PUBLIC ADMINISTRATION AND DEVELOPMENT

Public Admin. Dev. 23, 445^56 (2003)

Published online in Wiley InterScience
(www.interscience.wiley.com) DOI: 10.1002/pad.292

THE PERFORMANCE ERAMEWORK: A SYSTEMS APPROACH

TO UNDERSTANDING PERFORMANCE MANAGEMENT
DOUGLAS SALTMARSHE*, MARK IRELAND AND J. ALLISTER McGREGOR
University of Bath, UK

SUMMARY
The articie presents a framework for better understanding the nature of performance in organisations involved in the provision
of overseas development assistance (ODA). It uses a case study to illustrate the three main features ofthe framework which are:
goals, performance assessment and performance management. It is asserted that a vibrant performance culture is one which
links these features together to form an organisation capable of continual improvement through producing effective learning.
Organisational culture determines the nature of linkages between the three sub-systems. The article stresses how the notion of
performance may extend beyond various forms of evaluation and scrutiny to being part of a sentient learning system rooted in an
organisation's culture and structure. The article concludes with consideration of key issues associated with the generation of a
reflexive learning organisation operating in the ODA sector. These are concemed with understanding the role and nature of
systems, organisational vision, the embracing of diversity, training and accountability. Copyright © 2003 John Wiley & Sons,
Ltd.

INTRODUCTION
This article presents a framework which has been generated to better understand the notion of organisational per-
formance within the context of intemational aid. It derives from the study of and interactions with organisations
concemed with the delivery of overseas development assistance (ODA). This analytical approach to performance
has relevance for organisations across the aid sector, be they large or small, donors, multilaterals or NGOs. We also
believe it may have wider relevance for different types of organisations operating in complex policy arenas, char-
acterised by multi-agency responsibility for performance.
The article uses as a case study the Department for Intemational Development (DFID), the British govemment
department responsible for the delivery of ODA. Since the UK 1997 White Paper on Intemational development,
DFID has played an increasingly significant role intemationally in driving for greater accountability and effective-
ness in the delivery of ODA. In recent years, it has intemalised these concerns by strengthening its focus on per-
formance. This article draws upon an empirical research commissioned by DFID in 2002 which was concemed
with the role of training in strengthening performance management through training and staff development. The
study was grounded in a review of relevant literature and then entailed a series of interviews across the organisa-
tion, as well as the use of focus groups and a questionnaire.
In seeking to make sense of the notion of performance as it is currently being promoted, we found that it was all
too easy to drift into an obscurantist discourse where meanings are only readily understood by initiates. A challenge
therefore was to clarify for ourselves the implications associated with the drive for performance in these organisa-
tions. Performance has been a prominent consideration in the management systems of public organisations over the
past two decades, but has been slower to filter into intemational development. A key element in the accelerated focus
with performance concems in the ODA sector has been the adoption ofthe Millennium Development Goals (MDGs)
which firmly focuse signatory govemments on poverty eradication as the overriding goal of their development assis-
tance. Hitherto, systems of finance, policy-making and human resource management in intemational development

*Correspondence to: Dr D. Salttnarshe, 21 St James Park, Bath BAl 2ST, Englatid, UK. E-tnail: [email protected]

Copyright © 2003 John Wiley & Sons, Ltd.

446 D. SALTMARSHE ETAL.

organisations had been able to operate in a relatively detached relationship from organisational performance. It is
argued that an organisation capable of reflexive leaming aquires an improved understanding of how its activity sys-
tems interact to make up the whole. The performance framework presented here is thus set within the context of
systems thinking, a discipline for seeing wholes, inter-relationships and underlying structures (Flood, 1999).
The article examines the evolving notion of performance in the intemational development context and presents
a framework to identify and define a development organisation's performance architecture, in this case that of
DFID. The main components of the framework are described and are followed with a discussion of what in this
policy context might constitute a leaming organisation.

