Introduction
The growth of gig work has altered the practical meaning of employment. In a conventional model, work is performed for an identifiable employer, at an identifiable workplace, under reasonably visible managerial control, and in exchange for wages or salary. Sri Lankan labour law, like most labour law systems, developed around that image. The gig economy disrupts each of those assumptions. A delivery rider may not know the full chain of contractual relationships behind an application; a ride-hailing driver may be treated as an independent service provider while prices, ratings, access to trips and deactivation are determined by a platform; an online freelancer may work for a client abroad through a platform whose terms are drafted outside Sri Lanka, raising cross-border enforceability problems.1 The ILO Convention concerning Decent Work in the Platform Economy recognises that digital platforms are significantly transforming the world of work and that platform work has generated both opportunities and decent work deficits.2,3
The term “gig worker” is used in this paper to refer to persons who perform short-term, task-based or demand-driven work that is organised or facilitated through a digital platform. This includes location-based work, such as ride-hailing, courier services and food delivery, and online work, such as freelancing, data annotation, design, coding and customer support. Convention No. 193 uses a more precise concept: a “digital labour platform” is a person or entity that, through digital technologies and automated decision-making systems, organises or facilitates remunerated work for the provision of services, whether online or in a geographic location.4 A “digital platform worker” is a person employed or engaged to perform such work for remuneration, regardless of their classification or status in employment.5 That definition is important because it begins with the reality of work rather than the label chosen by the platform.
The problem addressed in this paper is therefore not merely semantic. If gig workers are employees, they may claim the protection of labour legislation. If they are independent contractors, many protections may disappear. If they are something in between, existing Sri Lankan law does not provide an adequate category. This classification uncertainty is not accidental; it is central to many platform business models. Platforms often insist that they are technological intermediaries rather than employers, while exercising power over prices, allocation of work, performance ratings, suspension and access to customers. The worker is formally “free” to log in or log out, but is practically dependent on a private digital infrastructure that determines whether work is available and how it is rewarded.
This paper argues that Sri Lanka needs a legal framework that protects gig workers without artificially forcing all platform work into a single employment category. The proposed solution is ratification and implementation of ILO Convention No. 193. The Convention applies to all digital labour platforms and all digital platform workers unless limited exclusions are justified after consultation, and it expressly covers workers in both the formal and informal economy.6 It therefore provides a suitable normative framework for Sri Lanka, where platform workers may be formally registered, informally engaged, locally employed, internationally contracted or treated as self-employed.7
The paper has four objectives. First, it explains why gig workers need rights despite claims of flexibility and autonomy. Secondly, it analyses the content of Convention No. 193 and the reasons for Sri Lanka’s adoption. Thirdly, it evaluates the compatibility of Sri Lankan labour law with the Convention. Fourthly, it proposes reforms that would allow Sri Lanka to protect gig workers while preserving useful aspects of platform-based work. The argument proceeds on the basis that decent work and innovation are not opposites. A platform economy that depends on unprotected work, hidden control and unilateral deactivation is not sustainable. A platform economy that formalises workers, clarifies responsibility, ensures minimum standards and permits collective representation can support both enterprise development and social justice. Future-oriented scholarship similarly emphasises that the growth of gig work will require worker classification rules, minimum wage protection, access to benefits and collective bargaining mechanisms.8
Doctrinal Method and Conceptual Framework
This paper adopts a doctrinal legal method. It analyses legal texts, legal concepts, institutional structures and normative standards rather than conducting field interviews or statistical surveys. The primary legal text is ILO Convention No. 193. The Sri Lankan legal materials considered include the broad architecture of labour law relating to industrial disputes, shop and office employment, wage regulation, work injury compensation, social security, gratuity, maternity protection, termination protection and trade unions. The paper also relies on doctrinal and socio-legal scholarship in the uploaded sources to identify the regulatory problem created by platform work. The purpose is not to measure the number of gig workers in Sri Lanka, but to determine whether the existing legal framework is normatively and structurally adequate. This reflects the broader research agenda in platform work scholarship, which places legal classification, protection gaps and algorithmic management at the centre of analysis.9
A doctrinal method is appropriate because the central issue is classification. Labour law normally distributes rights through legal categories. A person who falls within the category of “employee”, “worker” or “workman” receives statutory protection; a person placed outside that category may be left to contract law. The platform economy exposes the fragility of that structure. Stewart and Stanford identify several features that typify gig work: irregular schedules, provision of equipment by workers, work performed outside a traditional workplace, piece-rate payment and digital mediation.10 These features do not necessarily prove self-employment. Instead, they show that platform work combines autonomy and subordination in ways traditional tests do not easily capture.
The doctrinal framework used here has three elements. The first is the protective purpose of labour law. Labour law exists because ordinary contract law is not sufficient where bargaining power is unequal. A worker who depends on labour income cannot be treated as if they bargain on equal terms with a business that controls the conditions of market access. Shyamali notes that ILO Recommendation No. 198 addresses disguised employment relationships, including arrangements designed to conceal or distort employment so as to avoid labour and social security obligations.11 That idea is directly relevant to gig work. A platform may call itself an intermediary, but if it controls the essential conditions of work, the law must examine substance over form.
The second element is the concept of algorithmic control. Conventional employment tests often focus on visible human supervision: who gives orders, who monitors attendance and who disciplines the worker. Platform control is different. It may operate through ratings, acceptance rates, hidden performance scores, dynamic pricing, automated allocation, customer feedback and deactivation. Shyamali’s study records that digital platforms automate allocation, evaluation and payment, while shifting capital and operational costs to workers.12 Shepherd similarly emphasises that algorithmic management creates new forms of control and subordination that challenge employment tests based on human supervision.13 Therefore, doctrinal analysis must treat algorithmic control as legally relevant control.
