The production economics of economics production

• Date: 01 February 2021
• Author: Ben G. Li,Yushan Hu
• Published date: 01 February 2021
• DOI: http://doi.org/10.1111/jems.12399

J Econ Manage Strat. 2021;30:228–255.wileyonlinelibrary.com/journal/jems228

|

© 2020 Wiley Periodicals LLC

Received: 24 May 2018

|

Revised: 29 June 2020

|

Accepted: 23 July 2020

DOI: 10.1111/jems.12399

ORIGINAL ARTICLE

The production economics of economics production

Yushan Hu

1

|Ben G. Li

2

1

School of Economics, Renmin University

of China, Beijing, China

2

Department of Economics, University of

Massachusetts, Lowell, Massachusetts

Correspondence

Ben G. Li, Department of Economics,

University of Massachusetts, Lowell,

Massachusetts, MA 01854.

Email: benli36@gmail.com

Abstract

How is new economic knowledge produced over time? That depends on how

the expertise of authors is managed within economic journals. Using data from

41 major economics journals spanning 21 years (1994–2014), we find that both

the intensive margin (article length) and extensive margin (article number) of

the discipline have been growing. In particular, the extensive margin has

outgrown the intensive margin, such that each article produces absolutely

more but relatively less knowledge. This pattern is highly consistent with a

model of within‐journal specialization. As predicted by the model, the share of

an individual article shrinks less in general interest journals and in more

prestigious journals, where expertise is less substitutable across topics.

1|INTRODUCTION

Readers' willingness to pay for journals is declining. Discontinuing print journals is a current trend in university

libraries. Electronic journals are sold online, though their unit of sale is the article. This is anything but surprising.

Working papers now become freely accessible online long before being published, and many final drafts remain freely

accessible after being published. Journals are diminishing in their role as knowledge disseminators and instead are

transitioning to a role as quality‐controlled repositories of knowledge.

1

An average page in a journal receives less

attention now than in the preinternet era when researchers relied on journals to stay on the frontier of research

advancements.

This study is concerned with how the annual outputs of journals, including their length and number of published

articles, have evolved in the new age. Our data cover 41 prestigious economic journals, spanning two decades

(1994–2014) that coincided with the time when the World Wide Web and online search engines revolutionized in-

dividual information intake.

2

We find that journal length, article length, and number of published articles have been

steadily increasing, and more important, the page share of an average article within a journal has been decreasing.

Think of every journal as a knowledge‐making firm that charges consumers for every page of knowledge they read,

journal length as its total output, and articles as its knowledge‐making teams. The observed data pattern suggests that

every team produces absolutely more but relatively less.

The observed data pattern is consistent with what an analogical within‐firm specialization model predicts. As

readers' attention on every published page declines, their willingness to pay reduces. In response, journals raise their

outputs to compensate for the lost revenues. Suppose that introducing a team of experts (an article) incurs a fixed cost,

and letting a few teams produce all the output overstretches their expertise. Then, as a tradeoff, every journal has an

optimal number of teams (number of articles) every year, conditional on its annual output. When journals produce

greater annual total outputs, the fixed costs of additional teams (articles) are economized. As a result, a greater number

of articles are published, each being more focused on its core expertise. Therefore, both journal length and article

number grow, with the latter margin growing more to reduce every article's output (page) share; that is, the extensive

margin outgrows the intensive margin as the amount of economic knowledge increases over time.

Up to this point, within‐journal specialization is just one possible explanation. We next derive unique predictions

from it. Within a journal's scope, if the expertise is less substitutable across topics, the fixed costs of publishing more

articles will be less economized, so that the page share of individual articles should decrease less. Conceivably,

the expertise on one topic is more discounted on other topics (a) in general‐interest journals than in field journals, and

(b) in journals with stronger expertise overall. Thus, the page‐share decrease should be less in general‐interest journals

and in better ranked journals. These hypotheses are found to be supported by the data, and our findings are robust to

the use of different estimation methods, subsamples, and journal rankings.