BACKGROUND
The article is set against a background of what has been called 'new public management' (NPM), an approach to
public administration which links efficiency, effectiveness and accountability. The approach is influenced by man-
agement philosophies from the private sector. For public bureaucracies, the key requirement has been to shift from
an emphasis on the control of workers, to creating systems and structures that can capture their productivity and
innovation. NPM emphasises control by targets and performance indicators rather than by rules. Its acceptance by
public organisations increases concems over accountability.
As Minogue suggests in his discussion ofthe relevance of NPM in intemational development, this draws us into
debates over 'good govemance', democratic participation and empowerment (Minogue, 1998, p. 17). However, it
is a distinctive feature ofthe complexity of intemational development that the logic of good govemance forces us
to recognise that accountability for ODA must flow in two directions: back into the polities of donor countries, but
also into the political systems of recipient nations. This increased emphasis on greater accountability has resulted
in an upsurge of attention on results rather than on process (see Ireland et al., 2003). There has been increased
pressure to measure change and to establish value for money, with donors seeking ways to improve data on
how their resources are being utilised to reach publicly stated strategic objectives.
In the UK, a longstanding process of civil service performance-related reform has only recently come to bear
more substantially on DFID. Similar pressures for attention to performance are being felt across the intemational
donor community. The prime drivers for change have been Finance Ministries which have demanded improved
accountability for public funds from govemment departments charged with delivering ODA. Multilateral devel-
opment and financial organisations and NGOs have different and complex histories that have brought about dis-
tinctive organisational cultures. So although they share many performance challenges, their modes of
accountability are different, and accordingly so are their approaches to measuring and reporting on performance.
The tardy impact of the drive for improved performance on ODA donors can broadly be understood to have
occurred for technical and political reasons. On the technical side, assessment of donor activity presents particular
measurement challenges within a multi-donor environment. Even after narrowing the policy objectives to poverty
eradication, the measurement of the impact of development funding remains technically challenging. In addition,
there is a complex organisational environment where performance is primarily in the hands of recipient govem-
ments and organisations and where there may be many different donor organisations playing different roles. This
leads to problems of attribution when interrogating the relationship between aid giving and performance outcomes.
From the political perspective, it is important to recognise that the time lag between interventions and resultant
reductions in poverty tends not to map well onto the shorter political timescales to which donor govemments must
attend. For example, investments in improving primary education is unlikely to bear fmit within anything less than
20 years. Furthermore, ODA tends not to be faced with the same level of public scrutiny and pressure to deliver, as
do, for example, domestic health or education sectors where success or failure in achieving progress is of far
greater public concem. Along with this, ODA usually only represents a small fraction of national budgets so that
in 2001, UK ODA represented approximately 0.3% of GDP (£2.8 bn) while govemment expenditure on health
represented approximately 5.5% (£50bn) of GDP Given this combination of technical and pohtical setting, we
can begin to understand why govemments and agencies charged with delivering ODA have been slow to take
up the performance challenge.

Copyright © 2003 John Wiley & Sons, Ltd. Public Admin. Dev. 23, 445-456 (2003)
 THE PERFORMANCE FRAMEWORK 447

Nevertheless, the fact remains that large amounts of public money are spent on ODA with the clear and worth-
while policy objective of bringing a large number of the world's poor out of poverty. Commenting on development
endeavours, Albert Hirschman cogently observed that we have a tendency to underestimate the human problem-
solving capacities in this field (Hirschman, 1967). Thus we would argue that despite undoubted problems, it is
possible to move towards creating more effective performance-orientated development organisations,

PERFORMANCE
Performance assessment is a developing concept. It is being increasingly applied across public and private orga-
nisations. Within the international development context, it has assumed a number of meanings. The range of defi-
nitions encompass a broad understanding of development performance and possible impacts on a recipient country
down to definitions that emphasise organisational performance and task management. One current definition is
'Performance assessment consists of monitoring, evaluation and impact assessment activity associated with pro-
jects, programmes, strategies and policies' (PARC, 2001). While accurate, the definition begs a number of ques-
tions. Namely, what happens with performance information once it has been obtained? How are the findings from
performance data mediated within an organisation? What are the factors that create a responsive and dynamic
learning organisation that utilises performance information to improve the quality of its work?
During interviews with DFID staff, it became evident that there was uncertainty over the meaning and implica-
tions of performance assessment and only a vague understanding of the wider dimensions of performance. At that
stage, management had not fully explained the processes and terminology associated with the performance agenda.
As a consequence, the performance discourse lacked clarity and had failed to capture the understanding and ima-
gination of staff. However, consultancy experience indicates that it is not infrequent in organisations for the notion
of performance to be predominantly understood within the context of the individual's personal development. Per-
formance therefore tends to be conceptually linked to human resource management rather than being understood
as an issue for the organisation as a whole.
Clearly, in this case, 'performance' was being envisaged by senior management as being more than personal
development plans and more than monitoring and evaluation. If performance assessment is to extend beyond mea-
suring outputs and the quality of staff activity, then the establishment of systems orientated to providing an effec-
tive means of scrutinising the relationship between goals and outcomes is a much more complex undertaking.
Furthermore, if the information derived from such scrutiny is to improve performance, then it needs to be managed
and applied. The study highlighted how the capacity to leam and apply lessons is linked not just to organisational
structures, but also to the organisational culture which underlies them (Clegg, 1989),

ADDRESSING THE CHALLENGE

As our research progressed, a wealth of information was generated. What appeared to be a relatively straightfor-
ward assignment became increasingly complex. The range of complexity encompassed structures, drivers for
change, patterns of change and staff views and attitudes. We were therefore faced with the task of analysing
and systemising a large amount of diverse performance-related information and no framework within which to
contextualise the core issue of training needs. The research process threw up two issues. Firstly, that there was
no framework for understanding how performance operated within organisations and secondly that within a per-
formance context, 'training' could not be seen in isolation from a range of other factors and infiuences.
In order to address these issues, we set about producing a framework which explained the dynamics between the
various components we perceived as constituting a performance cycle. The three principal sub-systems were:
goals, performance assessment and performance management.
Our assertion is that a vibrant performance culture is one which links these three sub-systems to form an orga-
nisation capable of being its own teacher through producing effective learning. The organisational culture deter-
mines the nature of linkages between the three sub-systems.
Very briefiy, the notion is advanced that once a hierarchy of goals has been established, performance is assessed
across the organisation; performance information is managed; lesson learning and analysis takes place producing

Copyright © 2003 John Wiley & Sons, Ltd, Public Admin. Dev. 23, 445-456 (2003)
448 D, SALTMARSHE ET AL.

