Author: Irwin Collier

The following plea for more funding of economic research outside of official government agencies in general, but at universities like Harvard in particular, was written by public finance professor Charles Jesse Bullock and published in the Harvard alumni magazine in 1915. It left a deep enough impression to get mentioned at a meeting of the economics department’s visiting committee with faculty nearly thirty years later. 

I was somewhat surprised that after the long wind-up about the importance of large-scale research in the social sciences (especially in economics) for nothing less than “the future of civilization,” the essay ends up being little more than a pitch for a couple of paid research assistantships for the department. Still Bullock’s obiter dictum to the effect that they who pay the piper can choose the tune will be familiar to those living in our present age of partisan think-tanks and policy research institutions.

__________________________

THE NEED OF ENDOWMENT FOR ECONOMIC RESEARCH.

By Charles J. Bullock

The Harvard Graduates’ Magazine, vol. 23, June 1915, pp. 601-610

                  This is a time of economic unrest, and therefore of economic inquiry. Existing conditions are the object of incessant criticism, the fundamentals of the present order are often called in question, and nothing seems exempt from discussion, criticism, assault. Whereas a generation ago a mere handful of books and a few magazine articles represented the annual output of the United States in economics and sociology, today the output rises to approximately 1000 volumes, which is nearly one twelfth of the total book crop of the year; while the magazines and newspapers flood the market with their articles by the thousands and tens of thousands, to the despair of the cataloguer and expert indexer.

                  Recent conditions may have been unusual; perhaps another decade will see a change in popular interest. But it is not to be doubted that economic problems will continue to absorb their share of attention, and that economic inquiry will continue on a larger scale than was ever known before the 20th century. Equally clear is it that the importance of such inquiry cannot be gainsaid. If the 19th century was the century of the natural sciences, it cannot be doubted that the 20th, whatever else it may be, will be a century of social and economic inquiry. Modern life will doubtless grow more complex rather than less, more delicate and difficult economic adjustments will doubtless be necessary, projects for the reform and perfection of mankind will not become less numerous, and there will be great need of scientific investigation in economics and the other social sciences. Upon the success or failure of such inquiry, indeed, may depend in no small measure the future of western civilization.

                  To meet the need of the times, our existing equipment for scientific economic research is inadequate. For serious investigation in this field two agencies, and only two, are now available. On the one hand, we have the individual investigator working with such private means as are at his command, and in such leisure as he can snatch from his regular vocation. On the other hand, we have governmental agencies like statistical bureaus, commissions of inquiry, and certain administrative departments having to do with such matters as taxation, railroads, corporations, labor, commerce, agriculture, and the like. These yearly become more numerous, and perhaps more influential; and they supply materials of the greatest value to the private investigator. Undoubtedly, the economist of today commands a far larger mass of data than his predecessors.

                  But the greater part of this material is in very raw state, some of it is untrustworthy, and most of it requires careful verification, analysis and interpretation before it is fit for scientific use. Therefore the resources of the individual investigator are as inadequate as ever; indeed, not the least of his troubles is the enormous mass of material, — valuable, doubtful, or worthless, — which must receive patient and critical examination at his hands. On the whole, he is hardly better off than the economist of the last generation, and there can be no doubt that the progress of scientific economic investigation is greatly hampered at every turn by the lack of such provision as has been made in generous measure for the study of the physical and natural sciences.

