This brief is based on a research that investigates if there’s a synergy effect of procurement and decentralization reforms in Ukraine on procurement efficiency. The analysis shows the similarity between new and old administrative units procurement performance. Although the analysis does not provide evidence of a significant synergy effect, such a similarity could be considered as something positive (due to the lower market power and capacity of the newly created administrative units) that should be analyzed further.
Decentralization in Ukraine
In April 2014, the Ukrainian government launched a systemic decentralization reform – a delegation of a significant part of resources and responsibilities from oblast and raion-level executive branches of the government to the local self-government level. The key issue of the reform was the creation of a strong basic level of local self-government in line with the European Charter of Local Self-Government. This is done through the creation of amalgamated hromadas (AH), the merger of several settlements with a single administrative centre.
An AH is governed by a council of the amalgamated hromada (CAH), a representative body of a local self-government. It is elected by residents of territorial communities and is responsible to independently resolve local issues, develop and approve AH budgets. Particularly, the government redistributed tax revenues and expanded the system of state subsidies (medical and educational subvention, subvention on development of amalgamated hromadas etc.) that could be used according to AH decisions.
In 2015, 159 AH were established. As of September 2018, 831 AH made up by 3,796 hromadas with 7 million residents (decentralization.gov.ua).
Contemporaneously, a reform of public procurement has been implemented. According to the Law on Public Procurement of Ukraine, since August 2016, public procurement must be announced and executed through ProZorro, a public procurement web portal administered by the state enterprise. The e-procurement system consists of a central database, an auction module and commercial marketplaces (Figure 1).
In order to participate in public tenders, bidders can choose one of 22 commercial marketplaces (7 companies that provided initial investment for the project became the first marketplaces). The commercial marketplaces are web resources managed by private companies that provide access to the electronic procurement system.
Figure 1. ProZorro Architecture
The electronic procurement system does not completely cover the procurement cycle. Actually, it only covers the tendering process while planning and contract execution are mostly out of the system (plans are published online). Moreover, the existing legislation provides opportunities to manipulate tendering process by switching between different procedures. Within the ProZorro system there are 6 main procurement procedures that can be used by procuring entities depending on the volume and specifications of their needs.
Selection of procedures is based on a Threshold principle. There are three thresholds (Figure 2):
- Lower Threshold (LT). Contracting authorities are not obliged to report procurements in the electronic system if the total value of procurement is lower than UAH 50,000.
- Higher Threshold (HT). Contracting authorities are not obliged to use open competitive procedures if the total value of tender is lower than a defined level. This level is equal to UAH 200,000 for goods and services and UAH 1.5 million for construction.
- Euro Threshold (ET). The value of tender that requires applying the strictest competitive auction procedure. The euro threshold corresponds to thresholds used in EU public procurement legislation and is different for goods and services and construction works.
Transparency and efficiency indicators
Typically, a procurement process is divided into three stages: pre-tender stage, tender stage and post-tender stage. To measure efficiency and transparency of AH procurement, we constructed a system of eleven indicators that evaluate each stage of the procurement process (Table 1).
Table 1. Transparency and efficiency indicators used in the report
- Avoiding ProZorro. Both this and the following indicator would be associated with a decreased transparency, and, while not necessarily evidential, raise suspicions about procurement done in a less efficient and more collusive/corrupt way. AH are not inclined to avoid the ProZorro system. The share of procurements outside of ProZorro per AH is smaller than the corresponding indicators for other administrative units (Raion State Administrations, RSA, and unamalgamated hromadas, UH).
- Avoiding higher and euro thresholds. An analysis of ProZorro data shows that for AH, 84% and 11% of AH have at least one “suspicious” case for each corresponding threshold. For, RSA the indicators are 73% and 31% respectively.
- Unanswered questions. Having a productive dialogue with suppliers is of crucial importance for the success of the procurement. It helps to adjust the tender documentation so that it does not include discriminatory demands. Analysis shows that this practice is not dominant: relatively small proportion of AH and UH uses it.
- Level of competition. Higher competition is normally assumed to imply more efficient procurement deals. There is no difference in competition across administrative units and products (measured by the number of bidders per tender).
- High disqualification rates can be a consequence of ill prepared tender documentation with unclear technical specification or it can be a consequence of suppliers’ inexperience. It can also be a sign of corruption, when a tender committee is trying to find any reason to disqualify ‘unwanted’ suppliers. The analysis shows that disqualification is not a significant problem and, in fact, there are no significant differences across administrative units and products.
- Success rates. To successfully complete competitive procurement, the contracting authority has to determine the technical description of a good and its expected value based on their budget and market analysis. It also has to prepare and publish tender documentation and answer questions of potential suppliers. Finally, after the auction, the contracting authority has to evaluate documents of the auction winner and sign a contract. Failure in each of these steps will lead to unsuccessful or cancelled procurement. There is no significant difference across administrative unit groups in terms of procurement success rate
- Abnormal saving rate. Generally, a high saving rate (the difference between tender expected and contracted values) is regarded as a positive indicator, however, a too high rate is suspicious. It can be a sign of an inadequate expected value or an abnormally low price (suspicious behavior on behalf of the supplier). For the purpose of this study, we consider a saving rate abnormal if it is greater or equal to 30%. The analysis shows that AH had a significantly lower share of tenders with abnormally high saving rate than RSA. On average, 0.6% (in terms of value) of AH tenders are suspicious, for RSA this indicator equals 6.1%
- Contract termination. Frequent contract termination is a sign of significant inefficiencies in the procurement function of contracting authorities. The share of terminated contracts (as a percentage of the total contract value) is approximately similar for AH and RSA. On average, one AH has 5% of contract value terminated, while RSA indicator equals to 6%.
- Fixing the price with additional agreements. Although the Ukrainian Law on Public Procurement gives the right to amend the price per unit indicated, this right can be misused. It could lead to significantly higher costs. RSA are strikingly different from the other two groups – on average 20% of the RSA contract value stems from contracts with amended prices. This difference is the consequence of the different structure of goods and services procured by AH and UH on the one hand and RSA on the other.
- Share of largest supplier. Generally, it is considered to be a good practice, when contracting authorities are not overly dependent on one supplier. Approximately 30% of contract value of average AH and RSA belongs to one supplier. For UH this indicator is even higher (on average 48%) but it could be the consequence of the smaller number of contracts signed by UH.
Effect on prices
If contracts are successfully executed, the price of a good usually summarizes the efficiency of the procurement process.
There are many factors that affect the prices of goods in public procurements. On the one hand, AH (a) “realized” that they spent their own money and thus, they have more incentives to save and (b) have more power to choose where to spend. On the other hand, there are some factors that have the opposite effect: (a) because of low quantity demanded, the tenders announced are not interesting for large companies that could potentially provide lower price, and (b) the procurement officers could have insufficient capacity to negotiate lower price. Although, it is impossible to evaluate all these factors, we can assess their outcome – the contract price of a good.
For this analysis we looked at the prices on homogeneous goods such as food (potato, butter, eggs) and fuel (petrol A95, petrol A-92, diesel.
Table 2 summarizes the prices on the goods received by hromadas and compares it to the prices received by other types of entities (UHs and RSAs).
The data shows that for food products, AH average prices are lower or not different from UH, and slightly higher or not different from prices received by RSAs.
Even though there are some differences in prices of Petrol A-95 (partially due to inefficient planning and contracting at periods of higher prices), in general, the price level is very similar between all the entities.
In most cases, despite some warnings, there were no significant gaps between AH’s prices and UH or RSA. Moreover, the more competitive is the market of goods procured, the closer are prices received by different administrative units.
Table 2. Prices of goods by administrative units in 2017
The analysis shows the similarity between AH and RSA in terms of number of procurements, success and disqualification rates as well as competition level and share of terminated contracts. However, in cases when it is allowed by the Procurement Law, AH are more likely than RSA to choose direct selection of a supplier than a competitive procedure. Such behavior can be caused by a lack of professionalism (or even corruption), a desire to select a local trustworthy company or just because it is easier and faster to conduct uncompetitive procedure below the threshold.
On the other hand, AH are less inclined (in comparison to RSA) to avoid the ProZorro system (by using procurements below UAH 50 K) and to sign additional agreements that increase the price. Such behavior is potentially punishable by law. It can be suggested that procurement officers of AH only recently started to work with tenders above HT and are therefore more conscious of possible negative consequences of such actions.
The price per unit is the key indicator that summarizes information on procurement efficiency. Although AH show varying price efficiency, their prices of procured goods, in general, are not worse than in other administrative units’ groups. The more competitive the market, the closer are prices (especially in the case of fuels). Even if some gaps were observed, these differences are decreasing over time. Better planning can help to receive lower prices (better estimation of needs and choosing appropriate periods for procurement).
Currently, the analysis does not provide an evidence on a significant synergy effect of decentralization and procurement reforms. There are no significant differences between old and new administrative units. However, usually new communities have lower market power and capacity, and “no difference” could be considered as a positive sign that should be analyzed further.
- Shapoval, Natalia; Iavorskyi, Pavlo; Stepaniuk, Oleksa; Kovalchuk, Arthur. 2018. “An evaluation of decentralization impact on transparency and efficiency of public expenditures”. SKL International AB.
Disclaimer: Opinions expressed in policy briefs and other publications are those of the authors; they do not necessarily reflect those of the FREE Network and its research institutes.