2.1 Concepts of public sector productivity

The review did not begin with a precise definition of public sector productivity, but our orientation from the outset was derived from the proposition of Kalliola (2003) that

"the legitimacy of public services is derived from the capacity to respond to the needs of citizens in an economically efficient way." (p.113).

That concise statement encapsulates the emerging consensus that, while public sector productivity involves efficiency and outputs, it also involves effectiveness and outcomes (Pritchard 2003; Tolentino 2004).

However, beyond that general level, there is less consensus - not least about the meaning of those terms. As Halachmi (1999, p.9) notes:

"Even those who research performance and productivity in the public sector are not always in agreement when it comes to the exact distinction(s) or the functional relationships among concepts like: output, outcome, results, impact, objectives, accomplishments, direct/indirect (or primary/secondary) performance indicators."

In Kalliola's formulation, at least two points illustrate the issues that make the whole question of public sector productivity a highly contentious one in many countries: are citizens being provided with what they need, and how could the services concerned make better use of the resources at their disposal?

In the private sector's production of goods for the market, productivity is defined as the ratio of outputs to inputs (Samuelson & Nordhaus, 1989). That seems simple, but as the Organisation for Economic Cooperation and Development (OECD) has noted:

"Productivity is commonly defined as a ratio of a volume measure of output to a volume measure of input use. While there is no disagreement on this general notion, a look at the productivity literature and its various applications reveals very quickly that there is neither a unique purpose for, nor a single measure of, productivity." (OECD 2001, p.11).

Exploring the theme further, the OECD explains:

"There are many different productivity measures. The choice between them depends on the purpose of productivity measurement and, in many instances, on the availability of data. Broadly, productivity measures can be classified as single factor productivity measures (relating a measure of output to a single measure of input) or multifactor productivity measures (relating a measure of output to a bundle of inputs). Another distinction, of particular relevance at the industry or firm level, is between productivity measures that relate some measure of gross output to one or several inputs and those which use a value-added concept to capture movements of output." (OECD 2001, p.12).

So, for example, and with particular relevance for our purposes:

"Labour productivity changes reflect the joint influence of changes in capital, intermediate inputs, as well as technical, organisational and efficiency change within and between firms, the influence of economies of scale, varying degrees of capacity utilisation and measurement errors. Labour productivity only partially reflects the productivity of labour in terms of the personal capacities of workers or the intensity of their effort. The ratio between output and labour input depends to a large degree on the presence of other inputs." (OECD 2001, p.12).

The productivity and labour productivity of any organisation, therefore, are affected by a range of factors in their environment, both within and outside the organisation concerned. In the public sector environment, the matter is further complicated by at least two key characteristics of the public sector: that it produces services rather than goods (Morley 1987); and that it does so in response to governmental rather than market decisions (Rosen 1993; Kelly, R.M., 1988.). As Rosen puts it:

"In general terms, "productivity" is an efficiency measure: it tells how well resources have been used. The more produced with a given set of resources, the higher the productivity. We say someone is a productive worker if he or she turns out a lot of good work per day, per week, or per year. A productive organisation, similarly, is one that turns out a high level of good quality product with its resources. Public sector productivity focuses on the efficiency of governmental (that is, publicly authorised and funded) administrative agencies and their subunits.

"Public productivity as a field is different from the private sector productivity field. Public agencies operate under significantly different conditions. They are more tightly constrained: missions are legislatively fixed; operations are open to public scrutiny and reaction; volumes of rules and detailed procedures define options; civil service and budget systems limit freedom to redeploy labor and monetary resources. The major difference is in measurement: goods and services produced in the private sector can be measured in terms of their dollar value because they are sold in the market. Public agencies produce services that are not for sale. This makes measurement quite different and more difficult (but not impossible - valid and useful information is obtainable on the productivity of public sector organizations)." (Rosen 1993, pp.4-5).

Rosen's use of some key terms is not shared across the literature, and it begs many further questions. In relation to terminology, for example, it blurs the distinction emphasised by some between efficiency and effectiveness. As Allen Schick has remarked:

"In the burgeoning literature on performance, efficiency generally is associated with outputs - the goods and services produced by government - and effectiveness with outcomes - the impacts of government programmes on society. Efficiency has both quantitative and qualitative characteristics that include the volume and cost of services, response times and error rates, the accessibility of services and the courtesy with which they are provided, and citizen/customer satisfaction with services. Effectiveness means that programmes are in accord with the priorities and objectives of government, and produce the expected or desired impacts." (Schick 2006, p.3.).