The third element is the distinction between flexibility and vulnerability. Flexibility can be valuable. Many workers enter platform work because they need supplementary income, because standard employment is unavailable, or because platform work allows them to manage time and family obligations. Singh identifies technology, changing workforce preferences, economic necessity and the pandemic as factors driving the rise of gig work.14 However, flexibility should not be allowed to neutralise all labour rights. Hickson shows that platform companies frequently defend gig work through the language of freedom, flexibility and being one’s own boss.15 Yet the legal question is whether that freedom is meaningful when workers remain economically dependent and vulnerable to arbitrary platform power.
Accordingly, the doctrinal inquiry in this paper asks four questions: what rights does Convention No. 193 require; what rights does Sri Lankan law already provide; which rights are unavailable because gig workers are outside the employee-employer relationship; and what reforms are necessary to close the gap. This method does not claim that every gig worker must be treated identically. It does claim that legal protection must follow the reality of work and dependence, not only contractual labels.
Gig Work, Exploitation and the Limits of the Employee-Employer Spectrum
Gig work is often presented as a modern alternative to rigid employment. That description is only partly accurate. Some platform workers do enjoy autonomy over when to log in, which tasks to accept and whether to combine multiple sources of income. But the broader legal issue is that the platform may control the market architecture within which those choices are made. In work-on-demand systems, as Stewart and Stanford explain, platforms may retain control over selection, supervision and discipline, while crowdwork systems may appear more decentralised.16 The legal relationship is therefore triangular: the worker performs the task; the customer receives the service; and the platform controls the digital infrastructure through which the transaction occurs.17 This triangular structure makes it easy for the platform to deny employer status while still extracting value from labour. Platform work can also involve uncompensated data labour, because drivers, riders and online workers generate information that trains and improves the systems that manage them.18
The risk of exploitation arises because the worker is made responsible for costs and risks that would traditionally be borne by the employer. In ride-hailing and delivery work, the worker normally supplies the vehicle, fuel, mobile phone, data, insurance, maintenance and time spent waiting for assignments. In online freelancing, the worker supplies the computer, software, internet connection and often unpaid time spent searching and bidding for work. Shyamali notes that platform operations shift investment in capital assets and operational costs to workers, while platform operators reduce their own risk.19 The worker’s income may therefore appear higher than a wage in gross terms, but the net income after expenses, waiting time and accident risk may be significantly lower.
Legal classification is the mechanism through which this risk transfer becomes lawful. If the worker is an employee, the law may impose minimum wages, overtime, paid leave, social security contributions, safety duties, unfair termination remedies and collective bargaining rights. If the worker is an independent contractor, those protections may be absent. Lobel observes that classification carries enormous weight because employees receive a range of protections in areas such as wages, discrimination, privacy, unionisation, leave, safety and health, while independent contractors are generally outside those laws.20 This is why gig worker classification has become one of the defining labour law questions of the digital economy. Javits and Luby similarly describe the lack of collective bargaining, workers’ compensation, health insurance and retirement supports for independent contractors as a major vulnerability.21 Their discussion of sports and entertainment also shows that non-standard work can be governed through a blend of collective and individual safeguards rather than through pure individual bargaining.22
The binary classification problem is particularly acute because gig workers do not perfectly resemble either traditional employees or independent entrepreneurs. They may choose their hours, but not the price. They may provide their tools, but not control the customer base. They may reject tasks, but repeated rejection may reduce future access to work. They may be called independent, but they cannot negotiate the platform’s standard terms. Shepherd describes this as a failure of current binary classifications to capture platform work relationships.23 Stewart and Stanford therefore identify regulatory options such as enforcing existing laws, expanding definitions of employment, creating an intermediate category, granting rights to workers rather than employees and reconsidering who counts as an employer.24
The rhetoric of freedom must also be examined critically. Hickson argues that the pro-gig discourse focuses on non-interference and self-authorship, presenting gig work as the freedom to choose when and how to work.25 But a worker may be legally free from direct commands while being practically dominated by the platform’s ability to withdraw access to income. Hickson uses the republican concept of domination to show that a worker can be unfree even where no actual interference occurs, if another actor has the capacity to interfere arbitrarily.26 In platform work, this capacity may take the form of sudden deactivation, opaque rating systems, unexplained reduction in work offers, or unilateral changes to commission. The deeper concern is structural domination: the legal and economic environment itself can systematically disempower workers beyond any single platform-worker relationship.27
The COVID-19 pandemic exposed this contradiction. Delivery and transport workers became essential to consumers and businesses, but many lacked sick pay, health insurance, protective equipment and clear safety duties from platforms. Shepherd notes that gig workers were on the pandemic’s front lines without access to sick pay, health insurance or protective equipment typically provided by employers.28 Lobel similarly argues that the pandemic revealed the vulnerabilities of freelancers and the irrationalities of rigid classification tests.29 The lesson for Sri Lanka is clear. A labour market can depend on gig workers during crisis while denying them the legal status necessary to claim protection.
Sri Lanka’s own digital and e-commerce context reinforces the urgency. Kumarasinghe and Sachitra describe how e-commerce businesses use gig and remote work to tap flexible talent pools and reduce overheads, while raising questions about job security, labour rights and sustainable business practices.30 Labour process analysis adds that control, emotional labour and the point of production remain relevant even where work is digitally mediated.31 They also describe the platform as both market intermediary and “shadow employer”, a concept that captures the practical power of platforms that deny legal employer identity.32 This is particularly important in a developing economy where workers may enter platform work not because they prefer entrepreneurship, but because formal employment is insufficient or unavailable.
