Illinois keeps a careful account of what gun violence costs after it happens. It keeps almost no account of what it would cost to prevent the next one, or to rebuild the working life of someone who survived the last one. Those are comparable numbers. When you put them next to each other, the case for investment stops being a moral argument and becomes an arithmetic one.
Candace Metcalf — September 2026. Sociological and fiscal analysis. Sourced figures are linked. My own aggregations are labeled as estimates.
I am a sociologist and a senior rehabilitation counselor with the Illinois Division of Rehabilitation Services. I work with people with disabilities on the South Side of Chicago, including Englewood.
A single round of handgun ammunition retails for well under a dollar. What follows one of those rounds — when it lands in a person — has been measured with reasonable precision by health economists, trauma researchers, and public health departments. It is one of the best-documented cost cascades in American social policy.
What has not been measured, because it largely does not exist, is the parallel investment: a vocational service designed for the person who survives, and an early-intervention pathway for the teenager who is currently on track to be the next person shot.
This essay puts those two ledgers side by side. The first is expensive and involuntary. The second is cheap and optional. We have chosen the expensive one by default, mostly by not choosing at all.
Ledger One: What a shooting costs
The first hours
Cook County residents made 7,122 hospital visits for intentional assault gunshot wounds between 2018 and 2020, generating $342 million in charges and 24,894 hospital days across 89 Illinois hospitals. Roughly a third of those patients were admitted; 6.5% died in hospital (Feinglass et al., 2022).
That is three years, one county, hospital charges only.
Nationally, initial hospital treatment of firearm injuries cost approximately $7.7 billion between 2016 and 2021. Annual costs held near $1.2 billion through 2019 and then rose to $1.6 billion in 2021. Medicaid was the primary payer for 52% of all hospital costs, followed by private insurance at 20% (Lundberg et al., 2025).
The distributional fact matters more than the total. Because Medicaid reimbursement frequently falls below the true cost of care, safety-net trauma centers absorb the difference. The bill does not disappear. It is redistributed onto hospital operating margins, onto commercial premiums through cost-shifting, and onto the public balance sheet.
The year after
The initial admission is the smallest part of the medical cost, and it is measured because it is billed. What follows is larger and mostly invisible.
A Harvard-led cohort study matched 6,498 survivors of nonfatal firearm injury to 32,490 controls, and 12,489 family members of survivors to 62,445 controls, using commercial and Medicare claims. In the year after injury, survivors’ medical spending increased by $2,495 per person per month — a 402% increase — with a 51% rise in psychiatric diagnoses and an 85% rise in substance use disorders relative to controls (Song et al., 2022).
The finding that should reorganize how rehabilitation counselors think about this population: family members of survivors also showed significant increases in psychiatric disorders, alongside reductions in their own routine health care. A gunshot wound is not an individual injury with a household audience. It is a household-level health event.
This has a direct vocational implication. A survivor’s mother who develops a depressive disorder and leaves paid work to provide care is a second labor-market casualty of the same round, and she appears in no count of victims anywhere.
The lifetime
Here the cost moves from health care into human capital, and it is here that vocational rehabilitation should have something to say.
A young adult who sustains a spinal cord injury, a traumatic brain injury, or a disabling extremity injury at twenty-two faces a truncated earnings trajectory measured across four decades. The scarring literature in economics and sociology is consistent: interruptions to early career formation — through incarceration, injury, or extended unemployment — produce earnings penalties that persist long after the interrupting event resolves (Western & Pettit, 2005; Kirk & Sampson, 2013).
Layer onto that the hiring penalty. Audit studies have repeatedly found that a criminal record substantially reduces callback rates, with larger penalties for Black applicants (Pager, 2003; Pager, Western, & Bonikowski, 2009), and a meta-analysis of field experiments found no decline in hiring discrimination against Black applicants over roughly 25 years (Quillian et al., 2017).
Now stack them. A visible mobility impairment, a possible record, a resume gap that is actually a hospital stay, and a residential address in a community area employers read as risk. Each penalty is documented independently. Together they produce something close to labor-market exclusion.
