From Finland to Germany: What Japan's Immigration Policy Could Learn From How Other Countries Legislate Integration

Based on 山脇啓造 (Yamawaki Keizo), "第62回 世界の移民統合法制と日本の選択" ("Vol. 62: Global Migrant Integration Legislation and Japan's Choice"), part of the "多文化共生3.0の時代" column series published by CLAIR (the Council of Local Authorities for International Relations), January 19, 2026 (source)

Most of what I cover on this blog is about who gets let in and under what conditions. This piece, by comparison, is about a question Japan's immigration debate barely asks: once someone is here, what is the state actually obligated to do about it? The real value of this piece, for me, is less its conclusion about Japan than the cross-national comparative study it builds to get there — a survey of how more than a dozen countries have actually legislated integration, with real statutes, real dates, and real consequences for non-compliance. Cross-cultural comparative work like this is something I find genuinely useful in its own right, so I want to walk through how each country has actually handled this before asking the more practical question: what, if anything, could Japan actually take from each of them?

The argument

Yamawaki's starting point is a number: Japan now has roughly 4 million foreign residents. His argument is that at this scale, integration stops being a temporary accommodation and becomes a question of national institutional design. He's explicit that he isn't calling for Japan to declare itself an "immigration country" — he frames this as a realistic response to a settlement pattern that has already happened, not an ideological statement. His concluding call is for Japan to move past managing foreign residents within the framework of the Immigration Control and Refugee Recognition Act, and toward a dedicated integration statute that treats foreign residents as members of society with a state obligation attached to their participation and social cohesion. To make that case, he doesn't start from Japan at all — he starts by mapping how other countries have already answered the same institutional question, each in its own way.

Four models, two axes

The framework Yamawaki uses (which he says he's reconstructed from academic sources, not drawn from any single official classification) plots integration regimes along two axes: how much support a state provides for migrants' social participation, and how much obligation it imposes on migrants in return. Most real systems land somewhere in between as hybrids, but he groups them into four rough types:

• Social inclusion — integration treated as social policy: guaranteed language education, employment support, and participation conditions, without tying those benefits tightly to migrant obligations. Canada's 1988 Multiculturalism Act obliges the federal government to actively promote equal participation, though the rights and support that flow from it are generally limited to permanent residents rather than all foreign residents. Australia is grouped with Canada as sharing the same institutional structure, which is also why Yamawaki folds the older "multiculturalism" category into this one instead of treating it as separate. Finland, under its 1999 Integration Act, is the exception that proves the rule is flexible: it extends integration support to non-permanent foreigners too, not just those already on a path to permanent status. Zooming out to the supranational level, the EU's 2004 Common Basic Principles for Immigrant Integration Policy defined integration as a "two-way process," and the Council of Europe has promoted a related "intercultural" approach since 2008 — both attempts to generalize this model across member states. Yamawaki also notes that Sweden has been drifting from this model toward the inclusive-integration type discussed next, though he doesn't cite a specific statute marking that shift.

  
  

• Civic integration — integration treated as the migrant's legal obligation, with language and civic-values compliance tied to residence status or permanent residence. The Netherlands passed the original Civic Integration Act in 1998, then tightened it with a 2007 amendment that linked pre-entry tests taken abroad, permanent residence, family reunification, and naturalization all to passing integration tests. Denmark's 2002 Integration Act goes further: welfare benefits and naturalization are both conditioned on passing tests, family reunification carries restrictive language and asset requirements, and applicants face what's described as a "Danish values" test. Francehas layered several laws on top of each other — the 2003 contrat d'accueil et d'intégration, a 2005 Social Cohesion Plan Act, a 2006 Immigration and Integration Act, and a 2016 contrat d'intégration républicaine — each one tightening the link between integration commitments and residence rights. Austria started with voluntary integration agreements in 2002 and moved to a 2017 Integration Act with real teeth: sanctions for non-compliance include benefit cuts and, in some cases, revocation of the residence permit itself.

  
  

• Inclusive integration — support and obligation combined, under what Germany calls "Fördern und Fordern" (support and demand) — the idea that the state's support and the migrant's obligation are two sides of the same coin rather than opposing poles. Germany's 2005 Residence Act created a legal right to attend integration courses, and its 2016 Integration Act linked those courses to vocational training and made municipalities statutorily responsible for providing them; participation becomes mandatory in some cases, with benefit cuts or consequences for residence status for refusing. South Korea is placed in the same category but is, in Yamawaki's words, the more moderate version of it: its legal basis is the 2007 Basic Act on Treatment of Foreigners and the 2008 Multicultural Families Support Act, operationalized through the 2009 Social Integration Program, which is voluntary and carries no penalty for skipping it — the incentive runs the other way, with graduates getting exemptions from tests or interviews when they later apply for permanent residency or naturalization.