                  In the latter field it was long ago learned that the resources of the individual investigator, even when he coöperated with his fellow scientists, were inadequate for the work at hand; and it is today a matter of comparative ease to secure generous endowments for scientific research in physics, chemistry, biology, psychology, and medicine. But for economic science similar endowments are almost entirely lacking, and seem hardly to be regarded as necessary. For the most part the economist is expected to make bricks without straw, or at least with such few wisps as he can supply from his private resources, which are seldom large; and yet economic research, when conducted properly, is as expensive as research in any other field, and more expensive than in most others. The collection of the primary materials is often wholly beyond the ability of an individual investigator under present conditions, and must be entrusted to governmental agencies, which alone can gather comprehensive data concerning population, resources, production, commerce, labor, finance, and many other subjects. But such data frequently need to be supplemented by private inquiry, and they always need most searching and painstaking criticism; so that governmental agencies leave much to be done even in the collection of trustworthy primary materials. Then after the data are at hand must begin the process of analysis and interpretation, which is difficult and time consuming. Here, as in all other fruitful scientific inquiry, economic investigation is always reaching into new domains; and in any given domain must probe more and more deeply, and make its analyses increasingly minute. In all these respects the task of the economist is as difficult and exacting as that of his colleagues in any other branch of science. His province is vast, and a field for endless labor opens before him.

                  In some particulars, indeed, the task of the economist is even more difficult than that of the student of physical or natural science. The elements in any economic problem, the materials with which the science deals, are exceedingly mutable, and frequently change even while the economist is analyzing and classifying them. Work done by the mathematician, if well done, abides forever. The chemist or physicist may make his determinations so accurate that they will remain the closest approximations to the truth; and the biologist, even though he knows that species are not immutable, can safely assume that his beasts and plants are not going to change before his investigation is completed. But the economist’s phenomena are in the highest degree mutable. Some things, indeed, may not change. The law of diminishing returns is not likely to be modified in the near future even by act of Congress; nor does human nature, however modifiable by environment, change over night or even reconstitute itself within a year. But such things as laws and institutions, methods of production, available natural resources, the numbers and distribution of population, are in constant state of flux; and many an economist who lightheartedly begins a study of current problems presently finds himself writing a treatise on ancient history. Indeed, the economist’s task is never done. His materials must ever be collected anew, and his work must ever be repeated; the economic order changes, and the living specimens of today become in a few years the fossil remains of a bygone age. It will be noticed that I am speaking not of changes in theories about given economic phenomena, but of mutations of the phenomena themselves. In every field of science theories change, but in no field do the phenomena themselves change so generally and rapidly as in the social sciences.

                  A further difficulty is that the materials with which the economist deals are peculiarly liable to perversion, distortion, and even deliberate falsification. This fact enormously increases the investigator’s difficulties, and greatly adds to his labor. For this reason alone, the resources of the private investigator would surely be inadequate; and when to this is added the mass and complexity of the materials and their extraordinary mutability, the need of greater facilities than the individual economist can command is too apparent to require further comment.

                  One conceivable solution of the difficulty is to turn all large undertakings over to the State. Already the United States government is spending large sums for research, and the total cost of such work must amount to several million dollars annually. The Department of Agriculture, the Department of Commerce, the Bureau of Labor, the Census Bureau, the Bureau of Corporations, and the Interstate Commerce Commission have done, and are doing, work of the greatest importance to the economist, much of which, especially in the field of statistics, would be absolutely beyond the capacity of any individual investigator or private organization. In a similar manner various states and an occasional city are carrying on work of great importance, usefully supplementing the scientific work of the Federal government. Why not, then, depend upon these public agencies for such economic research as lies beyond the power of the individual investigator?

                  I have not the least desire to disparage governmental research; on the contrary, as indicated in the previous paragraph, I believe it to be highly useful, and in some fields indispensable. I believe also that the last 10 or 15 years have seen a distinct improvement in the quality of the work done in the United States, although such improvement has not everywhere kept pace with the increase of output. But after giving the most generous recognition to what the State is doing for the promotion of economic research, we must recognize that it would be highly unfortunate, and even dangerous, to permit the State to monopolize all economic inquiries that lie beyond the power of the individual investigator.

