Select Committee on Science and Technology First Report


4  SUPPORT FOR RESEARCH

Funding modes

20. ESRC told us in its memorandum that the "balance between directed and responsive mode funding is approximately 2:1". This contrasts with the roughly 1:2 ratio of managed: responsive mode funding at NERC, EPSRC and BBSRC, although ESRC suggests that this may partly be due to the fact that it is "far more definitive and limited in its definition of the responsive mode" than the other Councils.[41],[42] Professor Diamond admitted to us in oral evidence that "an increased proportion of our funding in the last few years has gone into the directed mode" and suggested that "the reason for that is that a large amount of the increase in budget that we have had over the last few years has been to engage in cross-Council directed programmes, for example on the rural environment and on energy".[43] Professor Diamond also argued that some of ESRC's programmes "are effectively responsive mode", describing the cognitive science programme being run in conjunction with EPSRC, BBSRC and MRC as "a responsive mode programme which is steered".[44]

21. Mr Davies also pointed out that ESRC spend on responsive mode was starting to rise: "in the coming year we will be increasing our spend on responsive mode from £18 million in the last year to something like £23 million".[45] Nevertheless, the stakeholder consultation on ESRC's new strategic framework revealed "overwhelming support for a shift to support of research through more bottom up, responsive provision and away from what is perceived as undue central prescription of research".[46] Professor Diamond acknowledged this and indicated that ESRC was exploring the most effective way of increasing responsive mode provision: "It is not quite as simple as saying we will put more money into grants between a quarter and a half million pounds. We also want to give the opportunity to have some bigger response mode grants addressing really big questions".[47] ESRC's 2:1 ratio of directed: responsive mode funding is out of step with other Research Councils and is unpopular with its research community. We recommend that ESRC respond without delay to the calls from its community to increase significantly the proportion of responsive mode funding.

RESEARCH CENTRES

22. ESRC provides ten year funding commitments to support Research Centres at a single location with the aim of fostering:

the development of partnerships with stakeholders.[48]

Examples of current Centres include the Centre for Business Relationships, Accountability, Sustainability and Society (BRASS) at Cardiff University, the Centre for Competition Policy (University of East Anglia) and the Centre for Public Organisation (Bristol University).

23. After the initial ten year funding commitment is complete, Research Centres can reapply for continued funding for further five year periods. Professor Diamond indicated that the ESRC council believed "closing something down simply because you can is not a very effective way to take forward research".[49] ESRC had therefore decided that "every five years a Research Centre can apply to continue in a proper market test […] Where the agenda for research is still vibrant, the leadership is still strong and peer review is such that it is important that it is funded as a Centre and it is world class then we will continue to fund on an on-going basis".[50] Nevertheless, there was an expectation that Centres should seek alternative funding sources and of the 21 Centres that ESRC has supported in the last decade, five Centres were granted a further period of ESRC funding whilst the remaining 16 had mostly secured support from other sources to enable them to continue, primarily within universities.[51]

24. Various memoranda questioned ESRC's use of this model of research funding. The Political Studies Association, for example, told us: "There needs to be continual and rigorous scrutiny of the value for money offered by research centres compared to other forms of funding. There is a danger that they may reinforce a fashionable orthodoxy rather than stimulating new and innovative thinking".[52] The British Psychological Society was of a similar opinion, telling us: "ESRC should reduce the number and/or nature of the research centres that it supports. The current concentration of its resources into 26 centres (11 of which focus on economics) in a limited number of institutions will lead to severe restrictions in research capacity and recruitment difficulties in the UK as a whole. It is vital to ensure the breadth of social science in the UK and to protect its broad institutional, regional, and departmental base".[53] Similar views were expressed by a number of organisations in response to ESRC's consultation on its new strategic framework.[54]

25. Thirteen per cent of total ESRC expenditure in 2003-04 was allocated to Research Centres, compared with 18% on managed programmes and 18% for responsive mode grants and fellowships.[55] The proportion of ESRC research funding allocated to Research Centres has increased from 16.6% in 2001-02 to 19.2% in 2003-04.[56] ESRC told us that the Research Evaluation Committee had overseen comparative assessments of different approaches to research funding and said that "recent evaluations have highlighted the particular achievements of our centres in relation to the development of dynamic and responsive research agendas, capacity building, career development, sustained user-engagement and the leverage of external co-funding".[57]

