The causal impact of HACCP on seafood imports in the U.S.: An application of difference-in-differences within the gravity model
Introduction
On December 18, 1997, the U.S. implemented the Hazard Analysis by Critical Control Points (HACCP) system for domestic and imported seafood products (FDA 1995). Since the 1960s, HACCP has become increasingly popular globally (Caswell and Hooker, 1996, Li et al., 2013; and Ropkins and Beck, 2000). Therefore, a proper evaluation of the impact of HACCP on trade is useful for evaluating similar policy changes beyond the U.S. context. Despite the importance of the seafood industry1 and the implementation of HACCP two decades ago, trade research assessing this policy has been limited, yielding contradictory evidence on the impacts of HACCP. More importantly, an assessment of HACCP is a window into the effects of non-tariff measures, specifically food safety policies, on international trade. Further, the case of HACCP reflects the challenge of assessing a complex policy with multiple components which may have conflicting effects on the trade flows. On the one hand, HACCP could increase the compliance cost for the producers and decrease trade flows. On the other hand, HACCP could aid food safety inspections and increase consumer confidence in the food safety, thus enhancing trade flows (FAO/WHO, 1995).
Empirical evidence, thus far, shows mixed effects on U.S. seafood trade flows (Anders and Caswell, 2009, Li et al., 2012). Beyond the differential response to the policy changes, methodologically, researchers, in previous trade papers, use a single dummy variable of HACCP implementation in the gravity model, which does not address the causal effect of HACCP. In this paper, we identify the causal effect of HACCP with a difference-in-differences (DID) strategy. We find that, contrary to previous literature, the impact of HACCP had a null net effect on U.S. seafood imports.
Unlike other food safety policies that focus on a product, HACCP is a preventive approach to control each stage of the food chain, from the primary production, processing, and storage, to marketing and consumption. HACCP has been considered an effective alternative to the conventional end-point-testing (FAO/WHO, 1984, Bryan, 1992, Ropkins and Beck, 2000, Khatri and Collins, 2007) and recommended for commercial use (Roberts et al., 1996). HACCP is an important regulation that aims to improve food safety while facilitating trade flows (Caswell and Hooker, 1996, Unnevehr and Jensen, 1996, Unnevehr and Jensen, 1999, USGAO, 2017). Like other food safety regulations, the goal of HACCP is to avoid food safety dangers and scares coming from unsafe production practices. Seafood has a history of food safety concerns including unsafe drugs residues, such as malachite green in shrimp imports from China. Other safety concerns are related to the unsafe concentrations of dioxins, polychlorinated biphenyls (PCB), polybrominated diphenyl ethers (PBDEs), toxaphene, pesticides (Foran et al., 2005, Leiss and Nicol, 2006) and methyl mercury (Roosen et al., 2009). While many countries ban the use of substances such as enrofloxacin, furazolidone, and malachite green in aquaculture, other countries still use these substances because they are effective and low cost (Zhang et al., 2012).
After the implementation of HACCP in the United States in 1997 for seafood products, the EU required HACCP for foods with EC 852/2004, the European Food Hygiene Regulations in 2006 (European Commission, 2004). We use this fact to develop our causal framework for analysis. However, the EU established an earlier food hygiene policy with the Council Directive 91/493/EEC on July 22, 1991 (EU, 1991), which outlined the rules for the application in Commission Decision 94/356/EC on May 20, 1994 (EU, 1994). EU member states enacted their own interpretation of food hygiene laws complying with the Council Directive 91/493/EEC (EU, 1991, Ropkins and Beck, 2000). The implementation of these “HACCP and HACCP-like” policies from the Directives was heterogenous across member states. The Council Directives 91/493/EEC or 94/356/EC do not mention HACCP. Council Directive 93/43/EEC (EU, 1993) mentions HACCP once. The EU Commission required HACCP with EC 852/2004, which references HACCP sixteen times (EC 2004). Thus, the food hygiene policies predate U.S. HACCP implementation for seafood products and evolved separately in each member state to become the 2006 HACCP policy.
While EU member states developed their policies separately under the Directive, Higuera-Ciapara and Noriega-Orozco (2000) and Lupin (1999) argue that the food hygiene policies in the EU and U.S. differ in terms of the critical limits, evaluation and content plans, etc. These differences in criteria result in different consequences on enforcement (Higuera-Ciapara and Noriega-Orozco, 2000). Thus, we argue that the heterogenouspolicies of EU member states, which predate U.S. seafood implementation, and differences in those policies relative to U.S HACCP are sufficient that we use the EU as one of our two control groups before the enforcement of the EC 852/2004.
