676 TRANSACTIONS OP SECTION F. 



present a real though by no means insuperable difficulty which will have to be 

 carefully borne in mind by those responsible for devising and working any 

 unemployment insurance scheme. 



The conclusion seems to be that the extent to which the risk of unemployment 

 due to industrial and local displacement is properly insurable depends partly on a 

 wise choice being made of the group of trades and of the geographical area to be 

 embraced by the scheme, partly on the judicious limitation of the benefits 

 payable thereunder. Our analysis points to the necessity of a large area, 

 both geographical and industrial, and further suggests that the groups of trades 

 included should be such as are unlikely as a whole to undergo wholesale and rapid 

 displacement, and within which any decay to be apprehended is likely to be only 

 local and partial and not on a scale too great to be compensated by the expansion 

 of other branches of trade within the insured group. 



There remain the risks due to personal causes. Of these we have already 

 ruled out the risks due to the wilful act of the workman, and to these we 

 must now add the personal risk attributable to exceptional deficiencies, physical, 

 mental, or moral. These are not properly trade risks, the burden of which ought 

 to fall in a special degree on those following a particular industry, and if they 

 were allowed to do so they would ruin any scheme of insurance based on the trade 

 group. There is still, however, one important class of personal risk to which all 

 are liable, rnd which is in the main beyond the control of the individual, viz., the 

 increasing liability to unemployment due to advancing years. I do not intend to 

 trench on the important but quite separate problems of national provision for old 

 age and invalidity as such. I am solely referring to the ascertained statistical 

 fact that the chance of unemployment is a function of age, and that beyond a 

 certain age the risk is materially increased. For example, among a body of nearly 

 right thousand engineers whose industrial records were analysed for the purpose, 1 

 found that whereas the average number of working days lost in the year by the 

 whole body was fifteen, that for members below the age of forty-five was less 

 than twelve, while for members between the ages of forty-five and fifty-five 

 it was twenty, and for members between fifty-five and sixty-five, thirty-three. 

 (Above sixty-five the figures are affected by superannuation.) The question 

 we have to ask is, how far this class of risk is insurable ? 



The answer depends again on the scope of the scheme. A voluntary scheme 

 which workmen are free to join and leave at their pleasure cannot deal satis- 

 factorily with a risk of this kind, especially as no scheme of graduating con- 

 tributions according to age is likely to be administratively feasible. Trade-unions 

 which give unemployment benefit are in an exceptional position, because they exist 

 primarily for trade protection purposes, and hence have a hold on their members 

 which no voluntary insurance scheme pure and simple could possess. Generally 

 speaking, personal unemployment due to advancing years is insurable, and only 

 insurable, under a scheme which applies compulsorily throughout the whole period 

 of the workman's industrial life. 



It results from our analysis that some of the risks of unemployment are 

 properly insurable and others are not, and the next step is to ascertain broadly 

 the relative importance of the insurable and non-insurable elements. Now an 

 examination of the available statistics indicates clearly that at all events as 

 regards certain large groups of trades in which unemployment is acute — ■ 

 namely, the building, engineering, and shipbuilding trades — the insurable 

 element in the risk of unemployment predominates largely over the non- 

 insurable element. 



The method of statistical proof of this proposition may be indicated as 

 follows : — 



1. The percentage of unemployment in these trades — taking an average of 

 good and bad years together — has not varied very widely during the period of 

 fifty years during which the statistics have been collected (the average for the 

 first decade of the period was 5'6; for the second, 4'5; for the third, 6'8 ; for 

 the fourth, 5'2 ; and for the fifth, 7'2. The average for the whole period was 

 5"9). As the period of oscillation is not exactly ten years, part even of the 

 differences shown above is accounted for by the presence of an excessive propor- 

 tion of good or bad years in particular decades. Thus we may fairly say that 

 the element of unemployment due to progressive expansion or contraction of the 

 demand for labour has been relatively small. 