26. ESRC, like all the Research Councils, has to make decisions about how best to distribute its limited funding. In the light of the low amounts of funding currently allocated to responsive mode grants and the low success rates for some of these competitions (see paragraphs 27-28), we find it surprising that ESRC continues to channel such a significant proportion of its funds into ten year commitments to Research Centres. We recognise the value in providing research groups with long-term, predictable funding, but do not believe that ESRC can afford to continue to fund such a large cohort of Research Centres. Moreover, whilst we support the policy of seed funding Research Centres, and note that most have continued once ESRC funding has been withdrawn, we believe that they should be left to stand on their own feet from an earlier age. This would also bring ESRC closer into line with other Research Council initiatives, such as EPSRC's Portfolio Partnerships and Platform Grants which provide funding for five years. ESRC should increase the funding available for responsive mode applications by reducing the size of its contingent of Research Centres. In addition, ESRC needs to consider whether a shorter time frame, such as five years, for Research Centre funding would give better value for money. The effect of Research Centres on the concentration of research is discussed in paragraph 33.

Grant application success rates

28. Success rates for responsive mode award applications vary significantly between disciplines (see Table 3). For example, in 2003-04 the success rate for linguistics applications was 54.5%, whilst for interdisciplinary studies and education the success rates were 16.7% and 15.3% respectively. ESRC told us that it had recently commissioned research looking at the reasons behind the variable success rates for applications in management (very low success rate), sociology (medium rate) and linguistics (high rate) in order to establish whether there was anything in its application system which was "biasing the outcome".[62] Although the results indicated that the system was not inherently weighted in favour of any particular subjects, ESRC told us that it did emphasise the need "to ensure we had a mix of referees who were both focussed on the very specific subject and had a broader view". A correlation was also identified between subjects with low success rates and those which did "less well in getting high graded departments in the RAE".[63]

29. In our Report on the work of BBSRC, we urged BBSRC to address its falling success rates by better management of demand for research grants. It appears that ESRC has already taken some steps to manage demand for its awards and we note that ESRC publishes grant application rates for individual institutions—one of our recommendations to BBSRC.[64] Mr Davies told us that ESRC had been concerned in the 1990s about the low levels of applications receiving an alpha rating: "At that time, the split was roughly about 60:40 [alpha: non-alpha rated proposals]. We engaged in discussions with universities. We changed our system of application to an open date system so that the reader had a chance to look at the application before they submitted it to us. That did work [and] over a period of about 18 months [resulted in] this 80:20 ratio".[65] Mr Davies also recognised the need to address specifically the quality of applications in subjects which had low success rates, telling us: "the AIM initiative and the Teaching and Learning Research Programme are working on the other side of ESRC in how we improve the quality of applications coming from, for example, management and education", where success rates are below average.[66] We are pleased to find that ESRC is attempting to tackle the low success rates for responsive mode grant applications by managing demand and improving the quality of applications. ESRC must now significantly enhance the funding available for responsive mode grants. Until it does so, overall success rates are unlikely to improve significantly.

Table 3: ESRC Responsive Mode Grant Awards by Discipline 1999-2000 to 2003-04

  
APPLICATIONS
SUCCESS RATE %
DISCIPLINE99/0000/01 01/0202/0303/04 99/0000/0101/02 02/0303/04
Area Studies76 636 0.066.750.0 0.033.3
Economic and Social History27 323035 2729.653.1 43.340.044.4
Economics6056 6075112 30.046.440.0 36.041.1
Education 6562 726059 32.327.427.8 18.315.3
Environmental Planning15 14121417 33.321.450.0 28.623.5
Human Geography4038 424546 32.528.947.6 22.234.8
Interdisciplinary Studies0 32414 240.00.0 16.742.916.7
Linguistics2217 182533 31.841.244.4 32.054.5
Management and Business Studies55 346138 5412.720.6 16.415.822.2
Political Science and International Relations 503636 476018.0 50.033.323.4 23.3
Psychology140137 167152180 24.335.026.9 29.642.2
Social Anthropology30 27263020 33.344.434.6 33.340.0
Social Policy3442 202732 14.726.240.0 18.528.1
Socio Legal Studies16 14161210 6.335.731.3 41.750.0
Sociology9068 749585 27.830.935.1 26.335.3
Statistics Computing and Methodology17 9128 723.522.2 50.050.028.6
TOTAL668595 676680772 2535.132.4 28.134.6

Source: ESRC

INTERDISCIPLINARY RESEARCH

30. We were particularly concerned to note that success rates for applications in interdisciplinary studies had fallen from 42.9% in 2002-03 to 16.7% in 2003-04 and asked ESRC whether this reflected difficulties with the assessment of these applications. Mr Davies described such applications as "an odd collection" for which the applicant could not identify a single discipline, or a mix of disciplines to describe their work.[67] Professor Diamond also pointed out that the Research Councils had recently reviewed their approach to interdisciplinarity and had come to the conclusion that they "did not have any bias across the Research Councils in the way we dealt with interdisciplinarity" although there was room for improvement in the way they publicised their approach to it.[68]

