Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

Saving on bureaucracy is a frequently made argument that I have never seen supported by a quantitative analysis. Do they exist? Such a claim seems untrue on its face.

You can't eliminate the welfare bureaucracy with UBI because:

- You won't want to be giving free money to literally everyone, only people within your jurisdiction. So you still need to figure out who those people are and whether they're authorised to receive UBI.

- It will act as even more of a magnet for illegal immigration, so you'd still need to spend on border reinforcements. Same problem as they have now in Germany and Sweden: lots of people turned up who can't work and sit on welfare forever. If you don't stop this it's in reality a policy of giving out enough free money to live by western standards to the entire world, which isn't how it's being advertised.

- You still need to deduplicate everyone and ensure people can't register for UBI over and over.

- You need to stop people creating fake identities to receive UBI for people who don't exist.

- You still need to ensure UBI stops being paid when someone dies, even though families and friends have strong incentives to stop the state finding out about a death.

- You still need to decide how much the UBI is, how much it should change, and whether the amount should vary by location. What's plenty for someone in the countryside may be considered insufficient in the city.

- You still need to track where people live, the moment you give way on the amount being adjusted regionally, which you will because the sort of people who most enthusiastically support UBI are all-in on the type of "fairness" that invariably means not treating people the same. Or are they going to advocate for a single income tax rate across all earning bands as well?

- You will need a huge rise in taxes to pay for this, meaning tax evasion and avoidance will also go up quite dramatically. You'll need to bulk up the tax bureaucracies to handle this.

UBI is just a modern rebranding of what in the 20th century was called the socialist utopia. People tried it, it doesn't work. That was Russia at the start of the 1900s. UBI advocates today seem to have forgotten about all that, but they're re-treading intellectually dead ground.



> "So you still need to figure out who those people are and whether they're authorised to receive UBI."

You need to figure out who they are to be able to give them money in the first place. That's not the thing you're saving on, but all the other checks to see if you're justifiably unemployed/disabled enough to be receiving this money.

A big problem with the welfare system is that people only get it when they don't work. They lose it when they get a job, which can often mean they actually lose purchasing power, so they're effectively being punished for working. UBI creates a system where you always get ahead by working, even if it's only a little bit.

Of course you still need the basic civil bureaucracy; every country needs that. But you don't need an extra layer of bureaucracy on top of that.


Yet people routinely choose to get off welfare and get a job, so clearly, the thresholding can't be that bad. Obviously there's going to be some cases where some jobs and welfare package combinations don't make financial sense in the instant, if you assume no chance that the job will ever lead to a pay rise. But it works well enough most of the time.

I've argued above that in fact you will still need a complex welfare bureaucracy that looks very similar to the one we have today, given even basic constraints like "there should not be excessive levels of fraud". Deciding if people have a job or not is a trivial cost compared to the entry-level costs of just making sure you're paying people who exist and you only pay them once.

Again - does anyone have a rigorous, convincing analysis of the cost of means testing specifically? Perhaps I'm wrong, but I don't think so. Especially because people who argue for UBI do not argue for the abolition of things like disability benefits, government subsidised health insurance and various other means-tested subsidies.


Some people do, some people don't. The threshold is surmountable for some people, but it could be a lot lower.

> "you will still need a complex welfare bureaucracy that looks very similar to the one we have today, given even basic constraints like "there should not be excessive levels of fraud". Deciding if people have a job or not is a trivial cost compared to the entry-level costs of just making sure you're paying people who exist and you only pay them once."

I strongly disagree. For one, there's a lot less room for fraud if it's something that everybody gets. Secondly, the government already needs to know that people exist. People get born, they go to school, they pay taxes, they vote, they die.

> "Especially because people who argue for UBI do not argue for the abolition of things like disability benefits, government subsidised health insurance and various other means-tested subsidies."

I would prefer those not to be means-tested, but equal for everybody. If you're disabled, you get disability benefits on top of your basic income, and you get to keep them even if you do manage to get a job despite your disability. Similarly, I'd like to see health insurance as part of that basic income.

The only place to test people's income would be for paying taxes. Basic income takes care of everything else. (I suspect even progressive taxation might be replaced with a single flat tax rate in the basic income is high enough.)


Disability benefits are means tested by checking if you're disabled, so I think we're talking at cross purposes. Abolishing means testing for that would mean abolishing disability benefits entirely. It'd just be rolled into general welfare or UBI. Is that what you want to see?

What about income tax? Shall we set a single income tax band for everyone? You say no but where's the consistency? Means testing is good when taking money but bad when giving it? Why?

The logic of a strictly egalitarian universal welfare system would also suggest governments be strictly egalitarian in all ways, meaning all taxes would be poll taxes. Everyone treated the same. But nobody is arguing for that, which makes it look like they're not working from a general principle of lower bureaucracy and more equality, just a general love of welfare.


Is means testing not about income? I mean to take income out of the equation, so people don't lose benefits when their income rises, because that would create perverse incentives. If you manage to hold a job despite being severely disabled, more power to you. That's not something that should be punished.

I think I already addressed all of this in the comment you're responding to: if UBI is high enough, I think there could indeed be a single tax rate for everybody. Because the UBI would already provide a comfortable living income, and only income for luxuries would be taxed.

But I think UBI will start lower than that, which means that at first, progressive taxation will still be necessary to some extent.


Alright, I've been using means testing to mean, a test of eligibility. But I just checked the dictionary and it's indeed defined in purely financial terms.

So we can add to the list of welfare bureaucracy that'd still be needed, disability tests. That's where a big part of the fraud comes from in any welfare system, because fraudulently obtained disability benefits is effectively a form of UBI: for those who have it, it never ends even if they get a job, unless the job is one that their pretend disability would prevent them from doing. Very attractive. Also, doctors and assessors don't pay for it so they tend to be willing to grant it to people in very lax ways.

This can be seen in the way people on disability benefits rises and falls with the economy in the USA.

As for a "comfortable income", again, you're dreaming. For everyone to have a comfortable income without having to work, with extra income only for "luxuries", isn't physically possible. It's literally the original communist vision that was abandoned because it didn't work. Comfortable income was simply redefined back then as subsistence income in a broken economy, and luxury was not only reserved for Party members but redefined as a standard of living lower than that available to ordinary people in the supposedly deeply unequal west. UBI is just centuries old political ideas re-heated and re-branded.


> Again - does anyone have a rigorous, convincing analysis of the cost of means testing specifically? Perhaps I'm wrong, but I don't think so.

We don't actually need a rigorous analysis of the costs of means testing to conclude that arguments that extending a benefit from <10% of the relevant population cohort (e.g. an unemployment benefit with eligibility testing) to 100% of it (i.e. all working age population) will cost more than the testing, unless governments spend 90%+ of the relevant component of their welfare budget on admin rather than payouts. Common sense as well as the published statistics confirms they generally don't.


Yes, you're right. I'm being rather over-generous to the UBI case here.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: