European public procurement: the system designed to be fair that ended up being slow
How procurement rules meant to ensure transparency systematically disadvantage speed, startups, and innovation. The structural barrier hiding in plain sight.
By VastBlue Editorial · 2026-03-26 · 15 min read
Series: The Chessboard · Episode 10
The €2 trillion machine nobody talks about
There is a number that should dominate every conversation about European competitiveness but almost never appears in them. The number is €2 trillion. That is the approximate annual value of public procurement across the European Union — the aggregate spending by national governments, regional authorities, municipalities, public utilities, hospitals, universities, and state-owned enterprises on everything from office supplies to fighter jets, from IT systems to highway construction, from consulting services to school lunches. It represents roughly 14 percent of EU GDP. It is, by a considerable margin, the single largest source of demand in the European economy.
To put this in perspective: the Inflation Reduction Act, which consumed months of transatlantic debate and reshaped European industrial policy discussions, commits $369 billion over a decade. European public procurement spends that amount every two months. The EU's entire cohesion policy budget for 2021-2027 — €392 billion — equals roughly ten weeks of procurement activity. When people ask where the money is in Europe, the answer is not venture capital (too small), not the European Investment Bank (important but insufficient), not even the EU budget itself. The answer is procurement. And the rules governing how that money is spent are among the most consequential, least understood, and most actively counterproductive pieces of economic architecture in the Western world.
The story of European procurement is a story about good intentions that calcified into institutional pathology. The EU's procurement directives — first codified in the 1970s, substantially reformed in 2004, and revised again in 2014 — were designed to solve a real and serious problem: the tendency of national governments to award contracts to domestic companies, often through opaque processes that rewarded political connections rather than competence. French ministries bought from French suppliers. German municipalities hired German contractors. Italian regional governments awarded contracts to firms whose principal qualification was proximity to the relevant political party. The result was a fragmented market where public money was poorly spent, taxpayers were overcharged, and the single market — the EU's foundational economic achievement — stopped at the procurement office door.
The directives changed this. They imposed transparency requirements, mandatory publication of contract notices above certain thresholds, standardised procedures, non-discrimination obligations, and judicial review mechanisms that gave unsuccessful bidders the right to challenge awards. The transformation was genuine. Cross-border procurement increased. Prices fell in sectors where competition intensified. Corruption, while not eliminated, became harder to sustain in the open. The European Court of Justice built a substantial body of case law enforcing the principles of equal treatment and transparency. By the standards of the problem they were designed to solve, the procurement directives worked.
The question that European policymakers have been reluctant to confront is whether a system designed for fairness in 1971 remains fit for purpose in 2026 — a world where technological change moves faster than procurement cycles, where the companies building the most innovative solutions are too small to navigate the compliance burden, and where Europe's strategic competitors use public purchasing as an instrument of industrial policy rather than an exercise in procedural correctness.
The anatomy of delay
To understand why European procurement is slow, you need to understand what happens when a public body decides it needs something. Not in theory — in practice. The process that follows is a masterclass in well-intentioned procedural accumulation, each step individually defensible, collectively paralysing.
Consider a mid-sized European city that wants to deploy an AI-powered traffic management system. The technology exists. Multiple companies — some large, some startups — offer solutions. The city's transport department has identified the need, assessed the options informally, and concluded that a modern system could reduce congestion by 15-20 percent and cut transport-related emissions significantly. In a private-sector context, the procurement process might take three to six months: issue an RFP, evaluate proposals, negotiate terms, sign a contract, begin implementation.
In the EU public procurement framework, the timeline looks different. First, the contracting authority must determine the applicable procedure. For contracts above the EU threshold — currently €143,000 for central government supplies and services, €221,000 for sub-central authorities — the full directive applies. The city must choose between an open procedure (anyone can bid), a restricted procedure (a pre-qualification stage narrows the field), a competitive dialogue (for complex contracts where the authority cannot define the technical specification in advance), or an innovation partnership (designed for R&D procurement but rarely used due to legal uncertainty). Each procedure has different rules, different timelines, and different legal risks. The choice itself can be challenged.
Assume the city chooses an open procedure — the most common route. It must publish a contract notice in Tenders Electronic Daily (TED), the EU's official procurement journal. The notice must comply with standardised forms specifying the subject matter, technical requirements, selection criteria, award criteria, estimated value, and a host of administrative details. The minimum time limit for receipt of tenders is 35 days from the date the notice is sent to TED — reduced to 30 days if tenders can be submitted electronically, or to 15 days in cases of "duly substantiated urgency" (a standard so high that invoking it invites legal challenge). In practice, most authorities allow 40-60 days to avoid any suggestion of insufficient notice.
While the tender period runs, the contracting authority must respond to clarification requests from potential bidders — and must share all clarifications with all bidders simultaneously, to maintain equal treatment. In technology procurement, where specifications are often ambiguous and evolving, the clarification process can become a secondary negotiation that consumes weeks and generates hundreds of pages of correspondence. Some authorities answer questions with such caution — terrified of inadvertently advantaging one bidder — that their responses are functionally useless, forcing bidders to make assumptions that introduce risk into their proposals and price into their bids.
Once tenders are received, the evaluation begins. The directive requires that contracts be awarded on the basis of the "most economically advantageous tender" (MEAT), which can include price, quality, technical merit, environmental characteristics, social considerations, innovation, and lifecycle costs. In principle, this is sensible — it prevents the race to the bottom that pure price competition produces. In practice, MEAT criteria introduce enormous evaluative complexity. Each criterion must be defined in advance with sufficient precision to be objectively assessable. Each must be weighted. Evaluation panels must score each tender against each criterion, document their reasoning, and be prepared to defend their assessment in court if a losing bidder challenges the award.
The evaluation phase typically takes 60-90 days for complex technology contracts. After evaluation, the authority must notify all bidders of the decision and observe a mandatory "standstill period" — ten days for electronic notification, fifteen for postal — during which unsuccessful bidders can request a debrief and, if dissatisfied, file a legal challenge. The standstill period is a safeguard against arbitrary awards. It is also a window during which the entire process can be frozen by a single dissatisfied bidder filing for interim relief. In some member states, legal challenges are so common that contracting authorities build an additional three to six months into their timelines as a contingency for litigation.
Add it up. From the initial decision to procure to the signing of a contract, a well-managed, unchallenged, moderately complex technology procurement in the EU takes 180 to 350 days. If challenged, add six to eighteen months. If the challenge succeeds and the process must be re-run, start the clock again. The AI traffic management system that the city's transport department identified as urgently needed in January 2025 might, if everything goes smoothly, begin implementation in early 2026. If a challenge intervenes, 2027 is more realistic. The technology, meanwhile, has moved on. The startup that offered the most innovative solution eighteen months ago may have pivoted, been acquired, or run out of runway.
European procurement does not fail because its individual safeguards are unreasonable. It fails because it layers reasonable safeguards upon reasonable safeguards until the cumulative weight makes speed structurally impossible. The system is not broken. It is functioning exactly as designed — and that is the problem.
Editorial observation
The startup exclusion machine
If procurement timelines disadvantage everyone, they are lethal for startups and SMEs. The numbers tell a story that procurement reform advocates have been reciting for a decade without sufficient effect. Small and medium-sized enterprises account for 99 percent of all businesses in the EU and approximately 65 percent of private-sector employment. Their share of public procurement contracts, by value, is estimated at 30-45 percent across the EU — a figure that sounds respectable until you consider that it includes subcontracting (where the SME does the work but a large integrator holds the contract and captures the margin) and that it varies wildly by member state and sector. In technology procurement — the area where innovative startups are most likely to offer genuinely superior solutions — SME participation is significantly lower.
The barriers are structural, not attitudinal. Most procurement officials would prefer more competition and more innovation. They are constrained by a system that makes selecting innovative, unproven solutions professionally dangerous and selecting established, proven ones professionally safe.
