We have analyzed the efficiency of land auctions and found their potential weaknesses.
How can the work of land auctions be improved?
The first thing is to prevent auction participants from going into a conspiracy and identify related persons.
Secondly, the starting price for repeat bidding process should be lowered if a land parcel was not sold on the first try.
A more detailed analysis of the efficiency and potential weaknesses of land auctions is given below.
Land reform is not limited to the draft law on land turnover, but it includes a package of at least eight bills which are under parliamentary consideration and a set of measures to increase the transparency of land governance. These measures include the development of the Land Governance Monitoring System and the Agrarian Registry and the creation of a credit guarantee fund.
A separate issue of the reform is the efficiency of using state and communal lands, which are a source of mass corruption and significant budget losses. The solution to this problem can be designed as follows:
Transfer of approximately 5.5 million hectares of state-owned land beyond settlements (out of the 7.2 million ha) into communal ownership of communities as part of the decentralization reform.
Establishing land distribution during privatization of agricultural land of state and communal companies. When privatizing this type of agricultural land, it is proposed that:
First. Up to 40% of the total area of agricultural land be transferred into private ownership of the company employees free of charge.
Second. Up to 20% of the total area of agricultural land be transferred into the ownership of the participants of the Anti-Terrorist Operation and the United Forces Operation free of charge
Third. No less than 40% of the total area of agricultural land be sold or leased via transparent land auctions.
State and communal land will be sold or leased exclusively via transparent land auctions.
Taking into account yesterday’s amendments to draft law No.2178-10, it is highly probably to talk only about leasing state and communal land, since the sale of state and communal agricultural land is indefinitely prohibited.
Since state and communal land will be sold or leased exclusively via e-auctions, let us consider the special features of land auctions, their effectiveness and potential weaknesses.
Since 2016, the sale of lease rights to state-owned agricultural land has been done exclusively through land auctions. This has led to significant increases in the rental costs of state-owned land.
In 2016, the price of renting land through land auctions was UAH 2,249.80 / ha, whereas the price of renting state land on a non-competitive basis, according to earlier rental agreements, was only UAH 940 / ha.
For comparison, an average price of renting private land shares in 2016 was UAH 1093.40 / ha against UAH 862 / ha in 2015. Moreover, the gap between the prices of renting private and state land widened in 2018 (UAH 3,431.5 / ha is the average price of renting state-owned agricultural land via auctions).
Despite the successful practices of auctions, there is still room for operational improvement of this instrument in terms of increasing revenues from the use of state and community land. Let us focus on the relationship among the number of participants, starting and final prices of a land auction.
One can analyze the final prices of the auctions that actually took place, i.e. auctions with at least two registered participants. However, if an auction never took place due to the lack of participants, the land parcel has to be put up for auction not earlier than one month and no later than six months after the unsuccessful bidding.
What has changed in the rules for repeated auctions?
If the auction had only one participant, he or she may obtain that land parcel at the starting price. At the same time, the starting price is not lowered. So, if the real market price is lower than the starting price, it is logical that the auction does not take place, and another attempt will be made to sell the parcel again.
Under such conditions, parcels can be put up for sale for years. The “record-breaking” parcels have accumulated more than ten unsuccessful auctions. Moreover, the statistics shows that in some cases the price for repeat auctions may even grow.
The solution is a possibility to lower the starting price for the next auction. Currently, the starting annual rent for state and communal land is 8-12% of the normative monetary valuation (NMV), although the law defines that the rental price of state and communal land is limited exclusively by the amount of land tax, which is actually 1% of the NMV.
A lower starting price can attract more bidders, which will increase competition at the land auction. As for the dependence of the final price on the number of participants, it will be demonstrated below.
The graph shows the ratio between the starting and final prices at the auction. An interesting picture can be seen: most of the observations are located along a single line. It means that there is almost no difference between the final value of lease rights and the starting price to these land parcels.
The most probable reason for the coincidence is the lack of competition. In December 2019, only two bidders took part in the majority of successful auctions (1,399 out of 2,668).
Different colors of the graph dots help to differentiate auction results by number of participants. While most auctions with two bidders are located along the same line, observations for auctions with more participating bidders are scattered at the top of the graph, which illustrates a significant increase in prices.
For auctions with two participating bidders, the median price increase is only 2%, and for auctions with more than three bidders – 214%, i.e. more than three times higher. While three is the median number of bids made at auctions with two participants, the median number of bids at auctions with more than two participants is 79, i.e. 26 times bigger.
The most likely explanation for such a striking difference is the participation of related parties in auctions with only two bidders. For an auction to take place, it can be applied by two related parties who are not de facto competing with each other.
The problem is that, unlike other online auctions held on ProZorro platform, bidders are not disclosed after land auctions which are held on the SETAM platform. Therefore, it is not possible to trace what persons are constantly bidding together and to check whether they are related or not.
The decision to conceal the information on bidders cannot be justified with personal data protection, because in the open register of property rights, using the cadastral parcel number, everyone can see the USREOU code and the name of the auction winner who signs a lease agreement.
Thus, everyone who is going to auction with the purpose of buying a lease right is already prepared for information disclosure. It should be added that the web-pages of the auctions held by the StateGeoCadastre have no information on the number of auction bids and participants in a machine format, which makes a substantial analysis of such auctions impossible.
In order to increase the efficiency of e-auctions, it is necessary to prevent conspiracy of the participants and spot related bidders by publishing data on the participants after the auction is over and to reduce the starting price for re-auctioning if a land parcel was not sold at the first attempt.
Co-authored by Oleg Nivievskyi, Kyiv School of Economics