Crime and poverty: Correlation or causation?

Question

it is well known that difficult socioeconomic conditions are associated with higher prevalence of crime, especially violent crime. Living through difficult economic conditions through childhood myself and having many friends and family members going through even worse conditions I have never fully bought the idea that "poverty breeds crime". At least not in a direct causal sense.

Are there good scientific studies that study this question? In particular, if the relation is causal and not just correlational? If no, what would be a viable mechanism to separate correlation from causation? Is there even a scientific consensus on this topic?

I guess the answer may depend also on the specific country, where the study may be conducted, but I would be glad to know about result anywhere in the world.

EDIT:

Thank you all so much for very elaborate answers. Unfortunately, I cannot accept more than one answer, otherwise I would.

It seems that my initial skepticism was substantiated. To summarize the answers, there is no a single universal answer. It all depends on the circumstances. Here are some theses I drew from the answers:

1. Sometimes poverty pushes people to commit crime.
2. Sometimes high crime in a poor community pushes the community to even more poverty, starting a vicious cycle.
3. Sometimes direct causal relationship does not exist
4. All of these statements are substantiated by the existing research.

Unsurprisingly, blanket statements (e.g "poverty breeds crime") are reductive and mostly wrong.

Answer 1

It really depends (but sometimes there is causation)

As noted in other answers, "crime" is not necessarily a philosophically coherent category. Some classic examples of activities that have been classified as crimes and that have generally not had a strong correlation with wealth include same-sex relations, where these are banned, which still includes much of the world—at the very least, sexual attraction is not a function of how much money one has—and certain types of illicit drug use; for instance, to some extent, being wealthier might actually encourage illicit drug use, as hinted at in Table 1 of this study, presumably in part because it means having enough money to buy these often expensive substances.

Still, focusing exclusively on that would be missing some important points. There is a correlation between certain major categories of crime—the poster mentioned murder, theft, and assault—and neighborhood-level income and wealth levels, and one that is not simply a consequence of arbitrary illegalization of an activity nor only the result of disparate enforcement, though disparate enforcement is another important factor here.

The kinds of crimes that are encouraged by poverty tend to have to do with what the sociologist Émile Durkheim called innovation, meaning any effort to pursue socially acceptable goals through socially unacceptable means. In this sense, poverty can, for some people, serve as a powerful motivation to engage in certain crimes that offer a financial benefit, such as selling illegal drugs. Because of diminishing returns—every additional dollar, yen, naira or rupee makes someone less happy after a certain point, such that it's not the same for a poor person to get USD 100 as for a rich person to get the same amount—the income from certain illegal activities, relative to the substantial risk it entails, is more attractive if one has limited money. I mentioned that drug use does not necessarily positively correlate with lower income, but selling drugs probably does. As noted in this article, which divided the study group into three classes, of which two ("delinquents" and "externalizers") were characterized by selling drugs more than the other ("dabblers," who were more drug users):

Members of the delinquents class were significantly more likely to reside in households earning less than $20,000 (RR = 2.92, 95% CI = 1.78-4.77) or between $20,000 and $49,000 per year (RR = 1.89, 95% CI = 1.29-2.76). Members of the externalizers class were also significantly more likely to reside in households earning less than $20,000 per year (RR = 1.63, 95% CI = 1.01-2.64).

If we assume that, all other things equal, inherent willingess to commit a crime is independent of income, as seems likely, then certain financially beneficial crimes will be proportionally more attractive to the same people if those people lack sufficient financial resources to fulfill their needs. Though, interestingly, sometimes attaining markers of social status can be more powerful than immediate needs in driving people to these types of "innovative" activities, as one study in Ciudad Juarez, Mexico suggests.

In any case, it is worth mentioning that this is not universal; for instance, certain kinds of theft, such as shoplifting and embezzling, are more common among higher income groups, perhaps largely because of opportunity. Indeed, one reason that wealthier people may be more likely to engage in crimes such as embezzlement is that the rewards are higher relative to the risk, and they need higher rewards to make it worthwhile in any case, because of those diminishing returns.

It's also important to note that some of these financially beneficial crimes that are concentrated in lower-income neighborhoods can also increase the incidence of violent crime, due to their connections with organized criminal groups and the perceived need to use violence to engage in economic competition when the legal system does not provide non-violent enforcement (e.g. corporations can sue each other, but the Latin Kings can't sue the Aryan Family for selling drugs in "their" territory). A substantial proportion of the violent crimes that are concentrated in lower-income areas are essentially financial, associated with gang membership, organized criminal groups, botched robberies, and so forth.

