@article{hendricks2022distortions,
  title={The effects of agricultural policy on supply and productivity: Evidence from differential changes in distortions},
  author={Hendricks, Nathan P. and Smith, Aaron and Villoria, Nelson B. and Stigler, Matthew},
  journal={Agricultural Economics},
  year={2022},
  volume={},
  number={},
  pages={},
	keywords={agriculture},
	abstract={Incentives in agriculture are highly distorted. It has long been argued that these distortions were a key explanation for differences in supply and productivity across countries, but the empirical evidence is limited. We revisit this issue using data on policy distortions across 63 countries for the period 1961–2011. We estimate the effects of differential changes in agricultural distortions across countries on supply and productivity. We highlight concerns in our analysis and previous work about endogeneity that biases the estimated effect downward—countries that lose comparative advantage are likely to increase support for agriculture. We address these concerns by including country and region-time fixed effects, along with a rich set of controls. Overall, we find evidence that enhanced incentives through policy changes can increase the rate of production growth, with about half of the increase due to productivity increases. This result is strongest in Sub-Saharan Africa where anti-agricultural policies on exports were reduced and in Europe where pro-agricultural policies on imports were reduced, driven largely by external pressure. Endogeneity appears to be strongest in Asia where countries have followed the typical pattern of raising support for agriculture during industrialization due to a rising farm-urban income gap.},
	url={https://files.asmith.ucdavis.edu/2022_AgEcS_HSVS_distortions.pdf}
}

Incentives in agriculture are highly distorted. It has long been argued that these distortions were a key explanation for differences in supply and productivity across countries, but the empirical evidence is limited. We revisit this issue using data on policy distortions across 63 countries for the period 1961–2011. We estimate the effects of differential changes in agricultural distortions across countries on supply and productivity. We highlight concerns in our analysis and previous work about endogeneity that biases the estimated effect downward—countries that lose comparative advantage are likely to increase support for agriculture. We address these concerns by including country and region-time fixed effects, along with a rich set of controls. Overall, we find evidence that enhanced incentives through policy changes can increase the rate of production growth, with about half of the increase due to productivity increases. This result is strongest in Sub-Saharan Africa where anti-agricultural policies on exports were reduced and in Europe where pro-agricultural policies on imports were reduced, driven largely by external pressure. Endogeneity appears to be strongest in Asia where countries have followed the typical pattern of raising support for agriculture during industrialization due to a rising farm-urban income gap.

@article{lark2022rfs,
  title={Environmental Outcomes of the US Renewable Fuel Standard},
  author={Lark, Tyler J. and Hendricks, Nathan P. and Smith, Aaron and Pates, Nicholas and Spawn-Lee, Seth A. and Bougie, Matthew and Booth, Eric G. and Kucharik, Christopher J. and Gibbs,Holly K.},
  journal={Proceedings of the National Academy of Sciences},
  year={2022},
  volume={119},
  number={9},
  pages={1-8},
	keywords={agriculture},
	abstract={The Renewable Fuel Standard (RFS) specifies the use of biofuels in the United States and thereby guides nearly half of all global biofuel production, yet outcomes of this keystone climate and environmental regulation remain unclear. Here we combine econometric analyses, land use observations, and biophysical models to estimate the realized effects of the RFS in aggregate and down to the scale of individual agricultural fields across the United States. We find that the RFS increased corn prices by 30\% and the prices of other crops by 20\%, which, in turn, expanded US corn cultivation by 2.8 Mha (8.7\%) and total cropland by 2.1 Mha (2.4\%) in the years following policy enactment (2008 to 2016). These changes increased annual nationwide fertilizer use by 3 to 8\%, increased water quality degradants by 3 to 5\%, and caused enough domestic land use change emissions such that the carbon intensity of corn ethanol produced under the RFS is no less than gasoline and likely at least 24\% higher. These tradeoffs must be weighed alongside the benefits of biofuels as decision-makers consider the future of renewable energy policies and the potential for fuels like corn ethanol to meet climate mitigation goals.},
	url={https://www.pnas.org/content/119/9/e2101084119}
}

The Renewable Fuel Standard (RFS) specifies the use of biofuels in the United States and thereby guides nearly half of all global biofuel production, yet outcomes of this keystone climate and environmental regulation remain unclear. Here we combine econometric analyses, land use observations, and biophysical models to estimate the realized effects of the RFS in aggregate and down to the scale of individual agricultural fields across the United States. We find that the RFS increased corn prices by 30% and the prices of other crops by 20%, which, in turn, expanded US corn cultivation by 2.8 Mha (8.7%) and total cropland by 2.1 Mha (2.4%) in the years following policy enactment (2008 to 2016). These changes increased annual nationwide fertilizer use by 3 to 8%, increased water quality degradants by 3 to 5%, and caused enough domestic land use change emissions such that the carbon intensity of corn ethanol produced under the RFS is no less than gasoline and likely at least 24% higher. These tradeoffs must be weighed alongside the benefits of biofuels as decision-makers consider the future of renewable energy policies and the potential for fuels like corn ethanol to meet climate mitigation goals.

@article{adjemian2021estimating,
  title={Estimating the Market Effect of a Trade War: The Case of Soybean Tariffs},
  author={Adjemian, Michael K  and Smith, Aaron and He, Wendi},
  journal={Food Policy},
  year={2021},
  volume={105},
  number={},
  pages={},
	abstract={In 2018, China retaliated to U.S. trade actions by levying a 25\% retaliatory tariff on U.S. soybean exports. That tariff shifted market preferences so that Chinese buyers—who make up a substantial share of total world consumption—favored Brazilian soybeans. We use the relative price of a substitute (RPS) method to estimate that the resulting trade disruption effectively drove a wedge into the world soybean market, lowering U.S. prices at Gulf export locations by 0.74 dollar/bu on average for about five months, and increasing Brazilian prices by about 0.97 dollar/bu, compared to what would have been observed without the tariff in place. By the end of that period, world markets adjusted and the soybean prices in both countries returned to the ex-ante state of near parity, even if U.S. export volume did not recover until the end of the following marketing year. Our price impact estimate is substantially lower than subsequent U.S. government “trade aid” payments to American soybean producers: although actual payments to producers varied based on county-level differences, USDA’s nominal calculation of the commodity-specific payment rate for soybeans under MFP summed to 3.70 dollars for two bushels produced over the course of two years. We project that USDA's near-8.5 billion dollars in trade aid to U.S. soybean producers exceeded the tariff damage by about 5.4 billion dollars. These difference could be attributed to USDA’s broader definition of “economic injury”, beyond the short-run price impacts we estimate.},
	keywords={agriculture},
	url={https://asmith.ucdavis.edu/news/trade-war}
}

In 2018, China retaliated to U.S. trade actions by levying a 25% retaliatory tariff on U.S. soybean exports. That tariff shifted market preferences so that Chinese buyers—who make up a substantial share of total world consumption—favored Brazilian soybeans. We use the relative price of a substitute (RPS) method to estimate that the resulting trade disruption effectively drove a wedge into the world soybean market, lowering U.S. prices at Gulf export locations by 0.74 dollar/bu on average for about five months, and increasing Brazilian prices by about 0.97 dollar/bu, compared to what would have been observed without the tariff in place. By the end of that period, world markets adjusted and the soybean prices in both countries returned to the ex-ante state of near parity, even if U.S. export volume did not recover until the end of the following marketing year. Our price impact estimate is substantially lower than subsequent U.S. government “trade aid” payments to American soybean producers: although actual payments to producers varied based on county-level differences, USDA’s nominal calculation of the commodity-specific payment rate for soybeans under MFP summed to 3.70 dollars for two bushels produced over the course of two years. We project that USDA's near-8.5 billion dollars in trade aid to U.S. soybean producers exceeded the tariff damage by about 5.4 billion dollars. These difference could be attributed to USDA’s broader definition of “economic injury”, beyond the short-run price impacts we estimate.

@article{smith2004computer,
  title={Computer and Internet use by Great Plains Farmers},
  author={Smith, Aaron and Goe, W Richard and Kenney, Martin and Paul, Catherine J Morrison},
  journal={Journal of Agricultural and Resource Economics},
  pages={481--500},
  year={2004},
	url={https://files.asmith.ucdavis.edu/2004_JARE_SGKM_computer.pdf},
	abstract={This study uses data from a 2001 survey of Great Plains farmers to explore the adoption, usage patterns, and perceived benefits of computers and the Internet. Adoption results suggest that exposure to the technology through college, outside employment, friends, and family is ultimately more influential than farmer age and farm size. Notably, about half of those who use the Internet for farm-related business report zero economic benefits from it. Whether a farmer perceives that the Internet generates economic benefits depends primarily on how long the farmer has used the Internet for farm business and for what purposes.},
	keywords={agriculture},
  publisher={Western Agricultural Economics Association}
}

This study uses data from a 2001 survey of Great Plains farmers to explore the adoption, usage patterns, and perceived benefits of computers and the Internet. Adoption results suggest that exposure to the technology through college, outside employment, friends, and family is ultimately more influential than farmer age and farm size. Notably, about half of those who use the Internet for farm-related business report zero economic benefits from it. Whether a farmer perceives that the Internet generates economic benefits depends primarily on how long the farmer has used the Internet for farm business and for what purposes.

@article{carter2007estimating,
  title={Estimating the Market Effect of a Food Scare: The Case of Genetically Modified Starlink Corn},
  author={Carter, Colin A and Smith, Aaron},
  journal={The Review of Economics and Statistics},
  volume={89},
  number={3},
  pages={522--533},
  year={2007},
	url={https://files.asmith.ucdavis.edu/2007_REStat_CS_StarLink.pdf},
	keywords={agriculture},
	abstract={In 2000, a genetically modified corn variety called StarLink that was not approved for human consumption was discovered in the food-corn supply. To estimate the price impact of this event on the U.S. corn market, we develop the relative price of a substitute method. This method applies not only to the StarLink event but also to rare events in other markets. We find that the contamination led to a 6.8\% discount in corn prices and that the suppression of prices lasted for at least a year.},
	addendum={\textbf{Winner of Quality of Research Discovery Award, AAEA, 2008.}},
  publisher={The MIT Press}
}

In 2000, a genetically modified corn variety called StarLink that was not approved for human consumption was discovered in the food-corn supply. To estimate the price impact of this event on the U.S. corn market, we develop the relative price of a substitute method. This method applies not only to the StarLink event but also to rare events in other markets. We find that the contamination led to a 6.8% discount in corn prices and that the suppression of prices lasted for at least a year.

@article{balagtas2007effects,
  title={Effects of Milk Marketing Order Regulation on the Share of Fluid-Grade Milk in the United States},
  author={Balagtas, Joseph V and Smith, Aaron and Sumner, Daniel A},
  journal={American Journal of Agricultural Economics},
  volume={89},
  number={4},
  pages={839--851},
  year={2007},
	url={https://files.asmith.ucdavis.edu/2007_AJAE_BSS_milk.pdf},
	keywords={agriculture},
	abstract={The share of raw milk meeting fluid quality (Grade A) standards in the United States rose steadily through the latter half of the twentieth century, but a shrinking portion of that was used in fluid products. Grade A milk exceeds the quality standards for the manufactured products for which it has been increasingly used. We present an econometric model that exploits regional and temporal variation in policy implementation to identify the effect of marketing orders on the Grade A share of milk. Results support the hypothesis that marketing orders significantly encouraged the growth in the Grade A share of milk.},
  publisher={Oxford University Press}
}

The share of raw milk meeting fluid quality (Grade A) standards in the United States rose steadily through the latter half of the twentieth century, but a shrinking portion of that was used in fluid products. Grade A milk exceeds the quality standards for the manufactured products for which it has been increasingly used. We present an econometric model that exploits regional and temporal variation in policy implementation to identify the effect of marketing orders on the Grade A share of milk. Results support the hypothesis that marketing orders significantly encouraged the growth in the Grade A share of milk.

@article{adjemian2012using,
  title={Using USDA Forecasts to Estimate the Price Flexibility of Demand for Agricultural Commodities},
  author={Adjemian, Michael K and Smith, Aaron},
  journal={American Journal of Agricultural Economics},
  volume={94},
  number={4},
  pages={978--995},
  year={2012},
	url={https://files.asmith.ucdavis.edu/2012_AJAE_AS_flexibility.pdf},
	keywords={agriculture},
	abstract={We estimate the general equilibrium price flexibility of demand for corn and soybeans using monthly changes in expected supply published by the USDA. Our estimates reflect the demand response to a one‐year supply shock and thus correspond to the inverse demand elasticity. We derive the conditions under which our estimates are consistent, and we show how demand flexibility varies by season, inventory, time horizon, and demand composition. At average inventory and without accounting for corn‐ethanol use, we obtain price flexibility estimates of −1.35 and −1.03 for corn and soybeans, respectively. Current corn‐ethanol production levels are associated with much larger absolute flexibilities for both commodities.},
  publisher={Oxford University Press}
}

We estimate the general equilibrium price flexibility of demand for corn and soybeans using monthly changes in expected supply published by the USDA. Our estimates reflect the demand response to a one‐year supply shock and thus correspond to the inverse demand elasticity. We derive the conditions under which our estimates are consistent, and we show how demand flexibility varies by season, inventory, time horizon, and demand composition. At average inventory and without accounting for corn‐ethanol use, we obtain price flexibility estimates of −1.35 and −1.03 for corn and soybeans, respectively. Current corn‐ethanol production levels are associated with much larger absolute flexibilities for both commodities.

