Social Security Meanderings

I intend (someday) to do a detailed critique of the current fight over Social Security.  As many of you may suspect, I'm not aligned with the Drummian* "Three Little Birds" view (for those who don't know their Marley, that's the song with the refrain "don’t worry about a thing / ’cause every little thing gonna be all right!"), for my deep-seated suspicion of the government's ability to adapt, revise, and actually pay for itself is too great for such lightheartedness.  Moreover, as Sebastian has pointed out in these pages, Drum's numbers are a little goofy.  This does not mean, however, that I'm much the fan of George Bush's approach, which seems loosely based upon Pink Floyd's "Wish You Were Here."    

But all that's for another day.  Today, I've just got an anecdote.  I received a letter from the government the other day,** which stated that my Social Security benefits had vested and put a pretty surprising number on the monthly payment that I can expect to receive during my golden years.  The amount wasn't as sneeze-worthy as I had expected.

Now, don't get me wrong, I'm still in favor of some form of privitization.  But I admit to feeling a little, well, good about having that benefit guaranteed.  A little, secure, you know?

And then I had the offending bit of humanity removed and replaced with version 2.1 of MicroSoft's(r) Windows Human Replicator, Center-Right. 

von

*It's to Drum's great credit, incidentally, that he's been writing about this subject for quite some time with such urgency and clarity -- establishing the meme, as they say.  Y'all know that he's one of the smartest liberal voices out there, doncha?

**Keeping to the musical theme, I admit recalling the opening lines -- though, clearly, not the intent -- of Public Enemy's "Black Steel in the Hour of Chaos" as I wrote this passage. 

Comments

[rilkefan]

Glad you're not going to starve in your old age, but every second we spend talking about this phony crisis - excuse me, current issue - is one not spent talking about Medicare.

[Sebastian Holsclaw]

The fact that we can't deal with the easier question of dealing with Social Security issues doesn't give me much hope that the much harder problem of Medicare is going to be addressed.

[Katherine]

Remember the "energy crisis" of 2001? There was a real problem, but it was blown out of proportion by the same untrustworthy people who had helped cause it, and whose proposed "solution" would have made the problem much, much worse. Meanwhile, they ignored the much greater danger: an obsolete electricity grid, where an accident in Ohio (or terrorist sabotage) could leave much of the east coast without power and much of Manhattan without running water.

For that matter, remember the grave and urgent danger we faced from Saddam Hussein's Iraq in March of 2003? How we couldn't afford to wait, lest the smoking gun be a mushroom cloud and over a hundred thousand American troops have to remain there for a long, hot summer? There was a real danger, but it was blown out of proportion by untrustworthy people whose solution put us, and especially the troops in much greater danger. Meanwhile, they ignored the much greater dangers: North Korea, Iran, loose nuclear weapons in the former USSR, a black market in Pakistan.

Just substitute "social security" and "medicare" in the right places--it's the exact same story. I hope the ending's different.

[Kevin Drum]

Reasonable people can disagree about what we should do about Social Security, but I assure you my numbers aren't goofy. My cats are, but my numbers aren't.

I responded to Sebastian way back when he wrote that post. He misread my original piece, thinking I was counting on the trust fund in my analysis when I wasn't.

The current projections by the Social Security trustees are almost certainly too conservative, and if that's the case we really don't need to do very much at all. But even if they turn out to be spot on, it's still a quite solvable problem. Overall, given the evidence, we're probably better off waiting ten years to find out whether we really have a problem or not and just how big it really is.

I too was once a victim of years and years of hearing the Social Security was doomed because of the demographic timebomb heading our way. It was only after I started looking into the numbers for myself that I discovered that Social Security's problems had been wildly exaggerated. Democrats as well as Republicans are responsible for this tate of affairs.

[praktike]

Damn the tate of affairs! Additionally, Kevin Drum is completely right.

[Edward]

So, given the degree of informed commenters we have here, is it simply an ideological desire to dismantle SS driving this effort? That's what Krugman wrote today. That's what I've believed for some time now.

Or, is there a sincere belief (aside from NewDealaphobia) that the system is fundamentally flawed and needs rethinking?

[Jesurgislac]

Have to say, Krugman's figures look a lot more sensible than the conservative omg-it's-doomed figures.

Given that no one claims that US Social Security is going to fail any time in the next 40 years, it seems a little premature to be dismantling it now on the basis that in 40 years time it may not work any more.

If I were an Evil Overlord out to wreck US Social Security, I'd probably do just exactly what the doomsayers are doing: persistently and consistently claim that it's not going to work in the future, and so it might just as well be got rid of now. Until it gets to the point where it actually sounds wildly optimistic to point at the figures and say "Well, actually, it's in pretty good shape."

