STATISTICAL COMPARISONS. 473 



alcoholic liquors again, I have never seen any statistics satis- 

 factorily connecting a relatively large consumption of alcoholic 

 liquors with drunkenness. On the contrary the consumption 

 in every community is probably at all times much more 

 largely the consumption of sober people than that of people 

 who drink to excess, and you may have much drunkenness 

 among a people who, like the Americans, are generally total 

 abstainers, and little among a people like the populations of 

 the Southern States of Europe, who are generally moderate 

 drinkei's. Thus the question of drunkenness, or the reverse, in 

 a population is not to be easily treated by statistics. 



The statistics of bankruptcy or insolvency again are often 

 quoted as a test of the comparative excellence of commercial 

 communities. Here again I have had in my mind some recent 

 comparisons at home between certain of the Australian colonies 

 and England as regards insolvency. These colonies, we have 

 been told, have twice as many failures per head of popula- 

 tion as England, or some such proportion. But the traps in 

 dealing with bankruptcy statistics are innumerable. Even in 

 England it is not easy to compare one pei-iod with another, 

 owing to difference of legislation making the conditions and 

 record of official insolvency different at one time from what 

 they ai'e at another. The law at one time makes whitewashing 

 so easy that debtors readily avail themselves of the courts to 

 make themselves officially insolvent, and so you have a large 

 number of bankruptcies in the official statistics. At another 

 time the law is so stringent that debtors evade the courts, while 

 creditors do not make them bankrupt because it is not worth 

 while to do so, and so the official bankruptcies diminish. At 

 one time also non-traders may be made bankrupt, at another 

 time they may not be ; and so the record varies. Unless, there- 

 fore, the whole basis of the bankruptcy law in each case is 

 studied, no comparison is possible either between period and 

 period in the same country or between different countries. 

 Further difficulties would arise in any comparison owing to the 

 length of the commercial cycle, which renders it most danger- 

 ous to take the figures of one year only or even of two 

 or three years for comparison. We can imagine, 

 then, what wild work is made by amateurs when 

 they compare the insolvency of Australia and England. 

 Apart from these diffiirences there are others which are due to 

 fundamental differences of economic condition. I believe, for 

 instance, that in England a larger proportion of the business 

 done is carried on by J oint Stock Companies than is the case 

 in Australia. This may or may not be the case. But, supposing 

 it to be the case, how can the failures of England be compared 

 at all with those of Australia, without taking account of the 

 liquidations of Joint Stock Companies, and to how many units 

 of individual failures is that of a Joint Stock Company to be 

 considered equal ? I would not go so far as to say that no useful 

 comparison could be drawn from existing data by those who go 

 carefully into the subject and study all the conditions. What 

 I am contending for is, that it is utterly impossible for writers 

 in a hurry to make anything of the first figures that come to 