"Pure" interpersonal crimes caused by anger or alienation—Durkheim's anomie—while they may be inversely correlated with income in some contexts (e.g., a study in Turkey found higher rates of domestic abuse of women in low-income partnerships, something that seems to be the case more broadly) are not necessarily the primary kind of crime driven by poverty (anomie aside, when it comes to hurting other people for irrational reasons, it seems unlikely that money makes people moral). In addition, while social alienation is likely a factor in some cases, its correlation with poverty is not clear-cut and binary; notably, in the USA, mass shootings, which seem like one of the purest expressions of crimes driven by anomie, occur in a wide range of contexts for a wide range of (apparent) reasons.

Finally, getting at the doubt in the question, this suggests that it is also the case that none of these factors are universal, and in particular, that even in cases where poverty does play a causal role, it may well more often serve as a license or an incentive for the same types of people who exist in every socioeconomic group to engage in actions that they otherwise might not have, rather than directly affecting the number of people willing to engage in antisocial behavior.

What is perhaps less obvious is that the causation works the other way around: certain kinds of crime can encourage poverty. As an article about the disproportionate impact of gun violence on poor communities of color in the USA notes:

What I did not detail in that essay were the specific community factors associated with concentrated poverty that often lead to negative outcomes for children as they grow into adults. Social science research suggests that childhood exposure to urban gun violence is the most important factor in this regard.

The cycle of violence is brutal, self-reinforcing and yet another perpetuator of poverty in our most disadvantaged communities. When violence is commonplace, schools cannot properly educate children who are in fear; parents keep their children inside out of fear for their safety; medical professionals cannot fully address the direct and indirect consequences of violence; enrichment activities suffer; and businesses and those residents who can leave flee.

Violent crime, then, can create a state of fear that actually helps produce or at least reinforce poverty in a community, even in cases where the community was initially relatively prosperous (in the USA, some such cases have been seen as the result of policies against Black communities and other communities of color).

So, to summarize:

• Crime is not a monolithic category, and it can be difficult to get a picture of the incidence of certain crimes due to disparate enforcement and self-reporting biases.

• Some crimes are certainly more likely to be committed in lower-income communities due to financial incentives.

• Some crimes are probably more likely to be committed in lower-income communities due to the alienating effects of social exclusion on some individuals.

• Some crimes have no correlation with income.

• Some crimes are more likely to be committed in higher-income communities.

• Some crimes may even precede and precipitate the transition of a community from high-income to low-income.

Answer 2

FWTW, there's a somewhat old [1997] World Bank study on Jamaica:

Disentangling the direction of causality between violence and urban poverty is difficult. There are many channels through which violence exacerbates poverty. For example, stigma attached to violent communities frequently prevents otherwise qualified residents from gaining employment. This study, however, also found causality running from poverty to violence. Thus, lack of employment leads to increased levels of violence. The example of unemployment leading to violence leading to less employment shows the circularity of the poverty-violence nexus. Hence, this study concludes that there is a vicious cycle: violence leads to more poverty, which in turn leads to more violence.

Unfortunately, violence also has a negative impact on the traditional interventions which are designed to reduce poverty. Schools, for example, often close during episodes of high violence when teachers refuse to report to work. [...]

The example of unemployment leading to violence leading to less employment shows the circularity of the poverty-violence nexus. Hence, this study concludes that there is a vicious cycle: violence leads to more poverty, which in turn leads to more violence.

Whether it's a "high quality study", IDK it's more like descriptive than based on regressions, but sure it's long at some 76 pages. They discuss various forms of poverty-related crimes, including e.g. stealing electricity, which was sometimes done violently, e.g. by threatening utilities workers. And the attractiveness of the drug trade relative to what regular jobs pay, in that country. But even the allocation of regular jobs themselves was sometimes controlled through violence by local 'dons' etc. And relatedly, on a higher level, there was political violence.

Ultimately, I think it's going to be subjective whether you assign whether someone stealing electricity is doing it because of their poverty or because of their 'culture of crime'. (That kind of discussion does come up when it comes to stealing the electricity cables, at least. Sometimes that is stated as poverty creating a culture of crime. But it's probably difficult to prove what creates any state of mind, or even if they exist as such.) Take that a step further and consider the looting of aid in various conflicts in poor countries etc.

