Volunteers met with their community to provide them feedback about the results of health facilities monitoring in Jalalabad in April 2020. Photo credit: Integrity Watch
Afghanistan is juggling multiple threats, including insecurity and political uncertainty after the 2019 elections and corruption amid the COVID-19 pandemic. As an aid-dependent country, Afghanistan has to raise funds to meet not only the growing demand for health services and equipment, it also has to feed a growing number of people who were already living under the poverty line.
However, both donor and public trust are very low. The prevalence of corruption makes it very difficult for the Afghan Government to raise the funds needed and to spend the resources in a satisfactory manner. There are already allegations of waste, mismanagement and corruption in handling the COVID-19 situation, including personal protection equipment and materials purchased at exorbitant rates, the supply of substandard hand sanitizers and other goods, and a lack of access to personal protection equipment in many of the hospitals and health centers we have surveyed.
Nevertheless, it is not a completely hopeless situation. The government can take certain measures to address the trust deficit among donors and the public as they raise, allocate, and spend the funds needed to tackle COVID-19. Open budgeting and open contracting (including transparency and public participation) could greatly assist the government to overcome the situation.
According to the results of the most recent Open Budget Survey, Afghanistan only provides limited budget information to its citizens. The country‘s overall budget transparency score is 50 out of 100, with the public participation score being 15 out of 100 and the oversight score 31 out of 100. While the scores did not improve in the last two years, there are many low-hanging fruits the Afghan Government can pick to increase budget transparency. This includes publishing budget documents in a timely manner, such as the pre-budget statement which the government produced last year but did not publish on time. In addition, providing more details in budget documents and reports and allowing civil society and the public to participate in the budget process could turn the budgeting process into a trust-building vehicle.
These reforms can go hand in hand with fighting the COVID-19 outbreak. Since the government is planning to amend this year’s budget in response, it should publish the proposed amendments before they are finalized to enable citizens and civil society groups to engage in a meaningful discussion with the government. The government should gather inputs from civil society and other stakeholders for these amendments, especially regarding the health sector because improving the health system is a top priority at the moment.
The information generated by independent sources, such as civil society, should be used to prioritize spending in the health sector. For instance, data from Integrity Watch’s study to assess the quality of health services in more than 50 hospitals, a survey of 41 health centers, and data collected through our community-based monitoring of 170 health centers could be used in the allocation of funds based on the needs identified through these studies. In addition, data generated through programs such as the Citizens Charter by the Afghan Government partners and the Community-Based Monitoring of health centers by Integrity Watch, which ensures public participation through monitoring of health services, could be used to allocate resources based on publicly identified needs in specific health centers.
Afghanistan has made great strides in improving transparency around public procurement and how the budget is spent. However, there are still many steps that need to be taken, such as initiating e-procurement and improving the publication of contract data based on the Open Contracting Data Standard. More specifically to the COVID-19 situation, transparency and accountability around emergency procurement have to increase, taking into account civil society recommendations around this issue. The government should take the following actions:
Emergency procurement should be minimized to only necessary situations and when such decisions are made, they should be justified, recorded and publicized to create transparency and accountability around such decisions.
The government should publish emergency procurement data through its procurement portal AGEOPS in a timely manner, including all emergency procurements made by the provincial governments.
When publishing data, full publication should be the norm, including publishing all documents such as price quotations, procurement decisions including justification for those decisions, invoices, prices, type and quality of goods, and receipts.
Carry out public audits of emergency funds and emergency procurement by the Supreme Audit Office, engaging civil society and the public in the audit process.
Enable the participation of civil society groups and journalists through the provision of timely and detailed information and opportunities for meaningful engagement in monitoring emergency procurement and carrying out social audits of such expenditures.
We hope that these actions will help connect the budget process and the procurement with the issues that people are facing at the grassroots level.
In November 2019, Argentina’s Dirección Nacional de Vialidad (DNV), the national road authority in the Ministry of Transport, launched an data portal to publish information on all its contracting processes. This is the conclusion of a journey that started in 2018 with the help of the World Bank. Development Gateway was selected to carry out the project which included data publication and visualizations.
Since DNV’s work consists of public infrastructure projects, the information published includes data from both infrastructure projects and contracting processes. In this post, we will do a brief exploration of the published dataset and highlight its main features.
The published dataset includes all contracting processes related to public works managed by the DNV between 2016 and 2019. Contracting processes and infrastructure projects have a clear one-to-one relationship in the data, meaning that a single contracting process is carried out for each project. Also, each contracting process is represented by a single release. As of 4 May, there are 156 releases (and projects) in the data. Technically a daily automatic update is possible.
We’ll start by looking at the following selection of the OCDS sections used, and the coverage for each one:
We can see that all of the OCDS stages are included and although we can’t show the complete list here, most of the fields of the OCDS core data are used as well. The coverage is high for most of the fields, although we do expect the latest stages (awards, contracts) to have less coverage due to an issue we’ll explain later.
The coverage shows there’s good availability of information, which is a good starting point for more detailed analysis. The figure below shows a few high-level graphs made using the data. A few features of the tender process always show the same values across releases (as Submission Method in the figure). This makes sense considering that all processes are about construction work, which can have specific requirements for the tender process like an open tender.
A set of local extensions has been used to provide data that does not fit within the OCDS core data. Most of the extensions have been created to provide project information (as we’ll see later), but there are two with a special purpose that deserve an explanation here. Due to the rising inflation in Argentina, prices change quickly and it is necessary to adjust monetary values from time to time. Adjustments are also needed when we need to compare values set in different times: for example, the same construction service can have a significantly higher value today than six months ago.
DNV provides inflation-adjusted values for some fields in the OCDS data using the following extensions:
Opening value: adds the budget value adjusted to the bid opening date.
Current value: adds the budget and award values adjusted to the current date.
Therefore, if we want to make summarizations of values, we need to use values to the current date, because they are adjusted to the same date. The figure below uses the award values adjusted to December 2019, to show the top 15 suppliers with total awarded values and total number of awards.
Both extensions include a field named coefficient that is calculated internally by the DNV and used to calculate the value to date. Since not every value in the data has been adjusted to inflation, we should be careful when using other values present in the data. Values to date and coefficients are updated each month.
The remaining local extensions are used to provide project information for each contracting process:
Work Details: adds the current status of the project and project type (construction, rehabilitation, maintenance).
Progress: adds the advance percentage for each project, and the investment and actual values.
Certificates: adds information of the construction work certifications that are emitted through the execution phase of the project.
The figure below shows the values for the status field in the Work Details extension. Note that there is no status provided when the project is in the planning stage.
Even when the DNV makes a wide use of most of the fields in OCDS, there are a few details that can cause confusion to users that are familiar with the data standard. One example is that the items block in the tender section only contains the deliveryLocation block, from the Location extension, to include in the data the location in which the work is carried out. Another approach that could be easier to understand for users would be to create a local extension to put location data at release level.
Another possible improvement in the publication is to use the releases/records model. This can be particularly beneficial in the case of contract amendments, because even if amendments are provided, there is no way to track the changes made on each one.
Lastly, there are a few details that make it hard to track the current status of the contracting process. First, tags are not used correctly (most use the planning tag only when they are clearly ahead in the contracting process); second, empty lists and structures are not omitted from data (which is not necessarily an issue, but it does help to determine the stage of the process and to reduce the size of files); and third, some fields are filled way ahead in time, like contract titles when the data suggest the process is still in the planning stage. The last one causes the coverage to go down for some fields in the latest stages, as shown at the beginning of the blog post. Perhaps the tracking of the status of the contracting process has been disregarded in favor of the project status, but these may not be equivalent in all cases.
DNV makes good use of the fields provided by OCDS to publish their data, and using extensions (both existing and new) to provide details on the related projects. The use of adjusted values is particularly remarkable, as it enables users to make basic calculations and comparisons that would be very hard to do without.
It is important to mention that DNV started working in their open data portal before the launching of OC4IDS. OC4IDS is a data standard developed with CoST, designed to disclose projects and their related contracting data. Using OCDS only for project data disclosure has some limitations: for example, the DNV associates each project to a single contracting process, while an infrastructure project may have many contracting processes associated with it. As you have seen here, it also make it necessary to use extensions (as Location) to fully cover project data. In future updates, we recommend exploring OC4IDS to improve modeling potential one-to-many relations and increase the detail of project-related coverage within the data.
This work is a great first step to disclose infrastructure data in the region. We hope to continue supporting their commitment with public work data disclosure, as is stated in Argentina’s Government last OGP Action Plan (Compromiso #4); as well as the continuous update of the data in a timely manner. Are you interested in analyzing infrastructure project data? Have a try with this dataset. Go to the portal and look for the CSV and JSON links in the top right of the page. If you have any questions or comments about OCDS and OC4IDS, or if you want to share your experience with DNV data, please contact us, we’d love to hear from you!
This post is part of our ongoing content on COVID-19 emergency procurement. The civic procurement monitoring initiative DOZORRO and Transparency International Ukraine have analyzed the demand and prices of medical gloves in Ukraine in 2020, using ProZorro’s open contracting data and OCDS-based business intelligence tool. With a global shortage of gloves, it is critical to keep prices transparent to avoid potential manipulation by procuring entities and speculation by suppliers. The Ukrainian approach is a great example of how the disclosure and use of open contracting data can help manage such risks.
Protecting healthcare workers should be a top priority in the fight against the epidemic. However, in Ukraine, more than a thousand medical workers have been infected with COVID-19. According to YouControl, not all primary hospitals are sufficiently equipped with personal protective equipment, including medical gloves. Transparency International Ukraine investigated how gloves have been procured through ProZorro in 2020 and in what quantities.
Types of gloves
Medical gloves are meant to protect medical staff and patients from exposure to dangerous substances. There are two main types of medical gloves, depending on the application: examination (diagnostic) and surgical. Surgical gloves are made in more precise sizes; they are thicker, stronger and longer than examination gloves. They are used for more dangerous procedures.
Vinyl, latex and nitrile are commonly used for manufacturing medical gloves. Gloves made from natural rubber (latex) are the most popular, because this material is particularly strong and elastic. Its main drawback is that it can cause allergic reactions.
Nitrile gloves are primarily an alternative to latex gloves, since they are neutral to the skin and more resistant to mechanical punctures.
Medical gloves made of vinyl are used to work with chemicals in medicine. This year, fewer than 15 purchases of vinyl gloves have been made through Prozorro, so we did not include them in this analysis.
How many medical gloves have been purchased through Prozorro
As of April 20, almost 21 million pairs of medical gloves worth UAH 68.6 million (around USD 2.5 million) have been purchased in Ukraine via Prozorro. Most purchases were for latex examination gloves — 10.1 million pairs (48% of the total number), followed by 9.6 million pairs of nitrile examination gloves (46%) and 1.3 million pairs of surgical gloves (6%).
In the first three weeks of April, public procuring entities bought 6.5 million pairs of medical gloves for UAH 27.6 million, in March — 5.7 million pairs for UAH 19 million, in February — 6.3 million pairs for UAH 16.3 million, and in January — 2.5 million for UAH 5.7 million.
The number of tenders announced on Prozorro has increased every month. In January, 108 purchases of medical gloves were made, in February – 301, in March — 747, and from April 1 to 20 — 739. It is worth noting that gloves, especially in the last few months, are purchased not only by hospitals. There are many new procuring entities interested in them: utility companies, the police, the border service, and so on.
While in January and February 2020, customers bought medical gloves mainly through competitive procedures (pre-threshold procurement, open bidding), in March and April, it was mainly done by entering into direct contracts, primarily through the designated COVID-19 procedure.
Prices of gloves
We studied the median prices for various types of gloves. This is a more suitable indicator of the general price trend, because it is less sensitive to bias in the presence of an abnormally high price than the average price. Thus, the price fluctuation can be established more objectively.
Prices for medical gloves in January–April 2020:
Latex examination gloves
Nitrile examination gloves
Suppliers: Same gloves, different prices
According to our calculations, MK MEDGROUP LLC is the largest supplier of medical gloves. In January–April 2020, the company signed 37 contracts for the supply of 1.5 million medical gloves for UAH 5.1 million total. MK MEDGROUP LLC was founded in 2016 by Liudmyla Lytkova. The company specializes in the sale of pharmaceutical products. MK MEDGROUP actively participates in public procurement. In 2019-2020, the company participated in 493 tenders, which resulted in 272 contracts for UAH 95 million.
The second biggest supplier is Medical Center M.T.K. In January-April 2020, it signed 107 contracts for the supply of medical gloves for a total amount of UAH 4.5 million.
Medical Center M.T.K. was founded in 1994 and specializes in wholesale trade in pharmaceutical products. According to Youcontrol, the company is a member of the group (one of the founders) of the Yuria-Pharm Corporation.
Medical Center M.T.K. is an active public procurement participant. In 2019–2020, the firm took part in 3,931 tenders, 3,196 of which were successful, with the total amount of UAH 688 million. The company supplies pharmaceutical products and medical equipment.
According to our calculations, in April, Medical Center M.T.K. concluded 26 agreements for the supply of medical gloves. Analysis of these tenders showed that the highest price for gloves was three times greater than the lowest price among the different procuring entities. We contacted some of them to find out what gloves were supplied by Medical Center M.T.K. Most of the surveyed procuring entities replied that the manufacturer of gloves is the company “Yuria-Pharm”; some of the procuring entities refused to answer and suggested that we send them a written request.
The last purchases were made by O.S. Lunacharskyi Kherson City Clinical Hospital and Shostka Municipal Primary Healthcare Center. The supplier may have sold them gloves with different technical characteristics, which are actually more expensive. These two procuring entities refused to provide information about the purchase over the phone. A representative of the Kherson hospital promised to provide a response on April 23. We prepared a written information request to Shostka Municipal Primary Healthcare Center. At the time of publication, neither facility had provided a response.
It is worth noting that Medical Center M.T.K. supplies not only medical gloves to public procuring entities with major price differences. Nashi Hroshi wrote about the major fluctuation of prices for linezolid-based Linelid antibiotics supplied by Medical Center M.T.K.
Methodology: BI Prozorro, filters, sample
The procurement of medical gloves was downloaded from the public business intelligence module: the period of data download is April 1 – April 20, 2020, the search request included variations of the keyword “gloves.” After downloading the data, a filter was set to exclude procurement of non-medical gloves: construction gloves, textile gloves, household gloves, etc. After that, information on each tender was manually entered into an Excel spreadsheet. It should be noted that when data from the sample was processed, we excluded the tenders for medical gloves without a specification or necessary information on the procurement item. In the course of analysis, we identified that public procuring entities mostly purchased latex examination gloves, nitrile examination gloves, and, less frequently, surgical gloves.
At the final stage, a summary table was created with information about the quantity and price of medical gloves, suppliers and procuring entities.
Sierra Leone was one of the countries worst hit by the Ebola epidemic in 2014. I worked on DFID’s emergency procurement project – a tiny exercise compared to what will be needed to respond to the new coronavirus, but much of what I learned then is still valid today. Countries with inefficient and already overstretched manual procurement systems have gone into this pandemic flying blind. We need an immediate and dramatic global change in the current supply chain mindset to position electronic procurement as a critical tool against COVID-19 to both predict and fulfill demand on a country, regional, hospital, clinic, and patient level.
One of the reasons that people need emergency approaches is because of the clunking inefficiency of most government procurement. In most countries, it remains a paper-based, box-ticking compliance exercise rather than a digital service for government, buyers and the private sector. I’d like to make the case for rethinking that approach here.
Pity the poor procurement and logistics teams in Sierra Leone who had to rely on basic tools such as Outlook, Excel spreadsheets, Word, Google, and the good old fashioned telephone to try and manage the purchase and delivery of critical supplies.
These ‘tools’ were of very little help and were often an obstacle in facing up to the huge challenges of rapidly changing shopping lists, specifications and the shortages of even basic medical items such as personal protective equipment, body bags and thermometers, and a long list of other items needed for the local medical staff and British military to run the seven Ebola Treatment Centres (ETCs).
Even during the Ebola emergency we faced long and congested international supply chains, with competing buyers willing to pay any price to get their hands on lifesaving equipment and drugs.
To make matters worse, when trying to keep up with the constant demands from the brave medical staff, we had an almost complete lack of data on prices, suppliers, lead times and specifications to call upon. Finally, there were complex logistical challenges in buying in bulk and then consolidating items from numerous suppliers into ETC packs to be flown out on air charters and then distributed locally.
I’m sure that many professionals in procurement departments across the globe are now having to rely on these same basic tools to try and manage their emergency procurement for COVID-19. However, governments should be considering the use of electronic procurement platforms, which are readily available off the shelf, often on a software as a service subscription basis, to do their emergency procurement much quicker, more transparently, and much more efficiently.
Direct contracts, competitive tenders, framework agreements, e-catalogues and call-off contracts can all be managed much better using electronic procurement, as can the publication of tender and contract award notices online. With many countries imposing lockdowns or travel restrictions, bidders are likely to find it difficult or impossible to deposit hard copies of their bids in a tender box, and in many instances, buyers may have already switched over to only accepting bids by email, which provides ample opportunities for corruption or fraud.
Issuing tenders and receiving bids online through an electronic procurement system instead offers significant advantages in terms of saving time (from factory to patient and medical staff), reducing corruption and increasing security and data capture and utilization.
Many procurement staff are likely to be self-isolating in their homes and electronic procurement offers a practical way to manage their procurement remotely and work as efficiently as possible as a virtual collaborative procurement team. This tool can create an automated, data-rich audit trail with all data and documents securely stored in one platform rather than disbursed in email systems (often including personal email accounts), unstructured documents and spreadsheets in the cloud, laptops, network drives and paper files in filing cabinets.
