
Average country-wide rate (%): 54; 
Individual company rates: Number of rates: 1; 
Individual company rates: Average (%): 54; 
Individual company rates: Minimum (%): 54; 
Individual company rates: Maximum (%): 54. 

Year and product: (1990) Industrial nitrocellulose; 
Average (all rates) (%): 78; 
Average country-wide rate (%): 78; 
Individual company rates: Number of rates: 1; 
Individual company rates: Average (%): 78; 
Individual company rates: Minimum (%): 78; 
Individual company rates: Maximum (%): 78. 

Year and product: (1990) Tapered roller bearings; 
Average (all rates) (%): 3; 
Average country-wide rate (%): 3; 
Individual company rates: Number of rates: 2; 
Individual company rates: Average (%): 3; 
Individual company rates: Minimum (%): 1; 
Individual company rates: Maximum (%): 5. 

Year and product: (1991) Chrome-plated lug nuts; 
Average (all rates) (%): 42; 
Average country-wide rate (%): 42; 
Individual company rates: Number of rates: 1; 
Individual company rates: Average (%): 42; 
Individual company rates: Minimum (%): 42; 
Individual company rates: Maximum (%): 42. 

Year and product: (1991) Hand tools; 
Average (all rates) (%): 36; 
Average country-wide rate (%): 36. 

Year and product: (1991) Oscillating fans and ceiling fans; 
Average (all rates) (%): 1; 
Average country-wide rate (%): 1; 
Individual company rates: Number of rates: 8; 
Individual company rates: Average (%): 1; 
Individual company rates: Minimum (%): 0; 
Individual company rates: Maximum (%): 2. 

Year and product: (1991) Silicon metal; 
Average (all rates) (%): 139; 
Average country-wide rate (%): 139. 

Year and product: (1991) Sparklers; 
Average (all rates) (%): 59; 
Average country-wide rate (%): 76; 
Individual company rates: Number of rates: 3; 
Individual company rates: Average (%): 54; 
Individual company rates: Minimum (%): 2; 
Individual company rates: Maximum (%): 94. 

Year and product: (1991) Sulfur chemicals; 
Average (all rates) (%): 28; 
Average country-wide rate (%): 28. 

Year and product: (1991) Tungsten ore concentrates; 
Average (all rates) (%): 151; 
Average country-wide rate (%): 151. 

Year and product: (1992) Carbon steel butt-weld pipe fittings; 
Average (all rates) (%): 114; 
Average country-wide rate (%): 183; 
Individual company rates: Number of rates: 6; 
Individual company rates: Average (%): 102; 
Individual company rates: Minimum (%): 35; 
Individual company rates: Maximum (%): 155. 

Year and product: (1992) Sulfanilic acid; 
Average (all rates) (%): 52; 
Average country-wide rate (%): 85; 
Individual company rates: Number of rates: 1; 
Individual company rates: Average (%): 19; 
Individual company rates: Minimum (%): 19; 
Individual company rates: Maximum (%): 19. 

Year and product: (1993) Compact ductile iron waterworks; 
Average (all rates) (%): 127; 
Average country-wide rate (%): 127. 

Year and product: (1993) Ferrosilicon; 
Average (all rates) (%): 138; 
Average country-wide rate (%): 138. 

Year and product: (1993) Helical spring lock washers; 
Average (all rates) (%): 89; 
Average country-wide rate (%): 129; 
Individual company rates: Number of rates: 2; 
Individual company rates: Average (%): 70; 
Individual company rates: Minimum (%): 70; 
Individual company rates: Maximum (%): 70. 

Year and product: (1994) Cased pencils; 
Average (all rates) (%): 14; 
Average country-wide rate (%): 45; 
Individual company rates: Number of rates: 4; 
Individual company rates: Average (%): 6; 
Individual company rates: Minimum (%): 0; 
Individual company rates: Maximum (%): 17. 

Year and product: (1994) Garlic; 
Average (all rates) (%): 377; 
Average country-wide rate (%): 377. 

Year and product: (1994) Paper clips; 
Average (all rates) (%): 73; 
Average country-wide rate (%): 127; 
Individual company rates: Number of rates: 3; 
Individual company rates: Average (%): 55; 
Individual company rates: Minimum (%): 46; 
Individual company rates: Maximum (%): 61. 

Year and product: (1994) Sebacic acid; 
Average (all rates) (%): 98; 
Average country-wide rate (%): 243; 
Individual company rates: Number of rates: 4; 
Individual company rates: Average (%): 61; 
Individual company rates: Minimum (%): 44; 
Individual company rates: Maximum (%): 85. 

Year and product: (1994) Silicomanganese; 
Average (all rates) (%): 150; 
Average country-wide rate (%): 150. 

Year and product: (1995) Furfuryl alcohol; 
Average (all rates) (%): 46; 
Average country-wide rate (%): 45; 
Individual company rates: Number of rates: 2; 
Individual company rates: Average (%): 47; 
Individual company rates: Minimum (%): 44; 
Individual company rates: Maximum (%): 50. 

Year and product: (1995) Glycine; 
Average (all rates) (%): 156; 
Average country-wide rate (%): 156. 

Year and product: (1995) Pure magnesium 1; 
Average (all rates) (%): 108; 
Average country-wide rate (%): 108. 

Year and product: (1996) Manganese metal; 
Average (all rates) (%): 33; 
Average country-wide rate (%): 143; 
Individual company rates: Number of rates: 4; 
Individual company rates: Average (%): 6; 
Individual company rates: Minimum (%): 1; 
Individual company rates: Maximum (%): 12. 

Year and product: (1996) Polyvinyl alcohol 1; 
Average (all rates) (%): 78; 
Average country-wide rate (%): 117; 
Individual company rates: Number of rates: 2; 
Individual company rates: Average (%): 58; 
Individual company rates: Minimum (%): 0; 
Individual company rates: Maximum (%): 117. 

Year and product: (1997) Brake rotors; 
Average (all rates) (%): 10; 
Average country-wide rate (%): 43; 
Individual company rates: Number of rates: 6; 
Individual company rates: Average (%): 5; 
Individual company rates: Minimum (%): 0; 
Individual company rates: Maximum (%): 16. 

Year and product: (1997) Collated roofing nails; 
Average (all rates) (%): 39; 
Average country-wide rate (%): 118; 
Individual company rates: Number of rates: 2; 
Individual company rates: Average (%): 0; 
Individual company rates: Minimum (%): 0; 
Individual company rates: Maximum (%): 0. 

Year and product: (1997) Coumarin; 
Average (all rates) (%): 87; 
Average country-wide rate (%): 161; 
Individual company rates: Number of rates: 2; 
Individual company rates: Average (%): 51; 
Individual company rates: Minimum (%): 31; 
Individual company rates: Maximum (%): 70. 

Year and product: (1997) Crawfish; 
Average (all rates) (%): 133; 
Average country-wide rate (%): 202; 
Individual company rates: Number of rates: 5; 
Individual company rates: Average (%): 120; 
Individual company rates: Minimum (%): 92; 
Individual company rates: Maximum (%): 157. 

Year and product: (1997) Melamine dinnerware; 
Average (all rates) (%): 2; 
Average country-wide rate (%): 7; 
Individual company rates: Number of rates: 4; 
Individual company rates: Average (%): 1; 
Individual company rates: Minimum (%): 0; 
Individual company rates: Maximum (%): 3. 

Year and product: (1997) Persulfates; 
Average (all rates) (%): 55; 
Average country-wide rate (%): 119; 
Individual company rates: Number of rates: 3; 
Individual company rates: Average (%): 34; 
Individual company rates: Minimum (%): 32; 
Individual company rates: Maximum (%): 35. 

Year and product: (1999) Mushrooms; 
Average (all rates) (%): 155; 
Average country-wide rate (%): 199; 
Individual company rates: Number of rates: 4; 
Individual company rates: Average (%): 144; 
Individual company rates: Minimum (%): 121; 
Individual company rates: Maximum (%): 162. 

Year and product: (2000) Apple juice; 
Average (all rates) (%): 19; 
Average country-wide rate (%): 52; 
Individual company rates: Number of rates: 7; 
Individual company rates: Average (%): 14; 
Individual company rates: Minimum (%): 0; 
Individual company rates: Maximum (%): 28. 

Year and product: (2000) Bulk aspirin; 
Average (all rates) (%): 57; 
Average country-wide rate (%): 144; 
Individual company rates: Number of rates: 2; 
Individual company rates: Average (%): 14; 
Individual company rates: Minimum (%): 11; 
Individual company rates: Maximum (%): 17. 

Year and product: (2000) Creatine; 
Average (all rates) (%): 47; 
Average country-wide rate (%): 154; 
Individual company rates: Number of rates: 6; 
Individual company rates: Average (%): 30; 
Individual company rates: Minimum (%): 0; 
Individual company rates: Maximum (%): 58. 

Year and product: (2000) Synthetic indigo; 
Average (all rates) (%): 96; 
Average country-wide rate (%): 130; 
Individual company rates: Number of rates: 2; 
Individual company rates: Average (%): 80; 
Individual company rates: Minimum (%): 80; 
Individual company rates: Maximum (%): 80. 