(Some natural experiments in allegedly homogenous populations claim to prove that (at least) a 'culture of [organized] crime' is a real thing, at least in terms of in-group vs out-group trust. However that's only so-so helpful for the claim that poverty creates something somewhat similar.)

---

And if you'd rather have regressions, perhaps "“Hunger makes a thief of any man”: Poverty and crime in British colonial Asia" is worth a read:

This study uses rainfall variation as an instrumental variable for rice production to estimate the impact of poverty on different types of crime across British colonies in South and South East Asia (1910-–1940). Using original primary sources retrieved from annual administrative and statistical reports, it provides some of the first evidence in a historical setting on the causal relationship between poverty and crime. Extreme rainfall, both droughts and floods, lead to a large increase in property crimes (such as robbery, petty theft, and cattle raiding), but not to an increase in interpersonal violent crimes (such as murder, homicides, and assault). In line with a growing body of literature on the climate-economy nexus, this study offers evidence that loss of agricultural income is one of the main causal channels leading to property crime.

[...] Figure 2 shows how the relationship looks when we scatter plot property crime (fitted values in grey) against rainfall deviation (x-axis).

The authors themselves discuss however that causality may come through other channels e.g.

if heavy rains, due to flooded roads for instance, reduce criminals’ likelihood of being detained by the police or hamper police's capacity to report crimes. Therefore, if such channels are present, IV estimates could misattribute the direct effects of rainfall to crime.

Answer 3

Short Answer

Individual Level Factors

The vast majority of serious crimes are economically motivated.

A serious lack of a capacity to support oneself economically through legitimate means, without poverty measured on on relative basis in the offender's society, is a much better predictor of propensity to commit serious crime, than absolute affluence or income or even short term relative poverty levels.

Usually, a serious lack of a capacity to support oneself economically through legitimate means, lead to crime only when accompanied by an additional "plus factor" such as substance abuse, a traumatic brain injury, or a moderate to serious mental health condition.

Furthermore, serious crime is overwhelmingly committed by adolescent and young adult men. This is highly related to testosterone. A man's likelihood of committing a crime falls by a little bit almost every year after age twenty-one, and is below average by middle age. Age and gender demographics are an important factor in high crime rates in Alaska and Wyoming, which have many young, single men who work in the oil and gas and coal industries.

In places where men in these circumstances are concentrated (e.g. homeless encampments, large public housing complexes, slums, and non-age restricted trailer parks), crime rates are consistently elevated.

Someone who is at high risk of committing a serious crime is at very high risk of committing many crimes. The vast majority of crimes are committed by people with criminal records, often serious and lengthy ones, and about half of serious crimes in the U.S. are committed by gang members. Someone's likelihood of committing a crime follows a power law with a minority of offenders committing a greatly disproportionate share of crimes.

Long Term Trends And International Differences

Long term trends in crime rates, and differences in crime rates between countries are highly speculative.

The causes of the long term trends in the murder rate in the U.S., for example, are a matter of great controversy.

Similarly, the major long term decline in crime rates overall in the U.S. over the last 30 years is a matter of great controversy.

Property crime is down 63% since 1980, and violent crime is down 33%, but we're not entirely sure why.

Some of the well-established causes of long term changes in crime rates and international crime rates are the size of the cohort of men at the peak age for committing crimes relative to the overall population, and the proportion of the population that is institutionalized (i.e. in prison, in involuntary inpatient mental health treatment, and in the military, combined). But other causes of long term and international differences in crime rates are more speculative.

Notably, crime rate trends in modern England and the U.S. show very similar timing and magnitude, which tends to disfavor U.S. specific explanations for these trends in the U.S.

Some crime reduction is due to making it harder to commit a crime (e.g. with anti-theft ignition systems in cars and PIN requirements to use credit cards). For example, a 2012 study found that requiring a personal identification number (PIN) to use a credit card reduced credit card fraud in the U.K. by 63% when this was done.

A dramatic reduction in crime rates in the U.S. for men of any given age from 1985 to 2019, however, makes clear that demographics alone do not explain crime rate trends over time.

Likewise, we aren't entirely sure why the number of felonies committed in New York City (whose population was basically constant over this time period) declined by 82% from 1990 to 2017 (with homicides down by more than 85%).

Differences in rates of gun crimes between countries are strongly linked statistically to gun control measures and gun ownership rates, and there is very little substitution between gun homicides and non-gun homicides, in places where gun control is particularly great, like the U.K. and Japan.

