Introduction

What determines who is appointed to a full professorship in the social sciences and humanities at German universities? Although the professorship remains the central career goal in German academia, comparatively little is known about the institutional selection process through which candidates are chosen. This is particularly relevant in light of the expected retirement wave: within 2024-2034 about 44% of Germany’s 43,078 full-time professors (2023) turn 65 years old (Hüsch 2025). A growing body of research has examined the supply-side determinants of academic careers in Germany, tracking how publication productivity, international mobility, gender, and social capital shape individual trajectories from the doctorate or Habilitation to the first professorial appointment (Lutter and Schröder 2016; Schröder, Lutter, and Habicht 2021; Schröder, Habicht, and Lutter 2024; Jungbauer-Gans and Gross 2013). Yet these longitudinal studies focus on who eventually obtains a professorship over the course of a career, rather than on what happens inside the appointment process itself. The demand side of academic hiring – the institutional procedures, evaluative practices, and situational dynamics through which appointment committees select among candidates – has received far less empirical attention (Musselin 2010; Hamann 2019). The few existing large-n studies of German professorial appointments have been limited to specific time periods, single disciplines, or individual universities: Hamann (2019) analyzes historical appointment decisions in the humanities from the 1950s to the 1980s, while Auspurg, Hinz, and Schneck (2017) study appointment outcomes at universities in one federal state between 2001 and 2013. A multi-disciplinary, multi-university study covering the contemporary period has been conspicuously absent.

This gap is particularly striking given the distinctive features of the German appointment system. Unlike Anglo-Saxon tenure-track models, where hiring decisions tend to occur at the point of initial appointment, the German system concentrates the decisive career bottleneck at the transition to a full professorship (W2 or W3). The process is governed by provincial appointment regulations (Berufungsordnungen), which prescribe public job announcements, the formation of appointment committees (Berufungskommissionen), external peer reviews, candidate hearings, and ranked short-lists (Wissenschaftsrat 2005). In practice, these procedures take on average about two years from announcement to appointment, involve considerable local variation, and remain largely opaque despite a formal requirement of transparency. The procedure culminates in the public candidate hearing – informally known as the “Vorsingen,” evoking the image of an audition before a choir – constitutes the most visible stage of this otherwise opaque process. Long-listed candidates are typically invited to present their research before the appointment committee and the broader faculty, and a question-and-answer session before the formal interview. Increasingly, candidates are also asked to demonstrate their teaching abilities in a mock lecture (Lehrprobe) before students. While many aspects of German professorial hiring remain deliberately discreet, the hearings themselves are public: universities now routinely announce candidate lists on faculty websites, in departmental communications, or via social media, and these announcements sometimes circulate further through student representatives’ pages or professional mailing lists. This growing visibility makes the hearing stage uniquely amenable to data collection and analysis.

In this article, we exploit this transparency to assemble a novel dataset of candidates invited to hearings for more than 500 professorial positions at German universities from the late 2000s to the present day, with a focus on the social sciences and, as a comparative reference category, the humanities. We complement this core dataset with two additional sources: a comprehensive archive of digitized job advertisements for social science and economics professorships published by the Deutscher Hochschulverband (DHV) since 2014, and a compiled record of accepted and rejected professorship offers from the DHV journal “Forschung & Lehre.” Together, these three datasets allow us to trace the appointment process from job announcement through candidate hearing to final appointment or rejection. Our descriptive analyses reveal that slightly more men than women are invited to hearings, approximately 25% of candidates hold foreign affiliations (though many of these are Germans who have spent time abroad), and more than half already carry a professorial title. In logit regressions estimating the determinants of being appointed, we find that most measurable candidate characteristics – including gender, academic age, prior professorial status, foreign or domestic affiliation, citation counts, and institutional prestige – lack significant predictive power, controlling for time, university type, and discipline. The few significant predictors relate primarily to geographic proximity: candidates applying from the same city or returning to the city of their doctorate are more likely to be appointed, a pattern suggestive of academic localism. The largely insignificant other findings are consistent with earlier work suggesting that the pool of hearing invitees is sufficiently homogeneous that observable credentials no longer differentiate among them (Auspurg, Hinz, and Schneck 2017). It appears that softer, harder-to-measure factors – such as thematic fit with the advertised position, the quality of the hearing performance itself, or potentially prior personal acquaintances – may play a more decisive role.

The article sheds new light on the key procedure and ritual through which German professorship positions are allocated, finding that, if nothing else, then local academic ties matter for passing this very end of the bottleneck. The article goes along with an interactive website, the “hearing monitor” (https://dataverse21.shinyapps.io/hearing_monitoring/), which presents most of the aggregate statistical descriptions and invites future candidates or jury members to share their hearing information - much as existing in the French “Wiki Auditions” 1 - for making the hearing processes inter-regionally more transparent in social science disciplines.

The article proceeds as follows. We first review the existing literature on career determinants, appointment procedures, and hiring inequalities in German and comparative academia. We then describe our three datasets and the variables constructed from them. The results section presents descriptive statistics on the demographics and qualifications of hearing candidates, followed by logit regressions estimating the determinants of appointment success. We conclude with a discussion of our findings in the context of the broader literature on academic labor markets and the tension between meritocratic ideals and the institutional realities of professorial selection.

Literature

In international comparison, Germany looks back on a tradition of fairly homogeneous research universities and a chair-based faculty system, which endows professors with above-average autonomy (Schimank 2005). Musselin (2010) provides a foundational comparative analysis of academic labor markets in France, Germany, and the US, showing that despite growing convergence in higher education policies, national recruitment systems retain strong distinctive features in terms of career structures, hiring procedures, and the balance between institutional and disciplinary control. Despite waves of New Public Management reforms and growing competitive pressures since the 2000s, the traditional German chair model persists, shaping how academic careers unfold (Matthies and Torka 2015). Unlike tenure-track systems common in the Anglo-Saxon world, the German path to a full professorship (W2 or W3) has traditionally required a Habilitation or equivalent qualification, followed by an external appointment at another university. The introduction of junior professorships (W1) in 2002 created an alternative pathway, though their effectiveness in opening up the system remains debated (Zimmer 2018). The appointment procedure itself is regulated by provincial Berufungsordnungen, which prescribe public announcements, the formation of appointment committees (Berufungskommissionen), candidate hearings, and short-list rankings, yet leave considerable room for local variation in how these steps are implemented in practice. The Wissenschaftsrat (2005) issued landmark recommendations for improving the transparency and quality of these procedures, calling for open competition and systematic evaluation criteria.

A growing body of discipline-specific studies has examined the supply-side determinants of academic career success in Germany through longitudinal analyses. In sociology, Lutter and Schröder (2016) track researchers from their first publication to first professorship over the period 1980–2013 and find that publication productivity and female gender significantly increase tenure chances. They juxtapose an internationalist English journal article publication to a German-network based career success profile. In political science, Schröder, Lutter, and Habicht (2021) show that international mobility, academic awards, and high-quality publications make appointment more likely. For economics, Schröder, Habicht, and Lutter (2024) analyze tenure determinants from 1984 to 2021 and find that PhD origin from a top department, publication in leading journals, male gender, and international post-doctoral experience are significant predictors. Similarly, Jungbauer-Gans and Gross (2013) survey Habilitierte in multiple disciplines and identify publication productivity and network connections as key success factors. Beyond discipline-specific analyses, Püttmann, Siewert, and Frantz (2022) investigate how human capital, university prestige, third-party funding acquisition, and gender shape productivity in German political science. Naegele and Ordemann (2023) examine whether age constitutes a barrier to permanent positions, finding that older PhD holders may actually reach professorships faster, albeit often at universities of applied sciences rather than research universities. Zimmer (2018) studies the transition from junior to full professorship and highlights the role of accumulated social and cultural capital in navigating this career stage.

Comparatively less research has focused on the demand side of academic hiring, that is, the institutional appointment process and evaluation practices through which candidates are selected. An important contribution is Auspurg, Hinz, and Schneck (2017), who conceptualize German appointment procedures as “tournaments” and analyze gender-specific chances of being placed on short-lists at one mid-sized German university (2001-2013). Their findings suggest that gender disadvantages in appointment outcomes are modest once publication records, time and discipline are controlled for. Hamann (2019) studies the peer review of professorial candidates through a discourse-analytic lens for 145 appointment history procedures between 1950 and 1985 at 16 universities, showing how assessment criteria shift across disciplines and how “recognition” is constructed through the appointment process. Lamont (2009) provides a broader comparative framework for understanding how evaluative cultures differ across disciplines and national contexts, analyzing how peer review panels construct academic worth. In a similar vein, Hammarfelt and Rushforth (2017) examine how bibliometric indicators function as “judgment devices” in Swedish appointment committees (through external export “sakkunniga” reports), demonstrating that non-specialist evaluators deploy citation metrics in ad-hoc and sometimes inconsistent ways to legitimate their assessments.

Gender inequality in academic appointments has received particular attention in the German context. Friebel, Fuchs-Schündeln, and Weinberger (2021) document that the share of female full professors in economics stands at only about 15%, placing Germany near the bottom in European comparison. They trace this to a “leaky pipeline” where women’s representation drops sharply at each successive career stage. Beyond economics, Ordemann and Naegele (2024) analyze gender differences in the determinants of becoming a professor in German psychology from 1980 to 2019, finding that publication output matters more for women while institutional networks play a larger role for men. Van den Brink and Benschop (2012) provide important interview-based evidence from the Netherlands on how gendered gatekeeping practices operate in professorial appointments: recruitment by invitation through homogeneous male networks restricts the candidate pool, and seemingly neutral evaluation criteria can inadvertently reproduce gender inequality. In the German context, the leaky pipeline rather than discriminatory practices of survivors in the system accounts for gender inequality in appointments outcomes (Hofmeister and Solga 2025).

An emerging strand of literature examines the role of social class and socioeconomic origin in shaping academic careers. Stansbury and Rodriguez (2024) provide large-scale evidence from US academia showing that first-generation college graduates are 13% less likely to achieve tenure at R1 institutions, earn less, and report lower job satisfaction, with the gap driven primarily by differences in social and cultural capital rather than research productivity. For the German context, Matthies and Torka (2015) compare two generations of scholars and identify distinct “habitus formations” that persist across institutional changes, suggesting that the academic system reproduces social inequalities through embedded dispositions and informal norms. Möller (2013) examines whether the German full professorship is open to social climbers and finds persistent class-based barriers to entry. Reuter et al. (2020) compile autobiographical accounts and sociobiographical analyses of professors from working-class families, documenting class-based mechanisms in hiring and career progression.

