NETWORKS OF HISTORIANS OF ECONOMICS: FIFTY YEARS OF HISTORY OF ECONOMICS SOCIETY CONFERENCES

I. INTRODUCTION

The History of Economics Society (HES) is neither the oldest nor the largest academic society of historians of economic thought.¹ However, no one would deny—certainly not when celebrating its fiftieth anniversary—the vital role the HES has had in establishing and shaping the field of history of economics. The goal of this paper is to go beyond mere favorable impressions to an evidence-based history grounded in analyzing the attendees of the exploratory conference in 1973 and the fifty annual HES meetings that followed.

By the time the first HES conference took place, academic conferences had become permanent fixtures of the scientific way of life (Bigg et al. 2023). This profoundly impacted the mode of operation of international academic communities. The Republic of Letters, the transnational network of scholars of the early modern period in Europe, primarily operated through print and epistolary channels. Scholars rarely met, and when they did, it was on local and later national levels. The emergence of international meetings, “by contrast, brought scientists together regularly, in the flesh and in great numbers. Their previously imagined community now became tangible” (Somsen 2023, p. 453). So was the case of the HES. It was born in a period when the field of history of economics was acquiring the trappings of a distinct subfield, such as annual conferences organized by an academic interest group, which was one of the markers that helped to form a scientific community.

Despite their ubiquity, and perhaps because of it, conferences have been studied by historians and sociologists of science primarily as scenes for larger scientific developments (Bigg et al. 2023). ² Only rarely did they focus on them as a manifestation of scientific practice. One such exception comes from our field. The role of conferences in knowledge production, dissemination, and community formation was the subject of a special double issue of the Revue d’économie politique in 2021. The opening study of the editors Beatrice Cherrier and Aurélien Saïdi provides the first systematic attempt to historicize and categorize seminars, conferences, and workshops in the history of economics “as lenses to study science in action” (Cherrier and Saïdi 2021, p. 611). ³

Therefore, this paper contributes to a more empirical understanding of the first half of a century of the HES and its conferences. It also endeavors to advance the approach to conferences as a historical subject in its own right by analyzing the patterns and dynamics of discipline formation. To this end, prosopography as a “historiographical method that identifies and draws relationships between various people within a specific, well-defined historical or social context by collecting and analyzing relevant biographical data” is used (Svorenčík 2018, p. 605). ⁴

The approach is not entirely novel but still rarely used. It was only with the work of Thomas Söderqvist and Arthur Silverstein (1994) that a prosopographic approach to scientific conferences was applied beyond merely counting who came and who did not. They studied the participants of postwar immunology conferences that took place between 1951 and 1972 and were able to detect clusters of these conferences, revealing the transformations and disciplinary dynamics within the field of immunology. Prosopography coupled with network analysis also found application in the historical study of economics conferences. Analyzing the European and American economists attending the “International Seminar on Macroeconomics” (1978 to 1993) showed the growing centrality of US macroeconomists and the adoption of American professional standards in Europe (Goutsmedt, Renault, and Sergi 2021).

When zooming in on historians of economics and the conferences they organize, John Davis (2002) was the first to adopt a statistical approach. He analyzed the programs of the HES conferences from 1978 until 2000 to study the role of these meetings in the gestation of the history of economics as a subdiscipline of economics. However, he faced two limitations. First, early HES programs until 1977 had been unavailable to him. Second, to avoid laborious analysis of individual attendees, he focused on the number of presentations per conference as a measure of participation. He observed that the participation of women and non-US-based scholars increased over time. While he concurred on the importance of HES in the internal development of the history of the economics subfield, he argued that it had a very limited impact on the field’s standing within its native discipline of economics.

Following the prosopographic approach of Maxime Desmarais-Tremblay and Andrej Svorenčík (2021), who analyzed a quarter-century of the European Society for the History of Economic Thought (ESHET) conferences (1997 to 2019), ⁵ half a century’s worth of HES conference programs is used in this paper to create a list of all attendees. For regular attendees with at least three participations, biographical data are collected, as described in the second section. Attendees with at least three participations are labeled “regular attendees” or, put shortly, “regulars.”

The HES was incorporated in 1974 without explicitly stating that it was an international society. From the start, it drew members from outside the US and even outside of North America. Only in 2016 was the HES constitution amended to reflect its international nature. The subsequent sections, therefore, provide a set of observations about the composition and evolving structure of HES attendees that help to explore the extent to which the HES has been a US-centric society. Applying Jaccard’s similarity measure allows for comparing past HES conferences in terms of their participants and identifying clusters of similar conferences. The engagement of scholars from and outside North America—in terms of education, employment, academic age, and gender—is analyzed. A comparison of the HES with ESHET follows and shows the different paths the two societies have taken. The final section summarizes the main findings and tries to make sense of what these observations mean and potentially imply.

