Friday, January 27, 2012

Gaming the system - Tendering Law

Here's an issue that I hadn't thought of before, but that has potentially serious implications for my work:

My research on trains tends to assume that public tendering can be arranged with a reasonable amount of efficiency all around. It may not always be a good idea, but the process itself is fairly idiot-proof. This confidence was seriously hurt today when I read a ruling by the Court of Appeals in Den Bosch, the Netherlands. In this case, the plaintiff wanted a tender result annulled on the grounds that the winning company had violated the terms of the tender by - put simply - gaming the system.

This is how the city of Helmond screwed up: they defined the financial side of the evaluation criteria (350 points out of 1000) based on three sub-criteria:
  • one-off implementation costs: 150 points
  • total annual costs over a period of 5 years: 150 points
  • average hourly rate: 50 points
In each of these categories, the best offer was awarded maximum points, and the others were given fewer in proportion. Given this setup, the approach of the successful applicant was (or should have been) quite foreseeable: construct a bid that involves € 0,- implementation costs, thus assuring 150 points in that category, and stick the costs of implementation in with the ordinary annual costs instead. That way, the gain in category one will almost certainly exceed the loss in category two.

One of the losing bidders, Centric, sued, arguing that the city should have excluded this bidder because it was a "manipulative bid". The bidder and the city, on the other hand, argued that it was merely a "strategic bid", which was permitted. Ultimately, the Regional Court and the Court of Appeals sided with the city.

What fascinates me about this case isn't so much the ruling itself, which is based on the General Conditions of the tender and is therefore quite tedious (not to mention limited to these facts), but the sheer ineptitude of the city. Why did they not realise that this was the optimal strategy, given the formula chosen? Why did they not choose a formula based on total cost instead? If local authorities are this incompetent, perhaps it is better to let them run the local transport services in-house. That way, no one has much of an incentive to game the system, and if errors are still made, both the costs and the benefits of the error end up in the city's coffers.

Hat-tip: and Banning Advocaten (in Dutch)

No comments: