Complain about one arithmetic error and drop from second to fourth: how a re-examination in Bastogne turned against the complainant
Euro Bureautic rightly pointed out to the city of Bastogne an arithmetic error of nearly 2,900 euros in its price assessment, but the re-examination that followed did not stop at that error: the successful bidder Jacli in turn flagged a mistake in the technical analysis, the city redid every technical score on the basis of data sheets found online instead of the tenders, and Euro Bureautic fell from second to fourth place — whereupon the Council of State found that the contested decision had been replaced and adjourned the case sine die.
What happened?
The city of Bastogne launched a general call for tenders for a supplies contract: the long-term rental over sixty months of 23 multifunction photocopiers, rental and maintenance included, at a global price. The needs were spread over eight categories of devices — seven initially, raised to eight by a corrigendum to the specifications concerning technical specifications and taxes. Seven companies tendered: Océ Belgium, Ricoh, Xerox NG Partners, EBS-Eicher Bureau Service, Jacli, Xerox Eurobureautic and Konica Minolta. All seven were found regular. The award criteria were price excluding VAT (35 points), technical capabilities of the equipment and compliance with the models requested (25 points), service and assistance (30 points) and ecology (10 points). On that last criterion every bidder received the maximum, because their ecological qualities were, according to the city, so close. On 24 June 2011 the municipal college awarded the contract to Jacli for 154,248.60 euros. The final standings were exceptionally tight: Jacli 89.98 points, Xerox Eurobureautic 89.78 and Ricoh 89.58. Of those three the city chose Jacli, and the reasoning is striking: Jacli was, according to the college, the only one offering a clear solution in the event of an extension of the copier fleet during the sixty months, the municipality had worked with Jacli ‘for several years to complete satisfaction’, and its equipment and service had already proved themselves. The decision was notified to Jacli by registered letter on 27 June 2011; the applicant received its notification the same day. On 1 July 2011 its counsel wrote that a mistake had been made: for the price criterion the city had used a figure of 35,579.76 euros, whereas page 35 of its tender stated 32,713.71 euros. It should therefore have received 27 points instead of 24.78, which mathematically placed it first with 92 %. Counsel asked the college not to notify and to withdraw the deliberation of 24 June, announced an extreme-urgency application, and indicated that a withdrawal and a fresh award to his client would ‘of course’ end the litigation. That same 1 July the college suspended its own decision and informed both Jacli and counsel. On 5 July the mayor asked for one more day; the college would meet the following afternoon. The reply came on 7 July, and it turned out differently from what was hoped. The city acknowledged the arithmetic error: Euro Bureautic’s price score went from 24.78 to 28.07. But Jacli, once informed, had in turn reported an error in the technical analysis. The analysis report stated that the applicant’s colour model 2, a WorkCentre 7775, had a warm-up time complying with the specifications (maximum 30 seconds), whereas the device’s technical data sheet found online indicated under 130 seconds. The IT department therefore redid the entire analysis of the technical criteria, this time on the basis of the technical data sheets of the machines offered rather than the declarations in the tenders. The result was carnage: of the 25 points for the technical criterion, no one scored more than five. Only the applicant’s model 1 proved compliant; for Jacli, models 1 and 4 were. Euro Bureautic ended on 69.07 points and fourth place, Jacli on 73.98 and first. Jacli refused to have the arithmetic error revisited without also revisiting the technical assessment, and since the contract had already been notified to it, the college considered it could not award to another company unless Jacli admitted the error and withdrew — which it did not. The Council of State found that the file did not contain the text of the deliberation of 6 July, but that the letter of 7 July showed a fresh examination had indeed taken place, not limited to the error raised by the applicant, that the assessments on the technical criterion differed systematically from those of the first examination, and that it had led to a different ranking. A new decision had therefore been taken awarding the contract to Jacli on grounds different from those of the contested act, and that decision had substituted itself for the contested act — which the college had itself, moreover, ‘suspended’. The application for suspension had thereby become without object. But because that replacement decision could not immediately be regarded as definitive, the Council did not rule: it stayed the proceedings, adjourned the case sine die, had the judgment notified by fax and reserved the costs.
Why does this matter?
This judgment concerns a risk one is rarely warned about: the re-examination you provoke is not yours. Whoever points out an error to a contracting authority hopes for a correction of that one error. What he gets is a reopened file — and in a reopened file the competitor can put its own objections on the table. Here Euro Bureautic’s complaint was entirely correct: there was an arithmetic error of 2,866.05 euros in its favour, and the city acknowledged it. Yet it ended two places lower than where it started. The procedural lesson matters just as much. The Council of State does not rule on a decision that no longer exists. As soon as the authority, after a re-examination, takes a new decision on different grounds, that decision replaces the contested act and the application for suspension loses its object — even where the new decision leads to the same outcome. But the Council does not close the door here: because the replacement decision could not yet count as definitive — a written deliberation was even missing — it does not declare the application without object, but stays the proceedings, adjourns the case sine die and reserves the costs. That is a pragmatic solution to a real problem: an applicant obliged to start over with every new decision would never obtain a ruling. Two findings in the statement of facts deserve attention as well, even though the Council did not rule on them. The original reasoning relied in part on satisfaction with Jacli as the incumbent supplier — a consideration that did not appear as an award criterion in the specifications. And the second technical analysis replaced the declarations in the tenders with technical data sheets from the internet, after which every bidder’s technical score collapsed from twenty-odd to at most five out of twenty-five. A change of method on that scale in mid-procedure raises questions about equal treatment and about what an authority may re-correct; the judgment does not answer them, but it does show where they arise.
The lesson
Before you point out an error to a contracting authority, weigh what a re-examination could unleash across your whole file — not only on the point you raise. Check whether your own tender is verifiable on every technical point against the data sheets of the devices you offer, and not merely against what you declared in your tender; that is precisely where Euro Bureautic came unstuck. Frame your complaint as narrowly as possible and expressly request a correction of the identified point, not a wholesale reassessment. If you are already in proceedings and the authority changes its decision, do not cling to the old act: it has gone, and you will have to challenge the new one. As a contracting authority, the lesson runs differently. If you acknowledge an error, confine your re-examination to what you can justify and record your new decision in writing — here the deliberation of 6 July was missing from the file, which the Council expressly noted. If you change the method by which you assess an award criterion, realise that you are thereby reassessing every tender and that you must be able to explain that to everyone. And do not base your choice between three almost equally ranked tenders on satisfaction with your current supplier: that is not in your specifications and is not an award criterion.
Ask yourself
Before raising an arithmetic error, have you considered what a full re-examination of your own tender would produce — including on the criteria you are not complaining about? Do your technical declarations match the official data sheets of the devices you offer? Are you requesting a targeted correction, or inadvertently opening the whole file? Do you know that a new decision after re-examination replaces the contested act, so that your application for suspension loses its object and you must challenge the new decision? As a contracting authority: do you record every replacement decision in writing and with reasons? Can you justify changing the assessment method of an award criterion during the procedure, and do you treat all bidders equally in doing so? And does your choice between almost equal tenders rest solely on criteria set out in your specifications?
About this database
The Council of State (Raad van State / Conseil d'État) is Belgium's supreme administrative court. In disputes over public procurement — from contract awards to tenderer exclusions — the Council of State is the final arbiter. The rulings in this database are summarised by TenderWolf in plain language, with practical lessons for tenderers and contracting authorities. View all rulings →