Keywords

Introduction

Every year, around 1.3 million people are killed on the road, and another 20–50 million are severely injured (WHO 2018). Data from the World Health Organization (WHO) show that around 90% of the fatalities occur in middle- and low-income countries. Approximately half of those killed are vulnerable road users, including pedestrians, cyclists and motorcyclists, children, and elderly people. At present, road traffic accidents are the dominant cause of death among people aged 5–29 years, and they are expected to be the seventh leading cause of death in 2030. Thus, road safety is one of the most critical global public health issues to date. In order to stabilize and then reverse the negative trend, the international community has responded with the adoption of road safety targets aimed at reducing the number of killed and seriously injured people on the road. In March 2010, the United Nations (UN) General Assembly proclaimed the period 2011–2020 as the “Decade of Action for Road Safety” (United Nations 2010). Since 2015, the UN member states have been bound by the Sustainable Development Goals (SDGs), which include the goal to halve the number of global deaths and injuries from road traffic accidents by 2020 (SDG 3.6; see also SDG 11.2). In November 2017, the UN agreed on a subset of global road safety targets designed to advance current efforts toward the 2020 goal (https://etsc.eu/un-agrees-on-road-safety-sub-targets-to-aid-progress-on-2020-sustainable-development-goals/. Accessed 20.01.2020). As noted by the Organization for Economic Co-operation and Development (OECD) and the International Transport Forum (ITF), this is the “strongest ever mandate for action on road safety” (ITF 2016, p. 17). Road safety targets have also been adopted at the national and city/state levels in Sweden (Tingvall and Haworth 1999; Johansson 2009; Belin et al. 2012; McAndrews 2013), Norway (Elvik 2008), the Netherlands (Wesemann et al. 2010), Australia (Corben et al. 2010), the United States (Cushing et al. 2016; Evenson et al. 2018; NYC 2018), the European Union (EU) (European Commission 2011; Tolón-Becerra et al. 2014), and many other countries (see Part 2 of this handbook).

The frequent use of goals and targets in road safety policy and management makes it interesting to investigate under what conditions the adopted goals and targets can be expected to be achieved. The preconditions for successful management by objectives (MBO) in the public sector have been studied extensively. As a result, an array of factors leading to inadequate goal fulfilment have been identified, including insufficiently operational goals and targets (Elvik 1993; Edvardsson and Hansson 2005) and insufficiently calibrated goal evaluation methodologies (Larsson and Hanberger 2016), gaming behavior (Smith 1995; Propper and Wilson 2003; Bevan and Hood 2006), and lack of adequate communication channels between different governance or administrative levels responsible for implementing such goals (Wibeck et al. 2006). Much of this literature has addressed other policy areas apart from transportation and road safety, such as environmental and climate policy (Lundqvist 2004; Edvardsson Björnberg 2009, 2013) or public health in a broader sense (Lager et al. 2007; Smith and Busse 2010). However, exceptions exist. As early as 1993, Elvik addressed the question of what criteria road safety targets must satisfy in order to better guide the rational choice of means (1993). A decade later, Rosencrantz et al. (2007) used a set of criteria, which resembled the politically more established SMART criteria, to analyze the rationality of the Swedish Vision Zero for traffic safety. Drawing on the literature in the field, Elvik (2008) identified seven conditions for successful road safety MBO, some of which concerned the goals themselves (“challenging, yet in principle achievable,” “there should not be too many targets in view of the available policy instruments designed to realize them,” etc.) while others concerned the policy context in which the goals were to be implemented (“responsible agencies should be supplied with sufficient funding to implement all cost-effective road safety measures,” “incentives should exist to ensure commitment to targets from all agencies responsible for realizing them,” etc.) (Elvik 2008, p. 1116). Sobis and Okouma (2017) analyzed the use of MBO in transportation service for the disabled in the municipality of Gothenburg (Sweden). They identified two factors that contributed to its successful outcomes: that MBO was used in combination with other steering practices and the establishment of a new organizational culture that emphasized the importance of information exchange between politicians and civil servants regarding goal evaluation and achievement.

This chapter summarizes the main themes and conclusions of research that have been conducted on goal-setting in road safety policy and management to date. The primary focus of the chapter is on the rationality of the goals and targets themselves, rather than on the policy context in which the goals are to be implemented. Drawing on previously published research on rational goal-setting (Edvardsson and Hansson 2005; Rosencrantz et al. 2007; Edvardsson Björnberg 2016), the chapter aims to address the following research questions:

  • What are the potential benefits of using goal-setting in road safety management?

  • When is a road safety goal or target rational (functional), that is, what criteria must the goal fulfil in order to be operative?

  • What is the ideal number of road safety goals within a system of goals?

  • How should road safety goals and sub-goals be aligned?

  • Why does a road safety goal have to be stable over time?

  • Who should be involved in the road safety goal-setting process?

Before proceeding to the research questions, the next section presents a brief history of the development of road safety policy and the use of road safety goals and targets in the United Kingdom and the United States. In the following, the terms “goal,” “target,” and “objective”’ will be used interchangeably.

A Brief History of Goal-Setting in Road Safety Management

The invention of the modern car is commonly attributed to the German engineer and designer Karl Benz. His Benz Patent-Motorwagen, for which he received a patent in 1886, is considered by many to be the first power vehicle using an internal combustion engine. In the early days of motorcar invention, only limited numbers of cars for each brand were produced and used on the roads, which meant the number of car accidents and fatalities was low (Accidents involving coaches and pedestrians were obviously a cause of concern even before the motorcar was invented, thus leading to the implementation of safety measures, such as prohibiting minors from being cart drivers (Gregersen 2016). In one tragic instance, Pierre Curie, husband of Marie Curie, was killed when he fell under a horse-drawn cart in Paris in 1906.). It was not until large-scale manufacturing was introduced in the early twentieth century that road traffic situations remotely similar to the present one arose. Among the early organizations that noticed the problem of road safety were the National Safety Council (NSC) in the United States and the London Safety First Council (known today as the Royal Society for the Prevention of Accidents, RoSPA) in the United Kingdom. The NSC was established in 1913 with the aim of promoting health and safety initially in the workplace until the organization eventually expanded its efforts to include road safety, among other things. The London Safety First Council was established in 1916 in connection with a meeting convened by the operating manager of the London General Omnibus to address the “alarming increase in traffic accidents, and the direct connection therewith of the restricted street lighting which had been necessitated by the War conditions” (www.rospa.com; Jackson 1995).

