288 THE RELIEF OF THE UNEMPLOYED 



printing, wood-working, building, and other trades, 

 make weekly payments of various amounts to their 

 unemployed members. Consequently they are bound 

 to keep a strictly accurate record of the persons 

 entitled to such benefit, and many of them are also 

 able to show how many members are still unemployed 

 after having exhausted their claim to unemployed 

 benefit. It is true that the members of the Trade 

 Unions which keep these records include but a com- 

 paratively small minority of the total industrial 

 population, but it forms as a whole a sufficiently 

 representative sample of that population to justify the 

 conclusion that changes in the state of employment 

 for the workpeople included reflect corresponding 

 changes in the state of employment as a whole. There 

 are two considerations which lessen the value of these 

 returns, but as they would modify the conclusions in 

 exactly opposite directions, there is ground for the 

 hope that the two errors neutralize one another. The 

 first consideration is that unskilled casual labour is 

 insufficiently represented in the returns, and unskilled 

 casual labour suffers most from irregularity of em- 

 ployment. The second consideration to be borne in 

 mind is that the unions which insure their members 

 against want of work are the unions in the unstable 

 trades. With these qualifications the Board of Trade 

 puts forward these statistics as trustworthy. The 

 figures given in the annexed table do not include the 

 sick and superannuated, who are not counted as un- 

 employed. From the membership on which percentage 

 is based the superannuated are excluded, but not the 

 sick, on the ground that the latter are only temporarily 

 disabled. In the annexed table I have given in the 

 first four columns the figures for the four trades 

 which in the official returns are classified as one 

 group, in order to show the extent of the fluctuations 

 in the precarious trades. The succeeding four 

 columns represent groups of trades. The last column 



