Public Audit and Post-legislative Scrutiny Committee

Meeting date: Thursday, March 8, 2018

Official Report 461KB pdf

---

“Early learning and childcare”

Agenda item 2 is an evidence-taking session with witnesses from Audit Scotland on the Auditor General’s report “Early learning and childcare”. I welcome to the meeting Caroline Gardner, Auditor General for Scotland; Antony Clark, assistant director of performance, best value and audit; Tricia Meldrum, senior manager; and Rebecca Smallwood, senior auditor.

I invite the Auditor General to make an opening statement.

Thank you, convener. As you know, the Scottish Government has a major policy of increasing the amount of early learning and childcare that children are entitled to with the aim of improving outcomes for children and helping their parents into work, study or training. From August 2014, entitlement to funded early learning and childcare rose from 475 to 600 hours a year for all three and four-year-olds and eligible two-year-olds. The Scottish Government and councils are now working towards further extending the entitlement to 1,140 hours per year by 2020.

This report looks at planning and implementation of the initial expansion to 600 hours from 2014 and progress towards implementing 1,140 hours by August 2020. It is the first in a planned series of reports and makes recommendations for the crucial next stage of the policy.

The Government and councils have worked well together to expand provision, and parents are positive about the benefits of funded early learning and childcare for their children. However, parents have reported a limited impact on their ability to work due to the number of hours available and the way in which they are provided, particularly their flexibility.

We found that the Government implemented the increase in hours without comparing the costs and outcomes associated with alternative ways of achieving the increase. Although since 2014 it has invested almost £650 million of additional funding in expanding funded early learning and childcare to 600 hours, it was not clear enough about the specific outcomes that it expected to achieve and it did not plan how to evaluate the impact of expansion. It is therefore not yet clear whether the investment is delivering value for money.

The Government has done more to plan how it will evaluate the expansion to 1,140 hours, but there are significant risks that councils will not be able to achieve this goal by 2020. In particular, it will be difficult to put the necessary infrastructure and workforce in place in time. Given the scale of the change required, the Government should have started detailed planning with councils earlier than it did.

Councils prepared initial plans for delivering 1,140 hours in the absence of some important information on things such as quality standards, the required flexibility and how funding will follow the child in future. Their initial estimates of the costs are around £1 billion a year. That is significantly higher than the Scottish Government’s figure of around £840 million, and the Government and councils are currently working together to refine those estimates.

We have made a number of recommendations to reduce the risks of failing to deliver the expansion by August 2020. In particular, the Government and councils urgently need to finalise their plans for recruiting and training the additional staff required and for funding and building the necessary infrastructure.

As always, convener, we will do our best to answer the committee’s questions.

Thank you very much. Iain Gray will open the questioning.

I want to focus on the plans for expansion to 1,140 hours. The report identifies three areas of concern with regard to the capacity to deliver that. Two of those areas are financial, but I want to ask about the third, which relates to workforce.

There seems to be a very significant discrepancy between the Scottish Government’s estimate of 6,000 to 8,000 full-time equivalent staff for the workforce required and the local authorities’ estimate in their plans of 12,000. Is there any plausible explanation for such a discrepancy in the expected workforce requirements?

I will ask the team to respond in a moment, but I think that it comes down to some of the guidance not being available to councils when they were required to put their plans together, particularly around the quality standards. That is particularly relevant to younger children—the two-year-olds—because of the higher staffing ratios that are needed for high quality in such care, the flexibility that is needed to meet the requirements of parents as well as children and the way in which the funding will follow the child in future. The team can give you more of a sense of where the differences lie within that.

Some of the difference lies in the fact that councils included central staff in their estimates, while the Scottish Government did not. Another possible reason for the difference—

I am sorry to interrupt, but do you mean administrative staff within the council?

Yes. I am talking about other staff in the council, not just the practitioners.

And those are additional staff that they believe they will have to employ in order to administer the system.

Yes.

Some of the difference might also be to do with different ways in which this has been modelled. We know that the Scottish Government has used a zero-based model. It has looked at how many hours of early learning and childcare a practitioner can deliver in a day, and it has worked out that that person will spend six of the seven hours for which they are employed directly delivering funded ELC and that they will do that for 11 months of the year, taking into account leave allowances and so on. It has assumed that existing members of staff and all new members of staff will deliver the same output, and it has then used a zero-based model to work out how many hours are needed. That is how it has concluded that this number of staff will be necessary to deliver the expansion.

However, councils have taken a variety of approaches and their expansion plans are not always explicit about how, exactly, they have modelled their future workforce. Where there is information, it looks as though they have taken their existing model of staffing and applied it forwards to work out the number of people necessary. The Scottish Government’s model assumes that there are potential efficiencies to be achieved in the existing staffing model, whereas the councils have not necessarily done that.

Is it fair to say that the Scottish Government’s estimate is entirely theoretical, as it takes no account of the programme’s management and administration, whereas the local authority estimate is based on the reality of current provision and therefore takes account of the administrative requirements?

I think that they have both taken a different approach. The Scottish Government has modelled its estimates on the basis of practitioners and has taken into account the time that it thinks they need to deliver the management aspects—in other words, the hour a day that the Government has given practitioners for doing the various things that they need to do but which are not about directly delivering funded ELC. Councils, on the other hand, have taken a variety of approaches. We are not clear on the detail for all the councils, but where that information has been made explicit, it looks as though they have just continued with what they currently do instead of making any changes to that.

We know that Scottish Government and councils are having a series of one-to-one meetings to discuss areas of discrepancy in the plans and to refine them. The councils’ plans were just initial ones and further work will be undertaken to refine them.

Auditor General, you say in your report that it will be difficult to achieve or recruit this level of workforce. Is that also your view with regard to the lower end of the estimate of what is required?

We think that it will now be difficult to achieve the workforce and the infrastructure that are needed. This was always going to be ambitious—and that is not a criticism. Obviously, by going from 600 to 1,140 hours, you are nearly doubling the provision, and it is a big thing to deliver. The concern that we raise in the report is that planning could have started earlier, when the decision was taken, given the timescale for training and recruiting staff to deliver this very important service and for building new buildings or refurbishing the buildings that are required. Taking those two things together, we think that it will be difficult to achieve the full expansion to 1,140 hours by August 2020.

Since the report was published, the workforce issue has been raised a couple of occasions in Parliament; indeed, it has been raised with the First Minister directly at First Minister’s question time, and she elaborated a number of measures that have been taken to increase training places. For example, she mentioned the work that Skills Development Scotland is doing to increase the number of apprenticeships in this area and what the Scottish Further and Higher Education Funding Council is doing to increase the number of graduates in this respect. Is your report entirely aware of those additional training places? For example, when you say:

“this will only provide a very small number of the additional staff that need to be trained”,

have you taken account of those changes?

Yes. We have taken account of all of the initiatives that were under way at the beginning of the year, when we were finalising the report, and we also talk about some initiatives that individual councils are taking such as what is being done here in Edinburgh to retrain the existing workforce to meet the demands of expansion. We recognise all of that, but we still think that it will be difficult to get all the staff required by August 2020.

So your view is that, as things stand right now, there is no possibility of training enough additional workforce to deliver those 1,140 hours.

We have not said that. We do not think that this will be impossible, but we do think that it will be difficult.

What, then, will make it possible?

Some of the things that individual councils are doing will help. In the report, we highlight a number of examples of councils—for example, Edinburgh and Perth and Kinross—that are working well to tap new groups of people who can become part of the future childcare workforce, and a pipeline is developing as a result of the things that are happening at a national level through the funding council and SDS and which the First Minister referred to. However, all of those things need to work as well as they can and individual councils need to work with Government to refine their estimates, ensure that their staff are in the right places and bottom out the clarity that is required on what the childcare will look like with regard to flexibility and quality standards.

Is that, in your view, something that really should have happened some time ago?

As we say in the report, we think that it could have started sooner.

A number of questions have been asked at general questions on the theme and issues that Iain Gray has highlighted; in fact, the first question that was asked at last Thursday’s questions was from Liam Kerr on the specific issue of workforce. However, both the questions and the answers seemed to be looking at Scotland as a whole. What is the differential in terms of progress—or lack thereof—across Scotland? Are some local authorities doing particularly well, and are there some whose progress you are particularly concerned about?

There is clearly a mixed picture across Scotland. It is worth noting that some councils started the expansion to 600 hours back in 2014, but the team can probably give you more information about the variation we are seeing across the country.

We have reviewed only councils’ initial plans, and they have probably already moved on from them. Given that councils are in discussions with the Scottish Government, it is at this stage hard to pull out a particular example of a council that is further ahead of the others or ones that are particularly challenged.

I was just wondering whether any were doing particularly well and could be emulated by other local authorities.

The questions on workforce planning that were asked last week related to Scotland, but there are also specific geographic areas of concern with regard to recruitment. Obviously, the economic picture varies quite considerably across Scotland, and I imagine that in some areas of Scotland it is relatively easy to recruit people, while in other areas—the rural parts and island communities in my own constituency, for example—it is considerably more difficult.

Exhibit 3 in the report gives you a sense of the extent to which individual councils are already able to provide more flexibility in the provision of childcare at the 600-hours level, and I think that those with more flexibility are in a better place to get to 1,140 hours by 2020. That is the starting point.

Beyond that, as we say in the report, no council has a clear commissioning strategy that sets out the demand from parents; the number of children in each of the year bands and how that will change over time; and the extent to which different types of flexibility are needed. It is quite foreseeable that that will be different in cities such as Edinburgh or Glasgow compared with the more rural and remote parts of Scotland. If local authorities were to set that out and be very clear about the current provision and how that needs to change and develop, they would have a great basis on which to say, “Here are the staff whom we need to recruit, retain or be thinking about changing from other services that we currently provide.” There are differences, but without that kind of commissioning strategy, it is not possible to say that one council does not have a problem and another has a big gap.

Can I ask one more wee question on the same issue, convener?

Just a wee one.

I take it from your previous answer, Auditor General, that there are no real areas of best practice that you can look to.

Finally—and I appreciate the leeway that I have been given on this from the convener—have any initiatives come out since the report was published that you are aware of and which would help deliver this target?

09:15  

I would say, first of all, that there is no single council that we think is doing everything right, but we have highlighted in the report a number of examples of good practice that we would want to pull out.

I will ask the team to respond to your question about new initiatives since publication.

There has been an announcement about increasing the number of modern apprenticeships, but that is the only initiative that we are aware of. Quite a lot of guidance on flexibility and so on is due to be published later this month, and that will clarify some issues.