Performance Performance
Goals assessment management

Processes &
Evidence Procedures
for policy
decisions

Management
Director'8 plans Practice
for SDA deUvery

Individual
Practice

Guidance and
Advice

Analysis and lesson learning

Figure 1, The performance framework.

evidence for policy and management decisions, improved performance against goals and leading to the re-setting
of goals, A performance culture is dynamic, built upon the creation of the three sub-systems and the establishment
of effective linkages between them. In particular, successful performance is associated with addressing the poten-
tial disconnect between assessment and management.
In respect of DFID, we devised the framework described in Figure 1, This model locates the various elements of
performance which emerged during the course of the study. The value of using such a schema is firstly that it pro-
vides analytical categories for performance information. Secondly, it enables these functions to be broken down
and clarified in respect of particular organisations. Thirdly, through presenting it as a holistic process, the model
enables the notion of performance to become more accessible across organisations. Finally, a systems model such
as this, provides baseline data against which the effectiveness of the organisational system can be measured. While
in reality the three elements are, or optimally should be, closely intermeshed, for analytical purposes there is value
in making them discrete. By breaking down the performance framework in this way, process becomes more expli-
cit and it is possible to more clearly locate institutional strengths and weaknesses, and understand how, for exam-
ple, training, advice and knowledge generation operate,

THE PERFORMANCE FRAMEWORK

Goals and goai-setting

The setting of goals is a primary feature of establishing a performance framework. Success in formulating effective
policies to support poverty reduction is highly dependent on achieving concurrence between intemational donor
community goals and those set under the host country's poverty reduction strategy, DFID has invested a great deal
in defining the goals towards which it is working and against which it will be measured. In two White Papers, DFID
has clearly defined its broad goals for poverty alleviation in relation to the achievement of the MDGs (DFID
1997, 2000).

Copyright © 2003 John Wiley & Sons, Ltd. Public Admin. Dev. 23, 445^56 (2003)
 THE PERFORMANCE FRAMEWORK 449

To assess performance, it is essential to have a clearly defined set of stratified targets against which to measure
performance. In DFID's case, these are linked by the Treasury to detailed MDG-orientated targets that are laid out
in rolling 3-year Public Service Agreements. Within each three yearly agreement, annual targets are defined by
means of a Service Delivery Agreement (SDA). The SDAs are used by DFID Regional Directors to produce deliv-
ery plans which inform the formulation of individual country assistance plans. The director's delivery plans have
rapidly become the most important internal driver of organisational performance by providing detailed targets
against which regional performance can be measured.
Within departments, formally constituted teams usually have a written scope of work or terms of reference
against which the progress is assessed. Such teams are an important means of improving communication and les-
son leaming across the organisation. They improve effectiveness and efficiency by breaking down sectoral barriers
and strengthening work on crosscutting issues. Individual personal development plans, based on the objectives of
the department and annual personnel appraisals, help staff locate their contribution to the organisation and to
understand how to advance their career prospects.
The capacity to re-set 'smarter' goals is a reflection of the capacity of organisational leaming and analysis. The
generation of reflexive capacity within an organisation enables it not just to meet goals but also to modify them in
the light of circumstances and experience. It also enables goals to be shared throughout an organisation. Moreover,
the more an organisation's goals are related to the personal and aspirational vision of its members, the more robust
and cohesive the organisation will be in its ability to address the challenges it has set for itself

Performance assessment
Effective performance assessment in the public sector is highly contingent on transparency and openness. Perfor-
mance assessment is a multi-tiered process involving sources both extemal and intemal to the organisation. Forms
of extemal scmtiny are firstly undertaken by statutory bodies responsible for organisations, secondly by the
sources which disburse funds and thirdly by regulatory bodies. Within the ODA sector, there are broadly four types
of organisations, donors, multilaterals, partner govemments and NGOs. The forms of scmtiny which take place are
dependent on whether the organisation is in the public or private sphere. Effective scmtiny requires accurate, inde-
pendent and impartial evidence.

External scrutiny
There are a range of ways in which DFID is monitored, assessed and its activities approved. Parliament has the
final sanction and is supported by the select committees. Within government, the Prime Minister and the Cabinet
must be satisfied with departmental performance. The Finance Ministry (the Treasury) holds DFID to account
through its meeting agreed targets for poverty reduction as outlined in a public service agreement. DFID is also
a member of the OECD's Development Assistance Committee (DAC). This donor grouping requires openness to
peer scmtiny in the form of an independent review of members' organisational practices and achievements. Scru-
tiny of DFID is also undertaken by the National Audit Office. Most recently, an independent and publicly available
extemal assessment has been introduced. The Development Effectiveness Report (DER) provided a forthright
study of DFID's overall performance (Flint et al., 2003).

