Global employment solutions: a taxonomy of EOR, PEO, global payroll, and contractor of record

Abstract

Over the past two decades, businesses hiring across borders have increasingly turned to third-party intermediaries to handle the legal and administrative complexity of employing people in other countries. Four categories of service now dominate this market: Employer of Record (EOR), Professional Employer Organization (PEO), Global Payroll, and Contractor of Record (COR). The problem is that all four terms get used interchangeably, by vendors, advisors, and sometimes regulators, despite describing arrangements with meaningfully different legal structures, use cases, and risk profiles. This paper proposes a formal taxonomy to distinguish the four. Drawing on regulatory definitions, the International Labour Organization’s (ILO) framework for employment relationships, academic literature on platform work and worker classification, and over a decade of direct observation in the global employment solutions market, the taxonomy is built around four dimensions: (1) legal employer or engagement status, (2) entity requirement, (3) jurisdictional portability, and (4) co-employment or shared-liability structure. A practitioner decision framework is also offered.

This is a conceptual paper. Testing the taxonomy empirically across jurisdictions is beyond the scope of this work and is an explicit priority for future research.

Keywords: employer of record, professional employer organization, global payroll, contractor of record, international HRM, taxonomy, employment outsourcing, worker classification, misclassification, triangular employment, cross-border hiring

JEL classification: M16 (International business administration), M54 (Labor management), J41 (Labor contracts)

Defined Terms

The paper’s central argument is that imprecise vocabulary causes compliance failures. It therefore needs to model precise vocabulary itself. The definitions below establish this paper’s terminological basis. They do not claim to reflect universal legal definitions, which vary by jurisdiction, but rather to establish the consistent usage this paper applies throughout. Where a definition differs materially from common vendor usage, that is noted.

Employer of Record (EOR): the third party that assumes full legal employer status under local law in the worker’s jurisdiction. The client has no direct employment relationship with the worker. In this paper, EOR refers specifically to arrangements where the intermediary is the sole legal employer. Services marketed as “global PEO” are treated as EOR arrangements in this paper unless co-employment is legally operative in the relevant jurisdiction.

Professional Employer Organization (PEO): a firm that enters co-employment with the client, with both parties holding employer status simultaneously, with responsibilities divided by a written Client Service Agreement. In this paper, PEO refers to arrangements where co-employment is legally recognised and operative. This is largely confined to the United States. The term is not used here to describe international EOR arrangements, regardless of how vendors label them.

Global Payroll: the outsourced administration of payroll operations across multiple jurisdictions. The client remains the full legal employer. No employment relationship transfers to the provider. This is an administrative consolidation service, not an employment intermediary.

Contractor of Record (COR): a third party that formally holds the commercial engagement with an independent contractor on the client’s behalf. No employment relationship exists between any party. This paper treats COR as a distinct construct from Agent of Record (AOR), which has separate origins in insurance and staffing markets and serves different purposes.

Co-employment: a shared employer arrangement in which two parties, typically a PEO and a client, simultaneously hold legal employer obligations divided by written agreement. Not a universally recognised legal concept; absent from most jurisdictions outside the United States.

Misclassification: engaging a worker as an independent contractor when the working relationship meets the legal definition of employment under applicable law, regardless of contractual labelling.

Triangular employment relationship: an arrangement in which one entity formally employs a worker while a second entity directs that worker’s daily activities (ILO, Employment Relationship Recommendation No. 198, 2006).

1. Introduction

When a company wants to hire someone in a country where it has no legal presence, it hits a wall. Employment law in most jurisdictions requires a local employing entity. Setting one up takes months and costs real money. For workers engaged as independent contractors rather than employees, a different set of legal obligations applies, and the line between contractor and employee is drawn differently in almost every country.

A market of intermediary services has grown up to solve these problems. Four models now dominate that market: Employer of Record, Professional Employer Organization, Global Payroll, and Contractor of Record. All four are widely used. All four are regularly confused with one another, sometimes in ways that create serious legal exposure for the companies using them.

Vendors routinely market “global PEO” services that are legally EOR arrangements. Payroll providers sell services that overlap with what EOR providers do. COR is sometimes called EOR-lite, or lumped in with agent of record models that work differently. Advisory content bundles all four under “global employment solutions” without explaining what separates them or when each one actually applies.