Gun Homicide Rate Per 100,000 people:

• U.S.: 3.48 (about 58 times as great as the U.K.)

• U.K.: 0.06 (more than 6 times more than Japan)

• Japan: less than 0.01 (more than 348 times less than the U.S.)

Non-Gun Homicide Rates Per 100,000 people:

• U.S. 1.22 (21% more than Japan and 10% more than the U.K.)

• U.K. 1.11 (10% more than Japan)

• Japan 1.01

Some gun control measures have some impact between U.S. states but the free travel between U.S. states makes those effects more modest. For example, “shall-issue concealed carry permitting laws were significantly associated with 6.5% higher total homicide rates, 8.6% higher firearm-related homicide rates, and 10.6% higher handgun-specific homicide rates compared with may-issue states.” See also the chart below:

(This chart from this April 4, 2018 article in the Annals of Internal Medicine, for non-Hispanic white men only, combining data from 2008-2016. (Washington D.C. and Hawaii omitted due to small sample sizes.)).

Organized crime and access to gun (when both are present together) are both important drivers of murder rates:

Taking a global view, the six countries with the highest age-adjusted rates of firearm homicides are:

1. El Salvador
2. Venezuela
3. Guatemala
4. Colombia
5. Honduras
6. Brazil

Research has found high levels of homicides in these countries are associated with drug cartels, the illegal trade in firearms from the US, and firearms flowing to civilians after conflicts end, as summarized in the Global Burden of Disease study.

(Source)

Another long term and comparative factor in serious crime rates is the extent to which a group of people has a cultural of honor and strong clan/group identities. Cultures of honor are associated with higher rates of crime (which is one reason that crime is more common in the South in the U.S.). This is also supported by historical evidence of the decline of cultures of honor in England:

The qualitative historical materials we present reveal the growth of more individualized conflicts evident in less third-party partisan intervention and a reduced concern with honor. More individualized conflicts were, in turn, a product of a more individualized society, one characterized by increased social distance and mobility. As conflicts individualized they became less lethal, resulting in declining aggregate rates of homicide. Although the case study is historical, our argument has implications for understanding contemporary criminal violence. . . .

Scholars of the past have proposed that over several centuries homicide rates across Western Europe declined steeply. Drawing on a variety of historical sources, Elias (1939: 190-217) described a reduction in European violence from the Middle Ages. After Gurr (1981) assembled some more systematic data showing a steep drop off in English homicide rates historians began to uncover a similar pattern in other regions of Europe, including Scandinavia (Sweden and Finland), Belgium, [and] the Netherlands. . . .

The decline in third-party participation we call the individualization of conflict. Instead of drawing in outsiders either as loyal partisans or as members of an audience who must be impressed, conflicts increasingly involve the two principals alone.

MarkCooney, JefferyPatterson "Individualization and the decline of homicide: England 1250–1750" Journal of Criminal Justice (November 3, 2022) https://doi.org/10.1016/j.jcrimjus.2022.101997 See also Jeffrey L. Jensen, Adam J.Ramey, "Going postal: State capacity and violent dispute resolution" 48(4) Journal of Comparative Economics 779-796 (December 2020) (showing the weak state capacity was associated with higher rates of deaths in duels in the 19th century).

However, the evidence on this is mixed. Many Islamic countries have strong cultures of honor (although this is a regional cultural matter and is not uniform across the Islamic world). But in a comparison of murder rates across 124 countries by the World Health Organization:

Regression results suggested no significant difference in lethal violence between predominantly Catholic and Protestant countries, although Islamic countries revealed significantly lower homicide, suicide, and overall lethal violence rates than non-Islamic countries. Countries with a high level of religious heterogeneity are subject to an increased suicide rate.

Don Soo Chon, "National Religious Affiliation and Integrated Model of Homicide and Suicide" (2016).

One possible explanation is that Islam countries greatly reduce access to alcohol and at least temporary lack of access to alcohol is associated with lower crime rates (although in the U.S., prohibition led to organized crime and higher crime rates).

On the other hand:

Democratic countries with the lowest levels of religious faith and participation today — such as Sweden, Denmark, Japan, Belgium and New Zealand — have among the lowest violent crime rates in the world and enjoy remarkably high levels of societal well-being. If secular people couldn't raise well-functioning, moral children, then a preponderance of them in a given society would spell societal disaster. Yet quite the opposite is the case.

