I can't find the original tweet, but someone (half?) jokingly proposed a law that all benefits must be defined as continuously differentiable functions (thus making cliffs impossible).
"Yeah, I made $1M last year. Here's my SNAP check for six cents."
They should be. And for the great majority of citizens, they should be calculated automatically along with taxes without the need for filing/paperwork. But we must think of the Intuit shareholders and the harm that that would cause them.
If someone in the coming years ran on taking down regulatory capture and returning that as social safety net funding, public goods, and lower taxes, and had the chops to actually deliver, they'd do well.
Though best to start local government first, obviously.
We're reaching breaking points in so many places...
“They’d do well.” That is not the lesson I’ve taken from the past 15 years of politics in the US and abroad.
Political candidates actually interested in taking on the very difficult and nuanced task of governing are routinely drubbed by edge lord culture warriors or candidates that simply promise the world without any regard to annoying facts like existing laws, budget deficits, or basic tenets of economics.
I definitely agree there is probably less disfunction and greater chance for reform at the local level.
As someone who formerly worked in this space, the issue is administration is orthogonal to legislature.
Administration and implementation is inherently technocratic in nature, but legislation is is often driven by short-term electoral needs.
This isn't to say autocracy is the answer (it isn't), but the dysfunction arises when people assume that every single administrative decision needs to be made by "elected officials" and constantly second guess administrators.
When administration from local to federal becomes politicized (as is increasingly the norm across the democratic world), implementation slows down severely because the "ideal" solution might not be political tenable.
A good example of this is welfare expansion in the US - the assumption is lower income voters who voted for Trump voted irrationally, but in action, the majority of Trump voters tended to be in the 25th-75th percentile income bracket, which in most cases put them outside of the bracket for a number of social services. Those services which were available such as the ACA/Obamacare have been protected by legislators for that very reason because they know they would lose their seats as a result.
You're in Japan so you've probably seen similar decisions being made with regards to Japanese rice protectionism [0] - unpopular with urban voters, but urban voters don't swing elections at scale.
Japan is 93% urban. 25% of the population lives in Tokyo! It's not that they can't outvote the farmers, it's a more complicated mix of agrarian nationalism that supports farm subsidies in a lot of countries, plus outright bribery with literal sacks of rice.
> administration from local to federal becomes politicized
Administration is politics. It's a mistake to go full technocrat and proceed without the consent of the administered. The problem comes when techocrats get rings run round them by misleading populists.
Japan is a parliamentary system, and the amount of Single Member Seats in the House of Representatives needed to flip an election are primarily small towns or rural.
This is what happened in the 2009 election when the LDP lost the farmer vote.
There's actually plenty of examples out in the real world with competent administration to learn from. Especially if you generalise enough to look at how pockets of competence work even in an otherwise abysmal system, instead of demanding overall competence.
What you end up seeing is those cases of "competence" only worked in cases where administration and legislation was aligned. But even in those societies you'd still see problems which are distinct, but problems nonetheless.
GP might be saying that analyzing _differences_ in how perf targets are met in say, VN&CL (or SG&CH IE&HK TW&SI KR&DK) might be most productive. As you mention, in particular, how are their sad paths ("problems") different? Or the same?
Now there are subdepartments of study devoted to this very question (empirical study of the legislative-administrative divide) , but in the US I'm hard-pressed to list the corresponding think-tanks :) I suppose some MBA level depts in the US will have to suffice
I wish in the US more areas of policy would be decided at the state level, and I wish more state government would flip coins to decide on their policy. (Even better, if we can push it down to county level.)
The first part is about subsidiarity, which is a good idea anyway. But together this is just a tongue-in-cheek plea for doing more randomised, controlled experiments. (Alas, we can't blind them.)
> I wish in the US more areas of policy would be decided at the state level, and I wish more state government would flip coins to decide on their policy. (Even better, if we can push it down to county level.)
The overwhelming majority of policy is decided at the state and local level in the US so long as it's not foreign policy or monetary policy related.
Whatever is decided on the Hill has no bearing once administration and implementation comes to play - which is overwhelmingly done by state and local government.
This is why the US has become increasingly dysfunctional over the past few years - state and local elections which were previously staid affairs became polarized partisan affairs because turnout is low and the overwhelming majority of decision making roles are elected.