[hilzoy]

I have no idea whether they genuinely think that the system is fundamentally flawed and urgently needs fixing. I do think, however, that only ideology can explain why they propose to "fix" those problems with a "solution" that would only make them worse.

[von]

So, given the degree of informed commenters we have here, is it simply an ideological desire to dismantle SS driving this effort?

For me: yes, in part it is.

And, it appears I missed Kevin's response to Sebastian -- my apologies. I usually try to link those sorts of things.

[Sebastian Holsclaw]

"So, given the degree of informed commenters we have here, is it simply an ideological desire to dismantle SS driving this effort"

It depends on what you mean by dismantle, it also depends on what you mean by crisis. As Kevin likes to say, it isn't a Social Security crisis its a general fund crisis.

Ok, and SS is a huge amount of money going out. I don't see anything wrong with dealing with it in part by slowly cutting out rich and upper middle class people from the payouts. Is making it a safety net instead of including a huge wealth transfer to rich people 'dismantling'. If you think so, I guess it is. But whatever you call it, I think it would be prudent.

I still don't understand why so many on the left are ok with paying huge amounts to rich people (who not only don't need it, but also live longer and draw benefits longer). Even if support to the program requires that it pay lots of people who don't need it (which I do not conceed) couldn't you still cut out payments to the top third? You still have payments going to the bottom 2/3. Isn't that enough to keep enough people voting for it?

I would also note that if you are looking at it as a general fund crisis, Medicare is easier to pay for if we are saving huge amounts of money on Social Security.

[Edward]

I still don't understand why so many on the left are ok with paying huge amounts to rich people (who not only don't need it, but also live longer and draw benefits longer).

Because it seems to be working and I have very little faith that our poor elderly will be taken care of should the system be dismantled, all promises to that end duly noted.

[Sebastian Holsclaw]

It seems to be working, for the moment, in a hugely wasteful way that ends up making huge payments to rich people for no apparent reason.

As for: "I have very little faith that our poor elderly will be taken care of should the system be dismantled, all promises to that end duly noted.", I don't think you would be responsive to the argument that Social Security is on the road to communism--and you know how bad communism is. This is a not-so-slippery slope.

There are a lot of steps to be taken between "We pay no one who is in the top third" to "We pay no one". Also, if this is about support, 2/3 is still a very large majority.

Its funny how conservative you all sound in this argument--sure it is bad in some areas, but we don't dare change anything, we have to cling to it just the way it is, even if it was based on totally different assumptions about life-spans and number of retirees. ;)

[Dantheman]

"There are a lot of steps to be taken between "We pay no one who is in the top third" to "We pay no one"."

Unfortunately, the people proposing the privatiziation have made no secret that paying no one is their ultimate goal. Failing to believe they will stop after one step is prudence.

[Katherine]

Yes, but there are people on the right who actively want to dismantle social security, and Medicare, and Medicaid, and always have. There's no Communist equivalent of Grover Norquist among the Dems. And while I don't know that Bush is a "starve the beast"-er, it's one of very few rational explanations for his economic policies (the other one being "candy for everyone and let my successor worry about it").

[praktike]

Research shows that means-testing makes rich people resentful.

[Edward]

Research shows that means-testing makes rich people resentful.

All the arguments for cutting out the rich who don't need it seem insincere to me...Dantheman sums it up perfectly "the people proposing the privatiziation have made no secret that paying no one is their ultimate goal."

[Sebastian Holsclaw]

"the people proposing the privatiziation have made no secret that paying no one is their ultimate goal."

Name them. Because that generalization is pushing it. And besides why limit ourselves to privatization? Why not say: "Privatization isn't a good idea, but saving money by not paying rich people is a good idea."?

Are you saying that my idea is fine, you just don't trust Republicans to implement it?

Fine.

Have Democrats propose it and win elections on the basis of fiscal responsibility. It will work better with their 'Two Americas' thing anyway.

You don't have to go along with Bush's ideas if you don't like him. You may notice that that my main proposal is totally independent of privatization concerns. If you want, you could use it as an alternative to fight Bush's proposal.

[von]

Oh, c'mon. Means testing will probably only reduce the amount of expenditures by a fraction -- wealthy enough not to notice the loss is a fairly small number.

Better would be a two-tiered system: Private accounts for everyone, guaranteed welfare payment kickers for folks who are truly in need. The rich and middle-income-who'd-miss-losing-SS get something (private accounts) to tide them over. The truly needy get private accounts plus weekly payments. (Incidentally, if the middle income folks screw up their accounts, they may be eligible for the welfare kicker.) No one starves, but the system is shaved.

[rilkefan]

One argument for paying the rich is to incentivize them not to use their disproportionate political power to stop payments to the poor (e.g., see the bribing of von above).

I don't question SH's sincerity in wanting a more ideal system - but I hold that the better is the enemy of the good.

