Policy Analysis Intentions of UK Farmers toward Biofuel Crop Production: Implications for Policy Targets and Land Use Change ELIZABETH H. A. MATTISON* AND KEN NORRIS Centre for Agri-Environmental Research, School of Agriculture, Policy and Development, University of Reading, Earley Gate, P.O. Box 237, Reading, RG6 6AR, United Kingdom
The United States and the European Union have set targets for biofuel production to decrease reliance on fossil fuels and to reduce fossil carbon emissions. Attainment of biofuel targets depends upon policy and infrastructure development but also on production of suitable raw materials. Production of relevant crops relies on the decisions that farmers make in their economic and political environment. We need to identify any farmer-related barriers to biofuel production and to determine whether novel policy and technology are required to meet targets. These aspects of the emerging biofuel industry are relevant across international barriers and have not yet been addressed quantitatively. We describe a case study from the UK of farmers’ intentions toward producing two biofuel crops for which refining capacity either exists or is under construction. Given farmers’ intentions, current land use, and conversion efficiency, we estimate potential biofuel production. These estimates indicate that EU targets are not achievable using domestically grown raw materials without policy intervention, use of alternative feedstocks, and either significant improvements in processing efficiency or largescale changes in land use.
Introduction Decreasing our dependence on fossil fuels and reducing fossil carbon emissions is a global issue, and one strategy gaining in prominence is the use of biofuels for transport. The U.S. Energy Policy Act of 2005 established the Renewable Fuel Standard, setting requirements for biofuel consumption. The U.S. Congress mandated a 4 billion gallon total for national biofuel consumption in 2006, with an increase to 7.5 billion gallons by 2012. Most of the biofuel produced in the U.S. comes from corn (maize), and the use of this crop for biofuel is estimated to more than double by 2015 (1). One part of the EU program for compliance with the Kyoto protocol is increasing biofuel consumption (2). Directive 2003/30/EC outlined reference targets for biofuel use of 2.0% of all petrol and diesel (by energy content) by the end of 2005 and 5.75% by the end of December 2010. EU member states must establish their targets in 2007. To increase biofuel use and hit policy targets, developed regions such as the EU and North America could import * Corresponding author phone: +441189875123; fax: +441193786067; e-mail:
[email protected]. 10.1021/es062211v CCC: $37.00 Published on Web 07/18/2007
2007 American Chemical Society
products from overseas (for example, bioethanol from sugarcane or biodiesel from palm oil). However, there is growing concern that imported biofuel threatens sustainability because increased production of crops such as palm oil and sugarcane comes at the expense of deforestation in areas of high biodiversity (3, 4). An alternative strategy is the rapid expansion of domestic biofuel production through an increase in the agricultural area devoted to appropriate crops, with an increase in refining capacity. Any barriers to biofuel target attainment should be identified early in the development of an emerging industry. Increased biofuel production depends on policy, infrastructure, and market accessibility, but it is also heavily reliant on farmers’ decisions. Do any aspects of farmers’ intentions hinder biofuel production? If farmers are willing to produce appropriate crops, do any other barriers exist and what policies are required to surmount these barriers? To assess these aspects of biofuel production, we can investigate the intentions of relevant farmers and analyze the drivers of these intentions. This framework allows estimation of what farmers can achieve in their current political and technological environment, enables problem identification, and provides evidence of whether novel policies and technology are required if biofuel targets are to be met. These aspects of the biofuel industry are relevant across international barriers and have not yet been addressed using a systematic, quantitative approach. Here we present a case study of farmers from the UK that explores intentions toward the introduction of biofuel crops on arable farms and provides evidence of potential biofuel production. Data were collected from questionnaires supplied to farmers in the county of Norfolk, UK. Farmers were asked about their intentions toward two crops for which processing capacity existed or was under construction at the time: oilseed rape and sugar beet. While the use of wheat for bioethanol