Property Crime Is Falling. Business Theft Losses Are Setting Records. Both Are True.

What the falling crime numbers actually say

The headline decline is real, broad, and well documented. The FBI's historic early release covering 2025 estimates property crime fell 12.4%, and its summary data tables show burglary declined in every population group, with the largest drop reaching 22.8%. This was not a one-year blip. The FBI's 2024 statistics had already shown the property crime rate falling 9% to 1,835.1 per 100,000 people.

Independent measures agree on direction. The Bureau of Justice Statistics reports that burglary and larceny-theft victimization rates both declined from 2023 to 2024. When reported-crime data and victimization data move the same way, the trend is usually trustworthy.

The release itself is worth noting. The FBI published this early look sooner than it has ever released annual figures, which means the 2025 numbers are estimates and may be revised as more agencies report. Even with that caveat, a broad decline confirmed by two independent measurement systems is about as solid as national crime data gets.

So anyone who tells you crime is surging across the board is arguing with the best available evidence. The important question is what these aggregate numbers are actually measuring, because the answer explains the paradox. National statistics count incidents wherever they occur, and most incidents have always involved households and individuals, not commercial operations. A measure built that way can improve dramatically while a specific, concentrated category of business theft gets materially worse.

Where business theft losses are setting records

While the averages fell, three categories of theft aimed squarely at businesses moved the other way, and each carries the fingerprints of organized, professional operations.

Cargo theft: fewer amateurs, bigger hauls

Verisk CargoNet estimates 2025 cargo theft losses at roughly $725 million, up about 60% from the prior year. The detail that matters most sits underneath the total: the average value per theft rose 36% to $273,990. Trade coverage in the Commercial Carrier Journal notes that overall theft activity was roughly flat while losses surged. Read those two facts together and the picture sharpens. The incident count did not explode; the sophistication and selectivity did. Crews are identifying higher-value loads, using strategic methods like identity fraud and fictitious pickups, and executing fewer but far more profitable thefts. Our companion breakdown of cargo theft statistics for 2026 goes deeper on how these operations work.

Organized retail crime: incidents down, organization up

The National Retail Federation's 2025 report on retail theft and violence found the same split inside retail. Shoplifting incidents overall declined 12.4% in 2025, yet organized in-store theft rose 52% among retailers that track it, and fraud schemes increased, according to a trade-press summary from Security Info Watch. Casual, opportunistic shoplifting is easing while coordinated crews and fraud operations grow. We cover what that shift demands operationally in our piece on organized retail crime and remote monitoring response.

Infrastructure and materials theft: an organized-crime business line

Copper theft shows the same professionalization. AT&T reported more than 10,000 copper theft incidents against its network in 2025, with losses above $80 million, and it explicitly cited organized crime as a driver. That is one company. Stripping live telecommunications infrastructure at that scale is not petty theft; it is a supply chain feeding scrap and resale markets, run by people who scout targets and return to what works.

The 2026 scorecard: what's falling vs. what's rising

The split becomes obvious when the numbers sit side by side. Everything in the left column measures incident volume across the whole population. Everything in the right column measures dollar losses or organized activity aimed at commercial targets. Same year, same country, opposite directions.

Why both trends are true: theft is consolidating, not disappearing

National crime statistics are volume-weighted. They count incidents, and the incident pool has always been dominated by high-frequency, low-value, opportunistic theft: the stolen bike, the smashed car window, the residential burglary. That pool is shrinking, which is genuinely good news, and it drags every national average down with it.

Business theft losses, by contrast, are dollar-weighted, and dollars now concentrate in a small number of professional operations. The cargo data is the cleanest demonstration: activity roughly flat, losses up about 60%, average haul up 36%. The NRF data repeats the pattern in retail: total shoplifting incidents down while organized crews grow. The AT&T copper figures show it in infrastructure. In each case, the amateur tail of the distribution is receding while the organized head grows heavier.

This is consolidation, not contradiction. Theft is behaving like an industry under competitive pressure: the casual participants exit, and the remaining operators get more selective, more capable, and more profitable per job. A shrinking incident count and record dollar losses are exactly what that structure produces.

This also explains why the public debate feels so unresolvable. One side points at FBI averages and says the problem is exaggerated; the other points at record cargo losses, organized retail crews, and stripped copper infrastructure and says the data must be wrong. Both sides are reading real numbers. They are just reading different distributions, and neither national averages nor loss records alone describe the risk facing any individual operation.

There is one honest caveat. Categories like organized retail crime depend on which companies track and report them, and early FBI releases are estimates subject to revision. But the direction is consistent across independent sources: government incident data, insurer-affiliated cargo analysis, retail industry surveys, and corporate disclosures all describe the same bifurcation.

The question is no longer "is crime up?" It's "am I a selected target?"

For a household, the falling averages are the whole story. For a business, the averages can be actively misleading, because organized theft is not distributed randomly. Crews select targets, and selection follows value and vulnerability, not crime-rate maps.

The profile of a selected target is consistent across the categories above: concentrated value in a predictable location with limited after-hours presence. A trailer holding $273,990 of freight in an unstaffed yard. A stockroom of high-resale merchandise. A cell site or substation with copper. A jobsite with equipment, tools, and wire staged for the next phase, a risk we quantify in our review of construction security numbers for 2026. If your operation fits that profile, the 12.4% national decline offers you very little protection, because the people who remain in the theft business are precisely the ones who plan.

Selection also changes what deterrence has to accomplish. Opportunists are discouraged by a lock and a light. Professional crews scout, test response times, and come back with the right tools. Defeating them is less about making theft inconvenient and more about making detection and response certain.

The financial math changes too. When the average cargo theft removes $273,990 in one event, a single incident can erase years of security budget savings, trigger insurance consequences, and disrupt customer commitments in ways a string of small losses never would. Budgeting security against the frequency of theft made sense when theft was frequent and small. Against consolidated, high-value theft, the relevant number is the severity of the one incident you fail to stop.

What selected targets need: verified detection and response, not averages

If organized crews are choosing targets, the practical goal is to fail their site-selection test. That takes three layers working together: visible deterrence that signals the site is actively watched, detection that catches intrusion in real time, and human verification that turns an alert into a response while the theft is still in progress. Passive recording, on its own, mostly produces footage of a completed loss; we explain the distinction in plain terms in what is remote video monitoring.

This is where Vision Detection Systems fits the 2026 threat picture. Solar-powered mobile surveillance trailers put overt, relocatable coverage on yards, jobsites, and lots without trenching or infrastructure, and 24/7 remote monitoring adds trained operators who verify what AI-flagged cameras detect, issue live audio warnings, and escalate real incidents with evidence in hand. The point is not to replace anyone. It is a force multiplier, so people, yours or your security partner's, handle verified incidents instead of chasing false alarms. And because the organized-theft problem is ultimately a numbers problem, we document how we track deterrence and response outcomes in how we measure.

The 2026 statistics carry one clear instruction for operators: stop benchmarking your risk against national averages that are falling for reasons that have nothing to do with your site. Ask instead whether your operation concentrates value in a way that professional crews select for, and if it does, build detection and verified response to match the adversary you actually face.