Policy Research Working Paper 11013 Land Price Effects of Informality, Farm Size, and Land Reform Evidence from More Than One Million Transactions in Ukraine Klaus Deininger Daniel Ayalew Ali Development Economics Development Research Group December 2024 Policy Research Working Paper 11013 Abstract This paper uses a rich set of geo-coded administrative and transaction costs of selling land and accessing mortgage remotely sensed data on more than 1 million agricultural finance, improving publicity of pending land sales, and use land transactions in Ukraine to explore how informality, of electronic auctions—to enhance the reforms’ impact on size, and recent land reforms affect land prices. Three main efficiency and equity. Third, size at the parcel, field, and findings are highlighted. First, absence of registered rights farm levels is associated with higher per hectare prices, generates large negative externalities, the size of which pointing to scope for market-based land consolidation plausibly exceeds the cost of registering all land. By con- and growth of medium-size farms to increase land values trast, informality of lease contracts is a choice that may and productivity. Achieving this potential will require mea- enable owners to evade regulatory obstacles that prevent sures to limit speculative land acquisition and exercise of them from renegotiating contracts to obtain more favorable market power by making local land markets more compet- terms. Second, while land market liberalization generated itive and using market-based land valuation as a basis for significant indirect benefits, gains are unevenly distributed. taxing land on a recurrent basis and any capital gains due Furthermore, competition in sales markets remains limited, to land appreciation. pointing to scope for measures—including reducing the This paper is a product of the Development Research Group, Development Economics. It is part of a larger effort by the World Bank to provide open access to its research and make a contribution to development policy discussions around the world. Policy Research Working Papers are also posted on the Web at http://www.worldbank.org/prwp. The authors may be contacted at kdeininger@worldbank.org and dali1@worldbank.org. The Policy Research Working Paper Series disseminates the findings of work in progress to encourage the exchange of ideas about development issues. An objective of the series is to get the findings out quickly, even if the presentations are less than fully polished. The papers carry the names of the authors and should be cited accordingly. The findings, interpretations, and conclusions expressed in this paper are entirely those of the authors. They do not necessarily represent the views of the International Bank for Reconstruction and Development/World Bank and its affiliated organizations, or those of the Executive Directors of the World Bank or the governments they represent. Produced by the Research Support Team Land Price Effects of Informality, Farm Size, and Land Reform: Evidence from More Than One Million Transactions in Ukraine Klaus Deininger Daniel Ayalew Ali Keywords: Land market, prices, land reform, market power, productivity JEL Codes: Q10, O13, H56, R14 The findings, interpretations, and conclusions expressed in this paper are entirely those of the authors. They do not necessarily represent the views of the World Bank and its affiliated organizations, or those of the Executive Directors of the World Bank or the governments they represent. Funding support from the European Union (ENI/2017/387-093 and ENI/2020/418-654) is gratefully acknowledged. We thank Denis Bashlyk, Florian Blum, Ben Hell, Thea Hilhorst, Roman Hrab, Sergyi Kubakh, Vasyl Kvartiuk, Daria Manzhura, Sevara Melibaeva, Irina Schuman, Karlis Smits, and Taras Vysotskyi for helpful discussions and insightful comments and as well as Tania Khorzovskaya for outstanding coordination and administrative support. Land Price Effects of Informality, Farm Size, and Land Reform: Evidence from More Than One Million Transactions in Ukraine 1. Introduction Land and house prices in high-income countries have increased significantly in real terms since the 1950s (Knoll et al. 2017), allowing owners to realize collateral benefits (Kumar and Liang 2024; Zevelev 2021) and take up entrepreneurial activities (Schmalz et al. 2017). A growing literature on urban land markets shows that market prices of property are affected by public goods such as connectivity (Ahlfeldt et al. 2017), road maintenance (Gertler et al. 2024), trees (Han et al. 2021), proximity to the coast (Severen and Plantinga 2018), or flood risk exposure (Bosker et al. 2019). Data on land prices is instrumental to estimate the cost of agglomeration (Combes et al. 2019) and analyze impacts of policies such as urban regeneration (Rossi-Hansberg et al. 2010) and low-income housing construction (Asquith et al. 2023), density restrictions (Hilber and Vermeulen 2016), building standards (Baylis and Boomhower 2023), disaster risk (Gourevitch et al. 2023; Hino and Burke 2021) and insurance subsidies (Garbarino et al. 2023). In rural areas, factor mis-allocation that is not related to endowments (Adamopoulos and Restuccia 2022) has been shown to be widespread (Gollin and Udry 2020) and to negatively affect productivity (Maue et al. 2020) and occupational choice (Adamopoulos et al. 2022). Measures to improve market functioning such as issuance of transferable certificates (Bu and Liao 2022; Chen et al. 2022) or removal of restrictions on land (Chari et al. 2021; Chen et al. 2023; de Janvry et al. 2015) or labor (Giles and Mu 2018; Zhao 2020) have been shown to reduce mis-allocation. Yet, although digital and remotely sensed data would allow to assess market functioning and put in place regulations to limit exercise of market power (Balmann et al. 2021) in near real time, rigid restrictions on land market operation remain in place in many countries (Shifa and Xiao 2023; Vranken et al. 2021). Use of geo-coded price data to study impacts of how land was acquired (Allen and Leonard 2021), surveyed (Libecap and Lueck 2011), assigned (Smith 2024), or can be transferred (Dippel et al. 2024) has largely been limited to historical cases in the US. This paper uses more than 1 million transactions in lease and sales markets from Ukraine, a country that adopted far-reaching reforms including opening of land sales markets and land price reporting in 2020/21, to explore the scope for using land price data to make inferences on land market functioning that could inform policy in a more contemporary setting. It shows that linking administrative data to remotely sensed ones to identify fields as basic units of agricultural production within a farm can provide insights on (i) impacts of market liberalization on prices and structure of land markets at the local level; (ii) effects of 2 different types of informality on land sale or lease prices; and (iii) the benefits of size at the level of cadastral parcel, field, or farm that can usefully complement existing insights and methodologies. Market liberalization and structure: We show that, beyond any direct effects, elimination of restrictions on land sales markets triggered short-term lease price increases of 10-20 percent, well above the direct effects of less far-reaching reforms reported in the literature (Lawley 2018). A plausible mechanism is that the possibility of land sales increased competition in lease markets. This interpretation is supported by two findings: First, higher transaction volumes at the village level are linked to higher lease prices. Second, while tenants who already have some registered use rights in a village pay lower rental fees, this likely reflects an informational advantage rather than an ability to exercise market power as the rental discount shrinks instead of growing with higher shares of land use, a notion that is inconsistent with market power. Legal entities pay significantly higher prices than individuals and on average lease land in 12 villages, consistent with studies showing that such entities are sufficiently professional and spatially diversified (Seifert et al. 2021) to make lease markets competitive. Evidence of frictions in land or other markets constraining land supply or demand is more pronounced for sales markets where possible market power is indicated by the fact that prices decrease with the share of village land transacted or with the share of village land used by the buyer prior to purchase. Mechanisms to increase competition in the land sale market, including through better knowledge, expanded options for mortgage lending including for land acquisition, and measures such as capital gains taxation to reduce incentives for speculative land acquisition and or exercise market power, e.g., by taxing or requiring electronic auction if use right concentration exceeds a certain level, may be gainfully explored. Informality: Most studies assume that the benefits from formalization accrue to those directly affected and assess the impact of informal land rights by comparing parcels with and without rights using discontinuities in space (Ali et al. 2014; Hawley et al. 2018) or time (Beg 2022) for identification although rights may return to informality (Galiani and Schargrodsky 2016; Gutierrez and Molina 2020) if the cost of subsequent registration is high (Ali et al. 2021). Informal rights can substitute for formal ones to some extent (Lanjouw and Levy 2002). They offer short-term advantages (Niu et al. 2021) but impose costs in the short (Tellman et al. 2021) and the medium term (Henderson et al. 2021; Navarro and Turnbull 2014). Yet, if it yields rents for intermediaries (Krishna et al. 2020), informality may be difficult to eliminate politically. 1 We depart from a view of informality as a purely private issue by using fields as a basic unit of agricultural land use. This allows estimating external effects of three types of informality that we expect to affect land prices via different channels. First, parcels not mapped in the cadaster or registered in the registry of rights 1 A detailed model and discussion of costs and benefits of informality in the labor market for Mexico (Bobba et al. 2022); biased response to infrastructure improvement in Ethiopia (Perra et al. 2024), and externalities due to tax reform in Pakistan (Waseem 2018). 3 cannot be transacted formally, something we expect to affect prices for lease or sale negatively. Second, informal lease contracts, i.e. a parcel registered in the owner’s name while most of the associated field is cultivated by a legal entity, are a choice that may allow an owner to re-negotiate lease prices more frequently than is possible for formal contracts given the 7-year minimum term and, if landlords communicate with each other, generate positive externalities in the lease but not the sales market. Finally, having a larger number of registered tenants on a given field increases the cost of contract enforcement and negotiation and is likely to reduce lease but not sales prices. The data show that (i) parcels in fields with more unmapped or unregistered parcels are consistently less valuable than those unencumbered by such informality; (ii) those in fields with more informal contracts fetch higher lease but not sales prices; and (iii) parcels in fields with more than 5 registered owners exhibit lower lease prices. A nuanced and differentiated assessment of informality along these lines is not only relevant from an academic but also from a policy perspective: most importantly, with externalities, expecting landowners to pay the full cost of reducing informality may be unrealistic or impractical. Farm size: By estimating if and to what extent the size of parcels, fields, or contracting parties’ farms affects land rental and sales prices, we add to the debate on scale economies and, insofar as prices reflect underlying productivity, the agricultural farm size-productivity relationship (Binswanger et al. 1995). A combination of labor market transaction costs with increases in machine capacity at larger sizes has been shown to result in a U-shaped relation in India (Foster and Rosenzweig 2021). Higher agricultural productivity by large mechanized farms is also found in China (Sheng et al. 2019), Brazil (Helfand and Taylor 2021), Australia (Sheng and Chancellor 2019), Kenya (Muyanga and Jayne 2019), Nigeria (Omotilewa et al. 2021), Ethiopia (Ali and Deininger 2021), and Pakistan (Ayaz and Mughal 2024). Beyond agriculture, contiguous pieces of land also provide benefits for urban expansion (Brooks and Lutz 2016), mineral extraction (Leonard and Parker 2021), solar farming (Abashidze and Taylor 2023; Klingler et al. 2024), or conservation. We expect size to affect agricultural productivity at the parcel, field, or farm level differently, namely (i) for any field size, larger parcels reduce the number of potential parties to be negotiated with (and may also be correlated with land quality or field size); (ii) larger fields reduce operational cost by minimizing the time needed for traveling between fields or the amount of borders to a field; and (iii) larger farms will be able to spread costs of fixed factors (machines, management, access to marketing channels) over a larger operation, a gain offset by diseconomies due to a limited span of control. With competitive lease markets, such changes in cost of production would translate directly into lease prices, although market imperfections and agents’ ability to exploit these strategically may result in frictions. We find that size at different levels indeed affects lease prices differently: while inter-quartile increases in field size are estimated to yield somewhat larger impacts than equivalent increments in parcel size, larger 4 farms, up to the limit of about 1,000 ha in our sample, pay significantly higher lease prices than smaller ones, pointing towards unutilized scope for growth in what, in Ukraine, are considered mid-size farms. The fact that, beyond size, tenants value complete control of a field, points to non-negligible cost of contract negotiation and enforcement. While differences in coefficients from equivalent regressions for land sales as well as buyer attributes suggest factor market imperfections may make it easier to exercise market power in sales markets, drawing strong conclusions from the limited sample for which data were available to us may be premature. Exploring differences between lease and sale markets based on better access to registry data will be an important area for follow-up research that could also help identify options for the future trajectory of Ukraine’s agriculture sector and ways in which these would be affected by different policies. The rest of the paper is structured as follows. Section 2 provides country context by describing the evolution of Ukraine’s rural property rights structure and variable construction, recent land market reforms, and the evolution of transaction volume in land sales and rental markets. Section 3 discusses descriptive data for rental and sales markets respectively and identifies the econometric approach as well as expected results. Section 4 explains results for lease prices separately for commercial (with and without registry data) or personal use land as well as land sales. Section 5 concludes with policy and research implications. 2. Country context To set the scene and justify the models applied, this section discusses key aspects of land tenure and land markets in Ukraine, focusing on the types of land and the land ownership structure created by land right privatization in the early 2000s; the sources of informality; and the nature and implementation timeline of land policy reforms in 2019 and 2020. We discuss evidence from the literature and our data on some 900,000 and 300,000 lease and sales transactions, respectively, that form the basis for subsequent analysis. 2.1 Land ownership structure and institutional reforms in Ukraine After the fall of the Berlin wall, transition countries followed very heterogeneous trajectories that continue to influence how agricultural land markets operate (Swinnen et al. 2016). With more than 40 million hectares (ha), Ukraine has an agriculturally useable area that is larger than that of Germany, Poland, and France combined. Of this, approximately 20 million ha is farmed by large farms with average sizes of about 2,000 ha, many of which have links to foreign capital markets (Deininger et al. 2018); about 12 million ha is estimated to be cultivated by small and household farms; and some 9.2 million ha remained under communal and state ownership. 2 2 An unknown share of this land was “privatized,” often in nontransparent ways that imposed significant losses to the public (Nivievskyi 2020). 