VoxEU Column Economic history Frontiers of economic research

The impact of the first professional police forces on crime

Determining whether increased policing reduces crime is a difficult task, in part because policing affects crime and crime also affects policing. This column approaches the problem historically, asking whether the introduction of the first professional police forces in the 19th century, from the London Metropolitan Police onwards, reduced crime. It finds that when police forces were sufficiently large and well-regulated, they had a crime-reducing effect that, while not immediate, did persist over time.

“The test of police efficiency is the absence of crime and disorder, not the visible evidence of police action in dealing with them.” What sounds current was in fact written in 1829, one of Sir Robert Peel’s Principles of Law Enforcement. Though professional police forces tasked with crime prevention are a fundamental component of today’s criminal justice systems, Peel’s ideas introduced a profound change to the nature of policing in the early 19th century. The 1829 introduction of the London Metropolitan Police (the ‘Met’) created the first ever professional police force tasked with deterring crime. Subsequent police forces, across the counties and cities of England and Wales but also in the US and around the world, were modelled after this innovative institution.

What do we know about the effectiveness of police in reducing crime almost 200 years later? Studying this relationship is difficult due to its simultaneous nature: policing affects crime and crime affects policing. To overcome identification problems, the contemporaneous literature often studies (mostly temporary) expansions to the size of an existing police force. Crime reductions were found in London (Draca et al. 2011) and Buenos Aires (Di Tella and Schargrosky 2004) following increased police deployment after terror attacks in both cities. Other studies have focused on spatial instead of temporal variation, originating from geographic boundaries, to evaluate the permanent effects of private policing (MacDonald et al. 2016, Heaton et al. 2016). Chalfin and McCrary (2017) find a general “consensus that increases in police manpower reduce crime”.

In our recent research (Bindler and Hjalmarsson 2019), we take this literature one step further to study the very beginnings of the relationship between police and crime. Did the introduction of professional and institutionalised police forces, first the London Metropolitan Police and later the county police forces in England and Wales, reduce crime? In contrast to the existing literature, studying the early days of policing allows us to evaluate the effect of the first (permanent) officers rather than the last (temporary) officers.

History of policing in London and the counties of England and Wales

The idea of policing existed before the formal institution. Informal policing through local watchmen dates back to the mid-1700s, and the first more organised forms of policing started with the Bow Street Runners in 1750 (Emsley 2009). By the end of the 1700s, the Bow Street Runners were essentially full-time policemen located at the Bow Street House, where they collected reports of crime incidents. In 1792, seven additional offices, modelled after the Bow Street Runners, were introduced in central London. These offices were staffed with three magistrates and up to 12 constables who primarily processed criminal cases (magistrates) and followed up on crime reports (constables).

The introduction of the Met on 29 September 1829 (with a catchment area of a seven-mile radius from Charing Cross, excluding the City of London) brought two fundamental changes to policing in London: 

  • First, a sharp increase in the sheer number of police (more than 20-fold compared to pre-Met times). Upon creation, 1,000 men were immediately hired and the Met grew quickly to 3,000 men by May 1830 (see Figure 1). 
  • Second, the primary task of the new police force was the deterrence of crime, in contrast to the reactionary tasks of previous (non-institutionalised) police. 

These changes were accompanied by strictly enforced behavioural guidelines (e.g. with respect to alcohol consumption on duty or violent policing). 

Figure 1 Weekly hires to the Metropolitan Police, 1830-1856

Figure 2 The roll-out of police forces for English and Welsh counties


Following the deemed success of the Met, police forces were rolled out to the rural counties of England and Wales following the County Police Act of 1839. The adoption of police forces was optional in 1839 but became mandatory in 1856 (see Figure 2 for an illustration of the roll-out). The 1856 Act further introduced ‘efficiency’ criteria for the new forces. To apply these criteria (including sufficiently large forces relative to the population), national inspectors were appointed to annually inspect and certify the efficiency of each county force.

Measuring historical crime

To evaluate the impact of these new professional police forces on crime, we make use of a number of historical sources. First, we exploit the rich data from the Old Bailey Proceedings Online – the digitised versions of the session reports from the Central Criminal Court of London and Middlesex – and geocode crime incidents for all burglary, homicide, and robbery trials between 1821 and 1837 (Hitchcock et al. 2013). Focusing on the most serious crimes allows us to abstract from changes in reporting behaviour (assuming that these crimes would have been reported either way), while geocoding the incidents enables us to allocate crimes to a treatment and control area (within and outside the catchment area of the Met, respectively).

