Research and advocacy of progressive and pragmatic policy ideas.
Countries around the world are grappling with how to provide gig workers social protection. We analyse the trends and extract salient lessons for Malaysia.
By Edwin Goh & Nelleita Omar14 October 2021
With the rapid growth of gig platforms globally has come several attempts at resolving the lack of worker benefits and social protection coverage for gig workers. Almost every week, a new legislative or policy development emerges somewhere in the world attempting to address these labour law deficiencies.
Policymakers in Malaysia are also contending with the problem, to the extent of mooting the idea of a gig worker legislation last July, mentioned again in the recently announced 12th Malaysia Plan. The stated aim of such legislation is to ensure inclusion in social protection schemes as well as developing a new regulatory framework for the sector.
But how should a new legislation or regulatory framework be designed without causing unwanted U-turns or unintended consequences? And, who exactly are gig workers? Would these include non-platform informal workers too? These questions, and many others, reflect how gig worker protection is still very much an evolving policy area.
Since early 2021, we have been tracking legislative and policy changes around the world pertaining to gig workers and have analysed the different measures deployed. The key takeaway from looking at the various measures is that it would be a costly mistake to replicate any policy or legislation wholesale in Malaysia but an overall approach needs to be selected to fit our legal framework, labour laws, the country’s tax base and current social safety net scheme, amongst other factors.
Based on our analysis of global developments there are three distinct approaches or trends in how countries are addressing gig worker social protection coverage; in this piece, we discuss these trends and its lessons for Malaysia. The three trends are (i) designation as employees, (ii) status quo with piecemeal policies, and (iii) development of a new employment category (see Figure 1 for a geographical mapping of these trends).
Several jurisdictions have classified gig workers as employees to afford employee-level benefits and protections, including Australia, Germany, Netherlands, Spain, Switzerland, and Uruguay. The classification as employees came about mostly via court ruling.
A key factor across these relevant court rulings is the judges’ assessment of employment status based on existing employment tests or indicators. Figure 2 below shows that many of the employment status indicators cited in the court rulings hinge on the judges’ assessment of worker autonomy and the level of control held by the gig platform over the workers. The workers were assessed as having relatively low job autonomy due to factors such as limited ability to negotiate better rates, to reject work without consequences and to decide time and place of work. Platforms were judged to have relatively high levels of control due to factors such as having a performance monitoring feature or algorithmic setting that can discipline, terminate or suspend gig workers.
Another key factor or commonality across these court cases is that many of them were brought by a relatively small and quite specific group of gig workers: an individual worker in the case of Uruguay and Switzerland, and a group of four workers in the case of Australia. All workers were Uber e-hailers. Action by a comparatively large worker union via class-action suit was brought in only one country: the Netherlands.
The impact of designating gig workers as employees can be problematic. In California, the passing of Assembly Bill No. 5 or AB5, which reclassified gig workers as employees, caused many freelancers to lose their jobs due to the overly broad definition of ‘gig worker’ and ‘employee’ used in the statutory employment status test. The quantum of additional labour costs can also be a problem for platforms to absorb, particularly when abruptly imposed. In countries where employee benefits are high, some gig platform companies have chosen to cease country operations entirely as seen in Germany and Spain.
Awarding employee status can also result in policy reversals. After the passing of AB5 in California, giant gig platform companies including Uber, Lyft, and Doordash, fought to keep certain types of gig workers as independent contractors by putting forward a private bill called Proposition 22 (Prop 22). Prop 22 was then overturned by the California Supreme Court which ruled it “unconstitutional and is therefore unenforceable” for taking away the labour rights of gig workers. The story is unlikely to end here.
A slightly different and somewhat complicated development is observed in Taiwan. Public outcry spurred by fatal road accidents involving two delivery riders led to demands for gig platform companies to be punished for not protecting the workers. Under public pressure, the Taiwanese Labour Ministry intervened with an investigation, subsequently concluding that delivery drivers were employees and the gig platforms responsible for their safety. However, the gig platform companies involved have since filed a lawsuit against the government decision.
The majority of countries within this trend grouping have continued to keep silent on the employment status of gig workers – making these workers independent contractors by default – but have embarked on piecemeal efforts.
Countries that have kept silent on gig worker employment status tend to be developing nations with large proportions of informal employment in the total workforce*, such as Indonesia (56%), Thailand (56%) and India (90%). Given the significant presence of informal employment, the risk of misclassifying gig workers is high and potentially extremely costly to workers, businesses and consumers alike. The level of effective worker organisation is also lower in these countries relative to the ‘Trend 1’ countries.
*Note: There are also developed countries in this list, such as Norway, that has yet to decide on gig worker employment status but which has a universal social protection system covering every person, regardless of their employment status. Nevertheless, these countries are the exception rather than the rule.
Given this environment, many of these countries can be described as taking (and being allowed to take) a cautious wait-and-see approach, offering ad hoc targeted measures for the more visible groups of gig workers. Malaysia is a typical example, with the government introducing measures specifically for e-hailing gig workers, including enrolment into voluntary retirement schemes, compulsory occupational licenses, and mandatory accident insurance.
There is however an interesting sub-group of countries within this grouping that has officially designated gig workers as independent contractors. The sub-group comprises mainly developed countries including Belgium, Finland, and New Zealand, which have a relatively small proportion of gig workers in their total labour force. The designation as independent contractor came about mostly via court ruling, where indicators relating to job autonomy were also cited by the judges as the basis. The Arachchige versus Uber BV court case in New Zealand ruled that the driver was self-employed as he had control over several employment aspects* mentioned in Figure 2.
This difference goes to show that judicial deliberations based on employment tests, such as those mentioned in ‘Trend 1 countries’ above, can vary across jurisdictions and produce diametrically opposite results depending on the judges’ assessment as well as the particular type of gig worker that has brought the suit.
