No government activity is more susceptible to corruption than public procurement. The process by which government decides what to buy and from whom is lengthy, technically complex, and riddled with decision points that give procurement officers enormous discretion. Oversight is thus especially difficult. Moreover, because so much money is involved, the temptations procurement offers corrupt public servants and their private sector accomplices are particularly great. Some developed countries spend as much as one-third of their budget on the purchase of public works, goods, and services, and the available data suggests the figure may even be higher in developing countries.
With so much at risk, it is no wonder that there has been an explosion of material on fighting corruption in public procurement. The OECD, the World Bank, the European Union, and Transparency International have each issued a slew of publications on how to prevent corruption in public procurement, and the latest edition of Matthew’s anticorruption bibliography catalogues more than 100 articles, pamphlets, and books by academics, think tanks, and advocacy organizations on corruption in public procurement. Indeed, the amount of material is so overwhelming that reporters, civil society groups, and even parliamentarians and government auditors may be discouraged from pursuing allegations of procurement corruption. For how does one know where to start?
Below are ten simple, straightforward questions that provide a starting-point. They can be asked about any government purchase: from “off-the-shelf” items like pencils and paper to a new road or bridge to the acquisition of customized IT systems. The answers will provide a telling first indication of whether the procurement merits further scrutiny.
1) Was a cost-benefit analysis of the procurement prepared? No individual or firm rushes out to buy a product or service without considering whether there is a need and whether there are alternatives available. Likewise, no government should commit public funds before determining if on balance the purchase is worthwhile. While for lower value, simple purchases, a short memo or spread sheet analysis will suffice, for larger more complex projects a full-blown cost-benefit analysis should exist that considers different ways to accomplish the project’s goals. If the goal is to reduce the costs of transporting goods between point A and point B, did planners analyze the costs of: i) improving the existing road, ii) constructing a new one; or iii) building a rail line between the two. If the project was chosen because of corruption, likely either no analysis was done or it was an after-the-fact justification that will be easy to see through.
2) Was the purchase put out for competitive bid? Competition is the surest way to ensure government gets the best product at the lowest price, and the UN model procurement law, the basis of virtually every national procurement statute now in force, makes competition the default option for government purchases. There will be times when an exception is appropriate: a national emergency, the existence of only one supplier who can meet the terms of reference. But anytime competition is bypassed, the procurement merits heightened scrutiny.
3) How long were potential bidders given to prepare their bids? Where the supplier is chosen by competition, the firms competing for the contract should be given sufficient time to prepare their tenders. The length will depend upon a variety of factor: the complexity of the procurement, the extent of subcontracting anticipated, and the time needed for transmitting tenders. Nigerian law provides a minimum of six weeks for large construction projects, and for World Bank-financed consultancy contracts consulting firms must be given at least 14 days to submit an expression of interest. If the time to prepare the bid is too short, qualified firms will either decline to participate or will be unable to prepare a quality response. At the same time, the corrupt firm will likely be given a copy of the tender long before it is made public, giving it plenty of time to develop a responsive bid.
4) Was the request for bids advertised? For a competitive tender to produce the best product at the lowest price, the request for bids or tenders (the terms are frequently used interchangeably) must be widely disseminated,not just e-mailed to a few cronies. That will ensure all firms that might be able to meet the government’s needs have a chance to respond. A notice should appear in at least one national newspaper with a large circulation, and for contracts for large public works and other purchases where foreign firms are potential suppliers, at least one major international publication as well.
5) Were the bids screened for signs of collusion? As I explain here, a variety of methods are available for analyzing whether the bids show signs two or more bidders agreed with one another on the amount of the bid each would submit. A screen should be employed before the winning bidder is selected, but it can also be used after as well and certainly should be on large public works projects where collusion is particularly common.
6) Did those on the committee that evaluated the bids submit a statement disclosing their financial interests? One way corruption can creep into the procurement process is when one or more of those evaluating the bids has a financial interest in one of the bidders. Submission of a financial disclosure statement (that is then verified) is an important preventive measure. The European Union’s antifraud office (OLAF) has a fine example of a model disclosure form that covers the interests that evaluators should be required to reveal.
