Autonomy and street-level bureaucrats’ coping strategies


Autonomy and street-level bureaucrats’ coping strategies

Evert Vedung*

Institute for Housing and Urban Research (IBF), Uppsala University, Uppsala, Sweden

Evert Vedung is an emeritus professor of political science, especially housing policy at Uppsala University’s Institute for Housing and Urban Research (IBF). His works on evaluation in English include Public Policy and Program Evaluation (author, 1997, 2016), Carrots, Sticks and Sermons (1998, 2003, co-editor) and Four Waves of Evaluation Diffusion (2010).


Frontline practitioners like teachers in public-sector education systems are not policy takers but policy makers, according to Michael Lipsky’s seminal treatise Street-Level Bureaucracy, first published in 1980. They make policy by using their wide autonomy to adopt coping mechanisms, such as limiting client demand and creaming (cherry-picking). Winter and Nielsen have developed this into (1) reducing demand for output, (2) rationing output and (3) automating output. These distinctions are briefly clarified in the article. Do they have relevance for school systems and other Nordic public-sector frontline activities? The question is raised but left to upcoming research to clarify.

Keywords: Street-level bureaucrat; coping strategy; creaming; cherry-picking; autonomy; discretion; frontline practitioner; Lipsky, Michael; Winter, Søren; Nielsen, Vibeke L

Citation: NordSTEP 2015, 1: 28643 -

Copyright: © 2015 Evert Vedung. This is an Open Access article distributed under the terms of the Creative Commons Attribution 4.0 International License, allowing third parties to copy and redistribute the material in any medium or format and to remix, transform, and build upon the material for any purpose, even commercially, provided the original work is properly cited and states its license.

Published: 3 July 2015

*Correspondence to: Evert Vedung, Institute for Housing and Urban Research (IBF), Uppsala University, PO Box 514, SE-75120 Uppsala, Sweden, Email:


Freedom to determine one’s own actions, behaviour, etc. (retrieved 25 February 2015)

Freedom to act or judge on one’s own.
Freedom or authority to make judgements and to act as one sees fit (esp in the phrases at one’s own discretion, at the discretion of).
Power to decide or act according to one’s own judgement. (retrieved 25 February 2015)

In public policy literature, there are several fascinating theories on autonomy of public bureaucracies. The most well known among them include regulatory capture, inertia, path dependency, bureaucratic entrepreneurship, firebrands, product champions, standard operating procedures (SOP), public choice (self-serving interest of the appointed official), and goal displacement. Many of these focus on frontline practitioners and how they use their delegated freedom to determine their own actions, that is, autonomy – or discretion, to employ the conventional expression used in public administration.1

Frontline practitioners are defined here as persons working directly with targets (clients, addressees, pupils, students, patients, regulatees) of the public policy of interest.2 In prescriptive administrative theory, the time-honoured role ascribed to frontline practitioners has been one of the operators at the far end of implementation processes, who apply policy rules of some generality to incoming individual singular cases (individuals and collectivities). The job of the administrators is to carry out policy adopted by decision makers, and the role of frontline practitioners is to carry out the policy administered by the administrators. Frontline practitioners put policy into practice; they are policy takers, not policy makers.

Practitioners’ degree of discretion varies considerably between different policy sectors. Issuers of driver’s licenses and dispensers of national old age pensions are very rule-bound in their jobs, enjoying only minute autonomy for unfettered manoeuvring of their own. They are basically policy takers. By contrast, teachers in secondary schools and university lecturers have broad latitude over how they work. In addition, they do not carry out their jobs in machine-like circumstances but in situations of human interaction. Perhaps, they are policy makers as well.


For this brief article, I will select only one particular discourse among the many on frontline practitioner autonomy and one theory on how some frontline actors handle this discretion. I will concentrate on the autonomy and discretion of the so-called street-level bureaucracy, a discussion that grew out of Michael Lipsky’s seminal treatise Street-Level Bureaucracy: Dilemmas of the Individual in Public Services, first published in 1980. In addition to works by Lipsky, I will use literature produced by my Danish political science colleagues Winter and Nielsen.3

Street-level bureaucrats as defined by Lipsky

According to Lipsky (1980, p. 3), street-level bureaucrats are ‘public service workers who interact directly with citizens in the course of their jobs, and who have substantial discretion in the execution of their work’. Typical street-level bureaucrats are school teachers, social workers, home eldercare providers, health workers, and many other public employees who grant access to government programmes and provide services within them. To this list of Lipsky’s, we might add school supervisors, food and drug inspectors, environmental inspectors, prison guards, and junior lineworkers at local offices of the Swedish Social Insurance Agency or the Employment Agency Office.