The individual cost, then, is not a hospital bill. It is a life.
The community
Violence is also an educational and developmental cost, and this is the part of the ledger policymakers most consistently omit.
Sharkey (2010) found that Black children in Chicago who were assessed within roughly a week of a local homicide performed significantly worse on cognitive tests than comparable children assessed at other times — the effect of a shooting registering on the reading scores of children who were not shot and did not witness it (PNAS, 107(26), 11733–11738). Ang (2021) found parallel effects on academic outcomes and high school completion from exposure to police violence (QJE, 136(1), 115–168).
And the mortality gradient. Chicago Department of Public Health data for 2024 shows life expectancy varying by 18.3 years across community areas — 88.2 years in the highest, 69.9 years in Englewood — with a Black/non-Black gap of 9.4 years citywide, narrowed from 12.7 years at its 2021 peak (CDPH, 2026). CDPH itself attributes a substantial share of that gap to homicide and opioid overdose combined.
The Illinois total
Everytown’s cost model, developed with health economist Ted Miller at the Pacific Institute for Research and Evaluation, estimates gun violence at $557 billion annually nationwide, of which roughly $12.62 billion falls directly on taxpayers (Everytown Research, 2022). For Illinois the annual figure is $18.6 billion (Everytown, EveryStat: Illinois).
Per incident, in Illinois:
| Total social cost | Taxpayer share | |
|---|---|---|
| One homicide | ~$16.4 million | ~$800,000 |
| One nonfatal shooting | ~$845,000 | ~$44,000+ |
These are modeled averages, not audits, and they include quality-of-life valuations of the kind used in injury litigation — a legitimate method with a real methodological debate behind it (Cook & Ludwig, 2000; Ludwig & Cook, 2001). Readers who reject contingent valuation can use the taxpayer column alone. The argument below survives either choice.
Why the cost concentrates
The geography is not random, and the sociology explaining it is settled enough to state plainly.
Social disorganization and collective efficacy. When employment leaves a neighborhood and institutions thin out, the informal social control that regulates public behavior weakens. Sampson, Raudenbush, and Earls (1997) demonstrated in Chicago that collective efficacy — social cohesion combined with willingness to intervene for the common good — predicted lower violence independent of poverty and residential instability (Science, 277(5328), 918–924). This is the mechanism, and it is measurable.
Concentrated disadvantage. Wilson (1987) and Massey and Denton (1993) described how joblessness, poverty, and racial segregation stack in the same census tracts, producing compounding effects that none of them produce alone.
Situational adaptation, not cultural essence. Anderson (1999) described how, where legal institutions do not reliably protect people, respect becomes a private currency enforced privately. Subsequent work has been explicit that this is adaptation to constrained conditions rather than a cultural trait (Small, Harding, & Lamont, 2010; Harding, 2010). The distinction is not academic delicacy. It determines whether you design a program or a lecture.
The economics of the street are worse than assumed. Levitt and Venkatesh (2000), working from an actual gang’s financial records, found that street-level participants earned at or near minimum wage while facing extraordinary mortality risk (QJE, 115(3), 755–789). The illicit economy does not outcompete legal work on wages. It outcompetes on speed of entry, certainty of payment, and immediate status.
That finding is the single most policy-actionable thing in this literature, and it is routinely misread. A legal pathway does not need to offer more money. It needs to offer a paycheck this week, without a gatekeeping process.
What our systems currently do
Illinois DRS is a substantial operation. In SFY2022 it served over 37,000 customers and placed 4,724 people in competitive integrated employment; the agency has reported 27,619 customers entering competitive employment across a recent four-year window, against a stated annual target of 6,000 (IDHS, Vocational Rehabilitation Program).
That is real production. The question is what it is built to produce.
VR is organized around a medical-vocational taxonomy — sensory impairment, intellectual disability, psychiatric disability, orthopedic impairment. Those categories determine eligibility, and eligibility shapes service design. This is Lipsky’s (1980) street-level bureaucracy in its ordinary form: the categories available to a frontline worker define what a client can be within the system.
