rate in the “present” is higher than the MFN applied rate from the past. To control for these errors, we
assume no change in the preferential rate from past to present whenever the present preferential rate is
higher than the past MFN applied rate. For this exercise, we used the preferential rates for the United
States and its partners under the Mexican and Canadian segments of NAFTA, the Australia-US FTA, and
the Singapore-US FTA. Preferential rates are displayed in tables A.39 through A.46.
Nontariff Barriers
Efforts by scholars to estimate ad valorem tariff equivalents of NTBs—the data we need to analyze
NTBs in GTAP—have been limited. Ferrantino (2006) surveys the work that has been done in this
field; Deardorff and Stern (1997) provide an earlier assessment of NTB data work. Creating ad valorem
equivalents of NTBs involves considerable guesswork. In general, authors try to determine the level of
NTB protection either from the wedge between domestic and international prices caused by the NTB
or from the shortfall in expected imports caused by the NTB. The level of sophistication varies widely
between estimates, and most efforts have been limited to either a few countries or a few sectors. However,
a recent database published by the World Bank, created by Kee, Nicita, and Olarreaga (2005), provides
ad valorem equivalents at the HS 6-digit level for over 4,500 commodities for 91 countries. Their
approach is to “predict import [values] using factor endowments and observe [the] deviations in the
presence of NTBs” (Kee, Nicita, and Olarreaga 2005). The authors then convert the deviations to price
effects to calculate ad valorem tariff equivalents of each NTB for each country.
The underlying data for the Kee, Nicita, and Olarreaga (2005) estimates of ad valorem tariff
equivalents of NTBs is complied from the UNCTAD TRAINS database, various WTO Trade Policy
Reports, a European Union dataset created by the Groupe d’Economie Mondiale at Sciences Po (Paris),
and notifications from WTO members of their domestic support programs. The following types of NTBs
are included in the analysis: nonautomatic licenses, quotas, prohibitions, administrative pricing, voluntary
export price restraints, variable charges, monopolistic measures, technical regulations, and domestic
support subsidies.7 Estimates of ad valorem equivalents of NTBs are made for one year for each country
using data from the most recent year available. The underlying NTB data roughly corresponds to the year
2000 for every country we consider; other data in their model (e.g., tariffs and trade) is more recent.
Using a concordance between HS 1996 and GTAP codes provided by the World Bank, we collapse
the Kee, Nicita, and Olarreaga (2005) NTB estimates, which are provided at the HS 6-digit level, into
7. The UNCTAD codes for these NTBs are as follows: nonautomatic licenses (6100), quotas (6200), prohibitions (6300),
administrative pricing (3100), voluntary export price restraints (3200), variable charges (3300), monopolistic measures
(7000), and technical regulations (8000). Domestic support subsidies are not included in the UNCTAD coding. More
specific NTBs are listed under each of these parent codes; see Ferrantino (2006) for a complete list. Variable charges are
often included in national tariff schedules; however, they usually cannot be converted into ad valorem equivalents, so
double counting between the ad valorem equivalents of applied tariff rates and NTBs is unlikely (Stawowy 2001).
43
More intriguing information
1. Investment and Interest Rate Policy in the Open Economy2. An Attempt to 2
3. Does adult education at upper secondary level influence annual wage earnings?
4. Concerns for Equity and the Optimal Co-Payments for Publicly Provided Health Care
5. Strategic Effects and Incentives in Multi-issue Bargaining Games
6. XML PUBLISHING SOLUTIONS FOR A COMPANY
7. TINKERING WITH VALUATION ESTIMATES: IS THERE A FUTURE FOR WILLINGNESS TO ACCEPT MEASURES?
8. Firm Closure, Financial Losses and the Consequences for an Entrepreneurial Restart
9. Poverty transition through targeted programme: the case of Bangladesh Poultry Model
10. Novelty and Reinforcement Learning in the Value System of Developmental Robots