In Lipsky’s view, to be a street-level bureaucrat does not necessarily mean that you must be a frontline practitioner who meets clients face-to-face like, for instance, teachers in municipal schools who interact with students. A street-level bureaucrat may physically intermingle with clients but she may also interact through email, letters, or the telephone and still be a street-level bureaucrat.

Coping strategies of street-level bureaucrats to ease unbearable cross-pressure

Lipsky’s work is about the unendurable cross-pressures that street-level bureaucrats experience and how they act to ease them (1980, p. 140). Street-level bureaucrats experience a gulf between the interminable client demands for their services and the limited resources available. Street-level bureaucrats cannot fully meet the quantity of demands from the clients, nor can they meet the substance of the individual client’s demands. They end up in unfavourable situations where they are compelled to use what Lipsky calls coping mechanisms.4 Otherwise, their working day would be psychologically exhausting with never-ending demands (Nielsen, 2006, pp. 864–866; Winter & Nielsen, 2008, p. 128 ff).

However, street-level bureaucrats harbour high ideals and strong ambitions to do good work. They feel bad about using coping mechanisms. Thus, both the bureaucrat and the client are thought to be in a lose–lose situation. The bureaucrat is forced to handle the client differently than she ideally wants to, and the clients are too weak to get what they want, the way they want it (Nielsen, 2006, p. 864; Winter & Nielsen, 2008, p. 122).

To avoid heavy caseloads, street-level bureaucrats use two coping strategies. They try to limit client demand for services. They reduce information dissemination about their amenities, ask clients and inspectees to wait, make themselves unavailable to contacts, or make ample use of referrals of difficult clients to other authorities. Or they resort to creaming. Frontline operators concentrate on a limited number of select clients, programme types, and solutions. They handpick easy, well-defined cases rather than difficult, amorphous, and time-consuming ones. They attend to cases that promise to be successful and downplay the trickiest ones.

Coping strategies are so common that they lead implementation astray. Thus, Lipsky (1980) concludes that street-level bureaucrats are policy makers: they actually create policy through the multitude of decisions they make in interacting with clients. In other words, policies are formed during the implementation process by programme operators as they develop routines and shortcuts for coping with their everyday jobs.

The Winter–Nielsen extensions of Lipsky’s coping strategies

The Danish political scientists Winter and Nielsen (2008, pp. 116–126) have expanded the idea of coping strategies into a notable scheme of fundamental types, of which I have selected the following three for brief presentations:

  1. Reducing demand for output (Lipsky: limit client demand).
  2. Rationing output (Lipsky: creaming).
  3. Automating output (Lipsky: not noticed).

The reducing-demand-for-output mechanism includes the following three subgroups:

  1. Limiting information to clients on services they are entitled to.
  2. Forcing clients to form long waiting lines before getting appointments.
  3. Making access to frontline workers cumbersome through incomprehensible application forms, short and inconvenient opening hours, inadequate signage to pertinent offices inside the building, and complex procedures in general (Winter & Nielsen, 2008, p. 116 f).

In the rationing output strategy, frontline workers focus their efforts on a limited number of selected clients or solutions, not on the whole target group or a wider range of solutions. Here, I shall concentrate on creaming.5 Winter and Nielsen (2008, p. 117 f) describe three different types:

  1. Creaming for substantive success.
  2. Creaming for cost efficiency.
  3. Creaming for quantitative improvement.

In line with Lipsky, all creaming involves cherry-picking. Cherry-picking refers to selecting or accepting the best or most desirable, for example, accepting the most desirable people or things in a group (see cherry-picking, 2015). In creaming for substantive success, frontline operators choose to work with clients with good prognoses for recovery and/or rehabilitation and where they may use well-acknowledged or prestigious treatment technologies. Substantive success implies selecting cases that promise rapid and inexpensive cures, not those who are most in need or first in line. The idea behind this type of creaming is to further the achievement of substantive outcome goals.

