Hanna Hottenrott, Cornelia Lawson · Hanna Hottenrotta,b c,dand Cornelia Lawson aDüsseldorf...
38
No 153 Flying the Nest: How the Home Department Shapes Researchers’ Career Paths Hanna Hottenrott, Cornelia Lawson July 2014
Transcript of Hanna Hottenrott, Cornelia Lawson · Hanna Hottenrotta,b c,dand Cornelia Lawson aDüsseldorf...
No 153
Flying the Nest: How the Home Department Shapes Researchers’ Career Paths Hanna Hottenrott, Cornelia Lawson
July 2014
IMPRINT DICE DISCUSSION PAPER Published by düsseldorf university press (dup) on behalf of Heinrich‐Heine‐Universität Düsseldorf, Faculty of Economics, Düsseldorf Institute for Competition Economics (DICE), Universitätsstraße 1, 40225 Düsseldorf, Germany www.dice.hhu.de
Editor: Prof. Dr. Hans‐Theo Normann Düsseldorf Institute for Competition Economics (DICE) Phone: +49(0) 211‐81‐15125, e‐mail: [email protected] DICE DISCUSSION PAPER All rights reserved. Düsseldorf, Germany, 2014 ISSN 2190‐9938 (online) – ISBN 978‐3‐86304‐152‐6 The working papers published in the Series constitute work in progress circulated to stimulate discussion and critical comments. Views expressed represent exclusively the authors’ own opinions and do not necessarily reflect those of the editor.
Flying the Nest: How the Home Department Shapes
Researchers’ Career Paths
Hanna Hottenrotta,b and Cornelia Lawsonc,d
aDüsseldorf Institute for Competition Economics (DICE), University of Düsseldorf, Germany bCentre for European Economic Research (ZEW), Mannheim, Germany
c Department of Sociology and Social Policy, University of Nottingham, UK d BRICK, Collegio Carlo Alberto, Moncalieri (Turin), Italy
July 2014
Academic researchers face mobility related decisions throughout their careers. We study the
importance of team and organisational characteristics of the home departments for career
choices of departing researchers in the fields of science and engineering at higher education
institutions in Germany. We find that the organisational environments – the nests – shape career
paths. Research funding, research performance in terms of patents and publications as well as
the industry ties of department heads shape job choices. In particular, public research grants
increase the probability that departing researchers take a research job at a university or public
research centre, while grants from industry increase the likelihood that they take a job in
industry. Publication performance of the department head relates to R&D jobs in public, but
not in industry and patents predict the probability that departing researchers will move to small
and medium-sized firms. For these firms seeking technological knowledge from former
university employees may be particularly crucial. Academic start-ups are more likely to be a
job destination for departing researchers from technical universities, from departments with
higher publication output and with a research focus on experimental development.
Keywords: Researcher Mobility, Research Funding, Science-Industry Technology Transfer,
Academic Entrepreneurship, Academic Careers
JEL codes: I23; J24; O3
Acknowledgements
We thank the Centre for European Economic Research (ZEW) for providing the survey data and Susanne
Thorwarth for help with the collection of publication and patent data. We thank participants at the “The
Organisation, Economics and Policy of Scientific Research” workshop organised by LEI & BRICK, Collegio
Carlo Alberto, Torino (Italy) and the “Beyond spillovers? Channels and effects of knowledge transfer from
universities” workshop at the University of Kassel (Germany) for helpful comments. Cornelia Lawson
acknowledges financial support from the Collegio Carlo Alberto Project ‘Researcher Mobility and Scientific
Performance’.
Hanna Hottenrott (corresponding author), Düsseldorf Institute for Competition Economics (DICE), Heinrich
Heine University Düsseldorf, Universitätsstrasse 1, 40225 Düsseldorf, Germany; Phone: +49 211 81-10266, Fax:
+ 49 211 81-15499; Email: [email protected]
Cornelia Lawson, School of Sociology and Social Policy, University of Nottingham, University Park, Nottingham,
NG7 2RD, UK; Email: [email protected]
1
1. Introduction
Academic researchers face mobility related decision points throughout their careers. The
first important decision point, and the one most often considered in academic literature, is the
completion of doctoral education that leaves researchers to choose their future career paths.
Many young researchers choose to remain in public research but might leave their home
institution and take up jobs at a public research institution or a university elsewhere. Especially
in the US system mobility is encouraged following the PhD and also in Germany, young
academics are usually expected (and have in the past even been required) to leave their home
department following habilitation. For the US it has further been shown that a large share of
young researchers will not remain in academia as more researchers aspire to academia than
positions are available (Fox and Stephan, 2001). Especially in science and engineering, many
PhD holders and post-doctoral researchers leave academia and move to an R&D career in
industry. The same is true for Germany, where the number of faculty positions for full
professors is much lower than that of qualified post-doctoral researchers and lecturers.
Thus, academia is very competitive and researchers may need to withdraw from academic
research and move to a job in industry or public research or a non-research related job. Several
studies on the destination of PhD holders in the US indeed showed that many leave academia
for industry. Of the life science researchers that completed their PhD in 1985 and 1986, only
38% were in tenure stream positions 10 years later, while 24% had taken up a position in
industry (Austin, 2002). This share is even higher for PhD holders in other disciplines. About
34% of physicists and 46% of chemists were employed by industry five to six years after
obtaining their PhD (Stephan, 2012).
Yet, few papers have investigated the factors driving the destination choices of academics.
Studies on the mobility from academia to industry have primarily focussed on academic
entrepreneurship of senior academics (e.g. Audretsch and Stephan, 1999; Stuart and Ding,
2006; Toole and Czarnitzki, 2010), while studies on the mobility between academic institutions
have focussed on international movements (e.g. Franzoni et al., 2012; Stephan, 2012). Recent
interest in career choices of young researchers has resulted in a series of studies on the choice
between a career industry and academia (Roach and Sauermann, 2010; Agarwal and Ohyama,
2013; Balsmeier and Pellens, 2014; Lam and Campos 2014).
None of these studies, however, explicitly considers the organisational environment in
which these career decisions are made. Most research happens in teams, which underlines the
2
importance of organisational characteristics of the home departments – the nests – for career
choices of departing researchers. Carayol and Matt (2004) have previously shown the
importance of organisational characteristics for knowledge creation. In addition to widening the
understanding of the processes behind knowledge production, research groups also shape the
careers of their members (Walsh and Lee, 2013).
This study adds to previous research by studying the outflow of researchers from 676
science and engineering research units at 46 universities in Germany. Mobility and career
decision points in Germany are very common and can still occur relatively late. The average
age of taking up a position as tenured full professor is 42 years (Schultze et al. 2008). This is
much later than in other academic markets and researchers can be seen to still drop out of
academia or to move abroad at a relatively late age. Also, the average rate of mobility is much
higher in Germany compared to other countries, with more than 50% of PhD holders having
moved at least once over a ten year period (Auriol et al. 2013).
Based on survey data from research units we analyse the factors that drive job destinations
of departing researchers during the years 1997-1999. We differentiate between non-research
and research-related jobs in newly formed firms, small and medium-sized firms (SMEs), large
firms, consulting companies, public research institutes and universities. We link the career
destinations to a broad set of home department characteristics ranging from publications and
patents, research unit composition in terms of employment and research focus, to funding of
researchers. Research funding from public or private sector in the form of research grants that
complement the research unit’s core funding has become increasingly important for German
universities (Hottenrott and Thorwarth 2011) and the sources from which researchers are being
funded may also determine their career paths (Lam and Campos 2014).
In line with earlier research on US researchers we find that by far not all research units
produce researchers solely for academic careers. For our set of research units, we observe that
only 6% saw all their departing employees move to purely academic jobs, while 32% saw all
their departing researchers move to industry. The majority of research units, however, trained
people for academe as well as industry. Results from the estimation of simultaneous equation
models underline the important role of home departments (nests) for the career building of
researchers. We show that the majority of research units see their departing researchers take on
R&D-related jobs in industry. This confirms the important role of former university employees
for industrial R&D. We also find that researchers trained in research units that have more links
3
to industry and receive more grants from industry are also more likely to move to a job in the
private sector. Conversely, researchers trained in research units that receive more public
funding, have more publications and have a more basic research orientation are more likely to
take up employment in the public sector.
The following section summarizes the literature on career choices of academic researchers
and presents our hypotheses regarding the role of team and research head characteristics that
impact career choices. Section 3 describes the data and section 4 sets out the econometric
framework and presents the results. Section 5 concludes.
2. Mobility and career decisions of academics: background and hypotheses
Movement of scientists and hence scientific knowledge between different academic
institutions, and between university and private institutions is believed to be vital to facilitate
research and innovation in both the private and public sector. It has been stated that researcher
mobility supports knowledge and technology transfer, the creation of networks and productivity
(OECD 2000, 2008). All these assumed positive effects of mobility are related to the embedded
character of researchers’ human and social capital (Granovetter 1985; Griliches 1973) which
can spread and increase through mobility (Schultz 1961; Becker 1962; Nelson & Phelps 1966;
Bourdieu 1986; Coleman 1988; Burt 1997). The movement of scientists aids the diffusion of
ideas across institutions and can result in increased knowledge flows (Azoulay et al., 2011).
Specifically the movement of PhD holders into firms presents an opportunity for knowledge
transfer (Mangematin 2000, Zellner 2003, Enders 2005) and the goal of improving knowledge
and technology transfer with industry has created an increased interest amongst policy makers
in researcher mobility between the public and private sector. At the same time, previous
literature has mourned the potential brain drain from academia to industry that may impede
knowledge production (Aghion et al., 2008; Auriol et al., 2013).
2.1 Individual factors
Industry vs. Academia
These developments in researcher mobility and the choice between careers in academia and
industry have gained increasing scholarly attention in recent years. The decision to remain in
academia or to move to industry in this context has been linked to a scientist’s research ability
and job preferences (David and Dasgupta, 1994; Stern, 2004; Stephan, 2012). Scientists differ
in their ability to produce new knowledge and not all may have the capacity for remaining in
4
an academic research environment but may instead prefer to move into industry or to a non-
research position. Additionally, scientists assign different levels of importance to monetary and
non-monetary rewards associated to career choices (Levin and Stephan, 1991). The importance
of non-pecuniary factors has been described as ‘taste for science’ that is independent of ability
or salary concerns (Levin and Stephan, 1991; Stephan, 2012). It has been measured as an
intrinsic preference for freedom of research that is accommodated better in an academic than
an industrial work environment (Sauermann and Stephan, 2013). Accordingly, Roach and
Sauermann (2010) and Agarwal and Ohyama (2013) show that young researchers with a higher
‘taste for science’ are more likely to remain in academia. In contrast, researchers that value
pecuniary gains, in terms of salary and access to equipment and funding, and are more interested
in downstream research, are more likely to move to industry (Roach and Sauermann, 2010;
Balsmeier and Pellens, 2014).