II. SOURCES AND DESCRIPTION OF DATA

For most years, the HES has not created or kept a complete attendance list or a list of people who paid the conference fee. Therefore, the primary data source about HES conference attendees are conference programs published on the HES website. ⁶ Marianne Johnson did a great deal of work assembling these resources during her tenure as HES secretary (2015 to 2022). Due to missing or non-existent digital files, many programs had to be located and scanned in the HES archive at Duke University. Nevertheless, four programs from the 1990s remained missing, and in February 2019, she sent a message to the SHOE list subscribers, asking for their help. Despite these efforts, programs for the 1991 conference in College Park, MD, the 1993 conference in Philadelphia, and the 1995 conference in Indiana were still missing in the fall of 2023. I contacted dozens of attendees of the 1990s conferences, but no one could find any of the missing programs. Fortunately, published versions were found hidden in the Notices and Communications section of old JHET issues. All conference programs are thus available.

Unfortunately, the full text of abstracts and presented papers remains unavailable for most years. This significantly limits the ability to analyze the changing popularity of various topics and methodological approaches within the HES community. Relying solely on session titles and presentation titles provides a much narrower window into these trends, making it impossible to conduct a comprehensive corpus or content analysis of the research presented.

The key concept of this paper is the notion of “conference attendee” or “participant," operationalized as anyone listed in the program, regardless of their scheduled role (chair, discussant, presenter, panelist). Notably, being included in the HES leadership roster, often found within the program, does not qualify as attendance.

This paper adopts the participation threshold of three, established by Desmarais-Tremblay and Svorenčík (2021), to ensure consistency and enable comparisons between HES and ESHET. While other studies have used lower thresholds, e.g., Söderqvist and Silverstein used two (1994), setting the bar at three offers several advantages over higher thresholds.

First, a threshold of three allows us to track scholars who might not have remained active in the field long term. This includes those who attended as PhD students but have not pursued academia further or transitioned to different subfields, but attended four or five times. Second, it also captures regulars, such as young scholars who recently started attending HES conferences. Third, ESHET and HES conferences do not have a strict admission process. ⁷ This “lack of selectivity in the program” has been criticized for not creating and enforcing high community standards (Weintraub, this issue). However, it removes a potential heterogeneity bias in attendance: a strict admission process would vary from one HES conference organizer to another and would be reinforced by the evolving criteria of scholarly quality. ⁸

The reliance on printed programs excludes those who attended but are not mentioned in a program. However, those mentioned in the program but who canceled at the last moment are included. Another type of overcounting occurs with co-authors of a presentation when a program does not specify the presenting author. In such cases, all co-authors are counted with the same weight as single-authored presentations. In order to distinguish intensity of attendance from frequency of attendance, the concept of very regular attendees is introduced—regulars who, between their first and last attendance, came to at least 40% of all conferences.

Biographical data on their educational and employment history were collected for the subset of regular attendees. The primary sources of this data were people’s online CVs, obituaries, various academic directories, personal websites, and occasionally publications about them. Dissertation databases were used to obtain information about thesis advisors and dissertation committee members. Sometimes, information was acquired via email or by consulting the acknowledgment section of published dissertations.

III. ATTENDEES AND REGULAR ATTENDEES

There are 2,154 unique individuals listed in the HES conference programs (Table 1). Due to missing full first names or insufficient information to identify scholars (no affiliation, missing full first name, or generic surnames), for 106, or 5% of all attendees, it was impossible to determine their gender. Just 18% of all attendees of known gender are women, and 16% are women regulars.

Table 1. HES Annual Conferences 1973–2023 Attendance Statistics

Notes: The fourth column contains individuals whose gender was not determined. For the calculation in the penultimate column attendees of unknown gender are excluded.

Most scholars—1,236 and 337—participated in just one or two conferences, respectively. The remaining 581 individuals, or 27% of all, attended at least three times, representing a group of regular attendees. Three regulars—John Davis, Robert Dimand, and David Levy—attended thirty-six conferences each, the highest observed participation. At least 388 regulars are alive, and eighty-six confirmed deceased (most recent being Todd Lowry in April 2023, the HES president in 1990–91; David Colander in December 2023, the HES president in 1998–99; and Mauro Boianovsky in February 2024, the HES president in 2016–17).

