Report on the Status of Social Media Use in Economics and Recommendations for Best Practices

Ad Hoc Committee on Social Media: Tim Bresnahan (Chair), Belinda Archibong, Florian Ederer, Sarah Jacobson and Corrine Low

Status of Social Media Use in Economics
Recommendations for Best Practices in Use
for Individuals, Institutions, and the Association:
Final Report, January 2, 2025

1.   Introduction

This Committee was formed in May 2024 by the President of the AEA, Janet Currie. Our charge was twofold. First, assess and document the opportunities offered by different social media platforms to our profession as well as the problems arising therefrom. Second, prepare a set of recommended best practices for Economists, for the AEA and other associations, and for Departments and other employers of economists.

The literature has shown social media to have both potentially positive and negative effects for academics. Positive effects include facilitating professional network formation, increasing citations and awareness of research work and opportunities for minoritized individuals to find community.[1] These benefits can vary depending on group identity; for example, in the context of academic networking through social media, a positive gender gap is identified for women as well as a racial gap favoring white men[2]. Negative effects include exposure to abuse and harassment, which may target some groups more than others[3] and is exacerbated under anonymity.[4] These findings are against the background of broad general studies of social media impacts, though this report is primarily concerned with professional use.[5]

To understand how these varied effects arise in Economics social media use, and to form strategies to capture benefits while reducing harms, we surveyed the AEA membership, interviewed several economists with relevant experiences and corresponded with others.[6]

Those efforts reveal several facts that form the basis for our policy recommendations. These are explained in more detail below and summarized here.

  • There is a valuable role for social media to play in scholarly communication in economics.
  • Positive impacts arise through connection of scholarship and scholars.
    • Learning about new work and having one's work learned about.
    • Finding new collaborators and even positions.
    • Discussing methods, research, and current economic events.
    • Particularly positive for people who may be considered "outsiders."
    • #EconTwitter was once a center of positive interaction but has declined.
  • There are also negative effects of social media use in Economics
    • These often fall outside the control of individuals impacted and can be severe.
    • These negative impacts often take the form of abuse, harassment, rumor-mongering, or other harmful forms of interaction.
    • These impacts are substantial enough that the overall impact of social media use in economics is far less satisfactory than one's own use.
    • The negative effects are worsened by anonymous posting.
    • The negative effects particularly strike women, people of color, and LGBTQ economists.
      • Social media attacks may form a significant barrier to their participation in our profession.
    • The costs of online harassment are large and cannot be easily evaded by individuals, as many arise without visits to the relevant social media site.
      • They spill out into time budgets and academic assessment.
  • The positive outcomes and the negative ones arise on different social media platforms.

Our interviews and open-ended questions also drew some specific complaints about the current state of social media use in economics and hundreds of suggestions. In decreasing order of frequency these include:

  • The Economics Job Market Rumors site is destructive for economics: the AEA should provide an alternative site for reporting and learning job market facts.
  • The substantive and methods discussions once found on #EconTwitter were valuable: the AEA should encourage and support efforts to create such discussions.
  • A number of practical suggestions for Associations, Departments and Scholars to get the benefits of social media while avoiding the problems.

We take this advice seriously in forming a body of potential best practices for students, scholars, and departments and suggest AEA actions.

A small but not trivial group of economists oppose any action by the AEA concerning social media. Most of these assume that the only feasible actions are restrictions on what Economists may say. We take the views of this group seriously, and do not recommend any restrictions on speech beyond the application of the existing AEA Code of Professional Conduct to the social media domain. The Committee is highly aware of the importance of free and open scientific discussion.

These findings lead us to a set of best practices we recommend to Economists as individuals, to departments and other employers, and to the AEA and other professional organizations.

2.   Background: intro to social media in our profession

The social media used by economists vary somewhat but share a number of common features.

All social media exist to let people communicate and/or share content with others.

Most social media platforms are advertising supported. The AEA sponsored platforms are not. Revenue from advertising rises with users who read content, and with their length of stay. User supplied content -- posts -- create revenue only through attracting readers who might be shown advertisements. Some platforms pay posters who succeed in attracting readers to view advertisers. Financial support for a platform arises directly from those who read content and indirectly from those who post content.

Social media have positive feedback loops across different constituencies, including users who post messages, people (users or not) who are written about in posts, readers of posts, and advertisers (if any.) A site or an account with more posts attracts more readers, which increases the incentive to post .... More readers make a social media platform more attractive to advertisers. These network effects result in substantial (social) economies of scale.

The feedback loops matter for the positive Economics of social media markets. For example, the AEA's effort, launched five years ago, to move many users to the EconTrack platform for job market information and the EconSpark platform for economics-related discussion was difficult because of established feedback loops around existing platforms. A new equilibrium of posting and reading on the new platforms was not established because few users wanted to migrate to a new platform with little activity.

Different social media used by economists vary in purpose, technology, structure, rules and the options available to users. Key features of rules and options include "account verification," "content moderation," and "platform balance."

Social media have different policies about account setup. Some make an effort to verify that accounts are genuine, with varying tools and degrees of effectiveness. As a result, most platforms have some degree of anonymous participation. Of course, scholars on social media often have an incentive to link their social media presence to scholarly identity, e.g. so they can post their own new research.

Yet not all posters and commenters are who they appear to be. Some bogus accounts have political or commercial interests, including pushing false information into economic discussions, engaging in bogus debate (e.g. "sealioning"), or currying online relationships in order to defraud participants. New forms of bad actors are invented regularly.

One form of anonymous social media account is not human. These "bots" can serve to repost particular content in order to increase the chance it will be widely read, i.e., go "viral". Or, programmed by the same "farms," multiple bots can introduce and/or amplify the same content.

"Content moderation" is carried out to varying degrees by different social media platforms. One form of content moderation has a simple commercial premise: blocking posts which are advertisements, but not paid ads. Other forms of content moderation seek to prevent illegal activity, e.g. displays of child sexual abuse material, or simply to prevent content that drives users away, such as pornography, graphic displays of violence, and so on. Effective content moderation is difficult and expensive.

Connections between users on a social media platform can be bilateral (e.g. Facebook "friends") or unilateral (e.g. Twitter or BlueSky "follows"). The platform's algorithms determine what a reader sees, to some degree, that depends on the platform, as do the reader's and others' choice of connections. The same applies to a poster's posts. Platforms vary considerably in how much control over this they give to posters, the algorithms, and readers. Some let readers block posts or posters they find offensive, an alternative to content moderation. Others offer more authority to posters and to the algorithm. Since a successful platform attracts both posters and readers (sometimes the same person) these "platform balance" policies matter.

The platforms used by economists vary not only along those policy dimensions but also in their focus.

Some social media used by Economists have a discussion focus. These include Twitter/X, BlueSky, Mastodon, EconSpark and Threads. On any of these, economists can post remarks about research, policy, economic controversies, data, methodologies, or the current economic state of the world or of particular countries, markets or industries as well as topics unrelated to economics. These discussion-focused platforms hold the potential to be the town square for Economics.

There are differences among these platforms. Mastodon, EconSpark and Threads have never had a large positive feedback loop among economists. Twitter/X and BlueSky are very similar technically, but have very different policies on content moderation and the authority delegated to users. Twitter/X has a platform tilt to posters, BlueSky to readers. Many users of both platforms note that these policy differences appear to lead to much larger presence of bots and bad actors on Twitter/X than on BlueSky.

There are also a number of platforms which, while they may be used for other purposes as well, play an important role in technical discussion among economists. The relevant discussions on these sites include econometric, statistical (including statistical package) and mathematical discussion. Such platforms include Reddit, which has a strong system of moderators to keep discussions on track and Stack Exchange, where experienced users help prevent off-topic or badly phrased questions (There are similar fora dedicated to particular statistical packages with similar structures.)