5 Before 2022, Ukraine’s agriculture sector contributed about 10% of gross domestic product and 42% of the country’s exports, making the country an important supplier of agricultural commodities, in particular sunflower oil, maize, and wheat to the global market. This importance raised concern that the Russian invasion of Ukraine would not only impair Ukraine’s economic welfare (Devadoss and Ridley 2024) but also food security among importers heavily dependent on Ukrainian supplies and globally (Lin et al. 2023). The formal initiation of EU accession negotiations in December 2023 also implies that Ukraine’s agricultural structure and market functioning is of importance for potential future expansion of the EU. To justify our empirical approach, a better understanding of Ukraine’s land policy environment is key with three aspects particularly relevant: First, the types of land that could be owned and the ownership structure were shaped by de-collectivization in the early 2000s when some 7 million individuals, most of them workers on former collective farms, became landowners by being issued shares to land parcels of 3-4 ha each. In this process, a distinction was made between land for personal use that is close to the village settlement area and often comprises home gardens and land for commercial use that is made up of the land cultivated by former collective or state farms. While the former was often used for subsistence production and may eventually be informally converted to non-agricultural uses, 3 the latter has fewer alternative land uses beyond agriculture. Given these structural differences, we distinguish between these two types (which we refer to as ‘commercial’ and ‘personal’ land for simplicity) throughout. Second, the way in which land rights were mapped, registered, and can be transacted, resulted in different types of informality. Although cultivated fields generally comprise several cadastral parcels, the land corresponding to individual owners’ shares was, after privatization in the early 2000s, mapped by donor- funded projects. These projects had varying levels of quality control, implying that parcels may have been left out of the cadaster by mistake or because they had a dispute pending at the time of cadastral mapping. Figure 1, taken from Martyn et al. (2022), illustrates this for an irrigated area in the country’s South: each of the circles corresponds to a field, the boundaries of which are defined by the reach of the associated center-pivot irrigation system. Simple visual inspection shows that fields comprise many cadastral parcels and that there are cultivated areas that are not mapped in the cadaster, a situation we refer to as informality of type I. In 2013, a new digital registry of rights was created but documentation on rights that had been created before was not transferred. Even for parcels that are mapped in the cadaster, rights may thus not be visible in the registry either because owners did not register them or did so before 2013 so that registration information is still contained in the old (analogue) database in the cadaster. Consulting registry data allows to identify this type of informality, which we refer to as informality of 3 A small part of land for personal use includes non-transferable use rights to ‘reserve land’ that had not been cultivated before that village councils could give to individuals. 6 type II, easily. Finally, a third type of informality relates to presence of informal lease contracts. We assume that an informal lease is present for parcels that are registered only in the name of the owner but located in a field more than 50% of the area of which is used by a legal entity. Third, while land rentals, with a maximum term of 49 years, have been allowed throughout, a ‘moratorium’ on land sales was put in place almost immediately after land was privatized in the early 2000s. 4 This prohibition of land sales made it difficult to use land as collateral or liquidate land-attached investments, reducing productivity (Nizalov et al. 2016). Together with land owners’ limited awareness of their rights or capacity to use them, this is argued to have created opportunities for manipulation (Zadorozhna 2020) and exercise of market power (Graubner et al. 2021) at the local level, often with officials’ connivance (Kvartiuk et al. 2022; Neyter and Nivievskyi 2022). Limited competition in rental markets thus often implied that land was rented not by producers with high levels of productivity (Kvartiuk et al. 2024) but by those who were able to exploit informational and other advantages. The land sales moratorium was lifted by 2020 legislation to open land markets in two stages: Ukrainian individuals were allowed to buy agricultural land up to a maximum of 100 ha each from July 1, 2021. From January 1, 2024, the ownership limit was increased to 10,000 ha and fully Ukrainian owned legal entities were allowed to purchase land as well. This step was expected to deepen domestic financial markets more broadly and, in the agriculture sector, support diversification away from land- and capital- intensive bulk commodities with limited value addition and employment generation by allowing use of agricultural land as collateral for credit. Together with land sales market opening, laws were put in place to mandate public reporting of land prices and establish a public land market monitoring system; restructure institutions to increase transparency via digital registry-cadaster interoperability; improve public land management by transferring ownership of such land to local governments subject to central land use controls; require transfer of lease rights to public land via electronic auction; 5 and establish a partial credit guarantee facility to ease small producers’ access to financial markets. 6 2.2 Data sources and variable construction To explore land price determinants since 2021 in land rental and, after the moratorium had been lifted for individuals or legal entities, land sales markets, we combine a range of administrative data and use remotely sensed imagery to generate land use maps and field boundaries using machine learning approaches. 4 A limited class of land was exempt from the moratorium. Also, legislation requiring a minimum lease term of 7 years was passed in 2015. While this may have improved predictability for the large farm sector, it likely pushed farmers interested in a more flexible lease term into informality. 5 The mandatory shift from centralized, in-person to fully electronic auctions run by local authorities for any transfer of use rights to public land instantaneously increased lease prices by 175% (Deininger et al. 2023). 6 The numbers of relevant laws are 552 (turnover law opening agricultural land sales market), 340 (anti-raider law, including price reporting), 711 (local planning), 554 (spatial data interoperability), 985 (state support), 1423 (institutional reform of SGC and decentralization), 1444 (mandatory e-auctions for public land ), and 3205 (partial credit guarantee facility). 7 Public parcel-level data on agricultural land sale and rental transactions that are regularly published as part of Ukraine’s land market monitoring system on the website of State Geocadaster (SGC) form the basis of our analysis. These data, which include the date of the transaction, the parcel’s cadastral number, price, the term for leases, and firm IDs for contracting parties that are legal entities, 7 have been used by others (Ibatullin et al. 2024; Kvartiuk and Martyn 2022). 8 We construct prices per hectare and convert to US$ using the monthly exchange rate in place at the time of the transaction.9 Parcels’ cadastral number allows overlaying parcel polygons with publicly available crop maps to obtain the share of any given parcel under different types of land use. Crop maps also allow identification of land that is cultivated but not mapped in the registry, our indicator of informality. We further use parcel centroids to compute distances from every transacted parcel to key infrastructure objects such as roads, ports railways, and settlements that are likely to affect land prices and which we sourced from open street map. Finally, to measure soil quality, SoilGrids data (Poggio et al. 2021) is used to obtain pH, soil density, and nitrogen as well as organic matter in the 0-60 cm horizon for each parcel’s centroid as well. 10 Beyond these indicators that are routinely used in hedonic analysis, we define fields as contiguous units that are cultivated with the same crop and bounded by terrain features such as roads, rivers, wind breaks, stretches of forests or settlements that are difficult to modify in the short term. This definition, which is comparable to what is used in EU member states for agriculture sector monitoring, allows fields to be identified using machine learning algorithms trained on EU data using medium resolution satellite imagery, in our case stacks of sentinel imagery, 11 rather than a possibly outdated or incomplete cadastral map. Access to field boundaries allows us to compare if field, rather than cadastral parcel, size affects land prices. Moreover, for a limited subset of fields, we were able to access registry information by using the commercial registry API to obtain information on registered rights to all the cadastral parcels in the field. This allows to analyze the impact of different types of informality (i.e., not only the presence of cultivated area not mapped in the cadaster but also area mapped but not registered or the number of registered owners or users) within a field. Finally, we use cadastral numbers to construct the number of lease and sale transactions in each village and year to identify if the volume of market transactions affects prices. 7 Full reporting of prices for sales transactions started only in January 2023 following legal changes in mid-2022. Before that, sales prices were reported for less than 50% of transaction, a threshold that remains to be crossed for leases. Data on contract terms is missing for about 7% of lease transactions with price data. For transacting parties that are not legal entities, only synthetic IDs are available to protect personal data. 8 Files for transactions can be accessed at https://land.gov.ua/monitorynh-zemelnykh-vidnosyn/. We exclude a total of 47,515 sale and 8,317 lease transactions with reported prices for ‘gardens’ (land use categories 01.05, 01.06 and 01.07) that may have other unobserved infrastructure attached to it throughout. 9 Data for personal and commercial use used in the regressions are winsorized prices at the 1st and 99th percentiles to reduce outlier-induced bias. 10 Crop maps for 2019-24, elaborated using the methodology described in Kussul et al. (2017); Shelestov et al. (2017), and Shelestov et al. (2020) were used and are available at https://ukraine-cropmaps.com/. Soil quality data is at from https://files.isric.org/soilgrids/latest/data. 11 Crop field delineation for 2022 (https://area-monitoring.sinergise.com/docs/markers/field-delineation/) is used, as natural features will not change in the short term. 8 This can be combined with data on all parcels owned by a subset of legal entities to test if concentration of land use at field or village level affects prices, as a potential indication of market power. 12 2.3 Evolution of sales and lease market volume and prices Table 1 shows that, since January 2021, 1.97 million new leases and close to 390,000 sales of agricultural land were registered. While reporting of transaction prices was legally mandated, weak enforcement implies that there are 307,083 sale and 933,102 lease transactions for which prices are reported, with a median price of US$ 981 per hectare for sales and US$ 89 per hectare per year for leases as a basis for subsequent analysis (panel A). The data also suggest that the number of registered mortgages using agricultural or any other type of land as collateral remained minuscule. 13 Appendix table 1 illustrates the time profile of transactions. After opening of land markets for individuals in July 2021, monthly land sales quickly increased to 27,070 for 73,818 ha in December 2021 before, in February 2022, all land sales were stopped and leasing activity reduced significantly as a result of the war.14 Median prices for sales and leases decreased by 14% from US$ 1,133 to US$ 989 and by 6% from US$ 90 to US$ 85, respectively, between 2021 and 2022 and remained close to this level thereafter. 15 Compared to individuals or FOPs who accounted for 5.7% and 8.4% of transactions and area,16 legal entities dominated in the lease market, accounting for 94% of all transactions and 92% of the land transacted in the period (table 1 panel B). Of all the land parcels leased, 81.4% were categorized as land for commercial and 18.6% as land for personal use. By contrast, as legal entities were not allowed to purchase agricultural land before 2024, 95% of land sales transactions that account for 94% of the area transferred in land sales markets were by individuals or FOPs and 42% of sale transactions (accounting for 22% of area transacted) involved personal rather than commercial land. Ceteris paribus, legal entities pay about a third more for leasing or buying land. Monthly data provide a more granular picture: As figure 2a illustrates graphically, the number of monthly sales, which had reached almost 30,000 in December 2021, dropped to zero due to the war in early 2022 and then gradually recovered to about 10,000 in March 2023 and 15,000 by March/April 2024. Once legal 12 The date of parcel acquisition is used to determine whether the tenant’s or buyer’s farm was operational in the year before the transaction as well as the size of their operational holding. 13 The most immediate reason is that even if land can be used as collateral for loans, in the Ukrainian context, the transaction costs of doing so for individual farmers are prohibitive. The absence of a response to the opening of land sales markets to legal entities and the low number of mortgages even for non-agricultural land (with a total of 9,289 mortgages involving residential, recreational, or industrial land registered in the entire period since January 2021) points to more deep-rooted regulatory or institutional issues. 14 To fend off potential cyber-security threats, all land-related digital registries were taken offline immediately after the Russian invasion with only leases allowed to be registered locally on paper using simplified procedures. 15 26% of lease transactions affecting 18.6% of leased area with an average size of 1.6 ha per lease involved land for personal farming and 81% of leased area was land for use by commercial agriculture (below referred to as ‘commercial land’ for simplicity). Similarly, 78% of land transacted in the sales market was commercial, though with an average parcel size of 3.2 ha as compared to 1.2 ha for personal land, commercial and personal land account for 42% and 58% of transactions in the sale market, respectively. 16 The difference in area points to a shorter maturity of individuals as compared to legal entities’ lease portfolio (often referred to as ‘land bank’). 9 entities were allowed to purchase land in 2024, they accounted for about 20% of total land sales. Figure 3a shows, that sales prices for land for commercial agricultural use decreased to about 80% of the pre-war level after mid-2022. Once they were allowed to enter the market in January 2024, legal entities paid prices for such land that were about 30% higher than those by individuals, reaching or even exceeding the pre- war price level. While a similar pattern emerges for personal agricultural land, prices paid for it are above those for commercial land and less affected by the war, consistent with the notion that such land is located closer to villages’ settled area, making it more suitable for non-agricultural uses including housing. Although, as a result of the land privatization process, land supply is atomistic with individual landowners owning at most a few parcels, demand may be more concentrated. Appendix table 2 makes use of registry data to provide information on the mean and median areas acquired by individual tenants or buyers in each year, separately for land leases (panel A) and sales markets (panel B). For legal entities for which registry information could be accessed (73% of tenants and 67% of buyers post-2024), this is complemented with data on registered area cultivated. In the lease market, legal entities not only account for the lion’s share of land rented but also pay prices that are significantly higher than those paid by individuals, as noted earlier. With a mean of 158 ha (from 95 ha in 2021 to 46 ha in 2024), the area rented by each legal entity over a period of four years is about eight times that (20 ha) by individuals (panel A). On average, legal entity tenants have registered rights to 1,031 ha of which they cultivated 895 ha. The differential between legal entities and individuals in median (US$ 1,250 vs. US$ 969) or mean prices (US$ 1,650 vs. US$ 1,777) as well as the mean and median area purchased per buyer (8.5 and 1.5 ha for individuals vs. 37 and 6 ha for legal entities) is smaller in sales than in lease markets although in both cases, the distribution of sold area is heavily skewed with maximum areas bought by the same buyer amounting to 3,195 for individuals and 2,233 for legal entities. With 2,100 ha registered and 1,850 ha cultivated, legal entities that buy land are more than double the size of tenants, a fact that may be suggestive of higher barriers to sale market participation due to ill-functioning financial markets and possibly also speculative elements. 3. Descriptive evidence and econometric approach To characterize the subjects and objects of market transactions and market structure, we complement price data with information from cadaster, registry, and remote sensing. In addition to highlighting differences between tenants’ and buyers’ farms, this also allows us to identify fields as the basic unit of crop cultivation within a farm, thereby allowing to use within-field variation in formal land rights. We use this as a basis for discussing the model and expected coefficients. 