Creaming for cost efficiency entails prioritising cases that contribute most to the composite cost-effectiveness of the whole office or unit. In this case, frontline operators direct their time and energy at cases whose solutions will probably promote goal achievement per working day or some other cost unit more than alternative cases. Environmental inspectors may choose to concentrate their inspections on large enterprises and neglect small ones to produce more improved environment per euro or Danish krone.

Frontline operators may also focus their cherry-picking on creaming for quantitative improvement. The aim is to carry out many outputs per unit of time, while ignoring the outcomes. In public health care, focus is on the quantity of services performed, for example, the number of surgeries performed in a clinic (outputs), not on whether patients have become healthier (outcome) through these surgeries. Library operators concentrate on lending as many books as possible (outputs), without bothering much about whether the borrowers get valuable entertainment or education out of reading them (outcome).

Research on how data from quasi-permanent indicator systems are used in after-the-fact auditing shows that adherence to data on output indicators instead of data on outcome indicators may produce very negative constitutive effects in affected agencies; agency leadership may begin to push for increasing the number of outputs without considering whether the outcome results are encouraging or not (Dahler-Larsen, 2007, 2011, 2012). Unintentionally, this may encourage goal displacement (Nielsen, 2006, p. 865). What would the mechanism be? Bohte and Meier have put it succinctly: ‘When agency performance is evaluated in terms of numerical outputs, bureaucrats have an incentive to maximize outputs, regardless of whether maximizing outputs is the preferred strategy for achieving desired social outcomes (a form of goal displacement). This incentive to maximize outputs may lead to organizational cheating, where public agencies purposely manipulate output levels to portray their work in the best light possible’ (2000, p. 173).

In automating output – coping strategy type 3, not noticed by Lipsky – frontline operators adopt SOP to achieve a reasonable workload (Winter & Nielsen, 2008, p. 117, 120f). SOP are work-saving institutions in decision situations that tend to surface regularly. ‘SOP are a detailed explanation of how a policy is to be implemented. The SOP may appear on the same form as a policy or it may appear in a separate document. The main difference between a SOP and a policy is details. An effective SOP communicates who will perform the task, what materials are necessary, where the task will take place, when the task shall be performed, and how the person will execute the task’ (SOP, 2015). ‘SOPs help government organizations, emergency response operation[s], and clinical research organization[s] achieve maximum safety and operational efficiency’ (SOP, 2014).

Within or even somewhat outside the substantive jurisdictional area created by the larger intervention, a local intervention is created that might govern the street-level bureaucrats and affect the continued implementation of the larger intervention.

Reasons for the delegation of power and wide autonomy

In adopting coping strategies, street-level bureaucrats use the large latitude some of them have in their work. Why do many (but not all) frontline practitioners have wide autonomy to make decisions of their own? The first answer is that higher level principals have delegated their decision-making power to them. To delegate means

  1. to entrust to another and to entrust authority to another, and
  2. to appoint as one’s representative or agent (see Delegate, 2015).

Why do principals delegate power within wide jurisdictions (i.e. autonomy) to lower level agents, including street-level bureaucrats? Among the several reasons for this, I will dwell upon just two.

First of all, delegation occurs because governing bodies lack competence to make appropriate local decisions. An overwhelming majority of all public-sector resolutions concern individual instances or specific situations. Cases in point are the placement of patients in lines for surgery, teachers’ decisions concerning grades for pupils, or the allocation of government housing loans to individual building commissioners. Often, these decisions have to be made on a continuous basis, for example, by the doctor in her clinic, the teacher in her classes, or the care provider to the clientele in municipal homes for the elderly. The resolutions require far more expert knowledge of the specific situation than higher level principals reasonably can possess. They lack the necessary professional knowledge and knowledge of the individual clients. The need for situational adaptation with regard to the quality of client information is so demanding that higher level officials have no competence to handle these types of decisions.