Much of the work carried out by the early road safety organizations targeted the behaviors of individual road users, to whom accidents were causally attributed. In line with this, it was agreed that the behavior of individual road users, particularly children and drivers, had to be modified to fit the current road transport system. Among the early road safety measures initiated by the RoSPA were educational campaigns and road safety competitions targeting school children and professional drivers, respectively (www.rospa.com). The “traditional” behaviorally oriented approach to road safety management became very influential in the early twentieth century and is still the predominant approach to road safety management in many countries (Johansson 2009; Belin et al. 2012). However, other types of road safety measures have also been implemented, including improvements in road infrastructure and vehicle design (Oster Jr and Strong 2013). For instance, electric red-green traffic lights were introduced in the United States in 1912–1914 (https://en.wikipedia.org/wiki/Traffic_light#History. Accessed 09 Jan 2020). In 1922, the London Safety First Council suggested improvements in street lighting and argued for the marking of road crossings frequently in use (www.rospa.com). The first seat belt patent in the United States was secured by Edward J. Claghorn in 1885, but it would take another 65 years before seat belts were made available in American-made cars. In Britain, vehicle braking requirements were introduced by the Motor Car Act 1903, and further requirements concerning the construction, weight, and equipment of cars were introduced by the Road Traffic Act 1930.

Therefore, the early road safety work was, to a significant degree, advocated for and implemented by voluntary organizations. Although national and federal state governments were involved in establishing laws on speed limits, vehicle registration, and driver licensing—notable examples were the British Locomotive Acts (1861–1878) and the New York drunk driving laws (1910)—it was not until much later, around 1930, that national and federal state governments started to organize their work with the pronounced aim to increase road safety (For more details on the specifics of these regulations, see The Locomotives on Highways Act 1861, The Locomotive Act 1865, and the Highways and Locomotives (Amendment) Act 1878. See also Motor Car Act 1903 and Road Traffic Act 1930.). A first step in this direction was taken in 1933 when the British Government decided to start investigating the causes of road accidents. In fact, the RoSPA had already put forward calls for accident causation analysis to the British Minister of Transport in as early as 1928.

In the decades that followed the Second World War, road safety work became increasingly systematic and institutionalized. Road safety policies and plans were developed, coordinated, and implemented by governmental agencies given a parliamentary or congressional mandate to improve road safety. In the United States, the Department of Transportation was created by the Congress in 1966 with the mission to “serve the United States by ensuring a fast, safe, efficient, accessible and convenient transportation system that meets our vital national interests and enhances the quality of life of the American people, today and into the future.” Four years later, in 1970, the NHTSA was established by the Highway Safety Act and was given the responsibility to reduce deaths, injuries, and economic losses resulting from motor vehicle accidents.

A crucial feature of the systematic and institutionalized approach to road safety policy was the introduction of quantified road safety targets. The idea of MBO had been developed by Peter F. Drucker in the 1950s as a way of managing business corporations (Drucker 1954) (In the academic literature, especially in the Nordic countries, the term “management by objectives and results” (MBOR) is frequently used (e.g., Lundqvist 2004; Steineke and Hedin 2008; Kristiansen 2015). “Performance management” is a related management philosophy (Ammons and Roenigk 2015).). The basic idea of MBO is that business corporations can be effectively and efficiently managed if work departures from and is evaluated against a set of objectives set by senior managers and then implemented by the employees. In the 1980s, MBO was gradually implemented in the public sector, where it subsequently became a central feature of the so-called new public management (NPM) (Hood 1991) (Kristiansen (2015) argues that the origins of MBO in the Nordic countries can be traced back to the 1960s, 1970s, or 1980s, depending on which aspect of MBO is accentuated. In Sweden, for example, MBO-related ideas were already discussed in the 1960s in relation to budget reforms, although a comprehensive MBO reform was not launched until the 1980s (Sundström 2006).). In a public sector context, MBO typically encompasses a mixture of political control and administrative autonomy and discretion (Christensen and Laegreid 2001; Lundqvist 2004; Sundström 2006):

  • Policy goals are adopted by politicians, while the responsibility for implementing them lies with the subordinate bodies. The implementing agencies enjoy a considerable degree of discretion in deciding on which measures to take to achieve the goals.

  • Progress toward goal achievement is monitored, measured, and reported back to the politicians on a regular basis.

  • The politicians are responsible for making policy adjustments and strategic decisions based on the reported results, for instance, on resource allocation (MBO systems typically reward agencies whose results are satisfactory and, conversely, punish agencies with insufficient goal achievement; however, this is not a pronounced feature of MBO in, for example, the Swedish public sector, where failure to achieve a policy goal is often regarded as a reason for adding funds (Steineke and Hedin 2008).

Thus, translated into a road safety policy context, MBO essentially means that road safety goals are formulated by the politicians, while the responsibility for implementation, evaluation, and feedback is delegated to another government body, in many cases the national or state road transport administration. The underlying rationale for governing by objectives rather than by rules or direct instructions is efficiency, both in economic terms and based on the belief that the administration knows how to best effectuate the intentions of the politicians.

It is obviously difficult to locate the exact point in time when a national government first quantified road safety targets. Nevertheless, the Lalonde Report, published by the Canadian Minister of National Health and Welfare in 1974, is sometimes identified as a milestone in this development (Belin et al. 2010). In the Lalonde Report, setting quantitative public health targets was introduced as a means of stimulating and coordinating governmental efforts toward reduced mortality and morbidity. The goal-setting strategy set out in the report was later taken up by the World Health Organization (WHO) and put to use by national public health authorities, which began to adopt quantified road safety targets (OECD/ITF 2016). Sweden was one of the early adopters, with quantified road safety targets being discussed as early as 1972 and eventually decided by the Parliament in 1982 (Belin et al. 2010).

Why Use Quantified Goals in Road Safety Management?

As noted above, early road safety work was not policy-driven but largely uncoordinated and implemented by different actors in an ad hoc fashion. Measures were certainly introduced with a clear goal in mind, namely, to reduce the number of deaths and injuries caused by road accidents. However, these measures were typically not part of a strategy or plan adopted with the aim of coordinating actions across time or among agents. Nevertheless, in many respects, the road safety measures introduced in the early days of widespread automobile use were successful in saving lives and averting harm. This brings forward the question of why road safety goals and targets had to be introduced at all. More generally, what are the potential benefits of introducing goal-setting as a management tool in road safety work?

