Agricultural Land Price Convergence: Evidence from Polish Provinces

Abstract

This research deals with the problem of agricultural land market efficiency using the spatial market integration concept as well as the present value (PV) model. Empirically, it aims to test the convergence of agricultural land prices across Polish provinces. In order to check the law of one price (LOP), good-quality, medium-quality and bad-quality land sales markets are examined separately. Furthermore, this study is complemented by an analysis of the drivers behind agricultural land price convergence. The main method of testing price convergence is the log t regression. The latter was performed in two configurations, i.e., based on trend components of time series extracted using the Hodrick–Prescott filter and the Hamilton filter. Additionally, traditional β- and σ-convergence tests were applied. The obtained results indicated that agricultural land prices tend to converge in relative terms, which means that the provinces share a common long-run growth path. This finding and estimates of traditional convergence tests prove the increasing integration in the agricultural land market in Poland. There is no evidence, however, to support the conclusion that the absolute version of the long-run LOP holds. Moreover, using dynamic fixed effects models, it was identified that for good-, medium- and bad-quality land prices almost the same drivers of convergence apply. The only differences concern the strength of the influence of independent variables on prices of farmland of various types. Additionally, bad-quality land prices are the only ones which are affected by livestock density. Furthermore, estimates of the present value model finally confirmed that the agricultural land sales market in Poland cannot be considered as efficient.