Part of this is that specifically social security has become a monstrous program which is in need of significant reform but consistently fails to receive it because it's the third rail, and then because it has been allowed to become so large, it's crowding out the rest of the government.
There are a lot of other de facto transfer payments in the government, but those programs do have the potential for significant efficiency gains, because they're not just transfer payments. They're each making relatively small payments and then have associated eligibility and means-testing bureaucracies. They would also be better replaced with tax credits for the same people for exactly that reason, but in a comparison against the historical programs, the current programs should be benefiting from significantly improved administrative efficiency as a result of computers etc.
And the budget items that aren't direct transfers generally aren't individually large budget items (the military being an obvious outlier), but cumulatively they're still most of the budget.
Well, to be clear, I'm not claiming that there's no room for more government efficiency. My point here was just that these efficiencies don't happen automatically, simply because the private sector figured out how to increase productivity (i.e. a simplification of what increases in per-capita GDP presumably proxy).
Your point that the current programs should be (emphasis mine) benefiting from improved efficiency "as a result of computers, etc" suggests that we are in agreement that it's not automatic (since it's not happening). Likewise, we agree that direct transfers are the larger individual items.
But, that's not to say that I agree with you 100%. For instance, recent data suggests that only 0.5% of the money paid to Social Security recipients goes to administrative costs [0]. So, I don't want to overlook the programs that are administered efficiently. And, if by "monstrous" you mean we pay out a lot of money, then this is only a potential issue relative to what we take in. That is, the implication is always that cutting benefits is the answer.
So, the other third rail seems to be having the wealthy pay more by raising caps on FICA or otherwise. I mean, at what point do we consider that billionaires and centibillionares are increasingly common, and we're well on our way to minting a trillionaire? Yet, we're considering cuts to programs that contribute to mere subsistence for others? (I'll leave aside the irony that it's the would-be trillionaire who may well drive the cuts, because it's almost too rich).
>They would also be better replaced with tax credits
I'm not sure that adding to our already insanely complex tax code/bureaucracy gains us much, or that it would approximate needs as accurately as the current means testing approaches (even if it were more efficient).
> My point here was just that these efficiencies don't happen automatically, simply because the private sector figured out how to increase productivity (i.e. a simplification of what increases in per-capita GDP presumably proxy).
You can argue that the increase in government productivity wouldn't be exactly the same percentage change as the increase in private sector productivity, but it's highly implausible that it would be zero.
> that it's not automatic (since it's not happening)
That's not necessarily what's happening though. Some of the efficiency improvements have been realized by the government. But instead of taking the efficiency improvement as lower spending, other spending increased to consume the surplus.
> For instance, recent data suggests that only 0.5% of the money paid to Social Security recipients goes to administrative costs
That doesn't necessarily imply a high level of efficiency given the massive outlay though. "Oh, it's only 0.5%... of $1.5T." That's still billions of dollars in overhead. If you could cut the overhead in half you could save billions of dollars.
It's also not even accounting for all of it because a lot of the overhead from Social Security comes from the administration of the tax, which falls under the IRS. Which then uses the same trick where they hide a huge administrative budget by comparing it to the entirety of all income taxes paid by everybody to turn a large number into a small percentage. And then the separately budgeted federal law enforcement resources that go into investigating social security fraud etc.
It's easy to make something look more efficient by shifting its costs into someone else's budget.
> And, if by "monstrous" you mean we pay out a lot of money, then this is only a potential issue relative to what we take in.
Well no, any given program should have to stand on its own. Just because you could hypothetically collect more tax revenue doesn't mean you should raise taxes, nor does it mean that you should spend the money on program A instead of program B. For example, which is a better use of tax dollars, sending a social security check to millionaires, or increasing the child tax credit?
> So, the other third rail seems to be having the wealthy pay more by raising caps on FICA or otherwise.
The far better solution is to eliminate FICA whatsoever and put Social Security and Medicare into the general budget. Having a separate tax for them is an anachronism and it would eliminate the wailing about the "social security trust fund" which was never anything but Congress immediately spending the money when social security was collecting a surplus and funding the shortfall out of general revenues (or, let's face it, deficit spending) now that it isn't.
> I mean, at what point do we consider that billionaires and centibillionares are increasingly common, and we're well on our way to minting a trillionaire?
Why does anyone think this has much to do with taxes? Taxes in general are a percentage of profits. If someone's company becomes worth three trillion dollars and as a result they now have $800B, is the problem actually that the government took 30% of it instead of 38% or something like that? No, the problem is that the market is so consolidated there is a three trillion dollar corporation roaming around, which there still would be even if you changed who owns it unless you do something about that, which is a competition problem rather than a tax problem.
> Yet, we're considering cuts to programs that contribute to mere subsistence for others?
The point is that the program's budget is so expansive because it's cutting larger checks to affluent people who are not relying on it for subsistence.
> I'm not sure that adding to our already insanely complex tax code/bureaucracy gains us much, or that it would approximate needs as accurately as the current means testing approaches (even if it were more efficient).
You don't have to create a separate tax credit for each individual program. The programs all do the same general thing: Transfer payments to lower income people. Create a single refundable credit in the combined amount that everybody gets (and parents can claim on behalf of minor children) and it replaces all the different programs at once.