Apart from type and ranking, there are two other journal characteristics that relate to the substitutability of

expertise across topics within a journal. One is whether a journal imposes a page limit on initial submissions. We find

the journals with such page limits highly conforming to our earlier hypotheses, while the patterns from the journals

without page limits are weaker. This is consistent with the thesis of Card and DellaVigna (2014) on revealed preferences

in submissions—authors shorten their submissions following the imposed page limits if they have no better outlets that

match their papers' qualities. A natural implication of their thesis is that journals would not impose page limits if the

risk of having authors extend beyond their expertise is not a significant concern. In other words, for journals that

impose page limits on initial submissions, submissions longer than the page limits overstretch the expertise that the

journals believe those submissions to have.

The other journal characteristic that relates to the substitutability of expertise across topics is the degree of spe-

cialization of the researchers. In research fields that have more specialized researchers, the cross‐topic expertise loss is

expected to be less, so that the decrease in the page share of individual articles should be greater. In other words, letting

more‐specialized experts author articles in their fields runs a lower risk of overstretching expertise. We use the

field‐specific specialization index compiled by Ellison (2002a) and find evidence in support of our hypothesis.

The key mechanism of within‐journal specialization is that journals, in a highly competitive environment, push

articles to be short and focal. Dixit (1998) has an eloquent anecdote on this tendency:

As an editor of the Review of Economic Studies, (Frank) Hahn asked the author to cut down his paper from

40 pages to its essential core of three pages. When the author wrote a long and indignant letter, Hahn

responded in two sentences: “Crick and Watson described the structure of DNA in three pages. Kindly explain

why your idea deserves more space.”

The analogy used by Frank Hahn may not be entirely appropriate, because the writing in economics, unlike that in hard

sciences that build on lab experiments, usually needs more detailed descriptions, explanations, and justifications. Never-

theless, his analogy exemplifies a key tradeoff in the business of academic publishing: a longer article is richer but easily

strays from its central insights, and delivering more insightsinfewerpagesisthekeytopublishingasuccessfuljournal.

The first contribution of this paper is demonstrating the specialization patterns in the production of economic

knowledge. Economics is a discipline about specialization and its own evolution exemplifies specialization. Starting as

“an inquiry into the nature and causes of the wealth of nations”(Smith, 1776), economics has grown to be a modern

social science that has 849 clusters of research topics (defined by JEL codes) and 2,527 journals (according to RePEc

registrations).

3

The 41 economics journals covered by our 21‐year sample published 55,244 research articles. These

articles specialized in distinct topics, which went far beyond the “nature and causes of the wealth of nations,”and no

modern economist can claim that she understands all of these articles even at a superficial level. Adam Smith referred

to such a phenomenon as a division of labor.

There exist various approaches to formulating Adam Smith's division of labor. We built a variant of Chaney and

Ossa (2013) to guide our empirical study.

4

The approach taken by Chaney and Ossa (2013), featuring monopolistic

competition, introduces different teams, instead of pure labor, into the firms that make differentiated products.

Launching more teams incurs more fixed costs, but letting teams specialize in certain parts of production reduces

variable costs. This informs the importance of optimizing the structure of teams (essentially, expertise units) in the

production of any expertise‐intensive product. As teams become more specialized, they each are responsible for less, but

more teams are contributing at the same time and every team is becoming more productive. Think of a research project

as a team that adds to the body of economic knowledge. We believe that most economists agree with us on the

increasing sophistication of economics papers, the narrowing breadth of research topics, and at the same time the

remarkably rapid growth of modern economic knowledge.

This paper also sheds light on the operation of the economics profession. Economists have studied the whole production

line in their own profession, from authorship (Hamermesh, 2013) and coauthorship (Laband & Tollison, 2000;Ray&

HU AND LI

|

229