A firearm injury is eligible under that taxonomy. But nothing in the standard case flow accounts for the specific features of this injury: that the disabling event was a crime; that the client still lives where it happened; that survivors of a nonfatal shooting face substantially elevated risk of a second one, with that risk concentrated in identifiable social networks (Papachristos, Wildeman, & Roberto, 2015); or that the client’s household is simultaneously absorbing the mental health shock documented by Song et al.
WIOA is the other pathway, and it is a workforce statute rather than a reconstruction authority. The international meta-analysis of active labor market programs is consistent: short classroom-training spells show near-zero or negative effects in the first year from lock-in, with gains appearing only over longer horizons and concentrated in programs with genuine employer attachment (Card, Kluve, & Weber, 2018).
Neither program is failing at its own assignment. Both are being asked to absorb an injury neither was designed to price.
Ledger Two: What intervention costs, and returns
This is the half of the comparison that usually goes unstated, and it has better evidence behind it than most people assume.
Teenagers: the cheapest intervention on the board
In a randomized controlled trial of 1,634 disadvantaged Chicago high school students, an eight-week summer jobs program at minimum wage with an adult mentor reduced violent-crime arrests by 43% over 16 months — 3.95 fewer violent-crime arrests per 100 youth. Critically, the decline occurred largely after the program ended, ruling out simple incapacitation (Heller, 2014, Science, 346(6214), 1219–1223).
A follow-up of the 2013 One Summer Chicago Plus cohort, which focused on young men including justice-system referrals, found a 33% reduction in violent-crime arrests — a smaller percentage but a larger absolute effect, 7.9 fewer arrests per 100 participants, because the population was higher-risk (University of Chicago Urban Labs).
Eight weeks. Minimum wage. Effects persisting after the money stopped.
This is the strongest argument in the entire policy space for early intervention, and Chicago generated the evidence itself.
Adults at highest risk
READI Chicago offered 18 months of transitional employment plus cognitive behavioral therapy to men at extreme risk of gun violence involvement — participants averaging 17 prior arrests, with 34% having already been shot. In a randomized trial of 2,456 men, the primary combined index was not statistically significant at 20 months. Shooting and homicide arrests declined 65% (p = 0.13 after multiple-testing adjustment). Among participants referred by street outreach workers, a pre-specified subgroup, shooting and homicide arrests fell 79% and victimizations fell 43%, both surviving multiple-testing adjustment (Bhatt, Heller, Kapustin, Bertrand, & Blattman, 2023).
I am reporting the null alongside the positives on purpose. This is what honest evidence looks like.
And here is the number that makes this essay’s argument: because shootings are so expensive, READI generated estimated social savings of $182,000 to $916,000 per participant (p = 0.03), implying a benefit-cost ratio between 4:1 and 20:1 — against a program cost of roughly $60,000 per participant.
A program whose headline outcome was statistically ambiguous still returned four to twenty dollars per dollar spent, because the thing it was preventing costs so much.
Survivors at the bedside
Hospital-based violence intervention programs enroll survivors during the admission itself. One California program reported a fourfold reduction in injury recidivism, avoiding roughly $41,000 in trauma care per averted injury; the accompanying cost-utility analysis found the program cost-effective across a wide range of assumptions (Juillard et al., 2015). A 2025 systematic review of 25 studies found HVIPs reduced violence-related injury and recidivism while improving employment rates and educational engagement, with cost savings from reduced re-injury (Injury, 2025).
Note that employment outcome. It appears in the HVIP literature — a medical intervention — and not in our own.
The rehabilitation evidence we already own
The Individual Placement and Support model — place first, then train and support, with no readiness screening and time-unlimited follow-along — has been tested in spinal cord injury specifically. In a randomized multisite trial of 201 veterans, IPS participants were 2.5 times more likely to obtain competitive employment than treatment-as-usual at the same sites, and 11.4 times more likely than at observational sites (Ottomanelli et al., 2012, Archives of Physical Medicine and Rehabilitation, 93(5), 740–747).