Societal norms can also influence the character of certain kinds of criminal activity. For example, in the U.S. bullying is mostly associated with low status perpetrators, while in Japan, bullying is mostly associated with high status perpetrators abusing the authority and trust that has been reposed in them.

Recurring Short Term Trends And Localized Effects

There are also some well-established short term or seasonal causes of changes in the crime rate. Serious crime rates increase significantly on very hot, dry days. See, e.g., Leah H. Schinasi, Ghassan B. Hamra. "A Time Series Analysis of Associations between Daily Temperature and Crime Events in Philadelphia, Pennsylvania." Journal of Urban Health (2017); DOI: 10.1007/s11524-017-0181-y Crime rates are higher on full moons relative to new moons. There is usually a spike in crime when bars close. And, the lowest crime day of the year in most years is Mother's Day.

Allowing adult entertainment businesses in a neighborhood reduces the incidence of rapes in that neighborhood. Similarly, legalizing prostitution materially reduces sexual assault and drug crime rates. (Source: citing multiple studies.) And, while this is only established a matter of correlation and not causation, the rate of child sexual abuse declined significantly precisely in line temporally with the rise of widely available free pornography over the Internet.

Counterintuitively, the strategy of killing or capturing the top leaders of organized crime groups dramatically increases crime as power struggles between factions within the organize crime group erupt and no one is in a position to regulate the aggregate level of criminal activity by organized crime groups.

Statistical Issues With Urban v. Suburban and Rural Crime Rate Comparisons

Serious crimes rates are higher in non-residential central city areas than in residential and rural neighborhoods, although some of this is an issue of statistical methodology, because non-residential central cities have much higher numbers of people in them during the course of an average day than the number of people who have their primary residences there, artificially elevating the number of crimes per person with a primary residence in that city or neighborhood of a city. Age and gender demographics, and the "central city" effect are one of the reasons that, for example, Las Vegas, Nevada has a high crime rate.

Long Partial Answer

This isn't a complete answer, but there is indirect evidence about the causes of crime from the characteristics of people who are incarcerated in prison (from the annual report of the Colorado Department of Corrections in 2009, where not otherwise noted, but not that far out of date) can be at least suggestive regarding the issue:

• Among inmates (the general population) 1% (43% including 33% of the Colorado population that has a BA or more) of an associate's degree or other college degree, 11% (22%) have some college but no degree, 17% have a high school diploma but no college (15%), 34% have a GED (9%), and 37% (11%) have neither a high school diploma nor a GED. About 8% (10%) had an IQ of under 81. At least 36% of inmates are functionally literate. (Notably, more intellectually and economically capable potential criminals are more strongly influenced in deciding to commit a crime by the likelihood of being caught and the likely length of the sentence than less able potential criminals.)

• Moderate to severe substance abuse is a problem for 79% of inmates. Only 30% of inmates without a substance abuse problem have a high school diploma and 24% have neither that nor a GED.

• Just 6% of Colorado inmates both have a high school diploma (as opposed to a GED) and don't have a moderate to severe substance abuse problem. A significant share of the inmates in that 6% have moderate to severe mental health problems or traumatic brain injury.

• About half are gang members. See also here (discussing gang membership and its consequences among Colorado prison inmates).

• A very high percentage of inmates had a criminal record when they committed the crime for which they are incarcerated (probably more than 80% and definitely more than 75%). Once released from prison an estimated 68% of released prisoners were arrested within 3 years, 79% within 6 years, and 83% within 9 years. (Source) In federal prison (which skews heavily towards white collar crime compared to state prisons), 70% of inmates have either no criminal record or only a minimal criminal record. (Source) This is true even for lesser crimes like shoplifting. For example:

Nearly a third of all shoplifting arrests in New York City last year involved just 327 people, the police said. Collectively, they were arrested and rearrested more than 6,000 times. . . . Some engage in shoplifting as a trade, while others are driven by addiction or mental illness[.]

(Source)

Indeed, even children of people with criminal records are at greatly elevated risk of committing crimes. According to one criminologist’s analysis of the National Longitudinal Study of Adolescent Health, 5 percent of families account for more than 50 percent of all arrests.