@article{hendricks2014crop,
  title={Crop Supply Dynamics and the Illusion of Partial Adjustment},
  author={Hendricks, Nathan P and Smith, Aaron and Sumner, Daniel A},
  journal={American Journal of Agricultural Economics},
  volume={96},
  number={5},
  pages={1469--1491},
  year={2014},
	url={https://files.asmith.ucdavis.edu/2014_AJAE_HSS_illusion.pdf},
	keywords={agriculture},
	abstract={We use field‐level data to estimate the response of corn and soybean acreage to price shocks. Our sample contains more than 8 million observations derived from satellite imagery and includes every cultivated field in Iowa, Illinois, and Indiana. We estimate that aggregate crop acreage responds more to price shocks in the short run than in the long run, and we show theoretically how the benefits of crop rotation generate this response pattern. In essence, farmers who change crops due to a price shock have an incentive to switch back to the previous crop to capture the benefits of crop rotation. Our result contradicts the long‐held belief that agricultural supply responds gradually to price shocks through partial adjustment. We would not have obtained this result had we used county‐level panel data. Standard econometric methods applied to county‐level data produce estimates consistent with partial adjustment. We show that this apparent partial adjustment is illusory, and we demonstrate how it arises from the fact that fields in the same county are more similar to each other than to fields in other counties. This result underscores the importance of using models with appropriate micro‐foundations and cautions against inferring micro‐level rigidities from inertia in aggregate panel data. Our preferred estimate of the own‐price long‐run elasticity of corn acreage is 0.29, and the cross‐price elasticity is −0.22. The corresponding elasticities for soybean acreage are 0.26 and −0.33. Our estimated short‐run elasticities are 37\% larger than their long‐run counterparts.},
	addendum={\textbf{Winner of Outstanding AJAE Article Award, AAEA, 2015}},
  publisher={Oxford University Press}
}

We use field‐level data to estimate the response of corn and soybean acreage to price shocks. Our sample contains more than 8 million observations derived from satellite imagery and includes every cultivated field in Iowa, Illinois, and Indiana. We estimate that aggregate crop acreage responds more to price shocks in the short run than in the long run, and we show theoretically how the benefits of crop rotation generate this response pattern. In essence, farmers who change crops due to a price shock have an incentive to switch back to the previous crop to capture the benefits of crop rotation. Our result contradicts the long‐held belief that agricultural supply responds gradually to price shocks through partial adjustment. We would not have obtained this result had we used county‐level panel data. Standard econometric methods applied to county‐level data produce estimates consistent with partial adjustment. We show that this apparent partial adjustment is illusory, and we demonstrate how it arises from the fact that fields in the same county are more similar to each other than to fields in other counties. This result underscores the importance of using models with appropriate micro‐foundations and cautions against inferring micro‐level rigidities from inertia in aggregate panel data. Our preferred estimate of the own‐price long‐run elasticity of corn acreage is 0.29, and the cross‐price elasticity is −0.22. The corresponding elasticities for soybean acreage are 0.26 and −0.33. Our estimated short‐run elasticities are 37% larger than their long‐run counterparts.

@article{hendricks2014environmental,
  title={The Environmental Effects of Crop Price Increases: Nitrogen Losses in the US Corn Belt},
  author={Hendricks, Nathan P and Sinnathamby, Sumathy and Douglas-Mankin, Kyle and Smith, Aaron and Sumner, Daniel A and Earnhart, Dietrich H},
  journal={Journal of Environmental Economics and Management},
  volume={68},
  number={3},
  pages={507--526},
  year={2014},
	url={https://files.asmith.ucdavis.edu/2014_JEEM_HSDSSE_water.pdf},
	abstract={High corn prices cause farmers to plant more corn on fields that were planted to corn in the previous year, rather than alternating between corn and soybeans. Cultivating corn after corn requires greater nitrogen fertilizer and some of this nitrogen flows into waterways and causes environmental damage. We estimate the effect of crop prices on nitrogen losses for most fields in Iowa, Illinois, and Indiana using crop data from satellite imagery. Spatial variation in these high-resolution estimates highlights the fact that the environmental effects of agriculture depend not only on what is grown, but also on where and in what sequence it is grown. Our results suggest that the change in corn and soybean prices due to a billion gallons of ethanol production expands the size of the hypoxic zone in the Gulf of Mexico by roughly 30 square miles on average, although there is considerable uncertainty in this estimate.},
	keywords={agriculture},
  publisher={Academic Press}
}

High corn prices cause farmers to plant more corn on fields that were planted to corn in the previous year, rather than alternating between corn and soybeans. Cultivating corn after corn requires greater nitrogen fertilizer and some of this nitrogen flows into waterways and causes environmental damage. We estimate the effect of crop prices on nitrogen losses for most fields in Iowa, Illinois, and Indiana using crop data from satellite imagery. Spatial variation in these high-resolution estimates highlights the fact that the environmental effects of agriculture depend not only on what is grown, but also on where and in what sequence it is grown. Our results suggest that the change in corn and soybean prices due to a billion gallons of ethanol production expands the size of the hypoxic zone in the Gulf of Mexico by roughly 30 square miles on average, although there is considerable uncertainty in this estimate.

@article{carter2016commodity,
  title={Commodity Storage and the Market Effects of Biofuel Policies},
  author={Carter, Colin A and Rausser, Gordon C and Smith, Aaron},
  journal={American Journal of Agricultural Economics},
  volume={99},
  number={4},
  pages={1027--1055},
  year={2017},
	url={https://files.asmith.ucdavis.edu/2017_AJAE_CRS_ethanol.pdf},
	keywords={agriculture},
	abstract={Legislation passed in 2007 by the U.S. Congress increased by about 1.3 billion bushels the net amount of corn required to be processed annually into ethanol for motor‐fuel use. We estimate that corn prices were about 30\% higher from 2006 to 2014 than they would have been without this demand increase. We develop a partially identified structural vector autoregression model. Our identification strategy is unique in the literature because it enables us to estimate the effects of transitory shocks, such as weather, separately from the effects of persistent shocks, such as the increased ethanol mandate. Moreover, by only partially identifying our model, we show how to generate robust conclusions without strong identifying assumptions.},
	addendum={\textbf{Winner of Outstanding AJAE Article Award, AAEA, 2018}},
  publisher={Oxford University Press}
}

Legislation passed in 2007 by the U.S. Congress increased by about 1.3 billion bushels the net amount of corn required to be processed annually into ethanol for motor‐fuel use. We estimate that corn prices were about 30% higher from 2006 to 2014 than they would have been without this demand increase. We develop a partially identified structural vector autoregression model. Our identification strategy is unique in the literature because it enables us to estimate the effects of transitory shocks, such as weather, separately from the effects of persistent shocks, such as the increased ethanol mandate. Moreover, by only partially identifying our model, we show how to generate robust conclusions without strong identifying assumptions.

@article{hendricks2015futures,
  title={Futures Prices in Supply Analysis: Are Instrumental Variables Necessary?},
  author={Hendricks, Nathan P and Janzen, Joseph P and Smith, Aaron},
  journal={American Journal of Agricultural Economics},
  volume={97},
  number={1},
  pages={22--39},
  year={2015},
	url={https://files.asmith.ucdavis.edu/2015_AJAE_HJS_futuressupply.pdf},
	keywords={agriculture},
	abstract={Crop yield shocks are partially predictable—high planting‐time futures prices have tended to indicate that yield would be below trend. As a result, regressions of total caloric production on futures prices produce estimates of the supply elasticity that are biased downwards by up to 75\%. Regressions of the world's growing area on futures prices have a much smaller bias of about 20\% because although yield shocks are partially predictable, this predictability has a relatively small effect on land allocation. We argue that the preferred method for estimating the crop supply elasticity is to use regressions of growing area on futures prices and to include the realized yield shock as a control variable. An alternative method for bias reduction is to use instrumental variables (IVs). We show that the marginal contribution of an IV to bias reduction is small—IVs are not necessary for futures prices in supply analysis.},
  publisher={Oxford University Press}
}

Crop yield shocks are partially predictable—high planting‐time futures prices have tended to indicate that yield would be below trend. As a result, regressions of total caloric production on futures prices produce estimates of the supply elasticity that are biased downwards by up to 75%. Regressions of the world's growing area on futures prices have a much smaller bias of about 20% because although yield shocks are partially predictable, this predictability has a relatively small effect on land allocation. We argue that the preferred method for estimating the crop supply elasticity is to use regressions of growing area on futures prices and to include the realized yield shock as a control variable. An alternative method for bias reduction is to use instrumental variables (IVs). We show that the marginal contribution of an IV to bias reduction is small—IVs are not necessary for futures prices in supply analysis.

@article{lybbert2014weather,
  title={Weather Shocks and Inter-Hemispheric Supply Responses: Implications for Climate Change Effects on Global Food Markets},
  author={Lybbert, Travis J and Smith, Aaron and Sumner, Daniel A},
  journal={Climate Change Economics},
  volume={5},
  number={04},
  pages={1450010},
  year={2014},
	url={https://files.asmith.ucdavis.edu/2014_CCE_LSS_interhemispheric.pdf},
	keywords={agriculture},
	abstract={Climate models predict more weather extremes in the coming decades. Weather shocks can directly reduce crop production, but their effect on food markets is partly buffered by storage and supply responses that can be complex and nuanced. We explore how inter-hemispheric trade and supply responses can moderate the effects of weather shocks on global food supply by enabling potential intra-annual arbitrage. Our estimates of this effect in the case of wheat and soybeans suggest that it may be considerable: 25–50\% of crop production lost to a shock in the Southern Hemisphere is offset six months later by increased production in the North. These results have implications for the potential effects of climate change on global food markets, for how we model these interactions and, possibly, for the design of trade and production-related policies that aim to leverage this inter-hemispheric buffer more effectively.},
  publisher={World Scientific Publishing Company}
}

Climate models predict more weather extremes in the coming decades. Weather shocks can directly reduce crop production, but their effect on food markets is partly buffered by storage and supply responses that can be complex and nuanced. We explore how inter-hemispheric trade and supply responses can moderate the effects of weather shocks on global food supply by enabling potential intra-annual arbitrage. Our estimates of this effect in the case of wheat and soybeans suggest that it may be considerable: 25–50% of crop production lost to a shock in the Southern Hemisphere is offset six months later by increased production in the North. These results have implications for the potential effects of climate change on global food markets, for how we model these interactions and, possibly, for the design of trade and production-related policies that aim to leverage this inter-hemispheric buffer more effectively.

@article{yu2018effects,
  title={Effects of Crop Insurance Premium Subsidies on Crop Acreage},
  author={Yu, Jisang and Smith, Aaron and Sumner, Daniel A},
  journal={American Journal of Agricultural Economics},
  volume={100},
  number={1},
  pages={91--114},
  year={2018},
	url={https://files.asmith.ucdavis.edu/2018_AJAE_YSS_cropinsurance.pdf},
	keywords={agriculture},
	abstract={Crop insurance premium subsidies affect patterns of crop acreage for two reasons. First, holding insurance coverage constant, premium subsidies directly increase expected profit, which encourages more acreage of insured crops (direct profit effect). Second, premium subsidies encourage farms to increase crop insurance coverage. With more insurance coverage, farms obtain more subsidies, and farm revenue becomes less variable as indemnities offset revenue shortfalls, so acreage of insured crops likely increases (indirect coverage effect). By exploiting exogenous policy changes and using approximately 180,000 county‐crop‐year observations, we estimate the sum of these two effects of premium subsidies on the pattern of U.S. acreage across seven major field crops. We estimate that a 10\% increase in the premium subsidy causes a 0.43\% increase in the acreage of a crop in a county holding the premium subsidy of its competing crop constant. Taking into account the small share of premium subsidies in expected crop revenue, this subsidy impact is analogous to an own‐subsidy acreage elasticity of 1.24, which exceeds own‐price acreage elasticity estimates in the literature. One explanation for the larger acreage response to premium subsidies is that insurance causes an indirect coverage effect in addition to a direct profit effect.},
  publisher={Oxford University Press}
}

Crop insurance premium subsidies affect patterns of crop acreage for two reasons. First, holding insurance coverage constant, premium subsidies directly increase expected profit, which encourages more acreage of insured crops (direct profit effect). Second, premium subsidies encourage farms to increase crop insurance coverage. With more insurance coverage, farms obtain more subsidies, and farm revenue becomes less variable as indemnities offset revenue shortfalls, so acreage of insured crops likely increases (indirect coverage effect). By exploiting exogenous policy changes and using approximately 180,000 county‐crop‐year observations, we estimate the sum of these two effects of premium subsidies on the pattern of U.S. acreage across seven major field crops. We estimate that a 10% increase in the premium subsidy causes a 0.43% increase in the acreage of a crop in a county holding the premium subsidy of its competing crop constant. Taking into account the small share of premium subsidies in expected crop revenue, this subsidy impact is analogous to an own‐subsidy acreage elasticity of 1.24, which exceeds own‐price acreage elasticity estimates in the literature. One explanation for the larger acreage response to premium subsidies is that insurance causes an indirect coverage effect in addition to a direct profit effect.

