AAEA 2017

Agricultural Technology Adoption and Staple Price Risk in Kenya, Samuel S. Bird

Time: Tuesday, August 1st, 4:45 pm – 6:15 pm
Location: Streeterville

Bird, Sam.jpg Agricultural development is often proposed as an approach to reduce rural poverty in less developed countries, yet many agricultural development interventions exclude poor farmers if they are are expected to be less responsive to interventions. I study whether land poor agricultural households will respond more than wealthier households to an intervention that increases production of staple foods due to its effect on household exposure to staple price risk. The empirical setting is a randomized control trial in western Kenya in which farmers were randomly assigned to receive inorganic fertilizer and access to hybrid seeds for maize, the staple food. Control group farmers produce less maize than they consume and face price risk as buyers of maize on average. Treatment decreases exposure to price risk among land poor households on average. Policymakers underestimate the willingness of land poor households to adopt agricultural technologies when they do not account for the role of price risk in household decision-making.

Agricultural Groundwater Markets: Understanding the Potential Gains to Trade and the Role of Market Power, Ellen M. Bruno

Time: Tuesday, August 1st, 3:00 pm
Location: Ohio State

Bruno, Ellen.jpg Groundwater is becoming a more widely regulated resource, yet little is known about the potential for groundwater markets. With a theoretical and simulation model of groundwater trading calibrated with data from the Coachella Valley, California, we estimate how the gains to groundwater trade change as market conditions and market structure vary. Market power may be a defining component of groundwater markets in certain areas due to the presence of large grower-shippers or competition among a few water agencies on a shared basin. Simulation results based on linear groundwater demand curves show that gains are large, despite losses from market power.

 

 

 

Agricultural Risk, Insurance, and the Income Share Effect in the Land-Productivity Inverse Relationship, Huang Chen

Time: Tuesday, August 1st, 1:15 pm – 2:45 pm
Location: River North

Chen, Huang.jpg China is attempting to improve upon existing land entitling policy to stimulate rural land aggregation trend for absorbing rural labors to support urban development. However, the potential negative impact of land size increase on farm productivity, i.e. the Inverse Relationship (IR), is triggering an intensive debate. Given the fact that China’s agricultural insurance market has grown rapidly in recent years, the purpose of this study is to investigate whether the insurance boosts large farms’ productivity more than smaller farms’, so that the establishment of an insurance market can complement the development of land markets, and mitigate the concern of IR. To answer this question, the first part of this study analyzes the role of risk in affecting productivity. A general farm profit model is developed and the result brings an additional layer to the conventional conclusion in literature about the risk and the IR: a constant relative risk averse (CRRA) farmer still suffers from IR problem. The second part of the model shows that insurance can indeed boost productivity, and the impact of the insurance for large farms is bigger than the small farms. In the last theory section, three explanations of IR are decomposed and compared to see how their different roles in affecting productivity. The theoretical findings are econometrically tested using a large-scale filed survey data from 6 provinces in northern China. Insurance policy is employed as IV for dealing with the endogeneity of farmers' insurance purchase behavior. Results show that insurance can significantly boost productivity by 25%, and substantially mitigate the IR. Policy implications for land and insurance market developments are discussed.

Food Price Variation over the SNAP Benefit Cycle, Xinzhe H. Cheng

Time: Tuesday, August 1st, 1:15 pm – 2:45 pm
Location: Belmont

Cheng, Xinzhe.jpg This paper asks whether the monthly distribution of SNAP benefits distorts shopping behavior by analyzing the prices participants pay for food over the benefit cycle. By leveraging the random­ization of survey interview dates relative to the timing of SNAP benefit receipt in the FoodAPS dataset, I find that SNAP recipients pay more for otherwise similar food at the beginning of the benefit cycle than at the end. Recipients pay about 11.9% less per unit of food in the last week of the benefit month than in the first. To account for local retail food prices faced by participants during the survey week, I link IRI retail scanner data to FoodAPS data in order to characterize the local price distribution. I find that the prices paid by SNAP participants shift 6 to 8 percentage points lower in the price distribution from the beginning of the benefit month to the end. I further investigate the mechanisms by which SNAP participants procure lower food prices at the end of benefit month, including shopping more intensively, going to stores further from their homes, go­ing to specific types of stores, using store discounts and coupons more extensively, buying private label brands, and stockpiling storable food. These findings can guide policymakers in designing more effective food assistance programs through a better understanding of benefit recipients' food­purchasing behavior.