Academic mobility and inbreeding constitute another important dimension of professorial hiring. The prohibition against internal appointments (Hausberufungsverbot) is a distinctive feature of the German system intended to prevent academic inbreeding and ensure external competition. Yudkevich, Altbach, and Rumbley (2015) provide a global comparative perspective, showing that inbreeding is remarkably widespread and that while inbred faculty are not necessarily less productive, the practice fosters hierarchy and limits innovative ideas. Bauder (2020) analyzes international mobility and social capital accumulation among academics in Canada and Germany, finding that mobility can build international networks but also isolate researchers from national social capital needed for domestic appointments. This resonates with the phenomenon of “localism” documented in French recruitment, where geographic proximity to the hiring institution significantly predicts appointment success (Alamel and Marega 2023). In a comparative typoloy of countries’ academic labor market systems, Southern European countries are generally grouped as closed to foreigners, while Germany is considered “closed and insecure” (Afonso 2016): having particularly low shares of international full-professor appointments, but high shares of non-tenured employees (“Mittelbau”).

Finally, comparative studies of recrutments in other countries illuminate the specificities of the German case. French academic recruitment has been studied through the lens of geographic localism and the reproduction of inequalities in competitive recruitment procedures (concours) for “maîtres de conférences” (Angermuller 2013). In the US, Elder and Kozlowski (2025) show that male-dominated and theoretically oriented subfields concentrate in elite departments, suggesting that disciplinary stratification shapes career opportunities beyond individual merit. Betancourt, Jochem, and Otner (2023) examine the role of star scientists in shaping their co-authors’ careers, pointing to the importance of mentorship networks in academic advancement. Janger and Nowotny (2016) use stated choice experiments to identify what researchers across Europe value most in academic positions, finding that salary, colleague quality, early independence, and tenure prospects are key decision factors, which helps explain why candidates may accept or reject German professorship offers. These comparative perspectives underscore that the German appointment system, while unique in its institutional design, shares common challenges with other academic labor markets regarding the tension between meritocratic ideals and the reproduction of social and institutional hierarchies.

Data and methods

By 2025, Germany had 423 higher-education institutions, of which 182 universities (or equivalent) and 241 universities of applied sciences (BuWiK 2025). We draw on three complementary datasets that together trace the professorial appointment process from initial job announcement through candidate hearing to final appointment or rejection. Our primary novel dataset consists of hand-collected hearing lists for professorial positions at German full universities from the late 2000s to the present day. By law, hearing invitations and appointment lists must be made public, but this requirement is met in heterogeneous ways: some universities publish candidate lists as PDFs on faculty websites, others post paper announcements on seminar room doors, and still others circulate the information through social media, email lists, or student representative pages. Many university administrations archive hearing records more systematically, but with a retention period of 30 years (Hamann 2019), which prevents systematic archival research beyond single universities (Auspurg et al. 2019). Drawing on these public channels, some detective work and personal communications with participating colleagues, we assembled hearing lists for 422 positions comprising a total of 2468 invited candidates, with a focus on the social sciences and, as a comparative reference category, the humanities. We restricted the sample to hearings for which the full names of all candidates were available, as these serve as the anchor point for linking hearing-level information to online CV data, Scopus and Google Scholar records, and university reputation data derived from the Shanghai ranking. Table 1 provides an overview of the variables constructed from these sources. The resulting dataset is non-random and is driven by a publicity bias: faculties with strong interest in protecting candidates’ names or in keeping hearings below the radar are systematically excluded.

Our second background dataset draws on the digitized archive of professorial job advertisements maintained by the Deutscher Hochschulverband (DHV), which publishes monthly bulletins covering an estimated 95% of all professorial vacancies. As DHV members, we were able to access the fully digitized archive for the “sciences of society” from 2014 until end of 2025, yielding a total of 5,995 job advertisements: 3,688 in the social sciences (including sociology, political science, and geography, but excluding psychology and humanities) and 2,307 in economics (excluding temporary guest professorships offered primarily through the DAAD). The advertisements are published as PDFs with recognized text and follow a loosely standardized format, beginning with the university name, position title and denomination, application deadline, and followed by an unstructured description of the position requirements. We use this DHV job-ad dataset mainly as background and benchmark to evaluate our hearing sample’s representativeness.

The DHV shows a slightly more bottom-heavy distribution with 23.5% W1 and 42.2% W3 positions, while the hearing sample is more top-heavy with 19.8% W1 and 46.8% W3, but both are comparable in terms of position status. In terms of geographical scope, the hearing sample is mostly in line with the DHV data, with a slight over-sampling of Berlin and undersampling of NRW and Hamburg. The total number of job-ads for full professor by discipline, generously defined, is displayed in Figure 1, amounting to ca. 150 in sociology plus political science per year. This relates to 265 collected hearings in these and related social science disciplines, mostly from the last 10 years, which comprises only a small share.

