So the PFI has been booked in for some much-needed surgery. For years it has been getting more unwieldy, more expensive and less attractive to the private sector. Finally, Gordon Brown is to do something to save an essential method of upgrading public services.

The proposals are contained in a Treasury paper accompanying this week's budget. No doubt the devil is in the detail, and as this document contains 128 pages of diabolical prose, published as Building went to press, there's no doubt that it will take us some time to appreciate its full significance. But the overarching message has to be welcome to this industry. First of all because it reaffirms the government's commitment to the PFI; just as well, given that there are £26bn worth of projects in the pipeline. The paper also acknowledges that the PFI has been successful, but could be more so.

The PFI, it says, will continue to account for 10-15% of public sector investment. But it has to be more efficient and more flexible. And it must do more than deliver hospitals or schools on time and to budget; they must be better designed, cleaned and catered for than their traditionally procured counterparts. Equally importantly, it acknowledges the problems that have occurred with PFI have not been a fault of the system. Rather the PFI has highlighted the weaknesses of public sector procurement in general. For example, there is a lack of training and understanding among public sector clients - a long-standing bugbear for the private sector professionals - so the paper proposes that public sector skills be bolstered with an operational taskforce, new models of secondments and the possibility of training civil servants for careers as clients. All of this is long overdue.

But the document also demands that the private sector make changes to the way it operates. One point it makes is that there should be an increase in competition between equity providers - so that those in a consortium don't automatically get to provide the funding. Obviously, this will set the cat among the pigeons. And then there is the interesting suggestion about the creation of "project delivery organisations". This involves putting together an organisation that will act as a kind of souped-up project manager that may also take an equity stake in the project. It's not too dissimilar to the Local Education Partnerships being put together for schools, and the aim is to get PFI providers to concentrate on their long-term service, rather than the short-term profit to be made from the initial delivery.

Finally, it addresses two controversial matters. The first is the struggle for control of the design process presently being fought between architects, who want a design in place before a consortium is appointed, and the contractors, who want the opposite. Here the Treasury's attitude to the architects' case has changed from hostile to agnostic: there will more consultation in due course. The second point is about the provision of soft services, which it says should not automatically be included in the initial PFI contract. Although the private sector hasn't done badly on running the finished building, it hasn't provided a great leap in performance either. Although Brown's line here may be a sop to the unions, it is up to the private sector to explain why it should get the gig. Time, perhaps, to unleash the private sector and allow it to rethink the whole process from first principles.