Year and product: (2001) Certain hot-rolled carbon steel flat products; 
Average (all rates) (%): 76; 
Average country-wide rate (%): 91; 
Individual company rates: Number of rates: 4; 
Individual company rates: Average (%): 73; 
Individual company rates: Minimum (%): 64; 
Individual company rates: Maximum (%): 91. 

Year and product: (2001) Foundry coke products; 
Average (all rates) (%): 109; 
Average country-wide rate (%): 215; 
Individual company rates: Number of rates: 4; 
Individual company rates: Average (%): 83; 
Individual company rates: Minimum (%): 49; 
Individual company rates: Maximum (%): 106. 

Year and product: (2001) Honey; 
Average (all rates) (%): 72; 
Average country-wide rate (%): 184; 
Individual company rates: Number of rates: 4; 
Individual company rates: Average (%): 45; 
Individual company rates: Minimum (%): 26; 
Individual company rates: Maximum (%): 57. 

Year and product: (2001) Pure magnesium 2; 
Average (all rates) (%): 165; 
Average country-wide rate (%): 306; 
Individual company rates: Number of rates: 1; 
Individual company rates: Average (%): 25; 
Individual company rates: Minimum (%): 25; 
Individual company rates: Maximum (%): 25. 

Year and product: (2001) Steel concrete reinforcing bars; 
Average (all rates) (%): 133; 
Average country-wide rate (%): 133; 
Individual company rates: Number of rates: 1; 
Individual company rates: Average (%): 133; 
Individual company rates: Minimum (%): 133; 
Individual company rates: Maximum (%): 133. 

Year and product: (2002) Automotive replacement glass windshields; 
Average (all rates) (%): 37; 
Average country-wide rate (%): 125; 
Individual company rates: Number of rates: 3; 
Individual company rates: Average (%): 8; 
Individual company rates: Minimum (%): 4; 
Individual company rates: Maximum (%): 12. 

Year and product: (2002) Certain folding gift boxes; 
Average (all rates) (%): 58; 
Average country-wide rate (%): 165; 
Individual company rates: Number of rates: 2; 
Individual company rates: Average (%): 5; 
Individual company rates: Minimum (%): 2; 
Individual company rates: Maximum (%): 9. 

Year and product: (2002) Folding metal tables and folding metal chairs; 
Average (all rates) (%): 28; 
Average country-wide rate (%): 71; 
Individual company rates: Number of rates: 2; 
Individual company rates: Average (%): 7; 
Individual company rates: Minimum (%): 0; 
Individual company rates: Maximum (%): 14. 

Year and product: (2003) Barium carbonate; 
Average (all rates) (%): 58; 
Average country-wide rate (%): 81; 
Individual company rates: Number of rates: 1; 
Individual company rates: Average (%): 34; 
Individual company rates: Minimum (%): 34; 
Individual company rates: Maximum (%): 34. 

Year and product: (2003) Certain malleable iron pipe fittings; 
Average (all rates) (%): 31; 
Average country-wide rate (%): 111; 
Individual company rates: Number of rates: 4; 
Individual company rates: Average (%): 11; 
Individual company rates: Minimum (%): 7; 
Individual company rates: Maximum (%): 16. 

Year and product: (2003) Cut to length carbon steel plate; 
Average (all rates) (%): 62; 
Average country-wide rate (%): 129; 
Individual company rates: Number of rates: 5; 
Individual company rates: Average (%): 49; 
Individual company rates: Minimum (%): 17; 
Individual company rates: Maximum (%): 129. 

Year and product: (2003) Ferrovanadium; 
Average (all rates) (%): 40; 
Average country-wide rate (%): 67; 
Individual company rates: Number of rates: 1; 
Individual company rates: Average (%): 13; 
Individual company rates: Minimum (%): 13; 
Individual company rates: Maximum (%): 13. 

Year and product: (2003) Lawn and garden steel fence posts; 
Average (all rates) (%): 6; 
Average country-wide rate (%): 16; 
Individual company rates: Number of rates: 3; 
Individual company rates: Average (%): 2; 
Individual company rates: Minimum (%): 0; 
Individual company rates: Maximum (%): 7. 

Year and product: (2003) Non-malleable cast iron pipe fittings; 
Average (all rates) (%): 30; 
Average country-wide rate (%): 76; 
Individual company rates: Number of rates: 2; 
Individual company rates: Average (%): 7; 
Individual company rates: Minimum (%): 6; 
Individual company rates: Maximum (%): 7. 

Year and product: (2003) Polyvinyl alcohol 2; 
Average (all rates) (%): 52; 
Average country-wide rate (%): 98; 
Individual company rates: Number of rates: 1; 
Individual company rates: Average (%): 7; 
Individual company rates: Minimum (%): 7; 
Individual company rates: Maximum (%): 7. 

Year and product: (2003) Refined brown aluminum oxide; 
Average (all rates) (%): 135; 
Average country-wide rate (%): 135; 
Individual company rates: Number of rates: 1; 
Individual company rates: Average (%): 135; 
Individual company rates: Minimum (%): 135; 
Individual company rates: Maximum (%): 135. 

Year and product: (2003) Saccharin; 
Average (all rates) (%): 288; 
Average country-wide rate (%): 330; 
Individual company rates: Number of rates: 3; 
Individual company rates: Average (%): 274; 
Individual company rates: Minimum (%): 249; 
Individual company rates: Maximum (%): 292. 

Year and product: (2004) Carbazole violet pigment 23; 
Average (all rates) (%): 74; 
Average country-wide rate (%): 218; 
Individual company rates: Number of rates: 3; 
Individual company rates: Average (%): 26; 
Individual company rates: Minimum (%): 6; 
Individual company rates: Maximum (%): 45. 

Year and product: (2004) Certain color television receivers; 
Average (all rates) (%): 27; 
Average country-wide rate (%): 78; 
Individual company rates: Number of rates: 5; 
Individual company rates: Average (%): 17; 
Individual company rates: Minimum (%): 5; 
Individual company rates: Maximum (%): 26. 

Year and product: (2004) Hand trucks; 
Average (all rates) (%): 105; 
Average country-wide rate (%): 384; 
Individual company rates: Number of rates: 4; 
Individual company rates: Average (%): 35; 
Individual company rates: Minimum (%): 26; 
Individual company rates: Maximum (%): 46. 

Year and product: (2004) Ironing tables; 
Average (all rates) (%): 99; 
Average country-wide rate (%): 158; 
Individual company rates: Number of rates: 3; 
Individual company rates: Average (%): 80; 
Individual company rates: Minimum (%): 9; 
Individual company rates: Maximum (%): 158. 

Year and product: (2004) Polyethylene retail carrier bags; 
Average (all rates) (%): 27; 
Average country-wide rate (%): 78; 
Individual company rates: Number of rates: 8; 
Individual company rates: Average (%): 21; 
Individual company rates: Minimum (%): 0; 
Individual company rates: Maximum (%): 41. 

Year and product: (2004) Tetrahydrofurfuryl alcohol; 
Average (all rates) (%): 137; 
Average country-wide rate (%): 137; 
Individual company rates: Number of rates: 1; 
Individual company rates: Average (%): 137; 
Individual company rates: Minimum (%): 137; 
Individual company rates: Maximum (%): 137. 

Year and product: Total; 
Average (all rates) (%): 65; 
Average country-wide rate (%): 111; 
Individual company rates: Number of rates: 158; 
Individual company rates: Average (%): 44; 
Individual company rates: Minimum (%): 0; 
Individual company rates: Maximum (%): 292. 

Source: GAO AD database. 

Note: The average (all rates) is calculated as the average of the 
country-wide rate and each of the individual company rates that 
Commerce issued in its order. 

[End of table] 

Table 5: Average Duty Rates for 25 Products with Orders against both 
China and Market Economies, 1980-2004: 

Product and year: Carbazole violet pigment 23 (2004); 
Country and number of orders: China; 
Average: (all rates) (%): 74; 
Average country-wide or all-others rate (%): 218; 
Individual company rates: Number of rates: 3; 
Individual company rates: Average (%): 26; 
Individual company rates: Minimum (%): 6; 
Individual company rates: Maximum (%): 45. 

Country and number of orders: Market economies (1); 
Average: (all rates) (%): 29; 
Average country-wide or all-others rate (%): 27; 
Individual company rates: Number of rates: 2; 
Individual company rates: Average (%): 30; 
Individual company rates: Minimum (%): 10; 
Individual company rates: Maximum (%): 50. 

Product and year: Carbon steel butt-weld pipe fittings (1992); 
Country and number of orders: China; 
Average: (all rates) (%): 114; 
Average country-wide or all-others rate (%): 183; 
Individual company rates: Number of rates: 6; 
Individual company rates: Average (%): 102; 
Individual company rates: Minimum (%): 35; 
Individual company rates: Maximum (%): 155. 