7) Who is the beneficial owner or owners of the winning bidder? A recent investigation in the Solomon Islands revealed what is all too common in many countries: the senior civil servant responsible for approving procurement contracts secretly owned a company that was awarded a number of contracts. The firm winning the bid should be required to disclose the name of any individual who owns or controls five percent or more of the firm. The disclosure should be verified.
8) Does the request for bids contain an audit clause? An audit, conducted before or after the contract is awarded, is a powerful tool for uncovering wrongdoing, from bid rigging through false or inflated invoices to the delivery of substandard equipment or material. Failure to include one in the bid documents and resulting contract not only makes it very difficult to investigate corruption allegations but may suggest that those preparing the bid documents or the contract are a party to corruption. The World Bank requires a provision be inserted in works contracts it finances giving it the right “to inspect [bidders] accounts and records and other documents relating to the bid submission and contract performance.”
9) How many technical audits will be conducted during contract implementation? For contracts for roads, bridges, IT systems, and other purchases that require months, if not years, to complete it is essential that the contractor be regularly audited to ensure the work meets the specifications in the contract. Once the road is built it can be very hard to determine if the subsurface is as deep as it should be. With a technical audit an independent engineer or other professional inspects the job while it is under construction. Do the layers beneath the surface of the road contain the amount of asphalt and the size of aggregate specified in the contract? Does the IT system have the security features the contract requires to prevent hacking? A leading authority on corruption in public works, engineer A.L.M. Ameer, recommends that in construction projects government spend $1 for every $1,000 budgeted for construction on technical audits.
10) Does the winning bidder have an ethics or anticorruption compliance program? It makes sense in today’s environment that companies instill a sense of ethics in their employees, and many countries require companies doing business with them or subject to their jurisdiction to have a program in place to prevent their employees from paying bribes. There is now an international standard against which to measure such programs. Does the winning bidder have any such program, and if not, why not?
The ways in which corruption can sneak into procurement are limited only by the imaginations of greedy, crooked public servants and their cronies in the private sector, and the only way to be sure it is kept at bay is to build a professional, honest corps of public procurement professionals. In the meantime, “wrong” answers to the questions above are likely to expose the more egregious procurement corruption schemes.
GAB | The Global Anticorruption Blog 
Thank you for this blog Rick, I agree with your eight questions. Unfortunately research conducted by TI’s Defence & Security Program found little evidence that countries other than the USA require a winning bidder have an ethics or anticorruption compliance program for procurements in high-risk sectors such as defence. Source: “Disseminating Best Practice on Supply-Chain Management: Does the government formally require that the main contractor ensures subsidiaries and sub-contractors adopt anti-corruption programmes, and is there evidence that this is enforced?” July 2016, http://ti-defence.org/wp-content/uploads/2016/07/160728-EU-Commission-Defence-Directive-Evaluation-Paper.pdf
Thanks for flagging the paper on best practice in defence (or defense) industry supply chain management. It is a fine analysis and begs for a follow up (hint, hint) that does the same type of comparative analysis of government prevention policies across procurements of all kinds.
Thanks too for confirming what I suspicioned — that the U.S. stands alone in requiring firms that do business with it to have an anticorruption compliance program. I hope to change this, through GAB posts making the case why it is in governments’ interest, particularly those of developing countries with weak domestic enforcement agencies, to impose such a requirement. I would appreciate readers flagging any cases they know of where a government has imposed such a requirement. (Extra credit if there is any sign whatsoever that anything written on this blog has contributed to the decision.)
The 2010 UK Antibribery Act, which several governments have now copied, achieves the same end by giving corporations subject to UK law a strong incentive to establish a compliance program. The act provides that if an individual “associated” with a corporation is caught paying a bribe, the corporation is guilty of the offense of “failure to prevent bribery” unless it can show that it “had in place adequate procedures” to prevent bribery. I know at least one reader of this blog considers this a “corporate get-out-of-jail-free card,” but if properly enforced it will do the same job as the U.S. rule. Indeed, it reaches even further as it applies whether the corporation does business with the government or not.