Sites for participants in labor markets include LinkedIn, Economics Job Market Rumors (EJMR), and EconTrack. LinkedIn serves a broader labor market and appears to be used more by non-academic than by academic economists. It is useful for posting new research work, posting a mini-cv, and declaring interests; in some circles it is also commonly relied upon for maintaining professional connections. EJMR (with a lot of use) and EconTrack (with AEA sponsorship but hardly any use) provide venues for users to give and receive updates on the state of the economics job market, particularly the new-Ph.D. market.

A number of social media platforms focus more on life than on work life, though some Economists use them for both. These include Facebook and Instagram. Sites like Facebook have a structure of "friends" (bilateral) rather than "followers" (which can be unilateral.)

3.   Information Gathering

The Committee sent a survey to the AEA membership. In addition to questions with restricted (e.g. scale) responses, we solicited comments in open ended questions, offered survey respondents the opportunity to be interviewed, and had a temporary email address for potential interviewees and other correspondence. We received useful information through all these channels. We are grateful to all our colleagues for their efforts informing us.

Our survey was sent to the AEA's 17,706 members with a follow-up email in the summer of 2024. The Survey Instrument is in the Appendix. There were 1375 replies. Two open-ended questions asked for broad comments, drawing almost 900 answers. Many of these were long and pointed.

Reading the comments, we are led to conclude that our survey oversampled Economists with knowledge of and opinions about social media. Many expressed strong views. Others, not as many, appear to have answered the survey out of professional obligation, writing that they had no knowledge that would be helpful on this topic. A similar selection is likely to apply to our interviews and the emails we received (i.e., strong views on all sides). Accordingly, we are going to rely on the survey more for identifying areas of broad consensus in our profession rather than quantifying Economists' view on close questions and on the detailed communications as telling us about what can happen in professional use of social media.

The survey asked respondents about themselves, about their social media use and goals, and about satisfaction with their own use and with the state of use in the profession more broadly. Questions covered both social media in general and anonymous social media. We also asked about the appropriateness of a list of social media behaviors in a professional context.

We interviewed 8 colleagues and received emails from dozens.

4.  Discussion of Survey and Interview Results

a.  Who Responded

We show some demographics of respondents in Table 1.

Table 1 Proportion of Respondents in various categories

American Indian or Alaska Native*

0.002

 

Female*

0.289

Asian*

0.130

 

Gender Other**

0.008

Black*

0.017

 

LGBTQ*

0.075

Hispanic or Latino*

0.085

 

 

 

Middle Eastern or North African*

0.029

 

Student

0.079

Native Hawaiian or Pacific Islander*

0.002

 

Nonacademic

0.142

White*

0.780

 

Employer Rank Top 50

0.566

Other Race or Ethnicity

0.021

 

 

 

 

 

 

 

 

N

1375

 

 

 

* Race/Ethnicity, Gender, and LGBTQ community membership also include “prefer not to say” (not shown.)
** Gender other includes non-binary, agender, and “something else.”

Respondents came from a wide variety of primary fields, with labor, macro, development public, behavioral and IO making up the top six.

The survey let respondents skip questions. Four respondents left the survey early and 306 skipped at least one question.  Some questions offered “prefer not to say” as an option: 118 preferred not to state their race or ethnicity, 58 preferred not to state their gender, and 100 preferred not to say if they were a member of an LGBTQ community.  (The “prefer not to say” answer tends to appear more than once for the same respondent if it appears at all.)

b.  Social Media Used by Economists

We provided a list of 12 social media platforms to prompt answers about which ones are used for professional purposes.[7]  Respondents could also list other platforms they used in open-ended questions, and many did.  

Professional social media use by survey respondents is frequent, evidenced by responses to the query.  We asked “which social media you use in your professional life as an economist” and whether the use was reading or posting, and whether past or present. Results for the most common behavior, reading social media in the present, are in Table 2.  Just over 50% (.504) of survey respondents currently read Twitter/X,[8] and just under 50% (.496) currently read LinkedIn. Other sites are far less frequently used or  read, though technical-advice sites Stack Exchange and Reddit are read by between 10 and 20% of respondents.  The same person can use multiple social media: Overall, 81.67% of respondents are current readers of some social media.[9]  Many use multiple social media.  

Table 2 Proportion of Respondents Currently Reading Social Media Platforms 

Platform

 

 

Platform

 

 Twitter/X

0.504

 

 

 LinkedIn

0.496

 

 BlueSky

0.068

 Facebook

0.265

 

 EJMR

0.058

 Instagram

0.170

 

 TikTok

0.043

 Stack Exchange

0.163

 

EconTrack

0.019

 Reddit

0.132

 

 Mastodon

0.015

 

 

 

EconSpark

0.013


A version of this Table with four proportions per platform (currently reading, currently posting, have read in the past, have posted in the past) appears in the Appendix.  Unsurprisingly, more Economists report that they are frequent readers of social media than frequent posters.  While recollections of past behavior are weak in surveys, we do note two changes over time.  Almost all the sites get more yes on “currently read” and “currently post” than on “have read/posted in the past”  The exception is EJMR, with “currently” falling below “in the past.” This is consistent with other indicators, such as Google Trends results on searches for “econjobrumors.com” which show seasonal fluctuations and a decline over the last eight years or so.

Graph showing worldwide searches for "econjobrumors"

Worldwide searches for “econjobrumors.com” from 1/1/2010 to 11/13/2014.

c.   Social Media Use Goals in Economics  

Social media are a useful tool for sharing research, for example through posting or reading announcements of new working papers.  In Table 3, we show the proportion of respondents who say they use social media, at least infrequently, for each of a list of goals.  The most common uses are learning about others’ research (79%), sharing one’s own research (67%) and expanding one’s network (63%).  It is unsurprising that these classic academic activities are important.  A variety of specific connections goals (connect with journalists or professional associations or learn about opportunities) make up less frequent uses.  Academic labor market uses and professional gossip follow, with personal advice less frequent though not uncommon. 

Table 3 Goals of Social Media Use

Goal, used frequently or infrequently

Proportion

Used to share research

0.666

Used to learn about others research

0.790

Used to expand network

0.626

Used to connect with professional associations

0.429

Used to connect with journalists

0.313

Used to learn about employment and research opportunities

0.427

Used to advertise employment or research opportunities

0.371

Used to receive professional advice

0.374

Used to give professional advice

0.294

Used to catch up on professional gossip or rumors

0.420

Used to receive personal advice

0.193

Used to give personal advice

0.183

 

We asked respondents which platform they used for each of those purposes.  We report the breakdown for each of seven platforms in Table 4, again summing frequent and infrequent use.  The core academic goals of sharing research and learning about it are pursued most often on Twitter and LinkedIn, with a bit of the activity on Facebook and only smatterings elsewhere.  That same pattern holds for most of the other goals, with one exception.  The "rumors" platform EJMR is used by just over six percent of respondents to catch up on gossip and rumors, and not by over one percent for any of the other purposes. 