Consider turnover requirements. It is standard practice for EU procurement notices to require bidders to demonstrate annual turnover of two to three times the contract value. For a €5 million IT contract, that means the bidder must show €10-15 million in annual revenue. This requirement eliminates most startups by definition — a three-year-old company with a superior product and €2 million in revenue cannot bid, regardless of its technology, its team, or its references. The 2014 directive explicitly states that turnover requirements should not exceed twice the contract value, but compliance is inconsistent, and even the directive's own ceiling is high enough to exclude the vast majority of young companies.
Then there are the track record requirements. Procurement notices routinely demand evidence of three to five comparable projects completed within the last three to five years. For a startup bringing a genuinely new approach — by definition, one that no one has deployed before — this requirement is a paradox: you cannot win a public contract without a track record, and you cannot build a track record without winning a public contract. The directive allows "equivalent" experience to be considered, but evaluators, conscious of audit risk and legal challenge, overwhelmingly prefer the safety of demonstrated identical performance. Innovation, by its nature, has no identical precedent.
The compliance burden itself functions as a barrier. A typical EU procurement tender for a technology contract requires the submission of a European Single Procurement Document (ESPD), financial statements for the past three years, evidence of professional registration, proof of insurance, health and safety certifications, environmental management certifications, quality management certifications, data protection compliance documentation, social responsibility statements, tax compliance certificates, and — for above-threshold contracts — a statement confirming the absence of grounds for exclusion under Article 57 of the directive, which covers criminal convictions, tax offences, professional misconduct, and a dozen other categories. Each document must be current, correctly formatted, and — in cross-border tenders — translated and, in some cases, apostilled.
For a large systems integrator with a dedicated bid team, the compliance burden is a cost of doing business — absorbed into the overhead that public-sector clients ultimately pay for. For a 20-person startup, the compliance burden is a strategic decision: divert two to three engineers from product development for four to six weeks to prepare a bid that has, statistically, a 15-25 percent chance of success, or focus those resources on the product and the private-sector clients who can sign a contract in two weeks. The rational choice, for most startups, is to avoid public procurement entirely. And they do — which is why Europe's most innovative companies build products for the private sector while Europe's public sector buys from the same large integrators it bought from twenty years ago.
The innovation paradox: how procurement rules penalise the new
The 2014 reform of the EU procurement directives was, in part, an explicit attempt to make procurement more innovation-friendly. It introduced the Innovation Partnership procedure — a mechanism allowing contracting authorities to procure R&D and the resulting product in a single process, without the legal requirement to re-tender at the production stage. It expanded the use of preliminary market consultations, allowing authorities to engage with potential suppliers before launching a formal procedure. It encouraged the division of contracts into lots to facilitate SME participation. It introduced lifecycle costing as an explicit award criterion, theoretically enabling authorities to value long-term innovation over short-term price.
The results, a decade later, are modest. Innovation Partnerships have been used sparingly — a few hundred across the entire EU since 2014, compared to millions of procurement procedures conducted in the same period. Preliminary market consultations, while more common, create their own legal anxieties: how much contact with potential suppliers is permitted before it constitutes an unfair advantage? How does an authority demonstrate that its specifications were not shaped by a particular supplier's capabilities? The risk of legal challenge deters precisely the kind of exploratory, open-ended engagement with the market that innovation requires.
The deeper problem is that innovation procurement requires something that the directive framework fundamentally discourages: risk tolerance. When a contracting authority procures an established product from an established supplier, the risks are known and manageable. The product works. The supplier will deliver. If something goes wrong, there is a track record to point to, a contract to enforce, a large company that will still exist next year. When an authority procures an innovative solution from a young company, every one of these certainties evaporates. The product may not work as promised. The company may not survive the contract period. If something goes wrong, the procurement official who championed the innovative choice faces audit scrutiny, media criticism, and potential legal liability.
