Productivity, Non-Tariff Measures and Openness

Final Report Summary - PRONTO (Productivity, Non-Tariff Measures and Openness)

Executive Summary:
The PRONTO (Productivity, Non-Tariff Measures and Openness) project has been concerned with the policy relevance of non-tariff measures (NTMs) affecting international trade and investment flows. It was a four-year collaborative project funded under the FP7 thematic area “SSH.2013.4.3-3: Untapped potential for growth and employment.” The project ran from 1 February 2014 until 31 January 2018. Project research has been focused on two broad sets of questions: (i) appropriate and meaningful measurement of NTMs; and (ii) the impact of NTMs on economic performance, social outcomes, and sustainability. In an effort to answer those questions, the PRONTO project team has been engaged a mix of regulatory data analysis, econometric analysis, applied theoretical analysis, and numerical model-based assessments of NTMs.

NTMs are a primary focus of the European Union’s external trade and investment policy strategies, and for this reason the research conducted under PRONTO has socio-economic policy relevance, in addition to the scientific outcomes of the project. The final publishable summary report brings together summary findings on the two broad sets of questions identified above.

The results of the project with respect to measurement include: (i) identification of existing methods and extension of those methods for better classification of NTMs; (ii) collection and organization of existing data and generation of new data on NTMs; (iii) mapping from raw regulatory data to policy-focused quantitative measures; (iv) and identification and analysis of broad patterns of NTM incidence, in terms of countries, regions, and sectors. The result of this research both fed into project work on the impact of NTMs, and into a publicly available dataset that reflects extensive processing, clean-up, and integration of individual datasets, and development of new composite measures based on principal component analysis (PCA) of the full, detailed (HS6) set of individual NTM indicator data series. The PCA measures combine information both unique to individual source data and following from the combined patterns of these data collectively. In this way, the project was able to compensate for significant variations in coverage and focus across source data, providing a new basis for identifying key measures in particular sector across countries, and key NTM users in particular sectors identifying and integrating information from multiple data sources for use in ranking and prioritizing NTMs.

While the analysis of NTM incidence provide guidance on what type of NTMs are present in which countries and sectors, the outcomes of project research with respect the
impact of NTMs on relevant socio-economic indicators identify expand our understanding with respect to the following NTM impact dimensions: (i) integration of NTM measurements in frameworks for quantitative impact assessments; (ii) quantifying the impact of NMTs on trade and value chains; (iii) quantifying the impact of NTMs on economic and social objectives; and (iv) quantifying the impact of NTMs on sustainable growth and climate related policy objectives and related indicators.
Project Context and Objectives:
Over the past fifty years, there has been significant progress in lowering tariff barriers to international trade. This has led to a growing awareness of the importance of what are termed Non-Tariff Measures (NTMs). In general terms, NTMs are all measures that can have an impact on trade and/or investment but that are not import tariffs. They are often referred to as “behind the boarder measures” that have a bearing on the production of foreign firms in other countries. Increasingly, they are also central to the negotiation of trade and investment treaties. Unlike tariffs, NTMs imposed by one country can affect the structure of production activities of the firms of other countries, irrespective of whether the measures are actually applied at the border.

In addition, NTMs can be imposed for perfectly valid reasons that have little or nothing to do with trade. Examples include measures taken for consumer health and safety reasons. To the extent these measures affect market access by foreign suppliers, NTMs take on a role as trade and investment non-tariff barriers, or NTBs. The barrier aspect of policy measures may be unintentional, for example when different regulatory solutions are applied for the same objective, leading to duplication of compliance costs. In other cases, NTBs may reflect a deliberate effort to restrict market access, for example by setting standards that can only really be met by domestic suppliers, or by imposing restrictive import license requirements. Using the term NTB rather than NTM highlights the negative market access aspect of regulatory measures.

The medium-scale focused research project "Productivity, Non-Tariff Measures and Openness" (PRONTO) has aimed for better data, better methodologies, and better understanding of the impact of NTMs on international investment and trade. The project team has been focused on the following over-arching objectives:

1. Collection of quantitative information on the regulatory measures influencing cross border trade and investment;
2. Developing new methodologies for quantifying NTMs and for both quantitative and qualitative analysis of their effects;
3. Mapping data availability, identifying scope for improvement by building on existing sources, and mapping onto other databases such as WIOD and EU-KLEMS.
4. Analysing the impact of NTMs on a range of economic and social indicators (e.g. the SDGs).

One outcome of the research targeting these objectives has been the PRONTO database itself, including its web-based portal and associated database(s). These are discussed in the core results section of the final publishable summary, Main Scientific and Technical Results and Foregrounds. A second outcome has been improved understanding of the ways in which NTMs interact with one another, again discussed in the core results section of the final publishable summary. A third outcome has been new evidence on the impact and relevance of NTMs, again as discussed in the core results section of the final publishable summary. This has involved use of the data developed within the project to empirically estimate the effects of NTMs on a variety of social and economic outcomes, including: their impact on income and inequality in the EU; their role in promoting sustainable growth in developing countries; the effects they have on technological growth; and the frictions they create in the global supply chain. Overall, these project outcomes provide both improved understanding of NTMs and their effects, allowing for more meaningful policy recommendations, and new data and methodologies suited for continued (post-PRONTO) work on international trade and investment policy.

The evidence from PRONTO, as summarized in the core results section of the final publishable summary, also points not only to areas where NTMs are clearly protectionist, but also to areas where NTMs may actually enhance market access, and areas where objectives are non-economic and where there is little or no impact on market access either way. This particular outcome makes it clear that setting public policy in this area requires that the benefits of measures be considered when imposed for public policy reasons (and if they are not only identified as disguised protectionist measures.) Indeed, NTMs can sometimes open up markets, thereby functioning as trade catalysts, even for producers in developing countries, where the trade-restricting effect of NTMs has usually been emphasized.

With respect to the first project objective, collection of quantitative information on the regulatory measures influencing cross border trade and investment, outputs have targeted: consolidation of the status quo of the data and methodologies used in NTM research; integration of existing primary data with new data developed during the project; use of these data on later elements of the overall project; and development of a final, end of project, public release dataset on NTMs. This process has translated into a platform provided as an online portal where the information is made easily accessible, i.e. a one-stop shop. It has also yielded a collaboration network, including several international organizations, that will allow for future updating of this resource.

With respect to the second and third over-arching project objectives, the focus of work has been on: expanding the state of the art by improving the quality and types of NTM data and methodologies; and connecting data on NTMs to other relevant socio-economic indicators; and establishing a database on non-tariff measures (NTMs) allowing one to analyse the effects of NTMs on cross-border trade in intermediate products, final goods, investment goods, and services trade. This has included linkages to the WIOD database and GTAP modelling framework, development of new composite NTM measures based on principal component analysis (PCA) of detailed, sector based NTM indicators from multiple sources, and development of new guidelines and methodologies for estimating the cost impact of NTMs The PCA measures combine information unique to individual source data, which otherwise vary in coverage and focus, providing a basis for identifying key measures across countries and sectors. The linkages to WIOD and GTAP data (which are global input-output databases) have been critical to work targeting the fourth over-arching objective, analysing the impact of NTMs on a range of economic and social goals.

With respect to the fourth of the PRONTO over-arching objectives, the outcomes of project research expand scientific understanding, but also with policy relevant findings, as discussed in detail the core results section of the final publishable summary. These include the new evidence on the following: the impacts of NTMs on vertical specialisation, trade, and competitiveness (productivity) in global supply chains for goods and services; the impact of NTMs at the firm level and particularly on the composition of exporters; the impact of NTMs, as well as possible changes in NTMs, on EU economic and social objectives; and
the broader impacts that NTMs have on society through the nexus of poverty reduction and environmental impacts.

The broad scope of project findings, with discussion of their scientific and societal relevance, is summarized in the part of the publishable summary report titled Main Scientific and Technical Results. The reader of this summary is also encouraged to follow up with the individual papers and reports from the project, available at the website:

http://www.prontonetwork.org

In addition, the reader is invited to explore the rich data on NTMs that are also available at the data portal on the PRONTO website.

Project Results:
-------
Please see attached pdf file, "4.1.3_main_S&T_results_foregrounds.pdf" for a formatted version of this text, with tables and charts.
-------

1. INTRODUCTION

This section is the “Description of the main S & T results/foregrounds” of the publishable summary report for the medium-scale focused research project "Productivity, Non-Tariff Measures and Openness" (PRONTO). The PRONTO project has aimed for better data, better methodologies, and better understanding of the impact of NTMs on international investment and trade. The project team has been focused on the following over-arching objectives:

Objective 1. Collection of quantitative information on the regulatory measures influencing cross border trade and investment;
Objective 2. Developing new methodologies for quantifying NTMs and for both quantitative and qualitative analysis of their effects;
Objective 3. Mapping data availability, identifying scope for improvement by building on existing sources, and mapping onto other databases such as WIOD and EU-KLEMS.
Objective 4. Analysing the impact of NTMs on a range of economic and social indicators (e.g. the SDGs).