• 96% of inmates in a high risk unit of the Denver jail have a history of traumatic brain injury. By comparison, 67%-80% of all jail and prison inmates, nationally, have a history of traumatic brain injury, and 6%-8.5% of the general population has suffered from a traumatic brain injury. In the Denver high risk jail unit, 100% had a prior criminal history, more than 90% had a history of mental illness and more than 90% had a history of substance abuse. More than 77% of the inmates in the high risk unit had all four risk factors. (Source) Traumatic brain injury (often undiagnosed) explains a significant share of the excess crime rates, homelessness, and suicide rates observed among veterans who experienced combat while in military service. Some of the effect of traumatic brain injury is direct and some is a result of its impact on the economic prospects of those who have it. For example, "nearly two-thirds of those experiencing chronic homelessness in metro Denver suffer from some kind of brain injury[.]" (Source)

This United States is not exceptional in this regard:

The distribution of criminal behaviour is highly skewed, with a very small proportion of the population accounting for a significant share of total crime. Incapacitation of this group through imprisonment results in a significant reduction in societal exposure to criminal activity.

(i) A study of the Swedish population born between 1958 and 1980 found that 3.9% of the cohort was convicted of a violent crime. A group of persistent offenders accounting for 1% of the total population accounted for 63.2% of all convictions. These offenders were relatively likely to commit offences early, use drugs, and display personality disorders.

(ii) A paper comparing convictions from a UK longitudinal study and a US dataset of self-reported delinquency found that in both cases, criminal behaviour was well described by a power law.

(iii) Some 70% of custodial sentences in England and Wales are handed out to offenders with at least seven previous convictions or cautions; 50% to those with at least 15.

(iv) Roughly 73% of 2016 US federal offenders had previous convictions. Among this group, the average number of convictions was ~6. Around 39.5% of these offenders had prior violent offences.

Offenders often possess characteristics which require greater support both in prison and in general society, and which may make standard models of rational choice a poor fit for their decision-making processes.

[Multiple studies show that people convicted of crimes have "high discount rates" which is to say that they disproportionately care about the short term and short change the long consequences.] . . . .

(iv) Some 53,109 adults in British prisons were being treated for alcohol or drug abuse issues in 2017-18, out of a total prison population of around 82,000.

(vi) The National Institute on Drug Abuse states that ”85% of the [US] prison population has an active substance use disorder or were incarcerated for a crime involving drugs or drug use”

(vi) The UK does not have up-to-date figures on mental health in prisons, but the Institute of Psychiatry estimated over half of prisoners have common mental disorders - PTSD, anxiety, or depression

• while another 15% have specialist mental health needs, and 2% acute and serious problems.

(vii) The APA estimates that “at least half of prisoners have some mental health concerns”, and 10-25% of US prisoners serious mental illnesses such as schizophrenia.

A study reviewing TBI incidence in Australian prisons from 2006 found an 82% TBI incidence rate. (My link to the original source for this statistic is broken, but other studies of Australian prison inmates have produced similar results.)

Psychopathy is a mental health condition that isn't recognized as such in the DSM-5 (anti-social personality disorder isn't quite the same thing), but is fairly well defined and validated.

of the approximately 6,720,000 adult males that are in prison, jail, parole, or probation, 16%, or 1,075,000, are psychopaths. Thus, approximately 93% of adult male psychopaths in the United States are in prison, jail, parole, or probation.

(Source)

In contrast, increased home ownership rates tends to significantly reduce property crime. See Richard Disney, John Gathergood, Stephen Machin, Matteo Sandi, "Does Homeownership Reduce Crime? A Radical Housing Reform from the UK" The Economic Journal uead040 (June 5, 2023) https://doi.org/10.1093/ej/uead040

Age and Sex

• Everywhere in the world, men are dramatically more likely to commit serious crimes, especially serious violent crimes, than women. About 93% of U.S. prison inmates are men, and women are disproportionately non-violent offenders. Also, while the sample sizes are small, the recidivism rates of castrated inmates is dramatically lower than other inmates and a significant share of those who reoffend did so while receiving testosterone booster shots.

• Crime is disproportionately committed by adolescents and young adults, in every ethnic subcategory (See, e.g., here which is the source for the chart below.) People generally "age out" of being a risk of committing crimes gradually with age, falling below the average crime rates by the time that they are middle aged.

Sex and age effects are less pronounced for non-violent crimes (e.g. embezzlement and drug dealing) than they are for violent crimes.

An example of the impact of demographics on crime rates comes from Medieval Oxford England:

The team behind the Medieval Murder Maps – a digital resource that plots crime scenes based on translated investigations from 700-year-old coroners’ inquests – estimate the per capita homicide rate in Oxford to have been 4-5 times higher than late medieval London or York.