Stern (2004) showed that the ‘taste for science’ is strongly correlated to research ability in
terms of publications (see also Sauermann and Roach, 2012). Researchers that value publishing
are more likely to look for a career in academia (Roach and Sauermann, 2010). Similarly, we
could expect that patenting numbers correlate positively with valuing pecuniary benefits
(Owen-Smith and Powell, 2001; Sauermann and Roach, 2012). 1 This ‘taste for
commercialisation’ is associated with favouring a career in industry (Roach and Sauermann,
2010). Balsmeier and Pellens (2014) indeed find for a sample of Flemish researchers that
publication numbers negatively affect the propensity to move to industry, while patent numbers
have a positive effect. Mangematin (2000) confirm that publication numbers increase the
chance of recruitment in academia; and Crespi et al. (2007), who look at the sector mobility of
academic inventors in the European context find that academics with more valuable inventions
are more likely to move to industry. Studies on sector mobility of senior academics instead find
that publication performance has a positive effect on labour mobility to industry (Zucker et al.,
2002; Toole and Czarnitzki, 2010). Thus, publications may not truly represent career
preferences but can indicate career opportunities.
The home department (nest) likely plays a vital role in shaping researchers’ tastes and thus
career preferences. Advisors and supervisors often have a fundamental impact on research focus
and research content. Their skills and experiences may be directly transferred to the young
1
Academic involvement in commercial activities correlates positively with both, pecuniary and non-pecuniary
benefits (Lam, 2011) and several papers have reported a positive link between patents and publications.
5
scientists with whom they work. Likewise, may a research unit’s professional network shape
career options of departing researchers. This is discussed in detail in section 2.2.
R&D vs. non-R&D jobs
While individual researchers may have a preference for a job in academe or industry, not
all positions in industry are alike. When selecting a job in industry, the departing scientists have
the choice between joining an established firm and starting their own company. Roach and
Sauermann (2010) show that PhD students with an intention to leave academe have a lower
‘taste for science’ and place higher value on pecuniary benefits. In later work Sauermann and
Roach (2012) qualify these findings and show that PhD students who plan to start their own
firm have a taste for both science and commercialisation, while those joining established firms
only value pecuniary benefits. This is also in line with the positive effect of publications and
patents on start-up formation found in the literature on academic entrepreneurship (e.g. Stuart
and Ding, 2006).
Also, differences in workplace characteristics within established firms may attract different
types of scientists. Organisational theory suggests that firms become more bureaucratic as they
grow in size which may result in less autonomy for scientists. They may, however, offer higher
wages and may be better equipped for supporting top research. Sauermann and Stephan (2013)
confirm that scientists at larger firms are less satisfied with their research autonomy but that
they receive higher salaries. Thus, scientists with a higher ‘taste for science’ may be better
advised to join small firms; scientists with a taste for pecuniary returns instead, should join
large firms. Also within academia, job heterogeneity plays a role. Public research institutions
offer research positions without teaching duties that may attract the most able researchers within
the field. They also offer a higher level of autonomy than universities and are more likely to
reward commercial research lines.
2.2 Nest factors
While the individual researcher holds a specific tendency to favour a job in academia or
industry, this is affected by the organisational environment within which he or she works.
Socialisation processes of academia inform scientists’ attitudes towards various pecuniary and
non-pecuniary gains associated with research. Especially the head of the research unit has a
very formative influence on the values and perceived opportunities of his or her research staff.
6
The research unit also has a strong effect on individual ability. Several papers have shown
that a high quality research environment has a positive effect on individual research
performance (Hall et al. 2007; Waldinger, 2012). Having more full professors, having a larger
percentage of department faculty working on research and having more ‘star professors’ all
contribute to enhanced individual research productivity (Smeby and Try, 2005). Researchers
may moreover gain early citation advantage if co-authored by a scientist with a high reputation
in the scientific community (Petersen et al., 2013).
Researchers trained as PhDs or postdocs in high quality departments will thus have a
publication and reputation advantage and may value academic career paths more than
researchers at other departments. Sauermann and Roach (2014) indeed find that those from
highly ranked PhD programme give higher importance to publishing, which (as seen above)
results in a higher probability to remain in academia.
H1: Research units with higher academic publication performance are more likely to see their
departing researchers take jobs at universities and public research institutions.
Research heads also shape the networks that are crucial for career advancement in academe,
but also beyond. Several papers have considered the prestige of PhD granting and hiring
institutions to study early career advancement. Most of these studies find that the prestige of
the university is more relevant for obtaining an academic position than the level of individual
productivity (Crane, 1965, 1970; Long, 1978; Baldi, 1995). Burris (2004) links this to elite
networks that have developed between top universities. Also in Europe the importance of social
ties for promotion within academia is very high and may impact mobility decisions (Pezzoni et
al., 2012; Zinovyeva and Bagues, 2014). These social ties may be observed through a research
unit’s greater access to prestigious public grants which may be indicative of its research quality
as well as standing within the research community.
H2: Research units with a higher share of public grants are more likely to see their departing
researchers stay in academe.
Also for research careers in industry, social ties are important. Several papers have shown
that academics in departments with links to industry and high levels of commercial activity are
more likely to engage with industry or be entrepreneurial themselves (Bercovitz and Feldman,
2008; Lawson, 2013; Aschhoff and Grimpe, 2014). This is confirmed by Sauermann and Roach
7
(2012) who find that those with commercially oriented advisors are more likely to value
pecuniary benefits, which in turn affect career choices and opportunities.
H3: Research units with higher patenting activity (also in terms of patents’ relevance for
industrial applications) are more likely to see their departing researchers leave for an R&D
job in industry or start their own business.
Lam (2007) further shows that large firms establish close links with universities, manifested
through hiring of PhDs and postdocs to engage scientists in joint knowledge production. In
doing so, firms establish closer knit networks across institutional boundaries that allow them to
directly tap on knowledge in these departments and influence their teaching and research
agendas. Mangematin (2000) finds for French PhD graduates that the collaboration with a
private-sector partner during the PhD-phase increases the probability of obtaining a position in
the private sector. He also stresses that networks created during the PhD are crucial for finding
a job. These networks are largely influenced by the head of the research unit and industry-ties
may be visible through collaboration with firms (contract research, joint publishing or
patenting). In particular, research funding from industry may constitute a channel through
which ties are maintained.
H4: Research units with stronger industry ties are more likely to see their departing researchers
take R&D jobs in industry, especially in large firms.
Moreover, some research lines may be better suited to meet the need of private firms.
Research units that pursue a higher share of applied research or experimental development may
be more likely to train researchers for industry than departments that focus on basic research
lines.
H5: Research units with focus on applied rather than basic research are more likely to see their
departing researchers start their own business or leave for an R&D job in industry.
3. Data
The empirical analysis is based on a unique dataset that was assembled using different data
sources. The core data were collected through a survey of research units at German higher
education institutions in the fields of science and engineering. The Centre for European
Economic Research (ZEW, Mannheim) conducted the survey of 3,507 research units in 2000
8
and the sampling method involved stratification by regions. The original sample included public
research institutions as well as universities, technical universities and universities of applied
sciences (Fachhochschulen). For the purpose of this study we exclude respondents from public
research institutions and focus on faculties at institutes of higher education, i.e. institutions that
also offer teaching. The questionnaire was addressed to the head of a research unit, who is
usually a full professor with budget and personnel responsibility. The overall response rate to
the survey was 24.4%. The survey data were complemented with publication and patent
information of the head of the research unit covering the five year period before the survey
(1994-1999). Data on publications were collected manually from the ISI web of science and
patent information from the German Patent Office in 2008. We also collected citations to
publications and patents with a citation window ending in 2008. Publication and patent data
were manually matched to survey respondents based on names and information collected from
university websites and the researchers’ CVs. We also collected information on the year of
doctoral degree for each research unit head to derive a measure for the academic age and
experience of the professor. The German National Library collects information on all doctoral
theses submitted to German universities. For professors with no PhD degree, who can teach at
Universities of Applied Sciences only, or professors that received their degree outside Germany,
we use the year of the first publication to measure the start of their academic career. The final
sample comprises 676 professor-research unit observations from 46 different higher education
institutions of which 56% are Universities (Uni), 23% are Technical Universities (TUs) and
21% are Universities of Applied Sciences (UAS). For each of the 16 German States (Länder),
the sample contains at least one observation. The survey data is further complemented by
university and regional characteristics that are likely to affect the job choice of departing
researchers. These include the number of new firms in the region, the purchasing power of the
regional population as well industry performance in the area. The indicators serve to proxy for
the outside options of the departing researchers.
3.1 Nest characteristics
The unit of analysis is the department, research unit or chair. Teaching and research at German
universities has traditionally been organized around “chairs” with one professor heading a
research group specialized in a certain sub-field. The number of researchers employed at such
chairs may differ substantially. Some may be labelled research unit or department when they
employ several post-doctoral researchers or more than one professor.
9
Table 1 summarized the main characteristics for the sample of research units that provided full
answers to the survey and for which patent and publication information could be
unambiguously identified.2 The average institution size as measured by the number of students
was about 18 thousand during the year of the survey. Since this distribution is skewed with few
large institutions and many smaller ones, we use the log of the student count in the econometric
analysis. At the level of the research unit, on average, eight researchers (not including the head)
were employed. The mean share of technical staff over all employees was 10 per cent. As could
be expected given that only department or research unit heads were surveyed, these researchers
are very experienced with an average of 22 years since completing their doctoral degree. With
regard to gender, we see that only about three per cent of research unit heads are female.
The surveyed research unit heads were also asked to indicate the importance of several factors
for their unit’s links with industry using a scale from zero to three (0 = not important to 3 =
very important). The first factor related to the head’s former jobs in industry, the second to the
relevance of contract research and the third to joint research with firms. Of these three
categories the first two were of higher importance, on average, than the latter. In terms of
research orientation, the research unit heads indicated the time the unit usually spent on basic
research, applied research and experimental development. In our sample, research units spent
on average about 42 per cent of their time on basic research, about 41 per cent on applied
research and less than 18 per cent on experimental development. By multiplying these time
shares with the number of researchers at the unit we derive indicators for the relative work force
attributed to each type of activity. Research orientation may also be reflected in the publication
and patent record of the unit heads. For the 5-year pre-sample period from 1995 to 1999 we
observe an average count of 11 publications and 1.4 patents. The aggregate number of citations
to journal articles published in that period until 2008 is about 237 at the mean, but the median
is much lower with about 24 citations. Patents got cited during the same period about 20 times
on average (the median is zero and the 75th percentile is just two citations). Thus, as is common
for these types of measures, the distribution of publications and patents is highly skewed.
Therefore, we take logarithms (+1) of these counts in the regression models. We further
2 In few cases for very common names like “Hans Müller” we could not unambiguously identify publications
and patent applications even when addresses, CV information, and research field were taken into account. We
dropped these observations from the sample.
10
calculate at the professor level the average number of citations per publication and patents and
include these variables as quality-weighted publication and patent indicators.