In this section we provide a relatively non-technical description of the objectives and outcomes of the entire project. Referenced papers, data, and related documents are available from the project website: www.prontonetwork.org. The discussion is organized following the four over-arching project objectives outlined above.

2. Objective 1: Collection of quantitative information on the regulatory measures influencing cross border trade and investment

Project Outcome: consolidation of data and sources

This research stream involved two goals: (i) generation of internal project data for use within the project; and (ii) generation of data and utilities for the end-of-project public dataset. The first goal was met with internal dissemination of data reflecting the following:
- Raw data compilation;
- Documentation/manuals of the various databases (if available);
- Cleaned datasets and corresponding scripts.
Following the initial compilation of raw data beginning in 2014 (i.e. the first year of the project), an update of the data was conducted mid-2017, including rerunning and testing the cleaning scripts developed in the work package. This served as a basis for server script development for the final database as discussed below.

As part of this line of research, a first version of Rau & Vogt (2017) provides a background paper/reference document to NTM data concepts and sources. There, NTM data sources are mapped to a classification/framework for NTM data (see Table 1 of the formatted pdf version of this summary). The paper provides summaries of the main databases, highlighting the type of data they contain, underlying collection methods, as well as country and temporal data availability. It further elaborates on the nature of NTM data and possible caveats in their application. Thus, PRONTO researchers from other WPs were able to get a quick overview of NTM data before starting their research. Note that a short version of the paper is forthcoming in a Cambridge University Press volume, following presentation at the 2016 World Trade Forum.

Project Outcome: development of a data platform hosting the PRONTO public access database

Following extensive clean up and processing of external data sources, along with mapping of available regulatory data to product data (in particular, traded goods classification schemes) we have both made the information on raw agency data of NTMs available on a publicly-accessible internet-based portal and have provided access on the same portal to PRONTO datasets generated within this and other work packages. The public release database is based on the internal data backbone of the project and to a large extend has been structured to facilitate future, post FP7 updates to the data. This will ensure a lasting, living legacy and focus point for the research network formed for the project. In particular, the update process has been partially automated to facilitate future periodic updates. The team working on the public data distribution platform also worked closely with agency representatives linked to primary source data, ensuring a cooperative and continued relationship in data updates after the FP7 phase has ended.

The three main building blocks of the NTM portal are: 1) information on NTM data and research; 2) an internal data backbone that organizes datasets and the data they were built on (i.e. the Dropbox repository); and 3) access to the NTM indicators and improved datasets generated in the PRONTO project. With respect to the latter, these include those datasets that improve (e.g. imputed WTO notifications) and consolidate (e.g. environmental taxes) original data. The indicators were constructed in the various WPs (e.g. PCA-based indicators in WP4). Documentation and reference to the source data is provided on the portal.

A key feature of the public database, as completed in the last phase of the project, is that not only the datasets themselves but also other elements on the website are script-based. Consequently, period updates will be easier to administered than otherwise, once new raw data are released or NTM and sectoral classification schemes are revised. The set-up thus is a data warehouse with specific features for easy and user-friendly data access. Interactive data availability sheets constructed from the indexed raw data are provided to help searching the available data. Parties interested in conducting NTM related research therefore have a one-stop shop that is easily accessible and guides towards raw data that may be applicable to their work.

Note that while the portal provides data and information on data availability (including guides to sources on classification and institutions active in NTM research), we have also extended the portal by an information component that builds on the work conducted in other parts of the overall PRONTO project. Summary information of the research undertaken within the PRONTO project is provided on the website, and as such the Portal also serves as a dissemination platform in its own right.

3. Objective 2: Developing new methodologies for quantifying NTMs and for both quantitative and qualitative analysis of their effects

Project Outcome: benchmarking the state of the art

The project study generated by Davies, Rau & Vogt (2015) takes stock of the different methodologies applied in NTM research. As part of the stock-taking exercise, they also provide a framework to classify research according how it makes use of NTM information.

Specifically, Davies, Rau & Vogt (2015) consider the following dimensions for making an inventory of the available literature on NTMs:
- Purpose:
o Construction of alternative measures;
o Studies of outcome of NTMs;
o Studies determining NTMs.
o Literature review
- Perspective & scope.

These categories are not a methodological classification per se but they stipulate the context in which the research under review takes place. Here, for the purpose of the study, i.e. the goal the researcher had in mind when starting her research, is divided into four sub-categories.

First, the construction of an alternative measure relates to the conversion of NTM data, which is oftentimes binary, to a measure more conducive to further analysis, e.g. an ad-valorem equivalent that can be used in simulation experiments. Mostly, these measures indicate a degree of restrictiveness and are based on methods of comparing price and quantity data in the presence versus the absence of an NTM.

Second, Davies, Rau & Vogt (2015) identify methods that are used to assess the impact of NTMs on an outcome variable or set of variables. Figure 1 presents an overview of these and their identified sub-categories. The two methods most often used are econometric regressions and simulation models. Particularly, gravity regressions augmented with NTM variables (dummies 0/1) as a control have been adopted widely in the literature. Furthermore, computable general equilibrium models are often used in trade policy analysis to assess the impacts of trade agreements or tariff concessions. Other methods have not yet been widely used in NTM-related research. However, given the non-trade dimension of NTMs, cost-benefit approaches may prove to be suitable to compare possible trade costs/restrictiveness with regulatory benefits.

Third, the purpose of some studies is to examine determinants of NTMs. Such studies are found specially in the field of political science/economy. Here the measure is the dependent variable and regressed against a set of explanatory variables that explain the presence of the measure (e.g. lobbying, import competition, political cooperation, etc.). Furthermore, this would cover qualitative approaches like case study designs that explain the incidence of an NTM. Fourth and finally, literature reviews survey research on certain aspects that are not further broken down.

Davies, Rau & Vogt (2015) further differentiate between the perspective and the scope of NTM studies. These two categories further narrow down the research focus as well as the methodological options. With regard to the research focus it can be distinguished between retrospective and predictive studies (or a combination thereof). Here, regression analyses can clearly be considered as being retrospective by their nature: data of events and variable realizations that have already taken place are usually applied in regression analyses. Others, such as CGE simulations are instead used for ex-ante assessments of NTMs. Finally, the scope further defines the economic agents, sectors, and country-level focus of the study. Two general categories emerge from this: partial and general equilibrium studies. The crucial distinction is that in a partial equilibrium setting spill-over effects are not accounted for.

Project Outcome: new methodologies for extending the TRI to include NTMs

Trade Restrictiveness Indexes (TRIs) and related extensions have been frequently used in the literature on NTMs and complicated tariff schedules to derive the tariff equivalent of tariff structures, tariffs and quotas, tariffs and domestic production subsidies, or tariffs and AVEs of other NTMs. However, until recently available measures have not allowed for market structure imperfections (including market power related to country size). Too address this problem and advance the state of the art a parsimonious framework has been developed to account for external effects and corrective policies addressing these effects, in the context of a TRI. The framework has been used in an empirical investigation demonstrating its operability. There are three principal results. First, the TRI calculated with the quantitative model in a small country setting is close to the TRI calculated with an approximate formula for smaller levels of trade costs, whereas the two deviate for larger levels of trade costs. Second, the impact of differences in market structure is limited. Third, the TRI in a small country setting deviates significantly from the large country TRI, which takes changes in the terms of trade into account.

Project Outcome: new methodologies for modelling NTM policy in the presence of GVCs

Pronto-based research also addressed the important question of the economic impact of NTMs in the presence of international process fragmentation, i.e. global supply chains. This line of research yielded a new database providing cumulative bilateral-trade restrictiveness indices using the ad-valorem equivalents of NTMs and tariffs considering backward linkages. In a three-stage approach, the cumulative impacts of trade policy measures along global value chains using the world input-output database (WIOD) were quantified. The next question addressed was then how to model changes in trade policy in the presence of global value chains and non-tariff measures. To do so, Bekkers and Francois (2015) proposed a way to incorporate NTBs for the four workhorse models of the modern trade literature in computable general equilibrium models (CGEs). CGE models feature intermediate linkages and thus allow us to study global value chains. Importantly, and this has been a major computational breakthrough, they have shown that the Ethier-Krugman monopolistic competition model, the Melitz ﬁrm heterogeneity model and the Eaton-Kortum model can be deﬁned as an Armington model with generalized marginal costs, generalized trade costs and a demand externality. The authors implement the different models in a CGE setting with multiple sectors, intermediate linkages, non-homothetic preferences and, importantly, detailed data on trade costs. They rely on the Melitz model to mimic changes in non-tariff measures with a ﬁxed cost-character and by analysing the effect of changes in ﬁxed trade costs.

