What are the patterns of world trade?

26.3.2.1.1 Skill type

The split into skill types, usually understood as level of education, responds to the huge amount of literature on skill-specific wage disparities and their possible reasons (skill-biased technological change and shifts in international trade patterns in the course of globalization). It acknowledges that wages do not always move in parallel, which becomes relevant in situations where differential effects on labor markets of different skill types are plausible. A typical example of a situation of this sort is trade liberalization, which changes the exposure of a country with imports from regions with different comparative advantages (e.g. Thierfelder and Robinson, 2002; Carneiro and Arbache, 2003). Similar effects can be the consequence of sectoral reallocations due to tax policy or climate policy measures.

Most of the literature on skill-specific labor market effects uses a split into two classes, high- and low-skilled (with a conventional cutoff point analogously to completed college education in the US). This follows a long tradition of attempts to estimate the substitution pattern between these two skill classes and capital (see Thierfelder and Robinson, 2002). Conceptually, it is easy to extend the skill split to more than two classes. However, the more skill classes, the more challenging labor demand estimation, which becomes more likely to produce implausible substitution patterns (see Section 26.4). In addition, the more skill classes, the less plausible the implicit claim that skill is an unchangeable attribute of the households (i.e. that individuals cannot switch from one skill class to another). Jung and Thorbecke (2003) and Cloutier et al. (2008) are two examples in which the choice of the skill type is endogenous, involving investment in education. Jung and Thorbecke (2003) work in a recursively dynamic context and let the education decision be governed by myopic expectations. The model of Cloutier et al. (2008) is static, representing a long-term equilibrium. Transformability of skills is imperfect (CET function) and the choice between skills is driven by contemporary wages.

A data-related issue with skill classes in multicountry models is the problem of comparability of skills data across borders. The larger the differences between educational systems, the higher the obstacles to finding comparable data. Dimaranan and Narayanan (2008) explain how the skills split is implemented in the GTAP (Global Trade Analysis Project) context. As detailed data are only available for a subset of the countries covered by GTAP, they estimate a functional relationship between the share of skilled labor payments, growth of GDP per capita and the average number of years of tertiary education. This is used to generate values for the countries with missing data.

International trade is the physical movement and electronic transfer of goods and services across national borders. The attributes of trade flows of interest to geographers include the direction, composition, and magnitude of trade. To analyze the geographical aspects, or ‘geographies,’ of international trade patterns, geographers make use of the following core spatial concepts: place, location, distribution, spatial interaction, spatial scale, change, region, and a variety of potentially constraining or limiting forces. Prior to the early 1990s, most of the research by geographers was focused upon commodity movements between countries, with little attention given to the evolving theory of international trade or to the interrelationships that exist between the changing global economy and the geographies of sub-national places. More recently, geographers have begun to intensify their research on international trade, and have been focusing this work at both the macro- and microlevel spatial scales. This research has important implications for world competitiveness, the internationalization and changing spatial structure of economic activities, global–local relationships, government policymaking, and national and regional economic growth and development strategies.

Economics: science of commonsensibles

Land, labour, and capital. Markets, money, prices. Private good, public good, merit good. Demand and supply. Individuals, households, business firms, central banks, government bureaus. Preferences, expectations, greed and fear. Cost and choice, budget and benefit. Competition, contract, convention. Auction, arbitrage, alertness. Risk, uncertainty, learning. Exchange and externality, property right and moral hazard. Wages, profits, taxes, subsidies, fairness. Saving and investment. Debt, mortgage, interest, trust. Unemployment, inflation, growth, recession. Trade, exports, imports. Competitiveness, comparative advantage, exchange rate. Gross Domestic Product, Sustainable Economic Welfare.

None of these look like electrons and their properties and behaviour. Indeed, there is a long tradition in economics of viewing the basic constituents of its subject matter as being familiar to us through commonsense experience. This has often been presented as a source of an epistemic advantage enjoyed by economics compared to physics. There is no direct access to the ultimate constituents of physical subject matter, so physics must infer to them from their effects, while economics deals with a domain that is more directly accessible through ordinary experience [Cairnes, 1888, 84]. This has implications for realism: “… the ultimate constituents of our fundamental generalizations are known to us by immediate acquaintance. In the natural sciences they are known only inferentially. There is much less reason to doubt the counterpart in reality of the assumption of individual preferences than that of the assumption of the electron” [Robbins, 1935, 105; see also 78–79]. Note that this optimism goes beyond mere minimal realism.

The idea should not be that economics deals with things of which we can have direct sense perception. The preferences and expectations of economic agents are unobservable in the sense of being inaccessible directly by senses. So are, say, multinational companies and the revenues they make and the institutional constraints they face. But they, just as money and prices, salaries and taxes, are familiar parts of our commonsense view of the social world within which we live our daily lives. These are commonsensibles rather than perceptibles or observables in any strict or pure sense. Commonsensibles involve concepts and inference, cultural meanings and shared interpretations - they involve the unavoidable hermeneutic moment of economics.