Goals are typically set because the individual or organization (henceforth “agent”) who sets the goal wants to achieve the state of affair referred to in the goal and because they believe that it becomes easier to achieve the desired states of affairs by setting the goal. Edvardsson and Hansson (2005) use the term “achievement-inducing” to denote those goals that contribute to their own achievement, i.e., goals that are rational (functional, or operative). There are two ways by which goals can be achievement-inducing. First, an adopted goal can be used to plan and coordinate action toward goal achievement over time and between agents. With the help of a goal, a road safety organization can plan and allocate work tasks to different departments or administrative units while ensuring that everyone knows what to do, when to do it, and how what is being done fits into what other departments or units are doing. For instance, a goal such as Vision Zero for road safety can be used to guide the selection of strategies or the adoption of goals and targets further down in the administrative chain (Tingvall and Haworth 1999). Moreover, an adopted goal can be used to induce and sustain action among the organization’s employees. It may invigorate the commitment of the organization and its employees to road safety. Thus, Wong et al. (2006) argue that the role of road safety targets is “to provide a basis for motivating and monitoring actions to reduce death and injury in road traffic crashes” (p. 997).

Past studies have provided considerable empirical evidence to support the idea that goal-setting can have a positive effect on an individual’s performance and, hence, be conducive to goal achievement (Locke and Latham 1990, 2002). Goals that are specific and challenging have, for instance, been shown to affect an individual employee’s choices, efforts, and persistence in such a way that goal achievement is furthered (Latham et al. 2008). Although significantly less empirical evidence exists on public policy goals and how they affect organizational output (Jung 2014), studies suggest that the positive relationship between goal-setting and goal achievement may also hold true in this case. Elvik (1993) analyzed how road safety performance differed among Norwegian counties during the years 1982–1985 and 1986–1989 based on whether quantitative or qualitative road safety targets had been adopted. He showed that quantified road safety targets were more successful than qualitative targets in reducing the accident rate per kilometer; moreover, “the best performance was achieved by counties with highly ambitious quantified targets” (p. 569). Wong et al. (2006) investigated the association between goal-setting and road safety improvement during the period 1981–1999. In their study, they investigated 14 countries that had adopted road safety management targets and examined the road fatalities before and after the setting of the target in each country. The results showed that there was a significant overall reduction in road fatalities after quantified road safety targets had been adopted. On the basis of these findings, the authors argued that road safety goal-setting “helps to raise concern about road safety in societies, encourages decision-makers to formulate effective road safety strategies, and ensures that sufficient resources are allocated to road safety programs” (p. 1004) (The results were later updated by Allsop, Sze, and Wong (2011), who made some changes in the numerical estimates used. However, to the author’s understanding and as pointed out by the authors of the 2011 publication, the changes did not alter the main argument of the previous paper.). In line with these studies, the ITF (2008) concluded that countries that have adopted quantitative road safety targets do better than countries with no such targets.

However, governance based on goals and targets also has some potential drawbacks. Bevan and Hood (2006) identified several ways in which public MBO systems (in their case, public health service targets) can be vulnerable to gaming. Among other things, they discussed the practice, common among public agencies, to adopt goals based on what has been achieved recently. However, this practice has an unfortunate consequence: managers and politicians who expect to gain a renewed term of office tend not to exceed adopted goals even if they could, as that would put greater expectations on their future performance. It remains unclear to what extent Bevan and Hood’s (2006) findings can be translated into a road safety management context. Their analysis focused on the English public health service, which at the time of writing was managed through a system combining goals and targets with awards and punishments for (in)sufficient goal achievement. Such awards and punishments are not used in Swedish road safety work, to take one example; hence, the risk of gaming is arguably smaller. However, it is important to keep in mind that gaming could potentially occur in any MBO system depending on how it is set up, specifically on whether it grants awards and dispenses punishments based on performance outcomes.

When Is a Road Safety Target Rational?

Goals, therefore, have an important role in directing and motivating action, over time and among agents. However, to fulfil this action-guiding and action-motivating role, the goals must satisfy certain criteria. In the management and psychology literature, those criteria are often referred to as the “SMART criteria,” according to which goals should be specific, measurable, attainable, relevant, and time-bound (See Rubin (2002) for an elaboration of what the SMART acronym stands for.). Edvardsson and Hansson (2005) proposed a set of rationality criteria that are similar to the SMART criteria. They argued that goals should be precise, evaluable, approachable, and motivational. Other criteria are also conceivable. Edvardsson Björnberg (2016) examined the extent to which it is considered rationally justified for goals to be stable over time. Rosencrantz (2008) and Edvardsson Björnberg (2009) suggested that additional criteria, such as consistency, comprehensiveness, and non-redundancy, may apply to systems of goals. Elvik (1993) identified two sets of requirements that a target must satisfy to serve as a basis for the rational choice of means. The first concerns the relationship between the target and the values or preferences that the target expresses. The target should be “operational,” which means that the underlying preferences should satisfy the requirements of transitivity and completeness. The second set of requirements concerns the relationship between the target and the state of affairs to which the target refers. Elvik (1993) identified three such requirements: (1) the targets should not be self-contradictory, by which he means that they should not be “formulated in a way that makes their fulfillment logically impossible” (cf. Hansson et al. (2016) on self-defeating goals); (2) they should not be tautologically fulfilled, i.e., “formulated in a way that will make any outcome fulfill the target”; and (3) they should refer to “targetable” outcomes, i.e., “not to outcomes that are essentially by-products” (p. 570).

Whether or not a goal will have the capacity to guide and motivate action does not only depend on the characteristics of the goal itself, such as precision or measurability. The context in which the goal is set and the process of setting such a goal could also determine how effective it will be in regulating action. This appears to be the case not the least in a road safety management context. Elvik (2008) identified seven conditions for successful road safety MBO, some of which concern the goals and targets themselves, while others relate to the policy context in which they are to be implemented. Among the contextual factors identified by Elvik are endorsement of the goals by the top management (e.g., politicians, ministries of transport, road safety authorities), availability of funding sufficient for implementing the goals, and the establishment of a system for monitoring progress and providing feedback to the responsible agencies, among others.

Below, the two categories of rationality criteria—for individual goals and systems of goals—will be discussed separately before proceeding to the contextual conditions.