Gunshot wounds produce spinal cord injuries. The evidence for the right service model already sits inside our own field. We are simply not deploying it for this population.
Community infrastructure
Sharkey, Torrats-Espinosa, and Takyar (2017), analyzing 264 cities across more than 20 years with a strategy addressing the endogeneity of nonprofit formation, estimated that every 10 additional community-focused nonprofits per 100,000 residents produced a 9% reduction in the murder rate and a 6% reduction in violent crime (American Sociological Review, 82(6), 1214–1240).
Community organizational density is not decoration around the real policy. It is part of it.
The comparison
| Cost | |
|---|---|
| LEDGER ONE — what we pay now | |
| One round of handgun ammunition | under $1 |
| One nonfatal shooting, Illinois (taxpayer share) | ~$44,000 |
| One nonfatal shooting, Illinois (total social cost) | ~$845,000 |
| One homicide, Illinois (taxpayer share) | ~$800,000 |
| One homicide, Illinois (total social cost) | ~$16.4 million |
| Illinois, annually | ~$18.6 billion |
| LEDGER TWO — what intervention costs | |
| One youth, 8-week summer job with mentor | low thousands |
| One READI participant, 18 months | ~$60,000 |
| Estimated social savings, per READI participant | $182,000–$916,000 |
| Trauma care avoided per prevented re-injury (HVIP) | ~$41,000 |
The taxpayer share of a single Illinois homicide — roughly $800,000 — exceeds the full 18-month program cost for thirteen READI participants. The taxpayer share of eighteen nonfatal shootings covers one year of a fully staffed survivor employment program serving a hundred people.
That is the whole argument. Not that violence is tragic, which everyone concedes and no budget responds to. That we are already spending the money, at the most expensive possible point in the sequence, after the injury, in a trauma bay, through Medicaid, forever.
What should be built
A Survivor Employment Pathway, specified from the evidence above:
- Enrollment at the bedside. A VR counselor embedded in Level I trauma, not a referral eleven months later. HVIPs have demonstrated this window works.
- IPS fidelity, not train-then-place. Rapid job search, no readiness screening, time-unlimited support. The spinal cord injury randomized evidence is ours already.
- Benefits counseling in week one. Deshpande (2016) found that youth removed from SSI suffered long-run earnings losses that did not offset the lost benefits (American Economic Review, 106(11), 3300–3330). Clients who fear the benefits cliff are reasoning correctly, and a program that cannot answer that question will lose them.
- Relocation assistance as a clinical tool. Kirk (2009), using Hurricane Katrina as a natural experiment, found parolees who relocated away from former neighborhoods had substantially lower recidivism (ASR, 74(3), 484–505). When the block is the barrier, funding the move is the intervention.
- First-hire lanes in public-sector and union employment, where hiring is rule-governed rather than discretionary.
And upstream, an early-intervention commitment with a number attached: guaranteed summer employment with mentorship for every high school student in the five CDPH priority community areas. Chicago produced the randomized evidence for this program. It should not have to be re-argued every budget cycle.
The theoretical warrant is age-graded. Uggen (2000), analyzing an experimental employment program, found that stable work functions as a genuine turning point reducing recidivism for participants past their mid-twenties (ASR, 65(4), 529–546) — consistent with Sampson and Laub’s life-course framework, in which employment is not merely an outcome of desistance but a cause of it.
Employment is not the reward for having stopped. It is the mechanism by which people stop.
What I would need to be wrong
A claim worth publishing should specify how it could fail.
I would abandon the core argument if Illinois VR outcome data disaggregated by injury etiology showed firearm-injury cases closing to employment at rates comparable to other traumatic injuries. I would revise substantially if an IPS trial in a violence-survivor population failed to replicate the spinal cord injury result. I would revise the early-intervention case if One Summer Chicago Plus effects failed to replicate at scale outside the trial.
