- Armament and disarmament
- Conflict, peace and security
- Peace and development
In November 2020, in the midst of the ‘second wave’ of COVID-19 and despite a projected 10.4 per cent decrease in the United Kingdom’s gross domestic product in 2020, the British Government announced a £16.5 billion budget boost for ‘defence’. The realities of this announcement however remain unclear. The definition of what constitutes the UK’s total military expenditure, continues to be probed in recent years by the House of Commons, think tanks and non-governmental organization experts alike. In particular, it remains opaque why there is a discrepancy between what the British Ministry of Defence (MOD) itself reports as the defence budget (£38 billion for 2019) and what it reports to the North Atlantic Treaty Organization (NATO) as ‘defence expenditures’ (£46.5 billion for 2019).
For its estimate of British military spending, SIPRI uses net cash requirement (NCR) data, as reported in the MOD’s annual reports and accounts since the British fiscal year 2000/2001 (‘the MOD’s NCR figures’). For the years from 1949 until 2000, SIPRI uses NATO defence expenditure data. This enabled SIPRI to publish a consistent time series following accounting changes in British Government spending in the early 2000s. While the MOD’s NCR figures were reasonably close to the UK’s ‘defence expenditure’ figures reported to NATO up to 2012, the two data series began to diverge significantly in subsequent years. In 2019 the difference between the two figures was £8.5 billion.
These discrepancies raise two questions: what are the additional spending items that the UK includes in its NATO defence expenditure figures? Moreover, is SIPRI’s estimate missing parts of British military spending that come from budgets other than the MOD’s NCR?
The MOD’s answer to a recent Freedom of Information (FOI) request indicated that the elements outside the core MOD budget—understood as the MOD NCR—included in the UK’s report to NATO were:
This SIPRI Topical Backgrounder examines these six categories to determine whether they fall under the SIPRI definition of military expenditure; whether they are in fact already included in the MOD core budget (MOD NCR figure) and are therefore already included in SIPRI’s estimate; and whether a consistent data series can be created.
As a result of this in-depth examination, SIPRI has revised its data series for the UK to include four of the six identified additional categories of military spending. This update will take effect in SIPRI’s upcoming military expenditure data launch on 26 April 2021.
SIPRI’s definition of military expenditure includes spending on ‘peacekeeping forces’. In its answer to the FOI request the MOD stated that ‘elements of the Conflict, Security and Stability Fund relating to peacekeeping activities undertaken by Defence’ are included in the UK’s defence expenditures reported to NATO.
The CSSF is an annual fund that allocates resources to various British government ministries to be used in response to fragility and conflict in 84 countries and territories around the world. Among other things, it includes funding for MOD activities related to peacekeeping, such as operations in Cyprus, Somalia and South Sudan. The CSSF annual report states that funding allocated to MOD peacekeeping activities amounted to £197.5 million for the fiscal year 2019/20.
There appears to be another link between the CSSF and military spending. The MOD’s 2019/20 Annual Report and Accounts shows that the MOD’s NCR figure already includes a line item related to the CSSF, amounting to £84.5 million in 2019/20. The figure of £84.5 million matches the MOD’s Departmental Resources statistical series for ‘MOD Operations and Peacekeeping Costs’, which indicates spending of £85 million for CSSF activities in the fiscal year 2019/20. However it remains unclear as to how this figure relates to the £197.5m reported in the CSSF annual report.
Only the $197.5m figure (the figure reported in the CSSF annual report) will be included in SIPRI’s revised estimate of British military spending. It remains uncertain as to whether or not the $197.5m includes the £84.5m figure reported in the MOD’s 2019/20 Annual Report and Accounts. The CSSF is an additional line item which is not part of the MOD NCR and was not previously included in SIPRI’s estimate.
The inclusion of this additional element should go back to 2001, when predecessor mechanisms for the CSSF were first created. The estimates from 2001–2020 will thus be revised to include this element. For the years 2015–19, information on CSSF allocations is available in ministers’ written statements to the Parliament and in CSSF annual reports. However, prior to 2015, the UK’s cross-government aid and peacekeeping spending was organized under a funding instrument known as the ‘Conflict Pool’. The annual reports for the Conflict Pool do not provide expenditure breakdowns to the same level of detail as the CSSF reports since 2015.
Due to the difficulty in reconciling the CSSF with the Conflict Resources Settlement (the funding mechanism for the Conflict Pool), SIPRI’s new assessment uses two different estimates, one for 2011–14 and one for fiscal years 2001/2002 to 2010/11, based on data availability.
In years prior to 2015, MOD operations accounted for 15 per cent of the total CSSF for 2015–19. For the years 2011–14, where data for the Conflict Resources Settlement exists, 15 per cent of that total is used. Another estimate has to be made for the years from 2001 until 2010, since no data series was uncovered by SIPRI for the Conflict Resources Settlement in this period. This is done by using the average 0.24 per cent of total British military spending that the Conflict Pool funding accounted for during 2011–15.