The conclusion of this section is that gig workers need rights not because they are identical to traditional employees, but because they are exposed to comparable and sometimes greater forms of vulnerability. Their risks include income insecurity, occupational injury, unfair deactivation, non-payment, discrimination, data extraction, surveillance, lack of collective voice and absence of social protection. A legal system that protects only the traditional employee will leave a growing segment of the labour market outside decent work, especially where work is poorly paid and no official mechanism addresses unfair treatment.33
ILO Convention No. 193 and the Case for Adoption by Sri Lanka
Convention No. 193 is significant because it does not begin by asking whether a platform worker is an employee. It begins by recognising the platform economy as a specific field of work requiring specific standards. The Convention requires Member States to respect, promote and realise fundamental principles and rights at work in the platform economy, including freedom of association, collective bargaining, elimination of forced labour, abolition of child labour, non-discrimination and a safe and healthy working environment.34 This approach is especially useful for Sri Lanka because it allows protection to be built around the reality of platform work while preserving the possibility of different employment classifications.
The Convention addresses occupational safety and health in a way directly relevant to ride-hailing and delivery work. It requires measures to prevent occupational accidents, occupational diseases and other injuries to platform workers’ health arising out of or linked with their work. It also requires allocation of responsibilities among public authorities, platforms, workers and other actors, taking account of national conditions and worker classification. It further recognises the right to remove oneself from a work situation presenting imminent and serious danger, and requires protection from violence and harassment, including third-party and online harassment.35 These provisions are important because platform work often occurs in public spaces, at night, in traffic, in customers’ premises or through online communication channels where harassment may be diffuse and difficult to report.
The Convention also links platform work to formalisation. Article 7 requires national policies that promote decent work opportunities and career and skills development, while Article 8 requires measures to facilitate formalisation of work via digital labour platforms, including registration of self-employed workers.36 This is particularly relevant to Sri Lanka because the difficulty is not only that some platform workers are outside labour law, but that they are often invisible to labour administration. Registration can enable social security contributions, accident insurance, tax clarity, dispute resolution and statistical planning. It can also reduce unfair competition between businesses that comply with labour standards and platforms that externalise obligations.
Article 9 is the doctrinal core of the Convention. It requires correct classification of platform workers in relation to the existence or non-existence of an employment relationship, guided mainly by the facts relating to work performance and remuneration, and considering the specificities of platform work.37 This is a substance-over-form approach. It does not compel every country to classify all platform workers as employees; rather, it prevents platforms from using contractual labels to pre-determine the legal outcome. Comparative scholarship shows that the main policy choices include better enforcement, wider definitions, intermediate categories and broader worker-based rights.38 The pandemic literature also cautions that an exclusive battle over employee status may obscure the need for protections across the labour market.39 For Sri Lanka, this means that labour tribunals, regulators and courts should be empowered to examine algorithmic control, economic dependence, exclusivity, pricing power, disciplinary systems and the worker’s real opportunity for profit or loss.
The remuneration provisions are also central. Article 10 requires timely, full and lawful payment of remuneration or payment due under law, collective agreement or contract. For platform workers in an employment relationship, remuneration excluding tips must not be lower than the applicable statutory or negotiated minimum wage, and such workers must be compensated for expenses or costs incurred in the performance of work. The Convention further requires consideration of whether minimum wage measures should extend to platform workers who are not in an employment relationship.40 Article 11 adds an information right, requiring accurate and understandable information on remuneration and deductions.41 This is a direct answer to opaque commission structures, dynamic pricing, hidden deductions and unexplained payment changes.
Article 12 deals with social security. It requires that platform workers have access to social security protection on terms no less favourable than other workers with the same classification of status in employment.42 The clause is modest but important. It recognises that social protection need not be identical for all statuses, but it cannot be denied merely because work is platform-mediated. In Sri Lanka, implementation could involve contributions to EPF, ETF, accident insurance or a new portable benefits fund depending on classification and earnings. The key point is portability. A platform worker may work for several platforms; protection must attach to the worker and income stream, not only to a single employer.
Articles 13 to 15 are among the most innovative parts of the Convention. They require platforms to inform workers and their representatives about the use of automated systems to monitor, evaluate or generate work-related decisions, and about the impact of those systems on working conditions and access to work. They also require responsible use of automated systems and access to written explanations and review, with appropriate human involvement, for significant adverse decisions such as non-payment, suspension, deactivation or termination.43 This is crucial because algorithmic management is the main form of control in platform work. A Sri Lankan implementation statute should therefore make algorithmic transparency a labour right, not merely a data protection issue.
The Convention further addresses personal data and privacy. Article 16 requires safeguards for workers’ personal data, including access, rectification and erasure, subject to applicable data retention laws.44 Articles 17 and 18 prohibit suspension, deactivation or termination on discriminatory or unlawful grounds and require timely, verifiable and understandable information on terms and conditions of employment or engagement.45 Article 19 states that terms and conditions should preferably be governed by the law of the country where the work is performed, unless otherwise provided by law, international instruments or agreements.46 This is particularly valuable for Sri Lankan online workers whose platform terms may select a foreign law and forum.
