F. A. Hayek vs. J. M. Keynes in Shackle's marginal gloss^*

* This research was conducted in the Institute of Management Studies, Goldsmiths, University of London and St. Edmund's College, University of Cambridge.

ABSTRACT

The intellectual rivalry of F. A. Hayek and J. M. Keynes has recently caught the attention of historians of economic thought, journalists and the broad public. However, how was it viewed at the time? This article uses archival material in the form of marginal annotations made by G. L. S. Shackle to determine contemporary reading responses to the theoretical developments of the 1930s. Shackle's unique reading style that includes legible, dated, annotations and the fact that a substantial part of his academic library survives gives us a unique vantage point from which to explore anew this period of intellectual history.

KEYWORDS:

• Hayek
• Keynes
• Shackle
• Common Reader
• marginal gloss

JEL CODES:

• B00
• B20
• B31

Acknowledgments

This work was supported by the G. L. S. Shackle fund and St. Edmund's College, where I was Shackle scholar from January to April 2015. I would also like to thank Geoff Harcourt, Tiago Mata, Nuno Martins, Roger Backhouse, Roberto Scazzieri, Ivano Cardinale and three anonymous referees, for useful discussion comments and feedback. Furthermore, I am grateful to participants of the relevant session in the ESHET 2015 in Rome and in the CAMHIST special seminar at Clare Hall for their feedback. Also, I would like to thank the staff of the following institutions for their support and help in locating material by G. L. S. Shackle: Manuscript room, University Library Cambridge; Archive and College Library, St. Edmund's College; Archives and Special Collections, LSE Library; Special Collection and Library, University of Liverpool Library; and the members of the Oxford University Archives, Bodleian Library. In relation to my efforts to find lost annotations or other primary material by Shackle, I would like to thank Victoria Chick, Patrick Zutshi, James Ford, Peter Earl, Bruce Littleboy, Alexandra Browne, Sussane Jennings, and Richard and Robert Shackle. Finally, I am particularly indebted to Stephen Littlechild for helping me locate (Meadows 1997).

Disclosure statement

No potential conflict of interest was reported by the author.

Notes

* This research was conducted in the Institute of Management Studies, Goldsmiths, University of London and St. Edmund's College, University of Cambridge.

¹ Jackson informs us that “in Charles Lamb's copy of the plays of Beaumont and Fletcher he wrote, ‘I will not be long here, Charles!—& gone, you will not mind my having spoiled a book in order to leave a Relic’”(Jackson 2001, 7). This book became a relic only because the margin was spoiled by a famous hand, otherwise this marginal note holds little use if it was made anonymously, as a rather trivial bit of marginalia.

² There has been some recent work on the concept of the common reader in economics, and the first study can be found in Lamm (1993). More recently, scholars have worked on the popularisations of contemporary economics and the development of a literature called “Economics for Fun” (see e.g., Vromen 2009; Fleury 2012; Maki 2012). The view of the economist as a public intellectual has also been explored in a recent annual supplement of HOPE (Mata and Medema 2013). Furthermore, Repapis (2014b) discussed the use of book reviews in the work of G. C. Harcourt in informing the academic community on the developments in economic theory. It is interesting, however, that marginalia have not yet received attention in this field of study.

³ By focusing on Shackle as a reader, we do not explore here the link of these readings and his annotations to the later development of his thought. Shackle's later work on expectations has been explored in some detail by scholars (for recent contributions in this literature, see Earl and Littleboy [2014], Zappia [2014] and Latsis [2015]). However, the early formation of his ideas during the 1930s is still an under-researched topic, partly because this material is not very widely known.

⁴ This list is not exhaustive, Shackle notes in a letter to Frowen on 1962 that “The Hayek seminar at LSE in my day included both Paul Chambers and Sir Robert Shone” (Frowen 2004, 105). Both Sir Paul Chambers and Sir Robert Shone became distinguished economists, working both in industry and academia.

⁵ Recollections on the importance of the two seminars vary between contemporary accounts. For example, Susan Howson in her biography of Lionel Robbins writes: “Most recollections are, however, of the ‘Grand Seminar’ especially in 1933/4 – 1936/7 when it took place on Monday afternoons” (Howson 2011, 251). Instead, it appears that Shackle found the seminar organised by Hayek for the LSE graduate students as more memorable and intellectually stimulating.

⁶ Incidentally, he also read Thomas’ article in the March issue of the Economic Journal on the Swedish economist David Davidson (Thomas 1935). This was at the same time that Shackle was attending Thomas’ lectures. In fact, the majority of his readings during the 1930s were of articles written by people who he knew personally to some degree, either as teachers, colleagues or even examiners for his PhD thesis. Being a PhD student at the LSE, visiting Cambridge for a research conference in 1935, and getting his first posting in Oxford, meant that he came into sustained contact with the majority of British academics working in economic theory at the time.