In addition, it is not practically possible either for principals at the pinnacles of pyramids to determine such low-level, street-level work in detail before the fact. The performance of the school teacher in her classroom and the social worker in his encounters with assistance seekers is too complicated to be regulated from the top. Partly, this is due to the joint production between practitioners and clients. Joint production means that the content of service delivery is determined by the frontline practitioners after dialogues and discussions with the clients; such collaborative content of human interactions cannot be specified in advance and across the board.

Second, delegation occurs because governing bodies lack time to make well-informed local decisions. Usually, higher level officials make fundamental decisions in the same substantive area only every 2, 3, or 4 years. However, in most functional policy domains, decisions have to be made several times a month, a week, and quite often even a day, in a continuous manner. The parliamentary resolutions in these cases must, therefore, be general and contain guidelines only. This in turn means that many extremely important decisions for citizens and clients must be made incessantly by bodies that have their powers delegated to them by parliamentary institutions or by authorities that in turn have their powers delegated to them by bodies that derive their power from parliamentary institutions.

Autonomy and information asymmetry advantage frontline practitioners

In adopting coping strategies, street-level bureaucrats run relatively small risks of being discovered. First, they are physically separated from their principals. Second, there are never enough resources to provide close, frequent and direct supervision of them. Third, there are no precise performance criteria that specify exactly how a teacher, an engineer, a public health nurse, or a social worker should do their job. Finally and most importantly, there is a considerable information asymmetry in their favour.

Lipsky’s view of coping behaviour among street-level bureaucrats seems to be in line with the so-called public choice theory. Public choice theory departs from the economic principal–agent theory. Agents (executives) may act unfaithfully towards their principals because of the enormous information asymmetry in their favour.

Teachers who intermingle on a day-to-day basis with students and pupils know infinitely more about the school’s operations at the street level than the rector (principal), who almost never meets a student in a teaching situation. It is information asymmetry of this type that enables self-regarding behaviour. This mechanism will influence implementation.

Coping behaviour is universally valid for all street-level bureaucrats, maintains Lipsky (Winter & Nielsen, 2008, p. 141 f). Once cross-pressure occurs, street-level bureaucrats of all sectors are compelled to adopt coping strategies to secure a fair and manageable workload. Against this, Winter and Nielsen (2008, p. 128 ff) argue that some street-level bureaucrats are enticed, not forced to cope. At least, this goes for a country like Denmark with a welfare-state regime of the Nordic type. Optimising job satisfaction or good relations with clients might be drivers, particularly in implementing regulation (Nielsen, 2006, p. 867 ff.) For such cases, mechanisms of optimisation might be the proper expression to use. Street-level bureaucrats may adopt both coping and optimising mechanisms.


Coping mechanisms, limiting client demand for output, creaming, rationing output, or automating output through SOP: Do these notions have any relevance for teachers’ behaviour towards students in schools and top–down governance from national authorities? Do they have any relevance for other Nordic public-sector frontline activities? In this context, I can only raise the question. Providing answers to it must be left to future upcoming scholarship and science.


Bohte, J., & Meier, K.J. (2000). Goal displacement: Assessing the motivation for organizational change. Journal of Public Administration, 60, 173–182.

Cherry-picking. (2015). Retrieved February 28, 2015, from

Dahler-Larsen, P. (2007). Constitutive effects of performance indicator systems. In S. Kushner & N. Norris (Eds.), Dilemmas of engagement. Evaluation and the new public management (pp. 17–36) (Advances in Program Evaluation; 10). Amsterdam: Heinemann.

Dahler-Larsen, P. (2011). Constitutive effects as a social accomplishment: A qualitative study of the political in testing. Education Inquiry, 3(2), 171–186.

Dahler-Larsen, P. (2012). The evaluation society. Stanford: Stanford University Press.

Delegate, delegation. (2015). Retrieved February 28, 2015, from

Durose, C. (2011). Revisiting Lipsky: Front-line work in UK local governance. Political Studies, 59(4), 978–995.

Ellis, K. (2011). Street-level bureaucracy’ revisited: The changing face of frontline discretion in adult social care in England. Social Policy & Administration, 45(3), 221–244.