10 3.1 Descriptive evidence for lease and sale transactions Descriptive information for the close to 900,000 transactions for which new leases were signed and prices reported after January 2021 are presented in table 2, overall (column 1) and for commercial and personal land in the entire sample (columns 2 and 3), for the roughly 60% of the sample where tenant data is available (columns 4 and 5), and for the 9% of the sample for which accessing registry data was feasible (column 6). Parcels for commercial land are slightly larger (2.3 ha) than those for personal land (1.6 ha). Mean and median annual lease rates are US$ 117.5 and 89.6 per ha, respectively. The average number of lease transactions per village over the entire period was close to 170 (192 and 69 for commercial and personal land, respectively), implying that new leases were recorded in about 10,889 villages in Ukraine Of nearly 340,000 leases conducted in 2021 only, about half were concluded before or after the moratorium on land sales was lifted for individuals. Although the fact that the date of land market opening was known well in advance makes increases of land rents in anticipation of land sales markets likely, implying our estimate is a lower bound, this allows us to use an indicator variable for lifting of the sales moratorium to obtain an estimate of the short-term effects of this policy change on lease prices. While the number of leases conducted in 2024 remains, with some 125,000, much smaller, a similar strategy can be applied to assess the impact of land sales market opening to legal entities on the lease price paid by such entities. While more than 70% of leases are for terms of less than 14 years, which provides the most flexibility given the current legal environment of a minimum lease term of 7 years, about 17% and 11% are concluded for terms between 14 and 20 or above 20-years, respectively. The average leased parcel is 2.1, 7.0, 12.1, and 27 km from the next primary, secondary, tertiary road or motorway, respectively and 9.9 and 12.9 km from the next railway station and grain elevator (appendix table 3). Remotely sensed data suggest that about 77% of the area of leased parcel is cultivated, 20% and 2% are covered with grass or trees, respectively, and there are small shares of built up and other land uses (appendix table 3). Linking cadastral records to field boundaries from remotely sensed imagery as discussed above points to an average field size of 63 ha (68 ha for commercial and 49 ha for personal land) with about 6% of field area not mapped in the cadaster. Tenant information, which is available for 60% (534,127/884,936) of lease transactions with reported prices, points to a mean size of 345 and 261 hectares of registered land for tenants of commercial and personal land, respectively with these tenants having registered rights to between 20% and 24% of cultivated land in the villages where transactions were recorded.17 Tenants who had rights to less than one third, between one third and half, and more than half of village land were involved in 62%, 14%, and 18% as well as 63%, 13%, and 13% of new lease contracts for commercial and personal land, 17 Note that 6.6% of transactions for commercial and 7.7% for personal land were concluded by tenants who did not previously have registered use rights to land in the village. 11 respectively. The mean number of new leases per village concluded over the entire period was close to 170 and 67 for commercial and personal land, respectively. For the subset of transactions (12% of commercial land) for which we obtained registry information, mean field area and tenant size are, with 95 and 515 hectares, respectively, are more than a third and nearly half above the means for the sample with tenant characteristics reported in column 4. For this sample, the tenant’s share of village land (36%) and mean and median prices (US$ 138 and US$ 116) are higher and the share of unmapped land lower (3.9%) than in the sample with tenant characteristics. Registry data show that, although use rights to some 80% of field area are registered by the field’s largest user, the mean number of users with registered rights in a field is 2.5 with 10% of fields having 5 or more registered users. Moreover, 81% of fields include land (8.6% of cultivated field area on average) that, albeit mapped in the cadaster, is not registered. For 16% of cases, the area involved is minuscule (<1% of area) whereas for 47% and 18%, it is either non-negligible (1%-10% of area) or significant (>10% of area). Also, 29% of fields, while having use rights to most of their area registered by a legal entity, include land that is registered in the name of the owner only, possibly indicating presence of an informal lease contract. In 6%, 20%, and 3% of cases, the share of owner-registered area is minuscule (<1%), not negligible (1%-10%), and significant (>10%), respectively. Sales markets: Parcels transacted in sales markets are slightly larger than those in the lease market, with a mean sales price of US$ 1,132 for commercial and US$ 1,986 for personal land that in the latter case is driven by some outliers, as evidenced by a median price that is almost equal to that for commercial land. Parcel attributes are not too different from those for leased parcels except for a higher share of grassland for personal land and vice-versa for commercial use (appendix table 4). While 93% of commercial land transacted in land sales markets was part of a field, this is true to less than 70% of land for personal use. The share of field area cultivated but not mapped in the cadaster was, with 5%, similar to what is observed for rental markets. With 7.5% as compared to 3.7%, the incidence of this type of informality was much higher for personal than commercial land. 3.2 Econometric approach and key issues to be discussed Denote transacted parcels, which are located within a field that in turn is located in a village, by i and years by t. We regress sale or lease prices for parcels of commercial or personal use on a set of parcel, field, and village characteristics, along with policy regime dummies, and hromada as well as year fixed effects. Formally, the equation to be estimated is = + + + + + + (1) 12 where is the lease or sales price for parcel i in field f transferred via lease or sale to party p in year t; Xit is a vector of parcel attributes including area, land quality, distance to key infrastructure objects, and land use (crops, forest, built up or grass/uncultivated) in the year before the transfer; F is a vector of field characteristics, including size and the share of a field area that is not mapped in the cadaster (informality of type I), not registered (informality of type II), or has an owner but not a user registered, the number of users with registered rights to the field, and the share of field area held by the largest registered user; T is a vector of acquirer characteristics, including its total cultivated area and the share of registered and cultivated land in the village land to which the acquirer has registered use rights; P is a vector of policy regimes especially if land sales to individuals or legal entities were allowed or not; s are hromada fixed effects; s are year fixed effects; and is a random error term. Our main interest is in how changes in market structure, informality, and size at different levels affect land prices. Regarding market structure, we are interested in (i) the short term effect of eliminating restrictions on land sales markets on land rental markets; (ii) the extent to which longer lease terms are associated with higher prices; and (iii) whether characteristics of local land markets, 18 in particular transaction frequency or local dominance of certain players affect prices paid (Harding et al. 2003) and whether, to the extent data availability allow robust inferences, these differ between land lease and sales markets and allow inferences on possible exercise of market power (Cotteleer et al. 2008). in contrast to evidence from the US where the fact that tenants capture about three quarters of public subsidies points towards imperfect competition in local farmland rental markets (Kirwan 2009). We are interested in exploring whether, beyond the direct effects documented in the literature, informality of land rights, as captured by different dimensions, has external effects. To do so, we assess whether prices for a given parcels are affected by characteristics of the field within which they are located, focusing on (i) the share of a field’s cultivated area that is either not mapped in the cadaster or not registered in the registry, thereby making formal contracting impossible; (ii) the share of a field’s area that is leased out informally rather than through a registered contracts that adheres to the legally stipulated minimum term, possibly making renegotiation of lease rights at field level easier; and (iii) the number of tenants other than the party that is acquiring rights that hold registered rights to a field as a proxy for potential contestability of rights and the ease of enforcing them. Finally, we expect size to affect agricultural productivity at parcel, field, or farm level differently as (i) larger parcels reduce the number of potential tenants to be negotiated with; (ii) larger fields will reduce the cost of mechanization by minimizing the time needed for traveling between fields or the amount of borders 18 See Balmann et al. (2021) for a discussion of the localized nature of land markets and the implications for possible exercise of market power and Pennerstorfer (2022) and Storm et al. (2015) for country-specific examples. 13 included in a field but may increase pest pressure; and (iii) subject to potential diseconomies from a limited span of control, larger farms will be able to spread costs of fixed factors (machines, management, access to marketing channels) over a larger operation. If land markets are competitive, a permanent change in cost of production would translate directly into prices, though market imperfections and agents’ ability to exploit these strategically may result in frictions. 4. Explaining land price determinants Beyond standard coefficients and short-term benefits from land market opening, lease price regressions for commercial agricultural land point towards positive effects of local land market competition, parcel, field, and tenant size, and a negative externality from un-mapped land within a field. Absence of registered rights or increases in the number of parties with registered rights in a field, all of which increase negotiation and enforcement cost, cause negative externalities whereas informality of lease agreements may, by offering greater flexibility regarding legally mandated minimum lease terms, increase leases but not sale prices. Personal land prices are U-shaped rather than increasing in field size and external effects of informality are less pronounced. A decrease in land sales prices with local (village level) transaction frequency or buyers’ share of local use rights suggests sales markets are more prone to imperfections. 4.1 Lease price regressions for commercial agricultural land Results from lease price regressions for commercial land are in tables 4 and 5 without and with registry information, respectively. We report results for the entire sample (cols. 1 and 2) and for subsamples that include data on lease terms (cols. 3 and 4) and tenant characteristics without (cols. 5 and 6) and with (cols. 7 and 8) lease terms. In each of these cases, results in the first column exclude and those in the second column include field characteristics, respectively. 19 Coefficients on land characteristics and time dummies are not reported in the main tables, they are included in appendix tables. For example, appendix table 5 shows that coefficients on distance to roads have expected negative sign and are highly significant. In most regressions, shares of built and cultivated area are significant and positive, with point estimate of around 0.9 for built area and 0.5 for cultivated area compared to bare land as the omitted category. Year dummies point to an average decline of prices by 11%, 16%, and 20% in 2022, 2023, and 2024, respectively, in line with war-induced reductions in agricultural profitability (Deininger et al. 2024). Effects of policy reform: There is evidence that, beyond any direct effects in land sales markets, lifting of the moratorium on land sales had a significant indirect effect on lease prices: allowing sales to individuals on July 1, 2021, is estimated to have been associated with a short term (within year) increase of rental prices 19 As information on tenant characteristics is available only for legal entities, no information on legal entities can be included in cols 5-8. 14 by between 13% and 17%, depending on the specification. Allowing land sales to legal entities on January 1, 2024, is estimated to have been associated with an added lease price increment of between 7% and 9% for legal entities only, beyond the lease price premium of 39% to 43% paid by legal entities as compared to individuals. While this effect can plausibly be explained as a result of increased competition, price data would need to be complemented with production data to ascertain whether it is indeed the most productive producers that access land markets (or what types of barriers prevent this, if not) and how resulting surplus is split between tenants and landowners. Local competition and length of lease terms: With an elasticity of between 3.0% and 4.4%, higher levels of land rental activity at village level (as captured by the number of transactions in the village in the previous year) are estimated to be associated with higher lease rates. If the amount of land available for lease in a locality is constant and tenants cannot renegotiate or cancel leases unilaterally, market volume is a function of the term structure of existing leases, implying that, as long as lease prices are expected to increase, greater flexibility in adjusting lease terms has a benefit (Grenadier 1995) as confirmed by empirical studies (Bond et al. 2008; Hüttel et al. 2016). Yet, including a measure of the length of the lease term in the regression produces counter-intuitive results as lease rates for contracts with terms beyond 20 years suffer a discount of 16 to 21 percent. Recalling that almost 75% of leases are for 7 (the legal minimum) to 10 years, this suggests that up to a quarter of tenants lack awareness of available options or the ability to resist contracts that not only lock them in for more than a generation but also punish them for doing so by paying lower prices than short-term contracts. A possible explanation for this is that the Soviet valuation system, albeit outdated and replaced by a market-based approach, still serves as a focal point for some. Field size and control: Table 4 suggests that parcel lease rates are increasing in field size: moving from the 25th to the 50th or from the 50th to the 75th percentile (from 15 to 45 or 96 ha) is associated with an increase in the lease rate of 4.5 or 3.4 percent, in line with economies of scale in using large machinery at field scale (Foster and Rosenzweig 2021). Although there is less heterogeneity in the size of cadastral parcels, a move from 3 to 5 ha, equivalent from a shift from the 75th to the 90th percentile, would be associated with an increase of between 2.5 and 3 percent in the rental rate, possibly due to reduced negotiation cost, suggesting that, for commercial agricultural land, there is no ‘small parcel premium’ as in the US (Brorsen et al. 2015). Contrary to what would be expected if tenants were to exercise market power, table 5 suggests that those that use a larger share of a field’s cultivated area pay higher lease rates: a tenant cultivating 80% of a field pays a lease rate about 10.8 percent above one without any registered use rights to land in the field, pointing to potential large benefits from avoiding fragmentation of use rights. Land use consolidation thus has clear benefits that are reflected in market prices, casting doubt on the need for an administrative-coercive 15 approach to consolation that, albeit (Loumeau 2022) and implying that reinforcing markets with appropriate policy measures may be more appropriate, equitable, and sustainable. Tenant size effects: Beyond economies of scale at field size, larger tenants (with more than 10 ha of total area) also tend to pay a substantial rent premium: moving from the 25th percentile to the median, from the median to the 75th or from the 75th to the 95th percentile (from 75 to 215, 450, and 1,054 ha, respectively) would, according to regression, be associated with rent premia of 7.2, 6.7, and 9.5 percent, other covariates held constant. While production data for different farm sizes will be desirable to assess the extent to which this price increase mirrors cost reductions in production or economies of scale beyond the production stage and the share of such benefits passed on to landowners, it suggests lease market are reasonably competitive. Limited evidence of market power: For the approximately 60% of the sample where information on tenant characteristics is available, including the share of registered land cultivated by the tenant in the village in which the parcel of interest is located can be used to more directly test for market power. Tenants that cultivate less than 50% of the land in the village indeed pay lease rates 11% lower than those paid by non- resident bidders. The fact that this discount decreases to between 6 and 7% for parties cultivating more than 50% of village land suggests that either land rental markets remain relatively competitive or that concentration of land use at village level is not the most appropriate indicator to measure market power. External effects of informality: The significant and negative coefficient on the share of a field’s area not mapped in the cadaster suggests that informality generates negative externalities as registered parcels in fields that include unmapped parcels fetch lower lease rates. If, as seems plausible, informality is associated with higher costs of concluding or enforcing contracts, such lower lease rates are a real economic cost rather than a transfer. Based on the point estimate, a program to map all the country’s agricultural land and thus reduce the level of unmapped but cultivated land from the current level of 5% to zero would imply a 1.7% increase in the lease rate. In other words, had such a program been implemented by 2021, it would have increased annual rental income received by landowners by US$ 8.71 million (118*0.017*4.34), implying a net present value of close to US$ 100 million. To the extent that it would cost less to map all agricultural land, a program to do so and, rather than charge the full cost to owners, recover them through a fee that takes into account the external effects from informality, may be more appropriate and effective. Although registry data is only available for about 10% of our sample (table 5), it allows to consider other forms of informality, beyond checking the robustness of earlier estimates. Coefficients on policy reforms (10% for the short-term effects of lifting the moratorium and up to 8% for allowing participation of legal entities), positive effects of local competition (estimated coefficients of number of transaction in village 16 ranging between 0.040 and 0.052) and the premium paid by legal entities (0.27-0.38) and the share of a field’s area not mapped in the cadaster (-0.24 to -0.34) are all significant and of a magnitude similar to what was obtained for the full sample. Several conclusions emerge. Lack of registered rights as detrimental as lack of mapping: A parcel in a field that has more than 1% of its area mapped but not registered is estimated to be suffering a lease price reduction by 2.4 to 3.4 percent, as they increase tenure uncertainty for the entire field. Parcels in fields to which more than 5 parties have registered use rights also trade at a lease price discount of between 2.1 and 2.9 percent, presumably a result of the negotiation and contract enforcement cost with more parties. Yet, lack of significance of the coefficient on a dummy for having less than 1% of field area unregistered suggests tenants may have ways of addressing very small levels of informality. No price reduction due to informal lease contracts: Although not always significant, the coefficient on the share of the field that, with most use rights registered by a tenant, has registered ownership but not use rights, which we use as a proxy for the presence of informal lease contracts, is positive throughout. Having more than 10 percent of a field’s area registered by owners is estimated to be associated with a lease price premium of up to 6.2 percent compared to a field that has all cultivated land registered in the name of users. Although this is prima facie surprising, a plausible explanation would be that owners who did not formally transfer their use right to a tenant will have greater flexibility in renegotiating lease terms than those who are locked into the statutorily prescribed minimum lease term of 7 years. 