Agricultural Markets for Ecosystem Services: Greenhouse Gases, Conservation Practice Adoption & Behavioral Responses
Recent research on the development and refinement of markets for ecosystem services, including markets for agricultural greenhouse gases, adoption of conservation practices, and behavioral responses to alternative program design, was the focus of a workshop Aug. 8, 2013. This workshop was a collaboration of USDA's Economic Research Service (ERS) and Farm Foundation, NFP.
The nation’s farmland produces more than just food and fiber. Environmental goods and service—such as clean air, wildlife and greenhouse gases—are also affected by agricultural activities. Development and refinement of ecosystem services markets is one tool to encourage farmers and ranchers to produce more environmental goods.
Directly or through cooperative research agreements with universities and other partners, USDA's Economic Research Service (ERS) has done extensive research on the opportunities and challenges of creating agricultural, market-based incentives for ecoservices. This includes work on markets for agricultural greenhouse gases, incentives for adoption of conservation practices, and behavioral responses to alternative conservation program design. This workshop highlighted those research findings.
Below are abstracts and PowerPoint presentations concerning the respective papers presented at the workshop. For more information on the workshop, contact Daniel Hellerstein at ERS.
Session I: Offset Supply
Markets for Agricultural Greenhouse Gas Offsets: The Role of Payment Design on Abatement Efficiency
Cloe Garnache, University of California, Davis, and Pierre Merel, Swiss Federal Institute of Technology, (ETH) Zurich
Abstract: This paper investigates the role of agricultural greenhouse gas (GHG) offset payment design on abatement efficiency. We develop a disaggregated positive mathematical programming (PMP) model of California's agriculture calibrated to both economic and agronomic information. Using a biophysical model, we estimate regional yield and GHG responses to production practices for the three principal agricultural GHGs. The model allows simultaneous and continuous changes in water, nitrogen fertilizer, and tillage intensities, and captures crop substitution effects. Results show that second-best policies that rely on regionally aggregated emission factors lead to small abatement efficiency losses relative to the first-best policy with fine-scale emission factors. Because the costs of such second-best policies are substantially lower, the findings suggest these policies would be cost-effective in California. In contrast, second-best policies targeting a single GHG or input use, such as nitrogen fertilizer or water, entail significant abatement efficiency losses.
Credit Stacking in Agri-Environmental Program: Water Quality Trading Programs and Carbon Markets
Adriana Valcu and Catherine Kling, both of Iowa State University, and Sergey Rabotyagov, University of Washington
Abstract: We explore the impact of participation in a carbon offset market on the cost efficiency of a water quality trading program designed for an agricultural watershed. The water quality trading program is a local cap-and-trade program, whereas the carbon market is a wider market (i.e. nationwide) with no specific cap requirement at farm or field level. We compare the outcomes of a point-based trading market for water quality when there is no carbon market with the trading outcomes in the presence of an outside carbon market. The empirical analysis is performed for two pollutants (nitrogen and phosphorus), three water quality goals, and three price levels of carbon offsets using data available for an agricultural watershed in Iowa. The empirical assessment shows that: (a) the realized abatement levels are not very different in the cases where a carbon market is available, (b) the farmers’ total cost of a point-based trading program decreases as the price of carbon increases, (c) as the price of carbon increases the total costs become negative, meaning that farmers obtain extra revenues by selling the carbon offsets, which offset the cost of implementing the abatement practices, (d) the costs of implementing the abatement actions increase, and (e) there are not significant changes in the distribution of the abatement actions. To generate our results, we use two different watershed based models (EPIC and SWAT) together with an optimization algorithm.
Session II: Adoption of Practices
Integrity and Agriculture Landowner Behavior in Afforestation for Carbon Sequestration as part of a Carbon Pricing Program
David Haim and Eric White, both of Oregon State University
Abstract: This research explores the importance, integrity, and adoption of agriculture afforestation within U.S. regions under a carbon market to offset GHG emissions. In particular, we quantify leakage behavior and investigate permanence issues of agricultural land afforestation under simulated national and region-specific carbon offset markets. We use the Forest and Agriculture Sector Optimization Model- Greenhouse Gases model to examine responses between the two sectors as part of the afforestation policy analysis. Modeling results suggest that region-specific policies for carbon offset allowances from afforestation can result in both a substantial land conversion between the two sectors and an intensification of agricultural production, with spillovers to other regions. Moreover, regional allowance of carbon offsets from afforestation activities can result in either net gains or net losses in sequestered carbon relative to a nationally-consistent policies, depending on regional characteristics as well as the policy period considered. In addition, we find that a permanence value reduction for afforestation leads to greater rates of future harvest of newly afforested stands in the U.S. South while increasing the area devoted to grassland pasture and cropland.
