Platform Work - How Helpling's bankruptcy exposes the flaws of the debate on the gig economy - January 26, 2023

MARTIJN ARETS (LinkedIn - Twitter)

After a long legal battle, platform for domestic cleaners Helpling in the Netherlands filed for bankruptcy. The bankruptcy marks the end of a not uncontested battle with Dutch trade union FNV, which claimed the bankruptcy as a "belated victory" in the Dutch media.

The big question that remains is: a victory for whom? In a blog in 2018, I predicted that a ruling in FNV's 'favour' would result in the company going bankrupt. And I already put quite a few question marks and exclamation marks on this case. In this blog, I outline the context of the case and Helpling's activities, discuss what is (not) new about a domestic cleaning platform and give an overview of what changes the platform has made over the years. To conclude with a proposal for a debate that honours both workers and society.

The rise of Helpling

Helpling, a startup from Germany's Rocket Internet, entered the Dutch market in June 2014. The platform where individuals can book a cleaner for their home is ambitious and is quickly seen as one of the big promises in the rise of the gig economy, a.k.a. 'gig economy', in the Netherlands. That the platform will not just stick to cleaning becomes clear to me when I interview then-Director of the Netherlands Floyd Sijmons at their Amsterdam office in 2015. When I ask him if it will stick to cleaning, he replies, "if we were to stick to cleaning only, we would have called our platform Cleanling". Crystal clear.

Yet the platform fails to expand its services. It is looking at window cleaners, but that ultimately does not get off the ground. The added value of the platform also turns out to be a lot less than with, say, Uber. It involves a pre-scheduled transaction (nobody wants a cleaner 'on demand') and the worker and client stay connected after an initial successful gig and can easily work around the platform. Due to the nature of the transaction, no high-level algorithm is needed and the benefits of data mining are limited. Therefore, there is no strong 'lock in' as with Uber and the platform has to work hard to offer an appropriate service to worker and client to bind them to the platform.

Retaining clients within national markets was more difficult than potentially expected at the outset. Where it was first thought that one strategy would suffice, it soon became clear that each country needed its own strategy. This is reflected in the recently published paper "Platform adaptation to regulation: the case of domestic cleaning in Europe", written by Koutsimpogiorgos, Frenken and Herrmann. For this paper, the authors compare Helpling's general terms and conditions in France, Germany, Ireland, the Netherlands and the UK and analysed the changes over the years. Their conclusion is that although Helpling starts with a general strategy, it soon has to adapt to the institutional conditions per country. Over the years, Helpling regularly adjusts its model in the Netherlands. For instance, the platform changes its commission structure and introduces a model where cleaners are allowed to set the price for a cleaning job themselves. A minimal fee should ensure that there is no 'race to the bottom'.

Home services arrangement

Helpling has never worked with freelancers. The platform, like other platforms where consumers hire services in and around the house, makes use of the ‘regeling dienstverlening aan huis’ ('service at home scheme'). This is a national regulation in the Netherlands that exempts private employers, which is what you are when you get a cleaner to come for your home, from a large set of employer obligations. However, a private employer must pay the minimum wage and 8 per cent holiday pay, continue to pay the cleaner in case of illness and provide a safe and healthy workplace.

In practice, little comes of this regulation in the private cleaning market in the Netherlands due to lack of enforcement. Or, as I wrote in an earlier blog: "So it looks like we have it well regulated for our domestic workers, but it is not. They fall into an 'excuse category', so to speak." Helpling, by using this arrangement, joins the standard in the Netherlands."

The end of Helpling

The fact that the private employer finds, books and pays a cleaner through a platform is reason for FNV to take the platform to court in 2018. FNV believes that the platform is an employer and must employ the cleaners. In the first instance, the subdistrict court ruled in 2019 that Helpling is not an employer but an intermediary. Two years later, however, the Amsterdam court of appeal sees things differently and rules that Helpling must treat the cleaners as temporary workers, with corresponding rights and obligations.

Helpling is aware that a temp model will not work for the platform: consumers will never want to pay these rates. The platform flips the model for the last time, rigorously discontinuing its original model where it deducts a commission on the hours paid out and introduces Helpling Select (where you pay a one-off fee to find a cleaner and then take care of everything yourself) and Helpling Premium (where a professional cleaner comes to be employed by a cleaning company).

In the end, the platform did not live up to expectations: the receiver's response stated that 10 employees were working for the platform on the day of the bankruptcy.

Looking back: what was new about Helpling?