To my knowledge, the first two studies have not been conducted. That absence is itself the finding. Illinois does not publish VR outcomes by injury etiology or by community area, which means we cannot evaluate a service we are not measuring, and we are not measuring it because no one has been required to ask.
So the first reform costs nothing: publish Illinois VR outcomes disaggregated by disability etiology and by community area. Every claim in this essay becomes testable the day that data exists. If I am wrong, the data will say so, and I would rather know.
The bullet is cheap. Everything after it is not. We already pay for the aftermath, in full, every time. The only open question is whether we will also pay, at a fraction of the price, for the part that comes before.
— Candace Metcalf
Chicago, September 2026
References
Anderson, E. (1999). Code of the street: Decency, violence, and the moral life of the inner city. W. W. Norton.
Ang, D. (2021). The effects of police violence on inner-city students. The Quarterly Journal of Economics, 136(1), 115–168. https://doi.org/10.1093/qje/qjaa027
Bhatt, M. P., Heller, S. B., Kapustin, M., Bertrand, M., & Blattman, C. (2023). Predicting and preventing gun violence: An experimental evaluation of READI Chicago. University of Chicago Crime Lab. https://crimelab.uchicago.edu/resources/readi-paper/
Card, D., Kluve, J., & Weber, A. (2018). What works? A meta analysis of recent active labor market program evaluations. Journal of the European Economic Association, 16(3), 894–931. https://doi.org/10.1093/jeea/jvx028
Chicago Department of Public Health. (2026). Life expectancy in Chicago, 2024. https://www.chicago.gov/
Cook, P. J., & Ludwig, J. (2000). Gun violence: The real costs. Oxford University Press.
Deshpande, M. (2016). Does welfare inhibit success? The long-term effects of removing low-income youth from the disability rolls. American Economic Review, 106(11), 3300–3330. https://doi.org/10.1257/aer.20151129
Everytown for Gun Safety Support Fund. (2022). The economic cost of gun violence. https://everytownresearch.org/report/the-economic-cost-of-gun-violence/
Everytown for Gun Safety Support Fund. (2025). Hospital-based violence intervention programs: A guide to implementation and costing. https://everytownresearch.org/
Everytown for Gun Safety. (2026, April). A $25 device is costing Illinoisans their lives and tens of millions of dollars. https://www.everytown.org/
Everytown for Gun Safety. (n.d.). Gun violence in Illinois. https://www.everytown.org/state/illinois/
Feinglass, J., Patel, T. R., Rydland, K., & Sheehan, K. (2022). Trends in hospital care for intentional assault gunshot wounds among residents of Cook County, Illinois, 2018–2020. American Journal of Public Health, 112(5), 795–802. https://doi.org/10.2105/AJPH.2022.306747
Harding, D. J. (2010). Living the drama: Community, conflict, and culture among inner-city boys. University of Chicago Press.
Heller, S. B. (2014). Summer jobs reduce violence among disadvantaged youth. Science, 346(6214), 1219–1223. https://doi.org/10.1126/science.1257809
Illinois Department of Human Services. (n.d.). Vocational Rehabilitation Program. https://www.dhs.state.il.us/page.aspx?item=146841
Juillard, C., Smith, R., Anaya, N., Garcia, A., Kahn, J. G., & Dicker, R. A. (2015). Saving lives and saving money: Hospital-based violence intervention is cost-effective. Journal of Trauma and Acute Care Surgery. https://pubmed.ncbi.nlm.nih.gov/25757108/
Kirk, D. S. (2009). A natural experiment on residential change and recidivism: Lessons from Hurricane Katrina. American Sociological Review, 74(3), 484–505. https://doi.org/10.1177/000312240907400308
Kirk, D. S., & Sampson, R. J. (2013). Juvenile arrest and collateral educational damage in the transition to adulthood. Sociology of Education, 86(1), 36–62. https://doi.org/10.1177/0038040712448862
Levitt, S. D., & Venkatesh, S. A. (2000). An economic analysis of a drug-selling gang’s finances. The Quarterly Journal of Economics, 115(3), 755–789. https://doi.org/10.1162/003355300554908
Lipsky, M. (1980). Street-level bureaucracy: Dilemmas of the individual in public services. Russell Sage Foundation.
