64 



National Resources Convmittee 



in one ctxKe ami "pride" in the other concerning the 

 research work undertaken and the expenditures for 

 such purposes. 



Transfer of Research Investigations 

 to Specialized Agencies 



Some Federal agencies are especially qualified by 

 equipment, personnel, and experience to undertake 

 research for other organizations. Hence they conduct 

 investigations on a reimbursement basis at the request 

 of other Government agencies. The outstanding ex- 

 amples, but by no means the only ones, are the National 

 Bureau of Standards, which caiTied on some research 

 for more than 14 other agencies in the fiscal year 1937 ; 

 and the Bureau of the Census, which received working 

 funds from 7 other agencies in addition to reimburse- 

 ments for work not formally provided for in advance. 

 Transfers of one undertaking may even be multiple, as 

 the following heading in the Budget illustrates: 

 "Working Fund, Commerce, Coast and Geodetic Survey 

 (Public Works Administration, Allotment to Interior, 

 Soil Erosion Prevention, Transfer to Agriculture, 

 Transfer to Commerce, Coast and Geodetic Survey)." 



In the data given in the present report, such reim- 

 bursable expenditures are usually included under the 

 organization actually doing the work, except in the 

 case of an agency that specifically requested otherwise. 



Expenditures Not a Measure 

 of Value of Research 



There is, of course, no necessary relationship between 

 the size of an appropriation for any given research 

 project and the usefulness of the results obtained from 

 it. Such a relationship is sometimes assumed by pro- 

 ponents of a specific research undertaking. Aside 

 from the obvious qualifications concerning direction, 

 personnel, time factors, etc., the amount of research 

 per dollar of expenditure varies with the kind of tech- 

 niques used. For example, economic studies adequately 

 planned, reviewed by outsiders, and replanned through 

 an extended period of time by experienced research 

 people in long-established agencies, are apt to be car- 

 ried out much more efficiently than studies hurriedly 

 undertaken by new personnel in a recently established 

 governmental unit. "Wlien a factor extraneous to re- 

 seai'ch proper comes into the picture, as was the case of 

 finding work quickly for the white-collar unemployed, 

 the discrepancies between the amounts of expenditures 

 and the value of the results obtained may have been 

 greater than in the case of a project cai-ried out solely 

 for research purposes and by purely research methods. 



Unfortunately, it might also be said that, in spite of 

 the best efforts of the guardians of the Treasury, 

 amounts for some periodic research may not always 



terminate when the particular need has passed. If 

 the amounts have been appropriated annually for many 

 years, the presumption is usually that the same appro- 

 priation should be continued. 



There are at least two distinguishable values, and 

 in most cases three, emanating from any research activ- 

 ity. First, there is the value to the public, which may 

 sometimes be roughly related to the amount of expendi- 

 tures on the study, since more generous funds may 

 secure a larger sample of data to analyze and, hence, 

 there may be a greater likelihood that variations in 

 different localities will be adequately treated. How- 

 ever, even in this case the point of diminishing returns 

 is often rather quickly reached so that a small sample 

 may be almost as satisfactory as a somewhat larger 

 one obtained at considerably greater cost, although the 

 difficulty of determining the size of an effective yet 

 economical sample in advance of the project may be 

 insuperable. The value of a research project to the 

 general public may, of course, be quite different from 

 the value to a specific group for the immediate benefit 

 of whom the project was undertaken. 



A second value of research projects is the benefit to 

 the administrative efficiency of the organization under- 

 taking the research work. Since research activities 

 constitute the planning and critical functions of the 

 agency, the existence of research within an organiza- 

 tion not only energizes the agency as a whole, but also 

 the individuals who are in immediate contact with it, 

 provided they have enough perspective to see the sig- 

 nificance of the result obtained. For this reason, some 

 organizations decentralize their research activities as 

 far as feasible rather than concentrate them in 

 specific research sections or divisions, although there 

 are obvious advantages in such segregation. For our 

 present purposes, it is enough to note that the energiz- 

 ing influences of research within the agency bear no 

 necessary relationship to the funds available for carry- 

 ing on the investigatory activity. 



The third possible value of a research program lies 

 in the usefulness of results obtained to the Federal 

 Government itself, either directly to the administrative 

 process involved in the research or to policy formation. 

 It is here perhaps that the relationship between the 

 amount of expenditures and the results obtained shows 

 the least direct correlation. 



Previous Estimates of Federal 

 Expenditures for Research 



Estimates of research expenditures of the Federal 

 Government have been made several times in the past. 

 Some of these compilations have been confined to the 

 natural sciences and technology, and some have used a 

 concept even broader than the one employed in the pres- 

 ent study. These differences in definition make the re- 



