I was happen to present testimony in support of a NJ bill to expand collection of DNA for human trafficking, burglary, arson, and car jacking arrests.
Here is my written submission.
Dear SLPS Committee,
I submit this testimony in support of S4312. Collecting DNA at arrest for these additional
enumerated offenses – human trafficking, arson, carjacking, and as amended, burglary – will
help New Jersey solve more crimes, solve them faster, and identify the right person. That
combination is the most reliable crime deterrent. Additionally, it helps ensure justice for victims and survivors as well as those accused.
Arnold Ventures is a national philanthropy working to maximize opportunity and minimize
injustice through a combination of funding highly rigorous, causal research and scalable policy grounded in that evidence. In criminal justice, that has meant supporting work to help police solve more crimes faster and to reduce unnecessary incarceration while making prisons a safer place for inmates and officers and offer a path for rehabilitation. We see those goals as connected. A system that solves few crimes tends to make up for it with severity. A system that solves more can lean less on long sentences.
What deters crime
Decades of research point to one finding. People are deterred mainly by the chance they will be
caught. The National Institute of Justice, summarizing Daniel Nagin’s review of the field, concludes that the certainty of being caught deters far more than the punishment itself, and that increasing the severity of punishment and applying it randomly does little to deter crime. Police deter crime when their work raises the perceived odds of getting caught.
Speed matters alongside certainty. A consequence that arrives quickly and reliably shapes behavior. One that arrives years later, if ever, mostly does not. The failure to apprehend an offender quickly risks allowing that person to reoffend and cause additional harm. Long
sentences handed down in the small share of cases that get solved amount to punishment by lottery, severe for the few who are caught, and absent for everyone else.
DNA is one of the few tools that raises certainty and speed together, and it does so cheaply. Economist Jennifer Doleac estimated that preventing one serious offense through DNA profiling cost under $555. Preventing one through longer sentences cost about $7,600, and through hiring more officers between $26,300 and $62,500.
New Jersey’s certainty gap
In 2025, New Jersey police cleared 52% of all violent crimes combined. So, your likelihood of
getting arrested for a violent crime was a coin flip. Homicide clearance rate was the highest at
67%, while rape was only 17%. Burglary, an offense this bill addresses, was only cleared at 17% as well. There is variation among which victims are more likely to have their cases solved by demographics. For instance, from 2020 to 2024, homicides with Black victims were solved 53% of the time, versus 84% for white victims. Each unsolved case leaves a victim without answers and leaves the person responsible free to offend again.
New Jersey’s DNA database is also narrower than it could be. The state collects DNA from
everyone convicted of a crime. At arrest, it collects only for a short list of offenses: murder,
manslaughter, second-degree aggravated assault, kidnapping, sexual assault, and certain crimes against children. Nineteen states and the federal government collect DNA at arrest for all felonies.
Because conviction-based collection already reaches these offenses, the real change in S4312 is timing. It moves collection from the end of a case, often months or years after arrest, to booking.
What the evidence shows
DNA solves crimes that would otherwise go unsolved. In a randomized trial run by John
Roman and colleagues at the Urban Institute with the U.S. Department of Justice, police in fivejurisdictions (Orange County and Los Angeles, California; Topeka, Kansas; Denver; and Phoenix) collected biological evidence at burglary scenes from 2005 to 2007. Testing that evidence identified a suspect in 31% of cases, versus 12% under usual investigation, and doubled the arrest rate from 8 to 16%. DNA was three times as effective as fingerprints at identifying a suspect. Each arrest of an otherwise unknown offender cost about $3,500. Those arrested averaged three prior felony convictions and nearly six prior felony arrests, many for serious crimes against people.
The national database is badly underused. CODIS now holds about 25 million profiles and
aids roughly 50,000 investigations a year. Against about two million serious violent and property crimes annually, that is about 2.5%. Roman attributes this mainly to eligibility rules. Most states add profiles only from people convicted of the most serious crimes, who then serve long sentences, so the database grows slowly. Research shows that the number of offender profiles in a database predicts how often DNA aids investigations.
DNA databases deter. Doleac, using criminal history data from seven states, found that being
required to give a DNA sample reduced the likelihood of a new conviction within five years by 17% for serious violent offenders and 6% for serious property offenders. A second study, by
Anker, Doleac, and Landersø, examined Denmark’s 2005 reform allowing police to add anyone charged with the equivalent of a felony. The chance of being in the database rose from 4% to nearly 40%. One-year reconviction fell by up to 42%, and police identified offenders 3 to 4 percentage points more often when they did reoffend. People added to the database were also more likely to be working or in school afterward. Denmark’s reform is charge-based collection, the same step S4312 takes.
