The Location of Industry

Location theory must be one of economic theory’s least applied branches. The theoretical literature on industrial location is extensive, the empirical evidence sparse. There are many reasons for the gap between theory and evidence. It has proved difficult to measure the impact of location on costs and returns since data are not available on enterprises with similar production functions, but different locations. The responses of industrialists to questions about location decisions give unreliable results since sample surveys of industrialists are inevitably biased by the exclusion of those who considered but rejected a particular location or who failed because they made the wrong locational decisions. Surveys also suffer from rationalizations and ex-post explanations. Even if a location decision was based on a rumor or accident, an industrialist may well develop a more logical explanation afterwards. Surveys also obtain a picture of the average reason for investment over time and it is difficult to determine the predominant reason at any particular time. Without a time profile of the reasons for decisions, one cannot link them to causal factors which undoubtedly changed over time. Dynamic and accidental factors are especially important in location decisions and are difficult to analyze. For instance, the first factory in an industry may be located because an empty building exists. Later investors in the new industry may locate near the pioneer just because they draw assurance from his success. Both the initial accident and the later nonrational elements are unlikely to be uncovered by a survey or other techniques.