Intra-industry trade in processed foods: economies of scale and market structure

¶ … intra-industry international trade within the standard international trade classification SITC6, which represents manufactured foods classified chiefly by material. The scope of this paper is limited to processed foods, and includes analytical frameworks from the gravity model, and classic approaches to product differentiation, product commoditization, pricing, and market structure.

References to market structures in the literature typically oversimplify the dynamics influencing the development of market types -- perfect competition, monopolistic competition, oligopoly, and monopoly -- and this tendency is exacerbated when the focus is on intra-industry trade. Further, the distinctions become considerably more obfuscated when companies that are in the same trade do business in both domestic and international markets. A global manufacturing strategy is the goal of most multinational companies, and rationalizing manufacturing strategies is an objective for nations, as well (Lee, 1984). For instance, assume that a country has the industrial infrastructure to produce canned beverages, but there are few natural resources to support the industry, productivity is low and is only growing slowly. These are the countries to which a company should locate new production. The reason for this is that global strategic advantage in manufacturing is generated predominantly by "logistical economies of scale, purchasing economies of scale, global experience, and production economies of scale" (Lee, 1986). The new competitiveness goes beyond one country producing all commodities cheaper than other countries -- outperforming international competitors is insufficient for the contemporary market (Lee, 1984). The metrics to watch today indicate whether a company is performing better relative to its own economy than a competitor performs relative to its own economy.

Economies of Scale

The literature points to a positive link between economies of scale and intra-industry trade (Brander and Krugman, 1983; Sharma, 2002). According to Harrigan (1984), examining the volume of trade directly is preferable to examining the proportion of intra-industry trade. In his words, "there is some evidence in favor of the proposition that the volume of trade is higher in sectors with large scale economies, but that inference is somewhat sensitive to the choice of proxy" (Harrigan, 1984). Consider the matter of transport costs: The welfare effects of intra-industry trade can increase if transport costs are low and remain low. However, if transport costs are high, the intra-industry opening up of trade can bring about a decline in welfare -- essentially, "the precompetitive effect is dominated by the increased waste due to transport costs" (Brander and Krugman, 1983). Where economies of scale produce benefit in an industry, countries are restrained from themselves producing a full range of products in that industry. This dynamic can shape a platform and scope for the exchange of similar products between countries (Sharma, 2002).

INTRA-INDUSTRY INTERNATIONAL TRADE CASE STUDY

The gravity model of trade is a "parsimonious and tractable representation" of global economic interaction. The gravity model has found favor with researchers as a method to examine bilateral trade flow patterns and to analyze the effects of regional trade cooperation (Anderson, 2011). The tractability of the gravity model is largely attributable to its modularity, such that, comparisons can be made across several countries. The model articulates the distribution of goods and variables across country spaces that are influenced by gravity forces -- which are conditioned in turn by the size of location-based economic activity (often represented by GDP) and the distances involved in the transactions (Anderson, 2011). The utility of the model's modularity extends capacity to disaggregate data independently from models of market structure or models of production in full general equilibrium (Anderson, 2011). Disaggregation is also independent of scale and can apply to any goods or any regions, allowing inferences to be made about trade costs (Anderson, 2011).

The conventional model of gravity is derived from an analogy with Newton's Law of Gravitation (Anderson, 2011). To whit: A corpus of labor, good, or other variables of production occurs at an origin (i) and (Yi) experiences attraction to a corresponding mass of demand at destination (j) (Ej), with the potential flow between the two locations is reduced by the distance between the origin-destination pairs (dij) (Anderson, 2007). The analogy is represented in a strict form as: Xij = YiEjdeij (Anderson, 2007).

The unit of analysis of intra-trade transactions is an origin-destination pair. Relative levels of attractiveness occur between the origin-destination pairs; the theory of bilateral flows is applied to the strength of the attraction between the elements of the pairs. Theoretically, any purchase can occur at multiple possible origins and any sale can have multiple possible destinations, such that, "any bilateral sale interacts with all others and involves all other bilateral frictions" (Anderson, 2011). The structural gravity model elegantly solves this problem of general equilibrium (Anderson, 2011).

