AGRICULTURAL PRODUCERS' WILLINGNESS TO PAY FOR REAL-TIME MESOSCALE WEATHER INFORMATION

Real-Time Weather Information 365

Table 3. Distribution of Respondents’ Willingness to Pay for Raw Weather Data and Raw
Data Plus Value-Added Weather Information

free. These were considered protest bids and were excluded from the estimation sample. The
final sample sizes for model estimation were 107 for the raw data model and 114 for the
value-added model. The higher number of protest bids given for the raw weather data as
compared with the raw data∕value-added combination suggests that respondents are more
likely to expect provision of the raw data as a public service.

Willingness to Pay for Mesoscale WeatherInformation

The two objectives of the study were to estimate average willingness to pay and to determine
the characteristics of producers who would pay to access and use mesoscale weather
information. To this end, two maximum likelihood models were estimated: one for the raw
Mesonet data and one for the raw data/value-added information combination. When protest
bids were removed from the regression data, as is standard in CV analyses, the final number
of observations for the raw weather data model was 107 and the final number for the
value-added model was 114. Results of the model estimation are shown in table 4. Chi-
squared statistics testing the joint significance of the models’ parameters indicated that both
models were significant at the 0.001 percent level.⁴

Variables representing payments for agricultural publications, full- versus part-time
farming, gross sales, use of irrigation, and weather-related crop income losses were found
to significantly impact the willingness to pay for raw mesoscale weather data. Specifically,
fanners paying more for agricultural magazines are likely to pay more for the mesoscale
weather data. Full-time farmers could be expected to pay about $0.55 per month less than
part-time fanners (all else constant) for the weather data; this may reflect that they have
more time available to obtain and study existing weather information sources and, as such,
perceive a limited benefit to the mesoscale information. Results suggest that those producers

⁴The fragility of the model results was tested using an abbreviated speci Iication for each model. The new model coefficients
and their significance were virtually identical to the longer models’ results.

<<   <   7   8   9   10   11   12   13   >   >>