Finally, the Convention requires safe, fair and effective dispute resolution and remedies, enforcement mechanisms, no less favourable treatment and implementation through consultation with employers’ and workers’ organisations.47 It also requires implementation in relation to platforms and intermediaries operating in the territory and workers working in the territory, with responsibilities allocated between platforms and intermediaries where intermediaries are used.48 These provisions show why ratification is preferable to isolated amendments. Platform work is not a single labour issue; it involves classification, payment, safety, social security, data, automation, contract terms, cross-border law and collective representation. Convention No. 193 provides a coherent structure for reform.
Compatibility with Existing Sri Lankan Labour Law
Sri Lanka’s labour law is not hostile to the protective aims of Convention No. 193. On the contrary, its labour legislation has historically been protective. The difficulty is that the protection is organised around statutory categories that may not include gig workers. Wanasinghe, Phuoc and Khatibi note that Sri Lanka has numerous labour statutes dealing with different aspects of employment, but virtual employment is not defined or explicitly addressed, creating a significant drawback in the legal system.49 The same critique applies with greater force to gig workers because platform work may be even more fragmented than virtual employment.
The Industrial Disputes Act No. 43 of 1950 is potentially the most compatible statute because its concept of “workman” is broader than a narrow employee concept. The uploaded Sri Lankan study notes that the Act defines a workman as a person who has entered into or works under a contract with an employer, whether express or implied, oral or written, and whether a contract of service, apprenticeship, or a contract personally to execute work or labour.50 This language may be useful for some gig workers, especially where the worker personally performs labour and the platform exercises control. However, the difficulty remains the need to identify an “employer” and a contract with that employer. Platforms may argue that the worker contracts with the customer, not the platform, or that the platform merely facilitates the transaction.
The Shop and Office Employees Act No. 19 of 1954 and Wages Boards Ordinance No. 27 of 1941 provide partial compatibility in relation to hours, leave, overtime and wage regulation. However, their application depends on statutory coverage and the character of the workplace or trade. Wanasinghe’s study observes that definitions in key legislation, including the Industrial Disputes Act No. 43 of 1950, Shop and Office Employees Act No. 19 of 1954 and Workmen’s Compensation Ordinance No. 19 of 1934, do not clearly apply to virtual employment.51 In platform work the problem is more complex because the “shop” or “office” may be an app, the workplace may be a road or a home, and remuneration may be calculated per task rather than by time. A rider waiting for work while logged into an app may be economically engaged, but the law may not treat waiting time as working time. The virtual employment literature similarly records uncertainty and non-payment for overtime in non-standard digital work.52
There is also partial compatibility regarding remuneration. Sri Lankan wage laws can regulate minimum wages, overtime and payment practices for covered workers. The uploaded Sri Lankan study states that the Wages Boards Ordinance No. 27 of 1941 concerns minimum rates for piece work, time work, overtime rates, working hours and working days, but also notes that it applies to identified trades.53 Gig work often uses piece-rate payment. Therefore, the conceptual tools already exist. What is missing is a platform-specific mechanism to calculate minimum earnings per active hour, waiting time, task time, expenses and deductions. Convention No. 193’s distinction between remuneration and expenses is helpful because it prevents platforms from treating gross payment as if it were take-home pay.
Sri Lankan law also contains important social security institutions. The EPF, ETF and gratuity regimes are designed to protect workers after long service, retirement, cessation of employment or premature retirement due to sickness or other reasons. The uploaded study records EPF employer and employee contribution rates and identifies ETF and gratuity as core welfare mechanisms.54 These institutions are broadly compatible with Article 12 of Convention No. 193. The incompatibility lies in access. A gig worker treated as self-employed may not receive employer contributions. A worker using multiple platforms may not have a single employer through whom contributions can be collected. Thus, Sri Lanka needs a portable contribution mechanism that can capture platform income across multiple platforms.
Maternity protection and leave rights show a similar pattern. Sri Lankan law grants maternity leave to covered female employees, but Wanasinghe’s study notes that employees working for companies outside Sri Lanka may not benefit from those Acts, and that online work from home should not be used to deny the mental and physical purpose of maternity leave.55 The lesson for gig work is that location flexibility cannot replace legal entitlement; the same is true of annual, sick and public holiday leave where foreign or app-based arrangements displace local norms.56 A woman working through a platform may be at home, on the road or online, but she still needs income security, health protection and protection against discrimination during pregnancy and maternity.
Trade union law presents a more serious compatibility problem. Convention No. 193 requires freedom of association and effective recognition of collective bargaining in the platform economy. Sri Lankan law constitutionally recognises freedom of association and the Trade Unions Ordinance No. 14 of 1935 provides a statutory framework for unions. However, Wanasinghe’s study notes that self-employed persons and unpaid employers and employees cannot form a trade union under the Ordinance.57 Shyamali’s regional study similarly observes that in Sri Lanka a trade union member had to be a “workman”, creating legal impediments for digital platform workers categorised as independent workers.58 If gig workers are treated as self-employed, they may form associations but not necessarily trade unions with bargaining rights.
The Sri Lankan experience with ride-hailing workers illustrates both possibility and limitation. Shyamali records that a group of drivers signed up to a ride-hailing platform in Sri Lanka formed an association with the assistance of a mainstream trade union and concluded a memorandum of understanding with the platform, but the MOU was not a collective agreement under Sri Lankan legislation because the workers were not considered employees and the platform was not considered an employer.59 This is a crucial compatibility finding. Sri Lanka already has collective organising in practice, but the law does not adequately recognise it. Convention No. 193 would require Sri Lanka to move from informal accommodation to legally enforceable collective voice.