Country and number of orders: Market economies (1); 
Average: (all rates) (%): 25; 
Average country-wide or all-others rate (%): 39; 
Individual company rates: Number of rates: 3; 
Individual company rates: Average (%): 21; 
Individual company rates: Minimum (%): 0; 
Individual company rates: Maximum (%): 51. 

Product and year: Certain hot-rolled carbon steel flat products (2001); 
Country and number of orders: China; 
Average: (all rates) (%): 76; 
Average country-wide or all-others rate (%): 91; 
Individual company rates: Number of rates: 4; 
Individual company rates: Average (%): 73; 
Individual company rates: Minimum (%): 64; 
Individual company rates: Maximum (%): 91. 

Country and number of orders: Market economies (7); 
Average: (all rates) (%): 24; 
Average country-wide or all-others rate (%): 23; 
Individual company rates: Number of rates: 11; 
Individual company rates: Average (%): 25; 
Individual company rates: Minimum (%): 3; 
Individual company rates: Maximum (%): 48. 

Product and year: Chrome-plated lug nuts (1991); 
Country and number of orders: China; 
Average: (all rates) (%): 42; 
Average country-wide or all-others rate (%): 42; 
Individual company rates: Number of rates: 1; 
Individual company rates: Average (%): 42; 
Individual company rates: Minimum (%): 42; 
Individual company rates: Maximum (%): 42. 

Country and number of orders: Market economies (1); 
Average: (all rates) (%): 8; 
Average country-wide or all-others rate (%): 7; 
Individual company rates: Number of rates: 2; 
Individual company rates: Average (%): 9; 
Individual company rates: Minimum (%): 6; 
Individual company rates: Maximum (%): 11. 

Product and year: Collated roofing nails (1997); 
Country and number of orders: China; 
Average: (all rates) (%): 39; 
Average country-wide or all-others rate (%): 118; 
Individual company rates: Number of rates: 2; 
Individual company rates: Average (%): 0; 
Individual company rates: Minimum (%): 0; 
Individual company rates: Maximum (%): 0. 

Country and number of orders: Market economies (1); 
Average: (all rates) (%): 22; 
Average country-wide or all-others rate (%): 3; 
Individual company rates: Number of rates: 3; 
Individual company rates: Average (%): 28; 
Individual company rates: Minimum (%): 3; 
Individual company rates: Maximum (%): 40. 

Product and year: Ferrosilicon (1993); 
Country and number of orders: China; 
Average: (all rates) (%): 138; 
Average country-wide or all- others rate (%): 138; 
Individual company rates: Number of rates: 0. 

Country and number of orders: Market economies (2); 
Average: (all rates) (%): 27; 
Average country-wide or all-others rate (%): 23; 
Individual company rates: Number of rates: 4; 
Individual company rates: Average (%): 29; 
Individual company rates: Minimum (%): 3; 
Individual company rates: Maximum (%): 89. 

Product and year: Ferrovanadium (2003); 
Country and number of orders: China; 
Average: (all rates) (%): 40; 
Average country-wide or all- others rate (%): 67; 
Individual company rates: Number of rates: 1; 
Individual company rates: Average (%): 13; 
Individual company rates: Minimum (%): 13; 
Individual company rates: Maximum (%): 13. 

Country and number of orders: Market economies (1); 
Average: (all rates) (%): 116; 
Average country-wide or all-others rate (%): 116; 
Individual company rates: Number of rates: 2; 
Individual company rates: Average (%): 116; 
Individual company rates: Minimum (%): 116; 
Individual company rates: Maximum (%): 116. 

Product and year: Furfuryl alcohol (1995); 
Country and number of orders: China; 
Average: (all rates) (%): 46; 
Average country-wide or all-others rate (%): 45; 
Individual company rates: Number of rates: 2; 
Individual company rates: Average (%): 47; 
Individual company rates: Minimum (%): 44; 
Individual company rates: Maximum (%): 50. 

Country and number of orders: Market economies (2); 
Average: (all rates) (%): 10; 
Average country-wide or all-others rate (%): 10; 
Individual company rates: Number of rates: 2; 
Individual company rates: Average (%): 10; 
Individual company rates: Minimum (%): 8; 
Individual company rates: Maximum (%): 12. 

Product and year: Helical spring lock washers (1993); 
Country and number of orders: China; 
Average: (all rates) (%): 89; 
Average country-wide or all-others rate (%): 129; 
Individual company rates: Number of rates: 2; 
Individual company rates: Average (%): 70; 
Individual company rates: Minimum (%): 70; 
Individual company rates: Maximum (%): 70. 

Country and number of orders: Market economies (1); 
Average: (all rates) (%): 32; 
Average country-wide or all-others rate (%): 32; 
Individual company rates: Number of rates: 3; 
Individual company rates: Average (%): 32; 
Individual company rates: Minimum (%): 32; 
Individual company rates: Maximum (%): 32. 

Product and year: Honey (2001); 
Country and number of orders: China; 
Average: (all rates) (%): 72; 
Average country-wide or all-others rate (%): 184; 
Individual company rates: Number of rates: 4; 
Individual company rates: Average (%): 45; 
Individual company rates: Minimum (%): 26; 
Individual company rates: Maximum (%): 57. 

Country and number of orders: Market economies (1); 
Average: (all rates) (%): 36; 
Average country-wide or all-others rate (%): 30; 
Individual company rates: Number of rates: 3; 
Individual company rates: Average (%): 38; 
Individual company rates: Minimum (%): 27; 
Individual company rates: Maximum (%): 55. 

Product and year: Industrial nitrocellulose (1990); 
Country and number of orders: China; 
Average: (all rates) (%): 78; 
Average country-wide or all-others rate (%): 78; 
Individual company rates: Number of rates: 1; 
Individual company rates: Average (%): 78; 
Individual company rates: Minimum (%): 78; 
Individual company rates: Maximum (%): 78. 

Country and number of orders: Market economies (6); 
Average: (all rates) (%): 37; 
Average country-wide or all-others rate (%): 37; 
Individual company rates: Number of rates: 6; 
Individual company rates: Average (%): 37; 
Individual company rates: Minimum (%): 4; 
Individual company rates: Maximum (%): 66. 

Product and year: Iron construction castings (1986); 
Country and number of orders: China; 
Average: (all rates) (%): 12; 
Average country-wide or all-others rate (%): 12; 
Individual company rates: Number of rates: 0. 

Country and number of orders: Market economies (3); 
Average: (all rates) (%): 11; 
Average country-wide or all-others rate (%): 12; 
Individual company rates: Number of rates: 10; 
Individual company rates: Average (%): 11; 
Individual company rates: Minimum (%): 0; 
Individual company rates: Maximum (%): 59. 

Product and year: Melamine institutional dinnerware (1997); 
Country and number of orders: China; 
Average: (all rates) (%): 2; 
Average country-wide or all-others rate (%): 7; 
Individual company rates: Number of rates: 4; 
Individual company rates: Average (%): 1; 
Individual company rates: Minimum (%): 0; 
Individual company rates: Maximum (%): 3. 

Country and number of orders: Market economies (2); 
Average: (all rates) (%): 18; 
Average country-wide or all-others rate (%): 6; 
Individual company rates: Number of rates: 6; 
Individual company rates: Average (%): 22; 
Individual company rates: Minimum (%): 0; 
Individual company rates: Maximum (%): 53. 

Product and year: Polyethylene retail carrier bags (2004); 
Country and number of orders: China; 
Average: (all rates) (%): 27; 
Average country-wide or all-others rate (%): 78; 
Individual company rates: Number of rates: 8; 
Individual company rates: Average (%): 21; 
Individual company rates: Minimum (%): 0; 
Individual company rates: Maximum (%): 41. 

Country and number of orders: Market economies (2); 
Average: (all rates) (%): 81; 
Average country-wide or all-others rate (%): 44; 
Individual company rates: Number of rates: 10; 
Individual company rates: Average (%): 88; 
Individual company rates: Minimum (%): 2; 
Individual company rates: Maximum (%): 123. 

Product and year: Polyvinyl alcohol 1 (1996); 
Country and number of orders: China; 
Average: (all rates) (%): 78; 
Average country-wide or all-others rate (%): 117; 
Individual company rates: Number of rates: 2; 
Individual company rates: Average (%): 58; 
Individual company rates: Minimum (%): 0; 
Individual company rates: Maximum (%): 117. 

Country and number of orders: Market economies (2); 
Average: (all rates) (%): 63; 
Average country-wide or all-others rate (%): 48; 
Individual company rates: Number of rates: 6; 
Individual company rates: Average (%): 68; 
Individual company rates: Minimum (%): 19; 
Individual company rates: Maximum (%): 77. 

Product and year: Polyvinyl alcohol 2 (2003); 
Country and number of orders: China; 
Average: (all rates) (%): 52; 
Average country-wide or all-others rate (%): 98; 
Individual company rates: Number of rates: 1; 
Individual company rates: Average (%): 7; 
Individual company rates: Minimum (%): 7; 
Individual company rates: Maximum (%): 7. 

