Behavioural Insights for Public Integrity

Chapter 2. The dynamics of moral decision making

Integrity policies are often based on the understanding that people will exploit an opportunity for misconduct whenever its profits are worth risking potential negative consequence. In other words, individuals are assumed to weigh the probability of getting caught (i.e. the strictness of internal and external control and detection mechanisms) and the ensuing sanctions against the undue gain they could obtain through action or inaction. The policy implications of such a view involve a high level of monitoring and enforcement. However, in everyday life, morality is often perceived in terms of character traits. When confronted with an integrity-related choice, it is assumed that a person will consult his or her own personal values and act in accordance with them. Indeed, corrupt behaviour is often interpreted as an extension of the “bad character” of those involved.

Behavioural research challenges this strictly rational cost-benefit analysis as well as the limited understanding of morality as a character trait. Instead, it paints a more nuanced picture of the dynamics behind moral choices. In a well-functioning integrity system, individuals use discretionary power wisely, judge fairly, lead by example and speak up against misconduct. Reality, of course, is more challenging. How can integrity policies encourage and support such moral choices? The following sections explore the dynamics of moral decision making and examine how policies can support individuals to act with integrity.

2.1. Moral choices

Who do policy makers have in mind when designing integrity policies? Put another way, who acts unethically? The answer to this question is central to the targeting of integrity policies. A key behavioural insight is that unethical choices are often made by people who in general want to behave ethically. However, many integrity policies do not take into account the possibility that the moral compass of these individuals may fail. Some target criminal-minded actors with a view to deterring unethical choices through strict sanctions. Others rely fully on the judgement and actions of trustworthy individuals. Behavioural insights make the case for integrity policies to be based on an understanding that even well-meaning decision makers sometimes fail to make the best ethical choices, and could work to prevent these ethical shortcomings.

Understanding who commits misconduct in which kind of situation often determines the type of prevention employed. However, misconduct is not always a deliberate decision or an action based on ill intent. Rather, moral choice is a dynamic process. Evidence shows that individuals balance their moral choices by continuously comparing their current moral self-perception with their own moral reference point. The moral reference point represents the level of integrity individuals perceive as morally acceptable for themselves. If they find that their own action deviates too much from their personal reference point, they counteract (see Box ‎2.1). Some people tend to have relatively high moral reference points, while others might allow themselves more selfish acts.

Box ‎2.1. Moral balancing

“Bad” actions often produce a desire on the part of individuals to perform an act of altruism or cleansing. Conversely, doing something “good” may produce a feeling of entitlement to cheating or egoistic behaviour. This process of moral balancing is often unconscious. For example, an individual might give a donation to charity to relieve a feeling of guilt without being fully aware of this process (Sachdeva, Iliev and Medin, 2009_[6]). Likewise, individuals that have acted in a particularly altruistic manner may feel licensed to be a little selfish.

 

Moral balancing is not limited to a specific context. The guilt acquired in one situation can make a person act in a moral fashion in a completely different one. For instance, an individual might give a larger tip at dinner because they feel guilty about stealing a colleague’s dessert from the fridge at lunch.

Furthermore, moral balancing is subjective. Irrespective of the actual moral consequences of their actions, people might feel more moral due to the weight of an experience. For example, enduring pain or being treated unfairly can make people feel licensed to be egoistic (Houser, Vetter and Winter, 2012_[7]; Ploner and Regner, 2013_[8]). “I deserve this” is a typical justification for corrupt behaviour (see also section ‎2.4).

Source: (Carlsmith and Gross, 1969_[9]); (Gneezy, Imas and Madarász, 2014_[10]); (Kalanthroff, Aslan and Dar, 2017_[11]); (Sachdeva, Iliev and Medin, 2009_[6]); (Zhong and Liljenquist, 2006_[12])_.

The following key conclusions can be drawn from behavioural research:

• The majority of people follow a moral compass, which orients them even when no one is watching. However, they also tend to give themselves some leeway and may wander away from the moral path when given an opportunity (Mazar, Amir and Ariely, 2008_[13]). This applies in particular when the immediate benefits are appealing and the costs are remote, unlikely or in the future (Gino et al., 2011_[14]).