production is now commonly discussed, at the time of this study there were no planned facilities for processing wheat. To date, except for experimental use, such facilities are not available in the UK. The region was chosen for two reasons. The study was linked to ecological research based in the area. Second, this region would probably affect domestic biofuel production because it is a center of arable production in the UK where both sugar beet and oilseed rape are common crops. The questionnaire was structured using the theory of planned behavior (TPB) a social-psychological model which attempts to predict and understand people’s behavior in specific contexts (5, 6). The results were used to predict biofuel production given the farmers’ willingness to provide raw materials. The implications for attainment of biofuel targets are discussed in light of these predictions. The main assumptions of TPB are that people behave rationally, their behavior is a function of the information or beliefs they have, and behavior is immediately preceded by behavioral intention. The greater the behavioral intention, the more likely it is an individual will perform the behavior. The individual weighs up all available information and influence; from personal instinct, policy, advisory services, the media, family, friends, and peers and forms their beliefs (5). Three aspects are considered to influence behavioral intention. Beliefs about the likely consequences of the behavior produce an attitude toward the behavior. Beliefs about others normative expectations result in perceived social pressure or subjective norm. Beliefs about the presence of VOL. 41, NO. 16, 2007 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
5589
FIGURE 1. The Theory of Planned Behavior (adapted from ref 13). issues that may help or hinder performance of the behavior give rise to perceived behavioral control (7). Importantly, each aspect can be weighted, as behavioral intention is predicted to be influenced only by those aspects which are salient to the individual, and this may vary from person to person (8). Figure 1 provides an overview of the theory. Attitude (A) toward a specific behavior is a measure of the degree to which it is positively or negatively valued. Attitude could be assessed directly by asking an individual whether a behavior is good or bad, or it can be determined indirectly using a set of behavioral beliefs, linking the behavior to various outcomes. For outcome i the strength of belief in that outcome (bs) is weighted by the evaluation of that outcome (oe). The products of each belief strength and outcome evaluation in the set are summed and considered to be directly proportional to attitude:
A∝
∑oe bs i
(1)
i
Subjective norm (SN) is a measure of the perceived social pressure to engage or not to engage in a behavior. An individual’s normative belief (nb) about social referent j is weighted by his or her motivation to comply (mc) with social referent j. Again, all the products of the set of normative beliefs are summed and assumed to be directly proportional to the individual’s subjective norm:
SN ∝
∑mc nb j
j
(2)
Perceived behavioral control (PBC) is a measure of people’s perceptions of their ability to perform a specific behavior. It can be measured directly by asking people how easy or difficult it would be to perform the behavior. PBC can also be determined using a set of control beliefs: beliefs about things that may facilitate or impede performance of the behavior. For control belief k the strength of belief (cb) is weighted by the perceived power (pb) of k to facilitate or inhibit behavior. The sum of the products in the set of control beliefs is assumed to be directly proportional to an individual’s perceived behavioral control.
PBC ∝
∑pb cb k
k
(3)
A large amount of literature discusses the development and application of TPB. Much applied research relies on the assumption that behavioral intention is an accurate proxy for actual behavior, i.e., the actual behavior of individuals is not recorded. Empirical evidence suggests that this assumption holds; when behaviors pose no serious control issues, then behavioral intention is a good predictor of behavior (e.g., refs 9, 10). When behaviors are not completely of the 5590
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 41, NO. 16, 2007
subject’s own volition then perceived behavioral control in addition to behavioral intention is an important predictor of actual behavior (11). TPB and the theory of reasoned action (TORA; which measures attitude and subjective norm but not perceived behavioral control) have recently been applied to several aspects of farmer decision making, including environmental management (12, 13) farm business management (14), and uptake of new technology (15, 16).