A further remove from being directly observable is due to the various modifications that the economically relevant commonsensibles undergo when theorized and modelled by economists. Goods exchanged in markets may become perfectly divisible in theoretical models. The messy local preferences familiar to us become transformed into the transitive and complete preferences of expected utility theory. Our local and flawed expectations become transformed into comprehensive rational expectations. Ordinary mortal people like us become infinitely lived agents in some models. Time-consuming and otherwise costly transactions become free and instantaneous. Internally structured business organizations with multiple goals become modelled as devoid of internal structure and pursuing nothing but maximum profits. Strategic rivalry between powerful price-making firms becomes non-strategic perfect competition among powerless price-taking firms. The (institutionally, culturally, politically and otherwise) complex structures of international trade of multiple goods between multiple countries become modeled as perfectly free trade of two goods between two countries with same technologies and same consumer tastes. Mathematical techniques of representation often make the items in such models even more poorly recognizable from the point of view of commonsense experience.

Given such theoretical transformations, one may wonder whether economics is really about commonsensibles after all. A natural doubt is that the items that appear to be talked about in economic theories and models are too far removed from ordinary experience and commonsense frameworks to qualify as commonsensibles, so it would be better to liken them with unobservables akin to electrons and viruses. In response to this doubt, one can argue that in modifying commonsensibles by various simplifications and idealizations the theorist does not thereby introduce entirely new kinds of entities and properties. There is no radical ontological departure from the realm of commonsense items when moving from boundedly rational to perfectly rational agents, or from messy preferences to well defined preferences, or from costly transactions to free transactions, or from time-consuming and permanently out-of-equilibrium processes to instant adjustments to market equilibria, or from particular market prices to inflation rates. Commonsensibles are modified by cognitive operations such as selection, omission, abstraction, idealization, simplification, aggregation, averaging. None of these amount to postulating new kinds of unobservable entities (but see the queries about macroeconomic entities by Hoover [2001]).

It is also possible to turn the above doubts about commonsensibles into doubts about existence. Not only do the idealizing modifications of ordinary items take them away from the commonsense realm, but they are taken away also from the realm of existents. Those idealized entities are fictions rather than candidates for real things. The philosophically minded economist Fritz Machlup has suggested just this idea. His prime example was the neoclassical or marginalist theory of the perfectly competitive firm [Machlup, 1967]. The theory depicts firms [1] as devoid of internal structure; [2] as perfectly informed; [3] as taking price signals as the only source of information; [4] as not interacting through rivalry; [5] as price takers rather than price makers; [6] as pursuing maximum profits as their only goal.

Machlup's reasoning is straightforward in concluding that in the neoclassical theory of the competitive firm, “all firms are pure fictions” [1967, 30], that the theory postulates “this purely fictitious single-minded firm” [1967, 10]. He warns against failing to keep apart these fictional theoretical firms and the familiar real organizations also called ‘firms’. This may be interpreted as a claim about reference, implying that the term ‘firm’ in the neoclassical theory of the competitive firm fails to refer to real firms; and not only does it fail to so refer, but it is not even purported to refer. In the old-style instrumentalist manner, it is just an “intermediate variable” that serves useful functions in scientific inference without itself being connected to any real entities.

This conclusion derives from an implicit premise, the description theory of reference [Mäki, 1998]. According to this theory, the factual reference of a term is determined by the associated descriptions. Whenever those descriptions do not fit with anything in the world, the term fails to refer. So, the term ‘firm’ in the neoclassical theory of the competitive firm fails to refer to real firms simply because the assumptions of the theory [1]–[5] are not true of any real entities in the social world. This is why the assumptions cannot be used for identifying any real objects of reference.

There is no established account of reference to social objects that could here be appealed to in response to Machlup's fictionalism, but an obvious point of departure would be Donnellan's [1966] distinction between the attributive and referential uses of definite descriptions, the latter enabling the use of false descriptions referentially. We could generalise on this idea and suggest an analogous distinction for general descriptive terms such as ‘firm’: the assumptions associated to the term in neoclassical theory can be used both referentially and attributively. Machlup's suggested primary use of ‘firm’ is attributive. This enables him to infer to the conclusion that because nothing in the world satisfies the description provided by neoclassical assumptions, the term in this theoretical context is not to be used to refer to any things in the world. The alternative is to use the term referentially. In this case we hold that even if nothing in the social world satisfies the attribute of being a perfectly informed atomistic price-taking maximizer, it is not out of the question that the description can be used to pick out a class of non-fictional entities, namely real business firms. Just as I can be mistaken about your age and shoe size and still be talking about you, economists may employ false assumptions about firms and yet talk about them.

Once reference to real things is secured, we are ready to ask for the rationale for false descriptions. The literature in the theory of reference usually cites error, ignorance, and incompleteness as sources of falsehood. However, when an economist makes unrealistic assumptions, she is often not making an error or being ignorant. She is instead deliberately employing strategic falsehoods in order to attain some epistemic and pragmatic gains. This is the point of idealization.