Country and number of orders: Market economies (2); 
Average: (all rates) (%): 103; 
Average country-wide or all-others rate (%): 54; 
Individual company rates: Number of rates: 5; 
Individual company rates: Average (%): 123; 
Individual company rates: Minimum (%): 39; 
Individual company rates: Maximum (%): 144. 

Product and year: Porcelain-on-steel cooking ware (1986); 
Country and number of orders: China; 
Average: (all rates) (%): 67; 
Average country-wide or all-others rate (%): 67; 
Individual company rates: Number of rates: 0. 

Country and number of orders: Market economies (2); 
Average: (all rates) (%): 15; 
Average country-wide or all-others rate (%): 17; 
Individual company rates: Number of rates: 8; 
Individual company rates: Average (%): 15; 
Individual company rates: Minimum (%): 2; 
Individual company rates: Maximum (%): 57. 

Product and year: Potassium permanganate (1984); 
Country and number of orders: China; 
Average: (all rates) (%): 40; 
Average country-wide or all-others rate (%): 40; 
Individual company rates: Number of rates: 1; 
Individual company rates: Average (%): 40; 
Individual company rates: Minimum (%): 40; 
Individual company rates: Maximum (%): 40. 

Country and number of orders: Market economies (1); 
Average: (all rates) (%): 5; 
Average country-wide or all-others rate (%): 5; 
Individual company rates: Number of rates: 1; 
Individual company rates: Average (%): 5; 
Individual company rates: Minimum (%): 5; 
Individual company rates: Maximum (%): 5. 

Product and year: Preserved mushrooms (1999); 
Country and number of orders: China; 
Average: (all rates) (%): 155; 
Average country-wide or all-others rate (%): 199; 
Individual company rates: Number of rates: 4; 
Individual company rates: Average (%): 144; 
Individual company rates: Minimum (%): 121; 
Individual company rates: Maximum (%): 162. 

Country and number of orders: Market economies (3); 
Average: (all rates) (%): 86; 
Average country-wide or all-others rate (%): 57; 
Individual company rates: Number of rates: 7; 
Individual company rates: Average (%): 98; 
Individual company rates: Minimum (%): 6; 
Individual company rates: Maximum (%): 244. 

Product and year: Silicomanganese 1 (1994); 
Country and number of orders: China; 
Average: (all rates) (%): 150; 
Average country-wide or all-others rate (%): 150; 
Individual company rates: Number of rates: 0. 

Country and number of orders: Market economies (1); 
Average: (all rates) (%): 41; 
Average country-wide or all-others rate (%): 18; 
Individual company rates: Number of rates: 1; 
Individual company rates: Average (%): 65; 
Individual company rates: Minimum (%): 65; 
Individual company rates: Maximum (%): 65. 

Product and year: Silicon metal (1991); 
Country and number of orders: China; 
Average: (all rates) (%): 139; 
Average country-wide or all- others rate (%): 139; 
Individual company rates: Number of rates: 0. 

Country and number of orders: Market economies (2); 
Average: (all rates) (%): 58; 
Average country-wide or all-others rate (%): 50; 
Individual company rates: Number of rates: 3; 
Individual company rates: Average (%): 63; 
Individual company rates: Minimum (%): 9; 
Individual company rates: Maximum (%): 93. 

Product and year: Steel concrete reinforcing bars (2001); 
Country and number of orders: China; 
Average: (all rates) (%): 133; 
Average country-wide or all-others rate (%): 133; 
Individual company rates: Number of rates: 1; 
Individual company rates: Average (%): 133; 
Individual company rates: Minimum (%): 133; 
Individual company rates: Maximum (%): 133. 

Country and number of orders: Market economies (4); 
Average: (all rates) (%): 57; 
Average country-wide or all-others rate (%): 37; 
Individual company rates: Number of rates: 12; 
Individual company rates: Average (%): 64; 
Individual company rates: Minimum (%): 17; 
Individual company rates: Maximum (%): 102. 

Product and year: Sulfanilic acid (1992); 
Country and number of orders: China; 
Average: (all rates) (%): 52; 
Average country-wide or all- others rate (%): 85; 
Individual company rates: Number of rates: 1; 
Individual company rates: Average (%): 19; 
Individual company rates: Minimum (%): 19; 
Individual company rates: Maximum (%): 19. 

Country and number of orders: Market economies (1); 
Average: (all rates) (%): 115; 
Average country-wide or all-others rate (%): 115; 
Individual company rates: Number of rates: 0. 

Product and year: Sulfur chemicals (1991); 
Country and number of orders: China; 
Average: (all rates) (%): 28; 
Average country-wide or all-others rate (%): 28; 
Individual company rates: Number of rates: 0. 

Country and number of orders: Market economies (2); 
Average: (all rates) (%): 75; 
Average country-wide or all-others rate (%): 75; 
Individual company rates: Number of rates: 2; 
Individual company rates: Average (%): 75; 
Individual company rates: Minimum (%): 50; 
Individual company rates: Maximum (%): 100. 

Product and year: Tapered roller bearings (1987); 
Country and number of orders: China; 
Average: (all rates) (%): 3; 
Average country-wide or all-others rate (%): 3; 
Individual company rates: Number of rates: 2; 
Individual company rates: Average (%): 3; 
Individual company rates: Minimum (%): 1; 
Individual company rates: Maximum (%): 5. 

Country and number of orders: Market economies (3); 
Average: (all rates) (%): 54; 
Average country-wide or all-others rate (%): 65; 
Individual company rates: Number of rates: 2; 
Individual company rates: Average (%): 36; 
Individual company rates: Minimum (%): 36; 
Individual company rates: Maximum (%): 37. 

Product and year: Total; 
Country and number of orders: China; 
Average: (all rates) (%): 67; 
Average country-wide or all-others rate (%): 98; 
Individual company rates: Number of rates: 50; 
Individual company rates: Average (%): 52; 
Individual company rates: Minimum (%): 0; 
Individual company rates: Maximum (%): 162. 

Country and number of orders: Market economies (54); 
Average: (all rates) (%): 44; 
Average country-wide or all-others rate (%): 37; 
Individual company rates: Number of rates: 114; 
Individual company rates: Average (%): 48; 
Individual company rates: Minimum (%): 0; 
Individual company rates: Maximum (%): 244. 

Source: GAO AD database. 

Notes: 

The average (all rates) is calculated as the average of the country-
wide rate and each of the individual company rates that Commerce issued 
in its order. 

The overall average individual duty rates listed in this table are for 
the 25 products with both China and market economy orders. However, 
only 18 of these products had individual rates imposed on both China 
and at least one other market economy. We use these 18 products for our 
comparison of individual company duty rates imposed on China and market 
economies, rather than the 25 products, in the rest of the report. The 
average individual company duty rate imposed on China for these 18 
products is 53 percent and for market economies it is 55 percent. 

[End of table] 

[End of section] 

Appendix III: Regression Analysis Results: 

In order to examine the difference between duty rates applied to China 
and those applied to market economy countries, we performed 
multivariate regression analyses on the cases in which the Department 
of Commerce (Commerce) applied duties to both China and at least one 
market economy country. These involved 25 different products, affected 
by 25 duty orders against China, and 54 duty orders against market 
economies. Multivariate regression analysis makes it possible to 
examine the simultaneous effect of several different factors on the 
duty rates and to determine the extent to which these factors, taken 
together, explain variation in these rates. To determine whether our 
analytical results for China held true for all nonmarket economy (NME) 
countries, we also identified six instances in which Commerce applied 
duties to a nonmarket economy other than China, and at least one market 
economy country, and reran our analyses using data for all 31 products. 

Table 6 shows the results of our multivariate regression analysis of 
variation in the dependent variable (the antidumping [AD] duty rate) 
attributable to the following independent variables: 

* China (a variable indicating whether the AD duty rate is for China or 
not): 

* the country-wide rate (a variable indicating whether the AD duty rate 
is a country-wide rate), and: 

* year (a variable indicating the year in which the duty went into 
affect). 

We also included a constant term. The regression involved 25 products 
covered by 25 orders against China and 54 orders against market 
economies and included a total of 243 duty rates (the dependent 
variable) from these 79 orders. 

The results show that the variable for China as the target country had 
a coefficient of 3.002 percent, indicating that duty rates against 
China tended to be about 3 percentage points higher than those against 
market economies, on average. However, this coefficient is not 
statistically significant, meaning that there was no statistically 
significant difference between the rates assigned to China and market 
economy countries, when the other factors in the regression are 
included. The coefficient for the country-wide rate, on the other hand, 
shows that there is a 52 percentage point difference between country-
wide rates against China and other rates. This result is statistically 
significant at above the 99 percent level. The variable for the year of 
the order is also statistically significant, but it has a small 
coefficient.[Footnote 56] The adjusted R-square measure shows that 
about 15 percent of the overall variation in duty rates is explained by 
the independent variables included here. 

Table 6: Results of Multivariate Regression Analysis of Duty Rates on 
Explanatory Variables for China, Country-Wide Rates, and Year: 

Dependent variable = AD duty rate. 

Independent variable: China; 
Unstandardized coefficients: (B): 3.002; 
Standard error: 7.511; 
Standardized coefficients (Beta): .028; 
t-statistic: .400; 
Significance of t-statistic: .690. 

Independent variable: Year; 
Unstandardized coefficients: (B): 2.095; 
Standard error: .497; 
Standardized coefficients (Beta): .252; 
t-statistic: 4.218; 
Significance of t-statistic: .000. 

Independent variable: Country-wide rate; 
Unstandardized coefficients: (B): 52.050; 
Standard error: 11.398; 
Standardized coefficients (Beta): .316; 
t-statistic: 4.567; 
Significance of t-statistic: .000. 

Source: GAO analysis of Commerce data. 

Note: R-square = 0.164; Adjusted R-square = 0.154; Observations = 243. 

[End of table] 

We then included additional variables for product groups, such as 
agriculture and steel, and, in separate regressions, individual product 
variables for each type of product. The additional variables generally 
improved the overall "fit" of the regression; the adjusted R-square 
measure with the individual product variables included showed that the 
regression explained between 24 and 31 percent of the overall variation 
in duty rates across the sample compared with 15 percent in the 
regression above. Also, certain types of products, such as agriculture 
products, tended to have higher duty rates relative to other types. 

Table 7 shows the regression results when individual product variables 
are included. Once again the coefficient for China is insignificant, 
while the coefficient for the country-wide rate is significant at the 
99 percent level. Some coefficients for individual products are 
significant (e.g., carbon steel butt-weld pipe fittings), but many are 
not. The overall adjusted R-square measure shows that this regression 
model explains about 31 percent of total variation in the duty rates. 

Table 7: Results of Multivariate Regression Analysis of Duty Rates on 
Explanatory Variables for China, Country-Wide Rates, and Individual 
Products: 

[See PDF for table] 

Source: GAO analysis of Commerce data. 

Note: R-square = 0.386; Adjusted R-square = 0.309; Observations = 243. 

[End of table] 

In order to examine the effect of applying adverse inferences and facts 
available (other than adverse inferences) on the duty rates, we added 
additional variables indicating when Commerce used these approaches. 
The results show that application of adverse inferences is a 
significant variable and has a large effect on the duty rates, but that 
application of facts available (other than adverse inferences) is not. 
When adverse inferences are introduced, this results in the country-
wide rate variable becoming insignificant (see table 8). However, this 
is likely due to the fact that the adverse inferences variable is 
highly correlated with the country-wide rate. Therefore, it is not 
surprising that the country-wide rate is no longer significant since 
the adverse inferences variable is already accounting for much of the 
variation. In addition, the variable for China once again becomes 
significant. As we discuss in the body of this report, Commerce uses 
adverse inferences in very few determinations for Chinese companies 
granted their own rates. Adverse inferences were applied in making only 
3 out of the 50 individual determinations used in this analysis. 
However, Commerce used adverse inferences in nearly half of its 
determinations against individual market economy companies. 

Since adverse inferences are already factored into this model, as is 
the country-wide rate, the remaining differences accounted for by the 
China variable in table 3 are between individual (noncountry-wide) 
Chinese rates and individual market economy rates in which adverse 
inferences are not used. Table 8 shows that there is a statistically 
significant 27 percentage point difference between these rates. 
However, because there are methodological differences between the NME 
and market economy methodologies for individual companies, it is not 
clear how often adverse inferences would be used against individual 
Chinese companies should they move to a market economy methodology. In 
other words, we cannot predict the extent to which, under a market 
economy methodology, individual Chinese companies would cooperate with 
Commerce or Commerce would find it necessary to use adverse inferences 
in its determinations against Chinese companies. It is possible that 
some Chinese companies that currently have an individually determined 
rate under the NME methodology would face adverse inferences under a 
market economy methodology, whereas others would not. This could 
produce a result similar to the market economy cases we have examined 
in which the overall average (for example, 55 percent) is the result of 
some companies receiving comparatively high duty rates (e.g., 77 
percent) when adverse inferences are used and others receiving 
comparatively low rates (e.g., 16 percent) when adverse inferences are 
not used (see table 2). In any case, these results show that there is a 
remaining difference between these two groups after accounting for the 
use of adverse inferences and the country-wide rate. 

Table 8: Results of Multivariate Regression Analysis of Duty Rates on 
Explanatory Variables for China, Country-Wide Rates, Adverse 
Inferences, Other Facts Available, and Individual Products: 

Dependent variable = AD duty rate. 

Independent variable: China; 
Unstandardized coefficients: (B): 27.342; 
Standard error: 6.856; 
Standardized coefficients (Beta): .252; 
t-statistic: 3.988; 
Significance of t-statistic: .000. 

Independent variable: Year; 
Unstandardized coefficients: (B): 4.831; 
Standard error: 17.572; 
Standardized coefficients (Beta): .582; 
t-statistic: .275; 
Significance of t-statistic: .784. 

Independent variable: Country-wide rate; 
Unstandardized coefficients: (B): 9.910; 
Standard error: 10.064; 
Standardized coefficients (Beta): .060; 
t-statistic: .985; 
Significance of t-statistic: .326. 

Independent variable: Adverse inferences; 
Unstandardized coefficients: (B): 60.196; 
Standard error: 6.845; 
Standardized coefficients (Beta): .587; 
t-statistic: 8.794; 
Significance of t-statistic: .000. 

Independent variable: Facts available; 
Unstandardized coefficients: (B): 10.135; 
Standard error: 8.495; 
Standardized coefficients (Beta): .069; 
t-statistic: 1.193; 
Significance of t-statistic: .234. 

Independent variable: Carbazole violet pigment 23 (2004); 
Unstandardized coefficients: (B): -40.728; 
Standard error: 229.493; 
Standardized coefficients (Beta): -.136; 
t-statistic: -.177; 
Significance of t-statistic: .859. 

Independent variable: Carbon steel butt-weld pipe fittings (1992); 
Unstandardized coefficients: (B): 39.749; Dependent variable 
= AD duty rate: Standard error: 26.189; 
Standardized coefficients (Beta): .165; 
t-statistic: 1.518; 
Significance of t-statistic: .131. 

Independent variable: Certain hot-rolled carbon steel flat products 
(2001); 
Unstandardized coefficients: (B): -57.704; 
Standard error: 176.678; 
Standardized coefficients (Beta): -.337; 
t-statistic: -.327; 
Significance of t-statistic: .744. 

Independent variable: Collated roofing nails (1997); 
Unstandardized coefficients: (B): -45.478; 
Standard error: 107.679; 
Standardized coefficients (Beta): -.152; 
t-statistic: -.422; 
Significance of t-statistic: .673. 

Independent variable: Ferrosilicon (1993); 
Unstandardized coefficients: (B): -3.176; 
Standard error: 49.788; 
Standardized coefficients (Beta): -.011; 
t-statistic: -.064; 
Significance of t-statistic: .949. 

Independent variable: Ferrovanadium (2003); 
Unstandardized coefficients: (B): -40.348; 
Standard error: 212.436; 
Standardized coefficients (Beta): -.114; 
t-statistic: -.190; 
Significance of t-statistic: .850. 

Independent variable: Furfuryl alcohol (1995); 
Unstandardized coefficients: (B): -13.868; 
Standard error: 73.344; 
Standardized coefficients (Beta): -.046; 
t-statistic: -.189; 
Significance of t-statistic: .850. 

Independent variable: Helical spring lock washers (1993); 
Unstandardized coefficients: (B): -18.938; 
Standard error: 41.354; 
Standardized coefficients (Beta): -.063; 
t-statistic: -.458; 
Significance of t-statistic: .647. 

Independent variable: Honey (2001); 
Unstandardized coefficients: (B): -31.946; 
Standard error: 176.875; 
Standardized coefficients (Beta): -.121; 
t-statistic: -.181; 
Significance of t-statistic: .857. 

Independent variable: Industrial nitrocellulose (1990); 
Unstandardized coefficients: (B): 8.668; 
Standard error: 25.589; 
Standardized coefficients (Beta): .040; 
t-statistic: .339; 
Significance of t-statistic: .735. 

Independent variable: Iron construction castings (1986); 
Unstandardized coefficients: (B): 25.459; 
Standard error: 89.838; 
Standardized coefficients (Beta): .118; 
t-statistic: .283; 
Significance of t-statistic: .777. 

Independent variable: Melamine institutional dinnerware (1997); 
Unstandardized coefficients: (B): -53.825; 
Standard error: 107.215; 
Standardized coefficients (Beta): -.242; 
t-statistic: -.502; 
Significance of t-statistic: .616. 

Independent variable: Polyethylene retail carrier bags (2004); 
Unstandardized coefficients: (B): -55.280; 
Standard error: 229.336; 
Standardized coefficients (Beta): -.310; 
t-statistic: -.241; 
Significance of t-statistic: .810. 

Independent variable: Polyvinyl alcohol 1 (1996); 
Unstandardized coefficients: (B): -5.196; 
Standard error: 90.169; 
Standardized coefficients (Beta): -.022; 
t-statistic: -.058; 
Significance of t-statistic: .954. 

