Is there a Term Structure in Land Lease Rates? Martin Odening Silke - PowerPoint PPT Presentation

Motivation Theoretical Background Empirical Analysis Conclusion Is there a Term Structure in Land Lease Rates? Martin Odening Silke H¨ uttel Matthias Ritter Viacheslav Esaulov Humboldt-Universit¨ at zu Berlin Energy Finance Workshop May 9, 2014 Anhang Martin Odening Is there a Term Structure in Land Lease Rates? 1/12

Motivation Theoretical Background Empirical Analysis Conclusion Motivation Land rental markets Land is an important factor for agricultural production Most farms gain access to land via rental markets Land rental prices are used to determine farmland values Unclear influence of the length of rental contract ( term ) Objectives Explore the role of the term Detect farmer’s implied expectation about future price development Anhang Martin Odening Is there a Term Structure in Land Lease Rates? 2/12

Motivation Theoretical Background Empirical Analysis Conclusion Theoretical Background Upward−sloping What is a term structure? Known from interest rates price Influence of length of contract Equilibrium: term Value of long-term contract Downward−sloping = value of several short-term contracts price Different possible shapes term Literature Single−humped Gunnelin/S¨ oderberg (2002): Term structures in Office Rental Market price Stanton/Wallace (2009): Theoretical derivation term Anhang Martin Odening Is there a Term Structure in Land Lease Rates? 3/12

Motivation Data Theoretical Background Descriptive Statistics Empirical Analysis Empirical Model Conclusion Results Data Land rental market in Saxony-Anhalt History of expropriation, land collectivization and socialist policy Privatization of state-owned land after German reunification 1990 Bodenverwertungs- und -verwaltungs GmbH (BVVG) Average share of rental land per farm in 2010: 77 % Fig. 1: The Federal state of Saxony-Anhalt in Germany Anhang Martin Odening Is there a Term Structure in Land Lease Rates? 4/12

Motivation Data Theoretical Background Descriptive Statistics Empirical Analysis Empirical Model Conclusion Results Data Rental prices for Saxony-Anhalt from BVVG Price and plot size (arable/grass/other), soil quality, location on county-level, length of contract Contracts running between 2006 and 2010 Data selection: contracts starting between 2002 and 2010 including arable land contracts starting in October outliers removed 2,504 observations Fig. 2: Average BVVG rent prices 2010 by county in Saxony-Anhalt Anhang Martin Odening Is there a Term Structure in Land Lease Rates? 5/12

Motivation Data Theoretical Background Descriptive Statistics Empirical Analysis Empirical Model Conclusion Results Descriptive Statistics Variable Mean Stand. Dev. Min. Max. Price arable land ( e per ha) 318.66 167.2 54 1075 Plot size (ha) 17.44 31.1 0.02 554 Arable land (ha) 14.14 25.7 0.001 364.65 Grassland (ha) 2.41 7.4 0 124.96 Other land (ha) 0.90 4.3 0 190.75 Share of arable land (%) 80.92 26.5 0.17 100 Soil quality (points [0,102]) 61.34 21.0 9 100 Length of contract (years) 3.43 2.1 0 18 Table 1: Descriptive statistics BVVG data 2002–2010 ( N = 2 , 504) Anhang Martin Odening Is there a Term Structure in Land Lease Rates? 6/12

Motivation Data Theoretical Background Descriptive Statistics Empirical Analysis Empirical Model Conclusion Results Empirical Model � 2 log( R arable ) = a 0 + a 1 A arable A arable � scaled + a 3 log � Q arable � + a 4 log � S arable � + a 2 i i i i i 11 � b c County ci + a 6 log � ¯ R lagged, county � + a 5 Trend ( i ) + i c =2 2010 2010 � � d t · Year ti · Term 2 + c t · Year ti · Term i + i + e i t =2002 t =2002 R arable – Price arable land ( e /ha) County ci – County dummies i A arable – Lot size in ha Year t – Year dummies i Q arable – Soil quality in points Term i – Length of the contract i S arable – Share of arable land in ha i – Contract index i Trend ( i ) – Yearly time trend e i – Error term R lagged, county ¯ – Average county price of i previous year Results Anhang Martin Odening Is there a Term Structure in Land Lease Rates? 7/12

Motivation Data Theoretical Background Descriptive Statistics Empirical Analysis Empirical Model Conclusion Results Regression results Land lease rate Estimated P-value Land lease rate Estimated P-value determinant Coeff. determinant Coeff. Size arable 0.002 0.000*** Location land (ha) dummies Size arable -0.011 0.001*** AMK Salzwedel – – land squared Anhalt- -0.007 0.822 (scaled by Bitterfeld county) Burgenlandkreis -0.007 0.857 Share arable 0.050 0.000*** B¨ orde 0.033 0.272 land (%, log) Harz 0.039 0.183 Soil quality 0.908 0.000*** Jerichower 0.076 0.009*** (log) Land Time trend 0.085 0.000*** Mansfeld- -0.134 0.000*** Lagged 0.056 0.047** S¨ udharz regional rental Saalekreis -0.156 0.000*** price (log) Salzlandkreis -0.135 0.000*** Stendal 0.050 0.046** Wittenberg -0.098 0.002*** Table 2: Results from the regression Note: *** and ** denote significance at the 1 and 5 per cent levels, respectively Anhang Martin Odening Is there a Term Structure in Land Lease Rates? 8/12