@article{andersen2018century,
  title={A Century of US Farm Productivity Growth: A Surge then a Slowdown},
  author={Andersen, Matthew A and Alston, Julian M and Pardey, Philip G and Smith, Aaron},
  journal={American Journal of Agricultural Economics},
  volume={100},
  number={4},
  pages={1072--1090},
  year={2018},
	url={https://files.asmith.ucdavis.edu/2018_AJAE_AAPS_productivity.pdf},
	keywords={agriculture},
	abstract={U.S. farm productivity growth has direct consequences for sustainably feeding the world's still rapidly growing population, as well as U.S. competitiveness in international markets. Using a newly expanded compilation of multifactor productivity (MFP) estimates and associated partial‐factor productivity (PFP) measures, we examine changes in the pattern of U.S. agricultural productivity growth over the past century and more. Considering the evidence as a whole, we detect sizable and significant slowdowns in the rate of productivity growth in recent decades. U.S. multifactor productivity grew at an annual average rate of just 1.16\% per year during 1990–2007 compared with 1.42\% per year for the period 1910–2007. U.S. yields of major crops grew at an annual average rate of 1.17\% per year for 1990–2009 compared with 1.81\% per year for 1936–1990. More subtly, but with potentially profound implications, the relatively high rates of MFP growth during the third quarter of the century are an historical aberration relative to the long‐run trend.},
	addendum={\textbf{Honorable Mention. Outstanding AJAE Article Award, AAEA, 2019}},
  publisher={Oxford University Press}
}

U.S. farm productivity growth has direct consequences for sustainably feeding the world's still rapidly growing population, as well as U.S. competitiveness in international markets. Using a newly expanded compilation of multifactor productivity (MFP) estimates and associated partial‐factor productivity (PFP) measures, we examine changes in the pattern of U.S. agricultural productivity growth over the past century and more. Considering the evidence as a whole, we detect sizable and significant slowdowns in the rate of productivity growth in recent decades. U.S. multifactor productivity grew at an annual average rate of just 1.16% per year during 1990–2007 compared with 1.42% per year for the period 1910–2007. U.S. yields of major crops grew at an annual average rate of 1.17% per year for 1990–2009 compared with 1.81% per year for 1936–1990. More subtly, but with potentially profound implications, the relatively high rates of MFP growth during the third quarter of the century are an historical aberration relative to the long‐run trend.

@article{metaxoglou2019productivity,
  title={Productivity Spillovers from Pollution Reduction: Reducing Coal use Increases Crop Yields},
  author={Metaxoglou, Konstantinos and Smith, Aaron},
  journal={American Journal of Agricultural Economics},
  volume={102},
  number={1},
  pages={259-280},
  year={2020},
	url={https://asmith.ucdavis.edu/research/coal-crop-yields},
	keywords={agriculture},
	abstract={Air pollution reduces crop yields by slowing down photosynthesis. We estimate the increase in US corn and soybean yields attributed to the recent dramatic reductions in emissions of nitrogen oxides (NOx) from electric power plants. In response to the observed changes in power plant NOx emissions over the eight‐year period from 2003–05 to 2011–13, we estimate that average corn yields improved by 2.46\% and soybean yields by 1.62\%. These improvements imply an increase in total surplus of 1.60 billion dollars annually across the two crops. The estimated yield improvements vary substantially across states depending on the change in NOx emissions. For corn, they range from 0.32\% to 6.87\% and for soybeans, they range from 0.21\% to 4.30\%. The demand for the two crops is quite inelastic, which means that prices decrease by more than production increases in response to this positive productivity shock and the implied rightward shift of the crop supply curve. Due to the low elasticities of supply and demand for U.S. corn and soybeans, we conclude from a welfare analysis that these changes made consumers better off and farmers worse off.},
  publisher={Oxford University Press}
}

Air pollution reduces crop yields by slowing down photosynthesis. We estimate the increase in US corn and soybean yields attributed to the recent dramatic reductions in emissions of nitrogen oxides (NOx) from electric power plants. In response to the observed changes in power plant NOx emissions over the eight‐year period from 2003–05 to 2011–13, we estimate that average corn yields improved by 2.46% and soybean yields by 1.62%. These improvements imply an increase in total surplus of 1.60 billion dollars annually across the two crops. The estimated yield improvements vary substantially across states depending on the change in NOx emissions. For corn, they range from 0.32% to 6.87% and for soybeans, they range from 0.21% to 4.30%. The demand for the two crops is quite inelastic, which means that prices decrease by more than production increases in response to this positive productivity shock and the implied rightward shift of the crop supply curve. Due to the low elasticities of supply and demand for U.S. corn and soybeans, we conclude from a welfare analysis that these changes made consumers better off and farmers worse off.

@article{irwin2020returns,
  title={Returns to Investing in Commodity Futures: Separating the Wheat from the Chaff},
  author={Irwin, Scott H and Sanders, Dwight R, and Smith, Aaron and Main, Scott},
  journal={Applied Economic Perspectives and Policy},
  volume={42},
  number={4},
  pages={583-610},
  year={2020},
	keywords={commodities},
	abstract={Commodity futures investment grew rapidly after its popularity exploded in the mid-2000s. However, real-time performance has been disappointing. Our analysis shows that the disappointing commodity returns were not driven mechanically by contango or negative ``roll yields.'' We show that the expected return to individual commodity futures is near zero before expenses, which implies net losses (before interest earnings) will be equal to order execution and operating costs estimated at 3-4\% per year. Finally, it is likely that rapid increases in commodity prices during 2004-2008 skewed investor return expectations upward much like it did in the early 1970's.},
	url={https://files.asmith.ucdavis.edu/2020_AEPP_ISSS_returns.pdf}
}

Commodity futures investment grew rapidly after its popularity exploded in the mid-2000s. However, real-time performance has been disappointing. Our analysis shows that the disappointing commodity returns were not driven mechanically by contango or negative ``roll yields.'' We show that the expected return to individual commodity futures is near zero before expenses, which implies net losses (before interest earnings) will be equal to order execution and operating costs estimated at 3-4% per year. Finally, it is likely that rapid increases in commodity prices during 2004-2008 skewed investor return expectations upward much like it did in the early 1970's.

@article{suenaga2008volatility,
  title={Volatility Dynamics of NYMEX Natural Gas Futures Prices},
  author={Suenaga, Hiroaki and Smith, Aaron and Williams, Jeffrey},
  journal={Journal of Futures Markets},
  volume={28},
  number={5},
  pages={438--463},
  year={2008},
	url={https://files.asmith.ucdavis.edu/2008_JFutMkt_SSW_NGfutures.pdf},
	keywords={commodities},
	abstract={We examine the volatility dynamics of NYMEX natural gas futures prices via the partially overlapping time‐series model of Smith (2005. Journal of Applied Econometrics, 20, 405–422). We show that volatility exhibits two important features: (1) volatility is greater in the winter than in the summer, and (2) the persistence of price shocks and, hence, the correlations among concurrently traded contracts, displays substantial seasonal and cross‐sectional variation in a way consistent with the theory of storage. We demonstrate that, by ignoring the seasonality in the volatility dynamics of natural gas futures prices, previous studies have suggested sub‐optimal hedging strategies.},
  publisher={Wiley Subscription Services, Inc., A Wiley Company Hoboken}
}

We examine the volatility dynamics of NYMEX natural gas futures prices via the partially overlapping time‐series model of Smith (2005. Journal of Applied Econometrics, 20, 405–422). We show that volatility exhibits two important features: (1) volatility is greater in the winter than in the summer, and (2) the persistence of price shocks and, hence, the correlations among concurrently traded contracts, displays substantial seasonal and cross‐sectional variation in a way consistent with the theory of storage. We demonstrate that, by ignoring the seasonality in the volatility dynamics of natural gas futures prices, previous studies have suggested sub‐optimal hedging strategies.

@article{smith2005partially,
  title={Partially Overlapping Time Series: A New Model for Volatility Dynamics in Commodity Futures},
  author={Smith, Aaron},
  journal={Journal of Applied Econometrics},
  volume={20},
  number={3},
  pages={405--422},
  year={2005},
	url={https://files.asmith.ucdavis.edu/2005_JAppEmet_S_POTS.pdf},
	keywords={commodities},
	abstract={In commodity futures markets, contracts with various delivery dates trade simultaneously. Applied researchers typically discard the majority of the data and form a single time series by choosing only one price observation per day. This strategy precludes a full understanding of these markets and can induce complicated nonlinear dynamics in the data. In this paper, I introduce the partially overlapping time series (POTS) model to model jointly all traded contracts. The POTS model incorporates time‐to‐delivery, storability, seasonality and GARCH effects. I apply the POTS model to corn futures at the Chicago Board of Trade and the results uncover substantial inefficiency associated with delivery on corn futures. The results also support two theories of commodity pricing: the theory of storage and the Samuelson effect.},
  publisher={John Wiley \& Sons, Ltd.}
}

In commodity futures markets, contracts with various delivery dates trade simultaneously. Applied researchers typically discard the majority of the data and form a single time series by choosing only one price observation per day. This strategy precludes a full understanding of these markets and can induce complicated nonlinear dynamics in the data. In this paper, I introduce the partially overlapping time series (POTS) model to model jointly all traded contracts. The POTS model incorporates time‐to‐delivery, storability, seasonality and GARCH effects. I apply the POTS model to corn futures at the Chicago Board of Trade and the results uncover substantial inefficiency associated with delivery on corn futures. The results also support two theories of commodity pricing: the theory of storage and the Samuelson effect.

@article{carter2011commodity,
  title={Commodity Booms and Busts},
  author={Carter, Colin A and Rausser, Gordon C and Smith, Aaron},
  journal={Annual Review of Resource Economics},
  volume={3},
  number={},
  pages={87--118},
  year={2011},
	url={https://files.asmith.ucdavis.edu/2011_AnnRev_CRS_boombust.pdf},
	keywords={commodities},
	abstract={Periodically, the global economy experiences great commodity booms and busts, characterized by a broad and sharp comovement of commodity prices. There have been two such episodes since the Korean War. The first event peaked in 1974 and the second in 2008, 34 years apart. Both created major economic and political shocks, including fallen governments and human suffering due to high food prices. Each occurrence raised serious concerns over food and energy security and led to more government intervention in the commodity markets. Although there is no simple explanation for what causes such complex events, they do share similar characteristics. We find at the core of these cycles a set of contemporaneous supply and demand surprises that coincided with low inventories and that were magnified by macroeconomic shocks and policy responses. In the next few decades, the world faces the prospect of continued increases in the demand for commodities and greater uncertainty about supply. However, because market participants are likely to respond by increasing inventory holdings and investing in new technologies, we see no reason to expect an increase in the frequency of dramatic commodity booms and busts.},
  publisher={Annual Reviews}
}

Periodically, the global economy experiences great commodity booms and busts, characterized by a broad and sharp comovement of commodity prices. There have been two such episodes since the Korean War. The first event peaked in 1974 and the second in 2008, 34 years apart. Both created major economic and political shocks, including fallen governments and human suffering due to high food prices. Each occurrence raised serious concerns over food and energy security and led to more government intervention in the commodity markets. Although there is no simple explanation for what causes such complex events, they do share similar characteristics. We find at the core of these cycles a set of contemporaneous supply and demand surprises that coincided with low inventories and that were magnified by macroeconomic shocks and policy responses. In the next few decades, the world faces the prospect of continued increases in the demand for commodities and greater uncertainty about supply. However, because market participants are likely to respond by increasing inventory holdings and investing in new technologies, we see no reason to expect an increase in the frequency of dramatic commodity booms and busts.