Table 4 Platforms used for different goals

 

Twitter/X

Facebook

LinkedIn

Instagram

StackExchange

Reddit

EJMR

...share research

.385

.068

.218

.007

.004

.003

.001

...learn others research

.483

.057

.201

.008

.014

.008

.006

...expand network

.265

.045

.291

.01

.004

.001

.001

...connect with professional associations

.185

.034

.138

.006

.003

.001

.002

...connect with journalists

.148

.017

.068

.004

.003

.001

.001

...learn about employment and research opportunities

.176

.015

.161

.001

.002

.001

.007

...advertise employment or research opportunities

.167

.02

.133

.002

.001

.001

.001

...receive professional advice

.188

.024

.062

.004

.009

.006

.010

...give professional advice

.121

.025

.055

.004

.004

.003

.004

...catch up on professional gossip or rumors

.197

.027

.050

.005

.004

.007

.061

...receive personal advice

.047

.023

.026

.007

.004

.004

.002

...give personal advice

.042

.023

.027

.005

.003

.002

.002

 

d. Economists' Satisfaction with Professional Social Media

Our survey respondents have a mostly neutral-to-positive level of satisfaction with their own use of social media.  However, they have a mostly neutral-to-negative level of satisfaction with the state of social media use in our profession. 

 

How satisfied are you with …

 

the results of your social media use?

the state of social media use            in our profession?

 

Percent

Percent

Satisfied or Very Satisfied

36.89

14.00

Neutral

46.60

48.91

Unsatisfied or Very Unsatisfied

16.50

37.10

 

 

 

           

This difference foregrounds the dual effects associated with professional use of social media.  Some elements are under the individual scholar’s control, and yield benefits.  Other elements, including many of the negative impacts, are not easily controlled by the individual scholar.

There is some variation in these assessments by Gender and LBGTQ community membership.  While men and women are similar, people who identify with another gender (non-binary, agender, etc.) are less satisfied than others, as are LBGTQ respondents.[10]  Still, each of these groups tends to report more satisfaction than dissatisfaction with their own use.  See Table 5.

Table 5 Gender, LGBTQ community, and Own Social Media Use Satisfaction

How satisfied are you with the results of your social media use?

What is your gender?

Are you a member of an LGBTQ Community?

Female

Male

Other*

PNTS**

No

Yes

Total

Very satisfied

7.08

7.54

0.00

13.73

7.49

5.49

7.61

Satisfied

29.18

29.81

20.00

23.53

30.24

25.27

29.29

Neutral

50.71

45.01

40.00

45.10

46.26

46.15

46.60

Unsatisfied

9.92

12.90

20.00

9.80

11.75

18.68

11.97

Very unsatisfied

3.12

4.74

20.00

7.84

4.27

4.40

4.53

"Other" is Non-Binary / Agender / Something else
"PNTS" is Prefer not to say (shown only for gender.)
Total includes prefer not to say for both categories

 

Respondents are generally less satisfied with the state of social media use in our profession overall than with their own use.  There are also differences across groups. Women are less satisfied than men, as are LGBTQ Economists and gender-nonconforming Economists.  See Table 6.

Table 6 Gender, LGBTQ community, and satisfaction with the state of Social Media Use profession-wide

How satisfied are you with the state of social media use in our profession?

What is your gender?

Are you a member of an LGBTQ Community?

Female

Male

Other*

 

No

Yes

Total

Very satisfied

0.54

1.23

9.09

 

1.14

1.09

1.13

Satisfied

11.96

13.73

0.00

 

13.22

6.52

12.72

Neutral

50.27

49.11

27.27

 

50.44

38.04

49.51

Unsatisfied

30.16

24.89

45.45

 

25.39

41.30

26.58

Very unsatisfied

7.07

11.05

18.18

 

9.81

13.04

10.05


"Other" is Non-Binary / Agender / Something else
"PNTS" is Prefer not to say (shown only for gender.)
Total includes prefer not to say for both categories

 

There is little variation across other observable groups with regard to satisfaction with their own social media use or use in our profession.  We looked at race/ethnic groups, date of highest degree, nature of professional appointment, field, and student status.  With little variation, we do not display tables with breakdowns by those variables. 

Turning now to anonymous social media use, respondents have a more negative view of its use in our profession than of social media generally.  59% find the use and existence of anonymous social media to be detrimental or very detrimental.  There are also differences across groups in the assessment of anonymous social media by  gender and LGBTQ status:

Table 7 Gender, LGBTQ community,  and the impact on Economics of anonymous social media use

Anonymous Social Media in Economics is …

What is your gender?

Are you a member of an LGBTQ Community?

Female

Male

Other*

 

Yes

No

Total

Very beneficial

0.82

1.59

0.00

 

1.11

1.32

1.75

Beneficial

5.16

12.24

0.00

 

2.22

10.33

10.55

Neutral

23.10

33.56

36.36

 

37.78

30.01

30.37

Detrimental

43.48

36.51

54.55

 

41.11

38.92

37.66

Very detrimental

27.45

16.10

9.09

 

17.78

19.42

19.67


"Other" is Non-Binary / Agender / Something else
"PNTS" is Prefer not to say (shown only for gender.) 
Total includes prefer not to say for both categories

These group differences are consistent with respondents’ assessment that harassment and other negative effects of social media and anonymous use fall primarily on minoritized groups, evidence on which we present in the coming sections.

e.   The Impacts of social media in our profession

We next examine a series of questions about the impact of social media within our profession. These questions let respondents evaluate 11 possible impacts, with space for write-ins of any other impacts.  In the following table, we have put the positive impacts on the left and the negative ones on the right.  The proportions are respondents who view the impact as present, either somewhat or a great deal.  Broadly speaking, there is widespread agreement that both the positive and negative impacts are present.

Impact of social media (proportion of respondents saying somewhat/a great deal)

Allows scholars to connect more extensively/intensively with each other

0.911

Allows unproductive and/or harmful conversations (e.g., gossip) to spread

0.929

 

Provides a venue for scholars and journalists to connect

0.854

Increases anxiety or other mental health issues among scholars

0.878

 

Provides a venue for scholars and policymakers to connect

0.676

Encourages scholars to waste time

0.860

 

Enables scientific discussion of research results and methodologies to improve current and future research

0.722

Provides a locus for discrimination and /or harassment

0.856

 

Allows equal access of economists to participate in important discussions within the profession, regardless of job or seniority

0.639

Increases stress on particular groups of economists, such as racial minorities, women, etc

0.811

 

Allows people to use “whisper networks” in a positive way to help other scholars stay safe or avoid harassment or discrimination

0.626

 

 

 

N

1375

 

 

 

 

To the extent there is variation across these reported impacts, we note that connectivity among scholars is the most-frequently cited positive impact, and unproductive/harmful conversations are the most frequent negative.  Both positive and negative equilibrium effects appear, with discrimination and harassment, plus their consequences in terms of stress on the groups discriminated against and mental health issues for those harassed appearing on the downside.

We asked on which social media platforms these impacts, both positive and negative, tended to fall.  This was an open-ended question where respondents could offer multiple answers; in most cases it was possible for us to identify all platforms a respondent referred to.  In the table below,  we show the three most frequently-mentioned platforms.  The “Obs” column is the number of respondents who cited the impact.  Thus, for example, 1270 respondents cited more scholarly connection as a benefit, and of these 1270 respondents 1.4% wrote that this impact arises on EJMR, while 42.8% wrote that it arises on Twitter/X.

Table 8 Platforms on which impacts arise

 

Obs

EJMR

LinkedIn

Twitter

Allows scholars to connect more extensively/intensively with each other

1270

0.014

0.156

0.428

Provides a venue for scholars and journalists to connect

1229

0.007

0.117

0.341

Provides a venue for scholars and policymakers to connect

1069

0.004

0.115

0.268

Enables scientific discussion of research results and methodologies to improve research

1058

0.02

0.084

0.405

Allows equal access of economists to participate in important discussions within the profession, regardless of job or seniority

991

0.025

0.08

0.339

Allows people to use “whisper networks” in a positive way to help other scholars stay safe or avoid harassment or discrimination

1156

0.041

0.017

0.123

Encourages scholars to waste time

1292

0.142

0.149

0.393

Increases anxiety or other mental health issues among scholars

1259

0.167

0.091

0.288

Allows unproductive and/or harmful conversations (e.g., gossip) to spread

1298

0.265

0.062

0.276

Provides a locus for discrimination and /or harassment

1242

0.245

0.039

0.18

Increases stress on particular groups of economists, such as racial minorities, women, etc.

