LGA response: Consultation on Local Government capital risk mitigation measures in the Levelling Up and Regeneration Bill: capital risk metrics

1. The Local Government Association (LGA) is the national voice of local government. We are a politically led, cross party membership organisation, representing councils from England and Wales.
2. Our role is to support, promote and improve local government, and raise national awareness of the work of councils. Our ultimate ambition is to support councils to deliver local solutions to national problems.
3. This response has been cleared by the lead members of the LGA’s Economy and Resources Board.

4. The Levelling Up and Regeneration Bill will give the Secretary of State significant additional powers to review and intervene in individual councils, and ultimately to be able to make strong directions, such as capping borrowing or forcing the sale of specific assets. The capital risk metrics in the current consultation will be used in the first step in triggering whether a local authority can be subject to intervention using those new powers.

5. It is important that the whole process around triggering an intervention is clear and transparent. In order to plan, councils need to know what they are being measured on, the standards and thresholds that the Government will apply, and the consequences if those standards and thresholds are not met.

6. The Government should be mindful that the chosen metrics will give an incomplete picture of total risk exposure as well as backstop counterparty risks to HM Treasury and other types of lenders, which may include private or foreign lenders. In addition, the metrics do not take account of other methods of assessing risk nor of possible input from other organisations that are capable of playing a role in assessing risk, for example credit reference agencies.

Metrics as specified in the Levelling Up and Regeneration Bill

7. The current consultation is solely about the measurement of the risk metrics specified in the Levelling Up and Regeneration Bill. The narrative of each metric is specified in the Bill (and will therefore be enshrined in the Act). The proposal for the measurement of each of these metrics shows severe limitations with what has been proposed – for metrics 1 and 2 the consultation proposal is to measure something different from what is specified in the Bill (and the same holds for the alternative options considered for metric 3).  The proposal is therefore effectively that alternative metrics are to be measured than those specified in the Bill. Surely these should have been specified in the original bill and been subject to Parliamentary scrutiny rather than being introduced after the Bill has been through Parliament? This is a grave concern. The Act will give current, and future, Secretaries of State the potential to take significant powers to intervene in local authorities and the proposal is that the measurement of the metrics to trigger this will not follow what is in the proposed Act.

8. The data to enable measurement of metric 2 is not currently collected in annual data returns by DLUHC. The recent review of capital data did cover proposals to collect more complete data in future (which have already been piloted and trialled). We urge that the laying of regulations relating to metric 2 be deferred until the results of this review have been fully implemented. This will allow the metric as specified in the bill to be measured, rather than something wholly unrelated (as is proposed in the consultation). If the proposals in the consultation are followed, this could open up authorities to intervention for undertaking activities that are unrelated to the what the metric in the bill is stated to measure.

Resources and data missed in the proposals

9. There is a general point that the proposed measurements of the metrics largely ignore income-based resources, as well as concentrating on revenue rather than capital. Proposals to measure expenditure net of income only will mean significant resources are excluded from consideration.

10. Further consideration needs to be given to local authorities with Housing Revenue Accounts (HRA), particularly for metric 1. The current proposals are to include HRA figures in the numerator of the calculation but not in the denominator. This will result in a significantly unbalanced calculation for HRA councils and is likely to make them far more likely to be subject to a trigger point. Suggestions for a resolution could be (i) exclude HRA from the numerator (ii) find some way of calculating HRA resources to be included in the denominator or (iii) perform a separate HRA calculation. Of these (i) seems to be the most practical. It is noted that HRA is excluded from the proposed metric 4 calculation (which is supported), so exclusion from metric 1 would be consistent with that.

Thresholds and trigger points

11. The measurement of all these metrics cannot be considered in isolation from the trigger points (or thresholds) within them that would place a local authority within the scope of the new powers in the Bill, and the process that will then follow once a local authority has been identified as passing a trigger point.

12. Officials have discussed possible proposals for setting trigger points with officers from a number of local authorities (and others) at events arranged as part of the current consultation. While informal consultation at this stage is welcome, when formal proposals are clearer, they should be formally consulted on before being included in regulations, in order to ensure transparency and also to avoid unintended consequences. It was unclear at the events whether the proposed trigger points would be workable in second and subsequent years of a new system, for example.  