Why these offenses
The main payoff from arrestee collection usually comes from matching the person arrested to other unsolved crimes. People who commit serious crimes rarely specialize. Former New Mexico Governor Susana Martinez, who prosecuted the Sepich case, put it simply: offenders “don’t stay in their lane.”
- Burglary. The evidence here is the strongest. A California Department of Justice analysis of the first 100 matches under Proposition 69 found that burglary arrests generated 25% of murder matches, 12% of rape matches, and 30% of robbery matches. In Texas, after the state began collecting at arrest for all felonies, burglary investigations were the second most frequently aided (236), after sexual assault (382). In New Mexico, burglary arrests have matched more unsolved crimes than any other category of arrest. Katie Sepich’s killer was arrested for aggravated burglary less than three months after her murder. Because New Mexico did not collect DNA at arrest, the match came three years
later. - Carjacking. Carjackings are often committed by small groups operating across county lines. A statewide database connects cases across jurisdictions in a way local investigation cannot. In Texas, arrests for unauthorized use of a motor vehicle produced 77 database hits in the program’s first years, and Kansas reports regular hits from stolen auto cases.
- Human trafficking. Trafficking frequently involves sexual violence. In Texas, sexual
assault was the investigation most often aided by arrestee DNA. Adding trafficking
arrestees gives New Jersey’s tested sexual assault kits more profiles to match against. - Arson. Fire destroys much of the biological evidence at a scene, and arson is too rare to
study on its own. The case for including it rests on the person arrested, who may be
linked to other crimes, and on the danger arson poses to life. Maryland’s statute, the one upheld in Maryland v. King, treats arson as a crime of violence.
The legal footing is settled
In Maryland v. King (2013), the U.S. Supreme Court held that taking a cheek swab at booking
from a person arrested for a serious offense is a reasonable search under the Fourth Amendment, comparable to fingerprinting and photographing. The Maryland statute it upheld covered crimes of violence, including arson, and burglary. The Court also found that the markers used in CODIS do not at present reveal information beyond identity. It noted that analyzing samples for traits like disease risk would raise different concerns. The safeguards recommended below take that line seriously.
In State v. O’Hagen (2007), the New Jersey Supreme Court upheld the DNA Database and
Databank Act under both the federal and state constitutions. The Court accepted that the Act
serves identification, solving crimes, deterrence, and exonerating the innocent. The Legislature
has since written the Court’s reasoning into the Act’s findings, including the parallel between
DNA and fingerprinting. New Jersey has collected DNA at arrest for its most serious offenses since its 2011 amendments.
Answering civil liberties concerns
I would like to address some common misperceptions raised against DNA collection at the point of arrest.
“Collecting DNA before conviction undermines the presumption of innocence.” The
presumption of innocence governs what the state must prove at trial before it can punish
someone. It has never barred booking procedures. Fingerprints, photographs, and pretrial detention all happen before conviction. A DNA swab at booking is an identification step, and a person whose charges are dropped can have the profile removed.
“DNA reveals health, ancestry, and family relationships.” This confuses the physical sample with the profile stored in the database. A CODIS profile records 20 short tandem repeat markers in non-coding regions of DNA, plus a marker for sex. It is a string of numbers that works like a serial number. The Supreme Court found that these markers reveal nothing beyond identity at present. Whole-genome sequencing, the kind consumer ancestry companies use, is a different technology that S4312 does not authorize.
“DNA is relational and exposes family members.” Standard CODIS searching looks for an exact match to one person. Familial searching and forensic genetic genealogy are separate techniques that this bill does not authorize. If the Legislature wants to regulate those, it can do so in its own legislation.
“A breach can’t be undone. You can’t change your DNA like a password.” The same is true
of fingerprints and photographs, which New Jersey already stores. CODIS profiles carry no
names. Each profile is tied to a specimen number, and the identity behind it stays with the New Jersey State Police laboratory. Federal law makes unauthorized database information disclosure a crime.
“Expanded collection puts communities of color under genetic surveillance.” Racial disparities in arrests are real and deserve attention. So do disparities in which crimes get solved. In New Jersey, homicides with Black victims are solved far less often than those with white victims, and those families are owed answers too. DNA is also the most objective evidence available. Mistaken eyewitness identification, which DNA helps replace, contributed to most of
the wrongful convictions later overturned by DNA. Collecting at arrest moves the discovery of a wrong arrest to the start of a case when it is easiest to correct.