With regard to Mr Gibson’s point about councils, the Auditor General said that some progress on recruitment had been made in Lothian and Perth and Kinross, and I think that Ms Smallwood said that some councils were better prepared than others. Is it possible to publish some of that information to give the committee a flavour of what is going on? As you know, we would want universal service provision to be of equal standard, so it is really important for us to know where the gaps are. Clearly some of that information exists, so can it be published?

The report contains some of the examples of good practice that we have referred to, and exhibit 3 sets out where councils currently stand. The councils’ plans were initial ones that, as Tricia Meldrum has said, are being discussed between Government and councils as we speak, and I am not sure whether we can do much more with them at this stage.

You might want to pick up that question with the Scottish Government and the Convention of Scottish Local Authorities when you seek advice from them on the progress that is being made, given the on-going discussions that Rebecca Smallwood has already mentioned.

Okay. So, apart from the exhibits that are in the report, there is no other information.

I call Willie Coffey.

Thank you very much, convener, for the opportunity to ask a supplementary.

On the issue of the Scottish Government’s estimate of 8,000 and the local authorities’ estimate of 12,000 for the additional workforce that will be required, I believe that I heard that the additional 4,000 might principally be in central staff admin.

No. As Rebecca Smallwood has said, that is one element, but the larger element is the way in which the modelling has been done. The Government has had to take a standardised approach to assuming the number of staff needed to cover an additional number of hours for an additional number of children, while each council has extended the model and the provision that it already has in place to cover the additional children who are affected. I do not think it surprising, therefore, that there is a difference between the two, but as we say in the report, some of that difference was inevitable, because councils did not have the guidance that they needed on quality standards, flexibility and how funding will follow the child, and that affected their ability to make assumptions about what things will look like in two years’ time.

But the 12,000 estimate does include admin staff.

Yes.

Does the 8,000 estimate include admin staff?

No.

Okay.

As I understand it, the funding that has been provided to councils in the current year for workforce expansion is about £21 million, but that is going up to £52 million next year. Presumably, that will help us bridge this gap.

The overall revenue gap between the Government estimate, which is £840 million, and the councils’ estimate, which is £1 billion when added together—and I stress again that these are estimates—is £160 million. In some ways, I do not find it surprising that there is a gap at this stage. As always, there will be an element of negotiation and moving towards a common vision. Our finding in the report, though, is that the gap would likely have been smaller had some of this guidance been available earlier as councils were preparing their plans.

Good planning needs good data, and we seem, again, to be talking about the quality of or lack of data. In paragraph 26 of the report, you say:

“There is no available information on children’s attendance or the numbers of hours of funded ELC they receive.”

Moreover, in paragraph 32 you say:

“This research highlighted that councils not knowing the details of exactly who is eligible was a major barrier.”

It goes on. How can we base anything on such a lack of data? There is just nothing there.

You are absolutely right that this is a common theme in the reports that we produce for the committee. In some ways, though, those data gaps are not the most significant challenge that councils and Government are facing in delivering the expansion to 1,140 hours. We make some recommendations particularly with regard to the eligible two-year-olds, because it is very difficult for councils to know which of the two-year-olds in their areas are among the estimated 25 per cent who are eligible as a result of coming from more disadvantaged backgrounds. That would be a big help in allowing councils to plan, and it would particularly help families be aware of their eligibility and ensure that they could access the childcare that is available for their children.

The bigger data problem that we have identified is that, with the expansion to 600 hours back in 2014, the Government did not set out how it would evaluate success. That is making it hard both to look back and see whether such a move represented value for money and to inform the decisions about how best to manage the expansion to 1,140 hours.

I take it that the most appropriate people to collect this data are the councils. Have they received any guidelines as to what data they should be collecting? The report seems to indicate that, where they are collecting information, they are all using different formulas.

For the expansion to 1,140 hours, a framework is still being developed for the measures needed to evaluate and monitor it over time. However, I do not want to underplay the problems with collecting some of this data. I have already touched on the difficulties of identifying the eligible two-year-olds; at the moment, the Department for Work and Pensions is not able to share with councils the information that is necessary to let them know which children are eligible. In paragraph 26, which you have referred to, we point out that the number of registrations is not the same as the number of children in receipt of funded early learning and childcare. Some of those niggles are simply a reflection of the way in which childcare is delivered rather than a failure to plan or monitor what is going on.

Rebecca Smallwood can probably give you a bit more information about the work that is under way to fill those gaps. However, they are due not to a lack of foresight, but to genuine challenges with regard to the data’s availability in the first place.

The Government is working to improve the situation with the registration data, which at the moment is collected every September in a census in which individual nurseries or childcare settings are asked to provide information about the numbers in those settings. There can be an issue with double-counting children who are registered in more than one setting, but you can get rid of that by collecting information at individual child level. That is what the Government is working towards, and I think that that is likely to be in place around the time that the 1,140 hours provision is introduced.

According to the report, though, basic data such as attendance is not even being collected.

That is right. Attendance is not currently collected in the census.

Paragraph 36 says that councils have

“different ways of apportioning the costs of teachers and head teachers in nurseries which are part of a school, and different ways of splitting up other whole school costs (for example the cost of cleaning) for these nurseries”.

Moreover,

“not all councils include spend on partner providers in the pre-primary section of the LFR”.

It just seems as though very basic data and information are not available. That must affect how you evaluate this.

In the report, we recommend that the Scottish Government and local authorities work together on gathering better cost data to inform judgments about which models are most cost effective and provide value for money. They are working on that at the moment.

Clearly, the information has to come from the local councils, but they need some guidance on what information they should be gathering. Is work on that going ahead?

It is part of the discussions that are taking place at the moment between the Scottish Government and local authorities.

Do you have any idea when those discussions will be concluded?

I would need to double-check and get back to you on that. It is certainly part of the work that is going on at the moment, but I would need to get back to you on the data issue.

Given the relatively short timescale for introducing the larger number of hours, they do not have much time to start getting this information together.

Can you come back to the committee on that, Mr Clark?

Yes.

Thank you. Colin, do you have any more questions?

No.

I am going to stay on the issue of data but perhaps come at it from a slightly different angle. First of all, at some point—it might have been in the item of business that Kenny Gibson referred to—there has been a discussion in Parliament about a discrepancy with regard to what these hours are actually for. There seems to have been some guidance, but it does not necessarily clarify things. Is the aim of providing a certain number of hours to improve children’s outcomes and thereby close the attainment gap, or is it to improve outcomes for parents? What is the bias in that respect?

That is a really good question. As we say in the report, in expanding provision to 600 hours from 2014, the Government was not clear which of those two outcomes it wanted to achieve. The easy response is that more childcare is clearly a good thing, and most people would support it.

The reason that we have highlighted this as an issue, though, is that the outcome that you are focusing on affects how you go about expanding childcare. There is very little evidence that simply providing more hours of childcare for children who are already receiving it actually improves their attainment or the quality of life that they achieve later on in their lives. If you are focusing on outcomes for children, it makes sense either to cover more two-year-olds with fewer hours or to concentrate more hours on the two-year-olds from the most disadvantaged backgrounds. Clarity on the focus for the expansion to 600 hours of free childcare would have had an effect on the way in which the childcare should have been delivered, but, as we have said, that options appraisal was not done.

I think that, with the expansion to 1,140 hours, the Government has been much clearer that the focus is on outcomes for children, but that is where the questions around quality standards and how they play against the flexibility that is available to parents become very important. Because we are talking about outcomes over a long period of time, the evaluation framework takes on more importance.

That is why we think it important to be clear about what the outcome that you are seeking to achieve. It is not simply because we are a bunch of bean counters; it is because that sort of thing affects the way you go about doing this.

That is a very important point, and I am glad that we have that clarity now. Depending on whether other committee members ask about it, I might come back in later on the issue of flexibility and the ability of parents to access hours.

If I have heard you right, there is a lack of research on the impact on children’s outcomes of a simple increase in hours. However, I understand that there is research showing that earlier access has a more positive outcome for children, particularly those in lower socioeconomic groups or with less positive learning environments. If that is right, should the Government, when it was looking to increase hours, have looked more at targeting them at, say, two-year-olds, where the uptake is not particularly big at the moment, instead of making a blanket “Here are some more hours” announcement?

As we say in the report, with the expansion to 600 hours, there was, first, no clarity about what was the prime objective and, secondly, no options appraisal of the objective and then decision that said, “This is the way we intend to invest the additional resources that we are putting in.” With the expansion to 1,140 hours, there is a clearer focus that it is about outcomes for children, but I will ask the team to talk about the evidence that we have seen about the way in which that is being planned and carried through.

The blueprint with regard to the 1,140 hours makes it very clear that the primary reason for the expansion is outcomes for children and that the impact on parents would be, if you like, a side effect. That is important, too, but it is not the primary goal of the expansion, and there has been a much clearer statement in that respect.

There has also been better planning at this stage about what those outcomes would look like, with some planning on the process measures that need to be in place with regard to what can be measured in the shorter term and what you expect that to achieve in the medium term and the longer term in eradicating child poverty. There are other ambitious longer-term outcomes, too.

However, although we have seen more clarity around what is expected and better planning around the information that has to be collected to find out whether it is working, there are still some gaps that need to be addressed around having more clarity on the long-term outcome measures, how that sort of thing will actually be measured and how people will ensure that they collect the baseline information either at this stage or prior to the implementation of 2020. We feel that the Government is in a better position, but there is further work to be done.

09:30  

I accept that. Everyone wants better outcomes for children. The primary reason for this move might be to reduce inequality at a later stage, but is there any research that says that simply increasing free provision from 475 to 600 hours—and now from 600 to 1,140 hours—will result in a positive measurable outcome, or could it be suggested that we have just thrown £650 million into increasing provision to 600 hours simply on the basis of “Let’s try this and see what happens”? Surely that is what is being suggested if there is no research that says this is going to work.

We have been critical of the expansion to 600 hours for that very reason. If the Government had been clear at that time about what it was trying to achieve and about the importance of outcomes, it might well have done things differently, either by funding more childcare for all two-year-olds but for a small number of hours or by increasing the number of hours for two-year-olds, which is the age group for whom the evidence is strongest that this approach makes a difference to outcomes.

We also say in the report that the percentage of two-year-olds who are actually accessing the early learning and childcare that they are entitled to is lower than we would expect at about 10 per cent instead of 25 per cent. That is partly because of the problems that councils have in knowing which children are eligible and partly because of parents not knowing that they might be entitled. We have made some recommendations for increasing that figure, but clearly the priority is to increase the number of eligible kids who are actually taking up their entitlement in that subsection of two-year-olds for whom the evidence is clear that this would make the biggest difference.

My question is on the same theme. I would like you to drill down a wee bit further into uptake. Your report tells us that almost all three and four-year-olds are accessing the funded hours but that the number is much lower for two-year-olds. Looking across the socioeconomic groups, can you say anything about take-up by more people in deprived communities? Are the numbers for three and four-year-olds pretty high right across Scotland, or is the picture more patchy?

The way that the information is collected makes it very difficult for us to answer that. It is collected from a census; we do not have individual, child-level data, so we cannot say anything about take-up by individual children within each council area. We just know that take-up is high for three and four-year-olds at a council level and at a Scotland level.

As part of your recommendations, presumably you will ask for that information to be collated to allow us to examine take-up more carefully. Is the eligibility of two-year-olds one of the factors? Is the take-up rate for two-year-olds so low because of confusion or because eligibility is based on birth dates and so on? What is the reason for take-up being so low?

The work with families of eligible two-year-olds has found that one of the main reasons why people do not take up provision is because they are not aware that their child is eligible. Where people know that their child is eligible, there is higher uptake. It is just about getting that information to people so that they know.

We have some good examples of personal engagement by health visitors, social workers, job centres and so on, which is very important. An important way of getting the message out is people being able to share that information with the people they are working with. One of the big barriers is that people just do not know that they are eligible.

Are you recommending that councils should be asked to try to do a wee bit more to raise awareness that the facility is there for families to use?

There is also the issue that councils themselves do not know exactly who is eligible because they do not have access to the information through Her Majesty’s Revenue and Customs and the DWP. We recommend that the Scottish Government and councils work to try to improve access to that information, which could help councils to target those families better.

You mention evaluation in the report. As Liam Kerr asked, what is it we are evaluating? Is it the impact on children’s attainment and outcomes, or is it the positive benefits for families of allowing them to go back to work? I suspect that it is probably both, and I think that the Scottish Government has accepted that and has agreed to work on evaluating those two strands. Have you made any further recommendations about whether that evaluation should be short, medium or long term? It is a relatively new policy, to be honest, and its impacts and the benefits may not be felt for some years to come. Is the evaluation’s scope short, medium or long term?

The Government itself is committed to a short, medium and long-term evaluation strategy—we set out some of that in the report. As we have already said, there are still some gaps in some of the baseline data that need to be filled, and there is still an on-going discussion with local authorities about what measures might be used in some of the important areas.

In doing that, are you asking for data to be collected on a per authority basis and on a per community basis within authority areas, so we can get a really clear picture of how the policy develops over the years?

We have not made recommendations in that respect, but we are very aware that those discussions are taking place between the Scottish Government and local authorities.

If he is still on the committee in a few years’ time, Mr Beattie is bound to ask the same question about what the data tells us about policy’s value, benefits and impact.

I will pick up a point that Tricia Meldrum mentioned in answer to one of Mr Coffey’s questions. Is there any barrier to the information from the DWP and HMRC going from them directly to local authorities? Does that not happen because there is a constitutional issue, or is it just that an arrangement has not been set up?

The way we phrase it in the report is that councils do not have a statutory duty to identify the eligible two-year-olds and they do not get information from the DWP and HMRC, which means that it has just not been happening so far. We have recommended that the Government should engage with the DWP and HMRC to see whether the issue can be overcome—it already has significant engagement with both of them around the new financial powers on taxation and social security. We do not yet know what the position is, but to us it seems key that if councils are going to make sure that the parents of every eligible two-year-old at least know that they are eligible, the councils will need to know who those families are, given that entitlement is not universal but covers about 25 per cent of all two year-olds.

It seems very frustrating that the information is there but is just not being passed on. If we take further evidence on the policy, perhaps we can raise that issue with COSLA and try to push that information sharing along. It also strikes me that the national health service has information on how many two-year-olds there are because they have data on how many children were born two years ago. Is there any obstacle under data protection legislation to the sharing of information between the NHS and local authorities?

We are seeing more information sharing for a number of purposes, but just knowing the number of two-year-olds, or which children are two years old, is not very helpful. A subset of that age group is entitled at two years rather than at three and four—children whose parents are in receipt of particular social security benefits or looked-after children, for example.

Of course.

Just knowing that they are two is not enough. You need to know whether they fall into one of those categories.

Therefore, information from the DWP and HMRC would be much more helpful.

That is right.

I have another, related question. On a few occasions, constituents who work have told me that their three and four-year-old children are losing their nursery places because those places are being given to eligible two-year-olds whose parents are not working. Did you find any examples of that in your research?

We found some examples where councils are having to prioritise children in particular groupings in order to meet the targets. I will ask the team to give you a bit more information about what we saw.

I do not think that we found any specific examples of the type that the convener mentioned, but there are examples of local authorities capping access to certain services, which we set out in the report. That capping can have an impact on whether individuals are able to access local authority services or private or third sector services.

What do you mean by “capping access to certain services”?

The approach limits the number of places that councils offer to families, in order to manage the market, if you like, so that they have confidence and certainty over how many places they are purchasing across private and third sector providers and make effective use of their own in-house services.

Might there therefore be a situation in a nursery where the places for three and four-year-olds are capped to make way for eligible two-year-olds?

I am not sure that that would happen, and the approach is not widespread. We set out the six or nine local authorities that cap at the moment. Colleagues may be able to direct you to the section of the report that details that.

Would anyone like to add to that?

I will look for it myself. [Laughter.]

At present, a third of councils cap places at their partner providers. However, that is not really to do with age. A child may already be attending a partner provider when they are below the age at which they may be eligible—when, say, they are one or two. The issue for parents arises because they may not be able to continue in that same nursery with a funded place if they become eligible, at two or at three, because the council has capped the number of funded places that it will offer. The parents will still be offered a place for early learning and childcare, but it might not be in the nursery where they want that place to be. It might be in a different nursery, such as one of the council’s own nurseries, rather than in the partner provider setting that is the parents’ preference.

I certainly know that that is happening. There are situations in which families are asked to send children to attend nursery outside their community, and perhaps to split up siblings between different providers. It can get very complicated. Thank you for answering my question. Does anybody else want to come back in?

In the work that we did with parents, we heard a number of similar stories from parents about not being able to get places at their first choice of provider due to capacity. Sometimes that was about prioritising four-year-olds over three-year-olds, or prioritising looked-after children or children in other priority groups. We did not find that there was a particular issue around two-year-olds compared to other age groups.

Yes. Nobody is denying that it is a complex area.

My colleagues have mostly asked questions about the body of the report. I got a little bit stuck on the summary page, which says:

“The Scottish Government and councils have worked well together to expand provision.”

That is followed by remarks such as:

“it is not yet clear whether this investment is delivering value for money ... Parents in our research said funded ELC had a limited impact on their ability to work ... There are significant risks that councils will not be able to expand funded ELC to 1,140 hours by 2020”,

and, finally that the cost is

“significantly higher than the Scottish Government’s figure.”

How is that working well?

We recognise the fact that 600 hours were available for all three and four-year-olds and for the two-year-olds who were entitled by the date when that policy was put in place. That is an achievement. However, we think that it could have been done better, in the ways that we set out in the report.

And, looking to the future?

It is all the more important that the lessons are learned and that the baselines that would enable value for money and the impact on children to be measured are in place as the money is being invested, not afterwards.

That is not the feeling that I get from reading in the summary that it is working well.

We said that the Scottish Government and councils worked well together to achieve the expansion in places. The places were available, and there is a benefit to that. We think that it could have been done better, and in the report we have been very clear about how that could have been done and the lessons that can be learned.

If you consider that it could have been done better, in what way has what they have done worked well?

The wording is clear: we think that they worked well together to achieve the expansion, not that everything was done well. Those are two slightly different things in relation to the way in which the report was drafted and the conclusion that I reached about the investment that was made.

Kenny Gibson asked for an example of what was going well. I think that you said there were no examples. Did I mishear that?

No. I think that we intended to say that there was no single council that was doing all of it well. There are a number of examples of things that are going well in particular councils around flexibility, training the workforce for the future, and retraining staff from other council services. We were not able to identify one council that was doing everything well.

So your report is a sort of curating.

Yes, and we have recommended a number of areas where things can be improved.

Finally, I understand that, when you finalise your reports, you have a discussion about factual accuracy.

Yes.

The Scottish Government was happy that this was all factually accurate.

I can confirm that.

Without necessarily giving names, who would have agreed that. What is their job title?

The director general is the accountable person.

Paul Johnston is the director general who signed off the clearance comments on the report.

Was that a personal discussion?

We met Paul Johnston and his colleagues as part of the clearance process, and then we received a formal letter confirming the factual accuracy of the report.

09:45  

I will take this opportunity to come back on the issue of parents and accessibility.

As I understand it, 90 per cent of nurseries have no full-time places, and many places are available only during the school term. As I recall, 19 councils have no nurseries that are open on a full-time basis, and 45 per cent of nursery places are for half days. Against all that, surely the Government can put in as many hours as it wants, but if parents cannot access them—if they cannot actually avail themselves of those hours—the policy is of limited value to the parents, who we have now established are the secondary target for outcomes, if you like, and there are limited outcomes for the children. Is that correct?

There is no doubt that childcare needs to be available in ways that meet the needs of parents, particularly those who are trying to use the childcare entitlement so that they can get themselves into work or can increase the number of hours that they work. We show in exhibit 3 that there is a variation in the amount of flexibility that individual councils are able to offer across the range of provision, and in many ways that is why we think that the absence of commissioning strategies in each of the 32 councils is so important. They need to understand what the children and the parents in their area need in order to be able to deliver, and commission from their partners, the services that will help parents get back into education and work and deliver the improvements for children that are the focus of the policy.

I have a very quick question—it might be slightly from left field, but it is just a matter of clarification. Does a cost arise only if a place is filled? Entitlement to 1,140 hours is a great headline, but it actually has a cost only if the place is drawn down against, if that makes sense. How does it work?

I think that the answer is that it depends. I ask Antony Clark to pick that up.

It is a very complicated area, but clearly the cost does not arise only when a place is created. There are marginal costs associated with the delivery of these services. One reason why local authorities are quite keen to make full use of their resources is because that creates efficiency. The Scottish Government’s thinking around the funding following the child is that that should ensure the efficient use of resources in the expansion of early learning and childcare.

My next question is similar to my first one. If full-time hours are not offered, parents will have to pay for their own childcare. In other words, they cannot get free childcare unless they pay for the extra. That suggests to me, and I think that there is research on this, that such an approach will disproportionately benefit the more wealthy in society, presumably because they are better able to access it because they are more likely to have a job in the background to pay for it. If I am right about that, the very children on whom this programme could have the most positive impact are less likely to be able to access it as a result of the hours not being flexible. Would it not have been better for the Government to examine the use and accessibility of the 600 hours, say, to ensure that it got maximum benefit from that in terms of the outcomes for children before it increased the hours to a figure that potentially cannot be accessed or will not achieve the attainment end game?

I think that it would have been better if the Government had the information to enable it to evaluate 600 hours when it was making decisions about 1,140 hours. However, that was not there. The process of planning is more difficult because it is not clear what councils are required to put in place around flexibility. We know that flexibility is key to parents being able to use the provision and make it work around the particular circumstances of their jobs and working hours, how far they have to travel and where their other children may be in terms of school or nursery provision. Those things all make it more difficult.

That said, evidence is coming through of increasing flexibility and of parents being able to use their funded entitlement as it moves to 1,140 hours. Then if they work longer, they can pay for top-up hours that enable them to have wraparound care, and to have it during school holidays as well as in term time. That flexibility is increasing, but in the absence of commissioning strategies it is hard for us to be sure that it is increasing where it is needed, and that how it is being delivered both meets the needs of parents and provides the best value for money possible.

This is very complex and we have tried to simplify it as much as we can. The policy has to meet the needs of parents, as you say, if children are going to benefit, which will undoubtedly mean an increase in flexibility. There are questions about that: how much more will it cost, and how many more staff will it need? We do not yet know the answer to either of those questions because of the gap between the estimates of the councils and the Government.

Just on that last point, is that not something that should have been planned? As you will know from previous meetings, I struggle with the idea that people can just say, “Okay, we are going to do this and we will worry about the cost and how we are going to implement it later.” Perhaps that is a comment rather than a question, but I will throw it to you anyway.

Thank you—I appreciate it. We have said a number of times in our work over the past few years that the Government’s outcomes approach is a good thing. There is no doubt that it is much better to think about the outcomes that you want to achieve with public investment and public services than not thinking about those things. However, setting the outcomes is only the first step, and we think that this policy is an example of an area where planning for how outcomes are going to be achieved could have been done better, in terms of the priority given to both outcomes for children and helping parents back into work, and then the details of how the expansion is to be achieved by 2020 and planning earlier for that within what was already a short timescale. Planning for outcomes matters, and it could have been done better in this case.

Thank you. That is clear.

Auditor General, keep me right. Is my understanding correct that the Scottish Government has not yet agreed to the totality of funding for the expansion programme?

First, the Government is still working with councils to refine the plans in order to try to close that gap between £840 million and £1 billion. It is in negotiations with COSLA about a multiyear settlement for funding both the capital and the revenue costs to 2020. Those negotiations were due to be concluded in November, and I think that they are now due to be concluded towards the end of this month, before the start of the new financial year.

It is my understanding that Dundee City Council has money in its 2018-19 budget to meet the 600 hours provision, but that it has no money to meet the expansion programme. Would that be correct?

I cannot comment on Dundee specifically, but we know that the multiyear settlement has not yet been agreed. The Government and COSLA decided to postpone that agreement to allow more time for negotiation.

Would that be true of every local authority in Scotland?

That is my understanding.

But the expansion programme target for 1,140 hours is 2021.

It is August 2020.

Okay—2020. If there is no money for the expansion programme in the 2018-19 budget, it will move into the budget for the year when councils are expected to meet the target. Is that correct?

Our understanding is that it is still intended that the multiyear settlement will be agreed very shortly—ideally before the start of the 2018-19 financial year. That is obviously urgent because the money is for training staff, building or refurbishing the buildings that are needed, and generally investing in the expansion.

Rebecca Smallwood is desperate to add something.

I want to say just one thing before Rebecca Smallwood comes in. It is my understanding that the councils expect agreement to be reached by May, which would be too late for the 2018-19 budget, and the money would, therefore, go into the 2020-21 budget. That is the same year when local authorities are expected to deliver the expansion programme.

I will ask Rebecca Smallwood to answer that before I dig myself in any more.

Our understanding is that the 2018-19 settlement has been agreed, but we do not have the full details of the distribution. There is no decision on how capital will be distributed yet. That decision is delayed until people have seen the details of the multiyear settlement.

Am I correct in saying that it will not be distributed on the normal formulaic basis?

Yes, I believe that there has been a separate decision about this funding.

Okay, so they are still in negotiation about what need is in different areas. Is that correct?

Yes—about how the money will be distributed between different areas.

So, there is a bit of a gap between need and allocation. Even if agreement is reached by the start of the financial year at the start of April and the capital money goes into the budgets in 2018-19, is that really sufficient time? There has to be new infrastructure, has there not? There have to be new buildings, which have to be planned and approved. Staff have to be trained to teach in those buildings. The end date—the target date—for this is 2020. Is there enough time to meet that deadline?

One of our key messages is that that will be difficult now. It was always a short timetable. It became shorter because planning started later. We think that, in terms of both recruiting and training the staff needed and investing in the buildings that are needed, it will now be very difficult. That is particularly the case for the buildings, because all councils will be looking to put out tenders, contract with builders and others and get spades in the ground where that is needed within a short, two-year period. As we say in the report, we think that that will be very difficult.

I understand why you have to use very diplomatic language around this, Auditor General. You say that it will be “very difficult”. However, given the amount of time that it usually takes councils to approve buildings and get them built, I would say that it will be impossible, frankly. Councils have no idea of the distribution, they are unable to plan projects and there is no chance that they will get approval by the deadline.

At the start of your report, you talk about the quality of childcare. You say that the Scottish Government

“stresses the importance of high-quality ELC but does not define high quality”.

How much of a problem is it when it comes to recruiting and training staff if there is no definition of what we expect from “high-quality” childcare?

We know that quality is key for parents. Obviously, every parent wants that, and the quality of ELC matters particularly in terms of outcomes for children. A lot of that comes down to staffing ratios and the outcomes that are being measured. I think that Rebecca Smallwood can put some more flesh on the bones of that for you.

What we were talking about there related specifically to the statutory guidance for 600 hours. The guidance talks about the importance of high-quality ELC, but it does not define what high quality is. We know that there are processes in place for quality assuring ELC through Care Inspectorate and Education Scotland inspections, but the guidance does not set out a baseline or a benchmark for quality that a provider has to meet to be able to deliver funded ELC. We know that that is being taken forward as part of the expansion to 1,140 hours. The Scottish Government is developing a quality standard, which will be a benchmark standard that providers have to meet in order to be able to deliver funded early learning and childcare. However, that approach was not adopted for the 600 hours provision.

On quality, my recollection is that Care Inspectorate data suggests that the quality of early years provision has fallen in recent times and that the percentage of preferred providers that are rated good or above is at its lowest point for half a decade. Given the expansion, there will be increased pressure on the system, so how confident can we be that the quality will stop declining and start increasing again?

In the report, we comment on quality in paragraph 57, where we say:

“The most recently available information from Education Scotland inspections is for January 2012 to June 2016. Almost all centres inspected received satisfactory or better grades across three quality indicators.”

On the specifics of the Care Inspectorate, which you just mentioned, the report says:

“Care Inspectorate grades for daycare of children services and childminders as a whole have remained constant since March 2014”,

which means that

“About 40 per cent of daycare of children services received very good or excellent grades for all indicators over this period.”

In a sense, our judgment was that the expansion has maintained quality; quality has not improved or deteriorated.

Right, but there will be a further expansion.

Indeed.

You are confident that that will not have a negative impact on quality as the pressure comes.

I would not want to speculate on what the impact of the expansion will be. Rebecca Smallwood has already mentioned that the Scottish Government is setting out a set of quality standards and criteria that providers, be they local authorities, third sector or private sector, will be expected to meet if they are going to be providing funded childcare in line with the 1,140-hour expectations.

Auditor General, at the start of the meeting you said that the Scottish Government had not done work on alternative business cases for how to deliver the policy—I think that that was roughly the language you used. The cabinet secretary was asked about that in the chamber during the debate last week, and his reply was that we know what we need to do and we are just getting on with it. What value would it have been to the Scottish Government to investigate other ways of delivering the policy, in terms of the finances and the outcomes for both children and their parents?

10:00  

I recognise that parents value additional funded childcare. There is no question about that, and the increase has been welcomed by parents, as we say in the report. The point that I was making in my opening remarks and in the report is that, depending on what is most important to you—whether it is outcomes for children or helping parents into education, training and employment—you will take a slightly different approach.

If your focus is on outcomes for children, the evidence suggests that, rather than expand the number of hours available for children who already access childcare, it makes much more sense to invest in starting children—particularly more disadvantaged children—earlier in childcare. Rather than a blanket expansion, that outcome would have led you down the route of increasing the number of eligible two-year-olds, either by giving all two-year-olds some entitlement or focusing on the most disadvantaged and giving them a larger entitlement. We think that having clarity about the outcome that you are focusing on and doing an options appraisal on how best to improve that outcome would have enabled the Government to demonstrate value for money, against the backdrop of the outcomes that it has set itself as a Government.

That is not in any way to downplay the importance of additional childcare to the families who receive it. We are saying that if you are looking to achieve a particular outcome, you need to do more planning about the best way of achieving it to minimise the risk of wasting money on things that do not affect it or of downplaying options that would have given you more bang for your buck.

There is a very tight timescale and we are going to struggle to meet the deadline anyway. Do you think that the business case stage was passed over because of that timescale, in order to try to achieve those targets in as short a time as possible?

My comment was about the expansion to 600 hours, so it goes back to 2014. That was still a speedy expansion, but it was not the same scale of expansion that we will see between now and 2020. I think that it is probably part of a broader learning within Scottish Government about the next steps, having set an outcomes framework. In response to Mr Kerr’s question, I referred to planning for outcomes being the more difficult stage and the one that really delivers the benefits of outcomes. We have seen that in this report very clearly; we have also seen it in some of our other reports.

Thank you. Members have no more questions for you, Auditor General, so I thank you and your team very much for your evidence this morning.

10:02 Meeting continued in private until 10:42.