Internal scrutiny
Within large organisations, there tend to be evaluative procedures at all levels of operation. DFID's intemal
mechanisms for scmtiny include a senior management board, a development committee and an audit department.
The evaluation department commissions extemal evaluations and compiles lessons from a range of evaluative
activities. The Development Committee is a repository of institutional knowledge with a pivotal strategic role.
Its purpose is to ensure that DFID's policies and programmes meet the strategic priorities set by the Management
Board. Through its discussion of policy coherence, it represents an important formal mediating influence in the
organisation. Although it is solely an advisory body, its members are participants on the management board or have
other influential roles in the organisation.

Copyright © 2003 John Wiley & Sons, Ltd. Public Admin. Dev. 23, 445^56 (2003)
 450 D, SALTMARSHE ETAL.

Evaluation
Evaluation is described by the DAC as the systematic and objective assessment of a project, programme or policy,
its design, implementation and results. The purpose of evaluation is therefore to determine the efficiency, effec-
tiveness, impact, sustainability and value of an intervention or policy. Evaluations assist in assessing how parts of
an organisation are following best practice with regard to processes and procedures, assessing and managing risk,
assessing the reliability of internal and external information and reporting on accountability. To have sufficient
objectivity and credibility, evaluations need to be undertaken by people external to the organisation that is being
evaluated.
Evaluations in the ODA context have a tendency towards being contentious. Rarely are they universally wel-
comed. They are commonly seen as a chore that has to be gone through. Evaluators become familiar witb having to
deal with various forms of mediation which indicate institutional defensiveness. A finding from the DFID commis-
sioned DER was that insufficient external evaluations were taking place across the board.
The way in which evaluations take place is evolving. Donors are finding that large ex post evaluations are
too cumbersome with reports often arriving too late for their contents to be useful. As a consequence, a quicker,
shorter review type process is increasingly being adopted. A common theme of external organisational evaluations
is a call for higher quality data, improved inter-disciplinary cooperation and the imperative for independent
evaluations.

Monitoring
At the programme and project level, donors use a number of mechanisms linked to indicators. A cmcial issue is the
extent to which progress against indicators is externally verified and the extent to which stakeholders, particularly
beneficiaries, are involved in the monitoring process. Monitoring takes time and resources, particularly human
resources. There is therefore always a choice between implementation speed and monitoring rigor and, despite
the rhetoric, there is always a tendency to choose the former. In DFID, monitoring of projects and programmes,
through tools such as output-to-purpose reviews, was valued by staff as a means of learning and applying lessons.
Donors have an extremely infiuential role in determining the nature of monitoring. In general, implementation
speed is stressed at the expense of monitoring and evaluation. Furthermore, ODA monitoring has historically suf-
fered through a tendency to over-emphasise input/output measurement at the expense of measuring qualitative
issues, behaviour change and higher level impact. Therefore it is crucial that monitoring assimilate multiple
sources of evidence and benchmarks and monitor target group behavioural cbange. Much remains to be achieved
in developing approaches for the monitoring of budget support and influencing activity associated with efforts to
support public sector reform in developing countries.

Manager assessments
Staff assessments are primarily concerned with how managers review the practice and achievements of individuals
and teams in relation to effective progress towards goals as set out in work plans and personal development plans.
More often than not, individuals are team members. The existence of teams, how they operate and are rewarded is
closely linked to organisational culture which will be discussed in more detail below.
In large organisations, it is common to find a lack of clarity in the mind of individuals about how tbeir contribu-
tions to teamwork are assessed and incorporated into personal development plans. The articulation of clear lin-
kages between the individual and team performance and thence to department plans and organisational goals is
a feature of positive and enabling management.

Performance management
This represents a crucial sub-system in that it embodies tbe organisation's history and culture. It is tbese that
condition the nature of goal setting, assessment procedures and organisational learning. Within the context of
the performance framework, we describe performance management as being composed of four features:

Copyright © 2003 John Wiley & Sons, Ltd, PubUc Admin. Dev. 23, 445^56 (2003)
 THE PERFORMANCE FRAMEWORK 451