This is not just a branding problem. A company that signs up a Global Payroll provider expecting it to act as the legal employer in a new market, without realising that Global Payroll requires an existing local entity, has a compliance gap it does not know about. A company that puts workers on contractor agreements in a country where those workers legally qualify as employees is carrying misclassification liability, regardless of what the contract says. Carré and Wilson (2004) document state-level tax losses from misclassification running to tens of millions of dollars per jurisdiction, with unpaid obligations falling on governments and misclassified workers alike. A COR does not eliminate that risk, but it reduces it by bringing proper structure and documentation to the contractor relationship.

The academic IHRM literature has produced solid frameworks for multinational workforce management (Brewster et al., 2007; Collings et al., 2019), non-standard employment (ILO, 2016), and HR outsourcing (Cooke et al., 2005). Researchers studying the gig economy have examined the employee-versus-contractor boundary in depth (De Stefano, 2015; Stewart and Stanford, 2017). But none of this literature treats EOR, PEO, Global Payroll, and COR as separate constructs with distinct definitions and boundary conditions. One market research firm estimates the EOR market at USD 4.42 billion in 2023, growing to USD 8.59 billion by 2030 at a 6.8% compound annual rate, though forecasts vary considerably across research providers (Verified Market Research, 2024). The market for contractor engagement intermediaries has grown alongside it. Neither has received the definitional clarity it needs.

This paper tries to provide it. It defines each model precisely, identifies the dimensions that separate them, and builds a decision framework practitioners can use when choosing between them. This paper does not evaluate specific vendors, does not offer legal advice, and does not make empirical claims. The goal is to lay conceptual groundwork that future applied and empirical research can build on.

Note: nothing in this paper constitutes legal advice. How EOR, PEO, global payroll, and COR are characterised legally varies by jurisdiction. Anyone making decisions in this space should work with local legal counsel.

2. Literature Review

2.1 Non-standard employment and triangular relationships

The clearest academic framework for understanding third-party employment comes from labour law, specifically the ILO’s concept of the “triangular employment relationship.” The structure is simple: one entity formally employs a worker, a second entity directs that worker’s day-to-day activities, and the worker sits in the middle, formally bound to one party while practically managed by another (ILO, Employment Relationship Recommendation No. 198, 2006).

This is the basic architecture of both EOR and PEO arrangements. The difference lies in how the employer role is divided. Davidov (2004) points out that when employer functions are split across parties, it becomes genuinely unclear which party carries which obligations, and workers often end up with less protection than either party intended. The ILO has flagged this repeatedly in its work on non-standard employment (ILO, 2016), particularly in countries where regulation has not kept pace with how these models actually work in practice.

2.2 HR outsourcing in international contexts

The HR outsourcing literature is also relevant here. Cooke et al. (2005) map a spectrum from transactional outsourcing, running payroll and administering benefits, to more strategic outsourcing where the provider takes on meaningful HR functions. EOR, PEO, Global Payroll, and COR occupy different positions on that spectrum. More importantly, they answer the two fundamental questions of international hiring, “do we need a local entity?” and “is this worker an employee or a contractor?”, in entirely different ways. Brewster et al. (2007) document the persistent tension in multinational HR between global consistency and local legal compliance. That tension is precisely why these four models exist and why they differ from one another. But the IHRM literature has not treated the four as separate constructs with distinct definitions.

2.3 The employer of record in emerging literature

The EOR model has been getting more academic attention. The World Economic Forum (2023) describes it as a structural innovation that gives smaller businesses access to international hiring without the cost and delay of entity establishment. Some countries have started formalising EOR-adjacent arrangements: the Netherlands has created a statutory framework for certain third-party employment structures, and the IRS in the United States has long maintained regulatory definitions distinguishing PEO co-employment from other third-party payer arrangements (IRS, 2024).

Dahan (2025) is among the first law review-level treatments of EOR, comparing frameworks across more than a dozen jurisdictions and mapping the inconsistent patchwork of domestic leasing laws, PEO regulations, and service intermediation rules within which EOR providers operate. This paper builds on that work by situating EOR within a four-construct taxonomy alongside PEO, Global Payroll, and COR, none of which Dahan addresses as distinct constructs in their own right.

