Contract Killings

An exclusive extract from my new book - out today!

The bulk of today’s post is an exclusive extract from my book “Failed State: Why Nothing Works and How We Fix It”, which is out today.

(Tomorrow, Friday 12th July, from 2-4pm, I will be at West End Lane Books in West Hampstead, signing and dedicating copies of Failed State. Do stop by if you're in the area!)

But before we get to that I wanted to welcome the many new subscribers who signed up during the election campaign. You’ve made us not just the biggest politics substack outside of the US, but the third biggest in the UK on any topic.

Hopefully new readers have clocked the basic idea of what we do. I write about British politics and policy – with occasional forays into Europe and America – and Lawrence (my father) writes about global conflicts and international relations.

I worked for government as an adviser a decade ago, and I’m now a Senior Fellow at the Institute for Government as well as writing for a wide variety of outlets. Dad is Emeritus Professor of War Studies at King’s College London, a department he helped to build up, and has written numerous books including the official history of the Falklands War, the bestselling “Strategy: A History”, and most recently, “Command: The politics of Military Operations from Korea to Ukraine”. He was also a member of the Iraq Inquiry.

During the election, with the support of a few brilliant helpers, I did daily posts. But our usual pace is two a week (normally one from each of us). The extra posts, though, got a great response. So we’re going to increase the volume of content, particularly from September, once the new government starts pumping out policy papers, and their first set of tax and spending plans. We will also cover the US election and its consequences for geopolitics. Plus there will be more guest articles and interviews.

Over the next few weeks I’ll be writing more on the composition of the new parliament, and what the backgrounds of MPs tells us about how Labour will govern and the Tories will react. There will also be a one-off special Olympics guide (because I’m an Olympics obsessive – we don’t usually cover sport). Dad will be writing, among other things, on the latest situation in Ukraine.

If you want access to all this a paid subscription costs £4.50 a month or £45 a year:

You can also get gift subscriptions here:

---

Failed State

Today I’m launching my new book. The core idea is that three big negative trends – centralisation, the loss of effective scrutiny, and changes to the media – have left our state far less effective. And that the combination of the three has been utterly toxic.

While individuals and ideologies make a difference, and a well-intentioned, competent, government will always be preferable to one containing Liz Truss, bad systems will always beat good people.

Our new government has the chance to fix those systems but that requires understanding what’s gone wrong and why over the last forty years.

The extract I’m sharing here is the opening section of chapter three, which looks at one of the worst consequences of centralisation: a growing reliance on a set of largely unaccountable and often poorly performing private companies to deliver taxpayer funded services. Writing this chapter left me in an almost permanent sense of fury at the misery caused by all the scandals I describe. Hopefully it gives you a taste of the scale of the problems I’ve tried to uncover throughout the book.

Contract Killings

Children in care are some of the most vulnerable people in our society. They have often experienced severe trauma in the first years of life, developed serious behavioural problems as a result, and need intensive specialist support. In most wealthy countries this is done by local government. Here, in at least one case, it is done by a former OnlyFans model and reality TV star, supported financially by her fiancée, a former pornographer.

Ampika Pickston, best known for her appearances on The Real Housewives of Cheshire, registered a care home in the summer of 2023. It was financed by her partner David Sullivan, who owns West Ham United football club and made his fortune from pornographic magazines and sex shops. Nearby councils were, according to one report, charged £10,000 per week to place children there.

In November 2023 Ofsted inspected the home and found major failings in a report that was later removed from their website. They suspended the home’s registration, and the management were not allowed to look after any children until they could prove they had fixed the problems. The suspension was lifted after an initial monitoring visit in January 2024 and one child was placed in the home.

But then a further Ofsted visit a few weeks later found that this child had been moved out almost immediately due to the home’s inability to provide adequate care. Ofsted wrote that the ‘child made an allegation following a physical intervention that they had suffered bruising and a knee injury, which required hospital attendance’. The home was suspended again.

It is an astonishing indictment of the British state that it no longer has the ability to provide care for those who need it most, and instead allows blatantly ill-qualified people to charge exorbitant fees to provide unacceptable levels of care. This is an extreme example but the care home market is startlingly dysfunctional. Three quarters of places are now provided by for-profit companies, many of them owned by private-equity firms who have seen an opportunity for big margins.

Councils, who have a statutory duty to provide support for children in care, are a captive market with few options. The amount they are spending on placements has risen by 72 per cent in the last five years, with an average weekly bill per child of £5,980. In many cases they are sending young people miles away from where they grew up because providers set up homes in towns where property is cheaper.

In a 2023 report the Competition and Markets Authority (CMA) found that the fifteen largest providers were making annual profits of 22.6 per cent on average. They also found that ‘some of the largest private providers are carrying very high levels of debt, creating a risk that disorderly failure of highly leveraged firms could disrupt the placements of children in care.’ An investigation in 2024 found hundreds of children had been put, illegally, in unregulated homes because of a lack of places.

The emergence of this care homes market is an example of one of the biggest changes in the way the British state, particularly in England, has changed over the past few decades. As state capacity has dwindled, central and local government has become ever more reliant on the private sector to cover the gap.

The theory was that they could provide services more efficiently due to competitive pressure. The reality has been very different.

The state we’re in

How did we end up with this unquestioned belief across Whitehall that the private sector can provide better-quality services at lower cost when, after forty years of practice, it is very clear that it often does not?

The topic is often debated in simplistic left/right terms – public good and private bad, or vice versa. Critiques from the left can often ignore the weaknesses in state-run provision and make unfounded assumptions that public employees automatically have high levels of intrinsic motivation. Anyone who has worked in the public sector knows that motivation varies across the workforce, as it does in any large organization.

But defenders of the extensive use of the private sector on the right (and sometimes the centre) often seem to have completely missed that much government business with private providers no longer conforms to anything resembling a functional market with properly aligned incentives.

The core principle of contracting out is that competition will lead to innovation and lower cost than doing it yourself. This is true enough when certain conditions are met. Firstly, there needs to be many organizations, public or private, who can provide the service to generate genuine competition on both price and quality. Secondly, it needs to be relatively easy to measure whether the service you want has been delivered to an acceptable standard. Thirdly, government needs to be able to hand over most of the risk of failure – if the taxpayer is still on the hook for picking up the costs if things go wrong then the premium paid to private firms is not justified.

If all these boxes are ticked, then it can work very well. Cleaning contracts are an example of something that typically meet all these criteria. It is a well-established service that many companies offer, and for which there is already a large private market. It is easy to check cleaning is happening to an acceptable standard. If it is not then another company can be found to do it instead without too much fuss.

The best evidence we have shows tendering cleaning services does save money, even if the in-house team ends up winning the tender. For more complex services, though, often none of these criteria apply. Take a specialist activity like running immigration processing. There is no existing market. That means there are few plausible bidders outside of a handful of huge multinational outsourcing firms whose sole purpose is to win new contracts, almost regardless of what they are for.

The ‘big four’  –  who dominate the UK market  –  are Serco, Capita, Atos and G4S. At the time of writing, Serco are responsible for, among other things, running: detention centres for asylum seekers; facilities management for several hospitals; cycle hire schemes in various cities; ferry services to the Orkney and Shetland islands; six prisons; our warning system for incoming nuclear missiles; programmes to help unemployed people back into work; and bin collection in several councils.

This is a shortened list, and one that does not take into account all the things they have run in the past. The other companies in the big four have similar lists. This is despite all of them being involved in major scandals involving failed contracts over the past decade. Successive governments have decided they do not have much choice but to keep giving them contracts –  regardless of success rates – because there now aren’t any alternatives.

David Laws was a Liberal Democrat minister, who worked closely with Nick Clegg during the Coalition years. In his diaries there is a revealing entry for 28 October 2013, when he was a minister in the Cabinet Office.

This was just after Serco and G4S had been involved in a scandal around charging the state for electronically tagging prisoners who had left the country or even died. The previous year G4S had been all over the press for messing up the Olympics security contract, and lying about it, forcing the army to step in at the last minute so the Games would go ahead. Capita had just screwed up a major contract to provide court translators, causing huge delays in bringing criminals to justice. Atos were in the doghouse over failures around assessing people for disability benefits.

Laws noted that, in a ministerial meeting, some cabinet ministers were worried about these companies bidding on upcoming contracts. But George Osborne ”said it would be incredibly damaging for government outsourcing and for the engagement of government in the private sector if we stopped giving contracts to these types of companies . . . the UK market is currently highly concentrated, with four main suppliers – Atos, Capita, Serco and G4S. Three of these now have big problems . . . Osborne and Philip Hammond pointed out that these companies would still be managing extremely sensitive contracts, including being part of the consortium operating the UK’s nuclear deterrent [so relations had to be maintained].”

So much for a competitive market.

Accurately measuring the delivery of a contract is an even bigger problem. The more complex a service is, the harder it is to measure whether it is being delivered properly. Take school inspections, which were partially outsourced in 2009 and then brought back in-house in 2015. You can tell if the inspections have happened but how do you assess whether they have made good decisions without repeating the inspections? And, indeed, when the schools inspectorate Ofsted took it back in-house one reason was that they wanted more control over quality.

If it is not possible to accurately assess quality then companies can ensure profit by cutting staff, paying them less, and providing a worse service, while staying within the rules. This happens all the time. Then there is the question of risk. If a company fails to deliver cleaning services to an acceptable level then it is usually manageable. The contract can be cancelled and another firm brought in at short notice. But if a prison or hospital fails completely that is going to be much harder to cope with. Firstly because the level of public concern will be much higher; secondly because it will be much harder, and usually impossible, to contract someone else within an acceptable time period.

For instance, Birmingham Prison was contracted to G4S in 2011. An inspection in 2018 found it in total disarray: ‘inspectors witnessed inmates being intimidated and prisoners squirting urine or throwing faeces through broken observation panels’. It was taken back into public control, initially temporarily, and then, because no one else wanted to take over the contract, permanently.

Similarly, Hinchingbrooke Hospital in Cambridgeshire was handed over to a private company called Circle in 2010, as a test case. It was a complete disaster, and it failed both to provide acceptable levels of care or save money. The Care Quality Commission rated it inadequate in 2014 and another government report found it was the least efficient hospital in the country. The management were so desperate that they launched a controversial scheme to, in effect, bribe GPs to send them patients. Again the hospital had to be taken back into public ownership at great expense.

The Public Accounts Committee of MPs noted that ‘the total deficit incurred during the franchise will be well above the level that Circle is contractually committed to cover, leaving the taxpayer to pick up the rest’. In neither case was risk taken away from the state. Unlike a normal business, a prison or hospital cannot be allowed to fail, so ultimately the taxpayer was left holding the bill.

At least, in these two examples, the state could step back in because the Ministry of Justice and the NHS still run most prisons and hospitals. If an entire service gets contracted out, and it fails, then you have the additional problem of there being no state capacity left to pick up the pieces. This means the proposed scheme just ends up being dropped altogether or, where that isn’t possible, the state has to scramble to rebuild capacity, causing more delays and generating more costs.

The most horrendous example of this in recent years was the failed privatization of a large part of the probation service – which is supposed to stop people who have left prison from reoffending. Up until 2013 it had been run, in various guises, as part of the public sector. At that point there were thirty-five regional probation trusts under the supervision of the National Offender Management Service, which was also responsible for prisons. They were working reasonably well and all were rated good or exceptional by the government.

Then Justice Secretary Chris Grayling, who is high up in the list of the least competent people to be given high office in British history, decided to partially outsource the service to the private sector.

The idea was that low- and medium-risk offenders would be looked after by ‘community rehabilitation centres’ that would be run by private companies. High-risk offenders would continue to be managed by the state, as Grayling was worried about the public reaction to high profile murderers and rapists being the responsibility of for-profit firms. 

The entire sector told Grayling it wouldn’t work. It failed to meet all three tests of sensible outsourcing: competitiveness; ability to measure; and risk management. The policy was not piloted, or built up carefully over time, it happened nationally and almost overnight. Probation officers were simply assigned to either the public sector, managing ‘high-risk offender’ teams, or the new ‘rehabilitation centres’, based on what they were working on at the time. Though still in the same buildings they suddenly had different bosses with very different approaches.

The lack of any preparation meant there was no opportunity to build a market of companies with any particular expertise in probation. Eleven of the twenty-one contracts went to two large outsourcing conglomerates: Sodexo, a French company best known as providers of cheap food to schools and hospitals, and Interserve, which mainly did construction work and subsequently went into administration.

Another winning contractor, Working Links, mainly ran programmes to get unemployed people back into work. It also went into a chaotic administration in 2019, surrounded by allegations of fraud and with one of their probation centres receiving an abysmal inspection with an ‘inadequate’ rating. G4S and Serco, who might have expected to win multiple contracts, did not bid due to an ongoing serious fraud investigation into their electronic tagging contracts (the ones where they had been charging to track dead people). They have, though, subsequently won numerous other Ministry of Justice contracts.