The data captured using electronic procurement can be used and reused, and help detect over-pricing, fraud and corruption, as well as providing valuable insights for sourcing and predicting future demand and supply. Business intelligence tools, driven by the transactional data captured by electronic procurement systems, will improve reporting, analysis and responsiveness to the needs of the healthcare system. These tools will help to provide a real-time picture of the COVID-19 supply chain and identify poor performing suppliers and bottlenecks. Electronic procurement can also help with aggregating demand from various procuring entities and minimizing the chances of the government competing with itself to secure scarce supplies of urgently needed items.
It can be used by governments to improve compliance by buyers and suppliers and to open up their contracting activities to scrutiny by civil society. And it can help governments to maintain public confidence and meet their transparency obligations enshrined in procurement laws and regulations with minimal effort and burden on civil servants.
National COVID-19 procurement strategies, if they exist, need to be rapidly integrated and transformed into a global digital one to enable the treatment of patients and the protection of medical staff in possibly every country in the world affected by the pandemic. Without high-quality machine-readable and transparent procurement data and tools to understand needs and prices and enable collaboration across agencies, there will be unnecessary and huge casualties in both groups. By adopting the ready-to-use Open Contracting Data Standard, procurement data should be of a higher quality and importantly can be shared, compared, aggregated and used by multiple stakeholders to improve global supply chain collaboration.
It might not seem that important in a crisis of this scale, but a simple example demonstrates the impact the current lack of standardization in data can cause: the many names for referring to the virus itself, including COVID-19, COVID19, Coronavirus, Covid-2019, and SARS-CoV-2. Buyers and suppliers scouring the internet using a variety of these terms to find each other may fail simply because of an inconsistency in terminology. Apply the same principle to all of the different data types in the full public procurement cycle and the resulting data chaos is bad enough in normal times, let alone during this global emergency, hence the need for the adoption of data-centric electronic public procurement which uses the Open Contracting Data Standard.
We have the conditions for an international perfect storm where people will lose their lives not necessarily just to COVID-19 but in addition to, or because of chronic failures in the national or local procurement systems, supply chains, logistics networks and the health systems that rely on them. The Just-in-Time supply chain systems that have served the health systems of many countries well enough for many years, including the UK, are tragically turning into Not-in-Time or Far-too-Late for patients and the medical staff, as we are already witnessing with the PPE shortages in the UK and elsewhere.
As the supply chain fails, there is an increasing risk that the infection rate of our heroic doctors and nurses will accelerate as they expose themselves to the additional risks of a high viral load without the necessary PPE, which will result in an even greater strain on health care systems and repercussions to patient care.
There needs to be an immediate redesign of the supply chain from factory to patient before it crashes. We cannot assume that countries with local manufacturing capacity for medical items will be willing to endanger their own healthcare systems and export to other countries, or indeed whether supply will match the rapidly changing demand we can expect to see in the coming days, weeks and months.Electronic procurement tools and other supply chain platforms need to be rapidly built and deployed to help predict and aggregate demand so the global manufacturing capacity can plan and ramp up production accordingly and so the right items can get to the right patient at the right time. Our lives depend on it right now but the pandemic also offers an opportunity to start making a quantum leap to truly digital public procurement across the globe so that it is fast, open and smart every day for everything everywhere.
A safe exit from the Ebola red zone in Sierra Leone – Joseph gives a thumbs up as he waits to be safely guided through the careful exit procedure from the “red zone” at the Makeni Ebola treatment centre in Sierra Leone (Picture: Jessica Seldon/DFID)
By Vasyl Zadvornyy, Chief Executive Officer, State-owned enterprise ProZorro,Ukraine
A global pandemic is a daunting challenge not only for the healthcare system and the economy but for procurement as well. The OCP, OGP, OECD, Transparency International and many other organizations that advocate for transparency and accountability have issued their recommendations and guidelines for maintaining the supply of healthcare during the outbreak, and I’d like to share some of the positive steps being taken in Ukraine.
As the new Coronavirus pandemic has grown exponentially, so has the demand for supplies. The lack of medicine and medical equipment is arguably the biggest procurement challenge. Thousands of procuring entities have experienced an unprecedented and unpredictable surge in demand. At the same time, resources are limited. Ukraine is addressing this issue with the help of open contracting data in three ways: demand analysis and consolidation; supplier engagement; and procurement planning and management.
The quarantine measures introduced in Ukraine on March 12 created a huge demand, not only from hospitals and healthcare entities, but also from a multitude of other sources: citizens who sought to buy their own masks and other personal protective equipment, antiseptic, medicines and even ventilators; private companies who needed to safeguard their employees; international financial institutions and donors who were repreparing to reshape their programs to address the pandemic; and volunteers, who actively organized and created their own order lists. Some of these buyers set out to purchase everything they might need in the future: “just buy everything, we still won’t be able to over procure”, one volunteer told me.
But this is obviously not the best way to conduct public procurement. So Ukraine introduced several tools to consolidate demand. These included a tool created by the Ministry of Health and the Medical Centralized Procurement Agency, which helped hospitals to predict demand on over 100 items based on hospital capacity, existing equipment, and current load; for example, as of April 6, the vast majority of core and non-core hospitals (more than 2000) have reported their demand and current stocks, and a public business intelligence tool has been launched to communicate this information to the market and citizens.
Donors.ua, an NGO that usually manages blood donations, also provided a platform for hospitals to create simple request lists for COVID-19 items https://donor.ua/aid/hospitals. Although this tool is not mandatory, it is already widely used by volunteers, who can find a particular hospital and provide assistance to the facility directly.
Another significant concern is the shortage of suppliers. An overburdened healthcare system is one of the greatest dangers to managing the pandemic. This is applicable not only to the capacity of hospitals and equipment like ventilators but also to whether suppliers can provide all required goods and services. Not surprisingly, this leads to shortages and prices skyrocketing. For example, recent research by civil society found mask prices in Ukraine have increased by more than 10 times since January. The pricing rally has put pressure on suppliers: companies are updating their prices several times a day in order to meet current demand. On the other hand, the situation is creating a huge incentive for those who didn’t fulfil orders to increase their production capacity. The civil society research also found that three of the top five suppliers of masks in March had not sold such goods previously.
Open data are being used to increase suppliers’ capabilities in several ways. First, the public procurement system Prozorro is a perfect resource for those who are looking for potential suppliers. All public procurement data from 2016 are available through an open API, including all bidders, their contact details, and the regions in which they operate.
Another way to simplify procurement is to publish all current proposals available online. Anyone can see current prices, changes and available goods. In Ukraine, Prozorro Market — an e-shop for public procuring entities, which was launched in April 2019 — is being used to collect all offers in one place. The Prozorro team is doing its best to boost the number of proposals: the supplier qualification process was radically simplified, and the most popular goods, such as test kits, sanitizers, masks, and protective equipment, were quickly added to the stocklist. Now Prozorro Market can be used as a tool for researching suppliers, price monitoring, and of course purchasing. By collecting all the proposals on one platform, we also reach a price equilibrium more rapidly: suppliers react to their competitors’ pricing strategy in a real-time manner.
And last but not least, through data openness, it is easy to show the procurement volume and empower new businesses to become suppliers. The quarantine was introduced in Ukraine on March 12. Within eight days, all necessary changes had been made both at the legislative and technical level to establish a special process for conducting public procurement to curb the outbreak. The procurement procedure may have been simplified, but still remains traceable and transparent. Thus, we can already tell from Prozorro data that Ukraine spent almost UAH 1 billion (USD 36 million) within 16 days (until April 5). Such data are widely used by businesses who might consider turning to the public procurement market in times when their regular business opportunities decline.
Having demand and supply consolidated makes planning procurement much easier. As I mentioned earlier, a fast, simple but still traceable procurement method was established by legislation. It doesn’t impose open bidding but requires announcements to be published along with all selection criteria and full, timely reporting. Thus, all procedures can be finalized in one day, while still providing a lot of data. And a comprehensive procurement strategy is being developed, based on data collected. The Medical Centralized Procurement Agency (established in 2018) is starting to conduct high-volume bulk procurement, which will increase a procure’s bargaining power and, hopefully, reduce the end price. Open data are vital here to understand both demand and supply.
Moreover, the same data are critical to international financial institutions, private volunteers and charity organizations, who need to understand an actual uncovered demand as well. This is simply addressed by procurement plans, which are being published on Prozorro as well.
Procurement risk minimization is another important use for data. A monitoring tool and legal entity, YouControl, provides free access for all procuring entities, so they can minimize their risk of dealing with an unreliable supplier.
Communication, a critical success factor
To be useful, data must be properly used and accompanied by stakeholder engagement, and communication. That’s why not only data collection but also investigating insights and regular communication are vital parts of our job.
