Skip to content
SVERIGEFAKTA.com

Kostnader · brottslighet

Welfare crime

Kanoniskt intervall: ca 11–27 mdkr/år i felaktiga utbetalningar, centralt estimat ~18 mdkr (KUT/FUT, SOU 2019:59; bekräftat av ISF 2020:9 och Utbetalningsmyndigheten 2024). Ingen myndighet publicerar en jämn årsserie — det är punktvisa granskningar med olika metod. NOA och Polisen varnar samtidigt för kriminella nätverk som utnyttjar systemen.

Sources: KUT/FUT (SOU 2019:59), ISF 2020, the Payment Authority 2024, NOA, Försäkringskassan.

Last updated: View changelog

Fact pageSources checked
How we measureISF/FUT — definitioner →

Publicerade totalskattningar — inte en årsserie

Myndigheterna publicerar punktvisa granskningar med olika metod och år. Nedan visas punktskattning + intervall per rapport — inte en interpolerad årsserie.

Systemvis fördelning för 2024 (assistans, bostadsbidrag, sjukpenning m.fl.) visas separat på /valfardsbrott-2024.

Don't mix up: The Police's estimate of the criminal economy (100–150 bn SEK/year) refers to the entire criminal economy — not erroneous payouts. The gap to welfare-crime figures is 5–10×.

Reported benefit-fraud crimes 2024 (Brå)
22 436
-3 % vs 2023 · Brå — reported crimes 2024 ↗

Assistansersättning — återkrav och polisanmälningar

Own compilation from Försäkringskassan's annual reports 2014–2024 — not an official time series.

2014

Återkrav (mnkr)
220
Polisanmälda brott
95

2018

Återkrav (mnkr)
410
Polisanmälda brott
180

2020

Återkrav (mnkr)
540
Polisanmälda brott
220

2022

Återkrav (mnkr)
690
Polisanmälda brott
260

2024

Återkrav (mnkr)
780
Polisanmälda brott
290

Nätverk och klanstrukturer — separerade källor

Three different figures from different measurements. They measure different things and should not be stacked on top of each other.

≈40 släktbaserade nätverk (arbetshypotes)
40
Polisen/NOA (Löfving 2020)
Arbetshypotes, inte officiell registrering. Kritiserad internt inom Polisen.
~14,000 people in criminal networks
14 000
NOA — lägesbild organiserad brottslighet
Avser samtliga personer bedömda tillhöra kriminella nätverk — inte specifikt klanmedlemmar.
~62,000 in or close to the criminal milieu (2024)
62 000
Swedish Police/NOA 2024
Broader definition: people in or close to the criminal milieu. Should not be added to the 14,000 — it is the same milieu measured more broadly.
  • Påverkan i kommunala bolag och föreningar
  • Systematisk välfärdsbrottslighet via närstående
  • Parallell rättsskipning och våldsmonopol i vissa bostadsområden

Geografisk fördelning per polisregion

NOA 2024 — polisens egen regionsindelning (Väst, Syd m.fl., inte länsindelning).

Polisregion Stockholm

Share
38 %

Polisregion Väst

Share
22 %

Polisregion Syd

Share
19 %

Övriga polisregioner

Share
21 %

Primary sources

The question this page answers

How much welfare crime is there actually — and how do we tell erroneous payouts apart from organised crime against the welfare systems?

At a glance

Key figures — the most important numbers on this page

Erroneous payouts/year (range)

11–27 bn SEK

SOU 2019:59 (KUT/FUT) — confirmed by ISF 2020:9 and the Payment Authority 2024

Central estimate

≈18 bn SEK

Midpoint of the KUT range — individual years may deviate

Share of total benefits paid

≈5 %

SOU 2019:59 — share of ~350 bn SEK in total welfare benefits

Share judged intentional fraud

≈9 %

Remainder is agency error or unintentional mistake by the recipient

Reported benefit-fraud crimes 2024

≈15,000

Brå — reports from Försäkringskassan, PES, municipalities and others

Assistance-fraud investigation rate

≈40 %

Försäkringskassan — share of checked LSS assistance cases with irregularities

Criminal networks tapping welfare

≈75

NOA situational report 2024 — networks systematically abusing benefit systems

Payment Authority operational

2024

New agency with consolidated responsibility to prevent erroneous payouts

What does the data show?

Objective observations — not interpretations

  • The central range from the KUT inquiry (SOU 2019:59) — about 11–27 billion SEK per year in erroneous payouts — has been confirmed by ISF 2020:9 and by the Payment Authority's 2024 situational report. Individual years may deviate depending on the cycle and control intensity.
  • Only a subset — about 9 % per SOU 2019:59 — is judged as intentional fraud. The rest is spread across agency error (wrong decision, wrong data) and unintentional mistakes by recipients. Conflating these categories often leads to misleading debate.
  • Organised welfare crime is a smaller but growing subset. NOA's 2024 situational report identifies about 75 criminal networks that systematically abuse the welfare systems — personal assistance, unemployment insurance, social assistance, dental subsidies, corporate support. These networks are often the same ones active in other organised crime.
  • Personal assistance is the benefit form that recurs in most investigations. Försäkringskassan estimates that about 40 % of checked assistance cases contain some form of irregularity — from formal errors to organised crime.
  • The Payment Authority (operational from 2024) is for the first time consolidating the check for cross-agency patterns between Försäkringskassan, the Public Employment Service, CSN, the Tax Agency and municipalities. Its effect on the level of erroneous payouts cannot yet be measured but is expected to show in 2026–2027.

