Date of Award


Document Type


Degree Name

Doctor of Philosophy (PhD)



First Advisor

Stephen Farber


The primary objective of this dissertation is to model the regional markets for traditional construction aggregates, sand, gravel, and crushed stone. The long-term dynamics of the construction industry related to the demand for aggregates is modeled through a partial adjustment mechanism. The dynamics of the construction industry is introduced into the estimation of a simultaneous equations model of the demand and supply for aggregates at district levels. An engineering-based cost function for extraction of sand and gravel is estimated based on data collected from an operator's survey in the states of Alabama, Arkansas, Louisiana, Mississippi, and Texas. A step-supply for each state is approximated by relating the estimated cost function to output for thirty six different mine types. Finally, regional trade is evaluated based on a multimarket spatial equilibrium model of the Samuelson-Enke type for a region of nine adjacent districts in the states of Alabama, Arkansas, Louisiana, Mississippi, and Texas. Construction activity is found to depend on lagged values of construction activity, population, and changes in per capita income. Two-stage least squares is used to estimate supply and demand for aggregates. The empirical estimates show the demand for aggregates to be relatively inelastic to its own price, positively correlated to construction activity, and positively related to construction workers' wages. The long-run supply for aggregates is established elastic. Supply of sand and gravel is negatively related to the lagged price of crushed stone and vice versa. The estimated engineering-based cost function shows operating economies in the sand and gravel industry. Geological factors are found to be important determinants of operating costs. A continuous approximation of the step-supply functions for extraction of sand and gravel is used to determine the short-run extraction supply elasticity, which is found inelastic. No trade is found across districts at this time. Relatively high transport costs are believed to account for this result. The spatial equilibrium model's results are confirmed by the survey since on average 80% of the shipments travel less than 75 miles in 1991.