Materials and Methods A TPB study requires exploratory work to identify potential salient beliefs for each TPB construct (16). A telephone interview was carried out with a pilot sample of 30 farmers chosen at random from the Norfolk issue of the Yellow Pages. The interview was semistructured to establish whether the farmer was aware of biofuel crops; whether the farmer currently grew any biofuel crops; what advantages and disadvantages the farmer thought were associated with biofuel crops; who he or she would go to for advice about new technology, and whom he or she would consult before deciding about the use of new technology. Literature on farmer adoption of new technology was reviewed to see whether any outcome evaluations and social referents were thought to be important in other studies but not mentioned by the interviewed farmers. Eleven outcome evaluation/belief strength statements plus a list of 12 social referents to use for normative belief/motivation comply statements about growing biofuel oilseed rape and biofuel sugar beet were compiled. We designed a written questionnaire that assessed the level of intention farmers had to grow biofuel sugar beet and/or biofuel oilseed rape on their farm. Perceived behavioral control was measured using two questions. These encompassed perceived difficulty (i.e., having the right training and resources) and perceived control (i.e., whether the farmer felt that they were in control of the decision to grow biofuel crops). Attitude and subjective norm were then assessed. All potential responses were on a fully anchored five-point scale (labeled tick boxes). The question structure and answer options for each TPB construct are outlined in Table S1, Supporting Information. Farm information including size and enterprise details was also recorded. The questionnaire was posted to farmers in February 2005. To identify the largest possible sample of farmers in Norfolk, both the Cambridge and Norfolk 2005 Yellow Pages were used, as many farms in Norfolk were listed in the Cambridge issue. Thus, 972 farms (with those questioned during the pilot phase excluded) were identified. Whereas the UK Department for Environment, Food and Rural Affairs (Defra) can provide a near 100% coverage sample of farmers within a required geographical area, this service is available only if the research is funded by Defra. The Yellow Pages
remains the alternative with the most comprehensive coverage of UK farmers (17, 18). Behavioral intention and perceived behavioral control answer options were numerically coded from -2 to +2 for analysis. Outcome evaluations were coded from -2 to +2, and belief strengths were coded from +1 to +5. Normative belief statements were coded from +1 to +5, and motivation to comply was coded -2 to +2. Outcome attitudes (outcome evaluation × belief strength) and subjective norms (normative belief × motivation to comply) were calculated. Thus the potential range of scores associated with each outcome evaluation statement and each subjective norm was weighted and ran from -10 to +10. For example, if a farmer was undecided about a particular outcome evaluation, it received a score of zero, thus the outcome attitude was also zero, representing the fact that this particular outcome was not salient to the farmer in question. In the same way, if a farmer was undecided about his motivation to comply with a social referent, it received a score of zero. Therefore, the subjective norm also scored zero, reflecting the fact that the social referent was not salient to the farmer in question. The outcome attitudes were summed and the subjective norms were summed to give overall attitude and subjective norm scores. The dependent variable “behavioral intention” was ordinal (categorical with a natural order). Ordinal logistic regression was used to investigate whether attitude, subjective norm, perceived difficulty, and perceived control were significant indicators of intention to grow biofuel sugar beet and biofuel oilseed rape. Analysis was carried out using SPSS version 12.0.1. Attitude and subjective norm scores were treated as continuous covariates; perceived difficulty and perceived control as categorical factors. Ordinal logistic regression uses a proportional odds model with the assumption that the effects of independent variables are consistent over all levels of the dependent variable (19). If this assumption was not met when factors had five categories, the top two categories and the bottom two categories of each were collapsed to convert them to three category factors. Perceptions of particular aspects of biofuel crops that are potential drivers of uptake by farmers may be identified if individual outcome attitudes are significantly positively correlated with behavioral intent. A significantly negative correlation would indicate a barrier to adoption (16). Spearman’s correlation coefficients were used to describe the relationship between behavioral intention and each outcome attitude. To identify social referents that would potentially have the most influence over farmers’ decisions to grow biofuel crops we calculated mean motivation-tocomply scores for each social referent.
Results We received 278 questionnaires that were useable for at least part of the planned analysis, giving an adjusted response rate of 29.32%. The total land area covered by questionnaire responses was 88 551 ha, 21% of the total agricultural area in Norfolk. Half of farmers intended to grow biofuel sugar beet, 34% intended to grow biofuel oilseed rape. In both cases, however, a third of the farmers sampled were undecided (Table 1). Most farmers felt that growing biofuel sugar beet and biofuel oilseed rape would be easy or very easy to do. Slightly more farmers fell into these categories for sugar beet than for oilseed rape (Table 1). As 78% of the farms surveyed were producing sugar beet while only 25% produced oilseed rape, the greater intention and perceived ease associated with biofuel sugar beet production was to be expected. This probably reflected the farmers’ previous experience with the crops and (in the case of oilseed rape) the difficulties associated with unsuitable soil types that exist in some parts of Norfolk. In addition, reforms to the EU sugar
TABLE 1. Questionnaire Response Summary: Percentage of Responses for each Level of Behavioral Intention, Perceived Difficulty and Perceived Controla crop
TPB sample construct size
-2
-1
9.35 4.68
6.83 33.45 34.17 16.19 3.24 9.71 44.60 37.77
0
1
2
intention perceived difficulty
278 278
intention biofuel oilseed rape perceived difficulty
278 278
12.95 17.99 34.89 19.06 15.11 4.68 11.87 16.91 35.25 31.29
biofuel crops
273
14.65 16.12 16.12 23.08 30.04
biofuel sugarbeet
perceived control
a Response category scoring in each case (-2 to +2) corresponds to the measurement scale detailed in Table S1, Supporting Information.