1. Processes and procedures: These refer to the organisation's knowledge systems, both overt and covert. It is
through these that an organisation draws upon its memory about what it is and how it does what it does. Formal
systems would include management information systems and organisational rules and regulations, however
these are informed and conditioned by a set of infonnal systems which represent the culture of the organisation.
They encompass its history and the way the organisation is embedded in the society of which it is a product.
Finally, the outputs of formal systems are mediated and influenced by the behaviour of individuals.
So, DFID, for example, is an organisation steeped in the values of the British civil service. The traditions of that
service extend over several centuries. Historically, it has been a conservative service, self-selecting from a nar-
row band of British society that produced heavily stratified institutions, full of informal networks within indi-
vidual Departments and upwards to the Cabinet office. Nowadays, the civil service is striving to be an equal
opportunity employer. Notwithstanding this, DFID tends to select a particularly competitive type of personality
for senior and middle management positions.
Organisational structure is a derivative of organisational culture, and change is often slow and incremental. In the
British civil service, the relationship between Ministers and Permanent Secretaries represents a crucial element in
an organisational structure, as does the relationship between boards of directors and chief executive officers in pri-
vate sector organisations. Such higher level relationships lay the foundation for policy and operational procedures
that cascade through the various organisational tiers. Performance drivers are therefore policies driven forward by a
management committed to the ideals of the organisation. Fxamination of performance takes place by means of
monitoring regimes, including evaluation and audit departments, and knowledge management systems. Evidence
suggests that in this highly complex environment, as soon as organisations reach a certain size and its members are
no longer able to have regular personal contact, it becomes difficult to institute effective feedback loops.
Processes and procedures need to be capable of being adapted to rapidly changing circumstances such as con-
fiict or changes of govemment. During the 1990s, DFID moved away from a project-based approach to devel-
opment towards implementing a more explicitly pro-poor policy achieved through working with partner
govemments and applying results-based management.
2. Management practice: A key element in the management of any organisation is the quality and alignment of the
relationships between management strata. This is vital in every sphere of operation, none less so than in relation
to performance. Performance agendas need to be consistent, clear and well communicated. It is important that
managers at all levels have a good understanding of the challenges associated with monitoring, evaluation and
impact assessment. The embedding of a performance ethos within an organisation takes time, leadership and
resources for building capacity.
During the research, DFID staff drew our attention to the links between performance and people management.
There were wide variations across the organisation in management capacity and quality. The way that managers
valued and included staff was often reflected in their approach to communication, delegation and provision of
leaming opportunities. In order to respond to management shortcomings, DFID have instituted a behavioural
approach to management training. Nevertheless, culture change is complex. It was not uncommon for staff to
comment on the persistence of strong hierarchies in DFID.
The evidence from this research indicated a clear correlation between people-centred management and team
performance. While it is stating the obvious, good management can be seen to make a big difference. It is asso-
ciated with high staff morale and an environment where individuals and teams can operate effectively.
While focussing on behaviour change can improve the interpersonal skills of those with management and lea-
dership responsibilities, there are other important management skills required to achieve and maintain high
levels of performance. These include prioritisation of objectives and tasks and the associated enhancement
of managers' ability to make difficult decisions, retaining a strong focus on organisational goals and maintain-
ing the motivation of staff and stakeholders. Consistent minimum standards in management and leadership
across large organisations are a key element in performance management. Weak management fuels inter-
departmental tensions and constrains inter-disciplinary activity.

Copyright © 2003 John Wiley & Sons, Ltd. Public Admin. Dev. 23, 445-456 (2003)
452 D. SALTMARSHE ETAL.

The contemporary development environment is complex, fast changing and increasingly less linked to pro-
ject activity. In order to operate effectively in the face of the challenges associated with these dynamic circum-
stances, there is a need to pay constant attention to strengthening both management and individual practice.
This requires the development of an adaptive approach to the provision of performance-related training and
increased access to good quality guidance and advice.
3. Individual practice: A positive feature of DFID, one that enhanced its performance, was the commitment of its
staff to the organisation's goal of global poverty elimination. Staff took their work seriously and were con-
cemed that organisational structures be optimally aligned to meet objectives. A survey indicated a strong degree
of ownership with 95% of respondents feeling that DFID's overall performance was in some way their respon-
sibility and with 69% feeling a very strong concem that this was so.
However, the downside of commitment in any organisation is the way in which it can be exploited by managers.
Where there is a competitive and individualistic work environment, programme quality can be reduced through
the application of excessive pressure on individuals to cut comers. In order to maximise the effectiveness of
their departments/teams, managers need to ensure that workloads are commensurate with numbers and capacity
of staff for whom they are responsible.
In terms of staff motivation, it is important for managers to ensure that personal objectives link with depart-
mental objectives and that teams are motivated to achieve. Personal factors need to be utilised in achieving new
ways of working with partners. This requires that managers understand individuals' Values and, as appropriate,
encourage the development of goals, attitudes and behaviours that are relevant to increasing individuals' capa-
city to work with partners (see Pasteur and and Scott-Villiers, 2004). It is in the behaviour and demeanour of
individual staff that organisational change ultimately impacts and where obstructionism can occur if change
processes are not well handled. There is therefore a need to ensure that the production of additional bureaucracy
and guidance does not demoralise staff through becoming over-onerous and that training enables the individual
to navigate the business process.
To be effective, members of 'people organisations' need to communicate. While technology has a large part
to play in this process, evidence indicates that staff primarily rely on various forms of human communication
for quality support, quality information and quality understanding. Key mechanisms include formal and
non-formalised networks that provide an important role in leaming, research, strategy definition and the
generation of a performance culture.
Our work with DFID indicated that staff were highly reliant on personal networks to keep abreast of changing
agendas and new initiatives. Over 90% of the survey sample ranked informal discussions as a key source of
information. Indeed, research indicated that the benefits deriving from informal networks were greater than that
achieved through formal lesson-learning activities such as reviews and evaluations. Informal lesson leaming
within departments was enhanced by people-centred management. While it is stating the expected, good
intra-departmental communication was seen to bring significant benefits in terms of individual performance
and departmental cohesion.
4. Guidance and advice: Training is an integral part of an organisation's knowledge system. There is a manage-
ment challenge in establishing incentives for individuals and teams both to gain knowledge and to transmit it.
The drive to increase performance is dependent upon an organisation's ability to ensure that effective analysis
and lesson leaming takes place and that it feeds into its knowledge system. To this end, management guidance
and mentoring were seen to be of benefit to staff. Research also constitutes an important means of innovation for
ensuring organisational and development effectiveness.
Training may be provided through courses associated with induction, office management, development prac-
tice, country planning and to address specific training needs. The availability of comprehensive development-
related written material is most useful. The use of computer technology has facilitated the availability of infor-
mation. The DFID survey indicated that 62% of the respondents found the organisation's intemal intranet to be
a useful source of information and guidance on performance assessment. However, rather than generate new

