國
立 政 治 大
學
‧
N a
tio na
l C h engchi U ni ve rs it y
76
The discussion regarding multifunctionality of agriculture towards agricultural trade has been present in the WTO and the OECD since the 1990s (although the concept is much older). The organization has defined it as part of the non-trade concerns, and with the support of several countries (mostly the EU), it has included it as one of the elements that trade negotiations must take into account when dealing with agricultural trade liberalization.
The WTO, in its webpage, states that “The Agriculture Agreement provides significant scope for governments to pursue important non-trade concerns such as food security, the environment, structural adjustment, rural development, poverty alleviation, and so on. Article 20 says the negotiations have to take non-trade concerns into account” (WTO Secretariat, 2001). Where the discussion has focused, is on whether multifunctionality should be protected under Green Box measures (non-price distorting), or if other policies could be legitimately adopted to protect a sector which bring so many benefits to the country in different aspects. Japan, South Korea and Norway have been particularly strong defendants of the multifunctional benefits of agriculture, and its need to defend it, while “many exporting developing countries say proposals to deal with non-trade concerns outside the “green box”
of non-distorting domestic supports amount to a form of special and differential treatment for rich countries. Several even argue that any economic activity — industry, services and so on — have equal non-trade concerns, and therefore if the WTO is to address this issue, it has to do so in all areas of the negotiations, not only agriculture” (WTO Secretariat, 2001).
The debate is far from over, institutionally and academically, as chapter 4 will showcase.
For the purpose of this research, the WTO definition of multifunctionality will be taken into account as a rationale for protectionism in order to analyse the potential policy alternatives.
4. Methodology and operationalization
Single case study with embedded levels of analysis is the methodology chosen for the current research. The rationale for selecting this methodology lies on two particular points: (1) Taiwan represents a unique case when it comes to viewing it as an entity that’s negotiating from a position of non-diplomatic recognition, but (2) at the same time it represents a typical case of agricultural protection in East Asia, along the lines of the protectionist policies established by the likes of Japan and South Korea. Therefore, single case study is justified under both these rationales, giving the case’s uniqueness and its parallel typicality.
This research understands case studies, in the words of R. Stake, as “the study of the particularity and complexity of a single case, while trying to understand its interaction with important circumstances” (Stake, 1995). In the case of this research, the Taiwanese case
‧
國立 政 治 大
學
‧
N a
tio na
l C h engchi U ni ve rs it y
77
study will serve as the basis for understanding the phenomena of partial protectionism.
Within it there are embedded three levels of analysis, which represent layers of influence towards policy-making. A case studies methodology is the best possible approach to reach a complete understanding of the dynamics that influence the Taiwanese meat market, firstly because the existing quantitative analysis fail to understand the particularistic aspect of Taiwanese politics of foreign trade, giving only a measurement of the level of protection that the market has in the sectors under analysis. Secondly, case studies allow for the development of comparative analysis between the perspective of the different bureaucracies under analysis, and of the different countries and IOs being researched as factors of protectionism.
As George and Bennet have pointed out (2005), “case studies are generally strong precisely where statistical methods and formal models are weak”. Available quantitative data is not always the best data for understanding policy preferences and policy choices. There are 4 strong advantages of case study methods that make them valuable in testing hypotheses and particularly useful for theory development:
1) Potential for achieving high conceptual validity 2) Strong procedures for fostering new hypotheses
3) Value as a useful means to closely examine causal mechanisms in individual cases
4) Capacity to address causal complexity
The hypotheses under this research require an analysis of causal complexity, which will most probably differ from level to level, giving rise to a global understanding of the structures that operate towards the meat market liberalization (or protectionism) of Taiwan. A simple statistical test of economic complementarity would hardly explain the reasons why different countries have different levels of market access; for example, for the HS codes 0202 (frozen beef) and 0207 (poultry), Brazil is the largest worldwide exporter, having large market shares in Japan, South Korea, China and Hong Kong, while it’s access to the Taiwanese market is denied. Descriptive statistics will be used nonetheless for comparative purposes in order to understand the market potential of the non-market access countries under analysis.