TABLE 2. Questionnaire Response Summary: Attitude and Subjective Norm Scores ( See Text for Details of Score Calculation)
attitude toward biofuel sugar beet attitude toward biofuel oilseed rape subjective norm for biofuel crops
sample size
range
mean
std. error
277
-18 - 95
31.23
1.30
277
-18 - 95
28.57
1.31
273
-84 - 120
27.24
2.41
sector, including a significant reduction in the price associated with sugar beet crops were in proposal form at the time (Council regulation (EC) no. 215/2006). This indicates that farmers were seeking an alternative sugar beet market. More than half the farmers sampled felt they had a lot or complete control over growing biofuel crops on their farm (Table 1) and mean attitude scores for each crop were very similar. Mean attitude and subjective norm scores were positive although subjective norm was more variable than attitude (Table 2). Are Attitude, Subjective Norm, and Perceived Behavioral Control Significant Indicators of Intention to Grow Biofuel Sugar Beet and Oilseed Rape? The models resulting from ordinal regression analysis are shown in Tables S2 and S3, along with notes on interpretation of the results. The initial ordinal regression model for intention to grow biofuel sugar beet did not fulfill the assumption that the effect of the independent variables was consistent over all levels of the dependent variable. Amalgamation of the five category factors describing perceived difficulty and perceived control into three categories altered this (test of parallel lines p ) 0.194). The resulting model explained a significant amount of the variation in the data (model chi-squared test p < 0.001). Both attitude and normative belief scores were significantly related to intention to grow biofuel sugar beet. A more positive attitude toward biofuel sugar beet indicated greater intention to produce this crop on the farm. Higher levels of perceived social pressure (subjective norm) to grow biofuel sugar beet were associated with greater intention. Farmers who felt that growing biofuel sugar beet would be difficult and farmers who were unsure how difficult it would be had significantly less intention of growing biofuel sugar beet than those who felt it would be easy. Farmers who felt they had no or little control over growing biofuel sugar beet actually had a greater intention to grow it than those who felt they had a lot of/complete control. This result indicated that farmers keen to grow biofuel sugar beet felt the decision was beyond their control. Measures of attitude, VOL. 41, NO. 16, 2007 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
5591
TABLE 3. Spearman’s Correlation Coefficients for Individual Outcome Attitudes and Intention to Grow Biofuel Sugar Beet (Significance Levels: ***p < 0.001, **p < 0.01, *p < 0.05) outcome attitude biofuel crops are a renewable source of energy biofuel crops are a good alternative to oil biofuel crops could safeguard the future of arable faming biofuel crops produce more fuel than is used during production biofuel crops have higher gross margins biofuel crops are good for the environment biofuel crops are easy to do the paperwork for biofuel sugar beet fits in with my current cropping plans biofuel crops are popular with the general public a viable market exists for biofuel crop products
n 277
correlation coefficient 0.319***
277
0.301***
277
0.375***
277
0.157**
277
0.140*
277
0.203**
277
0.074
276
0.634***
276
0.253***
276
0.182**
subjective norm, and perceived behavioral control were therefore all significantly related to intention to grow biofuel sugar beet. Nagelkerke’s r2 approximation for this model was 0.448, indicating that there was a moderately strong association between the independent and dependent variables. The initial ordinal logistic regression model of intention to grow biofuel oilseed rape fulfilled the assumption that the effects of independent variables were consistent over all levels of the dependent variable (test of parallel lines p ) 0.186). The model described a significant amount of variation in the data (model chi-squared test p < 0.001). Attitude toward biofuel oilseed rape was significantly positively related to intention to grow the crop. The perceived difficulty of growing biofuel oilseed rape was significantly negatively related to intention; the gradient of odds ratios produced by the model indicated that greater perceived difficulty was associated with decreasing intention to grow biofuel oilseed rape. Nagelkerke’s r2 approximation for this model was 0.573, indicating a fairly strong association between the independent and dependent variables. Both subjective norm and perceived control were not significantly related to intention to grow biofuel oilseed rape. This result indicated that other people’s views and perceived off-farm barriers were not important to the farmers sampled during this study when deciding whether to grow biofuel oilseed rape. Drivers of and Barriers to Biofuel Crop Production Tables 3 and 4 show that with the exception of attitude toward the statement that “biofuel crops are easy to do the paperwork for”, all outcome attitudes were significantly positively correlated with intention to grow biofuel sugar beet and oilseed rape. These results indicated that almost all the aspects investigated during this study were potential drivers of intention to grow biofuel crops. For both biofuel sugar beet and oilseed rape, the closest correlations between intention and attitude occurred when farmers considered whether the crops fit in with their existing cropping plans and whether biofuel crops could safeguard the future of arable farming in the UK. There were no obvious attitudinal barriers to producing biofuel crops although farmers generally disagreed with the idea that the paperwork associated with biofuel crops would be easy to do. The farmers questioned had some motivation to comply with the advice of a wide range of social referents with regard to biofuel crops (Figure S1). On average, the farmers were 5592