Independent variable: Polyvinyl alcohol 2 (2003); 
Unstandardized coefficients: (B): -22.256; 
Standard error: 211.996; 
Standardized coefficients (Beta): -.084; 
t-statistic: -.105; 
Significance of t-statistic: .916. 

Independent variable: Porcelain-on-steel cooking ware (1986); 
Unstandardized coefficients: (B): 32.037; 
Standard error: 89.998; 
Standardized coefficients (Beta): .133; 
t-statistic: .356; 
Significance of t-statistic: .722. 

Independent variable: Potassium permanganate (1984); 
Unstandardized coefficients: (B): 33.392; 
Standard error: 125.194; 
Standardized coefficients (Beta): .085; 
t-statistic: .267; 
Significance of t-statistic: .790. 

Independent variable: Preserved mushrooms (1999); 
Unstandardized coefficients: (B): 32.301; 
Standard error: 139.519; 
Standardized coefficients (Beta): .155; 
t-statistic: .232; 
Significance of t-statistic: .817. 

Independent variable: Silicomanganese 1 (1994); 
Unstandardized coefficients: (B): 3.633; 
Standard error: 59.124; 
Standardized coefficients (Beta): .008; 
t-statistic: .061; 
Significance of t-statistic: .951. 

Independent variable: Silicon metal (1991); 
Unstandardized coefficients: (B): 39.950; 
Standard error: 22.201; 
Standardized coefficients (Beta): .124; 
t-statistic: 1.799; 
Significance of t-statistic: .073. 

Independent variable: Steel concrete reinforcing bars (2001); 
Unstandardized coefficients: (B): -41.334; 
Standard error: 176.900; 
Standardized coefficients (Beta): -.216; 
t-statistic: -.234; 
Significance of t-statistic: .815. 

Independent variable: Sulfanilic acid (1992); 
Unstandardized coefficients: (B): -1.893; 
Standard error: 35.427; 
Standardized coefficients (Beta): -.004; 
t-statistic: -.053; 
Significance of t-statistic: .957. 

Independent variable: Sulfur chemicals (1991); 
Unstandardized coefficients: (B): -8.757; 
Standard error: 23.151; 
Standardized coefficients (Beta): -.025; 
t-statistic: -.378; 
Significance of t-statistic: .706. 

Independent variable: Tapered roller bearings (1987); 
Unstandardized coefficients: (B): 12.952; 
Standard error: 73.025; 
Standardized coefficients (Beta): .046; 
t-statistic: .177; 
Significance of t-statistic: .859. 

Source: GAO analysis of Commerce data. 

Note: R-square = 0.562; Adjusted R-square = 0.502; Observations = 243. 

[End of table] 

In order to examine whether the above results hold for all NMEs, we ran 
the same regressions for a larger set of 31 products (compared with the 
25 products above) in which we found matching cases between nonmarket 
economies other than China and market economies. The data set on these 
31 products included rates from 128 orders (26 on China, 82 on market 
economies, and 20 on NMEs other than China) that contained 355 duty 
rates (dependent variable). 

These analyses confirmed our China-market economy only analyses but 
also showed that other NME countries tend to have duty rates that are 
statistically higher than market economy rates for this sample of 
matching cases. (Note that the number of additional products--six--is 
relatively small.) Controlling for both the NME designation and the 
country-wide rate, the NME designation itself is a significant variable 
at the 97 percent level of confidence with a coefficient of 23 percent 
(the coefficient for China is not statistically significant). The 
country-wide variable is also significant (99 percent level) and larger 
with a coefficient of 48 percent. As additional variables are added for 
individual products, the NME designation continued to be significant 
along with the country-wide rate variable. 

There may be other systematic factors not accounted for in this 
regression model that would explain some of the variability not 
accounted for by the variables we included. As shown in table 7, our 
model accounted for about 50 percent (half) of the variation in rates. 
Some of this variation may be idiosyncratic and related to differences 
in individual companies' practices, other may relate to how Commerce 
has implemented its analysis. However, these unexplained factors do not 
appear to be systematically related to whether the case involved China 
or a market economy since the regression analysis already controls for 
that difference. 

[End of section] 

Appendix IV: Comments from the Department of Commerce: 

UNITED STATES DEPARTMENT OF COMMERCE: 
The Under Secretary for International Trade: 
Washington, D.C. 20230: 

DEC 8 2005: 

Mr. Loren Yager: 
Director: 
International Affairs and Trade: 
U. S. Government Accountability Office: 
Washington, D.C. 20548: 

Dear Mr. Yager: 

Thank you for providing us with your draft report on the potential 
effects on certain Chinese companies of eliminating the non-market 
economy (NME) antidumping duty (AD) calculation methodology. As there 
has been increased concern with China trade practices in general, as 
well as increased attention paid to Commerce's statutorily based NME 
methodology, we find your report both timely and helpful. This study 
demonstrates a strong effort by the GAO staff to understand Commerce's 
application of the NME methodology; the history of AD duties applied to 
China during the last 25 years compared to the rates applied to market 
economy countries; and the circumstances under which the United States 
might determine it appropriate to stop applying its NME methodology, 
along with and the potential consequences of such a decision. 

We have reviewed carefully the facts regarding the 25 cases in which 
there were both Chinese and market economy AD orders issued. We found 
some errors in the data and shared them with your staff. We agree with 
the report's observation that the starkest difference between China 
rates and market economy rates lays in the difference between the NME 
country-wide rate and the market economy "all others" rate. We further 
agree that this difference arises from both the NME methodology itself, 
which requires a country-wide rate, and from how the two rates are 
calculated. The former rate usually reflects corroborated adverse 
inferences based on non-cooperation, while the latter is a weighted 
average of only the calculated individual company rates. The GAO could 
have analyzed such differing data pools as the set of all final AD 
margins generated regardless of whether an order was put in place. It 
also could have chosen to highlight other comparisons, such as between 
the rates determined for cooperating individual Chinese respondents and 
those applicable to cooperating respondents in market economy 
countries. The noticeable difference between the two types of rates, 
however, would likely have remained the same. 

We note with interest the GAO's cautionary observations. First, it is 
true that increasing numbers of Chinese companies have been applying 
for, and receiving, separate, individual rates in AD investigations. 
This has the effect of removing those companies from coverage under the 
countrywide rate, and likely reducing the ultimate trade impact of 
eliminating the NME methodology, as the GAO surmises. Second, we note 
the GAO's explicit discussion of the relatively low explanatory power 
of its model, i.e., the model does not explain that much of the 
observed difference between China and market economy rates. In 
addition, the report acknowledges that it cannot predict the extent to 
which individual Chinese companies would cooperate with Commerce under 
a market economy methodology and thus avoid the use of adverse 
inferences. These limitations make clear that no definitive conclusions 
or recommendations could reliably be drawn from the findings, as the 
report itself seems to recognize. 

I appreciate the opportunity to provide comments on the draft report. 
Enclosed is an attachment with specific technical comments relating to 
the text of the report. 

Sincerely, 

Signed by: 

Franklin L. Lavin: 

The following are GAO's comments on the Department of Commerce's letter 
dated December 8, 2005. 

GAO Comments: 

1. We re-examined our data, making corrections as appropriate, and 
updated our analyses. The report reflects these corrections, though 
they did not have a significant impact on any of our findings. 

2. As discussed in the report, the overall difference between the duty 
rates applied to China and those applied to market economy countries is 
largely explained by the application of comparatively high country-wide 
rates to China. Therefore, the model allows us to conclude that 
elimination of the NME methodology--and thus these country-wide rates-
-would result in lower duties for some Chinese companies. Nevertheless, 
there would still be variation in duty rates among companies and 
products due to a range of other factors. 

[End of section] 

Appendix V: GAO Contact and Staff Acknowledgments: 

GAO Contact: 

Loren Yager (202) 512-4347: 

Staff Acknowledgments: 

In addition to the individual named above, Adam R. Cowles, Monica 
Ghosh, R. Gifford Howland, Michael McAtee, Richard Seldin, Ross 
Tuttleman, Roberto Walton, and Timothy Wedding made significant 
contributions to this report. 

(320327): 

FOOTNOTES 

[1] Both values are expressed in constant 2004 dollars. 

[2] In U.S. law, an NME is a country that does not operate on market 
principles "so that sales of merchandise in such country do not reflect 
the fair value of the merchandise." 19 U.S.C. §1677(18). China is one 
of 12 countries that Commerce has determined is an NME. Commerce 
classified China as an NME in Final Determination of Sales at Less than 
Fair Value: Natural Menthol from the People's Republic of China, 46 
Fed. Reg. 24614 (May 1, 1981). 

[3] H.R. Rep. No. 108-401, at 574 (2003), accompanying the Consolidated 
Appropriations Act, 2004, Pub. L. No. 108-199, 118 Stat. 3, 65. 