Procurement procedures should be awarded in an open and simple way in order to engage new suppliers and create incentives for potential businesses, with the information provided about market capacity, demand level, and competitiveness. Moreover, overall procurement statistics should be communicated to help involve in public procurement those who will be hardest hit by an economic recession caused by the outbreak.
Prozorro does this by actively communicating with firms through the media and business associations. On a weekly basis, we share information on the most important COVID-19 related procurement and the current state of public procurement (compared to what it was a year ago).
Another communication audience is procuring entities. They require guidelines and recommendations on a number of topics: what should be bought and how; which criteria are recommended and how to find suppliers. Civil society activists from the watchdog project Dozorro created guidelines that were distributed to procuring entities by representatives from the Ministry of Health.
Civic monitoring: Data analysis for transparency and accountability
Opening procurement data helps to establish transaction traceability so any wrongdoing can be easily found, disclosed and hopefully, prevented or corrected.
For this reason, data analysis tools were an important part of ProZorro from the very beginning. Similarly, when specific procurement procedures were established for managing the coronavirus outbreak, appropriate updates were made to the business intelligence tools (e.g, bi.prozorro.org and its description).
Transparency International Ukraine, based on its watchdog project Dozorro, started a dedicated project to focus specifically on COVID-19 procurement. The most important topics are analyzed several times per week, including procurement for masks, tests, and ventilators.
Another good example of open contracting data use is a medical equipment and supplies procurement map developed by journalists from Slidstvo.Info. With these simple dashboards, anyone can check how many test kits or ventilators were or will be procured by hospitals all over the country.
But it isn’t only the COVID-specific procurement that should be monitored. In tough times, the public becomes more sensitive to opaque or unjustified procurement purchases. So any significant spending should also be proactively communicated.
For example, after a public outcry, the Ministry of Internal Affairs canceled the procurement of around 600 vehicles, and the Kyiv city administration canceled a $2.5m contract for street cameras that would have allowed to recognize faces and measure people’s temperature on the streets.
Dealing with an emergency requires a holistic approach. Data openness and government accountability cannot be sacrificed, even when radical and urgent steps are needed. Rather than trying to seize as much control as possible, the only way to address such a situation is to build a stakeholder coalition around open and clear contracting data. Contracting data are only a part of a supply chain: the more understanding we have of them, the more we can make smart decisions based on them. In our current circumstances, it will directly impact our lives.
Michael Canares is a managing consultant at Step Up Consulting, a Philippines-based social enterprise which conducted research on an open contracting pilot project in the Indonesian city of Bandung, with financial support from HIVOS’ Open Contracting Program. The Bandung project sought to increase access to public procurement information and improve various actors’ ability to monitor the public contracting system, including government staff, vendors, civic groups and journalists. Despite strong initial interest and uptake of the data, maintaining engagement has proved challenging. In this post, Michael introduces the project and offers words of advice to those carrying out similar initiatives.
Bandung is a bustling, lively city in Indonesia with a population of over 2 million. In 2015, the local government began working with the national procurement agency to make the city’s public contracting data more accessible and more readily available, while also developing people’s skills to monitor the procurement system using data.
Despite reforms, Indonesia lacks accessible procurement data
Public procurement in Indonesia is often marred by inefficiencies, as well as poor transparency and accountability. This leads to massive state losses from a system that accounts for around a quarter of the national budget. To address these problems, the government has introduced several reforms since 2012, including e-procurement for both national and sub-national purchases, an e-catalogue system for commonly used supplies and equipment, and new legislation for better vendor management that simplified requirements for suppliers, with an eye to later introducing e-tendering.
The open contracting initiative: From publication to use
The project to adopt open contracting in Bandung had three components:
publish the Bandung government’s public contracting data and information in open data formats;
develop key performance indicators on public procurement and related data visualizations
facilitate citizen engagement and practical use of the data and statistics through the provision of ICT tools and targeted capacity-building to stakeholders from government, civil society and the private sector.
Technical and financial support for the project was provided by the World Bank, which commissioned a local consultant and Development Gateway to achieve the three objectives above. To facilitate component 3, the Bank enlisted the help of the World Wide Web Foundation’s Open Data Lab Jakarta.
As a result, Bandung published more than 40 000 procurement records from 2015 to 2018, along with online visualizations. The city government also worked with its different departments to improve the quality and availability of contracting data.
On its BIRMs portal (Bandung Integrated Resources Management System, the government published data on new and advertised tenders with sufficient details such as department, sources of funds, the deadline for applications, upper limits, terms of reference, start date, eligibility, supporting documents, among others. What the portal lacked, however, were the data on the award contracts, including company details of awardees, expected deliverables, and contracted amounts.
To improve engagement and use, the Open Data Lab Jakarta implemented a three-phased approach, starting with use case research, followed by user engagement activities, and a public launch. In the use case research phase, the Lab carried out an online survey to identify the potential user groups, their characteristics, motivations for engaging with contracting data, and their data needs. Identified user groups were then invited to a workshop to develop use cases that are relevant in addressing the key priority issues faced by Bandung City. Use cases were developed around specific challenges or benefits that open contracting data can positively impact.
The Lab developed engagement strategies based on the results of the research, which had highlighted a poor understanding of contracting processes and contracting data, even among stakeholders whose work or advocacy was affected by public contracting practices. The research also revealed a strong interest in public contracting, especially with data related to health, city planning, social development/poverty reduction, communication and informatics, and environment, coming from the civic tech community and journalists.
Data visualizations created by students
Communication design students from nine local universities took part in a competition (‘visualthon’) to create appealing and easy-to-understand infographics to raise the residents’ awareness of the City’s open contracting data. They produced 11 infographics which were presented at a public event in November 2018.
News articles on public contracting
24 journalists attended a two-day workshop in which they learnt how to use Bandung’s open contracting data portal and work with procurement data. They were then invited to take part in a two-week long competition to write articles using their new skills.
Representatives from government, businesses and civil society took part in a workshop to build prototypes of applications using open contracting data. This was followed by a period of mentoring and training to develop and test the solutions. The project teams produced an app on public transport procurement, a notification service for public procurement opportunities and a dashboard for disposable health supplies.
Here’s a summary of these engagement strategies:
University students across 10 design and IT universities in Bandung
Increase awareness among city constituents about the existence and importance of open contracting data disclosed by the city government (with a specific audience in mind – business community, transparency advocates, media, etc)
Visualization and communication materials based on available open contracting systems and data
Local journalists from print and broadcast media
Strengthen the capacity of infomediaries in using open contracting data to provide evidence-based reportage on contracting issues
3-minute-read contracting stories published
Incubation of projects
Local civic groups, activists and civic-minded technology experts
Demonstrate the value of open contracting data in longer-term engagements that have the potential for sustained positive social impact
3 apps developed:Push notification for contracting opportunities aimed towards businessesAnalytics dashboard on the procurement of disposable medical devicesAndroid-based mobile app on procurement activities in the transport sector
Results: Those typically excluded from procurement were involved
The engagement strategies sought to cast the net wide and invite all, what we call an “inclusive” approach. Although this approach was not targeted at marginalized communities, it did result in their participation, particularly women’s groups, such as the IWAPI Bandung (Ikatan Wanita Pengusaha Indonesia / Indonesian Business Women Association). But the non-purposive inclusive approach also meant that some organizations were unintentionally excluded such as the Bandung chapter of ASPEKINDO (Asosiasi Pengusaha Konstruksi Nasional Indonesia / Indonesian National Construction Entrepreneurs Association) and HIPMI Bandung (Himpunan Pengusaha Muda Indonesia / Indonesia Young Entrepreneurs Association) who were contacted but did not respond.
Non-participation was found out to be caused by at least three things: historical bias (e.g. some organizations do not necessarily see any use of engaging with the government because of past negative experiences), lack of incentives (e.g. engagement may not be the best option for them), and negative attitudes towards the change (e.g. maintaining the status quo works better for certain entities, especially the businessmen who are benefitting from the lack of transparency).
Nevertheless, the project has had a positive effect in involving different stakeholder groups that are habitually excluded from procurement processes. For example, in the past, journalists did not have access to procurement data. The project led to at least four reporters publishing stories that questioned the government’s procurement decisions. Given these outcomes, there is a certain degree of empowerment that took place following the relatively small step of disclosing data and building the capacity of users to apply it.
A year on: The challenge of maintaining momentum
The journalists continue to engage with public contracting. A local news organization ran an event to investigate open contracting and budget data for Open Data Day 2020 and a WhatsApp group remains active. Of the four who were able to publish stories, one continued to write about contracting activities. But scarcely a year after the engagement activities ended, much of the sustainability of the other initiatives is in question. None of the visualizations developed by the students were used by the government for its awareness activities, and the apps that were developed into prototypes never went further. This was a fear expressed early on by the Open Data Lab Jakarta team – the so-called ‘vapourware syndrome’ which also exists in the civic tech sector, which describes that situation when apps are developed but do not see deployment for several reasons, largely due to the lack of an enabling environment, according to a study by Cowater and UCT.