1227

0.207

0.036

0.161

 

There is a marked pattern in this Table that illustrates important differences in the role played by the different platforms.  In the upper part of the Table, where the impacts are positive, one sees greater impact on Twitter/X than on LinkedIn and far greater impact of either of them than on EJMR.  In the bottom part of the Table, where negative impacts are identified, EJMR plays a more important role.

In short, while social media brings some negative impacts, which we explore further in the next section, many respondents find it highly useful for specific work-related purposes, implying that there would be gains from improving the social media climate.  Positive and negative impacts arise on different platforms.

f.   Harassment on Social Media

Our survey asked respondents whether they, as individuals, had been bullied or harassed on social media, and, if so, on which platform(s).  19.8% of respondents had.[11]  We looked at the variation across groups:

Table 9 Gender, LGBTQ community, and harassment

Have you been bullied, harassed, etc. on social media?

What is your gender?

Are you a member of an LGBTQ Community?

Female

Male

Other*

 

Yes

No

Total

No

77.1

82.3

36.4

 

67.0

81.8

80.2

Yes

22.9

17.7

63.6

 

33.0

18.2

19.8


"Other" is Non-Binary / Agender / Something else
Total includes prefer not to say for both categories

There are different rates of harassment by gender, with men the least likely to be harassed.  In parallel, members of a LGBTQ community have a higher rate.  Variants of this table with harassment rates by race/ethnicity and by decade of latest degree can be found in the Appendix.  In short, those show no significant differences.

1.  On what platform were you harassed?

We asked those who reported they were harassed or bullied online on which platform it occurred.  We show the proportion of their answers in the pie chart, Figure 1

Figure 1 Locus of Social Media Harassment

Chart showing locus of social media harassment

Legend is least frequent at top, most frequent at bottom.


The great majority of harassment reports occur on Twitter and on EJMR. Twitter is the most-used social media platform for Economics. The share of harassment reports on Twitter is below the share of use, though both are high. EJMR’s share in harassment reports far exceeds its share in use. This follows from the now-familiar toxic nature of EJMR (see Wu, 2018) and the fact that a person does not need to use the site to be discussed on it.)

Harassment on EJMR often consists of comments from multiple anonymous accounts. Our discussions with and messages from those harassed on Twitter reveal that harassment there also comes from anonymous accounts. In a subset of cases, the anonymous accounts have clearly known details about the harassed person’s personal or professional life, indicating that the anonymous accounts likely are held by economists who are known to the person.

Dealing with the harassment of Economists on just these two platforms would go a long way toward alleviating the negative impacts of social media use on our profession.

2.   Upon whom do the impacts fall?

Our survey also asked the broader question "Do the positive or negative impacts [of social media] you noted here fall on any particular groups of scholars?" which had an open text field for answer. 190 respondents answered this question (not counting variants of "I don’t know.") The answers focus on the negative impacts (a very few list both positively and negatively impacted groups.) Accordingly, we categorized the comments about the groups impacted negatively. They fall into four groups:

Table 10 Proportion of Groups Mentioned by those answering

Mentioned Group

Proportion

Everyone

0.063

Conservatives, etc.

0.058

Some combination of women, people of color, and LGBTQ scholars

0.858

 

Answers not in any of the three groups above.

0.021

 

By far the largest group of answers list some or all of Economics' minoritized groups. Example answers in this area say "I think negative impacts fall more on people of color and women," "Women and minorities bear the negative impacts" "negative impacts clearly fall on under-represented groups" and "Women, minorities, immigrants, and LGBTQ+, all bear the brunt of these attacks on anonymous platforms."

Much smaller groups of answers, about 6% each, demur, either saying that the negative impacts fall on everyone or that they fall on conservatives (under various names, such as "classical liberalism.") These answers included some that appear to be gaslighting.[12]

The "other" category is far less frequent and contains other listed impacted groups, varying widely.

Minoritized groups in Economics gain from social media participation through finding community broader than that at their University or employer. They also run the risk of harassment and abuse. While they differ quantitatively, these groups share these two outcomes with all economists.

3.  Nature of Negative Impacts

We interviewed several economists who had been harassed on social media. Others declined to be interviewed but wrote about their experience in the open-ended questions.

The content of the harassment can arise from insults, often explicitly racist, sexist or homophobic insults. Sometimes these are derogatory academic insults (plagiarism, research misconduct, not deserving the placement of a publication, not deserving an appointment.) Given the language of the discussions in the more toxic Economics websites, this is unsurprising.

The immediate impact, perhaps desired by the harasser, is to make the harassed feel bad. Younger scholars tend to use "mental health" language to describe this impact.

More serious consequences come to some victims of harassment. Those in academic or other research appointments accused of plagiarism or other misconduct, however baselessly, must go into mandatory bureaucratic processes. The time lost is large, in the hundreds of hours for the accused and dozens of hours for their colleagues, even if the accusations were baseless. Victims in public roles, such as journal editors, government officials, and department chairs, report violent threats following harassment. The police are little help and home security systems are expensive.

Harassment, and the threat of harassment, cause some Economists to change behavior. Some withdraw from social media presence, though like many of our respondents they note that it is a valuable venue for finding employment and collaborations. A commenter, clearly one who saw positive benefits from social media use, wrote "I deleted my Twitter account and stopped posting due to harassment from a former classmate." Another interviewee said that they had found research collaborators and employment through social media, but had withdrawn after harassment. This forecloses the future benefits of social media use for them, but does not fully protect them from the harms, as the negative content remains online. The harassed person's colleagues, mentors, or potential hiring committees can still read the negative gossip about them. These are not empty concerns: one interviewee was told by recruiters at a major US University "we use EJMR to screen [job] candidates."

We learned that the victims think harassment follows women, people of color, or LBGTQ+ economists who have success or taken on a leadership position in our profession. Harassment is a tax on aspirations to lead, or even just to advance, in our profession. One interviewee told us that they had hoped to play a leadership role (which one we won't say for confidentiality reasons) but reconsidered after the harassment incidents. In these ways, social media harassment or targeting can stunt the professional development of scholars at any point in their career.

Support from colleagues and mentors is a weak force. Some victims feared a "blame the victim culture," and wanted to make sure that even the fact that they had talked to our Committee would be kept confidential. One, a graduate student, thought their advisers were supportive but that their reputation for being “serious about research” would be at risk, for example. Seeking support from senior colleagues or mentors is viewed by victims as potentially dangerous.

Harassment is a tax on entrants to the economics profession, and racist, sexist and homophobic harassment place a particular tax on people from minoritized identities. Commentators who are mentors of undergraduate students and/or of pre-docs noted the discouraging effects. One commenter wrote: "It pains me when I have to tell my undergrad students who want to go to grad school about EJMR. I work with majority minority students, queer students, and women. I have to inform them before they make a decision about grad school that we have an incredibly toxic, anonymous forum where they may some day be targeted. There are a lot of reasonable reasons that for students to not want to purs[u]e a PhD and it's my job to talk those reasons over with them, but fear of anonymous racist, sexist, homophobic, transphobic, ablest [sic. "ableist"] harassment should not be a reason my students decide not to go to grad school." Others note that harassing behavior in social media has driven some members of the same groups out of our profession. This echoes the findings of the AEA Climate Survey (Allgood et al., 2019)

Some survey respondents think the harassed should simply ignore social media, that all this is just part of academic life and criticism generally. "If you can't stand the heat, get out of the kitchen," wrote one. This is naïve. The personal attacks and real career costs cannot be avoided by being ignored, since others read the posts. While the harassers are anonymous, the victim is named.

