household, and farm resource/economic data. CEAP-ARMS was completed in 2004 for wheat
(882 farms across 16 States) and in 2005 for corn (489 farms across 4 States). By linking producer
behavioral, farm resource, and environmental data, USDA hopes that this survey instrument will
help to improve its ability to design, implement, and monitor conservation programs consistent with
its resource and environmental policy goals.
This study used the 2005 CEAP-ARMS data for corn production to first compare key
operator, field, farm, economic, and environmental characteristics of conservation program
participants with non-participants, by farm-size class. We hypothesize that producer decisions to
allocate acres to infield or perimeter-field conservation structures are likely correlated with the
acreage allocation decision for crop production on the field. We then estimate a cost-function based
technology adoption model of producer decisions regarding the allocation of field-level acres
between corn production and infield and perimeter-field conservation structures to examine how
these conservation choices differ between program participants and non-participants, while
accounting for differences in other field, farm, and environmental factors. Our null hypothesis is
that the average number of conservation structural practice acres across U.S. corn acres supplied by
growers participating in a conservation program is not different from non-participants. Infield
conservation structures include terraces, grassed waterways, vegetative buffers, contour buffers,
filter strips, and grade stabilization structures. Perimeter-field conservation structures include
hedgerow plantings, stream-side forest and herbaceous buffers, windbreaks and herbaceous wind
barriers, field borders, and critical area plantings. A Generalized Estimating Equations (GEE)
procedure is used to estimate two models. The cost-function models estimate field-level, producer
acreage allocation decisions, first, as a function of normalized production input prices (Model I),
and second, as a function of normalized input prices and several key exogenous variables reflecting
the potential influence of a variety of field, farm, and environmental characteristics (Model II). The