Policy Research Working Paper 10385
Land and Mortgage Markets in Ukraine
Pre-War Performance, War Effects,
and Implications for Recovery
Klaus Deininger
Daniel Ayalew Ali
Development Economics
Development Research Group
March 2023
Policy Research Working Paper 10385
Abstract
Almost throughout Ukraine’s independent history, agricul- land prices. Agricultural land market volume soon exceeded
tural land sales were prohibited. Measures to allow them and that of residential land and continued at a reduced level
make land governance more transparent in 2020/21 were and with prices some 15–20 percent lower even after the
expected to improve equity, investment, credit access, and invasion, with little sign of speculative land acquisition.
decentralization. This paper draws on administrative data Mortgage market activity and credit access remained below
and satellite imagery to describe land market performance expectations. The paper discusses reasons and options for
before and after the Russian invasion, assess changes in addressing them in a way that also factors in the needs of
land use for transacted parcels, and analyze determinants of post-war reconstruction.
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 or 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 and Mortgage Markets in Ukraine: Pre-War Performance, War Effects,
and Implications for Recovery
Klaus Deininger
Daniel Ayalew Ali
JEL Codes: Q10, O13, H56, R14
Keywords: Ukraine, land market, credit markets, conflict/war, agricultural production
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 Nataliia Kussul and Andrii Shelestov for support with data processing and Denis Bashlyk,
Markiyan Dmytrasevych, Ben Hell, Thea Hilhorst, Roman Hrab, Sergyi Kubakh, Vasyl Kvartiuk, Andrii Martin, Roman Neyter,
Oleg Nivievskyi, Taras Vysotskyi and Sergyi Zorya for helpful discussions and insightful comments and Daria Manzhura as well
as Tania Khorzovskaya for outstanding coordination and administrative support.
Land and Mortgage Markets in Ukraine: Pre-War Performance, War Effects,
and Implications for Recovery
1. Introduction
Legal reforms passed in 2020 to allow sale of agricultural land and establish the institutional basis for
transparent operation of land and financial markets were intended to lay the basis for Ukraine’s transition
towards a modern market economy by (i) increasing trust in state registries and their capacity to maintain
records properly without corruption, (ii) supporting financial market development by reducing the cost of
registration and providing broad address to land price information, and (iii) fostering decentralization by
empowering local governments to increase land values through planning and capturing part of the resulting
benefits via property taxes and land lease fees. Although the key laws were by and large completed in 2020
and land markets operated since July 2021, it was clear that completing relevant regulations and institutional
reforms would be a longer-term process.
The Russian invasion interrupted reform implementation in the short term and created challenges, the
resolution of which hinges on expeditious reform implementation: Protecting rights to property despite
destruction of physical registries in conflict-affected areas and loss of paper documents by displaced
persons requires easy and reliable access to digital records and scanned copies of these documents.
Compensation for lost land or property will be less arbitrary if based on objective price data and digital
access to pre-war records. Distress sales will be less likely if information on land market values is publicly
available or if any attempt to register transactions at prices that deviate too much from predicted prices can
be flagged and subjected to scrutiny automatically. Reconstruction will be faster and more likely to
maximize future economic potential if it leverages private finance and is informed by publicly available
local land use plans that are based on and can be enforced using cadastral and registry information.
This paper uses administrative and remotely sensed data to assess performance of land and mortgage
markets 18 months after the moratorium on agricultural land was lifted and uses this as the basis to assess
the most expeditious ways of confronting the challenges ahead. Three main conclusions emerge:
First, after market opening, the volume of agricultural land markets soon eclipsed that of residential ones
and exceeded it ever since. Prices for commercial agricultural land were more resilient to the invasion than
residential land prices or land for personal farming and are in line with fundamentals, presumably because
such land produces tradable goods. Neither registered transactions nor survey evidence raise concern about
a potential wave of distress sales included in the registry.
2
Second, although fears about chaotic land markets were unfounded, expected increases in mortgage lending
failed to materialize; with well below 1 percent of agricultural and residential land parcels mortgaged, it is
unlikely that lifting of the remaining restrictions on agricultural land sales will make much of a difference.
To address this constraint sustainably, there is need to (i) ensure that existing legislation mandating price
reporting is fully implemented and move towards market-based mass valuation as a basis for land taxation;
(ii) complete institutional reform of the registry and cadaster to establish a fully interoperable register of
real property that includes land and structures (based on municipal records) in rural and urban areas; and
(iii) reduce the cost of registering mortgages or foreclosing in case of default and ensure public land price
information allows the National Bank to properly value collateral and increase loan-to-value ratios if land
is pledged as collateral in line with Basel principles.
Third, although local planning and revenue collection have been superseded by more urgent priorities, the
doubling of land values due to a mandated shift from in-person to transparent electronic auctions (Deininger
et al. 2022) demonstrates the potential impact of land reform on local revenue. The destruction wrought by
the invasion and the associated need to deal with a vast number of compensation or restitution claims swiftly
and transparently create opportunities for rapid progress on the move towards electronic registries for
residential real estate and decentralized land management by (i) digitizing historical property and cadastral
records to use them as the basis for a seamless digital resolution and restitution processes; (ii) linking to the
underlying cadastral records and recent high-resolution imagery for a fully integrated electronic register of
real estate objects; (iii) publishing local land use plans digitally to streamline processes for enforcement as
well as updating and approvals to facilitate expeditious reconstruction in a way that matches private
investment with public services and avoids potentially irreparable damage to environmental and cultural
assets; and (iv) eventually using market rather than outdated normative values as a basis for property
taxation and establishing interoperability to support local efforts at better collection.
Our study links to three strands of literature. First, a large body of studies used land prices for hedonic
analysis to make inferences on the value of specific attributes and the effect of market structure. Residential
house prices have long been used to value size and incidence of benefits from public goods such as road
maintenance (Gertler et al. 2022), broadband access (Ahlfeldt et al. 2017), trees (Han et al. 2021), coastal
preservation (Severen & Plantinga 2018) or losses from hazards such as air pollution contamination (Chang
& Li 2021), radon exposure (Pinchbeck et al. 2020), proximity to nuclear plants (Coulomb & Zylberberg
2021) earthquake risk (Singh 2019), exposure to sea-level rise (Goldsmith-Pinkham et al. 2021), or species
extinction (Frank & Sudarshan 2022), zoning policies restricting on density (Hilber & Vermeulen 2016),
congestion pricing (Tang 2021).
3
For agricultural land, hedonic analyses are complemented by studies that use data on land prices to analyze
market structure and policy in several contexts. In Germany, the fact that institutional sellers achieve higher
prices (Seifert et al. 2021) and that even for foreclosure, auctions lead to higher prices (Hüttel et al. 2014)
points towards limited transparency, possibly warranting to consider measures such as mandatory price
disclosure that reduced price dispersion in urban settings (Ben-Shahar & Golan 2019). Elimination of
restrictions on outside buyers significantly increased land prices in Canada (Lawley 2018) while minimum
parcel size regulation increased land values in Taiwan (Chang & Lin 2016). Analysis of the incidence of
farm support between renters and owners point to monopsony power by renters (Kirwan 2009) who capture
greater shares of such support in less competitive markets, especially if they are large and able to enter into
long contracts (Kirwan & Roberts 2016). We add to this by showing that presence of unregistered land
exerts a negative externality on transaction prices; that basing tax liability on the administrative ‘normative
value’ rather than market values leads to public revenue loss . Once the current restriction of land
transactions to physical persons is lifted, it will be important to also assess market competitiveness by using
data on type, location, and size of buyers.
Second, while studies have explored the impact of implicit land price changes on the likelihood of conflict
(Berman et al. 2021) and the effect of war-induced commodity price booms in the 1920s on credit expansion
and bank failure in the US (Rajan & Ramcharan 2015, 2016; Jaremski & Wheelock 2020), evidence on the
impact of inter-state war on agricultural land markets and prices is limited. Besley and Mueller (2012) use
within-region variability in violent killings over time to assess the impact of such violence on residential
house prices which are then used to infer the value of the ‘peace dividend’ in Northern Ireland’s civil war.
Studies in rural settings point towards behavioral responses such as higher levels of (within-group) altruism
or risk-seeking and potential implications for savings and investments decisions (Voors et al. 2012) and
farmers’ adjustment to conflict by reducing exposure to markets (Arias et al. 2019). They also highlight
that conflict can be used strategically by elites to accumulate land at low cost, as documented by high levels
of forced displacement and land loss to paramilitaries in municipalities with cultivation of land-intensive
oil palm (Tellez 2022). While such issues are likely in areas outside of government control, land markets
in locations controlled by government proved resilient; in fact, prices for commercial agricultural land that
produces tradables declined only modestly, less than those for urban residential or personal farming land.
Finally, our study speaks to the institutional context for land governance in Ukraine, an issue the importance
of which for global food supply chains (Ihle et al. 2022) and security (Lin et al. 2023) has come in relief
with the invasion (Glauben et al. 2022). The country has long been subject to suppression of markets and
entrepreneurship (Yaremko 2022) and Ukraine’s agricultural sector differs significantly from that of its
neighbors (Deininger et al. 2018). Prohibition of land sales allowed political elites (Zadorozhna 2020) and
4
state institutions (Kvartiuk et al. 2022; Neyter & Nivievskyi 2022) to manipulate outcomes, exercise market
power (Graubner et al. 2021) and in doing so reduce investment (Nizalov et al. 2016) and dent productivity
and confidence in the state (Nivyievskyi et al. 2021). Documenting the components of land governance
reform and the status of their implementation highlights differences in the extent of reform implementation
and impact in practice, pointing to and suggests that improving land valuation, local planning, and links to
mortgage lending, could help expand land reform benefits and support decentralization and reconstruction.
The rest of the paper is structured as follows: Section two provides context by describing legal reforms to
improve land governance introduced together with the opening of the land sales market in June 2021 and
briefly summarizes ways in which the invasion affected the sector. Section three describes the registry data
used, discusses volume and prices for sales, rental, and mortgage transactions, and uses remotely sensed on
post-transaction land use change to make inferences on potential speculative land acquisition. Section four
provides results from parcel-level analysis of land transactions and compares market prices to ‘normative
value’. Section five concludes with implications for policy and research.
2. Institutional context
Legislation to allow agricultural land sales and to reform institutions to ensure transparency and reduce the
transaction cost of market operation passed in Ukraine in 2020/21 was also expected to contribute to
financial market development, investment, and decentralization. To appreciate potential impacts of this
reform, we discuss the background for its adoption, the content of individual laws, and the way in which
their implementation was affected by the unexpected outbreak of the Russian invasion of Ukraine in
February 2022. While the invasion vindicated the need for reform, it also created new challenges, most
importantly the need to transparently secure property rights for displaced people, especially in situations
where physical offices and paper records have been destroyed, to counter distress sales, and eventually to
expeditiously compensate for damages and aid private sector-led reconstruction.
2.1 Legal reforms to land governance in Ukraine
With 41.5 million ha of some of the most fertile agricultural land globally, Ukraine has traditionally been
a major source of food grains. Before the war, agriculture contributed about 10% to GDP and 42% of the
country’s exports. After de-collectivization in the early 2000s, when some 7 million landowners were
provided with land shares of about 4 ha each, the agricultural sector’s value added doubled. Some 20 million
ha of Ukraine’s agricultural land is farmed by large farms, often by firms with links to foreign capital
markets (Deininger et al. 2018); 12 million ha is cultivated by small and household farms. While about 9.2
million ha was state or communal land originally, repeated privatization is believed to have significantly
reduced the size of this segment (Nivievskyi 2020).
5
Fears that market-induced land concentration might undermine equity led to a ‘moratorium’ on land
sales in 2001. As the task of reforming land institutions to improve transparency and reduce corruption
was too daunting, the moratorium was regularly extended (often by one or two years), together with
measures aimed to deal with the symptoms of this situation, often introducing more distortions.1
Although there was agreement about the need to eventually lift this restriction, action was impeded
by concerns that market imperfections and endemic corruption in the land sector, land markets might
further exacerbate inequality. Allowing agricultural land sales, together with institutional reform to create
the basis for transparent market operation was expected to support financial market development, encourage
investment, and contribute to greater decentralization.
The scope for using agricultural land as a collateral for credit was expected to provide long term credit for
the economy as a whole and, together with more secure land rights, support agricultural diversification
away from the focus on producing land- and capital-intensive bulk commodities with limited value added
or employment generation. It was anticipated to foster equity by improving land market transparency and
competitiveness, increasing rent shares captured by landowners. Decentralization was to be improved by
reforms that transferred all public land to local governments to provide them with a predictable stream of
revenue for local service provision; by information sharing and interoperability to improve local bodies’
ability to collect local taxes (many indexed to land); and by allowing them to plan land use to increase the
level of economic activity and land values in the medium to long term.
To address these issues, in 2019 and 2020, Parliament passed legislation to allow land sales from July
1, 2021, a move supported by Ukraine’s international partners and complemented by a package of
seven additional laws to reform key institutions dealing with agricultural and urban land to level the
playing field and increase the likelihood that desired impacts would materialize. 2 Although a major
step, the law fell short of what was proposed by proponents in four respects, namely (i) land purchases by
foreign natural persons or legal entities remain prohibited unless such a step is approved by public
referendum;3 (ii) a land ownership limit of 10,000 ha is established; (iii) until Jan. 1, 2024, land that had
been in shares (i.e., land for commercial agriculture) can be purchased only by natural persons up to a limit
1
One example is legislation, passed in 2015, that prohibits registration of leases shorter than 7 years. See Kvartiuk and Herzfeld (2019) for
a more detailed discussion.
2
After prolonged and highly politicized discussion, Law 552-IX to this effect was passed on March 31, 2020.
3
This provision was highly controversial and resisted by foreign land users who claimed to have made significant investments in land improvements
and who feared to be left at a disadvantage especially with limited enforceability of lease contracts (renege and opportunistic behavior).
6
of 100 ha;4 and (iv) public land cannot be sold but only leased. Partly to allay concerns about distress sales,
a floor price for land sales is also set by a parcel’s ‘normative value’.5
Complementary laws aimed at institutional restructuring to increase transparency; decentralization;
and operation of and access to financial markets especially by small producers. First, transparency,
access to information, and competitiveness of markets was to be enhanced through laws (law 340 and law
554) that mandated price reporting and public access to relevant cadastral data. Also, legislation to reform
the cadaster and registry was passed to ensure integrity of land ownership information, eliminate the scope
for corruption in public and private land management, and reduce the transaction cost of registering land
transactions through digital interoperability (law 1423).
Second, local self-governance and decentralization were to be supported by transferring ownership of all
public land in within their boundaries from SGC to local governments (law 1423); providing local
governments with the tools to rationally plan land use (law 711); empowering them, via taxes on land users
and fees for public land leasing, to benefit from increased land values; and by mandating auctioning use
rights to public land via competitive electronic auction together with the digital infrastructure to implement
such auctions (law 1444).
Third, to improve access to finance by small producers whose inability to access foreign capital had earlier
put them at a distinct disadvantage, a digital farmer registry was established to transparently target and
transfer agricultural state support, serve as basis fora digital agricultural marketplace, and reduce transaction
cost and increase competition in agricultural credit markets (law 985). This was complemented by
establishment of an independent Partial Credit Guarantee Facility (PCGF) to reduce the risk of lending to
producers who still had to establish a credit history (law 3205).
The mandatory shift from centralized in-person to fully electronic auctions run by local authorities for
any transfer of use rights to public land (law 1444) was one of the first laws to be fully regulated. Analysis
shows that this instantaneously doubled lease prices received by local governments (Deininger et al.
2022).6 A simple count of the status of resolutions needed for individual reform laws as per Jan. 2023
suggests that progress with regulating other laws, especially those mandating institutional reform (law
1423) was less swift (see appendix table 1), creating a danger of far-reaching land sector reforms not
being effective due to gaps in implementation.
4
From Jan. 1, 2024, legal persons will be able to purchase land up to a total limit of 10,000 ha. Moreover, even before 2024, current tenants hold a
pre-emptive right that they can transfer in case legal restrictions do not allow them to exercise it themselves.
5
This provision was highly controversial and resisted by foreign land users who often had made significant investments in land improvements and
feared to be left at a disadvantage especially with limited de facto enforceability of lease contracts.
6
Electronic auctions that had earlier been introduced without a proper regulatory framework were in fact less effective than in-person auctions.
7
2.2 Impacts of the war
Beyond leading to a shutdown of the cadaster and registry from end Feb. to end May for cyber-security
reasons that made market transactions impossible, the war resulted in three changes.7 The first and most
immediate change was direct damage to farmers’ land and structures and disruption of logistics by
increasing the cost of seaborne grain exports and damaging grain storage infrastructure. Interpretation of
high resolution satellite imagery suggests that about 16% of Ukraine’s storage facilities had been damaged
by mid-2022 (Khoshnood et al. 2022). At that point, 18.7% of village councils had sustained damage to
cropland either through burning, shelling, or heavy vehicle movement (Deininger et al. 2023b). Aggregate
assessments put the direct damage at that point at US$ 4.29 billion for the agriculture sector alone (Neyter
et al. 2022b).8
Second, reduced supply, damage to logistic structures, and war-related impediments that reduced the
volume and increased the cost of seaborne grain exports and required use of more expensive overland routes
(von Cramon-Taubadel 2022) significantly reduced output prices. Together with higher prices for inputs,
this led to a marked reduction in farm profits. Data sourced from suppliers’ catalogues and websites points
towards significant increases in input prices with prices for fertilizer having almost doubled and those for
ploughing and harvesting services increased by between 10% and 30%. Evidence from a survey of small
and medium-scale farms (and households) highlights a drop in farmgate prices of 28% for barley, 23% for
wheat, 18% for sunflower, 15% for maize, and 11% for soybean. Together, these factors reduced profits
and output market participation compared to the pre-war situation. At the same time, willingness to sell
land was very low and even for those interested in selling, the asking price was high. A possible explanation
is that formal social support continued during the war period and informal safety nets remained strong
(Deininger et al. 2023a), suggesting that, for the time being, the danger of registered distress sales of
agricultural land seems to remain limited.
Third, by illustrating the importance of strong governance including IT systems that are up to par in terms
of cybersecurity to ensure business continuity and the scope for shifting to a digital environment, the war
reinforced the importance and benefits from quick implementation of reforms to help reform Ukraine’s
agriculture sector.9 As a result, MAPF expedited implementation of a full digital State Agrarian Registry
(SAR). The SAR was launched in August 2022 and successfully used to fully transfer, by Nov. 2022, € 50
million as cash grants to about 35,000 small producers (with less than 120 ha or fewer than 100 cows)
7
In addition to suspending land market activity, unharvested winter crops, 21% for machinery, 14% for stored products, 6% for storage facilities
and the remainder for livestock, perennial crops and inputs.
8
Of this total, about 50% is for land (mining pollution) and unharvested winter crops, 21% for machinery, 14% for stored products, 6% for storage
facilities and the remainder for livestock, perennial crops and inputs.
9
Key regulatory provisions to implement land sector reforms are included in the relevant chapter of the anti-corruption law (No. 2322) adopted in
June 2022 to smooth the way towards EU accession.
8
whose eligibility had been checked digitally against state registries and crop maps derived from satellite
imagery. This experience motivated MAPF to mandate that all state support, donor assisted programs, and
banks that benefit from state funds via guarantees use this platform, a call already heeded by some
organizations.10
3. Data and descriptive evidence
Registry data suggests that, following market opening, agricultural land markets quickly overtook those for
residential land in terms of volume and maintained this advantage during the war period when a drop in
volume was combined with a slight increase in the price of sold parcels. The potential of mortgage markets
was not realized, most likely due to institutional reforms’ incomplete nature. Changes in transacted parcels’
land cover provide little basis for concerns about speculative sales and could be useful to monitor rural as
well as urban land market activity in the future.
3.1 Data sources and variable description
We use parcel-level data on registered sales or leases of agricultural and residential land,11 prices for these
where reported,12 and mortgages, from the Registry of Rights of Ukraine to provide national and sub-
national statistics on transaction volume and prices regularly published by Ukraine’s State Geocadaster
(SGC). Transacted parcels’ cadastral numbers allow us to recover their location (either as a parcel shape or
centroid) and use it to compute spatial characteristics such as land use at the time of transfer, distance to
the next city, primary road, and park, a land quality index based on soil maps, altitude, and slope that are
likely to affect agricultural profitability and thus land prices. Crop maps for 2019-22 elaborated using the
methodology described in Kussul et al. (2017); Shelestov et al. (2017), and Shelestov et al. (2020)
are then used to obtain information on parcels’ land use before, at, and after the time of transfer. 13
As a proxy for local institutional quality, we use the share of unregistered but cropped land at the village
level. We obtain this by subtracting the area included in the cadaster in 2021 (as a precondition for
registering rights) from the total area cultivated with crops in a village in the same year. Experts believe
that such informality differs between private and public land. For private land, failure to register is often
attributed to general distrust of the state or fears that registration might incur additional tax liabilities. In
10
Following the PSG, the SAR has been used to manage distribution of short-term grain storage by FAO, for distribution of emergency support
(e.g., generators) by MAPF and programs to use it to distribute seeds and fertilizer for the 2023 spring planting season have just been launched.
SAR is also intended to provide the platform for farmers’ application for partial credit guarantees to be provided by an independent agency.
11
In addition to residential land, transfers and mortgages are reported for recreational and industrial land which are relatively minor (1,526
recreational and 1,991 industrial land sales were observed during the period). We include these figures in the number of land transactions but report
prices only for residential land to allow comparability and avoid issues with outliers.
12
Despite a legal obligation to register prices, prices are reported only for about 50% for sales and lease transactions (see appendix table 2).
Although ensuring compliance with the law is important to obtain reliable figures and for banks to accurately value land pledged as collateral,
below we use prices without adjusting for selection. Values of mortgages reported in the registry are not credible and are thus not analyzed.
13
Crop maps are available at https://ukraine-cropmaps.com/ and files for transactions at https://land.gov.ua/monitorynh-zemelnykh-vidnosyn/.
9
the case of public land, informality is more likely to be linked to corrupt practices that local authorities
either endorse or are unable to resist. While we lack the data to distinguish between these two,14 we note
that, irrespective of the ultimate reason, the presence of cropped land to which rights cannot be registered
may create negative external effects on registered land by reducing the scope for counting on local
authorities for contract enforcement and increasing the likelihood of contracts being disputed, challenged,
or reneged on.
3.2 Transaction volumes and price trends in land markets
Daily transaction volumes from July 1, 2021, when the agricultural land sales market opened to Feb. 25,
2022, when transfers were no longer possible as registries were taken offline for security purposes and from
May 31, 2022, to Dec. 31, 2022 are displayed in figure 1, separately for agricultural (panel A) and
residential (panel B) land.15 For agricultural land sales, we note a gradual increase from less than 200 to
more than 900 daily transactions in at year-end in Dec. 2021 followed by a drop to about 200 daily sales
during the war. Residential land sales volume was more stable at about 300 daily transactions before Dec.
2021 when a similar year-end blip is observed, stabilizing at about 150 daily transactions with the war.
Table 1 provides monthly averages of registered land sales and leases, prices, and the number of mortgages
for agricultural (panel A) and residential land (panel B). After a modest start in July 2021 when agricultural
land sales (5,041) numbered about half of those for residential land (9,831), they accounted for almost three
times residential (28,448 vs. 11,890) in December 2021. With the war, the volume of agricultural land sales
dropped: number and area of monthly sales after June amount to 44% and 34%, respectively, of the pre-
war level, a decline that is slightly more modest (50% for number and area) for residential land.
Markets for new leases are thinner and more seasonal than those for sales: compared to 173,606 and 97,545
sales, only 43,609 and 1,948 new leases were entered over the entire period for agricultural and residential
land, respectively. Low liquidity in the formal lease markets may be due to three factors, namely (i) most
agricultural enterprises locked in long-term leases (mostly for the legal maximum of 49 years) on favorable
terms in the early 2000s when land was perceived to have little value; (ii) measures such as an automatic
renewal of expiring leases for one growing season were introduced soon after the start of the war to maintain
agricultural production with the imposition of martial law (Kvartiuk & Martyn 2022); 16 and (iii) the legal
prohibition on registering leases with less than 7 years creates additional uncertainty and instead of entering
14
Records on the nature of unregistered land are only available in paper form at the local level and efforts to systematically gather these were
interrupted by the outbreak of the war.
15
The suspension of land transactions during the initial phase of the war (from Feb. 25 to May 31, 2022) when the registry was taken offline for
security reasons to prevent cyber-attacks or manipulation of information, making land transactions impossible, is marked by a black bar.
16
Note that this measure was repealed, restoring pre-war rules for lease contracting, on Nov. 19, 2022.
10
into long-term leases on what might later turn out to be unfavorable terms, land owners might prefer short-
term informal arrangements.
Table 1 also highlights that mortgage markets have been extremely thin not only for agricultural but also
for residential land with only 708 and 2,704 registered mortgages for both types, respectively, in the entire
period. While unfinished institutional arrangements (including absence of historical land prices), remaining
constraints on land market liquidity such as the ability to transfer only to natural persons, and banks’ risk
aversion or lack of familiarity with the market could explain this for agricultural land, the low number of
mortgages for residential land where such restrictions are absent suggests the transaction cost of using land
as collateral remains high. Issues such as complex and costly procedures for mortgage registration and
difficulties in executing collateral in case of default should be explored as a matter of priority if the scope
for financial market development to catalyze a private sector led reconstruction is to be realized.
Information on prices is needed to provide parties with an accurate picture of land market conditions, for
institutions including the central bank to properly value land pledged as collateral to secure their loan
portfolios, and for aggregate analysis. While legislation to make price registration mandatory for all sales
and leases was passed early in the reform process, appendix table 2 points to weak enforcement even before
the war when prices were reported for 57% (54%) of sales for agricultural and residential land. Enforcement
worsened with the war; after June 1, 2022, prices are reported only for 24% and 26% of agricultural and
residential land transactions, respectively.
Registered price data show that, in contrast to the drop in transaction numbers, agricultural sales and lease
prices were slightly higher (by 11% and 5%) during the war than in the pre-war period. Although surprising
at first, especially in light of the drop in agricultural profits, the average post-war price of land (US$ 1,369)
would be consistent with expectations under a standard capitalization formula (Deaton & Lawley 2022)
even under the assumption that such low prices would persist with any future increases in profits likely
resulting in appreciation of prices for agricultural land.
Agricultural land comprises two types, namely (i) land for personal farming (formerly household plots) that
is normally located close to owners’ house or the settled area of the village and used to provide crops for
home consumption (Kvartiuk & Martyn 2021) and (ii) land for commercial agriculture, located at greater
distance and farmed in larger fields, often based on rental. Figures by category of land in appendix table 2
show that mean reported prices increased during the war for the former and slightly decreased for the latter.
To properly interpret these data, analysis that controls for parcel-level characteristics is needed to adjust for
compositional effects (i.e., land entering the market being of higher quality or with structures).
11
To illustrate regionally differentiated war effects on transaction numbers, we regress volume of agricultural
and residential sales and total mortgages on regional indicators,17 interacted with a post-war dummy (see
appendix table 3).18 Coefficients are graphically illustrated in figure 2 which displays the pre-(blue) and
post-war (red) number of monthly transactions and associated 95% confidence interval for agricultural land
(panel A), residential land (panel B) and mortgages (panel C). For agricultural land, the pre-war volume of
transactions was highest in the Center, followed by the North, the South, the West, and the East. The war
significantly reduced transaction volumes everywhere except in the West, eliminated markets completely
in the East and nearly in the South, while triggering a more severe decline in the North than the Center. The
pattern of residential land sales is similar but characterized by a lower level of transactions in all regions.
Mortgage markets, though very modest, were most active before the war in the North.
Figure 3 illustrates pre- and post-war transaction volumes at the oblast level.19 The size of each pie chart is
proportional to transaction volume while blue and yellow segments indicate the distribution of registered
sales between the pre-war and the war period. In line with minimal market activity in the East and South,
no or very few registered land sales were reported during the war in Kherson, Luhansk, Donetsk, Kharkiv,
and Zaporizye.20 Agricultural land sales remained at less than a quarter of their pre-war level in Mykolaiv,
Sumy, Chernihiv, and Kyiv city. By contrast, the (limited) pre-war sales volume was exceeded during the
war in Ivano-Frankivsk, possibly reflecting a westward shift of producers displaced in the East.
3.3 Changes in transacted parcels’ land cover
While opening of agricultural land markets was expected to encourage investment via increased demand
(tenure security) and credit supply (mortgages), there was also concern of speculative land acquisition
whereby purchasers would acquire land in anticipation of land price increases and leave it idle, potentially
giving rise to negative external effects. Although fiscal measures such as a land tax that will encourage
more productive land use are a preferable policy response to more crude administrative measures that are
more costly to implement and prone to manipulation or corruption at local level, information on changes in
land use for all the 83,852 parcels transacted in 2021 based on remotely sensed imagery provides an
opportunity to test this empirically. To do so, we drop forest and water and collapse all crops into one
‘cropland’ category (Kussul et al. 2022).
17
Oblasts are assigned to regions as follows: The Center includes Cherkasy, Dnipro, Khmelnytsky, Kropyvnytsky, Poltava, Vinntsya, and
Zhytomyr; the East includes Donetsk, Kharkiv, and Luhansk; the North includes Chernihiv, Kyiv, and Sumy; the South includes Kherson,
Mykolayiv, Odessa, and Zaporizhje; and the West includes Chernivtsi, Ivano-Frankivsk, Lviv, Rivne, Ternopil, Volyn, and Zakarpattia.
18
Before the war, agricultural land markets were most active in the Center (with 1,285 weekly sales), followed by the North (941), South (546),
East (405), and West (388).
19
Oblasts are first-level administrative divisions equivalent to states or districts. There are a total of 24 oblasts in Ukraine (including Crimea).
20
Conflict-affected oblasts are Cherihiv, Sumy, Kharkiv, Luhansk, Donetsk, Zaporzhje, and Kherson.
12
Table 2 presents summary results for all as well as personal and commercial farmland in panels A, B, and
C for the entire country (cols. 1 and 2) and oblasts with and without conflict in cols. 3 and 4 or 5 and 6,
respectively. At the national level, a more than 50% increase in built up parcels is combined with a slight
reduction (-4%) in parcels under crops and a slight increase (+3%) in those covered by grass or weeds. The
notion of transacted parcels being used more intensively is reinforced by disaggregating parcels by conflict
status: in oblasts not affected by conflict, there is no change in the number of parcels under crops but a
significant increase in those with structures and a significant drop in those under grassland. By contrast, in
conflict-affected oblasts, there is still an increase in built up parcels but a reduction in cropped ones together
with an increase in the share of those under grass or uncultivated.