@article{garcia2015futures,
  title={Futures Market Failure?},
  author={Garcia, Philip and Irwin, Scott H and Smith, Aaron},
  journal={American Journal of Agricultural Economics},
  volume={97},
  number={1},
  pages={40--64},
  year={2015},
	url={https://files.asmith.ucdavis.edu/2015_AJAE_GIS_convergence.pdf},
	keywords={commodities},
	abstract={In a well‐functioning futures market, the futures price at expiration equals the price of the underlying asset. This condition failed to hold in grain markets for most of 2005‐2010, calling into question the ability of these markets to perform their price discovery and risk management functions. During this period, futures contracts expired up to 35\% above the cash grain price. We develop a dynamic rational expectations model of commodity storage that explains how these recent convergence failures were generated by the institutional structure of the delivery system. When delivery occurs on a grain futures contract, the firm on the short side of the market provides a delivery instrument (a warehouse receipt or shipping certificate) to the firm on the long side of the market. The firm taking delivery may hold the delivery instrument indefinitely, providing it pays a daily storage rate. The futures exchange sets the maximum allowable storage rate at a fixed value. We show that non‐convergence arises in equilibrium when the market price of physical grain storage exceeds the maximum storage rate on delivery instruments. We call the difference between the price of carrying physical grain and the maximum storage rate the wedge, and demonstrate theoretically and empirically that the magnitude of the non‐convergence equals the expected present discounted value of a function of future wedges.},
	addendum={\textbf{Winner of Quality of Research Discovery Award, AAEA, 2016.}},
  publisher={Oxford University Press}
}

In a well‐functioning futures market, the futures price at expiration equals the price of the underlying asset. This condition failed to hold in grain markets for most of 2005‐2010, calling into question the ability of these markets to perform their price discovery and risk management functions. During this period, futures contracts expired up to 35% above the cash grain price. We develop a dynamic rational expectations model of commodity storage that explains how these recent convergence failures were generated by the institutional structure of the delivery system. When delivery occurs on a grain futures contract, the firm on the short side of the market provides a delivery instrument (a warehouse receipt or shipping certificate) to the firm on the long side of the market. The firm taking delivery may hold the delivery instrument indefinitely, providing it pays a daily storage rate. The futures exchange sets the maximum allowable storage rate at a fixed value. We show that non‐convergence arises in equilibrium when the market price of physical grain storage exceeds the maximum storage rate on delivery instruments. We call the difference between the price of carrying physical grain and the maximum storage rate the wedge, and demonstrate theoretically and empirically that the magnitude of the non‐convergence equals the expected present discounted value of a function of future wedges.

@article{fishe2014hedging,
  title={Hedging and Speculative Trading in Agricultural Futures Markets},
  author={Fishe, Raymond PH and Janzen, Joseph P and Smith, Aaron},
  journal={American Journal of Agricultural Economics},
  volume={96},
  number={2},
  pages={542--556},
  year={2014},
	url={https://files.asmith.ucdavis.edu/2014_AJAE_FJS_hedging.pdf},
	keywords={commodities},
	abstract={Regulators and industry participants have expressed concern that excessive speculation harms agricultural futures markets. Such harm may arise if speculators cause prices to systematically differ from the price sequence that would arise in markets populated by equally informed traders with rational expectations (RE). We show theoretically that, when traders exhibit differences of opinion (DO) about the expected value of the commodity, futures prices may diverge from the RE equilibrium. Moreover, we develop a testable prediction, namely that positions held by different trader groups are correlated with prices in a DO equilibrium but not correlated in a RE equilibrium. We find strong empirical support for the DO‐type environment; changes in positions held by managed money traders are positively correlated with prices, and changes in positions held by producers are negatively correlated. In the context of our DO model, this finding implies that prices change by more on average than producers think they should and by less than managed money thinks they should. However, the evidence suggests that neither group is systematically more prescient than the other.},
	addendum={\textbf{Winner of Quality of Research Discovery Award, EAAE, 2015}},
  publisher={Oxford University Press}
}

Regulators and industry participants have expressed concern that excessive speculation harms agricultural futures markets. Such harm may arise if speculators cause prices to systematically differ from the price sequence that would arise in markets populated by equally informed traders with rational expectations (RE). We show theoretically that, when traders exhibit differences of opinion (DO) about the expected value of the commodity, futures prices may diverge from the RE equilibrium. Moreover, we develop a testable prediction, namely that positions held by different trader groups are correlated with prices in a DO equilibrium but not correlated in a RE equilibrium. We find strong empirical support for the DO‐type environment; changes in positions held by managed money traders are positively correlated with prices, and changes in positions held by producers are negatively correlated. In the context of our DO model, this finding implies that prices change by more on average than producers think they should and by less than managed money thinks they should. However, the evidence suggests that neither group is systematically more prescient than the other.

@article{adjemian2013non,
  title={Non-Convergence in Domestic Commodity Futures Markets: Causes, Consequences, and Remedies},
  author={Adjemian, Michael and Garcia, Philip and Irwin, Scott H and Smith, Aaron},
  journal={USDA-ERS Economic Information Bulletin},
  number={115},
	url={https://www.ers.usda.gov/publications/pub-details/?pubid=43780},
	keywords={commodities},
	abstract={From 2005-10, the price of expiring U.S. corn, soybean, and wheat futures contracts settled much higher than corresponding delivery market cash prices. Theories about why this unprecedented non-convergence occurred are examined along with policy options to prevent it in the future.},
  year={2013}
}

From 2005-10, the price of expiring U.S. corn, soybean, and wheat futures contracts settled much higher than corresponding delivery market cash prices. Theories about why this unprecedented non-convergence occurred are examined along with policy options to prevent it in the future.

@article{janzen2014deconstructing,
  title={Deconstructing Wheat Price Spikes: A Model of Supply and Demand, Financial Speculation, and Commodity Price Comovement},
  author={Janzen, Joseph and Carter, Colin A and Smith, Aaron and Adjemian, Michael},
  journal={USDA-ERS Economic Research Report},
  number={165},
  year={2014},
	url={https://www.ers.usda.gov/publications/pub-details/?pubid=45202},
	abstract={In 2008, wheat futures prices spiked and then crashed along with prices for other agricultural and non-agricultural commodities. This study uses an econometric model to explain the influence of various factors, including passive speculation by large traders, on wheat prices. Findings show that market-specific shocks related to supply and demand for wheat were the dominant cause of price spikes.},
	keywords={commodities}
}

In 2008, wheat futures prices spiked and then crashed along with prices for other agricultural and non-agricultural commodities. This study uses an econometric model to explain the influence of various factors, including passive speculation by large traders, on wheat prices. Findings show that market-specific shocks related to supply and demand for wheat were the dominant cause of price spikes.

@article{fishe2019speculators,
  title={Do Speculators Drive Commodity Prices Away from Supply and Demand Fundamentals?},
  author={Fishe, Raymond PH and Smith, Aaron},
  journal={Journal of Commodity Markets},
  volume={15},
  pages={100078},
  year={2019},
	url={https://files.asmith.ucdavis.edu/2019_JCommMkt_FS_speculators.pdf},
	keywords={commodities},
	abstract={No. We show that managed money traders tend to change positions in the same direction as prices, whereas commercial firms change positions in the opposite direction. Using insights from difference of opinion theory, we conclude that managed money traders have strong beliefs about the markets and trade aggressively. Commercial firms are willing to take the other side of these trades, and thus they provide liquidity to managed money firms. However, we find no evidence that this trading dynamic results in prices that deviate significantly from supply and demand fundamentals.},
  publisher={Elsevier}
}

No. We show that managed money traders tend to change positions in the same direction as prices, whereas commercial firms change positions in the opposite direction. Using insights from difference of opinion theory, we conclude that managed money traders have strong beliefs about the markets and trade aggressively. Commercial firms are willing to take the other side of these trades, and thus they provide liquidity to managed money firms. However, we find no evidence that this trading dynamic results in prices that deviate significantly from supply and demand fundamentals.

@article{main2018financialization,
  title={Financialization and the Returns to Commodity Investments},
  author={Main, Scott and Irwin, Scott H and Sanders, Dwight R and Smith, Aaron},
  journal={Journal of Commodity Markets},
  volume={10},
  pages={22--28},
  year={2018},
	url={https://files.asmith.ucdavis.edu/2018_JCommMkt_MISS_financialization.pdf},
	keywords={commodities},
	abstract={No. We show that managed money traders tend to change positions in the same direction as prices, whereas commercial firms change positions in the opposite direction. Using insights from difference of opinion theory, we conclude that managed money traders have strong beliefs about the markets and trade aggressively. Commercial firms are willing to take the other side of these trades, and thus they provide liquidity to managed money firms. However, we find no evidence that this trading dynamic results in prices that deviate significantly from supply and demand fundamentals.},
  publisher={Elsevier}
}

No. We show that managed money traders tend to change positions in the same direction as prices, whereas commercial firms change positions in the opposite direction. Using insights from difference of opinion theory, we conclude that managed money traders have strong beliefs about the markets and trade aggressively. Commercial firms are willing to take the other side of these trades, and thus they provide liquidity to managed money firms. However, we find no evidence that this trading dynamic results in prices that deviate significantly from supply and demand fundamentals.

@article{janzen2018commodity,
  title={Commodity Price Comovement and Financial Speculation: The Case of Cotton},
  author={Janzen, Joseph P and Smith, Aaron and Carter, Colin A},
  journal={American Journal of Agricultural Economics},
  volume={100},
  number={1},
  pages={264-285},
  year={2018},
	url={https://files.asmith.ucdavis.edu/2018_AJAE_JSC_cotton.pdf},
	keywords={commodities},
	abstract={Recent booms and busts in commodity prices have generated concerns that financial speculation causes excessive commodity‐price comovement, driving prices away from levels implied by supply and demand under rational expectations. We develop a structural vector autoregression model of a commodity futures market and use it to explain two recent spikes in cotton prices. In doing so, we make two contributions to the literature on commodity price dynamics. First, we estimate the extent to which cotton price booms and busts can be attributed to comovement with other commodities. Finding such comovement would be necessary but would not be sufficient evidence to establish that broad‐based financial speculation drives commodity prices. Second, after controlling for aggregate demand and comovement, we develop a new method to point identify shocks to precautionary demand for cotton separately from shocks to current supply and demand. To do so, we use differences in volatility across time implied by the rational expectations competitive storage model. We find limited evidence that financial speculation caused cotton prices to spike in 2008 or 2011. We conclude that the 2008 price spike was driven mostly by precautionary demand for cotton, and the 2011 spike was caused by a net supply shortfall.},
  publisher={Oxford University Press}
}

Recent booms and busts in commodity prices have generated concerns that financial speculation causes excessive commodity‐price comovement, driving prices away from levels implied by supply and demand under rational expectations. We develop a structural vector autoregression model of a commodity futures market and use it to explain two recent spikes in cotton prices. In doing so, we make two contributions to the literature on commodity price dynamics. First, we estimate the extent to which cotton price booms and busts can be attributed to comovement with other commodities. Finding such comovement would be necessary but would not be sufficient evidence to establish that broad‐based financial speculation drives commodity prices. Second, after controlling for aggregate demand and comovement, we develop a new method to point identify shocks to precautionary demand for cotton separately from shocks to current supply and demand. To do so, we use differences in volatility across time implied by the rational expectations competitive storage model. We find limited evidence that financial speculation caused cotton prices to spike in 2008 or 2011. We conclude that the 2008 price spike was driven mostly by precautionary demand for cotton, and the 2011 spike was caused by a net supply shortfall.

@article{ghanem2022causality,
  title={Causality in Structural Vector Autoregressions: Science or Sorcery?},
  author={Ghanem, Dalia and Smith, Aaron},
  journal={American Journal of Agricultural Economics},
  year={2022},
  volume={104},
  number={},
  pages={881-904},
	keywords={econometrics},
	abstract={This paper presents the structural vector autoregression (SVAR) as a method for estimating dynamic causal effects in agricultural and resource economics. Our paper has a pedagogical purpose; we aim the presentation at economists trained primarily in microeconometrics. We emphasize connections between SVARs and the classical instrumental variables (IV) model, both of which aim to extract exogenous variation from endogenous variables. We show that the population analogue of the Wald IV estimator is identical to the ratio of two impulse responses from an SVAR. We present an SVAR analysis of global supply and demand for agricultural commodities, which was previously examined using IV Roberts and Schlenker (2013).  We illustrate the additional economic insights that the SVAR reveals. We estimate that demand responds similarly to one-year and longer-run supply changes, whereas supply responds differently depending on whether a price change is driven by poor weather last year or a jump in consumption demand. We highlight the assumptions required to gain these insights and illustrate the robustness of our results to alternative assumptions.},
	url={https://asmith.ucdavis.edu/research/causality-structural-vector-autoregressions-science-or-sorcery}
}

This paper presents the structural vector autoregression (SVAR) as a method for estimating dynamic causal effects in agricultural and resource economics. Our paper has a pedagogical purpose; we aim the presentation at economists trained primarily in microeconometrics. We emphasize connections between SVARs and the classical instrumental variables (IV) model, both of which aim to extract exogenous variation from endogenous variables. We show that the population analogue of the Wald IV estimator is identical to the ratio of two impulse responses from an SVAR. We present an SVAR analysis of global supply and demand for agricultural commodities, which was previously examined using IV Roberts and Schlenker (2013). We illustrate the additional economic insights that the SVAR reveals. We estimate that demand responds similarly to one-year and longer-run supply changes, whereas supply responds differently depending on whether a price change is driven by poor weather last year or a jump in consumption demand. We highlight the assumptions required to gain these insights and illustrate the robustness of our results to alternative assumptions.