Scraping the Bottom of the Beer Barrel: Consumer Preferences for Localness and Responses to Brewery Acquisitions, Jarrett D. Hart

Time: Monday, July 31st, 10:00 am – 11:30 am
Location: Armitage

Hart, Jarrett.jpg This study will check for consumer preferences for localness by examining sales and ratings data, and it will test for changes in demand by investigating consumer reactions to acquisitions. I will examine the effects of localness and acquisitions on stated and revealed consumer preferences using U.S. data consisting of roughly 3 million consumer ratings over 900 million observations of weekly beer sales. Initial results provide clear evidence that consumers prefer small-scale, locally produced beer, and that the preference for localness results in a price premium. Ongoing work will focus on further robustness testing via model specification and random sampling as well as investigating regional style-specific price discrimination.

 

 

The Labor Supply of U.S. Agricultural Workers, Alexandra Hill

Time: Tuesday, August 1st, 1:15 pm – 2:45 pm
Location: Great America I

Hill, Alexandra.jpg This paper uses data from the National Agricultural Worker’s Survey to estimate the elasticities of labor supply for U.S. crop workers separately based on legal status and participation in a means-tested welfare program. The paper builds off of the previous literature by examining how legal status and welfare participation interact to affect the labor supply of workers. To overcome the issue of endogeneity between hours and wages, the paper uses changes in state minimum wages as an instrument for the agricultural wage rate so that the elasticities are estimated from exogenous shifts in wages.  Preliminary results suggest that between workers who are undocumented, citizens, and green card holders, undocumented workers have the lowest wage responsiveness, with wage elasticities near 0.1 for males, and near zero for females. The results also suggest that across all legal statuses, program participants are more wage responsive than non-participants.

 

Evaluating impacts of marine-based stimulus policies amid market imperfections in rural Indonesia, Amanda Lindsay

Time: Tuesday, August 1st, 4:45 pm – 6:15 pm
Location: Michigan Room

Lindsay, Amanda.jpg In recent years, countries all over the world have pledged commitments to managing marine ecosystems in a way that promotes both conservation and sustainable development. The Indonesian government has been a leader, introducing its Blue Economy Initiative at Rio +20 in 2012, a paradigm prioritizing marine policies that encourage the development of marine fisheries, marine conservation, and cultural appreciation. Many marine-based economic stimulus policies in developing countries are implemented without addressing underlying market failures driving overharvest and ecosystem degradation. Because market failures can distort household response to policies and market changes, the effectiveness of the policies is unclear. Utilizing a one of a kind household and local-economy data set, we measure direct and indirect effects of common marine policies on local economies and ecosystems using a bioeconomic, general equilibrium model. We develop a hybrid model of a Local Economy-Wide Impact Evaluation (LEWIE) model of a small, rural economy and a biological model of near-shore fisheries: together called a Bio-LEWIE. A LEWIE model is an applied computable general equilibrium (CGE) model that represents key components of an isolated economy, including key production activities, relationships between households, and market imperfections (Taylor and Filipski 2014). The model is constructed from econometric analysis of micro-survey data. Using the structural model, we run policy scenarios to illustrate how poor and nonpoor household groups are directly and indirectly impacted.

Indonesia is the second largest producer of marine fish in the world, with most production coming from households operating on a small-scale using artisanal fishery production methods. In addition to its role in global fisheries production, Indonesia is part of the Coral Triangle, a center of marine biodiversity, a global conservation priority. It is pertinent to consider household response to marine policies, and how responses of fishermen to these policies may affect near-shore ecosystems. Two of Indonesia’s recent marine stimulus policies include 1) provisioning of fishing boats, engine and gear to support pelagic fisheries, and 2) supporting organizations extending microcredit to small-scale fishermen. In this paper, we develop a bio-economic model of a rural Indonesian community to evaluate the ability of these marine policies to achieve both development and environmental objectives in isolation, and as a portfolio.

