Economics journals attract better submissions related to editors’ research��I.P.L. Png and Enqi Teo

Progress in science

• Two paradigms
• Ideal: Marketplace of ideas: Fair competition among new ideas---the best ideas win (Holmes 1919)
• Small circle (Merton 1968)
• Biomedical sciences:
• Death of star scientist => Rise of new topics which are better cited (Azoulay et al. 2019)
• Funding: NIH panels favour proposals in related areas, but those produced better cited papers (Li, 2017):
• Journals: Editor effects (favor related research) tiny relative to journal mission and trends in research (Krieger et al. 2024)

​

2

Research question

• What about economics? “Softer”, more discretion
• Less disciplined by empirical investigation
• Broader scope (imperialist)---climate change, history, politics,…
• Previous research revealed editor effects: Scholars connected to editors more likely to get published: (Laband and Piette 1994; Brogaard et al. 2014; Colussi 2018; Heckman and Moktan 2020; Ductor and Visser 2022)
• Our questions
• Do editors’ own research interests affect journal content?
• How?

3

Research question

• Mechanism
• Authors cater to editor interests (“author clientele”) (Heckman and Moktan 2020) or
• Editors blatantly favor related research (“editor favoritism”) (Colussi 2018; Ductor and Visser, 2022)

4

Findings

• Editor specialization affects likelihood of manuscript getting into review and then eventually receiving R&R
• Mechanism is author clientele, not editor favoritism

​

5

Additional findings

• Nature of author clientele
• No increase in overall number of related submissions, rather change in composition towards submissions more likely to meet threshold for publication
• Methodology/style rather than raw quality
• More productive authors more likely to submit related manuscripts

6

Contributions

• Progress of scientific research---particularly role of journals in economics
• Not perfect marketplace of ideas
• Gatekeeping favoring “related” research---maybe to further journal (rather than editors’ personal) agenda
• Scientific careers
• Matthew Effect: Better published scholars more attuned to journal preferences
• Economics of curation/taste: News---media slant responds to audience taste

​

7

Outline

• Model
• Empirical strategy
• Data
• Results
• Mechanism
• Discussion
• Future work

8

Editorial triage: �Related vis-à-vis unrelated submission

9

Editor: Initial screening

10

Editor: Initial screening cont’d

11

Referee: Recommendation

12

Referee: Recommendation cont’d

13

Outline

• Model
• Empirical strategy
• Data
• Results
• Mechanism
• Discussion
• Future work

14

Empirical strategy

• Exploit rotation of editors in journal
• Editor = any person with decision authority (Ductor and Visser 2022)
• Co-Editor (JEEA), Editor (QJE), Editor (REStat), Managing Editor (REStud)
• Editor specialization = field in which they published most papers in ten years up to year of manuscript under review.

​

15

Editor interests

16

Identification

• Reverse causality: Unlikely, editor papers written years before publication in turn before manuscript submission
• Confounding factors---trends in fields, trends in journals: Control for fixed effects for field x journal, field x year, and journal x year

​

17

Outline

• Model
• Empirical strategy
• Data
• Results
• Mechanism
• Discussion
• Future work

18

Sources

• Manuscripts (Card and Dellavigna, 2020)
• JEEA, QJE, REStat, REStud, 2005-2013
• Classified into 13 fields by JEL code (Card and Dellavigna, 2013)
• Editors (any person with decision authority (Ductor and Visser 2022))
• 45 editors: REPEC pages
• Classified publications into 13 fields (Card and Dellavigna, 2013)

​

​

​

​

19

Data

• Each record is a manuscript
• Possibly multiple fields, eg, development, labor, public economics: represent by equal fractions
• Possibly submitted to multiple journals
• Editors by year
• Specialization = field in which they published most papers in the 10 years up to that year
• Merge manuscripts with editors by journal and year
• Specialization indicator = 1 if at least one editor specialized in half or more of manuscript fields

20

Summary statistics

21

Sent for review (not desk rejected)

22

Sent for review (not desk rejected)

23

Sent for review (not desk rejected)

24

Outline

• Model
• Empirical strategy
• Data
• Results
• Mechanism
• Discussion
• Future work

25

Editor decisions (max likelihood logit)

26

Sent to review: Specification

27

Outline

• Model
• Empirical strategy
• Data
• Results
• Mechanism
• Discussion
• Future work

28

Editorial process: Mechanism

29

Editor favoritism, by field

30

Author clientele, by field

31

Robustness tests

32

Panel data estimates

33

Authors: Submitted manuscript related to editor specialization

34

Outline

• Model
• Empirical strategy
• Data
• Results
• Mechanism
• Consequences
• Discussion
• Future work

35

Cumulative effect up to R&R

• Cumulative effect = sent for review x positive report conditional on sent for review x R&R conditional on positive report
• Bootstrapped
• t-test: p < 0.001

36

Citations

37

Citations

38

Outline

• Model
• Empirical strategy
• Data
• Results
• Mechanism
• Consequences
• Discussion
• Future work

39

Findings

• Editor favoritism: Ruled out
• Author clientele: Sorting on pure quality or style and methodology?
• Results controlling for author characteristics are similar to those without controls
• Heterogeneity by field
• Citations of related papers were not higher than unrelated papers (with caution)

40

Policy implications

• Crowding out: Given limited journal space, papers outside editor research interests would be crowded out
• Implications
• Direction of research is channelled
• [Matthew effect] More productive authors are privileged
• Academic careers are constrained
• Interventions
• Increase diversity of editors
• Shorter terms
• Mandatory rotation

41

Outline

• Model
• Empirical strategy
• Data
• Results
• Mechanism
• Discussion
• Future work

42

Future work

• Author clientele only makes sense if there is favoritism
• Reasons for changes in editors: journals appointed editors to advance particular agenda (Onder et al. 2020)

​

43

End

44

Future work

• Survey experiment
• [[On authors]]
• Objectives
• Define specializations more finely
• Avoid confounds: (i) journals chose editors to advance agenda; (ii) author clientele
• What else?
• Other suggestions?
• Worth the cost of experiment?

45

Alternative explanations

• Within-journal manuscript assignment---journals assign manuscripts to editors with relevant expertise. We cast treatment at journal level (one or more specializing editors) not editor.
• Journals appoint editors to manage rising topics
• Development: Editors help junior scholars who work on related research. Estimated effect of editor specialization is almost the same with and without controlling for author characteristics.

​

46