Procurement fraud costs the UK £258 trillion annually*

Of course it doesn't. That's a made-up number. While it is deliberately ridiculous, it's there to illustrate a point. And that is - it's dangerous to present numbers as facts when they are not.

Here's the headline from Supply Management last week.

Not "might cost.." or "is estimated to cost..." It is very clear.  It "costs".

The UK public sector lost £2.3 billion last year to fraudulent purchasing practices such as bid-rigging, price-fixing, the submission of multiple invoices for the same work and diverted payments.

The Annual Fraud Indicator 2012, published by the National Fraud Authority (NFA), found procurement fraud made up the largest proportion of the overall total lost. It cost central government £1.4 billion and local government £890 million.

But when you actually look at the Annual Fraud Indicator report, it's clear that the £2.3 billion number is an estimate applied to another estimate then extrapolated (possibly wrongly) across central and local government. And being really cynical, the NFA have a vested interest in making the problem look as big as possible - that sustains their livelihood after all.

So the report takes total public sector procurement spend at £230M (estimate number one). It then applies a 1% figure that seems to largely come from an estimate that is used by the MOD, and is quoted with no evidence as to why it is seen as a validated estimate -

A one per cent ‘at risk’ figure used by the Ministry of Defence Police to estimate procurement fraud within their defence budget has been applied to this spend figure...

There's also some general evidence from a survey of procurement professionals - but not anything that stands up to scrutiny in terms of putting a figure to fraud losses. To be fair, the NFS report does talk about the numbers being an estimate - but Supply Management doesn't. It presents the £2.3 billion as fact. Which it isn't.

Back to the report.  For some reason I can't fathom, it then allocates the £2.3 billion as £1.4 billion central government, £890 million local government. I can't see why central is so much bigger than local - but where do health, emergency services, education etc. sit within that classification?

I have no doubt that fraud is a significant problem. I'd like to see more invested in seeking it out and understanding how to stop it. But pretending that we know exactly how much it costs us seems pointless, and of course estimates like these quickly - and wrongly - become quoted as scientific fact.

*Spend Matters estimate using the Hungarian Gamma-Cubed statistical methodology

First Voice

  1. RJ:

    I was recently involved in a tender for fraud risks consultancy. One of the bidders presented as “fact” that procurement fraud would be costing between 8 and 12% of total spend. Since the organisation I was working for spends several £bn each year that’s quite a figure!

    I have seen fraud and theft occur in several of the organisations where I have worked over a 20+ year career, the largest involving the theft of several million pounds of engineering spares and a fraud where £250K from non-existent was diverted into a senior manager’s bank account. I also acknowledge that global firms may be working in cultures and political environments where fraud and corruption are more endemic.

    Nevertheless to start quoting values of £100s of millions or billions of losses with little or no evidence to back up these figures is scaremongering of the worst kind. And as for journalists hyping up the situation, well…

    When asked to justify the numbers quoted, my potential supplier fell back on “industry standard assumptions”. They did not win the business.

Discuss this:

Your email address will not be published. Required fields are marked *