Distinguishing by type of land suggests that most of the shift from crop to grassland in conflict affected
oblasts is observed on commercial farmland and that in oblasts unaffected by conflict, a slight increase in
cultivated area on land for personal farming matches a slight decrease of commercial farmland. Appendix
table 4 presents transition matrices with land use in 2021 and 2022 for agricultural parcels transacted in the
sales market in 2021 at the national level (panel A) and separately for areas affected by conflict (panel B)
and those unaffected by conflict (panel C).
This suggests that any changes in transferred parcels’ cultivation status from cropped to uncultivated land
between the 2021 and 2022 cropping seasons were more likely due to conflict conditions than to
speculation. It also implies that continued monitoring of land cover in rural areas will be useful and could
possibly be extended to urban areas using high resolution imagery to assess war-related damages (Mueller
et al. 2021) and possibly also monitor progress with reconstruction.
4. Parcel-level land price analysis
Construction of a hedonic land price index allows us to obtain an index of intertemporal price changes net
of parcel attributes and to quantify how parcel characteristics affect market prices for agricultural land.
Comparing market price to administratively set ‘normative’ value points towards sizeable discrepancies
that limit the value of the normative value as a floor price to prevent distress sales even if it were enforced.
Use of predicted market prices to value land for administrative purposes could thus offer many advantages.
4.1 Evolution of land prices over time
The evolution of raw prices for (commercial) agricultural and residential land, normalized to equal 100 in
February 2022 as illustrated in figure 4 shows that, as indicated in table 1, the post-war decrease in prices
for residential land was more pronounced than that for commercial agriculture. Although data intensive
(Hill 2013), the potential of hedonic price indexes to control for a host of factors is a distinct advantage
13
(Diewert et al. 2020) that led to their widespread adoption in statistical agencies (Gouriéroux & Laferrère
2009) and research (Diewert & Shimizu 2022).
While lack of data on the nature of residential land sales (e.g., if structures were transferred together with
the land) precludes construction of a hedonic price index for this type of land, remotely sensed data on land
use and open-source information on infrastructure access construction of such an index for agricultural
land. To do so, we regress sale or lease prices for transacted parcels on their characteristics and rayon and
month fixed effects, overall and separately for land designated for personal and commercial farming.
Indexing parcels by i, and months by t, the equation to be estimated is
������������������ = ������������ + ������������������������ + ������������ + ������������������ (1)
where ������������������ is the sales or lease price for parcel i sold in month t; Xit is a vector of time invariant parcel
characteristics including area, slope, and altitude, land use at the time of transfer (crops, forest, built up or
grass/uncultivated), a land quality index, the distance to the next city and national or regional park, and the
share of unregistered land at village level.21 ������������ s are rayon fixed effects; ������������ s are month fixed effects; and ������������������
is a random error term. Standard errors are adjusted for clustering at rayon level throughout.
Descriptive statistics for the regression sample of 71,779 sold parcels with and 79,997 ones without prices
in appendix table 5 (panel A) and the 10,854 leased ones with and 15,994 without prices (panel B) point
towards vast differences between parcels for personal and commercial farming: the former are smaller (1.1
vs. 3.6 ha) and more valuable (US$10,775 vs. 1,178 per ha), possibly due to presence of structures. Land
use also differs between these groups: with 92.5% (43.2) of land for commercial (personal) farming covered
by crops; 5.6% (28.9%) covered by grass and 1.8% (22.2%) by forest; and little (5.8%) built up. Differences
between transferred parcels for which prices are or are not reported are comparatively small, we report
rayon and month fixed effects regression results below.22
Regressions for sales prices are presented in table 3, overall (col. 1) and separately for personal (col. 2) and
commercial (col. 3) farmland with the first set of (12) coefficients focusing on parcel characteristics. Col.
1 suggests that market prices for land in commercial agriculture are 31% higher than for those in personal
agriculture once parcel characteristics are controlled for. We discuss results separately for commercial and
personal farming land. For the former, a significant premium is estimated for larger parcels (coefficient of
0.106) at higher elevations (0.143) and those covered with crops (0.285) or trees (0.146). A one-point
increase of the land quality index is estimated to increase the sales price by 0.5%. While distance to the
next city is insignificant, access to roads matters, with the coefficient (-0.013) suggesting that doubling the
21
As (Kvartiuk & Martyn 2021) explain, land used for personal farming refers to house and garden plots and was not subject to the moratorium.
22
Results from a Heckman-type specification are not significantly different and included in the appendix/available on request.
14
distance to roads reduces prices for commercial farmland by 1.3%. A positive estimated coefficient on
distance to parks is consistent with the notion that, whether due to environmental regulations, a greater
number of odd-shaped parcels, or congestion externalities, proximity to a park reduces prices of commercial
farmland. Also, the coefficient on the share of unregistered village land is significant and negative (-0.124),
possibly reflecting a discount for the higher risk of rights being challenged in such a setting.
By comparison, for personal farmland coefficients of -0.464 on land size, -0.006 on slope, and -0.039 on
altitude indicate that smaller parcels of land for personal use fetch higher prices, consistent with a ‘small
parcel premium’ in US agriculture (Brorsen et al. 2015). Partly non-agricultural use of such land is in line
with premia for built up or forested land (with point estimates of 0.286 and 0.218) and highly significant
and large coefficients on distance to the next city and road (-0.021 and -0.024). The level of unregistered
land at village level is insignificant, in line with the notion that rights to land close to the homestead are
often secured through personal presence.
Conditional on controlling for other factors (parcel and time-invariant characteristics), the ������������ s provide an
estimate of a quality-adjusted monthly price index. Using July 2021 as the base month, we note that this
index almost halved over the period, a phenomenon driven by a drop in the value of land used for personal
farming. As can be seen in table 3 and figure 5, quality-adjusted prices for such land declined by 21% and
31% from their July 2021 value in Jan. and Feb. 2022 and bottomed out in Oct 2022, 62.2% below this
value. By comparison, commercial farmland prices were stable in the pre-war period (with a 5-7% gain in
Oct. and Nov. 2021) but dropped by about 15% after the start of the war.
Table 3 col. 4-6 report coefficients for lease price regressions where, given the thin and seasonal market
(with less than 11,000 valid transactions concentrated in July/August of each year as per table 1), we focus
primarily on commercial farmland (col. 6) compared to the sales regressions. Estimated coefficients on land
quality and road distance are much larger than in the sales regression, indicating the short-term impact of
these variables. Also, the coefficient on the share of unregistered land at village level is much larger (-0.320
for commercial and -0.513 for personal farmland) than for sales. This provides micro-level support for the
notion, first raised and tested in a cross-country context by Casas-Arce and Saiz (2010) that, if enforcement
via the legal system is costly, ownership rights are preferred. Given the seasonal nature of the lease market,
we only include an indicator variable for leases concluded during the war. Estimated coefficients on this
variable are slightly smaller in absolute terms than those for sales (-0.422 for personal farming and -0.155,
significant only at 10% for commercial farming land), but they are consistent in terms of a larger and more
significant war effect on prices for land under household plots than commercial farmland.
15
4.2 Comparing market prices to normative values
As registry data include the normative value for each transacted parcel,23 we can assess the relationship
between normative and registered sales price (market value) at parcel level. Deviations of this ratio from
unity are policy relevant as all land-related taxes, including those on presumed profits by commercial farms,
are linked to the normative value, any mismatch will affect the amount of tax revenue local governments
are able to obtain. Moreover, to prevent distress sales, the law prohibits land sales with prices below a
parcel’s normative value; and given its ready availability, the normative value may well be used as a basis
for compensation in the future.
As a first check, figure 6 presents scatterplots of normative value vs. market price (in logs) with each dot
representing a transaction, separately for commercial (panel A) and personal farmland (panel B) together
with the 45-degree line that should be the locus for all transactions if normative value and market price
coincide. Two observations stand out: First, although correlated with normative values, parcels’ sales prices
are often significantly above it. The simple correlation between normative value and sales price is 0.69 for
commercial farming while only 0.28 for personal farming, partly because the normative value is capped at
about US$ 2,500/ha (e7.8) whereas many transactions far exceed this value (see figure 6 panel B). Appendix
table 5 reinforces this by showing that, with 20%, the difference between sales price and normative value
remains modest for commercial farmland but, with mean market prices more than 6 times the normative
value, is vast for land in personal farms, suggesting that use of the normative value rather than market
values reduces revenue for local governments, similar to use value assessment in the US (Bigelow & Kuethe
2023). Interestingly, a relatively large share of land sales for the latter is at a price well below normative
value, implying that legislation establishing this as a minimum purchase price is widely ignored.
As we have information on current land use of transferred parcels including land cover from satellite
imagery, we can assess the extent to which the gap between normative and market value can be explained
by such attributes (vs. other dummies). To do so, we regress the ratio of the sales price to normative value
on parcel characteristics as well as oblast dummies and their interaction with a war dummy. Interacting
dummies for administrative units with a war dummy provides an indication whether with the war, the
normative value-to-price gap widened, implying an increase in the incidence of distress sales.
Table 4 reports results for commercial (col. 1 and 2) and personal farmland (col. 3 and 4). Constants of
0.986 for commercial and 0.675 for personal farmland confirm that sales prices for the latter are quite close
23
The ‘normative value’ which, by law, provides the basis for taxation and leasing out of public land, is based on the presumed rental income that
can be obtained from the parcel from effective use for its intended purpose. For agricultural parcels it accounts for land quality, distance to
infrastructure, parcel shape, and other environmental conditions. It is indexed to inflation with annual updates that adjust for inflation in the previous
year made on Jan. 15 of every year by State Geocadaster (SGC).
16
to normative values on average and that easily observable attributes (size or crop cover) increase the ratio
while attributes that are more difficult to measure such as distance to infrastructure, slope, or altitude (and
whether a parcel of farmland for personal use is built up) reduce it. This provides support to the notion that,
despite recent updates normative values continue to be based on a concept of productivity in terms of
physical output rather than economic value.