Motivation Data Theoretical Background Descriptive Statistics Empirical Analysis Empirical Model Conclusion Results Regression results (cont.) Land lease rate Estimated P-value Land lease rate Estimated P-value determinant Coeff. determinant Coeff. Term Structure D 2002 Term 2 D 2002 Term 0.062 0.009*** -0.002 0.164 D 2003 Term 2 D 2003 Term 0.067 0.005*** -0.003 0.156 D 2004 Term 2 D 2004 Term 0.016 0.497 -0.001 0.814 D 2005 Term 2 D 2005 Term 0.025 0.319 -0.001 0.785 D 2006 Term 2 D 2006 Term -0.029 0.134 0.004 0.175 D 2007 Term 2 D 2007 Term 0.071 0.000*** -0.003 0.481 D 2008 Term 2 D 2008 Term 0.148 0.000*** -0.012 0.000*** D 2009 Term 2 D 2009 Term 0.117 0.000*** -0.011 0.000*** D 2010 Term 2 D 2010 Term 0.083 0.000*** -0.008 0.000*** Constant 1.017 0.000*** Table 2 (cont.): Results from the regression Note: *** and ** denote significance at the 1 and 5 per cent levels, respectively Anhang Martin Odening Is there a Term Structure in Land Lease Rates? 9/12

Motivation Data Theoretical Background Descriptive Statistics Empirical Analysis Empirical Model Conclusion Results Term structure Year p -values Wald test Term Structure ( c t = d t = 0) 2002 0.005*** upward sloping 2003 0.001*** upward sloping 2004 0.410 n.s. 2005 0.328 n.s. 2006 0.325 n.s. 2007 0.000*** upward sloping 2008 0.000*** single-humped 2009 0.000*** single-humped 2010 0.000*** single-humped Table 3: Term Structure of Land Rental Prices Note: *** denotes significance at the 1 per cent level, n.s. = not significant Anhang Martin Odening Is there a Term Structure in Land Lease Rates? 10/12

Motivation Data Theoretical Background Descriptive Statistics Empirical Analysis Empirical Model Conclusion Results Average effect of the contract length 600 500 rent based on mean values 400 300 200 100 0 2002 2004 2006 2008 2010 2012 2014 2016 2018 2020 2022 years Term Structure 2002 Equation Fig. 3: Estimated significant term structures Anhang Martin Odening Is there a Term Structure in Land Lease Rates? 11/12

Motivation Data Theoretical Background Descriptive Statistics Empirical Analysis Empirical Model Conclusion Results Average effect of the contract length 600 500 rent based on mean values 400 300 200 100 0 2002 2004 2006 2008 2010 2012 2014 2016 2018 2020 2022 years Term Structure 2002 Term Structure 2003 Term Structure 2007 Equation Fig. 3: Estimated significant term structures Anhang Martin Odening Is there a Term Structure in Land Lease Rates? 11/12

Motivation Data Theoretical Background Descriptive Statistics Empirical Analysis Empirical Model Conclusion Results Average effect of the contract length 600 500 rent based on mean values 400 300 200 100 0 2002 2004 2006 2008 2010 2012 2014 2016 2018 2020 2022 years Term Structure 2002 Term Structure 2003 Term Structure 2007 Term Structure 2008 Equation Fig. 3: Estimated significant term structures Anhang Martin Odening Is there a Term Structure in Land Lease Rates? 11/12

Motivation Data Theoretical Background Descriptive Statistics Empirical Analysis Empirical Model Conclusion Results Average effect of the contract length 600 500 rent based on mean values 400 300 200 100 0 2002 2004 2006 2008 2010 2012 2014 2016 2018 2020 2022 years Term Structure 2002 Term Structure 2003 Term Structure 2007 Term Structure 2008 Term Structure 2009 Term Structure 2010 Equation Fig. 3: Estimated significant term structures Anhang Martin Odening Is there a Term Structure in Land Lease Rates? 11/12

Motivation Theoretical Background Empirical Analysis Conclusion Conclusion Summary Different shapes of term structure detected Upward-sloping term structure in 2002, 2003, 2007: Farmers expect growing rental prices Single-humped term structure after 2008: Market is expected to cool down Discussion Implications for agricultural policies and structural change in agriculture Term has to be considered when predicting future price development Careful with generalization because of illiquid market and local peculiarities Anhang Martin Odening Is there a Term Structure in Land Lease Rates? 12/12