Scope of the three-component model

Alonso’s Systemic Model: A Review and Representation

“…Naturally, people would like to ask whether the systemic model and this family are equivalent to each other-in other words, whether the former is indeed more general. Wilson (1980) and Ledent (1981b) regarded the two systems as equivalent, while Hua and Porell (1979) and Merkies and Meers (1989) did not agree with this view, considering that M i+ and M +j are treated as endogenous in the systemic model. Wilson (1980) quoted ( 14) and ( 15) in Hua (1980) as demonstrating that M i+ and M +j can indeed be separately determined.…”

“…Anselin (1982) took them as "quasi-elasticities," while Frotheringham and Dingnan (1984) took them as measuring the accuracy with which V and W represent M i+ and M +j , respectively. Merkies and Meers (1989) demonstrated α and β as two composite parameters containing substitution effects pertaining to relative prices.…”

“…Extension from cost to price is necessary if the equilibrium connoted in the systemic model is to be justified in economic theory. This was attempted by Merkies and Meers (1989), who tied in the systemic model with an international trade model by applying an extended version of Armington's (1969) model. In this application, α and β thus contain relative price information, although in a complex way.…”

“…When M i+ and M +j are unknown (endogenous), a case the systemic model intended to address, Ledent presented the procedure as described in the last section to identify the model. Merkies and Meers (1989) took a more elaborated discussion on the conditions for simultaneously determining M i+ , M +j , and M ij (which was also formulated by Bikker and De Vos [1980] as the three components or 3-C model). Counting the number of unknowns, including α and β, and the number of equations in the systemic model, they concluded that it is an underidentified system, unless extra exogenous restrictions are added.…”

“…For identifying Alonso's original systemic model, another proportionality should also be specified. In contrast to Merkies and Meers's (1989) pessimistic conclusion, Hua took up the issue in a constructive view: this underidentification leaves one or two degrees of freedom to a user who can bring the model to closure, choose to impose a restriction, add a postulation, or define a numeraire to suit particular needs in each subject area of application.…”