Week in Public Services: 14th January 2022

This week: the Omicron wave; geeking out over the social care impact assessment; and problems with the national tutoring programme


Watching the news, you’d be forgiven for thinking it’s endless №10 parties, rising energy prices, and the return to high inflation. Which are all big issues — but it’s going to be a big year of change in public services as well. We may be living through an Omicron wave causing having in schools and the NHS (more on that later) but this year is also due a schools white paper, a government response to the Augar review of further and higher education, the health and care bill, a consultation on social care reforms before they are implemented in October 2023…it’s going to be a busy year. This inaugural IFS podcast of 2022 is excellent on the upcoming challenges facing health, social care, and education, featuring Ben Zaranko and Christine Farquharson making good arguments for more evaluation in public policy. Recommended!

And from the tail end of last year: a good series from Sky’s Amy Borrett explores the links between councils with high levels of deprivation, the effects of Covid, and whether the councils most ‘in need’ of levelling up funding received it. TL;DR — those things don’t always interact in the way you would expect. We’re starting work on the relationship between social care, local authority spending, and the neighbourhood services that local authorities provide (buses, libraries, children’s centres etc.), so keep your eyes peeled later in the year…

Health and Care

Right, the Omicron wave — not great for public services. It’s led to rising infections among staff and residents in care homes, more infection and isolation among NHS workers and teachers, of course, more people in hospital with Covid.

John Burn-Murdoch’s thread is the single best simple explanation of the puzzle that this variant has led to fewer hospitalisations — but the NHS might well be more pressured than before. Why? High staff absences and fewer people staying away from hospitals (unlike the first wave) mean that the NHS is struggling to keep normal services running while caring for Covid patients. Matthew Taylor’s thread explains the problem. Several NHS Trusts have declared critical incidents — public statements that they may not be able to provide priority services — in light of staff shortages.

Has the NHS been overwhelmed, then? Like many things in life — it hinges on your definition of overwhelmed. The NHS has been able to provide care to all Covid patients who need it (and fortunately far fewer have required mechanical ventilation than past waves) — but at the expense of the normal volume (and perhaps quality?) of routine care. As I wrote in December — we’ll only start to see the impact of this when we get data on routine activity in February 2022.

Moving beyond the immediate crisis, let’s look at the longer-term. Last year I complained that the government hadn’t published an impact assessment alongside the announcement of the social care reforms in September… and so did the Treasury committee (and a whole other bunch of health policy wonks). So it’s good then, that the government released its full impact assessment for the social care reforms on 5th January. It’s 136 pages, so I’d highly recommend this summary thread from Simon.

I charted the government’s estimates of the costs (of course I did) — to try to understand how much different aspects of the reform cost — which makes for interesting viewing:

The big costs don’t really start coming until 2026/27 (this is when large numbers of older adults in particular start to hit the £86,000 cap and become eligible for entirely state-funded care). Paying a ‘fair cost of care’ — local authorities paying higher rates for care to eliminate the cross-subsidy where private payers currently pay more for their own care — has an instant and notable cost to local authorities, however.

One cost I hadn’t thought through was the way that introducing a ‘fair cost of care’ results in more people reaching the cap — the purple bar. The point is that state-supported people will meter towards the cap quicker because local authorities will be paying higher prices for care than they would have without the fair cost of care reforms, and these people will therefore reach the cap at an earlier point in time.

Some other interesting things I took from reading it:

  • The government thinks that the introduction of a cap will not meaningfully change the amount of unpaid care people provide (they make a good case for that assumption, but it’s notable given that reducing unpaid care is sometimes an argument made for reform) (pp. 49–50)
  • The government thinks that the cost of the reforms only reaches a steady state in 2027/28 (hence the big jump in costs between 2025/26 and 2027/28 — costs continue to increase steadily after that as a result of the ageing population, not the reforms)
  • Some of the biggest risks to delivery are the new burdens this creates on local authorities (ensuring local authorities have the technical ability to meter clients, and enough social workers to carry out the greater number of assessments)

This is a massive piece of work. And one that (finally!) allows us to compare how much the government thinks the reforms will cost against how much they’ve budgeted for them. The obvious thing that jumps out is just how much of the ‘health and social care levy’ is going to the NHS — £12bn a year is far more than the estimates of the costs of social care reform outlined in this impact assessment.

None of this, of course, removes the short-term problems facing care providers, where Omicron has made staffing shortages much worse.

In research news:

  • The health and social care select committee’s report into the elective backlog
  • Lancet summary of research suggesting the widespread introduction of remote appointments increased the share of same-day appointments, and that the difference in access to same-day appointments between patients in more and less deprived areas reduced (full online analysis here)
  • Chris Hopson from NHS providers makes the case for greater NHS capacity
  • The case for better data to address health inequalities
  • The 2020/21 NHS accounts should be published… at the end of June 2022

Children and Young People

The big schools story is how well (or not) the government’s national tutoring programme is going…which is not looking so great. The big finding is that only 10% of the target number of children for one-to-one or small group tuition actually received tuition through the national programme during the 2021–22 academic year — 52,000 so far compared to a target of 524,000 (full stats for 2020–22 and the first term of 2021–22 here). In contrast, the schools-led part of the programme has reached about 230,000 out of its 1,000,000 target — closer to 20%.

