CAB - Mexico's Distribution Infrastructure for Fresh Deciduous Fruits: Appendix A: Methodology

Research Publications

Mexico's Distribution Infrastructure for Fresh Deciduous Fruits After the North American Free Trade Agreement
by
Dr. Juan C. Batista, Lic. Jose Armando Martinez
and Antonio Soto
Appendix A: Methodology

CATI Publication #960301
© Copyright February 1996, all rights reserved

      The purpose of this study is to determine if any significant changes have transpired in the distribution infrastructure for the trade of fresh deciduous fruit between Mexico and the United States. Two theoretical models provide guidelines from which to d evelop a study methodology. One is the system model that takes the marketing chain for a commodity and tries to understand and analyze it in its entirety as a collection of interacting components. The components work toward a common goal of taking product s grown on a farm and, ultimately, selling them to consumers after several necessary marketing functions are accomplished, e.g., assembly, processing, packaging, inspecting, labeling, shipping, distribution, promoting, and merchandising. Goldberg (1968) c aptures the concept of the system model quite well. He noted that agribusinesses need to "...be fully aware of the total commodity system in which they participate, and they must understand the interaction of its parts" (p. 3). That is, "...an agribusine ss commodity system encompasses all the participants involved in the production, processing, and marketing of a single farm product. Such a system includes farm suppliers, farmers, storage operators, processors, wholesalers, and retailers involved in a co mmodity flow, from initial inputs to the final consumer. It also includes all the institutions which affect and coordinate the successive stages of commodity flow, such as government, futures markets (if they exist for the particular commodity) and trade associations" (p. 3).
      The other model that provides guidance is the marketing-margin model (hereafter referred to as the margin model). Tomek and Robinson (1981) define the margin model alternatively as "...(1) a difference between the price paid by consumers and that obtaine d by producers, or as (2) the price of a collection of marketing services which is the outcome of the demand for and supply of such services" (p. 120).
      Under the first definition, a marketing margin is simply the difference between primary and derived demand curves for a particular product. Primary demand for a product is determined by the response from the ultimate consumers. Empirical estimates of pri mary demand are made from retail price and quantity data. Derived demand is based on price-quantity relations that exist either at the point where the products leave the farm or at intermediate points where they are purchased by processors, wholesalers, a nd/or retailers.
      Primary demand is in some sense a joint demand for all inputs in the final product. Thus, a food product at retail may be divided (conceptually) into two parts: the farm-based components and the processing-marketing components. Given several simplifying assumptions, derived demand for a farm product is obtained by subtracting the per-unit cost (prices) of all marketing components from the primary demand function. Thus, the farm-level function represents the derived demand for the farm portion of the fina l product.
      The concepts of primary and derived supply are analogous to those for demand. Primary supply, however, refers to the relationship at the producer level. Empirical estimates are based on farm-level data. The supply relation at the retail is derived from t he primary relation by adding the appropriate margin.
      The retail price of a product is established at the point where primary demand and derived supply intersect. The farm-level price is determined at the intersection of primary supply and derived demand. The difference in the two prices is the marketing ma rgin. These intersections, prices and the margin are illustrated in Figure A-1 through a simple rendition of the margin model.

fig23

      A margin can also be defined as the price for a collection of services provided by the agents operating along the marketing chain. The services include assembly, processing, transportation, handling, packaging, merchandising, storage, advertising, and re tailing just to mention some of the more common ones. This definition parallels the concepts articulated by Goldberg.
      The system and margin models can be combined. Separating Goldberg's system into sectors (e.g., farm or processing or retailing) allows the separate sectors to be linked as a series of markets each having its respective margin. The combination is illustra ted, at least conceptually, in Figure A-2. The model illustrated suggests a commodity system consisting of five sectors (markets): farm inputs, farm, processing and manufacturing, and wholesaling and retailing. The sectors are linked by a series of market s in which farmers buy inputs and sell raw commodities to food processors and manufacturers who, in turn, produce food products for sale and distribution to consumers through a wholesale and retail sector. The symbols "p" and "q" represent the prices and quantities axis for each market. The solid lines within each of the graphs are the supply and demand curves for each market and they are denoted by the "S" and "D." The illustration does not show equilibrium; the graphs are used to convey concept rather than equilibrium. As a system, there would be price differences in equilibrium between adjacent markets/sectors. Markets would equilibrate as well as the system as a whole. The price differences between any two markets would represent the price of the col lection of services contributed by the various sectors.

