Since the Great Depression, the federal government has implemented agricultural programs by passing what is known as the farm bill. These farm bills typically contain sunset provisions, requiring new farm bills to be passed roughly every five years. These farm bill votes provide ample opportunities for the agricultural lobby to engage in political rent seeking behavior. There is a considerable literature on the impact of direct campaign contributions on farm bill amendment votes. The approach used in this literature is to identify a congressional floor level amendment vote to a farm bill that benefits a crop specific agricultural interest. Through the use of a simultaneous equations probit-tobit model, the relationship between donations and the amendment vote is estimated. This research uses a similar approach. Instead of looking at specific amendment votes, my paper looks at final farm bill votes in the House of Representatives, treating each farm bill vote as a repeated observation of the same event. In so doing, a time series is constructed, allowing for study of how historical events impact the actions of farming lobbies and legislators. Data on direct campaign donations comes from the Federal Election Commission. These data contain information on what industry the donating PAC represents, along with data on the recipients of campaign donations. The donation data is what is referred to in the political economy literature as “hard money” donations. These donations are highly regulated by the FEC, with a maximum contribution limit per PAC. Farming interests are able to bypass these contribution limits by creating more PACs, assuming organizational costs are sufficiently low. These data are merged with information on legislator characteristics, provided by a data repository maintained by Charles Stewart III. These data include chamber seniority, committee membership, committee seniority and party affiliation. A temporally consistent measure of political ideology comes from Lewis, Poole and Rosenthal. Information on congressional votes is provided by Civic Impulse LLC. Production data by crop is acquired from the USDA NASS, while data on farming demographics comes from the Bureau of Economic Analysis. Using reasonable assumptions, these data are converted from county level data to congressional district level geospatial shape files using data provided by Lewis, DeVine, Pitcher and Martis. Data are aggregated to the level of the crop lobby. That is to say, donations from multiple PACs representing the same crop to the same legislator are aggregated together. The unit of observation is a donation from a specific agricultural interest to a specific legislator in a given election cycle in which a farm bill vote take place. The time series consists of the 1985, 1990, 1996, 2002 and 2008 farm bills. Crop lobbies included in this model are the cotton, peanuts, rice, sugar beets and sugar cane lobbies. The model is in the form of a simultaneous probit-tobit model as outlined by Chappell (1982). The probit equation models the vote decision of the legislator on the farm bill. The probability that the legislator votes yes is a function of the amount of campaign donations received from various agricultural PACs, the initial policy position of the legislator (i.e. political ideology) and the farming related demographic attributes of the legislator’s district. The tobit equation represents the donation decision of the farming interest. This decision is a function of legislative power (such as committee membership and seniority), the probability of reelection, and the level of crop production in the legislator’s district. The model is identified through the use of exclusion restrictions. My model extends this framework through the use of a pooled cross section over multiple time periods. By treating each farm bill as a new observation of the same event, this approach allows for the use 1 temporal indicator variables to study the impacts of historical events on farm bill votes. Historical events studied thus far are the impacts of political regime changes in the House of Representatives and the impacts of legal regime changes in campaign finance law. The preliminary version of this model estimates the relationship between each crop lobby and the legislators separately. That is to say, a two equation model is estimated, with the donation of one crop lobby modeled in a tobit equation and the legislator’s vote modeled in a probit equation. The more innovative version of this model estimates the donation equations of all of the crop lobbies, along with the vote equation simultaneously. This better reflects the interconnected nature of the various farming interests and the legislator’s final farm bill vote decision. Results show that committee membership is a highly significant determinant of how much support a legislator receives. Two committees are tracked; membership on the House Agricultural Committee and the House Appropriations Committee. While the farm bill has little to do with appropriations, this is included because funding the programs of the farm bill requires separate legislation, which is drafted by the appropriations committee. This makes appropriations committee members important allies in any policy involving the disbursement of federal funds. In all versions of this model, agricultural committee membership has a positive and highly significant impact on the donation decision. For most crop lobbies, appropriations committee membership is also highly significant and positive. This suggests that farming PACs recognize that they also need the support of legislators that control funding. For most crop lobbies, the level of production of their crop in a legislator’s district has a positive and significant impact on the level of donations received. This suggests that farming lobbies support legislators that represent the districts that their members reside in. The impact of party majority is highly significant for most crop lobbies. The sign on the effect varies between farming interests, suggesting partisan heterogeneity among different groups of farmers. In the vote equation, political ideology, farming demographics and donations have significant impacts on the decision to vote yes, when each lobby is estimated separately and jointly. Results appear to be robust to specification, and the correlation coefficients between the equations imply that the system is, in fact, endogenous. The donation equations of the various farming lobbies are highly correlated with each other. These results extend the literature on agricultural political rent seeking by extending the previous, isolated analyses to a more broad analysis over a pooled cross section. These results suggest that political regime changes have a significant impact on both legislators and special interests, and demonstrate the importance of the appropriations committee to the agricultural sector, which has not been studied by previous research.