⁷ The degree was conferred on 5 June 1940. Shackle does not appear to have submitted these papers in a thesis format, and the Bodleian Library holds no copy of Shackle's thesis. The examiner's report notes six papers “which have appeared in learned journals”, and then adds “Three of these were elaborate statistical calculations on closely related subjects while two give a theoretical discussion of the trade cycle” (Oxford University Archives, FA 4/18/2/5). It is unclear if there was a sixth paper, it could be an earlier publication in the RES (Shackle 1937a) or Economica (Shackle 1936), or a typo in the examiner's report, and they considered only the five papers previously mentioned.

⁸ And while these datings are useful for anyone working on the intellectual development of Shackle thought, it would appear odd to suggest that Shackle developed this habit for this purpose from the very beginning of his career. Thus the motivation for this practice must be due to other reasons.

⁹ Harcourt informs us that “Shackle's mother said of him that he practiced the three Rs - reading, writing and rubbing out” (Harcourt 1981, 136).

¹⁰ Gabriel Harvey was a renaissance English humanist with an extensive private library.

¹¹ All Add. number references relate to University Library, Cambridge, classification system.

¹² A carefully reading of this note shows that Shackle was still changing and adapting how he read texts and interacted with them, hinting that he was using different approaches for different types of texts. Introduction to Algebraic Theories by A. Adrian Albert, which is a complex algebra textbook, has the most ornate system of annotations, dating notes and interleafing of any of his surviving books, so that we know, for example, when he attempted particular mathematical exercises.

¹³ This literature is very broad and cannot be adequately surveyed here. Perhaps an entry point into the literature for the non-specialist reader is Repapis (2014a).

¹⁴ There is no definitive evidence that Shackle read Kahn's The Relation of Home Investment to Unemployment at that time, and although the marginal annotations we have are undated, they most probably relate to a later period.

¹⁵ His pro-Hayek position becomes explicit in the following quote: “Dr. Hayek has thus shown that, starting from a position of equilibrium in which all the available resources are fully employed, a situation can arise through the self-intensifying character of disturbances in an elastic-money economy, where some of these resources are for the time being unemployable”(Shackle 1933, 37). Nevertheless, the stress is not that Hayek has shown this to be the case and therefore Keynes’ theory is wrong. Shackle instead is arguing that a “Hayekian type” cycle is theoretically possible, a much weaker statement than an acceptance that this alone explains the great depression.

¹⁶ In Repapis (2017), it is argued that the volumes from this period's Economic Journal that do not survive most probably did not hold any annotations by Shackle, and that is why they were not deposited in the archive.

¹⁷ He notes, for example, on one occasion, “While Prof. Hayek uses these two definitions as equivalent, it would seem that only the latter is correct” (marginalia on page 219 of Hayek [1934], Add. 7669, Box 17).

¹⁸ There is also a very interesting and thought provoking reading by Shackle of Pigou's Net Income and Capital Depletion (Pigou 1935) on September 1935. He reacts to Pigou's comment that “For economics the stock of capital existing at any time is a collection of objects…” (Pigou 1935, 235), with “The ‘size of the stock of capital’ possessed by an economy at a given moment of its life has meaning only in relation to a purpose, namely, the time-shape of the economy's intended evolution” (attached loose-leaf note to page 235 of Pigou [1935], Add. 7669, Box 17). Later on, Shackle eloquently writes in clear Hayekian spirit: “A measure of quantity of capital, which leaves it unaltered by a change in taste or knowledge but causes it to be reduced by physical wear and tear, is not interesting to the economist any more than to the business man. It is irrelevant” (loose-leaf note between pages 238 and 239 of Pigou [1935], Add. 7669, Box 17).

¹⁹ In Harcourt (1981, 144), we find that “Shackle does remember reading the Hayek-Knight exchanges on capital theory in the 1930s (he thinks Knight published essentially the same article in four or five different places!)”. This memory may be emanating from this reading on March 1935 of Knight (1935), which he appears not to have found impressive.

²⁰ I have argued elsewhere (Repapis 2011) that Hayek's thoughts on the Austrian business cycle were evolving through the 1930s. In fact, if we are looking for a picture to capture this complex process of Hayek's efforts towards articulation of his thought and its dissemination to his disciples, we need not look further than Appendix 1. Page 217 of Hayek's (1934) article in Shackle's archive fully captures the relationship between text and the different layers of marginal gloss, all efforts for more clarity and better articulation. We observe that Hayek not only takes about 40% of the page in footnotes, but that Shackle on top of that encrusts the whole text and Hayek's footnotes in his own gloss. And the gloss spills into loose leaf notes (see the markings of the paper clips on the page), as the effort to fully understand the text's message continues in repeated readings.

²¹ Therefore, he never returned to Hayek (1935) for a “fresh” reading, something which he occasionally had done with other articles that he heavily read in the past.

²² Shackle did not put a date at the end of the article to signify he finished reading it. While this is uncommon it is not the only time this happens.