Figure 1: Number of job ads by discipline

Figure 1: Number of job ads by discipline

Our third dataset is broader, by contrast, and comprises the lists of accepted and rejected professorship offers regularly published by the DHV journal “Forschung & Lehre” – in print before 2015, and online thereafter, amounting to ca. 700 per year across all disciplines. This relates to ca. 2000 annual new appointments. Each entry records the title, name, and institutional affiliation of the candidate who accepted or rejected a given offer. These data yield a relational network of university-to-university flows, allowing us to construct measures of university reputation and to examine the widespread phenomenon of candidates declining offers after being placed first (or second) on the appointment list.

To estimate the determinants of appointment success, we specify conditional logistic regression models (Chamberlain 1980), stratified by hearing, with the binary outcome variable indicating whether a candidate was ultimately appointed to the position (1) or not (0). The fundamental structure of the appointment process motivates this choice: candidates are not selected against some abstract population of academics but are evaluated relative to the specific pool of competitors assembled for a given position. A pooled logit with broad fixed effects cannot account for the many hearing-level confounders that shape appointment outcomes — the prestige of the advertised chair, the composition and preferences of the appointment committee, or whether the position called for a particular methodological or thematic profile. By conditioning on the hearing stratum, we exploit only within-hearing variation, comparing candidates who faced the same selection process and the same competitors, rendering hearing-level variables such as the share of professorial or male competitors collinear with the strata by construction. The key independent variables that vary across candidates within the same hearing — and are thus estimable — capture candidate demographics (gender, academic age, nationality), qualifications (prior professorial status, cumulative citation count from Google Scholar and Scopus, number of publications), institutional characteristics (domestic vs. foreign affiliation), and geographic factors (whether the candidate applies from the same city as the position, whether the position is located in the city of the candidate’s doctorate). Given the observational design, we interpret the estimated coefficients as conditional associations rather than causal effects.

Results

Who gets invited to German hearings in the social sciences and humanities? On average, there are almost 6 invited candidates per hearing and, subtracting the few cases without appointments, there is a 15% chance of being appointed. In about two-thirds of hearings, candidates present in German, otherwise in English. Figure 1 maps the geographic distribution of hearings across German-speaking universities. The spatial concentration of hearings in metropolitan areas – Berlin, Munich, Hamburg, and Cologne – mirrors the broader distribution of university infrastructure and faculty positions. On average (see Table 1), men are slightly over-represented among the candidates, which is driven by the W2, W3 and other status positions, whereas women are more frequently invited to W1-hearings. Women are least invited to economics and then social science hearings. While the majority of candidates does not apply with a professorship title (often printed on the hearing posters) or tenure-track position, it is noteworthy that a 44% of candidates apply from a prior professorship position, most in economics, least in the humanities. This is by definition almost non-existing for W1-hearings, but for W3-hearings almost two thirds apply with a professor title. While part of this is due to untenured W1 professors without tenure track applying elsewhere, a non-quantifiable larger share is due to the already-professors applying elsewhere to enter salary and endowment renegotiations with their home university.

Table 1: Key variables and descriptive statistics (hearing dataset)

Variable

Description

Descriptive Statistics

Gender

Gender of the candidate

man: 1305 (53.2%)

woman: 1150 (46.8%)

Professor

Already holds a professorial title

Already-professor: 1075 (43.6%)

No professor: 1393 (56.4%)

Affiliation

Domestic vs. foreign affiliation

Germany: 1809 (75.3%)

International: 592 (24.7%)

Nationality

German vs. non-German nationality

German: 1836 (79.2%)

Non-German: 482 (20.8%)

Language

Language of the presentation title

English: 805 (32.6%)

German: 1663 (67.4%)

Appointed

Whether candidate was appointed (outcome)

0: 1965 (84.7%)

1: 354 (15.3%)

Citations

Cumulative Google Scholar citations at hearing

Mean = 871.1, SD = 1950.5, Range: 0--35401

Scopus citations

Number of Scopus citations

Mean = 195.4, SD = 774.6, Range: 0--20126

This can also be seen in the age structure of applicants. Almost 40% of candidates have a publication age (since first publication) of up to 10 years, with all others falling equally into the 10-to-15 year olds and the publication seniors. This positively correlates with the W-status of the position. Economists are particularly publication-young when invited. This translates into an average of 10 years since PhD. In terms of biological age, where this is known, this implies a mean and median of 44 years, similar for men and women, foreigns and nationals, with few applicants even being in the 60s. The average is driven up by the humanities and driven down by economics.

Around 75% of candidates are affiliated with German universities (see map Figure 1), particularly in the humanities, the most frequent foreign affiliations being Swiss, Austrian and Dutch. Other countries mostly include (North-)European and Anglo countries, with very few exceptions. The share of national Germans, defined by self-report, undergrad country or native language, is slightly higher, where a number of German expats wants to repatriate from foreign universities. German nationals are more frequently found in W2 and W3 hearings. A bit more (less) than half of candidates does not have a current google scholar profile (Scopus entry), which may be due to them having dropped out of academia or being from disciplines were the citation cult is less widespread. The average google (Scopus) citations in the hearing year of 871 (195) are thus slightly upward biased. Older, higher-status, non-German male candidates outside of the humanities apply with higher citation scores.