Let’s see whether the enforcement of such laws over the next few years proves them to be “get-out-of-jail-free” cards or an important adjunct to government enforcement efforts.
Dear Rick, thanks for your comment, TI is currently in the process of creating an expert procurement working group that will research and publish a follow up comparative analysis of government prevention policies across defence and non-defence procurement, though it will be global in scope and not just limited to EU states this time. I would really like to make contact with you and discuss this further if you have time? My email is eva.anderson@transparency.org.uk
Re the Bribery Act s7 failure to prevent offence and being a “get-out-of-jail-free” card, I’ve not seen this so far. What we have seen is good use of s7 in convictions, guilty pleas and settlements/Deferred Prosecution Agreements (DPAs) since the law was enacted under the very active leadership of David Green QC at the SFO. This would indicate that companies and their lawyers feel it is actually fairly robust, though not yet fully tested. Last week senior members of the Serious Fraud Office reported in conversation that they have seen a significant rise in the number of self-reports by companies, particularly since the DPA with ROLLS-ROYCE. This is positive news and a good trend, as long as DPAs set a high bar for cooperation, remediation and compensation to victims of corruption.
In an earlier post, you explained how one can use statistical analysis to screen for collusion among bidders. That seems like a fairly easy, technical solution–I could imagine that a statistician and programmer could join forces and build some software that would catch likely instances of collusion reading data from an excel spreadsheet. Which, if any, of these other 10 issues do you think lend themselves to fairly simple technical fixes?
The questions are simple yes/no ones designed to make it easy for reporters, civil society groups, and others to zero in on whether a particular procurement deserves closer scrutiny. One might keep track of the answers over time to see whether any pattern emerges, say a particular agency rarely if ever using competive tendering or when it does setting an unusually short time period for responding to the tender.
One analysis closely related to the questions that I would very much like to see is a comparison of winning bids against the government’s estimate over time. For public works contracts in particular. On large works contracts government engineers prepare an estimate of what they think it will cost to do the work. Assuming the engineers are competent, in a competitive market one would expect some bids to come in below the estimate. Yet in several developing countries I have been told that rarely if ever does is a bid below the estimate submitted. This is a very strong indicator that collusion is rampant.
An example what such an analyis can show appears in box 4 of “Curbing Fraud, Corruption, and Collusion in the Roads Sector,” http://documents.worldbank.org/curated/en/975181468151765134/Curbing-fraud-corruption-and-collusion-in-the-roads-sector). For 46 World Bank-financed roads contracts in an Eastern European country the winning bid is compared to the engineer’s estimate. On all 46 the winning bid equales the estimate. Chances of this happening if the market were competitive? One in ten trillion plus or minus.
I’m curious who you think are the most appropriate institutional actors to ask these questions, e.g. an independent government accountability office, agency IGs, legislative actors, etc. Is procurement oversight most effectively accomplished by having investigative authority concentrated in one dedicated office or is a multi-stakeholder, multi-layered set-up more likely to catch wrongdoing? And is there general consensus on what sort of appeals process, if any, should be available to non-winning bidders? I would imagine in addition to the difficult of proving as a factual matter the existence of corrupt bidding practices, there are also structural impediments to raising challenges to contract awards.
There are many actors who could ask the questions. I had in mind the press (as my wife is an editor for a organization with reporters in 20 plus developing countries). But civil society groups too. For legislators and auditors the questions might help them identify priorities for further investigations. Procurement is so critical and so vulnerable to corruption at so points in the cycle, the more asking questions, the better.
Daniel Gordon wrote a nice piece on bid protest systems in 2006, “Constructing a Bid Protest Process: Choices Every Procurement Challenge System Must Make,” that provides a nice summary of bid protest mechanisms. It is readily available on the web.
Pingback: Exposing Procurement Corruption: Ten Questions to Ask | Matthews' Blog
Pingback: Questions To Ask In Exposing Procurement Corruption – EProcure
Pingback: Exposing Procurement Corruption: Ten Questions to Ask – CEGET