Among Oxford perpetrators with a known background, 75% were identified by the coroner as “clericus”, as were 72% of all Oxford’s homicide victims. During this period, clericus is most likely to refer to a student or member of the early university.

“A medieval university city such as Oxford had a deadly mix of conditions,” . . . “Oxford students were all male and typically aged between fourteen and twenty-one, the peak for violence and risk-taking. These were young men freed from tight controls of family, parish or guild, and thrust into an environment full of weapons, with ample access to alehouses and sex workers.

(Source)

Analysis

This tends to show that crime is highly concentrated in people who have very poor prospects of earning income legitimately, usually with some other mental health or traumatic brain injury or cognitive impairment plus factor in addition.

As an example of this interaction, recidivism is about one-third greater among released inmates who score higher on the mental health trait of impulsivity. See, e.g., here.

A general theoretical treatment of the link between poverty and crime (not necessarily strongly rooted in empirical evidence) can be found at Christopher Lewis, "Inequality, Incentives, Criminality, and Blame" 22(2) Legal Theory (2016) ("The disadvantaged have incentives to commit crime, and to develop criminogenic dispositions, that limit the extent to which their co-citizens can blame them for breaking the law. This is true regardless of whether the causes of criminality are mainly “structural” or “cultural.” We need not assume that society as a whole is unjust in order to accept this conclusion. And doing so would neither stigmatize nor otherwise disrespect the disadvantaged.")

In support of this hypothesis, investing a lot of money in "at risk" young men with employment, job training, and scientifically validated mental health treatment, for example, reduced homicide rates and shootings in a recent Chicago based study, although only at not quite a statistically significant level.

Economically motivated immigrants consistently have lower crime rates than demographically comparable native born people, because immigrants are drawn from a self-selected population that believes that they have strong legitimate economic prospects. See, e.g., Robert Adelman, et al., "Urban crime rates and the changing face of immigration: Evidence across four decades" 15(1) Journal of Ethnicity in Criminal Justice (November 21, 2016).

The very strong protective effect of even an associate's degree also suggests that mental health and substance abuse are decidedly second order factor that rarely comes into play unless someone has no economic prospects due to a lack of education, although, of course, severe mental health and substance abuse problems can make it difficult to obtain a higher educational degree.

It also bears noting that almost all factors that affect crime rates have less strong effects for comparatively minor crimes than for comparatively serious crimes (an exception that proves the rule is animal cruelty, which is often a misdemeanor but is a strong indicator of psychopathy).

At the very bottom of the seriousness range, traffic offenses and parking violations have only a slight relationship to risk factors for other serious crimes (except ADHD which is also a major risk factor for traffic offenses).

Relative Affluence Matters More Than Absolute Affluence

Absolute levels of income and assets aren't that relevant as shown by international comparisons and comparisons over long periods of time. People with income and assets that would put them at the poverty level in the U.S., but which are "middle class" in less affluent countries, are not at elevated risk of committing crimes.

For example, crime rates are quite low in most of the Islamic world, despite the fact that people in most countries in the Islamic world (outside a few oil rich countries in Arabia) are much less affluent than people in developed countries.

Similarly, crime rates were low in the 1950s, despite the fact that poverty rates were much higher than they are today (in excess of 20% in most years), and despite the fact that average incomes and net worth, even adjusted for inflation, were lower. However, in the 1950s, there was a much higher demand of less skilled labor and educational credentials were much less common, so far fewer people (in general, and young men, in particular) had very low prospects for earning income legitimately.

Relative measures of income and assets are more predictive of propensity to engage in crime.

Evidence From The Crimes Themselves

The vast majority of serious crimes are economically motivated.

Property crimes (larceny, car theft, burglary), fraud crimes, drug dealing, robbery, and human trafficking are almost always economically motivated.

Arson, kidnapping, extortion, and child pornography/child prostitution related offenses, while not always economically motivated, are often economically motivated. Most immigration offenses are at least partially economically motivated.

A significant share of murders and assaults, while not inherently economically motivated, are also connected to economically motivated criminal activity, often in connection with organized criminal activity as part of a gang.

Answer 4

The Head Start program is based on the idea that poverty is a cause of crime (and future low income, and health problems, ... ). It's a US program helping poor children. It started in 1965, was nationwide, sometimes went on and off, was sometimes in one county but not in a nearly identical next-door county -- so it's got lots of decent studies involving long-term effects, with control groups, etc... , which tend to show it works; that reducing the effects of poverty reduced crime and other bad effects.