However, the ordinal scale established for this analysis will be better suited to account for the specifics that Taiwan’s agricultural trade policy establishes. This ordinal scale of protectionism has been established from the standpoint of several alternative theoretical approaches.
‧
The research design can be understood through the following flowchart:
Fig. 7. Research Flowchart
A qualitative methodology will be used, considering the main criteria of scientific rigor as follows:
Construct validity: The concept being studied under Taiwan’s agricultural trade policy is that of protectionism. As Bhagwati argues, protectionism is influenced by ideas, interests and institutions. At the same time, Milner argues that the formulation of a nation’s trade policy involves a struggle among domestic groups, the national government, and foreign governments. In order to achieve a robust validity of measurements, the operational measures selected involve ideas, interests and institutions in the three levels of analysis. Firstly, the ideas influencing policy (liberal or protectionist) will be analysed as a source. Secondly, through the application of the concepts developed by rational choice theory, interests and power will be analysed according to gains as they are perceived by the players involved or affected by this policy. Thirdly, the theoretical propositions derived from historical institutionalism will serve the purpose of analysing institutions from the point of view of their historical dynamism. Construct validity will rely on the following tactics:
a. In order to strengthen it, this research will use multiple sources of evidence (documentation, archival records and interviews). Following Robert Yin’s book (2008, p. 114), the different sources of data will be triangulated and corroborated. “The most important advantage presented by using multiple sources of evidence is the development of converging lines of enquiry, a process of triangulation and corroboration. When you have really triangulated the data, the events or facts of the case study have been supported by more than a single source of evidence” (Yin, 2008).
Research
‧
國立 政 治 大
學
‧
N a
tio na
l C h engchi U ni ve rs it y
79
b. A chain of evidence will be developed in order to allow an external observer to follow the derivation of any evidence from initial research questions to ultimate case conclusions.
Internal validity: In order for causality in this research to be non-spurious, this research will use the explanation building method. The research will make initial theoretical statements about policy. Then it will compare the statements with the findings in order to revise it or approve it for both rival explanations. This approach is appropriate to this research of theory-testing through case study methods, since it devises a protectionism scale with the goal of re-defining the concepts of selective protection and selective liberalization; thus the theoretical framework gives causal explanations to the “why” question addressed, but they do not go as far as to define a particular policy, which builds up from the results of causality.
As Robert Yin acknowledges, with this kind of tactic the final explanation does not need to be fully stipulated at the beginning of the study, does differing from a pattern-matching approach. The method developed relies on iteration throughout the research, which will move from making theoretical statements to comparing the findings of the case against the statements, and then to revising the statements or propositions. The protectionism scale devised was established in order to better categorize the comparison between be theoretical statements and the findings. It will also serve to bring the embedded levels of analysis of the case under research back to a macro picture where final conclusions can be made regarding Taiwan’s agricultural trade policy and how to define it. The protectionism scale devised uses parameters that derive from ideas, institutions and interests, as it is the case with the unit of analysis.
External validity: Protectionism can be viewed from different theoretical prisms. This research has chosen the historical institutionalism and rational choice as they have been developed to address matters regarding trade policy formulation and change as it is influenced from the different levels of analysis. “In analytical generalization, the investigator is striving to generalize a particular set of results to some broader theory”
(Yin, 2008). Due to the fact that the current unit of analysis is complex thus needing an embedded (as opposed to holistic) case study, two theories have been chosen to give the research stronger external validity in its different propositions.
Reliability: To increase the reliability of this research, and show that the same results can be obtained with the same sets of data, the researcher will establish a
‧
國立 政 治 大
學
‧
N a
tio na
l C h engchi U ni ve rs it y
80
case study protocol which can be followed in its logic, and replicated with access to the same sets of data. The data which is not easily attainable (interviews) will be transcribed in full11, but at the request of many of the interviewees, it will not be in the annexes of the research in order to protect the identity and information shared by these persons or organizations.