                  For, in the first place, even scientific research, when turned over to a governmental agency, is brought directly within the domain of politics. I do not mean, of course, that all of our departments or bureaus carrying on scientific work are headed by practical politicians and manned by political workers. This sort of thing, as we all know, is becoming less common; and there are not a few cases in which it is possible to say that politics, in this sense of the word, has been largely, and for considerable periods even wholly, excluded. To be sure, even a president of Mr. Wilson’s antecedents has been guilty of placing in charge of an important bureau, the only work of which is of a scientific character, a man whose principal qualification evidently was that he had been chairman of the party committee of a certain state. But such occurrences are becoming less frequent, and we may fairly anticipate continued improvement in the matter of treating scientific positions as mere political spoils.

                  But even with the grosser forms of political influence eliminated, it is true, and must remain true, that political considerations or purposes can never be wholly eliminated from governmental research. Even such an apparently non-political bureau as the Geological Survey may become the storm centre of the conservation movement if official determination has to be made of the apparently simple question of the effect of forest destruction upon soil erosion, and the Weather Bureau may become surcharged with political lightning if a loquacious chief expresses uncalled for opinions concerning the influence of forests upon rainfall and the flow of rivers. Even chemical and physiological inquiries take on a political tinge if they relate to the use of benzoate of soda, the wholesomeness of oleomargarine, and the products of the Chicago stock-yards. In fact, a clever politician can extract a surprising amount of political capital from such scientific inquiries as these, and a scientific investigator may risk his official head if his inquiries lead to an unwelcome conclusion. Some years ago a physiological chemist who was so unfortunate as to determine that good oleomargarine is a perfectly healthful article of food was told that his institution need expect no further support from the State if its professors were to antagonize the farmers in this manner. Equally hard might be the lot of any other investigator whose scientific determinations in this, or any allied field, should prove unpalatable to the conservationist, the pure-food crusader, the farmer, the social reformer, or the big corporation that produced the articles subjected to scientific analysis.

                  What happens to such peaceful and apparently non-political sciences as chemistry and physiology when they come into contact with politics, is much more certain to happen to a science like economics, which from the very nature of the case must deal with questions that are political in character. Even if we grant that it is possible to eliminate absolutely the spoils system, it would still remain true that economic research under Republican auspices would necessarily be a somewhat different thing from economic research under Democratic guidance, or under the control of a Progressive, Socialist, or Prohibition administration. Messrs. Redfield and Davies, for instance, inevitably give a different tone to economic inquiries under their charge from that imparted by Messrs. Cortelyou and Smith. This is not by remotest implication a reflection upon the honesty or fairness of any of these gentlemen, but it is merely a statement of a condition that inevitably results from the personal equation and the political creed. Nor is it a reflection upon governmental research as such, for such work may be highly useful in spite of the allowance that has to be made for the personal or political equation. I maintain simply that we must not blink the patent fact that governmental research can never wholly lose a political character. Such research may be highly useful, and, in fact, is becoming increasingly necessary; but we should not on that ground indulge in any illusions concerning it. “Official statistics,” the “impartial findings” of a Federal commission, the “final and authoritative” determinations of a government bureau, are indeed entitled to respectful reception and careful consideration; but they do not give us necessarily the last word upon any subject.

                  I have spoken so far only of the inevitable defects that arise from personal or political bias, such as is bound to exist among the best of men, and is least harmful when frankly admitted. But beyond this, there is the possibility of deliberate perversion of governmental investigation for partisan purposes. Some branches of Census work have suffered seriously from this cause, particularly the statistics that used to be published concerning the average wages paid in manufacturing industries. The most notorious case occurred in 1892, when, by manipulating the divisor used in computing average wages, the Census was able to announce that the average remuneration had risen from $347 in 1880 to $445 in 1890. On the eve of the presidential election the Census issued a series of bulletins relating to wages paid in the leading cities of the country, and exploiting in the most conspicuous manner possible the increase alleged to have occurred during the decade ending in 1890. These bulletins purported to show that wages had increased nearly 53% in New York, 35% in Chicago, 45% in Boston, 52% in Philadelphia, 73% in Atlanta, 77% in Richmond, 77% in Syracuse, and so on through the list. It seemed as if the campaign committee had mobilized its forces at the Census Office, and was directing a hail of deadly statistical shrapnel at the enemy’s trenches. This may have been good politics, but it certainly was not good science; and even from the political point of view, it led to awkward consequences. The average wages for 1890 were placed at such a high figure that it was a foregone conclusion that, without deliberate falsification of the data, the statistics of 1900 could not exhibit a further increase. As a matter of fact, they showed a decrease, computed by the old method, from $445 to $438, which was perhaps a fortunate result in that it demonstrated the dangers of political wage statistics. It is gratifying to be able to add that there has been no time since the Census was made a permanent bureau when such a performance as that of 1892 would have been conceivable.