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 41, NO. 16, 2007
TABLE 4. Spearman’s Correlation Coefficients for Individual Outcome Attitudes and Intention to Grow Biofuel Oilseed Rape (Significance Levels: ***p < 0.001, **p < 0.01) outcome attitude
n
correlation coefficient
biofuel crops are a renewable source of energy biofuel crops are a good alternative to oil biofuel crops could safeguard the future of arable faming biofuel crops produce more fuel than is used during production biofuel crops have higher gross margins biofuel crops are good for the environment biofuel crops are easy to do the paperwork for biofuel crops are popular with the general public a viable market exists for biofuel crop products biofuel oilseed rape fits in with my current cropping plans
277
0.283***
277
0.335***
277
0.388***
277
0.172**
277
0.185**
277
0.215***
277
0.074
276
0.276***
276
0.212***
276
0.744***
most motivated to follow the advice of their agronomist about biofuel crops, followed by other experienced farmers and the crop research organization, the Arable Group. Farmers tended not to be interested in consulting websites.
Discussion Both attitude and perceived difficulty were consistently significant indicators of intention to grow biofuel crops. More positive attitude and less perceived difficulty indicated greater behavioral intention. The influence of subjective norm was less certain. While there was a significant positive relationship between subjective norm and behavioral intention in the case of biofuel sugar beet, there was not for biofuel oilseed rape. Perhaps farmers were more confident about the appropriateness of producing oilseed rape for biofuel as a small amount of oilseed rape is currently grown for biodiesel in the UK. The relationship between perceived control and intention was also only significant in the case of sugar beet. Farmers who intended to grow biofuel sugar beet felt they had less control over the decision than farmers who had little or no intention. This indicated a barrier to uptake of sugar beet production for the biofuel industry as farmers felt the decision was beyond their control, probably because at the time of the questionnaire there was no refining capacity for bioethanol from crops in the UK. However, a refinery is under construction in Norfolk (www.britishsugar.co.uk). Are EU Biofuel Targets Achievable and What are the Potential Changes in Land Use? Using the results of this survey scenarios were developed to predict potential biofuel production and land use change. The proportion of the sugar beet area sampled during this survey that was grown by farmers with the intention of growing biofuel sugar beet was used as the potential proportion of the total UK sugar beet area that would be turned over to bioethanol production. In the same way, the proportion of the oilseed rape area sampled that was grown by farmers intending to grow oilseed rape for biodiesel production was used as the potential proportion of the total UK oilseed rape area (not grown on set-aside land) that would be used for bioethanol production. In the UK farmers may grow oilseed rape on their set-aside allocation if the crop is used for biodiesel production (20). The potential proportion of the UK set-aside area which would be used to grow biodiesel was, therefore, also calculated. Farmers with more than 19.48 ha are required to
FIGURE 2. Potential UK production of bioethanol and biodiesel expressed as percentages of UK petrol and diesel consumption by energy content, based on intentions and land use of the farmers surveyed. Black bars indicate bioethanol production from sugar beet; white bars indicate biodiesel production from oilseed rape grown both in existing areas and on set-aside land. Calculations were based on the following information. Average UK sugar beet yield of 55.22 tons/ha, average UK oilseed rape yield of 3.08 tons/ha (24). Assuming 1 ton of bioethanol is produced from 14.61 ton sugar beet and 1 ton of biodiesel is produced from 2.699 tons of oilseed rape (22). Unleaded petrol net calorific content is 43.99 MJ/L, bioethanol net calorific content is 26.8 MJ/L, diesel net calorific content is 42.82MJ/L, biodiesel net calorific content is 32.27MJ/L (25). UK petrol sales in 2004 is 19 484 000 tons, UK diesel sales in 2004 is 18 514 000 tons (www.dti.gov.uk/energy/statistics/source/ oil/page18470.html). UK sugar beet area is 148 000 ha, UK oilseed rape area is 519 000 ha, UK set aside area ) 559 000 ha (24). set aside a proportion of their land based on the area they devote to production, excluding permanent pasture and permanent crops (20). The fraction of land