Project Outcome: new methods for application of border rejection data to study uncertainty

Disdier, et al. (2017) address the microeconomic impact of the risk of rejection at the EU border for Chinese exporters of food products. They combine information from the Rapid Alert System for Food and Feed with firm level export data from China by product-destination over the period 2000-2011. The results show that border rejections reduce the number of competitors, the number of varieties available to consumer and unevenly affect different-sized exporters. Thus, although border rejections may serve to improve product safety, it is likely to have additional, unintended consequences because of competition-inhibiting effects.

Project Outcome: new econometric frameworks for NTM assessment

Bekkers, Francois, and Rojas-Romagosa (2018a) provides an overview of the ways in which changes in non-tariff measures (NTMs) are mapped into welfare effects through changes in trade costs. This is done in two steps. First, methods to calculate the trade cost changes associated with changes in NTMs are discussed, concentrating on changes in NTMs as a result of the conclusion of FTAs. Second, the different ways to calculate the welfare effects of changes in trade costs are presented, compared, and assessed. In the first part the two main approaches to estimate NTM reductions associated with the implementation of FTAs are introduced and examined, the bottom-up and top-down approach. The report compares and analyses the main differences in estimates and how these differences affect the overall estimates of economic impact. One finding is that differences in expected NTM reductions can explain a large share of discrepancies in impact assessments regarding the overall potential economic effects. A second finding relates to differences in solution methods and baseline calibrations in current quantitative trade models (QTMs) used to evaluate the effects of counterfactual experiments on reductions in NTMs: computable general equilibrium (CGE) models, structural gravity (SG) models and models employing exact hat algebra (EHA). The report finds that the different solution methods generate identical results on counterfactual experiments with changes in NTMs if baseline trade shares or baseline trade costs are identical. SG models, calibrating the baseline to gravity-predicted shares, potentially suffer from bias in the predicted welfare effects due to misspecification of the gravity equation, whereas the other methods, calibrating to actual shares, potentially suffer from bias as a result of random variation and measurement error of trade flows. Simulations show that fitted shares calibration can generate large biases in predicted welfare effects if the gravity equation does not contain pairwise fixed effects or is estimated without domestic trade flows. Calibration to actual shares and to pairwise fixed effects based fitted shares display similar performance in terms of robustness to the different sources of bias.

Project Outcome: numerical frameworks for NTM assessment

The paper by Bekkers, Francois, and Rojas-Romagosa (2018b) provides an overview of numerical methods used for ex-ante policy assessment of NTMs, offering a comprehensive overview based on recent ex-ante policy assessments (particularly of the Transatlantic Trade and Investment Partnership or TTIP). These studies provide a broad range of applied numerical methods focused on non-tariff measures (NTMs), generally involving either computable general equilibrium (CGE) models or structural gravity (SG) models. In most of these studies, the bulk of the expected trade and welfare effects occur through changes in NTMs. The report compares different methodological approaches to predicting the welfare effects of trade policy experiments, with emphasis on those studies employing either CGE models or SG models. In the report the authors compare and critically discuss differences in the estimated trade cost reductions and in the economic models employed and how these can explain the relatively wide range of economic effects found in the different studies. The report also discusses the integration of firm theory with gravity estimation frameworks and the extension of the theoretical structure of CGE models with firms and NTMs.

4. OBJECTIVE 3: MAPPING DATA AVAILABILITY, IDENTIFYING SCOPE FOR IMPROVEMENT BY BUILDING ON EXISTING SOURCES, AND MAPPING ONTO OTHER DATABASES SUCH AS WIOD AND EU-KLEMS.

Outcome: GAP-Analysis (mapping our way beyond the state of the art)

As detailed in Rau & Vogt (2016), on the basis of the methodological inventory and data compiled, gaps with respect to actual raw data, as well as methods used in NTM research were identified, which also includes statements on necessary data improvements to facilitate deeper, more comprehensive NTM analysis.

Gaps in the raw data show that the geographic coverage of databases is also a function of the resource intensity of the underlying collection methodology, as well as to what degree NTM data collection is institutionalized. Thus, we see relatively low geographic coverage for large scale business surveys conducted by ITC, and fewer countries available for the full regulatory reviews of NTM TRAINS compared to WTO notifications, where countries are required to report policies to the WTO under various agreements. Other data sources’ country focus is by mandate as for example the OECD’s primary focus on OECD countries’ policies. Given the geographic scope, time coverage was also assessed. Note however that the characteristic of the time coverage can differ substantially even within databases (e.g. date of measure entering into force relative to the date of announcement and/or year of data collection, etc.). Furthermore, the coverage of measures defined by MAST chapter is highest for SPS, TBT and trade contingent protective policies. Rau & Vogt (2016) conclude with a discussion of data limitations and list a number of points where gaps in NTM data need to be filled ranging from adding more information to current measure (e.g. actual value of residue limit) to connecting NTMs to outcome variables enabling e.g. cost-benefit analysis.

In the second step, the gap analysis used information of more than a hundred studies entered into the Methodological Inventory Database for Non-tariff measures (called the MIND and described by Davies, Rau & Vogt (2015)) in order to assess the research on NTMs conducted in the last decade. The majority of studies reviewed seem to have looked at the outcome of an NTM by using regression analyses, thereby taking a retrospective perspective. Many studies focused on NTMs in goods with a particular emphasis on agri-food products, and in this regard SPS measures that are predominantly imposed on agri-food products, have been widely analysed. As mentioned above, data limitations seem to play a key role in that they restrict the use of certain methodologies; for example cost-benefit analyses that require measurement metrics and indicators such as estimates of the consumers’ willingness-to-pay in order to assess consumer benefits of higher SPS standards.

Project Outcome: NTMs, vertical specialisation and trade (linking WIOD data to NTMs)

Research on this theme developed a database for non-tariff measures (NTMs) for use by various teams in the project, and also to use these data directly examine how different types of NTMs affected global trade, how effects of NTMs can be compared across different types of NTMs and tariffs, and how important they are in the context of global value chains (GVC). The various steps in generating the database have been documented in a series of working papers (Ghodsi, Grübler, Reiter and Stehrer, 2017; Ghodsi, Grübler, and Stehrer, 2016a, 2016b; Ghodsi and Stehrer, 2017. The contributions of this work are many. First it amended the WTO Integrated Trade Intelligence Portal (I-TIP) to suit econometric analysis. Second, the effects of different types of NTMs on trade flows were estimated at the HS 6-digit product level for more than 100 countries applying a gravity approach. These results are differentiated by country and product characteristics. Third, import demand elasticities, which allow computing ad-valorem equivalents of NTMs, are estimated rendering NTMs comparable across types and with the level of tariffs. Fourth, a database linking NTMs in goods and global supply chains (based on WIOD) has been prepared, which provides bilateral trade restrictiveness indices that are used to estimate the impact of trade policy measures on labour productivity in goods and services industries. The data are part of the public database and also formed a data source for other teams.

The findings suggest: (i) Roughly 60% of all estimates point towards trade-impeding effects of NTMs, with stark differences between NTM types, where measures related to health are more likely to show positive effects than technical regulations. This is interesting as it co-validates the independently obtained results in other papers from the project. (ii) Highest average import demand elasticities are found for the economically biggest countries in their respective regions and intermediate goods, which appears particularly noteworthy in the context of global value chains. (iii) Simple average AVEs reach up to 8% for SPS measures, 11% for TBTs, and as high as 19% for Antidumping. When compared to an average tariff rate of 5%, this highlights the importance of the work carried out under the project as a whole. (iv) While the number of notifications and types of NTMs used increases with income, AVEs of richer countries seem to be lower. (v) Lowest AVEs are found for final consumption goods and highest for goods contributing to gross fixed capital formation. (vi) SPS regulations along GVCs seem to increase labour productivity, particularly in the services sector. Conversely, TBTs appear productivity decreasing, primarily in the non-services sector.