4.2 Lease price regressions for personal use land As noted earlier, agricultural land for personal use is more likely to be devoted to small-scale farming or non-agriculture including housing. Results from lease price regressions for such land in table 6 are similar to those from commercial land in two respects. First, with point estimates of 0.026 to 0.040, significant at 1% on the number of village level transactions, higher local transaction volume is associated with higher lease prices, similar to what was found for commercial land. Second, the lease price premium paid by legal entities is, with between 37% and 43%, almost identical to what was observed for commercial land. At the same time, as most of it was leased by individuals, no registry information is available for land for personal use and there are other important differences between land for personal and commercial use, including: Land market opening: With an estimated price increase of between 15% and 25%, the estimated short-term effect of the first stage of land market liberalization on lease prices for personal land is, on average, about one-third larger than that for commercial land, possibly because of higher land demand for personal land purchases from individuals rather than legal entities. This interpretation is supported by the insignificant coefficient on the indicator variable for the second stage land market opening which implies that, contrary 17 to what is observed for commercial land, most of the demand for personal land originates with individuals rather than legal entities. Parcel, field, and tenant size effects: The estimated coefficient of parcel size is positive and statistically significant for land leased by legal entities for which we have information on tenant characteristics (cols. 5-8). For this group of tenants, estimated parcel size effects (col. 6 in tables 4 and 6) are less pronounced than for similar tenants leasing for commercial use, with an elasticity of 2.1% compared to 7.2% at the median parcel size for each group (1.71 ha for personal and 1.97 ha for commercial use). The effects of field area and tenant size are of similar magnitude across both types of use: elasticities of 2.1% vs. 2.3% (or 2.4% at the median value for commercial use) and 9.2% (or 7.7% at the median value for commercial use) vs. 7.2% at median values of 23 ha vs 48 ha and 159 ha vs 215 ha for personal and commercial use, respectively imply that, so long as a parcel is suitable for commercial cultivation by legal entities, its formal classification thus seems to have limited relevance for market outcomes. For all personal use land transacted (cols. 1-4), we find a statistically significant negative parcel size effect that disappears and turns positive for transactions carried out by legal entities. Appendix table 8 reports results for land for personal use bought by individuals only to support the notion that the negative relationship between size and lease price is indeed confined to transactions involving individuals. This is consistent with the notion that personal land leased by individuals may be intended for purposes other than commercial farming and may contain features such as structures that are not fully captured in our data. External effect of informality is less pronounced: With a coefficient of -0.099 on the share of unmapped area in a field, which has similar mean values of 6.3% for personal use and 5.2 for commercial use (table 2), this effect is less than a third of that found in the equation for commercial land (table 4). Beyond the fact that in many cases, personal land is acquired as a discrete object rather than as part of a field, this may also indicate that local means of enforcement may be more important for such land. The lease price discount for long-term lease contracts is higher than for commercial land, implying that, unless compensated by non- monetary support, owners of personal land who enter such contracts will incur large welfare losses. Less evidence of local market imperfections: Although coefficients on dummies for tenants cultivating more than one-third of village area are negative and statistically significant throughout, this does not provide strong evidence of market power in lease markets compared to commercial land. Those who already cultivate any land in the village pay prices that are by between 7% and 3.1% lower than those paid by outsiders. This is, particularly for the 33%-50% and greater than 50% categories, lower than in the case of commercial land. There is thus little evidence of this discount increasing with the share of village area cultivated by the tenant, suggesting that structural market imperfections are limited. 18 4.3 Sales price regressions Results from land sales price regressions are in table 7 with columns 1 and 2 reporting results for the close to 175,000 and 100,000 sales of commercial and private land, respectively and the remainder focusing on commercial land purchases by legal entities the vast majority of which took place in 2024: regressions for some 26,000 sales with only registry information in columns 3 and 4 and some 8,500 and 2,000 with tenant without and with registry information in columns 5 and 6. 20 To interpret results from the land sales price regressions, note that the sale price is just the present value of the stream of rental payments, discounted at the market interest rate. If markets function well, we expect estimated coefficients from the land sale and rental price equations to be identical. Any differences thus point towards potential frictions that would be worth further investigation. Sale market competition depresses prices: With an estimated elasticity of -1.21 % for commercial and -4.3 % for personal land, higher levels of competition seem to be associated with lower rather than higher sales prices. This points to constraints on the demand (financial market frictions) or supply side (prices too low or an expectation of higher land prices in the future) that would be worth further investigation. Legal entities pay a premium, especially for personal land: While the 30% premium paid by legal entities for commercial land is in line with that for commercial land, a premium of 60% for personal land supports the notion that some of this land may be structurally different from commercial land, most likely as it is more suitable for non-agricultural uses. Use of satellite imagery to assess post-transfer uses of such lands is one way to explore if a possibility to ‘informally’ convert such land to residential or industrial use may have partly driven such a large premium. Informality results in a discount for commercial land: For commercial land, the magnitude of external effects of informality from unmapped land is similar to that observed in the rental markets (discount of 3.1% for a 10% increase in the share of unmapped area in the field where the transacted parcel is located). Non-registration and informality of lease contracts are both estimated to reduce sales prices: the discount of 6.3% for having more than 10% of a field’s area not registered and 1.9% for having more than 1%-10% of field area covered with informal lease contracts is much larger than in the rental regression, consistent with the notion that informality has the potential of creating more problems for owners and that the threat of, for example raider attacks, may be an impediment to investment. Field and parcel size are key price determinants: For commercial land, there is an optimum parcel size of 9.42 ha, which is more than three times the maximum observed in our data. The estimated coefficients 20 Coefficients remain remarkably stable even with a reduction of the sample by more than 98% in column 6 compared to column 1, results should be interpreted with caution. 19 suggest that doubling parcel size is associated with a 6% increase in price at the median (2.75 ha) and a 4% increase at the 75th percentile (4.15 ha). By comparison, the price of parcels for personal use exhibits a U- shape, but with the turning point at 6.8 ha prices are negatively associated with parcel size within the observed distribution. Doubling parcel size results in a price decrease by 18% at the median (1.37 ha) and 14% at the 75th percentile (2 ha). Price also follows a U-shaped relationship with field size for both land types. At the median field area (59 ha and 5.22 ha), the estimated size elasticities are 3.7% and 2.5% for commercial and personal land use, respectively. While it slightly increases to 4.3% for commercial use, it more than doubles for personal use when evaluated at the 75th percentile (109 ha for commercial use and 36 ha for personal use). The amount buyers are willing to pay increases with the share of field area they cultivate, with a 10% increase in this share raising prices by 2% to 3%. At the same time, buyers who are the largest land users in a field pay prices about 10 percent lower, consistent with the notion that, with atomistic supply, this is a buyer’s market. The same trade-off is apparent regarding market power: as observed for the case of land leases, large operators pay more for acquiring a given piece of land (e.g., doubling farm size increases price by 10%, 11% and 13% at the 25th percentile (24 ha), median (452 ha) and 75th percentile (721 ha)), plausibly passing on some of the advantages it enjoys in terms of access to markets for capital or output. At the same time, results in column 5 suggest that buyers who already use some village land benefit from a discount the magnitude of which is increasing with the share of village area used: farms that operate less than one-third, between a third and a half, or more than half of village land, respectively, would, based on regression results, enjoy discounts of 19%, 23% or 25% relative to one that has no operational presence in the village. F-tests rejects equality of estimated coefficients for buyers with less and more than 33% and 50% of village land in columns 5 and 6, respectively, suggesting study of market power in sales markets, ideally by using all registry data to ensure robust coefficient estimates, is warranted. 5. Conclusion and policy implications Our analysis shows that administrative data on prices and parcel characteristics that are routinely generated but often not systematically captured by land registries offer considerable promise to improve analysis of relatively rare events such as land sales or leases. Regulations to ensure such data are captured and reported without bias and can be linked to relevant spatial and textual attributes thus contribute to an important public good that is particularly relevant in settings undergoing structural transformation or being exposed to shocks that may make traditional modes of data collection more difficult. Three findings from the analysis are notable: First, land sales market opening increased competition, generating opportunities for landowners to benefit from higher rental prices. The fact that a quarter of 20 landowners failed to take advantage of this opportunity and instead entered long-term contracts on extremely unfavorable terms implies that, to allow all owners of land to benefit, additional efforts to disseminate land price information will be needed. Moreover, the fact that mortgage lending remained flat throughout the period suggests that the main expected benefit from liberalization of agricultural land markets remains to be realized. Reform to streamline registration and use of mortgages, reduce the transaction cost of selling land, and ensure market-based land valuation will be needed to harness such benefits and deepen domestic financial markets. Such transparent valuation can also avoid injustice of taxation due to the differential evolution of market prices for land across different regions or sectors. Second, informality is not only a private issue but associated with negative externalities. Global experience suggests that, by generating rents for local intermediaries, informality may reinforce resistance to reform and weaken governance. Public action to get all cultivated land registered is thus warranted. Recovering associated cost through a recurrent land tax, the size of which is proportional to the increments in land value resulting from such a step, would be more viable and fairer than asking credit constrained individuals to shoulder associated cost upfront and also mechanisms while avoiding the risk of indirectly incentivizing land grabbing. Analysis also suggests that the statutory minimum lease term may encourage informality and make it more difficult for landowners to share in the benefits from reform. Finally, while greater competition increases prices in rental markets, the opposite is the case in sales markets where dominant local players may be able to exercise some market power. Although the fact that land sales markets for legal entities opened only in 2024, together with the scant amount of registry data we were able to access makes it impossible to arrive at a confident conclusion, the importance of land sales markets as a determinant of Ukraine’s reconstruction and future development renders this an important issue for future research. Ideally such research would combine use of registry information with production data to link land market participation with productivity of land use and, in doing so, overcome some of the limitations that analysis based on market participation alone will not be able to overcome. 21 Table 1: Volume of registered sales and leases for agricultural: Jan. 2021-Aug. 2024 Land sales Land leases No. of Area (ha) Price No. of Area (ha) Price Term No. of Sales (US$/ha) Leases (US$/ha) (years) Mortgages Panel A: Entire sample Total 389,344 910,092 981.2 1,974,319 4,339,537 89.2 12.5 1,866 w price bef. 1/23 84,053 213,800 1,074.0 531,722 1,175,480 88.5 13.0 w price aft. 1/23 223,030 501,892 945.2 401,380 888,351 90.1 12.4 Panel B: By buyer type & land use category Individual/FOP 369,267 853,455 968.8 112,940 365,261 51.5 12.8 Legal entity 20,077 56,637 1,248.5 1,861,379 3,974,276 91.9 12.4 Personal 164,299 198,087 998.5 518,685 809,279 87.4 12.5 904 Commercial 225,045 712,004 975.6 1,455,634 3,530,258 89.9 12.5 962 Panel C: By year 2021 90,074 238,407 1,133.3 754,617 1,692,628 90.0 12.7 493 2022 74,825 167,362 988.7 377,931 763,162 85.2 12.7 398 2023 119,715 259,708 936.0 552,106 1,209,667 92.3 12.2 534 2024 104,730 244,615 956.2 289,665 674,080 86.6 12.1 441 Source: Own computation based on parcel-level statistics from Ukraine registry of rights on property. Note: Transaction prices are median prices if prices are reported. Data for sales starts from July 2021, i.e., after the opening of sales market for agricultural land. Mortgages are for agricultural land only. In addition to the 1,866 mortgages involving agricultural land, there were 9,289 mortgages involving non-agricultural land (7,439 for residential, 1,710 for industrial, and 140 for recreational land, respectively). 