associated with farmers intending to grow rape for biodiesel production was calculated and used as the potential proportion of UK setaside that would be turned over to biodiesel production. In the first scenario, only those farmers who had a strong intention were included in the calculations. In the second scenario, farmers who intended and who strongly intended were included. In the third scenario, farmers who were undecided were included as well as those who intended to grow biofuel crops. The results, expressed in terms of the percentage of UK petrol and diesel consumption by energy content, are presented in Figure 2 and compared with EU reference targets. Inclusion of the energy content factor was important as this is specifically stated in directive 2003/30/E and has not been included in similar published calculations (see ref 21). Despite the crude estimates, even the most optimistic scenario for biofuel oilseed rape and sugar beet uptake by farmers showed that the UK would fall short of EU reference targets for 2010. Given average yields, expected conversion figures and the difference in energy content between conventional fuels and biofuels (Figure 2), the UK would have to grow approximately 412 000 ha of sugar beet and 981 000 ha of oilseed rape to attain the recommended target. These figures indicate that even if all existing sugar beet and oilseed rape production was turned over to the biofuel industry, an increase in land area devoted to sugar beet and oilseed rape of 178 and 89% respectively would be required. There is the potential for large changes in agricultural land use, and additional feedstocks such as wheat and biomass are clearly required for bioethanol production. Because sugar beet is an important component of arable rotations in Norfolk, the intentions of 278 farmers surveyed during this study toward biofuel oilseed rape may not represent the country as a whole. Nonetheless, in the scenarios described above, the assumed intention levels of
farmers ranged between 15 and 69% of those producing oilseed rape and 16-84% of those producing sugar beet. At the time of the survey, these crops were the most profitable break crops (noncereal arable crops) available and were therefore grown wherever agronomically feasible. In addition, because of reforms to the EU sugar regime, farmers were seeking a new market for their sugar beet crops. A survey of arable farmers conducted elsewhere in the country is unlikely to yield results outside this range. However, further study is needed to identify the intentions of farmers in other regions and the likely land-use change due to policies designed to encourage biofuel production. Information for Policy Makers Several issues relevant to policy makers arose during this survey. Decisions on the best crops to encourage are needed. For example, during this study the strongest relationship between intention and attitude implied that farmers’ main concern about growing a biofuel crop was whether it would easily fit in with their current business plan. Interestingly, however, 50% of surveyed farmers who had some intention to produce oilseed rape for biodiesel were not producing oilseed rape at the time of the questionnaire. Most of these farmers were sugar beet producers, suggesting that the area of oilseed rape could increase at the expense of sugar beet. Only 11% of those who intended to grow biofuel sugar beet were not already producing sugar beet on their farm. This finding implies that the UK area devoted to sugar beet production is unlikely to increase due to farmers introducing the enterprise on their farm unless encouraging policy measures are in place. The amount of bioethanol produced per hectare of sugar beet compares favorably both with that produced per hectare of wheat and with the amount of biodiesel produced per hectare of oilseed rape (22). Although issues relating to refining capacity and potential improvements in refining efficiency also need to be considered, the need to increase biofuel production in the face of limited land availability suggests that sugar beet should not be ignored in preference to other crops. We must recognize that appropriate policies may vary geographically even within countries or states. For example, existing sugar beet production is centered in the east of England. Policies designed to encourage bioethanol production from sugar beet would be more efficient if targeted toward this region as the results from this survey indicated that few farmers who are not producing sugar beet intended to grow the crop in the future. In other areas of the country, it may be more appropriate to encourage bioethanol production from wheat or biomass or to concentrate on biodiesel production from oilseed rape or recycled vegetable oil. Decisions about infrastructure development are also required. For example, in the case of sugar beet, refinery capacity should be in the same region as crop production. If not, both infrastructure development and sufficient encouragement for farmers to introduce the crop in new areas would be required. We also must assess the potential impacts on water quality and biodiversity associated with different biofuel crops. The consequences of an increase in agrochemical use due to cultivation of set-aside land and potential changes in habitat structure for associated wildlife due to expansion of the area of biofuel crops need to be investigated. For example, production of oilseed rape at the expense of cereal crops could have a detrimental impact on farmland birds as cereal stubbles left available over the winter are an important food source (23). Appropriate promotion of related policies and training of farmers will be needed. Producers of biofuel crops must secure a contract with a processor prior to the growing season and fulfill detailed crop quality requirements with associated documentation. This “paperwork” may be sufficient to VOL. 41, NO. 16, 2007 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