The survey further provides information on the amount and composition of research grants that
complemented a unit’s institutional core funding. We differentiate between grants from public
sources, e.g. the German Research Foundation (DFG) and the federal state governments, and
income generated from industry sources. 61% of surveyed professors stated that their unit had
received funding from industry and 78% had acquired public research grants in addition to their
core institutional funding. The amount of industry funding and its share over the total budget
differ between institution types and research fields (see Table A.2). On average the share was
8.6% amounting to approximately 98 thousand Euros. The share of research grants received
from public sources is similar for universities and technical universities, but is considerably
lower at universities of applied sciences. On average, research units received 21.7% of their
total budget from public research grants, which corresponds to about 127 thousand Euros.
Table 1: Summary statistics (n = 676)
Variable description Variable name Mean Std. Dev. Min Max
Institution and research units
Institution size (total # of students) STUDENTS 17,913.4 11,850.79 1,451 59,599
Number of researchers LABSIZE 7.573 9.537 0 71
Share of technical staff (in % of
total) TECHS 9.943 13.710 0 80
Number of years since PhD EXPERIENCE 21.869 8.720 1 43
University UNI 0.558 0.497 0 1
Technical University TU 0.232 0.423 0 1
University of Applied Sciences UAS 0.210 0.408 0 1
Gender of unit head FEMALE 0.033 0.178 0 1
Head’s former job in industry FORMER_JOB 1.371 1.184 0 3
Contract research for industry CONTRACT_RESEARCH 1.348 1.116 0 3
Joint patenting/publishing with
industry JOINT_RESEARCH 0.851 0.920 0 3
Research orientation/output
Basic research (in %) BASIC 41.707 26.357 0 100
Applied research (in %) APPLIED 40.598 26.357 0 100
Experimental development (in %) DEVELOPEMENT 17.695 21.519 0 100
Number of publications PUBLICATIONS 11.167 20.448 0 243
11
Number of citations to publications PUB_CITATIONS 236.709 608.970 0 5907
Number of patents PATENTS 1.402 3.463 0 32
Number of citations to patents PAT_CITATIONS 20.054 124.968 0 2634
Research funding
Public grants (in % of total budget) PUBLIC GRANTS 21.779 20.123 0 100
Industry grants (in % of total
budget) INDUSTRY GRANTS 8.580 13.435 0 100
* Seven scientific field dummies not presented. See Table A.2 in the Appendix for details.
3.2 Job choices of departing researchers
The survey asked a series of questions about researchers that had left the unit during the two
years prior to the survey. On average, about six researchers left each research unit (median =
4). We distinguish between short-, medium- and long-term affiliation to the unit and find, in
line with our expectations, that drop-out is highest after four to five years (see Table 2). People
leaving after more than five years occurs much less. This observation points to the conclusion
that the majority of departing researchers leave after completing their doctoral degrees, their
habilitation (postdoc) or quit earlier. We see that this pattern is quite consistent for all institution
types. University of Applied Sciences, however, have fewer departing researchers due to
smaller overall team sizes (see Table A.1 in the Appendix).
Table 2: Departing researchers (n = 676 obs.)
Variable description Median Mean Std.
Dev. Min Max
Number of departing researchers 4 6.30 9.56 0 132
by duration of employment
1-3 years 1 2.63 6.65 0 105
4-5 years 1 2.89 4.68 0 40
> 5 years 0 1.01 3.50 0 60
Table 3 shows the research unit answers regarding the destinations of these departing
researchers. The survey question was formulated such that unit heads were asked to indicate
the type of job (R&D job or other activity) leavers had taken up, the type of firm or institution,
and whether leavers took up the new job in Germany or abroad. Multiple answers were possible
and respondents could also indicate that they did not know. We can broadly classify the job
destinations as jobs in industry and as employment at public institutions, e.g. universities, public
12
research centres, and other public institutions, which include government. We further
distinguish between different types of “industry jobs”: start-up companies, employment at small
and medium-sized firms (SMEs), employment at large firms, and jobs in consulting firms.3
The descriptive statistic already show that the higher education system provides an important
source for highly qualified employees for both the public and private sectors. Only 6% of
research units trained researchers for public jobs alone, and 31% reported that all their departing
researchers joined industry (numbers not reported in Table 3). The majority of units, however,
indicated destinations in both the public and the private sector. Academic start-ups as post-
employment job choice occurred in about 20% of the units. The foundation of a new firm by
former employees is highest at technical universities as well in the fields of physics and
mechanical engineering (see Table A.2 for a disaggregation by field and institution type). The
difference between SMEs and larger firms is not particularly pronounced, although researchers
that go abroad move to larger firms. A large share of departing researchers tend to stay in R&D-
related jobs as indicated by the generally relative small differences between the categories “any”
job and “R&D” job. When comparing jobs domestically and abroad, we find the overall
distribution pattern to be quite similar although going abroad is relatively less common. Finally,
universities and public research institutions also constitute important destinations for leavers.
We are interested in which department or unit-level factors explain the destination choices of
departing researchers. In the econometric set-up we account for the fact that post-employment
job choices are taken with different options in mind. In addition to the unit level characteristics
presented in Table 1, we control for geographical characteristics as local opportunities may
affect the decision to start a new firm and geographical proximity to large firms or consulting
companies may induce young researchers to move there. We therefore include three measures
for regional economic activity at the district level (Landkreis)4 in values referring to the pre-
survey year. The gross domestic product is included to account for industrial activity in the
region, income per capita to take demand factors into account and we calculate net entry (new
firm registrations minus exists) to control for regional structural change.
Table 3: Job choices of departing researchers by type of destination (n = 676)
Variable description Variable name Mean values
3 A further category was “unemployment“. This category had been selected by 7% of the departments, however,
always in combination with other categories. 4 Germany has 295 of these districts of which our sample covers 38.
13
ALL DOMESTIC ABROAD
any R&D any R&D any R&D
Industry job INDUSTRY_JOB 0.694 0.580 0.685 0.566 0.158 0.067
Start-up START_UP 0.194 0.090 0.192 0.89 0.033 0.004
SME SME 0.472 0.333 0.462 0.321 0.052 0.021
Large firm LARGE_FIRM 0.546 0.439 0.530 0.425 0.115 0.050
Consulting company CONSULTING 0.197 0.068 0.195 0.058 0.018 0.010
Public job PUBLIC_JOB 0.439 0.339 0.398 0.327 0.159 0.018
(Techn.) University /
UAS UNI_RESEARCH 0.358 0.260 0.300 0.251 0.136 0.010
Public research
institution PUBLIC_RESEARCH 0.204 0.160 0.189 0.155 0.047 0.004
other public institutions OTHER_PUBLIC 0.107 0.027 0.104 0.024 0.006 0.004
* Unemployment and unknown destination / unknown job type not presented.
4. Econometric Analysis
In the following econometric analysis, we aim to identify research unit factors that explain the
job decisions of departing researchers. It is important to note that job choices can be studied at
different levels of aggregation. We start by distinguishing between industry jobs and public
jobs and consider these options to be interdependent. Next, we consider the case that departing
researchers are settled on the decision to take an R&D job. In that case, we would like to know
what determines whether this job is taken in industry or in public research. Third, we consider
the case of a fixed decision to move to industry. In that case, we investigate what determines
the probability that an R&D job is taken up versus a non-research-related job. Finally, we
disaggregate our measures of public and industry jobs and estimate all employment options
(START_UP, SME, LARGE_FIRM, CONSULTING, UNI, PUBLIC_RESEARCH, OTHER_PUBLIC)
simultaneously.
4.1 Identification Strategy
For each of the first three scenarios, we estimate simultaneous (two-equation) discrete choice
models by maximum-likelihood that can be written as
𝑙𝑛𝐿 = ∑ 𝑙𝑛𝛷2(𝑞1,𝑗,𝑥1,𝑗𝛽1, 𝑞2,𝑗,𝑥2,𝑗𝛽2, 𝜌𝑗∗)𝑁
𝑗=1 (1)
where Ф2 is the cumulative bivariate normal distribution function and qi,j = 1 if yi,j ≠ 0 (for I =
1,2). We observe yi,j = 1 if y*i,j > 0 and yi,j = 0 otherwise. If 𝜌 = 0, then the log likelihood for the
14
bivariate probit model is equal to the sum of the individual log likelihoods of the independent
models.
Next, we disaggregate destinations and estimate the probability of a unit having former
researcher leaving to any of the seven destinations (START_UP, SME, LARGE_FIRM, CONSULTING,
UNI, PUBLIC_RESEARCH, OTHER_PUBLIC) and estimate h-equation multivariate probit models (h
= 7) that can be written as:
𝑦𝑚∗ = 𝑥𝑚𝛽𝑚 + 𝜀𝑚, 𝑚 = 1, … , ℎ (2)
𝑦𝑚 = 𝐷(𝑦𝑚∗ > 0), 𝑚 = 1, … , ℎ (3)
𝜖 = (𝜀1, … 𝜀ℎ)′~𝑁(0, Σ) (4)
with m representing the different destinations and the vector x summarizing unit level,
institutional and regional characteristics. The variance-covariance matrix ∑ has values of 1 on
the diagonal due to normalization and correlations ρjk = ρkj as off-diagonal elements. The log-
likelihood function is then given by:
𝑙𝑛𝐿 = (𝛽1, … 𝛽ℎ), 𝛴; 𝑦|𝑥 = ∑ 𝑙𝑛𝛷ℎ ((𝑞𝑖,1,𝑥𝑖,1𝛽1, … , 𝑞𝑖,ℎ,𝑥𝑖,ℎ𝛽ℎ); 𝛺)𝑁𝑖=1 (5)
The matrix Ω has values of 1 on the diagonal and ωj,k = ωk,j = qi,jqi,kρi,k for j ≠k and
and j,k = (1,..,h) as off-diagonal elements. In the multi-equation case Φℎ denotes the joint
normal distribution of order h.5
4.2 Results
Aggregate destinations
Table 4 presents the results, more precisely marginal effects at the means of all other variables,
from the first set of bivariate estimations on a research unit’s probability to train researchers for
industry and/or public jobs. Both models show that institution type and scientific field are
important factors in shaping the job decision of departing researchers, while regional
characteristics do not have a significant effect. Model 1 includes publication and patent counts,
while we include the citation-weighted publication and patent measures in Model 2. Models 1
and 2 both show that research jointly with industry translates into a higher probability for
5
The expression for the log-likelihood function thus involves an h-dimensional integral that does not have a closed
form. We employ a Maximum Simulated Likelihood Method using the GHK simulator to evaluate the integral
numerically (Geweke 1989, Hajivassiliou and McFadden 1998, and Keane 1994). See Roodman (2009) for details
on the implementation of the simulation method.
15
researchers to move to industry. Contract research also has a positive, but slightly weaker effect.
A higher publication performance during the five-years prior to the survey is associated with a
higher likelihood of researchers moving to other public institutions. This confirms Hypothesis
1 that research units with higher academic publication performance are more likely to see their
departing researchers take jobs at universities and public research institutions.
