Comparison of alternative sources of farmland values

Abstract

Purpose Consistent and reliable data on farmland values is critical to assessing the overall financial health of agricultural producers. However, little is known about the idiosyncrasies and similarities of standard land value data sources – US Department of Agriculture (USDA), Federal Reserve Bank land value surveys, and transaction prices. The purpose of this paper is to determine the differences and similarities of land value movements from three land value data sources. Design/methodology/approach In addition to Oklahoma transaction prices, two survey sources are considered: the USDA annual report and the quarterly Tenth District Survey of Agricultural Credit Conditions administered by the Federal Reserve Bank of Kansas City. The paper describes each data set and identifies differences in data sampling, collection, and reporting. Average values of Oklahoma farmland across data sources are examined. USDA estimates are regressed against quarterly Federal Reserve values across multiple states to determine the point in time represented by USDA estimates. Granger causality tests determine if Federal Reserve land value estimates anticipate movements in USDA land value estimates. Findings It is found that all three data sources are highly correlated, but transaction prices tend to be higher, especially for irrigated cropland and ranchland. USDA land values are reported as representing land values on January first, but instead they more closely represent first and second quarter land values according to a multi‐state comparison to changes in quarterly Federal Reserve land values. Given the finding that first quarter Federal Reserve Bank land values lead USDA land values and that they are published before the USDA release, Federal Reserve land values are a timely indicator of agricultural producers' financial position. Originality/value No previous research has addressed the topic of how various sources of agricultural land values compare.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 15 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.