5593
discourage potential growers and it is likely that additional training and communication resources will be required if the biofuel industry in the UK is to become a mainstream occupation for farmers. Our data regarding motivation to comply (Figure S1) indicated that support and information about administration, cropping plans, and market opportunities would be most helpful if provided by agronomists and other farmers (e.g., through workshops), rather than through websites or the farming press. A large proportion of farmers were keen to growing biofuel crops, a good indication that UK domestic biofuel production could be significantly higher than current levels. Even with the majority of arable farms contributing to production of crops for which some processing capability exists, however, our results imply that biofuel targets will not be achieved without additional feedstocks, large-scale changes in land use, or significant technological development. There is evidence that infrastructure development is too slow and policy support insufficient. For example, for wheat to become a major biofuel feedstock, certain conditions must be met. First, processing capacity requires significant investment and development. Second, upward shifts in the price of wheat in the UK from around 72 £/tonne at the beginning of 2006 to 95 £/tonne in 2007 (www.fwi.co.uk) significantly alters the economic viability of wheat as a feedstock without further policy intervention. Farmers feel that the decision to produce biofuel sugar beet is beyond their control and many farmers are undecided about biofuel crop production. Policy makers need to understand that efficient targeting of measures designed to promote biofuel crop production and infrastructure development should account for regional variation in cropping practice. Because of the potential for widespread changes in agricultural land use (particularly the use of land currently set aside from production), researchers need to identify the potential impacts on associated water resources and biodiversity to provide information on the most sustainable manner in which to promote biofuel production.
Acknowledgments E.H.A.M. was funded by the BBSRC. We thank Alison Bailey and Tahir Rehman for methodological advice and four anonymous reviewers for their helpful comments.
(8) Armitage, C. J.; Christian, J. From attitudes to behaviour: Basic and applied research on the Theory of Planned Behaviour. Curr. Psychol. 2003, 22, 187-195. (9) Fishbein, M.; Azjen, I. Attitudes and Voting Behaviour: An Application of the Theory of Reasoned Action. In Progress in Applied Social Psychology; Stephenson, G. M., Davis, J. M., Ed.; Wiley: London, 1981; Vol. 1, pp 253-313. (10) Manstead, A. S. R.; Proffitt, C.; Smart, J. L. Predicting and understanding mothers’ infant-feeding intentions and behavior: Testing the theory of reasoned action. J. Personality Social Psychol. 1983, 44, 657-671. (11) Ajzen, I.; Madden, T. J. Prediction of goal-directed behavior: Attitudes, intentions, and perceived behavioral control. J. Exp. Social Psychol. 1986, 22, 453-474. (12) Beedell, J. D. C.; Rehman, T. Explaining farmer’s conservation behaviour: Why do farmers behave the way they do? J. Environ. Manage. 1999, 57, 165-176. (13) Fielding, K. S.; Terry, D. A.; Masserb, B. M.; Bordiab, P.; Hogg, M. A. Explaining landholders’ decisions about riparian zone management: The role of behavioural, normative, and control beliefs. J. Environ. Manage. 2005, 77, 12-21. (14) Bergevoet, R. H. M.; Ondersteijn, C. J. M.; Saatkamp, H. W.; van Woerkum, C. M. J.; Huirne, R. B. M. Entrepreneurial behaviour of Dutch dairy farmers under a milk quota system: goals, objectives and attitudes. Agric. Syst. 2004, 80, 1-21. (15) Cook, A. J.; Fairweather, J. R.; Campbell, H. R. New Zealand Farmer and Grower Intentions to use Genetic Engineering Technology and Organic Production Methods; Research Report no. 243; Agribusiness and Economics Research Unit, Lincoln University: Christchurch, NZ, 2000. (16) Garforth, C.; Rehman, T.; McKemey, K.; Tranter, R.; Cooke, R.; Yates, C.; Park, J.; Dorward, P. Improving the design of knowledge transfer strategies by understanding farmer attitudes and behaviour. J. Farm Manage. 2004, 12, 17-32. (17) Emerson, H.; MacFarlane, R. Comparative bias between sampling frames for farm surveys. J. Agric. Econ. 1995, 46, 241-251. (18) Burton, R. J. F.; Wilson, G. A. The Yellow Pages as a sampling frame for farm surveys: assessing potential bias in agrienvironmental research. J. Rural Stud. 1999, 15, 91-102. (19) Norusis, M. In SPSS 13.0 Advanced Statistical Procedures Companion; Prentice Hall: London, 2005; pp 69-89.