Modeling village general-equilibrium effects can highlight the connection between economic sectors and natural resource stocks (Manning et al. 2016, 2013, Gilliland et al. 2016). It is clear that households participating in fishing activities can be directly affected by marine stimulus policies, but the general equilibrium model allows us to capture indirect effects as well. Indirect impacts occur through economic linkages – including wage labor markets and fish output markets, and biological linkages – including changes in fisheries health.

The Bio-LEWIE developed for analysis realistically models fishery production and incorporates real-world capital constraints to illustrate the fishermen’s response and economy impacts of marine stimulus policies. First, we have chosen to model household fishermen as allocating labor and fishing capital to two interconnected fishing activities with two separate output markets: nearshore fishing and offshore fishing. Though output markets may be separate, production is linked; small-scale fishermen in developing countries, employing less selective fishing gear and strategies, may using the same fishing gear for both fish production activities. Second, we have modeled household saving and investment amidst capital market imperfection to more accurately capture the household response to stimulus policy. Local fisheries are best characterized as open access resources, with low returns to household labor and capital. Despite low returns, fishing households continue to invest time and money in fishing capital, rather than increasing allocations to alternative livelihoods. Modeling savings and investments helps highlight household production decision strategies.

While the marine stimulus policies are being implemented in locations throughout Indonesia, this study focuses on Selayar Island, the main island of the District Selayar, located just south of Sulawesi Island.

Selayar Island’s main industries include marine capture fisheries and agriculture. Selayar is the epitome of a poor, isolated economy within a country characterized by low quality trade- and transport-related infrastructure. Because households of Selayar face high transaction costs, imperfect and missing markets, a Bio-LEWIE model is well suited to demonstrate how marine policies will impact fisheries and Selayar Island’s economy. More generally, Selayar is an excellent location to develop insights into the effects of government interventions on rural coastal economies.

The model is parameterized using a unique data set collected in Fall 2016. The random survey design resulted in a sample of representative households and businesses for the Island. We carried out 487 household and 256 business surveys in 12 of the 52 villages on Selayar Island. Household surveys collected data on household demographics, production (fishing, agriculture, livestock, enterprise), purchase, food security, and finance. If households engaged in fishing activities, additional information was collected on fishing behavior and harvests. Business surveys gathered information detailing business’ use of hired labor, expenses, sales and financing. When applicable, information on the use of fish inputs was collected.

Distribution of fishing capital can help poor households who cannot otherwise increase their investments in fishing activities. Offshore fishing, a more capital-intensive activity, is viewed as an alternative livelihood to fishing over vulnerable nearshore habitat: coral reef and sea grass. Fishing boats and gear provided by the government, however, can be used in both offshore and nearshore fisheries. Without additional regulation or establishment of fishing rights, this marine stimulus policy may increase, rather than reduce, pressure on local coral reefs. Increasing the availability of microcredit to poor fishing households may increase investment in fishing capital, potentially leading to outcomes similar to the boat provisioning policy. However, it also allows fishing households to invest in other production activities or human capital, reducing the burden on local fisheries. The specific outcomes these policies will have on the households and fisheries depend on the characterizations of the economy, underscoring the importance of developing a robust model to realistically capture Selayar’s features. The Bio-LEWIE model is uniquely capable of determining how fish stocks and households’ allocation of labor and capital in and beyond the fishing sector will be impacted by marine policies. Analyses highlight the policies’ impacts on fish stocks and welfare of poor and non-poor households on Selayar Island over a ten-year time horizon. We also highlight the differences when policies are implemented for short vs long periods, as many policies in developing countries (e.g., microcredit) are implemented for brief periods.

This study offers new perspectives on the biological and social impacts of marine-based stimulus policies implemented in rural coastal communities of Indonesia. The steps taken to characterize fishing production and household capital allow us to depict household response to policies. We demonstrate to what extent short-term support for small-scale producers can achieve poverty alleviation and conservation objectives. Estimates of local economy-wide effects can help us re-examine blue policy initiatives being employed in developing countries around the world.