Flipping GP's tangent on its head (after all states in the US are akin to countries in the EU/EA:)
If you can't blind them find them
I suggest that, absolutely, state governance has _already_ generated lots of excellent data >> federal level :) probably much more accessible than corporate governance data )
Of course YC's Seibel might have something to say here, it's probably not a new perspective
I wish you were right. NYC can't even charge people for their using their roads (whether that's driving or parking) without the state and federal governments getting involved.
The problem isn’t that such candidates would never accomplish anything let alone get elected due to the insurmountable amount of resistance from so many incumbents with a vested interest in that not happening, ever.
One of the biggest problem of a two-party system is that the two parties are thoroughly captured by lobbyists.
In a PR system, fresh parties do arise over election cycles, and it takes some time for them to be thoroughly infested. These can then push for some reforms that threaten entrenched interests, and sometimes succeed.
Most people don’t even know what regulatory capture really means. There’s no “brand” to rally around and I don’t know how you’d go about building one.
And without the concept spread through the population, where would you find the grass roots support you’d need for resisting the avalanche of interest-group pushback?
“Draining the swamp” or “revolving door” were sorta in the neighborhood, but still ineffective and counterproductive.
DOGE was a Potemkin organization. They destroyed USAID, smashed up a bunch of other institutions, then went home. Very little of which complied with Federal funding law, either.
I like my cliff. I earn enough that I and my family are ineligible for most welfare schemes. I do not want even 6 cents worth of SNAP. I spent my entire childhood on that, and it disgusts me. Thus, were this policy ever seriously proposed, I would do what I can to dissuade my legislators from voting it into law.
Right, so people who would use their trauma to prevent themselves from thinking logically about the problem and helping those in the middle instead of some arbitrary line your mind made up?
Never underestimate how many good policies go nowhere because some median voter is irrationally mad at them due to something that happened decades ago.
Because you live in a world where other people exist, and how they go about existing can impact you negatively, both directly and indirectly. The fewer people living in and dealing with poverty, the better off the everyone else is.
>so people who would use their trauma to prevent themselves from thinking logically
You're using your trauma here. You keep yearning for the kindergarten lessons to be true, that we're all the same people and we should all share and all the other nonsense that was only taught to you so that the elemntary school teachers didn't have a city-burning-down-riot on their hands every day.
Even if you insist that I'm the one being illogical here, it's wired pretty deeply into my brain, and I'm not going to change. Not only am I not going to change, I'm raising children to be like me, and one of the early and fundamental lessons is that fertility rates are top priority. Be fruitful and multiply. There will always be more like me than more like you, and it's only going to get worse (for you and those like you).
There is idea behind that, but continuous is not enough.
The variable is all transfers, taxes and benefits T = [all taxes - all benefits] as function of income per person (including children). T starts negative (benefits are negative taxes).
Goal: monotonously increasing effective marginal T rate.
This. No need for continuity in the derivative - the marginal tax rate should be fine to jump around - but should start low (I'm fine with negative) and increase as you earn more. Like US federal tax brackets, but with benefits also considered.
Continuously differentiable is fine with jumps in that derivative.
I agree in principle. And I also suggest that we should probably only have one bureaucracy that does means testing that can then administer both taxes and benefits. No need to duplicate the effort.
However most benefits and taxes aren't just means tested against income, but a myriad of other conditions. So net transfers aren't just a function of income.
To calculate the amount of your childcare benefit take the negative of the your adjusted gross income (box 26a), divide it by 10,000, and add 2. Then take the arctan of that result and multiply by 5,000. Add 2500π and write the result in box 46. You may round the value of π to 2 digits. If you are filing as head of household, divide the amount in box 46 by the definite integral from 0 to your adjusted gross income (box 26a) divided by 10,000 of sin^2(x)/(x^2 + 1) dx, and write that amount in box 48. If you are unable to find the function of the anti-derivative, IRS rules allow you to approximate with a Riemann sum using the midpoint rule and a rectangle width of 0.1.
I would guess that was not a joke. The benefits could drop all the way to 0 at a reasonable point unless you impose the stronger condition that the function be analytic.
You jest, but as an unemployed student I was approved for $6/mo of SNAP.
Like... for what possible purpose would the calculation even go that low? Just deny the application instead of wasting everyone's time and resources. I have no idea how long they put $6 into the account. I never bothered to look.
Suppose, e.g., that you can get $5k/yr in benefits if you make less than $10k/yr in other revenue and $0 otherwise. Unless you have a viable strategy for pushing past $15k/yr it's a strong financial disencentive against actually working, and even then your incremental ROI isn't very good past that cliff (if it takes an extra hundred hours to push to $15.1k/yr, then compared to your $10k/yr option you're only making $1/hr for the extra work).
