The average identity fraud victim lost about $1,517 in 2025, a figure calculated by dividing Javelin Strategy & Research's disclosed $27.3 billion in total US identity fraud losses by its count of 18 million victims, from the firm's 2026 Identity Fraud Study, published April 21, 2026 from a survey of more than 115,000 US consumers. That single number hides enormous variation: government survey data puts the typical victim's loss far lower, while FBI complaint records put it far higher, depending entirely on how "victim" and "average" get defined.

Average identity theft victim lost 1,517 dollars in 2025 $1,517 average loss per identityfraud victim in 2025, computed

What is the average cost of identity theft per victim?

There is no single official "average cost" figure, because no one source tracks every victim the same way. Three named organizations, each measuring a different population, arrive at three different numbers, and none of them is wrong once you know what each one counts.

Javelin Strategy & Research, an industry research and advisory firm that sells fraud-analytics services to banks and identity-protection companies, does not publish a per-victim average directly. Dividing its disclosed $27.3 billion in 2025 identity fraud losses by its disclosed 18 million victims works out to $1,517 per victim, a calculation this post is making from Javelin's own published totals rather than a figure Javelin itself states. The US Department of Justice's Bureau of Justice Statistics measures something different and does publish a direct average: among identity theft victims who experienced any financial loss in 2021, the mean loss was $1,160 and the median was $200, drawn from its Identity Theft Supplement to the National Crime Victimization Survey, a nationally representative household survey rather than an industry poll.

SourceWhat it measuresAverage lossYear
Javelin Strategy & ResearchTotal losses divided by total identity fraud victims (calculated)$1,5172025
US DOJ, Bureau of Justice StatisticsMean loss among victims with any financial loss$1,1602021
US DOJ, Bureau of Justice StatisticsMedian loss among victims with any financial loss$2002021
FBI, Internet Crime Complaint CenterTotal losses divided by total identity theft complaints filed$8,1462024

The Javelin and Bureau of Justice Statistics figures land in a similar range once you account for methodology, but the FBI's number is nearly seven times higher. That gap is the most important thing to understand about any "average identity theft cost" headline, and it comes down to who ends up in the denominator.

Average identity theft loss per victim, by source and year FBI IC3 complaints (2024)8,146$Javelin, calculated (2025)1,517$DOJ mean, with loss (2021)1,160$DOJ median, with loss (2021)200$

Figure 1: Four ways to measure the average cost of identity theft, none of them wrong, none of them comparable without context. Sources: FBI Internet Crime Complaint Center (2025), Javelin Strategy & Research (2026), US DOJ Bureau of Justice Statistics (2023).

Why does the average cost vary so much by fraud type?

The Bureau of Justice Statistics' 2021 survey breaks identity theft losses down by exactly what happened to the victim, and the differences are large. Having a new account opened in your name was the single costliest category, with a mean loss of $3,360 and a median of $1,000, more than five times the mean loss from having an existing credit card misused, at $620.

Mean loss per victim by identity theft type, 2021 (USD) New account opened3,360$Email or social media2,730$Other personal info misuse2,680$Multiple types, other1,920$Multiple existing accounts1,160$Bank account misused680$Credit card misused620$Other existing account550$

Figure 2: Mean financial loss by type of identity theft, among victims who experienced a loss. Source: US Department of Justice, Bureau of Justice Statistics, "Victims of Identity Theft, 2021," October 2023.

The pattern makes sense once you consider what each category actually involves. Misusing an existing credit card is usually caught fast, disputed with a bank, and reimbursed, capping the victim's real exposure. Opening a brand-new account or misusing an email or social media login gives a criminal more room to run before anyone notices, which is why those categories carry both a higher mean loss and, for new accounts specifically, a much higher median loss of $1,000 against $200 for existing-account misuse.