• Nobody likes to lose out. People make choices in their own favour both consciously and unconsciously and seek advantage for themselves and those similar to them.

• People do not like to be confronted with their own unethical behaviour. Instead, they may apply justifications to make the dissonance between their moral standards and their actual behaviour appear less grave (Barkan, Ayal and Ariely, 2015_[15]). When they do acknowledge the dissonance, they often feel bad about their behaviour, resulting in a desire to compensate (see Box ‎2.1).

The silver lining for policy makers promoting integrity is that most people want to be ethical. The opposite – exhibiting no concern for the impact of one’s actions on others – is a symptom of psychopathy (Hare, 2006_[16]). While it remains unclear to what extent “successful psychopaths” with these attributes are present in society (Boddy, Ladyshewsky and Galvin, 2010_[17]); (Stevens, Deuling and Armenakis, 2012_[18]), they do not constitute the majority. The average person has an intrinsic motivation for honesty and cares about others (Aquino and Reed, 2002_[19]). The strength of this intrinsic motivation depends on the individual’s moral identity, prevailing social norms and other outside factors, and varies between people and across situations and circumstances. Integrity policies should therefore target the majority and be designed to support the dynamics that encourage ethical choices. For example, Box ‎2.2 presents a framework used within the Australian integrity system to effectively target integrity policies to the majority of individuals who do not exhibit criminal tendencies.

Box ‎2.2. Australia’s values alignment model

The Australian values alignment model is based on the supposition that not everyone is the same, and that not all individuals in an organisation share the same values. The following three groups are assumed to exist and represent different relative shares among the employees of any organisation:

• Group A consists of people unlikely to act in a corrupt manner regardless of circumstances, perhaps as a result of a strong moral identity.

• Group B comprises people whose decision to act in a corrupt manner is dependent on the circumstances. In ideal conditions, this group is unlikely to act in a corrupt manner. However, the opposite may be true if personal or environmental circumstances are conducive.

• Group C consists of a small group of people who tend to act in a corrupt manner whenever they can avoid the consequences. This group is driven by self-interest and tends to respond only to effective deterrence.

An effective integrity framework accepts the existence of each of these categories of people and is accordingly designed to:

• provide a work environment for staff in which high professional standards are valued, opportunities for corrupt conduct are minimised and compliance with integrity measures is facilitated (providing Group A and B with ideal conditions);

• prepare for the existence of purely self-interested individuals (Group C) by putting in place effective detection and deterrence measures.

The goal of the integrity framework is to obtain an appropriate balance of measures by encouraging integrity without losing sight of the risks.

Source: Based on information provided by the Australian Commission for Law Enforcement Integrity, www.aclei.gov.au/corruption-prevention/key-concepts/values-alignment.

Essentially, integrity policies can impact the internal dynamics of moral choice making and encourage individuals to follow their moral compass by working to influence the moral reference point of an individual in two ways:

• Raising the moral reference point. Raising the moral reference point induces people to compare their own actions against higher internal standards. Clearly defining ethical expectations and emphasising the relevance of integrity and the trust placed in public officials can help to adjust peoples’ moral reference points upwards. Discouraging information, surveillance and distrust can, in turn, lower the moral reference point against which a person assesses his or her own behaviour.

• Emphasising the moral reference point. Even individuals with very high internal moral standards sometimes fail to follow them. If this happens, a dissonance arises between a (considered) behaviour and the moral reference point. However, people often ignore such dissonances. Addressing people’s personal morality and encouraging them to reflect on the ethical consequences of their actions can lead to more ethical choices.

2.2. Profiting from corruption can be a temptation

Staying true to one’s own internal moral reference point is not always easy. Acting against this moral compass is often prompted by temptation – a behaviour engaged in for immediate benefit that probably does not serve long-term goals. An integrity breach sacrifices future welfare for immediate gratification (Bazerman and Tenbrunsel, 2011_[20]). Actions that compromise integrity such as keeping undue travel expenses to buy a new watch, might not result in guilty feelings immediately – until the next morning when colleagues admire the watch. Integrity, in these situations, can be understood as a matter of self-control.