31. Nevertheless, ESRC's consultation on its new strategic framework clearly indicated a need for the Council to "make clear its position on supporting research within and across disciplinary boundaries" and "reduce actual and perceived barriers" to supporting interdisciplinary research.[69] We wholeheartedly endorse this assessment. Interdisciplinary research can engender ground-breaking developments but researchers in these areas frequently complain of difficulties in securing funding. It is not sufficient that ESRC is content with its procedures for assessing interdisciplinary applications: the research community must also be convinced that these applications will be given proper consideration. If this does not happen, the perceived lack of a level playing field will act as a disincentive for researchers to submit interdisciplinary applications, or indeed to engage in interdisciplinary research.

FULL ECONOMIC COSTS

32. The Government has recently announced that the Research Councils are to move to a funding model which requires them to pay the full economic costs of the research they fund.[70] At present, Research Councils pay 46% of the direct staff costs for a research grant. It is expected that the Research Councils will need to pay in excess of 60% of the direct staff costs in order to meet the full economic costs.[71] ESRC remarked on the fact that the changes entailed in the move towards full economic costs may also have implications for the success rates of grant applications: "We welcome the greater transparency the introduction of full economic costs will bring. However it is possible that this will lead to a considerable increase in the number and costs of grant applications received by ESRC, given the size of our community and the likelihood that a research council award will cover a greater proportion of costs than funding from some other sources".[72] The impact of the move towards full economic costs on grant application success rates needs to be carefully monitored at a cross-Council level.

Research concentration

33. ESRC stated in its memorandum that, in line with the other Research Councils, it does not have a "policy in favour of either institutional dispersal or concentration".[73] In fact, although ESRC currently funds research at over 120 different institutions, 44% of ESRC funding goes to ten institutions, as illustrated in the Table 4. ESRC's funding of Research Centres is likely to contribute to the concentration of research funding (see paragraphs 22-26), and the move towards the quota system for PhD studentships (see paragraphs 53-57) will undoubtedly reinforce this situation.

Table 4: Concentration of ESRC funding

Overall Expenditure on Research and Training 2003-04 (£k)
%
1.University of Essex
6,806
7.3
2.London: School of Economics and Political Science
5,204
5.6
3.University of Oxford
4,553
4.9
4.University of York
4,111
4.4
5.University of Manchester
4,076
4.4
6.University of Cambridge
4,023
4.3
7.University of London: University College
3,423
3.7
8.University of Sheffield
2,965
3.2
9.Cardiff University
2,903
3.1
10.University of Warwick
2,816
3.0
Total
40,880
43.9

Source: ESRC

34. We have expressed our concern over the effects of research concentration in previous Reports on the work of the Research Councils and, most recently, in our Report on the Research Assessment Exercise.[74] A review conducted by the Science Policy Research Unit at the University of Sussex reported "little if any convincing evidence to justify a government policy explicitly aimed at further concentration of research resources on large departments […] on the grounds of superior economic efficiency" and the Lambert Review emphasised the importance of local universities for the development of a healthy regional (and therefore national) economy.[75],[76] A study commissioned by Universities UK also demonstrated the value of investment in departments with RAE scores of 3 or 4 for enhancement of the performance of the research base.[77] The ten year investment framework for science and innovation published in July 2004 outlined a provision for the Higher Education Funding Council for England (HEFCE) to intervene in cases where departments are threatened with closure.[78] ESRC needs to acknowledge the role that it plays in the increasing concentration of research in a small number of institutions. We recommend that ESRC co-ordinates its activities with the other Research Councils and with HEFCE to ensure the continued funding of excellent social science across a wide range of departments, institutions and regions. This matter is discussed further in paragraph 66.

Grant administration

35. During his introductory session with the Committee on 22 January 2003, Professor Diamond commented on the fact that ESRC staff were struggling to cope with the administrative burden: "The one thing I hear from everywhere, all over the country from academics is that the ESRC staff are stretched to the limit […] it is absolutely clear in my mind that within the Council at the moment people are working flat out".[79] The Development Studies Association also told us that "with this new strategic framework and with increased funding it is important that the ESRC's administrative capacity increases to meet the challenges. As it stands there is a perception that the ESRC staff are fully stretched in seeking to ensure the quality and effectiveness of their systems. Any increase in ambition with this new framework must be matched by a realistic increase in administrative capacity".[80] In addition, it has been stated that administrative convergence of the Research Councils should result in a 10% reduction in ESRC recurrent costs .[81] ESRC, like all the Research Councils, aims to keep its administrative costs at less than 4% of its total expenditure and is expected to meet this target in this financial year.

