Humboldt-Universität zu Berlin - Department of Agricultural Economics

Current Research Projects

more research projects: research data base of Humboldt-Universität zu Berlin

Price Formation and Agricultural Land Markets

Land is a crucial production factor in agriculture. In developed countries this input factor is usually in short supply and its overall availability shrinks permanently. On the contrary, the increasing demand of growing farms and more recently of non-agricultural investors causes price pressure on the land market. Thus, it is not surprising that land prices increased in recent years. Though the analysis of land markets is a core topic in agricultural economics, many questions are still unanswered. For example, do land prices reflect the price boom for agricultural commodities and bio-energy? Can land prices be fully explained by fundamental factors or are they also driven by speculative bubbles? What is the role of non-agricultural investors? Should land markets be regulated and if so, what are the most efficient instruments?

Against this background the project aims at understanding the recent developments on land markets, particularly in Germany and the EU. The focus will be on quantitative modelling and empirical (econometric) analyses.

Contact: Prof. Dr. Martin Odening, m.odening(at)
Cooperation: Prof. Dr. Silke Hüttel, S.Huettel(at), Universität Bonn

Index-based insurance and data scarcity

Most markets are exposed to risk. Agricultural production is particularly affected because many crops are produced on arable land, which is susceptible and exposed to different types of risk. This makes the production very vulnerable and highly dependent on variables such as the weather and other natural factors. Most of the world’s poorest populations live in rural areas where they rely heavily on agriculture. These groups are also considered the most vulnerable to risk. At the same time, these groups have the lowest protection against risk. Although these populations would profit immensely from risk management strategies, such as insurance, the data situation often prevents insurance companies from developing and offering crop insurances. Especially developing countries are prone to lacking data, which plays a central role in the permeation and design of agricultural insurance. Low quantity and quality data potentially lead to ineffective and expensive insurance products. Hence, the objective of this study is to employ econometric tools and strategies for dealing with data scarcity against the background of agricultural insurance.

Contact: Katarina von Witzke, witzkevk(at)

Assessment of compensation options under quarantine pests occurrence

Climate change and international trade favor the introduction of harmful organisms that have not been established in a country yet. If such organisms have the potential to cause high economic or ecological damage, the European Commission classifies them as quarantine pests and takes them under official control. The occurrence of quarantine pests entails considerable costs due to obligatory measures that intend to prevent further spread, reduce and/or eradicate harmful organisms. These measures include, for example, the disposal of the infected plant material and/or restriction on planting an infected field.

The aim of this interdisciplinary research project is to assess compensation options for agricultural and horticultural farms. This will be exemplified for the following organisms: Xylella fastidiosa, Flavecence Dorée, Anoplophora chinensis, Tomato brown rugose fruit virus, Thrips palmi, Clavibacter michiganensis sp. Sepedonicus, Synchytrium endobioticum. The following compensation options will be considered and evaluated: insurance opportunities, mutual fund schemes, private organized funding structures and other compensation forms. Financial feasibility of compensation measures will take into account public support at the national and international level.

Contact: Anna Filiptseva, anna.filiptseva(at)
Cooperation: Prof. Dr. Carmen Büttner, carmen.buettner(at)

Liquidity on the agricultural land market

Liquidity constitutes a decisive factor for the efficient functioning of a market. It describes the ability of market participants to realize desired buy or sell transactions without a time delay. Poor liquidity bears the risk of either paying additional premia on top of a “fundamental” value or losing money if an immediate transaction shall be enforced. The determinants and impact of liquidity are well-explored in financial markets and real estate markets, both theoretically and empirically. When analyzing liquidity on agricultural land markets, one has to consider the special characteristics of farmland: It is limited,immobile, extremely heterogeneous, and in short supply. Since it is not traded on exchanges, common liquidity measures as the bid-ask spread cannot be derived for farmland markets. Moreover, there is no counterpart for exchange traded Real Estate Investments Trusts (REITs) on agricultural land markets. Further, there is little knowledge about the relationship between market liquidity and prices on land markets. Itis even unclear if this relationship is positive or negative. Thus, the objectives are to measure market liquidity on land markets with different indicators and methods as well as to explore its relationship with prices.

Contact: Marlene Kionka, marlene.kionka(at)

Machine Learning in Agricultural Risk Management and Agricultural Land Markets - Prediction, Classification and Causal Inference

Machine learning offers a great variability of tools and techniques for various tasks and allows with the flexible structure of most of the machine learning models an adaption to specific problems. Especially when it comes to prediction machine learning techniques in particular supervised machine learning techniques can provide an enormous improvement in comparison to classical statistical approaches. With higher computational power and the higher availability of large data sets, applications of machine learning models are rising. The thematic focus in agriculture is often on plant sciences or animal sciences. In the field of agricultural economics, the application of such methods is not as widespread as it is in other research areas and applications that are going beyond the basic prediction and classification tasks are rare, especially in agricultural risk management and agricultural land markets. Therefore, this research project aims to develop and apply machine learning models to these specific topics to address the existing shortcomings of traditional statistical approaches.

Contact: Lorenz Schmidt,