@article{ferguson2021import,
  title={Import Demand Elasticities Based on Quantity Data: Theory and Evidence},
  author={Ferguson, Shon M and Smith, Aaron},
  journal={Canadian Journal of Economics},
  year={2022},
  volume={55},
  number={2},
  pages={1027-1056},
	abstract={Correct estimates of import demand elasticities are essential for measuring the gains from trade and predicting the impact of trade policies. We show that estimates of import demand elasticities hinge critically on whether they are derived using trade quantities or trade values, and this difference is due to properties of the estimators. Using partial identification methods, we show theoretically that the upper bound on the set of plausible estimates is lower when using traded quantities, compared to the standard approach using trade values. Our theoretical predictions are confirmed using detailed product-level data on U.S. imports for the years 1993-2006. Our proposed method using traded quantities leads to smaller point estimates of the import demand elasticities for many goods and imply larger gains from trade compared to estimates based on trade values.},
	keywords={econometrics},
	url={https://files.asmith.ucdavis.edu/2021_CJE_FS_armington.pdf}
}

Correct estimates of import demand elasticities are essential for measuring the gains from trade and predicting the impact of trade policies. We show that estimates of import demand elasticities hinge critically on whether they are derived using trade quantities or trade values, and this difference is due to properties of the estimators. Using partial identification methods, we show theoretically that the upper bound on the set of plausible estimates is lower when using traded quantities, compared to the standard approach using trade values. Our theoretical predictions are confirmed using detailed product-level data on U.S. imports for the years 1993-2006. Our proposed method using traded quantities leads to smaller point estimates of the import demand elasticities for many goods and imply larger gains from trade compared to estimates based on trade values.

@article{ghanem2021which,
  title={What are the Benefits of High-Frequency Data for Fixed Effects Panel Models},
  author={Ghanem, Dalia and Smith, Aaron},
  journal={Journal of the Association of Environmental and Resource Economists},
  volume={8},
  number={2},
  pages={199-234},
  year={2021},
	keywords={econometrics},
	abstract={High-frequency panel data sets, where outcomes and regressors are observed at a daily or hourly frequency, are increasingly available in environmental and resource economics. To understand the potential gains from these richer datasets, this paper compares fixed effects estimators using high-frequency data with those using temporally aggregated data. We provide a set of conditions under which both estimators are consistent for the same parameter. Three departures from these conditions are: (1) response heterogeneity at the high-frequency dimension, (2) differential response to high- and low-frequency variation in the regressor, (3) nonlinearities in the relationship between the high-frequency outcome and regressor. Under these alternative conditions, the two estimators converge to different probability limits. In general, we recommend that empirical researchers think carefully about the features of the ``true'' high-frequency outcome equation to understand the effects of high-frequency data and temporal aggregation. We illustrate our results using an application to the energy-temperature relationship.},
	addendum={\textbf{Winner of Outstanding Publication in JAERE, 2021, AERE}},
	url={https://files.asmith.ucdavis.edu/2021_JAERE_GS_highfreq.pdf}
}

High-frequency panel data sets, where outcomes and regressors are observed at a daily or hourly frequency, are increasingly available in environmental and resource economics. To understand the potential gains from these richer datasets, this paper compares fixed effects estimators using high-frequency data with those using temporally aggregated data. We provide a set of conditions under which both estimators are consistent for the same parameter. Three departures from these conditions are: (1) response heterogeneity at the high-frequency dimension, (2) differential response to high- and low-frequency variation in the regressor, (3) nonlinearities in the relationship between the high-frequency outcome and regressor. Under these alternative conditions, the two estimators converge to different probability limits. In general, we recommend that empirical researchers think carefully about the features of the ``true'' high-frequency outcome equation to understand the effects of high-frequency data and temporal aggregation. We illustrate our results using an application to the energy-temperature relationship.

@article{engle1999stochastic,
  title={Stochastic Permanent Breaks},
  author={Engle, Robert F and Smith, Aaron},
  journal={Review of Economics and Statistics},
  volume={81},
  number={4},
  pages={553--574},
  year={1999},
	keywords={econometrics},
	url={https://files.asmith.ucdavis.edu/1999_REStat_ES_stopbreak.pdf},
	abstract={This paper bridges the gap between processes where shocks are permanent and those with transitory shocks by formulating a process in which the long-run impact of each innovation is time-varying and stochastic. In the stochastic permanent breaks (STOPBREAK) process, frequent transitory shocks are supplemented by occasional permanent shifts. Consistency and asymptotic normality of quasi-maximum-likelihood estimates is established, and locally best hypothesis tests of the null of a random walk are developed. The model is applied to relative prices of pairs of stocks and significant test statistics result.},
  publisher={MIT Press}
}

This paper bridges the gap between processes where shocks are permanent and those with transitory shocks by formulating a process in which the long-run impact of each innovation is time-varying and stochastic. In the stochastic permanent breaks (STOPBREAK) process, frequent transitory shocks are supplemented by occasional permanent shifts. Consistency and asymptotic normality of quasi-maximum-likelihood estimates is established, and locally best hypothesis tests of the null of a random walk are developed. The model is applied to relative prices of pairs of stocks and significant test statistics result.

@article{smith2005level,
  title={Level Shifts and the Illusion of Long Memory in Economic Time Series},
  author={Smith, Aaron},
  journal={Journal of Business and Economic Statistics},
  volume={23},
  number={3},
  pages={321--335},
  year={2005},
	URL={https://files.asmith.ucdavis.edu/2005_JBES_S_levelshifts.pdf},
	keywords={econometrics},
	abstract={When applied to time series processes containing occasional level shifts, the log-periodogram (GPH) estimator often erroneously finds long memory. For a stationary short-memory process with a slowly varying level, I show that the GPH estimator is substantially biased, and derive an approximation to this bias. The asymptotic bias lies on the (0, 1) interval, and its exact value depends on the ratio of the expected number of level shifts to a user-defined bandwidth parameter. Using this result, I formulate the modified GPH estimator, which has a markedly lower bias. I illustrate this new estimator via applications to soybean prices and stock market volatility.},
  publisher={Taylor \& Francis}
}

When applied to time series processes containing occasional level shifts, the log-periodogram (GPH) estimator often erroneously finds long memory. For a stationary short-memory process with a slowly varying level, I show that the GPH estimator is substantially biased, and derive an approximation to this bias. The asymptotic bias lies on the (0, 1) interval, and its exact value depends on the ratio of the expected number of level shifts to a user-defined bandwidth parameter. Using this result, I formulate the modified GPH estimator, which has a markedly lower bias. I illustrate this new estimator via applications to soybean prices and stock market volatility.

@article{smith2006markov,
  title={Markov-Switching Model Selection Using Kullback--Leibler Divergence},
  author={Smith, Aaron and Naik, Prasad A and Tsai, Chih-Ling},
  journal={Journal of Econometrics},
  volume={134},
  number={2},
  pages={553--577},
  year={2006},
	url={https://files.asmith.ucdavis.edu/2006_JE_SNT_msc.pdf},
	keywords={econometrics},
	abstract={In Markov-switching regression models, we use Kullback–Leibler (KL) divergence between the true and candidate models to select the number of states and variables simultaneously. Specifically, we derive a new information criterion, Markov switching criterion (MSC), which is an estimate of KL divergence. MSC imposes an appropriate penalty to mitigate the over-retention of states in the Markov chain, and it performs well in Monte Carlo studies with single and multiple states, small and large samples, and low and high noise. We illustrate the usefulness of MSC via applications to the U.S. business cycle and to media advertising.},
  publisher={North-Holland}
}

In Markov-switching regression models, we use Kullback–Leibler (KL) divergence between the true and candidate models to select the number of states and variables simultaneously. Specifically, we derive a new information criterion, Markov switching criterion (MSC), which is an estimate of KL divergence. MSC imposes an appropriate penalty to mitigate the over-retention of states in the Markov chain, and it performs well in Monte Carlo studies with single and multiple states, small and large samples, and low and high noise. We illustrate the usefulness of MSC via applications to the U.S. business cycle and to media advertising.

@article{metaxoglou2007maximum,
  title={Maximum Likelihood Estimation of VARMA Models Using a State-Space EM Algorithm},
  author={Metaxoglou, Konstantinos and Smith, Aaron},
  journal={Journal of Time Series Analysis},
  volume={28},
  number={5},
  pages={666--685},
  year={2007},
	url={https://files.asmith.ucdavis.edu/2007_JTSA_MS_VARMA.pdf},
	keywords={econometrics},
	abstract={We introduce a state‐space representation for vector autoregressive moving‐average models that enables maximum likelihood estimation using the EM algorithm. We obtain closed‐form expressions for both the E‐ and M‐steps; the former requires the Kalman filter and a fixed‐interval smoother, and the latter requires least squares‐type regression. We show via simulations that our algorithm converges reliably to the maximum, whereas gradient‐based methods often fail because of the highly nonlinear nature of the likelihood function. Moreover, our algorithm converges in a smaller number of function evaluations than commonly used direct‐search routines. Overall, our approach achieves its largest performance gains when applied to models of high dimension. We illustrate our technique by estimating a high‐dimensional vector moving‐average model for an efficiency test of California's wholesale electricity market.},
  publisher={Wiley Online Library}
}

We introduce a state‐space representation for vector autoregressive moving‐average models that enables maximum likelihood estimation using the EM algorithm. We obtain closed‐form expressions for both the E‐ and M‐steps; the former requires the Kalman filter and a fixed‐interval smoother, and the latter requires least squares‐type regression. We show via simulations that our algorithm converges reliably to the maximum, whereas gradient‐based methods often fail because of the highly nonlinear nature of the likelihood function. Moreover, our algorithm converges in a smaller number of function evaluations than commonly used direct‐search routines. Overall, our approach achieves its largest performance gains when applied to models of high dimension. We illustrate our technique by estimating a high‐dimensional vector moving‐average model for an efficiency test of California's wholesale electricity market.

@article{smith2005forecasting,
  title={Forecasting in the Presence of Level Shifts},
  author={Smith, Aaron},
  journal={Journal of Forecasting},
  volume={24},
  number={8},
  pages={557--574},
  year={2005},
	url={https://files.asmith.ucdavis.edu/2005_JFore_S_levelshifts.pdf},
	keywords={econometrics},
	abstract={This article addresses the problem of forecasting time series that are subject to level shifts. Processes with level shifts possess a nonlinear dependence structure. Using the stochastic permanent breaks (STOPBREAK) model, I model this nonlinearity in a direct and flexible way that avoids imposing a discrete regime structure. I apply this model to the rate of price inflation in the United States, which I show is subject to level shifts. These shifts significantly affect the accuracy of out‐of‐sample forecasts, causing models that assume covariance stationarity to be substantially biased. Models that do not assume covariance stationarity, such as the random walk, are unbiased but lack precision in periods without shifts. I show that the STOPBREAK model outperforms several alternative models in an out‐of‐sample inflation forecasting experiment.},
  publisher={Wiley Online Library}
}

This article addresses the problem of forecasting time series that are subject to level shifts. Processes with level shifts possess a nonlinear dependence structure. Using the stochastic permanent breaks (STOPBREAK) model, I model this nonlinearity in a direct and flexible way that avoids imposing a discrete regime structure. I apply this model to the rate of price inflation in the United States, which I show is subject to level shifts. These shifts significantly affect the accuracy of out‐of‐sample forecasts, causing models that assume covariance stationarity to be substantially biased. Models that do not assume covariance stationarity, such as the random walk, are unbiased but lack precision in periods without shifts. I show that the STOPBREAK model outperforms several alternative models in an out‐of‐sample inflation forecasting experiment.

@article{smith2012markov,
  title={Markov Breaks in Regression Models},
  author={Smith, Aaron},
  journal={Journal of Time Series Econometrics},
  volume={4},
  number={1},
  year={2012},
	url={https://files.asmith.ucdavis.edu/2012_JTSEmet_S_MB.pdf},
	keywords={econometrics},
	abstract={This article develops a new Markov breaks (MB) model for forecasting and making inference in linear regression models with breaks that are stochastic in both timing and magnitude. The MB model permits an arbitrarily large number of abrupt breaks in the regression coefficients and error variance, but it maintains a low-dimensional state space, and therefore it is computationally straightforward. In particular, the likelihood function can be computed analytically using a single iterative pass through the data and thereby avoids Monte Carlo integration. The model generates forecasts and conditional coefficient predictions using a probability weighted average over regressions that include progressively more historical data. I employ the MB model to study the predictive ability of the yield curve for quarterly GDP growth. I show evidence of breaks in the predictive relationship, and the MB model outperforms competing breaks models in an out-of-sample forecasting experiment.},
  publisher={De Gruyter}
}

This article develops a new Markov breaks (MB) model for forecasting and making inference in linear regression models with breaks that are stochastic in both timing and magnitude. The MB model permits an arbitrarily large number of abrupt breaks in the regression coefficients and error variance, but it maintains a low-dimensional state space, and therefore it is computationally straightforward. In particular, the likelihood function can be computed analytically using a single iterative pass through the data and thereby avoids Monte Carlo integration. The model generates forecasts and conditional coefficient predictions using a probability weighted average over regressions that include progressively more historical data. I employ the MB model to study the predictive ability of the yield curve for quarterly GDP growth. I show evidence of breaks in the predictive relationship, and the MB model outperforms competing breaks models in an out-of-sample forecasting experiment.