Estimating the Cost of Supplying Greenhouse Gas Offsets with Continuous Conservation Tillage
Silvia Secchi, Southern Illinois University, and Lyubov Kurkalova, North Carolina A&T University
Abstract: The study estimates the costs of adoption of continuous no till practices using simulations for the state of Iowa, a prime agricultural area. Because carbon is lost to the atmosphere once tillage occurs, and to reduce transaction costs, tillage offset contracts would have to last for several years. In our case, farmers choose whether to enter into 6 year continuous no till contracts on the basis of their soil productivity and expected net returns from alternative sets of management and crop choices. The choice of crops and tillage are discrete while fertilizer use is optimized using yield response functions on the basis of different sets of input and output prices and soil productivity. In order to capture the uncertainty and variability in the costs and prices that farmers face, the cost of labor and the price of corn and soybeans are randomized. We find that the relative price of corn heavily influences social costs, because continuous corn rotations typically are associated with higher tillage intensity. We also find that higher energy prices reduce the costs of the offsets, because they make low till and soybeans more competitive.
Estimation of Discrete Choice Models with Aggregate Data: An Application to the Adoption of Conservation Tillage
Tara Wade and Luba Kurlakova, both of North Carolina A&T University, and Silvia Secchi, Southern Illinois University
Abstract: The study estimates the costs of adoption of conservation tillage by using the method of empirical estimation of a logit model that combines the full information on the attributes of farmers and county-aggregated measures of farmers' choices. The methodology treats the aggregated data as an expected value—the area-weighted group average of individual probabilities of choosing conservation tillage-subject to a measurement error. Using the 2002 and 2004 county-average conservation tillage choice data, we estimate field-level costs of the adoption of conservation tillage and predict that the sample average subsidy payment required to entice farmers to use conservation tillage in these years would have been $13 per acre. We find that the subsidies vary significantly not only by crop but also by rotation: the analysis finds the sample-average subsidies of $42 per acre for corn after corn, $14 per acre for corn after soybeans, and $3 per acre for soybean fields. Additionally, the results indicate that conservation tillage adoption in Iowa is significantly affected by varying weather and soil characteristics.
Session III: Adoption & Additionality
Additionality and the Adoption of Farm Conservation Practices
Richard Woodward, Texas A&M University; Mariano Mezzatesta, FERC; and David A. Newburn, University of Maryland
Abstract: We use propensity score matching to estimate additionality from enrollment in federal cost-share programs for six practices. We analyze farmer adoption decisions based on farmer survey data in Ohio. We develop a new methodological approach to decompose the average treatment effect on the treated according to relative contributions of voluntary adopters and new adopters. Our results indicate that cost-share programs achieve positive levels of additionality for each practice. But percent additionality varies dramatically between practices. Specifically, percent additionality is highest for hayfield establishment (93.3%), cover crops (90.6%), and filter strips (88.9%), while it is lowest for conservation tillage (19.3%).
Additionality in Conservation Programs
Roger Claassen and Eric Duquette, both with ERS
Abstract: Payment programs that incentivize conservation practices produce additional environmental gains only if farmers receiving payments adopt practices that they would not have adopted without the payment. For some conservation practices, the “additionality” of payments may be low if programs do not differentiate between farmers who would only adopt with a payment and those farmers that may find adoption of the practice profitably even without a payment. We use a Propensity Score Matching method to estimate unobserved counterfactual adoption behavior in a nationwide survey of farmers and calculate the level of additionality for payments that target nutrient management, conservation tillage, soil conservation structures, and buffer practices. We find that roughly 80 percent of farm receiving payments for soil conservation structures and buffer practices would not have adopted these practices without the payment. For conservation tillage, we find that just over 50 percent of the farmers who received payments would not have adopted conservation tillage in the absence of the payment. We also find that nutrient management payments significantly reduced fall application of nitrogen in corn.
Luncheon Speaker: Bill Hohenstein, Director, Climate Change Program Office, USDA
Session IV: Experiments on Ag Participation in GHG Markets
Time Preference and Technology Adoption: A Single-Choice Experiment with U.S. Farmers
Nate Higgins and Eric Duquette, both of ERS, and John Horowitz, U.S. Department of the Treasury
Abstract: We elicit time-discounting behavior from U.S. farmers that are broadly known to be either late or early adopters of farming best management practices. Using a single-choice experiment, we estimate the mean discount rate for each farmer group and find that late adopters have a mean discount rate that is thirteen percentage points higher than the mean rate of early adopters. We argue through simulations that this difference is likely due to differences in time preference rather than risk aversion. There are policy implications to this finding. Government programs that aim to increase the flow of environmental services from agriculture, such as the Conservation Reserve Program, the Conservation Reserve Enhancement Program, and the Wetland Reserve Program, could potentially increase enrollment or reduce costs by changing the timing of payments to better reflect farmer preferences.