To estimate the determinants of appointment success, we fit a logistic regression model with the binary outcome variable Appointed (1 = appointed, 0 = not appointed). Table 2 shows four different models, with each testing a different set of variables (demographic, geography, publications) before the last column presents the full model, stratified by hearing. It is noteworthy that even at the 90% confidence interval, very few regressors are statistically significant. In the first model, only applying abroad can be considered borderline significant. Applying from the same city, however, or being closer to ones PhD city, significantly increases the appointment chances, while publication differences (for those having citation information) do not make a difference. When estimating the full model (though on fewer cases), only the geographical-proximity factors remain significant.

Table 2: Condition logit regression on appointment

## 
## =======================================================================================================
##                                                                                                        
##                                        Demography       Geography       Publications        Full       
##                                           (1)              (2)              (3)              (4)       
## -------------------------------------------------------------------------------------------------------
## Publication age 10-15 years (vs. <9) -0.109 (0.179)                    0.022 (0.238)    0.028 (0.264)  
## Publication years > 15               -0.066 (0.198)                    0.047 (0.262)    0.119 (0.292)  
## Applying from abroad                 0.267 (0.176)                                      0.073 (0.264)  
## PhD abroad                           0.058 (0.191)                                     -0.273 (0.274)  
## Woman                                0.081 (0.137)                                      0.039 (0.209)  
## No professor title                   -0.091 (0.171)                                    -0.161 (0.240)  
## Distance to PhD city (log)                          -0.229*** (0.036)                 -0.201*** (0.055)
## Applying same city                                   0.516** (0.263)                   -0.200 (0.419)  
## Scholar profile                                                        -0.019 (0.236)   0.053 (0.270)  
## Log Scopus citations                                                   0.065 (0.064)    0.065 (0.073)  
## -------------------------------------------------------------------------------------------------------
## Observations                             1,724            1,792            1,112             868       
## R2                                       0.002            0.031            0.001            0.019      
## Max. Possible R2                         0.401            0.403            0.377            0.379      
## Log Likelihood                          -440.218         -432.978         -262.434        -198.573     
## Wald Test                            3.220 (df = 6) 53.410*** (df = 2) 1.240 (df = 4) 16.010* (df = 10)
## LR Test                              3.198 (df = 6) 56.929*** (df = 2) 1.234 (df = 4) 16.465* (df = 10)
## Score (Logrank) Test                 3.231 (df = 6) 61.390*** (df = 2) 1.243 (df = 4) 17.262* (df = 10)
## =======================================================================================================
## Note:                                                                       *p<0.1; **p<0.05; ***p<0.01

As a robustness check, we complement the conditional logit with multilevel logistic regression models (Appendix Table A2), specifying random intercepts nested within hearings. Unlike the conditional logit, which conditions the likelihood on hearing strata and thereby eliminates all between-hearing variation, the multilevel approach models unobserved hearing-level heterogeneity as a variance component while retaining the full sample. This specification also allows us to estimate the effect of hearing-level compositional variables, namely the share of professorial and male competitors, which are constant within hearings and therefore not identified in the conditional logit. The findings are largely confirmed.

Because geography appears to be a central dimension in the candidate selection, we draw on our third and richer appointee dataset and take a closer look at the networks of who applies from where to which city in our sample combined with our supplementary dataset on who accepts/rejects the eventual appointment. Confirming the close alignment of selected and not selected candidates, the demographics of the ones selected for gender, foreign affiliation and prior professorship (title) are largely following our non-random hearing sample, at least for the disciplines under study (cf. Figure 4). The social sciences and law amount to only 20%, economics to 6% and the humanities to ca. 17% of all new appointments since 2015. The appointee population in these disciplines is close to 50% professors who previously held a professoral title, has a majority men (particularly in economics, the social sciences) and around 20% foreign-affiliation appointments, particularly in economics, but less in the humanities and social sciences. The inclusion of “law” in the social-science category makes the category more male, domestic and prior-professorship driven. The comparison to appointees in other disciplines shows that natural-science appointees are rather men, have less prior professorships and fairly high internationalization rates.

Figure 4: Appointee demographics in the social sciences

Figure 4: Appointee demographics in the social sciences

To see how the geographic structure of German professorial appointments reveals systematic localism we present two analyses based on these appointee data. First, the average distance between a professor’s origin and destination institution is substantially shorter than would be expected under a purely random assignment of candidates to positions, with a disproportionate share of appointments involving moves within the same city entirely (Figure 4). This pattern holds across candidate types — for already-professors and first-time appointees alike, and for men and women — suggesting that geographic proximity is a structural feature of the appointment system rather than an artifact of any particular subgroup’s behavior.

