During the last few years the economic relationships between developed and developing countries have been studied using models in which the world economy is disaggregated into geopolitical blocs, whose interaction provides a complete account of the behaviour of the world economy as a whole. These blocs are often taken to consist of the developed countries (the North), the oil-importing developing countries (the South), and OPEC. In previous CEPR Discussion Papers (Nos. 164 and 165), Research Fellow Michael Beenstock has reported estimates of separate econometric models for the Northern and Southern blocs. In Discussion Paper No. 230, he reports the results of simulations in which the two models are linked in order to investigate economic interdependence between North and South.
Beenstock argues strongly for what he terms the `econometric' approach to analysing interdependence between the rich and poor countries. He criticizes the `oracular' approach, which relies on simulation models in which `plausible' parameter values are gleaned from the empirical literature. We insist on empirical scrutiny of the models used to analyse national policies, and demand that such models describe the past to a reasonable degree. The same criteria should be used to evaluate policy at the international level as well, Beenstock argues.
In Beenstock's Northern model the main endogenous variables were GDP, inflation, interest rates and primary product prices. For the South the principal endogenous variables concerned the balance of payments and included exports, imports, capital flows, reserves and the exchange rate: no attempt was made to model the determinants of economic growth and inflation in the South, which were assumed to be exogenous. Imports and exports of each bloc were composed of primary non-oil products and manufactured products. Prices were disaggregated further into oil prices and non-oil primary product prices, as well as import and export prices.
In both of the separate models, economic activity outside each bloc was assumed to be given: thus the behaviour of the Northern economies was assumed to be given in the Southern model and vice versa. In the present Discussion Paper, Beenstock introduces interdependence in two stages. He first introduces shocks to the Northern variables in the Southern model, while these Northern variables remain exogenous (and vice versa). This helps reveal, for example, how an autonomous economic expansion in the North affects the South and how an autonomous increase in external debt in the South affects the North. In the second stage he simulates both models jointly, allowing Northern shocks to affect the South with further feedbacks to the North, while Southern shocks similarly rebound off the North back onto the South. In these joint simulations many of the previously exogenous Northern and Southern variables are endogenized, in the sense that they are allowed to affect each other. OPEC's behaviour is not modelled: in particular the price of oil is assumed to be exogenous.
Trade and capital linkages form the basis of North-South interdependence in Beenstock's models, but data deficiencies prevent him from analysing these directly. Instead he models the international (non-oil) primary product market and the international capital market, in which commodity prices and world interest rates are determined. Increased GDP in the North raises export demand in the South, for example, while the supply of Southern exports depends on relative prices and the South's productive potential. Increased export earnings raise the South's demand for external indebtedness. Balance of payments deficits also influence external indebtedness but if debt grows too rapidly the South devalues and this helps to boost exports and lower imports.
An increase in Southern indebtedness puts upward pressure on world interest rates. Higher interest rates reduce aggregate supply in the North via a fall in the capital stock. At the same time interest rates vary directly with Northern economic activity and with bond-financed fiscal deficits in the North, and inversely with OPEC surpluses deposited in the international capital market. Higher interest rates reduce the demand for inventories of primary products: commodity prices fall; this adversely affects the South's terms of trade and its international price competitiveness. Higher commodity prices adversely affect aggregate supply in the North. At the same time commodity prices vary directly with global economic activity. Finally, the relative price of non-oil commodities varies directly with oil prices, which also adversely affect economic activity in the North.
Beenstock cautions that the paper is essentially experimental and describes what happens when the two models are `docked': research of this type is in its infancy, and the results of the simulation exercises already suggest improvements to the design of the models. The simulations do suggest, however, that the North affects the South to a much greater extent than vice versa; that Southern demand for imports from the North does not appear to affect Northern economic activity to any significant extent, but that the benefits to the South of expansion in the North are magnified by the resulting increase in commodity prices and the associated improvement in Southern terms of trade.

An Econometric Investigation of North-South Interdependence
Michael Beenstock