Doing Economics

Citing only famous economists. This is related to the previous point. If
your manuscript cites only famous economists, and if editors start making a
list of would-be reviewers for your manuscripts by taking a glance at whose
work you cite, this is a surefire way to maximize your chances of getting
reviews written by referees found by a keyword search.
I realize that citing only famous economists might make you feel like
your paper is in the same league as those famous economists’ papers, but
one thing you have to understand is that often, like referees like. In other
words, if you are an early-career researcher, odds are your paper is getting
reviewed by early-career researchers, and if you are a very senior
researcher, odds are your paper is getting reviewed by very senior
researchers. This means that even if you submit your paper to a top-five
journal, your citing leading economists still does not mean that your paper
is going to get reviewed by those leading economists. The bottom line is
that although you certainly should cite famous economists wherever
appropriate, you should aim to cite those economists who (i) have worked
on topics similar or close to your topic, and (ii) are likely to referee for the
journal you are submitting to. Only you can answer who fits the bill for (i).
As for (ii), a good rule of thumb is whether those would-be reviewers have
published in the journal you are targeting.
Strategically avoiding citing certain people. The findings in your paper
might run counter to the priors (or worse, the ideology) of some of the
people doing the gatekeeping in your field. Or you may have a personal
conflict with someone in your field. One unfortunate feature of the peer-
review process—especially in economics, where single-blind reviewing has
become the norm at top journals—is that it involves human beings, and thus
human frailties. This means that if you step on the toes of someone who is

---

more senior than you, and whose ego is invested in a particular qualitative
finding (e.g., “School vouchers are bad”), you will predictably want to
avoid having them (or their cronies) as reviewers, which makes it tempting
to avoid citing their work. Avoid that temptation as much as possible. While
drawing those people as reviewers is certainly not ideal, a worse outcome
occurs when you get caught strategically citing the literature, which is at
best interpreted as you not knowing what you are doing and at worst as you
trying to obscure some part of the literature. The only right way to act here
is to cite honestly, and use your writing skills to simultaneously signal to
the editor (who makes the final decision, and can often see through people’s
biases when given a reason to do so) that you are stepping on some toes and
to soften the blow of your findings to your reviewer’s ego.
Here is an example. I once wrote a paper looking at the relationship
between farmers markets (number of farmers markets per capita in a given
state) and food-borne illness (aggregate number of outbreaks and cases of
food-borne illness per capita in a given state, and then the same broken
down by types of illness). The identification was obviously not up to
randomization standards, but it was pretty good, and we even supplemented
our core identification strategy (state fixed effects with a battery of means
of accounting for time: year fixed effects, linear time trends, state-specific
linear time trends, census region–year fixed effects, and coarsely
controlling for violations of the stable unit treatment value assumption)
with a weather shock-based instrumental variable. No matter which way we
approached the data, we kept finding the same thing: there was a positive
relationship between farmers markets and food-borne illness.
But we also kept running afoul of one specific reviewer who hated our
finding because that reviewer was invested in the goodness of farmers
markets, to a point where they simultaneously told us, in the same referee
report, that (i) our results were likely spurious, and that (ii) we were looking
at “too many” outcomes because we disaggregated. We resolved that
problem in our cover letter, by telling the editor that the reviewer was at
best not being earnest and at worst a crank when we wrote:
So we are in a Catch-22: If we hadn’t conducted a disaggregated
analysis by illness, we would have been told that our findings are
likely to be spurious. Now that we do conduct such a disaggregated

---

analysis, we are exposing ourselves to the “multiple hypothesis tests”
criticism. This really makes us wonder whether anything we do or say
will satisfy [that reviewer].
Ultimately, the editor saw through the reviewer’s bias, and decided to
accept our manuscript for publication. Though you should not write as
candidly in a manuscript as we did in our cover letter, it is still possible to
write your introduction in such a way as to communicate to the editor (and
the other reviewers) that your work, though it may contradict someone
else’s pet findings, is worthy of publication.