The incentive structure is devastating in its clarity. A procurement official who selects an established but mediocre solution and gets mediocre results faces no consequences. A procurement official who selects an innovative but risky solution and gets superior results receives modest recognition. A procurement official who selects an innovative but risky solution and gets poor results faces career consequences. The rational behaviour, given these incentives, is to select the safe option. And so European public procurement systematically selects the safe option, year after year, contract after contract, €2 trillion at a time.
A procurement official who buys the wrong innovative solution gets fired. A procurement official who buys the wrong established solution gets sympathy. That asymmetry, more than any directive or regulation, explains why European public procurement does not innovate.
Editorial observation
The contrast with the United States is instructive. The US federal procurement system is not a model of efficiency — it has its own pathologies, its own delays, its own incumbent advantages. But it has one mechanism that the EU lacks: Other Transaction Authority (OTA). OTAs allow US defence and civilian agencies to award contracts for prototype development and follow-on production outside the standard Federal Acquisition Regulation (FAR) framework. They can be awarded quickly — sometimes in weeks rather than months. They do not require the full compliance apparatus of a standard federal contract. They can be used with non-traditional contractors, including startups that have never held a government contract. The Department of Defense used OTAs to award over $50 billion in contracts between 2016 and 2023, including to companies like Anduril, SpaceX, Palantir, and Shield AI — companies that have fundamentally reshaped defence technology precisely because they were given a pathway into the government market that did not require them to become bureaucracies before they could serve one.
The Small Business Innovation Research (SBIR) programme is another instrument without a true European equivalent. SBIR mandates that US federal agencies with extramural R&D budgets exceeding $100 million set aside 3.2 percent for small business research contracts. The programme has channelled over $50 billion to small companies since its inception in 1982. Companies like Qualcomm, Symantec, and iRobot received early SBIR funding. The programme is not perfect — it has been criticised for creating "SBIR mills" (companies that subsist on grants without ever commercialising) — but it provides a structured pipeline from small-company innovation to government adoption that the EU has struggled to replicate.
The defence procurement gap
Nowhere is the cost of procurement dysfunction more strategically consequential than in defence. The European defence market is fragmented across 27 national procurement systems, each with its own defence industrial base, its own political imperatives, its own interpretation of Article 346 of the Treaty on the Functioning of the European Union (which allows member states to exempt defence procurement from single market rules when "essential security interests" are at stake). Article 346 was designed as a narrow exception. In practice, it has become a broad exemption — a legal basis for maintaining national defence procurement as a protected domain where competition, transparency, and efficiency are subordinated to sovereign industrial policy.
The consequences are measurable. Europe collectively spends approximately €240 billion per year on defence — a figure that has risen sharply since Russia's full-scale invasion of Ukraine in 2022. Yet European defence procurement produces significantly less capability per euro spent than American defence procurement. The EU operates 178 different weapons systems; the United States operates 30. Europe has 17 different main battle tank designs; the US has one (the M1 Abrams). Europe has 29 different destroyer and frigate classes; the US has four. This is not diversity in the sense of operational flexibility. It is fragmentation in the sense of duplicated development costs, incompatible logistics systems, reduced interoperability, and production runs too short to achieve economies of scale.
The European Defence Fund (EDF), established in 2021 with a budget of €7.9 billion for 2021-2027, was designed to incentivise collaborative procurement and joint development. It has funded promising projects in areas like drone technology, space surveillance, and cyber defence. But €7.9 billion over seven years is roughly one percent of total European defence spending over the same period. The fund can catalyse collaboration at the margins. It cannot restructure the fundamental dynamic in which each member state's defence ministry procures primarily from its own national champions, using Article 346 to insulate those purchases from competition.
The contrast with the speed at which Ukraine — a country at war — has procured and deployed new technologies is painful for European procurement professionals. Ukraine has adopted commercial drones for military use, deployed AI-powered targeting systems, integrated Starlink communications infrastructure, and fielded autonomous maritime vehicles, often moving from identification of a need to operational deployment in weeks rather than years. The urgency of combat compressed procurement cycles that peacetime bureaucracies stretch across fiscal years. This is not a recommendation that European procurement adopt wartime shortcuts — it is an observation that the gap between what is procedurally required and what is operationally possible has become so wide that it constitutes a strategic vulnerability.