The Bandung case also suggests that inclusion by design is insufficient, and even inclusion through implementation. An inclusive process will not necessarily bring about sustainable inclusive gains, especially when the underlying power dynamics do not change. While it is true that the project, along with the support of the city government, attempted to engage different types of users, the publication process was marred by inefficiencies, particularly by the reluctance of the city government to share the API (an interface that allows two different systems to “talk” to each other). They expressed concerns that sharing it would expose the city government to risks. And yet, without it, and without the resourcefulness of the civic tech activists to bypass authorization procedures, the mobile app prototypes could not have been produced.
Further, due to the lack of economic resources of the civic tech activists, the initiatives remained as prototypes because of their inability to pursue development and their failure to market the solution to intended audiences was hampered by the lack of financial resources. The support provided by the project at its initial stages was insufficient. Despite the fact that the prototypes could have helped the city government to further strengthen the open contracting initiative, the necessary financial support was not provided.
Engagement is a means to an end
This finding raises an important point in terms of inclusion: inclusion is not an end-goal but a process; a reiterative process of nurturing the sustainability of intermediaries and their efforts to create value out of data. While indeed, in the case of Bandung, disruption of data flows was brought about by the disclosure of contracting data by the city government of Bandung, it failed to lead to value creation. This was because of an absence of sustained support for the newly engaged intermediaries who were tackling the difficult topic of open contracting and public procurement. Had the intermediaries been established organizations, or well-funded private companies and tech start-ups and/ or individuals, the conversion of the opportunities they had identified and developed into actual value products could have been sustained and the development process could have been pursued after the initial incubation stage.
Sadly, this is not necessarily the case for a lot of local governments in Indonesia. Most of the potential users of contracting data do not have the resources, even the capacity to benefit from it. This project in Bandung required significant time from the implementers to build capacity and put in the incubation resources for actual use cases to emerge. While support is needed for data publication, more assistance is needed in ensuring that published data are understood and used to strengthen transparency and accountability in public procurement.
This post is part of our ongoing content on COVID-19 emergency procurement. The civic procurement monitoring initiative DOZORRO and Transparency International Ukraine have analyzed the demand and prices of surgical masks in Ukraine over the first quarter of 2020, using ProZorro’s open contracting data and OCDS-based business intelligence tool. With a global shortage of masks, it is critical to keep prices transparent to avoid potential manipulation by procuring entities and speculation by suppliers. The Ukrainian approach is a great example of how the disclosure and use of open contracting data can help manage such risks.
On April 1, the Government of Ukraine introduced stricter measures to fight COVID-19. Among other restrictions, all people must wear a mask or respirator in public places.
There are different masks
There are several types of medical-grade masks being procured by Ukrainian authorities. A single-use surgical mask reduces transmission of the virus through exhaled droplets. It is usually a three-layer mask with a filter between two external layers of non-woven fabric.
There are also three types of respirator masks offering varying levels of protection: FFP1, FFP2, and FFP3. They are usually used by medical staff or the military working on the frontlines of the emergency response.
Mask prices in the ProZorro system
Nationwide quarantine measures and the rapid spread of COVID-19 in Ukraine have led to price gouging and shortages of goods that were not in high demand before the outbreak. Medical masks are up to twenty times more expensive than they were in January 2020. So how many medical masks have been purchased in 2020 via the country’s e-procurement system ProZorro and for what price?
According to our calculations, public entities bought almost 2.5 million surgical masks for the total amount of UAH 25 million between January 1 and March 28, 2020 (± USD 1 million with an average exchange rate of 1 USD = 25 UAH). The month with the largest volume of purchases was March at 1.8 million masks. In February, entities bought 547,000 masks, and in January 108,000.
In mid-February, procuring entities started buying surgical masks more actively. For instance, on February 14, eight tenders were announced on ProZorro for the purchase of 228,000 units in total. Almost 90,000 units were purchased through direct agreements; the rest through open bidding.
The second peak of procurement happened on March 26, when over 1.1 million masks were purchased. One million of them were bought by Kharkiv Municipal Clinical Hospital No. 13 for UAH 14 million (± USD 560,000). The winning supplier, SOFI-MED LLC, must deliver them by April 30.
In January and February 2020, procuring entities mostly purchased surgical masks based on competitive bidding (below-threshold simplified competitive procurement, open tender). Most of them were successful and ended in the conclusion of agreements. In January 2020, suppliers bought surgical masks for UAH 0.68-1.5 on average, in February for UAH 0.7-3.2.
But in March 2020, 55% of competitive tenders were unsuccessful, meaning not a single participant made a bid. This means that there was a shortage of the product and/or that suppliers raised their prices. At the beginning of March, suppliers refused to sell medical masks for UAH 2-3 per unit, and in the middle of March, even UAH 7-12 per unit was not enough anymore. The average price jumped to UAH 10-25 per item.
For instance, ME ODESFARMsupplied 1,000 surgical masks for UAH 24.5 per item to the Department of Municipal Property of Odessa City Council. SP Zheleznova O.V.supplied 2,000 masks for UAH 25.5 per item to the Territorial Directorate of the State Judicial Administration of Ukraine in Donetsk Oblast. SP Bastiuchenko O.V.supplied 5,000 masks for UAH 23 per item to the Center for Traffic Organization.
The rapid price increases and the big gap between the average and maximum price may indicate price gouging. Although definitive conclusions cannot be drawn from the data alone, scrutinizing the data provides critical insights to be investigated further.
On March 12, Deputy Minister of Healthcare Viktor Liashko said that the Anti-Monopoly Committee of Ukraine would check prices for personal protective equipment. By the end of the month, on 30 March, the Kyiv Oblast Territorial Division of the AMCU opened a case against manufacturers, suppliers and pharmacy chains in Kyiv and Kyiv Oblast due to the significant increase in mask prices.
It is interesting that one of the suppliers involved in the case sold masks to public procuring entities in March for the same price as in January and February 2020 – around 3 UAH per unit. In March, Tekhnokompleks LLC supplied 5,000 masks to Okhmatdyt children’s hospital for UAH 2.99 per unit, 3,600 masks for UAH 3 per unit to the Head Directorate of the National Police of Ukraine in Cherkasy Oblast, and 1,500 masks for UAH 2.96 per item to Kyiv Municipal Children’s Infectious Hospital. While in January and February these were some of the highest prices in the system, in March they were low relative to others.
Who are the top 5 suppliers of masks?
For the period January to March 2020, we can identify five leading vendors that earned the most and supplied the largest number of masks, accounting for 67% of the total spending and 48% of the total quantity. They are SOFI-MED LLC, TRANSPORT COMPANY AUTOLINE LLC, ODESFARM municipal enterprise, SP Polozhii Yevhen Oleksandrovych, and DIMAKS VBK LLC.
The top supplier is SOFI-MED LLC, which sold one million medical masks for a total sum of UAH 14 million (that is, UAH 14 per unit) to Kharkiv Municipal Clinical Hospital No. 13. According to information provided by the analytical system YouControl, the primary activity of SOFI-MED LLC is wholesale trade in pharmaceutical goods.
Even though the company has existed for just over a year (it was registered on December 22, 2018), it actively participates in public procurement. From February 2019 to March 2020 inclusive, SOFI-MED LLC participated in 147 tenders, winning 61 tenders for a total amount of UAH 22.6 million. In 90% of the tenders, the goods supplied were medical supplies, equipment, and pharmaceutical items.
The main activity of the company is technical maintenance and repairs of motor vehicles. Unlike the top supplier, TRANSPORT COMPANY AUTOLINE was created eight years ago, but hasn’t been a very active participant in public procurement.
Between January 2018 and March 2020, the company participated in 55 tenders, with only nine of them competitive and the other 46 awarded through direct negotiations.
The main customer of TRANSPORT COMPANY AUTOLINE is the Odessa Municipal Electric Transport entity. The company entered into 38 agreements with this procuring entity, out of a total of 49 agreements concluded with any entity.
Before March 20 this year, the company specialized exclusively in car parts and car repair services in public procurement. But on March 20, TRANSPORT COMPANY AUTOLINE started supplying medical items, personal protective equipment, gloves and disinfectants. The company entered into 21 direct agreements for the supply of these goods, including 11 with the same Odessa Municipal Electric Transport entity. The company sold masks for UAH 17 and UAH 18 per item.
The third-ranked company is ODESFARM municipal enterprise, which sold 53,000 masks for UAH 785,000 to various procuring entities. ODESFARM is a municipal enterprise of Odessa City Council, which engages in retail sales of pharmaceutical goods. It is an active player in public procurement. The enterprise has participated in 442 tenders, with 332 concluded agreements. Between March 16 and March 27, the company sold masks for UAH 17 to UAH 24.5 per item, i.e. the municipal enterprise sold masks for higher prices than private companies.