The frequency of attendance of HES presidents—depicted in the last column of Table 1—shows that they make up a disproportionate share of attendees who came to HES conferences at least seventeen times. Put differently, they mainly belong to the most active participants with the longest exposure time to influence the HES in intellectual and organizational matters. On average, it took 25.7 years since earning the highest academic degree for a person to become an HES president. The share of women HES presidents—seven out of forty-seven (14.9%)—is below the share of women regulars. However, the presidential term increased to two years, starting with Evelyn Forget’s tenure in 2017. If the length of tenure is accounted for, the share of women leading the HES increases to 15.7%, which is comparable with the share of women regulars.

The exploratory conference in 1973 attracted just nineteen attendees, and the first regular conference a year later attracted forty-eight attendees. Until the end of the last century, participation oscillated around an upward-sloping trend. The 1987 conference hosted at the Harvard Business School scored a record with 205 attendees (Figure 1).

Figure 1. Number of HES Conference Attendees.

Notes: Number of all attendees (full line) with a ten-year moving average (dotted line), regulars (dashed line), and new regulars of HES conferences (double full line). New regulars are attendees who have just reached three participations in a particular year.

In the first fifteen years of this century, the HES faced a slow decline in attendance, reaching levels similar to the first half of the 1980s in 2015. The four conferences prior to the pandemic reversed the downturn. Even the 1987 participation record was broken during the 2019 conference in New York with 206 participants. Regrettably, HES attendance had not fully recovered from the pandemic yet when two virtual conferences were held in 2020 and 2021. A similar pattern can be observed regarding new regular attendees, i.e., those who became regular attendees (see Figure A.1 in the online appendix).

A meaningful measure of engagement with the HES is the number of years it takes an attendee to obtain the highest degree and become a regular. This measure can be reliably calculated only for those 422 regulars who received their highest degree since 1973. On average, it took them 9.41 years from graduation to become an HES regular. Women are slightly faster, with an average of 8.48 years compared with 9.62 years for men. However, if we exclude regulars whose academic age at becoming a regular was twice the average (nineteen years), the average for the 359 scholars was just 6.68 (for women, 6.52, and for men, 6.72). This is more in line with the overall average academic age of 6.1 years for ESHET regulars.

IV. HIERARCHICAL CLUSTERING ANALYSIS

Hierarchical clustering is a data analysis method that starts with individual data points and merges them progressively based on their similarity. Clustering analysis, in general, can be used as a heuristic device, with the identified clusters requiring interpretation in conjunction with other insights.

This paper uniquely applies the hierarchical clustering method to HES conferences, using their list of attendees as the starting data points. For each year, a vector of ones and zeros is created, indicating which of the 2,154 unique attendees from the fifty-year period attended that year. The similarity between any two conference years is determined using Jaccard’s measure, which is typically used for binary data. Further technical details can be found in the online appendix.

The results of the hierarchical clustering analysis are two trees or dendrograms of conferences depicted in Figure 2, and each should be viewed from right to left. The vertical axis represents HES conference years not ordered chronologically and is identical in all figures. What changes is the cut-off point on the horizontal axis that reveals a nested hierarchy of clustered conferences. Looking at the left part of Figure 2, we see that the fifty HES conferences can be split into two clusters, the first spanning 1973 through 1986 and the second covering the remaining period. This suggests a significant change in people attending the HES conference before 1986 and after. Lowering the cluster threshold by moving the cut-off point closer to zero on the horizontal axis retains the 1973 to 1986 cluster but splits the larger 1987 to 2023 cluster into two: the first covering all conferences until 2004 and the second covering all conferences thereafter.

Figure 2. Hierarchical Clustering (Largest Clusters) of HES Conferences.

Notes: Dendrogram of HES conferences based on Jaccard’s similarity measure and Ward linkage method. Highlighted are the largest clusters; vertical axis is identical.

Please note that within the clusters, the years do not follow a strict chronological order but are ordered according to their similarity. The non-trivial aspect of the clustering analysis is that each cluster contains all years within its range of years, and the three clusters partition the fifty-year period. Therefore, the key insight of clustering analysis in the HES case is that there are three main conference clusters that periodize HES conference history into three distinct, chronologically ordered groups. This is not a superficial result, as the attendance patterns could have been entirely different, and the partition into three subsets of conferences was not a guaranteed outcome. For instance, if, for whatever reason, attendees were split into two groups that came every other year (say conferences focusing on pre-World War II and postwar topics attract different crowds), the clusters would not partition the fifty years into distinct periods.

V. DOMINANCE OF REGULAR ATTENDEES

Regular attendees represent a substantial portion of HES conference participants. The dominance of regulars in this context refers to the high share of regulars among all conference participants, which takes on a different meaning depending on changes in overall attendance. In general, a stable low share of regulars—assuming the number of attendees is also stable—means a constant influx of new attendees who come once or twice but do not become interested in long-term engagement with an organization. Similarly, a stable high share of regulars implies an inability to attract and tie newcomers. However, if the number of attendees is growing, then a high share of regulars implies a constant ability to attract new attendees and turn them into regulars.