@article{hendricks2015grouped,
  title={Grouped Coefficients to Reduce Bias in Heterogeneous Dynamic Panel Models with Small \textit{T}},
  author={Hendricks, Nathan P and Smith, Aaron},
  journal={Applied Economics},
  volume={47},
  number={40},
  pages={4335--4348},
  year={2015},
	url={https://files.asmith.ucdavis.edu/2015_AE_HS_grouped.pdf},
	keywords={econometrics},
	abstract={We propose the grouped coefficients estimator to reduce bias in dynamic panels with small T that have a multilevel structure to the coefficient and factor loading heterogeneity. If groups are chosen such that the within-group heterogeneity is small, then the grouped coefficients estimator can lead to substantial bias reduction compared to pooled GMM dynamic panel estimators. We also propose using a Wald test that can be used to assess whether pooled estimators suffer from heterogeneity bias. We illustrate the usefulness of grouped coefficients with an application to labour demand in which the coefficients are grouped by sub-sector. Our results suggest that the standard pooled estimates are substantially biased.},
  publisher={Routledge}
}

We propose the grouped coefficients estimator to reduce bias in dynamic panels with small T that have a multilevel structure to the coefficient and factor loading heterogeneity. If groups are chosen such that the within-group heterogeneity is small, then the grouped coefficients estimator can lead to substantial bias reduction compared to pooled GMM dynamic panel estimators. We also propose using a Wald test that can be used to assess whether pooled estimators suffer from heterogeneity bias. We illustrate the usefulness of grouped coefficients with an application to labour demand in which the coefficients are grouped by sub-sector. Our results suggest that the standard pooled estimates are substantially biased.

@article{reed2017time,
  title={A Time Series Paradox: Unit Root Tests Perform Poorly When Data are Cointegrated},
  author={Reed, W Robert and Smith, Aaron},
  journal={Economics Letters},
  volume={151},
  pages={71--74},
  year={2017},
	url={https://files.asmith.ucdavis.edu/2017_EL_RS_unitroot},
	keywords={econometrics},
	abstract={Cointegration among time series paradoxically makes it more likely that a unit test will reject the unit root null hypothesis on the individual series. This occurs because at least one series in the system has a negative moving average component.},
  publisher={North-Holland}
}

Cointegration among time series paradoxically makes it more likely that a unit test will reject the unit root null hypothesis on the individual series. This occurs because at least one series in the system has a negative moving average component.

@article{bushnell2022food,
  title={Food vs. Fuel? Impacts of the North Dakota Oil Boom on Agricultural Prices},
  author={Bushnell, James B and Hughes, Jonathan E and Smith, Aaron},
  journal={Journal of the Association of Environmental and Resource Economists},
  year={2022},
  volume={9},
  number={1},
  pages={79-112},
	keywords={energy},
	abstract={Farmers and politicians in North Dakota and nearby states claim dramatic increases in shipments of crude oil by rail in 2013-14 caused service delays and higher costs.  We investigate these claims accounting for other potential sources of rail congestion.  We show that grain price spreads between the market hub and regional elevators  expanded significantly when crude oil shipments increased. However, the incidence of those effects was borne mostly by buyers paying higher prices at the hub, rather than farmers receiving lower prices. The effects differ by the type of grain being transported.  Wheat markets were affected much more than corn and soybeans, most likely because shipping delays were more costly for wheat than corn and soybeans. When rail capacity is scarce, railroads use railcar auctions to price discriminate over the time sensitivity of a shipment.},
  year={2022},
	url={https://asmith.ucdavis.edu/research/North_Dakota_oil}
}

Farmers and politicians in North Dakota and nearby states claim dramatic increases in shipments of crude oil by rail in 2013-14 caused service delays and higher costs. We investigate these claims accounting for other potential sources of rail congestion. We show that grain price spreads between the market hub and regional elevators expanded significantly when crude oil shipments increased. However, the incidence of those effects was borne mostly by buyers paying higher prices at the hub, rather than farmers receiving lower prices. The effects differ by the type of grain being transported. Wheat markets were affected much more than corn and soybeans, most likely because shipping delays were more costly for wheat than corn and soybeans. When rail capacity is scarce, railroads use railcar auctions to price discriminate over the time sensitivity of a shipment.

@article{novan2020residential,
  title={Residential Building Codes Do Save Energy: Evidence From Hourly Smart-Meter Data},
  author={Novan, Kevin and Smith, Aaron and Zhou, Tianxia},
  journal={Review of Economics and Statistics},
	keywords={energy},
  year={2022},
  volume={104},
  number={3},
  pages={483-500},
	abstract={In 1978, California adopted building codes designed to reduce the energy used for temperature control. Using a rich dataset of hourly electricity consumption for 158,112 houses in Sacramento, we estimate that the average house built just after 1978 uses 8\% to 13\% less electricity for cooling than a similar house built just before 1978. Comparing the estimated savings to the policy's projected cost, our results suggest the policy passes a cost-benefit test. In settings where market failures prevent energy costs from being completely passed through to home prices, building codes can serve as a cost-effective tool for improving energy efficiency.},
	url={https://asmith.ucdavis.edu/research/residential-building-codes-do-save-energy}
}

In 1978, California adopted building codes designed to reduce the energy used for temperature control. Using a rich dataset of hourly electricity consumption for 158,112 houses in Sacramento, we estimate that the average house built just after 1978 uses 8% to 13% less electricity for cooling than a similar house built just before 1978. Comparing the estimated savings to the policy's projected cost, our results suggest the policy passes a cost-benefit test. In settings where market failures prevent energy costs from being completely passed through to home prices, building codes can serve as a cost-effective tool for improving energy efficiency.

@article{metaxoglou2007efficiency,
  title={Efficiency of the California Electricity Reserves Market},
  author={Metaxoglou, Konstantinos and Smith, Aaron},
  journal={Journal of Applied Econometrics},
  volume={22},
  number={6},
  pages={1127--1144},
  year={2007},
	url={https://files.asmith.ucdavis.edu/2007_JAppEmet_MS_reserves.pdf},
	keywords={energy},
	abstract={We test the efficiency of the California electricity reserves market by examining systematic differences between its day‐ and hour‐ahead prices. We uncover significant day‐ahead premia, which we attribute to market design characteristics. On the demand side, the market design established a principal–agent relationship between the markets' buyers (principal) and their supervisory authority (agent). The agent had very limited incentives to shift reserve purchases to the lower priced hour‐ahead markets. On the supply side, the market design raised substantial entry barriers by precluding purely speculative trading and by introducing a complicated code of conduct that induced uncertainty about which actions were subject to regulatory scrutiny. We use a high‐dimensional vector moving average model to estimate the premia and conduct correct inferences. To obtain exact maximum likelihood estimates of the model, we develop a new EM algorithm that seamlessly incorporates missing data and applies directly to general moving average time series models. Our algorithm uses only analytical expressions: the Kalman filter and a fixed interval smoother in the E step and least squares‐type regressions in the M step. },
  publisher={Wiley Online Library}
}

We test the efficiency of the California electricity reserves market by examining systematic differences between its day‐ and hour‐ahead prices. We uncover significant day‐ahead premia, which we attribute to market design characteristics. On the demand side, the market design established a principal–agent relationship between the markets' buyers (principal) and their supervisory authority (agent). The agent had very limited incentives to shift reserve purchases to the lower priced hour‐ahead markets. On the supply side, the market design raised substantial entry barriers by precluding purely speculative trading and by introducing a complicated code of conduct that induced uncertainty about which actions were subject to regulatory scrutiny. We use a high‐dimensional vector moving average model to estimate the premia and conduct correct inferences. To obtain exact maximum likelihood estimates of the model, we develop a new EM algorithm that seamlessly incorporates missing data and applies directly to general moving average time series models. Our algorithm uses only analytical expressions: the Kalman filter and a fixed interval smoother in the E step and least squares‐type regressions in the M step.

@article{suenaga2011volatility,
  title={Volatility Dynamics and Seasonality in Energy Prices: Implications for Crack-Spread Price Risk},
  author={Suenaga, Hiroaki and Smith, Aaron},
  journal={The Energy Journal},
  volume={32},
  number={3},
  pages={27--58},
  year={2011},
	url={https://files.asmith.ucdavis.edu/2011_EnJnl_SS_crack.pdf},
	keywords={energy},
	abstract={We examine the volatility dynamics of three major petroleum commodities traded on the NYMEX: crude oil, unleaded gasoline, and heating oil. Using the partially overlapping time-series (POTS) framework of Smith (2005), we model jointly all futures contracts with delivery dates up to a year into the future and extract information from these prices about the persistence of market shocks. The model depicts highly nonlinear volatility dynamics that are consistent with the observed seasonality in demand and storage of the three commodities. Specifically, volatility of the three commodity prices exhibits time-to-delivery effects and substantial seasonality, yet their patterns vary systematically by contract delivery month. The conditional variance and correlation across the three commodities also vary over time. High price volatility of near-delivery contracts and their low correlation with concurrently traded distant contracts imply high short-horizon price risk for an unhedged position in the calendar or crack spread. Price risk at the one-year horizon is much lower than short-horizon risk in all seasons and for all positions, but it is still substantial in magnitude for crack-spread positions. Crack-spread hedgers ignore nearby high-season price risk at their peril, but they would also be remiss to ignore the long horizon.},
  publisher={International Association for Energy Economics}
}

We examine the volatility dynamics of three major petroleum commodities traded on the NYMEX: crude oil, unleaded gasoline, and heating oil. Using the partially overlapping time-series (POTS) framework of Smith (2005), we model jointly all futures contracts with delivery dates up to a year into the future and extract information from these prices about the persistence of market shocks. The model depicts highly nonlinear volatility dynamics that are consistent with the observed seasonality in demand and storage of the three commodities. Specifically, volatility of the three commodity prices exhibits time-to-delivery effects and substantial seasonality, yet their patterns vary systematically by contract delivery month. The conditional variance and correlation across the three commodities also vary over time. High price volatility of near-delivery contracts and their low correlation with concurrently traded distant contracts imply high short-horizon price risk for an unhedged position in the calendar or crack spread. Price risk at the one-year horizon is much lower than short-horizon risk in all seasons and for all positions, but it is still substantial in magnitude for crack-spread positions. Crack-spread hedgers ignore nearby high-season price risk at their peril, but they would also be remiss to ignore the long horizon.

@article{merel2014cars,
  title={Cars on Crutches: How Much Abatement do Smog Check Repairs Actually Provide?},
  author={Merel, Pierre and Smith, Aaron and Williams, Jeffrey and Wimberger, Emily},
  journal={Journal of Environmental Economics and Management},
  volume={67},
  number={3},
  pages={371--395},
  year={2014},
	url={https://files.asmith.ucdavis.edu/2014_JEEM_MSWW_crutches.pdf},
	keywords={energy},
	abstract={Not as much abatement as has been presumed. Smog check programs aim to curb tailpipe emissions from in-use vehicles by requiring repairs whenever emissions, measured at regular time intervals, exceed a certain threshold. Using data from California, we estimate that on average 41\% of the initial emissions abatement from repairs is lost by the time of the subsequent inspection, normally two years later. Our estimates imply that the cost per pound of pollution avoided is an order of magnitude greater for smog check repairs than alternative policies such as new-vehicle standards or emissions trading among industrial point sources.},
  publisher={Academic Press}
}

Not as much abatement as has been presumed. Smog check programs aim to curb tailpipe emissions from in-use vehicles by requiring repairs whenever emissions, measured at regular time intervals, exceed a certain threshold. Using data from California, we estimate that on average 41% of the initial emissions abatement from repairs is lost by the time of the subsequent inspection, normally two years later. Our estimates imply that the cost per pound of pollution avoided is an order of magnitude greater for smog check repairs than alternative policies such as new-vehicle standards or emissions trading among industrial point sources.

