Bitcoin, lemons and the reputation problem in public procurement

During the 3rd European Conference last week I tweeted this:

Crazy idea: What if we had a blockchain type of technology to keep track of procurement contracts?May 19, 2015

The more I think about this idea, the more I see merit in its farfetched possibilities. First, a technology primer. The biggest innovation in Bitcoin is the blockchain. This small piece of technology contains all the transactions ever done in that "platform". The currency bit is the least interesting bit (pun intended). To understand the untapped potential of the blockchain read this.

So what is the blockchain's importance for public procurement then?

Well, in public procurement we have a reputation problem, ie there is limited information readily available in the market to signal the trustworthiness of both suppliers and contracting authorities. We usually think about reputation issues affecting suppliers, hence the preference in some countries like the UK for detailed selection stages so that the contracting authority may be confident of each supplier trustworthiness. Yes, there has been some work done in that front in recent years to reduce the burden on suppliers but in my view that work does not address the crux of the problem: the selection stage exists to establish trustworthiness. That's its job to be done (well, that and screwing up the weaker suppliers). The European Procurement Single Document and other similar measures do not address this concern, at least not directly.

And the lemons?

This reputation problem is akin to what Akerlof described in the 1970s as the "lemon problem": on a market with limited information like the used car market there is an asymmetry between the information sellers and buyers have. The first has more information than the second, forcing the second to assume all second cars available in the market to be lemons and to discount whatever products are on offer. Public procurement suffers from exactly the same problem - suppliers (sellers) have information about their ability to do a project, information the contracting authorities (buyers) lack. Ergo, those damn long PQQs to establish trust and for the contracting authority to feel confident that it can get into a contract with the supplier.

However, the current selection system leaves contracting authorities only slightly better off. They have access only to the information that is provided by the supplier and other than confirming it there is little they can do without investing significant resources (ie, someone's time and by definition money).

At the same time, suppliers have the same trust/information issue: is this contracting authority reliable? Do they pay on time? Do they haggle afterwards? However, suppliers have no clear way to check out contracting authorities other than what is available on the grapevine.

How do we solve this conundrum? Enter the blockchain.

What if each and every public contract awarded in the EU was entered into the public procurement equivalent of the blockchain? What if it included a public feedback mechanism just like eBay does for its buyers and sellers? (See this paper if you want to dig into this)?

A ledger that included contract information *and* public feedback from both sides of the transaction. Yes I know people will cry foul at the prospect of this information being made public, but there is a reason why AirBnB, TripAdvisor, eBay, Alibaba and others work so well for cross-border transactions: they solve the damn trustworthiness problem. More about this here (gated).

The only way this would work is if:

i) the technology (ie the blockchain) is simple, automated and kept up to date without depending on a central service (which is the problem the blockchain solved beautifully);

ii) mandatory for both sides of the transaction, ie both have to lodge their bit in the blockchain, perhaps as part of the payment process;

iii) all contracts across the EU go into it;

The technology bit is "easy" (famous last words, particularly for someone without a technical background like myself - so it will not be easy at all) it is the middle bit that I find more difficult. However, Portugal provides a good example of an analogous situation. To make contracting authorities comply with the need to register all contracts awarded on a central repository, it made payments without that information an offense and presto compliance went through the roof. The stick approach could work here as well. However, I think that carrot bit would work as well: I get pestered by sellers on eBay and home owners on AirBnB to leave feedback, although my compliance is not what it used to be.

Can I see this happening anytime soon? No, but I am willing to invest sometime going through the drawbacks and thinking about solutions for the practical problems it would raise.

PS: Information asymmetry is behind another of public procurement big, white elephants - long term contracts (PPPs, concessions) particularly in natural monopolies. Like water.