This doesn't sound monotonic. This sounds like a mapping from pre-benefit income to post-benefit income which sends just under $10k/yr to just under $15k/yr, but sends just over $10k/yr to just over $10k/yr. So it sends a larger input to a smaller output.
I think I see the definitions we're disagreeing on. I'll lay out what I meant and then address your thing.
1. A couple levels up, the function somebody requested being continuously differentiable was the "benefits." You seem to be looking at the total post-benefit income instead.
2. It's not totally clear, but you _might_ also be using "monotonic" to refer to "monotonic increasing" or "strictly monotonic increasing" instead (the total income function in my example isn't even monotonic, so this is just me reading between the lines in the wording of your reply).
The cliff issue still exists though (and still exists in the differentiable version -- you want bounds on the derivative ideally to prevent the cliff issue, and for unrelated reasons you probably want other properties like the benefits not increasing as a function of pre-benefit income). Suppose you have a strictly monotonic increasing function mapping pre to post benefit income. That function can still, e.g., have a long region with a high slope, followed by a long region with a low slope, and some curvy thing connecting those. It's still continuously differentiable and strictly monotonic increasing, but the incremental value of work in the low-slope region is low (by definition). You might make a dollar in pre-benefit income and $0.10 in post-benefit income, so at minimum wage you're back to a <$1/hr situation unless you can make enough money to push substantially past the right side of that low-slope region (which we're assuming exists, else it basically says you have a 90% tax rate if you make too much money, and in-context "too much" would be near poverty levels, so even people wanting that sort of thing for the ultra rich wouldn't think that to be a good idea).
I'll go further and say what we probably want is for the derivative of net income as a function of earned income to be monotonic increasing but max out less than 1. So that there aren't ranges of income where you are receiving very little per dollar earned and then after some point start receiving more per dollar.
But solving benefit cliffs really just means having earned=>net income strictly increasing with the marginal rate reasonable, say at least 30 cents more net income per earned income. Under that constraint, you could have ranged where net income grows slower until you hit some higher dollar amount of earnings, but imo that should also not be desirable.
That can still have work disincentives; anywhere the magnitude of slope of the benefits is close to (or steeper than) the slope of the income as it phases out, then working more can get you no gain (or lose you money).
It's not sufficient to make cliffs impossible, you just never want the effective marginal tax rate go above a certain threshold.
It's insane that we cap marginal tax rates for the wealthy below 50% “because they need incentives to work harder” yet the working class family is facing an effective marginal rate near 100% because of reduced benefits.
The solution would simply be to stop making benefits decrease when salary goes up.
“But it's going to be insanely expensive” one may say, but it's an accounting illusion. All we need to do to break the illusion would be to stop counting gross public spendings and taxes and instead count the net public spendings/taxes for each individual (that is, over the whole population you take the difference between what they pay and what they receive and that gives you how much they contribute or how much they cost, instead of the current accounting system where we count people paying for their own benefits).
What's really expensive is the economic inefficiency of the current system.
> The solution would simply be to stop making benefits decrease when salary goes up.
That's an option, and I'd be interested in how the math works especially with predictions of how the economy would respond.
That said I don't think not decreasing benefits is an important requirement. But net income (benefits + earned income - taxes) should be strictly increasing with earned income, and probably at a rate of at least 50% but ideally higher than that up to some reasonable definition of middle class income. It should never be the case that if you earn more you can end up worse off.
> That's an option, and I'd be interested in how the math works
The math works because the derivative of a constant is zero, it's that simple.
> That said I don't think not decreasing benefits is an important requirement. But net income (benefits + earned income - taxes) should be strictly increasing with earned income, and probably at a rate of at least 50% but ideally higher than that up to some reasonable definition of middle class income.
Exactly, it doesn't need to be fixed, we could also have decreasing benefits, simply with a much lower rate of decrease (ideally the rate of decrease should be the marginal tax rate for this income). But in practice such a system would be much more complex than a flat benefit system for little gain (the more limited the decrease rate is, the closer to a flat rate your public spending would be).
Actually, optimal taxation models tend to show that a base transfer at zero income with an initially "high" marginal rate for clawing back the transfer at the lowest end (but still no higher than 100%, and falling very quickly as earnings increase!) works very well. (Marginal rates then rise progressively for higher incomes, and paradoxically become lower again at the very top end, trending towards zero at the extreme top-end of the scale. This is actually a consistent result in non-linear incomes taxation models; the very top earner should face zero marginal tax on the very last cent she earns!)