22 Table 2: Descriptive statistics for lease transactions Total By type With tenant char’s w. registry Commercial Personal Commercial Personal Commercial Parcel characteristics Parcel area (ha) 2.08 2.27 1.55 2.34 1.52 2.31 Lease price ($/ha*a) 117.48 117.94 116.24 123.10 121.27 137.52 Median lease price ($/ha*a) 89.58 90.42 87.44 94.78 92.65 115.98 Post-July 2021 0.810 0.810 0.809 0.838 0.830 0.809 Tenant legal entity 0.948 0.949 0.945 1.000 1.000 0.975 Data on contract duration exists 0.931 0.931 0.929 0.939 0.937 0.943 if yes, 14-20 yr contract 0.175 0.183 0.151 0.182 0.151 0.213 if yes, > 20 yr contract 0.108 0.108 0.108 0.114 0.108 0.137 Field characteristics Parcel located within field 0.902 0.910 0.882 0.907 0.889 1.000 if yes, field area (ha) 63.02 68.29 48.59 69.34 48.17 95.04 field area share not mapped in cadaster 0.055 0.052 0.063 0.050 0.067 0.039 unregistered land in RRP 0.807 if yes, field area share not registered 0.086 <1% unregistered area 0.162 1-10% unregistered area 0.465 >10% unregistered area 0.179 Only owner registered in RRP 0.294 if yes, field area (%) with owner only 0.047 <1% only owner reg. area 0.062 1-10% only owner reg. area 0.200 >10% only owner reg. area 0.031 % of field used by tenant 0.694 field area share of largest user 0.795 Tenant is the largest farm 0.792 No. of registered users 2.435 5 or more registered users 0.100 Tenant & village characteristics Tenant's reg. area (ha) 344.90 261.07 514.67 Tenant's cult. area (ha) 296.42 226.98 428.15 Tenant's share of village land 0.239 0.204 0.364 Tenant's sh. of v. land <33% 0.621 0.629 0.410 Tenant's sh. of v. land 33-50% 0.136 0.131 0.148 Tenant's sh. of v. land >50% 0.177 0.132 0.367 No. of villages tenant operates in 11.97 11.07 10.56 Leases of comm. land in village 169.63 191.56 68.86 170.94 67.02 247.30 Leases of comm. land in village 66.96 27.88 121.48 28.19 116.44 38.46 Time period of transaction Year 2021 0.382 0.376 0.397 0.341 0.367 0.377 Year 2022 0.193 0.182 0.224 0.184 0.228 0.184 Year 2023 0.284 0.296 0.252 0.320 0.275 0.305 Year 2024 0.141 0.146 0.127 0.156 0.130 0.134 Number of observations 884,936 648,292 236,644 382,653 151,474 79,316 Source: Own computation based on land monitoring data reported by SGC, complemented by administrative and remotely sensed data as discussed in the text. 23 Table 3: Descriptive statistics for sales transactions Total By type of land Commercial land with data on .. Commercial Personal Registry Buyer char’s (BC) Reg & BC Parcel characteristics Parcel area (ha) 2.42 3.07 1.27 2.74 3.00 2.86 Sale price ($/ha) 1,439 1,132 1,986 1,153 1,527 1,897 Median sale price ($/ha) 959 977 902 995 1,155 1,475 Buyer legal entity 0.065 0.078 0.042 0.092 1.000 1.000 Field Characteristics Parcel located within field 0.839 0.928 0.679 1.000 0.876 1.000 if yes, field area (ha) 72.34 83.66 44.70 102.93 109.60 116.02 area share not mapped 0.048 0.037 0.075 0.029 0.031 0.028 unregistered land in RRP 0.790 0.800 area share not registered 0.087 0.069 <1% unregistered area 0.173 0.189 1-10% unregistered area 0.458 0.489 >10% unregistered area 0.160 0.121 Only owner registered 0.322 0.304 area % w owner; no user 0.058 0.040 <1% only owner reg. area 0.051 0.058 1-10% owner reg. area only 0.228 0.218 >10% only owner reg. area 0.043 0.028 % of field used by buyer 0.060 0.716 area share of largest user 0.805 0.834 Buyer is the largest farm 0.067 0.808 No of registered users 2.332 2.147 5 or more registered users 0.085 0.071 Buyer & village characteristics Buyer's reg. area (ha) 498.76 600.50 Buyer's cult. area (ha) 446.12 526.16 Buyer's share of village land 0.371 0.540 Buyer's sh. of v. land <33% 0.352 0.160 Buyer's sh. of v. land 33-50% 0.155 0.174 Buyer's sh. of v. land >50% 0.385 0.622 No. villages buyer operates in 23.10 17.74 Sales of comm. land in village 27.66 28.41 20.35 43.65 47.54 54.87 Sales of pers. land in village 20.69 10.97 24.22 13.32 11.48 13.42 Year 2021 0.163 0.147 0.192 0.139 0.000 0.000 Year 2022 0.114 0.109 0.123 0.108 0.000 0.000 Year 2023 0.386 0.380 0.396 0.385 0.000 0.000 Year 2024 0.337 0.363 0.289 0.367 1.000 1.000 Number of observations 270,762 173,651 97,111 25,719 9,166 2,135 Source: Own computation based on land monitoring data reported by SGC. Note: All distances are in km. 24 Table 4: Lease price regressions for agricultural land for commercial use All transactions With contract With tenant Tenant details & contract duration characteristics duration Area (ha) 0.071*** 0.061*** 0.068*** 0.059*** 0.068*** 0.060*** 0.069*** 0.061*** (0.001) (0.001) (0.001) (0.001) (0.002) (0.002) (0.002) (0.002) Area^2 -0.004*** -0.004*** -0.004*** -0.004*** 0.009*** 0.009*** 0.009*** 0.009*** (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) Moratorium lifted 0.152*** 0.153*** 0.117*** 0.118*** 0.158*** 0.158*** 0.118*** 0.119*** (0.003) (0.003) (0.004) (0.004) (0.004) (0.004) (0.005) (0.005) Legal entity (renter) 0.353*** 0.333*** 0.361*** 0.341*** (0.005) (0.005) (0.005) (0.005) Legal entity * 2024 0.081*** 0.085*** 0.071*** 0.075*** (0.011) (0.011) (0.011) (0.011) # transactions 0.037*** 0.034*** 0.033*** 0.030*** 0.044*** 0.042*** 0.044*** 0.042*** in village (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) Parcel part of field -0.055*** -0.061*** -0.044*** -0.046*** (0.005) (0.005) (0.007) (0.007) Field area (ha) 0.014*** 0.018*** 0.016*** 0.018*** (0.002) (0.002) (0.003) (0.003) Field area^2 0.004*** 0.003*** 0.001** 0.000 (0.000) (0.000) (0.000) (0.000) Unmapped area -0.329*** -0.321*** -0.322*** -0.313*** share in field (0.012) (0.013) (0.016) (0.016) Lease term 14-20 0.019*** 0.017*** -0.053*** -0.053*** years (0.003) (0.003) (0.004) (0.004) Lease term > 20 -0.159*** -0.163*** -0.212*** -0.213*** years (0.003) (0.003) (0.004) (0.004) Tenant's cult. -0.052*** -0.052*** -0.037*** -0.038*** area (ha) (0.003) (0.003) (0.003) (0.003) Tenant's area^2 0.012*** 0.012*** 0.011*** 0.011*** (0.000) (0.000) (0.000) (0.000) Tenant uses <33% -0.105*** -0.102*** -0.116*** -0.113*** of village land (0.006) (0.006) (0.006) (0.006) Tenant uses 33-50% -0.112*** -0.112*** -0.110*** -0.109*** of village land (0.007) (0.007) (0.007) (0.007) Tenant uses >50% -0.064*** -0.067*** -0.064*** -0.066*** of village land (0.007) (0.007) (0.007) (0.007) No. of obs (farms) 648,292 648,292 603,648 603,648 382,653 382,653 359,323 359,323 Adj R-squared 0.338 0.341 0.344 0.347 0.388 0.390 0.398 0.399 Note: Parcel characteristics, year dummies, OTG fixed effects and a constant included but not reported (see appendix table 2 for estimated coefficients). Standard errors in parentheses (* p<0.10, ** p<0.05, *** p<0.010). 25 Table 5: Lease price regressions for commercial agricultural land; sample with registry information All transactions With contract With tenant Tenant details & duration characteristics contract duration Area (ha) 0.080*** 0.066*** 0.084*** 0.072*** 0.092*** 0.081*** 0.098*** 0.088*** (0.003) (0.003) (0.003) (0.003) (0.003) (0.003) (0.003) (0.004) Area^2 0.001 -0.002 0.004 0.001 0.002 0.001 0.005* 0.004 (0.003) (0.003) (0.003) (0.003) (0.003) (0.003) (0.003) (0.003) Moratorium lifted 0.102*** 0.101*** 0.078*** 0.079*** 0.095*** 0.094*** 0.070*** 0.070*** (0.008) (0.008) (0.009) (0.009) (0.008) (0.008) (0.009) (0.009) Legal entity (renter) 0.366*** 0.270*** 0.381*** 0.281*** (0.017) (0.018) (0.019) (0.019) Legal entity * 2024 0.045 0.069* 0.056 0.081** (0.037) (0.037) (0.037) (0.037) # transactions 0.052*** 0.046*** 0.046*** 0.040*** 0.048*** 0.046*** 0.045*** 0.042*** in village (0.003) (0.003) (0.003) (0.003) (0.003) (0.003) (0.003) (0.003) Field area (ha) 0.064*** 0.069*** 0.041*** 0.046*** (0.007) (0.007) (0.008) (0.008) Field area^2 -0.004*** -0.004*** -0.003** -0.003*** (0.001) (0.001) (0.001) (0.001) Unmapped area -0.335*** -0.307*** -0.274*** -0.242*** share in field (0.042) (0.043) (0.045) (0.047) <1% of field area -0.011 -0.014 -0.011 -0.014 unregistered (0.008) (0.008) (0.008) (0.009) 1-10% of field area -0.034*** -0.034*** -0.024*** -0.024*** unregistered (0.007) (0.007) (0.007) (0.008) > 10% of field area -0.027*** -0.023** -0.028*** -0.024** unregistered (0.009) (0.010) (0.010) (0.010) <1% of field area 0.015 0.025** 0.014 0.020* reg. by owner only (0.010) (0.011) (0.011) (0.011) 1-10% of field area 0.011* 0.011* 0.021*** 0.023*** reg. by owner only (0.006) (0.007) (0.007) (0.007) >10% of field area 0.032** 0.015 0.062*** 0.041*** reg. by owner only (0.014) (0.015) (0.015) (0.016) Largest user’s cult. 0.128*** 0.117*** 0.133*** 0.120*** share of field area (0.018) (0.018) (0.019) (0.020) Buyer is the largest 0.079*** 0.084*** -0.007 -0.004 user in field (0.007) (0.008) (0.008) (0.009) >5 registered users -0.021** -0.026*** -0.026*** -0.029*** In field (0.009) (0.010) (0.010) (0.010) Lease term 14-20 0.010 0.006 0.003 0.004 years (0.007) (0.007) (0.008) (0.008) Lease term > 20 -0.220*** -0.225*** -0.229*** -0.228*** Years (0.009) (0.009) (0.009) (0.009) Tenant's cult. -0.021** -0.020** -0.007 -0.006 area (ha) (0.010) (0.010) (0.010) (0.010) Tenant's area^2 0.013*** 0.012*** 0.010*** 0.010*** (0.001) (0.001) (0.001) (0.001) Tenant uses <33% 0.023 0.019 0.011 0.006 of village land (0.019) (0.019) (0.019) (0.019) Tenant uses 33-50% -0.035* -0.043** -0.015 -0.024 of village land (0.020) (0.020) (0.021) (0.021) Tenant uses >50% -0.022 -0.042** -0.005 -0.025 of village land (0.020) (0.020) (0.021) (0.021) No. of obs (farms) 79,316 79,316 74,773 74,773 71,312 71,312 67,305 67,305 Adj R-squared 0.439 0.444 0.450 0.455 0.466 0.468 0.477 0.479 Note: Parcel characteristics, year dummies, OTG fixed effects and a constant included but not reported (see appendix table 4 for estimated coefficients). Standard errors in parentheses (* p<0.10, ** p<0.05, *** p<0.010). 26 Table 6: Lease price regressions for agricultural land for personal use All transactions With contract With tenant Tenant details & duration characteristics contract duration Area (ha) -0.007** -0.017*** -0.006** -0.017*** 0.030*** 0.021*** 0.030*** 0.022*** (0.003) (0.003) (0.003) (0.003) (0.003) (0.004) (0.004) (0.004) Area^2 -0.006** -0.009*** -0.002 -0.005* 0.003 0.002 0.006** 0.005* (0.003) (0.003) (0.003) (0.003) (0.003) (0.003) (0.003) (0.003) Moratorium lifted 0.193*** 0.192*** 0.141*** 0.140*** 0.222*** 0.221*** 0.183*** 0.182*** (0.005) (0.005) (0.006) (0.006) (0.007) (0.007) (0.008) (0.008) Legal entity (renter) 0.331*** 0.316*** 0.360*** 0.344*** (0.009) (0.009) (0.009) (0.009) Legal entity * 2024 0.028 0.028 0.004 0.004 (0.018) (0.018) (0.019) (0.019) # transactions 0.036*** 0.030*** 0.031*** 0.026*** 0.040*** 0.036*** 0.038*** 0.034*** in village (0.002) (0.002) (0.002) (0.002) (0.002) (0.002) (0.002) (0.002) Parcel part of field -0.009 -0.008 0.033*** 0.026*** (0.007) (0.007) (0.009) (0.009) Field area (ha) -0.001 -0.001 -0.005* -0.002 (0.002) (0.002) (0.003) (0.003) Field area^2 0.006*** 0.006*** 0.004*** 0.004*** (0.000) (0.000) (0.001) (0.001) Unmapped area -0.086*** -0.091*** -0.099*** -0.095*** share in field (0.017) (0.017) (0.020) (0.021) Lease term 14-20 0.028*** 0.024*** -0.058*** -0.059*** years (0.006) (0.006) (0.007) (0.007) Lease term > 20 -0.279*** -0.281*** -0.297*** -0.298*** Years (0.006) (0.006) (0.008) (0.008) Tenant's cult. -0.011** -0.010** -0.008* -0.006 area (ha) (0.004) (0.004) (0.004) (0.004) Tenant's area^2 0.010*** 0.010*** 0.010*** 0.010*** (0.001) (0.001) (0.001) (0.001) Tenant uses <33% -0.070*** -0.071*** -0.068*** -0.069*** of village land (0.008) (0.008) (0.008) (0.008) Tenant uses 33-50% -0.040*** -0.044*** -0.031*** -0.035*** of village land (0.010) (0.010) (0.011) (0.011) Tenant uses >50% -0.036*** -0.042*** -0.031*** -0.038*** of village land (0.011) (0.011) (0.011) (0.011) No. of obs (farms) 236,644 236,644 219,904 219,904 151,474 151,474 141,921 141,921 Adj R-squared 0.346 0.347 0.354 0.356 0.379 0.379 0.388 0.389 Note: Parcel characteristics, year dummies, OTG fixed effects and a constant included but not reported (see appendix table 3 for estimated coefficients). Standard errors in parentheses (* p<0.10, ** p<0.05, *** p<0.010). 27 Table 7: Sales price regressions for agricultural land By type of land Commercial agric. land Commercial Personal With registry data With buyer data Area (ha) 0.1081*** -0.2159*** 0.2584*** 0.2377*** 0.1897*** 0.3439*** (0.0022) (0.0046) (0.0052) (0.0054) (0.0086) (0.0158) Area^2 -0.0241*** 0.0563*** -0.0946*** -0.0911*** -0.0554*** -0.1083*** (0.0015) (0.0019) (0.0043) (0.0042) (0.0063) (0.0129) Buyer is legal entity 0.2607*** 0.4668*** 0.2095*** 0.2685*** (0.0051) (0.0136) (0.0131) (0.0198) # transactions -0.0121*** -0.0427*** -0.0073* -0.0107** 0.0043 -0.0523*** in village (0.0015) (0.0029) (0.0042) (0.0042) (0.0080) (0.0179) Field area (ha) -0.0034 -0.0063*** 0.0269*** -0.0088 0.0152 (0.0028) (0.0024) (0.0075) (0.0093) (0.0290) Field area^2 0.0050*** 0.0096*** 0.0011 0.0036** -0.0001 (0.0004) (0.0005) (0.0011) (0.0016) (0.0040) Unmapped area -0.3055*** -0.0351 -0.1825*** -0.3038*** 0.1545 share (0.0197) (0.0236) (0.0649) (0.1099) (0.2026) <1% of area 0.0045 -0.0050 unregistered (0.0097) (0.0268) 1-10% of area -0.0003 -0.0263 unregistered (0.0086) (0.0239) > 10% of area -0.0631*** -0.0944** unregistered (0.0117) (0.0375) Owner area <1% -0.0122 0.1030*** (0.0136) (0.0358) Owner area 1-10% -0.0189** 0.0045 (0.0076) (0.0221) Owner area > 10% -0.0091 0.0967* (0.0156) (0.0518) Largest user’s 0.2054*** 0.3306*** cult. area share (0.0203) (0.0654) Buyer is the largest -0.0987*** -0.1227*** user in field (0.0227) (0.0366) >5 registered users 0.0120 0.0120 (0.0121) (0.0385) Buyer's area (ha) -0.0667*** -0.0566 (0.0095) (0.0810) Buyer's area^2 0.0147*** 0.0215** (0.0014) (0.0084) Buyer uses <33% -0.2075*** -0.2691* of village land (0.0348) (0.1456) Buyer uses 33-50% -0.2922*** -0.2200 of village land (0.0384) (0.1565) Buyer uses >50% -0.3082*** -0.3289** of village land (0.0377) (0.1554) No. of obs (farms) 173,651 97,111 25,719 25,719 9,166 2,135 Adj R-squared 0.3820 0.5067 0.4904 0.4999 0.6649 0.8173 Note: Parcel characteristics, year dummies, OTG fixed effects and a constant included but not reported (see appendix table 5 for estimated coefficients). Standard errors in parentheses (* p<0.10, ** p<0.05, *** p<0.010). 28 Appendix table 1: Land market transactions by time period Land sales Land leases No. Area (ha) Sale prices No. Area (ha) Rental price Median Mean Median Mean Jan 21 54,362 129,188 70 91 Feb 21 68,324 154,433 86 100 Mar 21 95,493 214,378 85 99 Apr 21 68,575 133,158 75 94 May 21 43,203 95,601 82 101 Jun 21 52,125 116,352 103 113 July 21 3,076 5,802 1,269 4,599 48,687 107,045 94 122 Aug 21 8,266 20,175 1,173 3,050 52,286 120,489 112 138 Sep 21 15,115 38,950 1,151 2,556 62,182 136,413 94 124 Oct 21 15,269 41,164 1,167 2,405 56,031 154,925 100 135 Nov 21 21,278 58,498 1,134 2,029 65,265 137,804 96 136 Dec 21 27,070 73,818 1,076 1,959 88,084 192,844 91 134 Jan 22 11,489 29,436 1,040 1,818 68,690 153,134 83 133 Feb 22 18,163 48,981 1,058 1,385 72,484 142,596 93 163 Mar 22 7,517 17,297 124 148 Apr 22 8,018 15,073 140 136 May 22 512 961 1,116 1,561 9,404 17,666 131 144 Jun 22 4,352 8,164 1,090 1,673 23,899 45,105 114 148 July 22 5,764 10,685 1,059 1,577 24,873 46,124 97 127 Aug 22 6,429 12,062 922 1,382 28,938 52,717 68 99 Sep 22 6,781 13,332 901 1,375 32,919 63,555 75 107 Oct 22 6,420 12,487 909 1,356 26,309 52,668 75 105 Nov 22 6,553 13,662 933 1,318 29,847 61,277 72 103 Dec 22 8,362 17,591 877 1,199 45,033 95,949 74 110 Jan 23 4,382 9,286 912 1,288 49,088 110,191 73 102 Feb 23 7,586 16,197 907 1,359 39,542 85,472 92 120 Mar 23 10,950 23,634 902 1,321 51,375 111,494 85 111 Apr 23 9,364 20,821 917 1,405 42,687 91,008 87 113 May 23 11,148 23,842 930 1,480 46,198 96,238 98 116 Jun 23 10,474 22,099 959 1,549 45,972 102,783 102 116 Jul 23 9,661 20,136 955 1,644 37,822 81,580 102 122 Aug 23 10,759 23,103 956 1,608 42,364 91,602 91 121 Sep 23 11,094 24,523 939 1,631 46,147 96,312 87 112 Oct 23 11,157 25,039 958 1,647 47,979 108,141 95 116 Nov 23 11,595 25,386 957 1,633 50,265 113,012 92 112 Dec 23 11,545 25,642 925 1,623 52,667 121,834 94 118 Jan 24 8,656 20,532 963 1,571 58,594 132,888 80 109 Feb 24 13,858 34,300 949 1,488 43,138 100,063 86 112 Mar 24 14,252 32,914 947 1,540 49,458 114,051 85 110 Apr 24 15,235 34,364 958 1,468 48,827 113,121 86 113 May 24 13,457 31,647 985 1,545 32,946 79,125 95 117 Jun 24 11,985 28,043 942 1,542 24,077 54,456 91 134 Jul 24 13,266 30,505 968 1,585 15,231 36,783 91 117 Aug 24 14,021 32,311 940 1,589 17,394 43,593 88 112 Total 389,344 910,092 981 1,658 1,974,319 4,339,537 89 118 Source: Own computation based on parcel-level statistics from Ukraine registry of rights on property. Note: Transaction prices are in US$ equivalent for the sample for which prices are reported. 29 Appendix table 2: Sale and lease transaction with buyer information Individuals Legal entities Total 2021 2022 2023 2024 Total 2021 2022 2023 2024 Panel A: Leases No. of transactions 112,940 38,862 18,782 30,580 24,716 1,861,379 715,755 359,149 521,526 264,949 Area transacted 365,261 131,994 54,378 97,514 81,376 3,974,276 1,560,635 708,784 1,112,153 592,704 Price median 51.5 49.7 53.2 52.7 51.6 91.9 92.7 87.9 94.9 90.3 Price mean 86.0 84.3 94.4 85.4 82.7 119.3 117.3 129.9 116.2 116.8 # unique tenants 18,761 8,815 4,215 6,734 5,455 25,187 16,506 12,837 14,972 13,019 Mean area/tenant 19.5 14.97 12.90 14.48 14.92 157.79 94.55 55.21 74.28 45.53 Median area/tenant 7.20 6.99 7.01 7.09 7.93 19.38 24.52 17.89 20.60 15.22 Max area/tenant 694.7 694.7 278.1 377.9 303.9 35,106 35,106 14,605 13,952 5,905 Share w. data 0.73 0.65 0.74 0.77 0.78 Reg. area tenant 1,031 1,072 1,107 946 1,013 Cul. area/tenant 895 933 988 803 874 Panel B: Sales No. of transactions 369,267 89,204 74,003 119,331 86,729 20,077 870 822 384 18,001 Area transacted 853,455 236,896 165,845 258,639 192,075 56,637 1,511 1,516 1,069 52,540 Price median 969 1,133 990 935 920 1,249 1,134 779 2,074 1,259 Price mean 1,650 2,320 1,458 1,524 1,515 1,777 3,742 2,142 6,336 1,651 # unique buyers 100,085 25,904 26,477 41,268 30,597 1,532 141 98 126 1,275 Mean area/buyer 8.5 9.15 6.26 6.27 6.28 37.0 10.72 15.47 8.48 41.21 Median area/buyer 1.48 2.00 1.78 1.00 1.00 6.00 2.00 2.35 1.01 7.42 Max area/buyer 3,195 286 179 285 3,195 2,233 154 570 521 2,233 Share w. data 0.59 0.33 0.37 0.32 0.67 Reg. area buyer 2,102 2,480 1,398 1,410 2,144 Cul. area/buyer 1,850 2,103 1,228 1,328 1,887 Source: Own computation based on parcel-level statistics from Ukraine registry of rights on property. Note: Figures on area are in hectares throughout. Over the four-year period, there were a total of 43,948 unique tenants (25,187 legal entities and 18,761 individuals) who, on average, rented 98.7 ha (157.8 ha for legal entities and 19.5 ha for individuals) and 101,617 unique buyers (1,532 legal entities and 100,085 individuals) who bought on average 9 ha (37 ha for legal entities and 8.5 ha for individuals). 