[mike_hearn]

Someone needs to make a website explaining why UBI doesn't work conceptually because this comes up over and over again on HN.

You cannot make UBI work from money saved by removing means testing. Even UBI is a welfare scheme and would require significant bureaucratic hoop jumping to check that a person claiming it isn't:

• Dead

• Non-citizen

• Already claiming it under a different name/bank account/etc

• In prison

• Moved abroad

and so on. All that is expensive, and yet the overheads of even existing welfare systems just aren't high compared to the amounts they pay out. Getting rid of means testing doesn't magically make the numbers balance.

Geohot is correct. UBI seems to only appeal to people who don't understand how the economy works. You can't have an economy in which one person earns all the money by definition.

[deaux]

This is a complete non-issue in basically every wealthy country bar potentially the US, all five things you named are already known to the government at all times. They also apply the exact same way to any other scheme, there's nothing new about it.

[mike_hearn]

Quite a few governments have trouble verifying identity reliably. But to the extent they can do it, it's because there are lots of people employed to do so. The UBI thesis outline above is that you can find the money to pay for it by eliminating all those job roles from the government, so you can't use their existence to justify UBI as affordable.

[deaux]

Let's look at other wealthy countries with large economies.

Germany, Japan, UK, France, Italy, Canada, Spain, Australia, South Korea, the Netherlands, Switzerland, Taiwan..

Which of these have trouble verifying identity reliably, to the extent that it would be a meaningful obstacle to UBI?

[mike_hearn]

Even if governments were perfect at ID verification it wouldn't change the argument above, right? Being perfect at verifying UBI eligibility would require a large government infrastructure, just like today, so you can't claim that the U part makes it super cheap to administer.

But no government is close to perfect. Here are some examples for your edification.

The UK doesn't even know how many people are living there, and it's an island. There's no centralized identity scheme and during COVID more people came forward for vaccination in some age bands than theoretically existed at all.

Germany fails at reliably verifying that people who turn up for a language test as part of naturalization are the same people being given citizenship: https://www.infomigrants.net/en/post/69787/germany-police-ar...

All countries struggle with basics like "is the recipient of the welfare dead". Here's a specific case where Italy didn't notice it should stop paying out a state pension (a form of UBI) for years after death, with the fraudster only getting busted when he tried to dress up as his own mother: https://www.theguardian.com/world/2025/nov/25/italian-man-dr...

Paying out money to dead people is a very common problem. Here's an EU report on all the basic ways countries get defrauded by failing to track basic facts about identities:

https://www.ela.europa.eu/sites/default/files/2025-03/SSC_fi...

"Common fraud and error cases include falsified documents (birth, marriage and death certificates), identity fraud and falsified non-payment certificates"

Even in UBI, there would be a strong expectation that each person only receives it once. But checking stuff like that requires a huge bureaucracy.

[tonyedgecombe]

>Here's a specific case where Italy didn't notice it should stop paying out a state pension (a form of UBI) for years after death, with the fraudster only getting busted when he tried to dress up as his own mother

I assume you want to stop the state pension as well then?

[deaux]

None of these include a mention of meaningful economic impact.

> Being perfect at verifying UBI eligibility would require a large government infrastructure, just like today, so you can't claim that the U part makes it super cheap to administer.

It's already in place regardless of UBI, so it doesn't add meaningful costs.

Of course it's never going to be perfect, absolutely nothing is. Why even mention that? What matters is the impact of it being imperfect.

Compare it to the impacts of tax evasion, or wage theft, and they'll be completely negligible.

[bombcar]

> Even in UBI, there would be a strong expectation that each person only receives it once. But checking stuff like that requires a huge bureaucracy.

Or you issue UBI all at once during the month, and you stamp everyone who receives it with an indelible ink mark that takes longer than a day to wear off; like they do in poor countries to prevent double voting.

It's a solvable problem. The problem is that the "cost of managing welfare" is a small percentage of the cost of welfare, you can't pay for doubling/tripling it by saving 5%.

[swiftcoder]

Most EU countries have national IDs, so the "only receive it once" is a solved problem.

The "still paying dead people" problem exists in the current pension system, so we already have bureaucracy in place to solve that one (yes, it's not 100% accurate, but it works sufficiently well en masse) so no need for new bureaucracy there.

[graemep]

In the UK a lot of that is solved by using the NI number that everyone has to have to work, claim benefits, get a state pension, or pay tax.

For people who are employed it could be done by existing systems already used to calculate tax (which is deducted automatically by employers here so the systems to do calculations exist).

Self-employed people already have to register with HMRC.

For the rest it is a far, far simpler than the requirements of the benefits system and less prone to fraud.

> You cannot make UBI work from money saved by removing means testing.

No one claims it can be made to work using ONLY money saved from means testing. Something like removing means tested benefits together with lowering tax thresholds could work though.