The intuition is that you only "feel" the high marginal rate if you earn very little, which impacts very few people; but the effect of making the break-even point more manageable with lower marginal rates for most earners is felt throughout the incomes scale. What really screws up things is higher-than-100% marginal rates, which are regrettably common in real-world systems amd completely useless.
I'd be curious to read more about that, because none of what you've written makes any sense to me right now. (Why would we want zero marginal tax on top earner who already get zero marginal utility for her money in the first place ?).
The zero marginal tax only strictly applies to the very last cent the extreme top-earner is expected to earn in the tax period. The idea is that at this theoretical extreme top-end tax bracket basically no one is paying infra-marginal taxes, hence the share of the incentive effect compared to the pure revenue effect is maximized. So you'll want to ensure that this extreme top-earner is only paying in lower, infra-marginal tax brackets, which have no incentive effect on her. This cannot be done literally because of randomness and uncertainty, but the broad effect of paradoxically lowering optimal marginal rates at the top end is nonetheless real.
It's only real if you assume top earners have monetary incentives to work more, which is a very bold assumption in a world where the top earners don't even earn money through their work in the first place.
And again, the utility of income for an individual is logarithmic with regard to their income, which means the marginal utility is the inverse function and that never stopped the top earners to want more.
Labor income is very important at the top end. The work a CEO performs in her superintendence of a large company generates what's economically labor income, even when paid as stock grants, options or the like. Basically all professional income (including that of devs) is labor income, not capital income.
(Besides, optimal taxation models also say that capital income should not be taxed at all, and you should concentrate "capital" taxes on sources of pure rent instead, with the rest of the burden falling on labor income and/or consumption! The intuition is that taxing invested capital is basically double taxation, since the apparent "returns" on capital are in fact wholly accounted for by time value and risk. There are important offsetting arguments, but these also become less relevant at the extreme top end of the scale.)
> Labor income is very important at the top end. The work a CEO performs in her superintendence of a large company generates what's economically labor income, even when paid as stock grants, options or the like. Basically all professional income (including that of devs) is labor income, not capital income.
If the CEO has a significant voting power in his company, then it's capital income even if the said income is “a salary”. Like it or not.
> Besides, optimal taxation models also say that capital income should not be taxed at all
It would be nice if people advocating for their political view could stop labeling their view as “optimal”, but hey, this is economics so here we are.
The optimality results are drawn from modeling assumptions that happen to be fairly general, not from any specific political views. For instance, much of the real-world political advocacy of UBI or cash transfers towards low-income folks is downstream from arguments about the predicted efficacy of this as a kind of redistribution; the arguments are not themselves politically motivated. The argument for taxing pure rents only, as opposed to productive capital, is structurally quite similar; as is the general argument for paying careful attention to incentive effects at the top end of the income distribution.
> The optimality results are drawn from modeling assumptions that happen to be fairly general, not from any specific political views
The idea that someone rich would “work less and thus produce less value to the society if their marginal tax rate was non-zero” isn't “fairly general” it's straight Randian and it's not how the world works. Same for the idea that individual income reflect the value they create to society.
The idea that you ought not to tax capital is also politically motivated.
Anything can be said to be “optimal” if you pick the hypothesis accordingly, and that's exactly what's being done here… It may work on paper for a world populated by a spherical John Galt in vacuum, but it tells you nothing about the real world.
> The idea that someone rich would “work less and thus produce less value to the society if their marginal tax rate was non-zero” isn't “fairly general” it's straight Randian and it's not how the world works.
People generally tend to work in exchange for money, and the more money they earn thereby, the more effort they'll want to put in their work. (This broad idea is sometimes known as "efficiency wages"; it often leads to paying workers more than they could directly compete for on the market!) Some independently rich folks may obviously choose not to work in any high-paid job at all, but given that they are, it makes sense to ask what motivates them.
> Same for the idea that individual income reflect the value they create to society.
That's obviously not true, since positive externalities, negative externalities and pure transfer rents all exist. Some people might well end up creating more value to society than their income accounts for, others less.
(I actually stated above that the zero-taxes-on-capital result is somewhat unrealistic on its own and there are arguments that do push the other way, so the current notion of what should be taxed may in fact be roughly adaptive. In practice, it's often more important not to push too far away from optimality with punitive levels of taxation than to precisely match the outcome of any given theoretical model.)