The Bureau of Justice Statistics also separates direct loss from indirect loss, a distinction most other trackers skip. A direct loss is the dollar value of goods, services, or cash the offender actually obtained; an indirect loss is a related cost, like legal fees or notary charges, that was not reimbursed. Across all identity theft victims in 2021, the mean direct loss was $880 and the mean indirect loss was $280, and after accounting for reimbursement from banks and card issuers, the mean out-of-pocket loss that stayed with the victim was $790.

Figure 3: How a single identity theft incident splits into no-loss, direct-loss, and out-of-pocket outcomes. Source: US DOJ, Bureau of Justice Statistics, "Victims of Identity Theft, 2021."

Do FBI complaint records show a different average than survey data?

Yes, and the difference is the largest gap in this entire dataset. The FBI's Internet Crime Complaint Center, known as IC3, logged 21,403 identity theft complaints in 2024 with $174,354,745 in associated losses, which works out to $8,146 per complaint, a figure this post calculated from IC3's published totals. IC3 also tracks a separate "Personal Data Breach" category, 64,882 complaints and $1,453,296,303 in losses in 2024, which works out to roughly $22,399 per complaint, even higher.

Neither IC3 figure is comparable to a survey mean, and the gap is not a sign that FBI data is more accurate. IC3 complaints are self-selected: someone has to notice a loss, decide it is serious enough to report to a federal agency, and complete a complaint form. That filters out the large share of low-loss or no-loss identity theft incidents that show up in a random household survey like the Bureau of Justice Statistics' but never reach IC3 at all. The result is a smaller, more severely affected pool with a much higher average loss per case.

FBI IC3 reported losses by crime type, 2024 (USD billions) 6.574.172.771.461.450.17Investment fraud6.57All other reported crime types4.17Business email compromise2.77Tech support fraud1.46Personal data breach1.45Identity theft0.17

Figure 4: Identity theft is a small slice of total FBI-reported cybercrime losses, which is part of why its per-complaint average swings so much year to year. Source: FBI, Internet Crime Complaint Center, "2024 Internet Crime Report."

How much do account takeover victims lose compared to new account fraud victims?

Javelin's 2026 study separates identity fraud into two main categories that behave differently. Account takeover, where a criminal hijacks an existing account rather than opening a new one, affected 6 million victims in 2025 with losses topping $15 billion, an implied average just over $2,500 per victim, calculated from Javelin's disclosed totals. New account fraud, where a criminal opens a fresh account using stolen information, affected 5.4 million victims with $7 billion in losses, an implied average of about $1,296 per victim.

That ordering, account takeover averaging more per victim than new account fraud, differs from the Bureau of Justice Statistics' 2021 finding that opening a new account carried the highest mean loss of any single category. The discrepancy is a reminder that Javelin's categories and the Bureau of Justice Statistics' categories are not defined the same way: Javelin's account takeover bucket includes high-value targets like brokerage and cryptocurrency accounts that a government household survey does not separately break out, while the Bureau of Justice Statistics' new-account figure covers a narrower slice of consumer credit and banking accounts. Neither number is wrong, but neither should be quoted as though it settles the question of which fraud type costs more on its own.

For a business collecting personal data from customers, the practical lesson is the same regardless of which agency's average you use: the accounts and personal details a site stores are the raw material every one of these fraud categories runs on, and a site that is unclear about what it collects, retains, or shares makes itself a more attractive and less defensible target when something goes wrong. A current, accurate privacy policy generator that spells out data collection, retention, and third-party sharing practices is a direct, low-cost way to reduce that exposure and demonstrate accountability if a breach does happen.

How many total victims sit behind these dollar figures?

The average-cost figures on this page sit inside a much larger count of affected people. Javelin's 2026 study found 18 million identity fraud victims in 2025, and the Department of Justice's own household survey put the broader figure at 23.9 million people affected by any form of identity theft in its most recent count. That underlying data privacy exposure, how much personal information companies collect and how well they protect it, is the upstream driver of every dollar figure in this post; a related angle worth tracking is how much individual victims recover through data breach class-action settlements, which shows what happens after a breach turns into a payout rather than a loss.

