BBC business editor Robert Peston on government waste
Sir Philip Green told me that if he ran his businesses the way the government does, the lights would go out.
What he says he's uncovered is a culture where there are grotesquely inadequate controls of spending on everything from mobile phones, to travel, to property and paper.
The budgets he looked at are worth £191bn in total - and he believes billions could be sliced off that, if the government did simple things like centralising purchasing and monitoring expenditure properly.
For example, he argues that the public sector is spending around £700m a year too much on telecoms alone.
It's the many smaller examples of waste that some will see as shocking, such as that on a single IT contract, 400 private sector employees are being hired at a daily rate of more than £1,000 per day, for tasks that Sir Philip was unable to identify.
He also highlights:
1) the existence of 71,000 central government buyers, who can spend up to £1,000 a month using "procurement cards" without any checks or oversight;
2) in general, the government permits spending of up to £1,000 "without monitoring or authorisation";
3) £104m a year is spent on printing, with some departments paying 11p per leaflet while others pay £1.31;
4) there is £551m of annual travel spending by central government, with very little co-ordination of the purchases;
5) even more is spent on travel by the rest of the public sector, but the precise amount was impossible for Sir Philip to obtain;
6) some £38m a year is spent on 400,000 overnight hotels in London alone, with prices per room varying from £77 to £117;
7) some £84m goes on office supplies, with prices paid for ink cartridges varying from £86 to £398 and on paper from £8 per box to £73;
8) £61m is spent on laptop and desktop computers, from 13 different IT services providers, with no standard specification across departments and prices paid per laptop varying from £353 to £2,000.
9) one provider takes 98% of the £21m a year spent on mobile phones, which has nonetheless negotiated 68 different contracts with government departments - implying that the government is not benefitting from any economies of scale;
10) £25bn a year is spent on property, but only 6% of the entire estate is overseen by a central team;
11) the government is poor at taking advantage of break clauses in property leases, to rationalise use of office space.
That is just a selection of the myriad examples where Sir Philip believes the government can make very substantial savings, perhaps as much as 50%, on £191bn of procurement and property costs.
He is keen to point out that the savings he is suggesting would not involve a single public-sector redundancy: all of the squeeze would be on private sector suppliers (who probably won't thank him).
Perhaps his most controversial recommendation - which he made in his interview with me but is not explicit about in the report - is that government departments pay their suppliers far too quickly.
He says that the norm in most departments is to pay suppliers in five days, compared with the standard 30 days payment period for most private-sector transactions, and 45 days demanded by some bigger companies (like his own).
If the government demanded a minimum of 30 days of credit from suppliers, it would save hundreds of millions of pounds in financing costs.
This would of course be painful for those suppliers, especially smaller companies.
The previous government made it a matter of explicit policy to pay suppliers as quickly as possible, to help them through the recession.
Sir Philip Green is saying that priorities should now change - and that the government should take advantage of its sheer size and credit-worthiness to extract the best possible terms from all those who sell to it (and never mind if they don't like it).
Update 1430: I've had a brief chat with Peter Gershon, who advised the previous government from 2000 to 2004 on how to make savings in procurement.
He says that he is supportive of Sir Philip Green's central argument, that there should be more centralisation of purchasing by the public sector (or what Gershon calls more "aggregation of demand").
He says that in his work for the then Chancellor of the Exchequer, Gordon Brown, he put in place systems to facilitate the centralisation of purchasing.
But Brown did not order departments to centralise purchases.
And nor was there any serious move to standardise specifications for goods and services bought by different departments.
Different parts of government retained considerable discretion to choose different kinds of computers, and stationery, and mobiles phones, among others.
Which of course meant that their individual orders were smaller, and therefore there was less scope to yield savings from placing huge orders.
The corollary is that taxpayers arguably ended up paying more than was necessary to equip the public sector.
You can keep up with the latest from business editor Robert Peston by visiting his blog on the BBC News website.