Skip to content
SVERIGEFAKTA.com

Inquiry

Work decides.

By far the single most important variable for the public finances of immigration.

Allt på den här webbplatsen landar i samma punkt. Skillnaden mellan grupper i KI:s och Ruists beräkningar förklaras i deras modeller nästan helt av en sak: andelen i arbete. KI och Ruist mäter inte etnicitet, religion eller kulturell distans — deras data kan varken bekräfta eller utesluta sådana faktorer.

En procentenhets högre sysselsättning i en grupp om 100 000 personer flyttar nettot med ungefär 300–500 miljoner kronor om året.

Räkneexempel: 1 procentenhet av 100 000 = 1 000 fler i arbete. Skatt på svensk medianlön minus bortfallna transfereringar (a-kassa/försörjningsstöd) ≈ 300–500 tkr per person och år → 300–500 mnkr/år för gruppen. Antaganden: SCB medianlön 2022 och KI:s transfereringsprofil för nyanlända.

Employment vs. net contribution — 2022

Sysselsättningsandelar för inrikes/utrikes födda: SCB AKU 2022 (20–64 år). Uppdelning på flykting + anhörig samt MENA i etableringsfas: SCB register + egen härledning ur KI Specialstudie 117. Nettobidrag per person: KI 2022/2025.

Born in Sweden

Employment
84%
Net / person / year
−700 kr
Source
SCB AKU

Other foreign-born (work + EU/EEA + studies)

Employment
76%
Net / person / year
+19 800 kr
Source
SCB AKU

Foreign-born, total

Employment
69%
Net / person / year
+2 100 kr
Source
SCB AKU

Refugee + family

Employment
58%
Net / person / year
−27 500 kr
Source
SCB register

Refugees from MENA, establishment phase

Employment
42%
Net / person / year
−55 000 kr
Source
Beräkning

SCB AKU 2022 · SCB register (STATIV) · Konjunkturinstitutet, Specialstudie 117 (2025). Raden "MENA i etableringsfas" är egen beräkning ur KI:s modell.

Method note — how per-person reconciles with bn SEK

Per-person-siffrorna ovan (KI 2022) fördelas per individ i respektive grupp. Totalsumman för utrikes födda landar 2022 kring +4,6 mdkr (≈ +2 100 kr × 2,19 milj.). När vi på andra sidor skriver ”ca +6 mdkr” inkluderar vi även härledda effekter på inrikes föddas nettobidrag (sysselsättningseffekter på arbetsmarknaden, dvs. KI:s totaltabell, inte bara individraderna). De ~1 100 kr/förvärvsarbetande på /jamforelser är samma 6 mdkr utslaget på 5,3 milj. förvärvsarbetande — en illustrativ skala, inte en faktisk skatteförändring.

Why is work so dominant?

  1. Sweden has high taxes and generous transfers. Det innebär att en heltidsanställd betalar in mycket — och att en arbetslös tar emot mycket. Spannet mellan netto-bidragsgivare och netto-tagare är därför större här än i lågskatteländer.
  2. Establishment takes a long time for the refugee group. SCB:s registerdata för kohorter mottagna 1997–2015 visar att det tog ca 8–10 år innan halva flyktingkohorten var i arbete. För kohorter efter 2015 har tiden kortats något (SCB Integration 2022). Källa: SCB Integration — utrikes föddas arbetsmarknad, senaste utgåvan.
  3. Other foreign-born are selected into the labour market. Arbetskraftsinvandring, EU-rörlighet och studieinvandring kräver per definition antingen ett jobb, studier eller försörjning. Då blir sysselsättningen 70–80 % från start.

Reflektion

Vad frågan handlar om

Nedan är redaktionens tolkning, inte statistik.

Den politiska frågan är inte ”ska vi ha invandring eller inte” utan ”vilken sysselsättning kan vi realistiskt åstadkomma i mottagarens första 10–15 år, och vilka grupper tar vi då emot”. Allt annat är derivat av det.

What the data doesn't say

  • • KI:s och Ruists modeller mäter inte etnicitet, religion eller kulturell distans — bara sysselsättning, ålder, utbildning och transfereringar. Frågan om andra faktorer bidrar kan varken bekräftas eller avvisas med den här datan.
  • • Sysselsättningssiffrorna är genomsnitt. Enskilda personers utfall varierar kraftigt — och kohorter mottagna vid olika tidpunkter har olika arbetsmarknad att möta.
  • • Kalkylen fångar inte icke-monetära effekter (social sammanhållning, trygghet, kulturell mångfald) — varken positiva eller negativa. De ligger utanför en offentligfinansiell analys.

The question this page answers

Why is employment the single most important variable for public finances — and why is the employment rate alone not enough?

At a glance

Key figures — the most important numbers on this page

Employment gap 20–64

≈ 15–17 pp

SCB LFS 2024 — Swedish-born ≈85 %, foreign-born ≈68 %. Stable for 15 years.

Per pp, group of 100,000

≈ SEK 300–500 M

Rule-of-thumb: 1 pp × 100,000 = 1,000 more employed → median wage minus transfers.