2.4 Worker classification and contractor engagement intermediaries

To understand COR, two pieces of background are needed: the labour law work on worker classification, and the scholarship on platform work and the gig economy.

Classification, specifically who counts as an employee and who counts as an independent contractor, is among the most litigated questions in employment law. The OECD’s position is consistent: status is determined by how the relationship actually works, particularly the degree of control, whether the worker can substitute someone else, and whether there is mutual obligation, not by what the contract calls the person (OECD, 2019). The US, UK, Germany, France, and Australia have all developed their own classification tests, and those tests regularly produce different results for the same factual scenario (Hogan Lovells, 2023).

Platform work has made this messier. De Stefano (2015) noted that gig workers do not fit neatly into either category: they are not employees in the traditional sense, but they are not genuinely independent the way independent contractor status implies. Courts have started responding. The UK Supreme Court’s finding that ride-hailing drivers qualified as workers entitled to minimum wage and holiday pay is one example of a broader pattern where operational reality is overriding contractual labelling (Stewart and Stanford, 2017).

For companies hiring contractors across multiple countries, the regulatory variation creates what Carré and Wilson (2004) call a misclassification risk landscape: the same arrangement that is legally fine in one jurisdiction may be illegal misclassification in another. The COR model addresses this by inserting a specialised intermediary that checks classification eligibility, drafts compliant commercial contracts, handles cross-border payments, and keeps the documentation that regulators want to see. It does not change the nature of the relationship. It makes a genuinely independent contractor engagement easier to prove and manage.

3. Proposed Taxonomy

3.1 Methodology

This paper uses a conceptual taxonomy approach consistent with Doty and Glick’s (1994) typology criteria. The goal is not to generate empirical findings but to establish the definitional foundation that applied and empirical research can build on.

Scope definition. The taxonomy covers third-party employment and contractor engagement intermediaries used for cross-border hiring. Explicitly out of scope: staffing and recruitment agencies (which source workers but do not assume ongoing employer obligations), umbrella companies as used in UK domestic contracting (a distinct domestic structure), Agent of Record models in insurance and staffing (different origins, different legal function), and platform-mediated gig work as the primary subject of study (addressed here only as context for worker classification debates).

Construct identification. The four constructs were selected because they are the four commercially dominant intermediary models with meaningfully different legal structures and compliance implications. Other intermediary types exist but either overlap substantially with these four or occupy niche markets that do not yet warrant separate construct-level treatment. The goal was a parsimonious taxonomy, not an exhaustive catalogue.

Dimension selection. Four dimensions were selected because they are the ones that most directly determine: (a) what compliance obligations transfer to the intermediary, (b) what the client remains legally responsible for, and (c) whether a model is even legally applicable in a given scenario. Other dimensions exist, including cost structure, service scope, and technology platform, but they do not do the same definitional work as the four selected here. The dimensions were refined iteratively against practitioner cases and regulatory documentation until each construct occupied a distinct position on each dimension and no dimension was redundant.

Validation approach. The taxonomy is presented as a conceptual framework. Whether the four constructs and four dimensions adequately capture real-world practitioner decision-making across jurisdictions is an empirical question deferred to future research, as noted in Section 6.

Terminological note. Two of the four constructs draw on definitions already established in regulatory sources: PEO is defined by IRS statute and NAPEO guidance; the triangular employment relationship is formalised in ILO Recommendation No. 198. The other two involve definitional contributions by this paper. Global Payroll as a formal construct distinct from EOR is clarified here by the explicit statement that it is an administrative consolidation service that does not transfer employer status, a distinction that vendor marketing consistently blurs. COR as a formally defined construct distinct from AOR and from EOR-lite descriptions is introduced here as a contribution of this taxonomy. The four-dimension framework itself, applying legal employer status, entity requirement, jurisdictional portability, and co-employment structure as the four boundary-defining axes, is proposed in this paper.

3.2 Dimension definitions

The taxonomy uses four dimensions:

• 1. Legal employer or engagement status: who is the legally recognised employer of the worker, or, in contractor arrangements, who formally holds the commercial engagement?
• 2. Entity requirement: does the client need its own legal entity in the worker’s jurisdiction?
• 3. Jurisdictional portability: to what extent does the model’s legal structure function across different jurisdictions? This is a property of the underlying legal doctrine, not of market reach. High portability means the structure functions wherever the intermediary can establish a local presence. Low portability means the structure depends on legal concepts absent from most jurisdictions.
• 4. Co-employment or shared liability structure: is employer responsibility shared between intermediary and client, held entirely by the intermediary, or absent because no employment relationship exists?

These four dimensions are not the only ways these models differ, but they are the ones that most directly determine compliance obligations, liability exposure, and legal applicability in a specific situation.

3.3 Employer of Record (EOR)

Definition: An employer of record is a third party that assumes full legal employer status for a worker on behalf of a client company. Under local law in the worker’s country, the EOR is the formal employer of record. The client retains operational direction: it sets tasks, manages performance, and makes hiring and termination decisions. The legal infrastructure of employment, contracts, payroll, tax obligations, statutory benefits, and compliance, sits with the EOR.

How it works in practice:

The EOR has, or establishes, a legal entity in the worker’s country. It issues the employment contract, runs payroll, handles tax withholding and remittance, administers statutory benefits, and maintains compliance with local labour law. The client has no direct legal employment relationship with the worker. What the client has is a commercial services agreement with the EOR.

The EOR is the formal employer under local law. The client retains operational direction and may bear certain liabilities, including workplace safety and anti-discrimination obligations, depending on jurisdiction. Country-specific legal advice is always required.

Where it fits: EOR works when a company wants to hire an employee in a country where it has no entity and is not planning to open one, whether that is because the team is small, the market is a test, or the business model is permanently distributed. Some countries restrict this: Germany’s Arbeitnehmerüberlassungsgesetz (AÜG), amended in April 2017, limits how long an individual worker can be leased to the same hirer to 18 consecutive months, after which continuing the arrangement risks being treated as having created a direct employment relationship (AÜG, §1 Abs. 1b). Country-specific legal advice is always necessary.

Common failure mode: EOR selected without checking for jurisdiction-specific leasing restrictions, or used as a permanent staffing solution in markets like Germany where duration caps apply. In Germany, engagement beyond the 18-month AÜG limit risks creating a deemed direct employment relationship and associated back-liability.

3.4 Professional Employer Organization (PEO)

Definition: A professional employer organization is an HR services firm that shares the employer role with a client company. The PEO and the client are both employers of the client’s workers simultaneously, with the division of responsibility set out in a Client Service Agreement (CSA).

How it works in practice:

The PEO model was built in the United States and works most cleanly there. The IRS treats the PEO as a “statutory employer” for tax purposes: it files and pays payroll taxes under its own Employer Identification Number, while the client remains the common law employer who directs the workers’ actual work (IRS, 2024). The client must have its own legal entity in the jurisdiction. The PEO augments what the client already has. It does not substitute for it.

NAPEO reports the industry collectively covers over 200,000 businesses and roughly 4.5 million workers in the United States. These figures are reported by the industry’s own trade association and reflect association rather than causation; independent peer-reviewed evidence on PEO performance outcomes is limited (NAPEO, 2024).

“Global PEO” is probably the single biggest source of confusion in this market. Most services sold under that label are EOR arrangements. Co-employment, the structural core of a domestic PEO, does not exist as a legal concept in most countries outside the United States. Where local law does not recognise shared employer status, the intermediary either becomes the sole legal employer, which is EOR, or the arrangement has no legal foundation at all (ILO, 2006). “Global PEO” is a marketing label, not a legal structure.

Where it fits: PEO works when a company already has a legal entity and an existing workforce in a jurisdiction, and wants to hand off HR administration, payroll, benefits, compliance, and workforce risk management to a specialist. It does not enable hiring in new markets. It cannot replace entity establishment.

Common failure mode: “Global PEO” selected for international hiring where no local co-employment doctrine exists. The company believes it has transferred employer liability to an intermediary; legally, no such transfer has occurred. Compliance exposure remains entirely with the client.

