Research

We measure U.S. agricultural supply news using the reaction of field-crop futures prices to scheduled USDA reports, generating a monthly, expectations-based instrument dated to the moment belief change. In an external-instrument SVAR, adverse harvest news raises crop and food prices, core CPI, and inflation expectations while lowering real GDP and consumption — effects larger than agriculture's small share of output implies. In common-sample comparisons, only the USDA instrument recovers this full pattern; this result survives residualizing the USDA series on realized-harvest and narrative shock benchmarks, and observed weather measures explain very little of the series. It also improves real-time forecasts of crop and food prices and, at short horizons, of output and consumption.

Drought-induced low water on the Mississippi River in fall 2022 disrupted barge shipments from interior grain regions to Gulf export terminals. Using the relative price of a substitute (RPS) method on daily cash bids from 68 U.S. counties and the New Orleans export hub, we estimate the economic incidence of the disruption on producers and downstream purchasers. Producer prices fell by $0.54/bu (95% CI:-$0.62, -$0.44) for corn and $1.02/bu (95% CI: -$1.51, -$0.51) for soybeans, while hub prices rose by $0.26 per bushel (95% CI: $0.12, $0.39) for corn and $0.06 (CI: -$0.10, $0.22) for soybeans. Translated into aggregate dollar terms, the disruption reduced producer revenues by around $134 million for corn and $414 million for soybeans, against procurement-cost increases of $64 million for corn purchasers and $26 million for soybean purchasers. The asymmetric incidence, particularly stark for soybeans, is consistent with relatively inelastic U.S. supply meeting elastic foreign demand. Local heterogeneity is substantial. Counties near ethanol plants (corn) or crushing facilities (soybeans) experienced 25-46 percent smaller producer losses through access to alternative domestic outlets, while counties near barge-loading terminals or Gulf ports, those most dependent on the river export route, bore proportionally larger losses. 

Identifying the causal effect of agricultural production conditions on food prices is challenging because commodity prices respond simultaneously to supply and demand shocks. Exploiting high-frequency field crop futures prices and USDA crop reports, we estimate impacts of agricultural supply news shocks on food markets. A poor harvest news shock that raises field crop prices leads to an increase in food-at-home and food service prices, a decline in food expenditures. Using household-level food price indexes constructed across income quintiles, we find that welfare losses are approximately seven times larger for the lowest-income quintile than for the highest. Counterfactual simulations indicate that, without these shocks, U.S. food inflation during 2021-23 would have peaked up to 1.9 percentage points lower.

We study how U.S. biofuel policy affects food markets and consumer prices. Drawing on the institutional structure of U.S. biofuel policy and high-frequency price movements, we isolate biofuel policy shocks and estimate their effects on upstream commodity markets and downstream retail prices. A stringent biofuel policy shock generates statistically significant inflation across upstream and downstream markets, raising ethanol, oil, and field crop prices, as well as consumer food prices and the CPI. Real food expenditures decline in response to the shock. Weaker statistical evidence indicates that the biofuel policy shock may also raise fats and oils prices more than those of other food subcategories and that retail fuel and gasoline prices increase. Consumer inflation expectations rise following the shock. The welfare burden fails disproportionately on lower-income households—nearly five times larger than the highest income group—reflecting their disproportionately high food expenditure share. These findings reveal a fundamental distributional trade-off in biofuel policy: while higher commodity prices may support agricultural producer revenues, consumers finance these gains through higher retail prices for fuel and food, with lower-income households bearing a disproportionate burden.

Public information not only reduces fundamental uncertainty in financial markets; it also coordinates dispersed private beliefs around a common reference point. We study this coordinating role in the U.S. corn futures market by developing a Bayesian learning framework in which traders form posterior price expectations as a weighted average of private priors and the public signal from USDA crop reports. Using a novel inverse quantile mapping procedure to extract price expectations from agent-level supply forecasts, we estimate that market participants place an average weight of 15 percent on USDA reports relative to their private priors. This anchoring weight is highly dynamic. Traders rely less on the public signal when USDA forecasts become noisier, and substantially more when private-analyst dispersion rises. Using Local Projections with Instrumental Variables (LP-IV), we decompose the effect of each source of uncertainty on realized volatility, and uncover a striking asymmetry. Public uncertainty raises volatility primarily by weakening the anchor. Private uncertainty, by contrast, does not raise total volatility. Its destabilizing direct effect is fully offset by the stabilizing indirect effect of stronger anchoring. The market’s apparent resilience to private uncertainty is therefore a product of public information itself.  These findings reveal a trade-off facing public agencies between maximizing forecast accuracy and preserving market stability, and show that the value of credible public statistics rises with the disorder of private information environments.

Surging food prices disproportionately burden low-income households, affect inflation expectations, and influence political outcomes. Between January 2020 and August 2025, U.S. food prices increased 30%, the steepest rise in more than four decades. We show this episode was structurally different from prior surges, including the 2008 Global Financial Crisis. Decomposing prices into supply- and demand-driven components and attributing each to macroeconomic channels, we find that labor scarcity and logistics disruptions—not commodity prices—dominated the supply side, while fiscal transfers and then wage growth account for an unprecedented demand-side contribution. Wage growth sustained elevated prices for four years through both labor costs and household income, a pattern absent in earlier episodes. Our work reveals underappreciated vulnerabilities in food-system resilience, illuminates how compound shocks generate persistent inflation, and shows why effective response must address logistics, labor, and wage pressure—not commodity availability alone.

The Southeastern region of the United States is a major livestock producer but faces a grain deficit. Analyzing transportation demand for field crops in this region is important for evaluating the factors influencing the livestock industry’s profitability. Among the available transportation modes, rail is one of the most commonly used and cost-efficient options for shipping grain from the Midwest to the Southeast. We estimate rail transportation demand for grains and oilseeds, which are primarily used as animal feed. We find that higher rail rates between a state outside the Southeast and a state within the Southeast reduce grain rail shipments to the region, although the price elasticity of rail demand is inelastic. Furthermore, increased demand for poultry feed in the southeast drives up grain rail shipments to the region.