That way of defining the difference between efficiency and effectiveness has parallels with the distinction that can be made between "efficiency" and "effectiveness", and the confusion between them overcome, by using the distinction between "allocative efficiency" and "technical efficiency" in economics theory. Allocative efficiency is concerned with whether the right things are being produced, and technical efficiency with whether they are produced in the right way. The term "effectiveness" can be understood, therefore, as analagous to 'allocative efficiency' (Kelly, R.M., 1988; Leathers 1979), while the term "efficiency", when it is applied in the public sector, usually means "technical efficiency". The analogy can be taken only so far, however, because of the particular role of the public sector, as Rita Kelly explains:

"In terms of classical economic theory, the goal is to maximize satisfaction of the subjective demands of all citizens. In the public sector of a democracy, however, the rights, needs and wishes of minority groups and individuals need to be reckoned with as well as the needs and desires of the majority."

Therefore, it is significant that:

"Often the public sector is engaged in creating and delivering public goods and services that are not desired, needed or preferred by the majority, but rather are necessary for some particular group - such as the poor or elderly. Hence the term 'efficiency' in its allocative sense is not easily applied. In such cases, we need a broader and more objective standard for evaluation than just maximizing the satisfaction of the subjective preferences of the aggregate of individuals as revealed through the market. Indeed, even if there is a strong consensus of individuals, such non-subjective standards are essential for acceptable allocation of public goods, because such allocation must take account of the abstract principles of liberty, merit, equality and human rights." (Kelly, R.M., 1988, p.8).

It is here that the term "public value" (Moore 1995) can come usefully into the picture. The word "value" is as significant as the word "public" in the "public value". The concept is related to that of shareholder value in that it is concerned with the comparison of the value of what is produced with the costs (including the opportunity cost - that is, the value that could otherwise be created from the same resources) of the resources deployed in its production. In other respects, however, and notably in its understanding of the value created by government through services, laws, regulation and other activities, it is necessarily a rather more complex concept than shareholder value, for reasons that are clear from Rita Kelly's points above.

Public value has been defined in simple terms as that which is valued by the public, not only in terms of aggregated individual preferences but in terms of collective preferences expressed through social and cultural norms (Kelly, G., et al. 2002). Gavin Kelly et al. identify three interconnected components of public value: services, outcomes and trust. The public value of services is expressed not only in terms of user satisfaction but also in terms of the ethos and culture expressed in their delivery. Outcomes are produced in part directly by services but also by the effects of those services interacting with other social and cultural activities and forces. Therefore, the concept of the "co-production" of outcomes by government and citizens is significant, and this can be traced back to the way in which some services are organised in partnership with those who use them, or depend for their outcome on the varying use made of them by recipients. The third component of public value, trust, is yet more difficult to pin down precisely, but it is perhaps the most important of all. As Kelly et al. put it:

"Trust is at the heart of the relationship between citizens and government. It is particularly important in relation to services which influence life and liberty - health and policing. But it also matters for many other services - including social services and education. In these cases, even if formal service and outcome targets are met, a failure of trust will effectively destroy public value." (Kelly, G., et al. 2002, p.17).

2.2 Public sector productivity in historical perspective

We can also approach the question of public sector productivity by tracing the historical development of the debates about what the public sector is for, how it should operate and how its value should be evaluated. Bouckaert (1990) relates the history of the public sector productivity "movements" to phases in prevailing models of public management, categorising the last century into four phases:

• "Government by the efficient", from 1900 to 1940, when the important distinction between political and administrative roles were defined by Max Weber and Woodrow Wilson, and the approach to efficiency in public administration drew heavily upon the "scientific management" associated in the private sector with the work of Frederick Taylor, whose techniques set out to identify the single best way to carry out any particular function.
• The second phase, from 1940 to 1970, is characterised as "government by administrators" and marked a shift of focus from quality of government to control of expenditure.
• That paved the way for a third phase, "government by managers", which can be understood as a synthesis of the first two into a focus on getting "more bang for the public buck" (Bouckaert 1990, cited in Green 2004, p.11).
• The fourth phase, from 1980, "government by the private sector", can be seen in its turn as building on the third, not only by reinstating in modern form an idea that characterised the first (that private sector management techniques provide a model for the public sector), but also taking it a step further and actually transferring responsibility to the private sector or, failing that, restructuring the public sector in ways that stimulate public managers to operate as though they were in the private sector.

"Two trends were dominant during this period. One continued the approach pursued in the 1970s and the other was a "new" approach. This new approach was ideologically motivated and it advocated a private sector-inspired approach to productivity in the public sector and the privatization of many government services." (Green 2004).