>but it's highly implausible that it would be zero
I agree that it's likely not technically $0.00. But, as the entirety of the rest of your comment suggests, many of the opportunities for increased efficiency will come from structural or other changes, specific to how we administer these programs.
>Some of the efficiency improvements have been realized by the government...other spending increased to consume the surplus
I would have to see the data on all three assertions here: that we have realized significant efficiency gains; that they were owed in any significant way to private sector gains; and that we simply spent the gains. Actually, that last point is likely incalculable, as it'd be buried in overall increases in spending, which is a separate issue in any case.
>doesn't necessarily imply a high level of efficiency given the massive outlay though
But it does. I rightly focused on percentages, not size, to convey efficiency. You can argue that even tiny improvements in efficiency can yield meaningful numbers when the outlay is massive, which you also did...
>"Oh, it's only 0.5%... of $1.5T.
but it's impossible for either of us to know how much room is there. We can reasonably say, however, that 0.5% reads as "impressive", given that it's far below the percentages for private retirement annuities (see previous citation), and there's numerically just not a lot of room there.
>lot of the overhead from Social Security comes from the administration of the tax, which falls under the IRS.
As it should, since tax collection is the IRS's function. Look at it this way, if instead this were on the SSA's budget, people would say the IRS was pushing off its collection budget onto the SSA, and they'd be more correct.
Same with some part of fraud enforcement falling to the federal law enforcement budget (i.e. the actual LE-related part). The SSA does have its own investigation office, as it should. That office then coordinates with federal law enforcement, as needed to enforce the law.
So, you can't simply zero out the federal LE budget there and I'm sure you don't want to replicate an entirely new SSA-only LE apparatus within the SSA.
>Which then uses the same trick where they hide a huge administrative budget by comparing it to the entirety of all income taxes paid
But, it's not a trick. Income taxes paid is a relative proxy for the number of taxpayers involved, the implicit complexity of tax returns, etc.βall related to the IRS's function. Now we can say the administrative overhead is 80% of taxes paid or 0.5% or whatever as a way to measure relative efficiency. There's nothing nefarious going on here.
>The far better solution is to eliminate FICA...Having a separate tax for them is an anachronism
It's an anachronism with value, in that it represents what's supposed to be happening here. But, now that it is off of that original design, we do need to recalibrate, particularly WRT to having the wealthy more fully subsidize the program; whether that's by increasing the corporate share, raising the cap on collection and reducing the max benefit, or adding it to the overall general tax. I do think keeping it separate has the benefit of clarifying the subsidy effect and the focus on subsistence. I also think it's protective of its intent, because adding it to the general budget/tax collection is the gateway to nullifying the subsidy effect as the wealthy continue to lobby for (and receive) tax cuts.
You could perhaps set a baseline percentage of the general tax for this purpose, but it gets really murky, really fast.
>Why does anyone think this has much to do with taxes?
Because, as it is, there are maximums and loopholes, etc. that mean the wealthy and these massive corporations pay a much lower percentage of their profits in taxes. I'm all for anti-trust for multiple reasons, but litigation and restructuring our economy are not the most efficient paths for addressing this tax problem. Increasing the corporate tax rate is.
>cutting larger checks to affluent people who are not relying on it for subsistence
I'm not sure an increased budget size, owed to higher payments to wealthier people, is really the issue here. Remember, that the point is subsistence and that we don't know who is going to end up wealthy. So, we try to calibrate how much we collect over the working lifetime of each person by collecting based on what they're making at any point.
So, wealthier people are likelier to receive higher payments, but we do cap their payments. So, I'm not sure exactly what you're proposing as the efficiency gain here. Are you saying don't pay people anything if, at the end of their lives, it turns out they don't need it? If so, that sounds like means-testing and the current lack thereof is exactly why the SSA is as efficient as it is today. So, I don't see what we gain by adding means testing, simply so we can say the budget is lower. That reads like a likely net efficiency loss.
>Create a single refundable credit
You won't find me arguing against simplifying the tax code, but I have a suspicion that it's not quite this simple. Too many variables, including people's specific circumstances (and regular changes therewith) affect this. For instance, how does a tax credit help when someone suddenly needs (or no longer needs) SNAP benefits? OK, so you start prepaying the credit as-needed, then try to settle up at tax-time (oof) or just carve it out of the tax credit altogether and keep SNAP. But, you can start pulling threads like this everywhere and what you weave out of those threads is a picture of how we got here (e.g. the problems we're trying to address). I agree that improvements can be gained by looking at the entire system in hindsight versus its piecemeal construction, but I'd wager the solution set is far more complex than simply moving everything to a single tax credit.
There are a lot of other de facto transfer payments in the government, but those programs do have the potential for significant efficiency gains, because they're not just transfer payments. They're each making relatively small payments and then have associated eligibility and means-testing bureaucracies. They would also be better replaced with tax credits for the same people for exactly that reason, but in a comparison against the historical programs, the current programs should be benefiting from significantly improved administrative efficiency as a result of computers etc.
And the budget items that aren't direct transfers generally aren't individually large budget items (the military being an obvious outlier), but cumulatively they're still most of the budget.