interests (Broockman and Skovron 2018;Hertel-
Fernandez et al. 2019). These studies suggest that the
wealthy and well-organized interests enjoy greater access
because politicians incorrectly believe their self-interests
are congruent with constituents’interests.
Yet, the process by which this unequal access converts to
policy influence is unobserved (McKay 2022). So, we de-
signed an experiment to directly test how much campaign
donors, corporate lobbyists, and consumer group lobbyists
affect legislative action in addition to seeking access alone.
Like similar experiments (Chin et al. 2000;Hertel-
Fernandez et al. 2019;Kalla and Broockman 2016), our
study tests the counterfactual that constituents are less
likely to gain access to congressional offices than donors
and lobbyists. The innovation in our study is to also
measure staffers’likelihood to use information from, or to
side with, the petitioner’s request. Using an original
survey of over 400 DC-based congressional staffers in
2017, the vignette experiment hypothetically requests a
meeting with a member of Congress’s Washington office.
The script varies the identity of the petitioner, whether
they are seeking legislation to be introduced or blocked,
and the nature of the information offered in support of the
request. The outcomes we measure include the likelihood
of taking a meeting, using the information, siding with the
petitioner, and the perception that the petitioner is rep-
resentative of geographic constituent preferences.
Thus, our outcomes measure additional, more nuanced
responses to campaign donors and lobbyists beyond merely
granting access. Seeking and gaining access has long been
considered a prerequisite condition for influencing legislation
(Austen-Smith 1995;Wright 1995). Thus, the experiment is
designed to test whether the cost of granting a constituent,
donor, or lobbyist a meeting is relatively low compared to
other actions in the legislative process. There is anecdotal
evidence that the norm in congressional offices is that
members and staffers should grant meetings with as many
requests as possible as a matter of professional courtesy. For
instance, a senior policy advisor in a House members’office
who was interviewed for this project remarks:
We try to make sure that we do a lot of stakeholder outreach.
The [member of Congress] will tell anybody who’s listening
we have an open door. We’renot gatekeeping in terms of who
gets to come have meetings, who doesn’t. If you reach out
and if we can fit it in, we’re going to set up a meeting for you.
The norm of having a relatively low level of scrutiny to
grant meetings is common. The default to grant access to
nearly anybody who asks is borne out in an interview with
a legislative director in a personal office for a House party
leader. In this interview, the staffer is emphasizing how
unusual the party leaders’personal office operations are
compared to rank-and-file members’offices.
It’s a leadership office, and the way we prep our member and
everything we do is in that style. This office is quite possibly one
of the most demanding, and I’ve worked in a lot of different
offices […] [The member of Congress] has a briefing book
every single day. She’s very prepared in what she does. There
are a lot of internal meetings that have to do with her schedule,
making sure that she is where she needs to be. A lot of thought
goes into what meetings that she’s taking, so things just don’t
happen. There are some offices where it’sjustlike,“Well, a
constituent just came in. We’lljust put them on the schedule”or,
“This looks like a good meeting for the member.”None of that.
In this office specifically, we have an actual, legitimate
scheduling process. We call it the rallying process. It’s a form
that staff fill out, and do research, and make sure that this is the
right person that she needs to be speaking with, and why it is.
These comments from senior staff do not necessarily
establish how common it is to grant access or not. But they
do suggest that measuring access alone partly reflects a
meaningful political choice and partly reflects a profes-
sional custom on Capitol Hill.
In addition to professional courtesy, it stands to reason that
the stakes are low when deciding to whom to grant access,
especially among staff hired in part to be the eyes and ears of
members’legislative enterprises (Salisbury and Shepsle
1981). Legislators’time—especially given their limited
presence in Washington—is pressed between committee
meetings, floor voting, fundraising, media appearances, and
other events. Thus, observations of a meeting being granted
offer limited insight into legislative enterprises’choices to
engage in more intense forms of legislative participation (Hall
1998). In this case, the costs for acting after a pitch is made
during a meeting is higher. Costs include the time and effort
to follow through, the reputation risk for recommending the
“wrong”position to take, the potential publicity associated
with advocating on behalf of the interest or policy in question,
and so on. These actions are potentially more transparent to
attentive sub-constituencies, policy stakeholders, journalists,
and real and imagined political rivals. And, acting on an issue
being requested by a petitioner means not acting on hundr eds
of other possible issues, which is primarily what private
interests seek from legislators (Hall and Deardorff 2006).
The cost-benefit distinction between granting access
and then acting on a request is clear from two House chiefs
of staff who were lobbyists themselves prior to working in
Congress. The chief of staff in a majority member’soffice
I had been [lobbying] for about three or so years, and I
realized I didn’t know what happened when we left a
member’soffice meeting with staff or a member of Congress.
Like, what did they do? How did they make decisions once we,
as a lobbying team, left? Because I’d never worked on the Hill.
Furnas et al. 349