You notice it first as a scheduling problem. Someone in the hiring loop has found the right candidate, the candidate lives in a country where the business has no presence, and the question of how to actually pay them arrives on your desk with a start date already attached. Nobody set out to expand internationally. They set out to hire one person.
That gap between the hiring decision and the employment structure is where most international teams get built. Rarely from a strategy document. Usually from a run of individual hires that each made sense on their own terms, until the map of where people sit stops matching anything anyone planned.
The entity question arrives before anyone is ready for it
The sequence tends to go the same way. Recruitment finds someone good. Finance asks how payroll will run. Legal asks what exactly the business is signing and who is on the hook for it. And somewhere in that exchange, a sentence appears about needing a company in that country, usually said with more confidence than evidence behind it.
Some of the time that is right. A lot of the time it is a reflex. The practical reading on hiring globally without a local legal entity tends to start further back, with what the entity is actually for, because an entity is not a hiring tool. It is a legal person you have created and now have to maintain, and hiring happens to be one of the things it lets you do.
Once you separate those two ideas, the decision gets easier to hold. You are choosing a structure, and the number of people you want to employ is only one input into it.
What a local entity actually commits you to
Setting one up is the part people price. Running one is the part they underestimate.
An entity usually brings with it some combination of the following, in shapes that vary by country:
- Incorporation and a registered address that has to be real
- Local directors or a legal representative, sometimes resident
- Corporate tax registration and periodic filings
- Statutory accounts, sometimes audited, on the local calendar
- Payroll registration with the tax authority and the social insurance bodies
- A local bank account, opened under local identification rules
- Employment policies and documentation in the local language
None of that stops once the hire is made. It recurs, on local deadlines, in a language someone has to read. And closing an entity is a process with its own filings and its own waiting period, which matters more than people expect, because the countries you enter early are not always the countries you stay in.
You can feel the weight of it about a year in, when a small filing deadline in a country with one employee lands on the same week as something material at home. The entity does not care which one is more strategic.
Three routes, and the questions that separate them
There are broadly three ways to put a person in another country on a proper footing.
You employ them through your own local entity. You employ them through a third party that acts as the legal employer in that country while you keep day to day direction of the work. Or the relationship genuinely is a contracting one, and you engage them as a contractor, often with an agent of record handling the agreement, the classification review and the payments.
The useful questions are not about cost first. They are about duration, concentration and control. How long do you expect to have people there, five months or five years? Is this one person or the beginning of a function? Does the role need the kind of direction that only fits an employment relationship? Would you be comfortable explaining the arrangement to a labour inspector who already thinks they know the answer?
Teams that ask those four questions before they price anything tend to arrive at a structure they can defend. Teams that price first tend to arrive at one they have to keep re-explaining.
Where employer of record arrangements actually sit
This is the part that gets flattened in most summaries, and flattening it causes problems later.
Employer of record is not a single statutory category that exists the same way everywhere. It is a commercial description of an arrangement, and what sits underneath it depends entirely on the country. Some jurisdictions have purpose-built rules for it. In much of Europe the arrangement runs through licensed staffing or temporary work agency frameworks, which carry their own conditions. Germany is the clearest example, where the licensed temporary work agency regime under the AüG governs how a third party may supply a worker to another business, with licensing requirements and expectations about equal treatment attached.
Several countries also place limits on how long that kind of arrangement can run for the same person in the same role, or require that the worker be treated no less favourably than a comparable direct employee. That is not a reason to avoid the route. It is a reason to ask, country by country, what licence or framework the arrangement is operating under, and to get an answer with a name in it.
Vague answers here are informative. If nobody can tell you which regime applies in a given country, you have learned something about the arrangement.
The contractor route, used honestly
Contracting gets treated as the shortcut, which is unfair to it and dangerous for the people who use it that way.
There are relationships that genuinely are contracting relationships. The person works for several clients, sets their own hours, uses their own equipment, carries their own commercial risk, and delivers against a defined scope rather than sitting inside your management line. That is a real category, and engaging someone on that basis is a legitimate choice rather than a compromise.
What gets businesses into trouble is using it as a cheaper version of employment. Authorities almost everywhere look at the substance of the relationship rather than the label on the agreement, and the markers they look for are not subtle. Fixed hours. A single client. Line management. A laptop you issued. Work that looks exactly like the work your employees do.
An agent of record arrangement helps on the administrative side, keeping agreements consistent, running a classification review before engagement, handling payment across currencies and closing the relationship cleanly. What it cannot do is make a contracting label fit an employment relationship. That part is decided by the facts.
What the entity-free route does not take away
It is worth being precise about the boundary here, because the marketing around global hiring tends to blur it.
You remain the business directing the work, which means management practice, working time expectations and conduct issues stay yours. Intellectual property assignment still has to be handled properly in the agreement chain, and the rules on what an employer can claim vary. Data protection obligations follow the person, not the structure. Immigration and right to work do not soften either, since an employment arrangement is not work authorisation and never has been.
Permanent establishment exposure also remains a question that has to be assessed on the facts of what the person actually does, how they present themselves and whether they conclude contracts on your behalf. Nobody can tell you in the abstract how that lands for your business, and anyone who says the question disappears is selling you something. That one goes to your own tax and legal advisers, with the real details in front of them.
Termination is the other place people get surprised. Notice periods, severance entitlements and process requirements stay local, and in a fair number of countries they are considerably more protective than whatever the hiring manager is used to.
Deciding by the shape of the commitment
The clearer framing on when to use an employer of record treats it as a question about duration and commitment rather than headcount, which is why two businesses with the same number of people abroad often land in different places quite reasonably.
A team hiring two specialists in three countries for work that could move is in a different position from a team building a permanent engineering function in one city. The first is buying optionality. The second is eventually going to want the control, the cost base and the employer brand that comes with being the direct employer, and will set something up when the volume justifies the overhead.
Most businesses end up running a mix, and the mix is usually a better sign of judgement than consistency would be. Own entities where the presence is deep. Third party employment where you are still learning the market. Genuine contracting where the work really is project shaped.
The thing you start watching for, after a couple of years of this, is whether the structure still matches what the team has become. It drifts quietly. A country you entered for one contractor now has four employees and a manager, and nobody has revisited the arrangement since the first offer letter went out.