Time pressures and tiredness reduce self‑control and have also been found to reduce ethical behaviour (Kouchaki and Smith, 2014_[21]); (Shalvi, Eldar and Bereby-Meyer, 2012_[22]). Similarly, intense workloads, performance pressure and short-term contracts can undermine integrity. Such conditions give public employees reasons to cut corners and circumvent rules, and may also provide them with justifications for their misconduct: “If my employers don’t care about me, why should I care about them?”

In the face of an immediate temptation, people often act in a short-sighted manner and exhibit a lack of willpower. Behavioural approaches therefore try to help individuals incorporate long-term benefits into their decision making. For instance, people are psychologically better equipped to withstand a temptation when they know it is coming. Research has shown that this is also true for ethical temptations – anticipating them helps to resist them (Sheldon and Fishbach, 2015_[23]). From a policy perspective, this implies an emphasis on identifying typical ethical dilemmas (e.g. in Codes of Conducts or integrity trainings). If individuals have the opportunity to consider a dilemma beforehand from a sober perspective, they are more likely to act with integrity when confronted with it directly. A similar argument can be made for the systematic identification and disclosure of potential conflicts of interest. The acknowledgement of a conflict of interest and the anticipation of a potential moral temptation reduce the risk of misconduct.

Instance of corruption may also be avoided by creating a commitment to integrity beforehand. Integrity policies could provide opportunities for people to commit to certain behaviour and refrain from others. The binding effect of such commitments will be even stronger if they are made publicly. General commitments to integrity and against corruption are usually less effective than concrete actionable statements, such as “I will not use these funds for any other than the dedicated purpose”. Such actionable statements leave smaller moral wiggle room in the moment of decision making. For example, in the Philippines, vote-buying was reduced significantly after people made a promise not to sell their vote in the upcoming elections (Hicken et al., 2015_[24]).

Finally, so-called moral reminders have also been shown to be an effective tool to counteract unethical behaviour. Ethical choices can be encouraged by reminding people of moral norms in the moment of decision making (Bursztyn et al., 2016_[25]); (Mazar and Ariely, 2006_[26]). Inconspicuous messages such as “thank you for your honesty” can have a striking impact on compliance (Pruckner and Sausgruber, 2013_[27]). Such interventions have the most impact when timed carefully to just before the moment of decision making (Gino and Mogilner, 2014_[28]). Making the moral norm and self-perception salient in the decision-making moment helps morality, the long-term goal, outweigh acute temptation (see Box ‎2.3). However, the effect of a moral reminder may be reduced if an individual is exposed to it on a frequent basis, and at some point the message may no longer be noticed. Nonetheless, when policy makers place such messages in the proximity of decisions with integrity risks – which are not taken frequently by the same person – they can make a significant difference. Processes and procedures could therefore be assessed for the possibility of inserting timely moral reminders.

Box ‎2.3. Moral reminders

Small message, strong effect

One straightforward strategy to induce ethical behaviour is to remind decision makers of moral standards. Research has shown that inconspicuous messages can have a striking impact. In Austria, it is common practice to buy newspapers from a self-service box in public places. The papers have a fixed price, but there is no control of how much money customers actually place in the box when they take a paper. Researchers from the Universities of Linz and Vienna conducted a field experiment on these paper sale boxes. On some of the signs indicating the price of the paper they added, “Thank you for being honest”. The amount paid for a paper increased significantly in these cases. Other studies have shown similar effects. For example, students cheated less when reminded of a moral norm, such as the Ten Commandments, just prior to taking a test.

In Mexico, the Secretaría de la Function Pública (SFP) in co-operation with the research centre CIDE (Centro de Investigación y Docencia Económicas) applied this behavioural insight to their gift registration policy, in order to enhance compliance. SFP sent out reminder emails to public employees required to register their received gifts. They randomly varied the text of the message. Five different types of reminder messages were sent:

• Legal: It is your legal obligation to register received gifts.

• Honesty: We recognise your honesty as a public official. You are required to register gifts. Show your honesty.

• Impartiality: Receiving gifts can compromise your impartiality. When you receive a gift, register it.

• Social: More than 1 000 registrations per year are made by your colleagues. Do the same!

• Sanction: If you receive a gift and you do not inform us, someone else might. Don’t get yourself punished. Register your gifts.