30 Appendix table 3: Parcel attributes of lease transactions Total By type w. tenant characteristics w. registry Comm. Personal Comm. Personal Comm Panel A: Distances Tertiary road 2.09 2.08 2.12 2.05 2.06 2.01 Secondary road 7.04 7.21 6.57 7.14 6.47 7.06 Primary road(km) 12.11 12.00 12.42 12.35 12.17 11.55 Motorway 26.95 27.73 24.81 25.74 23.86 27.59 Railway 9.93 9.65 10.69 9.74 10.43 9.83 Port 408.29 397.94 436.64 392.97 441.97 408.69 Grain elevator 12.95 12.74 13.52 12.59 12.77 11.70 Nearest city 54.54 54.69 54.13 55.13 53.96 55.24 Panel B: Area shares Share cultivated area 0.774 0.803 0.694 0.804 0.704 0.902 Share grassland 0.198 0.170 0.275 0.168 0.264 0.086 Share forested area 0.021 0.021 0.022 0.021 0.023 0.009 Share built-up area 0.002 0.002 0.002 0.002 0.002 0.001 Share other land 0.005 0.005 0.007 0.005 0.007 0.002 Panel C: Soil indicators Bulk density (g/cm^3) 1.310 1.311 1.306 1.308 1.306 1.296 Total nitrogen (g/kg) 4.145 4.074 4.342 4.083 4.340 4.137 Soil pH) 6.770 6.762 6.791 6.764 6.786 6.772 Organic carbon (g/kg) 5.674 5.710 5.575 5.707 5.551 5.743 Number of observations 884,936 648,292 236,644 382,653 151,474 79,316 Source: Own computation based on land monitoring data reported by SGC and complemented by administrative and remotely sensed data as discussed in the text. 31 Appendix table 4: Parcel attributes for parcels involved in sales transactions Total By type Commercial land with data on .. Comm. Personal Registry Buyer char’s (BC) Reg. & BC Panel A: Distances Tertiary road 2.06 2.19 1.83 2.11 2.25 2.09 Secondary road 6.48 6.75 6.00 6.44 7.04 5.87 Primary road(km) 11.66 12.17 10.76 11.50 13.12 13.76 Motorway 26.04 27.22 23.93 29.94 34.78 33.21 Railway 9.82 10.05 9.42 9.79 10.63 10.48 Port 360.97 340.43 397.72 388.22 381.39 385.75 Grain elevator 12.18 11.81 12.84 11.14 11.70 11.10 Nearest city 51.26 53.57 47.14 54.51 51.57 44.12 Panel B: Area shares Share cultivated area 0.767 0.882 0.560 0.936 0.891 0.948 Share grassland 0.194 0.099 0.364 0.053 0.090 0.043 Share forested area 0.024 0.013 0.044 0.007 0.013 0.006 Share built-up area 0.009 0.002 0.022 0.002 0.002 0.002 Share other land 0.006 0.004 0.010 0.002 0.004 0.002 Panel C: Soil indicators Bulk density (g/cm^3) 1.309 1.308 1.311 1.294 1.290 1.285 Total nitrogen (g/kg) 4.007 3.945 4.117 4.112 4.210 4.301 Soil pH) 6.822 6.866 6.742 6.832 6.865 6.892 Organic carbon (g/kg) 5.844 5.951 5.653 5.923 6.101 6.275 Number of observations 270,762 173,651 97,111 25,719 8,544 2,064 Source: Own computation based on land monitoring data reported by SGC and complemented by administrative and remotely sensed data as discussed in the text. 32 Appendix table 5: Lease price regressions for agricultural land for commercial use All transactions With contract With tenant Tenant details & duration characteristics contract duration Area (ha) 0.071*** 0.061*** 0.068*** 0.059*** 0.068*** 0.060*** 0.069*** 0.061*** (0.001) (0.001) (0.001) (0.001) (0.002) (0.002) (0.002) (0.002) Area^2 -0.004*** -0.004*** -0.004*** -0.004*** 0.009*** 0.009*** 0.009*** 0.009*** (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) Moratorium lifted 0.152*** 0.153*** 0.117*** 0.118*** 0.158*** 0.158*** 0.118*** 0.119*** (0.003) (0.003) (0.004) (0.004) (0.004) (0.004) (0.005) (0.005) Legal entity (renter) 0.353*** 0.333*** 0.361*** 0.341*** (0.005) (0.005) (0.005) (0.005) Legal entity * 2024 0.081*** 0.085*** 0.071*** 0.075*** (0.011) (0.011) (0.011) (0.011) No of transactions 0.037*** 0.034*** 0.033*** 0.030*** 0.044*** 0.042*** 0.044*** 0.042*** in village (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) Parcel part of a field -0.055*** -0.061*** -0.044*** -0.046*** (0.005) (0.005) (0.007) (0.007) Field area (ha) 0.014*** 0.018*** 0.016*** 0.018*** (0.002) (0.002) (0.003) (0.003) Field area^2 0.004*** 0.003*** 0.001** 0.000 (0.000) (0.000) (0.000) (0.000) Unmapped area -0.329*** -0.321*** -0.322*** -0.313*** Share in field (0.012) (0.013) (0.016) (0.016) Lease term 14-20 years 0.019*** 0.017*** -0.053*** -0.053*** (0.003) (0.003) (0.004) (0.004) Lease term >20 years -0.159*** -0.163*** -0.212*** -0.213*** (0.003) (0.003) (0.004) (0.004) Tenant's cultivated area (ha) -0.052*** -0.052*** -0.037*** -0.038*** (0.003) (0.003) (0.003) (0.003) Tenant's cultivated area^2 0.012*** 0.012*** 0.011*** 0.011*** (0.000) (0.000) (0.000) (0.000) Tenant's uses <33% of -0.105*** -0.102*** -0.116*** -0.113*** village land (0.006) (0.006) (0.006) (0.006) Tenant uses 33-50% of -0.112*** -0.112*** -0.110*** -0.109*** village land (0.007) (0.007) (0.007) (0.007) Tenant uses >50% of -0.064*** -0.067*** -0.064*** -0.066*** village land (0.007) (0.007) (0.007) (0.007) Dist. to tertiary road (km) -0.021*** -0.025*** -0.021*** -0.025*** -0.014*** -0.016*** -0.013*** -0.015*** (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) Dist. to secondary road (km) -0.009*** -0.010*** -0.009*** -0.010*** -0.016*** -0.016*** -0.016*** -0.017*** (0.001) (0.001) (0.001) (0.001) (0.002) (0.002) (0.002) (0.002) Dist. to primary road (km) -0.030*** -0.030*** -0.028*** -0.029*** -0.025*** -0.024*** -0.023*** -0.022*** (0.001) (0.001) (0.001) (0.001) (0.002) (0.002) (0.002) (0.002) Dist. to motorway (km) 0.012*** 0.012*** 0.012*** 0.013*** 0.024*** 0.024*** 0.026*** 0.027*** (0.002) (0.002) (0.002) (0.002) (0.002) (0.002) (0.002) (0.002) Dist. to railway (km) 0.003** 0.003** 0.001 0.001 -0.007*** -0.007*** -0.010*** -0.010*** (0.001) (0.001) (0.001) (0.001) (0.002) (0.002) (0.002) (0.002) Dist. to port (km) 0.061*** 0.067*** 0.060*** 0.064*** 0.289*** 0.289*** 0.288*** 0.287*** (0.020) (0.020) (0.021) (0.021) (0.026) (0.026) (0.027) (0.027) Dist. to elevator (km) -0.037*** -0.035*** -0.035*** -0.034*** -0.028*** -0.028*** -0.023*** -0.023*** (0.002) (0.002) (0.002) (0.002) (0.003) (0.003) (0.003) (0.003) Dist. to nearest city (km) 0.068*** 0.065*** 0.070*** 0.067*** 0.090*** 0.088*** 0.097*** 0.095*** (0.005) (0.005) (0.005) (0.005) (0.006) (0.006) (0.006) (0.006) Share built-up area 0.919*** 0.908*** 0.944*** 0.933*** 0.633*** 0.630*** 0.643*** 0.641*** (0.038) (0.038) (0.039) (0.039) (0.052) (0.052) (0.052) (0.052) Share cultivated area 0.575*** 0.531*** 0.571*** 0.525*** 0.517*** 0.495*** 0.512*** 0.489*** (0.019) (0.019) (0.019) (0.019) (0.022) (0.022) (0.022) (0.022) Share forested area 0.307*** 0.281*** 0.303*** 0.277*** 0.208*** 0.184*** 0.209*** 0.185*** (0.022) (0.022) (0.022) (0.022) (0.025) (0.025) (0.025) (0.025) Share grassland 0.290*** 0.291*** 0.282*** 0.281*** 0.255*** 0.261*** 0.249*** 0.253*** (0.020) (0.019) (0.020) (0.020) (0.022) (0.022) (0.022) (0.022) Soil bulk density (g/cm^3) -0.731 -0.450 -0.915** -0.647 -3.860*** -3.672*** -3.155*** -2.994*** 33 (0.446) (0.445) (0.464) (0.463) (0.564) (0.564) (0.583) (0.583) Density^2 -1.650* -1.753** -1.467 -1.540* 3.917*** 3.886*** 2.426** 2.435** (0.871) (0.870) (0.906) (0.904) (1.109) (1.108) (1.145) (1.144) Total nitrogen (g/kg) 0.526*** 0.493*** 0.454*** 0.420*** 0.443*** 0.412*** 0.442*** 0.411*** (0.064) (0.064) (0.066) (0.066) (0.083) (0.083) (0.085) (0.085) Nitrogen^2 -0.235*** -0.213*** -0.205*** -0.183*** -0.186*** -0.169*** -0.181*** -0.165*** (0.022) (0.022) (0.023) (0.023) (0.028) (0.028) (0.029) (0.029) Soil pH (pH) 1.092 2.621* 1.474 2.984** -3.879** -2.449 -2.438 -1.106 (1.366) (1.364) (1.415) (1.413) (1.779) (1.778) (1.834) (1.834) pH^2 0.101 -0.370 0.003 -0.464 1.465*** 1.045** 1.070** 0.678 (0.368) (0.367) (0.381) (0.381) (0.479) (0.479) (0.494) (0.494) Soil organic carbon stock (g/kg) -1.216*** -1.188*** -0.594*** -0.555** -2.006*** -1.956*** -1.579*** -1.529*** (0.214) (0.213) (0.222) (0.222) (0.272) (0.272) (0.280) (0.280) Organic carbon^2 0.393*** 0.374*** 0.210*** 0.187*** 0.603*** 0.583*** 0.482*** 0.462*** (0.062) (0.061) (0.064) (0.064) (0.078) (0.078) (0.081) (0.081) year=2022 -0.118*** -0.119*** -0.123*** -0.124*** -0.143*** -0.142*** -0.147*** -0.146*** (0.003) (0.003) (0.003) (0.003) (0.004) (0.004) (0.004) (0.004) year=2023 -0.177*** -0.178*** -0.181*** -0.182*** -0.245*** -0.245*** -0.253*** -0.253*** (0.003) (0.003) (0.003) (0.003) (0.004) (0.004) (0.004) (0.004) year=2024 -0.225*** -0.230*** -0.221*** -0.228*** -0.227*** -0.228*** -0.238*** -0.238*** (0.011) (0.011) (0.011) (0.011) (0.005) (0.005) (0.005) (0.005) Constant 1.318 0.078 0.568 -0.643 5.987*** 4.756*** 4.231** 3.095* (1.274) (1.273) (1.320) (1.318) (1.658) (1.657) (1.710) (1.710) No. of obs (farms) 648,292 648,292 603,648 603,648 382,653 382,653 359,323 359,323 Adj R-squared 0.338 0.341 0.344 0.347 0.388 0.390 0.398 0.399 Note: OTG fixed effects are included. Standard errors in parentheses (* p<0.10, ** p<0.05, *** p<0.010). 34 Appendix table 6: Lease price regressions for agricultural land for commercial use All transactions With contract duration With tenant Tenant details & characteristics contract duration Area (ha) 0.080*** 0.066*** 0.084*** 0.072*** 0.092*** 0.081*** 0.098*** 0.088*** (0.003) (0.003) (0.003) (0.003) (0.003) (0.003) (0.003) (0.004) Area^2 0.001 -0.002 0.004 0.001 0.002 0.001 0.005* 0.004 (0.003) (0.003) (0.003) (0.003) (0.003) (0.003) (0.003) (0.003) Moratorium lifted 0.102*** 0.101*** 0.078*** 0.079*** 0.095*** 0.094*** 0.070*** 0.070*** (0.008) (0.008) (0.009) (0.009) (0.008) (0.008) (0.009) (0.009) Legal entity (renter) 0.366*** 0.270*** 0.381*** 0.281*** (0.017) (0.018) (0.019) (0.019) Legal entity * 2024 0.045 0.069* 0.056 0.081** (0.037) (0.037) (0.037) (0.037) # transactions 0.052*** 0.046*** 0.046*** 0.040*** 0.048*** 0.046*** 0.045*** 0.042*** in village (0.003) (0.003) (0.003) (0.003) (0.003) (0.003) (0.003) (0.003) Field area (ha) 0.064*** 0.069*** 0.041*** 0.046*** (0.007) (0.007) (0.008) (0.008) Field area^2 -0.004*** -0.004*** -0.003** -0.003*** (0.001) (0.001) (0.001) (0.001) Unmapped area -0.335*** -0.307*** -0.274*** -0.242*** share in field (0.042) (0.043) (0.045) (0.047) <1% of field area -0.011 -0.014 -0.011 -0.014 unregistered (0.008) (0.008) (0.008) (0.009) 1-10% of field area -0.034*** -0.034*** -0.024*** -0.024*** unregistered (0.007) (0.007) (0.007) (0.008) > 10% of field area -0.027*** -0.023** -0.028*** -0.024** unregistered (0.009) (0.010) (0.010) (0.010) <1% of field area 0.015 0.025** 0.014 0.020* reg. by owner only (0.010) (0.011) (0.011) (0.011) 1-10% of field area 0.011* 0.011* 0.021*** 0.023*** reg. by owner only (0.006) (0.007) (0.007) (0.007) >10% of field area 0.032** 0.015 0.062*** 0.041*** reg. by owner only (0.014) (0.015) (0.015) (0.016) Largest user’s cult. 0.128*** 0.117*** 0.133*** 0.120*** share of field area (0.018) (0.018) (0.019) (0.020) Buyer is the largest 0.079*** 0.084*** -0.007 -0.004 user in field (0.007) (0.008) (0.008) (0.009) >5 registered users -0.021** -0.026*** -0.026*** -0.029*** in field (0.009) (0.010) (0.010) (0.010) Lease term 14-20 0.010 0.006 0.003 0.004 years (0.007) (0.007) (0.008) (0.008) Lease term > 20 -0.220*** -0.225*** -0.229*** -0.228*** Years (0.009) (0.009) (0.009) (0.009) Tenant's cult. -0.021** -0.020** -0.007 -0.006 area (ha) (0.010) (0.010) (0.010) (0.010) Tenant's cult. area^2 0.013*** 0.012*** 0.010*** 0.010*** (0.001) (0.001) (0.001) (0.001) Tenant uses <33% 0.023 0.019 0.011 0.006 of village land (0.019) (0.019) (0.019) (0.019) Tenant uses 33-50% -0.035* -0.043** -0.015 -0.024 of village land (0.020) (0.020) (0.021) (0.021) Tenant uses >50% -0.022 -0.042** -0.005 -0.025 of village land (0.020) (0.020) (0.021) (0.021) Dist. to tertiary -0.003 -0.006** -0.002 -0.005* -0.003 -0.005* -0.002 -0.003 road (km) (0.003) (0.003) (0.003) (0.003) (0.003) (0.003) (0.003) (0.003) Dist. to secondary -0.021*** -0.024*** -0.019*** -0.022*** -0.024*** -0.026*** -0.020*** -0.022*** road (km) (0.003) (0.003) (0.003) (0.003) (0.003) (0.003) (0.004) (0.004) Dist. to primary -0.063*** -0.059*** -0.060*** -0.056*** -0.052*** -0.050*** -0.047*** -0.046*** road (km) (0.003) (0.003) (0.004) (0.004) (0.004) (0.004) (0.004) (0.004) Dist. to motorway (km) -0.016*** -0.017*** -0.018*** -0.020*** -0.016*** -0.017*** -0.017*** -0.018*** (0.005) (0.005) (0.005) (0.005) (0.005) (0.005) (0.005) (0.005) Dist. to railway (km) -0.014*** -0.016*** -0.020*** -0.021*** -0.027*** -0.027*** -0.031*** -0.031*** 35 (0.004) (0.004) (0.004) (0.004) (0.004) (0.004) (0.004) (0.004) Dist. to port (km) 0.980*** 0.908*** 1.068*** 1.002*** 0.876*** 0.874*** 0.980*** 0.982*** (0.115) (0.114) (0.118) (0.117) (0.122) (0.122) (0.125) (0.125) Dist. to elevator (km) 0.063*** 0.060*** 0.057*** 0.055*** 0.056*** 0.053*** 0.051*** 0.049*** (0.007) (0.007) (0.007) (0.007) (0.007) (0.007) (0.007) (0.007) Dist. to nearest 0.079*** 0.079*** 0.095*** 0.097*** 0.107*** 0.105*** 0.123*** 0.122*** city (km) (0.016) (0.016) (0.017) (0.016) (0.017) (0.017) (0.017) (0.017) Share built-up area 0.199 0.205 0.209 0.210 0.125 0.128 0.128 0.133 (0.145) (0.145) (0.147) (0.146) (0.150) (0.150) (0.152) (0.152) Share cultivated area 0.148* 0.074 0.128 0.053 0.175** 0.124 0.164* 0.111 (0.085) (0.085) (0.086) (0.085) (0.088) (0.088) (0.090) (0.090) Share forested area -0.077 -0.103 -0.051 -0.079 -0.036 -0.069 0.003 -0.028 (0.097) (0.097) (0.099) (0.099) (0.101) (0.101) (0.103) (0.103) Share grassland 0.011 0.003 -0.009 -0.019 0.043 0.035 0.033 0.023 (0.086) (0.085) (0.087) (0.086) (0.089) (0.089) (0.091) (0.090) Soil bulk density 2.444** 2.447** 2.362* 2.314* 2.528** 2.425** 2.745** 2.615** (g/cm^3) (1.179) (1.175) (1.216) (1.211) (1.213) (1.211) (1.250) (1.249) Density^2 -6.867*** -6.472*** -6.433*** -5.959** -7.618*** -7.153*** -7.670*** -7.176*** (2.360) (2.351) (2.432) (2.423) (2.426) (2.423) (2.500) (2.498) Total nitrogen (g/kg) -0.552*** -0.538*** -0.473** -0.449** -0.493** -0.483** -0.458** -0.449** (0.208) (0.207) (0.214) (0.213) (0.218) (0.217) (0.224) (0.224) Nitrogen^2 0.107 0.111 0.080 0.080 0.079 0.081 0.066 0.068 (0.071) (0.071) (0.073) (0.073) (0.075) (0.075) (0.077) (0.077) Soil pH (pH) -15.301*** -12.574*** -15.936*** -13.019*** -10.389** -9.357* -8.681* -7.621 (4.742) (4.728) (4.913) (4.897) (4.927) (4.923) (5.119) (5.116) pH^2 4.521*** 3.732*** 4.689*** 3.845*** 3.202** 2.891** 2.751** 2.430* (1.265) (1.262) (1.311) (1.307) (1.315) (1.314) (1.366) (1.365) Soil organic carbon -2.749*** -2.554*** -2.617*** -2.411*** -3.037*** -3.018*** -2.942*** -2.921*** stock (g/kg) (0.582) (0.580) (0.603) (0.601) (0.597) (0.597) (0.619) (0.618) Organic carbon^2 0.842*** 0.779*** 0.805*** 0.738*** 0.934*** 0.923*** 0.910*** 0.897*** (0.167) (0.166) (0.173) (0.172) (0.171) (0.171) (0.177) (0.177) year=2022 -0.079*** -0.080*** -0.086*** -0.086*** -0.116*** -0.112*** -0.122*** -0.118*** (0.008) (0.008) (0.009) (0.008) (0.009) (0.009) (0.009) (0.009) year=2023 -0.216*** -0.214*** -0.223*** -0.220*** -0.279*** -0.272*** -0.287*** -0.280*** (0.008) (0.008) (0.008) (0.008) (0.008) (0.008) (0.008) (0.008) year=2024 -0.246*** -0.270*** -0.264*** -0.290*** -0.277*** -0.271*** -0.285*** -0.278*** (0.037) (0.036) (0.037) (0.037) (0.010) (0.010) (0.010) (0.010) Constant 13.134*** 10.924** 13.068*** 10.641** 9.394** 8.450* 7.017 6.023 (4.451) (4.437) (4.607) (4.593) (4.629) (4.626) (4.806) (4.803) No. of obs (farms) 79,316 79,316 74,773 74,773 71,312 71,312 67,305 67,305 Adj R-squared 0.439 0.444 0.450 0.455 0.466 0.468 0.477 0.479 Note: OTG fixed effects are included. Standard errors in parentheses (* p<0.10, ** p<0.05, *** p<0.010). 