3.5 Global Payroll Services

Definition: Global payroll is the outsourced management of payroll operations across multiple countries: wage calculation, tax withholding and remittance, statutory reporting, and benefits administration. A global payroll provider does not become the legal employer. The client company stays the employer in every jurisdiction and keeps all the compliance obligations that come with it. Global payroll is an administrative consolidation service. It does not change who the employer is.

How it works in practice:

The client needs a legal entity in every country where it has workers. Global payroll does not solve the “no local entity” problem. It handles payroll for workers the client already has the legal right to employ. The value is consolidation: instead of managing separate payroll vendors, systems, currencies, and compliance requirements in each country, the client works through a single platform or managed service.

The provider’s relationship with the client is a commercial services agreement. There is no employment relationship between provider and worker, no co-employment, and no transfer of employer liability.

In practice, many companies use global payroll alongside EOR: EOR covers markets where they have no entity, global payroll handles markets where they do. A single provider can consolidate both, giving the company unified reporting regardless of the underlying employment structure in each jurisdiction.

Where it fits: Global payroll suits established multinationals, or growing companies that have set up entities in their main markets, that need accurate, compliant, centralised payroll management across multiple countries. It is not a substitute for EOR when no entity exists. It is what you add once the entity problem is already solved.

Common failure mode: Global payroll selected for a market where the client has no legal entity, under the mistaken impression that the provider assumes employer status. The company has no legal employer in that market, creating a compliance gap that may not surface until an enforcement event.

3.6 Contractor of Record (COR)

Definition: A contractor of record is a third party that formally holds the commercial engagement with an independent contractor on behalf of a client company. There is no employment relationship: not between the COR and the contractor, and not between the client and the contractor. The COR manages the contract, handles compliance assessment, processes payments across borders, and maintains the documentation that supports the contractor classification under local law.

In this paper, COR is treated as a distinct construct from Agent of Record (AOR). AOR originates in insurance and staffing markets, where it refers to the entity authorised to negotiate policies or contracts on behalf of a client, and it does not carry the classification compliance and cross-border payment functions that define COR as used here. Vendors sometimes use the two terms interchangeably; this paper does not.

How it works in practice:

The COR signs a commercial services agreement with the contractor: the contractor delivers defined services for agreed payment. The client contracts with the COR to access those services. The COR does not employ the contractor. It structures and administers the engagement in a way that reflects its genuine commercial nature and holds up to scrutiny from tax authorities and labour regulators.

A COR does not fix a misclassified relationship. If a client directs a contractor’s daily work, sets their hours, provides their tools, integrates them into operations, and engages them on an exclusive or near-exclusive basis, that relationship looks like employment in most jurisdictions, regardless of what any contract says or what intermediary sits in the middle. The COR belongs in genuinely independent contractor relationships. Its job is to document and administer that independence, not to create the appearance of it where it does not exist.

The underlying risks are real. The OECD is clear that classification depends on how a relationship actually functions, not on how it is labelled (OECD, 2019). Misclassification penalties can include retroactive payroll tax obligations, back payment of statutory benefits, fines, and in some jurisdictions personal liability for directors. Carré and Wilson (2004) document state-level misclassification losses in the tens of millions of dollars per jurisdiction. Enforcement is getting tighter: the UK’s IR35 reforms put classification responsibility on the engaging client rather than the contractor (Hogan Lovells, 2023), and the EU Platform Work Directive, adopted in 2024, creates a rebuttable presumption of employment for platform workers, putting the burden of proof on the company that wants to maintain contractor classification.

Permanent establishment risk is a separate concern, and not one the COR model addresses. Under international tax law, having workers in a jurisdiction, whether employees or contractors, can create a taxable presence even without a formal entity. This is a question any company should assess alongside the classification question when engaging cross-border contractors.