Whereas the patent count is not a good predictor for any of the destination options, citation
counts to patents filed in the five-years prior to the survey are positively associated with
industry jobs.
We additionally control for research orientation (basic, applied and experimental) and see that
basic research orientation is equally as important for both destination types. When splitting up
the unit’s grant-based financing into public grants and industry sponsoring, we find that industry
grants predict industry jobs and public grants predict public jobs. Finally, we see that older unit
heads see their former employees more often move to public jobs.
Models 1 and 2 of Table 5a repeat Models 1 and 2 of Table 4 for R&D jobs in industry and/or
public institutions, thus excluding all non-R&D job destinations. Just as previously, we find
that researchers trained in research units with a higher share of industry grants are more likely
to move to industry, while public grants are associated with public sector jobs. Publication
numbers have a positive effect on the propensity of departing researchers to take up jobs in
public institutions, while citations do not. Patent numbers have a positive effect on research
jobs in industry, but the patent citation measure turns insignificant. This partially confirms
Hypothesis 3 that research units with higher patenting activity are more likely to see their
departing researchers leave for an R&D job in industry. We cannot confirm that patent
relevance for industrial applications (as measured through patent citations) matters. The
analysis of the disaggregated destinations provides a more refined view as we describe in the
next section.
Table 4: Simultaneous bivariate probit estimation results on industry versus public sector employment (n = 676)
Model 1 Model 2
INDUSTRY_JOB PUBLIC_JOB INDUSTRY_JOB PUBLIC_JOB
dy/dx s.e. dy/dx s.e. dy/dx s.e. dy/dx s.e.
Institution and research unit
ln(STUDENTS) -0.037 0.037 0.021 0.049 -0.036 0.037 0.023 0.049
TECHS -0.003 * 0.002 0.001 0.002 -0.003 * 0.002 -0.001 0.002
EXPERIENCE 0.004 0.037 0.009 *** 0.003 0.004 * 0.002 0.009 *** 0.003
16
FEMALE -0.070 0.104 -0.066 0.117 -0.072 0.101 -0.080 0.116
FORMER_JOB 0.010 0.021 -0.023 0.021 0.008 0.021 -0.030 0.021
CONTRACT_RESEARCH 0.036 * 0.021 0.035 0.025 0.037 * 0.021 0.035 0.025
JOINT_RESEARCH 0.051 ** 0.025 -0.034 0.026 0.051 ** 0.026 -0.033 0.027
Research orientation/output
BASIC 0.016 ** 0.007 0.011 * 0.006 0.016 ** 0.007 0.012 ** 0.006
APPLIED -0.011 * 0.006 -0.001 0.007 -0.011 * 0.006 -0.001 0.007
DEVELOPEMENT 0.004 0.011 -0.001 0.012 0.004 0.011 -0.002 0.025
ln(PUBLICATIONS) 0.019 0.017 0.057 *** 0.021
ln(PATENTS) 0.048 0.030 0.025 0.030
ln(PUB_CITATIONS) 0.016 0.015 0.023 0.019
ln(PAT_CITATIONS) 0.044 ** 0.020 0.039 * 0.020
Research funding
PUBLIC GRANTS 0.002 ** 0.001 0.004 *** 0.001 0.002 ** 0.001 0.004 *** 0.001
INDUSTRY GRANTS 0.004 * 0.002 -0.002 0.002 0.003 * 0.002 -0.002 0.002
Log likelihood -685.75 -687.04
rho (s.e.) 0.454 (0.060)*** 0.451 (0.062)***
Joint sign. field dummies 26.50*** 33.13***
Joint sign. of inst. type dummies 36.02*** 36.52***
Joint sign. of regional variables 4.07 4.23
Note: Institution type and field dummies not presented. All models contain a constant. Standard errors clustered by institution
type, field and region (171 clusters). Marginal effects are calculated at means of all other variables. * (**, ***) indicate
significance levels of 1% (5%, 10%).
Finally, Models 3 and 4 of Table 5a compare R&D jobs in industry and other, non-research
jobs in industry. The results for R&D jobs are similar to those reported in Models 1 and 2. But
refining our previous results, we observe that basic research orientation has a positive marginal
effect for R&D jobs in industry but not for other jobs in the private sector.
17
Table 5a: Simultaneous bivariate probit estimation results on private versus public sector R&D job and type of job in the private sector (n = 676)
Model 1 Model 2 Model 3 Model 4
R&D_INDUSTRY R&D_PUBLIC R&D_INDUSTRY R&D_PUBLIC R&D_INDUSTRY OTHER_INDUSTRY R&D_INDUSTRY OTHER_INDUSTRY
dy/dx s.e. dy/dx s.e. dy/dx s.e. dy/dx s.e. dy/dx s.e. dy/dx s.e. dy/dx s.e. dy/dx s.e.
Institution and research
unit
ln(STUDENTS) -0.061 0.046 -0.014 0.054 -0.059 0.046 -0.013 0.053 -0.061 0.046 0.028 0.042 -0.058 0.046 0.026 0.043
TECHS -0.004 ** 0.002 -0.001 0.002 -0.004 ** 0.003 -0.001 0.002 -0.004 ** 0.002 0.001 0.002 -0.004 ** 0.002 0.001 0.002
EXPERIENCE 0.003 0.003 0.009 *** 0.002 0.003 0.003 0.009 *** 0.002 0.003 0.003 0.004 0.002 0.002 0.003 0.004 * 0.002
FEMALE -0.065 0.124 -0.028 0.107 -0.069 0.121 -0.041 0.107 -0.060 0.129 0.008 0.110 -0.064 0.127 0.003 0.111
FORMER_JOB 0.007 0.023 -0.016 0.020 0.002 0.024 -0.022 0.021 0.006 0.023 0.006 0.023 0.001 0.024 0.007 0.023
CONTRACT_RESEARCH 0.036 * 0.021 0.034 0.024 0.036 * 0.021 0.034 0.024 0.036 * 0.021 0.001 0.022 0.036 * 0.021 0.001 0.022
JOINT_RESEARCH 0.031 0.025 -0.025 0.023 0.036 0.025 -0.021 0.023 0.033 0.025 0.032 0.023 0.038 0.025 0.029 0.023
Research orientation/output
BASIC 0.018 *** 0.006 0.013 ** 0.005 0.018 *** 0.006 0.015 *** 0.005 0.018 *** 0.006 0.006 0.005 0.019 *** 0.006 0.006 0.005
APPLIED -0.007 0.006 0.002 0.006 -0.007 0.006 0.001 0.006 -0.006 0.006 0.008 0.007 -0.007 0.006 0.008 0.007
DEVELOPEMENT -0.007 0.012 0.001 0.009 -0.006 0.012 0.001 0.009 -0.008 0.013 0.004 0.011 -0.007 0.013 0.003 0.011
ln(PUBLICATIONS) 0.012 0.019 0.043 ** 0.019 0.015 0.019 0.036 * 0.021
ln(PATENTS) 0.055 * 0.033 0.025 0.027 0.051 0.034 -0.033 0.028
ln(PUB_CITATIONS) -0.004 0.017 0.014 0.017 -0.004 0.018 0.034 * 0.019
ln(PAT_CITATIONS) 0.033 0.021 0.024 0.019 0.031 0.021 -0.013 0.019
Research funding
PUBLIC GRANTS 0.002 0.001 0.003 *** 0.001 0.002 * 0.001 0.003 *** 0.002 0.002 0.001 0.002 ** 0.001 0.002 * 0.001 0.002 * 0.001
INDUSTRY GRANTS 0.005 *** 0.002 -0.001 0.002 0.005 *** 0.002 -0.001 0.002 0.004 *** 0.002 -0.001 0.002 0.004 *** 0.002 -0.007 0.002
Log likelihood -710.12 -712.87 -771.31 -772.20
rho (s.e.) 0.352 (0.061)*** 0.356 (0.055)*** 0.195 (0.068)*** 0.196 (0.068)***
Joint sign. field dummies 48.20*** 54.43*** 41.77*** 42.00***
Joint sign. of inst. type
dummies 35.19***
36.52***
41.71***
42.30***
Joint sign. of regional
variables 4.24
4.31
13.77**
13.41**
Note: Institution type and field dummies not presented. All models contain a constant. Standard errors clustered by institution type, field and region (171 clusters). Marginal effects are calculated at means of all other
variables. * (**, ***) indicate significance levels of 1% (5%, 10%).
18
Table 5b present the results from a set of models distinguishing between the types of jobs taken
in the public sector. The results shows that publication performance matters in terms of quantity
but not necessarily quality (see Model 6) for R&D tasks in the public sector, but not for non-
research related jobs. Similarly basic research orientation and public grants matter only for
R&D jobs. The latter confirms Hypothesis 2 that research units with a higher share of public
grants are more likely to see their departing researchers to stay in academe. Contract research,
on the other hand, appears to predict non-R&D public jobs.
Table 5b: Simultaneous bivariate probit estimation results on type of job in the public sector (n = 676)
Model 5 Model 6
R&D_PUBLIC OTHER_PUBLIC R&D_PUBLIC OTHER_PUBLIC
dy/dx s.e. dy/dx s.e. dy/dx s.e. dy/dx s.e.
Institution and research unit
ln(STUDENTS) -0.013 0.054 0.037 * 0.014 -0.012 0.053 0.036 ** 0.015
TECHS -0.001 0.002 0.001 0.001 -0.001 0.002 0.001 0.001
EXPERIENCE 0.009 *** 0.002 0.001 0.001 0.008 *** 0.002 0.001 0.001
FEMALE -0.029 0.108 -0.022 0.052 -0.041 0.108 -0.020 0.053
FORMER_JOB -0.015 0.020 0.007 0.009 -0.021 0.020 0.007 0.009
CONTRACT_RESEARCH 0.035 0.024 0.026 *** 0.009 0.035 0.024 0.025 *** 0.009
JOINT_RESEARCH -0.025 0.023 -0.024 * 0.012 -0.021 0.023 -0.025 ** 0.011
Research orientation/output
BASIC 0.013 ** 0.005 0.002 0.002 0.015 *** 0.005 0.002 0.001
APPLIED 0.002 0.006 0.003 0.002 0.001 0.006 0.003 0.002
DEVELOPEMENT -0.001 0.009 0.002 0.004 0.001 0.009 0.002 0.004
ln(PUBLICATIONS) 0.044 ** 0.019 0.007 0.008
ln(PATENTS) 0.028 0.027 -0.006 0.013
ln(PUB_CITATIONS) 0.016 0.017 0.010 0.008
ln(PAT_CITATIONS) 0.024 0.019 0.001 0.008
Research funding
PUBLIC GRANTS 0.003 *** 0.001 -0.001 * 0.000 0.003 *** 0.001 -0.001 * 0.000
INDUSTRY GRANTS -0.001 0.002 -0.001 0.001 -0.001 0.002 -0.001 * 0.000
Log likelihood -512.47 -514.58
rho (s.e.) 0.254 (0.097)** 0.253 (0.097)**
Joint sign. field dummies 32.01*** 41.22***
Joint sign. of inst. type dummies 3.73 4.12
Joint sign. of regional variables 6.86 6.88
19
Disaggregate destinations
Tables 6a and 6b show the results for all seven disaggregate destinations. Table 6b includes the
quality-weighted publication and patent measures. As in the previously presented models,
research field and institution type matter. For instance, research units at both universities and
technical universities are more likely than research units from universities of applied sciences
to see former employees move to large firms. For SMEs the institution types do not differ. We
find that technical universities are more likely to see employees moving to start-up firms. Again,
research units at universities and technical universities are more likely to produce future
university researchers. In addition, we now find that regional factors are jointly significant.