²³ However, there are few substantive annotations from the period on Keynes’ articles, even when there are a number of readings of the article. This is possibly because these were not Keynes’ most important theoretical pieces. It would have been interesting to have his notes on Keynes QJE article, but to this as to all his readings of American journals we can only speculate. Furthermore, he did not read everything that Keynes published at the time. To give one example, there is no dated reading for Keynes (1936a). In this case, it may be that he was not interested in empirical matters on investment, or he thought it to be not an important contribution as it was in the notes and memoranda section of EJ, or for some other reason lost in the mists of time. But the important point is Shackle did not appear to feel the need to read everything that came from Keynes’ pen at the time, and especially articles that did not directly relate to theoretical developments.

²⁴ While implicitly also dismissing it as irrelevant in practice. He writes in the introduction of his thesis, “Professor Hayek assumed an initial state of ‘full employment’, and the investment-boom which he describes is therefore at an expense of consumption. In this essay, however, I have tried to suggest how an investment-boom can develop from the conditions of severe underemployment which in historical cases has always preceded it” (Shackle 1937b, 6).

²⁵ We know that Shackle sent his book (Shackle 1938 [1968]) to Keynes, and Keynes replied thanking him. Keynes’ letter dated 30 April 1938, survives in the Shackle archive (and has been published in Littlechild [2000], 362–363) and shows that Keynes approved of the book, agreeing with a number of ideas, and writing in his reply, “I quite agree with you that the next step is to try to ‘build up a chain of situations growing one out of another representing a process in time’” (Littlechild 2000, 362).

²⁶ He did, however, make a very positive reference to Hayek's work on knowledge in Shackle (1937a, 108, footnote 1). Presumably this is a reference to Hayek's (1937) article. We don't know when Shackle read the article during the 1930s as that volume of Economica is missing from the Shackle archive.

²⁷ The book appears completely unread, and in fact, pages 154–155, 158–159, 250–251 and 254–255 still remained uncut when I visited the library in July 2015. The only way we know this was owned by Shackle is because in the front inside cover there is a bookplate of Liverpool University noting that it is a gift to the library by him.

²⁸ An indication of using insights from academics from both camps can be seen in footnote one in Shackle (1937a, 108). There Shackle thanks Hayek for his insights on his work on knowledge and its diffusion in the economy, Keynes on the evolution of the concepts of investment and income, and Keynes jointly with Kahn for their development of the multiplier.

²⁹ Admittedly Hayek's account is both open-ended and somewhat unclear. I am not aware of Hayek developing this thesis more fully elsewhere. Moreover, the above quote continues with the following line: ”it is certainly more than a coincidence that The General Theory appeared only three years after the foundation of the Econometric Society - even Keynes had no particular sympathy with the members of that group” (Hayek 1995, 60). This seems to suggest that Hayek was conflating, to some degree, the Keynesian revolution with the rapid growth of econometric techniques that was taking place about the same time, and considering the latter as the central change during the 1930s. In the rest of this section, I do not follow this line of argument and consider macroeconomic developments from the perspective of The General Theory and the Cambridge Keynesians working with Keynes as the central change.

³⁰ He may also have read Meade's The Rate of Interest in a Progressive State during this period. We know through Ford's biography (Ford 1994, 14) that he read it at least once towards the end of 1934, but as it has been lost we do not know Shackle's full reading history for this volume.

³¹ He did also read articles that relate to individual behaviour, like Irrationality in Consumers' Demand, by Reddaway, and The Concept of Hoarding by Robinson, as well as articles on uncertainty and risk for example Steindl's On Risk. All of these readings, however, are firmly connected with the Keynesian revolution and can be seen more are extensions of the research project brought forth by Keynes’ General Theory, and related to Shackle's own work on expectations, which would come together in his (1949) book on Expectations in Economics.

³² Shackle's reading can say something about these last two topics – which are interlinked. Thus, it is not only that he worked extensively on algebra in the late 1930s and early 1940s, even writing that the topic fascinates him and adding “One day I shall strike out from it some new ideas for my own subject” (see Appendix 2 for full quote), but also by the fact that he read Tinbergen's work (Tinbergen 1937, 1939) on the new type of econometric analysis for business cycle research. Interestingly, he read on 28/5/43 on the same day, parts of Tinbergen (1939) and the exchange in the Economic Journal between Tinbergen and Keynes, which voiced Keynes’ concerns on the limits of this type of testing. He returned to Tinbergen (1939) for a more in-depth reading in August–September of 1945, only days after the end of WWII.

³³ I would like to thank Professor Roger Backhouse and the participants of the 2015 ESHET session for making this point.

³⁴ Lipking writes that “the gloss outlasts the text” (Lipking 1977, 651). In many ways, this can ring true in economics, where the standardisation of an abstract idea may be very different to its original loose and speculative articulation embedded in a text.

Additional information

Funding

G.L.S. Shackle Fund [grant number GLSHACKLE2015]