132 SECTIONAL ADDRESSES 



is a better indication than an unweighted one, for a high or low location 

 factor in a large area represents a greater degree of concentration or 

 otherwise than the same factor in a small area. Weighting is particularly 

 necessary where the official region is used as a basis since these regions 

 range from Greater London with 21 -8 per cent, of the occupied population 

 of England and Wales to North Wales with i • 6 per cent. If the weight 

 used be these percentages of occupied population, a short-cut device can 

 be used avoiding the calculation of the regional location factors. This 

 is not the place for the demonstration of this device ; but a warning must 

 be issued that, apart from methods of calculation, the coefficient is liable 

 to be less significant the larger the areas used.' 



The coefficient of localisation can vary from o-oo for no localisation 

 to 2 -00 for extreme localisation. When the official regional areas are 

 used as a basis, tin-plate shows the highest coefficient with i -81. Lace, 

 I -68 ; cotton weaving i -49 ; cotton carding and spinning i -45 ; hosiery 

 1-45 ; cutlery 1-45 ; and pottery 1-44 are not far behind. At the 

 other extreme are local government 0-05 ; dealing 0-12 ; mineral waters 

 0-13 ; bread and flour 0-15 ; and beer-breweries 0-22 — ^all well dis- 

 tributed. 



Industries in the middle ranges of the coefficient which offer the best 

 hope of large development in the depressed areas include artificial silk 

 spinning (coefficient 1-15), cocoa and chocolate (coefficient o-86), mining 

 machinery (coefficient 0-82), toys, games and sports requirements (coeffi- 

 cient 0-74), rubber (coefficient 0-70), biscuits (coefficient 0-64), electrical 

 apparatus (coefficient o • 62), sugar confectionery (coefficient 0-53).^ The 

 mere possession of a middling coefficient is of course not enough to cast 

 the industry for development in a depressed area. Research must take 

 account of the special circumstances of the area and the industry and must 

 work on some theory of the reasons for location checked by statistical 

 measures and itself checking those measures. Where the fuel, materials 

 and markets of a particular industry are concentrated at distant places, 

 for instance, theories such as Weber's as to their relative force in deter- 

 mining the most efficient site must be tested and checked by reference to 

 the statistical factors and coefficients. If, as seems to be the case in 

 England to-day, the forces that determined the present localisations are 

 changing and the market is becoming more important at least than fuel, 

 this should be shown by a halt in the progress of localisation and of the 

 more localised industries. Areas seem to be depressed to-day largely 

 because their industries were too self-centred and too far from the centre 

 of the country's population. Delocalisation may soon be taking place 

 in the sense that there will be a smaller proportion of men employed in 

 an industry where that particular industry used to be localised. Realistic 

 research in siting problems has not yet however proceeded much beyond 



' Mr. Wensley has worked out a ' truer ' coefficient of localisation for a sample 

 of industries based on a division of the country into very small localities. The 

 differences in the case of the same industry between the truer localisation coeffi- 

 cients and the coefficient based on the official twelve regions of England and 

 Wales did not, in practice, however, turn out to be great. 



8 See The Second Industrial Survey of South Wales, Pt. Ill, chap. 3. for list of 

 Suggested New Industries. 