The study then observed the number of gifts registered around the Christmas period (peak season for gifts), and compared this with previous years and against a control group who did not receive any of the messages. The study demonstrated that receiving a reminder email increased the number of gifts registered. However, some messages were more effective than others: reminding public officials of their legal obligations and appealing to their impartiality and honesty encouraged more people to register gifts than referring to sanctions or registrations made by colleagues.

This example shows that: (i) small behavioural nudges can increase compliance with an existing policy, and (ii) appealing to values and integrity changes behaviour more effectively than threatening sanctions.

Source: (Fellner, Sausgruber and Traxler, 2013_[29]); (Pruckner and Sausgruber, 2013_[27]); Santiago Steta (SFP), presentation at the meeting of Senior Public Integrity Officials (SPIO) in Paris on 26 November 2017.

2.3. Objectivity is an illusion

Objectivity and consistency in judgement are expected from public officials; unfortunately, they cannot always be expected from human decision makers. Both are often lacking in moral choices, in part due to the dynamics of moral choice making. Literature in this field highlights the ways in which human judgement is misled by unconscious bias. For instance, structures within an integrity system and throughout the public sector should secure the independence of regulatory judgement (OECD, 2017_[30]; OECD, 2017_[31]). Biases and the resulting discrimination most affect those decisions that rely heavily on human judgment. Indeed, the success of some integrity policies depends heavily on objective judgement, for example in areas such as conflict-of-interest management, auditing, and internal control or recruiting. The design of such policies therefore needs to take into account the biases in human decision making.

A famous experiment by (Loewenstein et al., 1993_[32]) illustrates biased judgement. All participants were given the same detailed information about a car accident in which a car had hit a motorcyclist. Some were assigned the role of plaintiff (on behalf of the motorcyclist), while others acted as the defendants of the car driver. The participants were then asked how they would expect a judge to rule in this case. If their answer came close to the correct ruling they could earn money. Despite the incentive to be objective, the two groups provided significantly different interpretations of a fair settlement. Although the participants had no real interest in the case, simply adopting the hypothetical position of either defendant or plaintiff biased their judgement.

Self-serving bias can lead to favouritism not only towards oneself, but also with respect to a group. People tend to assess members of their own group more forgivingly. In an experiment, British and US American participants were presented a media article on torture and later asked what they thought. The participants were found to be more likely to judge the torture as justified when the article stated that it had been carried out by their own security services, rather than those of another country (Tarrant et al., 2012_[33]). Additionally, psychological closeness can lead to lenient judgements. Some people feel closer than others, either due to direct personal knowledge or because they seem similar in some respects. Judgement therefore tends to be biased when making decisions concerning these people (Bazerman and Tenbrunsel, 2011_[20]); (Chugh, Bazerman and Banaji, 2005_[34]). Holding someone in high regard but meanwhile recognising an unethical behaviour of this person creates a cognitive dissonance – a condition that is psychological burdensome. The mind tends to immediately seek justifications to harmonise the conflicting observations. As a result, people are less likely to notice unethical behaviour that contradicts their beliefs about a person or group they are sympathetic towards.

In light of this research, policy makers should be aware that bias can lead to conflicts of interest that might not be obvious at first sight. Aspects that are usually desirable for public officials, such as a strong identification with the institution, high trust and close bonds between team members, can present an integrity risk in some situations. While it is rarely feasible to isolate individual decision makers from such potential invisible conflicts of interest, risk mapping could recognise unconsciously biased decision making as an integrity risk.

Because self-serving bias is not deliberate it is not effectively addressed through harsh punishments or continuous control. However, the way in which choices are presented can prevent biased decisions. Integrity policies could ensure that at-risk decision makers are able to distance themselves from potentially biasing factors. In some settings, blind judgment is an option. When participants in the plaintiff-defendant experiment read the information before they were assigned their role as plaintiffs or defendants, they came to more convergent assessments of the case (Babcock et al., 1995_[35]). The same could apply, for example, to procurement officers reviewing an offer without knowledge of the identity of the provider.

Requiring argumentation both for and against a choice before making a final decision can guide individuals towards better judgement. Asking subjects to list the weaknesses in their own argument has been shown to significantly reduce self-serving bias (Babcock, Wang and Loewenstein, 1996_[36]). Where relevant, two individuals could be asked to argue for one option each and then bargain over the best choice.

