The Harm Gap SOC 3333 · Crime Lab · Midwestern State University
Everything you have measured so far has been street crime. Chapter 4 is about the offenses that do not fit the machinery: crime in suites, crime as an enterprise, crime that scales through a screen. They share one feature that makes them hard to see. The bigger the harm, the less likely it is to appear in the number people call the crime rate. This page measures that gap in dollars, then shows you exactly where it hides. Guess before each reveal, and keep your own notes. You will need them for Forum 4.
Every robbery in America, added up
Robbery is the offense most people picture when they hear the word crime: a person, a threat, property taken by force. The FBI once published what all of it costs. In the last year it reported the figure this way, robberies nationwide accounted for about $390 million in stolen property, at an average of $1,190 per robbery. Hold that number.
Internet-enabled fraud reported to the FBI in a single recent year cost victims $20.9 billion. How many years of every robbery in America would it take to add up to that one year of reported online fraud?
About 54 years. One year of reported internet fraud costs roughly what every robbery in the United States would cost across more than half a century. And the three big bars are all undercounts: the fraud figures are only what victims reported, and the wage theft figure counts one narrow violation, employers paying below the legal minimum, in a study of the ten most populous states. It excludes unpaid overtime, stolen tips, and misclassification entirely.
Now the part that matters for this course. Robbery, the smallest bar on this chart by a factor of fifty, is the one that appears in the crime rate, gets its own FBI category, drives police deployment, and shapes elections. The other three barely register in public conversation about crime. That asymmetry is not an accident of arithmetic. It is a question about power, and Chapter 1 gave you the two competing answers. The consensus perspective says law tracks shared values, and people genuinely do fear the mugger more than the payroll department. The conflict perspective says law tracks whose interests are threatened, and asks who wrote the rules that make one a felony and the other a wage claim.
Sources: FBI, Uniform Crime Reports, robbery topic page (about $390 million in losses nationwide, average $1,190 per robbery, 2015, the last year the FBI published a national robbery loss total in this form). FBI Internet Crime Complaint Center, 2025 Internet Crime Report ($20.877 billion in reported losses, 1,008,597 complaints, a 26 percent increase in losses over the prior year). Federal Trade Commission, Consumer Sentinel Network, 2024 ($12.5 billion in reported fraud losses across 2.6 million reports). Economic Policy Institute, minimum wage violation study (over $8 billion annually in the ten most populous states, extrapolating to more than $15 billion nationally, 2.4 million affected workers, based on 2013 to 2015 data). The 54-year figure is arithmetic from the first two numbers. These figures come from different years and different collection systems, so treat the comparison as orders of magnitude rather than a precise ratio.
Which of these is in the crime rate?
When a news story says crime rose or fell, it almost always means the FBI's violent crime total (homicide, rape, robbery, aggravated assault) plus the property crime total (burglary, larceny-theft, motor vehicle theft, and arson). That is the number. Eight real harms follow. For each one, decide whether it lands inside that number.
Look at which four made it in. Robbery, burglary, shoplifting, and the assault. Direct, physical, individual, and committed by people without institutional cover. Now look at the four that did not: wage theft, investment fraud, illegal dumping, and health care billing fraud. Every one of them is unlawful. Every one has identifiable victims. Several dwarf the first four in dollars and in bodies. They are absent because of decisions made about what the counting system would count, not because anyone decided they were fine.
One honest clarification, because this is the kind of claim students overstate. Not counted in the crime rate does not mean not counted anywhere. NIBRS, the incident-based system you met in Chapter 2, does record fraud, embezzlement, and bribery when police handle them. Federal agencies keep their own enforcement statistics, the FTC and the FBI's complaint center collect victim reports, and regulators track violations. The problem is that these live in separate systems that do not add up to anything, are never combined into a headline number, and mostly capture cases only when someone with authority decided to pursue them. There is no national harm rate. There is a crime rate, and it is built out of street offenses.