Boundary conditions for COR applicability:

COR is appropriate when the following conditions are met:

• The engagement is deliverable-based and project-scoped, not time-based or open-ended
• The contractor controls their own method, tools, and work schedule
• The relationship is non-exclusive, or near-exclusive engagement is clearly justified by the nature of the work
• The contractor is not integrated into the client’s operational hierarchy or day-to-day management structure
• The contractor bears genuine economic risk: they can profit from efficiency and bear the cost of remediation

COR is not appropriate, and should not be used as a substitute for EOR, when:

• The client directs daily work activities, sets hours, or provides the primary tools and equipment
• The engagement is exclusive or near-exclusive with no genuine independence
• The contractor is integrated into the client’s team in ways that mirror an employment relationship
• The engagement is time-based (hourly or daily rate) without defined deliverables

COR reduces documentation risk and administrative complexity for genuinely independent contractor relationships. It does not transfer classification liability to the intermediary. If a COR-structured relationship is found by a regulator or court to be employment in substance, the client remains exposed regardless of the COR structure in place.

Where it fits: COR works when a company wants to engage a genuinely independent contractor in a country where it has no entity, and the nature of the work actually supports contractor classification under local law. It is not a cheaper version of EOR. Choosing COR over EOR to reduce cost, when the working relationship resembles employment, creates more legal risk than it removes.

Common failure mode: COR applied to a relationship where the client directs daily work and provides equipment. The COR creates a documented record of a misclassified arrangement, typically worsening liability by making the misclassification explicit rather than ambiguous.

4. Comparative Summary

4.1 Comparison table

4.2 Liability allocation

For each model, the table below maps the three-party relationship (Worker, Client, and Intermediary) across four critical liability points.

4.3 Structural observations

Three things stand out from this comparison.

First, the legal employer status dimension and the entity requirement dimension are not independent. They follow from each other. Where the intermediary takes full formal employer status (EOR), the client needs no entity because the intermediary’s entity provides the legal foundation. Where the client keeps employer status, either fully (global payroll) or partially (PEO), it needs its own entity because it is still a legal party to the employment. COR looks like EOR on the entity requirement dimension, but for a different reason: there is no employment relationship at all, so no employment entity is required on the client’s side.

Second, EOR and COR are the only two models designed for companies without a local entity. They are not interchangeable. EOR applies when the worker is an employee. COR applies when the worker is a genuine independent contractor. Choosing between them should be determined by the nature of the working relationship. Using COR for a relationship that is substantively employment does not reduce legal risk. It creates a documented record of a misclassified arrangement, which typically makes things worse.

Third, the “global PEO” labelling problem has a parallel in the contractor space. Some providers use “agent of record” (AOR) and COR interchangeably. The terms come from different industries, AOR has roots in insurance and staffing, and they are not precisely the same thing. Companies comparing providers under these labels may not realise they are looking at structurally different services. The same dynamic that produced the global PEO confusion is producing similar confusion in the contractor intermediary market.

5. Practitioner Decision Framework

5.1 Decision map

The two questions that determine model selection are: what is the worker’s status, and does the client have a local entity? The matrix below maps these two questions to the appropriate model.

Worker classification is the first question and must be answered before the entity question. A company that routes an employment relationship through a COR because it has no entity has not solved a compliance problem; it has created a larger one.

5.2 Decision logic

Step 1: Is this worker an employee or an independent contractor?

Classification is determined by how the relationship actually works, not by what the contract says. If the client sets the worker’s hours, provides their equipment, manages their daily tasks, integrates them into the team, and is their only or main source of work, that relationship looks like employment in most jurisdictions.

• Relationship resembles employment: go to Step 2.
• Relationship is genuinely contractor-based, project-scoped, deliverable-driven, non-exclusive: go to Step 3.

Step 2: Does the client have a legal entity in the worker’s country? (Employee models)

• No entity: EOR is the compliant option. Global payroll and PEO require an entity and do not apply.
• Entity exists: go to Step 2a.

Step 2a: Does the client want to transfer formal employer status to a third party, or keep it while outsourcing administration?

• Transfer employer status: EOR. Even with an existing entity, EOR can make sense for speed or compliance management, but compare costs against entity-based options.
• Keep employer status: go to Step 2b.

Step 2b: Is the main need payroll and compliance reporting, or broader HR administration including benefits?

• Payroll and compliance reporting: global payroll provider.
• Broader HR including benefits and workforce management: domestic PEO, where co-employment is legally recognised.