The 1970s and the first half of the 1980s saw an initial build-up of a base of regular attendees as more and more participants came three or more times. In 1985, for the first time, the share of regulars on all participants surpassed 50%. The steep decline in the share of HES regulars in the second half of the 1980s (down to 30% in 1988) resulted from a substantial increase in attendees. Once enough of the new attendees became regulars, the share reached its highest peak ever of 69% in 1993. From the early 1990s until the pandemic, regular attendees made up an average of 56% of all attendees. As the number of attendees declined between 2009 and 2015, the share of regulars remained stable, indicating an inability to attract new scholars. This changed in the second half of the 2010s when growing attendance was met with a declining share of regulars (Figure 3).

Figure 3. Share of Regulars at HES Conferences.

Notes: Share of regulars (full line without markers), very regular participants (dotted line), and regulars with at least five participations (dashed line) on all participants in any given year. Future regulars are attendees in a given year who are in their first or second attendance and, by the end of 2023, become regulars. The top full line with markers represents future regulars and regular HES attendees together. The future regulars are thus the difference between the two full lines.

To illustrate the dominance of regular attendees, it is helpful to restrict the set of regular attendees. If we consider only regulars who came to at least 40% of all conferences between their first and last attendance, we get a subset of 282 regulars. These very regular attendees represent over 48% of all regulars but account for 44% of all attendees since 1990 (Figure 3).

The large group of non-regular attendees can be split into those who came just once or twice over fifty years of HES and those who eventually became HES regulars. The latter group of future regulars, i.e., attendees who are in their first or second attendance in a particular year but by 2023 became regulars, is also depicted in Figure 3. On average, 21% of all attendees in any given year are future regulars, including the early years when there were no regulars yet and years shortly before and during the pandemic. The average for the period 1979 to 1999 is 24%, and for the period 2000 to 2018, 16%, indicating a lower ability of the HES to turn attendees into regulars. That might be caused by increased co-authorship of presented papers, more frequent attendance of scholars outside the traditional geographic and disciplinary boundaries of typical HES regulars, and increased opportunities to attend other national and regional conferences for historians of economics.

VI. A NORTH AMERICAN AFFAIR?

The HES originated in North America, which is also reflected in the location of the annual conferences. The annual conference was typically held at a university where the current president worked. ⁹ Thirty-seven of them took place in the USA and twelve in Canada. Until 2023, only one conference was supposed to occur outside North America. Due to the pandemic, the 2020 conference scheduled in Utrecht was turned into a virtual conference, and then its redux in 2021 was made virtual at the last moment as well. ¹⁰

More than half of the conferences—twenty-eight in total—took place in just nine cities: Toronto leads with five, Vancouver and Boston hosted four each. Chicago, Durham, and Fairfax follow with three each. The list ends with Montreal and New York, with two each.

Given this geographic concentration of conference location, it is natural to ask to what extent HES regulars are geographically concentrated. To study this issue, I will consider the location of studies, especially of the highest degree but also of lower academic degrees, and the country of most recent employment of regulars. Visiting positions were excluded. In terms of education, information about the country of undergraduate and graduate studies and the institution of the highest earned degree were coded.

Regulars’ Education: American Dominance and Its Decline

Over the twentieth century, doctoral degrees became essential for academic careers worldwide. This can be seen when examining the highest-earned degree of the 581 HES regulars. Information about the highest degree is missing just for eighteen regulars. Twenty-eight regulars have not been conferred a doctoral degree, and a master’s degree is their terminal degree. Just two regulars were pursuing a PhD at the end of 2023, not in economics but in history. The remaining 533 regulars earned a PhD (91.6%). ¹¹ The earliest earned PhD was granted to William Jaffe from the Sorbonne in 1924, half a century before the first HES conference. Therefore, information about the highest degree achieved is available for 561 regular attendees. ¹²

Figure 4 visualizes the changing shares of countries and regions of the highest degree of origin for all attendees. Two are large groups of attendees. First, non-regular participants make up the largest share initially, which is natural as it took time for a set of regulars to emerge. ¹³ The other large group is regulars who were educated in the US. They reached their highest share in 1986 and 1991 with 40% and 41%, respectively, coinciding with periods of high attendance and a large number of regulars graduating in the US (Figure 6). Excluding the pandemic 2020 conference, the lowest share was reached in 2019, with just 17%. The 2019 conference in New York was very specific—it attracted a record number of attendees and a high share of regulars (highest since 2015) especially from Europe and Latin America. This is a share not seen since the 1980s and resulted from a gradual decline in the 2010s, indicating a potential coming to an end of a strong generation of US-educated regulars. The most recent peak in 2023 might result from the fiftieth-anniversary conference in Vancouver, which many past HES presidents attended.