Previous insights that public grants are a good predictor of public sector R&D employment are
confirmed. The effects from industry grants disappear, however. Interestingly, we find
publication performance to matter also for employment in industry, in particular in start-ups,
large firms and in consulting probably reflecting institutional reputation.
For movements to SMEs contract research and patent numbers are highly significant. When
accounting for industrial relevance of the patents (Table 6b), we find a significant positive
impact of citations-weighted patents for SMEs, large firms and public research institutions, but
the marginal effect is largest for the first destination. Thus, we cannot confirm the part of
Hypothesis 3 that presumed that research units with higher patenting activity (also in terms of
the patent relevance for industrial applications) are more likely to see their departing
researchers start their own business, but rather found academic patenting to be associated with
jobs in established SMEs.
Similarly, we found only partial support for our Hypothesis 4 that research units with stronger
industry ties are more likely to see their departing researchers take R&D jobs in industry,
especially in large firms. While Models 1 and 2 in Table 4 confirmed the first part of the
hypothesis, we second part is less clear. Contract research is associated with SME employment
and the effect of joint research is only significant at a 10% confidence level.
Research units with a research focus on experimental development are more likely to see their
former employees move to start-up firms, while for SMEs and larger firms a focus on basic
research appears to be attractive. Thus, Hypothesis 5 that research units with focus on applied
rather than basic research are more likely to see their departing researchers start their own
business or leave for an R&D job in industry is again only partially confirmed.
20
Table 6a: Multi-equation simultaneous probit estimation results on separated destinations (n = 676)
Model 1
START_UP SME LARGE FIRM CONSULTING UNI_RESEARCH PUBLIC_RESEARCH OTHER_PUBLIC
dy/dx s.e. dy/dx s.e. dy/dx s.e. dy/dx s.e. dy/dx s.e. dy/dx s.e. dy/dx s.e.
Institution and research unit
ln(STUDENTS) -0.041 0.029 -0.008 0.05 -0.059 0.045 0.008 0.029 -0.012 0.020 0.030 0.039 0.043 ** 0.017
TECHS 0.002 0.001 -0.001 0.002 -0.001 0.002 -0.001 0.001 -0.002 0.002 -0.001 0.001 0.002 * 0.001
EXPERIENCE 0.005 *** 0.002 0.004 0.003 -0.001 0.003 0.002 0.002 0.008 *** 0.002 0.004 ** 0.002 0.001 0.001
FEMALE 0.156 0.123 -0.012 0.138 -0.199 * 0.112 0.087 0.094 -0.022 0.100 -0.129 *** 0.041 -0.045 0.037
FORMER_JOB 0.002 0.015 -0.030 0.022 0.022 0.026 0.005 0.016 -0.024 0.020 0.007 0.016 -0.003 0.012
CONTRACT_RESEARCH 0.014 0.017 0.065 *** 0.023 0.028 0.023 -0.008 0.015 0.009 0.024 0.028 * 0.017 0.027 *** 0.011
JOINT_RESEARCH -0.006 0.018 0.012 0.027 0.052 * 0.029 0.027 0.017 0.005 0.024 -0.003 0.018 -0.018 0.016
Research orientation/output
BASIC 0.005 * 0.003 0.016 *** 0.005 0.018 *** 0.007 0.006 * 0.003 0.012 ** 0.005 0.008 *** 0.003 0.002 0.002
APPLIED 0.006 0.018 0.005 0.007 -0.007 0.007 0.009 * 0.005 -0.001 0.007 0.002 0.005 0.001 0.003
DEVELOPEMENT 0.017 * 0.009 -0.001 0.016 0.009 0.013 -0.011 0.009 0.001 0.012 -0.002 0.009 0.005 0.005
ln(PUBLICATIONS) 0.031 ** 0.014 0.001 0.019 0.060 *** 0.019 0.029 ** 0.014 0.032 * 0.017 0.018 0.015 0.011 0.010
ln(PATENTS) 0.001 0.020 0.063 ** 0.029 0.036 0.033 -0.023 0.022 0.014 0.030 -0.008 0.019 -0.007 0.015
Research funding
PUBLIC GRANTS 0.001 0.001 0.002 * 0.001 0.002 0.001 0.001 0.001 0.003 ** 0.001 0.002 ** 0.001 -0.001 0.001
INDUSTRY GRANTS 0.002 0.001 -0.001 0.002 0.003 0.002 0.001 0.001 -0.002 0.001 -0001 0.002 -0.001 0.001
Institution type
UNI 0.084 0.060 0.107 0.082 0.533 *** 0.068 0.173 *** 0.066 0.211 ** 0.082 0.086 0.067 0.043 0.041
TU 0.120 * 0.072 0.145 0.090 0.445 *** 0.062 0.195 * 0.107 0.201 ** 0.098 0.091 0.093 0.051 0.052
UAS Reference category
Log likelihood -2051.471
Joint sign. field dummies 152.61***
Joint sign. of inst. type dummies 56.71***
Joint sign. of regional variables 67.80***
Note: Field dummies not presented. All models contain a constant. Standard errors clustered by institution type, field and region (171 clusters). Marginal effects are calculated at means of all other
variables. * (**, ***) indicate significance levels of 1% (5%, 10%). See Table A.3 for correlations between equations.
21
Table 6b: Multi-equation simultaneous probit estimation results on separated destinations with quality-weighted measures research performance (n = 676)
Model 2
START_UP SME LARGE FIRM CONSULTING UNI_RESEARCH PUBLIC_RESEARCH OTHER_PUBLIC
dy/dx s.e. dy/dx s.e. dy/dx s.e. dy/dx s.e. dy/dx s.e. dy/dx s.e. dy/dx s.e.
Institution and research unit
ln(STUDENTS) -0.046 0.029 -0.004 0.045 -0.054 0.044 0.003 0.029 -0.017 0.043 0.040 0.038 0.044 ** 0.018
TECHS 0.001 0.001 -0.001 0.002 -0.003 0.002 -0.001 0.001 -0.002 0.002 -0.001 0.001 0.001 * 0.001
EXPERIENCE 0.005 *** 0.002 0.004 0.003 -0.002 ** 0.003 0.002 0.002 0.008 *** 0.002 0.004 ** 0.002 0.001 0.001
FEMALE 0.138 0.120 -0.008 0.136 -0.219 *** 0.111 0.072 0.090 -0.036 0.096 -0.128 *** 0.039 -0.046 0.038
FORMER_JOB 0.001 0.015 -0.029 0.022 0.012 0.026 0.003 0.016 -0.028 0.020 0.007 0.016 -0.002 0.012
CONTRACT_RESEARCH 0.013 0.017 0.065 *** 0.023 0.028 * 0.023 -0.008 0.016 0.008 0.024 0.028 * 0.016 0.027 *** 0.011
JOINT_RESEARCH -0.001 0.017 0.011 0.026 0.054 0.028 0.029 * 0.017 0.014 0.025 -0.009 0.017 -0.021 0.015
Research orientation/output
BASIC 0.006 * 0.003 0.016 *** 0.005 0.019 *** 0.007 0.006 ** 0.003 0.012 ** 0.005 0.008 *** 0.003 0.002 0.002
APPLIED 0.006 0.005 0.005 0.016 -0.008 0.007 0.008 * 0.005 -0.001 0.007 0.003 0.005 0.001 0.003
DEVELOPEMENT 0.018 ** 0.009 -0.002 0.016 0.009 0.014 -0.011 0.009 0.003 0.012 -0.006 0.009 0.003 0.004
ln(PUB_CITATIONS) 0.025 * 0.013 0.019 0.019 0.017 0.017 0.019 * 0.011 0.017 0.017 0.006 0.013 0.010 0.010
ln(PAT_CITATIONS) -0.013 0.013 0.055 ** 0.022 0.038 ** 0.022 -0.022 0.015 -0.010 0.019 0.027 ** 0.013 0.008 0.010
Research funding
PUBLIC GRANTS 0.001 0.001 0.002 ** 0.001 0.002 ** 0.001 0.001 0.001 0.003 ** 0.001 0.002 ** 0.001 -0.001 0.001
INDUSTRY GRANTS 0.002 0.001 -0.001 0.002 0.002 0.002 0.001 0.001 -0.002 0.001 -0.001 0.002 -0.001 0.001
Institution type
UNI 0.095 0.061 0.085 0.082 0.547 *** 0.065 0.185 *** 0.064 0.231 *** 0.080 0.076 0.066 0.040 0.041
TU 0.130 * 0.075 0.128 0.089 0.447 *** 0.062 0.210 ** 0.107 0.214 ** 0.097 0.074 0.090 0.044 0.052
UAS Reference category
Log likelihood -2051.67
Joint sign. field dummies 163.75***
Joint sign. of inst. type dummies 62.25***
Joint sign. of regional variables 70.26***
Note: Field dummies not presented. All models contain a constant. Standard errors clustered by institution type, field and region (171 clusters). Marginal effects are calculated at means of all other
variables. * (**, ***) indicate significance levels of 1% (5%, 10%). See Table A.3 for correlations between equations.
22
5 Conclusions
In this paper we studied the importance of team and organisational characteristics of the
home departments for career choices of departing researchers. Young researchers may move
to a different university or move out of academia all together and into industry. Previous
literature has linked these decisions to the young academic’s taste for science and academic
performance (Scott, 2004). Research groups have been argued to affect individual performance
as well as shaping the careers of their members (Walsh and Lee, 2013), by providing young
researchers with a certain skill set and by giving access to professional networks.
Based on survey data from 676 science and engineering research units at 46 universities in
Germany we differentiate between R&D and non-R&D jobs in newly formed firms, SMEs,
large firms, consulting companies, public research institutes and universities. The results from
multiple simultaneous equations models on the likelihood that researchers from a focal
department or research unit take a specific follow-on career decision confirm the important role
of home departments (nests) for career building of researchers.
We find that the performance of the nest in terms of access to research grants and
publication numbers as well as industry ties shape job choices of departing researchers. In
particular, we find that the share of the research unit’s total budget from public grants and its
publication performance increases the probability that departing researchers take a research job
in the public sector. On the other hand, grants from industry increase the likelihood that they
take a job in industry. Patents correlate positively with R&D jobs in SMEs indicating that for
these firms seeking technological knowledge from former university employees may be crucial.