Figure 5: Distance between origin and destination of accepted appointments

Figure 5: Distance between origin and destination of accepted appointments

The flow network of accepted appointments reveals a clear hub structure in the German academic labor market. As Figure X shows, most cities cluster closely around the diagonal, indicating roughly balanced import and export flows, but several systematic deviations stand out. Hamburg, Gießen, Heidelberg, Kiel, and Halle emerge as the clearest net importers in absolute terms, consistently receiving substantially more professors than they send. Berlin, despite its size, is a notable net exporter with a negative flow of 120, sending far more professors to other universities than it attracts — consistent with its role as a stepping stone rather than a final destination in academic careers. Frankfurt, Münster, Tübingen, and Bonn similarly send more than they receive. Foreign institutions collectively constitute the single largest source category, though as noted this is structurally asymmetric since German exports abroad are not recorded. Acceptance rates are uniformly high across all cities, typically above 95%, suggesting that rejection of offers is relatively rare and does not drive the imbalance between importers and exporters.

When flows are adjusted for city size using location quotients, the picture shifts substantially toward smaller university towns. Heidelberg, Jena, and Kiel emerge as the strongest over-attractors relative to population, while Göttingen, Tübingen, Marburg, and Bayreuth — despite their traditional academic prestige — are actually net over-senders once size is accounted for, meaning they produce more outgoing professors than their population share would predict. Hamburg moves close to the population baseline after adjustment (LQ = 1.11), and Berlin, though not in the size-adjusted top 20, remains a net exporter in absolute terms. This suggests that the prestige advantage in German academic appointments accrues most clearly to mid-sized university towns like Heidelberg, Jena, and Kiel rather than to the largest metropolitan centers — a pattern consistent with the decentralized, historically rooted character of the German university system.

Figure 5: Hub universities in German professorial appointments

Figure 5: Hub universities in German professorial appointments

The regression evidence that geographic proximity predicts appointment success is corroborated at the aggregate level by the structure of appointment flows: moves are systematically shorter than chance would predict, and a small number of hub institutions dominate both as senders and receivers of professorial talent.

Discussion

The results can be interpreted differently. They are in line with Auspurg et al. (2019), who also found that candidates placed (first) on the list hardly differ from others in terms of gender, publications, etc. They are largely consistent with the hypothesis that the pool of hearing invitees is sufficiently pre-selected to render most observable candidate characteristics non-predictive of appointment success. As Table 2 shows, gender, prior professorial status, domestic versus foreign affiliation, nationality, and the language of the presentation do not significantly predict whether a candidate is appointed. The estimated odds ratios for these variables hover near 1, with confidence intervals spanning the null value in each case. The only variable that approaches or reaches conventional significance thresholds is geographic proximity: candidates who apply from the same city as the advertising university are more likely to be appointed. This pattern of academic localism resonates with comparable findings from the French context (Alamel 2023) and may reflect either network advantages – candidates in the same city are more likely to have prior personal acquaintance with committee members – or lower perceived risk for committees selecting a locally known quantity.

The overall picture is one of a remarkably level playing field among hearing invitees. This does not mean that meritocratic factors are irrelevant to professorial appointments; rather, it suggests that these factors have already done their sorting work at earlier stages of the process: in the decision of whom to invite to the hearing in the first place. By the time candidates reach the hearing stage, they constitute a sufficiently narrow and homogeneous pool that the measurable characteristics captured in our data no longer differentiate among them. The decisive factors at this stage are likely those that are hardest to measure: the quality and fit of the hearing presentation itself, thematic alignment with the advertised position, and the interpersonal dynamics within the appointment committee.

The geographic and institutional structure of German professorial appointment flows connects to a broader literature on hiring networks in academia, while differing from it in important respects. Studies of PhD placement networks in the United States — most prominently Clauset et al. (2015) and subsequent work — have documented steep prestige hierarchies in which a small number of elite departments produce the overwhelming majority of faculty at research universities, with placement success strongly predicted by the prestige rank of the doctoral institution. The German pattern revealed here is both similar and distinct. Similar, in that appointment flows are far from random: a small number of cities systematically over-attract relative to their size, and the system as a whole is characterized by considerable geographic concentration. Distinct, however, in at least two respects. First, because more than half of the candidates in our data already hold a prior professorship, the relevant hierarchy is not one of doctoral training but of mid-career repositioning — professors moving laterally or upward within an established chair system rather than freshly minted PhDs entering the market for the first time. The prestige logic operating here is therefore closer to what Musselin (2010) describes as the “market for academics” governed by reputation accumulated over a career, rather than the credentialing logic of PhD placement. Second, the cities that emerge as over-attractors after size adjustment — Heidelberg, Jena, Kiel — are not necessarily the largest or most research-intensive universities by conventional metrics, which diverges from the US pattern where elite research universities dominate placement. This may reflect the more decentralized and historically path-dependent character of German academic prestige, where traditional university towns retain reputational capital that is not easily reducible to contemporary ranking indicators. The finding that Berlin functions as a net exporter is particularly striking in this regard: despite its concentration of universities and research institutions, it appears to serve more as a launching pad than a destination, consistent with the observation that large metropolitan locations may offer attractive outside options — think tanks, policy organizations, international institutions — that draw academics away from the university system rather than anchoring them within it. Finally, the geographic localism documented in our regression results — that applying from the same city increases appointment chances — finds its aggregate-level counterpart in the short-distance distribution of appointment flows, resonating with Alamel and Marega’s (2023) findings on geographic localism in French academic recruitment and suggesting that such patterns may be a general feature of European academic labor markets rather than a French peculiarity.

