Winner concentration

Winner concentration is one of the indicators used to calculate the zIndex score for evaluating contracting authorities.

The winner concentration indicator measures whether a single bidder, or a very limited group of bidders, is awarded a significant portion of the authority's contracts.

Best practice guides note that „avoiding the concentration of key areas in the hands of a single individual is fundamental in the prevention of corruption“ (OECD (2009), p. 35).

A high winner concentration (which gives a low indicator rating) is usually caused by:

• One large contract being awarded in place of several smaller ones. A preference for larger contracts tends to distort market conditions, is more legally problematic (as major contracts are more prone to challenges in court) and does not always lead to cost savings. On the contrary, in limited market conditions it raises the risk of cartel arrangements and may increase costs. The general rule of thumb should be to prefer several smaller contracts to single large contracts (although this does not apply to purposeful splitting in order to achieve contracts so small they are below the legal limits for public tender). As the EU states in its Code of Best Practices Facilitating Access by SMEs to Public Procurement Contracts, splitting large contracts into smaller contracts not only enables smaller bidders to participate, but also raises competition level, to the contracting authority's benefit. For more on this topic, see here (in Czech).

• Unfair tendering processes, which discriminate or disqualify some bidders in favour of others.

• Technological or legal restrictions preventing changes of supplier. These may be inherently related to particular contracts, but more often than not they are the consequences of uneconomic conduct on the part of the contracting authority. This may include errors made in the subject matter specification (the contracting authority should generally specify its needs rather than a specific solution to them, which limits competition) or previous strategic mistakes (so-called vendor lock-in; for more on this topic see in this presentation (in Czech)).

• Exceptionally high quality tenders submitted by one or more bidders. This may occasionally happen, however most of the market is currently exposed to such a strong competition that maintaining a significant competitive advantage over several years proves very difficult for the majority of suppliers. If a single bidder wins contracts repeatedly, this is more probably caused by one of the reasons listed above.

The winner concentration indicator is complementary to the Bidder participation indicator. The number of tenders submitted for a given contract tells us about the number of competitors participating in the procedure, while the winner concentration indicator monitors which of those bidders is awarded the contract, checking for the occurrence of a single repeated winner (despite apparent competition).

Winner concentration is measured as the value of all contract(s) awarded to each individual bidder, divided by the total value of all contracts awarded by the respective contracting authority. Formula:

$$z_4 =\Big(\frac{(\sum\:value\:of\:all\:contracts)^2-\sum(value\:of\:contracts\:awarded\:to\:each\:bidder^2)}{\sum(value\:of\:all\:contracts\:^2)-(\sum\:value\:of\:all\:contracts)^2}\Big)^4 $$

The calculation is inspired by the standard Herfindahl index for measuring supplier concentration, adapted for the purpose of public procurement contracts. The formula has been modified in order not to discriminate smaller contracting authorities with lower number of contracts. The resulting index is then raised to the power of four in order to emphasize the difference between highly rated and lowly rated contracting authorities.

• If all the contracting authority's contracts were awarded to a single contractor, the indicator value equals 0.
• If each single contract has been awarded to different bidder, no matter its value, the indicator value equals 1.

This indicator puts minor contracting authorities that award fewer contracts at a disadvantage, because they have a more limited range of bidders. It should therefore be used only for the benchmarking of comparably sized contracting authorities.

Authorities whose previous management left a dubious legacy, having allowed a supplier to establish an irreplaceable position (so called vendor lock-in, see above) would also be unjustly penalized by this indicator, as would authorities who award contracts several times to the same supplier because the supplier's bids are genuinely of unmatched quality.

• Establish non-discriminatory criteria. Consider the usefulness of every qualification criterion you set, ensuring it does not restrict competition or prevent new bidders from entering the procedure.
• Communicate with potential bidders. Efficiency-minded contracting authorities should consult their intended investment projects prior to issuing calls for tender. Appropriate non-discriminatory ways of getting feedback from potential bidders include holding open day consultations and notifying bidders of prospective contracts through other means (implementing a machine-readable buyer profile, and accurately completing contract notice registration forms in the Journal - especially in terms of CPV codes, contract titles and descriptions).
• Split contracts. Split large projects into smaller lots (reasonably selected according to subject matter), so each lot may be separately contracted and awarded to the best bidder.
• Make long-term plans. If there is a risk of vendor lock-in (see above) – this can happen for example in the ICT sector – strategic long-term planning is essential. The contractual framework should be designed in such a way as to prevent the contracting authority from becoming dependent on the winning bidder.