The proposed European Defence Industrial Strategy (EDIS), announced in March 2024, set a target of 40 percent of defence equipment being procured collaboratively by 2030 — up from approximately 18 percent in 2023. The ambition is laudable. The mechanism — voluntary coordination among sovereign states, each with domestic political incentives to protect national industries — has a track record that suggests the target is aspirational rather than achievable. European defence procurement reform has been discussed in Brussels since the establishment of the European Defence Agency in 2004. Two decades later, the fundamental structure — 27 separate buyers, each prioritising national suppliers, each using Article 346 as a shield — remains substantially unchanged.
What reform actually looks like
It would be dishonest to diagnose the problem without examining the solutions — both those that have been proposed and those that have been implemented. Procurement reform is not a new idea in Brussels. It is a recurring theme in competitiveness reports, European Council conclusions, and Commission communications. The Draghi Report of September 2024 identified procurement reform as a key lever for European competitiveness. The Letta Report on the single market made similar recommendations. The question is not whether reform is needed — that consensus exists. The question is what kind of reform is possible within the EU's institutional architecture.
Some reforms are already underway. The European Commission has promoted the use of "innovation-friendly" procurement practices through guidance, training programmes, and the Innovation Procurement Observatory. Several member states have established innovation procurement agencies or units: the UK's Small Business Research Initiative (SBRI) — modelled on the American SBIR — has been operational since 2009 and has funded over 4,000 contracts with innovative companies, though its scale remains modest. France's Direction Générale de l'Armement (DGA) has experimented with "flash" procurement processes for defence innovation, reducing timelines to weeks for small-scale prototype contracts. The Netherlands and Denmark have implemented below-threshold procurement portals that simplify procedures for smaller contracts.
The most ambitious proposal on the table is the so-called "Buy European Act" — a concept that would create preference mechanisms for European suppliers in public procurement, analogous to the Buy American Act that gives domestic suppliers a price preference of up to 20-30 percent in US federal procurement. The proposal is politically potent: it appeals to industrial sovereignty narratives, domestic employment concerns, and the widely shared frustration that European procurement rules are more open to foreign competition than those of any major trading partner. The WTO Government Procurement Agreement (GPA) — to which the EU is a signatory — commits its members to open their procurement markets to other GPA signatories. But China is not a GPA member. The practical effect is that European companies face significant barriers accessing Chinese public procurement while Chinese companies can, in principle, compete for European public contracts on equal terms.
The International Procurement Instrument (IPI), which entered into force in August 2022, gave the Commission limited power to restrict access to EU procurement for bidders from countries that do not offer reciprocal access to European companies. It is a step toward levelling the playing field, but a cautious one — the IPI has been invoked sparingly, and its mechanisms are designed to encourage negotiation rather than impose exclusion. The gap between the IPI's measured approach and the aggressive domestic preference mechanisms used by the United States, China, India, and other major economies remains wide.
Europe runs the most open procurement market among major economies — a principled position that becomes a strategic liability when no other major economy reciprocates. Fairness that is not mutual is not fairness. It is unilateral disarmament.
Editorial observation
The deeper reform challenge is not legislative but cultural. European procurement culture — shaped by decades of audit scrutiny, legal challenge, and political scandal — is structurally risk-averse. Changing the rules is necessary but insufficient if the people applying the rules continue to optimise for process compliance rather than outcome quality. This requires investment in procurement professionalism — better training, higher status, clearer career pathways, and — critically — protection for officials who take well-reasoned risks that do not succeed. The private sector accepts that a certain percentage of innovation investments will fail. The public sector, under current accountability frameworks, accepts no such thing. Until it does, procurement reform will change the rules without changing the behaviour.
