Långsiktiga välfärdskostnader
Långsiktiga välfärdskostnader — utveckling över tid per kohort
Sidan fokuserar på tid, livsfaser och generationer. Den kompletterar/totalkostnad, som förklarar metodfältet. Registerbaserad forskning och långsiktsmodeller — Ruist (ESO 2018:3), Aldén & Hammarstedt, KI Specialstudie 117, Långtidsutredningen — visar hur offentliga netton utvecklas över decennier för olika kohorter.
1980–1989
Tidigare kohorter visade snabbare sysselsättningskonvergens; offentligfinansiellt nettoutfall förbättrades efter ~10 år.
1990–1999
Längre tid till sysselsättning. Fiskalt nettoutfall stabiliseras kring –50 000 till –70 000 kr/år enligt Ruist (tvärsnittsvärde, modellerat).
2000–2015
Bibehållet negativt netto efter 15 år för kohorten som helhet. Stora inomkohortskillnader beroende på födelseland, utbildning och sysselsättning.
2015–
Mottagningstoppen 2015. Registerdata visar snabbare sysselsättning än 90-talskohorten vid motsvarande vistelsetid, men fortsatt nivåskillnad mot inrikes födda. Långsiktsprognosen är känslig för sysselsättningsantagandet.
The question this page answers
How are public welfare costs and revenues calculated over time, per generation and for different population groups — and which assumptions drive the results?
At a glance
Key figures — the most important numbers on this page
Annual fiscal net effect
≈ SEK –74,000
Ruist (ESO 2018:3) — cross-section per refugee per year. Modelled value, denominator: the entire refugee population 2015.
Life-cycle net (NIER)
≈ –74 %
NIER Special Study 117 — life-cycle present value for the 2015 refugee cohort, share of the Swedish-born equivalent net. Modelled.
Discount rate (NIER/ESO)
3.0 %
Real discount rate used in main scenarios. Sensitivity analyses at 2 % / 4 %.
Dependency ratio 2070
≈ 90 %
SCB Population Projection 2024 — persons 0–19 + 65+ per 100 aged 20–64. Rising from ≈ 76 (2024).
Individual public consumption
≈ 25 %
SCB National Accounts — health, care and education directed to identifiable individuals, share of GDP.
Collective public consumption
≈ 27 %
Defence, judiciary, administration — harder to allocate per individual in life-cycle calculations.
Ageing costs EU 2070
≈ +9 %
EU Ageing Report 2024 — projected rise in age-related public spending as share of GDP.
Self-support threshold
≈ SEK 15,300/mo
ESO 2023 (Eklund) — work income above the social-assistance norm. Used as outcome measure in life-cycle models.
What does the data show?
Objective observations — not interpretations
- Long-term welfare costs are almost always calculated as model-dependent results, not exact forecasts. Every figure results from assumptions about employment, wages, taxes, transfers, public consumption, productivity, longevity and discount rate. Two studies with different assumptions can reach very different results for the same population.
- Most people are a public net cost during childhood and old age, and a public net contributor during parts of working life. The central question is therefore not whether a group uses welfare, but the relation between tax payments, employment, transfers, education, health, care, pensions and longevity.
- Cross-section calculations (Ruist ESO 2018:3) measure the public net in a given year. Life-cycle calculations (NIER Special Study 117) sum over a person's whole life in present value. Generational calculations (Long-Term Survey, Fiscal Policy Council) also cover future generations and their debts/assets. The methods answer different questions.
- Collective public consumption (defence, judiciary, central administration, interest) is hard to allocate per individual. Different studies make different choices: equal per capita, by another key, as a fixed cost, or excluded from marginal analysis. This choice can shift the result by tens of thousands of kronor per person per year.
- Group averages do not describe individuals. A long-term net result for a group ('refugees' or 'Swedish-born') hides large within-group variation. A calculation showing SEK –500,000 as life-cycle net for a group does not mean every member has that outcome — the spread can be several million kronor.
Method note
The page strictly separates (1) public-fiscal effect — impact on state, municipal and regional budgets — from (2) macroeconomic effect — impact on GDP, productivity, consumption and labour supply. A life-cycle calculation showing a negative fiscal net for a group may simultaneously be macroeconomically positive (e.g. by relieving labour shortages) or vice versa. The calculations also do not capture social, humanitarian or cultural effects — they measure only money. Underlying differences in discount rate (2 %, 3 %, 4 %), time horizon (lifetime, 30 years, generations), employment assumptions (current state, convergence, alternative paths), and treatment of public assets (pension system, infrastructure) explain most of the divergence between studies. No single calculation is 'right' — several scenarios must be compared. Projections over 40–60 years (EU Ageing Report, Long-Term Survey) carry wide uncertainty intervals and always report sensitivity analyses.
