- July 20, 2024
Only 5.7% of the Treasury’s payments made in 2023 have reliable data on procurement in the Prozorro system. This creates significant difficulties in monitoring the efficiency of the use of funds and identifying potential corruption. This was stated by Viktor Maziarchuk, head of the Center for Fiscal Policy Research, during his speech at a meeting of the parliamentary provisional investigatory commission to investigate violations in public procurement during the war.
He noted that within the framework of the Cost of War project, the Center’s team presented an interactive system for monitoring the expenditures of the Fund for Elimination of the Consequences of Armed Aggression. This tool allows public authorities and the public to track every hryvnia spent from the Fund and see who paid the money and who received it. The team is currently working on visualizing local budget expenditures of all territorial communities in Kyiv region.
However, according to him, there are a number of systemic problems that impede effective monitoring of procurement and ensuring transparency of the use of funds. “The first and key problem for our team is the quality of data generated by government institutions and local governments in terms of procurement and payment for these procurements,” he emphasized.
In particular, out of 3.7 million tenders in 2023, the Treasury’s payment orders contain information on only 212 thousand, or 5.7%. “Only a very limited part of payments can be found by the procurement identification number by simply comparing it in two state databases – Prozorro and the Treasury,” explained Maziarchuk.
He also drew attention to problems with the quality of data in payment orders. Thus, out of 244 thousand payment orders in 2023 made for major repairs and reconstruction, less than a third contain information on procurement (74 thousand).
“You will laugh, but our Treasury allows us to pay for goods for procurements that have not yet been made but will take place in the future. But I’m not laughing, because this all seriously affects the quality of the data we work with,” he added. ”The scale of this problem is not very large compared to the budget as a whole, but in 2023 there were such purchases for UAH 19 million, made by 91 institutions at the expense of state and local budgets.
The expert also emphasized the need to synchronize the Treasury and Prozorro databases, which has been discussed for years. “In an ideal world, documents in the Prozorro system should be presented in the same way as laws on the website of the Verkhovna Rada. This is the standard that everyone should follow,” summarized Maziarchuk.
Full text of the speech:
Dear Mr. Chairman, Members of the Temporary Investigation Commission of the Verkhovna Rada of Ukraine.
Thank you for the opportunity to share our proposals and ideas on public procurement and efficiency of use of taxpayers’ funds.
Our team’s area of interest is public finance and big data. As part of the Cost of War project, we collect and verify information on actual expenditures made from the state budget, local budgets, and international partners’ funds to overcome the consequences of the war.
Recently, we presented an interactive system for monitoring the expenditures of the Fund for Elimination of the Consequences of Armed Aggression, which shows all the expenditures of the Fund in 2023 (except for those with restricted access). This tool will allow you to find every hryvnia spent from the Fund and show who paid the money and who received it. You will be able to see all the objects for which the money was spent. We are currently working on visualizing the expenditures of local budgets of all territorial communities of Kyiv region.
Mr. Chairman, dear MPs, members of the PIC, those present, you may ask what is the purpose of the tool for monitoring recovery expenditures that we have developed?
The answer is very concise and simple. Our task is to link three basic steps:
Government (local government) decision to allocate funding;
procurement;
payment for works and services performed.
And this is exactly what we have been doing in recent months. Given the large amount of open information generated by the Government, and despite the incredible transparency in certain issues of spending, there are still a number of systemic problems that I would ask you to consider in your report.
These problems are a deterrent that does not allow for the best possible monitoring of procurement and ensuring the efficient use of funds. These limitations of the system make it possible to hide inefficiency and corruption.
That is why I will speak today in the broader context of corruption-related offenses in public procurement, where procurement is one of the elements.
The first and key problem for the team of the Center for Fiscal Policy Research is the quality of data generated by government institutions and local governments in terms of procurement and payment for these procurements.
Mr. Chairman, as a person who knows how and loves to analyze large amounts of data, you understand the principle of trash in – trash out, or let me rephrase: the inability to identify potential corruption and prevent corruption.
Here are some figures for the year 2023.
The number of successfully conducted tender procurements is almost 3.7 million tenders. We know the exact number because we have the full Prozorro database, which we work with when making analytics or individual interactive products. Out of 3.7 million tenders in 2023, the Treasury’s payment orders contain information on only 212 thousand, or 5.7%. That is, only a very limited part of payments can be found by the procurement identification number, simply by comparing it in two state databases – Prozorro and the Treasury.
We all understand that payments for capital construction or for the purchase of goods and services should be made by the Treasury based on information about the tender held, this is one of the mandatory fields.
Of the 244 thousand payment orders in 2023 made for capital repairs, reconstruction, etc. (capital expenditures), less than a third contain procurement information (74 thousand).
Dear colleagues, you will laugh, but our Treasury allows us to pay for goods for purchases that have not yet been made but will take place in the future. But I’m not laughing because this seriously affects the quality of the data we work with. The scale of this problem is not very large compared to the budget as a whole, but in 2023, there were UAH 19 million worth of such purchases made by 91 institutions at the expense of state and local budgets.
For example, the Treasury makes such payments to the Semen Kuznets Kharkiv National University of Economics for almost UAH 400 thousand, which is state budget funds. I give you this example to show you several aspects of the data problem:
- First, the Treasury does not ensure verification of the correctness of the tenders under which payments are made, as it pays for electricity to the University under a tender held by the Bohdan Khmelnytsky National Academy of the State Border Guard Service of Ukraine for the purchase of chairs. Bohdan Khmelnytskyi National Academy of the State Border Guard Service of Ukraine for the purchase of chairs. Other customer and subject of procurement (2023-12-19-021629-a).
- The second – the Treasury pays the University for a non-existent tender, namely the procurement of services for the installation of a video surveillance system, which is to be held in October 2323 (2323-10-23-009341-a)
Sometimes simple math allows us to look for violations of budget legislation, but it does not allow the Treasury to avoid making payment orders that contain errors.
Dear members of the TSC, you are professionally engaged in procurement, you know that the tender ID has its own unique number, which usually consists of 22 characters. So, according to 115 thousand payment orders of the Treasury (6% of the total), the length of the tender ID character varies from 2 to 49 characters. The most common and incorrect practice of writing this identifier is simply two English letters UA.
We can only be glad that government employees know not only Ukrainian but also English. Because in the procurement identifiers of Prozorro and the Treasury, you can find the letter “a” in both Ukrainian and English. This also distorts the quality of the data. There are characters in two languages in 317 thousand payment orders out of almost 2 million in the procurement ID, which is 16% of all data.
Dear Mr. Chairman, members of the PIC, those present, why don’t I say that estimates, construction plans and other procurement items should be in a machine-readable format? Because this issue is overripe, and I am grateful to the MPs who have taken it up.
I would also like to draw your attention to the fact that documents in the Prozorro system should be presented in the same way as laws on the website of the Verkhovna Rada of Ukraine. This is the standard that everyone should follow.
In addition, if we want to prevent corruption, we must clearly understand the financial source of procurement and payment for it. Prozorro should implement the functionality of synchronization with the budget and budget programs. In the context of recovery monitoring, which is now becoming extremely relevant, there should be clearly defined fields with the address of the construction, which should be formed from the names of the region, territorial community, settlement and street.
In conclusion, I would like to draw your attention, dear members of the PIC, to the work of such bodies as the Accounting Chamber and the SASU. It is very important that these institutions publish reports on their work and justifications for the audit plan in a machine-readable format.
It is important to resolve the issue of publishing SASU reports on audits in general. It is also important to draw the attention of the Accounting Chamber to the fact that the Classifier of Budget Violations has been developed but not yet approved since 2021. This document could be the basis for identifying violations and reducing discretion in the decisions of public authorities.
Dear Mr. Chairman, members of the Verkhovna Rada provisional investigative commission, I would like to thank you for the opportunity to present our findings. I hope that our experience in analytics and working with big data will be useful to you.
I will be happy to answer any questions you may have.
Follow us
Subscribe
Subscribe to our newsletter and receive a selection of interesting articles and research every week