The fourth place on the list of top mask suppliers belongs to Polozhii Yevhen Oleksandrovych – a sole proprietor, who won one tender in Prozorro and sold 40,000 masks to Kharkiv Heating Networks municipal enterprise for a total of UAH 600,000, that is, for UAH 15 per item.
The primary activity of Polozhii is wholesale trading in clothes and footwear, as well as production of leather clothes, work uniforms, etc. Before February 2017, Polozhii had engaged in wholesale trading in meat and meat products but had not participated in public procurement.
The fifth largest supplier is DIMAKS VBK LLC. The company sold 21,500 masks to public entities for a total amount of UAH 366,000.
The company is quite young and has existed for only 1.5 years. It specializes in the construction of housing and non-residential buildings, electrical and other construction works.
DIMAKS VBK has participated in just four public tenders. In the first two, it supplied channels and pipes, in two others – medical masks for UAH 17 per item.
With the insights from open data on contracts and prices, we can monitor which procuring entities are purchasing which items, when and how. We can also monitor who wins these deals and all the anomalies in the system to keep our public spendings accountable. Based on similar research, the Anti-Monopoly Committee has already started its investigations. Moreover, such analytics and use of data (including our business intelligence tool) empowers procuring entities. It gives them an opportunity to better understand the market and plan their procurement purchases properly.
This post in our technical series explains how to generate an JSON in the Open Contracting Data Standard from tabular data.
There isn’t a single approach when a publisher, for example, a public procurement agency, decides to start a transparency process by implementing the Open Contracting Data Standard (OCDS), and converting the data managed by their e-procurement systems into OCDS. The architecture and tools chosen to implement this process will strongly depend on the internal technical capacity to build the Extract, Transform and Load (ETL) procedures, or evaluate the need to hire external consultants or development teams.
From the variety of tools available at the building stage, chances are that SQL is one of them. In public institutions, it’s common to have an IT department that deals with the administration of internal systems. Members of the IT department may or may not have knowledge of programming languages and frameworks, but it’s almost a given that they would be familiar with some DBMS (DataBase Management System) and its SQL flavor.
Even when programming gives more flexibility when building systems, it’s totally possible to develop a full transformation process using SQL only. Now, you may know that producing JSON from SQL queries and functions is technically possible but takes a lot of time and effort to achieve in real life; especially with large instances of JSON, as OCDS can have. Thankfully, there is an alternative: producing tabular results that can be easily transformed into one of the OCDS secondary serialization formats, and then converting the files to JSON using an existing transformation tool.
In this blog, we will present the steps to use this method. At the helpdesk, we have been fortunate to follow two real-world cases in which the approach is already being implemented in practice: first we have the case of Paraguay’s DNCP (the national procurement agency), where the transformation process is being written in SQL to take advantage of the local resources for future maintenance; and the second case is MIGA (Movimiento de Integración Gastronómico Boliviano in Spanish), a Bolivian NGO that wants to use OCDS to share their collected procurement data but also have the data in a friendlier format for non-technical users.
The approach and uses
We’ll start by explaining what the OCDS secondary serializations are. To be compliant with OCDS your data must be published in JSON format, but the standard allows alternative formats to be published alongside JSON. Secondary serializations can be represented in any tabular file format like CSV, but the transformation between the nested structure set by the standard to a tabular format is not straightforward.
OCDS documentation proposes two approaches for the transformation: simplified single table, in which the output is a single table, and multi-table with the output is distributed into several tables. Ideally, we would like a multi-table approach if we want to transform all the available fields in the data.
The figure below gives a glimpse of how the transformation should be done from a single JSON source.
As you can see, arrays with multiple items within the JSON structure can be represented as tables of their own. The linkage between tables is done through keys shared between tables, where the ocid and release id are shared between all tables and additional ids are used where needed, as in the items table which has a reference to the awards table through the awards/id column.
As said before, using SQL to produce JSON can be next to impossible, but producing results that can be stored as CSV files that look like the example above should be significantly less difficult. The difficulty of writing the SQL scripts needed to achieve this would depend on the size and complexity of the source(s) system(s). Below is an example taken from Paraguay DNCP’s open source project:
Once we have the scripts in place, the next step will be to transform these CSV files into JSON. There is more than one tool to do this, but we recommend the Flatten Tool. The Flatten tool is a generic JSON to CSV/Excel transformation tool (and vice versa) that can be used as a Python library and also provides a CLI interface. The transformation takes a single line command; you can go to the documentation of Flatten Tool for OCDS for use examples.
The following figure describes the architecture to do the exports from source to OCDS as explained:
In Paraguay, the DNCP is currently extending the architecture proposed to upgrade their publication to OCDS 1.1. After the last step, the JSON output is inserted into an Elasticsearch server, to provide an API service as well as static files for direct downloading. It is expected that any future modifications needed because of new data or upgrades could be done in the SQL scripts, so the DNCP can use their own internal resources for maintenance of the process. The project has an open source version that can be found here.
Another potential use of the tabular version of OCDS is to create database representations. Since it already defines relations between tables through keys, the structure and data can be easily taken into a database, and keeping the original structure would ease the transformation between the database and the official JSON format: from tables to CSV, and then from CSV to JSON.
A practical example for this comes from MIGA in Bolivia. They have been collecting procurement data on food products for the monitoring of public programs, and opted to load their data into a database with OCDS CSV-like tables. This would facilitate the transformations to OCDS, sharing their data when needed, and keeping everything in a database that will allow journalists and other less tech-savvy users to use the data without being familiar with JSON. Paraguay’s DNCP is also considering the possibility of offering a “database backup” version of their data using the tabular OCDS structure.
Challenges and considerations
Despite the advantages of this approach, there are some considerations to keep in mind if you want to implement it.
First, it is important to know that the names of the columns in the tabular format can get quite large for fields that are deeply nested. See the example of the image below, with shows the table for the contracts/implementation/milestones/documents block:
In most database engines, names of tables and columns are limited in length. The example in the image is from MySQL 5.7 which has a limit of 64 characters for both. The field:
has 65 characters already (and it’s not fully displayed in the image). Depending on the source data, it’s possible that you won’t reach the name limits of your database engine. But if you do, you should take provisions to ensure that the tables and column names are correctly generated when dumping data to CSV files.
Another challenge is with the Flatten Tool itself. Users have reported to have run out of memory when attempting to convert large volumes of data. To use this tool you will have to decide a sensible segmentation criteria to generate the JSON (release/record) packages. For example, you may not be able to generate a single package for a full year of data, so you will have to divide your data using another criteria, like by month or year and buyer.
Using a tabular OCDS representation to store and generate OCDS JSON can be a very useful approach in more than one scenario, as proven by the cases of Paraguay and Bolivia. We are hoping that more publishers find it useful, and help us learn what could be improved as well as what tools and techniques we could create together to make the publication and use of OCDS data easier.If you think this strategy might be good for you or your organization, get in touch with us! Send an email to firstname.lastname@example.org and we’ll be happy to assist you.
Moldova has partially fulfilled its commitment to increase public procurement transparency. Since October 2018, all above-the-threshold procurement is conducted through the MTender system. A survey conducted by IDIS Viitorul revealed that 53.3% of civil society representatives believe that transparency significantly increased in 2019. However, MTender does not cover the entire procurement cycle. Civil society currently has access to all procurement-related documents from planning to contract award stage, but not the contract implementation stage.
There’s another important factor we should be considering: does transparency solve all issues in a procurement system? We don’t believe so. According to the open contracting implementation cycle and principles, the publication of data is only the first step! And then comes engagement with stakeholders, analysis, data-driven changes in policy and constant improvements, again and again. That is why we, at the civil society organization IDIS Viitorul, continuously monitor not only individual tenders but developments in the procurement system in general. We also engage with different stakeholders and are focusing our efforts on a systemic change!
The monitoring results reveal that only 10% of the actions set out in the 2019 plan have been accomplished (that is one out of the 10 commitments, which aimed to inform potential vendors about opportunities to participate in public procurement procedures and curb anti-competitive practices). Another 40% were partially accomplished, while the remaining half of the actions were deemed unaccomplished. In addition, some 60% of the actions were evaluated as producing no impact.
Advancing procurement reform in Moldova is important to ensure the efficient spending of limited financial resources and proper economic growth. One of the main objectives of the National Strategy is to eradicate corruption, in particular in the public procurement sphere. Also, the country has to fulfill its commitments under the Association Agreement with the EU. Accountability and integrity must be added as priorities, in addition to transparency. As an organization, we understand the political challenges and constraints and try to provide constructive feedback with a clear set of actions to improve the situation.
What do we recommend now to improve the anti-corruption efforts?