  
  

• Market-based — no comprehensive integration law at all; integration left to labor markets, schools, and civil society to sort out. The United States has no federal integration statute; Yamawaki notes federal refugee resettlement support has been shrinking and that English-language instruction access is uneven from state to state, depending entirely on local funding and policy choices. The United Kingdom is similar in kind: there's no integration law, and funding for ESOL (English for Speakers of Other Languages) programs was cut sharply during the 2010s. Yamawaki points to the 2001 Bradford riots and the 2011 London riots as the cautionary evidence for this model — instances he reads as what happens when integration is left to drift with no institutional backbone underneath it.

  
  

Where Japan actually sits

A few things stand out once you line all four models up side by side. Every single one of them ties something concrete to the migrant's legal status — a benefit, a residence permit, a path to naturalization — so the real divide isn't support versus no support, it's whether that link runs through tests and sanctions (civic integration, inclusive integration) or stays a softer expectation (social inclusion, market-based). And almost none of these eleven countries is a pure type: Sweden is apparently sliding from social inclusion toward inclusive integration, Korea sits in the inclusive-integration column but behaves more like social inclusion in practice since it drops the penalties, and the US and UK aren't really a fourth "model" so much as the absence of one. Yamawaki's point is that this hybrid, shifting quality is normal for countries that have actually legislated on this — and that Japan, by comparison, hasn't picked a lane at all.

Yamawaki's read is that Japan's existing municipal-level efforts — Japanese-language education, support for children's learning, medical interpretation, local exchange programs — already resemble the social inclusion model in substance. What's missing is the law underneath them. None of this is a national legal obligation; it runs on subsidy programs (補助金事業) and local government initiative, which means the services that exist depend on budget cycles and which municipality someone happens to live in, not on an entitlement anyone can point to.

What Japan could actually take from each model

If I set aside the "which model is best" question and ask what's actually transferable, a few things stand out. From the social inclusion countries, the lesson isn't the values — Japan's municipalities already operate almost exactly this way, with language classes and child-support programs — it's the legal backbone underneath them. Finland's choice to extend support to non-permanent residents, not just those on a path to permanent status, is directly relevant here too: most of Japan's long-term foreign residents aren't permanent residents either, so a Finland-style law would actually reach the population Japan already has.

From the civic integration countries, I'd take the caution more than the method. The Netherlands and Denmark show that once a test is written into law, it tends to migrate — from a landing requirement to a residence requirement to a naturalization requirement to a family-reunification requirement. Japan is already adding language and civics elements into its new permanent residency guideline, so this group reads less like a model to adopt and more like a preview of where that kind of testing regime can end up if it isn't bounded early.

Germany's "Fördern und Fordern" pairing — support tied to a genuine, enforceable obligation — strikes me as the most honest version of what Yamawaki is actually proposing for Japan: legislate both the support and the obligation together, so neither side can claim the other owes it nothing. South Korea's version of that same model, built on incentives rather than penalties, is probably the more realistic starting point for Japan specifically: it's closer to the voluntary, program-based way Japan already does things, and it still gets measurable buy-in without a politically difficult sanctions regime.

The market-based countries are the clearest warning. Japan's current setup — language classes and integration support running on local subsidy programs with no statutory guarantee — is structurally closer to the US and UK's "no law" model than to any of the three model-countries Yamawaki is actually calling on it to emulate. The lesson from the Bradford and London riots isn't that Japan is headed for unrest; it's that a system this dependent on grants and goodwill is exactly the kind of system that erodes quietly until something forces the question.

I'll say upfront that this is one academic's argument, not a government position — and it's worth noting that the piece names foreign laws (Canada's 1988 act, Korea's 2007-2009 statutes, Germany's 2016 law) without citing the actual statute text, so I'd treat the comparative framework as a useful lens rather than a verified legal survey.

What strikes me is the direction it's pointing in, set against everything else I've been tracking this year. The permanent residency revocation guideline, the new compliance question going onto the entry card, the talk of a possible population cap — all of that is about tightening the terms on which people get in or get to stay. Yamawaki's proposal runs the opposite way: it's an argument for building out obligations the state owes to the roughly 4 million people who are already here, on the theory that a country this deep into settlement doesn't get to treat integration as optional just because it never said yes to being an "immigration country" out loud. Whether or not Japan ever legislates something like this, the gap he's describing — a lot of local practice, no national law backing it — is real, and it's a different kind of gap than the ones I usually write about here.

Source: CLAIR (公益財団法人自治体国際化協会), "第62回 世界の移民統合法制と日本の選択," column by Yamawaki Keizo, January 19, 2026 (clair.or.jp).