Method note
The 11–27 bn SEK estimate is based on the KUT method (‘correct payouts from the welfare systems’) which combines review of randomly sampled cases with register data. This yields a range, not a point estimate. Reporting and conviction statistics from Brå capture only what is detected and reported — the dark figure is large. NOA's situational report is built on police intelligence and is therefore reserved for actors where a network structure has been confirmed.

Definitions

How the numbers are counted — and what they do not cover

Erroneous payouts
Payment from the welfare systems that should not have occurred, or should have occurred at a different amount, had all data been correct. Covers three categories: intentional fraud, agency error and unintentional mistakes.
Welfare crime
Crimes directed at the welfare systems — benefit fraud, assistance fraud, document forgery in contact with authorities, money laundering of criminal proceeds through benefits. Constitutes the intentional share of erroneous payouts.The term is sometimes used more narrowly for organised crime that systematically abuses welfare — NOA distinguishes this as ‘organised welfare crime’.
KUT method
‘Correct payouts from the welfare systems’ — method for estimating the level of erroneous payouts through sample-based review of cases at the agencies concerned. Developed for the FUT delegation and later by the KUT delegation (SOU 2019:59).
Benefit Fraud Act (2007:612)
Law criminalising the provision of incorrect information — or the omission to report changes — that leads to benefits being paid out incorrectly. Penalty scale: fines to imprisonment; aggravated offence up to six years.
Assistance fraud
Fraud with personal-assistance compensation under LSS — e.g. inflated time use, fictitious next-of-kin assistance, double invoicing. A large share of detected cases involve family- or network-based arrangements.
Payment Authority
New central agency operational from 2024 with consolidated responsibility for identifying and preventing erroneous payouts across Försäkringskassan, the Public Employment Service, the Pension Authority, CSN, the Tax Agency and municipal social assistance.
Shell company
Company without real operations used to channel benefits, salaries or contracts between persons or to create fictitious employment giving rise to unemployment insurance or sickness benefit. Common in organised welfare crime.
Nominee (bulvan)
Person formally responsible for a company, tenancy or benefit application but in practice acting on behalf of another. Nominees are used to hide the real principal and to circumvent control systems.
Cross-agency control (MGK)
Cooperation between Försäkringskassan, the Tax Agency, the Police, the Economic Crime Authority, the Migration Agency and others to detect and investigate organised welfare crime. Systematised from 2018 and now part of the Payment Authority's toolkit.
Intentional fraud vs agency error
SOU 2019:59 estimates that about 9 % of erroneous payouts constitute intentional fraud by the recipient, about 20 % are agency errors (wrong decision or wrong data) and the rest are recipients' unintentional mistakes. The mix varies substantially between benefit types.

Primary sources

Agencies and research institutions behind this page

6

agencies/institutions

6

reports & studies

3

primary datasets

Frequently asked questions

Short answers to what is most often discussed

No — 27 bn is the upper bound of the KUT range for ALL erroneous payouts, including agency errors and unintentional mistakes. Only about 9 % of that (roughly 2–3 bn) is judged intentional fraud per SOU 2019:59. Organised welfare crime is in turn a subset of that fraud category. These levels must be kept apart in the debate.

Related topics

Logical next steps if you want to understand the background

Sources for this page

The data comes from official primary sources.

Report an error on this pageSupport Sverigefakta
Share this page:XFacebookReddit
Cite this page

Content and our own calculations are CC BY 4.0. Please also cite the primary source.

APA 7
Sverigefakta.com. (2026). Sverigefakta. Retrieved 2026-08-15, from https://sverigefakta.com/en/welfare-crime
Harvard
Sverigefakta.com (2026) Sverigefakta. Available at: https://sverigefakta.com/en/welfare-crime (Accessed: 2026-08-15).
Simple web citation
Sverigefakta.com, "Sverigefakta", https://sverigefakta.com/en/welfare-crime (accessed 2026-08-15).
BibTeX
@misc{sverigefakta_2026_en_welfarecrime,
  author       = {Sverigefakta.com},
  title        = {Sverigefakta},
  year         = {2026},
  howpublished = {\url{https://sverigefakta.com/en/welfare-crime}},
  note         = {Dataset v2026.2, accessed 2026-08-15}
}
Permanent link
https://sverigefakta.com/en/welfare-crime
More citation formats and rules