Definitions
How the numbers are counted — and what they do not cover
- Welfare cost
- Public expenditure on welfare services (health, care, education) and transfers (pensions, disability, benefits) directed to an individual or group. Normally calculated net of taxes and fees.
- Public consumption
- Government, municipal and regional purchases or own production of services — mainly health, care, education, defence and judiciary. Distinguished from transfers and public investment.
- Individual public consumption
- Public consumption attributable to an individual — mainly health, care and education. Allocated in life-cycle calculations based on actual use by age, sex and need.
- Collective public consumption
- Public consumption not attributable to an individual — defence, judiciary, central administration, general infrastructure. Requires methodological choice in life-cycle calculations (per capita, marginal or excluded).
- Transfer
- Income transfer from state or municipality to households without a counter-service — pensions, disability, income support, housing benefit. Distinguished from public consumption (which is service production).
- Tax revenue
- Government, municipal and regional income from direct taxes, indirect taxes, employer contributions and capital taxes. Forms the revenue side of the fiscal calculation.
- Net contribution
- Tax revenue from an individual or group minus transfers and public consumption directed to the same group. Positive net = the individual finances more than they use; negative = the opposite.
- Life-cycle profile
- Calculated average net contribution by age — negative in childhood and old age, positive in working life. Built from register data on actual use and income by age.
- Generational calculation
- Forward-looking calculation estimating how much each future generation will pay to and receive from the public system, given current rules and demographic projection.
- Cohort
- Group with a common starting point — usually birth year or year of immigration. Followed forward in time. Distinguished from cross-section (one year, all ages simultaneously).
- Discounting
- Method to make future kronor comparable to today's by scaling them down with an interest rate. One krona in 30 years is worth about 0.41 kr today at a 3 % real rate.
- Present value
- The discounted value of a future payment stream expressed in today's money. Used to sum life-cycle nets across decades.
- Discount rate
- The interest rate used in discounting. Standard values: 2 % (cautious), 3 % (NIER, Long-Term Survey), 4 % (market rate). The choice strongly affects long-term nets.
- Demographic dependency ratio
- Number of persons 0–19 and 65+ (dependent) per 100 aged 20–64 (supporting). Rising from ≈ 76 (2024) to ≈ 90 (2070) per SCB projection.
- Fiscal sustainability
- Balance between future public revenue and expenditure, given current rules and demographic path. Measured via the S1/S2 indicator (European Commission) — how much today's tax ratio needs to shift for the system to hold long-term.
- Primary financial balance
- Public sector balance excluding interest income and payments. Used in long-term calculations to separate demographic/structural effects from interest-cost effects.
- Sensitivity analysis
- Systematic variation of key assumptions (employment, discount rate, productivity) to show how much the result changes. Should always be reported in serious long-term calculations.
- Baseline scenario
- The calculation's main path — the scenario the author considers most likely given current information. Distinguished from alternative scenarios (optimistic, pessimistic, sensitivity tests).
- Alternative scenario
- Calculation path with changed assumptions — e.g. higher employment, faster integration, changed tax rules. Highlights how sensitive the baseline is to the assumptions.
Primary sources
Agencies and research institutions behind this page
7
agencies/institutions
8
reports & studies
4
primary datasets
Frequently asked questions
Short answers to what is most often discussed
Related topics
Logical next steps if you want to understand the background
Cost of immigration
The methodological field — Ruist, NIER and the foundations of fiscal calculations.
Work decides
Why employment is the central variable in all long-term calculations.
Tax burden
The revenue side of the public-fiscal calculation.
Welfare dependency over time
The transfer side across cohorts and time.
Enforcement gap
Population composition affects long-term calculations.
Where does the money go?
Where public expenditure actually lands in the state budget.
Sources for this page
The data comes from official primary sources.
KI · Irregular
Specialstudie 117 — Invandrades nettobidrag till de offentliga finanserna 1983–2022Nettobidrag till offentliga finanser per person och år, efter grupp (endast KI:s egna beräknade värden)
Latest data point · Source published