What's the ROI on Cold Case Investigations?

The field of forensics has grown by leaps and bounds over the past several years, so much so that decades-old crime cases can sometimes be solved with DNA testing and other modern technology. In an effort to increase case clearance rates (and catch bad guys long gone) police departments have slowly opened more ‘cold case’ units over the last 20 years; a phenomenon that has been documented and dramatized on TV. 

In a new RAND paper, researchers Robert C. Davis, Carl Jensen, and Karin E. Kitchens set out to measure the effectiveness of cold case units by posing a simple question, though one that’s rarely asked of police work: What’s the return on investment? They write:

[D]espite the increasing number of cold-case units and the expenditure of significant resources to fund them, we know virtually nothing about the return on this investment. Does it make sense for law enforcement agencies to devote significant resources to solving cold cases, or are those resources better deployed in solving recent cases? How can agencies best organize and support cold-case work? What kinds of cold cases are most susceptible to being solved? What is the payoff to solving a cold case: Does solving a case always lead to arrest, prosecution, and conviction?

The researchers mailed out 5,000 surveys to police departments across the country, from which they got 1,050 responses. They found that only 20 percent of agencies have any sort of protocol for cold cases, and that on average, clearance rates for these cases are low, with only 1 in 5 turning up a “conclusively identified” suspect. However, clearing a case doesn’t always mean a successful prosecution – the prime suspect could be identified, but missing for any number of reasons. Therefore, only 1 in 20 cold cases resulted in an arrest, and 1 in 100 resulted in conviction. Shockingly, many departments had no idea how many of their own cold cases had been cleared or closed.

The majority of cold case departments (56 percent) are funded by grants or supplemental funds rather than directly from the police budget; the median cold case allocation is about $35,000. The consistent budgeting for these departments is also very much in question – only half of grant-funded units indicated that they could rely on a steady source of money. Compounding the scatter-shot finances is the lack of systematic policies for those departments to follow. The authors write: “Only 13 percent of agencies reported having specific policies on the types of cold cases or circumstances under which DNA samples are submitted for matching.”

Examining the cases themselves, the authors found different success rates based on different variables. Cases reopened due to family or media inquiries were the most expensive, and unsuccessful. Cases reopened because of available forensic testing were inexpensive and more successful, though they still had clearance rates below 50 percent. Cases reopened because of a confession or plea deal were obviously both inexpensive and the most successful. In all three scenarios, however, the perpetrator is often dead, in prison, or part of the plea bargain.

Beyond that, there are some strange circumstantial variables that contribute to a successfully solved cold case. 

Some factors associated with crime context—whether the victim was a drug user and age of the case were associated with clearances: A case was more likely to be solved if it was more recent, if it did not involve a drug user, and if the victim had been found in a private residence. It was more likely to be solved if the victim was younger or male. Also, some factors associated with the initial investigation—whether there was a known motivation for the crime and whether a prime suspect had been identified—also predicted success…

The researchers end with a call to conduct more analysis of cold cases as a whole, in particular on investigator time spent on cold cases versus newer ones, and whether prosecutor involvement leads to higher rates of conviction.  They conclude:

We were surprised at the lack of accountability in cold-case work. Specifically, there is little emphasis on convictions as a goal of cold-case investigations. If obtaining a conviction were the ultimate goal, then it would seem logical for cold-case investigators to work closely with prosecutors when screening cases so they could decide whether, if the case were to be solved, there would likely be a prosecutable case.

Matt Price

None of this tells me anything if I don't know what the clearance/arrest/conviction rates are for current cases in the same cities. I vaguely recall from The Wire (a fictional show), the Baltimore Homicide had less than a 50% clearance rate. Compared to that, the cold case unit doesn't seem too bad.


The spam-blocker is preventing any commenting to the Election Stories post.

Joe Dokes


I think you need to re-read the summary. While cases using new forensic techniques had a clearance rate under 50%. Other cases had a clearance rate of under 5% and a conviction rate of 1%. Which is clearly well below the Baltimore rate of 50%.

Although Baltimore might have a low clearance rate, not all metropolitan areas have such a low clearance rate. For example, here in Long Beach Ca. the clearance rate is 90%. Considering the number of active gangs, and poverty in the city I believe this is quite an accomplishment.

Clearly when or if someone comes in to a police department with NEW information regarding a crime, then the police should investigate. Otherwise, it appears that cold case departments are not a good use of limited resources.


Joe Dokes

Enter your name...

This naturally leads to the question, "Is obtaining a conviction the ultimate goal?" It's entirely possible that the ultimate goal is something else: to satisfy the stakeholders (law enforcement professionals' curiosity, family members' desire to know what happened, etc.), to look tough on crime during an election cycle, or (by identifying a deceased suspect) to rule out other possible suspects.


In other words once a case becomes "cold" the chances to solve it are...small. If you kill someone and leave no DNA evidence and there aren't any witnesses and you are not related to that person...chances are you get away. This is why professional killers (I was about to say assassin but that's a name for a former sect of muslims) exist.


Point number 1: IF police did an adequate job of real investigation into criminal investigations, then there would be no cold case files or funds needed for a separate "feel good" unit to set up a tip line....and wait for a hand out from a tipster needing a reduced sentence. No one knows how to investigate anymore any.

Where should the focus of interest be? If there is a 30 year old homicide case of an unidentified person, would it go the the infamous "file 13" (not interested) IF there is little to no hope of grandstanding in a court room? That is a stunning mindset!

Most missing person cases end up being a homicide case - unsolved - generally with a case file of misinformation or incomplete information in either a medical examiner's office or stuffed away in some DA'S desk. Today's Ninga Warrior who got his police training in a gym vs top notch criminology courses (no, not online!) does not consider missing person cases real police work. Yes, I know it is not a criminal offense for an adult to want to zone out and leave to unknown parts of the world, most meet up with foul play along the way. But then, who cares? Not much chance of finding out who dumped a human corps along a major highway!

We remain an uncivilized society, where work is a burden and people without a pedigre or some form of notoriety are throw aways.

Just when I feel my work as a volunteer cold case researcher is haping little impact on the big picture, I read something like this and say:" Thank G0d I have not lost my compassion for the victims and their families. If ONLY ONE missing/unidentified cased is resolved with the identity known and the family knowing what happened to their loved one, then all cold cases are important.

Most important point? There is no such thing as a "cold case". There are only cases which have gone ignored because no one knos how to take the information and turn it into a new lead.