Project Outcome: better understating of subsidy and procurement policies as NTMs

The project has also yielded better information (with new data) and better understanding of NTMs related to beggar-ty-neighbour subsidies and discrimination against foreign companies in public procurement. Concerning the latter, despite the large value of such government contracts, very little empirical research has been undertaken on the trade effects of public procurement policy changes, reflecting a dearth of underlying data. Hence the PRONTO data in this area should underpin future research in this area.

Two inter-linked research streams involved first constructing detailed product-level (6-digit level of the harmonized classification of traded products – HS6) bilateral trade database for which there are available reports on subsidies in the Global Trade Alert database. This was done using the BACI database, a harmonized and reconciled matrix of bilateral trade flows covering all countries in the world and derived from COMTRADE. The database provides with stylized facts across sectors, country type, and time (up to 2016). Of particular interest is the incorporation of the time dimension, permitting the analysis of whether there are domino effects in which countries match subsidies in certain sectors with subsidies of their own. Ultimately, the economic impact of these two sets of data on subsidies and public procurement policies has been evaluated using a gravity-type estimation of the responses of the different margins of exports. The result is a database and analytical report.

In the second research stream, data providing a mapping from GTA-based notifications of public procurement, centred on the years before and after the crisis, product groups and trading partners, was also collected and assembled into a larger dataset including trade flows at the HS6 level by PSE. The data are available on the PRONTO portal.

The data are first exploited in Evenett and Shingal (2016). This paper summarises public procurement policy changes undertaken since November 2008, based on data from the GTA project. A particular focus is on policy changes that alter the relative treatment of domestic firms vis-à-vis foreign rivals. The ultimate goal of this paper is to inform other, ongoing data collection efforts, public policy deliberations on crisis-era policy response in particular as they relate to state purchasing policy, and discussions on the relative merits of strengthening disciplines on public procurement matters in trade agreements.

The data on public procurement NTMs are further exploited in the second paper (Fontagné et al., 2017). In their paper, the expanded database, including individual products and trade flows, as well as procurement based NTM measures, is included in the NTM data platform. Although representing a significant share of GDP in many countries, little was known on how public procurement policies affect international trade in goods and services until PRONTO addressed this issue and provided a novel database. Part of the difficulty was indeed data availability. The contribution of PRONTO is to rely on a recently developed database mapping a treatment of the Global Trade Alert database focusing on obstacles to public procurement policies with the matrix of world trade flows at the bilateral and product level (BACI). Considering the 2009-2016 period, the paper shows that the most active restrictive policies are enforced in large markets, that the most successful exporters are targeted by these policies, and finally that these policies significantly deflect sales of targeted exporters for the targeted goods in imposing countries.

Project Outcome: new priority rankings of NTMs

The PRONTO team also focused on integrating the outcome of the various PRONTO work streams to generate a ranking and prioritization of NTMs based along different vectors – incidence of use, direct economic impacts, indirect impacts along value chains, and impacts in social and sustainability dimension. This work ran alongside database development, wherein data were made available to other teams working on development of NTM indicators, analysis of detailed firm data, econometric (gravity) analysis of NTMs, mapping of NTMs to value chain data based on input-output tables, and analysis of economic performance, social, and sustainability impact of NTMs. The resulting data extends the NTM data developed in the project through principal component analysis of NTMs. The resulting report, Francois et al (2018), reflects contributions from all project consortium teams. It serves as a synthesizing report from the PRONTO project teams, bringing together findings from work packages focused on NTM measurement and on the impact of NTMs on economic performance, social outcomes, and sustainability. As such, it provides an integrated assessment from the broad PRONTO project team, who have been engaged in a mix of regulatory data analysis, econometrics, and numerical assessment of NTMs. Chapter 2 of the report provides a conceptual overview of NTMs. This includes methods of classification, data collection on NTMs, and the mapping from raw regulatory data to quantitative measures. The Chapter draws heavily on work by the PRONTO project team under the work packages on database development. Chapter 3 of the report examines broad patterns of NTM incidence, in terms of countries, regions, and sectors, working with a large-scale dataset that follows from the data elements of the PRONTO project. The merged dataset also reflects extensive processing and clean-up of individual datasets, as well as mapping to detailed HS6 product categories and to the standard MAST classification scheme. Chapter 4 presents composite NTM measures based on principal component analysis (PCA) of the full, HS6 based set of NTM indicators. The PCA measures combine information unique to individual source data, which otherwise vary in coverage and focus, providing a basis for identifying key measures in particular sector across countries, and key NTM users in particular sectors. In other words, the PCA exercise provides a basis for integrating information from multiple data sources for use in ranking and prioritizing NTMs. Chapter 5 is focused on the impact of NTMs. Here, the report draws on lesson from work by the PRONTO team on the economics, social, and sustainability impacts of NTMs.

5. OBJECTIVE 4: ANALYSING THE IMPACT OF NTMS ON A RANGE OF ECONOMIC AND SOCIAL INDICATORS (E.G. THE SDGS).

Project Outcome: evidence on the impact of regulatory linkages in services

Leitner et al (2018) provide evidence of the role of business services and trade barriers on services on manufacturing productivity growth. Here, the resulting estimates point to a negative impact of services trade restrictiveness on manufacturing labour productivity growth might exist. In a second paper, (Pellizzari and Pica, 2018) have studied the impact of occupational licensing which can be viewed as a barrier to entry. In a theoretical framework it is shown that lower barriers to entry lead to higher-quality service providers depending on the ability of the provider. In an empirical analysis on a regulatory reform in 2006 in Italy it is shown that reducing the level of barriers to entry had very strong short run (and most likely long-run) impacts on exits, thus supporting the hypothesis and the interpretation of such licenses as NTMs. This work also yielded data contributions to the final PRONTO dataset.

Project Outcome: evidence on the impact of NTMs on firms

Two papers focus on the impacts of NTMs on productivity at the firm level and the population of exporters based on the hypothesis derived from a heterogeneous firm model: Behrens, et al. (2017) and Barba-Navaretti et al. (2017b). The link between NTMs and domestic market conditions depends on whether they involve new standards and technical speciﬁcations imposed on both domestic and foreign ﬁrms or the extension to foreign ﬁrms of standards and technical speciﬁcations already adopted by domestic ﬁrms. Data confirms this. In a second paper reflecting the same theme, it is shown that a liberalization of trade in services will have stronger impacts than a liberalization of trade in goods on EEA countries’ productivity. However, gains (and losses) remain modest an in all cases below 1 % and are distributed unevenly across countries. In an additional counterfactual exercise to isolate the role of NTMs concerning the exit of the UK form the EEA (Brexit) the paper ﬁnds sizeable losses for many EEA countries and in particular for the UK and Ireland for services trade. Overall, the results point towards heterogeneous effects of NTMs on trade depending on the exact measure (e.g. TBT or SBS), the industry (or product) and the countries considered. Concerning the effects on productivity and competitiveness results are similarly dependent on the exact nature of the measure.

Project Outcome: evidence on the impact of NTMs on labour demand

Leonardi and Meschi (2017) develop a quasi-experimental approach to identify the effect NTMs have on labor demand. Rising import competition from low-income countries has been an important cause of the decline in manufacturing employment in many countries. Since tariffs on international trade have been progressively liberalized over the last decades, developed countries have increasingly relied on Non-Tariff Measures (NTMs) to protect their industries from foreign competition. This analysis uses a quasi- experimental approach and exploits a novel database on NTMs to study the effects of NTMs on labor demand, composition of the workforce and wages. The results indicate that NTMs protection managed to mitigate the negative employment effect of import exposure, but has no effect on local wages, which is consistent with mobility of workers across local areas until wages are equalized. These results are potentially important for policy makers in many countries.

Project Outcome: evidence on the impact of NTMs on gender and skill inequality

Research in the PRONTO project also focused on the impact of NTMs on wage inequalities between genders and workers characterized by different skills and/or educational attainments, as well as the specific contribution of NTMs on inequalities under the the “anything but guns” provisions of EU trade barriers.

In the first of three resulting papers, Davies and Mazhikeyev (2016) use firm level data across 99 developing and transition economies to explores the productivity differences between firms depending on their export status and the gender of their owners. Findings are that female-owned exporters have roughly half the exporter productivity premium of comparable male firms. This is particularly true for larger firms, suggesting that this difference may reflect greater difficulty in implementing learning by exporting for female-owned firms. Nevertheless, there is also evidence consistent with selection into exporting where female-owned firms face relatively higher export costs. Together, these point to significant discrimination barriers female firms face when exporting.