Bouckaert's schema clearly offers only a very rough approximation to actual historical development in which phases are neither as distinct from each other nor as universal as it suggests. However, it does offer a useful platform for understanding the various approaches to definition of public sector productivity and their development in the context of changing norms of public administration and management. It also points to a fifth, emerging phase of public sector productivity conceptualisation, corresponding to the "public value" approach to public governance and management.

Stoker (undated) offers, in effect, an abbreviated version of the Bouckaert phases by noting the current emergence of a public service management model that is distinct not only from "traditional public administration" (TPA) but also from the "new public management" (NPM). (NPM would correspond clearly with Bouckaert's fourth phase, overlapping also with his third, while TPA does not distinguish between the earlier phases.) It can be seen as building upon certain elements of NPM that have been effective in overcoming the weaknesses of TPA under contemporary conditions while, in turn, transcending NPM's own weaknesses. According to Stoker:

"What marks out the approach from NPM is that an ethic of public service is seen as vital to the system. There is not a specific public sector ethic but there is a public service ethos."

Ethos is clearly closely related to trust, and Schick expands on the issue and its relationship with public sector productivity and how it is improved:

"A public service ethic is the bedrock of governmental performance which depends at least as much on people as on machinery and process. This view clashes sharply with the principal-agent model popularised by new institutional economics (NIE) and imported into the public sector by New Public Management (NPM). NIE and some versions of NPM teach that public employees are self-interested, opportunistic agents, slackers who feather their own nest at the expense of the public interest. In this view, public agents can be made to perform only if they are actively monitored, given clear instructions as to what is expected of them, and strong incentives to do the job right. The notion that agents might do more than is formally expected of them because they have internalised public service values may be alien to NIE/NPM, but it is familiar to generations of students who overcame education handicaps because of teachers who stayed after class to help them, the police officer who coached the community sports team and never asked for pay, the visiting nurse who dropped by shut-ins after her daily rounds were done, and in countless other ways. Of course, this was never the whole story of public employment, or even the larger part, but it was the stuff out of which governments performed, earned the trust of their people, and communities and states were built." (Schick 2006, p.7.).

It is striking that Schick's three examples all involve public servants doing extra work for no extra pay, implying that their ethical commitment consists in doing so. Another interpretation of that behaviour by public servants might be that they are compensating for insufficiently productive organisations whose deficiencies place them in the invidious position of having to work after hours in order to meet the standards they believe their organisations should meet. In that context, it is significant that Aldridge and Stoker (2002) include sustainable systemic characteristics, sufficient capacity and "responsible employment practices" among the five elements that, in their view, should structure the practice of public sector management, namely:

• A performance culture: a strong commitment to service for individuals and the community reflected in world class service delivery and reinforced by training, support and systems to ensure a sustainable service culture and continuous improvement.
• A commitment to accountability: an emphasis on providing open access to information to individuals and to groups of interested citizens, as well as accountability to the electorate at large.
• A capacity to support universal access: recognition of a special responsibility to support the rights of all service users in an environment in which their choice of service is restricted.
• Responsible employment practices: the maintenance of a well-trained, well-managed and well-motivated body of staff that acts professionally and is fairly rewarded.
• Contribution to community well-being: a recognition of the need to work in partnership with others across the public, private and voluntary sectors in order to contribute to the promotion of community well-being and to meet the needs of individuals.

2.3 Measurement of public sector productivity

Both an historical and a conceptual tour of the terrain have taken us, then, from a very simple proposition about productivity as the ratio of outputs to inputs to a highly complex set of inter-connecting relationships that produce public sector productivity. A major problem associated with that spectrum is that the further along it you travel, the more complicated and difficult the task of productivity measurement becomes. As Schick notes:

"It is widely accepted that outcomes are the more important dimension of performance, but it is also recognised that outcome data are often unavailable or costly to obtain, and that even when data are available, the causal relationship between government policy and social conditions may be problematic. In countries that take performance seriously, reforms that aim to improve outcomes tend to end up focusing on outputs instead." (Schick 2006, p.3.).

This implies a tension between what the public sector should be doing, as understood by contemporary "public value" thinking about its role and its complex relationship with citizens, and measurement of the productivity with which it is doing it. Not only are outcomes of government action the products of factors external to a public service organisation, but so are the outputs of public sector workers and their organisations. As Schick notes,

"Outputs and outcomes are not sufficient measures of government performance. They are snapshots of what government is doing or accomplishing at a particular point in time; they do not uncover the factors that contribute to or retard the results, nor do they indicate whether government will have the capacity to perform in the future" (Schick 2006, p.3.).