In the aforementioned blog from 2018, I asked the question: what's new about Helpling? I wrote:

"Platform Helpling enables households to find a domestic cleaner through an online marketplace. In addition, the platform supports with quality control (by phone intake, ID check and reputation system), back office (complaints, queries, replacement help in case of holidays) and payment module. The company makes money by skimming a margin from each transaction. A common used model.

Helpling is certainly not unique; this kind of service has been around for years. Take the company HomeWorks. A company that has been doing exactly the same thing as Helping for 25 years. The only difference is that this is not an online marketplace, but the company manually matches supply and demand. It also has service coordinators who oversee quality and even go with the initial cleaning. All this makes it an expensive model: the commission the company charges is therefore around 39% (compared to 23% at Helpling)."

The platform competes with intermediaries with a similar model and strategy that have been operating in the market for many years. It is therefore questionable why the platform was then suddenly seen as an undesirable player.
Where was the informal market in the debate?

In the lawsuit against Helpling, FNV pretends that the platform competes with cleaning companies. That, of course, is nonsense. The only thing Helpling was competing with is the informal market. My prediction in my earlier blog: "if FNV wins and Helpling has to employ domestic workers, the price of help for the customer will go up to a minimum of 25 euros. Practically no individual is willing to pay this amount per hour. No one will book through the platform anymore and Helpling will be bankrupt in no time. Does that solve the problem? No. That cleaning will continue to be done afterwards. But then through the informal market." And that is exactly what will happen now.

And that is a shame. The case concerning Helpling could have sparked a debate on how we as a society value work in and around the house. That is a discussion we should have in society that should culminate in a political choice. Because that's what it is: a choice. That politics can indeed do something to improve the situation is proven by the countries around us: Belgium, France, Scandinavia. There, the government subsidises this kind of work, in Belgium, for example, through service vouchers. I talk to many people who recognise that the current Dutch rule is worthless and we should introduce a system like in Belgium. But they fail to get the issue on the agenda. Apparently, a structural debate is less interesting than a random court case.

Towards a broader view of 'mediation' for home cleaning

With today's knowledge, I want to dig a little deeper into what was new about Helpling. The fact that it is a platform made Helping distinctive from other ways for private employers to find domestic cleaners, but the question is what impact this has on workers' autonomy, satisfaction and income. The moment something is new, my first question is always: how was it done before? And in what areas does the 'new' make things better, worse or stay the same. In the case of Helpling: what are the conditions of domestic cleaners who find their private employers through other channels? There are always different ways to find a domestic cleaner: via via via, a note in the supermarket, via social media, mediation websites and through platforms.

Quite coincidentally, on the day of Helpling's bankruptcy, I attended an academic workshop on 'platform work' organised by Koen Frenken of Utrecht University. One of the speakers was Juliet Schor, economist and Sociology Professor at Boston College. Schor has some very interesting research in the sharing and clustering economy to her name and presented a study in Utrecht where she interviewed workers in private home care who find their clients via a Facebook group or via a platform. She was curious whether these workers would experience their work and conditions differently.

She found that the outcomes between the two ways of finding work were very similar. The only differences she discovered were traceable to differences in the workers' circumstances, not the way they had found work. Workers' income was higher through the platform than through a Facebook group and "our interviews found reveal high levels of work and platform satisfaction among these workers". To which Schor did add the note that the 'market conditions' for this target group are predominantly positive.

Her conclusion: "We conclude that factors external to the technical functioning of platforms, such as market and regulatory conditions, workforce composition and the nature of the work, should be more systematically examined to understand what forms of control platforms themselves can be subjected to."

Conclusion and proposal debate

As predicted in 2018, Helpling is now out of business and cleaners have to find their way back into the informal market. The fact that there has never been a serious attempt to explore how a central platform can improve the situation of a vulnerable group of workers without becoming an equal employer is a missed opportunity (and responsibility). Not only of FNV, but also of all the institutions that are responsible about this agenda and have not spoken out. That FNV is claiming a belated victory is sour, because you have to wonder who won. Certainly not the cleaners and certainly not the debate.

I hope that this case (and this blog) will encourage policymakers, but also trade unionists, to still take up the important questions surrounding this group of workers. That research is now done into the differences that the different ways of 'matching' entail and to see how to really do something for this group of workers. This starts with (re)recognising the context, to make (political) choices along these lines. Perhaps the closing of Helpling's chapter will now free up space for this. Let's hope so.