I searched "head start effectiveness crime" and found Early childhood education and crime reduction which summarizes a paper titled “The Effect of Early Childhood Education on Adult Criminality: Evidence from the 1960s through 1990s”, which "appears in the February 2023 issue of the American Economic Journal: Economic Policy."

The relevant quote (in the web-page's summary of that paper) is: "Overall, the authors find that Head Start lowered the conviction rate by 1.3 percentage points in high-poverty areas—a reduction of roughly 20 percent."

Answer 5

Poverty and crime certainly correlate, but poverty isn't so much a causal factor as a gateway to criminal activity. There are a few causal factors to consider here…

1. Impoverished people have more incentive to commit minor crimes — mainly theft of various sorts — because they don't have the means to secure luxuries, or sometimes even the necessities of life. Rich people have no credible reason to steal a loaf of bread from a market (though some do anyway). They can afford to buy one, so they are never confronted with that particular moral dilemma.
2. Impoverished people lack access to the 'normal' modes of economic
   self-improvement. They can't afford college tuition (certainly not at elite schools), rarely have well-off friends or relatives they can network for job placements, can't get loans for businesses and such because they lack collateral. So people with a drive towards economic self-improvement are often tracked into 'black-market' modes of commerce. These modes necessarily involve varying degrees of criminality, and often harsh competition for resources and clients, which can lead towards violence.
3. Law enforcement (in Liberal nations, at least) is generally construed in 'property' terms, meaning that law is geared towards protecting property more than people. This gives a tremendous legal advantage to the wealthy. An impoverished individual who (allegedly) kills someone will have a difficult time defending himself, while a corporation that (allegedly) kills hundreds through malignant practices will at best face a stiff fine. People and entities with property have skewed the law to work in their interests, so the impoverished tend to suffer more from the law's 'fair' execution.

If we're talking specifically about violent crime, and separate out teleological violence — violence committed merely in pursuit of a criminal goal, like muggings or shootouts between competing drug gangs — then the primary causal factor of violence is ego. Someone becomes enraged because the fell belittle, shamed, disrespected, swindled, deceived, etc, and commits violence to redress that injury. There's slight advantage to the wealthy here as well, since they have the resources to do economic or reputational violence instead of physical violence, but I don't think it's significant. A certain percentage of the population — largely males, but irrespective of wealth — is ego-bound in a way that isn't satisfied with mere tit-for-tat retribution. Remember billionaire Trump's assertion that: "Anybody who hits me, we're gonna hit them 10 times harder".

Answer 6

You probably need to define what precisely you mean by "crime".

Because technically "crime" is just what society deems "undesirable" and what that is can vary from society to society. So in the worst case poverty is directly linked to crime, because poverty is MADE a crime. Idk if poverty itself is seen as undesirable and if the blame for poverty is assigned to the poor, idk lazyness, aversion to knowledge, inability to follow rules, etc. then the very existence can be defined as crime.

Idk a homeless person could be charged with loitering or violating rules to beg or sleep on park banks or whatnot. In which case you could ask what else should they do if they have no house, food or place to sleep.

And poverty in general makes it more likely for situations to occur in which the law is hostile towards the poor, but where the poor lack the financial means and visibility to speak out against it or in which they are so forgotten that they aren't even made aware of the law that they are violating to begin with. So in order to comply with the rules of society you kinda need to be part of that society and the more the poor are marginalized and pushed out of society the more you might have a development towards sub and counter cultures that are seen by mainstream culture as a threat outside of it's control and which are hence criminalized.

Answer 7

One interesting idea I've been pondering.

Let's suppose there is a class of working poor people in a society like USA. They have a reasonably stable employment with comparatively small wage, they can't save to buy a house so they have to rent.

They live in a crime-infested area because that's what they can afford. If crime goes away (e.g. by policing), gentrification is going to happen to the area where they live, and such people will be pushed away to another, non-gentrified, crime-affected area. Different classes will get to occupy this newly crime-free neighbourhood.

In this sense, their social model needs crime to function. One may say, they depend to have some criminally inclined youth (or whatnot) for their class to function.

The society has demand for their labor at the price tag, and let's suppose they mostly cannot escape this class and/or it is being replenished (such as, by recent illegal immigrants).

I acknowledge that the model works only within some bounding box and that phenomenon may fall apart outside of it, but it may be seen as persistent unless that happens.

So here, "high functioning" poverty and violent crime may be seen to be in symbiosis, or client-patron relationship where patron happens to be the crime part.