36. We asked ESRC whether it expected to have to increase its administrative capacity. Professor Diamond told us: "If we increase our budget significantly, I think it is only likely that we would have to increase the number of people to administer that".[82] Whilst we do not dispute that there may be a need to expand the numbers of staff at ESRC, it does not logically follow that increasing the budget will automatically necessitate a larger body of staff.

PEER REVIEW

37. During our scrutiny of the various Research Councils we have encountered a number of problems with the peer review systems being employed. ESRC utilises different peer review systems depending on the size of the grant. Small grants (below £45,000) are reviewed using a streamlined process that involves members of both a virtual college of reviewers and the Research Grants Board, which comprises 24 academic members and two "user" members drawn from the public or private sector, covering a range of different disciplines. We support ESRC's decision to use a streamlined peer review process for small grant applications.

38. Large grants (above £45,000) enter into a three stage process. In the first instance they are sent for review to individual researchers with relevant expertise. The number of reviewers used at this stage tends to vary between three and five depending on the size of the grant. If the average grading awarded to the proposal by the reviewers exceeds a certain threshold it will pass to the next stage: consideration by two members of the Research Grants Board. If successful at this stage, the proposal will then be reviewed by the Research Grants Board as a whole. Thus ESRC uses neither the committee system favoured by BBSRC for reviewing grant proposals, nor the peer review college system employed by NERC and EPSRC.

39. Applicants can nominate two reviewers, of which at least one will be selected to review the proposal. This is similar to the system used by EPSRC, whereby at least one out of the three reviewers requested by the applicant will be used. Although reviewers' comments are generally made available for the applicant and other reviewers to see, ESRC does not normally give applicants the chance to respond to these comments. ESRC has recently commenced a trial to give applicants an opportunity to respond to the referees' comments on grant applications for more than £200,000. EPSRC already has a policy in place to allow applicants the chance to give feedback on referees' comments. We note that there are still a number of discrepancies in the peer review processes utilised by the various Research Councils. It is not clear why best practice in peer review should vary significantly between the Research Councils. Moreover, harmonisation of the peer review systems used by the Research Councils should facilitate the evaluation of applications for interdisciplinary research that fall within the remit of more than one Research Council.

Impact of the Research Assessment Exercise

40. ESRC identified in its written memorandum four main challenges that it was likely to face over the next five to ten years, the first of which was:

Professor Diamond then picked out development studies and education as two disciplines that had been particularly adversely affected by the RAE.[84] We also heard that lack of appropriate recognition in the RAE was acting as a disincentive for researchers to undertake knowledge transfer activities. However, Professor Diamond seemed optimistic that previous problems had been resolved, telling us "we support the plans [for the 2008 RAE] as they are".[85] ESRC also noted that the Research Councils had been acting collaboratively to address problems with the RAE and would continue to do so. In our recent Report on the RAE, we called for panels to give "equal weight to pure and applied research" and urged HEFCE to ensure that "this is understood by everyone".[86] We also called for the RAE to "recognise that excellent research may not be internationally significant but it may transform the fortunes of a local business or the provision of public services".[87] We are pleased that HEFCE has accepted these recommendations in principle; it must now ensure that these are put into practice. ESRC must continue to work closely with the other Research Councils and the Funding Councils to reduce the deleterious side-effects of the RAE and encourage the RAE subject panels to be open and clear about what they will be measuring and how they intend to do it.

Research resources

41. Another of the major challenges identified by ESRC for the next five to ten years concerned data access and usage. ESRC described data resources to us as "the very lifeblood of economic and social research" and Professor Diamond was keen to point out that the costs of building and maintaining a world class data infrastructure mean that "Social science is no longer a cheap option".[88],[89] In 2003-04 ESRC spent £11.62m on resources and, following a major review of its longitudinal datasets, has agreed to commit an additional £22m over the next six years to further sweeps of the British Household Panel Study, the National Child Development Study, the 1970 Birth Cohort Study and the Millennium Cohort.[90] ESRC also asserted that "if the UK is to maintain its position at the leading international edge, then it needs to do more than simply commit to the long term support of existing data sets".[91] The collection and analysis of major new datasets would clearly require significant additional funding.