Self-support rate, foreign-born

≈ 73 %

ESO 2023 (Eklund) — work income above the social-assistance norm, 20–64.

Establishment time, refugees

8–10 years

IFAU 2016; KI 2024 — time to 50 % employment from year of arrival.

Tax wedge, median earner

≈ 45 %

OECD Taxing Wages 2024 — Sweden near the OECD top.

Dependency ratio 2024

≈ 0.74

SCB — children + elderly per working-age person (15–64).

Underemployed

≈ 5 %

SCB LFS 2024 — want to work more but cannot.

Long-term unemployed

≈ 144,000

Public Employment Service Oct 2024 — registered >12 months.

What does the data show?

Objective observations — not interpretations

  • Group differences in KI's and Ruist's net contribution figures are explained almost entirely by the employment rate in their models. Other variables (age, education) matter — but employment dominates.
  • Sweden has a high tax level and generous transfers. That means a full-time employee pays in a lot and a non-worker receives a lot — the gap between contributors and recipients is therefore wider here than in low-tax countries.
  • The employment rate alone does not describe self-support. LFS counts one hour of work in the reference week as 'employed'; RAMS uses an annual income threshold; ESO's self-support measure requires work income above the social-assistance norm. The three measure different things.
  • Establishment time for refugees is about 8–10 years to 50 % employment (IFAU, KI). After the 2-year establishment programme, a large share continues on other public support rather than reaching independent income.
  • Other foreign-born (labour, EEA, studies) have high employment from the start because the residence permit requires it. Lumping them together with refugees/family hides the reality of both groups.

Method note
This page separates three different employment measures: SCB LFS (monthly sample survey, ILO definition, one-hour threshold), SCB RAMS/LISA (annual register data, annual-income threshold) and the ESO/Eklund self-support measure (work income above social-assistance norm). All three are correct — but they answer different questions. Fiscal calculations (KI, Ruist) use register data, not the LFS. Employment affects public finances through two channels: (1) the individual pays tax when working, (2) the individual does not receive social assistance / unemployment insurance / establishment benefit when working. The difference between the two (the tax wedge plus the transfer) is roughly SEK 300–500 k/year for a median earner. The models say nothing about why employment differs between groups — that is a separate question (education, language, discrimination, matching, cohort composition).

Definitions

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

Employed (LFS)
Person aged 15–74 who worked at least one hour as an employee, self-employed or unpaid family worker in the reference week, or was temporarily absent from such work. ILO definition.Counts one hour of work as 'employed' — good for international comparison, but does not capture whether the person actually supports themself.
Employment rate
Share of the employed in a population age group. The EU headline is 20–64; Swedish debate usually uses 15–74. Different denominators give different percentages even with the same population.
Labour-force participation
The sum of employed and unemployed as a share of the population 15–74. Captures those active on the labour market — regardless of whether they have a job. Sweden historically has a high level (~74 %).
Work income
Income from work (wages, business income) before tax. Distinguished from capital income and transfers. Based on tax returns in SCB LISA.
Self-support
ESO 2023 (Eklund): individual with annual work income above the social-assistance norm (≈ SEK 15,500/month in 2020). About 73 % of foreign-born 20–64 met the threshold in 2020, versus ~88 % of Swedish-born. Not official SCB statistics.
Tax wedge
The difference between what the employer pays and what the employee keeps after tax and social contributions. OECD Taxing Wages: ~45 % for a Swedish median earner 2024. Among the higher in the OECD; affects labour supply and employment in low-margin jobs.
Disposable income
Household income after tax and including transfers. Captures the household's economic reality — but mixes work income with benefits.
Public support (main income)
Person whose main income is a public transfer (social assistance, sickness benefit, activity compensation, establishment benefit, etc.). SCB LISA uses annual income data to determine what counts as main income.
Establishment
That a newcomer has reached own subsistence (work or studies with student aid) after a given number of years. Studies use different thresholds; 8 price base amounts per year is a common proxy.The 'establishment programme' is a 2-year Public Employment Service programme — not the same as establishment in the statistical sense.
Long-term unemployment
SCB/EU: unemployed in the LFS for at least 12 continuous months. The Public Employment Service uses 'registered openly unemployed or in activity-support programmes for at least 12 months' — a different population, often with different levels.
The work-first principle (arbetslinjen)
Political principle that public support should be temporary and that work should pay better than benefits. Implemented through earned-income tax credits, activity requirements, time-limited benefits. Not a statistical measure.
Dependency ratio
Number of persons in non-working age (0–14 + 65+) per person in working age (15–64). SCB 2024: ~0.74. Rises with demographic ageing — affects the sustainability of public finances independent of migration.
Tax capacity
Taxable income per inhabitant in a municipality. Forms the basis of the municipal equalisation system. Municipalities with low tax capacity (e.g. Malmö, Södertälje) receive transfers from those with high (e.g. Stockholm, Danderyd).

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

The LFS counts one hour of work in the reference week as 'employed'. Someone working 4 hours a week on an hourly contract counts — but cannot support themselves. The gap between the employment rate (~82 % for 20–64) and the self-support rate (~73 % for foreign-born per ESO) measures how many are weakly anchored on the labour market.

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