36 Appendix table 7: Lease price regressions for agricultural land for personal use All transactions With contract duration With tenant Tenant details & characteristics contract duration Area (ha) -0.007** -0.017*** -0.006** -0.017*** 0.030*** 0.021*** 0.030*** 0.022*** (0.003) (0.003) (0.003) (0.003) (0.003) (0.004) (0.004) (0.004) Area^2 -0.006** -0.009*** -0.002 -0.005* 0.003 0.002 0.006** 0.005* (0.003) (0.003) (0.003) (0.003) (0.003) (0.003) (0.003) (0.003) Moratorium lifted 0.193*** 0.192*** 0.141*** 0.140*** 0.222*** 0.221*** 0.183*** 0.182*** (0.005) (0.005) (0.006) (0.006) (0.007) (0.007) (0.008) (0.008) Legal entity (renter) 0.331*** 0.316*** 0.360*** 0.344*** (0.009) (0.009) (0.009) (0.009) Legal entity * 2024 0.028 0.028 0.004 0.004 (0.018) (0.018) (0.019) (0.019) # transactions 0.036*** 0.030*** 0.031*** 0.026*** 0.040*** 0.036*** 0.038*** 0.034*** in village (0.002) (0.002) (0.002) (0.002) (0.002) (0.002) (0.002) (0.002) Parcel part of field -0.009 -0.008 0.033*** 0.026*** (0.007) (0.007) (0.009) (0.009) Field area (ha) -0.001 -0.001 -0.005* -0.002 (0.002) (0.002) (0.003) (0.003) Field area^2 0.006*** 0.006*** 0.004*** 0.004*** (0.000) (0.000) (0.001) (0.001) Unmapped area -0.086*** -0.091*** -0.099*** -0.095*** share in field (0.017) (0.017) (0.020) (0.021) Lease term 14-20 0.028*** 0.024*** -0.058*** -0.059*** years (0.006) (0.006) (0.007) (0.007) Lease term > 20 -0.279*** -0.281*** -0.297*** -0.298*** Years (0.006) (0.006) (0.008) (0.008) Tenant's cult. -0.011** -0.010** -0.008* -0.006 area (ha) (0.004) (0.004) (0.004) (0.004) Tenant's area^2 0.010*** 0.010*** 0.010*** 0.010*** (0.001) (0.001) (0.001) (0.001) Tenant uses <33% -0.070*** -0.071*** -0.068*** -0.069*** of village land (0.008) (0.008) (0.008) (0.008) Tenant uses 33-50% -0.040*** -0.044*** -0.031*** -0.035*** of village land (0.010) (0.010) (0.011) (0.011) Tenant uses >50% -0.036*** -0.042*** -0.031*** -0.038*** of village land (0.011) (0.011) (0.011) (0.011) Dist. to tertiary -0.014*** -0.019*** -0.015*** -0.020*** -0.019*** -0.022*** -0.020*** -0.023*** road (km) (0.002) (0.002) (0.002) (0.002) (0.002) (0.002) (0.002) (0.002) Dist. to secondary -0.009*** -0.011*** -0.008*** -0.011*** -0.028*** -0.030*** -0.026*** -0.028*** road (km) (0.002) (0.002) (0.002) (0.002) (0.002) (0.002) (0.003) (0.003) Dist. to primary 0.006** 0.005** 0.009*** 0.008*** 0.016*** 0.015*** 0.020*** 0.019*** road (km) (0.002) (0.002) (0.002) (0.002) (0.003) (0.003) (0.003) (0.003) Dist. to motorway (km) -0.029*** -0.027*** -0.025*** -0.023*** -0.044*** -0.043*** -0.042*** -0.042*** (0.003) (0.003) (0.003) (0.003) (0.004) (0.004) (0.004) (0.004) Dist. to railway (km) 0.012*** 0.011*** 0.015*** 0.014*** 0.024*** 0.024*** 0.027*** 0.026*** (0.002) (0.002) (0.002) (0.002) (0.003) (0.003) (0.003) (0.003) Dist. to port (km) 0.208*** 0.226*** 0.258*** 0.278*** 0.209*** 0.222*** 0.320*** 0.335*** (0.041) (0.041) (0.042) (0.042) (0.051) (0.051) (0.053) (0.053) Dist. to elevator (km) -0.045*** -0.045*** -0.044*** -0.044*** -0.042*** -0.042*** -0.042*** -0.042*** (0.004) (0.004) (0.004) (0.004) (0.005) (0.005) (0.005) (0.005) Dist. to nearest city (km) -0.134*** -0.131*** -0.148*** -0.145*** -0.134*** -0.132*** -0.139*** -0.136*** (0.008) (0.008) (0.008) (0.008) (0.011) (0.011) (0.011) (0.011) Share built-up area 0.501*** 0.494*** 0.510*** 0.504*** 0.446*** 0.446*** 0.503*** 0.502*** (0.072) (0.072) (0.074) (0.073) (0.092) (0.092) (0.093) (0.093) Share cultivated area 0.303*** 0.264*** 0.331*** 0.291*** 0.398*** 0.368*** 0.407*** 0.376*** (0.029) (0.029) (0.029) (0.029) (0.035) (0.035) (0.035) (0.035) Share forested area 0.139*** 0.146*** 0.157*** 0.165*** 0.179*** 0.198*** 0.184*** 0.201*** (0.034) (0.034) (0.034) (0.034) (0.040) (0.040) (0.041) (0.041) Share grassland 0.155*** 0.151*** 0.175*** 0.171*** 0.251*** 0.246*** 0.258*** 0.253*** (0.029) (0.029) (0.030) (0.030) (0.035) (0.035) (0.035) (0.035) Soil bulk density 1.140 1.094 2.092** 2.037** -1.836* -1.929* -1.086 -1.170 37 (g/cm^3) (0.874) (0.873) (0.909) (0.908) (1.069) (1.068) (1.104) (1.103) Density^2 -3.333* -3.024* -5.002*** -4.665*** 1.797 2.166 0.408 0.763 (1.714) (1.712) (1.780) (1.777) (2.091) (2.090) (2.158) (2.156) Total nitrogen (g/kg) 0.325*** 0.338*** 0.427*** 0.433*** 0.592*** 0.602*** 0.673*** 0.679*** (0.098) (0.098) (0.102) (0.102) (0.121) (0.121) (0.126) (0.126) Nitrogen^2 -0.175*** -0.173*** -0.210*** -0.205*** -0.270*** -0.268*** -0.301*** -0.298*** (0.032) (0.032) (0.033) (0.033) (0.040) (0.040) (0.041) (0.041) Soil pH (pH) -10.174*** -8.827*** -10.577*** -9.235*** -5.053 -4.921 -7.414** -7.348** (2.776) (2.774) (2.884) (2.882) (3.443) (3.443) (3.579) (3.578) pH^2 2.969*** 2.566*** 3.050*** 2.644*** 1.622* 1.555* 2.248** 2.198** (0.740) (0.740) (0.769) (0.768) (0.918) (0.918) (0.954) (0.953) Soil organic carbon -2.128*** -1.979*** -2.002*** -1.846*** -0.443 -0.423 -0.369 -0.342 stock (g/kg) (0.348) (0.348) (0.361) (0.361) (0.442) (0.441) (0.456) (0.456) Organic carbon^2 0.655*** 0.596*** 0.620*** 0.559*** 0.183 0.167 0.163 0.146 (0.102) (0.102) (0.106) (0.106) (0.129) (0.129) (0.133) (0.133) year=2022 -0.138*** -0.136*** -0.139*** -0.137*** -0.160*** -0.157*** -0.162*** -0.160*** (0.005) (0.005) (0.005) (0.005) (0.007) (0.007) (0.007) (0.007) year=2023 -0.192*** -0.189*** -0.194*** -0.191*** -0.240*** -0.236*** -0.240*** -0.236*** (0.005) (0.005) (0.005) (0.005) (0.006) (0.006) (0.006) (0.006) year=2024 -0.221*** -0.216*** -0.209*** -0.204*** -0.258*** -0.248*** -0.269*** -0.259*** (0.018) (0.018) (0.018) (0.018) (0.008) (0.008) (0.008) (0.008) Constant 13.135*** 11.833*** 13.067*** 11.760*** 7.127** 7.011** 8.536** 8.472** (2.604) (2.603) (2.705) (2.703) (3.233) (3.233) (3.360) (3.359) No. of obs (farms) 236,644 236,644 219,904 219,904 151,474 151,474 141,921 141,921 Adj R-squared 0.346 0.347 0.354 0.356 0.379 0.379 0.388 0.389 Note: OTG fixed effects are included. Standard errors in parentheses (* p<0.10, ** p<0.05, *** p<0.010). 38 Appendix table 8: Lease price regression for agricultural land for personal use purchased by individuals All transactions With contract duration Area (ha) -0.033** -0.039** -0.029* -0.033** (0.015) (0.015) (0.016) (0.016) Area^2 -0.040*** -0.038*** -0.045*** -0.043*** (0.012) (0.012) (0.013) (0.013) Moratorium lifted 0.069** 0.068** 0.019 0.018 (0.028) (0.028) (0.033) (0.033) No of transactions in village 0.028*** 0.027*** 0.022** 0.023** (0.009) (0.009) (0.009) (0.009) Parcel part of a field 0.051** 0.058** (0.025) (0.026) Field area (ha) 0.004 -0.004 (0.008) (0.008) Field area^2 -0.001 -0.000 (0.002) (0.002) Unmapped area share in field -0.079 -0.057 (0.074) (0.078) Lease term 14-20 years -0.055* -0.054* (0.032) (0.032) Lease term > 20 years -0.268*** -0.268*** (0.029) (0.029) Dist. to tertiary road (km) -0.012 -0.013 -0.013 -0.012 (0.008) (0.008) (0.008) (0.008) Dist. to secondary road (km) 0.002 0.002 0.007 0.007 (0.009) (0.009) (0.009) (0.009) Dist. to primary road (km) -0.008 -0.008 -0.015 -0.015 (0.010) (0.010) (0.011) (0.011) Dist. to motorway (km) -0.031** -0.031** -0.031** -0.031** (0.015) (0.015) (0.015) (0.015) Dist. to railway (km) -0.010 -0.010 -0.005 -0.005 (0.010) (0.010) (0.011) (0.011) Dist. to port (km) -0.546*** -0.548*** -0.544*** -0.545*** (0.167) (0.167) (0.176) (0.176) Dist. to elevator (km) -0.066*** -0.065*** -0.066*** -0.065*** (0.017) (0.017) (0.018) (0.018) Dist. to nearest city (km) -0.074* -0.073* -0.065 -0.065 (0.041) (0.041) (0.042) (0.042) Share built-up area 1.608*** 1.626*** 1.218*** 1.233*** (0.228) (0.228) (0.244) (0.244) Share cultivated area 0.657*** 0.640*** 0.636*** 0.623*** (0.142) (0.143) (0.143) (0.143) Share forested area 0.684*** 0.698*** 0.652*** 0.667*** (0.172) (0.172) (0.174) (0.175) Share grassland 0.526*** 0.517*** 0.492*** 0.482*** (0.143) (0.143) (0.144) (0.144) Soil bulk density (g/cm^3) -11.583*** -11.259*** -12.867*** -12.641*** (4.273) (4.275) (4.469) (4.471) Density^2 18.976** 18.453** 20.749** 20.382** (8.243) (8.246) (8.615) (8.619) Total nitrogen (g/kg) 0.707 0.679 0.834* 0.809* (0.455) (0.455) (0.471) (0.471) Nitrogen^2 -0.268* -0.257* -0.310** -0.300* (0.151) (0.151) (0.157) (0.157) Soil pH (pH) 25.064* 24.902* 36.596*** 35.850*** (13.273) (13.281) (13.707) (13.718) pH^2 -6.232* -6.195* -9.300** -9.104** (3.518) (3.520) (3.634) (3.637) Soil organic carbon stock (g/kg) -0.927 -0.958 0.523 0.490 (1.511) (1.511) (1.593) (1.594) Organic carbon^2 0.264 0.272 -0.134 -0.124 (0.428) (0.428) (0.452) (0.452) 39 year=2022 -0.034 -0.033 -0.044* -0.043* (0.026) (0.026) (0.026) (0.026) year=2023 -0.162*** -0.157*** -0.174*** -0.170*** (0.026) (0.026) (0.026) (0.026) year=2024 -0.124*** -0.108*** -0.137*** -0.120*** (0.028) (0.029) (0.028) (0.029) Constant -16.136 -15.983 -28.061** -27.365** (12.470) (12.478) (12.876) (12.886) No. of obs (farms) 13004 13004 12116 12116 Adj R-squared 0.478 0.478 0.488 0.488 Note: OTG fixed effects are included. Standard errors in parentheses (* p<0.10, ** p<0.05, *** p<0.010). 40 Appendix table 9: Sales price regressions for agricultural land By type of land With registry information With tenant characteristics Commercial Personal Commercial Area (ha) 0.1081*** -0.2159*** 0.2584*** 0.2377*** 0.1897*** 0.3439*** (0.0022) (0.0046) (0.0052) (0.0054) (0.0086) (0.0158) Area^2 -0.0241*** 0.0563*** -0.0946*** -0.0911*** -0.0554*** -0.1083*** (0.0015) (0.0019) (0.0043) (0.0042) (0.0063) (0.0129) Buyer is legal entity 0.2607*** 0.4668*** 0.2095*** 0.2685*** (0.0051) (0.0136) (0.0131) (0.0198) # transactions in village -0.0121*** -0.0427*** -0.0073* -0.0107** 0.0043 -0.0523*** (0.0015) (0.0029) (0.0042) (0.0042) (0.0080) (0.0179) Parcel part of a field -0.0506*** -0.1418*** (0.0071) (0.0087) Field area (ha) -0.0034 -0.0063*** 0.0269*** -0.0088 0.0152 (0.0028) (0.0024) (0.0075) (0.0093) (0.0290) Field area^2 0.0050*** 0.0096*** 0.0011 0.0036** -0.0001 (0.0004) (0.0005) (0.0011) (0.0016) (0.0040) Unmapped area share -0.3055*** -0.0351 -0.1825*** -0.3038*** 0.1545 (0.0197) (0.0236) (0.0649) (0.1099) (0.2026) <1% of area unregistered 0.0045 -0.0050 (0.0097) (0.0268) 1-10% of area unregistered -0.0003 -0.0263 (0.0086) (0.0239) >10% of area unregistered -0.0631*** -0.0944** (0.0117) (0.0375) Owner area <1% -0.0122 0.1030*** (0.0136) (0.0358) Owner area 1-10% -0.0189** 0.0045 (0.0076) (0.0221) Owner area >10% -0.0091 0.0967* (0.0156) (0.0518) Largest user's cult. area share 0.2054*** 0.3306*** (0.0203) (0.0654) Buyer is the largest user in field -0.0987*** -0.1227*** (0.0227) (0.0366) >5 registered users 0.0120 0.0120 (0.0121) (0.0385) Buyer's area (ha) -0.0667*** -0.0566 (0.0095) (0.0810) Buyer's area^2 0.0147*** 0.0215** (0.0014) (0.0084) Buyer uses <33% of village land -0.2075*** -0.2691* (0.0348) (0.1456) Buyer uses 33-50% of village -0.2922*** -0.2200 land (0.0384) (0.1565) Buyer uses >50% of village land -0.3082*** -0.3289** (0.0377) (0.1554) Dist. to tertiary road (km) -0.0132*** -0.0243*** -0.0011 -0.0042 -0.0206*** -0.0071 (0.0013) (0.0025) (0.0034) (0.0034) (0.0054) (0.0103) Dist. to secondary road (km) 0.0011 -0.0185*** 0.0064* 0.0031 0.0034 0.0219* (0.0014) (0.0028) (0.0039) (0.0038) (0.0063) (0.0116) Dist. to primary road (km) -0.0117*** -0.0251*** -0.0092** -0.0107** -0.0346*** -0.0459*** (0.0015) (0.0030) (0.0043) (0.0042) (0.0072) (0.0162) Dist. to motorway (km) -0.0085*** -0.0171*** 0.0123* 0.0084 -0.0016 -0.0420* (0.0019) (0.0037) (0.0063) (0.0063) (0.0112) (0.0233) Dist. to railway (km) -0.0118*** -0.0055* -0.0036 -0.0008 -0.0068 0.0045 (0.0017) (0.0033) (0.0047) (0.0047) (0.0080) (0.0206) Dist. to port (km) 0.0534** 0.0938* -0.1666 -0.2334 0.1315 -1.3476** (0.0224) (0.0515) (0.1474) (0.1461) (0.2009) (0.5469) Dist. to elevator (km) -0.0070** -0.0275*** -0.0157** -0.0199*** 0.0191 -0.0072 (0.0028) (0.0056) (0.0078) (0.0077) (0.0137) (0.0285) 41 Dist. to nearest city (km) -0.0078 -0.1110*** -0.0173 -0.0193 -0.0199 -0.1570*** (0.0049) (0.0083) (0.0158) (0.0157) (0.0247) (0.0532) Share built-up area 0.6308*** 0.3376*** -0.1655 -0.2078 0.5256*** -1.1449** (0.0512) (0.0422) (0.1466) (0.1453) (0.1692) (0.4786) Share cultivated area 0.5038*** 0.3322*** 0.0873 -0.0006 0.3988*** -1.4858*** (0.0271) (0.0359) (0.1022) (0.1018) (0.1066) (0.3849) Share forested area 0.3623*** 0.2188*** -0.0085 -0.0653 0.1752 -2.1104*** (0.0314) (0.0392) (0.1325) (0.1316) (0.1241) (0.4687) Share grassland 0.3250*** 0.2192*** -0.2149** -0.2226** 0.1608 -1.8231*** (0.0275) (0.0359) (0.1040) (0.1033) (0.1086) (0.3865) Soil bulk density (g/cm^3) -2.7403*** -2.2536** -2.0124 -1.6065 10.6046*** -1.9599 (0.5441) (0.9717) (1.6023) (1.5890) (2.6995) (5.3322) Density^2 2.7976*** 1.9991 1.7497 1.5165 -19.4749*** 7.4346 (1.0658) (1.9171) (3.1772) (3.1502) (5.4271) (10.4732) Total nitrogen (g/kg) 0.5217*** 0.9681*** 0.7147*** 0.6363** -0.4215 0.6306 (0.0799) (0.1579) (0.2563) (0.2542) (0.4455) (0.8302) Nitrogen^2 -0.2139*** -0.4255*** -0.2589*** -0.2287*** 0.1277 -0.1803 (0.0283) (0.0541) (0.0887) (0.0879) (0.1512) (0.2802) Soil pH (pH) -6.4245*** -11.0528*** -33.4228*** -28.0613*** -26.2017*** -52.7881** (2.1347) (3.3933) (6.6996) (6.6453) (9.7158) (25.0955) pH^2 2.1031*** 3.2783*** 9.2032*** 7.6864*** 7.2066*** 13.7788** (0.5690) (0.9107) (1.7780) (1.7639) (2.5844) (6.6188) Soil organic carbon stock (g/kg) -2.2043*** -3.6027*** -2.4920*** -2.4106*** -4.3408*** -0.7619 (0.2690) (0.5305) (0.7250) (0.7187) (1.0341) (1.8209) Organic carbon^2 0.6897*** 1.0905*** 0.7599*** 0.7300*** 1.3041*** 0.2940 (0.0760) (0.1535) (0.2046) (0.2028) (0.2890) (0.5014) year=2022 -0.1386*** -0.2753*** -0.1423*** -0.1421*** (0.0048) (0.0094) (0.0120) (0.0119) year=2023 -0.1976*** -0.3293*** -0.1856*** -0.1895*** 2.2497*** (0.0039) (0.0075) (0.0097) (0.0096) (0.6690) year=2024 -0.2288*** -0.3415*** -0.2105*** -0.2137*** 1.9166*** (0.0041) (0.0082) (0.0103) (0.0102) (0.5930) Constant 12.8313*** 18.9354*** 40.0989*** 35.5640*** 30.3493*** 67.5655*** (2.0008) (3.1700) (6.2980) (6.2468) (9.2779) (23.7195) No. of obs (farms) 173,651 97,111 25,719 25,719 9,166 2,135 Adj R-squared 0.3820 0.5067 0.4904 0.4999 0.6649 0.8173 Note: OTG fixed effects are included. Standard errors in parentheses (* p<0.10, ** p<0.05, *** p<0.010). 42 Appendix table 10: Sales price regressions for agricultural land before January 2024 All transactions With registry information Commercial Personal Commercial Commercial Area (ha) 0.1171*** -0.2220*** 0.2393*** 0.2227*** (0.0027) (0.0056) (0.0064) (0.0066) Area^2 -0.0274*** 0.0510*** -0.0924*** -0.0896*** (0.0018) (0.0023) (0.0051) (0.0051) Buyer is legal entity 0.6105*** 0.3740*** (0.1328) (0.0297) # transactions in village -0.0172*** -0.0505*** -0.0165*** -0.0199*** (0.0019) (0.0034) (0.0054) (0.0054) Parcel part of a field -0.0230* -0.1622*** (0.0121) (0.0106) Field area (ha) -0.0038 -0.0084*** 0.0340*** (0.0032) (0.0026) (0.0106) Field area^2 0.0049*** 0.0090*** 0.0003 (0.0005) (0.0006) (0.0015) Unmapped area share -0.2659*** -0.0326 -0.0352 (0.0231) (0.0263) (0.0803) <1% of area unregistered -0.0092 (0.0119) 1-10% of area unregistered -0.0099 (0.0106) >10% of area unregistered -0.0596*** (0.0143) Owner area <1% -0.0207 (0.0172) Owner area 1-10% -0.0234** (0.0094) Owner area >10% -0.0102 (0.0188) Largest user's cult. area share 0.2043*** (0.0246) >5 registered users -0.0022 (0.0148) Dist. to tertiary road (km) -0.0145*** -0.0300*** -0.0061 -0.0089** (0.0016) (0.0030) (0.0042) (0.0042) Dist. to secondary road (km) 0.0042** -0.0191*** -0.0023 -0.0049 (0.0017) (0.0033) (0.0048) (0.0048) Dist. to primary road (km) -0.0100*** -0.0211*** 0.0056 0.0042 (0.0019) (0.0035) (0.0052) (0.0052) Dist. to motorway (km) -0.0106*** -0.0115*** 0.0173** 0.0131* (0.0024) (0.0043) (0.0077) (0.0076) Dist. to railway (km) -0.0137*** -0.0024 -0.0050 -0.0026 (0.0021) (0.0039) (0.0057) (0.0057) Dist. to port (km) 0.0700*** 0.0593 -0.0251 -0.1176 (0.0264) (0.0610) (0.1844) (0.1831) Dist. to elevator (km) -0.0133*** -0.0266*** -0.0249*** -0.0293*** (0.0035) (0.0066) (0.0094) (0.0094) Dist. to nearest city (km) -0.0107* -0.1244*** -0.0113 -0.0141 (0.0060) (0.0100) (0.0190) (0.0189) Share built-up area 0.9075*** 0.4571*** -0.2899 -0.3083 (0.0828) (0.0536) (0.3838) (0.3809) Share cultivated area 0.5582*** 0.3911*** 0.2234 0.1692 (0.0360) (0.0452) (0.1775) (0.1761) Share forested area 0.4186*** 0.2448*** 0.0982 0.0821 (0.0405) (0.0484) (0.2009) (0.1993) Share grassland 0.3607*** 0.2662*** -0.0825 -0.0630 (0.0362) (0.0450) (0.1795) (0.1780) Soil bulk density (g/cm^3) -2.5201*** -2.4086** -3.2283* -2.8201 (0.6333) (1.1551) (1.8849) (1.8699) Density^2 2.2925* 2.8329 4.5623 4.1983 43 (1.2463) (2.2812) (3.7556) (3.7259) Total nitrogen (g/kg) 0.6336*** 0.6623*** 0.8035** 0.7151** (0.0962) (0.1856) (0.3168) (0.3145) Nitrogen^2 -0.2474*** -0.3059*** -0.2904*** -0.2581** (0.0341) (0.0636) (0.1099) (0.1091) Soil pH (pH) -6.9811*** -13.1285*** -37.3043*** -32.0775*** (2.5877) (4.0412) (8.2927) (8.2331) pH^2 2.2586*** 3.8653*** 10.2133*** 8.7388*** (0.6900) (1.0843) (2.2025) (2.1870) Soil organic carbon stock (g/kg) -1.8621*** -3.4117*** -1.6789* -1.4657* (0.3256) (0.6252) (0.8909) (0.8841) Organic carbon^2 0.5837*** 1.0312*** 0.5284** 0.4633* (0.0918) (0.1804) (0.2519) (0.2500) year=2022 -0.1324*** -0.2772*** -0.1394*** -0.1389*** (0.0047) (0.0094) (0.0120) (0.0119) year=2023 -0.1884*** -0.3340*** -0.1835*** -0.1864*** (0.0039) (0.0077) (0.0098) (0.0097) Constant 12.7905*** 20.9680*** 42.1951*** 37.7684*** (2.4244) (3.7800) (7.7989) (7.7427) No. of obs (farms) 110,589 69,037 16,271 16,271 Adj R-squared 0.3766 0.5287 0.4375 0.4472 Note: OTG fixed effects are included throughout. Standard errors in parentheses (* p<0.10, ** p<0.05, *** p<0.010). 44 Appendix table 11: Sales price regressions for agricultural land after January 2024 All transactions With registry information With tenant characteristics Commercial Personal Commercial Commercial Commercial Commercial Area (ha) 0.0944*** -0.1864*** 0.2746*** 0.2484*** 0.1894*** 0.3439*** (0.0037) (0.0080) (0.0088) (0.0090) (0.0086) (0.0158) Area^2 -0.0210*** 0.0700*** -0.0871*** -0.0852*** -0.0559*** -0.1083*** (0.0025) (0.0033) (0.0075) (0.0075) (0.0063) (0.0129) Buyer is legal entity 0.2687*** 0.4154*** 0.2145*** 0.2697*** (0.0058) (0.0169) (0.0165) (0.0222) # transactions in village -0.0127*** -0.0237*** -0.0291*** -0.0324*** 0.0038 -0.0523*** (0.0028) (0.0061) (0.0084) (0.0084) (0.0080) (0.0179) Parcel part of a field -0.0532*** -0.1149*** (0.0120) (0.0169) Field area (ha) -0.0051 -0.0060 0.0266** -0.0090 0.0152 (0.0056) (0.0062) (0.0107) (0.0093) (0.0290) Field area^2 0.0054*** 0.0112*** 0.0011 0.0036** -0.0001 (0.0008) (0.0012) (0.0015) (0.0016) (0.0040) Unmapped area share -0.4206*** -0.0148 -0.4570*** -0.3025*** 0.1545 (0.0373) (0.0536) (0.1086) (0.1099) (0.2026) <1% of area unregistered 0.0298* -0.0050 (0.0164) (0.0268) 1-10% of area unregistered 0.0089 -0.0263 (0.0146) (0.0239) >10% of area unregistered -0.0773*** -0.0944** (0.0201) (0.0375) Owner area <1% 0.0026 0.1030*** (0.0216) (0.0358) Owner area 1-10% -0.0110 0.0045 (0.0126) (0.0221) Owner area >10% -0.0009 0.0967* (0.0274) (0.0518) Largest user's cult. area share 0.1550*** 0.3306*** (0.0354) (0.0654) Buyer is the largest user in field -0.1028*** -0.1227*** (0.0254) (0.0366) >5 registered users 0.0307 0.0120 (0.0208) (0.0385) Buyer's area (ha) -0.0668*** -0.0566 (0.0095) (0.0810) Buyer's area^2 0.0148*** 0.0215** (0.0014) (0.0084) Buyer uses <33% of village land -0.2090*** -0.2691* (0.0348) (0.1456) Buyer uses 33-50% of village -0.2957*** -0.2200 land (0.0384) (0.1565) Buyer uses >50% of village land -0.3105*** -0.3289** (0.0377) (0.1554) Dist. to tertiary road (km) -0.0108*** -0.0095** 0.0065 0.0029 -0.0208*** -0.0071 (0.0023) (0.0047) (0.0059) (0.0059) (0.0054) (0.0103) Dist. to secondary road (km) -0.0022 -0.0166*** 0.0214*** 0.0176*** 0.0030 0.0219* (0.0024) (0.0053) (0.0065) (0.0065) (0.0063) (0.0116) Dist. to primary road (km) -0.0152*** -0.0345*** -0.0358*** -0.0373*** -0.0341*** -0.0459*** (0.0027) (0.0055) (0.0073) (0.0073) (0.0072) (0.0162) Dist. to motorway (km) -0.0068** -0.0256*** 0.0091 0.0057 -0.0018 -0.0420* (0.0033) (0.0068) (0.0112) (0.0111) (0.0112) (0.0233) Dist. to railway (km) -0.0062** -0.0139** -0.0004 0.0019 -0.0071 0.0045 (0.0029) (0.0060) (0.0084) (0.0083) (0.0080) (0.0206) Dist. to port (km) 0.0396 0.0683 -0.3182 -0.3219 0.1210 -1.3476** (0.0423) (0.1152) (0.2620) (0.2594) (0.2009) (0.5469) Dist. to elevator (km) 0.0018 -0.0268** -0.0072 -0.0092 0.0190 -0.0072 (0.0049) (0.0105) (0.0134) (0.0133) (0.0137) (0.0285) 45 Dist. to nearest city (km) -0.0063 -0.0924*** -0.0160 -0.0163 -0.0209 -0.1570*** (0.0086) (0.0153) (0.0285) (0.0283) (0.0246) (0.0532) Share built-up area 0.4927*** 0.2004*** -0.0869 -0.1137 0.5245*** -1.1449** (0.0740) (0.0683) (0.1824) (0.1808) (0.1691) (0.4786) Share cultivated area 0.4163*** 0.2374*** -0.1174 -0.2078* 0.3983*** -1.4858*** (0.0416) (0.0589) (0.1257) (0.1257) (0.1065) (0.3849) Share forested area 0.2598*** 0.1604** -0.1401 -0.2195 0.1745 -2.1104*** (0.0497) (0.0674) (0.2110) (0.2100) (0.1241) (0.4687) Share grassland 0.2835*** 0.1544*** -0.4995*** -0.4592*** 0.1604 -1.8231*** (0.0423) (0.0592) (0.1320) (0.1312) (0.1085) (0.3865) Soil bulk density (g/cm^3) -3.9544*** -3.2592* -2.8462 -2.3353 10.7028*** -1.9599 (1.0353) (1.7970) (2.9728) (2.9513) (2.6984) (5.3322) Density^2 5.4099*** 2.5342 2.5419 2.4528 -19.7245*** 7.4346 (2.0097) (3.5333) (5.8434) (5.7967) (5.4252) (10.4732) Total nitrogen (g/kg) 0.3730*** 1.7328*** 0.6837 0.6261 -0.4137 0.6306 (0.1402) (0.3028) (0.4304) (0.4265) (0.4453) (0.8302) Nitrogen^2 -0.1674*** -0.7202*** -0.2513* -0.2222 0.1242 -0.1803 (0.0493) (0.1033) (0.1481) (0.1467) (0.1512) (0.2802) Soil pH (pH) -6.0489 -1.1151 -29.5696*** -25.5410** -26.3812*** -52.7881** (3.7154) (6.2710) (11.0959) (10.9900) (9.7114) (25.0955) pH^2 1.9742** 0.5486 8.1950*** 7.0217** 7.2562*** 13.7788** (0.9896) (1.6833) (2.9408) (2.9130) (2.5832) (6.6188) Soil organic carbon stock (g/kg) -2.5657*** -5.6439*** -3.1576*** -3.4336*** -4.3343*** -0.7619 (0.4683) (1.0139) (1.2157) (1.2043) (1.0336) (1.8209) Organic carbon^2 0.8083*** 1.6878*** 0.9520*** 1.0233*** 1.3023*** 0.2940 (0.1324) (0.2952) (0.3417) (0.3384) (0.2888) (0.5014) Constant 13.0400*** 11.2649* 38.1574*** 34.8256*** 32.4758*** 67.5655*** (3.4830) (5.8460) (10.4383) (10.3394) (9.2547) (23.7195) No. of obs (farms) 63,062 28,074 9,448 9,448 9,163 2,135 Adj R-squared 0.4295 0.4894 0.5951 0.6035 0.6651 0.8173 Note: OTG fixed effects are included throughout. Standard errors in parentheses (* p<0.10, ** p<0.05, *** p<0.010). 