Europe's Self-Employment & Freelance Visas, 2026: Why 'Freelance Visa' Means Something Different in Every Country

"Freelance visa" is not one thing

Type "freelance visa" or "self-employment visa" into a search engine and you'll get a list of countries, a column of monthly income figures, and the strong impression that these programs are interchangeable — pick the cheapest threshold and go. They are not interchangeable. The number in that column is answering a different question in every country, and the question matters far more than the number.

Consider one person: a freelance software consultant with their own clients and, say, €4,000/month in net income. In France, they'd apply under the Passeport Talent's liberal-profession track, where the bar is essentially the national minimum wage — they clear it easily. In Norway, they'd have to restructure as a sole proprietorship (not a limited company) and prove the business itself generates NOK 341,373/year. In Germany, they'd discover there are two self-employment visas and they need the right one or they're refused. In Austria, income wouldn't be the test at all — the question would be whether they can transfer €100,000 of capital or create local jobs. In Denmark, an expert panel would judge whether their business idea is innovative enough, and even if it is, they might be turned away because the year's 75-permit cap is full.

Same person. Same laptop. Same clients. Five completely different gates. This piece maps the fifteen self-employment and entrepreneur programs we track across Europe by the model each one uses to decide whether you're in — because once you see the model, the country shortlist sorts itself. Jump to the model you're closest to, to the freelancer section, to the founder section, or straight to the full table.

The six models of "how much is enough"

Across the fifteen programs, the qualifying test falls into six families. Most programs sit primarily in one; a few combine two (income and a committee, for example).

1. Income / means-tested — you prove a recurring figure. This is the model most people picture. Portugal's D2 (€920/month for 2026), Spain's self-employed work visa (200% of IPREM, ~€1,200/month), France's liberal-profession Passeport Talent (a SMIC-pegged ~€21,900/year), Italy's Lavoro Autonomo (€8,400/year), Estonia's Startup Visa (€800/month), Norway's self-employed permit (NOK 341,373/year), and Poland's self-employed TRC (12× the provincial average wage) all run on a number.

2. Capital transfer — you move money in. Austria's Red-White-Red Card for Self-Employed Key Workers wants €100,000+ of capital transfer (or job creation, or technology transfer). France's economic-investor Passeport Talent wants €300,000 invested in fixed assets through a company you direct or hold ≥30% of.

3. Bank balance — you show savings you keep, rather than money you spend or transfer. Sweden's self-employed permit asks for SEK 200,000 of your own funds in the bank (SEK 100,000 more for a spouse, SEK 50,000 per child) to cover your first two years.

4. No fixed number — a viability test or local formula — Germany's freelance visa for liberal professions sets no statutory minimum at all. The Berlin immigration office instead applies a working formula: your rent, plus health insurance, plus a roughly €563/month subsistence floor. The "threshold" is whatever your own cost of living comes to. Germany's commercial self-employment route (§21 Selbständige) also carries no statutory figure — the old €250,000 capital test was abolished back in 2012, so today it turns on a case-by-case assessment of whether your business plan, financing, and projected local impact make the venture viable.

5. Points — the Netherlands' points-based self-employed permit routes you through the RVO (the enterprise agency), which scores you out of a possible 300 across personal experience, business-plan quality, and added value to the Dutch economy. You need 90 points with at least 30 in each category.

6. Panel / committee approval, sometimes with an annual cap — Denmark's Start-up Denmark runs your business idea past an expert panel before you can even apply for the residence permit, and grants only 75 permits a year. Estonia's startup committee makes a similar innovation judgment. Italy's Lavoro Autonomo sits behind the Decreto Flussi quota — only 500 self-employment slots per year (1,500 across 2026–28), released on a first-come "clickday."

Notice that the models sort applicants by what they have, not by what they do. A capitalized founder and a cash-flow freelancer are the same "self-employed" person to a search engine and completely different applicants to an immigration officer.

If you're a freelancer (keep your clients)

If your situation is "I have my own clients, I bill them remotely, I want to live in Europe and keep doing exactly that" — your natural home is the income/means-tested cluster, because those programs test the thing you actually have: recurring income.