Global oligopoly (Strategic Trade Policy)

Companies interact strategically in an oligopoly. The conventional theories regarding the economy of trade present models that differ for industries depending upon whether they are compete in business through importation, in which case they are likely to prefer protectionism, or whether they compete through exportation, which can result in a preference for industry protectionism (Milner & Yoffie, 1989). Given these positions, a large number of multinational industries look to the strategic trade policies that are currently undergoing scrutiny (Milner & Yoffie, 1989). The trade policies are considered strategic because they tend to support domestic free trade only if foreign markets open or the governments of foreign countries reduce the subsidies that they charge to firms (Milner & Yoffie, 1989).

Agglomeration Economies (External Economies of Scale)

Economies of agglomeration are described as the benefits that become available when firms locate in close proximity to one another -- in other words, they agglomerate (Arribas, et al., 2009). These firms -- through the economies of scale achieved, often realize savings and network effects -- since suppliers and customers may cluster near the plants, too (Arribas, et al., 2009). In the case of processed foods, consider that various companies that lack the infrastructure for bottling and canning processed foods may all employ the services of a common bottler or canner (Arribas, et al., 2009). Locating the various companies close to the bottling or canning operation can help to keep transport costs down (Arribas, et al., 2009).

Case Study

Food habits are becoming internationalized and developing economies are experiencing rapid urbanization (Sharma, 2002). Intra-industry trade of processed foods, particularly, is subject to influence by these factors and the variables of product differentiation and economies of scale. The Australian data includes one data set that consists of manufacturing commodity (SITC subgroups 5-8) and a second data set -- categorically is referred to as broad manufacturing -- combines commodity (SITC subgroups 5-8) and processed food (SITC subgroups 0-1) (Sharma, 2002). A 7-percent increase in broad manufacturing was seen from the 20-percent level achieved in 1979/1980 to 38% in 1992/1993 (Sharma, 2002). The primary driver of this increase is believed to be due to robust growth in intra-industry trade in commodity manufacturing that reached 29% in 1992/1993 from the 15-percent level of 1979/1980 (Sharma, 2002). The processed food sub-sectors that showed the most dramatic rise in Australia during the same time period were cereals, confectioneries, meat products, and tobacco products (Sharma, 2002).

As stated earlier in this paper, Brander and Krugman (1983) demonstrated that product differentiation generates intra-industry trade between countries, given between-consumer preference diversity and between-country demand similarity (Sharma, 2002). Product differentiation can be a form of horizontal differentiation, with different attributes -- a form of vertical differentiation -- which considers qualities -- and a form of technological breakthrough, which results in an improved product range (Sharma, 2002).

The research findings of Sharma (2002) support the hypothesis that economies of scale and product differentiation contribute to intra-industry trade, but that trade protection tends to contribute to diminished levels of trade (Sharma, 2002). Australia and New Zealand exhibit close economic links and have a "long history of close economic integration, which has been further strengthened in recent years" (Sharma, 2002). Regardless, New Zealand's small share of Australia's overall trade appears not to significantly impact intra-industry trade in Australian manufacturing (Sharma, 2002).

Conclusion

According to the comparative advantage theory of trade, countries will focus on trade characterized by transactions that are complimentary with regard to access to raw materials and competitive advantage. Put simply, countries will tend not to trade the same products. There is no real advantage to be had by such transactions and -- as the gravity theory of trade illustrates -- when distances are great in the supply-demand exchange -- the cost of transport can create disadvantage. When entire industry expands production in a particular geographic area, the average, long-run cost will be lower for each company in the industry in that region (Pugel, 2012). The difference between constant returns to scale and economy of scale is important: constant returns to scale indicate that increases in output and total cost will also increase proportionately, but average cost -- as a result of this ratio -- will be constant (Pugel, 2012). With economy of scale, however, the average cost can decrease as the quantity of output and the total cost increase (Pugel, 2012).

Ordinarily, economists consider the degree of product differentiation and economies of scale to be related to the level of intra-industry trade (Sharma, 2002). As Sharma (2002) and others have shown in their research that the coefficient of economies of scale is significant and does have an expected positive sign. However, the premise of this paper is that companies engaging in oligopolistic interactions bring about "trade in the absence of any of the usual motivations for trade: neither cost differences nor economies of scale are necessary" ((Brander and Krugman, 1983).