Figure 4. Shares of Countries of Highest Degree of HES Attendees.

Notes: Share of countries with PhD degrees (or highest degrees) of HES attendees at conference in any given year.

Figure 4 provides several additional insights. Regulars educated in Canada form a very small share, consistently outpaced by UK-educated scholars. Therefore, North American dominance should be understood as mainly US dominance, at least when considering the educational background.

While far fewer than the US regulars, UK-educated regulars were the first significant foreign (from the US perspective) group to appear in the 1980s, lasting until the first half of the 2000s. From the second half of the first decade of the current century, French-educated regulars took over as the largest country outside the US, almost overcoming US-educated regulars in the second half of the 2010s. In the anomalous 2020 online conference, they overtook them.

Regulars educated in European countries other than France and the UK and all remaining regulars (mainly educated in South America and Japan) have been present from the 1980s and became a permanent fixture of HES conferences from the 1990s, reflecting the HES’s internationalization but also decreasing costs of long-haul air travel. Regulars received their highest degree from twenty-eight countries in total. When one is considering their undergraduate studies, the number of countries rises to forty-two (online appendix, Table A.1).

Table 2 breaks down the number of regulars with a PhD across various countries, gender composition, and the number of institutions involved. Looking at a subset of regulars who graduated in the past twenty-five years reveals a dramatic decline in all Anglo-Saxon countries, with Australia and the United States experiencing the most prominent decline. The geographic distance might explain the Australian case. However, the US decline is indicative of the overall demise of the history of economics field. No country has experienced an increase in the number of PhD-granting institutions to HES regulars in the past twenty-five years (last column).

Table 2. Countries of Doctoral Origin of HES Regulars

Notes: Only completed PhDs (N=553) are counted. Four dual PhDs are counted in the country, each with a weight of 0.5 and not as a separate institution. Column 9 refers only to regulars who were awarded a PhD degree after 1998 (the second half of the fifty-year period).

* Anglo-Saxon countries are in bold in third-last column.

There are twenty-four PhD-granting institutions, from which almost half of the 533 regulars with PhDs graduated (Table 3). Nine are outside the US, with Paris 1 Panthéon-Sorbonne being the most productive of all universities. The last two columns show information about a regular’s earliest and latest graduation year (in and outside of economics departments) and allow the division of all universities into four groups. First, universities that had conferred PhD degrees to scholars who became HES regulars before HES’s establishment in 1973 but failed to produce a regular in the past ten years (2013 or later) contain most American universities: Harvard, Chicago, Wisconsin, Berkeley, Columbia, Princeton, University of North Carolina, and Utah; but Cambridge and Lyon as well. Second, universities that were early producers and have remained so are Paris 1 Panthéon-Sorbonne, Oxford, London School of Economics (LSE), Erasmus University in Rotterdam, and Duke as the only long-term American representative—though with uncertain prospects.

Table 3. Most Active Universities Conferring PhD Degrees to Regular Attendees

Notes: Dual degrees are shared equally between institutions. Universities producing five regulars include LSU, NYU, Toronto, MSU, Illinois, Freiburg, and Notre Dame. Universities producing four regulars include MIT, Stanford, Virginia, and Edinburgh.

Third, universities that started producing regulars after HES’s establishment but did not manage to do so in the past ten years include Pittsburg, Michigan, Massachusetts, Amsterdam, and Florence. The last two are examples of former European centers of HES that disintegrated. Fourth, relative newcomers still producing regulars include São Paolo and Paris Nanterre outside the US and George Mason University and the New School in the US.

The once-diverse landscape of US PhD-granting institutions for HES regulars has shrunk to just three centers: the economics departments at George Mason University, Duke, and the New School, in order of most recent graduates who became regulars. The last column of Table 3 shows a shift underway in American academe—regulars getting their PhD from non-economics departments such as Harvard, Chicago, and Princeton.

Another way of depicting the changing characteristics of HES regulars is to look at the stock of regulars. A moving window is used to create a cohort of regulars who graduated during a period of twenty-five years. More specifically, I start with the stock of regulars and the data about their PhD graduation year. For each year from 1973 through 2022, I look at a window of the previous twenty-five years and calculate the number of regulars who graduated in that period, regardless of their attendance in that year (or any previous twenty-five years). The resulting Figure 5 shows that France, the Netherlands, Italy, and Brazil gained share at the expense of the US. Meanwhile, the UK, Germany, and Canada have maintained their share over a fifty-year period. These changes relate to the stock of regulars, not to any trends in attendance, as attendance in specific years is not part of the calculation.