The friendliest math is in Portugal. The D2 sets means-of-subsistence at €920/month for 2026 (annualized to about €11,040), pegged to the national minimum wage — the same basis the D7 uses, and a figure AIMA assesses at the rate in force on your appointment date, not when you started the file. It's framed as an entrepreneur visa, but solo professionals qualify; there's a recommended €5,000 of startup capital but no fixed investment minimum, per AIMA's primary guidance. Spain's self-employed work visa asks for 200% of IPREM — with IPREM at €600/month for 2026, that's about €1,200/month, plus 50% per dependent — and a business plan endorsed by a recognized self-employed-workers' association (such as ATA or UPTA; membership of a professional colegio is a separate step, only for regulated professions). France's liberal-profession Passeport Talent pegs its means test to the SMIC minimum wage (around €21,900/year — and note the SMIC itself rose twice in 2026), but since the June 2025 reform the deciding question is whether your activity is economically viable, not just whether you clear the wage line.

Two traps to watch in this cluster. Italy's Lavoro Autonomo looks gentle — about €8,400/year, or a documented engagement proposal from Italian clients — but it sits behind the Decreto Flussi quota: just 500 self-employment visas a year, released on a clickday where the slots can be gone in minutes. The income bar isn't the obstacle; the calendar is. And Norway's self-employed permit is demanding in a structural way: you must operate as a sole proprietorship (enkeltpersonforetak), not a limited company (AS), the work has to require skilled qualifications, and the business itself has to clear a real income bar — NOK 341,373/year in pre-tax profit (a figure UDI re-set on 1 September 2025, so older guides quoting NOK 325,400 are already stale). The legal form is the trap here: a consultant who has been billing through a limited company has to restructure before this visa is even on the table.

I'll admit a bias here: of all six models, I find the moving-benchmark income thresholds the most honest in principle. A figure pegged to a minimum wage or the SMIC is meant to rise with the local cost of living, so the bar means roughly the same thing year to year — when the peg is actually allowed to move (Spain's IPREM, awkwardly, has been frozen since 2023, which is its own kind of warning). The fixed-dollar figures you see in aggregator listicles — "$2,000/month!" — are the ones that quietly go stale, because nobody re-indexes them. When a country ties its threshold to its own wage statistic, it's at least being internally consistent.

If you're a founder (base your company)

If your situation is "I'm building or running a company and I want to base it in Europe" — the means-tested visas are often the wrong door, and the capital, points, and panel models are the right ones.

The capital-transfer route is the most transactional. Austria's Red-White-Red Card for Self-Employed Key Workers is explicitly not points-based: you qualify by transferring €100,000+ of investment capital, creating or securing jobs, transferring know-how or technology, or demonstrating significance to an Austrian region. France's economic-investor Passeport Talent wants €300,000 invested in fixed assets (tangible or intangible) through a company you direct or in which you hold at least 30% — purely financial, portfolio-style investments don't count.

The judgment-based routes ask whether your business is good, not just whether you're funded. The Netherlands' points-based permit scores experience, business-plan quality, and Dutch-economic value, needing 90 of 300 points. Estonia's Startup Visa sends you to a startup committee that decides, in about ten working days, whether your venture is genuinely tech-driven, scalable, under ten years old, and Estonian-registered. Denmark's Start-up Denmark is the most selective program in the entire set: an expert panel at the Danish Business Authority must approve your idea before you can apply for the SIRI residence permit, passive shareholders are explicitly excluded, and the country issues only 75 of these permits a year. Finland's entrepreneur permit runs a two-stage assessment (an Economic Development Centre opinion — the body that took over from the ELY Centres in the 2025 regional-administration reform — then the Migri permit) and, a trap many freelancers miss, explicitly excludes "light entrepreneurs" who invoice through services like UKKO or Eezy rather than running a registered business.

The pattern across the founder cluster: capital buys you a transactional path (Austria, France-investor), but the panel programs (Denmark, Estonia, Finland) are buying conviction in your business, and conviction is rationed — sometimes by a literal annual quota.

The full comparison (15 programs)

Every program below is one we track with a primary-source citation on its page. "PR" = years of continuous residence before permanent-residence eligibility. Thresholds are the figures verified in our visa data as of late May 2026; several are moving benchmarks (see the next section).

Put the table next to your own situation and a shortlist appears: a cash-flow freelancer reads the top six rows; a capitalized founder reads the middle three; an innovative-startup founder reads the bottom four. The visa comparison tool lets you line any of these up side by side, and the remote-work explorer ranks the underlying countries on the broader living factors the visa page can't (connectivity, cost, healthcare).