[4] GAO, U.S.-China Trade: Textile Safeguard Procedures Should be 
Improved, GAO-05-296 (Washington, D.C.: Apr. 4, 2005); GAO, U.S.-China 
Trade: Commerce Faces Practical and Legal Challenges in Applying 
Countervailing Duties, GAO-05-474 (Washington D.C.: June 17, 2005); 
GAO, U.S.-China Trade: US-China Trade: The United States Has Not 
Restricted Imports Under the China Safeguard, GAO-05-1056 (Washington, 
D.C.: Sept. 29, 2005). 

[5] We also collected and examined data on duties imposed against other 
NME countries. Appendix III briefly discusses our analyses of this 
data. 

[6] U.S. antidumping duty laws are found in 19 U.S.C. §§1673 et seq. 
The general international framework for application of antidumping 
duties can be found in the WTO Agreement on Implementation of Article 
VI of the General Agreement on Tariffs and Trade 1994--commonly known 
as the "Antidumping Agreement"--and in article VI of the General 
Agreement. 

[7] For example, 19 U.S.C. §1673a sets forth criteria for initiating an 
AD investigation. It calls for petitions to be filed "by or on behalf 
of" the affected domestic industry and requires petitioners to submit 
"reasonably available" information to support their allegations that 
dumping has occurred and that they have suffered injury as a result. 

[8] U.S. law defines material injury as harm that is "not 
inconsequential, immaterial, or unimportant." 19 U.S.C. §1677(7)(A). 

[9] Under U.S. law, Commerce sets antidumping duties equivalent to 
dumping margins. 19 U.S.C. §1673. WTO rules encourage, but do not 
require, member states to apply antidumping duty rates that are less 
than the identified dumping margins if such lower rates are found 
sufficient to remedy the injury suffered by the domestic industry. WTO, 
Agreement on Implementation of Article VI of the General Agreement on 
Tariffs and Trade 1994, Art. 9.1. 

[10] 19 U.S.C. §1677a(a). 

[11] 19 U.S.C. §1677b establishes Commerce' authority to apply these 
approaches to establishing normal value. Article 2.2 of the WTO 
antidumping agreement also provides that normal value determinations 
shall generally be based on a product's sales price in its home market 
but permits investigating officials to refer to prices in another 
export market or to "constructed" prices, in certain circumstances. 

[12] 19 U.S.C. §1677f-1(c)(1) provides that, as a general rule, 
Commerce shall determine a dumping margin for each known exporter and 
producer of the subject merchandise. Article 6.10 of the WTO 
antidumping agreement is similar. 

[13] Commerce is required to verify all of the information that it 
employs in making its final AD determinations. 19 U.S.C. §1677m(i). 

[14] 19 U.S.C. §1677e provides that if necessary information is not 
available on the record, or an interested party withholds requested 
information, fails to provide it by the deadline for submission, 
submits information that cannot be verified, or otherwise significantly 
impedes an investigation, then Commerce shall use "facts otherwise 
available" to make its determinations. Article 6.8 of the WTO 
Antidumping Agreement specifies that when an interested party "refuses 
access to, or otherwise does not provide, necessary information ..or 
significantly impedes the investigation" the importing country may base 
AD duty rates on the facts available. Annex II of the agreement 
elaborates on this basic point. 

[15] 19 U.S.C. §1677e(b). 

[16] Final Determination of Sales at Less Than Fair Value: Certain Hot-
Rolled Carbon Steel Flat Products, Certain Cold-Rolled Carbon Steel 
Flat Products, Certain Corrosion-Resistant Carbon Steel Flat Products, 
and Certain Cut-to-Length Carbon Steel Plate from Brazil, 58 Fed. Reg. 
37091 (July 9, 1993). 

[17] 19 U.S.C. §1677b(c). This provision specifies that surrogate 
countries should be (1) at a level of economic development comparable 
to the NME in question and (2) a significant producer of the product 
being examined. Wage rates are determined by reference to wages 
prevailing in market economy countries at the per capita income level 
of the NME country being investigated. 19 C.F.R. §351.408(3). 

[18] 19 U.S.C.§1677b(c)(2) states that if the available information 
does not permit Commerce to apply this methodology, Commerce may base 
its normal value determination on the price at which a comparable 
product, produced in a market economy country at a level of development 
similar to the exporting country, is sold in other countries, including 
the United States. In practice, however, Commerce has never resorted to 
this alternative. 

[19] Commerce first developed and applied this test in a 1991 case 
involving sparklers from China--Final Determination of Sales at Less 
Than Fair Value: Sparklers from the People's Republic of China, 56 Fed. 
Reg. 20588, (May 6, 1991). Commerce elaborated further upon its 
criteria in Final Determination of Sales at Less than Fair Value: 
Silicon Carbide from the People's Republic of China, 59 Fed. Reg. 22585 
(May 2, 1994). 

[20] Department of Commerce, Import Administration Policy Bulletin 
Number 05.1 (Washington, D.C.: Apr. 5, 2005). 

[21] 19 U.S.C. § 1677f-1(c)(2) provides that if it is not practicable 
to establish individual rates for all known exporters or producers, 
Commerce may limit its individual examinations to a statistically valid 
sample of companies or to those companies that account for the largest 
volume of the subject merchandise. Article 6.10 of the WTO Antidumping 
Agreement provides that when large numbers of exporters are involved, 
authorities may limit their examination to a reasonable number of 
parties. 

[22] 19 U.S.C. §1673d(c)(5)(A). This section also specifies that in 
making such calculations Commerce shall exclude any duty rates that 
were determined entirely by applying facts available, as well as any 
rates that are zero or de minimis (less than 2 percent). Article 9.4 of 
the WTO antidumping agreement is similar. 

[23] Commerce did this in 15 of the 68 China cases that we examined. 
For example, in one recent case Commerce calculated individual rates 
for 7 Chinese companies and also assigned a weighted average duty rate 
to 19 other Chinese companies that passed Commerce's separate rates 
test but were not fully examined. See Notice of Final Determination of 
Sales at Less Than Fair Value: Polyethylene Retail Carrier Bags from 
the People's Republic of China, 69 Fed. Reg. 34125 (June 18, 2004). As 
in market economy cases, weighted average calculations exclude 
individual duty rates that are determined entirely by applying facts 
available, or that are or de minimis. 

[24] When Commerce identifies all NME exporters and producers, when all 
of these companies cooperate, and when Commerce can establish 
individual rates for each company, Commerce establishes country-wide 
rates that are weighted averages of the individual company rates. 
However, such cases are not common. Commerce employed this alternative 
in only 5 (about 7 percent) of the 68 AD orders it issued against China 
from 1980 through 2004. For an example, see Notice of Final 
Determination of Sales at Less than Fair Value: Furfuryl Alcohol from 
the People's Republic of China, 60 Fed. Reg. 22544 (May 8, 1995). 

[25] During the 1980s, Japan was the single most frequently targeted 
country. About 17 percent of all the AD orders that Commerce issued 
during the decade applied to Japan. 

[26] China had the highest percent of investigations that resulted in 
orders (62 percent of 110 investigations) of any country subject to 
more than 15 AD orders. Japan (58 percent), Korea (51 percent), Taiwan 
(47 percent), and Germany (38 percent) followed China. Some countries, 
such as Latvia and Bangladesh, were subject to only one or two 
investigations and all of them resulted in orders, giving them a 100 
percent rate. Ukraine, another NME, was subject to 13 antidumping 
orders, and 69 percent of them resulted in orders. 

[27] We reviewed all antidumping orders between 1980 and 2004 and 
identified, from the 68 orders put in place against China, any in which 
there was also a corresponding order against the same product from at 
least one market economy within one year of the order against China. We 
found similar market economy orders for 25 of the 68 orders against 
China. See appendix I for more information. 

[28] For details on this case, see the duty orders as follows: 69 Fed. 
Reg. 48201 (China), 69 Fed. Reg. 48203 (Malaysia), and 69 Fed. Reg. 
48204 (Thailand), all published on Aug. 9, 2004. 

[29] This was not the case in two of the 25 cases that we examined in 
detail. In these cases, Commerce calculated individual rates for two 
Chinese companies and, since these companies represented all known 
exports of the subject products to the United States, calculated a 
weighted average of these two rates. See, for example, Notice of Final 
Determination of Sales at Less Than Fair Value: Furfuryl Alcohol From 
the People's Republic of China, 60 Fed. Reg. 22544 (May 8, 1995). In 
addition, in 3 of the other 43 orders against China Commerce calculated 
a weighted average of individual rates rather than a country-wide rate. 

[30] See Notice of Final Determination of Sales at Less Than Fair 
Value: Carbazole Violet Pigment 23 from the People's Republic of China, 
69 Fed. Reg. 67304 (Nov. 17, 2004). In some cases, Commerce has set a 
country-wide rate equivalent to the duty rate that it calculated for 
one fully investigated company. Appendix II provides more information 
on these cases. 

[31] In certain cases, Commerce has used facts-available-based 
individual company rates to establish all-others rates. 19 U.S.C. 
§1673d(c)(5)(B) provides that if all of the individual rates in a case 
were determined through application of facts available (or are zero or 
de minimis) then Commerce may use "any reasonable method" to establish 
the all-others rate. 