The Bottom Line

Three credible sources, four different numbers, and one clear pattern: the average cost of identity theft depends almost entirely on how the average is built. Divide Javelin's total 2025 losses by its total victim count and you get $1,517. Use the Bureau of Justice Statistics' direct household-survey mean among people who actually lost money and you get $1,160, or its median of just $200 if a handful of large losses should not dominate the picture. Narrow the population further to only the people who filed a federal complaint, and the FBI's IC3 data puts the average at $8,146. None of these numbers is more "correct" than the others; each answers a different question about a different group of people. For a site owner, the type-level breakdown is the most actionable part of this data: new-account fraud and account misuse carry the highest per-victim losses, and both start with a criminal having enough personal information to pass as the real customer, which is exactly what disciplined data collection and disclosure practices are meant to limit.

Frequently Asked Questions

What is the average cost of identity theft per victim? About $1,517 in 2025, calculated by dividing Javelin Strategy & Research's disclosed $27.3 billion in total identity fraud losses by its count of 18 million victims, from the firm's 2026 Identity Fraud Study published April 21, 2026. The US Department of Justice's 2021 household survey found a lower mean of $1,160 among victims who lost any money, and a median of just $200.

Why is the median identity theft loss so much lower than the average? Because a small share of high-loss cases pull the mean upward. The Department of Justice's Bureau of Justice Statistics found a $1,160 mean loss but only a $200 median loss among identity theft victims in 2021, and victims who had a new account opened in their name lost a mean of $3,360, nearly three times the overall average.

Do FBI complaint records show higher average losses than survey data? Yes, by a wide margin. Identity theft complaints filed with the FBI's Internet Crime Complaint Center averaged $8,146 in losses each in 2024, based on $174,354,745 across 21,403 complaints, compared with the Department of Justice household survey's $1,160 mean. The gap reflects who chooses to file: FBI complainants are a self-selected group of people motivated enough to report, not a random sample of everyone affected.

Which type of identity theft costs victims the most? Having a new account opened in your name is the costliest, with a mean loss of $3,360 and a median of $1,000, according to the Bureau of Justice Statistics' 2021 survey. Misuse of an existing credit card was both the most common and the least costly type measured, with a mean loss of just $620.

Where the Numbers Come From

  1. Javelin Strategy & Research. (2026). "2026 Identity Fraud Study: The Illusion of Progress." 23rd annual edition, survey of more than 115,000 consumers, published April 21, 2026. 18 million identity fraud victims, $27.3 billion in total losses, 6 million account takeover victims with losses exceeding $15 billion, 5.4 million new account fraud victims with $7 billion in losses. Javelin is an industry research and advisory firm that sells fraud-analytics services to banks and identity-protection companies; the $1,517 and other per-victim averages in this post are calculated from its disclosed totals, not figures Javelin itself publishes.
  2. US Department of Justice, Bureau of Justice Statistics. (2023). "Victims of Identity Theft, 2021." National Crime Victimization Survey, Identity Theft Supplement, published October 2023 (NCJ 306474), authors Erika Harrell and Alexandra Thompson. 23.9 million victims, 58.9% with a financial loss of $1 or more totaling $16.4 billion, mean loss $1,160 and median $200 among victims with a loss, mean direct loss $880, mean indirect loss $280, mean out-of-pocket loss $790, and loss figures by type of identity theft ranging from $550 to $3,360 mean.
  3. Federal Bureau of Investigation, Internet Crime Complaint Center. (2025). "2024 Internet Crime Report." 21,403 identity theft complaints with $174,354,745 in losses, 64,882 personal data breach complaints with $1,453,296,303 in losses, $16.6 billion in total losses reported to IC3 across all crime types in 2024.

Note: All figures verified as of September 2026. The Bureau of Justice Statistics' household survey updates on a multi-year cycle, so its 2021 figures remain the most recent available even though Javelin and the FBI publish updated figures annually. Per-victim and per-complaint averages calculated in this post from each source's own disclosed totals are marked as calculated throughout rather than presented as a published average.