The quality-weighted patent count, however, also matters for employment in larger firms and
public research institutions. In line with these result, we find the department head’s network
with industry partners, especially when established through contract research or joint research
projects, to increase the propensity of her staff to move to industry. Research orientation (basic
versus applied), however, is not a precise predictor of academic or industry job choice. Instead,
researchers from applied units are more likely to go into consulting, while those from units
with a focus on experimental development are more likely to start their own firm.
From a policy perspective these results shed some light on the factors behind the job and
destination choices of academics. Young researchers have been recognised as contributing to
innovation and knowledge production and their mobility as supporting knowledge and
technology transfer. The results show that nest characteristics are important for explaining the
23
job choices of young researchers and point at the importance of department links and prestige
for job prospects of young researchers. Although the results are based on a single academic
market they can be extended to other countries as temporary positions in academia are
increasing and more academics are trained than positions are available in academe.
We strongly encourage future research on job choices of departing researchers. Little is still
known about which nest designs may contribute to keeping the most able scientific researchers
in academe, while at the same time facilitating knowledge transfer to industry and public
institutions. While our study focussed solely on the home department, future studies may in
addition consider individual characteristics of the researchers. Further, it would be worth
studying more explicitly the mechanisms that shape career decisions and how individual career
preferences change or do not change during initial academic employment. Future studies may
also attempt to measure the selection of certain individuals into particular research
environments and how that affects the nest culture.
24
References
Aghion, P., Dewatripont, M., Stein, J. (2008), Academic Freedom, Private-Sector Focus, and
the Process of Innovation, Rand Journal of Economics 39, 617–635.
Agarwal, R., Ohyama, A. (2013), Industry or Academia, Basic or Applied? Career Choices and
Earnings Trajectories of Scientists, Management Science 59, 950–970.
Aschoff, B., Grimpe, C. (2014), Contemporaneous Peer Effects, Career Age and the Industry
Involvement of Academics in Biotechnology, Research Policy 43, 367-381.
Audretsch, D., Stephan, P. (1999), Knowledge Spillovers in Biotechnology: Sources and
Incentives, Journal of Evolutionary Economics 9, 97-107.
Auriol, L., M. Misu, Freeman, R. (2013), Careers of Doctorate Holders: Analysis of Labour
Market and Mobility Indicators, OECD Science, Technology and Industry Working Papers.
2013/04, OECD, http://dx.doi.org/10.1787/5k43nxgs289w-en.
Austin, A.E. (2002), Preparing the Next Generation of Faculty: Graduate School as
Socialization to the Academic Career, Journal of Higher Education 73, 94-122.
Azoulay, P., Zivin J.S.G., Sampat B.N. (2011), The Diffusion of Scientific Knowledge Across
Time and Space: Evidence from Professional Transitions for the Superstars in Medicine,
NBER Working Paper No 16683.
Baldi, S. (1995), Prestige Determinants of First Academic Job for New Sociology PhDs: 1985-
1992, Sociological Quarterly 36, 777–789.
Balsmeier, B., Pellens, M. (2014), Who Makes, who Breaks: Which Scientists Stay in
Academe?, Economics Letters 122, 229-232.
Becker, G.S. (1962), Investment in Human Beings, Journal of Political Economy 70, 9–49.
Bercovitz J., Feldman M. (2008), Academic Entrepreneurs: Organizational Change at the
Individual Level, Organization Science 19, 69-89.
Bourdieu, P. (1986), The Forms of Social Capital, In. Richardson, J.G. ed. Handbook of Theory
and research for the sociology of education, New York, Greenwood.
Burris, V. (2004), The Academic Caste System: Prestige Hierarchies in PhD Exchange
Networks, American Sociological Review 69, 239–264.
Burt, R.S. (1997), A Note on Social Capital and Network Content, Social Networks 19, 355–
373.
Coleman, J.C. (1988) Social Capital in the Creation Human Capital, American Journal of
Sociology 94, 95–120.
Crane, D., (1965), Scientists at Major and Minor Universities, A Study of Productivity and
Recognition, American Sociological Review, 30, 699–714.
Crane, D., (1970), The Academic Marketplace Revisited: a Study of Faculty Mobility Using
the Carter Ratings, The American Journal of Sociology 75, 953–964.
Crespi, A.G., Geuna, A., Nesta, J.J.L. (2007), The Mobility of University Inventors in Europe,
Journal of Technology Transfer 32, 195-215.
25
Enders, J. (2005), Border Crossings: Research Training, Knowledge Dissemination and the
Transformation of Academic Work, Higher Education 49, 119-133.
Fox, M.F., Stephan, P. (2001), Careers of Young Scientists: Preferences, Prospects and
Realities by Gender and Field, Social Studies of Science 31, 109-22.
Franzoni, C., Stephan, P., Scellato, G. (2012), Foreign Born Scientists: Mobility Patterns for
Sixteen Countries, Nature Biotechnology 30 (12), 1250-1253.
Granovetter, M.S. (1985), Economic Action and Social Structure: The Problem of
Embeddedness, American Journal of Sociology 91, 481-510.
Griliches, Z. (1973), Research Expenditures and Growth Accounting. In: Williams, B. R. (ed.)
Science and technology in economic growth, London, Macmillan, 59–95.
Hottenrott, H., Thorwarth, S. (2011), Industry Funding of University Research and Scientific
Productivity, Kyklos 64 (4), 534–555.
Lam, A. (2007), Knowledge Networks and Careers: Academic Scientists in Industry-
University Links, Journal of Management Studies 44(6), 993-1016.
Lam, A. (2011), What Motivates Academic Scientists to Engage in Research
Commercialization: ‘Gold’, ‘ribbon’ or ‘puzzle’?, Research Policy 40, 1354-1368.
Lam, A., Campos, A. (2014), Content to be Sad’ or ‘Runaway Apprentice’? The Psychological
Contract and Career Agency of Young Scientists in the Entrepreneurial University, Human
Relations, forthcoming.
Lawson, C. (2013), Academic Patenting: The Importance of Industry Support, Journal of
Technology Transfer 38 (4), 509-535.
Levin, S.G., Stephan, P.E. (1991), Research Productivity Over the Life Cycle: Evidence for
Academic Scientists, American Economic Review 81, 114–132.
Long, J. S. (1978), Productivity and Academic Position in the Scientific Career, American
Sociological Review 43, 889–908.
Mangematin, V. (2000), PhD Job Market: Professional Trajectories and Incentives During the
PhD, Research Policy 29(6), 741-756.
Nelson, R., Phelps, E. (1966), Investment in Humans, Technological Diffusion and Economic
Growth, American Economic Review 61, 69-75.
OECD (2000), Mobilising Human Resources for Innovation. Paris: OECD.
OECD (2008), The Global Competition for Talent: Mobility of the Highly Skilled. Directorate
for Science Technology and Industry, Paris: OECD.
Owen-Smith, J., Powell, W.W. (2001), To Patent or Not: Faculty Decisions and Institutional
Success at Technology Transfer, Journal of Technology Transfer 26, 99-114.
Petersen, A., Fortunato, S., Pan, R.K., Kaski, K., Penner, O., Riccaboni, M., Stanley, H.E.,
Pammolli, F. (2013), Reputation and Impact in Academic Careers, arXiv:1303.7274v3
[physics.soc-ph] 23 Jan 2014.
Pezzoni, M., Sterzi, V., Lissoni, F. (2012), Career Progress in Centralized Academic Systems:
an Analysis of French and Italian Physicists, Research Policy 41, 704-719.
26
Roach, M., Sauermann, H. (2010), A Taste for Science? PhD Scientists’ Academic Orientation
and Self-Selection into Research Careers in Industry, Research Policy 39, 422-434.
Sauermann, H., Roach, M. (2012), Taste for Science, Taste for Commercialization, and Hybrid
Scientists, Paper presented at The Organisation, Economics and Policy of Scientific
Research, Torino, Italy.
Sauermann, H., Roach, M. (2014), Not all Scientists pay to be Scientists: PhDs’ Preferences
for Publishing in Industrial Employment, Research Policy 43, 32-47.
Sauermann H., Stephan, P. (2013), Conflicting Logics? A Multidimensional View of Industrial
and Academic Science, Organization Science 24, 889-909.
Schultz, T. (1961), Investment in Human Capital, The American Economic Review 51(1), 1-17.
Schultze, G.G., Warning, S., Wiermann, C. (2008), What and How Long Does It Take to Get
Tenure? The Case of Economics and Business Administration in Austria, Germany and
Switzerland, German Economic Review 9 (4), 473-505.
Smeby, J. C., Try, S. (2005), Departmental Contexts and Faculty Research Activity in Norway,
Research in Higher Education 46, 593–619.
Stephan, P. (2012), How Economics Shapes Science, Harvard University Press: Cambridge,
MA.
Stern, S. (2004), Do Scientists Pay to be Scientists?, Management Science 50, 835–853.
Stuart, T., W. Ding. (2006), When do Scientists Become Entrepreneurs? The Social Structural
Antecedents of Commercial Activity in Academic Life Sciences, American Journal of
Sociology 112, 97–144.
Toole, A.A., Czarnitzki, D. (2010), Commercializing Science: Is There a University “Brain
Drain” from Academic Entrepreneurship?, Management Science 56, 1599-1614.
Hall, B.H., Mairesse, J., Turner, L., (2007). Identifying Age, Cohort, and Period Effects in
Scientific Research Productivity: Discussion and Illustration Using Simulated and Actual
Data on French Physicists, Economics of Innovation and New Technology 16(2), 159-177.
Waldinger, F. (2012), Peer Effects in Science - Evidence from the Dismissal of Scientists in
Nazi Germany, Review of Economic Studies 79 (2), 838-861.
Walsh, J.P., Lee, Y.-N., (2013), The Bureaucratization of Science, Paper presented at The
Organisation, Economics and Policy of Scientific Research, Torino, Italy.
Zellner, C. (2003), The Economic Effects of Basic Research: Evidence for Embodied
Knowledge Transfer via Scientists’ Migration, Research Policy 32, 1881-1895.
Zinovyeva, N., Bagues, M., (2014), The Role of Connections in Academic Promotion,
American Economic Journal: Applied Economics, forthcoming.