A second conclusion emerging from table 4 is that normative values seem to be updated without accounting
for changes in land use such as building of new structures. This implies that shifting from normative value
to market values, possibly determined using an equation such as (1) above would be preferable to increase
local government’s land revenue. It also implies that use of the normative value as a floor price may not be
the most effective way to address distress sales. Instead, flagging observations that deviate significantly
from market prices after controlling for observable attributes and singling them out for attention (e.g., in
terms of automatically informing the seller about market prices and/or establishing a waiting period) might
be a more effective approach that would be less open to manipulation.
5. Conclusion and policy implications
Our analysis shows that fears of land market opening causing havoc did not materialize and that even under
war conditions, the positive results associated with moves to transparent and digital service delivery, via
mandatory e-auctions and the SAR, support the general approach taken to reforms. At the same time, the
war caused interruptions, especially to institutional reforms, that prevented the realization of benefits in
terms of financial market development and decentralization. Evidence on the resilience of markets,
especially for commercial agricultural land, suggests that fast-tracking institutional reforms, including
whatever regulations are needed on the financial and safeguards side to prevent exacerbating the impact of
war-related disturbances, can provide a basis for realizing these benefits, especially with full opening of
agricultural land markets around the corner.
The impact of land reforms in rural areas can be multiplied by expanding these reforms to urban areas,
where constraints to clear property records and data interoperability are more severe and potential benefits
even larger than in rural ones. Recent legal initiatives suggest that the challenge of re-building large parts
of Ukraine’s urban areas decimated by the war may provide an opportunity to extend property reform
benefits to urban areas. Reform initiatives include (i) moving towards a fully digital register of real property
objects; (ii) moving towards market-based mass valuation of land and associated property to provide a
buoyant tax base for urban governments; and (iii) planning land use in ways that maximize local revenue,
allow effective service provision, and respect social and environmental safeguards.
17
Table 1: Monthly volume of registered sales, leases, and mortgages for agricultural and residential land Jul. 2021-Dec. 2022
Land sales Land leases No. of
No. of Area (ha) Price No. of Area (ha) Price Mortgages
Sales (US$/ha) Leases (US$/ha)
Panel A: Agricultural land
July 21 5,041 5,922 4,146 13,830 34,852 70
Aug 21 9,934 18,303 1,493 10,143 26,884 76
Sep 21 14,523 32,763 1,311 3,191 7,801 56
Oct 21 17,054 41,514 1,309 983 2,261 64 19
Nov 21 21,159 52,194 1,233 498 1,370 76 144
Dec 21 28,448 71,229 1,165 4,596 11,364 82 145
Jan 22 12,190 29,661 1,114 75 294 63 109
Feb 22 16,794 43,437 1,124 67 284 46 91
Jun 22 4,261 6,857 1,601 131 288 53 35
July 22 6,978 10,966 1,528 4,809 9,116 84 8
Aug 22 7,740 12,289 1,375 3,264 6,487 68 18
Sep 22 6,104 12,167 1,214 20 102 N/A 17
Oct 22 7,511 13,400 1,367 1,837 3,863 68 31
Nov 22 7,695 15,059 1,524 127 341 33 32
Dec 22 8,174 16,760 1,164 38 195 53 59
Before war 15,643 36,878 1,228 4,173 10,639 73 101.6
After war 6,923 12,499 1,369 1,461 2,913 77 28.6
Ratio 0.44 0.34 1.11 0.35 0.27 1.05 0.28
Total 173,606 382,521 43,609 105,502 708
Panel B: Residential land
July 21 9,831 1,206 33,071 619 73 7,009
Aug 21 8,908 1,089 34,187 364 135 8,427
Sep 21 8,868 1,141 32,536 122 24 6,756
Oct 21 9,111 1,344 31,771 47 17 18,551 140
Nov 21 9,121 1,188 30,250 36 7 10,825 364
Dec 21 11,890 1,603 29,309 140 17 7,139 514
Jan 22 5,176 654 33,807 5 2 7,332 275
Feb 22 5,046 647 29,915 14 2 27,149 328
Jun 22 2,788 369 24,807 8 7 27,147 64
July 22 4,543 596 22,408 304 25 13,806 132
Aug 22 4,987 668 21,612 132 10 7,905 152
Sep 22 3,958 537 20,898 6 1 9,941 187
Oct 22 4,551 597 17,202 118 13 3,786 136
Nov 22 4,604 679 16,918 21 4 12,886 201
Dec 22 4,163 549 17,792 12 1 21,880 211
Before war 8,494 1,109 31,676 168 35 7,869 324.2
After war 4,228 571 20,033 86 9 9,737 154.7
Ratio 0.50 0.51 0.63 0.51 0.26 1.24 0.48
Total 97,545 12,867 1,948 338 2,704
Source: Own computation based on parcel-level statistics from Ukraine registry of rights.
Note: Data on mortgages for July-Sept. 2021 are not available. Transaction prices are median prices for cases where prices are
reported.
18
Table 2: Changes in land cover for transacted parcels
Entire country Oblasts with no conflict Oblasts with conflict
2021 2022 2021 2022 2021 2022
Panel A: All farmland
Built up 2,779 4,377 2,271 3,625 508 752
Cropland 58,720 56,539 36,613 36,655 22,107 19,884
Grassland 22,353 22,936 16,525 15,129 5,828 7,807
No. of obs. (parcels sold 2021) 83,852 83,852 55,409 55,409 28,443 28,443
Panel B: Personal farmland
Built up 2,762 4,314 2,255 3,573 507 741
Cropland 19,710 19,992 13,750 14,341 5,960 5,651
Grassland 19,553 17,719 15,214 13,305 4,339 4,414
Panel C: Commercial farmland
Built up 17 63 16 52 1 11
Cropland 39,010 36,547 22,863 22,314 16,147 14,233
Grassland 2,800 5,217 1,311 1,824 1,489 3,393
Source: Own computation from 2021 and 2022 crop maps (available on https://ukraine-cropmaps.com) using field coordinates for
83,852 agricultural parcels sold between July and December 2021 as summarized in the transition matrices in appendix table 5.
Note: For all included parcels, the table provides the total number of parcels in each land use class in 2021 and 2022.
19
Table 3: Land price determinants at parcel level
Sales Leases
All Personal Comm. All Personal Comm.
Parcel area in ha -0.343*** -0.464*** 0.106*** 0.100*** 0.006 0.115***
(0.003) (0.005) (0.003) (0.008) (0.015) (0.009)
Commercial agriculture 0.309*** -0.017
(0.008) (0.019)
Used for crops at time of transfer 0.148*** 0.071*** 0.285*** 0.133*** 0.149*** 0.132***
(0.009) (0.012) (0.010) (0.019) (0.027) (0.025)
Built up at time of transfer 0.645*** 0.286***
(0.020) (0.023)
Forested at time of transfer 0.367*** 0.218*** 0.146*** -0.074** 0.003 -0.086**
(0.012) (0.014) (0.018) (0.033) (0.053) (0.041)
Land quality index -0.001 -0.001 0.005*** 0.009*** 0.010*** 0.011***
(0.001) (0.001) (0.000) (0.001) (0.003) (0.001)
Dist. to next city (km) -0.038*** -0.021*** 0.001 -0.048*** -0.089*** 0.014
(0.003) (0.003) (0.003) (0.010) (0.015) (0.014)
Dist. to primary road (km) -0.022*** -0.024*** -0.013*** -0.006 -0.010 -0.034***
(0.003) (0.004) (0.002) (0.008) (0.013) (0.010)
Dist. to next park (km) 0.005* -0.003 0.018*** 0.017** -0.007 0.038***
(0.003) (0.005) (0.002) (0.007) (0.011) (0.008)
Village share of unregistered land -0.205*** -0.053 -0.124*** -0.534*** -0.513*** -0.320***
(0.031) (0.046) (0.025) (0.089) (0.150) (0.109)
Slope (%) -0.012*** -0.006** -0.011*** -0.007** -0.009 -0.003
(0.002) (0.003) (0.001) (0.004) (0.006) (0.004)
Altitude (m) 0.070*** -0.039*** 0.143*** 0.170*** -0.232*** 0.219***
(0.010) (0.015) (0.009) (0.036) (0.061) (0.045)
War dummy -0.269*** -0.422*** -0.155*
(0.059) (0.081) (0.085)
August 2021 -0.036* 0.008 0.035*
(0.021) (0.028) (0.019)
September 2021 -0.099*** -0.044 0.029
(0.020) (0.027) (0.018)
October 2021 -0.079*** -0.030 0.068***
(0.020) (0.027) (0.018)
November 2021 -0.176*** -0.110*** 0.047***
(0.020) (0.027) (0.018)
December 2021 -0.231*** -0.190*** 0.002
(0.019) (0.026) (0.018)
January 2022 -0.255*** -0.216*** -0.000
(0.021) (0.029) (0.019)
February 2022 -0.277*** -0.314*** 0.018
(0.020) (0.028) (0.018)
June 2022 -0.387*** -0.412*** -0.060**
(0.031) (0.042) (0.027)
July 2022 -0.426*** -0.520*** -0.096***
(0.028) (0.039) (0.024)
August 2022 -0.521*** -0.603*** -0.165***
(0.028) (0.038) (0.024)
September 2022 -0.518*** -0.601*** -0.191***
(0.029) (0.042) (0.024)
October 2022 -0.522*** -0.622*** -0.148***
(0.029) (0.040) (0.024)
November 2022 -0.501*** -0.544*** -0.185***
(0.028) (0.039) (0.025)
December 2022 -0.463*** -0.491*** -0.167***
(0.043) (0.064) (0.032)
Constant 7.072*** 7.738*** 5.660*** 3.066*** 5.301*** 2.477***
(0.059) (0.084) (0.051) (0.187) (0.324) (0.231)
Number of observations 71,779 37,244 34,535 10,854 3,819 7,035
Within R-squared 0.237 0.318 0.121 0.048 0.041 0.065
Note: Rayon fixed effects included throughout. Standard errors in parentheses, * p<0.10, ** p<0.05, *** p<0.010.