Unfair termination and deactivation are also partially compatible. Labour tribunals can grant just and equitable relief for unreasonable termination in employment relationships, and local virtual employees working for local employers are generally within the reach of Sri Lankan labour laws.60 However, Wanasinghe’s study also highlights jurisdictional difficulties when the employer is outside Sri Lanka, noting that practical enforcement may be futile unless reciprocal arrangements exist.61 In platform work, deactivation may function like dismissal, but it may be described as termination of access to an account under platform terms. Sri Lankan law must therefore treat serious suspension or deactivation as a labour-impacting decision requiring reasons, human review and remedy.
The overall compatibility assessment is therefore mixed. Sri Lanka is normatively compatible with Convention No. 193 because its labour law tradition values worker protection, social security, trade unionism and remedies. It is structurally incompatible because its protections are not designed for algorithmic, triangular, platform-mediated and cross-border work. The law protects many workers once they enter the statutory gate; the problem is that gig workers may be kept outside the gate by contractual classification.
What Needs to Change: A Reform Agenda for Sri Lanka
The first reform should be a dedicated Platform Work Act implementing Convention No. 193. Piecemeal amendments to existing labour statutes would not be enough because platform work raises interlocking issues of classification, payment, social security, safety, data, algorithms, termination, intermediaries and jurisdiction. The Act should define “digital labour platform”, “digital platform worker”, “intermediary”, “remuneration”, “active time”, “waiting time”, “deactivation” and “automated decision-making system” consistently with the Convention. The definition of platform worker should apply regardless of contractual label, as Convention No. 193 does. This would prevent platforms from excluding workers through standard terms.
Secondly, Sri Lanka should adopt a rebuttable classification framework. Article 9 requires classification to be guided mainly by facts relating to performance of work and remuneration, considering platform specificities. The Platform Work Act should provide that a worker is presumed to be in an employment relationship where the platform controls price, allocation, performance standards, ratings, appearance, customer communication, sanctions, or access to work. The platform may rebut the presumption by showing genuine entrepreneurial independence. This approach preserves flexibility for truly independent freelancers while preventing disguised employment. It also reflects Stewart and Stanford’s option of clarifying or expanding employment definitions.62
Thirdly, certain core rights should apply to all platform workers regardless of status. This is necessary because some workers may remain genuinely self-employed but still face non-payment, opaque deductions, data misuse, unsafe work and arbitrary deactivation. It also avoids treating employee status as the only gateway to public labour protection.63 Stewart and Stanford identify the option of creating rights for workers rather than employees, and Lobel similarly argues that some protections should extend to non-employees whether work is digital or offline.64 The core rights should include timely payment, transparent deductions, access to earnings records, protection from discrimination, safety duties, protection from violence and harassment, data rights, reasons for deactivation, human review and access to dispute resolution.
Fourthly, remuneration regulation must account for expenses and waiting time. Convention No. 193 requires timely and full payment and requires employees to receive at least the applicable minimum wage and compensation for expenses, while also asking states to consider minimum wage measures for non-employees. Sri Lanka should introduce a minimum earnings standard for platform work based on active time and a fair formula for waiting time in platform-controlled systems. For location-based work, vehicle costs, fuel, mobile data, insurance, maintenance and platform commissions must be considered. A platform should not be allowed to advertise gross earnings while transferring essential business expenses to the worker.
Fifthly, Sri Lanka should create a portable social security mechanism for platform workers. One model is a Platform Worker Social Protection Fund into which platforms contribute a percentage of payments processed through the app, with workers also contributing a smaller percentage depending on income. Contributions should follow the worker across platforms. The fund could cover accident insurance, sickness support, maternity-related income protection, retirement savings and emergency assistance. This model would build on Sri Lanka’s existing EPF, ETF and gratuity logic while adapting it to multi-platform work. It would also respond to Shyamali’s finding that workers often bear accident risks and have no social security system to rely on in emergencies.65
Sixthly, Sri Lanka must recognise collective representation for platform workers. The Trade Unions Ordinance No. 14 of 1935 should be amended so that platform workers, including those classified as dependent self-employed or economically dependent contractors, may form and join trade unions or registered platform worker organisations. The law should allow such organisations to bargain collectively with platforms without being defeated by the argument that workers are not employees. Shyamali’s study notes that there are no collective agreements covering digital platform workers in the six-country study and that Sri Lankan platform drivers used an association and MOU rather than a statutory collective agreement.66 Ratification of Convention No. 193 would require this gap to be addressed.
Seventhly, algorithmic transparency must be legislated as a labour right. Articles 13 to 15 require information about automated systems, written explanations for significant adverse decisions, review of non-payment, suspension, deactivation or termination, and appropriate human involvement.67 A Sri Lankan Platform Work Act should require platforms to disclose, in understandable terms, the main factors affecting task allocation, pay, ratings, incentives, penalties and deactivation. Workers should have the right to contest automated decisions before a human decision-maker within a short time. A platform should not be able to hide behind trade secrecy where the decision affects livelihood. Article 24 allows protection of commercially sensitive information, but that cannot defeat the worker’s right to reasons.68
Eighthly, platform contracts should be regulated. Convention No. 193 requires understandable terms and preferably applies the law of the country where the work is performed.69 Sri Lankan law should prohibit unfair platform terms, including unilateral variation without notice, foreign forum clauses that practically deny remedies for local work, broad waivers of labour rights, unexplained deductions, and terms allowing deactivation without reasons. For Sri Lankan workers performing work in Sri Lanka, Sri Lankan mandatory labour protections should apply irrespective of the governing law clause. For online cross-border work, Sri Lanka should at least regulate platforms operating or marketing services in Sri Lanka and provide local complaint channels. The statute should also address migrant and refugee platform workers, as required by Convention No. 193, because platform work often crosses citizenship and territorial boundaries.70
Ninthly, occupational safety and health rules should be platform-specific. Ride-hailing and delivery workers face road accidents, assault, harassment, weather exposure, fatigue and pressure created by time-based incentives. Online workers face eye strain, mental stress, surveillance, payment insecurity and harassment through digital channels. Convention No. 193’s safety and harassment provisions should be implemented by requiring risk assessments, accident reporting, insurance, emergency assistance, customer misconduct procedures, safe working guidelines, and a right to refuse dangerous tasks without penalty. This is not an excessive burden; it is the minimum responsibility of businesses that organise work through digital infrastructure.