[imtringued]

UBI in the form of CO2 dividends works extremely well. Be more creative about the potential applications of the pay out mechanism.

[jrimbault]

Except a lot of actual, very smart, economists are for UBI or similar arrangements (it's not a settled matter). And geohot might be smart, but he's just a self described hacker.

If we're going to use authority arguments.

[tonyedgecombe]

I think the problem is that a lot of the proponents are arguing for a level of UBI that is pretty close to the median wage whereas what would be affordable is probably a quarter of that.

[fragmede]

I'll come out and get kicked out of communism club to say that I don't support UBI on the basic fact that money is exchanged for goods and services is just so foundational that I can't support UBI. I think everybody should have a roof over their heads and 3 square meals a day, but UBI isn't the way to get there.

[UncleMeat]

I am not aware of any mechanism to ensure that everybody has a roof over their heads and 3 square meals a day besides direct payments to the poor.

[fragmede]

Welfare and SNAP may be imperfect, but they seem to be working.

[UncleMeat]

Welfare is a direct payment to the poor and SNAP is very close to that. Work requirements and other administrative hurdles are the primary thing that keeps these programs from truly ensuring that everybody has basic dignity.

[maybewhenthesun]

> money is exchanged for goods and services is just so foundational

it is in all large societies. That's true. But it is not in most small primitive societies. SO it's not like a law of nature, but more that we haven't found a system that works as well for large groups.

The trouble is that capitalism also has it's problems, and they're getting exacerbated by technological advancements. If you automate everything at a certain point there's just nothing to do for a large part of the population. And at that point the system stops working.

In science fiction we would get the 'post scarcity society', but nobody knows how that should work.

The UBI systems I've seen proposed that just might work are a sort of golden middle between those two. Not that different from the current welfare system we have in NL, but taking out the stress factor and stigma of receiving welfare.

[DangitBobby]

You don't support it on principle or because you don't think it works?

[fragmede]

The exchange of money for goods and services is foundational to capitalism, and UBI seems like a divide by zero kind of trick that's isn't going to work out. Let's fire a bunch of people to make the government more efficient with this one neat trick is just the most Republican thing ever.

[Obscurity4340]

Well what is then? Respectfully, please have an alternative or otherwise how is this not astroturfing

[fragmede]

Astroturfing? If I don't have an alternative, I'm secretly being paid by "them" to tear down UBI? Who would "them" even be? How would that even work?

Anyway, subsidized jobs programs is my answer. Pay people to do jobs. Plant trees! There's so many places that could use some reforesting. There's no shortage of work to do.

[tartoran]

I think the whole premise for UBI is that there won’t be any jobs left, jobs to make a living at least.

[flammafex]

Guess we'll starve then. Good luck dealing with hundreds of millions of hungry angry people.

[mike_hearn]

The idea you're suggesting here is 19th century era Marxism, and isn't based on historical or economic realities. There has never been a famine caused by new technologies creating unemployment, and food security is much higher now than at any other time in the past.

[maybewhenthesun]

> There has never been a famine caused by new technologies creating unemployment

That's a bold statement. And depending on how you define famine it might even be true-ish. But I'm quite certain most poor people in the 1920s would disagree with you.

Which is why we got the russian revolution and both world wards at that time. All 3 were partially caused by 19th century industralisation creating an poor, hungry underclass.

[mike_hearn]

It's not particularly bold, it's just a fact. It only seems bold because our society is still flooded with Marx-era propaganda.

Poor people in 1920 would agree with me. We know this because in this era Marxist ideas were already quite old, well known and widespread. He argued for what was effectively a UBI or very generous welfare system under a different name. He predicted the poor would rise up as capitalism drove their wages to zero and "alienated" them, then they'd overthrow it. But Marx became bitterly disappointed in his lifetime when capitalism didn't fail, wages didn't fall to zero and the working classes rejected his call to revolution.

Technological progress hasn't caused famines, it's ended them. What has caused several famines, though, is left wing people reorganizing society via force to solve imagined problems. If you're worried about people not having enough to eat, you should be terrified of a UBI.

[flammafex]

As an Irish person, I want you to explain to me how "left wing" the British were when they had us export our food at gunpoint.

[mike_hearn]

You're misinterpreting what I wrote. I didn't say all famines were caused by left wing policies. Obviously historically most were caused by natural disaster, like the Irish potato blight. But when famines have been caused by social change, it wasn't due to the march of technology.

The British never forced anyone in Ireland to export food "at gunpoint".

[flammafex]

https://www.ers.usda.gov/topics/food-nutrition-assistance/fo...

"Food security is much higher now than at any other time in the past", huh?

18 million households in the U.S. were food insecure in 2024.

And that's from 2024. How much was oil then?

[tartoran]

Now with ICE and tarrifs things are going to hit really hard.