As a driving engine for growth, the construction industry plays an important role in a country’s economic development process. Despite that, the industry is vulnerable to cyclical oscillation and at times more fundamental changes in work volume when the regional or global economy is hit by unforeseen events. In order to formulate appropriate policies and directions to help ease the impact of a fluctuating volume of construction work, a model that can reliably predict the work of various construction sectors after any economic turbulence would be extremely useful. In this study, the Box–Jenkins approach is used for model development due to its simplicity and sound theoretical background. The results illustrate that the Box–Jenkins models can reliably predict the medium‐term total construction demand and residential demand covering a turbulent period of ups and downs in construction demand. A multiple regression model is also developed to compare against the modelling reliability of the Box–Jenkins model.Construction demand, forecasting, Box–Jenkins technique, economic environment,

As an economy approaches maturity, the relative importance of the construction sector would gradually decline. Without effective policies and strategies, the construction industry will suffer irreversably and this may trigger a knock-on effect to the overall economy. The experiences of selected advanced economies are examined, including Australia, Japan, Singapore, South Korea and the UK in coping with structural changes in construction. Strategies employed by the government and the construction industry of these advanced economies to reinvent the construction industry were captured through an extensive literature review and a series of interviews with indigenous industry practitioners. In order to revitalize the construction industry, governments would accelerate publicly funded projects; provide financial support to ease the burden of industry stakeholders; and stimulate the market demand. In contrast, the industry would explore various market alternatives while companies would sharpen their competitive advantage locally and internationally through merger and acquisition. The identified strategies are corresponded to a construction industry development framework, which could therefore serve as a valuable reference for policy makers and practitioners to rejuvenate construction demand when a country approaches an urban service economy.Construction industry, advanced economies, recession, recovery, strategy,

No abstract

Acknowledging the importance of the private construction market and a close linkage between private construction investment, public sector output and general economic conditions, there is a strong motivation to develop reliable models to forecast private construction investment. Based on the Hong Kong scenario, two modelling approaches, namely the vector error correction (VEC) and the multiple regression models are developed and compared for their modelling accuracy and ability to handle non‐stationary time series data. The result suggests that private construction investment in Hong Kong can be predicted by reference to public investment in construction, gross domestic product (GDP) and unemployment rate. All in all, the VEC model is considered more accurate and robust in handling non‐stationary data. Through the VEC model, it is possible to confirm that the crowding‐in effect of public work programmes, though minimal, is discernible in private construction investment in Hong Kong. Yet private construction investment is more sensitive to general economic conditions, as represented by GDP and unemployment rate. The GDP could represent the ability of investors to pay for construction items, while the unemployment rate is used as a proxy for the willingness of end‐users to purchase the construction items. The models proposed should help policy and decision makers formulate suitable policies and strategies to sustain the construction industry in the medium to long run.Private construction investment, vector error correction model, regression analysis, stationarity, crowding‐in effect,