Høybye-Mortensen, M. (2011). Velfærdsstatens dørvogtere: Procesregulering af visitationer på socialområdet [Gatekeepers of the welfare state: Administrative procedures in social work decision-making]. Copenhagen: Department of Political Science, University of Copenhagen.

Jewell, C.J. (2007). Agents of the welfare state: How caseworkers respond to need in the United States, Germany, and Sweden. New York: Palgrave Macmillan.

Johansson, R. (1992). Vid byråkratins gränser: Om handlingsfrihetens organisatoriska begränsningar i klientrelaterat arbete [At the boundaries of bureaucracy: organizational constraints on discretion in client-related work]. Lund: Arkiv.

Kaufman, H. (1967). The forest ranger: A study in administrative behavior. Baltimore, MD: Johns Hopkins Press.

Kelman, S. (1981). Regulating America, regulating Sweden: A comparative study of occupational safety and health policy. Cambridge, MA: MIT Press.

Lipsky, M. (1980). Street level bureaucracy: Dilemmas of the individual in public service (30th Anniversary expanded edn in 2010). New York: Russell Sage Foundation.

Lundberg, L. (1982). Från lag till arbetsmiljö [From law to work environment]. Malmö: Liber.

Municio, I. (1987). Från lag till bruk: Hemspråksreformens genomförande [From law to use: Implementation of the home language reform]. Stockholm: Statsvetenskapliga institutionen, Stockholms universitet.

Nielsen, V.L. (2006). Are street-level bureaucrats compelled or enticed to cope? Public Administration, 84(4), 861–889.

Sevä, M. (2015). The decisive role of street-level bureaucrats in environmental management. Luleå: Luleå University of Technology.

Skolverket. (2015). Reform i skolpraktik: Om implementeringen av reformer i grundskolan 2011–2014 [Reform in school practice: on implementaion of reforms in elementary schools 2011–2014]. Stockholm: Skolverket (Marika Sanne).

SOP (Standard Operating Procedure). (2014). Retrieved August 30, 2014, from

SOP (Standard Operating Procedure). (2015). Retrieved January 2, 2015, from

Stjernquist, P. (1973). Laws in the forest: A study of public direction of Swedish private forestry. Lund: CWK Gleerup.

Vedung, E. (1997). Public policy and program evaluation. Piscataway, NJ: Transaction Publishers.

Vedung, E. (2016). Implementering i politik och förvaltning [Implementation in public policy and public administration]. Lund: Studentlitteratur.

Wilson, J.Q. (1978). The investigators: Managing FBI and narcotics agents. New York: Basic Books.

Winter, S. (2003). Studying the implementation of public programs. In B.G. Peters & J. Pierre (Eds.), Handbook of public administration (pp. 205–212). London: Sage.

Winter, S. (2006). Implementation. In B.G. Peters & J. Pierre (Eds.), Handbook of public policy (pp. 151–156). London: Sage.

Winter, S., & Nielsen, V.L. (2008). Implementering af politik [Implementation of public policy]. Århus: Academica.


1For an overview of these theories, see Vedung (1997, pp. 209–245), Vedung (2016), Winter (2003, 2006), and Winter and Nielsen (2008).

2Synonyms of frontline practitioners include frontline staff, lineworkers, operators, and field officers. In Swedish: närbyråkrat, gräsrotsbyråkrat, baspersonal, handläggare, fältpersonal, personal; in Danish: markarbejdere.

3Winter and Nielsen (2008, pp. 103–143), Nielsen (2006). Other Lipsky followers besides Winter and Nielsen include Durose (2011), Ellis (2011) and Høybye-Mortensen (2011). Jewell (2007), Johansson (1992), Kaufman (1967), Kelman (1981), Lundberg (1982), Municio (1987), Sevä (2015), Skolverket (2015), Stjernquist (1973) and Wilson (1978) are examples of empirical studies of frontline staff.

4In Swedish: bemästringsstrategier; in Danish: mestringsstategier, afværgemekanismer.

5I have excluded the other three categories due to interpretation difficulties. In Swedish: gräddskumning, gräddskummeri, skumma-grädden-av-mjölken and plocka-russinen-ur-kakan.

About The Author

Evert Vedung


Article Metrics

Metrics Loading ...

Metrics powered by PLOS ALM

Related Content