SIPRI includes the military activities of intelligence services in its definition of military spending. The UK’s three intelligence agencies are distinct from the MOD and are funded through the SIA budget. The SIA’s NCR figure is available in the British Treasury’s main supply estimates, but is not part of the MOD’s core budget. However, there is no detailed information as to how the allocated amount is distributed between MI5 (the Security Service), MI6 (the Secret Intelligence Service) and GCHQ (Government Communications Headquarters). These intelligence agencies engage in a range of both military and civilian activities. For SIPRI to include part of this budget in its military expenditure figures for the UK, an estimate of the proportion of military activities for these three agencies as against their civilian activities must be made.
MI5 is mainly concerned with internal security, such as counterterrorism, counterespionage and the protection of critical infrastructure. It should therefore be entirely excluded from the scope of British military spending. MI6 performs both civilian and military activities, with the latter including counterterrorism and the disruption of ‘hostile state activity’. GCHQ’s missions include counterterrorism; cybersecurity; ‘managing threats from hostile states and shaping the international environment’; countering serious and organized crime; and support to defence. Given their mission statements, an assumption can be made that about 50 per cent of the respective activities of MI6 and GCHQ are military in nature.
The distribution of resources between these three intelligence agencies can be estimated based on an assessment of their respective personnel numbers. The Intelligence and Security Parliamentary Committee’s annual reports provide the number of staff per agency. For the years where data was available, MI5 employed on average 31 per cent of total SIA staff, MI6 employed 23 per cent and GCHQ employed 46 per cent. Assuming that the budget distribution is equivalent to that of personnel, and based on the assumption that 50 per cent of the respective activities of MI6 and GCHQ are military in nature, SIPRI assesses that 34.5 per cent of the total SIA budget should be added to the British military spending estimate. In 2019 this amounted to £1.1 billion or 2.4 per cent of total British military spending.
SIPRI’s definition of military expenditure includes ‘retirement pensions of military personnel’. In the British system, military pensions are divided into two types. On the one hand, there is the ‘War Pensions Scheme’, which is already part of the MOD’s core budget and therefore included in SIPRI’s estimate. On the other hand, the ‘Armed Forces Pension & Compensation Schemes’ has a separate budget entity and is not included in the MOD’s core budget figure. Spending on the AFPS should be added to SIPRI’s estimate of British military expenditure. The data for the AFPS is available (in NCR figures) in the yearly main supply estimates. In 2019 AFPS spending was £1.4 billion accounting for 3.0 per cent of total British military spending.
SIPRI’s definition also encompasses pensions for civilians who have been working at the MOD, as their service contributed to the UK’s military activities. There is no indication as to which budget the MOD’s civilian pensions originate from. According to the MOD’s Departmental Resources statistical series, the MOD’s ‘resources DEL’ for ‘civilian personnel cost’ include pensions contributions. This would suggest that civilian pensions are already included in the MOD’s core budget. This is also noted in the Main Estimates memorandum for 2019–20. Moreover, this memorandum indicates that in fiscal year 2019/20, the MOD’s resources DEL received additional funding of £708 million ‘to reflect the increase in employers’ contributions for the current valuations of both military and civilian pension schemes’. At the same time, the MOD’s answer to the FOI request mentioned above stated that civilian pensions are added to the MOD budget in the NATO defence expenditure report. This would mean that civilian pensions are not already in the MOD’s core budget.
However, no available official data could be uncovered by SIPRI. A 2015 report by the Royal United Services Institute (RUSI) indicated that civilian pensions for the MOD were ‘perhaps around £200 million’ but this was not verifiable by SIPRI. Due to the uncertainty of the data, SIPRI will not include an estimate for MOD civil service pensions in its revised assessment.
The MOD generates some income by providing services to other organizations. This includes military support to foreign governments or other British government departments. Other sources of income are rental income, proceeds from the sale of property and equipment, dividends, and income from investments.
This data is available in the MOD’s departmental resources statistical series published yearly by the MOD back to fiscal year 2005/2006. For previous fiscal years (2000/2001 to 2005/2006) the figure is available under ‘operating income’ in archived MOD annual reports and accounts. In 2019 income generated by the MOD amounted to £1.7 billion or 3.7 per cent of total British military spending. Assuming that this income is then re-used by the MOD for military activities, it serves the purpose of funding military activities and is included in SIPRI’s revised estimate of British military spending.
Expenditure on military operations falls under SIPRI’s definition of military expenditure. In its budget request each year, the MOD includes initial costs of military operations. As operations unfold during the British fiscal year (from April to March each year), they generate additional costs. These ‘net additional costs’ are funded by the HM Treasury Special Reserve. However, as shown by the MOD’s annual report and accounts, these additional costs are already included in the yearly MOD NCR (in tables entitled ‘Statement of Parliamentary Supply (SoPS) Note 1.1 – Analysis of Net Resource Outturn, and Note 3 – Reconciliation of the Net Resource Outturn to Net Cash Requirement’).
Adding the net additional cost of operations to the MOD’s NCR would therefore lead to double counting. Thus, this category of spending is not included in the new SIPRI estimate.
In the 2015 Strategic Defence and Security Review (SDSR) the MOD announced the Joint Security Fund (JSF)—a shared fund between the MOD, the Foreign and Commonwealth Office, the Department for International Development and the intelligence agencies. The MOD stated that the JSF would be used ‘to fund and deliver the MOD’s SDSR commitments in full, maintaining the current levels of the Armed Forces and building four new submarines to renew the nuclear deterrent’. However, there is little available information on it and no additional information could be found by SIPRI on the origins of the fund or how it is spent.
A commentary by RUSI in 2019 noted that: ‘the £41 299 million settlement for 2020/21 includes Joint Security Fund allocations which, at Main Estimates, are due to be transferred out of the MOD to the SIA cyber and other non-MOD budgets. The 2019/20 equivalent to this transfer, projected to amount to some £480 million for 2020/21, has already been made in the baseline figures for 2019/20 used in Spending Round 2019.’ This would mean that the total JSF is first included in the MOD’s total spending, and then part of it is transferred to the other departments.
The 2015 Spending Review stated that, over five years, the MOD would receive £2.1 billion or £420 million per year from the JSF and the SIA £1.3 billion, with £100 million saved for reserves. However, SIPRI could not locate a data series that provided a breakdown of the JSF by annual spending by department or by spending category.
Given the uncertainty about whether the JSF is initially included in the MOD’s NCR and the lack of information on annual departmental spending, there is insufficient available data to make a yearly estimate of the JSF. It is therefore not included as a category in SIPRI’s new estimate of British military expenditure.
Table 1 shows the components of the new estimate as outlined above: the MOD’s NCR; elements of the CSSF dedicated to MOD peacekeeping operations; estimates of military intelligence expenditure based on the SIA NCR; the AFPS’s NCR; and MOD external income.
|MOD’s NCR/SIPRI’s old estimate||41481||The MOD’s annual NCR as reported in the main supply estimates.|
|Peacekeeping operations and the Conflict, Stability and Security Fund||198||The portion of the CSSF dedicated to MOD involvement in peacekeeping operations, as reported in the CSSF annual report.|
|Military-related spending by the security and intelligence agencies||1085||This represents 34.5 per cent of the SIA NCR as reported in the main supply estimates. The figure is based on the average share of total personnel of MI5, MI6 and GCHQ and the assessed proportion of military-related activities.|
|Expenditure on Armed Forces Pension & Compensation Schemes||1369||The AFPS NCR as reported in the main supply estimates.|
|Civil services pensions||–||Not added to the new estimate. No sufficient information available.|
|External income generated by the Ministry of Defence||1700||Based on data for the last available year, as reported in the MOD Departmental Resources Financial Bulletin.|
|Additional costs of operations||–||Not added to the new estimate. This category of spending is already included in the MOD’s NCR.|
|Joint Security Fund||–||Not added to the new estimate. No sufficient information available.|
Notes: – = figure not added to new SIPRI estimate; AFPS = Armed Forces Pension and Compensation Scheme; CSSF = Conflict, Stability and Security Fund; GCHQ = Government Communications Headquarters; JSF = Joint Security Fund; MOD = Ministry of Defence; MI5 = Security Service; MI6 = Secret Intelligence Service; NCR = Net Cash Requirements; SIA = Security and Intelligence Agency.
The new SIPRI estimate is higher than the old one, which uses only the MOD NCR, but remains slightly lower than the figure in the UK’s report to NATO for 2019 (£46.5 billion). This difference could be explained by the fact that SIPRI excludes civilian pensions, the JSF and the net additional cost of operations from its revised estimate, which the MOD stated are added to its NATO data.
The new estimate takes into consideration more categories that do fall under the SIPRI definition of military expenditure and therefore can be seen as an improvement to the data series. The full new data series will be available with the upcoming release of the updated SIPRI Military Expenditure Database, on 26 April 2021.
This SIPRI Topical Backgrounder shows that, even in a highly developed democracy (such as the UK), which releases numerous budgetary documents, it is not always possible to fully estimate total military spending. This makes it more difficult to assess opportunity costs at a time when, in a possibly more fiscally constrained environment after Covid-19, the allocation of resources—in particular to the healthcare system—will come under greater scrutiny.