> People generally tend to work in exchange for money,
When they need it. Healthy retired people regularly do work for free in various charities and community work. Working people also don't usually seek to maximize their income, but balance it with plenty of factors.
> and the more money they earn thereby, the more effort they'll want to put in their work. (This broad idea is sometimes known as "efficiency wages"; it often leads to paying workers more than they could directly compete for on the market!)
This only works up to a point, you can't pay a taxi driver a million bucks in the hope of getting to destination 10 000 times faster. People do work better and harder when they feel they are fairly compensated vs when they feel they are being abused but that's it.
> Some independently rich folks may obviously choose not to work in any high-paid job at all, but given that they are, it makes sense to ask what motivates them.
And the answer is not the income proper, but everything that comes with such jobs and income, particularly “I'm making more money than Bob”. As long as Bob and him are being taxed similarly, then the actual amount makes little difference.
As I said above the marginal utility of money converges to zero the more money you have: a million dollar would change the life of the median US worker but wouldn't even be noticed by Tim Cook. (And even a billion doesn't change Jeff Bezos perception of worth an inch).
> In practice, it's often more important not to push too far away from optimality with punitive levels of taxation than to precisely match the outcome of any given theoretical mode
The concept of “punitive taxation” is in itself a political one, and again I can build a model in which the optimal income taxation is 100% above 1M and then say “we shouldn't push too far from the optimal” and that would be equally nonsense.
I agree with you, but for what it’s worth I live somewhere top marginal rates are about 50% and indeed 32 hour weeks are very common. I think this is a good thing though.
Yes, looking at the net transfers makes more sense.
However, net transfers aren't a function of income alone. They depend on lots of factors, because they are plenty of benefits that depend on more than income and plenty of taxes like that, too. Eg capital gains taxes and property taxes and sales taxes and tariffs don't depend on your income.
If you continue with the idea of transfers, which is, in effect, the current government buying off parts of its constituents, then "democracy", whatever has remained of it in the West, will die for good. What happens if I receive money from the Government but in the same time I'm also actively opposed to said Government's actions? Will I be allowed to speak against the Government that is, as a matter of fact, paying me? Will I have second thoughts of doing it? Will the Government cut off aid to me if I'm too vocal against said Government's actions?
All this to say that all we're doing is turning most of our countries' citizens into de facto slaves, people with no political free-will and who are well-aware that if they were to speak out against the powers that be they risk destitution.
> What happens if I receive money from the Government but in the same time I'm also actively opposed to said Government's actions?
Nothing? That's how it's worked in literally all functioning democracies. The purpose of any government, democratic or not, is to benefit its constituents. If it doesn't do that, we have a moral obligation to destroy the tyrants. Those benefits range in their tangible value, from hard to quantify things, like establishing public expectations of behavior (laws), to easy to quantify (subsidies like wellfare, farming subsidies, etc).
I'm truly baffled by your take that seems to insist that helping the needy somehow makes you moreso a slave to the government than the whole monopoly on legitimate use of force thing.
You can have net transfers to the poor without turning most of your country's into net recipients.
I also don't understand why you assume that people who are net recipients would stop complaining or voicing their opinion? That's not at all what happens in practice.
You're turning the "poors" into slaves, into slaves to the Government that is keeping them (the "poors") on the Government's payroll. I'm not "mixing" anything, so if we can please leave that condescending tone behind that would be best.
> who are net recipients would stop complaining or voicing their opinion
Why? You're asking why the "poors" will have second thoughts about openly criticising the hand that, literally, feeds them?
> Why? You're asking why the "poors" will have second thoughts about openly criticising the hand that, literally, feeds them?
No, I'm not asking why the poor would stop complaining. I am pointing out that empirically we observe that people who receive government largess don't shut up! (And that's true for all kinds of government largess, eg also for subsidies for rich people or companies.)
No, the "why" question I am asking is: why do you think that recipients of government handouts would shut up, when in the real world they haven't done so?
> You're turning the "poors" into slaves, into slaves to the Government that is keeping them (the "poors") on the Government's payroll.
Would you say the same about eg car drivers, if the government provides roads free of charge to the user? Or about anyone who benefits from national defense? Or are in-kind benefits excluded from your calculus? In that case, would giving poor people food and clothing and shelter instead of money mean they are no longer 'slaves' by your definition?
(For comparison: real-world, actual slaves like in the American South were usually provided with in-kind benefits to keep them alive. Money rarely changed hands, partially because it leads to autonomy.)
"Yeah, I made $1M last year. Here's my SNAP check for six cents."