42. The British Psychological Society emphasised the national benefit that could be gained through these research resources, commenting that ESRC "collects and analyses data, not only to discern patterns of behaviour but also to identify processes and behaviours that could contribute to informed and evidence-based decisions on the effectiveness of specific interventions and initiatives on, for example, reducing drug-related crimes".[92] However, the British Psychological Society expressed concern that ESRC's "vital role in underpinning evidence-based policies" was "not fully understood in the 'corridors of power'", accounting for "the relatively small financial support that the ESRC receives from the Government".[93]

43. The Department for Trade and Industry (DTI) and Department for Work and Pensions (DWP) do at least seem to recognise the importance of national datasets. DTI told us: "Some of the projects funded by the ESRC provide evidence for policy appraisal. Examples include the Workplace Employee Relations Survey which is co-funded with DFES, ESRC and DTI".[94] DWP also noted its intention to increasingly exploit "the capacity to utilise government administrative databases and major surveys data in its analytical work for policy purposes".[95] DWP further told us that it was "considering with ESRC the possibility of opportunities for co-funding of longitudinal data collection".[96] Longitudinal studies and the collection and maintenance of national datasets are essential to build an evidence base to inform effective policy development. Such work is, however, costly and it is unreasonable to expect ESRC to fund these major activities out of its modest budget. Where a dataset is of particular relevance to a Government Department, the Department should shoulder the majority of the financial burden, whilst taking advantage of ESRC's skills and experience. We recommend that ESRC pursues this issue with the relevant Departments. In addition, given the reliance on these datasets by both Government and the research community, we recommend that future Spending Review allocations ensure that the collection and maintenance of national datasets are fully funded to prevent ESRC having to cut funding for other research activities to preserve these important statistics.


41   Excluding core funding to institutes for BBSRC. Back

42   New Stakeholder Framework: Consultation with Stakeholders, Preliminary Analysis and Discussion, ESRC, 2004 [not published] Back

43   Q 35 Back

44   Q 35 Back

45   Q 35 Back

46   New Stakeholder Framework: Consultation with Stakeholders, Preliminary Analysis and Discussion, ESRC, 2004 [not published] Back

47   Q 35 Back

48   Ev 20 Back

49   Q 36 Back

50   Q 36 [Professor Diamond] Back

51   Ev 39 Back

52   Ev 34 Back

53   Ev 37 Back

54   New Stakeholder Framework: Consultation with Stakeholders, Preliminary Analysis and Discussion, ESRC, 2004 [not published] Back

55   Ev 18 Back

56   Ev 27 Back

57   Ev 40 Back

58   Ev 20; the figure of 831 refers to responsive mode research grant and research fellowship applications in 2003-04. 772 of these applications were for research grants. This explains the apparent discrepancy between the figure cited here and the total number of applications given in Table 3. Back

59   Ev 20 Back

60   Ev 20 Back

61   Q 60 Back

62   Q 65 [Mr Davies] Back

63   Q 65 [Mr Davies] Back

64   Third Report of the Science and Technology Committee, Session 2003-04, The Work of the Biotechnology and Biological Sciences Research Council, HC 6 Back

65   Q 61 Back

66   Q 65 [Mr Davies] Back

67   Q 67 Back

68   Q 67 Back

69   New Stakeholder Framework: Consultation with Stakeholders, Preliminary Analysis and Discussion, ESRC, 2004 [not published] Back

70   HM Treasury, DTI and DfES, Science & innovation investment framework 2004 - 2014, July 2004 Back

71   Eleventh Report of the Science and Technology Committee, Session 2003-04, Research Assessment Exercise: a re-assessment, HC 586 Back

72   Ev 20 Back

73   Ev 19 Back

74   HC (2003-04) 586 Back

75   SPRU, The Effects of Size on Research Performance: A SPRU Review, June 2003 Back

76   HM Treasury, Lambert Review of Business-University Collaboration, 2003 Back

77   Evidence Ltd, Funding Research Diversity: The impact of further concentration on university research performance and regional capacity, 2001 Back

78   HM Treasury et al, Science & innovation investment framework 2004-2014, July 2004 Back

79   Science and Technology, Minutes of Evidence, ESRC Introductory Session, Session 2002-03, HC 277-i Back

80   Ev 37 Back

81   Ev 26 Back

82   Q 80 Back

83   Ev 26 Back

84   Q 78 Back

85   Q 79 Back

86   HC (2003-04) 586 Back

87   HC (2003-04) 586 Back

88   Ev 21 Back

89   Q 30 Back

90   ESRC, Annual Report and Accounts 2003-04, 2004 Back

91   Ev 21 Back

92   Ev 37 Back

93   Ev 37 Back

94   Ev 17 Back

95   Ev 33 Back

96   Ev 33 Back


 
previous page contents next page

House of Commons home page Parliament home page House of Lords home page search page enquiries index

© Parliamentary copyright 2004
Prepared 20 December 2004