Precision

A goal can guide action only when the implementing agent knows what the state of affairs referred to by the goal is and to what extent the actions undertaken bring him/her closer to those state of affairs. This requires that the goal is both precise and possible to evaluate. Edvardsson and Hansson (2005) argued that, of these two criteria, precision is more fundamental, because evaluability presupposes that the desired state of affairs referred to by the goals is reasonably clear. The criterion of precision can be divided into at least three subcategories. Edvardsson and Hansson (2005) distinguished between directional, completive, and temporal precision, where directional and completive precision correspond to the SMART criterion of specificity and temporal precision to the criterion that a goal should be “time-bound.” Directional precision means that the goal specifies in what direction the implementing agent should go in order to reach the goal. Consider the following example:

  1. 1.

    There should be a decrease in the number of people killed on the road.

    Assuming that the number of fatalities can be determined in a reasonably authoritative way from 1 year to another, the goal can be said to have directional precision. However, the goal does not tell the agent to what degree the goal should be achieved. Thus, it lacks completive precision. Consider instead the following goal:

  2. 2.

    Nobody should be killed in the road traffic system.

    This goal has both directional precision (decrease in the number of fatalities) and completive precision (nobody). However, it lacks temporal precision, because it does not tell the agent within what time period the goal should be achieved. The following goal has directional, completive, and temporal precision:

  3. 3.

    The number of people killed on the roads should be decreased by 5% annually between 2020 and 2030.

Similar distinctions between different forms of goal precision have been made by other authors.

Elvik (1993) distinguished between “open targets,” “semi-open targets,” and “closed targets.” Open targets are qualitative and are not specified to any degree or time. They correspond to Edvardsson and Hansson’s (2005) directionally precise targets. Semi-closed targets are either quantified in time or have a quantified level component. They correspond to Edvardsson and Hansson’s (2005) temporally and completely precise targets. Finally, closed targets have both temporal and completive precision.

Imprecise goals, or “goal ambiguity,” refer to a fairly common phenomenon in public organizations (Chun and Rainey 2005; Rainey 2014) (See Chun and Rainey (2005, p. 2) for a definition of “organizational goal ambiguity.”). Carrigan (2018) identified a number of reasons behind the relative proliferation of ambiguous goals in public policy. Public policy goals are typically the result of political compromise. In political contexts, keeping a goal vague is advantageous, because one may more easily gain broad support for it than if it is specified in greater detail. From a political point of view, it may be more expedient to propose a vague goal on the basis of which political unity can be sought and then work out the details at a later point when a reasonable degree of consensus surrounding the importance of the issue has been achieved. Thus, Chun and Rainey (2005) concluded that “goal clarification is often considered ‘managerially sound’ but ‘politically irrational’ in the public sector” (p. 23). Moreover, goals can be specified as part of the implementation process. Sometimes, adopting a goal and working out the details are more efficient, especially after accumulating knowledge about the policy issue and how it can be addressed. Many implementation issues can be difficult to foresee; therefore, it may be more efficient to allow the implementing agencies to specify precisely what to achieve (cf. Lindblom 1959). Finally, politicians may find it helpful to keep goals imprecise, as policy pronouncements in the form of clearly stated goals are often used as benchmarks against which performance is measured and votes are driven.

Organizational goal ambiguity has been considered problematic in the goal-setting literature. Imprecise goals are obviously more difficult to follow up and evaluate, and it may be more difficult for a group of agents working together toward the goals to organize their activities if the goals are imprecise and the agents lack the necessary background knowledge to coordinate their efforts efficiently. Imprecise goals may also decrease organizational commitment. Thus, Jung and Ritz (2014) argue that an ambiguous goal could “make it difficult for public employees clearly and effectively to determine how much and what kinds of effort they should make for the organization and the attainment of organizational goals, or in which direction to give effort and how to make task plans” (p. 467) (See Jung and Rainey (2011) for a discussion of the relationship between goal precision and government employee motivation.).

How precise a goal needs to be in order to have the capacity to guide action typically varies depending on the social context and the implementing agents’ background knowledge. Psychological studies have shown that for individuals, precise goals are typically more conducive to goal achievement than “do-your-best” goals; however, there are exceptions (Locke and Latham 2002). When the implementing agent is in a learning process and does not yet have sufficient knowledge to work toward an outcome-oriented goal, it may be better to set “do-your-best” outcome goals and then supplement them with specific learning goals. In those situations, specific high-performance outcome goals can be detrimental to goal achievement, as they can detract the agent from the search for an appropriate strategy. Consequently, Latham et al. (2008) concluded that “when effective behavioral routines have yet to be developed, a specific high learning goal rather than a performance one should be set” (p. 390). These findings may have some bearing on the issue of goal-setting in road safety management. Arguably, in countries and organizations with little previous knowledge about road safety management and where those responsible for implementing the targets are in a learning process, it could be more effective to adopt specific high learning goals (related to road safety management) rather than some specific performance outcome goal.

Evaluability

While precision concerns the goal (desired end-state) itself, evaluability concerns the agent’s actions and their effects on goal achievement. In this sense, precision is a more fundamental criterion, because evaluability depends on it. Goal-setting theory assumes that goals regulate performance more reliably when work is evaluated and information about how far one has come in relation to the goal is fed back to the goal-setter and/or implementer. Feedback and feedback mechanisms operate on different levels (Edvardsson and Hansson 2005). First, an agent or group of agents who are given information about where they stand in relation to the goal can more easily adjust their actions so as to further goal achievement more effectively. Second, such information can also be used to revise the goal itself. In many situations, it is difficult to know in advance whether or not a certain goal, for example, a road safety management target, is reasonably ambitious (see below). In such cases, information from evaluations may be necessary to adjust the level of goal difficulty. Third, adopting goals that are both precise and possible to evaluate is a prerequisite for establishing accountability for insufficient goal achievement. (Hence, for the incentive for political decision-makers to adopt vague goals that are difficult to evaluate, see above.) Finally, goals that are evaluable and evaluated could have a motivating function; see below.

Successful goal evaluation presupposes both that the goal itself (the desired end state) is clear and that it is possible to assess the degrees of goal achievement. In some situations, one and the same goal can be evaluated on the basis of more than one parameter. This is, for instance, the case with the Swedish Vision Zero for road traffic safety. The Swedish Vision Zero states that nobody should be killed or seriously injured on the road. One could imagine a situation wherein there is a decrease in the number of killed people while the number of seriously injured people is increasing at the same time, or vice versa. It is not entirely clear how such a mixed result should be interpreted in terms of actual goal achievement (Rosencrantz et al. 2007).

