A Scandinavian cautionary tale on lot division and distortions of competition for public contracts (E-3/16)

I have recently written a case comment on the EFTA Court's Judgment of 22 December 2016 in the case Ski Taxi SA, Follo Taxi SA og Ski Follo Taxidrift AS v Staten v/Konkurransetilsynet, E-03/16, which will soon be published in the Journal of European Competition Law & Practice.

The case concerned an instance of joint tendering by two competing taxi companies and it is interesting from a competition law perspective because the EFTA Court treated the joint bid as an anticompetitive price-fixing agreement by object (which limits the need to assess its effects in the market).

Thinking about the case from a public procurement perspective, I think that it also offers a cautionary tale about the restrictions of competition that can derive from decisions on the division of a single procurement into lots. This second perspective is the focus of this post.

The relevant facts of the case are as follows. In 2010, Oslo University Hospital (OUH) ran a public procurement tender for the award of framework agreements for the provision of patient transport services. The object of the tender was divided into nine geographical lots, which related to different catchment areas in the vicinity of the hospital.

For two of those lots, OUH only received a single tender, which was jointly submitted by two taxi companies that OUH would have expected to compete for the contracts. In view of this situation, which OUH interpreted as a privately-created restriction of competition for those lots (and thus insufficient to enable it to obtain value for money), it decided to cancel the procedure for those two lots. It also reported the joint bidders to the Norwegian Competition Authority,which eventually led to the imposition of fines for a price-fixing agreement (as discussed in the case comment).

OUH then launched a new tender procedure. In this occasion, OUH redesigned the geographical coverage and divided the object of the procurement in five areas instead of two. Interestingly, the taxi companies that submitted the joint bid in the previous tender also submitted a joint bid for all five lots, as did two competing taxi companies. OUH eventually entered framework agreements with all three companies, and assigned the joint tenderers second priority in all five areas.

On reflection (and hindsight, of course), it seems plausible that the limited competition in the first round of procurement derived not solely from the decision to submit a joint bid by two of the taxi companies active in those areas, but probably also from the decision to create too broad geographical catchment in the initial design of the lots. If the re-run of the procurement on the basis of smaller geographical lots attracted more competition (while still not changing the strategy of the tenderers that decided to bid jointly), it seems clear that the design of the object of the procurement is key in the prevention (or creation) of publicly-initiated restrictions of competition. From that perspective, more thought (and more market intelligence) is needed if the design of the procurement process is not to result in insufficient competition and thus limit the opportunities for the contracting authority to obtain value for money without distorting competitive trends in the market.

In the specific case, if the single bid for two of the lots had been submitted by a single taxi company (or if an anticompetitive agreement between the joint bidders consisted in an allocation of lots rather than a price-centered strategy), OUH may not have been able to spot the existence of any problems, but it may still have suffered the consequences of the limited competition for the contract that derived from the design of the procurement.

Overall, then, I think that this case offers a valid cautionary tale for contracting authorities regarding the need to make more extensive use of market intelligence and to approach lot division with a more competition-oriented mindset.