Supporting Information Available
(20) Defra Set-aside Handbook and Guidance for England 2006 edition, 2005. http://www.defra.gov.uk/farm/capreform/pubs/ pdf/Setaside2006.pdf (accessed 2 July 2006).
Description of survey question structure. Statistical modeling results. Additional survey results. This material is available free of charge via the Internet at http://pubs.acs.org.
(21) Powlson, D. S.; Richie, A. B.; Shield, I. Biofuels and other approaches for decreasing fossil fuel emissions from agriculture. Ann. Appl. Biol. 2005, 146, 193-201.
Literature Cited (1) USDA. USDA Agricultural Baseline Projections to 2015, 2006. http://www.ers.usda.gov/publications/oce061/oce20061a.pdf (accessed 4 September 2006). (2) Commission of the European Communities. Communication from the Commission: An EU strategy for Biofuels, 2006. http:// register.consilium.eu.int/pdf/en/06/st06/st06153.en06.pdf (accessed 14 July 2006). (3) Lynas, M. High Tide: News from a Warming World; Harper Collins: London, 2005. (4) WWF Oil palm plantations and deforestation in Indonesia. What role do Europe and Germany play? WWF, 2002. http:// assets.panda.org/downloads/oilpalmindonesia.pdf (accessed 14 July 2006). (5) Ajzen, I. From Intentions to Actions: A theory of Planned Behaviour. In Action-Control: From Cognition to Behaviour; Kuhi, J., Beckmann, J., Ed.; Springer: Heidelberg, 1985; pp 1139. (6) Ajzen, I. The theory of planned behaviour. Organ. Behav. Hum. Decision Process. 1991, 50, 179-211. (7) Ajzen, I. Perceived behavioral control, self-efficacy, locus of control, and the theory of planned behavior. J. Appl. Social Psychol. 2002, 32, 665-683. 5594
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 41, NO. 16, 2007
(22) Elsayed, M. A.; Matthews, R.; Mortimer, N. D. Carbon and Energy Balances for a Range of Biofuels Options Project no. B/B6/ 00774/REP, Department of Trade and Industry, 2003, http:// www.dti.gov.uk/files/file14925.pdf?pubpdfdload)03%2F836 (accessed 3 July 2006). (23) Gillings, S.; Newson, S. E.; Noble, D. G.; Vickery, J. A. Winter availability of cereal stubbles attracts declining farmland birds and positively influences breeding population trends. Proc. Royal Soc. Ser. B 2005, 272, 733-739. (24) Defra “Agriculture in the United Kingdom,” 2005. http:// statistics.defra.gov.uk/esg/publications/auk/2005/Complete.pdf (accessed 7 July 2006). (25) Woods, J.; Bauen, A. Technology status review and carbon abatement potential of renewable transport fuels in the UK Project no. B/U2/00785/REP,” Department of Trade and Industry, 2003, http://www.dti.gov.uk/files/ file15003.pdf?pubpdfdload)03%2F982 (accessed 10 June 2006).
Received for review September 15, 2006. Revised manuscript received June 4, 2007. Accepted June 6, 2007. ES062211V
