Social norms and the enforcement of laws

Daron Acemoglu, Matthew Jackson

19 September 2014



Human societies rely on social norms for coordinating expectations, encouraging some actions and discouraging others. But once in place, norms are durable (Helliwell et al. 2014). They are powerful constraints on social interactions and may conflict with institutions and laws attempting to sanction certain behaviours. The conflict between prevailing norms and new laws often renders such laws ineffective.

A deadly example: Duelling in Europe

The history of duelling in Europe illustrates the power of norms in shaping the enforcement of laws.

  • Duelling was outlawed in France by Louis XIII in 1626; both Louis XIII and Louis XIV vigorously sought to enforce this ban, and went so far as executing officers taking part in duels.
  • But duelling – a key pillar of the social norms of French military officers and aristocrats – remained widespread and these laws went unenforced.

Estimates suggest that over 10,000 duels among French military officers (with over 4,000 deaths) took place during the last 30 years of Louis XIV's reign.

The history of duelling in the 18th and early 19th century in the UK is similar. Despite bans of duelling, several leading British politicians and prime ministers engaged in it. Similarly, in the US, the former Treasury Secretary, Alexander Hamilton, was fatally wounded by the then Vice-President Aaron Burr in a duel; and the future President Andrew Jackson is reported to have taken part in several duels. Even though duelling appears to have declined in the North-eastern US after the early 19th century, the tradition remained strong for a long time in the South.

Laws against duelling were ineffective not just because they went against a deep-rooted norm among officers and citizens, but also because such bans necessitated some of the participants or witnesses to a duel to blow the whistle and inform the authorities – against their own preferences.

Smoking regulations and paying taxes

While laws that conflict with norms are likely to go unenforced, laws that alter behaviour -- either because they do not create too much tension with prevailing norms or because they are enforced with sufficient vigour -- also change norms. An illustrative example comes from smoking regulations, which have gradually banned smoking in public places such as restaurants, movie theatres and bars, and then in shared and private offices. In the process, these regulations have changed social norms.

Many economic decisions also depend on prevailing norms. Many more people evade taxes in societies where such evasion is socially acceptable. For example, 30% of taxes were evaded in Greece in 2011, compared to only 7% in the UK. High rates of tax evasion in Greece appear to be related not only to the fact that evaders view their behaviour as ‘normal’, but also to the lack of whistle-blowing from others against the behaviour that has come to be socially acceptable.

New research

In recent work (Acemoglu and Jackson 2014), we study the interplay of laws and norms. As in our discussion of duelling and tax evasion, whistle-blowing plays an important role in the interplay between norms and laws.

Consider the following stylised example.  

  • Firms make a choice, for example, about the quality of their products or what to report to tax authorities. 
  • Laws ban certain types of behaviour: Product qualities lower than a threshold, or not reporting transactions of more than a given amount for taxes. 
  • After choosing how to run their businesses, each firm matches with another firm as part of a business partnership – for instance with one supplying inputs to another, or collaborating on marketing or distribution, etc.  

Business partners observe each other's behaviour and one can whistle-blow the other about any illegal activity. 

They may do so because of the negative externalities imposed on society by unreliable products, tax evasion, etc. Moreover, a mismatch between two producers in terms of product quality or how much of their business is illicit is problematic for each. For example, both sides have to take part in avoiding some forms of value-added taxes. The costs of mismatches, as well as general externalities, create incentives for businesses to whistle-blow on law-breaking partners. However, note that businesses that are law-breakers have incentives not to whistle-blow, since they are already better matched with other law-breakers, and because they may anticipate being audited and punished themselves if they whistle-blow. Because authorities may not have the resources to audit more than a small fraction of agents to enforce laws, private whistle-blowing can be central to the enforcement of laws. This implies that a society in which laws conflict with social norms will be unable to leverage private enforcement and will have less effective laws.

Our formal analysis characterises equilibria and clarifies the interaction between laws and norms. The private enforcement of laws leads to a novel source of multiplicity. When many agents break the law, there is little whistle-blowing and this encourages further law-breaking, which creates a feedback that can result in multiple equilibria – consistent with the observation that similar laws have very different levels of effectiveness across societies. Despite this multiplicity, the equilibria are relatively simple and have intuitive comparative statics. For instance, tightening a law (banning more behaviours) can lead to lower levels of compliance, while increasing fines has the opposite effect.  Moreover, there is a non-monotonicity as fines are increased. Those who still break the law are those who have preferences for more extreme behaviours.  Then, since law-breakers only get to break the law when they are matched with other law-breakers (as otherwise they are whistle-blown upon), the more extreme preferences among remaining law-breakers feed back to induce those who do break the law to choose even more extreme behaviours than they would when matched with agents who have more moderate preferences.

Key findings

Our important findings include:

  • Laws that are in a strong conflict with existing norms backfire.
  • Abrupt tightening of laws causes significant lawlessness. Gradual imposition of laws that are more in accord with prevailing norms can successfully change behaviour and thus future norms.

The dynamic model also generates ‘social multipliers’ in law-breaking:

If there is high law-breaking, then in the next period there will be less private cooperation with law enforcement, increasing law-breaking further.

We also show how different types of laws interact: Law-breakers in one type of behaviour can then not whistle-blow on another type of behaviour as they may risk being found to be law-breaking themselves. In particular, our analysis shows how badly-designed – excessively tight – laws for one type of behaviour (e.g., small-scale drug crime in inner cities) can make laws against other types of behaviours completely ineffective (e.g., laws against larceny or gangs).

This extension thus provides a potentially new perspective on the debate on the broken windows theory of Kelling and Wilson (1982), which claimed that the high incidence of serious crime in inner cities was a result of permissive attitudes towards small-scale crimes such as vandalism, graffiti, fare-dodging on the subway, or certain types of anti-social behaviour. That theory calls for much stricter enforcement of laws against small-scale crimes, and was the inspiration of the tough policing strategies used in high-crime cities such as New York. Our theory suggests that pervasive law-breaking on such things as drugs or small-scale vandalism can indeed spill into law-breaking in other dimensions. Yet crucially, it sees the problem not in the fact that there is such behaviour in inner cities, but in the presence of laws that criminalise certain fairly common behaviours in these communities, thus criminalising a large fraction of the community. This perspective then suggests that it might be much more effective, according to our theory, to decriminalise some behaviours that have small externalities or costs to society, so that a large fraction of individuals do not automatically become law-breakers and might then have greater willingness to cooperate with law enforcement in other dimensions of behaviour of greater import to society.


Acemoglu, D and M Jackson (2014), “Social norms and the enforcement of laws”, NBER Working Paper No. 20369.

Helliwell, J, S Wang and J Xu (2014), “New evidence on the durability of social norms”,, 12 March

Wilson J Q and G L Kelling (1982), “Broken Windows”, The Atlantic, 1 March.



Topics:  Frontiers of economic research

Tags:  social norms, whistle-blowing, law enforcement

Professor of Applied Economics, MIT

William D. Eberle Professor of Economics, Stanford University