Step 3: Does the client have a legal entity in the worker’s country? (Contractor model)

• No entity: COR, provided the contractor relationship meets the boundary conditions set out in Section 3.6.
• Entity exists: the client can contract directly with the contractor under local law, or use a COR for administrative consolidation and compliance documentation across multiple markets.

In practice, several things complicate this picture. As a rough practitioner rule of thumb, EOR starts to look expensive relative to entity establishment once a company has around 10 to 15 employees in a single country, though the actual break-even depends heavily on the jurisdiction and provider pricing. Co-employment has no legal standing in many jurisdictions, which limits the PEO’s usefulness internationally. COR is only as protective as the underlying classification is sound. And some countries have specific restrictions on both EOR and contractor arrangements that require specialist legal input. This framework is a starting map, not a substitute for local legal advice.

6. Limitations and Directions for Future Research

This is a conceptual paper with no empirical claims. The limitations are significant. The taxonomy draws on regulatory and practitioner sources primarily from the US, Europe, and global-platform markets. Classification law, co-employment doctrine, and triangular employment frameworks vary considerably by jurisdiction. EOR in countries with protective labour regimes, France, Germany, Brazil, or with specific restrictions on third-party employment, China, South Korea, involves complications that country-specific legal research would need to unpack. The COR construct is especially sensitive to this: a contractor arrangement that passes classification tests in the US may not hold up in France or Germany, where the definition of employment is drawn more broadly.

The commercial boundaries between these models are also contested, deliberately so. Vendors have good commercial reasons to stretch category definitions, and many service offerings span multiple categories or use non-standard terminology. The “global PEO” and “agent of record” labelling problems discussed above are symptoms of this. The taxonomy is an analytical tool for understanding these models structurally, not a classification system for evaluating providers.

This paper does not address what these arrangements mean for workers. That is a real gap. The ILO has been clear that triangular employment structures, whatever their operational advantages, can fragment workers’ relationships with the organisations that direct their work and weaken the protections employment law is meant to provide (ILO, 2006; ILO, 2016). De Stefano (2015) and Stewart and Stanford (2017) have made the same argument specifically about contractor classification: it can be used to deny protections workers are legally entitled to. Whether COR engagements improve or worsen outcomes for workers relative to unmediated contracting is an open question this paper cannot answer.

Four specific research questions follow from the constructs proposed here:

1. How do practitioners, legal advisors, and HR professionals in different regions actually distinguish EOR, PEO, global payroll, and COR when making hiring decisions, and where do their definitions diverge? Survey or interview research across jurisdictions would test whether the four-construct model reflects real-world practice or needs adjustment.
2. How do regulators’ formal definitions map onto the four taxonomic dimensions? Comparative regulatory analysis would clarify which dimensions are settled in law and which remain contested, and would make the jurisdictional variation this paper currently treats as a caveat more precise.
3. How do workers in EOR and PEO arrangements experience employer identity? Who do they understand themselves to be employed by, who do they hold responsible for their working conditions, and does that change their sense of security or commitment? Qualitative research here would connect this taxonomy to non-standard employment debates and organisational behaviour.
4. What actually happens when COR-structured arrangements are challenged by tax authorities or labour courts? Administrative data or case study research on enforcement outcomes would provide empirical grounding for the COR construct that practitioner literature currently cannot supply.

7. Conclusion

Four models dominate the market for international employment and contractor engagement: employer of record, professional employer organization, global payroll, and contractor of record. All four are routinely conflated, in vendor marketing, in advisory content, and in the academic literature, which has not treated them as separate constructs with distinct definitions.

This paper has proposed a taxonomy built on four dimensions: legal employer or engagement status, entity requirement, jurisdictional portability, and co-employment or shared liability structure. The distinctions these dimensions reveal are not academic hairsplitting. They determine what compliance obligations transfer to the intermediary, what the client remains responsible for, whether an employment or contractor relationship exists at all, and whether the chosen model is even legally valid in the relevant jurisdiction. Picking the wrong model can mean carrying compliance obligations a company thought it had outsourced, or misclassification exposure it did not know it had created.

