Substances and Violence: The Numbers Behind the Story

Association is not causation. Most people who use drugs never commit a violent crime. These figures separate three different relationships, and show why the market, not just the molecule, drives so much of the violence.

1. Most drug-related killing is market violence, not intoxication

Goldstein's analysis of drug-related homicides, New York City, 1988. Of all homicides studied, 52.7% were drug related, and this is how those broke down.

Systemic (turf, debts, dealer robberies)
74%
Psychopharmacological (intoxication)
14%
Multidimensional
8%
Economic-compulsive (crime to buy drugs)
4%

In plain terms: nearly three quarters of drug-related homicide in this study came from how an illegal market settles disputes, not from drugs changing behavior.

2. The same pattern, 35 years later and an ocean away

England and Wales, April 2018 to March 2023 (Office for National Statistics, applying Goldstein's framework to the Home Office Homicide Index), compared with New York City, 1988. Shares are of drug-related homicides in each study.

New York City, 1988

Systemic
74%
Psychopharmacological
14%
Economic-compulsive
4%

England and Wales, 2018-2023 (of 1,627 drug-related homicides)

Systemic (464 cases)
28.5%
Psychopharmacological (399)
24.5%
Economic-compulsive (159)
9.8%
Ambiguous link (789)
48.5%

Same ordering through time and place: systemic largest, economic-compulsive smallest, in both studies. The UK shares look smaller because nearly half of UK cases could not be classified from administrative records, while Goldstein coded full case files, and because UK cases can meet more than one category. The UK psychopharmacological count covers suspects only; Goldstein also counted victims. Of all 3,148 homicides in England and Wales in the period, 27% met at least one Goldstein category and 52% counted as drug-related under the broad Home Office definition.

3. Alcohol is the largest intoxication contributor, and it is legal

37%of people convicted of violent crimes reported being under the influence of alcohol when they committed them (Bureau of Justice Statistics surveys)
48%of homicide offenders had been drinking beforehand (meta-analysis of homicide studies)
~7,000alcohol-attributable homicides per year in the US (CDC Alcohol-Related Disease Impact application)

4. Who is in prison for drugs

Share of sentenced prisoners whose most serious offense was a drug offense. The federal system is heavily a drug enforcement system; state prisons mostly hold people convicted of violent offenses.

Federal prison, Sept 30, 2023
45%
State prison, yearend 2022
13%

About 1.9 million people are incarcerated in the US. In state prison, 62% are serving sentences for violent offenses.

5. The demand-side evidence: treatment beats enforcement on cost

RAND, 1994: cost to reduce cocaine consumption by 1%. Treatment of heavy users is the cheapest option by a wide margin.

Treatment of heavy users
$34M
Domestic enforcement (7.3x treatment)
$246M
Interdiction (10.8x treatment)
~$367M
Source-country control (23x treatment)
~$782M

Measured against crime and lost productivity, RAND found treatment about 15 times more cost effective than enforcement. A later RAND study found treatment reduced serious crime roughly 10 times more per dollar than conventional enforcement, and 15 times more than mandatory minimums.

6. The business around the status quo

Annual spending, United States. Private prisons are a small slice; the system is mostly public money, and families pay a large share too.

Full criminal justice system
$445B
Direct corrections spending
~$81B
Paid by families (fines, fees, bail, commissary, calls)
$27.7B
Private prisons
$3.1B

Total war on drugs spending since the 1970s exceeds $1 trillion. Enforcement also falls unequally: Black Americans are arrested for cannabis possession at nearly 4 times the rate of white Americans despite comparable use (ACLU).

What a demand-side approach means

Treatment on demand, including medication for opioid use disorder. Harm reduction that keeps people alive: naloxone, fentanyl test strips, syringe services, overdose prevention sites. Diversion from arrest toward services: drug courts, deflection, crisis response. And the longer work: stable housing, employment, trauma care, and community investment where the trade and its enforcement concentrate. Plus a democratic fix: people harmed by the drug war should have a real say in how opioid settlement funds are spent.