Second, we digitised daily crime reports from the pre-existing police stations/magistrates. The original files (Report or Account of the Proceedings of the several Police Offices) come from the National Archives and are available for January to April of 1828, 1830, 1831, and 1832. The main advantage of these reports is that they contain both cleared crimes (charges with a known suspect) and uncleared crimes (incidents with an unknown offender), as well as all types of crime including misdemeanours.

Third, for the counties of England and Wales, we coded the year of police-force formation and its initial size from a book of the Police History Society (Stallion and Wall 1999). To consistently measure crime for all counties, we digitised the number of charges by year, county, and offense category reported in the Judicial Statistics (1832-1865). 

Evaluating the impact of the new police on crime 

Our analysis of the London data begins with a difference-in-difference design using the Old Bailey data, which compares the number of crimes in the treatment (i.e. catchment) areas with the number of crimes in the control areas before and after the introduction of the Met, respectively. We find the following results:

  • The number of robbery trials decreased by 40% after the introduction of the Met 
  • This decrease is statistically significant and persistent over time 
  • There were no significant changes in the number of burglary and homicide trials.

The second London analysis is based on the daily reports data and emphasises the difference between cleared and uncleared crimes. In the absence of a control group (as all offices were located within the Met’s jurisdiction), we estimate the total effect on crime in a pre-post design, zooming in to observe a small window around the reform:

  • Again, there was a significant reduction in violent crime – both cleared and uncleared 
  • There is a significant decrease in uncleared property crime (incidents)
  • But at the same time a significant increase in cleared property crime (charges).

The two opposing effects for property crime may explain the absence of any visible property crime reduction (burglary) in the Old Bailey trial data. 

Finally, the roll-out of county police allows again for a difference-in-difference setup, which yields the following conclusions:

  • Creating just any police force does not result in a significant decrease in crime 
  • But, once they are large enough (close to the recommended threshold of 1,000 people per officer), the new police forces reduce crime (19% overall, 18% for violent and 14% for property crime)
  • The crime-reducing effect is not immediate but does persist over time.

Lessons from history

What lessons can be learned from this historical setting for today’s policy landscape? 

  1. With local police departments closing due to budget concerns in countries such as the US and the UK, the extensive margin effect of police on crime comes into focus. Our results highlight the importance of having (large enough) professional police forces to reduce crime.
  2. With the aim of reducing corruption, police forces in some developing countries are being disbanded and new forces created (recent examples include Ukraine and the Republic of Georgia). Our historical study emphasises features of police forces (such as their size, but also enforced behavioural guidelines) that may help new forces to successfully reduce crime.


Bindler, A and R Hjalmarsson (2019), “The Impact of the First Professional Police Forces on Crime”, CEPR Discussion Paper No. 14068.

Chalfin, A and J McCrary (2017), “Criminal Deterrence: A Review of the Literature”, Journal of Economic Literature 55(1): 5-48.

DiTella, R and E Schargrodsky (2004), “Do Police Reduce Crime? Estimates Using the Allocation of Police Forces After a Terrorist Attack”, American Economic Review 94(1): 115-133.

Draca, M, S Machin and R Witt (2011), “Panic on the Streets of London: Police, Crime, and the July 2005 Terror Attacks”, American Economic Review 101(5): 2157-2181.

Emsley, C (2009), The Great British Bobby: A History of British Policing from the 18th Century to the Present, Trafalgar Square.

Heaton, P,  Hunt, J MacDonald and J Saunders (2016), “The Short-and Long-Run Effects of Private Law Enforcement: Evidence from University Police”, Journal of Law and Economics 59(4): 889-912.

Hitchcock, T, R Shoemaker, C Emsley, S Howard, J McLaughlin et al. (April 2013), “The Old Bailey Proceedings Online, 1674-1913”, www.oldbaileyonline.org (version 7.1).

MacDonald, J, J Klick and B Grunwald (2016), “The Effect of Private Police on Crime: Evidence from a Geographic Regression Discontinuity Design”, Journal of the Royal Statistical Society: Series A (Statistics in Society) 179(3): 831-846.

Stallion, M and D Wall (1999), The British Police: Police Forces and Chief Officers 1829-2000, Athenaeum Press, p. 262.

4,829 Reads