20
Table 4: Median regression for ratio of normative value to sales price:
Land for personal farming Commercial farmland
War dummy War dummy
interaction interaction
Log area in ha 0.0874*** 0.0004***
(0.0033) (0.0001)
Value of land quality, index 0.0032*** 0.00004***
(0.0003) (0.00001)
Crop 0.0307*** 0.0075**
(0.0074) (0.0032)
Built up -0.2818***
(0.0233)
Forest and low trees -0.0201* -0.0933**
(0.0107) (0.0449)
Log distance to the nearest city in km 0.0066*** -0.00003***
(0.0024) (0.0000)
Log distance to primary road in km -0.0002 -0.00003***
(0.0026) (0.0000)
Log distance to parks in km 0.0003 0.00002*
(0.0019) (0.00001)
Share of unregistered land at village level -0.0636** -0.0006***
(0.0256) (0.0002)
Log slope -0.0026* -0.00002***
(0.0016) (0.0000)
Log altitude meters 0.0203*** -0.0002***
(0.0074) (0.00004)
Vinntsya -0.1216*** -0.4634*** -0.0224 0.0194
(0.0243) (0.0687) (0.0238) (0.0772)
Rivne, Volyn -0.0232 0.0061 0.0049 0.0002
(0.0152) (0.0213) (0.0149) (0.1093)
Dnipropetrovsk 0.0057 -0.0255 0.0045** -0.0001
(0.0218) (0.0465) (0.0023) (0.0000)
Donetsk -0.0417 0.0045*
(0.0260) (0.0023)
Zhytomyr -0.0260 -0.4405*** 0.0042** -0.4908***
(0.0209) (0.1263) (0.0022) (0.1244)
Zaporozhje 0.0474** 2.2062*** 0.0047** -0.0003
(0.0218) (0.1173) (0.0023) (0.0002)
Chernivtsi, Ivano-Frankivsk, Zakarpatie -0.3976*** 0.0498 0.0004 0.0030
(0.0311) (0.0426) (0.0056) (0.0060)
Kiev -0.5336*** -0.0573** -0.1638*** 0.1588***
(0.0165) (0.0250) (0.0210) (0.0253)
Kirovohrad -0.0357* 0.0323 0.0044* 0.0000
(0.0185) (0.0452) (0.0023) (0.0001)
Luhansk -0.0097 0.0048**
(0.0556) (0.0023)
Lviv -0.5934*** 0.0314 -0.2915*** 0.1827**
(0.0242) (0.0278) (0.0647) (0.0845)
Mykolayiv 0.0226 -0.5432*** 0.0046** -0.0002
(0.0225) (0.0591) (0.0023) (0.0002)
Odessa -0.2431*** 0.1749*** 0.0043* -0.0000
(0.0202) (0.0357) (0.0023) (0.0001)
Poltava -0.0116 0.0444 0.0041* -0.0029**
(0.0163) (0.0304) (0.0023) (0.0012)
Sumskaya -0.0040 0.0050 -0.0078 0.0022
(0.0159) (0.0894) (0.0056) (0.0300)
Ternopil -0.4843*** 0.0580 0.0005 -0.2764***
(0.0445) (0.1107) (0.0035) (0.0615)
Kharkiv 0.0086 -0.0351 0.0044* -0.0001
(0.0180) (0.0934) (0.0023) (0.0021)
Kherson 0.0946*** 0.0047**
(0.0248) (0.0023)
21
Khmelnytsky -0.1524*** 0.0992*** 0.0043* -0.0002
(0.0173) (0.0247) (0.0023) (0.0021)
Cherkasy -0.0562*** 0.0170 0.0041* -0.0001
(0.0196) (0.0322) (0.0023) (0.0005)
Chernihiv 0.1556** 0.0055**
(0.0713) (0.0023)
Constant 0.6747*** 0.9864***
(0.0379) (0.0044)
N 10,729 21,128
Standard errors in parentheses. * p<0.10, ** p<0.05, *** p<0.010
22
Figure 1: No. of daily land sales
Panel A: Agricultural land
Panel B: Residential land
23
Figure 2: Coefficients (and confidence intervals) from total and regional regressions for land market activity
Panel A: No. of registered sales of agricultural land
Panel B: No. of registered sales of residential land
24
Panel C: No. of registered mortgages (agric. & residential)
25
Figure 3: No. of monthly registered sales in pre-war and war periods by oblast
Panel A: Agricultural
Panel B: Residential
Note: Colors indicate how oblasts are mapped into regional groupings used in the paper.
26
Figure 4: National sales price index for agricultural land (for commercial use) and residential land
27
Figure 5: Hedonic log sales price series net of observed characteristics and Rayon fixed effects
28
Figure 6: Scatterplot of normative against market value for agricultural land parcels sold
Panel A: Land for commercial agriculture
Panel B: Land for personal farming
Note: Each dot corresponds to a transaction
29
Appendix table 1: Status of subsidiary regulation for major reform laws
Law Topic No. regulations No. of regulations by status in the process
# required Not started Drafting Consultation Adopted
552 Turnover law 4 4
340 Anti-raider 5 5
711 Planning 10 1 9
554 Spatial data interoperability 3 3
985 State support 6 1 5
1423 Institutional reform (SGC ) 23 12 4 7
1444 E-auctions for public land 1 1
3205 Part. Credit Guarantee Fac. 9 2 7
Source: Own elaboration based on monitoring by legal reform working group
Note: Status is as of Jan. 26, 2023. Empty cells mean 0.
30
Appendix table 2: Monthly volume of registered sales, leases, and mortgages for agricultural and residential land Jul. 2021-Dec. 2022
2021 2022 Relative to invasion
Jul Aug Sep Oct Nov Dec Jan Feb Jun Jul Aug Sep Oct Nov Dec Before After
Agricultural land
No of sales registered 5,041 9,934 14,523 17,054 21,159 28,448 12,190 16,794 4,261 6,978 7,740 6,104 7,511 7,695 8,174 15,643 6,923
Total area (ha) 5,922 18,303 32,763 41,514 52,194 71,229 29,661 43,437 6,857 10,966 12,289 12,167 13,400 15,059 16,760 36,878 12,499
Share reporting price 0.50 0.60 0.59 0.59 0.57 0.56 0.58 0.56 0.28 0.25 0.25 0.26 0.22 0.22 0.21 0.57 0.24
Median sale price (US$/ha) 4,146 1,493 1,311 1,309 1,233 1,165 1,114 1,124 1,601 1,528 1,375 1,214 1,367 1,524 1,164 1,228 1,369
No of new leases 13,830 10,143 3,191 983 498 4,596 75 67 131 4,809 3,264 20 1,837 127 38 4,173 1,461
Total lease area in ha 34,852 26,884 7,801 2,261 1,370 11,364 294 284 288 9,116 6,487 102 3,863 341 195 10,639 2,913
Share reporting price 0.47 0.49 0.43 0.37 0.41 0.41 0.48 0.36 0.24 0.50 0.41 0.15 0.47 0.41 0.34 0.46 0.46
Median lease price (US$/ha) 70 76 56 64 76 82 63 46 53 84 68 N/A 68 33 53 73 77
No. of mortgages registered 19 144 145 109 90 35 8 18 17 31 32 59 101 29
Personal farm
No of parcels sold 3,413 5,559 7,393 7,897 9,480 12,936 5,579 6,676 2,394 3,875 4,186 3,105 3,902 3,958 3,923 7,367 3,620
Total sale area in ha 2,282 5,289 9,897 10,280 11,524 16,879 7,812 9,370 2,123 3,069 3,241 3,447 3,067 3,581 3,484 9,167 3,144
Share reporting price 0.47 0.61 0.59 0.60 0.59 0.57 0.58 0.56 0.29 0.23 0.26 0.27 0.23 0.25 0.22 0.58 0.25
Median sale price (US$/ha) 5,536 2,400 1,518 1,553 1,209 1,152 1,031 1,052 2,410 2,317 2,084 1,777 2,394 2,030 1,781 1,316 2,071
No of new leases 4,610 3,215 1,063 328 116 1,628 29 9 53 1,423 1,047 5 548 22 5 1,375 443
Share reporting price 0.47 0.51 0.45 0.38 0.48 0.47 0.62 0.78 0.19 0.52 0.41 0.40 0.53 0.50 0.60 0.48 0.48
Total lease area in ha 7,206 5,205 1,776 488 202 2,684 56 10 113 1,900 1,566 7 756 34 7 2,204 626
Median lease price (US$/ha) 73 91 84 79 116 109 101 111 112 93 68 N/A 68 55 N/A 84 78
No. of mortgages registered 6 32 47 42 79 32 6 15 8 9 25 36 41 19
Commercial farm
No of parcels sold 970 3,466 6,101 8,141 10,415 13,559 5,800 9,418 1,480 2,505 2,898 2,526 3,064 3,188 3,729 7,234 2,770
Total sale area in ha 3,494 12,488 22,038 30,204 38,594 50,234 20,305 32,745 4,043 6,899 8,168 7,226 8,736 9,521 11,007 26,263 7,943
Share reporting price 0.55 0.57 0.57 0.57 0.54 0.54 0.58 0.56 0.23 0.24 0.21 0.24 0.19 0.16 0.18 0.54 0.26
Median sale price (US$/ha) 1,173 1,198 1,192 1,210 1,207 1,133 1,119 1,124 1,189 1,079 921 940 938 958 959 1,164 969
No of new leases 8,656 6,533 2,017 575 359 2,657 40 54 67 3,295 2,098 15 1,228 101 33 2,611 977
Total lease area in ha 24,368 19,066 5,631 1,506 1,125 7,495 176 262 152 7,032 4,540 95 2,854 251 188 7,454 2,159
Share reporting price 0.46 0.49 0.41 0.35 0.37 0.36 0.40 0.30 0.30 0.50 0.40 0.07 0.45 0.40 0.30 0.45 0.45
Median lease price (US$/ha) 70 70 49 67 60 73 38 43 35 80 70 N/A 75 33 71 68 76
No. of mortgages registered 1 102 81 45 5 0 0 0 1 17 2 17 47 5
Residential property
No of properties sold 9,831 8,908 8,868 9,111 9,121 11,890 5,176 5,046 2,788 4,543 4,987 3,958 4,551 4,604 4,163 8,494 4,228
Total sale area in ha 1,206 1,089 1,141 1,344 1,188 1,603 654 647 369 596 668 537 597 679 549 1,109 571
Share reporting price 0.53 0.54 0.56 0.56 0.55 0.53 0.55 0.55 0.31 0.27 0.26 0.26 0.24 0.26 0.24 0.54 0.26
Median sale price (US$/ha) 33,071 34,187 32,536 31,771 30,250 29,309 33,807 29,915 24,807 22,408 21,612 20,898 17,202 16,918 17,792 31,676 20,033
No of new leases 619 364 122 47 36 140 5 14 8 304 132 6 118 21 12 168 86
Total lease area in ha 73 135 24 17 7 17 2 2 7 25 10 1 13 4 1 35 9
Share reporting price 0.52 0.54 0.61 0.64 0.61 0.56 0.80 0.50 0.88 0.56 0.53 0.83 0.59 0.52 0.33 0.54 0.56
Median lease price (US$/ha) 7,009 8,427 6,756 18,551 10,825 7,139 7,332 27,149 27,147 13,806 7,905 9,941 3,786 12,886 21,880 7,869 9,737
No. of mortgages registered 124 311 438 236 270 53 92 133 149 113 173 211 276 132
Source: Own computation from Registry of Rights of Ukraine, January 2023
Note: The columns “Before” and “After” refers to the months before and after the Russian invasion on February 24, 2022. In these columns, number and area of transactions are, for
comparison purpose, monthly averages. Median prices are not computed for months with very few reported prices.