The problem of harassment on social media has broader negative effects which are harming the Economics profession.

5.  Economists' Advice to the Committee: I

Part of the Committee’s charge was to recommend “best practices” for scholars, departments, and the AEA. Accordingly, we asked whether several online activities are appropriate.[13] There was broad consensus about more of these than we expected.

This table reports, for that list of activities, the proportion of respondents who thought each was sometimes appropriate or always appropriate.

Activity

Appropriate*

(Post) Positive views about the content of a conference or seminar presentation you attended

0.967

(Post) A photo you took of a conference or seminar presentation

0.833

(Post) Negative views about the content of a conference or seminar presentation you attended

0.826

(Post) Criticism of another scholar’s work on social media if that person is senior to / higher rank than you

0.783

(Post) Complaints about bad behavior you have observed by another member of the economics profession

0.749

(Post)  A video you took of a conference or seminar presentation

0.701

(Post) Criticism of another scholar’s work if that person is junior to / lower rank than you

0.589

(Post) Conduct a scientific study using social media without informing participants that they are in a study

0.466

(Post)  A discussion of a job market candidate’s portfolio, seminar presentation, or prospects at your institution

0.350

Frequent sites where one reads racist, sexist or otherwise discriminatory anonymous comments about other scholars

0.254

Use anonymous posts as a sole basis for form judgments about other scholars’ research achievements

0.088

(Post) Comments about another economist’s personal life, appearance, etc

0.083

Repost or otherwise disseminate anonymous criticism of other scholars without verifying the content

0.079

* Appropriate means the respondent chose "always appropriate" or "sometimes appropriate".

The bottom of the table is where the Committee sees broad consensus that there are some online activities that are not professionally appropriate. These are commenting on colleagues’ personal life and appearance, reposting anonymous criticisms without verification of the contents, and basing judgments about others’ research achievements solely on anonymous posts. We note that there is also a near consensus that it is appropriate to post positive or negative comments about the contents of a seminar presentation you attended.

These areas of consensus support continuing Economics' strong tradition of debate and discussion about ideas. They also suggest it is unprofessional to extend that discussion to online comments on colleagues' personal life and appearance, whether or not such comments violate the AEA's existing Standards of Conduct. They also, realistically, suggest that there is a professional problem associated with the viral spread of baseless anonymous criticisms of other scholars. The Committee recommends that all Economists note this broad consensus before engaging in those behaviors.

There is less clear consensus on other areas. A strong majority of respondents feel that "punching up" type of criticism of the work of a criticism of a scholar who outranks one, and in general calling out "bad behavior," is or can be appropriate, though there is less consensus on these.

The lack of consensus on other common elements of social media use imply that important conversations still need to be had in our profession. For example, while most (but not all) respondents feel it’s appropriate to post a photo from a presentation, fewer feel it is appropriate to post a video from a presentation. This implies that people should consider obtaining permission before posting such media (as conference organizers typically do in order to comply with relevant law) and associations should consider establishing policies about posting behavior at conferences to establish norms. Opinion is also mixed about whether scholars should use social media to criticize the work of those more junior to them, use social media to conduct a study without participants’ knowledge, or discuss job market candidates.

6.  Economists' Advice to the Committee: II, Perspectives on policy expressed in survey comments.

The Committee also received a remarkable volume of advice through the survey’s open-ended questions and through emails to the official Committee account. A subset of our interviewees and email correspondents also had advice.

The survey had open-ended questions at several junctures. We interpret 881 of these comments as offering the Committee advice on what our report should say about the use of social media in economics.[14] Many of the comments written by our colleagues also pointed to aspects of social media use in our profession that, while not explicitly policy advice, clearly suggested that the respondent saw a problem to fix or an opportunity to exploit. We report these as part of the policy-advice total. We emphasize that these comments are not answers to questions asked in the survey, but instead reflect what survey respondents have on their minds. The Committee thanks all the colleagues who took the time to write these comments.

a.   The most common advice: EJMR

Far more than any other point, open-ended comments cite the negative impact of the Economics Job Market Rumors website and call for Economists to stop using it. These are 16% of comments on all subjects. Some merely express that sentiment. Others call for AEA action. Yet others have specific ideas on how the Association should act in this regard.[15] Three representative quotes from different respondents will give you some of the flavor of these remarks.

"…it seems like the takeaway from the issues regarding EJMR is that a forum that is both anonymous and unmoderated creates a lot of space for bad behavior, and in fact there may be a Gresham's law in effect. There is a role for non-anonymous forums and for moderated anonymous forums, but an unmoderated anonymous forum creates perverse incentives.” Another version of this kind of quote “I think harassment on EJMR has driven women and minority scholars out of the profession, and that EJMR is generally an embarrassment to the Economics profession/makes us look bad to the outside world."

"*PLEASE* do something about EJMR. It is actively terrible for the women/minorities in our profession, and creates a chilling effect - there are many students who see the environment and decide not to go into academia."

"I know a few people who got harassed, had nasty gossip written about them, or saw comments about their appearance on EJMR. This is hurtful, damaging, and must stop. I don't know what the AEA can do about it in practice, but any action towards the goal of uprooting such behavior in the profession will be much appreciated."

A smaller body of comments, about 10, defends anonymous, unmoderated (or lightly moderated) social media including EJMR. Some colleagues point to supposed benefits of anonymous posting. Others assert that Economics has become a left-wing conspiracy or say that those harassed on social media deserved it. Perhaps the most important point about this perspective is that, while clearly deeply held, it is quite rare.

Neither the problems with EJMR, nor the arguments of its few defenders are surprising. Economists have been protesting against EJMR for years. This led to protests and to a decision some years ago by the AEA to sponsor two alternative social media platforms. One, EconTrack, was intended to provide a job-market information site free of abuse. Another AEA-sponsored platform, EconSpark, was designed to provide a safe forum for economics discussions.

The two efforts of the AEA to move Economists off EJMR to safety six years ago were ineffective.[16] EJMR usage is trending down, but not because of effective competition from the AEA sites. We propose a revised plan of action to deal with this problem below.

b.   Second Most Common Advice: Decline of #EconTwitter

The second largest group of comments concerns the decline over time in positive network effects experienced on “#EconTwitter.” Twitter was and is also one of the most important platforms for Economics discussion, frequently with the hashtag #EconTwitter. That was a hashtag used on Twitter (now X) to flag posts as of particular interest to economists, economic journalists, etc.

One colleague wrote:

"I used to enjoy EconTwitter quite a bit. The platform is a husk of what it used to be so I no longer am part of it. Would be nice to get a similar experience at one of the newer alternatives."

Another:

"This was a really hard question to answer because I think there was a time when Twitter had critical mass and achieved the best of what could be. Now I don't think any site has that critical mass and so none of the best potential outcomes are occurring. I miss the golden days."

Twitter itself has been declining in usage and it is not clear whether the decline in the volume and quality of economics discussion on that platform is merely part of that or reflects more specific causes.

The positive effects of social media in Economics are declining with the decline of #EconTwitter because of lost network effects. This, too, is an issue where the AEA and the Economics profession should act.