fig24

A Practical Approach

      The marketing-chain margin model suggests the course a commodity takes to get from a farm to a consumer as well as the prices (or costs) associated with the subsequent steps the commodity passes through along the way. One modification needs to be incorpor ated to this model for the purpose of this study, however. Trading between countries adds an element to a commodity's marketing system that does not exist in a domestic sale or transaction: a border. Figure A-3 shows a simple modification to include an in ternational border between processors and wholesalers. The implication here is that processors are, say, American firms, while wholesalers are Mexican firms, i.e., the wholesalers operate exclusively in Mexico and not in the U.S., although they participat e in the marketing system that takes U.S. grown commodities and sells them to Mexican consumers. In addition, this extended model incorporates border agents (e.g., customs brokers and officials, freight forwarders, etc.), who are not required for a "domes tic" transaction.

fig25

      To obtain information on the infrastructure that handles, transports, inspects, etc. commodities involved in trade between Mexico and the U.S., the model in Figure A-3 suggests all sectors need to be surveyed. Unfortunately, a census of the agents in the system would be too costly. An alternative is needed. The principles and concepts embodied in the aforementioned model provide the impetus for a practical approach to collect information and data.
      With a few assumptions, the model implies that a smaller sample can be surveyed while still obtaining rich information on the infrastructure. Since most product passes through the middle sectors of the marketing chain, it can be assumed these firms are w ell versed in the operation of the entire system. This is not an unreasonable assumption because the livelihood of these firms depends on product passing through their facilities. Furthermore, the number of firms operating within the middle portion of the system is relatively smaller than the number of farms and consumers. It is assumed, therefore, that the relative cost of surveying only the "middle-level" firms would be less than surveying all agents in the system.
      The approach can be further simplified while still maintaining a high degree of robustness. If it is assumed the agents who have been in business successfully for some time and/or handle substantial volumes of product passing through the system are at le ast as well informed as any competing agent, then the sample can be reduced by omitting smaller and/or newer firms. Thus, a sample of large, well-established firms should suffice for the survey. Of course, border agents will need to be included in the sam ple.
      Survey of Agents: A sample of marketing and border agents was selected from lists obtained from the Mexican government, the USDA, and trade associations affiliated with the fresh fruits included in this study. Firms and agents that exhibited success and/ or longevity were selected for the sample.
      A questionnaire was designed and tested. It was designed to capture any new changes since Batista and Hagen did their study of Mexico in 1993. Their study serves as a bench mark for Mexico's infrastructure prior to the implementation the NAFTA. The surve y instrument appears in Appendix A. The questionnaire was administered in person or by telephone. Respondents were asked questions regarding many aspects of the infrastructure, but in general, the queries pertained to type of enterprise, mode of transport ation, cost of transporting, time in transit, port of entry, border crossing procedures and protocol, warehousing capacity and facilities, documentation, customs, obstacles, and to the devaluation of the peso. (The portion of the questionnaire devoted to the devaluation of the peso was added because the devaluation occurred in the middle of the study. In addition, it was apparent the value of the peso was affecting the volume of trade between the neighboring countries, which in turn impacts the use of the infrastructure).
      Secondary data and information were also collected. This served two purposes. First, it supplemented the information collected through the survey. Second, it can facilitate cross checking information collected through the questionnaire.

{ page top }

CAB Research Publications , Table of Contents , Previous page

{ CATI , also CAB , CFSNR , CIT , VERC }