31
Appendix table 3: Mean war effects on no. of registered transactions & prices per ha of transferred land by region
No of transactions Prices (US$/ha)
Agric. land Resid. land Mortgages Agric. land Resid. land
Center 1310.824*** 562.706*** 13.048*** 1496.940* 16053.249***
(49.468) (16.604) (4.119) (856.664) (3496.674)
Center # War -573.617*** -192.982*** -2.530 -360.524 -7354.489
(72.912) (24.472) (5.409) (1262.646) (5153.786)
East 417.676*** 156.088*** 7.286* 2793.398*** 25613.209***
(49.468) (16.604) (4.119) (856.664) (3496.674)
East # War -393.470*** -140.295*** -6.458 2509.741* 21815.248***
(72.912) (24.472) (5.409) (1287.637) (5203.217)
North 971.765*** 553.088*** 29.619*** 5089.350*** 46782.731***
(49.468) (16.604) (4.119) (856.664) (3496.674)
North # War -627.110*** -339.019*** -16.205*** 868.517 -17570.138***
(72.912) (24.472) (5.409) (1262.646) (5153.786)
South 548.265*** 241.324*** 6.762 2050.455** 27645.946***
(49.468) (16.604) (4.119) (856.664) (3496.674)
South # War -401.885*** -171.703*** -5.486 -13.245 -2989.776
(72.912) (24.472) (5.409) (1262.646) (5153.786)
West 456.559*** 500.118*** 17.476*** 2784.245*** 34340.177***
(49.468) (16.604) (4.119) (856.664) (3496.674)
West # War -84.386 -177.842*** -4.235 33.248 -5656.125
(72.912) (24.472) (5.409) (1262.646) (5153.786)
No obs. 315 315 250 313 314
R-squared 0.844 0.935 0.340 0.335 0.706
Standard errors in parentheses. * p<0.10, ** p<0.05, *** p<0.010.
Note: Regressions are for 315 week-region observations for land sales except for mortgages where the no. of obs. is 250 and for
prices where no prices are reported for 1 or 2 weeks, respectively.
32
Appendix table 4: Land cover transition matrix for parcels transacted in 2021
I: All agricultural land
2022
2021 Built up Crop Grass Garden/parks Total %
Panel A: National
Built up 2,131 77 73 498 2,779 3.31
Crop 246 52,259 5,289 926 58,720 70.03
Grass 1,198 3,646 9,658 999 15,501 18.49
Garden/parks 802 557 1,209 4,284 6,852 8.17
Total 4,377 56,539 16,229 6,707 83,852
% 5.22 67.43 19.35 8.00
Panel B: Areas not directly affected by conflict
Built up 1,747 76 63 385 2,271 4.10
Crop 205 33,594 2,228 586 36,613 66.08
Grass 1,022 2,509 6,888 708 11,127 20.08
Garden/parks 651 476 1,045 3,226 5,398 9.74
Total 3,625 36,655 10,224 4,905 55,409
% 6.54 66.15 18.45 8.85
Panel C: Conflict-affected areas only
Built up 384 1 10 113 508 1.79
Crop 41 18,665 3,061 340 22,107 77.72
Grass 176 1,137 2,770 291 4,374 15.38
Garden/parks 151 81 164 1,058 1,454 5.11
Total 752 19,884 6,005 1,802 28,443
% 2.64 69.91 21.11 6.34
II. Land used for commercial agriculture
2022
2021 Built up Crop Grass Garden/parks Total %
Panel A: National
Built up 10 2 3 2 17 0.04
Crop 12 35,716 3,040 242 39,010 93.27
Grass 33 761 1,468 93 2,355 5.63
Garden/parks 8 68 114 255 445 1.06
Total 63 36,547 4,625 592 41,827
% 0.15 87.38 11.06 1.42
Panel B: Areas not directly affected by conflict
Built up 9 2 3 2 16 0.07
Crop 7 21,903 881 72 22,863 94.51
Grass 30 355 587 36 1,008 4.17
Garden/parks 6 54 68 175 303 1.25
Total 52 22,314 1,539 285 24,190
% 0.21 92.24 6.36 1.18
Panel C: Conflict-affected areas only
Built up 1 1 0.01
Crop 5 13,813 2,159 170 16,147 91.55
Grass 3 406 881 57 1,347 7.64
Garden/parks 2 14 46 80 142 0.81
Total 11 14,233 3,086 307 17,637
% 0.06 80.70 17.50 1.74
33
III. Land used for personal farming
2022
2021 Built up Crop Grass Garden/parks Total %
Panel A: National
Built up 2,121 75 70 496 2,762 6.57
Crop 234 16,543 2,249 684 19,710 46.90
Grass 1,165 2,885 8,190 906 13,146 31.28
Garden/parks 794 489 1,095 4,029 6,407 15.25
Total 4,314 19,992 11,604 6,115 42,025
% 10.27 47.57 27.61 14.55
Panel B: Areas not directly affected by conflict
Built up 1,738 74 60 383 2,255 7.22
Crop 198 11,691 1,347 514 13,750 44.04
Grass 992 2,154 6,301 672 10,119 32.41
Garden/parks 645 422 977 3,051 5,095 16.32
Total 3,573 14,341 8,685 4,620 31,219
% 11.44 45.94 27.82 14.80
Panel C: Conflict-affected areas only
Built up 383 1 10 113 507 4.69
Crop 36 4,852 902 170 5,960 55.15
Grass 173 731 1,889 234 3,027 28.01
Garden/parks 149 67 118 978 1,312 12.14
Total 741 5,651 2,919 1,495 10,806
% 6.86 52.30 27.01 13.83
Source: Own computation from 2021 and 2022 crop maps (available on https://ukraine-cropmaps.com) using field coordinates for
parcels transacted in the agricultural land sales market between July and December 2021.
Note: Grass includes bare and uncultivated land.
34
Appendix table 5: Characteristics of parcels transferred in the July 2021 – Dec. 2022 period
Price reported Price not reported
Total Personal Commercial Total Personal Commercial
Panel A: Sales
Price (US$/ha) 6,157.7 10,774.7 1,178.5
Parcel area (ha) 2.285 1.096 3.568 2.198 1.051 3.415
Normative value (US$/ha) 1,205.4 1,650.9 978.9 1,092.7 1,465.1 909.7
Commercial agric. 0.481 0.000 1.000 0.485 0.000 1.000
Used for crops at transfer 0.669 0.432 0.925 0.684 0.450 0.921
Built up at transfer 0.030 0.058 0.000 0.031 0.061 0.000
Forested at transfer 0.123 0.221 0.018 0.111 0.200 0.020
Grassland at of transfer 0.177 0.289 0.056 0.174 0.289 0.059
Land quality index 40.490 37.933 43.247 39.883 38.346 41.440
Dist. to next city (km) 13.117 11.338 15.035 13.766 11.760 15.799
Dist. to prim. road (km) 7.353 6.105 8.700 8.148 6.833 9.481
Dist. to parks (km) 29.167 27.695 30.754 29.253 28.058 30.463
% unreg. cropland in village 0.204 0.218 0.190 0.205 0.213 0.196
Slope (%) 2.592 2.939 2.219 2.513 2.878 2.144
Altitude (m) 172.118 177.124 166.720 166.234 174.648 157.706
Transacted during war 0.113 0.131 0.094 0.377 0.408 0.344
Number of parcels 71,779 37,244 34,535 79,997 41,178 38,819
Panel B: Leases
Price (US$/ha) 91.115 94.655 89.193
Parcel area (ha) 2.478 1.665 2.919 2.262 1.491 2.635
Normative value (US$/ha) 680.7 623.5 717.6 665.1 501.9 728.0
Commercial agric. 0.648 0.000 1.000 0.673 0.000 1.000
Used for crops at transfer 0.772 0.704 0.809 0.760 0.667 0.806
Built up at transfer 0.000 0.000 0.000 0.000 0.000 0.000
Forested at transfer 0.053 0.051 0.054 0.056 0.069 0.050
Grassland at of transfer 0.175 0.246 0.137 0.184 0.265 0.144
Land quality index 36.579 36.740 36.492 35.913 33.573 37.077
Dist. to next city (km) 14.631 14.537 14.682 17.088 16.008 17.625
Dist. to prim. road (km) 8.538 8.143 8.753 9.848 9.109 10.215
Dist. to parks (km) 32.051 28.118 34.185 30.976 27.382 32.762
% unreg. cropland in village 0.225 0.213 0.232 0.234 0.219 0.241
Slope (%) 2.241 2.409 2.150 2.260 2.418 2.181
Altitude (m) 173.310 175.759 171.981 171.789 176.108 169.641
Transacted during war 0.018 0.021 0.016 0.262 0.248 0.269
Number of parcels 10,854 3,819 7,035 15,994 5,227 10,767
Source: Own computation using Registry of Rights of Ukraine (Jan. 2023); State Statistics of Ukraine (Form 50); Crop cover
estimates based on remotely sensed data; Open Street map. Note: Col. 1-3 are for observation used in the regression analysis.