Earthbound Labor and Incomplete Exit from Farming in China: Multiple Distortions and Nonseparable Decisions, Meilin Ma

Time: Tuesday, August 1st, 1:15 pm – 2:45 pm
Location: Great America I, 6th Floor

Ma, Meilin.jpg Smallholder farming remains the predominant production mode in China, despite continuing urbanization and economic transformations. Previous research has found land markets to be relatively inactive, but not counted for seasonal and partial features of land use. Based upon recent survey data from Southwest China, I adopt an innovative method to reveal frequent and incomplete exit from farming. I propose an augmented household model to explain the preference for incomplete exit under multiple distorted incentives and nonseparable factor decisions. The use-based value of contract land, due to its safety-net and asset functions, induces households to retain labor with comparative advantages in farming on the farm and causes inefficient factor allocations at the household and the sector levels. A wedge exists between productivity of full-time and part-time farming labor and the corresponding nonfarm wage rate, which cannot be eliminated by providing perfect factor markets. Key market and institutional determinants to land and labor decisions are identified at the household and individual levels.  Using an instrumental variable, I show empirically how a household adjusts labor allocations as the cultivation size changes. Regarding policy implications, I stress that policies enhancing welfare of peasant households or promoting labor-saving technologies can impede farmland consolidation.

The Economic Viability of Suppressive Crop Rotations for the Control of Verticillium Wilt in Organic Strawberry Production in California, Aleksandr Michuda

Time: Monday, July 31st, 10:00 am – 11:30 am
Location: Iowa

Michuda, Aleksandr.jpg Organic strawberry production totaled 94 million dollars in farm level sales in 2012 (Klonsky & Healy, 2013). However, its potential for expansion is limited by soil borne diseases and the level of nitrogen in the soil. Verticillium wilt, for example, has historically been the major disease that affected strawberries, and fusarium wilt and charcoal rot have emerged as important diseases in recent years (Shennan, 2016). These diseases are particularly challenging for organic growers because they cannot apply fumigants to disinfect the soil prior to planting. Our analysis is part of a multi-year fixed trial that tested a set of suppressive organic crop rotations coupled with either ASD (anaerobic soil disinfestation) or mustard seed cake (MSM). Each rotation was evaluated based on its capacity to suppress disease and its economic viability.

The economic viability of suppressive crop rotations is of interest for conventional as well as organic production. Classic rotations such as processing tomatoes and wheat in California and corn and soybeans in the Midwest are part of many growers’ integrated pest management programs. Additionally, chemical pesticide use is increasingly regulated, particularly in California. As the number and variety of regulatory constraints increase, efficacious non-chemical pest management techniques are becoming an increasingly important part of growers’ pest management toolkit. The rotations we examine could be adopted by conventional growers to reduce their use of pre-plant soil fumigation, a highly regulated type of pesticide application.

Suppressive crop rotations are an important application of dynamics to agricultural production. There are dynamic tradeoffs at play, such as the decision of whether to grow a cash crop for immediate revenues versus planting a low revenue non-host crop which suppresses pathogens, pests, and disease and, thus, can improve the yields of a high revenue crop in the future. Evidence from our analysis shows that there are significant benefits to investing in a non-host crop such as broccoli, on future strawberry yields, and hence, revenues. It also shows the benefit of cash crops when the cost of suppressing disease becomes even higher due to low non-host crop prices.

Ex-Ante Investment Under Risk: Drought Tolerant Maize Adoption in East Africa, Laura Paul

Time: Tuesday, August 1st, 11:30 am – 1:15 pm
Location: Chicago Ballroom D-H

Paul, Laura.jpg Adoption of improved seed has been slow in Eastern and Southern Africa, where only 22% of maize farmers are using any type of improved seed, despite their introduction decades ago (LSMS, 2013). The lag in adoption could stem from increased risks, higher costs, or information failures. This paper conducts an ex ante analysis to inform why improved seed adoption is low.

If riskiness has indeed slowed adoption, then developing traits in seeds which reduce risk might increase adoption rates of improved seed. The International Maize and Wheat Improvement Center (CIMMYT) has developed over one hundred new maize varieties through the Drought Tolerant Maize for Africa (DTMA) program in the past two decades (CIMMYT, 2012). These new hybrids and open pollinated varieties of maize can withstand moderate drought with only small yield penalties, in addition to providing yield increases over traditional varieties in normal rain conditions (Fisher et al., 2015). These characteristics are advertised as particularly advantageous to the farmers in the mid-altitude areas of Tanzania, which have experienced drought and subsequent crop failure in one in five years; nonetheless, only around one in ten farmers has adopted drought tolerant maize (Fisher et al., 2015).  Maize yields are highly responsive to solar radiation and the temporal distribution of rainfall within a season. Drought tolerant varieties developed under the DTMA program have a particular advantage of being able to withstand flowering stage drought stress, in addition to being hybrids, which, in general, have yield gains over traditional varieties. However, the advantage of drought tolerant maize over other varieties is not always increasing with drought pressure– these varieties cannot survive an extreme drought event. The DTMA program promotes their maize varieties as less risky, and therefore their introduction should accelerate adoption. However, the risk of crop failure under the most severe drought outcomes could reduce adoption of drought tolerant maize among the most risk averse farmers (Lybbert and Bell, 2010).