An Experimental Analysis of Market Based Agricultural Greenhouse Gas Instruments: Alternative Market Designs, Monitoring and Enforcement
Brian Scott, Washington College
Abstract: One of the hurdles of agricultural activities entering into a greenhouse gasmarkets is uncertainty, best exemplified in the Market for Lemons problem. The unique wrinkle associated with this type of market is that the buyer (nor the seller) is interested in quality. The buyer of the potential lemon, in this case an uncreated carbon sequestration certificate, is only interested in satisfying the government mandated carbon sequestration policy by surrendering adequate certificates. We experimentally explore several market-based mechanisms to alleviate the problem of society receiving the lemons, where the government mandate is satisfied with little real sequestration taking place. In our experiments sellers attempt to create GHG certificates. The actual number of permits created may be any number up to the amount attempted, with each possibility having an equal chance of happening. The sellers then sell GHG certificates with their player number identity tied to each offer, while buyers are required to purchase certificates, but have an outside option of buying an unlimited number of certificates at a set price. The buyer is monitored for quality certificates 40% of the time. If the buyer is found to have a "lemon" they must purchase a replacement certificate from the outside option. In the control condition the sellers are unable to signal quality. In the treatment condition there are two simultaneous double auctions, one where certificate quality is known to pass monitoring for quality, and one where signaling is not allowed. Sellers incur a cost to sell in the known quality market. Allowing sellers to signal quality increases the total number of certificates sold by 5%, increases the amount of actual GHG sequestration, and decreases the number of lemons sold by 10%. Furthermore in the treatment condition prices for certificates of known quality were higher than prices in the control condition, and prices for certificates of unknown quality in the treatment 1 condition were lower than in the control condition. In the treatment 1 condition 70% of the certificates sold in the unknown quality market were lemons. This indicates that there was a fairly obvious market for created certificates, and a market for uncreated certificates.
Why do farmers adopt conservation tillage: An experimental investigation of framing effects
Ben Gramig and Leigh Raymond, both of Purdue University
Abstract: In this paper, we investigate framing effects in a new context: farmer decision-making about conservation tillage practices. Our primary hypotheses are: (1) frames (i.e., different arguments about or conceptions of an issue) portraying conservation tillage as “profitable” will generate more interest in the tillage technique among farmers than a control frame presenting only basic information; (2) frames discussing potential payments for “environmental benefits” will generate more positive attitudes than frames discussing payment for “storing carbon” to limit climate change; and (3) framing effects will vary based on subjects’ prior beliefs and experiences. We test these hypotheses using a survey-based experiment administered to a national sample of row-crop farmers. Contrary to our expectations, the profit frame and both payment frames had no effect on farmers’ interest in conservation tillage across our entire sample. Consistent with our third hypothesis, however, we found a negative framing effect for the profit frame on “non-adopters” who reported no use of no-till in the past two years. These results support our argument regarding the importance of prior beliefs in reactions to frames. They also suggest the possibility of modest financial payments “crowding out” intrinsic motivations for contributions to public goods such as soil conservation. From a policy perspective these findings also suggest the relative inefficacy of offers of modest conservation payments or profitability frames in promoting no-till farming, especially among non-adopters, and the need to find alternative frames that avoid reinforcing an argument that non-adopters appear to have already considered and rejected.
Session V: Experiments on Ag Participation in Environmental/Conservation Markets or Programs
Auctions Vs. Fixed-Prices: Experimental Evidence
Brian Roe, Ohio State University, and Nate Higgins, ERS
Abstract: ERS has made a commitment to the use of experimental methods in its economic research. To facilitate the use of experiments, both by cooperators and by USDA researchers, ERS has recently obtained a generic clearance to conduct economics experiments from OMB. Among the first studies ERS plans to conduct under its generic clearance is a study of procurement mechanisms. The primary purpose of the study is to compare two mechanisms that might be used to procure environmental services from farmers and other rural landowners: a competitive mechanism (an auction) and a posted-price offer. This presentation reviews some theoretical and practical motivation for the study and presents the experimental design.
Payment for Environmental Services with Spatial Externalities
Jacob Fooks, University of Delaware
Abstract: Conserving contiguous areas often enhances environmental benefits. However, most conservation efforts are voluntary, incentive-based, and do not reward landowners for contiguity. Thus, achieving optimal contiguity of conserved parcels is unlikely especially with very limited budgets where cost effectiveness is desired. Using experiments in the laboratory and in the field, this research evaluates two mechanisms in the context of reverse auctions for achieving optimal contiguity: spatial targeting and network bonuses. Results suggest that spatial targeting alone improves the aggregate environmental and social welfare outcomes while network bonuses alone result in lower outcomes. The most effective approach derives from a combination of both mechanisms. These unexpected results arise because of the interplay of complex incentives in the reverse auction, arising from information asymmetry and auction competitiveness. The bonuses encourage landowners of highly valued parcels to place offers more often; spatial targeting permits the best selection from among these highly valued parcels.
Options for Improving Conservation Programs: Evidence from Economics Experiments on the Design of Enrollment Mechanisms
Daniel Hellerstein, ERS
Abstract: USDA funds a number of voluntary programs that pay farmers and landowners to provide environmental services. Since these programs typically can not fund all interested parties, they must use some mechanism to select applicants. This largest USDA conservation program, the Conservation Reserve Program (CRP), uses a ranking mechanism combined with bid caps. The paper considers drawbacks of this approach, and whether use of more sophisticated auctions mechanisms could yield improvements in cost effectiveness. The results of several lab experiments, that test "share" and "reference price" mechanism are reported, and shown to offer noticeable improvements. However, a field experiment was also tried, but failed in a fashion indicating the need for more careful design when the controls of a lab environment are not available.