Copyright © 2003 John Wiley & Sons, Ltd. Public Admin. Dev. 23, 445-456 (2003)
 THE PERFORMANCE FRAMEWORK 453

material, a key issue was to find ways to improve gateways and linkages to existing information, including that
located in other organisations.

A LEARNING ORGANISATION
The performance framework presented above outlines the nature of processes associated with a leaming organisa-
tion. Deriving from this, the article now briefly discusses key issues that have arisen, associated with the generation
of a responsive leaming organisation operating in the ODA sector. These are concerned with understanding the role
and nature of systems, organisational vision and the embracing of diversity, training and accountability.

Systems
Systems thinking implies understanding inter-relationships. In the context ofthis article, it can be examined in two
ways. Firstly, it is concemed with understanding the nature of systems operating within an organisation and the
operational location of an organisation. Secondly, systems thinking can be applied to the cluster of issues which
form the problematique associated with intemational development and which ODA is intended to address.
In the case of the first, it requires an understanding of boundaries between systems within organisations and
between organisations and the external world (see Miller and Rice, 1967). A key intemal management system
is the performance framework, the subject of this article. If it is to operate optimally within an organisation, it
requires effective monitoring and evaluation, accurate, timely and reliable data and good information flows.
In respect of the second, a strength of the systems approach is that it permits improved understanding of
dynamic complexity as opposed to detail complexity (Senge, 1993, p. 71). Dynamic complexity is composed
of a number of causal variables and associated feedback loops which operate either subtly and/or over extended
periods of time. For example, the way in which the developed world has started to understand its interaction with
the environment has been rooted in systems thinking (e.g. see Carson, 1965).
By viewing poverty as a failure of livelihood strategies, Joy transposes a systems analysis into ODA (Joy, 2000).
Understanding how livelihoods are gained requires understanding causal networks and variables and the nature of
feedback loops within a particular polity (and perhaps further). Joy sees improved, responsive govemance as being
key to the reduction of poverty. He describes govemance as being a holarchy, that is a stmctured whole or unity, of
self-organising collectivities interacting through formal and informal mles. By improving the inter-relationships
between the various socio-political strata, knowledge about the system can be used to create coherent programmes
of mutually reinforcing action to enhance the responsiveness of the polity and in so doing, bring about positive
changes in livelihoods. However, he notes that the acceptance of an analysis associated with systems thinking and
process consultation will require a considerable value shift by donors and technical cooperation agencies.
The implementation of a systems approach to reducing global poverty is thus linked to organisational vision and
the way in which goals are generated and owned within ODA related organisations. Systems coherence is brought
about by understanding the 'how' of livelihoods operation and applying that knowledge to identify points where
interventions can be strategically focused for maximum impact. Reductionist thinking, associated with handling
problems arising from detail complexity, directs people to seek solutions in terms of causal factors rather than
through systemic awareness (Hood, 1999, p. 85). System coherence requires avoiding over-reliance on short-term,
quick-flx, symptomatic responses.

Linkages
With the constantly changing context of intemational development, there is a need to move beyond training and
individual leaming to more collective forms of leaming which reduce duplication and conflicting effort (Dearden,
2002). This starts with mechanisms and approaches to bring about more effective teamwork and action research.
However, it can usefully extend to include the whole organisation in the leaming cycle by means of configuring
systems to move information across organisational boundaries to allow collective interpretations which lead to
action and the feedback of lessons leamt (Dixon and Ross, 1999). In this process, management information
mechanisms have a significant role to play, but there are indications that successful sharing of management

Copyright © 2003 John Wiley & Sons, Ltd. Pubiic Admin. Dev. 23, 445-456 (2003)
454 D. SALTMARSHE ETAL.

information cannot rely solely on electronic communication but requires a strong element of human interface to
provide context for information and to encourage the sharing of perspectives (ibid., p. 438).
However, imposing a systems approach to performance management on unreformed structures can be counter-
productive where there are strong 'silo' tendencies or sectoral compartmentalisation. The institutionalisation ofthe
capacity for inter-disciplinary cooperation and other elements of intemal exchange have to take place before leam-
ing can effectively take place across an organisation. The prerequisite for cooperation is a willingness to recognise,
value and embrace diversity (Pedler et al., 1991, p. 114). To this end, other forms of linkage, consistent with a
learning organisation in the ODA sector, would include participating in leaming experiences with other aid actors
be they donors, multilateral organisations, partner govemments or NGOs.