Tenthly, dispute resolution must be made accessible. Article 21 requires safe, fair and effective dispute resolution and appropriate remedies. Sri Lanka should create a specialised Platform Work Desk within the Department of Labour, with power to receive complaints on non-payment, deactivation, algorithmic decisions, discrimination, harassment and social security contributions. Labour tribunals should be given jurisdiction over platform work disputes where the platform operates in Sri Lanka, the worker works in Sri Lanka, or the work is substantially connected to Sri Lanka. Remedies should include reinstatement of account access, payment of sums due, compensation, correction of ratings or records, and orders requiring human review.
Finally, implementation must be consultative. Article 24 requires implementation in consultation with representative employers’ and workers’ organisations. Sri Lanka should include trade unions, platform worker associations, platform companies, Labour Department officials, data protection authorities, consumer representatives and road safety authorities in drafting regulations. Comparative discussion of gig workers also supports creative combinations of political action, individual negotiation and collective organisation to protect flexible workers.71 This is necessary because platforms differ. A ride-hailing platform, a food delivery platform, an e-commerce logistics platform and a global freelancing website do not raise identical issues. However, the minimum principles of decent work, transparency, social protection and remedy must apply across the sector.
Conclusion
The expansion of gig work in Sri Lanka creates a choice. Sri Lanka can allow platform work to grow in a legal grey zone, with workers classified as freelancers or independent contractors while platforms control access to work through algorithms and standard terms. Or Sri Lanka can adopt a modern framework that recognises flexibility while preventing exploitation. This paper has argued for the second path.
Convention No. 193 provides the appropriate legal foundation because it does not rely solely on the traditional employee-employer binary. It recognises digital labour platforms and digital platform workers as legal subjects of regulation, covers workers regardless of status, requires correct classification based on facts, and introduces specific protections on remuneration, social security, automated systems, personal data, deactivation, terms and conditions, applicable law, remedies and enforcement.72 It therefore addresses the precise features that make platform work difficult for ordinary labour law.
Sri Lanka’s existing law is compatible in principle but incomplete in structure. Its labour law tradition already values minimum standards, welfare funds, collective representation and labour tribunal remedies. However, the law is not sufficiently adapted to triangular platform relationships, algorithmic control, multi-platform income, cross-border online work and worker invisibility. Wanasinghe’s analysis of virtual employment showed that even earlier digital work arrangements were not adequately addressed by labour legislation and that foreign employers created serious enforcement problems.73 Shyamali’s study shows that Sri Lankan platform workers have organised in practice but remain outside conventional collective bargaining structures.74 These findings confirm that reform is not merely desirable but necessary.
The key reform should be a dedicated Platform Work Act implementing Convention No. 193. Emerging regulatory models in other jurisdictions also suggest that hybrid categories and portable benefit systems can respond to the limits of binary classification.75 Such an Act should create a broad definition of platform worker, a rebuttable employment presumption where control and dependence exist, core rights for all platform workers, minimum earnings standards, portable social security, collective representation rights, algorithmic transparency, data protection, fair deactivation procedures and accessible dispute resolution. It should also amend existing labour laws where necessary so that platform workers are not excluded from trade union rights, social security, safety protection and remedies merely because of contractual labels.
The central claim of this paper is that flexibility should be protected, but not at the cost of legality. A worker can be flexible and still deserve payment transparency. A worker can choose hours and still need accident protection. A worker can work through an app and still have freedom of association. A worker can be self-employed in some respects and still be protected from arbitrary deactivation. The law must therefore move from status formalism to functional protection. Ratifying and implementing Convention No. 193 would enable Sri Lanka to build that framework before platform work becomes too large, too informal and too difficult to regulate.
*****
Footnotes
1. Ranaraja Shyamali, ‘Trade Union Responses to Organizing Workers on Digital Labour Platforms: A Six-country Study’ (2022) 11 International Journal of Labour Research 59, 65.
2. International Labour Organization, Decent Work in the Platform Economy Convention, 2026 (Convention No. 193) (12 June 2026) preamble.
3. Ranaraja Shyamali, ‘Trade Union Responses to Organizing Workers on Digital Labour Platforms: A Six-country Study’ (2022) 11 International Journal of Labour Research 59, 64.
4. ILO Convention No. 193 art 1(a).
5. ILO Convention No. 193 art 1(b).
6. ILO Convention No. 193 art 2.
7. Ranaraja Shyamali, ‘Trade Union Responses to Organizing Workers on Digital Labour Platforms: A Six-country Study’ (2022) 11 International Journal of Labour Research 59, 62.
8. Haridarshan Singh, ‘Gig Workers: A Comprehensive Analysis of the Rise, Challenges, and Future of the Gig Economy’ (2024) 6 International Journal for Multidisciplinary Research 1, 6–7.
9. Nicki James Shepherd, ‘Gig Economy Workers’ Rights: Legal Classification and Social Protection in Digital Labour Markets’ (2025) 2 International Journal of Law and Societal Studies 24, 26.
