Does the government get ripped-off when buying goods and services?


In brief:

  • The government typically pays 4% less than retail price for IT equipment.
  • But it often pays far more than this, pointing towards flaws in the procurement process.

Most government contracts are bespoke. That makes it hard to assess whether the government pays a fair price for the goods and services it procures. As there isn’t a retail market for replacement nuclear submarine parts, for example, we don’t know if the government is getting a fair price for its new periscopes.

But, occasionally, the government does purchase goods which have a retail market: new laptops, phones, monitors and other IT equipment for its staff. We’ve looked at how much the government pays for these purchases to get a sense for whether they generally get a fair price.

The chart below summarises the results. We’ve identified 50 different contracts for the bulk purchase of IT goods, and then compared the price paid to their retail price. For example, say the government pays £38,000 for 100 Apple iPhone SEs (2020). That is £380 per phone, 95% of the retail price of £400. And so that contract will appear in the 95% segment of the chart below. As it is getting a slight discount, we classify that as a “good deal”. Any contract where the government pays more than retail price is coloured red and labelled a “bad deal”.

Typically the government gets a slight discount on retail price, saving around 4%. That’s to be expected: bulk purchases normally attract a discount.

But there are multiple examples of the government paying more than retail price. It’s possible that in these instances the government has purchased a higher-specification machine than we’ve assumed (e.g. buying laptops with 16 GB RAM rather than a standard 8 GB). But this seems unlikely: if such requirements are not specified in the contract details, it would not be in suppliers’ interest to provide more expensive-goods.

An alternative – less generous – explanation is that the procurement process is flawed. IT equipment is often purchased through “framework agreements”. That means government departments run “mini-competitions” where only pre-approved “framework” providers can bid for the contract. In a closed-competition, its plausible that prices can rise above those seen in the (competitive) retail market.

Ideally we’d have looked at more than just 50 contracts. But unfortunately government agencies often provide insufficient about the IT equipment they’re purchasing for us to identify the right retail price. If the contract simply says “iPhone”, that could be a relatively cheap iPhone SE or the much more expensive iPhone 12 Pro. And so our analysis only includes contracts which provide enough detail about the purchased equipment’s specification, so that we can find the right retail price.

But this sample does suggest that the government regularly pays more than retail price for goods that it procures. That is problematic: particularly in the context of