Committee for Finance and Personnel Report (2007-2011 Mandate)

Research Paper: The Barnett Formula: Appendix C

HCS/104/09

Our Ref: COR/21/09

Audrey Nelson 
Clerk, Select Committee on the Barnett Formula 
House of Lords 
LONDON 
SW1A 0PW

17 February 2009

Dear Audrey

WRITTEN EVIDENCE: SUBMISSION FROM HEAD OF THE NORTHERN IRELAND CIVIL SERVICE

Thank you for extending the invitation to submit evidence to your Select Committee. You will appreciate that the Barnett Formula and it’s outworking is of great interest to the Northern Ireland Executive. I therefore welcome the opportunity to comment on the issues identified in the evidence call note. I have tried to follow the guidance instructions as closely as possible.

1. Application of the Formula in practice:

When referring to ‘present disparities’ in per capita public expenditure allocations across the UK countries I presume the Committee is referring to the Identifiable Public Expenditure (PE) per head data presented within HM Treasury’s annual Public Expenditure and Statistical Analysis (PESA) publication. The trend data from this publication (table below) does suggest that spend per head is indeed higher in the Devolved Administrations but there are some concerns about how this table is compiled in terms of how expenditure is attributed to the regions. These concerns have existed for some considerable time. The extent of HM Treasury mis-allocation of expenditure to the devolved regions is considerable. An example of these anomalies was the inclusion of English and Welsh prison spend in Scottish estimated spend. These anomalies were acknowledged by HM Treasury under challenge from the Scottish Executive. The expenditure estimates for the devolved administrations were then subsequently revised downward.

It is also worth noting that the table highlights evidence of some degree of convergence in Northern Ireland with the UK average – presumably an outworking of the ‘Barnett Squeeze’.

Total Identifiable Expenditure Per Head Index: 2002-03 to 2007-08 ( UK=100)

I would also make the important point that drawing conclusions from this table on whether a region is over or under-provided for in terms of public expenditure allocations is very risky and potentially misleading. The PESA data in no way reflects variations in regional needs. These needs reflect a range of factors such as demographic profiles (which influence spend in key service areas such as education and health), mortality, economic structure, peripherality etc. All these factors vary by region and determine the demand on public services.

The Barnett Formula process accounts for some 90.5 per cent of the Northern Ireland Executive budget. In terms of flexibility to allocate public expenditure, the Devolved Administrations have considerable autonomy. In the Northern Ireland case a Budget is approved by the Assembly which allocates money to spending areas according to the priorities defined in the Executive’s Programme for Government. There are, unfortunately some constraints on flexibility imposed by HM Treasury such as the discretion to switch between current and capital definitions.

2. Formula By-pass and the Barnett Squeeze:

Barnett convergence hastens during periods when public expenditure is growing above long term trend rates. It is therefore not surprising that there has been some degree of convergence over the last decade when the annual average growth rate of public expenditure was 7.6 per cent. The pace of convergence is ameliorated to some extend by receiving Exchequer funding outside of the Barnett process. This ‘bypass’ was a common feature of Northern Ireland allocations when significant resources were received for various EU structural funds / PEACE programmes. The scale of these funds is now diminishing and so, therefore, is the influence of ‘bypass’. One would also expect the pace of any convergence to slow over the coming years as the growth rate in UK public expenditure over the next few years is likely to fall below trend.

3. Data Quality and availability:

The HMT Statement of Funding Policy explains the mechanics of the Barnett Formula in a very transparent manner. There should not therefore be any misunderstanding about what it is trying to achieve over a long term time scale. Where confusion does exist is in understanding how HMT determines the extent of ‘comparability’ in the tables annexed to the Statement of Funding Policy. These tables provide some insight into the degree to which English determined funding is replicated in the Devolved Administrations. The manner in which this data is presented and the inability to source the English expenditure attributions back to specific areas in Whitehall departments is a concern. There remains a concern that some of the comparability percentages are arbitrarily set. Clearly full transparency in the application of the Barnett Formula process requires better information on how the comparability percentages are derived.

One solution to this concern over how comparability is set would be for HMT to publish additional tables which explicitly link expenditure spending areas to clearly identifiable functions within the individual Whitehall departments. To ensure that the whole exercise was conducted in an objective manner, this task could be given to the Office of National Statistics (ONS) who have a statutory duty to remain independent in discharging such functions.

4. Need for reform/alternatives to the existing formula:

The Barnett Formula mechanism does bestow many advantages. It should give the Devolved Administrations greater certainty in terms of the quantum of resources that they have to incorporate within their budgets. It also has a degree of transparency of application (not withstanding the concerns over the comparability factors) and it reflects population variations.

There are however two major concerns over the application of Barnett. The first relates to the fact that the mechanism cannot account for regional variations in need. Issues such as demographic profiles, mortality rates, economic structures, peripherality etc all impose demands on public expenditure in the UK regions. In the Northern Ireland context, there are needs, particularly within the education and health sectors that cost considerably more to meet than the Barnett consequentials deliver.

The second concern about the Barnett Formula relates to the application process. HM Treasury arbitrarily determine the outcome without independent validation. This then allows HM Treasury to arguably abuse the Formula to ensure that bypass results in non-allocation to the Devolved Administrations. A recent notable example of this was the decision to exclude 2010 Olympics funding from the Barnett process.

On the specific issue of reforming or replacing Barnett, in the absence of an alternative process, I would suggest that the Formula process can undoubtedly be improved upon. Ensuring independent validation of its application (i.e. ONS) and creating some form of appeals procedure to allow a challenge to HM Treasury would be helpful.

The difficulty with any needs-based approach to determining public expenditure allocations between regions is reaching agreement on the actual definition of need and how it might be measured. The original HM Treasury Needs Assessment Model (NAS) contained a vast array of need indicators across many spending areas. These indicators then had weightings attached to determine relative priority within the NAS model but this determination was quite subjective in nature. Any future allocation model based on relative regional need will still have to overcome this qualitative allocation of weights.

If such as model is to be constructed, it is imperative that it is done so in a collaborative fashion with the full participation of the Devolved Administrations. Anything produced unilaterally by HM Treasury will be treated with healthy scepticism.

5. Decision making and dispute resolution:

I have already alluded to the rather arbitrary approach adopted by HM Treasury in allocating funds to the Devolved Administrations. There is often a lack of clarity in how funds are attributed to these regions. Furthermore, there is an increasing number of HM Treasury decisions on public expenditure that result in reducing the spending discretion of the Devolved Administrations either by imposing efficiency targets / cuts to baselines or tightening rules on issues such as resource / capital switching, access to end-year-flexibility (EYF) stocks accumulated etc. These all effectively reduce the spending power of the Devolved Administrations – thus undermining the principle of devolved governance.

There is also a concern about the manner in which HM Treasury impose alterations to expenditure allocations. There tends to be little in the way of advance consultation with the Devolved Administrations and even after public announcement, it is often very difficult to obtain a full understanding of what HM Treasury has actually done.

These concerns highlight the need for greater engagement between HM Treasury and the Devolved Administrations, at least at ministerial level, when amendments to financial allocations are proposed. Ideally there should be some form of formal mechanism established to allow for appeal/resolution. This mechanism could then be reflected within an amended Statement of Funding Policy document.

I trust the Select Committee will find this submission helpful in their deliberations.

Yours sincerely

BRUCE ROBINSON
Head of the NI Civil Service

hcc Leo O’Reilly
Michael Brennan