Data sources and quality

13. As a general point, the design of the measurement of the metrics should be such that it does not add a significant data burden to councils. This does appear to be the approach taken with the main proposals all being based on existing published statistics that are submitted by councils. As mentioned above, we urge that for metric 2 this should be deferred until the results of the capital data review are fully implemented.

14. While most of the data to be used is readily identifiable from published statistics (any exceptions are raised in comments on individual metrics in answer to the specific consultation questions) the proposals do not make it clear which year’s data will be used nor which iteration, nor at what point the measurement will take place (nor, even, whether it will be a one off or recurring process).. For example, it is not clear whether published provisional or final data will be used, or whether it will be taken from the most recent year only and if so, at what point will the metrics be measured – several iterations and updates of published data are issued during the year, often to allow the inclusion of late or queried returns. Alternatively, will it cover a series of years, rather than one only? In all cases, the data will be backward looking and by the time publication is finalised some of it will relate to activities from up to two years previously. The proposed measurement of metric 3 includes data that is submitted by local authorities on a quarterly basis, if this used, the timing and quarter used will need further consideration.

15. There are also concerns with missing data and with quality and interpretation of data. For example, metric 1 is total debt compared with resources available. However, using data from the most recent year available in August 2023 (relating to the year 2021/22) and the recommended definition together with the trigger point calculation suggested at the consultation event would not have triggered a particular local authority that recently had a section 114 notice issued in connection with very high debt costs. This is for the simple reason that the local authority in question is missing from the specific piece of published data. If trigger points had been set on a comparative basis (as was proposed at the consultation event) the absence of that particular local authority’s data would impact on, and make lower, the trigger point for other local authorities. This would mean that individual local authorities would be subject to an unintended trigger point.

Impact on local authorities’ behaviour

16. It is important that none of the proposals encourage any perverse behaviours. More than one of the proposed measurements penalise local authorities who generate higher amounts from external income and the expectation must be that if they are enacted local authorities will be encouraged to reduce the amount of income they receive. This cannot be good for the overall public purse and provision of public services.

17. Such problems are clearly solvable, but they need to be resolved before regulations are laid and implemented that could enable the Secretary of State to enact the significant intervention powers in the bill.

18. We look forward to further dialogue with the Government on all these general points and to a further formal consultation on proposed trigger points and process for enacting them.

Local authorities in scope

19. The consultation document proposes separate calculations for what they call “upper tier” and “lower tier” local authorities, and states that ONS definitions of upper tier and lower tier will be used.

20. This is problematical. Under these ONS definitions single tier councils (metropolitan districts, English unitaries and London boroughs) are included in BOTH upper tier and lower tier, which implies they will be included in both groups. It would make a great deal more sense if single tier councils were only to be measured once. Alternative groupings for measurement for councils would be (i) shire districts and (ii) single tier and county councils.

21.Careful consideration needs to be given to including councils that don’t have all the same services, and therefore expenditure base (denominator) and potentially have different borrowing needs. An alternative for councils could be to measure county councils as a grouping separate from single tier. However, if the trigger points are calculated using outliers, as was proposed in the consultation event, a separate grouping would guarantee that one or two of those counties would be triggered for intervention each time the metrics are measured, whereas current figures would suggest that using the trigger points shared at the consultation events no county council would be triggered (except for metric 4) if compared with single tier councils.

22. It is also unclear how other authorities will be affected, particularly fire and rescue authorities and combined authorities. These are covered in the definitions included in the bill, however there is nothing in the consultation document on how these metrics would be measured for these types of authority. If they are measured as separate groups then at least one or more of each will be triggered on each measurement if the comparative approach to trigger points is taken; if they are included with councils, then the capital-intensive nature of combined authorities is likely to distort the figures for outliers. This needs further consideration.

Risk Metric 1 the total of a local authority’s debt (including credit arrangements) as compared to the financial resources at the disposal of the authority.

Question 1. Considering the objectives set out in this document, and the principles set out, do you agree that the proposed calculation (calculation 1(a)) should be the basis for this metric? YES/NO. Please explain.

23. NO.

24. The proposal is that calculation 1(a) should be calculated by dividing the Capital Finance Requirement (CFR) by the Total Service Expenditure - net service expenditure (TSE). While both these measurements are very useful pieces of data, CFR is not the same as total debt and TSE does not pick up all financial resources at the disposal of the authority.

25. The CFR is the measurement of the authority’s underlying need to borrow, even if that borrowing has not yet been undertaken (eg by being covered by internal borrowing). As such it is a very good indicator of overall borrowing need but that is not the same as debt. Actual local authority outstanding debt is reported (eg in the quarterly returns) but the level of debt will fluctuate with ongoing cash flow management and other time factors and so is not such a good comparative measure.

26. TSE does not wholly capture the resources available to a local authority. It would be wrong to assume that the revenue resources that make up TSE are the only resources available to pay down debt.  TSE is also net of any income received (so in this calculation it will penalise local authorities with high levels of income, including sales fees and charges) and it takes no account of one-off figures such as usable reserves and capital receipts.

27. Use of capital receipts is highly restricted but since one of the allowed uses of capital receipts is to repay debt, it is surprising that the proposal to measure “financial resources available” does not take any account of them.  

28. Reserves are a complex subject. Reserves are held for a variety of reasons, and they are crucial to the financial stability of local authorities. But if the metric is to measure “financial resources available to the local authority” then the measurement of the metric has to include some consideration of the level of reserves available.

29. There are specific issues relating to HRA councils (see point made in general points, above). CFR includes HRA indebtedness, but TSE does not include HRA spending. We suggest that HRA figures should be excluded for the CFR; they are not included in any of the calculations for financial resources available.

Question 2 Are any of the alternative calculations more appropriate than the proposed calculation? YES/NO. Please explain.

30. NO

31. The alternatives do not individually offer better options. Alternative measurements of debt are worse than the CFR offered in option 1 (a) (as, for example, they suggest measurement of debt servicing costs, which is further away from the definition in the Bill). Two of the alternatives suggested use core spending power (CSP) as the measure of resources available to the local authority. It would be completely unacceptable to use CSP. Apart from being very incomplete as a measure of financial resources available (covering only amounts included in the annual revenue local government finance settlement) it may be redefined each year by the Secretary of State. It would be entirely inappropriate for the Secretary of State to set the underlying figures to be used in a metric used to measure whether a local authority will be subject to intervention by that same Secretary of State.

Question 3. Considering the objectives set out in this document, and the principles set out, is there an alternative calculation/s you think is more appropriate? YES/NO. Please explain.

32.YES

33. It is accepted that CFR is probably the best measure available of overall need to borrow and that this is actually a more useful measure than overall debt (it is used in the calculations associated with the Prudential Code). However, the wording of the Act will be “Total debt” and such a figure is available.

34. An alternative calculation of financial resources that takes account of sales fees and charges and other income, capital receipts and reserves would be a much more accurate measurement of the financial resources available to a local authority than any of the options given in the consultation. This would also be less likely to encourage local authorities to seek to cut income generation.

Risk Metric 2 The proportion of the total of a local authority’s capital assets which is investments made, or held, wholly or mainly in order to generate financial return.

Question 4. Considering the objectives set out in this document, and the principles set out, do you agree that the proposed calculation (calculation 2(a)) should be the basis for this metric? YES/NO. Please explain.

35. NO

36. The proposal is to divide investment income by TSE. This in no way makes any attempt to measure “proportion of capital assets” and seems wholly unrelated to the metric stated in the Bill. Neither investment income nor a measure of service expenditure can be seen as a proxy for capital assets.

Question 5 Are any of the alternative calculations more appropriate than the proposed calculation? YES/NO. Please explain.

37. NO.

38. Most of the alternatives are just as far from the metric as the preferred option. Some of the options propose using CSP; see answer to Question 3 above for further problems with that. Given that this metric clearly specifies “capital assets” any measurement must be of capital assets, but the proposal and most of the alternatives given here use revenue figures, for example using a measure of service expenditure as proxy for total capital assets. That is clearly wrong.

39. The one exception is option 2 c which divides commercial (capital) expenditure by total capital expenditure. This at least looks at capital and so is nearer to the actual metric than any of the others. The problems with it are (i) that it will look at one year’s expenditure rather than overall assets as specified in the metric and (ii) there are questions of interpretation of commercial capital expenditure, and it is unlikely that all capital expenditure in this category will be “wholly or mainly for a financial return” (for example, in some cases it could include investment in a leisure centre or a car park). This could therefore penalise local authorities and open them up to intervention for undertaking activities that are wholly unrelated to the what the metric is stated to measure.

Question 6. Considering the objectives set out in this document, and the principles set out, is there an alternative calculation/s you think is more appropriate? YES/NO. Please explain.

40. NO

41. The information specified in this particular metric is not currently collected as part of the annual data collection managed by DLUHC. However, the capital information collected has been subject to a review and there are proposals to collect additional data that have been trialled. It was understood at the time that the review being undertaken that it was linked to the need to collect data for a process similar to that now specified by the metrics in the bill. It would therefore make sense to postpone making the regulations for this metric alone until the capital data review has been fully implemented and it is clear that the data is understood, complete and can be relied on. At that point it should be possible to implement measurement of the metric as it is specified in the bill. Laying regulations to measure something else could trigger interventions in local authorities for undertaking activities wholly unrelated to those specified in the bill.

Risk Metric 3 The proportion of the total of a local authority’s debt (including credit arrangements) in relation to which the counterparty is not central government or a local authority.

Question 7. Considering the objectives set out in this document, and the principles set out, do you agree that the proposed calculation (calculation 3(a)) should be the basis for this metric? YES/NO. Please explain.

42. YES

43. The proposal is to take total borrowing less borrowing from PWLB and local authorities and to divide the result by total borrowing. As a calculation this is straightforward and clearly meets the measurement as specified in the bill.

44. The problem with this is that the metric does not act as an indicator of any particular risk and the justification for it in the consultation document is unconvincing (and not linked to the measurement). It is not unlawful or even regarded as bad practice for councils to borrow from commercial sources. It leads to a balanced portfolio of debt. It is not clear what problem is trying to be addressed here. Nevertheless the measurement as proposed is clearly “correct”.

Question 8. Are any of the alternative calculations more appropriate than the proposed calculation? YES/NO. Please explain.

45. NO

46. The proposal is better than the alternatives offered. The alternatives take the result further away from the metric as stated in the bill.

Question 9. Considering the objectives set out in this document, and the principles set out, is there an alternative calculation/s you think is more appropriate? YES/NO. Please explain.

47. Perhaps.

48. Consideration does not appear to have been given to using more of the data in the quarterly borrowing and investment returns published by DLUHC (also called “live tables”) that contain more up to date and detailed data than that in the annual capital return suggested in the proposed options. This might offer a better option for measurement and certainly more detail, which might, on balance, be more useful.

49. However, as stated above, it is hard to see that this metric actually measures any real risk. This is one where the approach taken to the trigger will be particularly important and it would be hoped that the trigger point will be set in such a way that the number of local authorities passing a trigger point on this will be very small and preferably zero. Consideration could therefore be given to whether there is a current need to lay regulations to specify how this metric is measured. Instead it could be deferred and only laid if circumstances should change so much in the future that this ever becomes an area of real measurable risk.

Risk Metric 4 The amount of minimum revenue provision charged by a local authority to a revenue account for a financial year.

Question 10. Considering the objectives set out in this document, and the principles set out, do you agree that the proposed calculation (calculation 4(a)) should be the basis for this metric? YES/NO. Please explain.

50. YES

51. The proposal is to compare minimum revenue provision (MRP) with CFR less any HRA figures. This meets the metric as specified in the bill.

52. The only issue with this is that the consultation document states that the MRP figure can be taken from either capital or revenue returns. However, investigation shows that these two figures are not the same, and advice from some of our members is that they are not supposed to be the same. There should therefore be further investigation into which of the two figures should be used and data quality checked before any formal measurement is included in regulations. Advice from some of our members is that the capital figure is likely to be the better one for this measurement.

53. A further issue with this is how the trigger point will be calculated. The proposals presented at the consultation event would mean 188 councils would pass the trigger point. As outlined in the general points above, the whole measurement of the trigger points and implementation of subsequent action needs careful consideration and be subject to full and transparent public consultation.

Question 11. Considering the objectives set out in this document, and the principles set out, is there an alternative calculation/s you think is more appropriate? YES/NO. Please explain.

54. NO.

Bevis Ingram

Senior Adviser Finance

Phone: 079 2070 2354

Email: [email protected]