Appendix

Table A1: Proportional Distribution of Professors across German Bundesländer – DHV Sample vs. Hearing Sample

Bundesland DHV Sample Hearing Sample Difference
Baden-Württemberg 14.0% 14.2% +0.2 pp
Bayern 13.1% 14.2% +1.1 pp
Berlin 6.2% 20.0% +13.8 pp
Brandenburg 3.5% 1.6% -1.9 pp
Bremen 1.5% 1.6% +0.1 pp
Hamburg 3.3% 0.8% -2.5 pp
Hessen 7.2% 9.3% +2.1 pp
Mecklenburg-Vorpommern 1.6% 3.0% +1.4 pp
Niedersachsen 10.0% 6.0% -4.0 pp
Nordrhein-Westfalen 21.7% 19.7% -2.0 pp
Rheinland-Pfalz 4.6% 5.5% +0.9 pp
Saarland 0.6% 0.5% -0.1 pp
Sachsen 5.5% 4.7% -0.8 pp
Sachsen-Anhalt 2.3% 1.1% -1.2 pp
Schleswig-Holstein 2.7% 2.2% -0.5 pp
Thüringen 2.1% 2.2% +0.1 pp

Note: Hearing sample includes only German cities; foreign cities excluded. Percentages may not sum to 100% due to rounding.

Table A2: Multilevel-model on appointment with candidates nested in hearings

## 
## ==============================================================================================================================
##                                                                                                                               
##                                         Demography         Education         Geography       Publications          Full       
##                                             (1)               (2)               (3)               (4)               (5)       
## ------------------------------------------------------------------------------------------------------------------------------
## Publication age 10-15 years (vs. <9)  -0.157 (0.154)                                        -0.049 (0.202)    -0.056 (0.226)  
## Publication years > 15                -0.139 (0.153)                                        -0.104 (0.208)    -0.026 (0.236)  
## Applying from abroad                  0.287* (0.164)                                                           0.102 (0.231)  
## PhD abroad                             0.079 (0.174)                                                          -0.249 (0.247)  
## Woman                                  0.138 (0.128)                                                           0.028 (0.181)  
## No professor title                                      -0.066 (0.173)                                         0.057 (0.192)  
## % professorial competitors                              -0.002 (0.002)                                                        
## % men competitors                                       -0.003 (0.002)                                                        
## Distance to PhD city (log)                                               -0.195*** (0.030)                   -0.151*** (0.044)
## Applying same city                                                        0.479** (0.221)                     -0.059 (0.340)  
## Scholar profile                                                                             -0.032 (0.189)     0.015 (0.209)  
## Log Scopus citations                                                                         0.018 (0.044)    -0.020 (0.051)  
## Constant                             -1.663*** (0.202) -1.465*** (0.218) -0.630*** (0.167) -1.636*** (0.175)  -0.532 (0.452)  
## ------------------------------------------------------------------------------------------------------------------------------
## Observations                               1,724             2,319             1,792             1,112              868       
## Log Likelihood                           -795.617          -989.635          -792.597          -492.182          -403.739     
## Akaike Inf. Crit.                        1,605.234         1,989.270         1,593.193          996.365           831.479     
## Bayesian Inf. Crit.                      1,643.401         2,018.014         1,615.158         1,026.448          888.673     
## ==============================================================================================================================
## Note:                                                                                              *p<0.1; **p<0.05; ***p<0.01

References

Alamel, Alexis, and Oumar Marega. 2023. “Une géographie du recrutement des maîtres de conférences en France: trajectoires des candidats et spatialisation des recrutements.” Annales de géographie 749–750: 90–123.

Afonso, Alexandre. 2016. “Varieties of Academic Labor Markets in Europe.” PS: Political Science & Politics 49 (4): 816–821.

Angermuller, Johannes. 2013. “How to become a philosopher: Academic discourse as a multilevelled positioning practice.” Sociología histórica 3: 263–289.

Auspurg, Katrin, Thomas Hinz, and Andreas Schneck. 2017. “Berufungsverfahren als Turniere: Berufungschancen von Wissenschaftlerinnen und Wissenschaftlern.” Zeitschrift für Soziologie 46 (4): 283–302.

Bauder, Harald. 2020. “International mobility and social capital in the academic field.” Minerva 58 (4): 597–616.

Betancourt, Nathan, Torsten Jochem, and Sarah Otner. 2023. “Standing on the shoulders of giants: How star scientists influence their coauthors.” Research Policy 52 (1): 104624.

BuWiK. 2025. Bundesbericht Wissenschaftlerinnen und Wissenschaftler in einer frühen Karrierephase (BuWiK) 2025. Bielefeld: wbv Media.

Chamberlain, Gary. 1980. “Analysis of Covariance with Qualitative Data.” Review of Economic Studies 47 (1): 225–238.

Clauset, Aaron, Samuel Arbesman, and Daniel B. Larremore. 2015. “Systematic inequality and hierarchy in faculty hiring networks.” Science Advances 1 (1): e1400005.

Elder, Timothy B., and Austin C. Kozlowski. 2025. “Status and subfield: The distribution of sociological specializations across departments.” Poetics 109: 101973.

