There has been a buzz recently about the idea of a Universal Basic Income (UBI) and why it could be a solution to technological unemployment.
VOX reported a few days ago that Y Combinator, a start-up incubator, is about to start a five-year research project on how a UBI could work. Noah Kulwin over at re/code also has a useful article on this including a broad overview over existing work on the basic income. In the spirit of discussion I’d like to bring together some of the arguments I have made over recent years for why the basic income is not a suitable solution for technological unemployment but another option – a job guarantee – could be. So let’s start with the assumptions.
The idea of a basic income now being discussed assumes that people have to live on it entirely as work effectively disappears for most people other than the most highly skilled. This presumes a reasonably high level of income and that most work will indeed disappear. I am not so sure about this. There are wildly different research results about how many jobs technology will destroy and/or create and the honest answer is: nobody knows. There will be displacements and job losses and policy-makers need to prepare but we don’t really see this yet in the productivity figures. If anything recent experiences in the US and the UK in particular point to a slowing down of productivity increases although there are issues around measuring methods. We are also currently witnessing record employment numbers in Germany so the big labour market impact has not yet materialised and it remains to be seen how it will affect different economies in the future.
That said there seems to be general agreement that specific areas of work that rely on problem solving capacity, creativity and social capital as well as personal services are relatively safe from automation, at least in the near future. This to me suggests that work as we know it (wage labour) might well be significantly less than we are used to now but it is unlikely to disappear for most people. A reallocation of the remaining work through fewer working hours seems to me to be the first step that should be considered in this context. But I don’t think you should assume at this stage that most work (what is most anyway? 70%? 80%? 90%?) will disappear any time soon. Clarifying these differences in assumptions is essential. So what is wrong with the basic income? Here are my five main criticisms.
Five reasons against a basic income
- It reduces the value of work to mere income. It is of course very welcome if back-breaking or otherwise bad jobs can be automated and disappear. But we should not underestimate the social value of work. It is where most people spent much of their waking hours and where a lot of social interaction takes place. Work is not just wage labour and there is value in preserving the social aspects of work.
- It is an inefficient use of public resources. The UBI is not means-tested and is paid to everybody in society, from the social benefit recipient to the billionaire. As a significant part of society doesn’t need this income this watering-can approach distributes public resources (most likely most of it) to people who don’t require it. The general idea is that this money can be recuperated via the tax system but this is much easier said than done. If you look at how tricky it is even today to tax the rich you can be forgiven for assuming that you won’t see much of that money again once it is paid out.
- It doesn’t solve the inequality issue – it just shifts it. The progressive idea of a basic income assumes that it frees people from (bad) wage labour. The UBI makes sure that you can lead a decent life and if you would like to work on top of this you are free to do so. Well, if the point of departure is that many jobs might no longer be available so you are not free to choose to work a few hours on top, you are in trouble. In this case a lot of people would be stuck on whatever the basic income level is and the rest, the ones who can still work and benefit disproportionately from productivity gains, would run off with the spoils. As social inequality is relative and not absolute, a UBI would only shift the level rather than help to eradicate inequality.
- It won’t work in the EU. At the moment there is no real evidence that migration within the EU is driven by different welfare provisions in different countries – people move around to work, not to claim benefits – but this would probably change if a country introduced a basic income. Under non-discrimination rules every EU citizen who is legally a resident of a country could presumably claim the basic income. Now assume Germany were to introduce a 1000 Euro a month UBI – it is not hard to see what kind of pull factor this would create.
- It comes with the abolition of the welfare state. At least potentially that is. What worries me is that at least some Silicon Valley based proponents of the UBI argue that the way to pay for it should be by dismantling government-provided services. This in essence means the abolition of the welfare state and of the collective securing of life risks (such as unemployment, ill health and old age). This might sound appealing to libertarians but is completely at odds with what we understand as a social market economy in Europe. Read this paper if you interested in more of these issues.
Is a public job guarantee a solution?
On the basis of these issues that I have with the UBI, what would a more practical solution look like? Given that we don’t know the exact impact of the digital revolution on work, thinking along the lines of a public job guarantee seems much better to me. To be clear, this does not mean forced work or necessarily direct employment by the state. People can get work if they want or need to and could be employed by a variety of institutions paid for by the state. I am thinking in the direction of the state as the ‘employer of last resort‘ model that creates meaningful employment at a decent wage if the private sector does not provide it (have also a look at the second part of this interview). What would the benefits of this direction of travel be? Here are six points:
- A job guarantee does not reduce work to just income. The important social aspects of work I mentioned above are preserved as some level of regular employment is guaranteed by the system.
- It is a much more efficient use of scarce public resources. The job guarantee kicks in for people who ask for work and doesn’t allocate resources where they are not needed.
- It preserves and develops workers’ skills. Nobody knows how people would spend their time on a basic income handout but keeping people in work surely helps to maintain and improve their skills – especially in an age in which permanent upskilling is crucial.
- It is easier to implement in the EU. Implementing a job guarantee in the EU won’t be easy either but we already have policy initiatives – the EU youth guarantee – that go in a very similar direction. So politically we are much closer to a job guarantee than to a basic income.
- It retains the character of the social market economy. There is no need or threat to dismantle welfare systems altogether.
- It provides an additional public policy tool. This is a crucial argument for me. If the state guarantees employment it can create employment in areas that are underserved and important for other reasons. If you look at the demographic makeup of our ageing societies for instance it is quite easy to see that we will have labour shortages in the entire health and old age care sector. This incidentally is one of the sectors where the real added-value of work is rooted in social capital – or would you like to see robots console terminally ill patients? The ability to guide employment creation in this way might prove to be a very useful tool to address other social problems.
There you go. These are the reasons why I would say that thinking along the lines of a job guarantee makes more sense than pondering a basic income. The debate has only started and it will run for quite some time…
This article first appeared at Social Europe and is reproduced with permission