Would a farmer adopt drought tolerant maize with complete information of the production function and a known distribution of rainfall outcomes, given di↵erent risk preferences?  How does the distribution of drought outcomes impact adoption? Tanzania’s largely rural population depends on rain-fed staple production for their welfare and food security. Maize accounts for almost half of the calories and protein consumed in Eastern and Southern Africa (Shiferaw et al., 2011); in Tanzania, maize is the most common crop grown by households (Tanzania Bureau of Statistics, 2012). Understanding the economic consequences of drought and the potential of improved seed is critical for policy development and implementation.  This paper has its foundation in the literature on the economic benefits of new seed technology, household response to climate risk, specifically drought, and household adoption of new technology. This paper contributes a new analysis of CIMMYT On-Farm trial data which can give a greater understanding of the drought tolerant maize advantages. Further, this paper incorporates risk preferences to provide insights into seed technology adoption.

The Dynamic Migration Game: A Structural Econometric Model and Application to Rural Mexico, Ruben Irvin Rojas Valdes

Time: Tuesday, August 1st, 1:15 pm – 2:45 pm
Location: Great America I, 6th Floor

Rojas, Irvin.jpg The migration decisions of households in a village can be thought of as a dynamic game in which each household optimally decides how to allocate its members across distinct activities, taking into account dynamic considerations about the future and strategic considerations about what neighbors in the village are doing. We develop and estimate a structural econometric model of this dynamic migration game. The structural econometric model enables us to examine how natural factors, economic factors, institutions, government policies, and strategic interactions a ect the migration decisions of households in rural Mexico. We use this model to simulate the e ects of counterfactual policy scenarios, including those regarding wages, schooling, crime rates at the border, precipitation, and government policy, on migration decisions and welfare.

 

 

The Effect of Immigration on Native Workers: Evidence from the US Construction Sector, Zachariah Rutledge

Time: Tuesday, August 1st, 1:15 pm – 2:45 pm
Location: Great America I

Rutledge, Zachariah.jpg There are over 6 million construction workers in the United States (US). Our data indicates that 16 percent of these workers are immigrants. In recent decades, the US has seen a large increase in the number of immigrants from Mexico and other Central and South American countries, raising concerns about the employment opportunities of natives. Many immigrant workers are considered low-skilled and find work in the construction sector. Using US Census and American Community Survey data between 1990 and 2011, we estimate the recent effects of immigration on the labor market outcomes of native-born construction workers in the US.

The main challenge with estimating such effects is that increased immigration is plausibly correlated with unobservable demand pull factors which in turn influence native workers' employment and income levels. This causes upward biases in estimates of the effects of immigration on the income of natives, which can give the perception that immigration is actually beneficial (or at least not harmful) to native workers. To address this issue, our identification strategy uses a 2SLS fixed-effects panel data model with so-called \imperfect instrument" to establish an upper bound for the effect of immigration on the full-time employment and annual income of native workers (Nevo & Rosen, 2012). An imperfect instrument is correlated with the endogenous regressor and is "less" correlated with the error term than the regressor itself. In our application, our regressor is a variable that measures the proportion of immigrants working in the construction sector in each Metropolitan Statistical Area (MSA) in each year. This is a commonly used measure of supply shift in the literature. Our imperfect instrumental variable is a variable that measures the proportion of immigrants working in all occupations in each MSA in each year. This instrument is \imperfect" because it might still be correlated with demand pull factors in the construction industry (to the extent that demand pull factors are correlated across sectors of the economy), yet it is less correlated with demand pull factors in the construction industry than the proportion of immigrant workers in that industry. This strategy allows us to establish an upper bound for the negative effects of immigration on native workers' employment conditions.