Approachability

In goal-setting theory, it is commonly argued that goals ought to be realistic in the sense that it should be possible to at least approach them to a meaningful degree (Edvardsson and Hansson 2005). Locke and Latham (1990, 2002), for instance, provide empirical support for the so-called high performance cycle, that is, the idea that better goal achievement can be reached when the goals are precise and challenging yet not excessively difficult to achieve (see also Latham and Locke 2007). Elvik (1993) followed this line of argument when he stated that road safety management targets should be “challenging, yet in principle achievable” (p. 1116) and that “any quantified target is a compromise between idealism and realism” (p. 579).

Formulating challenging but sufficiently realistic goals is not an easy task, especially not in the public sector wherein goal achievement is dependent on many different factors. This is also the case for road safety management targets. Setting “optimally challenging” road safety targets presupposes that the goal-setter has some knowledge about what social developments affecting transportation and traffic safety can be envisaged during the period between goal-setting and projected goal achievement, what (if any) road safety management measures will likely be introduced during the indicated time period (what measures are politically and economically feasible), and how effective the introduced measures will be in furthering goal achievement (Wesemann et al. 2010; see also Corben et al. 2010).

A common argument against Vision Zero for road traffic safety is that it is an unrealistic goal, because we will never be able to achieve zero fatalities and serious injuries in road traffic (e.g., Long 2012). This argument is analyzed by Abebe et al. 2021, Edvardsson Björnberg, and Hansson (2021, Chap. Arguments Against Vision Zero: A Literature Review of this handbook). As noted there, several counter-arguments could be made against this way of reasoning. It could be argued that, when it comes to saving lives and avoiding serious injuries, no goal above zero should be considered ethically permissible. The Swedish Vision Zero for road safety policy is premised on the ethical assumption that it is unacceptable for people to be killed or seriously injured in the road transport system (Elvebakk 2007). Thus, Belin et al. (2012) argued that, “politically and humanly speaking, it was difficult to stipulate any other long-term goal” (p. 173). Trying to figure out what constitutes an optimal target level in terms of the number of killed and seriously injured people simply does not seem like a morally acceptable approach. Rosencrantz et al. (2007) drew parallels on how goals are set in workplace health and safety. Although few would deny that compromises between costs and workplace safety are unavoidable, government agencies are seldom (if ever) instructed to find out what constitutes an economically optimal level of fatal workplace accidents. Instead, it is assumed that a serious workplace accident should always be regarded as a failure.

Moreover, it is worth noting that many important political goals, such as liberty, social justice, or sustainable development, are highly idealistic in the sense that they concern end states that cannot be achieved once and for all but will have to be fought for indefinitely (Rosencrantz et al. 2007). This has not prevented political leaders and movements from formulating and using them as goals. Tingvall and Haworth (1999) appear to follow this line of reasoning when they write that zero (as in Vision Zero) is not a target to be achieved by a certain date. As noted by Kerr and LePelley (2013), highly ambitious goals, sometimes referred to as “stretch goals,” have also been adopted in business organizations in order to stimulate creativity and “out-of-the-box thinking” among the organization’s employees (see also Sitkin et al. 2011).

Motivity

A goal can be achievement-inducing not only by virtue of its action-guiding capacity; it can also facilitate goal achievement by motivating the agent to work toward the goal. The action-motivating function of goals is central to goal achievement. Merely knowing what has to be done to reach the goal is not sufficient for it to be achieved; the agent(s) responsible for working toward the goal must also want, or be motivated, to do so.

Arguably, inducement to take measures toward goal achievement could come from sources other than the goal itself (Edvardsson and Hansson 2005). For example, government agencies are often bound by legal rules prescribing that certain measures be taken, in which case the agencies and their employees have little choice but to follow the rules. Inducement to take measures that facilitate goal achievement could also arise if the organization and its employees stand to gain financially or in some other way from efforts taken to reach the goal. Moreover, leadership style and the process through which goals are adopted within an organization can have an impact on the employees’ motivation and willingness to work toward set goals (Bronkhorst et al. 2015; see below). Although these are important factors significantly affecting motivation within organizations and among employees, it is possible that such motivation could also be generated from the goal itself.

What makes a goal motivating is, of course, largely due to its content. Vision Zero for road traffic safety is motivating because many people perceive its end state to be desirable or worth striving for. However, psychological studies confirm that the motivating capacity of a goal can also come from other aspects of the goal than its end state. Locke and Latham (1990, 2002), among others, have showed that the motivating capacity of a goal is tightly linked to the other goal criteria discussed above, that is, it largely depends on how well the goal satisfies the criteria of precision, evaluability, and achievability. For instance, some studies have shown that many agents’ goals that are precise and challenging exert a higher degree of motivation (both in terms of intensity and durability) than, for example, do-your-best goals (Locke and Latham 1990, 2002; Wright 2004). Studies have also presented evidence suggesting that goals that are evaluable and evaluated have a positive effect on performance, which is believed to originate from the motivational effect experienced by individuals as they are able to determine where they stand in relation to the goal (Locke and Latham 1990, 2002).

Assuming that a goal’s motivating capacity is a function of how well it satisfies the other rationality criteria, it could be questioned whether motivity should really be considered an independent goal criterion. For a goal to be motivating, it would then suffice to ensure that it is sufficiently precise, evaluable, and possible to achieve to a satisfactory degree. However, some academic scholars have argued that there could be additional dimensions of goal motivity that are not adequately captured by the other criteria. Nutt and Backoff (1997), for example, discuss what makes a corporate or organizational vision motivating. Their argument is also relevant for organizational goals. In the authors’ view, the degree of motivation exerted by a vision partly depends on its articulation. Here, articulation is understood in broader terms than precision. By formulating a vision through expressive images that directly “crystallize in people’s mind what is wanted,” a shared understanding of an organization’s direction can be promoted among its employees (Nutt and Backoff 1997, p. 314).

Vision Zero for road traffic safety (i.e., nobody should be killed or seriously injured on the road) has the advantage of being both precise in the sense discussed in previous sections and formulated through an expressive image, as elaborated by Nutt and Backoff (1997) (See Rosencrantz et al. (2007) for a more extensive analysis of the precision of the Swedish Vision Zero.). The term “zero” is used here to communicate the message that it can never be ethically acceptable that people are killed or seriously injured in the road transport system (Tingvall and Haworth 1999).