Chapter 2 and Chapter 4 just met. A measurement system is a theory of what matters, written in the form of a database schema. The dark figure you studied last time was about victims choosing not to report. This is a deeper version: harms that no reporting decision could ever place inside the number, because the box does not exist.
Who actually gets hit
Cybercrime has a stubborn public image: a young person with technical skill, targeting institutions. The complaint data tells a different story about who loses money. First, where the losses come from.
Ransomware gets the headlines and investment fraud takes the money. Most of that investment fraud is what practitioners call pig butchering: a stranger builds a relationship over weeks or months, then introduces a fake trading platform that shows the victim fake gains until the victim sends everything. It is a confidence scheme with the patience of a long con and the reach of a call center. Note also that the ransomware figure counts only the losses victims reported to the FBI, and most organizations that pay do not file a complaint.
Which age group of victims reported losing the most money in total?
Victims over 60 accounted for roughly 37 percent of all reported losses. They are not the least sophisticated targets. They are the group most likely to hold retirement savings, to own a home outright, and to be reachable by phone. Offenders are doing what Chapter 3 called instrumental crime, and they are selecting targets rationally. Chapter 14 has a name for this way of thinking about crime, and it is not about the offender's character at all.
The newest wrinkle. The same annual report included its first dedicated section on artificial intelligence, covering 22,364 complaints with more than $893 million in adjusted losses, of which over $632 million came from AI-assisted investment scams. Cloned voices, generated video, and synthetic identities lower the cost of running a convincing fraud. The offense is ancient. The scale is new. That has been the story of cybercrime since it started, and it is why Chapter 4 keeps insisting that these are not new crimes so much as old crimes with better logistics.
Source: FBI Internet Crime Complaint Center, 2025 Internet Crime Report. Investment fraud $8,648,617,756. Business email compromise $3,046,598,558. Tech and customer support scams $2,134,675,818. Personal data breach $1,314,923,988. Ransomware 3,611 complaints with losses exceeding $32 million. Victims age 60 and over: $7.748 billion across 201,266 complaints, a 37 percent increase over the prior year. Artificial intelligence: 22,364 complaints, adjusted losses over $893 million, of which AI-assisted investment scams exceeded $632 million. Bar lengths are drawn relative to the largest category. All figures are complaints filed with the FBI, so they are a floor rather than a measure of total cybercrime.
Forum 4: your post
Write roughly 250 to 300 words for your initial post, then reply to two classmates. Check that week’s module in the LMS for the posting and reply deadlines.
The question: the harms with the largest dollar figures receive the least enforcement attention. Why?
- Open with the gap in your own words, using at least two figures from this page. Say which comparison surprised you and which one did not.
- Bring one real case as evidence. Find a news story from the past year about a white-collar, corporate, cyber, or organized crime case. ProPublica and the Marshall Project are linked in the module and are good hunting grounds. Summarize it in two or three sentences and classify it with this chapter's vocabulary: is it occupational or corporate, entrepreneurial or ideological?
- Answer the measurement question directly. Would your offender appear anywhere in the crime rate? If not, where would the case show up instead, and who would have to decide to pursue it before it appeared anywhere at all?
- Take a position on why. Use the consensus and conflict perspectives from Chapter 1 by name. The strongest posts make the best case for one and then say honestly what the other explains better. The weakest posts pick a side and stop.
For your replies: push on the causal claim rather than the outrage. If a classmate says the powerful write the rules, ask what evidence would distinguish that from ordinary public fear of violence, which is real and not manufactured. If a classmate says the public simply cares more about violence, ask why wage theft is handled by a civil agency while shoplifting the same dollar amount is handled by police. Both readings have real evidence, and the point of the forum is to make you say what would settle it.
Looking ahead: Exam 1 covers Chapters 1 through 5 and there is no forum during exam week. A good study move this week is a one-page comparison sheet of the data systems from Chapter 2, since every chapter since has leaned on it, including this one.