Training/learning
Training and staff development are central to improved performance and the creation of a leaming organisation.
There is increasing evidence that deepening understanding about performance assessment requires a fresh
approach to the whole leaming process. A learning organisation requires that staff be provided with time to take
up training opportunities and for this to be built into work plans. Training needs to be timely, practical and relevant
to the individual and groups so that knowledge can be applied before key messages get forgotten.
Staff need to take time to reflect on action and experience. This enables them to reframe problems or issues and
gain relevant insights which can lead to improved future action and performance (Pasteur and Scott-Villiers, 2004).
Relevant to this, the DFID research threw up an interesting example of innovative thinking whereby informal net-
works were semi-formalised into cross-disciplinary networks. These proved highly useful in debating key perfor-
mance issues, disseminating ideas and assisting new advisers and managers to become established and confident in
their roles. Success derived from the self-interest of participants in leaming and sharing views. This enabled them
to define and integrate the strategic implications of rapid policy change. Rather than obtaining an 'expert' view
from outside, the process allowed ownership through an intemal synthesis that had direct and rapid influence on
strategies and policy.
There is a need to ensure that transfer of assessment information/leaming is treated as a priority and that it
adequately informs performance management. Current systems, guidance and practice are catching up with the
requirements of the performance agenda, driven as it is by a need in both public and private management to
improve organisational performance and policy outcomes. The consequence of increased emphasis of donor staff
on 'upstream' policy issues has resulted in them spending increased time and resources on activities that are not
easily captured by centralised management information systems. A similar situation exists in assessing capacity
building undertaken by multilaterals and NGOs.
The generation of useful data in an information-scarce environment is of cmcial significance in influencing the
development of policy. In this respect, research is a valuable tool, not just for knowledge gathering and innovation,
but also for bridge building and providing leaming opportunities for research participants. It allows donors, host
govemments and NGOs to gain greater insight into circles of causation in the reproduction of poverty and the
effectiveness of their work. It permits the establishment of baseline data for project monitoring, evaluation impact
assessment. Finally, the use of participatory data collection provides one of the means by which donors, their part-
ner govemments and other aid actors can institute downward accountability.
Our research indicates that strengthening performance management and increasing the focus on training in per-
formance assessment will bring about improved analysis and lesson learning. We suggest that the feedback from
more timely and higher quality lesson leaming will produce improved performance against goals as well as
improved evidence for policy and management decisions.

Accountability
The human cost of skewed and inequitable development is poverty. Poverty is to do with poor housing, education
and health care. It is to do with the results of conflict, exclusion, environmental degradation and poor economic
opportunities. It is to do with poor govemance and the negative effects of global economic regimes that are insti-
tuted to favour the rich. These are the factors that ODA, in one way or another, is seeking to mitigate, or at least

Copyright © 2003 John Wiley & Sons, Ltd. Public Admin. Dev. Ii, 445-456 (2003)
 THE PERFORMANCE FRAMEWORK 455

should be. There is a normative element in this process, the implications of which can all too easily become lost in
vested interests of governments, business, academics and those who deliver ODA in one form or another.
While the drive for perfomiance is concemed with accountability, all too often this is translated in terms of
upward accountability and not accountability to those who form the 'poor', be they called recipients, beneficiaries
or clients. Participation is often more evident in its rhetoric than its practice. Chambers describes the generation of
poverty and exclusion as being a result of the continued predominance of institutions which sustain and reinforce
'upper-lower' dominance (Chambers, 1997, pp. 221-222). To reverse such tendencies, he suggests a pro-poor
agenda that demands radical systemic institutional change in the procedural, professional and personal domains,
of which the personal is key (Chambers, 1998, pp. 117-131).
Ugandan NGOs have played an important part in widening definitions of poverty and in alerting the develop-
ment community on some of its key causes. Wallace (2000) describes the nature of donor support to these NGOs.
Donor preference was for large NGOs undertaking service delivery. There was donor wariness of radical issues
associated with capacity building, civic education and working with marginalised people. Most training was pro-
vided around procedures, finances, reporting and logframes. Little was offered in the way of improving NGO per-
formance in poverty alleviation through developing listening and analytical skills, managing change and thinking
and working strategically. NGOs found donors dominating with onerous reporting procedures, dismissive of PRA
findings, high handed and arrogant. Donor focus was on planning with little time given to evaluation and leaming.
Thus it is suggested that without downward accountability, leaming is attenuated, the effectiveness of aid
becomes compromised and performance suffers. A leaming organisation operating within the realm of ODA
requires a clear vision of how it is to be accountable to those it is seeking to serve, and not solely to donors
and the taxpayers who fund them. Without transparent strategies which demonstrate the dual nature of account-
ability, the performance of organisations whose mandate is to relieve poverty will inevitably be constrained.