The second paper (Barba-Navaretti, et al. 2017a) provides an empirical assessment of the effect of NTMs on the size of firm-level wage skill premia and on the skill composition of labour demand, making use of detailed firm level matched employer-employee data with information on exports by destination country and Specific Trade Concern (STC) data released by the WTO to measure trade restrictive non-tariff measures. Findings are that NTMs have little impact on skill premia, while still affecting the skill composition of employment. In particular, Technical Barriers to Trade (TBTs) raise the share of managers at the expense of white collars and professionals, while Sanitary and PhytoSanitary (SPS) measures raise the share of qualified blue collars and reduce the share of white collars.

The third paper provides analysis of labour market effects specific to the “anything but guns” provisions of EU trade policies. The Everything But Arms agreement, introduced by the EU in 2001, eliminated duties on most imports from the least developed countries. To avail of these benefits, however, the exported product must contain a sufficiently large share of local content. Thus, the agreement may have affected both the quantity and the factor content of exports from the least developed countries to the EU. Using a panel of sector-level data across countries, Davies and Desbordes (2016), report estimates suggesting that, contrary to expectations, the agreement may have increased the skill-content of these exports, benefitting the lowest-skilled EU workers at the expense of their highest- skilled counterparts. This result, however, is entirely driven by textile trade; when omitting this industry, the authors find no significant effects. This suggests that the EBA may have led to the local provision of higher-skill inputs in the textile industry.

Project Outcome: evidence linking NTMs to aspects of Europe 2020 objectives

Research also focused on the nexus between trade policy and the locations of economic activities with the European space. Behrens et al. (2017) develop a computable general equilibrium model featuring love of variety, heterogeneous firms, labour mobility, as well as endogenous markups and productivity, that can be applied to analyse the impacts of trade costs changes, and in particular NTMs changes, on the location of economics activities across space. The model is calibrated to goods and services trade data, as well as GDP and population data, for European Economic Area (EEA) regions plus other OECD countries. Finally, the study assesses the importance of NTMs and other trade costs by performing a series of counterfactual experiments. Specifically, it evaluates the impact of implementing the Transatlantic Trade and Investment Partnership (TTIP) between the EEA and the US and the impact of Brexit. The paper separately considers a liberalization/restriction of trade in goods and a liberalization/restriction of trade in services (as well as a join liberalization/restriction) with the latter being a much cleaner instance in which NTMs represent the main existing obstacle to trade. Concerning TTIP results indicate that a liberalization of trade in services (essentially NTMs) will have stronger impacts than a liberalization of trade in goods on EEA countries' productivity. However, gains (and losses) remain modest an in most cases below 1%. Interestingly, countries in the core of the EEA (Germany, Belgium, the Netherlands, etc.) will mainly loose from TTIP while peripheral countries will gain. At the same time, large city-regions (Paris, London) tend to gain less from deeper service trade integration. The reason is that their large size confers them an advantage that is larger the harder it is to trade. As for population changes, they roughly mirror the pattern of productivity changes and are overall modest. When considering Brexit, while focusing on trade in services, results suggest sizeable losses for many EEA countries, with the UK and Ireland particularly impacted (about -1.5% productivity each and with a decrease in population of respectively 1.12% and 1.35%). Furthermore, results suggest changes induced by Brexit are likely to favour the larger city regions at the expense of smaller regions.

Project Outcome: evidence linking NTMs to the role of technology in Development

Research also focused on the linkage between NTMs, technology, and development. Bagayev and Davies (2017b) examine whether trade protection in the form of SPS or TBT regulations affect innovation in a dataset of 4750 firms across 13 developing countries. Specifically, the study examined two types of innovation: that related to the development of new products and that affecting the production method (process innovation). This was embedded in the infant Industry argument for trade protection which posits that by using short term trade protection, it encourages local firms to develop their productivity so that they can successfully compete even after protection is removed. This argument is often used to justify trade protection – including NTMs – as a means of sparking development. Critics of the Infant Industry Argument, however, counter that the lack of competition may actually reduce the need to innovate in order to survive, hurting growth. The findings validate this argument by using the WTO notification database developed within the project to examine the relationship between self-reported product and process innovation in firm level data obtained from the World Bank. In short, no evidence is found supporting a pro-innovation effect of protection. Whereas SPS regulations are completely uncorrelated with either type of innovation, TBT protection lowers product innovation and tariffs reduce both types of innovation. Thus, the evidence indicates that NTMs may inhibit growth by slowing innovation.

In a second paper, Bagayev and Davies (2017a) examines the relationship between NTMs and the distribution of firm productivities. As posited by Melitz (2003) among many others, an increase in trade barriers limits import competition, thereby permitting low-productivity firms to survive. In contrast, trade liberalization would drive those firms under, reallocating their resources to more productive firms, enhancing aggregate productivity and thereby economic development. Using data on 28 industries across 99 countries, this study examines two features of the productivity distribution: its mean and its skewness. The reason for considering both of these is that although economic development may depend on the average productivity across firms, in a political economy setting, the median voter likely holds a great degree of influence. Therefore, to understand the political economy of trade liberalization, it is necessary to consider not just the average firm but the difference between the average and the median firm. As predicted by Melitz (2003), the estimates indicate that NTM protection (as measured by the datasets produced in the PRONTO project) lowers average productivity. Further, they suggest that NTM protection increases the skewness of the distribution, widening the gap between the median and average firm. This suggests that although liberalization might spur development via reallocation, it is likely to encounter significant political resistance.

Project Outcome: evidence linking NTMs to government revenues and development

Although the link between trade policy and government revenues, particularly for developing countries, may not seem obvious, it must be recognized that developing countries rely on trade taxes (which include both tariffs and revenue-generating NTMs such as anti-dumping duties or export fees) for up to a third of their government revenues. As such, a push to liberalize trade policy may have unexpected detrimental effects if it lowers revenues and thus the ability to provide infrastructure or other services associated with development. In particular, one important government service associated with development is the enforcement of regulations, including the enforcement of trade policy. This sets up a nexus of trade taxes, government revenues, enforcement capability, and development in which each may well support the other.

To explore this issue, Bagayev, et. al (2017) estimated the impact of revenue-generating antidumping duties on imports and how this varies with proxies of government enforcement capability, including the size of government expenditures relative to GDP and the size of the shadow economy. They do so using PRONTO NTM data together with trade data for 108 countries at the 6-digit product level. The results indicate that, as predicted, the effect of NTMs are larger when government enforcement is higher. This is especially true for emerging economies. A key implication of this finding is that if, for example, international pressure forces governments to relax revenue-generating NTMs, this has both a direct negative effect on government revenues (as less is collected given enforcement capability) as well as an indirect one (since less revenue can weaken enforcement efforts). Together, these suggest the need for caution in recommending trade liberalization in emerging economies where trade taxes form a significant part of the government expenditure needed to promote development. This then mirrors the tariff/VAT debate elsewhere in the literature.

Project Outcome: evidence on the impact of NTMs on the environment

Davies, Edwards, and Mazhikeyev (forthcoming) consider the energy intensity of 11,000 firms in 32 developing countries in Africa and Asia as it depends on whether or not they are in a special economic zone, among other factors. As special economic zones are an increasingly important NTM, this then fits into the broader NTM and development objectives of this WP. In particular, the analysis finds that even after controlling for other factors, firms operating within a zone are significantly more electricity intensive, with the gap about 4% on average. As this is a common proxy for polluting, this suggests that this NTM may have negative environmental impacts even as it aids other types of development by lowering tariffs and export barriers (something considered in WP2). Going further, the study finds that this effect is especially large when firms are financially constrained, suggesting that part of the mechanism may be that firms in zones are more able to access credit and invest in more energy intensive technologies than others.

In another paper providing evidence on SEZs and emissions, the paper on Processing and Special Economic Zones by Francois and Davies (2015) The study first describes a database of WTO Members that employ special economic zones as part of their industrial policy mix. This is based on WTO notification and monitoring through the WTO’s trade policy review mechanism, supplemented with information from the ILO, World Bank, and primary sources. The paper athen provides a first analysis of the relationship between use of export processing zones and the carbon intensity of exports based on GVC and carbon accounts data. Thus, this provides a further tie between economic development and the environmental impact of NTMs, and corroborates the findings of Davies Edwards, and Mazhikeyev (forthcoming).

Pantelaiou, et. al (2016) develop a model in which two competing nations have multiple policy instruments, including public abatement and ERSs which affect both trade and resource use (i.e. pollution). Each of these policies has its advantages and disadvantages as they result in different trade-offs between trade levels and pollution. The authors find that while public abatement, funded by an emissions tax, is the best at encouraging economically beneficial trade between countries, ERSs and other policies do a better job at reducing pollution for a given level of trade. Such intuition is buttressed by simulation evidence indicating the impact on emissions, resource usage, and trade levels. Therefore, the optimal policy depends on a government's relative valuation of trade and environmental quality. In particular, building on the Bagayev, et. al (2017) results, this might suggest that when trade is an important revenue source, environmentally effective ERSs may be an unlikely policy choice.

