Simultaneous estimation of supply and demand for United States lodging
Abstract
The objectives of this study were: (1) to specify and estimate models of demand and supply for U.S. lodging (2) use the models to evaluate changes in supply and demand simultaneously. The sub-objectives were to (1) include variables to measure the effects of economic and governmental policy changes; (2) estimate the price and income elasticities, and; (3) use the models for simulations. The data included a 24 year time series of quarterly data (1968-1991). The method of estimation used was two stage least squares regression. The simulations were performed using the Simlin procedure in SAS. Four structural equations were hypothesized to explain four endogenous variables; supply, demand, price and occupancy. Supply was estimated as a function of lagged prices and the cost of capital inputs. Demand was estimated as a function of prices, income and price of other goods. The price equation was used to represent short term supply and was estimated as a function of demand, previous quarters prices and cost of labor inputs. The system of equations did a good job of explaining the variation in the endogenous variables with all of the adjusted R$\sp2$'s being over 0.95. All of the exogenous variables were highly significant and had the appropriate signs. The simulation results were also quite good with all three simulations having a mean absolute error of less than 5.0 percent. The results indicated that recessions have a significant negative effect on the lodging industry with the latest 1990 recession having a slightly larger effect than the previous three recessions. The price elasticity of demand was estimated at $-$0.918, and the supply elasticity was estimated at 0.587. The income elasticity was 0.40. The conclusions indicate that lodging consumers are more price sensitive, and lodging suppliers are less price sensitive, than previously thought. It also appears that the lodging industry is following questionable pricing policies by increasing room rates in the first quarter of each year. In addition, variables for tax incentives during the 1980's, exchange rates, and wage and price control variables were found to have significant impacts on the behavior of the lodging industry.
Degree
Ph.D.
Advisors
Hiemstra, Purdue University.
Subject Area
Business costs
Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server.