Discrete Choice Modeling of Freight Outsourcing Decisions of Canadian Manufacturers

Abstract

Behavioral freight transportation modeling is an approach to enhance the quality of freight and logistics policy assessments. Outsourcing of freight activities is one of the essential decisions that firms make. Such decisions are influenced by factors such as economic conditions, competition, the industry, firm strategies, and firm characteristics. In this paper, the authors introduce a set of discrete choice models—binary and multinomial logit—that quantify the effect of some of these factors on outsourcing of the freight-related activities of goods production and logistics for Canadian manufacturers. The models were estimated with the use of data from the Survey of Innovation and Business Strategy obtained from Statistics Canada. The models consider firm characteristics (e.g., employment, number of products or services, and supplier locations), economic conditions, and local and international competition. The models also show the influence of the use of innovation and advanced technologies and measure the impact of government support programs for businesses (e.g., government grants) on outsourcing decisions. Models that explain international freight outsourcing are also presented. Model results highlight that freight outsourcing culture differs from one industry to another. Firms that use government training programs are more likely to outsource freight operations locally, whereas those that use government grants outsource their freight activities internationally. Model validation indicated acceptable predictive capabilities. Simplified models that were based on industry classification, location, and firm size were estimated and are to be used for future microsimulation purposes.