*Note: These include the control over hours of work, the use of the worker’s tools and equipment for work, personal responsibility for tax deductions, personal choice in deciding the type of vehicle and services to perform, freedom to work for other services and more.
Interestingly, the Philippines is one of very few developing countries which has enacted a new law to categorise all gig workers as independent contractors. In 2021, the Philippines Congress passed the Freelance Workers Protection Act to ensure contractual rights and certain protections for freelancers and gig workers.
There are of course serious issues and grey areas in designating gig workers as independent contractors. As the New Zealand Employment Court said in a salutary reminder to the above-mentioned court ruling, “simply because a worker is labelled an independent contractor does not mean they actually are”. The same employment test can be applied to a group of e-hailers or delivery riders, and it can yield different assessments on whether they are truly independent contractors or employees, depending on hours worked, the gig platforms’ specific policies and algorithm, and a host of other factors.
Also, and most importantly, unless there is a very good universal safety net or very good social protection schemes for the self-employed, or both, designating gig workers as independent contractors would not address the precarity of gig workers or informal workers more generally. Today, many gig workers in ‘Trend 2’ countries are still excluded from benefits and rights that were mainly designed with full-time employees in mind.
The concept of a third worker category is not new. It was first introduced in 1965, where Canadian law professor Harry Arthurs used the term ‘dependent contractor’ to describe those who are economically dependent on a company, controlled by the company to some extent, and yet possess some level of job autonomy. Some countries have opted for a third employment classification to capture this particular mix of employment characteristics exhibited by gig workers (see Figure 3).
The countries in Figure 3 created a third worker category either by extending from existing traditional dual employment classifications or introducing a completely new and separate classification. For example, Italy and the UK expanded the term ‘employee’ by adding ‘para-subordinate worker’ and ‘worker’ respectively, while Ireland included “false self-employed” and “fully dependent self-employed” under the self-employed category.
Notably, Canada is one of the first countries to create a new employment status by recognising ‘dependent contractor’ in the Ontario labour legislation, even establishing a dependent contractor status test in the early 2010s.
What is a dependent contractor status test?
The Canadian dependent contractor status test is a two-step test to determine whether a worker is a dependent contractor.
The first step determines whether the worker is a contractor or an employee by evaluating various employment indicators pertaining to the worker’s job autonomy. These factors include:
– The employer’s control of the worker’s work
– Whether the worker owned his or her own tools
– Whether the worker had a chance of profit
– The manner in which parties conducted business
– Whether the worker hired his or her own staff
– The extent to which the worker assumed financial risk
– Whether the worker invested capital in the enterprise
– Whether the worker had management responsibility
– Whether the worker worked exclusively for the employer
The second step checks if the worker is an independent contractor or dependent contractor by evaluating three main factors: economic dependence, work exclusivity and job permanency.
Other ‘Trend 3’ countries have started to consider a third employment classification in recent years. As of July 2021, China will designate workers into employees, independent contractors and a third ‘intermediate’ worker group. Accompanying the new regulations is an upcoming algorithmic recommendation law to regulate gig platforms’ algorithms in job allocation, remuneration setting and payment, work time and more.
Adding a third worker category appears to sidestep some of the pitfalls presented by ‘Trend 1’ and ‘Trend 2’ approaches but it does open up fairly new policy territory and adds to complexity. A novel approach towards mitigating this implementation risk is currently underway In Denmark, where the world’s first collective agreement was reached between gig platform Hilfr and Danish union 3F to conduct a one-year trial to test out a new classification method for gig workers. During the trial period, Hilfr will automatically re-categorise all gig workers who meet the minimum 100 working hours requirement as workers with certain employees’ benefits, unless they opt to remain self-employed.
One key takeaway from our analysis of the three global trends above is that falling back on traditional employee vs. self-employed or contractor classifications will not suffice. Not all gig workers are the same, however we define them, and sweeping all gig workers into a broad ‘employee’ or ‘self-employed’ category ignores the specific employment dynamics that truly define them.
As ‘Trend 1’ countries have shown, the impact of designating broad types of gig workers as employees depends on the locale. In countries where worker benefits are relatively generous, the level of additional costs borne by the gig platforms could result in the reduction of available jobs or worse, business closure. The details of implementation are also hotly contested; for example, should waiting hours be counted as part of working hours? Questions of this nature are by no means automatically settled even when gig workers are classified as employees. Finally, despite calls by some quarters for the employee designation, it is not at all clear if different segments of gig workers are united in wanting this if it means that as a result, opportunities for casual or part-time gig workers are reduced, time flexibility is curtailed, or extra hours are not compensated.
On the other side of the coin, keeping gig workers as independent contractors with piecemeal benefits, as per ‘Trend 2 countries, does not effectively plug gaps in current labour laws and social protection schemes, particularly in developing countries like Malaysia which does not have a universal social safety net. In addition, this regulatory approach risks having a multitude of piecemeal measures, adding more administrative burdens to workers and companies alike. The approach by the Philippines could be worth adapting though, namely in amending legislation to improve rights and benefits for contractors.
The ‘Trend 3’ approach, i.e. designating a third worker category appears as a much more promising policy option in maintaining worker autonomy while awarding a better level of protections commensurate with the degree of dependence and control held by gig platforms as well as non-platform employers (including the government). However, this option would be a relatively uncharted policy territory for Malaysia, demanding much deeper understanding from policymakers and legislators not only on the nature of employment classifications and corresponding tests, but also the practicalities of different gig platforms’ commercial policies, algorithms and business models. We like the experimental sandbox approach currently being undertaken in Denmark on testing this new classification; Malaysia should consider doing the same.
This piece is part of a larger research series on decent work and promulgating a Fair Work Act in Malaysia.