Neither of those are good stats — although the government maintain that they expect tuition enrolment to increase quickly during this spring 2022 term. The wonk in me also really worries about how those tuition figures are being counted. In August 2021, SchoolsWeek found that some pupils were counted as being enrolled, despite not receiving tutoring. To be counted as starting a tuition package in these stats, the DfE states that “there must be an order record and at least one timesheet of tuition attributed to that [tutoring] package”. That’s a better definition — but one timesheet of tuition is not a high bar to overcome…

Frustratingly, the DfE ad-hoc stats do not show whether the pupils receiving tuition are eligible for pupil premium — despite Ranstad having a target that 65% of pupils who receive tutoring should be eligible for pupil premium. Apparently they will be published — which is good, given it is specified as a key performance indicator in the contract!

All of this reinforce worries about the administration of the national tutoring programme, run by the Dutch HR company Ranstad. They’ve come in for a lot of flak. In December last year, a group of charities told the education secretary that they thought the DfE should break the contract with Ranstad after one year.

It’s not particularly great news then, that a DfE-commissioned research report into how schools tried to catch up in the 2020–21 school year found that “externally developed interventions were less popular, as was the National Tutoring Programme (NTP) […] school leaders believed that more tailored approaches, fine-tuned by the teachers to meet specific learning needs, were more effective”. If there’s any glimmer of light, it is that more schools were planning to use the national tutoring programme when asked in March 2021, compared to when asked in October 2020.

In other research news:

  • Great summary of the state of education research on teacher effectiveness from Simon Burgess
  • Very helpful Commons library summary of home education, including some useful polling on the reasons parents choose to home educate from November 2021. Covid-related concerns were the single biggest reason but only made up 19% of choices — concerning
  • Interesting SchoolsDash analysis of how pupils accessed Oak academy, which provided a lot of online tuition, during the pandemic
  • Fascinating blog from the Education DataLab, showing that when Ofsted inspections restarted in 2021, primary schools previous graded outstanding were more likely than secondary schools to be downgraded
  • Useful overview of school disruption from the Sutton Trust showing, among other things, that staff absence is higher in state schools than in private schools

They also find that 29% of state school teachers reported more than 10% of their school’s isolating pupils still don’t have access to devices for remote learning. That’s nuts. A reminder: we are almost two years since the first lockdown!

Other things to keep an eye on — IfG senior fellow Sam Freedman makes the case as to why the government risks sleepwalking into a big exams problem in 2022. The current plan to fix the distribution of results between pre-pandemic results in 2019 and 2021’s inflated results will result in sharp drops — which are not evenly distributed across the country. Schools in parts of the country where Covid has been more prevalent will probably see results fall further — where pupils have had to isolate for longer periods of time and miss more face-to-face teaching. Sam recommends “that results be held at the same level as last year to reduce confusion as much as possible” — and notes that the government will need to make a choice soon if they want to change, given universities will be making offers to pupils soon.

Less children’s social care news, but Kathy Evans makes a good case for why some public services should be exempted from market competition — which feels very pertinent in light of the Competition and Market Authority’s findings of placement shortages and excessive profits among private providers in the children’s care market.

Law and order

In law and order news, a new Crest and Police Foundation report on policing during the pandemic, which contains interesting data from 12 forces across England and Wales on how crime changed during the pandemic, giving a more detailed breakdown than some of the aggregated Home Office and ONS data available. The most interesting finding, I thought, was that “police did not get more time to proactively investigate serious offences due to a rise in non-crime demand (associated with enforcing Covid restrictions), which largely offset the [reduction in demand from] falls in crime”. I am on board with their recommendation that the Home Office should require all forces to standardise police incident data collection and reporting. The standard anecdote is that non-crime demands on the police are rising but it is extremely difficult to know from what base, and exactly what these demands are!

The latest Prisons and Probation Service annual report and accounts are out. Lots of interesting details — including that probation providers were unable to deliver anywhere near as much unpaid work and accredited programmes in 2020/21 (during the pandemic) compared to 2019/20.

The Independent Advisory Panel on Deaths in Custody have produced an important analysis of deaths in custody, showing that around one in five deaths in detention are self-inflicted, and deaths in all custodial settings are much higher than the background all-cause mortality for the general population of similar age and sex — an important area to shine a light on.

Local government

In local government news, a big new IFS report explores what happened to local authority finances during the pandemic (using now-published outturn data, rather than the survey of financial pressures which was available during the pandemic). The headline finding is that in aggregate local authorities increased their reserves during the first year of the pandemic.

Surprising? Sort of — the story is that councils overestimated the additional costs of Covid when they attempted to measure them during the pandemic (and they were never asked to identify areas where spending fell). The IFS recommend that should audit a proportion of survey responses and enquire about areas where spending is expected to be lower next time, in order to get a better response.

The standard caveat — that the aggregate picture is not the individual — and that not all councils increased their reserves, of course still applies.

Last but not least, DLUHC have published the external assurance reviews — independent reviews of governance and finance, led by CIPFA — of local authorities who requested additional support during the pandemic. There’s probably an interesting machine learning project in there for an enterprising data scientist looking to see if there were any common factors in why local authorities requested more support.



Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store
Graham Atkins

Graham Atkins

Senior Researcher @instituteforgov: public services, infrastructure, other things. Too often found running silly distances in sillier weather.