Hacker News Reader

Comment by dragonwriter | Hacker News Reader

> Biggest issue with disability is you cannot start working again, even as an experiment, without losing your disability. It’s a lifetime disability.

One of the worst things about the US social safety net, almost as bad as how weak it is even in total and in theory, is how piecemeal it is. Workers Compensation (and, for veterans, the VA) provide benefits that apply to partial and temporary disability, whereas Social Security provides benefits for only long-term total disability (that both prevents your past work and prevents transitioning to new work.) But evolution in jobs and working conditions, evolution in medical and assistive technology, and, frankly, refinement of knowledge about the initial injury can change what once seemed to be in one category to the other, and vice versa.

View on HN

The whole thing is one giant trap of playing the game right, or falling through the giant cracks never to be seen again. Worse than that, all the means testing and adjudications and determinations are actually /expensive/ - mountains of paperwork and appeals and this and that and the other all because the US is simple obsessed that someone, somewhere, might be getting something they don't deserve.

(Without, ever, of course, facing that there are a lot of people getting what they didn't work for, they aren't generally trying to scrape by on a disability check or some unemployment though)

We can't have nice things because of massive fraud. There is always tension between ease of receiving services and those looking to exploit the benefit.

We can and should do better, but it cannot be perfect.

Which still happens?

If people think TSA is theater…

> We can’t have nice things because of massive fraud.

Creating multiple separate bureaucracies, often with similar functions, creates more fraud vulnerabilities and more cost to preventing fraud of any given type.

Our system is not designed to prevent fraud, but to enable it.

Yes, exactly. Simplicity is what prevents fraud, not more complexity. Our systems need to be fundamentally redesigned for simplicity at the core.

Comment from a heretic economist was that in the US you have gnashing of teeth and rending of garments over raising capital gains a few percent or the marginal tax rates for people with seriously enough income. And yet for low income people the loss of benefits that comes with earning a bit more money can by on the order of 40%.

Whether intentional or not, the system is really good at keeping poor people poor and stuck, and rich people richer and mobile. It is anti-human.

Some metaphors of demon-lords with hordes of indentured servants come to mind. The servants bound in fiery chains mucking through tar, and the demon-lords flying around with fiery wings between castles.

Won’t take many more steps for that imagery to be real. It would seem making money our collective master may have had some unintended consequences.

> We can't have nice things because of massive fraud. There is always tension between ease of receiving services and those looking to exploit the benefit.

Massive? Source?

Imagine the government put a giant bowl full of money in front of every city hall, and said that you should feel free to take from it if you're starving for food. It's enforced by the honor system. This system is extremely easy to use and efficient in delivering funds. How long does the bowl last, and what proportion of funds go to the intended recipients?

Obviously, not long and not much, so we build systems around the bowl of money to prevent fraud. These systems inherently cost money, making it less efficient, and reduce ease of use, meaning people who genuinely need the money have a much harder time accessing it.

SSDI in particular is rather infamous in that it's rife with fraud[0] but also extremely difficult for the genuinely disabled to utilize[1], to the point where it's pretty accurate to say that the SSA is willfully blind to people in genuine need. The article we're all talking about is an example of that.

[0] https://www.nationalreview.com/2013/06/rogues-gallery-social...

[1] https://www.kshb.com/news/local-news/lawsuits-lay-out-accusa...

So what you've done here is invent a hypothetical, state by fiat that it would end poorly, then used that complete fiction to argue that we build systems to prevent fraud then point to massive fraud in the system we built to prevent the fraud that you stated we need to build to prevent it.

Like I dunno, maybe if the thing we're doing recursively fails to work we could try the other thing for a bit and check.

I'm pretty sure you don't actually reject the outcome of my hypothetical.

And, yes, fraud is endemic, we try lots of systems, and some work better than others. Building ones that attempt to satisfy multiple constraints at once makes things harder.

I do actually. Although it's pretty easy to reject because both the hypothetical and the outcome are pretty vague like obviously a big bowl of cash is gonna lose most of it's money to the wind. But to be clear I do think a system where any one could receive money without first demonstrating "need" could be successfully continuously funded and would help a lot of people that "need" help. Although in my system "need" is in scare quotes because I don't think that its only the starving for food that have a moral claim to aid.

I'd also point out that in a system where I just ask the mayor for a hundred bucks and he gives it to me, there can't be fraud. There's nothing to defraud. It's only once we start adding shit like "prove you're poor enough" or "prove you can't get paid for work" that we have something to commit fraud about. Indeed the more constraints you add the more frauds become possible.

At the very least, you shouldn't be spending more on gatekeeping to prevent fraud than you'd lose to said fraud. I'm pretty sure that our system is way over that line.

It's not all because of fraud.

There's also a lot of crossover between the people defining the programs and the people administering the programs. So programs get rules written in a way that the software from the firm they are close to can handle it, but are hard for their competition to work with. None of it is actually defined with the taxpayer and beneficiary's interests at heart. (Source: worked for an agency in the space and saw it first hand).

Bit odd that the US tolerates, even celebrates, massive tax avoidance but spends a lot of money on preventing relatively small sums of benefit fraud.

This doesn’t seem correct to me. We can’t have nice things because of fear of massive fraud. I’ve heard that programs like SSDI have around a 1% fraud rate. I’ve also heard they deny something like 70% of claims on first application and 90% on appeal. That seems like way too much denial for a system that’s doing pretty good fraud-wise.

A high false positive rate doesn't imply a low false negative rate. If you're building a well-functioning system you have to consider trade-offs between the two, but a badly-functioning system can achieve both high false positives and high false negatives.

Reply on news.ycombinator.com