27
APPENDICES
Table A.1: Departing researchers by institution type (means, n = 676)
Variable description UNI TU UAS
Number of researchers 8.53 10.55 1.74
Number of departing researchers 7.61 7.52 1.46
by duration of employment
1-3 years 3.46 2.25 0.86
in % 34.38 22.61 27.90
4-5 years 3.34 4.21 0.25
in % 42.22 50.14 7.47
> 5 years 1.14 1.36 0.29
in % 10.94 15.78 9.70
Table A.2: Job choices by field and institution type (n = 676)
Field # % INDUSTRY JOBS PUBLIC JOBS
START_UP SME
LARGE_
FIRM
CON-
SULTING
UNI_
RESEARCH
PUBLIC_
RESEARCH
OTHER_
PUBLIC
by field
Physics 106 15.68 0.26 0.55 0.65 0.30 0.58 0.42 0.15
Mathematics /
Computer Science 107 15.83 0.12 0.36 0.46 0.23 0.31 0.12 0.06
Chemistry 95 14.05 0.15 0.62 0.65 0.23 0.52 0.31 0.23
Biology 58 8.58 0.17 0.47 0.26 0.05 0.45 0.19 0.05
Electrical Engineering 101 14.94 0.19 0.45 0.60 0.09 0.20 0.13 0.08
Mechanical Engineering 108 15.98 0.24 0.51 0.58 0.22 0.21 0.08 0.05
Other Engineering 101 14.94 0.21 0.36 0.50 0.18 0.29 0.19 0.12
by institution type
University 377 57.32 0.20 0.52 0.65 0.24 0.47 0.27 0.13
Technical University 157 23.68 0.29 0.54 0.63 0.25 0.35 0.20 0.11
University of Applied
Sciences 142 19.00 0.08 0.27 0.17 0.02 0.08 0.04 0.04
PREVIOUS DISCUSSION PAPERS
154 Hottenrott, Hanna and Lopes-Bento, Cindy, R&D Partnerships and Innovation Performance: Can There be too Much of a Good Thing?, July 2014.
153 Hottenrott, Hanna and Lawson, Cornelia, Flying the Nest: How the Home Department Shapes Researchers’ Career Paths, July 2014.
152 Hottenrott, Hanna, Lopes-Bento, Cindy and Veugelers, Reinhilde, Direct and Cross-Scheme Effects in a Research and Development Subsidy Program, July 2014.
151 Dewenter, Ralf and Heimeshoff, Ulrich, Do Expert Reviews Really Drive Demand? Evidence from a German Car Magazine, July 2014.
150 Bataille, Marc, Steinmetz, Alexander and Thorwarth, Susanne, Screening Instruments for Monitoring Market Power in Wholesale Electricity Markets – Lessons from Applications in Germany, July 2014.
149 Kholodilin, Konstantin A., Thomas, Tobias and Ulbricht, Dirk, Do Media Data Help to Predict German Industrial Production?, July 2014.
148 Hogrefe, Jan and Wrona, Jens, Trade, Tasks, and Trading: The Effect of Offshoring on Individual Skill Upgrading, June 2014.
147 Gaudin, Germain and White, Alexander, On the Antitrust Economics of the Electronic Books Industry, May 2014.
146 Alipranti, Maria, Milliou, Chrysovalantou and Petrakis, Emmanuel, Price vs. Quantity Competition in a Vertically Related Market, May 2014. Published in: Economics Letters, 124 (1) (2014), pp.122-126.
145 Blanco, Mariana, Engelmann, Dirk, Koch, Alexander K., and Normann, Hans-Theo, Preferences and Beliefs in a Sequential Social Dilemma: A Within-Subjects Analysis, May 2014. Published in: Games and Economic Behavior, 87 (2014), pp.122-135.
144 Jeitschko, Thomas D., Jung, Yeonjei and Kim, Jaesoo, Bundling and Joint Marketing by Rival Firms, May 2014.
143 Benndorf, Volker and Normann, Hans-Theo, The Willingness to Sell Private Data, May 2014.
142 Dauth, Wolfgang and Suedekum, Jens, Globalization and Local Profiles of Economic Growth and Industrial Change, April 2014.
141 Nowak, Verena, Schwarz, Christian and Suedekum, Jens, Asymmetric Spiders: Supplier Heterogeneity and the Organization of Firms, April 2014.
140 Hasnas, Irina, A Note on Consumer Flexibility, Data Quality and Collusion, April 2014.
139 Baye, Irina and Hasnas, Irina, Consumer Flexibility, Data Quality and Location Choice, April 2014.
138 Aghadadashli, Hamid and Wey, Christian, Multi-Union Bargaining: Tariff Plurality and Tariff Competition, April 2014.
137 Duso, Tomaso, Herr, Annika and Suppliet, Moritz, The Welfare Impact of Parallel Imports: A Structural Approach Applied to the German Market for Oral Anti-diabetics, April 2014. Forthcoming in: Health Economics.
136 Haucap, Justus and Müller, Andrea, Why are Economists so Different? Nature, Nurture and Gender Effects in a Simple Trust Game, March 2014.
135 Normann, Hans-Theo and Rau, Holger A., Simultaneous and Sequential Contributions to Step-Level Public Goods: One vs. Two Provision Levels, March 2014. Forthcoming in: Journal of Conflict Resolution.
134 Bucher, Monika, Hauck, Achim and Neyer, Ulrike, Frictions in the Interbank Market and Uncertain Liquidity Needs: Implications for Monetary Policy Implementation, July 2014 (First Version March 2014).
133 Czarnitzki, Dirk, Hall, Bronwyn, H. and Hottenrott, Hanna, Patents as Quality Signals? The Implications for Financing Constraints on R&D?, February 2014.
132 Dewenter, Ralf and Heimeshoff, Ulrich, Media Bias and Advertising: Evidence from a German Car Magazine, February 2014. Forthcoming in: Review of Economics.
131 Baye, Irina and Sapi, Geza, Targeted Pricing, Consumer Myopia and Investment in Customer-Tracking Technology, February 2014.
130 Clemens, Georg and Rau, Holger A., Do Leniency Policies Facilitate Collusion? Experimental Evidence, January 2014.
129 Hottenrott, Hanna and Lawson, Cornelia, Fishing for Complementarities: Competitive Research Funding and Research Productivity, December 2013.
128 Hottenrott, Hanna and Rexhäuser, Sascha, Policy-Induced Environmental Technology and Inventive Efforts: Is There a Crowding Out?, December 2013.
127 Dauth, Wolfgang, Findeisen, Sebastian and Suedekum, Jens, The Rise of the East and the Far East: German Labor Markets and Trade Integration, December 2013. Forthcoming in: Journal of European Economic Association.
126 Wenzel, Tobias, Consumer Myopia, Competition and the Incentives to Unshroud Add-on Information, December 2013. Published in: Journal of Economic Behavior and Organization, 98 (2014), pp. 89-96.
125 Schwarz, Christian and Suedekum, Jens, Global Sourcing of Complex Production Processes, December 2013. Published in: Journal of International Economics, 93 (2014), pp. 123-139.
124 Defever, Fabrice and Suedekum, Jens, Financial Liberalization and the Relationship-Specificity of Exports, December 2013. Published in: Economics Letters, 122 (2014), pp. 375-379.
123 Bauernschuster, Stefan, Falck, Oliver, Heblich, Stephan and Suedekum, Jens, Why Are Educated and Risk-Loving Persons More Mobile Across Regions?, December 2013. Published in: Journal of Economic Behavior and Organization, 98 (2014), pp. 56-69.
122 Hottenrott, Hanna and Lopes-Bento, Cindy, Quantity or Quality? Knowledge Alliances and their Effects on Patenting, December 2013.
121 Hottenrott, Hanna and Lopes-Bento, Cindy, (International) R&D Collaboration and SMEs: The Effectiveness of Targeted Public R&D Support Schemes, December 2013. Forthcoming in: Research Policy.
120 Giesen, Kristian and Suedekum, Jens, City Age and City Size, November 2013. Forthcoming in: European Economic Review.
119 Trax, Michaela, Brunow, Stephan and Suedekum, Jens, Cultural Diversity and Plant-Level Productivity, November 2013.
118 Manasakis, Constantine and Vlassis, Minas, Downstream Mode of Competition With Upstream Market Power, November 2013. Published in: Research in Economics, 68 (2014), pp. 84-93.
117 Sapi, Geza and Suleymanova, Irina, Consumer Flexibility, Data Quality and Targeted Pricing, November 2013.
116 Hinloopen, Jeroen, Müller, Wieland and Normann, Hans-Theo, Output Commitment Through Product Bundling: Experimental Evidence, November 2013. Published in: European Economic Review 65 (2014), pp. 164-180.
115 Baumann, Florian, Denter, Philipp and Friehe Tim, Hide or Show? Endogenous Observability of Private Precautions Against Crime When Property Value is Private Information, November 2013.
114 Fan, Ying, Kühn, Kai-Uwe and Lafontaine, Francine, Financial Constraints and Moral Hazard: The Case of Franchising, November 2013.
113 Aguzzoni, Luca, Argentesi, Elena, Buccirossi, Paolo, Ciari, Lorenzo, Duso, Tomaso, Tognoni, Massimo and Vitale, Cristiana, They Played the Merger Game: A Retrospective Analysis in the UK Videogames Market, October 2013. Forthcoming in: Journal of Competition and Economics under the title: “A Retrospective Merger Analysis in the UK Videogames Market”.
112 Myrseth, Kristian Ove R., Riener, Gerhard and Wollbrant, Conny, Tangible Temptation in the Social Dilemma: Cash, Cooperation, and Self-Control, October 2013.
111 Hasnas, Irina, Lambertini, Luca and Palestini, Arsen, Open Innovation in a Dynamic Cournot Duopoly, October 2013. Published in: Economic Modelling, 36 (2014), pp. 79-87.
110 Baumann, Florian and Friehe, Tim, Competitive Pressure and Corporate Crime, September 2013.
109 Böckers, Veit, Haucap, Justus and Heimeshoff, Ulrich, Benefits of an Integrated European Electricity Market, September 2013.
108 Normann, Hans-Theo and Tan, Elaine S., Effects of Different Cartel Policies: Evidence from the German Power-Cable Industry, September 2013. Forthcoming in: Industrial and Corporate Change.
107 Haucap, Justus, Heimeshoff, Ulrich, Klein, Gordon J., Rickert, Dennis and Wey, Christian, Bargaining Power in Manufacturer-Retailer Relationships, September 2013.
106 Baumann, Florian and Friehe, Tim, Design Standards and Technology Adoption: Welfare Effects of Increasing Environmental Fines when the Number of Firms is Endogenous, September 2013.
105 Jeitschko, Thomas D., NYSE Changing Hands: Antitrust and Attempted Acquisitions of an Erstwhile Monopoly, August 2013. Published in: Journal of Stock and Forex Trading, 2 (2) (2013), pp. 1-6.
104 Böckers, Veit, Giessing, Leonie and Rösch, Jürgen, The Green Game Changer: An Empirical Assessment of the Effects of Wind and Solar Power on the Merit Order, August 2013.
103 Haucap, Justus and Muck, Johannes, What Drives the Relevance and Reputation of Economics Journals? An Update from a Survey among Economists, August 2013.
102 Jovanovic, Dragan and Wey, Christian, Passive Partial Ownership, Sneaky Takeovers, and Merger Control, August 2013. Revised version forthcoming in: Economics Letters.
101 Haucap, Justus, Heimeshoff, Ulrich, Klein, Gordon J., Rickert, Dennis and Wey, Christian, Inter-Format Competition Among Retailers – The Role of Private Label Products in Market Delineation, August 2013.
100 Normann, Hans-Theo, Requate, Till and Waichman, Israel, Do Short-Term Laboratory Experiments Provide Valid Descriptions of Long-Term Economic Interactions? A Study of Cournot Markets, July 2013. Forthcoming in: Experimental Economics.