2.4. Justifications open up moral wiggle room

To avoid revisiting their moral self-image, people often seek justifications for their behaviour. Justifications overcome the gap between one’s moral ideal and one’s actual behaviour. They function as explanations or re-interpretations of a situation that help people re-align their moral reference points with their moral self-perceptions.

The literature provides various examples of typical justification mechanisms:1

• Linguistic euphemisms: Using benign labels for corrupt acts lowers the burden of ethical responsibility. For example, “cookie jar reserves” refer to the use of surpluses from profitable years to improve the balance sheet during leaner years (Economist, 2010_[37]).

• Metaphor of the ledger: People re‑interpret benefits from an integrity breach (e.g. embezzlement) as a well‑deserved reward for their good efforts, in particular when they feel under‑appreciated.

• Denial of victim, injury or responsibility: People engaging in unethical acts do not acknowledge the harm done or do not see it as a direct result of their own action. This enables them to avoid experiencing the full moral responsibility.

• Legality/Rule-bending/Loop-holing: When an act is unethical but legal or allowed, the responsibility is shifted to those making the law or rules. Conversely, a rule that seems unfair might be circumvented with less moral compunction.

• Cash substitution: Stealing or receiving non-monetary goods (e.g. taking resources from the work place) can seem less unethical than stealing or receiving money as a bribe or cheating for a monetary amount.

• Reframing an ethical decision as an economic one: This mechanism shifts the focus away from the moral damage to the measurable benefits.

• “Others do it”: The behaviour of a direct peer is automatically perceived as more acceptable. Use of this mechanism reinforces misconduct within a group.

• “Everybody does it”: When a certain integrity breach is tempting, people convince themselves that many others engage in this behaviour, which makes it more acceptable for them to do so.

• Delegation: Superiors delegate “dirty tasks” or ask for results that are not feasible without breaching integrity, thereby relinquishing personal responsibility for the actions of their subordinates. The subordinates “do as they are told”, thereby neglecting personal responsibility for the moral consequences of their behaviour.

All justification mechanisms ease the burden of immoral behaviour for decision makers, allowing them to make unethical choices in front of others, while avoiding personal feelings of guilt. They create moral “wiggle room” (Dana, Weber and Kuang, 2007_[38]) –situations in which objective moral assessment can be avoided. Evidence suggests that dishonest behaviour depends on the extent to which such justifications are (Dana, Weber and Kuang, 2007_[38]); (Pillutla and Murnighan, 1995_[39]) and (Shalvi et al., 2011_[40]). Individuals will only cheat if there is sufficient moral wiggle room to justify their behaviour to themselves and others.

This finding has a key implication for integrity policy makers: reducing moral “wiggle room” lessens the tendency to take immoral decisions. Public sector employers can contribute in this regard by defining clear and feasible expectations of proper conduct for their employers. Tasks and incentives should be aligned with the organisational goals and not contradict or undermine integrity. Employees who are well aware of the purpose of their tasks and the consequences of their actions will also have less moral “wiggle room” to justify unethical behaviour.

2.5. Corruption often feels like a victimless crime

Experimental evidence shows that the perceived (im)morality of an action decreases with the distance between the action and the harm caused by that action. Indeed, indirect agency (e.g. when an unethical task is delegated) or uncertainty regarding whether one’s own actions are pivotal for the outcome reduces the sense of moral responsibility (Falk and Szech, 2013_[41]); (Hamman, Loewenstein and Weber, 2010_[42]).

Corrupt transactions often involve intermediaries, clandestine causalities and invisible damages. The harm done by corruption is often far removed from the corrupt actor. This makes it easier to deny responsibility and reduces the moral burden of engaging in corruption. In addition, guilt becomes smaller when shared. When the responsibility for an immoral decision is borne by several actors, the actions of one individual seem less pivotal. Common reflections in such situations include “My action doesn’t make a difference” or “If I don’t do it, someone else will”.