10. Andrew Stewart and Jim Stanford, ‘Regulating Work in the Gig Economy: What are the Options?’ (2017) 28 The Economic and Labour Relations Review 1, 2.
11. Ranaraja Shyamali, ‘Trade Union Responses to Organizing Workers on Digital Labour Platforms: A Six-country Study’ (2022) 11 International Journal of Labour Research 59, 63–64, discussing the ILO Employment Relationship Recommendation, 2006 (No. 198).
12. Ranaraja Shyamali, ‘Trade Union Responses to Organizing Workers on Digital Labour Platforms: A Six-country Study’ (2022) 11 International Journal of Labour Research 59, 61–62.
13. Nicki James Shepherd, ‘Gig Economy Workers’ Rights: Legal Classification and Social Protection in Digital Labour Markets’ (2025) 2 International Journal of Law and Societal Studies 24, 25–26.
14. Haridarshan Singh, ‘Gig Workers: A Comprehensive Analysis of the Rise, Challenges, and Future of the Gig Economy’ (2024) 6 International Journal for Multidisciplinary Research 1, 1–2.
15. James Hickson, ‘Freedom, Domination and the Gig Economy’ (2024) 29 New Political Economy 321, 321–322.
16. Andrew Stewart and Jim Stanford, ‘Regulating Work in the Gig Economy: What are the Options?’ (2017) 28 The Economic and Labour Relations Review 1, 2–3.
17. Andrew Stewart and Jim Stanford, ‘Regulating Work in the Gig Economy: What are the Options?’ (2017) 28 The Economic and Labour Relations Review 1, 3.
18. Ranaraja Shyamali, ‘Trade Union Responses to Organizing Workers on Digital Labour Platforms: A Six-country Study’ (2022) 11 International Journal of Labour Research 59, 65–66.
19. Ranaraja Shyamali, ‘Trade Union Responses to Organizing Workers on Digital Labour Platforms: A Six-country Study’ (2022) 11 International Journal of Labour Research 59, 61–62.
20. Orly Lobel, ‘We Are All Gig Workers Now: Online Platforms, Freelancers & the Battles Over Employment Status & Rights During the Covid-19 Pandemic’ (2020) 57 San Diego Law Review 919, 920–921.
21. Joshua M Javits and Matthew L Luby, ‘Gig Workers: Walking a Tightrope Without a Safety Net’ (2022) Journal of Dispute Resolution 27, 28–29.
22. Joshua M Javits and Matthew L Luby, ‘Gig Workers: Walking a Tightrope Without a Safety Net’ (2022) Journal of Dispute Resolution 27, 29–30.
23. Nicki James Shepherd, ‘Gig Economy Workers’ Rights: Legal Classification and Social Protection in Digital Labour Markets’ (2025) 2 International Journal of Law and Societal Studies 24, 24–25.
24. Andrew Stewart and Jim Stanford, ‘Regulating Work in the Gig Economy: What are the Options?’ (2017) 28 The Economic and Labour Relations Review 1, 12–17.
25. James Hickson, ‘Freedom, Domination and the Gig Economy’ (2024) 29 New Political Economy 321, 328.
26. James Hickson, ‘Freedom, Domination and the Gig Economy’ (2024) 29 New Political Economy 321, 321.
27. James Hickson, ‘Freedom, Domination and the Gig Economy’ (2024) 29 New Political Economy 321, 329–333.
28. Nicki James Shepherd, ‘Gig Economy Workers’ Rights: Legal Classification and Social Protection in Digital Labour Markets’ (2025) 2 International Journal of Law and Societal Studies 24, 25.
29. Orly Lobel, ‘We Are All Gig Workers Now: Online Platforms, Freelancers & the Battles Over Employment Status & Rights During the Covid-19 Pandemic’ (2020) 57 San Diego Law Review 919, 920–922.
30. Wewaldeniyage Shanika Lakmali Kumarasinghe and Vilani Sachitra, ‘Redefining Work in the Digital Era: Exploring the Role of Gig Economy and Remote Employment in E-commerce’ (2025) 26 Advances in Research 551, 552–554.
31. Wewaldeniyage Shanika Lakmali Kumarasinghe and Vilani Sachitra, ‘Redefining Work in the Digital Era: Exploring the Role of Gig Economy and Remote Employment in E-commerce’ (2025) 26 Advances in Research 551, 552–553.
32. Wewaldeniyage Shanika Lakmali Kumarasinghe and Vilani Sachitra, ‘Redefining Work in the Digital Era: Exploring the Role of Gig Economy and Remote Employment in E-commerce’ (2025) 26 Advances in Research 551, 552.
33. Ranaraja Shyamali, ‘Trade Union Responses to Organizing Workers on Digital Labour Platforms: A Six-country Study’ (2022) 11 International Journal of Labour Research 59, 64–65.
34. ILO Convention No. 193 art 3.
35. ILO Convention No. 193 arts 4–6.
36. ILO Convention No. 193 arts 7–8.
37. ILO Convention No. 193 art 9.
38. Andrew Stewart and Jim Stanford, ‘Regulating Work in the Gig Economy: What are the Options?’ (2017) 28 The Economic and Labour Relations Review 1, 13–14.
39. Orly Lobel, ‘We Are All Gig Workers Now: Online Platforms, Freelancers & the Battles Over Employment Status & Rights During the Covid-19 Pandemic’ (2020) 57 San Diego Law Review 919, 938–945.