Fig. 8. Case Study Protocol Table
Overview of the case study project
-Background information: Taiwan’s agricultural trade policy is relatively closed, with only few foreign players able to participate in Taiwan’s meat markets.
-Issues to be investigated: Causal explanations for Taiwan’s partially protectionist agricultural trade policy in different levels of analysis including domestic institutions, bilateral interactions and ideational elements.
-Relevant reading about the issue: Agricultural protection in Taiwan has been treated from economic perspectives (Anderson, 2010) (Francks, 1999), as well as from political perspectives (Chen H.-H. , 2004) (Jen, 2006). But when aiming for causal explanations, only economics literature has looked for answer to Taiwan’s behaviour leaving the research incomplete.
Introduction to the case study
-Case study question: Why is a partially protectionist trade policy applied to Taiwan’s meat trade?
-Hypotheses: Six hypotheses were devised, in order to examine through rival explanations (historical institutionalism and rational choice) the causal elements behind a partially protectionist trade policy.
-Theoretical framework for the case study:
-I. Historical institutionalism: Institutions have an inertia that requires an analysis of the historical structure in order to establish the causal elements
-II. Rational choice: Actors are self-interested profit maximizers, thus powerful actors will tend to impose their preferred alternatives given that there are power asymmetries
11 Given that the interviewees grant permission to publish these transcriptions.
‧
國立 政 治 大
學
‧
N a
tio na
l C h engchi U ni ve rs it y
81 Data collection
procedures
-Interviews: Open-ended interviews with players involved in Taiwan’s meat trade
-Archival records and official documents: Documents published by the relevant authorities, as well as IOs (WTO, APEC, OCDE) will be collected and analysed
-Media: Online databases of media will be used to test and correlate existing data
-Prior research: While specific prior research on meat protectionism is inexistent, several studies have dealt with Taiwan’s economic development, its agricultural policies, and its agricultural trade policies, which are highly relevant for this research
This research applies an interpretative paradigm, which considers that the ontological reality depends on the meanings that people attribute to it, this is, what people do or say is a product of how they define their world, and it will be particularly relevant when analyzing the different actors involved in the meat market; in terms of epistemology is subjectivist, stating that the researcher studies the process of interpretation that social actors make of their reality emphasizing the process of understanding it.
In order to be able to carry out this research, the researcher will rely on multiple sources of evidence, focusing on archival records, documents and more importantly interviews. The first source to be relied upon are the documents, through a literature review and analysis. A review of the existing literature on the topic will be covered with the goal in mind of identifying the main actors on the field, previous causal explanations, and the trends on research on the topic. While this particular topic registers no previous publications in English, protectionism and trade liberalization, as well as NTBs and agricultural price distortions, have ample bodies of literature available, to which this research aims to contribute from the Taiwanese case study. This review will cover all the journals relevant to the research, book publications, media and official statements. A content analysis will not be carried out for the simple reason that the meaning behind the words will not be explained without its context.
Therefore, a more exhaustive method of analysis will be required to comply with the goals of this research. Babbie has qualified this methodological typology as “comparative and historical research”. “The comparative and historical researcher must find patterns among the voluminous details describing the subject matter of study” (Babbie, 2007).
To complement the documents, a trade data analysis will be necessary to correlate the existence of complementarity between different countries with Taiwan’s meat market in order
‧
國立 政 治 大
學
‧
N a
tio na
l C h engchi U ni ve rs it y
82
to make sure that there is a viable interest on the parties to enter the market. Using descriptive statistics, and gathering trade data from Trade Map (in order to have a unified source), the research will first look at the main worldwide exporters of the HS codes under analysis. Then, these will be correlated with their presence in the northeast Asian markets of Japan, South Korea, Hong Kong and China, in order to see if they are able to comply with the sanitary and phytosanitary standards of these countries at the same time that they can arrive in the region with a competitive price in order to have a viable distribution. Finally, these lists will be correlated with Taiwan’s imports of meat, in order to see which countries have potential to be a player in the Taiwanese market, but are impeded due to the barriers erected, or possibly (and less likely) a lack of interest.