This problem did not originate in Economics-related social media use but eventually spilled over into our area of interest. In October 2022, one of the most successful social media platforms, Twitter, changed hands. New management undertook a series of policy changes that have led to declining usage and declining revenues. Some of the policy changes, notably weakening the power of users to block posts and comments and changes in the algorithm that chooses posts for users to see, have reduced the value of the platform for academic purposes. Many economists have migrated elsewhere.

As we will note below, it appears that Bluesky is a viable alternative that many economists have already migrated to, and the creation of a new equilibrium on that platform could be facilitated by institutional support that could serve as a coordinating mechanism.

c.   "Don't Do Anything, AEA"

A minority of respondents argues that the AEA should not have any best practices for social media use. This is a small but diverse body of comments. We quote many of them in the Appendix and summarize here.

  • One group, while agreeing that there are negative impacts, opposes any effort to reduce them because limits on free speech will be counterproductive.
  • Another group sees any AEA action which it cannot enforce in a court of law as irrelevant.
  • Another group thinks that the toxic behavior seen on social media by economists is merely an accurate reflection of the culture of our profession so any best practices will be ineffectual.
  • Another group is particularly concerned that new “regulations” will suppress the free speech of economists who are not "progressives."

This final group, a small minority, asserts that our profession has racial bias against white people, gender bias against men, and intellectual bias against pro-market arguments. This group tends to offer ugly comments.

Some of the arguments are defeatist. Some are fringe. Still, taken in the aggregate, there is meaningful opposition to any social media "speech rules" that the AEA might consider.

We take this seriously. We will not recommend any AEA "best practices" for what an individual economist should post in social media beyond existing policies, such as the Code of Professional Conduct, the Policy on Harassment, Discrimination and Retaliation, and so on. We will make recommendations from this committee about professional behavior on social media, but those are advisory to the individual scholar, not AEA rules, much less speech codes.

7.  Committee Recommendations

We suggest social media policy changes by the AEA, by other professional organizations in Economics, by employers of Economists and by Economists ourselves. These changes respond to an outcry from Economists against the negative elements of social media use today and for the positive elements in the recent past. They preserve the benefits of social media use, including open discussion of economic ideas and methods. Their adoption will improve the professional climate in Economics.

Our recommendations include no new limitations on Economists’ speech, on social media or elsewhere. We call on the AEA to use its coordinating function rather than its standard-setting function.

a.  Recommendations for the AEA as an Organization

1.  Affirm that social media use can be a valuable part of the economist's toolbox.

Many economists and departments engage with social media to disseminate and learn about research, to expand networks, and to advertise and seek collaborations and positions. These activities are not detrimental to our research and teaching activities but complementary to them.

2.   Reaffirm that the AEA policy on Harassment, Discrimination and Retaliation applies to social media and to the rest of the online world.

Particularly important elements include

"Economists have a professional obligation to conduct civil and respectful discourse in all forums, including those that allow confidential or anonymous participation."

"…a professional environment with equal opportunity and fair treatment for all economists, regardless of age, sex, gender identity and expression, race, ethnicity, national origin, religion, sexual orientation, disability, health condition, marital status, parental status, genetic information, political affiliation, professional status, or personal connections."

"Economists have both an individual responsibility for their own conduct, and a collective responsibility to promote professional conduct. These responsibilities include developing institutional arrangements and a professional environment that promote free expression concerning economics. These responsibilities also include supporting participation and advancement in the economics profession by individuals from all backgrounds, including particularly those that have been historically underrepresented."

3.  Broaden AEA announcements and messages on social media

These communications should broaden beyond Twitter, LinkedIn and Facebook to the other platforms read by economists, especially BlueSky.

4.   Encourage other institutions in Economics which post on social media to put their content on multiple platforms.

5.  Update AEA job market guidance to potential hiring institutions

AEA should recommend against use of social media to comment on job market candidates. Negative assessments, particularly, can be communicated through a decision not to interview, not to fly out, or not to hire.

6.  Add questions concerning social media to future professional climate surveys.

7.  EconTrack

The AEA site EconTrack was introduced five years ago to transparently provide information about the Economics labor market without the toxicity of EJMR. It did not gain large usage, in part because of design issues and in part because of the difficulty of shifting from the old (EJMR) equilibrium to a new EconTrack equilibrium. Many survey commentators noted both the value of the goal and the problems of achieving it.

Another attempt at making EconTrack, in a better environment and with a better user interface, rules and policies, could move a positive feedback loop to the AEA site.

8.  EconSpark vs. Other Alternative Platforms

The AEA site EconSpark was introduced five years ago to provide a moderated, safe forum for sharing information relevant to economics. As with EconTrack, many survey commentators noted both the value of this goal for EconSpark and the problems of achieving it.[17]

The AEA succeeded in getting some very large discussions going on EconSpark by inviting marquee economists to seed them. There were also some other large discussions. But there was never critical mass, much less success at the goal of creating a large-scale “non-swampy” discussion board. Initially it was very hard to use, but the AEA listened to some of the complaints, for example lowering the number of clicks to get to the site from five to two. Still, it required registration, a user cost. It ended up as one of the many sites that did not replace #EconTwitter. Those included not just EconSpark, but Mastodon and Threads.

It is not clear to this Committee that EconSpark had any role to play in competing against EJMR, which is unimportant as a site for Economics discussion other than of the job market gossip and rumors. (See Table 4.) It is also not clear that the advantages of an official, AEA-sponsored discussion forum outweigh the problems. Additionally, we think a large, expensive effort would be required to reinvigorate EconSpark (e.g., hosting exclusive content on the platform), whereas there is natural momentum toward the use of other tools, such as BlueSky.

We recommend instead that the AEA, and other Economics profession institutions support a policy of mulitihoming content to BlueSky.

BlueSky has experienced a large influx of economists since our survey. After experimentation with various alternatives, it has now emerged as the alternate platform of choice for economists, with a robust “EconSky” community. Economists have created a number of public goods, including guides to migration, “starter packs” for over 100 fields and subfields in Economics, and a number of Economics-dedicated feeds. It is not only economists who are migrating, BlueSky users are up 7x from January to November 2024. Both the overall usage and the economics usage appeared to be accelerating further in November.

Twitter has an algorithm-based feed; users can either read this feed or posts by the accounts they choose to follow. BlueSky permits anyone to create “feeds” which users can follow. Twitter limits users’ ability to block or “mute” accounts or posts whose material they wish not to see, including recently changing Block so a blocked user can still see your posts, while BlueSky has a very effective block. Twitter has very limited content moderation, i.e., there are not many employees or volunteer moderators attempting to limit harmful content and there is an ineffectual “community notes” system flagging some material as misleading or harmful, while BlueSky has both the “block” option and a modest “trust and safety” department undertaking moderation. BlueSky also prioritizes higher quality content. It allows users to hide abusive replies to their posts and detach their original post from quote-posts creating pile-ons.[18] Additionally, BlueSky still allows for robust verification of users’ identities through owning website domains–if you own a domain with your name, you can use that in your BlueSky handle to verify your identity.We believe these additional trust and safety features are linked to users reporting a better experience on BlueSky, with less harassment and bullying, especially from those outside the intended audience of their posts. Yet, unlike Mastodon, which overly siloed communities, BlueSky still allows rich interaction across disciplines and interest areas. Additionally, because BlueSky’s functionality is so similar to Twitter (partially because it originally started as an in-house open source Twitter project, run by Twitter’s founder Jack Dorsey), users have found the experience of switching over to have low fixed costs. In repeated interviews, the committee heard the BlueSky had become the destination of choice for social media-using economists who wanted to leave, or engage less, on Twitter.