Figure 5. Stock Perspective on the PhD Background of Regulars in 25-year Moving Window.

Notes: Each year, a share of regulars graduating from a country or a region is computed by considering only PhD degrees from the past twenty-five years, regardless of attendance.

The observed decline in the share of US-educated HES regulars during the 2010s is likely a manifestation of the ending academic life cycles of two different American cohorts of HES regulars. Figure 6 shows a smoothed distribution of years when US-educated regulars graduated, showing two waves. The first wave started in 1949 and peaked in 1960 before dropping to zero graduates in 1965—this is the cohort of forty-nine founding figures of the HES. On average, their last participation was in 1997, and the last one was in 2017. The second wave started in the second half of the 1960s, plateaued in the 1980s, and ended in the early 1990s. This cohort represents the largest group of 147 HES regulars. Seventy-nine, or 54%, had their last attendance before 2010. For 100 of them, the last attendance was before 2015.

Figure 6. Smoothed-Out Distribution of PhD Graduation Years of US and French HES Regulars.

Notes: A five-year moving average is applied to the distribution of PhD graduation years of HES regulars educated in the US (full line) and France (dotted line).

To compare with French-educated regulars, Figure 6 also depicts their smoothed distribution of graduation years. Two French generations can be discerned: one not as old as the generation of American HES founders that coincides with the second American generation, and a strong second French generation emerging in the 2000s that interacts more with the HES. Since 2000, there have been forty-four French-educated regulars while only thirty-three US-educated regulars.

Employment: Low Mobility?

A common feature of prosopographies of academic economists is that they are employed in more locations than the number of locations where they earned the highest degree, which reflects the hierarchical structure of the discipline, i.e., that there are many more academic economics departments or research institutions such as central banks without a PhD program than with one (Cherrier and Svorenčík 2018; Hoover and Svorenčík 2023; Svorenčík 2014, 2018). That is also the case of HES regulars.

For 535 HES regulars, I reconstructed their career paths from the country of PhD education (not necessarily their country of birth) to the most recently available place of employment. The pattern of academic mobility can be depicted in an export-import table (Table 4). The rows of this table list all thirteen countries in which at least three HES regulars graduated. Fourteen additional PhD-granting countries are grouped as the Rest of the World education category. ¹⁴ The columns denote countries of employment. There are thirteen core countries in which 489 regulars work, and the remaining forty-six regulars worked in eighteen countries grouped in the column Rest of the World employment. ¹⁵ The export-import table can track how many regulars educated in country X work in country Y. Three-quarters of all regulars worked in the same country as they received their doctorate. The overall mobility rate is 25.1%, but does not factor in any undergraduate studies abroad.

Table 4. Export-Import Table of HES Regulars

Notes: Rows denote the country of PhD degree. Columns denote the country of latest employment. Work in two countries is shared equally. Values in bold on the diagonal represent the number of regulars working in the same country as they received their PhD. Other values in bold are countries exporting or importing more than half of PhDs. The categories “Rest of the World education” and “Rest of the World employment” are explained in the text.

There are differences among countries in their relationship between educating regulars and keeping or sending them away to work abroad. Switzerland, the UK, and the Netherlands are the only countries that exported more graduates than those who stayed—as depicted in bold in the last column of Table 4. Japan has a 0% export share. The lowest non-zero export share had the United States, which is in stark contrast to the fact that the US has been an exporting country for economics PhDs in the postwar period (Aslanbeigui and Montecinos 1998). Again, Switzerland, Canada, Australia, and the Rest of the World region have more than half of their regulars imported, i.e., educated in other countries. Japan and the US have the lowest import ratios.

The “balance of trade”—exports less imports—sheds light on which countries are net exporters or importers of historians of economics from the perspective of HES regulars and on the strength of these countries in their ability to self-sustain their local historians of economics communities. The UK had the largest positive surplus of thirty-eight regulars. The largest deficits were run by the Rest of the World (twenty regulars), Canada (sixteen regulars), Italy (seven regulars), and Australia (four regulars). They were driven in large part by scholars coming from the UK. The US was another country welcoming UK scholars—a result of the strength of its three main programs at Oxbridge and the LSE and the openness of the local academic job markets. The Canadian deficit is mainly the result of US-educated regulars moving to Canada. The Italian case might be explained by the late adoption of the PhD degree as an academic prerequisite, leading to too few Italian scholars with a PhD who could seek employment abroad.