The second source of evidence to be utilized is archival records. Official documents will be reviewed from the all the sub-units that are going to be researched from the three levels of analysis. These include the domestic institutions (MOFA, BOFT, FDA, BAPHIQ, COA), the trade-related organizations of the countries under analysis (those with market access, and those without), and the two international organizations (APEC and WTO). The documents to be considered differ from one level to the next. For the case of the domestic institutions the documents utilized are the following:
- Official policy statements
- Laws related to agricultural trade - SPS regulations
- Decrees
- Internal bureaucratic regulation documents - Trade policy review
For the countries under analysis, the documents to be reviewed are the following:
- Official documents regulating food exports - Barriers to trade cadaster
- Market intelligence reports - Trade policy reviews
For international organizations, the documents to be reviewed are the following:
- Founding documents
- Official statements regarding agricultural trade - Policy recommendations
‧
國立 政 治 大
學
‧
N a
tio na
l C h engchi U ni ve rs it y
83
The third source of evidence will be represented by interviews to the relevant actors involved in each level. In order to fulfill the qualitative open-ended interviews, a set of relevant areas of enquiry will be devised to cover all the relevant topics that could deliver the information needed. As previous research has shown, qualitative interview is best served by
“including the topics to be covered, but not a set of questions that must be asked with particular words and in particular order” (Babbie, 2007). This tactic of data collection will give the research an open area of discussion, where further causalities can be explored through the answers of the interviewees. An approach will be made to trade-related bureucrats and diplomats from Taiwan as well as from the countries under analysis, private sector players and academics. The process of interviewing will include the design of the topics, the interview itself, transcribing of information, analyzing of the data, verification of the information and reporting the gathered data. Throughout this process, the door will be open for a second or third visit to the same interviewees, given that further inquiries are raised concerning the results of the analysis. For the interviews of scholars, a similar process will be followed. The aim is to discern the process that leads the different agents to establish a protectionist or a liberal strand towards the meat market, and the reason and operationalization of it.
The abovementioned description of the data collection techniques does not reflect the order in which they will be used during the research, as they will be used according to the needs of the different chapters to be covered. In order to operationalize the current research, and take it from the abstract to the empirical level, a four step research process will be conducted in order to establish a clear pattern of research.
1. Literature review and compilation:
The compilation of data will be the first step to be taken, in order to have a clear picture of what has been researched and what has not. The compilation of the literature will follow an ordered distinction between the different issues analyzed, giving the researcher the chance to organize the materials. Besides giving the basis for research, the compilation of data will give the researcher the chance to enter the existing dialogues of the academic field.
This process will also include the compilation of quantitative data. It can be subdivided in two separate sets of data. The first one involves the compilation of comparable trade data.
In order to do that, the researcher will use the uniformed source provided by Trade Map, gathering all the trade data available regarding meat trade to Taiwan, between Taiwan and the relevant case studies, between the relevant case studies and the Northeast Asian region,
‧
國立 政 治 大
學
‧
N a
tio na
l C h engchi U ni ve rs it y
84
and from the relevant case studies towards the rest of the world. To complement this first set of data, detailed data of domestic meat production will be required, as the local producers would represent (according to rational choice and economic theories) the main target of protection.
The second set of data is the one gathered by Anderson and Nelgen, where the nominal rates of assistance (NRA) and the relative rates of assistance (RRA) for Taiwan will be compiled for the years between 2001 and 2011 (last year available). This data will allow the researcher to first proof that there is a certain level of protection in Taiwan’s agricultural
The second set of data is the one gathered by Anderson and Nelgen, where the nominal rates of assistance (NRA) and the relative rates of assistance (RRA) for Taiwan will be compiled for the years between 2001 and 2011 (last year available). This data will allow the researcher to first proof that there is a certain level of protection in Taiwan’s agricultural