To see how the criterion of motivity could encompass additional dimensions than goal precision, consider the following two goals:

  1. 1.

    In 2020, Sweden should have the lowest number of road traffic fatalities and serious injuries within the EU.

  2. 2.

    Sweden should decrease the number of road traffic fatalities and serious injuries to 15% between 2015 and 2020.

Given that adequate accident data are available, both goals could be said to be precise. However, goal 1 might in addition have the advantage of being slightly more motivating, assuming that the competitive feature of the goal (being ranked at the top of the EU nation list) has this effect. Empirical studies are obviously needed to confirm this. However, the example serves to illustrate that there could be more to the criterion of motivity than what can be captured by the criteria of precision, evaluability, and approachability.

Balancing the Criteria

In summary, a goal can be achievement-inducing (rational, functional, or operative) by virtue of being action-guiding or action-motivating, or both. Edvardsson and Hansson (2005) argued that improved satisfaction of each of the four criteria, namely, precision, evaluability, approachability, and motivity, ceteris paribus makes the goal function better in the achievement-inducing sense. However, conflicts between the criteria could occur. The criteria that make a goal action-guiding could, for instance, make it less motivating and vice versa. Edvardsson and Hansson (2005) provided some examples of such conflicts. Although they will not be repeated here, the main message is that the four criteria need to be balanced from case to case in order to optimize the achievement-inducing function of the goal. Factors beyond the goal itself, such as the implementing agent’s background knowledge, will determine the extent to which the four criteria need to be satisfied in each individual case to further goal achievement.

One way of balancing the action-guiding and action-motivating properties of a goal is to adopt goal systems in which some goals primarily motivate action, while others guide action. As an example, in the case of the Swedish Vision Zero, the Swedish Parliament has adopted an overarching, assumingly motivating, vision—“Vision Zero”—that is operationalized through a more precise interim target: “In 2020, no more than 220 people shall die and 4,100 people shall be seriously injured on the Swedish roads.”

Rationality Criteria for Systems of Goals

Public policy goals are seldom adopted in isolation but are parts of a system of goals addressing a certain policy problem or issue. One such goal system is the Swedish system of transport policy objectives. It consists of a general transport policy objective (“To ensure the economic efficiency and long-term sustainability of transport provision for citizens and enterprise throughout Sweden”) that is operationalized into a “functional objective” and an “impact objective,” which, in turn, are specified through time-bound interim targets (Government Offices of Sweden 2016, p. 6) (The functional objective states that “The design, function and utilisation of the transport system are to provide everyone with a basic level of accessibility of good quality and usability and to contribute to the development potential of the entire country. The transport system is to be gender equal, meaning it is to meet the transport needs of women and men in an equivalent manner.” The impact objective states that “The design, function and utilization of the transport system are to be adapted in such a way that no one is killed or seriously injured and further the achievement of the overarching generational goal for the environment and environmental quality objectives and contribute to improving human health.”).

Another example is the UN’s SDGs, which consist of 17 broad goals and 169 targets to be achieved by 2020. For systems of goals, such as the Swedish transport policy objectives and the SDGs, additional rationality criteria to those discussed above may apply. Below, three such criteria will be discussed: completeness, number of targets, and consistency.

Completeness

In public policy goal systems, goals on a higher administrative level are often operationalized or broken down into sets of sub-goals on a lower level. The question then arises: How should the goals and targets in the system be aligned? A commonly defended idea is that the goals and targets should be aligned such that the higher goal is achieved or at least approached to a significant degree if all targets on a lower level are reached. This requires that the targets, taken together, are complete in the sense that they capture the most salient aspects of the overarching goal. If this does not hold, achieving the targets may give the impression that work toward the desired end state is progressing, while the opposite is true in reality (See Tingvall et al. (2010) for a similar discussion of the relationship between road traffic safety performance indicators (SPIs) and the overarching goal of creating a safe road transport system.).

The adequate operationalization of an overarching goal requires that the operationalizing agents possess a good understanding of what causes the policy issue, or problem, and how it can be addressed, among other things. For example, the operationalization of the overall goal of road safety (i.e., to avoid fatalities and serious injuries in road traffic) presupposes that one knows who is involved in those accidents and what factors contribute to causing harm to the people involved. Mononen and Leviäkangas (2016) pointed out that one serious shortcoming in road safety work globally is that, although half of all deaths that occur in road traffic are suffered by vulnerable road users, mainly pedestrians and cyclists, very few road safety goals target those groups. Instead, the adopted goals emphasize in-vehicle safety, among others, which primarily affects driver and passenger safety.

When the overarching goals are qualitative, additional problems may arise in the process of goal operationalization. Qualitative goals, such as “maintaining a flourishing flora and fauna” or “providing a high-class education to everyone,” are often operationalized through precise and more easily evaluable quantitative targets. In those situations, goal displacement may occur. This happens when the goal-setting and implementing agencies lose sight of the overarching goal and instead treat the quantitative targets as if they were the “ultimate” goals (Bohte and Meier 2000). Goal displacement has at least two unfortunate implications. Unless the quantitative targets are complete in the sense that they can be said to cover the most central aspects of the qualitative goal, achievement of the subordinate targets does not represent achievement of the overarching goal. Moreover, focusing on the quantitative aspects of a policy problem could conceal its political nature and make it appear as if the problem is strictly technical in nature (Cortner 2000). Framing a policy problem as predominantly “political” versus “technical” has consequences for whom is expected and allowed to participate in the efforts to solve the problem.

Admittedly, goal displacement may not be a serious problem in road safety policy, as many road safety targets are of a quantitative kind, even when adopted on an overarching level, as in the case of the Swedish Vision Zero. However, the problem of goal displacement could hypothetically occur if qualitative road safety goals are formulated and operationalized through quantitative sub-goals or targets.

The Number of Targets

One way of ensuring that all aspects of the overarching goal are adequately operationalized is to adopt as many targets as possible, with each target representing some aspect of the overarching goal. Thus, the requirement of completeness appears to favor goal systems consisting of a large number of sub-goals or targets. However, such goal systems have also been criticized, thus raising the question of whether there is a limit to how many goals a goal system should contain to function well in an achievement-inducing sense.