[32] These averages are based on a subset of the 25 products in which 
individual company rates were calculated for both China and at least 
one market economy. We found 18 products in which this was the case. 
However, the average duty rates for individual companies from China and 
market economies are still similar for the full set of 25 products--the 
average duty rate for individual Chinese companies was 52 percent 
compared with 48 percent for market economy companies. 

[33] To determine whether our results held for all NMEs, we collected 
and performed regression analyses including data on all AD actions from 
1980 through 2004 in which Commerce applied duties to both a nonmarket 
economy other than China--such as Ukraine--and at least one market 
economy country. This increased the number of products we analyzed from 
25 to 31. 

[34] GAO-05-474, pages 10-15. 

[35] 19 U.S.C. § 1677(18). The criteria include the extent to which the 
country's currency is freely convertible and the extent to which wage 
rates are determined by free labor-management bargaining. 

[36] Since 1993, Commerce has reclassified Russia and nine other 
formerly communist countries as market economies. 

[37] According to Commerce, the working group has held two meetings--in 
July 2004 and May 2005. The United States and China established the 
Joint Commission in 1983 to serve as a forum for high-level dialogue on 
bilateral trade issues. 

[38] For details, see Sulfanilic Acid from the Peoples Republic of 
China, 57 Fed. Reg. 9409 (Mar. 18, 1992). Commerce's criteria for 
designation as a market-oriented industry are (a) virtually no 
government involvement in setting prices or amounts to be produced, (b) 
industry characterized by private or collective (not state) ownership, 
and (3) market-determined prices for all significant inputs whether 
material or nonmaterial (e.g., labor and overhead). 

[39] WTO Protocol on the Accession of the People's Republic of China, 
Art. 15(a). 

[40] WTO Protocol on the Accession of the People's Republic of China, 
Art. 15(d). China acceded to the WTO in December, 2001. The protocol 
also specifies that countries determining that market economy 
conditions prevail in China as a whole or in individual Chinese 
industries will cease applying third-country information in AD duty 
investigations against the country as a whole or against such 
industries. This provision is similar to U.S. trade law. 

[41] The General Agreement on Tariffs and Trade, Annex I, Ad Art. VI, 
para. 1.2. 

[42] The duty rate reported here (17 percent) differs slightly from the 
rate reported in table 2 (16 percent) since this statement refers to 
the full 25 cases and the table refers just to the 18 cases in which 
individual rates were calculated for both China and at least one market 
economy. 

[43] Companies may decline to cooperate, at least in part, because 
analysis of their own pricing practices leads them to conclude that 
cooperation will only result in AD duty rates that are as high, or 
higher, than those that would be imposed by application of adverse 
inferences. 

[44] While Commerce may apply adverse inferences to calculate 
individual rates in China cases, it does so infrequently. For example, 
in the 25 comparable cases Commerce applied adverse influences in only 
3 of 50 individual rate determinations for Chinese companies. Commerce 
officials stated that this may be due to Chinese companies that might 
otherwise have adverse inferences applied to them declining to 
participate in the investigation, and thus accepting application of the 
country-wide rate. 

[45] Regression analyses employing variables to account for (1) 
application of a country-wide rate, (2) whether or not the duty was 
applied to China or a market economy, (3) the year of application, (4) 
whether or not Commerce applied adverse inferences, and (5) the type of 
product involved, taken together, accounted for 50 percent of the 
variation in duty rates. Appendix III provides more information on 
these results. 

[46] One factor that remains poorly documented and is not reflected in 
our analysis is the effect that any Chinese government subsidies may 
have on Chinese export prices. For more discussion on this topic see 
GAO-05-474, pages 18-19. 

[47] For details, see Notice of Final Determination of Sales at Less 
than Fair Value and Negative Final Determination of Critical 
Circumstances: Certain Color Television Receivers from the People's 
Republic of China, 69 Fed. Reg. 20594 (Apr. 16, 2004). 

[48] Commerce provides internet access to all of its determinations 
since July, 1995, and to detailed information on its antidumping 
procedures at http://ia.ita.doc.gov/ia-decisions-and-data.html. 

[49] The 68 cases resulted in 72 individual orders since the hand tools 
case included four separate types of hand tools and the ceiling and 
oscillating fans case involved orders against both ceiling and 
oscillating fans. 

[50] There were more than 68 country-wide rates because in two cases 
Commerce calculated more than 1 country-wide rate for China. For hand 
tools (1991), Commerce calculated 4 separate country-wide rates based 
on the four different types of hand tools under investigation. 
Similarly, for oscillating and ceiling fans (1991), Commerce issued 2 
separate country-wide rates--1 for oscillating fans and 1 for ceiling 
fans. Technically, these two cases resulted in six separate orders, 
although we count them as two orders. Since our analysis includes all 
of the rates in these six orders, the results are not affected either 
way. 

[51] Notice of Final Determination of Sales at Less Than Fair Value: 
Non-Malleable Cast Iron Pipe Fittings from the People's Republic of 
China, 68 Fed. Reg. 7765 (Feb. 18, 2003). 

[52] Notice of Final Determination of Sales at Less than Fair Value: 
Glycine from the People's Republic of China, 60 Fed. Reg. 5620 (Jan. 
30, 1995). In two cases, Chinese companies did provide information but 
nonetheless did not receive an individually determined rate. In the 
first case, Chinese companies did not request consideration for an 
individual rate. In the second, Commerce denied eligibility on the 
grounds that the company in question was a state-owned enterprise. See 
Notice of Final Determination of Sales at Less than Fair Value: Pure 
Magnesium and Alloy Magnesium from the People's Republic of China, 60 
Fed. Reg. 16437 (Mar. 30, 1995) and Final Determination of Sales at 
Less than Fair Value: Certain Compact Ductile Iron Waterworks Fittings 
and Accessories Thereof from the People's Republic of China, 58 Fed. 
Reg. 37908 (July 14, 1993). 

[53] Commerce officials clarified they have recently stopped including 
in duty orders the names of individual companies that have been 
assigned the country-wide rate because they have failed the separate 
rates test. However, Commerce still lists individual companies that 
pass the separate rates test but for other reasons receive an adverse 
inference-based rate--which may be the same as the country-wide rate. 

[54] Notice of Final Determination of Sales at Less than Fair Value: 
Refined Brown Aluminum Oxide from the People's Republic of China, 68 
Fed. Reg. 55589 (Sept. 26, 2003). No other countries were included in 
this investigation. 

[55] See Bruce Blonigen, "Evolving Discretionary Practices of U.S. 
Antidumping Activity" National Bureau of Economic Research, Working 
Paper #9625 (April 2003). 

[56] In other analyses in which we added variables for the types of 
product involved, the variable "year" becomes insignificant. Overall, 
we found mixed evidence of whether there is a positive trend over time 
in duty rates for this group of cases. (See appendix II.) 

GAO's Mission: 

The Government Accountability Office, the investigative arm of 
Congress, exists to support Congress in meeting its constitutional 
responsibilities and to help improve the performance and accountability 
of the federal government for the American people. GAO examines the use 
of public funds; evaluates federal programs and policies; and provides 
analyses, recommendations, and other assistance to help Congress make 
informed oversight, policy, and funding decisions. GAO's commitment to 
good government is reflected in its core values of accountability, 
integrity, and reliability. 

Obtaining Copies of GAO Reports and Testimony: 

The fastest and easiest way to obtain copies of GAO documents at no 
cost is through the Internet. GAO's Web site ( www.gao.gov ) contains 
abstracts and full-text files of current reports and testimony and an 
expanding archive of older products. The Web site features a search 
engine to help you locate documents using key words and phrases. You 
can print these documents in their entirety, including charts and other 
graphics. 

Each day, GAO issues a list of newly released reports, testimony, and 
correspondence. GAO posts this list, known as "Today's Reports," on its 
Web site daily. The list contains links to the full-text document 
files. To have GAO e-mail this list to you every afternoon, go to 
www.gao.gov and select "Subscribe to e-mail alerts" under the "Order 
GAO Products" heading. 

Order by Mail or Phone: 

The first copy of each printed report is free. Additional copies are $2 
each. A check or money order should be made out to the Superintendent 
of Documents. GAO also accepts VISA and Mastercard. Orders for 100 or 
more copies mailed to a single address are discounted 25 percent. 
Orders should be sent to: 

U.S. Government Accountability Office 

441 G Street NW, Room LM 

Washington, D.C. 20548: 

To order by Phone: 

Voice: (202) 512-6000: 

TDD: (202) 512-2537: 

Fax: (202) 512-6061: 

To Report Fraud, Waste, and Abuse in Federal Programs: 

Contact: 

Web site: www.gao.gov/fraudnet/fraudnet.htm 

E-mail: fraudnet@gao.gov 

Automated answering system: (800) 424-5454 or (202) 512-7470: 

Public Affairs: 

Jeff Nelligan, managing director, 

NelliganJ@gao.gov 

(202) 512-4800 

U.S. Government Accountability Office, 

441 G Street NW, Room 7149 

Washington, D.C. 20548: