might trump those factors with the relatively small sums that PACs, even bundled PACs,
typically give to a multi-million dollar campaign seems implausible. Were vote-buying the
purpose of PAC contributions, moreover, the principal hypothesis would be that PACs will
shower their funds on legislators likely to be on the fence on group-relevant issues. Again, the
evidence suggests that this simply is not true. PACs contribute heavily to their natural allies,
those whose support they least need to buy.
Second, the above studies suggest that the proximate political objective of the PAC
manager is to enable his lobbyists to gain access to legislators, albeit for purposes not always
specified. Interview-based research with PAC-managers, as well as more quantitative research
supports this view (e.g., Langbein 1986; Wright 1990). But the research on this is not extensive,
and the findings are not consistent. Differently motivated, some of the best studies of lobbying
do not investigate the effects of money (e.g., Carpenter, Esterling, and Lazer 1998).
Hojnacki
and Kimball (2001), in turn, challenge the conventional wisdom. They argue that the analysis of
access has failed to control for the constituency connections that PAC-affiliated groups enjoy;
omitted variable bias thus leads to a mistaken finding that PAC contributions buy access. A
second limitation of past research is that none take up the matter of soft money, nor theorize
about the different behavioral implications of soft versus hard money. We try to make progress
on these limitations, using a dataset on Senate consideration of the MMA developed specifically
for our purposes.
II.
A Theory of Lobbying
Empirically, our goal is to provide a systematic examination of the access gained by both
private and public sector interest groups in the Senate’s consideration of the MMA. We take this
to be the first but not the last step in establishing whether this important legislation was bought
and bossed by the private health sector. To this end, we place the MMA case in the context of a
6
We do not, in turn, investigate the effects of weak or strong ties among lobbyists in an interest group
network. For instance, the possibility that access gained by money for one group might improve the access
of another, connected or aligned group warrants further work.
7
Convention