Ludwig, J., & Cook, P. J. (2001). The benefits of reducing gun violence: Evidence from contingent-valuation survey data. Journal of Risk and Uncertainty, 22(3), 207–226. https://doi.org/10.1023/A:1011144200419
Lundberg, A., Stey, A., et al. (2025). Hospital costs of firearm injuries in the United States, 2016–2021. JAMA Health Forum. Summary: https://ihpi.umich.edu/
Massey, D. S., & Denton, N. A. (1993). American apartheid: Segregation and the making of the underclass. Harvard University Press.
Ottomanelli, L., Goetz, L. L., Suris, A., McGeough, C., Sinnott, P. L., Toscano, R., Barnett, S. D., Cipher, D. J., Lind, L. M., Dixon, T. M., Holmes, S. A., Kerrigan, A. J., & Thomas, F. P. (2012). Effectiveness of supported employment for veterans with spinal cord injuries: Results from a randomized multisite study. Archives of Physical Medicine and Rehabilitation, 93(5), 740–747. https://doi.org/10.1016/j.apmr.2012.01.002
Pager, D. (2003). The mark of a criminal record. American Journal of Sociology, 108(5), 937–975. https://doi.org/10.1086/374403
Pager, D., Western, B., & Bonikowski, B. (2009). Discrimination in a low-wage labor market: A field experiment. American Sociological Review, 74(5), 777–799. https://doi.org/10.1177/000312240907400505
Papachristos, A. V., Wildeman, C., & Roberto, E. (2015). Tragic, but not random: The social contagion of nonfatal gunshot injuries. Social Science & Medicine, 125, 139–150. https://doi.org/10.1016/j.socscimed.2014.01.056
Quillian, L., Pager, D., Hexel, O., & Midtbøen, A. H. (2017). Meta-analysis of field experiments shows no change in racial discrimination in hiring over time. PNAS, 114(41), 10870–10875. https://doi.org/10.1073/pnas.1706255114
Sampson, R. J., & Laub, J. H. (1993). Crime in the making: Pathways and turning points through life. Harvard University Press.
Sampson, R. J., Raudenbush, S. W., & Earls, F. (1997). Neighborhoods and violent crime: A multilevel study of collective efficacy. Science, 277(5328), 918–924. https://doi.org/10.1126/science.277.5328.918
Sharkey, P. (2010). The acute effect of local homicides on children’s cognitive performance. PNAS, 107(26), 11733–11738. https://doi.org/10.1073/pnas.1000690107
Sharkey, P., Torrats-Espinosa, G., & Takyar, D. (2017). Community and the crime decline: The causal effect of local nonprofits on violent crime. American Sociological Review, 82(6), 1214–1240. https://doi.org/10.1177/0003122417736289
Small, M. L., Harding, D. J., & Lamont, M. (2010). Reconsidering culture and poverty. The ANNALS of the American Academy of Political and Social Science, 629(1), 6–27. https://doi.org/10.1177/0002716210362077
Song, Z., Zubizarreta, J. R., Giuriato, M., Paulos, E., & Koh, K. A. (2022). Changes in health care spending, use, and clinical outcomes after nonfatal firearm injuries among survivors and family members: A cohort study. Annals of Internal Medicine, 175(6), 795–803. https://doi.org/10.7326/M21-2812
Uggen, C. (2000). Work as a turning point in the life course of criminals: A duration model of age, employment, and recidivism. American Sociological Review, 65(4), 529–546. https://doi.org/10.2307/2657381
Western, B., & Pettit, B. (2005). Black-White wage inequality, employment rates, and incarceration. American Journal of Sociology, 111(2), 553–578. https://doi.org/10.1086/428816
Wilson, W. J. (1987). The truly disadvantaged: The inner city, the underclass, and public policy. University of Chicago Press.
Leave a Reply