The rest of the paper is organized as follows. Part 2 summarizes the relevant literature and part 3 describes the data. Part 4 is the model specification. Part 5 discusses the findings, and part 6 offers a discussion and conclusion.
Section snippets
Non-tariff measures
An older trade literature suggests that technical measures for food quality and safety are trade-impeding (Henson and Loader, 2001, Disdier and Marette, 2010, Otsuki et al., 2001a, Otsuki et al., 2001b, Disdier et al., 2008, Wilson and Otsuki, 2004, Winchester et al., 2012). Wilson and Otsuki (2004b) study the impact of food regulation stringency and find a negative effect of pesticide chlorpyrifos notified by the Organisation for Economic Co-operation and Development (OECD) countries on banana
The data
We build a panel database of seafood imports affected by HACCP as well as data for food imports unaffected by HACCP between the U.S., EU-15, and 217 partner countries from 1991 to 2006. We use data on U.S. seafood imports (03), other non-HACCP U.S. edible imports (6 and 9–18), and EU-15 (3, 6,and 9–18) imports from all 217 countries
Empirical framework
Previous work evaluating the impact of HACCP (and similar programs) uses the gravity framework and a single dummy variable of HACCP implementation to establish if the policy affected trade (Anders and Caswell, 2009, Li et al., 2012, Li et al., 2013). That approach has been criticized since it does not estimate a causal impact, and the results may be attributable to other factors that have changed in time such as shocks, and economic cycles (Silva and Hassani, 2015) rather than the policy itself.
Results
A simple assessment of U.S. seafood imports prior to and after HACCP shows that in 1997, the U.S. imported seafood from 127 countries of which 95 (75%) were developing countries and 32 (25%) were developed countries. By the end of the study period, after HACCP implementation, the U.S. imported seafood from 110 trading partners of which 85 (77%) were developing countries and 25 (22%) were developed countries, representing declines in importers of 10.5% and 21.9%. Despite the decline in the
Discussion
Based on a program evaluation approach, we find evidence of a null effect of HACCP on U.S. seafood imports. This finding is important for at least two policy reasons: (1) Non-tariff measures are not necessarily indicative of protectionism, in part, because NTMs may redistribute export sources which may be neutral to development status. (2) NTMs on food safety may have little effect on trade because consumers may have minimal responsiveness to the policy shift.
A long-standing debate in the
Conclusion
The previous literature used a simple dummy variable for HACCP implementation to study its effect on trade and found mixed results. This paper is the first to disentangle the causal effect of HACCP implementation in the U.S. by controlling the unobserved factors affecting U.S. seafood imports using DID. Specifically, we chose imports of U.S. non-HACCP commodities and EU-15 seafood as the control groups separately to estimate the casual effect of HACCP implementation. To check the robustness of
Acknowledgement
The authors thank the editors and reviewers for useful comments that strengthened the paper. We also appreciate discussions with multiple colleagues on EU policy. Nevertheless, the authors take responsibility for any errors.
References (62)
- et al.
Quantitative analysis of the benefits and risks of consuming farmed and wild salmon
J. Nutr.
(2005) - et al.
Barriers to agricultural exports from developing countries: the role of sanitary and phytosanitary requirements
World Dev.
(2001) - et al.
Mandatory aspects of the seafood HACCP system for the USA, Mexico and Europe
Food Control
(2000) - et al.
A choice experiment model for beef: What U.S. consumer responses tell us about relative preferences for food safety, country-of-origin labeling and traceability
Food Policy
(2007) Producing to achieve HACCP compliance of fishery and aquaculture products for export
Food Control
(1999)- et al.
Saving two in a billion: quantifying the trade effect of European food safety standards on African exports
Food Policy
(2001) - et al.
Differential impacts of country of origin labeling: COOL econometric evidence from cattle markets
Food Policy
(2014) The treatment effect, the cross difference, and the interaction term in nonlinear ‘difference-in-differences’ models
Econ. Lett.
(2012)- et al.
Evaluation of worldwide approaches to the use of HACCP to control food safety
Trends Food Sci. Technol.
(2000) - et al.
Product standards and developing country agricultural exports: the case of the European Union
Food Policy
(2013)