Yet the need for governments to be accountable for what they do and for how they spend the money levied by compulsion from their citizens continually recreates the drive to measure, and to improve measurement techniques in order to bridge the imformation gaps between causes and effects. This issue too can be approached historically as well as conceptually. Van de Walle and van Douren point out:

"The use of indicators in government has a long history: censuses and moral statistics in the nineteenth century, the scientific management movement in the first decades of the twentieth century, the work of the New York Bureau of Municipal Research, the social indicators movement in the 1960s and 1970s, PPBS [Planning, Programming and Budgeting System] experiments, etc. New public management was certainly not the first reform paradigm that relied extensively on performance measurement." (Van de Walle & van Douren 2006, p.445).

Rosen (1993) notes:

"Public sector productivity measurement systems began by measuring efficiency, the ratio between quantity of output and quantity of input. Output quality was disregarded or assumed to be constant. ... But the question of 'how good' could not be ignored for long. As early as 1971, Harry Hatry of the Urban Institute [in the USA], in collaboration with Donald Fisk of the Bureau of Labor Statistics, urged that quality be taken into account in measuring efficiency." (Rosen 1993, p. 93, citing Hatry & Fisk 1971).

This led to the development of quality indicators - categories such as cleanliness, comfort, timeliness and accessibility - and further to the attachment of various quantified weights and measures to different qualities and their extent, and to productivity formulae based on dividing inputs into a function of outputs and quality weightings. But as Van de Walle and van Douren again note:

"What is considered as performance at one time and place may elsewhere be seen as a despised aberration, as an administration gone astray, as an administration working towards the wrong ends using unacceptable means."

And further:

"A focus on one definition or approach over another ultimately leads to disastrous results in other areas. Ignorance of multi-faceted challenges inherent in performance measurement and management may well lead to wrong decisions, financial loss, and demotivated staff." (Van de Walle & van Douren 2006, p.445).

Rita Kelly made similar observations nearly two decades earlier:

"In recent years, private sector models of efficiency and productivity have been applied in the public sector by means of a number of different analyses: cost-effectiveness analysis, cost-benefit analysis, planning-programming-budgeting systems, operations research, and introduction of new incentive systems. These applications have been hailed by some as being instrumental in bringing about costs savings, more and better services and goods, and higher standards for workers and managers in the public sector. They have also been the focus of controversy and criticism, on grounds that, at best, they bypass the critical issues involved in the production and delivery of public goods and services, and, at worst, they yield a false precision that is misleading." (Kelly, R.M., 1988, p.6).

Rosen (1993, p.55) points out that

"measuring productivity requires, first, specification of what the 'products' or services are,"

and this in turn means weighting the difficulty of different tasks, such as the variable scale of the challenge of placing different unemployed people in jobs. Citing Hatry and Fisk (1971), Morley (1987, p.12) notes:

"Each output measure selected should reflect a fairly uniform degree of employee effort to provide the service. In other words, outputs should be basically homogenous."

2.4 The impact of productivity measurement on productivity itself

It is here that we see the impact that measurement itself can have on service delivery and public management practice and, therefore, on productivity itself. The homogenisation of outputs for measurement can lead to homogenisation of services themselves, particularly in a management context in which personnel are judged and rewarded against their performance in meeting output targets. The problem is that this collides immediately with the modern "public value" approach to service provision as an element in the creation of public value, in which a critical aspect of service quality is the extent to which socially and culturally heterogeneous needs are met within a context of social and cultural norms mediated through democratically accountable political processes. Bentley and Wilsdon (2003, pp.15-16) emphasise this point:

"Successful reform does not only depend on the level and scale at which decisions are taken or performance is measured; it will require greater adaptive capacity in organisations at every level of the system. Public services in diverse societies must offer far greater flexibility to meet personal needs, while keeping the ability to connect resources and activities across entire systems of governance. This is the only way to serve diverse needs equally well, and to make specialist knowledge and resources available to everybody."

This means, they argue, that

"we need systems capable of continuously reconfiguring themselves to create new sources of public value."

If this approach to public service reform is accepted, it has profound implications, not only for the way productivity is defined and measured, but also for both the design of the processes by which it is defined and measured and the ways in which resulting data are used. Bentley and Wilsdon address that issue in the British context:

"The current preoccupation with setting national standards as a basis for accountability obscures a tension that the process of adaptive reform must address; the specification of performance standards often narrows the scope for organisational innovation. This is partly because it encourages risk aversion, but more importantly because it establishes rigid parameters of organisation and formal responsibility. In fact, some of the most significant performance gains may arise from cross-boundary collaboration - something that is hard to design into the formal functions of bounded organisations. This does not mean that targets and standards are not essential. But they must be used judiciously, and owned by the participants, rather than used primarily as an instrument of control." (Bentley & Wilsdon 2003, p.23).