The Germany trap: two visas, one wrong choice

Germany deserves its own section because it is the clearest illustration of why the label fails you — and because it generates more confused searches than any other country in this set ("germany self-employment visa," "germany section 21," "§21 startup visa").

Germany has two self-employment visas, both living in §21 of the Residence Act (AufenthG), and they are not interchangeable:

• Freelance Visa — Freiberufler, under §21(5), is for the liberal professions (freie Berufe): lawyers, architects, engineers, IT consultants, doctors, journalists, artists, and similar. It has no fixed income minimum — the Berlin foreigners' office applies the formula model (your rent + health insurance + a ~€563/month living floor). It's the route most independent knowledge-workers actually want.
• Section 21 Self-Employment — Selbständige, under §21(1), is for commercial entrepreneurs running a trade (Gewerbe). There's no statutory figure here either — the old €250,000 capital benchmark (and the five-jobs rule that went with it) was scrapped in a 2012 reform, so you'll still see it quoted in stale guides but it is no longer the law. The office instead weighs qualitatively whether your business plan, financing, and likely economic impact make the venture viable.

The distinction is whether German law treats your work as a liberal profession or a commercial trade — a categorization that has nothing to do with how an English-speaking freelancer thinks about their own work. An "IT consultant" usually lands in Freiberufler; an "online shop owner" lands in Gewerbe. If you apply on the wrong track, you don't get gently redirected — you get refused, and you reapply.

This is the part I find genuinely under-explained everywhere else: the English word "freelancer" maps onto a German legal taxonomy that predates the gig economy by a century, and the burden is on you to know which side of the Freiberufler / Gewerbe line your work falls — before you book a Bürgeramt appointment. The German missions' own visa-service portal↗ and the federal Make-it-in-Germany↗ guide are the primary sources to read first; a local tax adviser (Steuerberater) confirming your category is cheap insurance against a wrong-track refusal.

Verify before you commit

Here's the brand-voice part, and it's load-bearing for this article in particular: almost every number above is a moving target.

The SMIC (France), the SMI (Spain's minimum wage, behind the digital-nomad threshold), Portugal's minimum wage (behind the D2), and Poland's provincial average wages all re-index — most of them every January, and France's SMIC moved twice in 2026. Spain's IPREM is the cautionary counter-example in the other direction: it's been frozen at €600/month since 2023 because no new national budget passed to lift it, so a benchmark you'd assume tracks inflation simply hasn't. Norway's income bar is reset by UDI on its own schedule (it jumped on 1 September 2025); Sweden's is a fixed krona sum. Austria's and France's capital thresholds are policy numbers a regulation can change — and Germany's commercial route has no number at all, just a case-by-case viability judgment that varies by Land. An aggregator that copied "€2,150/month" from another aggregator two years ago is now simply wrong, and it has no idea, because nobody re-checked the underlying instrument.

So treat the table as a map of models, not a quote. For the live figure on any program, open its page on this site — every threshold there carries the primary-source link and the date we last verified it — and then click through to the issuing government's own page before you commit a euro or a flight. The income tests are pegged to benchmarks that update; the capital tests are policy numbers that move; the panel programs can fill their annual caps mid-year. The model is durable. The number is perishable.

One more cross-check worth doing: confirm what kind of income each program will accept. We've written separately about how some visas exclude "speculative" or remote-employment income even when the headline threshold is met — the same scrutiny applies to self-employment income that's really disguised employment. And before you book anything, check where you're allowed to file from: several of these are consulate-from-origin programs, not apply-on-arrival ones.

Where to go next

If you're a freelancer keeping your clients, start with Portugal's D2, Spain's cuenta propia, or France's liberal-profession track, and read the Norway and Italy traps before you fall in love with their numbers. If you're a founder, decide first whether you're buying a transactional path (Austria, France-investor) or submitting to a judgment (Denmark, Estonia, Finland, the Netherlands), because those are different applications with different timelines. And if you're in Germany or thinking about it, settle the Freiberufler-versus-Gewerbe question before anything else.

Then put your two or three finalists into the visa comparison tool, check the full visa directory for the live thresholds, and rank the countries themselves on the factors a visa page can't show you in the remote-work explorer.

Don't trust a list. Read the model, not the label — and read the law, not last year's number.