In terms of actually switching to BlueSky, there are multiple tools to reduce costs. Users can identify their previous follows using Sky Follower Bridge. They can find users to follow using the multiple econ starter packs, compiled by a user here. And users can seamlessly multihome using tools such as Buffer, Fedica, and Touitomamout.

BlueSky offers rich and exciting functionality that can be used by the AEA and other economics institutions to promote engagement around economics research. The AEA can create a feed of AEA members, so anyone can see all posts of verified AEA members. AEA could create additional lists around member's self-identified specialties (labor, environment...) or identities (women, LGBTQ+, etc...). Some of these activities are already being undertaken by individual users. Media and policy people looking to connect with economists fitting some profile might have a particular interest in this. The AEA could also create starter packs for people in different fields, by asking for social media handles of AEA conference registrants and members. Conferences could have a conference starter pack, where users could follow everyone speaking at a conference, and a custom feed to aggregate all posts about a conference, suggested by a user here. The AEA membership has exhibited a demand for alternatives beyond Twitter, and thus the AEA has an opportunity to serve its members by reducing fixed costs and generating network effects for users looking to multihome, with low cost and zero externality (for people not interested in BlueSky) interventions.

Multihoming for Economics Discussion

With input from the Association, this Committee participated in the accelerated migration of economics discussion from Twitter/X to BlueSky in the late summer and early autumn of 2024. We nudged economics institutions, suggesting multihoming on both platforms.

The migration was already in slow progress, and network effects had declined on Twitter. Many of our survey respondents expressed concerns about the decline of #EconTwitter. See Section 6.g. Our conversations with some of them and examination of current social media use revealed that, while many platforms were being considered and tried, the largest migration was to BlueSky. A lively group of economist public goods providers has sought to make this migration easier. Much of their work is summarized in https://aaronsojourner.org/for-bluesky-curious-econ-lovers/

We engaged in mild coordinative activities focused on widely followed institutions and scholars. We informed them of the ongoing migration of economists and of the activity of other institutions. The substance of declining economics discussion on Twitter was familiar to some. Others use broad, general social media usage measures, not focused on economists.

We encouraged those institutions posting to a Twitter account to consider adding a BlueSky account with the same content, i.e. to multihome. We provided links to the services that make multihoming very easy for posters. This led several important economics institutions (NBER, CEPR, AEA, IZA, SIEPR, etc., etc.) and journals to multihome. Widely followed scholars also chose to multihome and a few chose to migrate outright.[19] This effort was visibly successful, so scholars in other disciplines asked for our suggestions and a general academic migration now appears to be underway.

While our survey found many economists mourning the loss of network effects on #EconTwitter, today many publicly celebrate regaining those same benefits on BlueSky. For example:

Paul Krugman @pkrugman.bsky.social [11/12/2024]

It looks as if this place is becoming what Twitter used to be; for now, at least, we can post stuff to a sigificant [sic] audience without being overrun by trolls and bots.

This is a success for light-touch coordinative efforts and for making information about Economics widely available. We thank all the scholars, institution leaders and communications directors who joined the effort.

b.  Recommendations for other Professional Associations

1. Mention social media in standards of conduct.

2. Set norms explicitly for social media use in conferences (typically as part of conference code of conduct).

3. Broaden dissemination of announcements of meetings, publications, research opportunities, etc. to Twitter, LinkedIn and BlueSky.

c.   Recommendations for Individual Scholars

1. Consider social media to disseminate research-related opportunities broadly.

This is consistent with the AEA’s Best Practices in Conducting Research, “Share research opportunities broadly” and includes announcements of conferences, jobs, research assistantships, pre-docs, post-docs, fellowships and grants, inter alia.

2. Consider using social media to share research.

a. Your own (using your own name) and your students’ (especially JMP) are good candidates.
b. Reposting announcements of papers where you are excited about content.

3. Consider using social media to find collaborators.

4. If you feel the need to ask questions about economics or economics methodology anonymously, prefer to use a moderated site where discussions stay on-topic.

a. Such sites include Stack Exchange, Reddit, and more specialized sites focused on specific tools (e.g. Stataist, or Wolfram Community)
b. EJMR is very little used for such discussions. See Table 2.

4. If you wish to report bad behavior by an economist, either officially or through a “whisper network,” consider alternatives to anonymous public posting, which has little credibility

6. Consider the quality of the evidence presented before crediting anonymous posts

7. Please refrain from reposting or otherwise disseminating anonymous criticism of other scholars without verifying the content.

This behavior is viewed as inappropriate by a supermajority of AEA survey respondents.

8. Please refrain from posting comments about another economist’s personal life, appearance, etc.

This behavior is viewed as inappropriate by a supermajority of AEA survey respondents.

9. Please refrain from posting discussion of a job market candidate’s portfolio, seminar presentation, or prospects at your institution.

This behavior is viewed as inappropriate by a supermajority of AEA survey respondents.

10. Please avoid social media sites where one reads racist, sexist or otherwise discriminatory anonymous comments about other scholars.

Today, EJMR is the leading example to avoid. This behavior is viewed as inappropriate by a supermajority of AEA survey respondents. It may be less obvious that going to a site to read toxic material is problematic. However, social media sites typically earn income from advertising, so reading provides financial support. Also, posters of toxic materials enjoy having an audience, so it is helpful not to join that audience.

11. Please do not use anonymous posts as a sole basis from which to form judgments about other scholars’ research achievements or character.

This is the elementary information economics of anonymous posts. Anonymous posts are unreliable without verification. Also, this behavior is viewed as inappropriate by a supermajority of AEA survey respondents.

12. Do not violate the AEA’s professional code of conduct on social media.

13. In mentoring and in assessment of junior scholars’ work, do not view use of social media as a negative.

d.   Recommendations for Departments and Organizations

1. Post to several social media, the Economics audience is on multiple platforms

a. BlueSky, LinkedIn, Twitter, possibly Facebook

Recruiting and labor market

2. Use the AEA Job-Market facilities, including the revised EconTrack

a. Post job opportunities, interview lists, flyout lists, offers and hiring, or confirm true rumors and correct false ones when an EconTrack reporter calls.  Job market information will be publicly known whether you post them or not.
b. Leaving this information to anonymous sites with weak (at best) moderation puts discussion of your potential future colleagues into a cesspool.

3. Ensure that hiring, promotion and appointment recommendations are based on the recommenders’ knowledge of the candidate’s characteristics, not on unverified and unreliable gossip.

a. Consider posting a policy statement to this effect.

4. If your university or other organization will permit it, ask reviewers to certify the basis of their assessment in the work, not in anonymous gossip about the person.

Use social media in a manner consistent with the AEA’s existing best practices for recruiting

5. Consider sharing research and job opportunities on social media to broaden the list of possible candidates.

6. Conduct job searches without unnecessary stress for candidates.

a. Don't post pictures or videos of candidates’ job talks
b. Don't post assessments of candidates publicly
c. This can unfairly prejudice them at places they might later interview/visit

7. Be sensitive to cultural differences / outsider status of candidates.

 

8.   References

Ajzenman, N., Ferman, B., & Sant'Anna, P. C. (2023). Discrimination in the formation of academic networks: a field experiment on# EconTwitter. Working paper.

Alberti, C. T., & Morris, L. (2023). Online Toxic Communication about the Accounting Academic Job Market. Forthcoming, Accounting Horizons.

Allcott, H., Braghieri, L., Eichmeyer, S., & Gentzkow, M. (2020). The welfare effects of social media. American Economic Review, 110(3), 629-676.

Allcott, H., Gentzkow, M., & Song, L. (2022). Digital addiction. American Economic Review, 112(7), 2424-2463.