                  Another celebrated feat of official statistics was the so-called Aldrich Report of 1893, which purported to give, among other things, statistics showing the general course of wages in the United States from 1840 to 1891. These statistics were immediately accepted as “official,” and incorporated in the economic literature of this and other countries; but it later developed that they had been gathered and handled by methods that would not bear the slightest careful criticism, and that some of the things done by the makers of the Report were so preposterous as to bring in question the investigators’ honesty of purpose. In one establishment, a brewery, the investigators found a brewer whose wages had increased from $6.39 a day in 1860 to $23.96 in 1891, or something like 285%; and they adopted a method of averaging which made the wages of this typical proletarian count for as much in determining the general result as those of 133 common laborers found in another industry whose wages had increased only 29% during the period of 31 years.

                  These examples show what official investigators can do even with such comparatively simple and definite things as statistics. When it comes to inquiries into complicated industrial conditions and the investigation of large questions of public policy, the opportunity for deliberate bias is greatly increased. Some 14 years ago, we had a Federal Industrial Commission which investigated almost every conceivable subject except white slavery and the recall of judges, but was particularly concerned with the trust problem and the protective tariff. The final report of the commission, in some 19 formidable volumes, has been widely used by both American and European investigators as a repository of economic information. Yet it was perfectly evident to the discerning at the time, and today would probably not be questioned by anybody, that, so far as the trusts and the tariff were concerned, the work of the commission was fundamentally partisan and political, and that its report contains fully as much misinformation as information. Certainly an economist with a professional reputation to maintain would today be chary of citing the findings of this commission as high authority upon either the trust or the tariff problem.

                  At the present writing, we have with us another industrial commission appointed a year or two ago as a result of the recent social unrest. In the closing months of his administration, President Taft named a commission, but his nominations aroused violent protest on account of the alleged conservative views of the nominees; and they were not confirmed by the Senate. President Wilson a few months later named another commission, against which the charge of conservatism can hardly lie; and this body is now making an official investigation of social conditions. On the eve of an important investigation the chairman, in a public address, denounces roundly the institutions he is about to investigate. Some months before the commission’s inquiries are concluded he announces that the country can never prosper “as long as the banks handle the wealth of the nation purely to make it pay the largest dividends,” and makes the “definite” suggestion that “autocracy in business” must go. When such performances arouse discussion and criticism, the grand inquisitor then announces that his “position is not a judicial one,” and that “judicial poise” is “a great bar to human progress.” Yet two or three years from now we shall be asked to accept the findings of this commission as “official.”

                  Cases are not wanting where investigations that yielded inconvenient results have been wholly suppressed. This happened, for instance, with an investigation of the sugar beet industry in the United States, which was made for one of the departments of the Federal government only a few years ago. And other similar, but less well authenticated, cases will doubtless occur to persons familiar with Washington affairs. Actual suppression, however, is probably a comparatively rare thing. What usually happens is that the administration in charge of national or state affairs is committed to certain policies, and that the expert investigators of such an administration are unlikely to reach inconvenient results. This is true not only of the political policies of national or state administrations, but also of the general policies of public departments in matters that are not immediately of political moment. A scientific student who turns to the reports of any public department, whether it has to do with taxation, banking, railroad administration, labor, or any other economic interest, must always be careful to make due allowance for the settled policies of the department. This is not a reflection upon the integrity of administrative departments, but is a necessary allowance for the personal equation which enters into all human affairs, public and private.