@article{knittel2015ethanol,
  title={Ethanol Production and Gasoline Prices: A Spurious Correlation},
  author={Knittel, Christopher R and Smith, Aaron},
  journal={The Energy Journal},
  volume={36},
  number={1},
  year={2015},
	url={https://files.asmith.ucdavis.edu/2015_EnJnl_KS_spurious.pdf},
	abstract={Ethanol made from corn comprises 10\% of U.S. gasoline, up from 3\% in 2003. This dramatic increase was spurred by recent policy initiatives such as the Renewable Fuel Standard and state-level blend mandates and supported by direct subsidies such as the Volumetric Ethanol Excise Tax Credit. Some proponents of ethanol have argued that ethanol production greatly lowers gasoline prices, with one industry group claiming it reduced gasoline prices by 89 cents in 2010 and 1.09 dollars in 2011. The 2010 figure has been cited in numerous speeches by Secretary of Agriculture Thomas Vilsack. We show that these estimates were generated by implausible economic assumptions and spurious statistical correlations. To support this last point, we use the same statistical models and find that ethanol production "decreases" natural gas prices, but "increases" unemployment in both the U.S. and Europe. We even show that ethanol production "increases" the ages of our children. Overall, we see no compelling reason to believe that the effect of ethanol use on gasoline prices has been more than 10 cents per gallon.},
	keywords={energy},
  publisher={International Association for Energy Economics}
}

Ethanol made from corn comprises 10% of U.S. gasoline, up from 3% in 2003. This dramatic increase was spurred by recent policy initiatives such as the Renewable Fuel Standard and state-level blend mandates and supported by direct subsidies such as the Volumetric Ethanol Excise Tax Credit. Some proponents of ethanol have argued that ethanol production greatly lowers gasoline prices, with one industry group claiming it reduced gasoline prices by 89 cents in 2010 and 1.09 dollars in 2011. The 2010 figure has been cited in numerous speeches by Secretary of Agriculture Thomas Vilsack. We show that these estimates were generated by implausible economic assumptions and spurious statistical correlations. To support this last point, we use the same statistical models and find that ethanol production "decreases" natural gas prices, but "increases" unemployment in both the U.S. and Europe. We even show that ethanol production "increases" the ages of our children. Overall, we see no compelling reason to believe that the effect of ethanol use on gasoline prices has been more than 10 cents per gallon.

@article{lade2018policy,
  title={Policy Shocks and Market-Based Regulations: Evidence from the Renewable Fuel Standard},
  author={Lade, Gabriel E and Lin Lawell, C-Y Cynthia and Smith, Aaron},
  journal={American Journal of Agricultural Economics},
  volume={100},
  number={3},
  pages={707--731},
  year={2018},
	url={https://files.asmith.ucdavis.edu/2018_AJAE_LLS_rfspolicyshocks.pdf},
	abstract={The Renewable Fuel Standard mandates large increases in U.S. biofuel consumption and is implemented using tradable compliance credits known as RINs. In early 2013, RIN prices soared, causing the regulator to propose reducing future mandates. We estimate empirically the effect of three “policy shocks” that reduced the expected mandates in 2013. We find that the largest of these shocks decreased the value of the fuel industry’s 2013 compliance obligation by 7 billion dollars. We then study the effects of the shocks on commodity markets and the market value of publicly-traded biofuel firms. Results show that the burden of the mandate reductions fell primarily on advanced biofuel firms and commodity markets of the marginal compliance biofuel. We argue that the policy shocks reduced the incentive to invest in the technologies required to meet the future objectives of the RFS, and discuss alternative policy designs to address the problems that arose in 2013.},
	keywords={energy},
  publisher={Oxford University Press}
}

The Renewable Fuel Standard mandates large increases in U.S. biofuel consumption and is implemented using tradable compliance credits known as RINs. In early 2013, RIN prices soared, causing the regulator to propose reducing future mandates. We estimate empirically the effect of three “policy shocks” that reduced the expected mandates in 2013. We find that the largest of these shocks decreased the value of the fuel industry’s 2013 compliance obligation by 7 billion dollars. We then study the effects of the shocks on commodity markets and the market value of publicly-traded biofuel firms. Results show that the burden of the mandate reductions fell primarily on advanced biofuel firms and commodity markets of the marginal compliance biofuel. We argue that the policy shocks reduced the incentive to invest in the technologies required to meet the future objectives of the RFS, and discuss alternative policy designs to address the problems that arose in 2013.

@article{novan2018incentive,
  title={The Incentive to Overinvest in Energy Efficiency: Evidence from Hourly Smart-Meter Data},
  author={Novan, Kevin and Smith, Aaron},
  journal={Journal of the Association of Environmental and Resource Economists},
  volume={5},
  number={3},
  pages={577--605},
  year={2018},
	url={https://files.asmith.ucdavis.edu/2018_JAERE_NS_energyefficiency.pdf},
	keywords={energy},
	abstract={Many households pay a marginal price for electricity that exceeds the marginal social cost of supplying that electricity. We show evidence that such pricing schemes can create an incentive to overinvest in energy efficiency. Using hourly smart-meter data for households facing time-invariant increasing block prices, we estimate how air conditioner upgrades affect electricity use. We find that the average participating household reduces consumption by 5\%, which provides private savings in the form of lower electricity bills and social cost savings by decreasing generation and pollution costs. The private savings exceed the social savings by an average of 140\%, so the average household is faced with an incentive to overinvest in energy efficiency. This incentive to overinvest in energy efficiency would be cut in half if consumers faced any one of three alternative pricing plans with lower marginal price but the same average price.},
  publisher={University of Chicago Press Chicago, IL}
}

Many households pay a marginal price for electricity that exceeds the marginal social cost of supplying that electricity. We show evidence that such pricing schemes can create an incentive to overinvest in energy efficiency. Using hourly smart-meter data for households facing time-invariant increasing block prices, we estimate how air conditioner upgrades affect electricity use. We find that the average participating household reduces consumption by 5%, which provides private savings in the form of lower electricity bills and social cost savings by decreasing generation and pollution costs. The private savings exceed the social savings by an average of 140%, so the average household is faced with an incentive to overinvest in energy efficiency. This incentive to overinvest in energy efficiency would be cut in half if consumers faced any one of three alternative pricing plans with lower marginal price but the same average price.

@article{lade2018designing,
  title={Designing Climate Policy: Lessons from the Renewable Fuel Standard and the Blend Wall},
  author={Lade, Gabriel E and Cynthia Lin Lawell, CY and Smith, Aaron},
  journal={American Journal of Agricultural Economics},
  volume={100},
  number={2},
  pages={585--599},
  year={2018},
	url={https://files.asmith.ucdavis.edu/2018_AJAE_LLS_climatepolicy.pdf},
	keywords={energy},
	abstract={Many policies mandate renewable energy production to combat global climate change. These policies often differ significantly from first‐best policy prescriptions. Among the largest renewable energy mandates enacted to date is the Renewable Fuel Standard (RFS), which mandates biofuel consumption far beyond what is feasible with current technology and infrastructure. We critically review the methods used by the Environmental Protection Agency to project near‐ and long‐term compliance costs under the RFS, and draw lessons from the RFS experience to date that would improve the program's efficiency. The lessons are meant to inform both future RFS rulemaking and the design of future climate policies. We draw two lessons specific to the RFS. First, incorporate uncertainty into rulemaking; second, implement multi‐year rules. Multi‐year rulemaking allows for longer periods between major regulatory decisions and sends greater certainty to markets. We also provide two more general recommendations: tie waiver authority to compliance costs or include cost containment provisions, and fund research and development of new technologies directly rather than mandating them. Future technological advancement is uncertain, and mandating new technologies has proven to be largely ineffective to date, particularly in fuel markets.},
  publisher={Oxford University Press}
}

Many policies mandate renewable energy production to combat global climate change. These policies often differ significantly from first‐best policy prescriptions. Among the largest renewable energy mandates enacted to date is the Renewable Fuel Standard (RFS), which mandates biofuel consumption far beyond what is feasible with current technology and infrastructure. We critically review the methods used by the Environmental Protection Agency to project near‐ and long‐term compliance costs under the RFS, and draw lessons from the RFS experience to date that would improve the program's efficiency. The lessons are meant to inform both future RFS rulemaking and the design of future climate policies. We draw two lessons specific to the RFS. First, incorporate uncertainty into rulemaking; second, implement multi‐year rules. Multi‐year rulemaking allows for longer periods between major regulatory decisions and sends greater certainty to markets. We also provide two more general recommendations: tie waiver authority to compliance costs or include cost containment provisions, and fund research and development of new technologies directly rather than mandating them. Future technological advancement is uncertain, and mandating new technologies has proven to be largely ineffective to date, particularly in fuel markets.

@article{fishe2012identifying,
  title={Identifying Informed Traders in Futures Markets},
  author={Fishe, Raymond P.H. and Smith, Aaron},
  journal={Journal of Financial Markets},
  volume={15},
  number={3},
  pages={329--359},
  year={2012},
	url={https://files.asmith.ucdavis.edu/2012_JFinM_FS_informed.pdf},
	keywords={finance},
	abstract={We use daily positions of futures market participants to identify informed traders. These data contain 8,921 unique traders. We identify between 94 and 230 traders as overnight informed and 91 as intraday informed with little overlap. Floor brokers/traders are over-represented in the overnight informed group. The intraday informed group is dominated by managed money traders/hedge funds and swap dealers, with commercial hedgers under-represented. We find that characteristics such as experience, position size, trading activity, and type of positions held offer significant predictive power for who is informed. An analysis of daily trader profits confirms that we select highly profitable traders.},
  publisher={North-Holland}
}

We use daily positions of futures market participants to identify informed traders. These data contain 8,921 unique traders. We identify between 94 and 230 traders as overnight informed and 91 as intraday informed with little overlap. Floor brokers/traders are over-represented in the overnight informed group. The intraday informed group is dominated by managed money traders/hedge funds and swap dealers, with commercial hedgers under-represented. We find that characteristics such as experience, position size, trading activity, and type of positions held offer significant predictive power for who is informed. An analysis of daily trader profits confirms that we select highly profitable traders.

@article{metaxoglou2017state,
  title={State Prices of Conditional Quantiles: New Evidence on Time Variation in the Pricing Kernel},
  author={Metaxoglou, Konstantinos and Smith, Aaron},
  journal={Journal of Applied Econometrics},
  volume={32},
  number={1},
  pages={192--217},
  year={2017},
	url={https://files.asmith.ucdavis.edu/2017_JAppEmet_MS_SPOCQ.pdf},
	abstract={We develop a set of statistics to represent the option‐implied stochastic discount factor and we apply them to S\&P 500 returns between 1990 and 2012. Our statistics, which we call state prices of conditional quantiles (SPOCQ), estimate the market's willingness to pay for insurance against outcomes in various quantiles of the return distribution. By estimating state prices at conditional quantiles, we separate variation in the shape of the pricing kernel from variation in the probability of a particular event. Thus, without imposing strong assumptions about the distribution of returns, we obtain a novel view of pricing‐kernel dynamics. We document six features of SPOCQ for the S\&P 500. Most notably, and in contrast to recent studies, we find that the price of downside risk decreases when volatility increases. Under a standard asset pricing model, this result implies that most changes in volatility stem from fluctuations in idiosyncratic risk. Consistent with this interpretation, no known systematic risk factors such as consumer sentiment, liquidity or macroeconomic risk can account for the negative relationship between the price of downside risk and volatility.},
	keywords={finance}
}

We develop a set of statistics to represent the option‐implied stochastic discount factor and we apply them to S&P 500 returns between 1990 and 2012. Our statistics, which we call state prices of conditional quantiles (SPOCQ), estimate the market's willingness to pay for insurance against outcomes in various quantiles of the return distribution. By estimating state prices at conditional quantiles, we separate variation in the shape of the pricing kernel from variation in the probability of a particular event. Thus, without imposing strong assumptions about the distribution of returns, we obtain a novel view of pricing‐kernel dynamics. We document six features of SPOCQ for the S&P 500. Most notably, and in contrast to recent studies, we find that the price of downside risk decreases when volatility increases. Under a standard asset pricing model, this result implies that most changes in volatility stem from fluctuations in idiosyncratic risk. Consistent with this interpretation, no known systematic risk factors such as consumer sentiment, liquidity or macroeconomic risk can account for the negative relationship between the price of downside risk and volatility.

@article{chen2013nonlinear,
  title={The Nonlinear Multidimensional Relationship Between Stock Returns and the Macroeconomy},
  author={Chen, Pian and Smith, Aaron},
  journal={Applied Economics},
  volume={45},
  number={35},
  pages={4985--4999},
  year={2013},
	url={https://files.asmith.ucdavis.edu/2013_AE_CS_SIR.pdf},
	keywords={finance},
	abstract={We use nonparametric dimension-reduction methods to extract from a set of 15 macroeconomic variables the risk factors that are priced in the stock market. The dominant factor moves with the business cycle but, because it is a nonlinear function of observed macroeconomic variables, it captures a rich set of interactions. Low-credit risk and low-inflationary expectations have a greater positive effect on stock returns when leading macroeconomic indicators are high relative to current economic activity, i.e. early in the business cycle as the economy emerges from recession. High-stock returns also arise in periods when the economy is booming relative to its leading indicators, but such periods tend to portend crashes.},
  publisher={Routledge}
}

We use nonparametric dimension-reduction methods to extract from a set of 15 macroeconomic variables the risk factors that are priced in the stock market. The dominant factor moves with the business cycle but, because it is a nonlinear function of observed macroeconomic variables, it captures a rich set of interactions. Low-credit risk and low-inflationary expectations have a greater positive effect on stock returns when leading macroeconomic indicators are high relative to current economic activity, i.e. early in the business cycle as the economy emerges from recession. High-stock returns also arise in periods when the economy is booming relative to its leading indicators, but such periods tend to portend crashes.