The model to watch may be Estonia — a country that appears repeatedly in this series for good reason. Estonia's e-procurement system processes virtually all public procurement electronically, with standardised templates, automated compliance checks, and radically simplified procedures for below-threshold contracts. The system reduces administrative burden for both authorities and bidders, shortens timelines, and generates data that enables real-time monitoring of procurement performance. Estonia processes procurement at a speed and cost that larger member states — with their layered bureaucracies and legacy IT systems — struggle to match. The technology to modernise European procurement exists. The political will to deploy it at scale has, so far, been the binding constraint.
The structural barrier hiding in plain sight
European public procurement is not a policy failure in the conventional sense. It is a success that has become a liability. The directives achieved what they set out to achieve — open, transparent, non-discriminatory public purchasing across a single market of 450 million people and 27 sovereign states. That is a genuine accomplishment, one that no other regional bloc has replicated. The World Trade Organization, when it discusses procurement transparency, uses European practice as the benchmark.
But benchmarks from 2004 are not benchmarks for 2026. The world in which the current procurement framework was designed is not the world in which it now operates. In 2004, the strategic question was whether French and German governments would buy from each other's companies. In 2026, the strategic question is whether European governments can buy innovative technology fast enough to remain competitive with blocs that are spending faster, building faster, and deploying faster — and whether the companies developing the most innovative solutions can survive long enough to navigate the procurement labyrinth.
The €2 trillion annual procurement spend is Europe's largest lever for industrial policy. It dwarfs any subsidy programme, any structural fund, any innovation grant. If even a modest percentage of that spend were redirected toward innovative European companies — through simplified procedures, reduced compliance burdens, innovation-specific procurement tracks, and a cultural shift toward outcome-based rather than process-based accountability — the effect on European competitiveness would be transformative. Not hypothetically transformative. Measurably, concretely, immediately transformative.
The barrier is not ignorance. Every stakeholder — from procurement officials to startup founders, from Commission bureaucrats to national defence ministries — understands the problem. The barrier is institutional inertia, reinforced by legitimate concerns about corruption, political interference, and the potential abuse of simplified procedures. These concerns are real. Corruption in public procurement costs the EU an estimated €120 billion per year. Any reform that weakens safeguards against corruption is worse than the disease it treats.
The task, then, is not to choose between fairness and speed. It is to build systems that deliver both — procurement frameworks that are transparent and fast, accountable and innovative, open and strategically intelligent. This is difficult. It requires institutional design of a sophistication that matches the complexity of the problem. It requires political courage to challenge procurement cultures that have optimised for risk avoidance over outcome quality. And it requires an honest acknowledgment that Europe's current procurement system, for all its procedural elegance, is failing the very competitiveness objectives that the single market was built to serve.
The system designed to be fair ended up being slow. The challenge for the next decade is to make it both fair and fast — before the cost of slowness becomes a price that European competitiveness can no longer afford to pay.
Sources
- European Commission — Public Procurement Indicators 2022 — https://single-market-scoreboard.ec.europa.eu/policy_areas/public-procurement_en
- Directive 2014/24/EU — Public Procurement Directive — https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=celex%3A32014L0024
- Mario Draghi — The Future of European Competitiveness (September 2024) — https://commission.europa.eu/topics/strengthening-european-competitiveness/eu-competitiveness-looking-ahead_en
- European Defence Agency — Defence Data 2022-2023 — https://eda.europa.eu/publications-and-data/defence-data
- European Commission — Innovation Procurement Observatory — https://digital-strategy.ec.europa.eu/en/policies/innovation-procurement
- International Procurement Instrument (IPI) — Regulation (EU) 2022/1031 — https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=CELEX%3A32022R1031
- US Government Accountability Office — Other Transaction Authority Usage (2023) — https://www.gao.gov/products/gao-20-84
- OECD — Government at a Glance 2023: Public Procurement — https://www.oecd-ilibrary.org/governance/government-at-a-glance-2023_3d5c5d31-en