The UN’s SDGs are an example of a goal system that has been criticized for being “so sprawling and misconceived that the entire enterprise is being set to fail” (The Economist 2015). One problem with having adopted a large number of goals is that priorities have to be made somewhere along the line. However, in the absence of clear directions for how priorities should be made, there is a risk that the agencies responsible for implementing the goals will focus on less relevant policy aspects than others.

Elvik (2008) appeared to follow this line of reasoning when he argued that one must limit the number of goals in order to create an effective system of road safety management goals. Instead of trying to address all policy problems at once, one should concentrate on a few key areas that have a major impact on safety (see also Smith and Busse 2010). Here, high-quality accident data can play an important part, thereby rendering road safety policy, including the prioritization of road traffic policy goals, to be more evidence-based.

Consistency

When goals are adopted as parts of a goal system, it may be argued that the goals should ideally be consistent in the sense that they do not conflict with one another. The problem with conflicting goals is that efforts to attain one goal make it more difficult to achieve another goal. Thus, the implementing agent will have to prioritize between the goals, which can be resource consuming unless some guidance on how priorities ought to be made has been provided by the goal-setter (e.g., the politicians). Goal conflicts, their formal definition, and the possible ways of addressing them in public policy contexts are issues that have been addressed in past works (Rosencrantz 2008; Nilsson et al. 2016; Carrigan 2018). They will not be discussed at length in this chapter. However, two observations will be briefly presented before proceeding.

First, it is hard to determine the degree of consistency required for a goal system to be rational in the achievement-inducing sense. Complete consistency may come with its own cost. Hansson (1998) warned that attempts to avoid goal conflicts altogether can lead to the adoption of goals with very low ambition levels. Such goals may be easy to co-achieve; however, their achievement will not represent any significant change in the present state of affairs. Put differently, they are not particularly meaningful.

Second, it is worth noting that goal conflicts are addressed in different ways in various policy areas. In some policy areas, compromises and adjustments between (conflicting) goals are made before the goals are set. This is, for instance, common in economic policy. In other policy areas, the goals are set first and compromises are made afterward. This is often the case in health and safety policy. Although traffic and transportation usually follow the first pattern of decision-making, road traffic safety is an exception (Rosencrantz et al. 2007). Road traffic safety work, particularly when discussed in terms of Vision Zero, follows the second pattern. Like public health issues in general, this is based on the ethical premise that it is morally unacceptable that people are killed or seriously injured on the road when mitigating measures can be taken to avoid these.

Why Does a Road Safety Target Have to Be Stable Over Time?

If goals are to fulfil their typical function of regulating action toward goal achievement, they need to have a certain stability. This holds true for both goals set by individuals and goals adopted by organizations. Frequent goal revision makes it difficult for the individual/organization to plan their activities over time. It also makes it more difficult for them to coordinate their actions with other individuals/organizations. Thus, the non-reconsideration of adopted goals appears to be the default position both in private life and in public policy.

However, it is not difficult to think of situations in which, for instance, a government agency has reason to revise one or several of its goals. For example, a road safety target could turn out to be much more (or less) difficult to achieve than initially believed and therefore in need of some modification. To take one example, Wesemann et al. (2010) referred to a set of road safety goals adopted by the Dutch government for which the level of ambition was eventually sharpened as the government found out that an unusually strong decrease in traffic fatalities occurred some years after the goals were adopted.

Baard and Edvardsson Björnberg (2015) identified two types of considerations that could give an organization (or individual) a reason to reconsider its goals: achievability- and desirability-related considerations. On the one hand, the organization could realize that the goal’s level of ambition is too high or too low (example above) and thus decide to adjust the goal accordingly (achievability-related considerations). On the other hand, an organization could realize that the value premises upon which the goal was initially set no longer hold (i.e., the end state is no longer considered valuable or is considered much less valuable than when the goal was adopted).

In relation to road safety targets, it is more likely that achievability-related considerations may come into play and justify adjustments in the targets. It would be more difficult to envision that people’s preferences and values regarding loss of human life and well-being would change significantly, even over longer time periods, although other social developments leading to more severe losses of human life, such as severe pandemics, could obviously lead to politicians prioritizing other policy measures over road traffic safety.

Obviously, new data can justify the adjustments of the adopted road safety targets. However, Elvik (1993) argued that frequent goal revisions on the basis of accident history and traffic forecasts are problematic, because, over time, such a practice can lead to the present number of accidents being adopted as the target (cf. Bevan and Hood 2006, on ratchet and threshold effects). This clearly reduces the value of the goals as policy-making instruments.

Who Should Be Involved in the Goal-Setting Process?

It is sometimes argued that participatory goal-setting is conducive to goal achievement, at least in employer-employee settings. When an employee is involved in the goal-setting process, he/she will not only be more motivated to reach the goal but also have a better idea of what to do in order to reach the goal, or so it is argued (O’Connell et al. 2011). Jung and Ritz (2014, p. 468), for instance, argued that involving employees in the goal-setting process can improve their “affective organizational commitment.” Various psychological mechanisms contribute to this. As noted by the authors, involvement in the goal-setting process sends a signal to the employees that “they are important, worth-while, and valued by their superiors and the organization, which in turn can enhance their self-esteem” (ibid.). Moreover, such involvement can boost the employee’s trust in and identification with the organization’s goals and its values.

However, mixed empirical results have been obtained in relation to the hypothesis that participatory goal-setting leads to better goal achievement. Certainly, some studies confirmed that employees who participate in the goal-setting process generally perform better than their colleagues who are simply assigned to achieve some goals. However, Latham et al. (2008) pointed out that the reason why performance levels were increased in some of those studies was that significantly more ambitious goals were adopted when participatory goal-setting processes were used. Therefore, these studies indicate that it is the level of ambition, rather than the method of goal-setting, that determines how wide the goal-outcome gap will be. To further support their argument, Latham et al. (2008) cited a study by Latham and Steele (1983), which concluded that “when goal difficulty is held constant, there is no motivational benefit to one method of goal setting versus the other provided that the logic or rationale for an assigned goal is given” (Latham et al. 2008, p. 388). The findings suggest that, when goals are assigned, a clearly stated rationale may fulfil a similar function to that allegedly served by participatory goal-setting, namely, to foster trust, legitimacy, and commitment to the goal.