Hatzipanayotou, et. al (2017) directly link the use of ERSs and development via the so-called "resource curse." In the resource curse, a resource abundant country is actually damaged as trade barriers decline because the increase in exports results in an income-reducing terms of trade effect. As such, an ESR which limits resource usage and exporting can be used to offset this effect. They begin by using a growth model to illustrate this benefit of environmental NTMs. This is then supported by empirical evidence on income and poverty data from the least developed countries where ISO14001 certification is used as the measure of ESRs (something obtained from the environmental database work in support of the public PRONTO database).

6. References

Bagayev, I. and Davies, R. (2017b) "The Impact of Protection on Observed Productivity Distributions," Mimeo. University College Dublin Working Paper WP17/05. Pronto project report D6.1.
• Open access version: https://prontonetwork.org/database/resources/papers/The%20Impact%20of%20Protection%20on%20Observed%20Productivity%20Distributions.pdf

Bagayev, I., Davies, R., Hatzipanayotou, P., Konstantinou, P., and Rau, M. (2017) "Non-Tariff Barriers, Enforcement, and Revenues: The Use of Anti-Dumping as a Revenue Generating Trade Policy," Mimeo. University College Dublin Working Paper WP17/06. Pronto project report D6.2.
• Open access version: https://prontonetwork.org/database/resources/papers/Non-Tariff%20Barriers,%20Enforcement,%20and%20Revenues.pdf

Barba-Navaretti G., L. Fontagné, G. Orefice, G, Pica and A.C. Rosso (2017a) “NTMs, Income Inequality and Social Cohesion.” Pronto project report D5.3.
• Open access version: https://prontonetwork.org/database/resources/papers/NTMs%20Income%20Inequality%20and%20Social%20Cohesion.pdf

Barba-Navaretti, G., Felice, E. Forlani, and P.G. Garella (2017b). “Non-tariﬀ measures, competitiveness and the population of exporters.” Pronto project report D3.3
• Open access version: https://prontonetwork.org/database/resources/papers/Non-tariff%20measures,%20competitiveness%20and%20the%20population%20of%20exporters.pdf

Behrens K., G. Mion (2017) “Estimating the costs and gains of TTIP and BREXIT for EU regions.” PRONTO project report D5.1.
• Open access version: https://prontonetwork.org/database/resources/papers/Estimating%20the%20costs%20and%20gains%20of%20TTIP%20and%20BREXIT%20for%20EU%20regions.pdf

Bekkers, E., Francois, J. and H. Rojas-Romagosa (2018a), “Econometric framework for NTM assessment: comparing methods to map NTMs into trade costs and welfare,” PRONTO project report D4.1.
• Open access version: https://prontonetwork.org/database/resources/papers/Econometric%20framework%20for%20NTM%20assessment%20-%20comparing%20methods%20to%20map%20NTMs%20into%20trade%20costs%20and%20welfare.pdf
• Journal version: forthcoming (2018) as “The welfare effects of free trade agreements in quantitative trade models: A comparison of studies about Transatlantic Trade and Investment Partnership.” World Economy.
https://doi.org/10.1111/twec.12670.

Bekkers E., and Francois J. (2015). “Calibrating a CGE model with NTBs that Incorporate Standard Models of Modern Trade Theory.” PRONTO project report D2.3.
• Open access version: https://prontonetwork.org/database/resources/papers/Calibrating%20a%20CGE%20model%20with%20NTBs%20that%20Incorporates%20Standard%20Models%20of%20Modern%20Trade%20Theory.pdf
• Journal version: forthcoming (2018) as ““A Parsimonious Approach to Incorporate Firm Heterogeneity into CGE Models,” in the Journal of Global Economic Analysis, December.

Bekkers, E. and Francois, J. (2018). “Amending the Trade Restrictiveness Index to Account for Market Imperfections: market structure and country size.” PRONTO project report D2.4.
• Open access version: https://prontonetwork.org/database/resources/papers/Amending%20the%20Trade%20Restrictiveness%20Index%20to%20Account%20for%20Market%20Imperfections.pdf

Bekkers, E., Francois, J. and H. Rojas-Romagosa (2018b), “Numerical Methods for Ex Ante Policy Assessment of NTMs and the Impact of NTMs in Models of Firm Heterogeneity,” PRONTO project report D4.2.
• Open access version: https://prontonetwork.org/database/resources/papers/Numerical%20Methods%20for%20Ex%20Ante%20Policy%20Assessment%20of%20NTMs%20and%20the%20Impact%20of%20NTMs%20in%20Models%20of%20Firm%20Heterogeneity.pdf

Davies R.B. and R. Desbordes (2016) “The Impact of Everything But Arms on EU Relative Labour Demand.” University College Dublin Working Paper WP2016/14. Pronto project report D5.3.
• Open access version:
https://prontonetwork.org/database/resources/papers/The%20Impact%20of%20Everything%20But%20Arms%20on%20EU%20Relative%20Labour%20Demand.pdf

Davies, Ronald B., Marie-Luise Rau, and Achim Vogt. “A Methodology Inventory for Studies Using NTM Data.” Pronto project report D1.3.
• Open access version:
https://prontonetwork.org/database/resources/papers/A%20Methodology%20Inventory%20for%20Studies%20Using%20NTM%20Data.pdf

Davies R.B. and A. Mazhikeyev (2016) “The Glass Border: Gender and Exporting in Developing Countries.” University College Dublin Working Paper WP15/25. Pronto project report D5.3.
• Open access version:
https://prontonetwork.org/database/resources/papers/The%20Glass%20Border%20-%20Gender%20and%20Exporting%20in%20Developing%20Countries.pdf

Davies, R., Edwards, T.H. and Mazhikeyev, A. (2018) "The Electricity Intensity of Firms in Special Economic Zones," PRONTO project report D6.3 published in The Energy Journal. 39(SI1):5–24.
• Open access version:
https://prontonetwork.org/database/resources/papers/The%20Impact%20of%20Special%20Economic%20Zones%20on%20Electricity%20Intensity%20of%20Firms.pdf
• Journal version: The Energy Journal. 39(SI1):5–24.
https://doi.org/10.5547/01956574.39.SI1.rdav

Disdier A.C. Beestermöller M. and L. Fontagné (2018). “Impact of European food safety border inspections on agri-food exports: Evidence from Chinese firms.” PRONTO project report D2.2. published in China Economic Review, 48, 66-82.
• Open access version: https://prontonetwork.org/database/resources/papers/Uncertainty%20and%20Exports.pdf
• Journal version: China Economic Review, 48, 66-82.
https://doi.org/10.1016/j.chieco.2017.11.004

Evenett S., and Shingal A. (2016). “Gauging Procurement Policy Change During the Crisis-Era: Evidence from the Global Trade Alert.” Pronto project report D2.5.
• Open access version:
https://prontonetwork.org/database/resources/papers/Gauging%20Procurement%20Policy%20Change%20During%20the%20Crisis-Era.pdf

Fontagné L., Disdier A.-C. and Tresa E. (2018). “Public Procurement-Related Protection: Insights from the Global Trade Alert Database.” Pronto project report D2.6.
• Open access version:
https://prontonetwork.org/database/resources/papers/Public%20Procurement-Related%20Protection%20-%20Insights%20form%20the%20Global%20Trade%20Alert%20Database.pdf

Francois, J. editor (2018), “Ranking and Prioritizing NTMs,” PRONTO project report D4.3.
• Open access version:
https://prontonetwork.org/database/resources/papers/Ranking%20and%20Prioritizing%20NTMs.pdf

Francois, Joseph F., and Miriam Manchin (2016). “Technical Note on Services NTM Database.” Bern: WTI. PRONTO project report D1.4.
• Open access version:
https://prontonetwork.org/database/resources/papers/WB%20Services%20NTM%20database.pdf

Ghodsi, M., J. Grübler, O. Reiter and R. Stehrer (2017). „The Evolution of Non-Tariff Measures and their Diverse Effects on Trade.” wiiw RR 419, May 2017. PRONTO project report D1.4.
• Open access version:
https://prontonetwork.org/database/resources/papers/The%20Evolution%20of%20Non-Tariff%20Measures%20and%20their%20Diverse%20Effects%20on%20Trade.pdf