35
Appendix figure 1: No. of daily agricultural land sales per week and oblast, before and after the start of the war
Panel A: Agricultural for personal farming
Panel B: Agricultural for commercial farming
36
Appendix figure 2: Land sales price of agricultural land per ha
Note: Excluding the first five weeks after the opening of the agricultural land sales market due to the very erratic nature of the price
information.
37
References:
Ahlfeldt, G., Koutroumpis, P., Valletti, T., 2017. Speed 2.0: Evaluating Access to Universal Digital Highways. Journal of the
European Economic Association 15, 586-625
Arias, M.A., Ibáñez, A.M., Zambrano, A., 2019. Agricultural production amid conflict: Separating the effects of conflict into shocks
and uncertainty. World Development 119, 165-184
Ben-Shahar, D., Golan, R., 2019. Improved information shock and price dispersion: A natural experiment in the housing market.
Journal of Urban Economics 112, 70-84
Berman, N., Couttenier, M., Soubeyran, R., 2021. Fertile Ground for Conflict. Journal of the European Economic Association 19,
82-127
Besley, T., Mueller, H., 2012. Estimating the Peace Dividend: The Impact of Violence on House Prices in Northern Ireland.
American Economic Review 102, 810-833
Bigelow, D.P., Kuethe, T., 2023. The Impact of Preferential Farmland Taxation on Local Public Finances. Regional Science and
Urban Economics 98
Brorsen, B.W., Doye, D., Neal, K.B., 2015. Agricultural Land and the Small Parcel Size Premium Puzzle. Land Economics 91,
572-585
Casas-Arce, P., Saiz, A., 2010. Owning versus Renting: Do Courts Matter? Journal of Law and Economics 53, 137-165
Chang, H.-H., Lin, T.-C., 2016. Does the Minimum Lot Size Program Affect Farmland Values? Empirical Evidence Using
Administrative Data and Regression Discontinuity Design in Taiwan. American Journal of Agricultural Economics 98,
785-801
Chang, Z., Li, X., 2021. How Regulation on Environmental Information Disclosure Affects Brownfield Prices in China: A
Difference-in-Differences (DID) Analysis. Journal of Environmental Planning and Management 64, 308-333
Coulomb, R., Zylberberg, Y., 2021. Environmental Risk and the Anchoring Role of Mobility Rigidities. Journal of the Association
of Environmental and Resource Economists 8, 509-542
Deaton, B.J., Lawley, C., 2022. A survey of literature examining farmland prices: A Canadian focus. Canadian Journal of
Agricultural Economics 70, 95-121
Deininger, K., Ali, D., Nivyievskyi, O., 2023a. Changes in Ukrainian Farm Profitability 2021 to 2022: Farm Survey Evidence In:
Policy Research Working Paper. World Bank, Washington DC
Deininger, K., Ali, D.A., Kussul, N., Shelestov, A., Lemoine, G., Yailimova, H., 2023b. Quantifying war-induced crop losses in
Ukraine in near real time to strengthen local and global food security. Food Policy 115, 102418
Deininger, K., Ali, D.A., Neyter, R., 2022. Impacts of Transparent Online Auctions on Public Land Lease Revenue: Evidence from
Legal and Administrative Changes in Ukraine. In: Policy Research Working Paper 10201. World Bank, Washington DC
Diewert, W.E., Shimizu, C., 2022. Residential Property Price Indexes: Spatial Coordinates versus Neighborhood Dummy
Variables. Review of Income and Wealth 68, 770-796
Diewert, W.E., Shimizu, C., Watanabe, T., Nishimura, K.G., 2020. Property Price Index: Theory and Practice. Springer, Tokyo.
Frank, E.G., Sudarshan, A., 2022. The Social Costs of Keystone Species Collapse : Evidence From The Decline of Vultures in
India. University of Warwick, Department of Economics, The Warwick Economics Research Paper Series (TWERPS)
Gertler, P.J., Gonzalez-Navarro, M., Gracner, T., Rothenberg, A.D., 2022. Road Maintenance and Local Economic Development:
Evidence from Indonesia’s Highways. University of California, Berkeley
Glauben, T., Svanidze, M., Gotz, L., Prehn, S., Jamali Jaghdani, T., Duric, I., 2022. The War in Ukraine, Agricultural Trade and
Risks to Global Food Security. Review of European Economic Policy 57, 157-163
Goldsmith-Pinkham, P., Gustafson, M.T., Lewis, R.C., Schwert, M., 2021. Sea Level Rise Exposure and Municipal Bond Yields
Yale University, New Haven, CT
Gouriéroux, C., Laferrère, A., 2009. Managing hedonic housing price indexes: The French experience. Journal of Housing
Economics 18, 206-213
Graubner, M., Ostapchuk, I., Gagalyuk, T., 2021. Agroholdings and Land Rental Markets: A Spatial Competition Perspective.
European Review of Agricultural Economics 48, 158-206
Han, L., Heblich, S., Timmins, C., Zylberberg, Y., 2021. Cool Cities: The Value of Urban Trees ∗.
Hilber, C.A.L., Vermeulen, W., 2016. The Impact of Supply Constraints on House Prices in England. Economic Journal 126, 358-
405
Hill, R.J., 2013. Hedonic Price Indexes for Residential Housing: A Survey, Evaluation and Taxonomy. Journal of Economic
Surveys 27, 879-914
Hüttel, S., Jetzinger, S., Odening, M., 2014. Forced Sales and Farmland Prices. Land Economics 90, 395-410
Ihle, R., Bar-Nahum, Z., Nivievskyi, O., Rubin, O.D., 2022. Russia's Invasion of Ukraine Increased the Synchronisation of Global
Commodity Prices. Australian Journal of Agricultural and Resource Economics 66, 775-796
Jaremski, M., Wheelock, D.C., 2020. Banking on the Boom, Tripped by the Bust: Banks and the World War I Agricultural Price
Shock. Journal of Money, Credit & Banking 52, 1719-1754
Kirwan, B.E., 2009. The Incidence of U.S. Agricultural Subsidies on Farmland Rental Rates. Journal of Political Economy 117,
138-164
Kirwan, B.E., Roberts, M.J., 2016. Who Really Benefits from Agricultural Subsidies? Evidence from Field-Level Data. American
Journal of Agricultural Economics 98, 1095-1113
38
Kussul, N., Deininger, K., Shumilo, L., Lavreniuk, M., Ali, D.A., Nivievskyi, O., 2022. Biophysical Impact of Sunflower Crop
Rotation on Agricultural Fields. Sustainability 14, 3965
Kvartiuk, V., Herzfeld, T., 2019. Welfare effects of land market liberalization scenarios in Ukraine: Evidence-based economic
perspective. In: Discusion paper 2019 # 186. Leibniz Institute of Agricultural Development in Transition Economies
(IAMO), Halle/Saale
Kvartiuk, V., Herzfeld, T., Bukin, E., 2022. Decentralized public farmland conveyance: Rental rights auctioning in Ukraine. Land
Use Policy 114, 105983
Kvartiuk, V., Martyn, A., 2021. Ukraine’s Agricultural Land Sales Market: First Outcomes and Monitoring Challenges. In: Agro
Policy Report APD/APB/11/2021. German-Ukrainian Agicultural Policy Dialogue, Kyiv
Kvartiuk, V., Martyn, A., 2022. Ukraine’s agricultural land sales market: An update and the effect of Russian war against Ukr aine
In: Agro Policy Report APD/APB/20/2022. German-Ukrainian Agicultural Policy Dialogue, Kyiv
Lawley, C., 2018. Ownership Restrictions and Farmland Values: Evidence from the 2003 Saskatchewan Farm Security Act
Amendment. American Journal of Agricultural Economics 100, 311-337
Lin, F., Li, X., Jia, N., Feng, F., Huang, H., Huang, J., Fan, S., Ciais, P., Song, X.-P., 2023. The impact of Russia-Ukraine conflict
on global food security. Global Food Security 36, 100661
Mueller, H., Groeger, A., Hersh, J., Matranga, A., Serrat, J., 2021. Monitoring war destruction from space using machine learning.
Proceedings of the National Academy of Sciences 118, e2025400118
Neyter, R., Nivievskyi, O., 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
Nivievskyi, O., 2020. Where state agricultural land disappears to. URL https://www.epravda.com.ua/columns/2020/06/4/661385/
Nizalov, D., Thornsbury, S., Loveridge, S., Woods, M., Zadorozhna, O., 2016. Security of property rights and transition in land
use. Journal of Comparative Economics 44, 76-91
Pinchbeck, E.W., Roth, S., Szumilo, N., Vanino, E., 2020. The Price of Indoor Air Pollution: Evidence from Radon Maps and the
Housing Market. In: IZA DP No. 13655, Bonn
Rajan, R., Ramcharan, R., 2015. The Anatomy of a Credit Crisis: The Boom and Bust in Farm Land Prices in the United States in
the 1920s. American Economic Review 105, 1439-1477
Rajan, R., Ramcharan, R., 2016. Local Financial Capacity and Asset Values: Evidence from Bank Failures. Journal of Financial
Economics 120, 229-251
Seifert, S., Kahle, C., Hüttel, S., 2021. Price Dispersion in Farmland Markets: What Is the Role of Asymmetric Information?
American Journal of Agricultural Economics 103, 1545-1568
Severen, C., Plantinga, A.J., 2018. Land-use regulations, property values, and rents: Decomposing the effects of the California
Coastal Act. Journal of Urban Economics 107, 65-78
Singh, R., 2019. Seismic risk and house prices: Evidence from earthquake fault zoning. Regional Science and Urban Economics
75, 187-209
Tang, C.K., 2021. The Cost of Traffic: Evidence from the London Congestion Charge. Journal of Urban Economics 121, 103302
Tellez, J.F., 2022. Land, Opportunism, and Displacement in Civil Wars: Evidence from Colombia. American Political Science
Review 116, 403-418
Voors, M.J., Nillesen, E.E.M., Verwimp, P., Bulte, E.H., Lensink, R., Van Soest, D.P., 2012. Violent Conflict and Behavior: A
Field Experiment in Burundi. American Economic Review 102, 941-964
Yaremko, V., 2022. The Long-Term Consequences of Blacklisting: Evidence from the Ukrainian Famine of 1932-33. University
of California, Berkeely, CA
Zadorozhna, O., 2020. Clientelism and Land Market Outcomes in Ukraine. Eastern European Economics 58, 478-496
39