In public policy settings, goals are commonly set by one agent, in many cases a political decision-maker, and implemented by one or several other agents. In road safety policy, those other agents can be government agencies or various road safety professionals (e.g., national road administrations), local councils, business corporations (e.g., vehicle producers), and individuals. It is sometimes argued that greater acceptance of the goals will be fostered by allowing those other actors to participate in the goal-setting process. Thus, Kristianssen et al. (2018) argued that a road safety target that is “formulated from below by actors within the policy area, can enhance the internal legitimacy of the vision” (p. 266).

Corben et al. (2010) provided an example of a participatory bottom-up approach to the adoption of road safety targets. They described how the Western Australian Government and the Road Safety Council of Western Australia proceeded with the development of a new road safety strategy for the period 2008–2020. The WA Road Safety Council decided that the new road safety strategy should be developed “in a consultative and transparent way to maximize stakeholder and community acceptance” (p. 1085 f.). The development of the strategy thus involved a great degree of stakeholder consultation: over 4,000 people participated in the consultation process, which involved three phases. In phase 1, the community’s views on road safety were gathered through a number of community forums. In phase 2, community members were given the opportunity to comment on a recommended package of initiatives developed by Monash University Accident Research Centre (MUARC). This was carried out through community forums and a survey of a representative sample of the population. In phase 3, the endorsed strategy was communicated to the public, the stakeholders, and the Parliament. Using both community forums and survey samples was considered an effective way of gathering the views of those who were specifically interested in road safety issues for some reason (forum attendants) and those who were not (survey). The authors concluded that “to promote community acceptance of the Safe System philosophy and a bold, long term vision for road safety, it has been important to share the research with the community and listen to its views” (Corben et al. 2010, p. 1095).

However, involving the public in the goal-setting and implementation process, as opposed to letting the road safety experts and politicians decide, comes with its own dangers. In the case of Western Australia, analyzed by Corben et al. (2010), the consulted community had divided views on the desirability of speed limit reductions. As a result of public opposition, the WA Road Safety Council did not implement the reductions in speed limit suggested by the MUARC. Instead, a decision was made to focus on demonstration projects to illustrate the effects of speed limit reductions. According to the MUARC’s estimates, the proposed speed limit reduction would have reduced the number of killed or seriously injured people by 1,600 over the 12-year life of the strategy. Therefore, in this case, it could be argued that wider community support and greater legitimacy were gained at the potential cost of 1,600 deaths and serious injuries.

In the context of health targets, Smith and Busse (2010) acknowledged that public consultations can be useful in identifying priorities for improvement. However, they cautioned against uncritically accommodating every interest group in the process of target setting. Regardless of policy area, a wide range of considerations and interests come into play when setting a target, including the respective interests of future users, taxpayers, and users of other services. It is the role of the government to balance those (sometimes) conflicting interests and demands. When inviting stakeholders to participate in the goal-setting process, the government must bear this responsibility in mind and actively discourage potential attempts by some stakeholder groups to “hijack” the agenda in order to further their own interests.

What About Contextual Rationality Aspects?

The extent to which goals can further their own achievement depends not only on how they are formulated or how goals and sub-goals are aligned. Instead, factors that are external to a goal may also have a significant impact on how well that goal will be able to guide and motivate action toward its achievement. A number of such factors can be envisaged, and some of them will be outlined below.

A system for monitoring and evaluating progress: One of the conditions for the successful MBO in road safety policy identified by Elvik (2008) is that there should be a system in place for monitoring progress toward the targets and providing feedback to responsible agents. Thus, effective MBO presupposes not only that the goals themselves are evaluable but also that they are, in fact, evaluated on a regular basis. There must be a system of monitoring and evaluating in force that can provide feedback to those responsible for adopting the goals, and in the case of road safety, these include mostly governments at the national and local levels. Sometimes, feedback systems include incentives and disincentives (awards and punishments) that supplement the motivating effects of the goals themselves. These could be official announcements of the performance output, “naming and shaming” or other reputational measures, award of bonuses, etc. (Bevan and Hood 2006). As noted above, evaluation systems that give feedback in the form of awards and punishments may be vulnerable to gaming.

Clear communication channels: Another prerequisite for effective MBO, as pointed out by Wibeck et al. (2006), is that there should be well-developed communication channels between agents operating at various government levels. Effective communication is needed when both implementing and evaluating adopted goals, not least to avoid misunderstandings and controversies concerning the goal content, time frames, and potential goal conflicts. Arguably, this is especially crucial when the MBO system contains qualitative goals. As an example, Wibeck et al. (2006) discussed the case of the Swedish environmental quality objective “a good built environment,” which is interpreted in different ways by various agents depending on the administrative context and the ideological perspective. Here, Vision Zero has the advantage of being formulated in quantitative terms, which could make communication regarding goal achievement easier than in many other public policy contexts. However, effective communication channels between agents at different societal levels are nevertheless important in road safety work, not the least to facilitate discussions about the effectiveness of various road safety measures.

Strong political commitment: Finally, though perhaps trivially, for road safety goals to be achieved, it is not sufficient that the goals have the capacity to guide and motivate action, that there are well-designed systems in place for the assessment and evaluation of the goals, and that effective communication channels exist between road safety agents operating at different societal levels; beyond these, there must also be a strong political commitment to the goals in the form of sufficient resource allocation and political prioritization (Elvik 2008).

Conclusions

In this chapter, it has been argued that goal-setting can be an effective management technique in road safety policy. Goal-setting theory, developed by psychologists and management theorists over the last 40 years, supports this argument. Empirical evidence from the research field of road safety confirms that the theoretical claims made in the goal-setting literature are also relevant in a road safety policy context: ambitious quantified targets can indeed reduce the number of dead and seriously injured people on the road. However, for this to hold true, a number of requirements must be satisfied. Not only must the goals be formulated in such a way that they can guide and motivate agents to act in ways that are conducive to goal achievement. In addition, goal-setting requires that there are effective evaluation systems and practices in place; that implementation measures, assessment, and evaluation outputs are communicated and discussed among the road safety agents concerned; and that adequate financial means are allocated to those responsible for implementing and evaluating the goals. In addition, governments that wish to organize their road safety work around one or several road safety targets should keep in mind that, in many public policy settings, goals and goal-setting are vulnerable to gaming. Thus, given that gaming corrupts the point of adopting and working toward goals, it is vital to counteract such behaviors at an early stage in the goal-setting process.

Cross-References