This was as clear an example as you could want that competitive markets cannot be conjured out of thin air. Probation is the definition of a complex service. Its performance is closely tied to many other parts of the public sector, most obviously policing and prisons but also health, housing and education.

Grayling decided to add a ‘payment by results’ element to the contracts on the assumption that companies would be incentivized to reduce reoffending rates if their profits were determined by that metric. Unfortunately reoffending rates are only partly down to the actions of probation officers, and have a lot to do with all those other services private firms had no control over. They could do nothing to stop austerity cuts to housing benefits or adult education schemes from pushing up reoffending rates.

At the same time, cuts to the police meant fewer crimes were resulting in someone being charged, especially in less serious cases. That meant less business for the probation companies. A problem that had not been considered in the contract design.

It was a car crash. Some of the companies, trying to stem losses, stopped bothering trying to reduce reoffending, given how many other factors were involved, and tried to do the absolute minimum possible to meet the requirements of their contracts. They resorted to making occasional calls to offenders without any meaningful contact. The chief inspector of probation wrote, in the damning inspection report on Working Links, that they were ‘completing individuals sentence plans to meet performance targets, without actually meeting the offender’. And ‘the professional ethos of probation has buckled under the strain of the commercial pressures put upon it’.

By this time the Ministry of Justice had acknowledged the entire programme had been a catastrophe and decided to re-nationalize probation. As a critical service, the state had never handed over any risk and it was still responsible for making probation work. So, they had to scramble around for a solution. But a huge amount of expertise had been lost, staffing shortages were severe, and the constant disruption of reorganizations had broken the service and its morale.

As of 2023 the newly unified system was still struggling badly, with the chief inspector writing of ‘very inexperienced staff being handed inappropriately complex cases with minimal management oversight’. He reported that he could not honestly say the public was safe.

The brutal sexual assault and murder of Zara Aleena, a 35-year-old court official, in 2022 at the hands of a man who was wrongly assessed by probation officers as not high risk, is just one real-world example of this ongoing failure. The chief inspector, in an independent review, wrote that her murderer: ‘Should have been considered a high risk of serious harm offender. If he had, more urgent action would have been taken to recall him to prison, after he missed his supervision appointments on release from custody. The Probation Service failed to do so, and he was free to commit this most heinous crime on an innocent young woman. Our independent review brings into sharp focus the consequences of these missed opportunities and reveals a Probation Service, in London, under the mounting pressure of heavy workloads and high vacancy rates.’

On top of this risk to public safety, the additional cost to the taxpayer from contracts having to be changed in 2017–18 and then cancelled early was almost half a billion pounds, according to the National Audit Office.

This was a particularly egregious failure, even if most of the British public remain unaware of it, but these problems – lack of proper competition, the inability to write contracts that take complexity into account, and the absence of any meaningful risk-sharing – are repeated over and over again. It is clear we have ended up with the worst of all worlds: all the costs without the benefits of competition, innovation, or risk-sharing.

We have a small group of vast conglomerates, with minimal expertise in any given area, but well-practised in bidding for contracts, that the government has become entirely reliant on. But despite repeated failures they are now so embedded into the state that they are effectively part of it, except without any of the accountability or transparency that should come with that. So they cannot be jettisoned without completely rethinking the way the state works.

Calling this farrago ‘a market’ is a gross misnomer and criticism should not be the preserve of the political left. It should also infuriate any true Conservative who cares about misuse of taxpayer money. Sometimes there is a flash of recognition. In 2014, Mark Wallace, who was then editor of the Conservative Home website for the Tory party faithful wrote: ‘For far too long, a small number of huge firms have won vast numbers of government contracts with huge price tags attached and delivered relatively poor value for money.’ Tory MPs on parliamentary committees investigating these scandals have put their names to many reports making similar points or slamming a particularly disastrous contract.

However, they can never quite take the next logical step, arguing instead that the state just needs to get better at contracting. Of course, there are plenty of examples of incompetence in contracting, though that is inevitable given large multinationals are always going to be able to pay more to the people writing the bids than the government will to those assessing them. But the bigger problem is that it is just not possible to design effective contracts for highly complex services like probation or hospitals. Nor is it desirable to put these services at such high risk, or lose state capacity to deliver them. Attempts to do so have led to many of the most dramatic examples of state failure in recent years, explored in the rest of this chapter.