Oxford University Gazette, 20 October 2005
The University's Resource Allocation Method (RAM) (updated for 20056)
1. This note provides an overview of the University's Resource Allocation Method (RAM) for 20056. It outlines the underlying principles, and describes the operational detail. It also highlights changes to the method which have been made in this, the fifth year of its operation.
2. Further details on any aspect of the method may be obtained from the relevant staff within the Planning and Resource Allocation Section in the University Offices: Mr M.D. Sibly (telephone: (2)80303, e-mail: firstname.lastname@example.org); Mr P.F. Clark (telephone: (2)70078, e-mail: email@example.com); or Dr K.J. Dexter (telephone: (2)70563, e-mail: firstname.lastname@example.org). Information on the RAM's operation, and time series of the principal datasets, can also be found on the Planning and Resource Allocation Web site: http://www.admin.ox.ac.uk/pra/.
3. It should be noted, at the outset, that the RAM is only one component of the
University's annual planning and budgeting cycle. The RAM determines the allocation of
income; alongside this, the academic divisions and the services draw up expenditure plans.
Together, these two elements feed into the overall University budget, and also inform the
five-year plans and annual operating statements drawn up by the divisions and services.
4. The Resource Allocation Method (RAM) was introduced in 20012 to complement the University's new governance structures. It is intended to provide a transparent and systematic way of allocating the University's income to its constituent divisionsto whom budgetary responsibility has been devolvedaccording to strategic priorities. The University RAM makes allocations at divisional level: although these represent the aggregate total of lower-level allocations (i.e. to individual departments or faculties), divisions are free to adopt their own internal allocation models which may deviate, to a greater or lesser degree, from the central RAM. Details of divisional RAMs may be obtained from the relevant divisional office.
5. Two of the principal aims of the RAM are to encourage increases in the University's income, and to support the delivery of the University's strategic objectives. These are achieved by building suitable incentives into the RAM, both through the direct allocation of income and through the use of appropriate drivers in the formulaic allocations.
6. Although the design of the RAM broadly reflects the way in which income flows
into the University, it is not a pure 'income-as-earned' model. Thus, the RAM does not
mirror changes in external funding formulae (particularly in the calculation of the HEFCE
grant) if these are inconsistent with Oxford's own strategic priorities. This allows for internal
stability by acting as a buffer between the divisions and the fluctuating external environment.
It also enables the University to redistribute funds towards activities which are deemed to be
academically desirable, but which may not otherwise be financially self- sustainable.
The Resource Allocation Method
7. The basic method by which income is allocated to the academic divisions, 1 services and colleges in 20056 is as follows:
(a) Directly allocate income where appropriate.
(b) Determine the sum available for formulaic allocation.
(c) Top-slice funds for specific purposes (including the College Quantum transfer).
(d) Allocate formulaic funds to divisions (including differential salary cost and transitional adjustments).
(e) Levy infrastructure charge.
(f) Levy capital charge.
8. Although the RAM is often perceived as comprising only the more formulaic
elementsi.e. steps (c) to (f)this is not the case:
the RAM encompasses total University income, and it is important to consider
components within the context of the overall allocation method. The remainder of this note
describes each stage in more detail.
9. The total estimated University income for 20056 is £488.4m. Of this, £294m is directly allocated to divisions and services: this includes both earmarked or specifically-earned income, and certain streams of general income which are allocated directly in order to create particular incentives.
10. The largest single element is research grant and contract income, including all overhead payments received on such grants [20056: £165.6m]. The direct allocation of overheads (which are not earmarked in the same way as the actual grants themselves) is intended to encourage departments to maximise their overhead recovery rates: this will become increasingly important with the advent of Full Economic Costing (fEC) for research. The overheads are intended, in part, to support general research infrastructure (and, under fEC, research sustainability); as such, they should be regarded as contributing towards elements of the central infrastructure and capital charges, which are dealt with below.
11. For 20056, the University received a one-off sum of £9.2m from the Government (through the Research Councils) as part of its Science and Innovation Investment Framework and the move to fEC. This was allocated directly to divisions (outside the RAM formula) pro-rata to a three-year average of Research Council income.
12. A significant proportion of the University's tuition fee income is directly allocated to
divisions: this includes fee income from all overseas students, and Home/EU fee income
from certain specific courses (in particular, those within OUDCE, or those which charge
premium-rate fees). In the majority of cases, the University is able to set the fee rates of the
courses concerned itself: direct allocation is intended to encourage divisions to set them at
appropriate levels, and to ensure that all costs are recovered (including those reflected in the
infrastructure and capital charges).
Determine the sum available for formulaic allocation
13. During the first three years of the RAM's operation, the funds available for distribution to the divisions, services and colleges on a formulaic basis (rather than through direct allocation) were explicitly linked to specific income streams. This changed in 20045, following Council's decision to apply an equal percentage uplift to all three sectors (derived from an analysis of the year-on-year increase in non-specific elements of the HEFCE grant).
14. The principle of an equal uplift for all three sectors has also been adopted in 20056. However, a more budget-led approach has been used to determine the rate of uplift. Whereas the percentage increase applied in 20045 required the injection of some £2m from the strategic reserve, the uplift for 20056 4.1 per centis the maximum which can be supported without drawing on the strategic reserve (it is thus lower than the actual year-on-year increase in non- specific HEFCE grant). This is in line with the University's financial strategy, which states that the strategic reserve should no longer be used to fund recurrent operating costs.
15. Applying this 4.1 per cent uplift produces a total of £162.2m available for
formulaic allocation in 20056.
Top-slice for specific purposes
16. First, a proportion of these formulaic funds is 'top-sliced', or retained centrally; in
20056, this totals £52.9m. There are two components to this: the Quantum
transfer to the colleges [20056: £40.5m]; and a contribution to specific services
budgets which either do not vary in direct relation to academic activity, or have a significant
external or 'heritage' function [2005 6: £12.4m].
17. The remaining £109.3m is then allocated formulaically to the academic divisions.
18. Part of this sum (40 per cent) is allocated according to teaching-related criteria; the remainder (60 per cent) according to research-related criteria. The two-part allocation structure mirrors the HEFCE formula, and the 40:60 split between teaching and research reflects the historic balance of funds within the University's block grant from HEFCE. Changes to the way in which HEFCE funds postgraduate research (PGR) students mean that the 'research' component formed a higher percentage of the 20056 grant; however, in order to maintain internal stability, the 40:60 split has been retained in the RAM.
19. Teaching-related funds [20056: £43.9m] are allocated pro-rata to weighted Home/EU student load. Full-time equivalent student load is distributed between departments according to the teaching apportionment percentages supplied annually by the divisions. The apportioned load is then weighted to reflect broad cost differentials between subjects. New weights have been adopted for 20056: 4.0 for clinical subjects; 1.7 for laboratory- based subjects; 1.3 for subjects with a laboratory or fieldwork element; 1.0 for all other subjects. This change brings the internal RAM in line with the HEFCE formula, which adopted these weightings in 20045. 2 A further weighting of 1.25 is applied to postgraduate taught (PGT) students to reflect the additional costs to divisions associated with their teaching (a result of the fact that most PGT courses run for twelve months, rather than nine months).
20. Four drivers are used to distribute the research-related funds [20056: £65.9m]. These serve as representative proxies for the volume of research activity in a given subject area, and also broadly reflect the basis on which funds are received from HEFCE.
21. The three main volume drivers are weighted to reflect the relative cost and quality of research. Cost weightings mirror those used in the HEFCE research funding formula: 1.6 for clinical and high-cost laboratory subjects; 1.3 for intermediate cost subjects; 1.0 for all other subjects. As in the teaching-related allocation, the cost weights reflect average cost differentials (across the sector), rather than actual costs. Research quality is measured according to a subject's performance in the most recent Research Assessment Exercise (RAE 2001). The weights used in Oxford (which differ from the HEFCE funding formula) are: RAE rated 4 = 2.0; RAE rated 5 = 3.0; RAE rated 5* = 4.0.
22. For each of the minor research volume drivers, the sum distributed through the RAM is equal to the sum generated by the equivalent driver in the HEFCE research formula (QR). In this sense, they can be regarded as 'fixed', or 'ring-fenced' sums. For 20056, the relevant amounts are:
(a) £13.6m is allocated pro-rata to a three-year average 3 of external research grant and contract income received from UK-based charities, weighted to reflect both cost and quality;
(b) £5.8m is allocated pro-rata to a three-year average 3 of non-charitable UK research grant and contract income, weighted to reflect cost and quality;
(c) £873k is allocated pro-rata to the full-time equivalent number of research fellows employed within each department (as returned to HEFCE each year in the Research Activity Survey). Research fellows are defined as 'academic staff who hold specific awards on the basis of their own research record or research proposals'. This allocation is not weighted for either cost or quality.
23. The bulk of the research-related funds [20056: £45.6m] is distributed pro- rata to a composite volume driver: PGR student load , weighted at 0.15, plus HEFCE-funded staff. The resultant full-time equivalent totals are then weighted to reflect both cost and quality. HEFCE-funded staff are defined as: staff returned as research active in RAE 2001 (Category A, or Category A* in post), who were funded either from general University funds or NHS- specific funds. The PGR element includes both Home/EU and overseas students, and has been retained within the 20056 RAM despite changes this year to the way in which HEFCE funds PGR students.
24. For 20056, HEFCE once again awarded additional QR funding to institutions with units of assessment which either gained a 5* rating in both the RAE 1996 and the RAE 2001, or gained a 5* rating for the first time in RAE 2001 without a decrease in research volume compared to RAE 1996. All of Oxford's 5* units of assessment are eligible, generating some £4.4m of so- called 'Best 5*' funds for the University. This is allocated directly to the earning divisions: both the total received and the internal distribution are identical to 20045.
25. An adjustment is then made to the total formulaic allocations (i.e. teaching-related plus research-related) to reflect the differential salary cost to departments of certain types of (joint) academic appointmentin particular the cost of a university lecturer relative to a CUF lecturer or faculty lecturer. A per capita deduction is made in respect of each CUF or faculty lecturer. This is designed to equalise the net unit cost of teaching, irrespective of appointment type, and thereby to remove the financial incentive to convert academic posts from one type of appointment to another. The resultant funds [20056: £5.0m] are used to offset the top-slice. 4
26. Finally, each division's adjusted formula allocation (i.e. teaching-related allocation plus research-related allocation less CUF/FL deduction) is subject to a transitional adjustment (also known as 'moderation'). The aim of this is to enable divisions to move smoothly, over a number of years, from their funding position under the old General Board model to their position as calculated under the RAM. By capping the year-on- year percentage gain/loss in their formulaic allocation, divisions are protected from unexpectedly large changes in funding.
27. When the RAM was first implemented, Council's intention was to phase out the process of moderation within five years: 20056 is therefore the last year in which moderation will be applied.
28. For 20056, moderation is calculated by comparing each division's formulaic
allocation with its 20045 formulaic allocation (including any supplementary funding).
The maximum gain permitted is +6.96 per cent: any excess is used to support 'losing'
divisions, whose loss is capped at 5 per cent.
In line with the principle that the
strategic reserve should no longer be used to used to fund recurrent operating costs, the
upper limit is set at the level required to produce a cash-neutral outcome, whereby the funds
released from 'gaining' divisions is equal to the sum required to support 'losing' ones.
29. A number of other (mostly non-recurrent) allocations, totalling £30.9m, have also
been made to divisions, services and the colleges in 20056. These include strategic
investment in preparation for RAE 2008; Restructuring and Investment Funds (RIF); and
certain central budgets (e.g. specific scholarships).
30. For 20056, the total PRAC budget for the services (i.e. excluding directly- allocated income) is £68.6m, of which £12.4m is funded through the initial top- slice (see paragraph 16). The remaining £56.2m is funded through an infrastructure charge, which is levied on the academic divisions' total incomei.e. both their formulaic and directly-allocated income (see paragraph 10, above, for a discussion of the relationship between directly-allocated research overheads and the infrastructure charge).
31. Each individual service budget is recharged across the divisions using an appropriate
driver which is intended to serve as a broad proxy for usage (e.g., the Careers Service
budget is apportioned on the basis of student load). The principal drivers used for
20056 are: student load, staff load (academic staff only or all staff, as applicable) and
area occupied or maintained.
32. A capital charge is also levied against total divisional income. The charge is based on area occupied (with certain exclusions, such as recently- constructed space funded from external donations) and its purpose is essentially two-fold. First, to encourage the efficient use of space, ensuring that departments do not expand to larger premises without being exposed to the financial consequences which such expansion has for the University as a whole. Second, to promote long-term sustainability and move towards the eventual goal of recovering the full economic cost of infrastructure.
33. The second objective underlies two significant changes to the operation of the capital charge for 20056. First, the rate has been increased from £20 per sq.m. to £40 per sq.m. Second, whereas the funds generated by the capital charge were previously redistributed to the divisions on a (modified) formulaic basis, the £6.3m generated in 20056 has been retained in the University's strategic reserve to contribute towards long- term sustainability. It is likely that further modifications to the capital charge will take place within the context of Full Economic Costing.
34. The position of the services in relation to the capital charge remains under review. In 20056, a notional capital charge was calculated for the services for the first time, although it was not actually levied in practice. However, it is likely that from 20067 onwards the services will become liable to some form of capital charge.
Further development work36. The operation of the RAM is kept under regular review, to ensure that it continues both to support the University's strategic priorities and to reflect adequately how the University receives its income. In the immediate future, it is likely that implementation of Full Economic Costing for research (fEC) will have a significant impact on the RAM. Proposals for these (and any other) changes to the method are brought to PRAC and Council for approval. Work on a Joint Resource Allocation Method (JRAM)covering both the University and the collegesis proceeding in parallel with this.
Footnotes1 Throughout this note, the term 'divisions' includes the Department for Continuing Education (OUDCE), unless otherwise stated.
Return to text
2The previous weights
were: 4.5, 2.0, 1.5 and 1.0 respectively.
This is a change from previous years, when a two-year average was used.
Thus, the £109.3m distributed formulaically comprises: teaching-related funds
[£43.9m] plus research-related funds [£65.9m] plus Best
5* funds [£4.4m] less the CUF/FL adjustment [£5.0m].
As agreed by Council, the funding allocated to the Medical Sciences Division as a result
of its successful bid to HEFCE for Additional Student Numbers (ASNs) is excluded from
moderation and passed to the division in full.
This represents a significant change from 20045: then, gains/losses were capped
at +/5 per cent; and an additional £1.15m was injected from the strategic
reserve because funds released from 'gaining' divisions were insufficient to support the
Supplement (1) to Oxford University Gazette No. 4746. Wednesday, 19 October 2005.