Accordingly, a central consideration in the design of integrity policies could be to reduce the distance between action and harm. The design of organisational structures, communications, stakeholder engagement and human resource management in the public sector can evoke a sense of responsibility for integrity. For example, the distance between action and harm can be reduced by speaking about integrity as a personal matter rather than in abstract terms. The clearest assignment of responsibilities in the case of many integrity policies would be to hold a single actor responsible for choices. A second person can play a controlling function in decision making, but should not share the moral responsibility. Communications focusing on outcomes and dialogue with the affected citizens can also increase a sense of pivotality among public officials – that is, the perception that their choices make a difference.

2.6. Control bears a hidden cost

The introduction of a control mechanism is a signal of distrust. Employers who rigidly monitor the actions of their staff are sending a signal to their employees that they expect misconduct. In such situations, the risks of detection might deter those with bad intentions from engaging in misconduct. However, the evidence highlights the adverse impact on intrinsic motivations for integrity (Schulze and Frank, 2003_[43]): emphasising controls and sanctions may give or reinforce the impression that misconduct is common and thus more acceptable. Box ‎2.4 describes an experimental set-up that showed the deterrent effects of control.

Box ‎2.4. How control can destroy intrinsic motivation for honesty

Schulze and Frank (2003_[43]) undertook an experiment to investigate whether the introduction of control reduces honesty. The experiment was carried out in the context of a movie club at a university. Just before the movie was shown, the following scenario was presented to the students:

A 200 Deutsche Mark (DM) bill belonging to the club fell into a pipe and cannot be removed without the help of a plumber. The student commissioned by the club must select the best offer from among 10 plumbing companies. The offers consist of: (i) the price the movie club will have to pay, and (ii) a bribe for the student in exchange for choosing this company.

After the film, a sheet was selected at random and the chosen student anonymously received a payment in accordance with the choice he or she had made.

The researchers organised two versions of the experiment – one with controls and one without. When controls were in place there was a chance of detection, which increased with the amount of the bribe. If the student was not detected, he or she would receive the payment according to the choice made. If the student was detected, the cheapest company won the contract and the student would receive nothing. Table ‎2.1 provides an overview of the situation.

Table ‎2.1. Experimental scenario

Company	Firms’ bids of price and bribe		Probability that the student is detected	Expected value of the bribe (probability of detection x bribe)
	Price which the movie club has to pay (DM)	Amount which the student could receive (bribe in DM)
A1	20	0	0	0
A2	40	16	0	16
A3	60	32	1/6	26.7
A4	80	48	1/6	40
A5	100	64	1/3	42.7
A6	120	80	1/3	53.3
A7	140	96	1/2	48
A8	160	112	1/2	56
A9	180	128	2/3	42.7
A10	200	144	2/3	48

The results were astonishing. The average amount of bribes was somewhat higher in the risk version then in the non-risk version of the experiment. On average, students behaved more corruptly with a risk of detection than without. In the non-risk version, 9.4% of the students were completely honest, choosing the most efficient company without asking for a bribe in return, compared to only 0.9% in the risk treatment. Control had reduced honesty. In other words, the intrinsic motivation for honesty had been crowded out. This result clearly contradicts any simple recipe for controlling corruption through stricter law enforcement. However, deterrence works too. The risk of detection had reduced the percentage of particularly ruthless students who selected the maximum bribe. This share fell from 28.8% in the non-risk treatment to 12.6% in the risk treatment (Table ‎2.2).

Table ‎2.2. Main results of the experiment

	No bribe asked (Alternative A1)	Highest bribe asked (Alternative A10)	Average bribe
risk treatment	9.40%	12.60%	91.4
non-risk treatment	0.90%	28.80%	87.9

In conclusion, the experiment showed that the addition of controls destroys the intrinsic motivation for honesty, but also acts as a deterrent to corruption.

Source: Schulze and Frank (2003_[43]).

Consideration of detection risks in an ethical decision implicitly turns an ethical question into a mere cost-benefit consideration. If the expected benefit from the unethical action outweighs the expected cost, taking into account the probability of detection and the sanction, then the individual will likely choose this option. Experimental evidence suggests that such a reframing (see section ‎2.4) shifts people’s focus away from the moral implications, making corruption more likely.