Brailovskaia, J., & Margraf, J. (2018). What does media use reveal about personality and mental health? An exploratory investigation among German students. PloS one, 13(1), e0191810.

Braghieri, L., Levy, R. E., & Makarin, A. (2022). Social media and mental health. American Economic Review, 112(11), 3660-3693.

Brock, A. (2012). From the blackhand side: Twitter as a cultural conversation. Journal of Broadcasting & Electronic Media, 56(4), 529-549.

Bursztyn, L., Handel, B. R., Jimenez, R., & Roth, C. (2023). When product markets become collective traps: The case of social media (No. w31771). National Bureau of Economic Research.

Chan, H. F., Önder, A. S., Schweitzer, S., & Torgler, B. (2023). Twitter and citations. Economics Letters, 231, 111270.

Ederer, F., Goldsmith-Pinkham, P., & Jensen, K. (2024). Anonymity and Identity Online. Working paper.

Faelens, L., Hoorelbeke, K., Cambier, R., Van Put, J., Van de Putte, E., De Raedt, R., & Koster, E. H. (2021). The relationship between Instagram use and indicators of mental health: A systematic review. Computers in Human Behavior Reports, 4, 100121.

Klar, S., Krupnikov, Y., Ryan, J. B., Searles, K., & Shmargad, Y. (2020). Using social media to promote academic research: Identifying the benefits of twitter for sharing academic work. PloS one, 15(4), e0229446.

Lee, S. Y., & Kim, J. H. (2023). What makes people more polarized? The effects of anonymity, being with like-minded others, and the moderating role of need for approval. Telematics and Informatics, 76, 101922.

Mohammadi, E., Thelwall, M., Kwasny, M., & Holmes, K. L. (2018). Academic information on Twitter: A user survey. PloS one, 13(5), e0197265.

Moore, A., Fredheim, R., Wyss, D., & Beste, S. (2021). Deliberation and identity rules: The effect of anonymity, pseudonyms and real-name requirements on the cognitive complexity of online news comments. Political Studies, 69(1), 45-65.

Rowlands, I., Nicholas, D., Russell, B., Canty, N., & Watkinson, A. (2011). Social media use in the research workflow. Learned Publishing, 24(3), 183-195.

Sugimoto, C. R., Work, S., Larivière, V., & Haustein, S. (2017). Scholarly use of social media and altmetrics: A review of the literature. Journal of the association for information science and technology, 68(9), 2037-2062.

Vannucci, A., Ohannessian, C. M., & Gagnon, S. (2019). Use of multiple social media platforms in relation to psychological functioning in emerging adults. Emerging Adulthood, 7(6), 501-506.

Wu, A. H. (2018, May). Gendered language on the economics job market rumors forum. In AEA Papers and Proceedings (Vol. 108, pp. 175-179). 2014 Broadway, Suite 305, Nashville, TN 37203: American Economic Association.

 

View Report Appendix

 


[1] See Rowlands et al., 2011; Klar et al., 2020; Chan et al., 2023; Sugimoto et al., 2017; Mohammadi et al., 2018. For community finding benefits, see Brock, 2012.

[2] See Ajzenman et al., 2023.

[3] See Wu, 2018; Alberti and Morris, 2023; Ederer et al., 2024.

[4] Lee and Kim, 2023; Moore et al., 2020.

[5] There are impacts of social media addiction on mental health (Braghieri et al., 2022). Excessive use is often linked to lower well-being, but more generally, negative and positive effects of usage can depend on features like what platforms are considered or what well-being metrics are used (Brailovskaia and Margraf, 2018; Vannucci et al., 2018; Faelens et al., 2021). Recent work also indicates that traditional measures of welfare overstate the value of social media to consumers (Allcott et al., 2020) and do not capture negative impacts imposed on non-consumers (Bursztyn et al., 2023); and suggests that self-control problems, habit formation and inattention drive a large percentage of usage (Allcott et al., 2022).

[6] While we quote some of their comments below, we respect the confidentiality of all respondents’ identities.  We are grateful to all who responded to the survey, sat with us for interviews, or sent emails, and especially to the many survey respondents who wrote substantive answers to the open-ended questions.  We are also grateful to Liz Braunstein and Michael Albert of the AEA staff who assisted with our survey and Rachel Bi for research assistance. 

[7] These were (alphabetically) BlueSky, Economics Job Market Rumors (EJMR) the two AEA-sponsored platforms EconTrack and EconSpark, Facebook, Instagram, LinkedIn, Mastodon, Reddit, Stack Exchange, TikTok, and Twitter.  Twitter has been rebranded as X, but in many comments our colleagues refer to it as Twitter.

[8] Several events have lessened the importance of Twitter/X for Economists since our survey.  See discussion in section 7.j.10.

[9] If we restrict attention to the twelve social media listed in the Survey, 81.16% of respondents use social media.  There was an open-ended question about other social media survey respondents use. We counted most of these as social media, including messaging systems with some social features (Discord, Telegram, etc.), blogs and Substack’s with discussion capabilities, overseas platforms (Weibo, WeChat, WhatsApp, etc.), Threads and YouTube.   We excluded email lists.  With that expanded definition, overall 81.67% use social media.

[10] The other group that appears different is “Prefer not to say.”  While they are not perfectly coincident, the same people tend to “prefer not to say” about multiple demographic characteristics.  Also, these respondents tend to be more positive about profession-wide social media use and anonymous social media use as well as their own use. 

[11] We exclude from this count those who answered “yes” but included behaviors (“plagiarism. Is plagiarism harassment?”) or platforms (email, a university’s course-evaluation system) that are outside the scope of this report.

[12] We included the answer “It falls on people who are less-qualified than others to be in the positions they are. Blame the DEI policies, not the critics!” in that last group.

[13] A small minority of survey respondents objected to this line of enquiry.   They do not see a role for a professional association here, characterizing such considerations as leading to  “censorship,” “regulation,” limits on “freedom of speech,” etc.

[14] The survey’s final question, a very open-ended one, drew 504 responses, mostly with policy advice. Earlier open-ended questions, one about “other impacts” of social media (265 entries) and the location and “other impacts” of anonymous social media (355 entries) also drew some essays. Other open-ended questions (“on what platforms does this impact occur”) also had some policy advice for us. We do not count answers to specific questions (e.g., “Upon what groups … “ answered with a group or groups or “Other platform, please name”) as containing policy advice. We also do not include complaining about the survey, as opposed to complaining about the AEA doing anything (including the survey) as policy advice.

[15] The third class of comment changed, not in frequency but in form, after an email from Prof. Anya Samek to many colleagues on the subject. Later commenters often echoed Prof. Samek’s specific suggestion. Criticisms of EJMR’s impact and calls for the AEA to do something about it were the most frequent kind of comment before that, and policy suggestions were more varied. We thank Prof. Samek for letting us attribute the letter to her by name and for the suggestions in it.

[16] Despite continued hard work by AEA staff and leadership on EconTrack, it does not have enough information about job market posts to interest readers. Despite the participation of a number of leading economists as discussion facilitators and hard work by AEA leadership and staff, this site as well failed, after an initial burst of interest, to create a sufficient volume of participation for long-term survival. Today, it has a very low flow of posts and limited usage.

[17] Some commenters in our Survey recommended changing EconSpark to a model that permits anonymous posting but is moderated, that is, not changing anything. Defenders of EJMR also wrote that EconSpark is non-anonymous, incorrectly.

[19] The structure of the network effects is complex. Some economists have primarily academic economics connections, while others engage with scholars in other disciplines, journalists, and policymakers. Thus far the coordinative efforts have largely worked with the first group. We recommend multihoming to institutions seeking to reach many economists as a result.