Forecasting construction manpower demand: A vector error correction model

Manpower demand forecast is an essential component to facilitate manpower planning. The purpose of this paper is to establish a long-run relationship between the aggregate demand for construction manpower and a group of inter-related economic variables including construction output, wage, material price, bank rate and productivity, based on dynamic econometric modelling techniques. The Johansen co-integration procedure and the likelihood ratio tests indicate the existence of a long-run and stable relationship among the variables. A vector error correction (VEC) model is then developed for forecasting purposes and is verified against various diagnostic statistical criteria. The construction output and labour productivity are found to be the most significant and sensitive factors determining the demand of construction manpower. The model and the factors identified may assist in predicting manpower demand trend and formulating policies, training and retraining programmes tailored to deal effectively with the industry's labour resource requirements in this critical sector of economy.