Terminological contributions. Two definitional contributions in this paper go beyond restating established regulatory or academic sources. First, COR is formalised here as a distinct construct separate from AOR and from informal “EOR-lite” descriptions, with explicit boundary conditions for applicability. Second, the four-dimension taxonomy, using legal employer status, entity requirement, jurisdictional portability, and co-employment structure as the boundary-defining axes, is proposed in this paper as a framework for distinguishing these four models. Researchers and practitioners building on this work should note which elements are the author’s own formulation and which draw on prior regulatory or academic sources, as set out in Section 3.1.

One thing this paper has not fully addressed is what these models mean for the people working inside them. They do not only shift compliance obligations between legal entities. They also shift, and sometimes break up, the employment relationship itself. Under EOR, a worker’s legal employer is not the organisation managing their work. Under COR, there may be no employer at all. The ILO has consistently argued that the employment relationship exists to make sure workers get the protections they are owed, and that structures which obscure who the real directing party is undermine that purpose (ILO, 2006). That is the most obvious gap this paper leaves open.

References

Arbeitnehmerüberlassungsgesetz (AÜG) [German Temporary Employment Act], §1 Abs. 1b, as amended 1 April 2017. Bundesministerium der Justiz.

Brewster, C., Sparrow, P., & Vernon, G. (2007). International human resource management (2nd ed.). Chartered Institute of Personnel and Development.

Carré, F., & Wilson, R. (2004). The social and economic costs of employee misclassification in the Maine construction industry. Construction Policy Research Center, Harvard Law School and Labor and Worklife Program, Harvard Law School.

Collings, D. G., Scullion, H., & Caligiuri, P. M. (2019). Global talent management (2nd ed.). Routledge.

Cooke, F. L., Shen, J., & McBride, A. (2005). Outsourcing HR as a competitive strategy? A literature review and an assessment of implications. Human Resource Management, 44(4), 413–432.

Dahan, S. (2025). Global workforce compliance and the employer of record model. SSRN Working Paper.

Davidov, G. (2004). Joint employer status in triangular employment relationships. British Journal of Industrial Relations, 42(4), 727–746.

De Stefano, V. (2015). The rise of the “just-in-time workforce”: On-demand work, crowdwork, and labour protection in the “gig economy.” Conditions of Work and Employment Series No. 71. International Labour Organization.

Doty, D. H., & Glick, W. H. (1994). Typologies as a unique form of theory building: Toward improved understanding and modelling. Academy of Management Review, 19(2), 230–251.

Hogan Lovells. (2023). Global guide: Independent contractors, potential misclassification issues, and labor implications. Hogan Lovells International LLP.

Internal Revenue Service (IRS). (2024). Third party payer arrangements – professional employer organizations. U.S. Department of the Treasury.

International Labour Organization (ILO). (2006). Employment relationship recommendation, No. 198. International Labour Conference, 95th Session, Geneva.

International Labour Organization (ILO). (2016). Non-standard employment around the world: Understanding challenges, shaping prospects. International Labour Office.

McKinley, W. (2010). Organizational theory development: Displacement of ends? Organization Studies, 31(1), 47–68.

National Association of Professional Employer Organizations (NAPEO). (2024). PEO industry overview.

Organisation for Economic Co-operation and Development (OECD). (2019). OECD employment outlook 2019: The future of work. OECD Publishing.

Stewart, A., & Stanford, J. (2017). Regulating work in the gig economy: What are the options? The Economic and Labour Relations Review, 28(3), 420–437.

Verified Market Research. (2024). Employer of record market size, scope, trends and forecast.

Wikipedia. (2025). Employer of record.

Wikipedia. (2025). Professional employer organization.

World Economic Forum. (2023). Why the ‘employer of record’ is a model for global working.

Author

Robbin Schuchmann is co-founder of Employ Borderless, an independent advisory and research platform in the global employment solutions sector, headquartered in Singapore. He has over a decade of direct experience advising companies on international employment and has personally evaluated dozens of EOR, PEO, global payroll, and COR providers.

Contact: [email protected] - Entity home: https://robbinschuchmann.com

AI-use disclosure: Drafts prepared with assistance from AI writing tools. All subject matter expertise, analytical reasoning, taxonomic claims, and practitioner judgements are the author’s own.