                  A final difficulty with the scientific work of governments is that it is generally confined to what are considered practical ends, by which is usually meant undertakings that give promise of immediate practical results. This is seen in appropriations for state universities which readily obtain money for agricultural, engineering, and other practical subjects, but have difficulty in securing meagre allowances for pure science, philosophy, and the humanities. It is evidenced also by the large appropriations the Federal Government makes for agricultural research, labor, and similar practical interests. In time, conditions may change, but for the present there is slight prospect of securing public support for research outside of economic questions of immediate practical concern.

                  Useful and even indispensable as it may be, therefore, governmental research in the field of economics needs to be supplemented by adequate private agencies. We need to place beside the Census Office, the Bureau of Labor, the Interstate Commerce Commission, the Bureau of Corporations, and the other excellent boards and bureaus, both state and national, now engaged in economic research, a number of private agencies that shall be free from political stress and disturbance, relieved from the necessity of confining themselves to investigations of immediate practical value, and amply equipped for the most thorough, painstaking, and accurate research in both pure and applied economics. Since scientific work of such a character cannot possibly be remunerative in the pecuniary sense, it is evident that such agencies can be provided only by endowments.

                  It also seems clear that a university devoted to scientific studies and dedicated to the pursuit of truth is a most fit institution to receive such endowments. Here the investigator will not be obliged to confine himself to inquiries that promise immediate practical results. Here he may be free from political or other pressure, and may benefit from association with scientists engaged in other fields of work, especially the older and more exact sciences.

                  The work that might be accomplished by such endowments can hardly be overestimated. Never yet in the history of the science has the economist been given the resources and equipment really necessary for his work. To fashion bricks without straw were a light and attractive task compared with his. If Harvard University could receive during the next few years an endowment adequate to make even a respectable beginning of organized research, it might within a generation do more than any private agency has ever done to advance the frontiers of economic science.

                  Such a tremendous vista of useful investigations would open before a department properly equipped for economic research that a very large endowment is thoroughly justified and even urgently needed. This may not be the time for undertaking large enterprises that call for money, but it is possible to begin the work in a modest way by the endowment of one or more research assistantships, which would permit the Department to prospect the field. Such endowment would enable the University to provide a professor with competent assistants like those provided for investigators in other fields. The sum of $30,000 would endow such assistantships and provide for the incidental expenses that always arise in connection with scientific work. They would certainly justify themselves by their results, and further endowments would then be easier to secure.

                  Another excellent plan would be the provision of funds for the investigation of particular subjects. There is great need, for instance, of searching investigation of the recent increase of public expenditures in the United States, an undertaking that would certainly prove fruitful in both theoretical and practical results. Even greater is the need of a searching investigation of the present world-wide increase of prices, which, like similar price movements of former times, is producing economic disturbances of vital practical moment and the greatest theoretical interest. Then there are the troublesome problems of the day, — socialism, single tax, labor legislation, the extension of public industries, public regulation of private industry, the tariff problem, the problem of large-scale production, and all the others, — that occasion so much discussion at the present time. We hear much about what other countries have done in this direction or that, but we have comparatively little first-hand investigation, impartial and absolutely scientific, of the actual results of such experiments. At every hand topics of fascinating scientific interest and great practical importance abound. Competent workers are not numerous, and their resources are painfully inadequate.

                  In a new undertaking of this character, the first step is usually the hardest. The endowment of economic research at Harvard University is a thing that can be finally and conclusively justified only by its results, and such results in turn are impossible without an endowment. The Department of Economics, however, believes that a strong case can be made out in favor of an experiment in this direction. It is now in quest of endowments for research assistantships and funds to defray the expense of particular investigations. I am grateful for the opportunity to bring this matter to the attention of the Harvard Graduates‘ Magazine.

Image Source: Portrait of Charles J. Bullock from the Harvard Class Album 1915. Colorized with image enhancement by Economics in the Rear-view Mirror.