VII. ADVISOR-ADVISEE NETWORKS

For 379 out of the 533 regular scholars with a completed PhD, I obtained information about their PhD advisors. The primary sources of information were dissertation databases, most notably Proquest and Theses.fr, and dissertation title pages with the signatures of all advisors. ¹⁶

In total, 568 people were involved in the supervision of HES regulars. Following Svorenčík (2014), a distinction between extensive and intensive supervision is applied. The former measure counts the number of graduates with whom an advisor was involved within the set of HES regulars with a PhD. The latter accounts for co-supervision or thesis committee participation by adding weights to individual committee members supervising an HES regular. The assignment of weights follows Desmarais-Tremblay and Svorenčík (2021) as it allows the making of comparisons with advisors of ESHET regulars. Details can be found in the online appendix. The extensive supervision measure overvalues the intensity of advisors’ involvement in the supervision process. The intensive measure, with its imperfect weights, potentially misattributes the actual contribution of specific committee members. However, their combined usage sheds enough light on the advisor-advisee networks to make their usage worthwhile. There are nineteen advisors who supervised four up to ten HES regulars (Table 5). Fifteen of the nineteen most active advisors are HES regulars themselves, but only one of them (Pedro Duarte) was supervised by another HES regular.

Table 5. Most Active Supervisors.

Notes: Most active supervisors who were involved in supervising at least four HES regulars (extensive measure). Scholars in bold are HES regulars themselves. The third column refers places where the students of a supervisor received a PhD degree.

* denotes people who served as HES presidents.

Several groups of advisors and their students can be identified. There are two US networks of the same size (extensive measure) and with some overlap. The first can be described as the Duke University group, which originated with Joseph Spengler (not a regular himself), who supervised Craufurd Goodwin, who, in turn, was involved in the supervision of eight HES regulars, some of whom—Pedro Duarte, Robert Leonard, and Arjo Klamer ¹⁷—went on to become advisors of other HES regulars—creating a multi-generational network of nineteen HES regulars. The second US group originated in Chicago with Frank Knight and Jacob Viner (also supervised at Princeton), who, through their students Martin Bronfenbrenner and George Stigler, were involved in supervising further HES regulars—Mark Blaug, David Levy, and Arjo Klamer—creating a group of nineteen HES regulars. ¹⁸ Merging these two US groups into one supergroup and adding Neil De Marchi and Roy Weintraub, as they were involved in supervising some of the scholars in both groups, leads to thirty-five HES regulars in total.

There are also two French groups. The largest group consisting of eighteen HES regulars can be traced back to Hubert Brochier (1923–2017), who supervised HES regulars Annie L. Cot and Claude Menard and a non-HES regular, Jérôme Lallement. A smaller French group of fourteen HES regulars can be traced back to Arnaud Berthoud, Jacques Wolff, and André Lapidus, with the first two not being HES regulars. In total, they supervised or were on committees of eight HES regulars, including Philippe Fontaine, who supervised another four HES regulars. There are two advisors primarily based in the UK. One is Geoffrey Harcourt, who supervised five HES regulars. The other is Mary Morgan, who has been involved in the supervision of ten HES regulars in the UK and the Netherlands.

VIII. COMPARISON WITH ESHET

The prosopography of the first quarter-century of ESHET conferences (1997 to 2019) remarked that “in spite of the decline of HET in some [primarily European] universities, ESHET constitutes a lively community with reproduction mechanisms in place that has allowed it to grow over the quarter-century of its existence” (Desmarais-Tremblay and Svorenčík 2021, p. 1006). Sadly, this is not the case with HES. As demonstrated in previous subsections, the HES has been declining regarding its largest constituency in the US. In the following comparison of data between HES and ESHET, unless stated otherwise, the reference period is from 1997 until 2019.

ESHET conferences are larger than HES conferences. From the first ESHET conference in 1997 until 2019, 1,777 unique individuals, including 476 regulars, attended them. In contrast, there were 1,313 unique HES attendees and 348 regulars. ¹⁹ A larger share of women participated in ESHET (24.1%) than in HES (19.0%). Even if one restricts to regulars, the share of ESHET women is 22.9% while HES is 19.0%. However, HES fares better with female presidents, 18.2% versus 16.6%.

The average annual ESHET attendance was 201 scholars (minimum eighty-four, maximum 306), with an increasing trend over the entire period, while for the HES, it was 147 scholars (minimum eighty-four, maximum 206), with a decreasing trend until 2015.

There is a slight overlap between the places where HES and ESHET regulars most frequently graduate. Seven of the twenty-four HES all-time most productive universities are among the twenty universities producing the most ESHET regulars. Only one of these schools is in the US (the New School), and the rest are in Europe: Université Paris 1 Panthéon-Sorbonne, Cambridge, LSE, Florence, Université Paris Nanterre, and Lyon. Somewhat unsurprisingly, from the most active HES supervisors, Annie L. Cot, Geoffrey Harcourt, Mary S. Morgan, André Lapidus, and Jérôme Lallement—all based in Europe—are among the most active supervisors of ESHET regulars (Desmarais-Tremblay and Svorenčík 2021, p. 1020).