Furthermore, control has been shown to significantly reduce the efforts of the person being controlled (Falk and Kosfeld, 2006_[44]). It forces people to provide only the minimum effort necessary to pass the control, but removes the element of positive reciprocity: employees subject to controls might feel less obliged to remain faithful, even when they are being treated well overall (Lambsdorff, 2015_[45]).

Lastly, while limiting discretion in decision making is a standard recipe of traditional anti-corruption measures, limiting a person’s scope of action through controls and reporting regulations might be counterproductive. Public officials with strictly reduced discretionary power, for example, have less freedom to draw on their expertise and experience to decide in the public interest.

However, the fact that controls have an associated cost does not mean that they should be removed. The detection of misconduct and the enforcement of legal norms are essential for building and maintaining trust in integrity. The policies that could be reconsidered in the light of this cost are those that induce a strong burden of distrust while offering a limited chance of detection. When compliance with a rule is generally high and a full control is less feasible, trust might be a better choice than control. Vicious control of a trust-based rule might even drive people to clandestinely disregard the rule, which creates an entry point for severe misconduct. For example, sanctioning an employee for accepting a box of chocolates, rather than trusting that the gift will not unduly influence them, might create resentment. If employees feel distrusted, they might begin to secretly accept small gifts, knowing that such policies are difficult to monitor, and thereby develop a tolerance of non-compliance with the gift policy.

2.7. Precepts for strengthening integrity in moral choices

Individuals make all kinds of unethical choices and in many cases may not even be aware of doing so. Consideration of the main dynamics impacting moral choices allows public policies to be rethought in a way that supports moral decision making and strengthens integrity. The ways in which behavioural insights on moral choices can enrich integrity policies are summarised in Table ‎2.3.

Table ‎2.3. Precepts for strengthening moral choices in integrity systems

Precepts for strengthening moral choices in integrity systems	See section
a) Expect ethical failures from anyone b) Raise moral reference points c) Emphasise moral reference points	2.1
d) Provide actionable training and commitment e) Encourage ethical reflection (e.g. through moral reminders)	2.2
f) Anticipate biased assessment	2.3
g) Counteract justifications	2.4
h) Invoke individual responsibility i) Reduce the distance between action and harm	2.5
j) Avoid framing ethical questions as matters of control k) Avoid communicating distrust	2.7

“We are not as ethical as we think we are” is one conclusion from research into behavioural ethics. Policies that aim to strengthen moral choice making could therefore target the majority of public officials who are motivated to act ethically, yet sometimes go astray. The fact that moral choice is dynamic and can be impacted by external factors also means that policies can intervene.

According to the theory of moral balancing, policies can affect moral choice making in two ways – by emphasising moral reference points and raising moral reference points. While the former involves building long-term norms and strengthening integrity values, the latter is concerned with on-the-spot advice and reminders. ‎Chapter 4. takes a closer look at how both approaches can be integrated into an integrity system in a behaviourally informed way.

Why do people deviate from their own moral standards? Integrity can be seen as a matter of self-control. As such it can be strengthened by creating pre-commitment and mentally preparing people for ethical temptations. Furthermore, there is strong evidence that moral reminders affect ethical choices. A small message can be sufficient to induce ethical reflection – a quick mental check of an intended action against one’s own moral reference point. Moments of ethical reflection can be integrated in many policies.

Sometimes, people are not even aware that their behaviour diverts from ethical standards. Typical justifications and biased judgment can hide integrity breaches in personal choices or the local environment. Behaviourally informed policy design can help erect lines of defence against such conduct.

Lastly, the research in behavioural ethics shows that overly strict controls can have adverse effects. A careful approach to control policies that avoids communicating prevent the discouraging effects of control without abandoning the use of strict and strong measures to enforce integrity.

Note

← 1.  (Ashforth and Anand, 2003_[119]); (Barkan, Ayal and Ariely, 2015_[15]); (Bazerman and Tenbrunsel, 2011_[20]); (Bicchieri and Xiao, 2009_[120]); (Erat, 2013_[118]); (Falk and Szech, 2013_[41]); (Hamman, Loewenstein and Weber, 2010_[42]); (Kouchaki et al., 2013_[114]); (Lambsdorff, 2012_[55]); (Paharia et al., 2009_[117]); (Pittarello et al., 2015_[116]); (Tenbrunsel, 1998_[115]).