Ghodsi, M., J. Grübler and R. Stehrer (2016a). “Import Demand Elasticities Revisited.” wiiw WP 132, November 2016. PRONTO project report D1.3.
• Open access version:
https://prontonetwork.org/database/resources/papers/Import%20Demand%20Elasticities%20Revisited.pdf

Ghodsi, M., J. Grübler and R. Stehrer (2016b), Estimating Importer-Specific Ad Valorem Equivalents of Non-Tariff Measures, wiiw WP 129, September 2016. PRONTO project report D1.3.
• Open access version:
https://prontonetwork.org/database/resources/papers/Estimating%20Importer-Specific%20Ad%20Valorem%20Equivalents%20of%20Non-Tariff%20Measures.pdf

Ghodsi, M. and R. Stehrer (2017), NTMs in the Presence of Global Value Chains and their Impact on Productivity, PRONTO project report D3.1 (revised version).
• Open access version:
https://prontonetwork.org/database/resources/papers/NTMs%20in%20the%20Presence%20of%20Global%20Value%20Chains%20and%20their%20Impact%20on%20Productivity.pdf

Gourdon, Julien. (2014). “NTM-MAP: A Tool for Assessing the Economic Impact of Non-Tariff Measures.” Technical Paper. Paris: CEPII Working Paper 2014-24. PRONTO project report D1.4.
• Open access version:
https://prontonetwork.org/database/resources/papers/NTM-MAP.pdf

Hatzipanayotou, P., Konstantinou, P., Pantelaiou, I., and Xepapadeas, A. (2017) "ERSs and Trade in Natural Resources: The Impact on Economic Growth and Poverty in LDCs," AUBE Working Paper 1809. PRONTO project report D6.5.
• Open access version: https://prontonetwork.org/database/resources/papers/ERSs%20and%20Trade%20in%20Natural%20Resources.pdf

Kee, H.L. A. Nicita, and M. Olarreaga (2009). “Estimating Trade Restrictiveness Indices.” Economic Journal.” (119): 172-199. https://www.jstor.org/stable/20485299

Leitner, S., O. Pindyuk and R. Stehrer (2017). “Services trade restrictiveness and manufacturing labour productivity growth.” PRONTO project report D3.2.
• Open access version:
https://prontonetwork.org/database/resources/papers/Services%20trade%20restrictiveness%20and%20manufacturing%20labour%20productivity%20growth.pdf

Leonardi M. and E. Meschi (2017) “Do Non-Tariff barriers to trade save jobs and wages?” Pronto project report D5.3.
• Open access version:
https://prontonetwork.org/database/resources/papers/Labour%20Demand%20Effects%20of%20Non%20Tariff%20Measures.pdf

Melitz, M. (2003). "The Impact of Trade on Intra-industry Reallocations and Aggregate Industry Productivity." Econometrica, (71): 1695-1725. https://www.jstor.org/stable/1555536.

Pantelaiou, I., Hatzipanayotou, P., Konstantinou, P., and Xepapadeas, A. (2017) "Intra (Inter) Regional Effects of Environmental Policies as NTMs in an Economic Union.” Pronto project report D6.4.
• Open access version:
https://prontonetwork.org/database/resources/papers/Intra%20(Inter)%20Regional%20Effects%20of%20Environmental%20Policies%20as%20NTMs%20in%20an%20Economic%20Union.pdf

Pellizzari, M. and G. Pica (2018). “On the effects of occupational licensing: evidence from Italian Lawyers.” IGIER Working Paper n. 372. http://www.igier.unibocconi.it/files/372_1pdf.pdf.

PRONTO (2018), “PRONTO Data Portal – Codebook 31 January 2018,” PRONTO project report D1.4.
• Open access version:
https://prontonetwork.org/database/resources/docs/Pronto%20Database%20Codebook%202017.pdf

Rau, Marie-Luise, and Achim Vogt (2016). “A GAP Analysis of NTM Data.” Working Paper. PRONTO Working Paper. The Hague/Bern: WUR/WTI. Pronto project report D1.3.
• Open access version:
https://prontonetwork.org/database/resources/papers/A%20GAP%20analysis%20of%20NTM%20data.pdf

Rau, Marie-Luise, and Achim Vogt. “NTM Data: Sources and Concepts.” Working Paper. PRONTO Working Paper. The Hague/Bern: WUR/WTI, 2017. Pronto project report D1.4.
• Open access version:
https://prontonetwork.org/database/resources/papers/NTMs%20-%20Data%20concepts%20and%20sources.pdf

Potential Impact:
----------
Please see attached pdf file "4.1.4_impact_dissemination_exploitation.pdf" for a formatted version of this text
----------

1. Potential Impact and Relevance

As noted in the project summary, the PRONTO (Productivity, Non-Tariff Measures and Openness) project has been concerned with the policy relevance of non-tariff measures (NTMs) affecting international trade and investment flows. NTMs are a primary focus of the European Union’s external trade and investment policy strategies, and for this reason the research conducted under PRONTO has socio-economic policy relevance, in addition to the scientific outcomes of the project.

Although the research team has been European, the results will have impacts well outside the EU. First, it must be remembered that the non-tariff measures (NTMs) that are the focus of the project have impacts on both EU and non-EU trading partners, in terms of both restricting their trade and investment directly and by diverting trade and investment towards member states which may find it easier to overcome the NTM barriers studied. Furthermore, if NTMs influence the productivity of European firms, this will have an impact on EU exports to the US, Japan, and other countries. Thus, understanding the impact of NTMs on the EU has important implications for understanding their impact outside the EU and for the potential for mutually beneficial coordination on NTMs. In addition, by laying out methodologies for the study, quantification, and use of NTMs in research, the projects findings will have a large influence on the study of these issues even when the focus is elsewhere. Thus, the analysis will both enrich our understanding of the global issues surrounding NTMs and on the academic study of these policies. Finally, although the research team is European, among the goals of the project has been identification of the impacts of NTMs on other countries, most notably the EU’s developing country trading partners.

The results of the project with respect to measurement include: (i) identification of existing methods and extension of those methods for better classification of NTMs; (ii) collection and organization of existing data and generation of new data on NTMs; (iii) mapping from raw regulatory data to policy-focused quantitative measures; (iv) and identification and analysis of broad patterns of NTM incidence, in terms of countries, regions, and sectors. The result of this research both fed into project work on the impact of NTMs, and into a publicly available dataset that reflects extensive processing, clean-up, and integration of individual datasets, and development of new composite measures based on principal component analysis (PCA). The data series hosted on the database portal, as discussed in other parts of this publishable summary, will be a lasting legacy of the project. The project team will work to update the database on the portal after the FP-7 funded phase of the project. The network will continue to support collaboration between in international agencies and researchers.

The project outcomes themselves, again as summarized elsewhere in this report, are relevant for both the policy and scientific communities, and their impacts will live beyond the 4-year life of the project. The data themselves have been structure to allow for regular updating and extension to serve the needs of the policy and research community. The data and methodology elements of this project should benefit the broader public by improving the ability of our elected representatives to negotiate new, deeper trade and investment agreements by better understanding the role of NTMs in affecting economic integration. Civil society should benefit as well from a better understanding of the impact of NTMs on broad socio-economic objectives. In general, adhering to the belief that democratic societies benefit from more information, as this is an important public policy area, the information generation and dissemination tasks under this project should improve the public policy dialog on NTMs.

2. Dissemination

Overview

The entire purpose of the dissemination elements of the project been to ensure that the output of the PRONTO project is made known to interested parties from academic and policy circles, as well as the public at large. This was done through three key methods: construction of a webpage, routine efforts to advertise the project, and finally the organisation of events the engaged all relevant stakeholders. The first of these, development of a dissemination plan, was completed at the beginning of the project. The other activities linked to dissemination continued through the life if the project.

The dissemination plan itself was developed during the first reporting period, setting the strategy for the project communication and exploitation activities. The plan identified the project stakeholders (PRONTO partner networks, the Economic and the Scientific Community and policy makers, ex: European Commission: DG Trade, WTO, ITC, UNCTAD, UNITC, World Bank, OECD, central banks, etc) and outlined the tools and channels to be used to reach the relevant targeted audiences and the general public.