The last row of Table 4 contains information about the internationalization of various countries from the perspective of HES regulars. This index is the sum of exports and imports divided by the number of HES regulars working in the country where they were conferred their PhD. Although the index refers to half a century of HES data, each country’s value of the index is comparable to index values obtained in the ESHET prosopography covering just twenty-five years. The reasons for this are unclear (Desmarais-Tremblay and Svorenčík 2021, pp. 1018–1019). Comparison of the HES and ESHET Young Scholar programs is available in the online appendix.

IX. CONCLUSIONS

This paper provides a new historiographic perspective on the professionalization of the history of economics discipline through the analysis of the attendees of the first fifty HES annual conferences. The results are somber. Initial growth in the number of conference attendees and the share of regular attendees lasted until the 1990s but was followed by long stagnation and eventual decline in the early 2010s. The subsequent short-lived rebound was interrupted by the COVID-19 pandemic, and the past few conferences have not recovered to the pre-pandemic levels.

Hierarchical clustering analysis yielded a non-trivial result—half a century of HES conferences can be split into three consecutive periods: 1973 to 1986, 1987 to 2004, and 2005 to 2023. A more detailed analysis of the educational and employment background of regular attendees reveals that until 1986 the HES conferences were primarily US-centered. Then, while the US maintained dominance until 2004, scholars from other countries came to represent another large portion of HES regulars. Since 2005 the participation of US-educated HES regulars is gradually declining. The ensuing stagnation and decline are a result of a concurrent decline in attendance of US-educated and employed regulars, which has in large part been compensated by new regulars educated and employed abroad in Europe and France in particular, and more recently in Latin America as well. The paper thus provides the first ever—at least to my knowledge—quantitative assessment of the well-known and protracted demise of historians of economics in the United States.

The exact mechanism of how non-US historians of economics were drawn to HES conferences in North America cannot be determined based solely on the data used in this paper. Other elements of this story include intellectual leadership and pull of the American economics academe, the History of Political Economy journal and the HES-sponsored Journal of the History of Economic Thought, global emphasis on publishing in such indexed journals, availability of domestic and regional history of economics conferences, and, last but not least, the decreasing costs of international travel. ²⁰

Despite the early and persistent claims of insufficient standing of the history of economics field within economics, the fate of the field in the US was not sealed by the time HES was established. The generation of scholars who entered the field from the late 1960s until the early 1990s is far larger than the founding cohort of HES regulars and represents the bulk of US-educated HES regulars. The lack of their overall reproduction—something not observed in a similar study of the first quarter-century of ESHET conferences—is a significant outlier, considering that the US has been a major producer and exporter of economics PhDs across all subfields.

From a practical perspective, the HES continues to face a generational challenge that requires extraordinary leadership and deliberate efforts. ²¹ This generational challenge has been long anticipated and understood. Already in 2002, a HOPE conference volume entitled The Future of the History of Economics compared the development and future prospects of history of economics in various countries (Weintraub 2002b). Alas, the grim prospects identified by the volume’s numerous contributors have materialized and the HES has not been able to countervail the decline of history of economics in North America, its birthplace. In the coming decade or two, the group of North American historians of economics as identified in my analysis will continue to age and dwindle. It is not guaranteed that they will be replenished by historians from other parts of the world willing to engage in the activities of the HES. True, cultivating these new grounds will be vital if the HES wants to remain an agenda-setter and relevant player in the global disciplinary field of the history of economics. Yet, it might be tempting to be satisfied if such efforts actually succeed and the HES international character is bolstered.

I consider three tasks even more important than seeking refuge in international waters. Again, much of this has been noted by others (e.g., Schabas 1992; Weintraub 2002a, 2002b, 2007). The first task is not abandoning hope for the future of the field in the US and Canada. This constituency still remains the largest for the foreseeable future but requires renewed efforts to reinvigorate, better connect, and fuse with scholars in other domains interested in the history of economics. Second, without the unabated pursuit of scholarly excellence and innovation, the HES will lose its long-held edge compared with other national or regional associations, as so clearly spelled out by Loïc Charles (this issue). Third, the HES must continuously strive to become a more competent, better-managed, and result-oriented society with a shared sense of a mission. This entails, for instance, encouraging high retention and growth of its membership base, ensuring compliance with its own rules and regulations, providing unique and valuable services to its members, and achieving high participation in HES activities.