Friebel, Guido, Nicola Fuchs-Schündeln, and Alisa Weinberger. 2021. “Statusbericht zum Frauenanteil in der Volkswirtschaftslehre an deutschen Universitäten.” Perspektiven der Wirtschaftspolitik 22 (3): 237–250.

Janger, Jürgen, and Klaus Nowotny. 2016. “Job choice in academia.” Research Policy 45 (8): 1672–1683.

Hamann, Julian. 2019. “The making of professors: Assessment and recognition in academic recruitment.” Social Studies of Science 49 (6): 919–941.

Hammarfelt, Björn, and Alexander D. Rushforth. 2017. “Indicators as judgment devices: An empirical study of citizen bibliometrics in research evaluation.” Research Evaluation 26 (3): 169–180.

Hofmeister, Sophie, and Heike Solga. 2025. “Bleiben oder Gehen? Eine fächerspezifische Analyse von Geschlechterunterschieden im Anstreben einer Professur an deutschen Hochschulen.” WZB Discussion Paper SP I 2025-501. Berlin: Wissenschaftszentrum Berlin für Sozialforschung.

Hüsch, Marc. 2025. “Demographische Entwicklung bei Professorinnen in Deutschland.” DatenCHECK* 2/2025. https://hochschuldaten.che.de/demographische-entwicklung-bei-professoren/.

Jungbauer-Gans, Monika, and Christiane Gross. 2013. “Determinants of success in university careers: Findings from the German academic labor market.” Zeitschrift für Soziologie 42 (1): 74–92.

Lamont, Michèle. 2009. How Professors Think: Inside the Curious World of Academic Judgment. Cambridge, MA: Harvard University Press.

Lutter, Mark, and Martin Schröder. 2016. “Who becomes a tenured professor, and why? Panel data evidence from German sociology, 1980–2013.” Research Policy 45 (5): 999–1013.

Matthies, Hildegard, and Marc Torka. 2015. “Academic habitus and institutional change: Comparing two generations of German scholars.” Minerva 53 (3): 245–268.

Musselin, Christine. 2010. The Market for Academics. New York: Routledge.

Möller, Christina. 2013. “Wie offen ist die Universitätsprofessur für soziale Aufsteigerinnen und Aufsteiger?” Soziale Welt 64 (4): 341–360.

Naegele, Laura, and Jessica Ordemann. 2023. “Zu alt für die Wissenschaft? Wissenschaftliche Karrierewege von älteren Promovierten in die Entfristung.” In Polarisierte Welten: Verhandlungen des 41. Kongresses der Deutschen Gesellschaft für Soziologie in Bielefeld 2022, edited by Paula-Irene Villa Braslavsky, 1–11. Bielefeld: Deutsche Gesellschaft für Soziologie.

Ordemann, Jessica, and Laura Naegele. 2024. “Forty and over the academic hill? Biological and academic age and the race for tenure.” Soziale Welt Sonderband 26: 457–489.

Püttmann, Vitus, Markus B. Siewert, and Christiane Frantz. 2022. “How human capital, universities of excellence, third party funding, mobility and gender explain productivity in German political science.” Scientometrics 127: 4223–4249.

Schimank, Uwe. 2005. “‘New public management’ and the academic profession: Reflections on the German situation.” Minerva 43 (4): 361–376.

Schröder, Martin, Mark Lutter, and Isabel M. Habicht. 2021. “Publishing, signaling, social capital, and gender: Determinants of becoming a tenured professor in German political science.” PLoS ONE 16 (1): e0243514.

Schröder, Martin, Isabel M. Habicht, and Mark Lutter. 2024. “What leads to a professorship in German economics? A longitudinal analysis of tenure determinants (1984–2021).” Studies in Higher Education 50 (1): 1–18.

Reuter, Julia, Markus Gamper, Christina Möller, and Frerk Blome, eds. 2020. Vom Arbeiterkind zur Professur: Sozialer Aufstieg in der Wissenschaft. Autobiographische Notizen und soziobiographische Analysen. Bielefeld: Transcript.

Stansbury, Anna, and Kyra Rodriguez. 2024. “The class gap in career progression: Evidence from US academia.” Working Paper, MIT Sloan School of Management.

Van den Brink, Marieke, and Yvonne Benschop. 2012. “Gender practices in the construction of academic excellence: Sheep with five legs.” Organization 19 (4): 507–524.

Wissenschaftsrat. 2005. Empfehlungen zur Ausgestaltung von Berufungsverfahren. Drs. 6709-05. Cologne: Wissenschaftsrat.

Yudkevich, Maria, Philip G. Altbach, and Laura E. Rumbley, eds. 2015. Academic Inbreeding and Mobility in Higher Education: Global Perspectives. Basingstoke: Palgrave Macmillan.

Zimmer, Lena M. 2018. Das Kapital der Juniorprofessur: Einflussfaktoren bei der Berufung von der Junior- auf die Lebenszeitprofessur. Wiesbaden: Springer VS.

  1. See the Wiki Auditions for economics (https://wikiauditionseco.fr), sociology (http://sociologuesdusuperieur.org), political science (http://ancmsp.com), and history (https://afhe.hypotheses.org).↩︎