Those observations are well supported in the literature. Atkinson and McGrindel (1997, p.10) note that

"government performance measurement systems do not generally provide a reliable means of assessing how well (public) services and the individuals, groups and organizations that deliver them, contribute to the broader objectives of government."

Rosen (1993, p.68) makes a similar point at the level of the interface between service and user:

"Without the dollar as universal surrogate, each public sector output has to be measured directly by counting the actual amount of service delivered and gauging its quality in terms of selected criteria. Given the number of public services, this is a tremendously important fact."

Different systems must be developed for each service, and there is usually more than one per agency, she points out - the fire service is responsible for fire prevention and fire fighting, for example - with the result that:

"It is a vast understatement to say that measuring public sector productivity is complex .... The best way to ensure that a measurement system will be accepted and used appropriately is to include as many viewpoints as possible in its development. Participation generates more ideas, permits more interests to be accommodated in the measurement system, and facilitates the collection of good data and their constructive use as feedback." (p.69).

Rosen goes on to stress the importance of the knowledge of frontline employees in particular:

"It is often not the managers but the workers, at every level and sector of the organisation, who discover that things are not working as it was assumed they would, who first encounter the unexpected difficulties, and who are the first to hear from the clients about needs that the program is not meeting. In short, workers know the operations most intimately and are in the most immediate contact with the clientele. Workers are not only the natural source of feedback on how things are going, but also the natural source of ideas and insights into the specifics of operations." (p.141).

Those observations are also well supported in the literature, and yet governments evidently find it difficult to follow their logic through into the way services are organised and productivity is defined and evaluated, perhaps because the timescale of accountability within the political cycle does not fit will with the longer term and complex nature of organisational development and transformation. For example, the UK Treasury states:

"As normally defined, productivity simply describes the amount of public service (output) we get for the money, labour and know-how we put in. But the public and the Government are more concerned about the impact of public services on society (outcomes). We have therefore interpreted our remit on productivity in this wider sense, to cover economy, efficiency and effectiveness (often referred to as value-for-money)." (UK Treasury, 2000).

It goes on to chart the productivity path through "economy" ("paying the right price for inputs"), "efficiency" ("technical efficiency") and "effectiveness" ("providing the right outputs, [and providing them] well"). It links that path to a parallel track connecting funding to input to output to outcome.

It neglects, however, the role of employees in designing and implementing productivity measurement. Yet much of the literature, both before, during and since the NPM wave, warns against doing so. For example, Stoker (2002) argues:

"In TPA the trade-off was that democracy provided the inputs and bureaucracy the efficient solutions. NPM came close to implying that management processes can do it all, defining preferences and the best means of meeting them. For the public value paradigm, faith is placed in the system of dialogue and exchange associated with network governance. It is through the construction, modification, correction and adaptability of that system that democracy and efficiency are reconciled. Vigilance by all the partners in the system is central to ensuring that the promise of both democracy and efficiency is delivered." (Stoker 2002).

Harvey (1996) emphasises the importance of measurement, but stresses the importance of adaptability to successful measurement, and the role of employees in enabling such an adaptable approach to productivity measurement:

"As the just-in-time philosophy and the total quality movement are spreading to the public sector, public professional service organizations are increasingly aware of the necessity to empower professionals, to strive for continuous process improvement, and to focus on the client." (pp.197-8).

This, she goes on, requires "systematic feedback mechanisms that allow professionals to learn from the system and thus find ways to improve delivery processes", which means "keeping it simple at first and letting the measurement system and productivity improvements grow together."

Harvey's work concerns public sector professionals, and, clearly, the way in which particular categories of public sector employees are involved in creating public value and can evaluate the productivity with which they and their organisations are doing so varies. Other literature suggests, however, as we shall see later in this report, that the same general point applies to public sector employees generally. For that reason, we would suggest that the exploration of the nature of public sector productivity and how to evaluate it that we have made in this section suggests a significant role for workplace partnership in productivity improvement. We will explore further evidence for that proposition later in the report. In the following section, however, we lay the second foundation stone on which the analysis can be made - defining and discussing what is meant by "public sector productivity" having been the first - by exploring what is meant by "workplace partnership".

Previous | Contents | Next