CONCLUSION
The performance framework presented in this article enables the establishment of a clearer definition between
structure and process and the identification of their dual role in determining performance. The framework is simple
and seems obvious once set out. But without the model, we were unable to adequately understand the nature of the
performance arena and what was taking place within it. Once we applied the framework, an area of seeming com-
plexity became much more comprehensible. While the framework described here refiects DFID's structures, we
have applied it to other organisations operating in the ODA sector to good effect.
By clarifying the nature of an organisation's performance system and subsystems, it is possible to attain an
improved understanding ofthe role of culture, process and procedures as they infiuence the delivery of performance
against goals. A dynamic performance culture requires quality management and opportunities for quality leaming.
It requires sufficient human, financial and technical resources. Performance-related training cannot be seen as being
separate from management since management conditions approaches and access to leaming opportunities.
In respect of organisations operating in the ODA sector, we conclude that the analysis of an organisation's per-
formance and the application of lessons leamt need to take place in a variety of ways. The larger and more complex
the organisation, the greater the need for diverse methods of scmtiny and assessment which are consistently prac-
ticed through the organisation. Effective performance assessment is much more than the measurement of outputs
by means of monitoring and evaluation and individual staff performance by means of assessments. Each of the
three elements of the performance framework and the linkages between them need constant review. The nature
and interplay of the features which constitute performance management is as yet poorly understood and needs
further research.
When set within a performance framework, there is an identified need for a greater appreciation of the role of
systems thinking in improving performance. Performance management is embedded in the norms and institutional
disciplines which form an organisation's management culture. However, management practice is heavily mediated
and informed by individual practice. Performance management represents a capacity to understand and apply
knowledge gained through performance assessment and other lesson-leaming tools, so as to improve performance

Copyright © 2003 John Wiley & Sons, Ltd. Pubiic Admin. Dev. 23, 445-456 (2003)
456 D. SALTMARSHE ETAL.

in meeting goals and achieving accountability—accountability both to those who provide resources and to those
who are intended to benefit from them.

ACKNOWLEDGEMENTS

We are grateful for the cooperation provided by DFID staff in undertaking the original research.

REFERENCES

Carson R. 1965. Silent Spring. Harmondsworth: London.

Chambers R. 1997. lWio.se Reality Counts: Putting the Last First. Intermediate Technology: London.
Chambers R. 1998. Public management: towards a radical agenda. In Beyond the New Public Management, Minogue M, Polidano C, Hulme D
(eds). Edward Elgar: Cheltenham; 117-131.
Clegg S. 1989. Frameworks of Power. Sage: London.
Dearden P. 2002. Monitoring, Evaluation and Leaming in DFID—Moving Towards Being a Learning Organisation! CIDT: University of
Wolverhampton.
DFID. 1997. Eliminating World Poverty: A Challenge for the 21st Century. White Paper on Intemational Development. HMSO: London.
DFID. 2000. Eliminating World Poverty: Making Globalisation Work for the Poor White Paper on Intemational Development. HMSO: London.
Dixon N, Ross R. 1999. The organisational leaming cycle: turning knowledge creation into a self-changing system. In The Dance of Change:
The Challenges to Sustaining Momentum in Learning Organizations, Senge P, Kleiner R, Roberts C, Ross R, Roth G, Smith B (eds). Double-
day: New York; 435-444.
Flint M, Cameron C, Henderson S, Jones S, Ticehurst D. 2003. How Effective is DFID? An Independent Review of DFID's Organisational and
Developmental Effectiveness. Evaluation Department Study EV640. DFID: London.
Flood R. 1999. Rethinking the Fifth Discipline: Learning Within the Unknowable. Routledge: London.
Hirschman A. 1967. Development Projects Observed. The Brookings Institution: Washington, DC.
Ireland M, McGregor A, Saltmarshe D. 2003. Challenges for donor agency country level performance assessment: a review. Mimeo, University
of Bath.
Joy L. 2000. The changing role of international technical co-operation. In Applying Public Administration in Development: Guideposts to the
Future, Collins P (ed.). John Wiley: Chichester; 249-263.
Miller E, Rice A. 1967. Systems of Organisation: The Control of Task and Sentient Boundaries. Tavistock: London.
Minogue M. 1998. Changing the state: concepts and practice in the reform of the public sector. In Beyond the New Public Management,
Minogue M, Polidano C, Hulme D (eds). Elgar: Cheltenham; 17-37.
PARC. 2001. Meanings, http://www.parcinfo.org/meanings.asp
Pasteur K, Scott-Villiers P 2004. Minding the gap through organisational leaming. In Inclusive Aid: Changing Power and Relationships in
Intemational Development, Groves L, Hinton R (eds). Earthscan: London.
Pedler M, Burgoyne J, Boydell T. 1991. The Learning Company: A Strategy for Sustainable Development. McGraw-Hill: London.
Senge P. 1993. The Fifth Discipline: The Art and Practice of the Leaming Organization. Century Business: London.
Wallace T. 2000. Is the role played by Donors in Supporting Uganda's NGO sector Enabling it to Develop Effectively? In Applying Public
Administration in Development: Guideposts to the Future, Collins P (ed.). John Wiley: Chichester; 227-248.

Copyright © 2003 John Wiley & Sons, Ltd. Public Admin. Dev. 23, 445-456 (2003)