46 Figure 1: Typical pattern of land use in irrigated lands Source: Martyn et al 2022 47 Figure 2: No. of monthly agricultural land sales and rentals Panel A: Agricultural land sales Panel B: Agricultural land rentals 48 Figure 3: Mean sale and lease prices or agricultural land Panel A: Sales prices Panel B: Lease prices Note: All prices are in US$ equivalent per hectare. 49 References Adamopoulos, T., L. Brandt, J. Leight and D. Restuccia. 2022. "Misallocation, Selection, and Productivity: A Quantitative Analysis with Panel Data from China." Econometrica 90 (3): 1261-1282. Adamopoulos, T. and D. Restuccia. 2022. "Geography and Agricultural Productivity: Cross-Country Evidence for Micro Plot- Level Data." Review of Economic Studies 89 (4): 1629-1653. Ahlfeldt, G., P. Koutroumpis and T. Valletti. 2017. "Speed 2.0: Evaluating Access to Universal Digital Highways." Journal of the European Economic Association 15 (3): 586-625. Ali, D. A., K. Deininger and M. Goldstein. 2014. "Environmental and Gender Impacts of Land Tenure Regularization in Africa: Pilot Evidence from Rwanda." Journal of Development Economics 110 (0): 262-275. Ali, D. A. and K. Deininger. 2021. "Does Mechanization Reverse the Inverse Farm Size-Productivity Relationsip? Evidence from Ethiopia." Policy Research Working Paper, the World Bank, Washington, DC. doi:10.1596/1813-9450-9306. Ali, D. A., K. Deininger, G. Mahofa and R. Nyakulama. 2021. "Sustaining Land Registration Benefits by Addressing the Challenges of Reversion to Informality in Rwanda." Land Use Policy 110: 104317. Allen, D. W. and B. Leonard. 2021. "Property Right Acquisition and Path Dependence: Nineteenth-Century Land Policy and Modern Economic Outcomes." Economic Journal 131 (640): 3073-3102. Asquith, B. J., E. Mast and D. Reed. 2023. "Local Effects of Large New Apartment Buildings in Low-Income Areas." The Review of Economics and Statistics 105 (2): 359-375. Ayaz, M. and M. Mughal. 2024. "Farm Size and Productivity: The Role of Family Labor." Economic Development & Cultural Change 72 (2): 959-995. Baylis, P. and J. Boomhower. 2023. "The Economic Incidence of Wildfire Suppression in the United States." American Economic Journal: Applied Economics 15 (1): 442-473. Beg, S. 2022. "Digitization and Development: Property Rights Security, and Land and Labor Markets." Journal of the European Economic Association 20 (1): 395-429. Binswanger, H. P., K. Deininger and G. Feder. 1995. "Power, Distortions, Revolt and Reform in Agricultural Land Relations." Handbook of development economics 3B: 2659-2772. Bobba, M., L. Flabbi and S. Levy. 2022. "Labor Market Search, Informality, and Schooling Investments." International Economic Review 63 (1): 211-259. Bond, S. A., P. Loizou and P. McAllister. 2008. "Lease Maturity and Initial Rent: Is There a Term Structure for Uk Commercial Property Leases?" Journal of Real Estate Finance & Economics 36 (4): 451-469. Bosker, M., H. Garretsen, G. Marlet and C. v. Woerkens. 2019. "Nether Lands: Evidence on the Price and Perception of Rare Natural Disasters." Journal of the European Economic Association 17 (2): 413-453. Brooks, L. and B. Lutz. 2016. "From Today's City to Tomorrow's City: An Empirical Investigation of Urban Land Assembly." American Economic Journal: Economic Policy 8 (3): 69-105. Brorsen, B. W., D. Doye and K. B. Neal. 2015. "Agricultural Land and the Small Parcel Size Premium Puzzle." Land Economics 91 (3): 572-585. Bryan, G., J. de Quidt, T. Wilkening and N. Yadav. 2023. "Market Design for Land Trade: Evidence from Uganda and Kenya." London School of Economics, London. Bu, D. and Y. Liao. 2022. "Land Property Rights and Rural Enterprise Growth: Evidence from Land Titling Reform in China." Journal of Development Economics 157: 102853. Chari, A., E. M. Liu, S.-Y. Wang and Y. Wang. 2021. "Property Rights, Land Misallocation, and Agricultural Efficiency in China." Review of Economic Studies 88 (4): 1831-1862. Chen, C., D. Restuccia and R. Santaeulàlia-Llopis. 2022. "The Effects of Land Markets on Resource Allocation and Agricultural Productivity." Review of Economic Dynamics 45: 41-54. Chen, C., D. Restuccia and R. Santaeulalia-Llopis. 2023. "Land Misallocation and Productivity." American Economic Journal: Macroeconomics 15 (2): 441-465. Combes, P.-P., G. Duranton and L. Gobillon. 2019. "The Costs of Agglomeration: House and Land Prices in French Cities." Review of Economic Studies 86 (4): 1556-1589. Cotteleer, G., C. Gardebroek and J. Luijt. 2008. "Market Power in a Gis-Based Hedonic Price Model of Local Farmland Markets." Land Economics 84 (4): 573-592. de Janvry, A., K. Emerick, M. Gonzalez-Navarro and E. Sadoulet. 2015. "Delinking Land Rights from Land Use: Certification and Migration in Mexico." American Economic Review 105 (10): 3125-3149. Deininger, K., et al. 2024. "Micro-Level Impacts of the War on Ukraine’s Agriculture Sector Distinguishing Local and National Effects over Time." World Bank Policy Research Working Paper 10869, Washington DC. Deininger, K. W., D. A. Ali and R. Neyter. 2023. "Impacts of a Mandatory Shift to Decentralized Online Auctions on Revenue from Public Land Leases in Ukraine." Journal of Economic Behavior & Organization 213: 432-450. Devadoss, S. and W. Ridley. 2024. "Impacts of the Russian Invasion of Ukraine on the Global Wheat Market." World Development 173: 106396. Dippel, C., D. Frye and B. Leonard. 2024. "Property Rights without Transfer Rights: A Study of Indian Land Allotment." Journal of Political Economy: Microeconomics. Foster, A. D. and M. R. Rosenzweig. 2021. "Are There Too Many Farms in the World? Labor Market Transaction Costs, Machine Capacities, and Optimal Farm Size." Journal of Political Economy 130 (3): 636-680. Galiani, S. and E. Schargrodsky. 2016. "The Deregularization of Land Titles." Man & the Economy 3 (2): 169-188. 50 Garbarino, N., B. Guin and C.-H. Lee. 2023. "The Effects of Subsidized Flood Insurance on Real Estate Markets." Bank of England Working Paper No. 995, London. Gertler, P. J., M. Gonzalez-Navarro, T. Gračner and A. D. Rothenberg. 2024. "Road Maintenance and Local Economic Development: Evidence from Indonesia’s Highways." Journal of Urban Economics 143: 103687. Giles, J. and R. Mu. 2018. "Village Political Economy, Land Tenure Insecurity, and the Rural to Urban Migration Decision: Evidence from China." American Journal of Agricultural Economics 100 (2): 521-544. Gollin, D. and C. Udry. 2020. "Heterogeneity, Measurement Error, and Misallocation: Evidence from African Agriculture." Journal of Political Economy 129 (1): 1-80. Gourevitch, J. D., et al. 2023. "Unpriced Climate Risk and the Potential Consequences of Overvaluation in Us Housing Markets." Nature Climate Change 13 (3): 250-257. Graubner, M., I. Ostapchuk and T. Gagalyuk. 2021. "Agroholdings and Land Rental Markets: A Spatial Competition Perspective." European Review of Agricultural Economics 48 (1): 158-206. Grenadier, S. R. 1995. "Valuing Lease Contracts: A Real-Options Approach." Journal of Financial Economics 38 (3): 297-331. Gutierrez, I. A. and O. Molina. 2020. "Reverting to Informality: Unregistered Property Transactions and the Erosion of the Titling Reform in Peru." Economic Development & Cultural Change 69 (1): 317-334. Han, L., S. Heblich, C. Timmins and Y. Zylberberg. 2021. "Cool Cities: The Value of Urban Trees∗." Harding, J. P., S. S. Rosenthal and C. F. Sirmans. 2003. "Estimating Bargaining Power in the Market for Existing Homes." Review of Economics & Statistics 85 (1): 178-188. Hawley, Z., J. J. Miranda and W. C. Sawyer. 2018. "Land Values, Property Rights, and Home Ownership: Implications for Property Taxation in Peru." Regional Science and Urban Economics 69: 38-47. Helfand, S. M. and M. P. H. Taylor. 2021. "The Inverse Relationship between Farm Size and Productivity: Refocusing the Debate." Food Policy 99. Henderson, J. V., T. Regan and A. J. Venables. 2021. "Building the City: From Slums to a Modern Metropolis." The Review of Economic Studies 88 (3): 1157-1192. Hilber, C. A. L. and W. Vermeulen. 2016. "The Impact of Supply Constraints on House Prices in England." Economic Journal 126 (591): 358-405. Hino, M. and M. Burke. 2021. "The Effect of Infromation About Climate Risk on Property Values." Proeedings of the National Academy of Sciences 118 (17). Hüttel, S., M. Ritter, V. Esaulov and M. Odening. 2016. "Is There a Term Structure in Land Lease Rates?" European Review of Agricultural Economics 43 (1): 165-187. Ibatullin, S., et al. 2024. "Agricultural Land Market in Ukraine: Challenges of Trade Liberalization and Future Land Policy Reforms." Land, 13(3), 10.3390/land13030338: 10.3390/land13030338. Kirwan, B. E. 2009. "The Incidence of U.S. Agricultural Subsidies on Farmland Rental Rates." Journal of Political Economy 117 (1): 138-164. Knoll, K., M. Schularick and T. Steger. 2017. "No Price Like Home: Global House Prices, 1870-2012." American Economic Review 107 (2): 331-353. Krishna, A., E. Rains and E. Wibbels. 2020. "Negotiating Informality--Ambiguity, Intermediation, and a Patchwork of Outcomes in Slums of Bengaluru." Journal of Development Studies 56 (11): 1983-1999. Kumar, A. and C.-Y. Liang. 2024. "Labor Market Effects of Credit Constraints: Evidence from a Natural Experiment " American Economic Journal: Economic Policy. Kvartiuk, V., T. Herzfeld and E. Bukin. 2022. "Decentralized Public Farmland Conveyance: Rental Rights Auctioning in Ukraine." Land Use Policy 114: 105983. Kvartiuk, V. and A. Martyn. 2022. "Ukraine’s Agricultural Land Sales Market: An Update and the Effect of Russian War against Ukraine ", Agro Policy Report APD/APB/20/2022, German-Ukrainian Agicultural Policy Dialogue, Kyiv. Kvartiuk, V., E. Bukin and T. Herzfeld. 2024. "“For Whoever Has Will Be Given More”? Land Rental Decisions and Technical Efficiency in Ukraine." Land Use Policy 146: 107336. Lanjouw, J. O. and P. I. Levy. 2002. "Untitled: A Study of Formal and Informal Property Rights in Urban Ecuador." Economic Journal 112 (482): 986-1019. Lawley, C. 2018. "Ownership Restrictions and Farmland Values: Evidence from the 2003 Saskatchewan Farm Security Act Amendment." American Journal of Agricultural Economics 100 (1): 311-337. Leonard, B. and D. P. Parker. 2021. "Fragmented Ownership and Natural Resource Use: Evidence from the Bakken." The Economic Journal 131 (635): 1215-1249. Libecap, G. D. and D. Lueck. 2011. "The Demarcation of Land and the Role of Coordinating Property Institutions." Journal of Political Economy 119 (3): 426-467. Lin, F., et al. 2023. "The Impact of Russia-Ukraine Conflict on Global Food Security." Global Food Security 36: 100661. Loumeau, G. 2022. "Land Consolidation Reforms: A Natural Experiment on the Economic and Political Effects of Agricultural Mechanization ", Cener of Economic Researchat ETH Zurich Working Paper 22/376, Zurich. Martyn, A., A. Koshel, L. Hunko and L. Kolosa. 2022. "Land Consolidation in Ukraine after Land Reform: Voluntary and Forced Mechanisms." Acta Scientiarum Polonorum, Administratio Locorum 21 (2): 223-229. Maue, C. C., M. Burke and K. J. Emerick. 2020. "Productivity Dispersion and Persistence among the World's Most Numerous Firms." National Bureau of Economic Research, Inc, NBER Working Papers: 26924. 51 Muyanga, M. and T. S. Jayne. 2019. "Revisiting the Farm Size-Productivity Relationship Based on a Relatively Wide Range of Farm Sizes: Evidence from Kenya." American Journal of Agricultural Economics 101 (4): 1140-1163. Navarro, I. A. and G. K. Turnbull. 2014. "Property Rights and Urban Development: Initial Title Quality Matters Even When It No Longer Matters." Journal of Real Estate Finance & Economics 49 (1): 1-22. Neyter, R. and O. Nivievskyi. 2022. "Local Versus Central Public Land Governance: Evidence from Spatial Analysis of Land Auctions in Ukraine." Kyiv School of Economics, Center for Food and Land Use Research, Kyiv. Niu, D., W. Sun and S. Zheng. 2021. "The Role of Informal Housing in Lowering China’s Urbanization Costs." Regional Science and Urban Economics (91): 103638. Nivievskyi, O. 2020. "Where Is the State Agricultural Land Disappearing?", Economic Pravda, Kyiv. Nizalov, D., et al. 2016. "Security of Property Rights and Transition in Land Use." Journal of Comparative Economics 44 (1): 76- 91. Omotilewa, O. J., et al. 2021. "A Revisit of Farm Size and Productivity: Empirical Evidence from a Wide Range of Farm Sizes in Nigeria." World Development 146: 105592. Pennerstorfer, D. 2022. "Farm Exits and Competition on the Land Market: Evidence from Spatially Explicit Data." Johannes Kepler University of Linz - Working Paper 2209, Linz, austria. Perra, E., M. Sanfilippo and A. Sundaram. 2024. "Roads, Competition, and the Informal Sector." Journal of Development Economics 171: 103339. Poggio, L., et al. 2021. "Soilgrids 2.0: Producing Soil Information for the Globe with Quantified Spatial Uncertainty." SOIL 7 (1): 217-240. Rossi-Hansberg, E., P.-D. Sarte and R. Owens, III. 2010. "Housing Externalities." Journal of Political Economy 118 (3): 485-535. Schmalz, M. C., D. A. Sraer and D. Thesmar. 2017. "Housing Collateral and Entrepreneurship." Journal of Finance 72 (1): 99- 132. Seifert, S., C. Kahle and S. Hüttel. 2021. "Price Dispersion in Farmland Markets: What Is the Role of Asymmetric Information?" American Journal of Agricultural Economics 103 (4): 1545-1568. Severen, C. and A. J. Plantinga. 2018. "Land-Use Regulations, Property Values, and Rents: Decomposing the Effects of the California Coastal Act." Journal of Urban Economics 107: 65-78. Sheng, Y. and W. Chancellor. 2019. "Exploring the Relationship between Farm Size and Productivity: Evidence from the Australian Grains Industry." Food Policy 84: 196-204. Sheng, Y., J. Ding and J. Huang. 2019. "The Relationship between Farm Size and Productivity in Agriculture: Evidence from Maize Production in Northern China." American Journal of Agricultural Economics 101 (3): 790-806. Shifa, A. B. and W. Xiao. 2023. "Urban Bias, Migration Control and Rural Land Policy: The Case of Hukou in China." Journal of Economic Geography 23 (2): 251-274. Smith, C. 2024. "Land Concentration and Long-Run Development in the Frontier United States." Univerity of Maryland, College Park, MD. Sood, A. 2022. "Land Market Frictions in Developing Countries: Evidence from Manufacturing Firms in India ", Working Paper, University of Toronto, Toronto. Storm, H., K. Mittenzwei and T. Heckelei. 2015. "Direct Payments, Spatial Competition, and Farm Survival in Norway." American Journal of Agricultural Economics 97 (4): 1192-1205. Swinnen, J., K. Van Herck and L. Vranken. 2016. "The Diversity of Land Markets and Regulations in Europe, and (Some of) Its Causes." Journal of Development Studies 52 (2): 186-205. Tellman, B., et al. 2021. "The Role of Institutional Entrepreneurs and Informal Land Transactions in Mexico City’s Urban Expansion." World Development 140: 105374. Vranken, L., E. Tabeau, P. Roebeling and P. Ciaian. 2021. "Agricultural Land Market Regulations in the Eu Member States." Publications Office of teh European Union Luxembourg. Waseem, M. 2018. "Taxes, Informality and Income Shifting: Evidence from a Recent Pakistani Tax Reform." Journal of Public Economics 157: 41-77. Zadorozhna, O. 2020. "Clientelism and Land Market Outcomes in Ukraine." Eastern European Economics 58 (6): 478-496. Zevelev, A. A. 2021. "Does Collateral Value Affect Asset Prices? Evidence from a Natural Experiment in Texas." The Review of Financial Studies 34 (9): 4373-4411. Zhang, X., L. Hu and X. Yu. 2023. "Farmland Leasing, Misallocation Reduction, and Agricultural Total Factor Productivity: Insights from Rice Production in China." Food Policy 119: 102518. Zhao, X. 2020. "Land and Labor Allocation under Communal Tenure: Theory and Evidence from China." Journal of Development Economics 147: 102526. 52