Date of Award
Doctor of Philosophy (PhD)
Economics / Economic Policy
This study has had two major objectives. The first was to construct an econometric model of the world tea market, paying particular attention to its noncompetitive market structure. The second was to apply the model to examine whether there might be potential gains for the producing countries from cartelization of the supply of estate tea. The estate tea market was found to be monopsonistic, and the econometric model was developed wtihin this market structure. To the best of my knowledge, this is the first econometric model of a world commodity market derived within a theoretical framework of quantity and price determination in a monopsonistic market structure.
Within the model, a distinction was drawn between estate tea (or input or producer tea) and consumer tea (or output or retail tea). Estate tea is the tea which is produced as a finished product ready, except for blending, for consumption, but this estate tea does not usually reach the consumer directly. It is generally bought from the producers by an intermediary who blends, packages, brands and distributes it to the final consumers through retailers. This is the tea which is designated as consumer tea.
To examine whether cartelization is worthwhile and to estimate the potential gains from cartelization, a theoretical formulation was developed explicitly within the framework of quantity and price determination in a monopsonistic market structure. The relation between this formulation and the relevant formulation in the case of perfect competition is also shown. The econometric model was specifically designed to allow the estimates of potential gains from cartelization under alternative versions of a tea producers cartel to be calculated. The model allows one to estimate potential gains in the short-run as well as in the long-run.
Alternative versions of the econometric model were estimated by two stage least squares (2SLS) and constrained nonlinear least squares (CNLS). In the CNLS method, estimates of the structural parameters were obtained via simultaneous estimation of constrained reduced form equations. Further, a sub-model of tea acreage response equations for Sri Lanka, India, African countries, and the Rest Of the World was also estimated by ordinary least squares (OLS).
Two alternative cartels were considered: 1) Sri Lanka acts alone as a cartel, and 2) Sri Lanka and India together form a cartel. Potential gains seem to exist only in the short-run in the former case while there seems to be substantial gains in both the short-run and long-run in the latter case.
Dayananda, Muhandiramge D., "Potential Cartelization in a Monopsonistic Market Structure: A Model of the World Tea Market With Special Reference to Sri Lanka" (1977). Open Access Dissertations and Theses. Paper 3153.