Finalize and conduct a public consultation on regulations governing the procurement of goods, works, and services by state-owned enterprises or by joint-stock companies with full or majority state capital. This huge piece of the procurement pie is still conducted behind closed doors. We suggest elaborating on an assessment of the benefits of centralization (in different sectors/regions) with an analysis of the best international practices (piloting one to two sectors/regions).
Assess the opportunity to (re)introduce preventive data-driven control based on automated risk indicators (including data from beneficiaries registry, asset declarations etc.). Open contracting data generated by MTender and structured according to the OCDS schema opens a lot of opportunities in this direction.
Amend the legal framework to be able to enforce violations. Those who offended the rules have to be punished, there should be no impunity!
Create even more transparency. All contracting authorities have to use MTender, even for low-value purchases.
Eliminate all inconsistencies of MTender with the provisions of the legal framework.
Finally, we suggest completing the development of the electronic procurement system in order to cover the full contracting cycle, from planning to contract implementation.
For us, it is obvious that making the e-procurement system transparent and publishing data are only the first steps of the open contracting journey. We, at IDIS Viitorul, want to make our procurement sphere open, accountable, efficient and corruption-free. That is why we are continuing our monitoring activities and offering our permanent support and expertise to our partners in government, civil society, and business.
“These are my contracts, and if you don’t like them… I have others,” reads the Todos Los Contratos website which scrutinizes Mexico’s public contracts, riffing off a joke by the comedian Groucho Marx. “Mexican authorities seem to follow this maxim when awarding contracts and reporting on them,” it continues, before explaining that around 1% of the country’s GDP is wasted due to corruption in procurement, according to conservative estimates.
PODER, the civil society organization behind the platform, is committed to bringing more transparency and accountability to contracting processes in Mexico, and has built an innovative solution for this purpose, centered around a single database of public contracts, which they created by aggregating millions of records from three different government sources. Covering almost MXN 30 trillion (US$1.5 trillion) worth of public spending, the database connects to several digital tools, which PODER developed to explain corruption and mismanagement in the Mexican procurement system. They relied on open contracting technology and free software solutions, and published detailed guidelines on how the software works to make it easy for journalists, programmers and analysts elsewhere to develop their own versions of the tools.
It’s an extraordinary approach, employed by an interdisciplinary team of developers, journalists and researchers, so it’s no surprise that the project recently won a Sigma Award, which recognizes outstanding data journalism from around the world.
PODER wanted to show that “it is possible to measure corruption based on public contracting data,” the team said. “We are starting to see the possibility of one day no longer relying on corruption perceptions surveys.”
Creating the database was a painstaking process that involved cleaning data on four million public contracts awarded by the federal government from 2001 and 2019, and standardizing them according to the Open Contracting Data Standard format.
The website TodosLosContratos.mx (“All The Contracts” in Spanish) answers important questions about Mexico’s procurement with explainers and simple tables outlining how the procurement process works, along with typical “bad” practices and a ranking of procuring entities scored using algorithms, while the search engine QuiénEsQuién.wiki and an API offer access to the full database.
PODER says the project has simplified the work of conducting investigations about public contracts, with at least a dozen Mexican and international news outlets (such as El Universal, Proceso, and EMEEQUIS) using the tools for their own reporting. It has also encouraged greater transparency in public contracting — three government agencies approached PODER with an interest in improving or uploading new data to the platform and the parties are discussing how improvements to their open data strategies could feed back into the civil society organization’s tools.
Ordinary people are better informed about public procurement too. PODER says visits to the QuiénEsQuién.wiki platform are on the rise and every week they receive messages from people who have concerns or questions about contracts or their participants.
Key technical features:
Data Import: an importer and web scraper orchestrator were developed based on the free software Apache NiFi. This modular software offers a simple setting for reusable components like the data cleaning module or the data update module.
Platform and API: QuiénEsQuién.Wiki is based on a mongoDB+node.js, all the data is hosted in a Kubernetes cluster of MongoDB databases and then exposed through a public API which is documented both in Spanish and English. Plus a model client in node js is usable with the NPM package registry. The website consumes the API and is compatible with desktop, tablets and mobile devices.
Data analysis: In order to fine tune the parameters of the algorithmic analysis engine we have combed through the data with the help of Kibana, an open source data visualization dashboard based on the ElasticSearch database engine, which helped us to quickly recognize patterns and detect deviations.
Data visualization: Our data is nicely presented using custom designed web-based interactive graphs and maps using primarily the D3.js library.
The European procurement market is a huge beast. In 2017, the total value of public purchases of goods and services across the European Union was estimated to be a whopping €2 trillion (or about 13% of GDP). Although the EU took significant steps in 2016 to open up the procurement market, it still recognizes that a lot of work remains to be done to level the playing field so that small and medium-sized enterprises (SMEs) can compete with big business for tenders in a fair and transparent fashion.
This transparency principle, a key element of a good democracy, demands that data related to taxpayers’ money should be as open as possible. Open data is good for everyone: only an economy that is open and fair can provide opportunities for new and small businesses. This is incredibly important as the best innovations often start small! Being life-long transparency advocates ourselves, we align with the OCP’s mission ‘to use the power of open data to save governments money and time and deliver better goods and services for citizens’. This is a big job which requires tackling the issue from multiple angles.
One way of bringing some well-needed transparency into the procurement market is to visualize who is doing what, when and how: where do public funds go and how are stakeholders connected? How have these networks evolved over time? Presenting this data in a concise manner is not just a way to uncover unrealized business opportunities but also helps detect patterns of collusion and corruption. Data visualization is a fantastic approach to make sense of data and to provide better insight into the complex world of procurement. But no one size fits all and for practitioners who work in the fields of open data, procurement and transparency it is important to sift through the good, the bad and the ugly.
So what are the principles of a good data visualization strategy? Here are some basics that any open data advocate and procurement professional should consider:
Be clear about what you want to communicate
If you work in the area of procurement, you’re likely to sit on tons of heterogeneous datasets. Your data is probably numerical (e.g. tender volume), related to time or categorical (e.g. sector of spending), and spatial (e.g. geographical coordinates). Not all of this data will be equally important for all of your stakeholders to see. And not all of your data will require visualization. The first step in getting your visualization strategy right is to think carefully about what users will want to see and what insight you wish to get across in the most efficient way.
Once you’ve worked out what data you want your stakeholders to consume, you should think of some appropriate analyses, statistics, indicators and aggregates that best summarize and communicate your data. You may wish to provide intel on how information about contract dates, volumes, etc. has evolved over time, or present a comparison between bidders and suppliers, or evaluate some KPIs, or show how data is geographically distributed. We recommend thinking about this carefully at the design stage. And consider how these variables hang together. A good data visualization strategy is clear about what it wants to achieve. Here are some examples of how a simple and clear message could be communicated visually:
Think about every element of your visualization design – it matters
Data visualizations are designed to make it easy to compare data or tell a story – both of which can help users to understand a topic and make decisions (the Trifecta checkup provides a general framework to evaluate if the visualization works). This includes selecting a chart type that fits your data and the message you want to convey with and about your data. You will want to think about the graphical elements, who would see it and where (e.g. on a desktop browser or on small screens, such as a mobile device). But you will also need to think about the type of font you use, if you use icons or not, the colour choices and if they are readable but also if they convey a message. You should consider the axes and scales you use in the chart and how to label them. The issue of legends and annotations and whether people can actually read and make sense of what you created is important too. As is thinking about whether your users could face constraints, such as physical ones (e.g. colour blindness) or in terms of skills (e.g. avoid technical terms) or of bandwidth and screen size (e.g. avoid large charts that require scrolling or interactive visualizations that don’t convey the message when it does not load fast enough).
Avoid chart junk
Chart junk describes all visual elements in charts and graphs that are not necessary to comprehend the information on the graph, or that will otherwise distract the viewer from this information. This means avoiding graphical excesses that don’t ‘respect’ the data and the audience. The visual layout should help to deliver the message and should contain the minimum amount of elements on a chart necessary to achieve this. So when you have chosen the data, the message and a chart type, review the visualization and see if you can reduce unnecessary lines, background images or elements. You can use custom styles and shapes to make data easier to understand at a glance, in ways that suit users’ needs and context. Explain your charts, bars and indicators well, e.g. by labelling your axes clearly and to the right amount of detail – for instance by providing units of measurement. Think about the language your audience is most likely to understand and don’t forget that communicating things as easily as possible includes the largest range of users to consume your content. Avoiding unnecessary complex words is also a way to reduce chart junk.
Design a dashboard if your story is more complex
If you want to display a series of multiple charts you can design a dashboard. As we said before, you should keep your visualization as simple as possible. That’s why separate but connected charts can sometimes better tell a story than one complex chart. Dashboards are often used to monitor changes in data (for more guidance on dashboard design, see Stephen Few’s Information Dashboard Design). A good dashboard can contain different types of charts, such as bars, time series or indicators presented in an aesthetically pleasing way that strikes the right balance between keeping things tidy and simple but providing valuable insight (see https://datavizproject.com/ for an overview of the data visualizations commonly used for this purpose).