Our study is unique for several reasons. First of all, we investigate the effects of immigration on natives' incomes in the sector of the economy that is likely to have been impacted the most by immigration. Second, in contrast to previous results, we find that immigration negatively affects the annual income of native workers. Third, our analysis suggests that this negative effect is channeled through less time worked by natives rather than through lower wages per unit of work time. That is, native workers are induced to reduce actual labor supplied, perhaps because fewer of their bids get accepted, but their wage rate per unit of labor does not seem to be affected. Fourth, we propose a policy-relevant application of the imperfect instrument theory recently proposed by Nevo & Rosen (2012).

Impacts of the U.S. Ethanol Boom on Corn Transportation Markets, Heidi Schweizer

Time: Monday, July 31st, 4:15 pm – 5:30 pm
Location: Ballroom D-H, 5th Floor

Heidi Schweizer The ethanol boom began in the early 2000s in anticipation of major changes in U.S. energy policy. This project focuses on the effects of the 2005 Energy Policy Act and the 2007 Energy Independence and Security Act. These legislations include the Renewable Fuel Standard program, which requires a minimum volume of biofuels to be used in transportation fuel. This project is about understanding how energy policy changes geographic pricing patterns and understanding how these changes influence agricultural freight markets. Northwestern Iowa is identified as a good area to study because two crops dominate farm acreage, there is unlikely to be rail market power, it is relatively far from the Mississippi river, and there is spatial variation in ethanol plant locations over time. I have assembled much of the geospatial data needed to consider whether the ethanol boom has caused a persistent change in the relative basis of corn and soybeans. These data include production, prices, and ethanol plant locations. Particularly, I have obtained confidential Waybill Sample data from 1989 to 2015. Waybill Sample data is a stratified sample of rail receipts and the confidential set includes unmasked rail revenues and precise origin and destination information. It is too early to draw firm conclusions but my sense of these data is that the ethanol boom is most evident in the cross-price elasticity of corn and soybean transportation demand.

Response of North Dakota Soybean Flows to the Panama Canal Expansion: A Positive Mathematical Programming Model, Heidi Schweizer

The Panama Canal Expansion opened in June 2016 and approximately doubled the capacity of the Panama Canal. The expansion will likely have direct and indirect effects on grain and oilseed transportation markets because U.S. grain and oilseed exports dominate the East to West dry bulk traffic through the canal. This project examines how the Panama Canal Expansion may change soybean flows between North Dakota and U.S. export locations. There are four main steps. First, an intertemporal partial equilibrium model of soybean flows from North Dakota is specified. Second, this model is calibrated using positive mathematical programming techniques. Third, distributions of soybean production, prices, and transportation costs are empirically determined. Fourth, the calibrated programming model accounting for the Panama Canal Expansion and is solved with simulated parameters to find soybean shipments from North Dakota production regions throughout the marketing year. Sensitivity analysis shows that storage and fuel costs are important factors in determining seasonal shipping patterns. Analysis of the effects of Panama Canal Expansion is currently underway to determine how it may change destinations of North Dakota soybean shipments.

Are Farmers Good Neighbors? Self-Regulation of Pesticide Applications near Schools and Daycares in California, Tor Tolhurst

Time: Tuesday, August 1st, 10:00 am – 11:30 am
Location: Wisconsin

Tolhurst, Tor.jpg We test whether California agricultural producers self-regulate their pesticide applications near public schools and licensed daycares using administrative data on all field-level applications with grower identifiers from 2009 to 2014. Agricultural producers can self-regulate their pesticide applications near schools and daycares by voluntarily conducting their applications on evenings and weekends, thereby minimizing the potential harm of drift. As a policy option, self-regulation is almost always within the regulator's choice set; however, it is typically difficult to measure the extent of self-regulation because data may not be available at the relevant observational level and strategic considerations may confound inference about decision-making. Our setting and administrative data allows us to overcome these challenges. The data suggest fields with more schools and daycares in their proximity are less likely to be sprayed between 6am and 6pm on weekdays; however, the magnitude of the effect is very small. Imposing the admittedly strong assumption of all else equal, our results suggest farmers would reduce their applications on school days from 56.0% to 53.2%. Whether or not the extent of this self-regulation is sufficient, or publicly optimal, remains an open question.