An Uncertain Start1
The Institute of Statistics at the University of Oxford was established in 1935 with the appointment of the econometrician Jacob Marschak to a Lectureship in Statistics at All Souls College.
In 1933, the Rockefeller Foundation had given the University a grant of $350,000 to create such an institute. The money was to be spent over the next seven years in establishing statistics teaching and research at Oxford. From the beginning, it was believed that ‘it is undesirable to divorce the study of statistics and statistical method from the cognate study of Economic Theory and Organisation', and the appointment of Marschak as its first Director confirmed this orientation in the work of the Institute (Arena 2011: 110-111).Marschak was an unusual choice for this position. He had been born in Kiev in 1898 and had participated as a Left Menshevik in the October Revolution, rising to the position of a Minister in the Caucasian Republic of Terek. Marschak left Russia in 1919, and went to Berlin, where he studied under Ladislaus von Bortkiewicz, and then under Emil Lederer in Heidelberg. He went on to work with Adolph Lowe at the Kiel Institute for the World Economy
1 I wish to thank David Hendry and the Librarians of the Bodleian Library for assisting me with finding the papers of the Oxford Institute of Statistics. I am grateful to Robert Cord for directing me to other material on the history of the Institute. Any remaining errors are mine.
J. Toporowski (*)
Department of Economics, SOAS University of London, London, UK e-mail: jt29@soas.ac.uk
© The Author(s), under exclusive license to Springer Nature Switzerland AG 2021 147
R. A. Cord (ed.), The Palgrave Companion to Oxford Economics, https://doi.org/10.1007/978-3-030-58471-9_6
(Institut fur Weltwirtscha.fi). German economic theory was highly regarded at that time in Britain, and the Kiel Institute was an outstanding example of a new direction of research that had emerged in the economic and financial turbulence that followed the First World War, gathering economic data and interpreting it with a view to assisting business and government in decision-making and other strategies.
Similar institutions had been established in France, Poland and even Russia. The one in Vienna, the Institut fur Konjunkturforschung that had been established by Ludwig von Mises, was already world famous. A notable feature was that such institutes were not based in universities, but were supposed to be closer to business and government.Apart from their service in providing information to business and government, the new research institutes reflected another trend in economic theory towards mathematical formalisation and the systematic creation of data on changes in the economy, in particular the business cycle. At the League of Nations, a British economist Alexander Loveday headed an Economic Intelligence Service that published important data on the changing economic fortunes of member countries.
Britain lagged behind Europe in this kind of research. John Maynard Keynes wrote on commodities and stocks for the London and Cambridge Economic Service that brought together contributions from the London School of Economics and Cambridge University (see Cord 2017). But national income estimates were not collected until the Second World War. In part, this was the outcome of a system of higher education that was organised around teaching undergraduates, in particular with the tutorial system in the ancient universities of Oxford and Cambridge where college fellowships were the prize for postgraduate research, rather than the doctorates that had recently been introduced to bring British academic standards up to the level of German universities. When Michal Kalecki's Rockefeller Fellowship expired in 1937, his friends in Cambridge found only a handful of university departments where research in economics was being conducted (see Toporowski 2013: 111).[25]
The Oxford Institute of Statistics was supposed to remedy this, at least for the University of Oxford. However, it seems to have had difficulty in getting itself started. The possible explanations for this range from the personal to the institutional.
Among the latter was the alternative focus of research in economics at Oxford provided by Hubert Henderson at All Souls College. Henderson had assisted with the establishment of the Institute of Statistics. But he also had his own research agenda. He brought together a group of the younger, most research-ambitious economists at Oxford to enquire into prices and interest by means of interviews with businessmen. Henderson’s collaborators included Maurice Allen, Eric L. Hargreaves, Frank A Burchardt, Marian Bowley, P.W.S. Andrews, Arthur J. Brown, George Shackle, Roy Harrod, James Meade and others (see Besomi 1998). The relative size of their undertakings was reflected in the staff employed by the two research centres. Marschak wrote to Richard Kahn at Cambridge that at the Institute ‘only 1½ of its researchers are paid...the rest are voluntary workers...[whereas] two men work under the “Economists’ Research Group”’ (Marschak quoted in Toporowski 2013: 112).A possible personal factor in the slow start was Marschak’s own insecurity, or at least his desire to be elsewhere. Having reached Berlin to escape the Russian Civil War, and then England to escape the Nazi takeover in Germany, Marschak was unwilling to test the limits of Germany’s expansion, while in America even larger funds (from Rockefeller and the Cowles Commission) were available to beef up economics research with statistics. In December 1938, he left Oxford to go to the United States on a Rockefeller Travelling Scholarship. This left the Institute without a Director. Eventually, on 28 August 1939, two days before Germany’s attack on Poland, Marschak resigned his position. However, he then changed his mind and four weeks later, on 26 September, asked for a further year’s leave of absence. The Standing Committee of the Institute decided that no further leave would be granted, whereupon Marschak resigned for a second time (see Papers of the Oxford Institute of Statistics (hereafter OIS), University Archives, Bodleian Library: UR/SI/1, File 1).
2