@article{fishe2016foreign,
  title={Foreign Central Bank Activities in US Futures Markets},
  author={Fishe, Raymond PH and Robe, Michel A and Smith, Aaron},
  journal={Journal of Futures Markets},
  volume={36},
  number={1},
  pages={3--29},
	url={https://files.asmith.ucdavis.edu/2016_JFutMkt_FRS_centralbanks.pdf},
	abstract={We analyze the daily positions of 31 foreign Central Banks in US interest rate futures markets between 2003 and 2011 for targeted hedging or informed profit‐making decisions. Central Bank positions before the financial crisis of 2007–2009 are consistent with hedging some underlying balance sheet exposure. During and after the crisis, the pattern suggests an attempt to enhance returns. In particular, Central Banks held and profited from directional positions in 5‐ and 10‐year T‐Note futures in a manner indicative of a non‐hedging strategy. We also examine whether Central Bank position changes are synchronized in the sense that they tend to occur simultaneously. We identify differences before and after the onset of the financial crisis: Euro‐linked Central Banks become more synchronized, whereas non‐European Central Banks show no significant change during the crisis. We document that Central Bank positions generally account for a small fraction of the overall size of the futures markets, so it is unlikely that these institutions' goal is to influence US interest rates.},
	keywords={finance},
  year={2016}
}

We analyze the daily positions of 31 foreign Central Banks in US interest rate futures markets between 2003 and 2011 for targeted hedging or informed profit‐making decisions. Central Bank positions before the financial crisis of 2007–2009 are consistent with hedging some underlying balance sheet exposure. During and after the crisis, the pattern suggests an attempt to enhance returns. In particular, Central Banks held and profited from directional positions in 5‐ and 10‐year T‐Note futures in a manner indicative of a non‐hedging strategy. We also examine whether Central Bank position changes are synchronized in the sense that they tend to occur simultaneously. We identify differences before and after the onset of the financial crisis: Euro‐linked Central Banks become more synchronized, whereas non‐European Central Banks show no significant change during the crisis. We document that Central Bank positions generally account for a small fraction of the overall size of the futures markets, so it is unlikely that these institutions' goal is to influence US interest rates.

@article{metaxoglou2016option,
  title={Option-Implied Equity Premium Predictions via Entropic Tilting},
  author={Metaxoglou, Konstantinos and Pettenuzzo, Davide and Smith, Aaron},
  journal={Journal of Financial Econometrics},
  volume={17},
  number={4},
  pages={559--586},
  year={2019},
	url={https://files.asmith.ucdavis.edu/2019_JFinEc_MPS_tilting.pdf},
	keywords={finance},
	abstract={We propose a new method to improve density forecasts of the equity premium using information from options markets. We obtain predictive densities from stochastic volatility (SV) and GARCH models, which we then tilt using the second moment of the risk-neutral distribution implied by options prices while imposing a non-negativity constraint on the equity premium. By combining the backward-looking information contained in the GARCH and SV models with the forward-looking information from options prices, our procedure improves the performance of predictive densities. Using density forecasts of the U.S. equity premium from January 1990 to December 2014, we find that tilting leads to more accurate predictions using statistical and economic criteria.},
  publisher={Oxford University Press}
}

We propose a new method to improve density forecasts of the equity premium using information from options markets. We obtain predictive densities from stochastic volatility (SV) and GARCH models, which we then tilt using the second moment of the risk-neutral distribution implied by options prices while imposing a non-negativity constraint on the equity premium. By combining the backward-looking information contained in the GARCH and SV models with the forward-looking information from options prices, our procedure improves the performance of predictive densities. Using density forecasts of the U.S. equity premium from January 1990 to December 2014, we find that tilting leads to more accurate predictions using statistical and economic criteria.

@article{metaxoglou2017forecasting,
  title={Forecasting Stock Returns Using Option-Implied State Prices},
  author={Metaxoglou, Konstantinos and Smith, Aaron},
  journal={Journal of Financial Econometrics},
  volume={15},
  number={3},
  pages={427--473},
  year={2017},
	url={https://files.asmith.ucdavis.edu/2017_JFinEmet_MS_forecast.pdf},
	keywords={finance},
	abstract={Options prices embed the risk preferences that determine expected returns in asset pricing models. Therefore, functions of options prices should predict returns. In this paper, we show that the State Prices of Conditional Quantiles (SPOCQ)—functions of options prices introduced in Metaxoglou and Smith (2016)—exhibit strong predictive ability for the U.S. equity premium. These SPOCQ series provide estimates of the market’s willingness to pay for insurance against outcomes in various quantiles of the return distribution. They also relate to expected returns in prominent asset pricing models. Our SPOCQ series that captures relative risk aversion exhibits strong predictive ability for S\&P 500 returns at horizons between 6 and 18 months, both in the full sample, 1990–2012, and out of sample. Our SPOCQ series that captures volatility aversion, however, exhibits no predictive ability due to the lack of skewness in the return distribution for the horizons considered.},
  publisher={Oxford University Press}
}

Options prices embed the risk preferences that determine expected returns in asset pricing models. Therefore, functions of options prices should predict returns. In this paper, we show that the State Prices of Conditional Quantiles (SPOCQ)—functions of options prices introduced in Metaxoglou and Smith (2016)—exhibit strong predictive ability for the U.S. equity premium. These SPOCQ series provide estimates of the market’s willingness to pay for insurance against outcomes in various quantiles of the return distribution. They also relate to expected returns in prominent asset pricing models. Our SPOCQ series that captures relative risk aversion exhibits strong predictive ability for S&P 500 returns at horizons between 6 and 18 months, both in the full sample, 1990–2012, and out of sample. Our SPOCQ series that captures volatility aversion, however, exhibits no predictive ability due to the lack of skewness in the return distribution for the horizons considered.

@article{boucher2007impacts,
  title={Impacts of Policy Reforms on the Supply of Mexican Labor to US Farms: New Evidence from Mexico},
  author={Boucher, Stephen R and Smith, Aaron and Taylor, J Edward and Y{\'u}nez-Naude, Antonio},
  journal={Review of Agricultural Economics},
  volume={29},
  number={1},
  pages={4--16},
  year={2007},
	url={https://files.asmith.ucdavis.edu/2007_RAE_BSTY_labor.pdf},
	keywords={other},
	abstract={The availability of immigrant farm-workers from Mexico is a critical factor affecting the U.S. fresh fruit and vegetable sector. This paper uses retrospective panel data from rural Mexico to examine the impact of the North American Free Trade Agreement and the Immigration Reform and Control Act on the supply of migrant labor to the United States. We find that, in contrast to expectations, both policies were associated with an increase in migration to U.S. farm jobs from rural Mexico.},
  publisher={Oxford University Press}
}

The availability of immigrant farm-workers from Mexico is a critical factor affecting the U.S. fresh fruit and vegetable sector. This paper uses retrospective panel data from rural Mexico to examine the impact of the North American Free Trade Agreement and the Immigration Reform and Control Act on the supply of migrant labor to the United States. We find that, in contrast to expectations, both policies were associated with an increase in migration to U.S. farm jobs from rural Mexico.

@article{lybbert2012does,
  title={Does the Red Flag Rule Induce Risk Taking in Sprint Finishes? Moral Hazard Crashes in Cycling's Grand Tours},
  author={Lybbert, Travis J and Lybbert, Troy C and Smith, Aaron and Warren, Scott},
  journal={Journal of Sports Economics},
  volume={13},
  number={6},
  pages={603--618},
  year={2012},
	url={https://files.asmith.ucdavis.edu/2012_JSE_LLSW_redflag.pdf},
	keywords={other},
	abstract={Sprint finishes in professional cycling are fast, furious, and dangerous. A “red flag rule” (RFR) seeks to moderate the chaos of these finishes, but may induce moral hazard by removing the time penalty associated with crashing. To test for moral hazard, the authors use a 2005 rule change that moved the red flag from 1 km to 3 km from the finish. Data from Europe’s Grand Tours indicate that, after the rule change, both the incidence and the size of crashes nearly doubled in the 1–3 km from the finish zone. There was no such increase in crashing rates in the 3–5 km zone.},
  publisher={Sage Publications Sage CA: Los Angeles, CA}
}

Sprint finishes in professional cycling are fast, furious, and dangerous. A “red flag rule” (RFR) seeks to moderate the chaos of these finishes, but may induce moral hazard by removing the time penalty associated with crashing. To test for moral hazard, the authors use a 2005 rule change that moved the red flag from 1 km to 3 km from the finish. Data from Europe’s Grand Tours indicate that, after the rule change, both the incidence and the size of crashes nearly doubled in the 1–3 km from the finish zone. There was no such increase in crashing rates in the 3–5 km zone.

@article{taylor2012immigration,
  title={Immigration and the US Farm Labour Supply},
  author={Taylor, J Edward and Boucher, Stephen R and Smith, Aaron and Fletcher, Peri L and Y{\'u}nez-Naude, Antonio},
  journal={Migration Letters},
  volume={9},
  number={1},
  pages={87--99},
  year={2012},
	url={https://files.asmith.ucdavis.edu/2012_ML_TBSFY_migration.pdf},
	abstract={This paper uses unique data from rural Mexico to examine the supply of immigrant hired labour to US farms. Econometric evidence indicates that immigration policy reforms had unintended consequences for farm labour supply. The long-term trend in migration from rural Mexico to US farms is decreasing, and in recent years, US farms have drawn more labour from remote and less developed areas of rural Mexico. Other high income countries, as well as some developing nations, mirror the US in reliance on foreign agricultural workers. Our analysis questions the sustainability of an agricultural system that depends on foreign sources of labour, and highlights the importance of labour productivity-enhancing technological change.},
	keywords={other}
}

This paper uses unique data from rural Mexico to examine the supply of immigrant hired labour to US farms. Econometric evidence indicates that immigration policy reforms had unintended consequences for farm labour supply. The long-term trend in migration from rural Mexico to US farms is decreasing, and in recent years, US farms have drawn more labour from remote and less developed areas of rural Mexico. Other high income countries, as well as some developing nations, mirror the US in reliance on foreign agricultural workers. Our analysis questions the sustainability of an agricultural system that depends on foreign sources of labour, and highlights the importance of labour productivity-enhancing technological change.

@article{macewan2016supplemental,
  title={The Supplemental Nutrition Assistance Program, Energy Balance, and Weight Gain},
  author={MacEwan, Joanna P and Smith, Aaron and Alston, Julian M.},
  journal={Food Policy},
  volume={61},
  number={},
  pages={103-120},
  year={2016},
	url={https://files.asmith.ucdavis.edu/2016_FP_MSA_SNAP.pdf},
	keywords={other},
	abstract={Between 2001 and 2006, women participating in the U.S. Supplemental Nutrition Assistance Program (SNAP) weighed 15.5 lb more than eligible nonparticipants on average. Using a dataset (NHANES) that contains detailed information on a wide variety of demographic, socioeconomic, health and behavioral characteristics for 2018 SNAP-eligible women, we make three contributions to the growing literature that addresses this weight difference. First, we specify a physiological model of weight gain with which we show that average differences in caloric intake and physical activity are much too small to explain the weight difference. Second, we estimate regression models that show SNAP participation is not associated with changes in energy balance (i.e., caloric intake and physical activity). Third, we estimate regression models that show SNAP participation does not predict significant weight gain. Our results provide no support for the hypothesis that SNAP participation causes weight gain.},
  publisher={Elsevier}
}

Between 2001 and 2006, women participating in the U.S. Supplemental Nutrition Assistance Program (SNAP) weighed 15.5 lb more than eligible nonparticipants on average. Using a dataset (NHANES) that contains detailed information on a wide variety of demographic, socioeconomic, health and behavioral characteristics for 2018 SNAP-eligible women, we make three contributions to the growing literature that addresses this weight difference. First, we specify a physiological model of weight gain with which we show that average differences in caloric intake and physical activity are much too small to explain the weight difference. Second, we estimate regression models that show SNAP participation is not associated with changes in energy balance (i.e., caloric intake and physical activity). Third, we estimate regression models that show SNAP participation does not predict significant weight gain. Our results provide no support for the hypothesis that SNAP participation causes weight gain.