40. ILO Convention No. 193 arts 10(2), 10(3).
41. ILO Convention No. 193 arts 10–11.
42. ILO Convention No. 193 art 12.
43. ILO Convention No. 193 arts 13–15.
44. ILO Convention No. 193 art 16.
45. ILO Convention No. 193 arts 17–18.
46. ILO Convention No. 193 art 19.
47. ILO Convention No. 193 art 21.
48. ILO Convention No. 193 arts 22–24.
49. W M V Wanasinghe, Jeong Chun Phuoc and Ali Khatibi, ‘Application of Sri Lankan Labor Laws on Sri Lankan Virtual Employees’ (2017) 3 International Journal of Arts and Humanities 29, 30–31.
50. W M V Wanasinghe, Jeong Chun Phuoc and Ali Khatibi, ‘Application of Sri Lankan Labor Laws on Sri Lankan Virtual Employees’ (2017) 3 International Journal of Arts and Humanities 29, 31.
51. W M V Wanasinghe, Jeong Chun Phuoc and Ali Khatibi, ‘Application of Sri Lankan Labor Laws on Sri Lankan Virtual Employees’ (2017) 3 International Journal of Arts and Humanities 29, 31–33.
52. W M V Wanasinghe, Jeong Chun Phuoc and Ali Khatibi, ‘Application of Sri Lankan Labor Laws on Sri Lankan Virtual Employees’ (2017) 3 International Journal of Arts and Humanities 29, 35.
53. W M V Wanasinghe, Jeong Chun Phuoc and Ali Khatibi, ‘Application of Sri Lankan Labor Laws on Sri Lankan Virtual Employees’ (2017) 3 International Journal of Arts and Humanities 29, 36.
54. W M V Wanasinghe, Jeong Chun Phuoc and Ali Khatibi, ‘Application of Sri Lankan Labor Laws on Sri Lankan Virtual Employees’ (2017) 3 International Journal of Arts and Humanities 29, 36–37.
55. W M V Wanasinghe, Jeong Chun Phuoc and Ali Khatibi, ‘Application of Sri Lankan Labor Laws on Sri Lankan Virtual Employees’ (2017) 3 International Journal of Arts and Humanities 29, 37.
56. W M V Wanasinghe, Jeong Chun Phuoc and Ali Khatibi, ‘Application of Sri Lankan Labor Laws on Sri Lankan Virtual Employees’ (2017) 3 International Journal of Arts and Humanities 29, 35–36.
57. W M V Wanasinghe, Jeong Chun Phuoc and Ali Khatibi, ‘Application of Sri Lankan Labor Laws on Sri Lankan Virtual Employees’ (2017) 3 International Journal of Arts and Humanities 29, 37–38.
58. Ranaraja Shyamali, ‘Trade Union Responses to Organizing Workers on Digital Labour Platforms: A Six-country Study’ (2022) 11 International Journal of Labour Research 59, 69–72.
59. Ranaraja Shyamali, ‘Trade Union Responses to Organizing Workers on Digital Labour Platforms: A Six-country Study’ (2022) 11 International Journal of Labour Research 59, 71.
60. W M V Wanasinghe, Jeong Chun Phuoc and Ali Khatibi, ‘Application of Sri Lankan Labor Laws on Sri Lankan Virtual Employees’ (2017) 3 International Journal of Arts and Humanities 29, 38–39.
61. W M V Wanasinghe, Jeong Chun Phuoc and Ali Khatibi, ‘Application of Sri Lankan Labor Laws on Sri Lankan Virtual Employees’ (2017) 3 International Journal of Arts and Humanities 29, 39–40.
62. Andrew Stewart and Jim Stanford, ‘Regulating Work in the Gig Economy: What are the Options?’ (2017) 28 The Economic and Labour Relations Review 1, 10–12.
63. Orly Lobel, ‘We Are All Gig Workers Now: Online Platforms, Freelancers & the Battles Over Employment Status & Rights During the Covid-19 Pandemic’ (2020) 57 San Diego Law Review 919, 944–945.
64. Andrew Stewart and Jim Stanford, ‘Regulating Work in the Gig Economy: What are the Options?’ (2017) 28 The Economic and Labour Relations Review 1, 15–16.
65. Ranaraja Shyamali, ‘Trade Union Responses to Organizing Workers on Digital Labour Platforms: A Six-country Study’ (2022) 11 International Journal of Labour Research 59, 69–70.
66. Ranaraja Shyamali, ‘Trade Union Responses to Organizing Workers on Digital Labour Platforms: A Six-country Study’ (2022) 11 International Journal of Labour Research 59, 71.
67. ILO Convention No. 193 arts 15, 17.
68. ILO Convention No. 193 art 24(2)–(3).
69. ILO Convention No. 193 arts 18–19.
70. ILO Convention No. 193 art 20.
71. Joshua M Javits and Matthew L Luby, ‘Gig Workers: Walking a Tightrope Without a Safety Net’ (2022) Journal of Dispute Resolution 27, 30–35.
72. ILO Convention No. 193 art 23.
73. W M V Wanasinghe, Jeong Chun Phuoc and Ali Khatibi, ‘Application of Sri Lankan Labor Laws on Sri Lankan Virtual Employees’ (2017) 3 International Journal of Arts and Humanities 29, 40.
74. Ranaraja Shyamali, ‘Trade Union Responses to Organizing Workers on Digital Labour Platforms: A Six-country Study’ (2022) 11 International Journal of Labour Research 59, 72–73.
75. Nicki James Shepherd, ‘Gig Economy Workers’ Rights: Legal Classification and Social Protection in Digital Labour Markets’ (2025) 2 International Journal of Law and Societal Studies 24, 29–32.