99 Dertwinkel-Kalt, Markus, Haucap, Justus and Wey, Christian, Input Price Discrimination (Bans), Entry and Welfare, June 2013.
98 Aguzzoni, Luca, Argentesi, Elena, Ciari, Lorenzo, Duso, Tomaso and Tognoni, Massimo, Ex-post Merger Evaluation in the UK Retail Market for Books, June 2013.
97 Caprice, Stéphane and von Schlippenbach, Vanessa, One-Stop Shopping as a Cause of Slotting Fees: A Rent-Shifting Mechanism, May 2012. Published in: Journal of Economics and Management Strategy, 22 (2013), pp. 468-487.
96 Wenzel, Tobias, Independent Service Operators in ATM Markets, June 2013. Published in: Scottish Journal of Political Economy, 61 (2014), pp. 26-47.
95 Coublucq, Daniel, Econometric Analysis of Productivity with Measurement Error: Empirical Application to the US Railroad Industry, June 2013.
94 Coublucq, Daniel, Demand Estimation with Selection Bias: A Dynamic Game Approach with an Application to the US Railroad Industry, June 2013.
93 Baumann, Florian and Friehe, Tim, Status Concerns as a Motive for Crime?, April 2013.
92 Jeitschko, Thomas D. and Zhang, Nanyun, Adverse Effects of Patent Pooling on Product Development and Commercialization, April 2013. Published in: The B. E. Journal of Theoretical Economics, 14 (1) (2014), Art. No. 2013-0038.
91 Baumann, Florian and Friehe, Tim, Private Protection Against Crime when Property Value is Private Information, April 2013. Published in: International Review of Law and Economics, 35 (2013), pp. 73-79.
90 Baumann, Florian and Friehe, Tim, Cheap Talk About the Detection Probability, April 2013. Published in: International Game Theory Review, 15 (2013), Art. No. 1350003.
89 Pagel, Beatrice and Wey, Christian, How to Counter Union Power? Equilibrium Mergers in International Oligopoly, April 2013.
88 Jovanovic, Dragan, Mergers, Managerial Incentives, and Efficiencies, April 2014 (First Version April 2013).
87 Heimeshoff, Ulrich and Klein Gordon J., Bargaining Power and Local Heroes, March 2013.
86 Bertschek, Irene, Cerquera, Daniel and Klein, Gordon J., More Bits – More Bucks? Measuring the Impact of Broadband Internet on Firm Performance, February 2013. Published in: Information Economics and Policy, 25 (2013), pp. 190-203.
85 Rasch, Alexander and Wenzel, Tobias, Piracy in a Two-Sided Software Market, February 2013. Published in: Journal of Economic Behavior & Organization, 88 (2013), pp. 78-89.
84 Bataille, Marc and Steinmetz, Alexander, Intermodal Competition on Some Routes in Transportation Networks: The Case of Inter Urban Buses and Railways, January 2013.
83 Haucap, Justus and Heimeshoff, Ulrich, Google, Facebook, Amazon, eBay: Is the Internet Driving Competition or Market Monopolization?, January 2013. Published in: International Economics and Economic Policy, 11 (2014), pp. 49-61.
82 Regner, Tobias and Riener, Gerhard, Voluntary Payments, Privacy and Social Pressure on the Internet: A Natural Field Experiment, December 2012.
81 Dertwinkel-Kalt, Markus and Wey, Christian, The Effects of Remedies on Merger Activity in Oligopoly, December 2012.
80 Baumann, Florian and Friehe, Tim, Optimal Damages Multipliers in Oligopolistic Markets, December 2012.
79 Duso, Tomaso, Röller, Lars-Hendrik and Seldeslachts, Jo, Collusion through Joint R&D: An Empirical Assessment, December 2012. Forthcoming in: The Review of Economics and Statistics.
78 Baumann, Florian and Heine, Klaus, Innovation, Tort Law, and Competition, December 2012. Published in: Journal of Institutional and Theoretical Economics, 169 (2013), pp. 703-719.
77 Coenen, Michael and Jovanovic, Dragan, Investment Behavior in a Constrained Dictator Game, November 2012.
76 Gu, Yiquan and Wenzel, Tobias, Strategic Obfuscation and Consumer Protection Policy in Financial Markets: Theory and Experimental Evidence, November 2012. Forthcoming in: Journal of Industrial Economics under the title “Strategic Obfuscation and Consumer Protection Policy”.
75 Haucap, Justus, Heimeshoff, Ulrich and Jovanovic, Dragan, Competition in Germany’s Minute Reserve Power Market: An Econometric Analysis, November 2012. Published in: The Energy Journal, 35 (2014), pp. 139-158.
74 Normann, Hans-Theo, Rösch, Jürgen and Schultz, Luis Manuel, Do Buyer Groups Facilitate Collusion?, November 2012.
73 Riener, Gerhard and Wiederhold, Simon, Heterogeneous Treatment Effects in Groups, November 2012. Published in: Economics Letters, 120 (2013), pp 408-412.
72 Berlemann, Michael and Haucap, Justus, Which Factors Drive the Decision to Boycott and Opt Out of Research Rankings? A Note, November 2012.
71 Muck, Johannes and Heimeshoff, Ulrich, First Mover Advantages in Mobile Telecommunications: Evidence from OECD Countries, October 2012.
70 Karaçuka, Mehmet, Çatik, A. Nazif and Haucap, Justus, Consumer Choice and Local Network Effects in Mobile Telecommunications in Turkey, October 2012. Published in: Telecommunications Policy, 37 (2013), pp. 334-344.
69 Clemens, Georg and Rau, Holger A., Rebels without a Clue? Experimental Evidence on Partial Cartels, April 2013 (First Version October 2012).
68 Regner, Tobias and Riener, Gerhard, Motivational Cherry Picking, September 2012.
67 Fonseca, Miguel A. and Normann, Hans-Theo, Excess Capacity and Pricing in Bertrand-Edgeworth Markets: Experimental Evidence, September 2012. Published in: Journal of Institutional and Theoretical Economics, 169 (2013), pp. 199-228.
66 Riener, Gerhard and Wiederhold, Simon, Team Building and Hidden Costs of Control, September 2012.
65 Fonseca, Miguel A. and Normann, Hans-Theo, Explicit vs. Tacit Collusion – The Impact of Communication in Oligopoly Experiments, August 2012. Published in: European Economic Review, 56 (2012), pp. 1759-1772.
64 Jovanovic, Dragan and Wey, Christian, An Equilibrium Analysis of Efficiency Gains from Mergers, July 2012.
63 Dewenter, Ralf, Jaschinski, Thomas and Kuchinke, Björn A., Hospital Market Concentration and Discrimination of Patients, July 2012 . Published in: Schmollers Jahrbuch, 133 (2013), pp. 345-374.
62 Von Schlippenbach, Vanessa and Teichmann, Isabel, The Strategic Use of Private Quality Standards in Food Supply Chains, May 2012. Published in: American Journal of Agricultural Economics, 94 (2012), pp. 1189-1201.
61 Sapi, Geza, Bargaining, Vertical Mergers and Entry, July 2012.
60 Jentzsch, Nicola, Sapi, Geza and Suleymanova, Irina, Targeted Pricing and Customer Data Sharing Among Rivals, July 2012. Published in: International Journal of Industrial Organization, 31 (2013), pp. 131-144.
59 Lambarraa, Fatima and Riener, Gerhard, On the Norms of Charitable Giving in Islam: A Field Experiment, June 2012.
58 Duso, Tomaso, Gugler, Klaus and Szücs, Florian, An Empirical Assessment of the 2004 EU Merger Policy Reform, June 2012. Published in: Economic Journal, 123 (2013), F596-F619.
57 Dewenter, Ralf and Heimeshoff, Ulrich, More Ads, More Revs? Is there a Media Bias in the Likelihood to be Reviewed?, June 2012.
56 Böckers, Veit, Heimeshoff, Ulrich and Müller Andrea, Pull-Forward Effects in the German Car Scrappage Scheme: A Time Series Approach, June 2012.
55 Kellner, Christian and Riener, Gerhard, The Effect of Ambiguity Aversion on Reward Scheme Choice, June 2012.
54 De Silva, Dakshina G., Kosmopoulou, Georgia, Pagel, Beatrice and Peeters, Ronald, The Impact of Timing on Bidding Behavior in Procurement Auctions of Contracts with Private Costs, June 2012. Published in: Review of Industrial Organization, 41 (2013), pp.321-343.
53 Benndorf, Volker and Rau, Holger A., Competition in the Workplace: An Experimental Investigation, May 2012.
52 Haucap, Justus and Klein, Gordon J., How Regulation Affects Network and Service Quality in Related Markets, May 2012. Published in: Economics Letters, 117 (2012), pp. 521-524.
51 Dewenter, Ralf and Heimeshoff, Ulrich, Less Pain at the Pump? The Effects of Regulatory Interventions in Retail Gasoline Markets, May 2012.
50 Böckers, Veit and Heimeshoff, Ulrich, The Extent of European Power Markets, April 2012.
49 Barth, Anne-Kathrin and Heimeshoff, Ulrich, How Large is the Magnitude of Fixed-Mobile Call Substitution? - Empirical Evidence from 16 European Countries, April 2012. Forthcoming in: Telecommunications Policy.
48 Herr, Annika and Suppliet, Moritz, Pharmaceutical Prices under Regulation: Tiered Co-payments and Reference Pricing in Germany, April 2012.
47 Haucap, Justus and Müller, Hans Christian, The Effects of Gasoline Price Regulations: Experimental Evidence, April 2012.
46 Stühmeier, Torben, Roaming and Investments in the Mobile Internet Market, March 2012. Published in: Telecommunications Policy, 36 (2012), pp. 595-607.
45 Graf, Julia, The Effects of Rebate Contracts on the Health Care System, March 2012, Published in: The European Journal of Health Economics, 15 (2014), pp.477-487.
44 Pagel, Beatrice and Wey, Christian, Unionization Structures in International Oligopoly, February 2012. Published in: Labour: Review of Labour Economics and Industrial Relations, 27 (2013), pp. 1-17.
43 Gu, Yiquan and Wenzel, Tobias, Price-Dependent Demand in Spatial Models, January 2012. Published in: B. E. Journal of Economic Analysis & Policy, 12 (2012), Article 6.
42 Barth, Anne-Kathrin and Heimeshoff, Ulrich, Does the Growth of Mobile Markets Cause the Demise of Fixed Networks? – Evidence from the European Union, January 2012. Forthcoming in: Telecommunications Policy.
41 Stühmeier, Torben and Wenzel, Tobias, Regulating Advertising in the Presence of Public Service Broadcasting, January 2012. Published in: Review of Network Economics, 11/2 (2012), Article 1.
Older discussion papers can be found online at: http://ideas.repec.org/s/zbw/dicedp.html
ISSN 2190-9938 (online) ISBN 978-3-86304-152-6