This means on a dashboard you’ll need to arrange your content in a meaningful and effective way. Prioritize the most important information using layout structures. Research that tracks how users navigate on a website has revealed that people associate different levels of relevance and importance depending on where content is located on a page. Users tend to pay more attention to the top-left and centre sections of a screen; your most important visualizations should go here. Material in the bottom-right corner, however, is often considered less relevant. Also make sure that items that are logically close to one another are in proximity on display too, e.g. avoid spreading spatial data all across the screen.
Here are some examples of dashboards for procurement data:
A lot more can be said about data visualization and we have compiled more detailed guidelines under https://theybuyforyou.eu/visualizationtool/guideline.html. In the next post we’ll provide an overview of the TheyBuyForYou tool that we’ve been working on as an example of how we have taken these guidelines to heart when working with procurement data.
Elena Simperl is Professor of Computer Science at King’s College London. She leads several projects on TheyBuyForYou, an EU-funded research consortium where she and her team built a cross-European knowledge graph of procurement data alongside a series of interfaces and tools that enable SMEs to realize new procurement opportunities. Elena is also the Principal Investigator of the Data Stories project that develops novel ways for citizens to engage with data in a post-truth society.
Laura Koesten is a researcher at King’s College London. She was a Marie Curie Skłodowska fellow at the Open Data Institute and at the University of Southampton, UK investigating the user perspective in dataset search. She is part of the TheyBuyForYou and the Data Stories project. In her research she is looking at ways to improve Human Data Interaction by studying sensemaking with data, data reuse and collaboration and human factors in data science.
This is part of a new series of country data reviews looking at progress in publishing and using open contracting data by our partners. We will focus on: i) what the goal of using the data is; ii) what progress has been made so far; and iii) what’s working and what the challenges have been.
This blog looks at Afghanistan’s efforts to implement the OCDS to tackle corruption. We take a deep dive into the data available on the public procurement agency’s new open contracting platform, which is one of the government’s key tools in its efforts to reform the public procurement system.
In a very challenging context, Afghanistan has begun publishing Open Contracting Data Standard (OCDS) data to its new public procurement portal AGEOPS. The initiative is part of a campaign to tackle corruption in public procurement, launched by President Ashraf Ghani after he was elected in 2014. The National Procurement Authority (NPA) also has a dedicated team of open contracting data technicians in-house that regularly makes improvements to the portal and proactively engages with support providers at the Open Contracting Partnership and our helpdesk.
To understand how the OCDS data from AGEOPS could help support the goal of curbing corruption and other potential uses of the data, we downloaded the full dataset from AGEOPS in January 2020 and calculated some basic metrics on public contracting in Afghanistan. The data covers 6,952 contracting processes involving 71 procuring entities and 1,392 suppliers. Data is disclosed at the planning, contract and implementation stages and includes some information at all stages of the contracting process, to varying degrees of coverage. Much of the data relates to the implementation stage of the contracting process – publication of which is typically less common than tender and award information – and includes more than 28,000 milestones, 8,000 documents and 4,000 transactions. Note the slight variances from the metrics from the OCDS dataset and the live AGEOPS portal: the former is from a point in time, and the latter includes live data not yet exported as OCDS.
The public procurement challenges in Afghanistan can’t be overstated: rampant corruption, mafia-run companies, donor fatigue, and severe instability which affects project implementation, maintenance and monitoring. With very limited resources, the government must prioritize projects that have a realistic chance of success and develop transparent mechanisms for engaging with businesses and community groups. Since 2014, Afghanistan has introduced a series of procurement reforms, including the creation of the country’s first ever national procurement authority, to professionalize and modernize procurement processes and practices. Open contracting and open government commitments have been used as vehicles for delivering these reforms and tracking their implementation. We look at the influence these reforms are having on the public procurement system in a feature story here.
Progress in publishing open contracting data
Afghanistan has been publishing OCDS data since August 2018, with a focus on detecting corruption. In its open contracting commitments, the country pledged to set up a dedicated procurement portal. So far, the procurement agency has upgraded its existing system, the Procurement Management Information System (PMIS), to publish OCDS data via a new portal, the Afghanistan Government Electronic Open Procurement System (AGEOPS). This portal allows users to search, download and visualize procurement contracts, applying the Open Contracting Partnership’s releases and records model. It also publishes a change history of the contracting processes so that any modifications or new steps in the process are visible. A dedicated AGEOPS OCDS portal provides individual downloads of releases and records, bulk monthly downloads and an API.
Data availability & quality
Please note: While Afghanistan is continuously working to improve the quality and completeness of its open contracting data, we cannot make any guarantees as to the accuracy of the information in this data analysis; any findings should be verified with the National Procurement Authority.
The data contains information on 3,185 contracts with start dates ranging between October 2007 and January 2020 and end dates ranging between September 2006 and September 2023. This discrepancy between the earliest start and end dates suggests that there are some issues with data quality or completeness for contract periods in the data.
Buyers & Suppliers
The data contains 71 unique procuring entity names. Most procuring entities are identified using an identifier drawn from the Afghanistan Chart of Accounts register (AF-COA), however some procuring entities do not have an identifier. Some procuring entities also share the same identifier, in which case one may be a department or unit of the other.
These are the contracting processes for the top 10 procuring entities in the data, showing the وزارت دفاع ملی or National Defence Ministry, as the largest procurer:
The data contains details of 1,392 suppliers. Most suppliers are identified using an organization identifier drawn from the Afghanistan Tax Identification Number (AF-TIN) registers, however there are some examples of invalid identifiers (e.g. AF-TIN-0, AF-TIN-), which suggests organization identifiers may not be validated at source against the register.
The top supplier with 53 awards (out of a total of 3,362 awards) is the Afghan Telecom Company.
The data on contract implementation includes more than 28,000 milestones, 8,000 documents and 4,000 transactions. Providing data on implementation transactions allows data users to track both fine-grained and aggregated details of the spending transactions made against contracts. As a fine-grained example, the chart below shows the cumulative value of transactions and the total value of a contract between the Department of Finance and the Shiladiyeh Societe Enk for advisory services for supervision of Kabul highway reconstruction at Jalalabad with an ID of MPW/1527/ADB/QCBS.
Note: We found that transactions between 2014 and May 2017 for this contract appeared to be duplicated. Duplicates (based on date and transaction value) were removed to produce the above analysis.
Aggregated implementation transactions give a view of the scale of contracting processes captured in OCDS. Here is a summary of the published data on transactions for 3 currencies, AFN, USD and EUR, converted to USD:
Note: For convenience, we converted all currencies to USD at the current exchange rate (January 2020), however for more accurate results currency conversion should be based on the date of each transaction. Duplicates (based on date and transaction value) were removed to produce the above analysis.
We noted a number of transactions in the data where the currency was not stated, so these were excluded from the conversion. Omitting the currency from the data means that it is not possible to get a complete picture of spending.
Recently, NPA has been working with the OCDS Helpdesk, engaging with questions and clarifications, to map more fine-grained detail on implementation milestones using the OCDS metrics extension [https://extensions.open-contracting.org/en/extensions/metrics/master/]. This will enable users to understand the planned and actual progress of contract implementation, both in terms of physical and financial progress.
Afghanistan’s OCDS data includes documents as well as data: 16,150 documents attached to the contract section, and 8,112 documents attached to the implementation section.
Publishing documents as well as data is important because they allow end-users to perform detailed scrutiny of contracting processes, which is particularly relevant for detecting fraud and corruption, a priority goal for Afghan stakeholders.
What can’t we tell from the data yet?
Although data on 6,952 tenders was published, only the following fields are provided:
This limits some uses of the data, for example, it isn’t possible to understand what procurement method was used or how many bidders a tender attracted. Afghanistan is also planning to expand the awards section data in the future.
Finer-grained detail about which items are being purchased is not available since Afghanistan does not yet provide item-level details in tender, award or contract stages using an item classification scheme. This detail would allow end-users to calculate what types of goods, works, and services the government spends money on, as shown in our recent blog on Buenos Aires. However, we should note that Buenos Aires uses a local classification scheme, which supports comparisons between different contracting processes within the city, but not with data from other jurisdictions/publishers. For Afghanistan, as can be seen from the implementation section, more data on milestones such as date modified, title and description would provide further opportunities for analysis.
The system uses a mix of legacy and new electronic government procurement (eGP) modules to cover the entire procurement cycle, which does have its challenges; for example, a lack of shared identifiers (unique ID numbers for each contracting process) across modules makes linking up data more difficult.
Finally, while organization identifiers are provided for all organizations, there is room for improvement in how these are modeled in some cases, for example by ensuring that all award suppliers are listed in the parties array in compiled releases.
The NPA is constantly working on upgrades to the system that we hope will address these issues and will continuously improve the quality and completeness of Afghanistan’s open contracting data.