202 



National Resources Committee 



Bureau represents inquiries on subjects which had been 

 covered only at the decennial census prior to the per- 

 manent establisliment of the Bureau. The fact that 

 within a given subject-matter field the Bureau is, in 

 many cases, just one of several agencies collecting data 

 is explained, of course, by the rapid growth of statis- 

 tical agencies attached directly to the administrative 

 arms of the Government. 



An itemized chronological account of the acquisi- 

 tion of the Bureau's reporting functions has been pre- 

 ])ared for the writer by Dr. Joseph A. Hill, chief of 

 the Bureau's Division of Kesearch. It appears as Ap- 

 pendix A of this report. Reference to this appendix 

 will show that upon the establishment of the perma- 

 nent Bureau, in 1902, the following subjects formerly 

 covered in the decennial census were detached to spread 

 the woi'k load of the new Bureau or to permit more 

 frequent reporting : 



Mortality. 



Valuation, taxation, and public indebtedness. 



Religious bodies, or churches. 



Social statistics of cities. 



Transportation by water. 



Defective, dependent, and delinquent classes. 



In addition to shifting numerous inquiries from the 

 regular decennial census, the act of 1902 also provided 

 for an additional mid-decennial Census of Manufac- 

 tures (made biennial in 1919), for a quinquennial 

 Census of Electrical Industries, for current reporting 

 of cotton production, and for statistics of births as 

 well as deaths. The first mid-decennial Census of 

 Agriculture was taken in 1925. The only important 

 new fields entered by the Bureau since 1902 are repre- 

 sented by the Census of Distribution (wholesale and 

 retail trade, service establishments, etc.) and the Cen- 

 sus of Unemployment — both authorized as part of the 

 1930 decennial census — and the current industrial re- 

 porting which developed mainly in the decade 1920-30. 

 Many present inquiries, however, have a form quite 

 different from that contemplated in the beginning. An 

 example is the current reports on judicial criminal 

 statistics, which were authorized by special legislation 

 in 1931. The principal intercensal publications of the 

 Bureau are summarized in Appendix A to this report. 

 Thus the Bureau of the Census took on the form 

 of a huge statistical factory. It developed efficient 

 methods of mass production and kept a heavy flow 

 of work passing through its mechanical tabulation ma- 

 chinery. In its mechanical laboratory some of the 

 machinery was invented and mechanical improvements 

 were constantly added. The mechanical equipment 

 also was at the service of other agencies in the Gov- 

 ernment for special tabulations, on much the same basis 

 as in the Government Printing Office for printing. 



Statistical Work of Other Agencies 



Yet, the Bureau, with this impressive body of inter- 

 censal work, its priority as the statistical agent in 

 almost every subject-matter field, its experienced staff, 

 and its unrivaled mechanical equipment was not, as 

 we have seen, to share extensively in the rapid growth 

 of the Federal statistical services. In 1935, the Com- 

 mittee on Government Statistics counted the following 

 nmnber of separate Federal agencies, in addition to 

 the Bureau of the Census, which collected or compiled 

 statistics in selected fields, most of these agencies hav- 

 ing begun their activity within the last 25 years : ^' 



Production 18 Finance 25 



Distribution , 19 Population 4 



Labor 19 



Out of more than 2,000 economists, statisticians, and 

 political scientists in the professional and scientific 

 classifications of the civil service, in Washington, at 

 the beginning of 1938, the Bureau of the Census had 

 only 42. 



Reasons for Bureau's Failure to Become 

 Central Statistical Agency 



Why did not the Bureau acquire more of the fimc- 

 tions of current reporting and research which instead 

 were developed in other agencies? The answer to this 

 question would require much more extensive study 

 than has been possible for the preparation of this 

 rejiort. On the basis of present information, one can 

 only speculate, and a number of guesses are here of- 

 fered without supporting evidence : 



1. The statistical data sometimes flowed into an agency iu 

 the direct course of law enforcement. Example. — The income 

 tax returns coming to the Treasury. 



2. When the data did not come iu automatically, a regu- 

 latory agency often could use for the collection and analysis 

 of data .some of the funds appropriated to it for administration 

 of particular legislation, even if funds were not specifically 

 earmarked for research. 



3. Such an agency might have wide popular support or 

 benefit by the activity of pressure groups on its behalf. Thus 

 bureaus in the Department of Agriculture were in a strategic 

 position for securing appropriations, while a general-purpose 

 statistical agency like the Bureau of the Census, lacking a 

 popular dramatic appeal or an opportunity to appear indis- 

 pensable to particular special interest groups, was In a weak 

 strategic position for getting appropriations. 



4. Some agencies developed large and expert field staffs, 

 which could be deployed quickly for the collection of iuforma- 

 tlon, even though the primary function of the field staff was 

 that of inspection or other regulation. 



5. Most agencies probably wanted to grow and hence were 

 reluctaut to "farm out" their work to another. Even If the 

 Bureau of the Census had been ideally equipped to supply 

 the needed data, the normal tendency of many agencies, unless 

 checked, doubtless would have been to augment tlieir own 



^Report, Appendix A. 



