Transparency in non‐tariff measures: An international comparison

• Author: Lili Yan Ing,Olivier Cadot,Janine Walz
• Published date: 01 March 2018
• DOI: http://doi.org/10.1111/twec.12552
• Date: 01 March 2018

ORIGINAL ARTICLE

Transparency in non-tariff measures:

An international comparison*

Lili Yan Ing

1

|

Olivier Cadot

2

|

Janine Walz

3

1

Economic Research Institute for ASEAN and East Asia (ERIA), and University of Indonesia, Jakarta, Indonesia

2

Faculty of Business and Economics, Universit

e de Lausanne, CEPR and FERDI, Dorigny, Switzerland

3

The World Bank, Washington, DC, USA

1

|

INTRODUCTION

Transparency is fundamental to modern administrative law in many industrial countries and is key

to prevent arbitrariness and capture by special interests. Transparency in administrative rules and

processes is also crucial for transactions that involve lengthy and complex procedures like interna-

tional shipments and is thus particularly relevant as a trade facilitator; as non-tariff measures

(NTMs) are often complex legal instruments, transparency is particularly important in their case.

Indeed, a growing literature has shown that improved transparency can generate substantial

gains in trade and investment flows (Francois, 2001; Helble, Shepherd, & Wilson, 2009; Kerr,

2008; Lej

arraga & Shepherd, 2013; Wolfe, 2003). In these papers, the determinants of trade vol-

umes typically follow the gravity equation, while the causal effects of transparency on trade are

identified using variation either in perception-based transparency indices across countries or in

transparency provisions across regional trade agreements (RTAs).

In spite of growing interest in measuring transparency in trade policy, most of the proxies used

in the literature are broad in scope, relating to general perceptions of government transparency or

to “WTO+”provisions in RTAs. In this paper, we follow a conceptual framework laid out in

Wolfe (2013) which identifies three “generations”of transparency requirements at the WTO. The

first generation, essentially “right-to-know”provisions adopted in 1947 and elaborated over subse-

quent years, includes, most importantly for us, notification requirements. The second generation,

which includes monitoring and surveillance mechanism such as trade policy reviews, is more

specific to the WTO and of less direct concern to us. The third generation covers not just the pro-

duction of information, but also its dissemination, in particular through web-based instruments, and

is a key part of our analysis through NTM inventories and web-based portals.

Based on this framework, rather than relying on perception-based indices, we adopt a more direct

approach drawing on what governments actually do in the area of NTM transparency. We draw on

various sources, including “off-the-shelf”data, on country-level compliance with WTO requirements.

For instance, among we use a normalised count of notifications of potentially trade-restricting

*Support from ERIA under project ERIA-RD/RA-001-001-406 is gratefully acknowledged. Cadot also gratefully acknowledges

support from France’s Agence Nationale de la Recherche under “Investissement d’Avenir”grant ANR-10-LABX-14-01.

DOI: 10.1111/twec.12552

884

|

©2017 John Wiley & Sons Ltd wileyonlinelibrary.com/journal/twec World Econ. 2018;41:884–912.

measures (a general requirement for all WTO members), a first-generation transparency instrument

in Wolfe’s terminology. We also use the creation of a trade portal (a requirement under the recent

Trade Facilitation Agreement), or the public availability, in the World Bank’sWITSportal,of

NTM inventories under the MAST classification in the new UNCTAD database, third-generation

transparency instruments in Wolfe’s terminology. As the focus of the paper is on NTMs, which

cover goods, we do not include measures covering trade in services. Our off-the-shelf variables

coverawiderangeofNTMs,includinginter alia, licensing, quantitative restrictions (QR), anti-

dumping and countervailing duties, price-control measures, preshipment inspection, and restric-

tions on government procurement markets, and our experimental question is relatively open-ended,

asking about technical regulations and any licensing or other requirement. However, we focus the

discussion on sanitary and phytosanitary (SPS) and technical barriers to trade (TBT) measures,

which account for the bulk of both notifications and MAST inventories and are handled by SPS/

TBT enquiry points.

In order to get closer to what really happens on the ground, we combine these measures with

the results of an original experiment, in which a Swiss producer of food containers and logistics

solutions sent a standardised request on relevant import regulations to trade agencies and SPS/

TBT enquiry points (the creation of such enquiry points is a first-generation transparency

requirement in Wolfe’s terminology) to 187 countries in the world. The information requests

were in French to French-speaking countries, in Spanish to Spanish-speaking countries and in

English to all other ones. They were sent in repeated waves in order to smooth out cases of

non-response due to temporary unavailability of particular staff members or other non-persistent

factors. When responses were received, we rated them by their quality; for instance, some agen-

cies simply acknowledged reception, while, at the other extreme, Slovenia sent a highly detailed

technical explanation that was translated into German (the Swiss company’s home language) by

the Slovenian embassy in Switzerland. The different components of our transparency index are

not highly correlated, as some countries may, for example, lack the administrative capability to

identify the right measures to notify to the WTO while willing to respond to a company’s

request for information.

1

We then aggregated the experimental scores with those under the other criteria, using as

weights the variable loadings from principal factor analysis, and then ranked countries by decreas-

ing order of the index value. As our scores have a limited range of integer values, the resulting

index has many ties. Within those ties, rankings are arbitrary; in order to lift the indeterminacy,

based on the observation that transparency, correlates with income levels, we rank countries by

decreasing order of GDP per capita.

Our index correlates with the government transparency score of the World Economic Forum’s

Global Competitiveness Index. It also reveals plausible but nevertheless interesting variation across

income and regional groups. Unsurprisingly, OECD countries typically have the highest scores;

non-OECD high-income countries—which include a number of oil producers—have low ones. The

lowest scores are observed in low-income and lower-middle-income countries. Among developing

countries, ASEAN has the highest average scores, followed (albeit which much more heterogene-

ity) by European and Central Asian countries. The lowest scores are observed in Middle East and

North Africa (MENA) and sub-Saharan African countries.

The rest of the paper is organised as follows. Section 2 briefly reviews the literature on trans-

parency and trade. Section 3 details the construction of the index. Section 4 analyses the properties

of the resulting ranking. Section 5 concludes.

1

We are grateful to a referee for attracting our attention to this.

ING ET AL.

|

885