The Website

From the beginning, the dissemination plan featured the setup and regular updating of the project website. The project webpage can be found at http://www.prontonetwork.org . This website hosts PRONTO-driven research, a database on NTM methodology, and the PRONTO data portal. The last two points reflect support work related to dissemination of the methodology inventory and databases developed under the PRONTO project. The site not only works to advertise the output of the project, bus also serves as a springboard for further activity among PRONTO members as well as the general research community. In particular, the one-stop-shopping data portal benefits both academic and non-academic researchers (including government institutions, other policy bodies, and the private sector). The database portal provides, in our view, a unique, valuable access to data on NTMs for the research and policy community alike that will extend well beyond the FP7 funding period for this project. It has been structured, in cooperation with a number of agencies that provide primary source data, to allow for periodic post-FP7 updating of the indicators (such as the PCA-based NTM indexes) generated in the PRONTO project.

A key feature with respect to the website has been the data portal, which brings together the data used and developed in the project makes the link to the NTM data that have been collected by international organisations. The cooperation with the international organisations has been much appreciated and highly productive, in addition to the provision of data by the project partners.

The NTM Data portal has the following features:
• Documentation of source data;
• Datasets and guide to source data indexed by importing country, partners (where applicable), product codes (HS, WIOD/ISIC, GTAP), and MAST classification chapters;
• PRONTO datasets that will be updated periodically to reflect changes in the HS classification and eBOPS schemes;
• Concordance files that help working with NTM data;
• Interactive tables and graphs that support assessing data availability and NTM patterns;
• R and Stata scripts that access a MySQL in-stance of the PRONTO data, and thus load data directly into these programs;
• Sample script for replications exercises.

The NTM Data portal has been presented to the community of researchers, practitioners and policy-makers and was promoted at various events, like the ETSG conference, the World Trade Forum, the MAST meeting and the EAAE conference. This then serves to implement all the objectives in WP8.

Logos and Newsletters

Our dissemination strategy has also included distribution of tangible material (PRONTO logo coffee mugs, a PRONTO brochure, etc.) and intangible (widely circulated biannual newsletters) items. As with the website, while these called attention to the research output of the project, they were also geared to disseminate the data availability via PRONTO to relevant researchers.

A project e-newsletter was produced every six months by CEPR outlining the latest updates from each work package, information on upcoming and past events and access PRONTO working papers. Furthermore, a researcher’s profile section featuring key PRONTO investigators was included in several editions.

Seven newsletters were sent out throughout the project. The e-newsletters have been based on partners’ contributions, designed and edited by CEPR and distributed to different stakeholders (scientific organisations, policy makers and other interested parties).

Publication and Presentation

Another element of dissemination and exploitation has been the publication of project research, as working papers, articles, and also as a (forthcoming) volume with Cambridge University Press. PRONTO participating institutions have published working papers based on PRONTO research, while working papers based on PRONTO research and resulting outputs have also been published on the PRONTO website itself (with periodic updates). Final versions of all working papers and data are also hosted on the WTI working paper repository (linked to the University of Bern’s open access platform). This includes all papers/reports and related data for the project. Project partners have also present project results in international conferences, while related papers have also been submitted for publication at scientific journals. Given the time lag when publishing in economics, these papers will generally not be in print yet during the final reporting period, though several are already in press. Finally, based on PRONTO team research, a book volume with the Cambridge University Press has been accepted and is under editorial review.

Together, these various efforts have served to get the PRONTO name in front of a wide-ranging group of interested parties. These efforts will continue to pay off as the resulting data are updated and exploited by the research community.

Events and Conferences

Throughout the project, several dissemination events and conferences took place. Project partner CEPR organised a Kick Off Meeting in Brussels and the Annual and the Final Conferences in Vienna with the local support of partner WIIW. The public events were well attended by practitioners and policy makers, as well as by prominent researchers from outside of the project network. Apart from that, two Dissemination Workshops were organised by Ecorys in Amsterdam and by PSE in Paris. Here we describe these events.

The First PRONTO Dissemination Workshop was organized by partner ECORYS over 27-28 May 2015. This event was held at Vrije Universiteit Amsterdam. The main topics discussed were the representation of NTMs across different data-sources were held, as well as recent developments with regard to data improvement. The event was organised in close cooperation with stakeholder agencies (WTI, UNCTAD, ITC).

We also organized annual conferences, as well as a final capstone conference. The project started with a Kick Off Meeting, organised by CEPR, which was held the 12-13 May 2014 in Brussels. During the 2-day event the PRONTO Team presented an overview and summary of immediate work-plan and deliverables, followed by a group discussion. Among other topics, the meeting focused on how to consolidate and develop the data and how to quantify economic, social and environmental impacts of NTMs.

The PRONTO Annual Conference, 25-26 February 2016 took place at the WIIW in Vienna and was jointly organised by CEPR and WIIW. This dissemination event aimed at promoting the project activities among different stakeholders and gathering feedback from partner organisations, the scientific community and practitioners from the field. A round table discussion on data availability was organised. The conference was followed by a meeting with the Supervisory Board and a management meeting. During the conference, participants from within and outside the PRONTO network presented research, including some papers resulting from the project's deliverables.

Pronto organized its Third Conference in cooperation with the European University Institute. In particular, the September 2016 World Trade Forum was organized around the theme of PRONTO research on behind the border measures. Several dedicated PRONTO sessions were organized, and the resulting papers (based on PRONTO deliverable reports discussed in this report) will form the basis for a forthcoming book published by Cambridge University Press on behind the border trade policy.

The Second PRONTO Dissemination Workshop, took place the 16 and 17 March 2017 at the Paris School of Economics. Six sessions were held over the course of the two days where speakers and participants discussed topics including NTMs data concepts and sources; trade margins, price, and product quality adjustments to NTMs; the effectiveness of cash subsidies on export performance: NTMs skills and inequality; EU firms’ experiences with NTMs and trade defence in the Trump era, amongst others. As with the other events, it included presentations by PRONTO members (including their work within the project) as well as non-members.

The PRONTO Final Conference was held at the WIIW in Vienna, the 18-19 January 2018, co-organised by WIIW and CEPR. Among the participants were PRONTO project members, Supervisory Board representatives, invited guests and external partners from the European Commission, WTO, UNCTAD, ITC and central banks. Fourteen speakers presented their papers in six sessions, speakers which included PRONTO members and non-members and papers derived from within the projects. Distinguished economists from outside of the project consortium got acquainted with the project results and added to the discussion, presenting their research on NTM topics. The final project results and the NTM Database were also introduced to the public in a dissemination seminar. Following the conference, the project partners met with the Supervisory Board to discuss the future of PRONTO, opportunities for potential cooperation and follow-up PRONTO research activities.

Beyond these events, throughout the four years of the project, the consortium partners promoted and contributed to a series of events and conferences, related to NTMs. These served as forums to call attention to the project's data collection and resulting research, as well as the name of the project overall.

"Non-tariff measures: Data, methods and future challenges" was the first of these public events. The workshop took place on the 24 September 2014 and it was held at the WTO in Geneva and co-organised by ITC and Pronto partner Ecorys. The topic of the workshop was methodologies for NTM measurement. On the 25 September, the event was followed by an internal methodology workshop and consortium meeting. This first annual public event served to advertise PRONTO to the stakeholder community, invite presentations and discussion on NTM measurement methodology, and solicit input from stakeholders and the broader public at an early stage in the formation of the PRONTO project. A second workshop on "Methodologies for NTM measurement" was held in Geneva on 25 September 2014 at the UNITC. Note that PRONTO members were heavily involved in the proceedings of this event. A comparable PRONTO presence contributed to an UNCTAD meeting in the September 2015 on NTM classification under the MAST system, the conference on NTM measurement and quantifying socioeconomic impacts held over 5-8 October 2016, and the Mandatory Trade Regulations and MAST Meeting at the UNCAD in Geneva during 25 - 26 September 2017.

The Pronto team also organised a session on NTMs for the World Trade Forum in September 2015. This included several presentations by PRONTO members (including some work within the project) as well as external stakeholders.

Further, one of the PRONTO partners, UCD, hosted the Society for International Trade Theory conferences in November 2014, April 2016, and April 2017. These events, all in Dublin, Ireland, provided stage for promotion of the PRONTO research activities. Each included both PRONTO researchers (presenting PRONTO-driven research) and non-members.

Finally, PRONTO members have been heavily engaged in conference presentations (including the European Trade Study Group, the world's leading conference on international trade) and invited seminars. Although too numerous to list here, each of these presentations further disseminated the output of the project.

List of Websites:
http://www.prontonetwork.org
Project coordination:
Prof. dr. Joseph F. Francois
World Trade Institute,
University of Bern
3012 Bern, Switzerland
Phone: +41 31 631 32 70
Fax: N/A
Email: joseph.francois@wti.org

Final Report Summary - PRONTO (Productivity, Non-Tariff Measures and Openness)

Related documents

Share this page

Download