[rilkefan]

"Have Democrats [...] win elections on the basis of fiscal responsibility."

Kerry tried that. Oh well.

[carpeicthus]

Sebastian, even though it's more fun to argue against you because you have the better ideas, there are big things going on with Social Security right now and they have nothing to do with keeping its payment structure but means-testing it. The argument of consequence is against the Republican leadership, which is unfortunately not you.

[Sebastian Holsclaw]

"Oh, c'mon. Means testing will probably only reduce the amount of expenditures by a fraction -- wealthy enough not to notice the loss is a fairly small number. "

Not really, hitting the top 1/3 would save at least $20 billion a year. That is a significant portion of the GDP (much more the government's budget).

As for 'wealthy enough not to notice the loss', that number is heavily influenced by the fact that upper-middle class people have been structuring their retirements around getting SS. That is why I propose a long phase out time for them--(3 1/3% per year for 30 years perhaps).

[Sebastian Holsclaw]

Sebastian, even though it's more fun to argue against you because you have the better ideas, there are big things going on with Social Security right now and they have nothing to do with keeping its payment structure but means-testing it. The argument of consequence is against the Republican leadership, which is unfortunately not you.

So steal my ideas then. Easier to fight with a good plan dontcha think? I'll convince my conservative friends to go along. We end up with more money, a fairer system, and you win against a Bush proposal. What could be better for you?

[Sebastian Holsclaw]

Or, look at it this way. If the Republicans are so powerful that you are almost helpless, isn't it better to push them toward a means-testing proposal which cuts out the hated wealthy people rather than have them go into a privatization scheme that you think is horribly risky?

[von]

As for 'wealthy enough not to notice the loss', that number is heavily influenced by the fact that upper-middle class people have been structuring their retirements around getting SS. That is why I propose a long phase out time for them--(3 1/3% per year for 30 years perhaps).

OK, I might be able to come around to this plan. Incidentally, a long lead time is essential, lest we deal out (among other things) severe demand shocks.

One argument for paying the rich is to incentivize them not to use their disproportionate political power to stop payments to the poor (e.g., see the bribing of von above).

Well, if you're basing this on me, you better up the payments. 'Cause the "bribe" apparently ain't enough. ;-)

[Dantheman]

""Have Democrats [...] win elections on the basis of fiscal responsibility."

Kerry tried that. Oh well."

And Al Gore. And Ross Perot. And Paul Tsongas. And Walter Mondale. And George H.W. Bush (in 1980). People just want "free candy".

"I'll convince my conservative friends to go along."

How many members of Congress is that, exactly? If under 218 Representatives, and 51 senators, no deal.

[Dantheman]

"isn't it better to push them toward a means-testing proposal which cuts out the hated wealthy people rather than have them go into a privatization scheme that you think is horribly risky? "

No. The wealthy have their own privatization scheme. It's called private investments. If wealthy people (who few to no liberals actually hate, speaking of overbroad generalizations) want to take risks with their own money, that's fine. Liberals are more concerned with keeping a floor for the people who do not have sufficient funds to invest.

[rilkefan]

"Well, if you're basing this on me, you better up the payments. 'Cause the "bribe" apparently ain't enough. ;-)"

Patience, von, these things take a while to sink in.

SH: "If the Republicans are so powerful that you are almost helpless, isn't it better to push them toward a means-testing proposal which cuts out the hated wealthy people rather than have them go into a privatization scheme that you think is horribly risky?"

I'll skip the vector analysis analogy that leaps to mind, but pushing for a change in a reasonable direction is partly a push for a change in a bad direction. The system is ok as it stands - suggesting ways to change it empowers people who want to do away with it. As von suggests, you're for turning SS into a welfare system, which would make it more politically vulnerable. And I don't see how you make up for the loss of support for the program when you reform away the bribe aspect.

[liberal japonicus]

Research shows that means-testing makes rich people resentful.

Shouldn't this be more resentful?

[JerryN]

Just thought I'd poke my nose in again on means testing. Don't forget that SS benefits above a threshold income level are taxable. In effect, this already reduces the benefit to wealthy recipients.

Sebastion mentioned that the upper-middle class have currently structured their retirement around SS. It would be an interesting think tank sort of project to see how these people approach the current income tax provisions to get some sense of how effectively means testing could be gamed.

[rilkefan]

Compare the 1% administrative cost of SS to the 73% admin cost of reconstruction in Iraq...

[Sebastian Holsclaw]

"And I don't see how you make up for the loss of support for the program when you reform away the bribe aspect."

Do you need 100% support? Isn't 60% support enough?

[rilkefan]

"Do you need 100% support? Isn't 60% support enough?"

What's the RMS on that 60%? If there are 11% fluctuations I'm going to be unhappy.

[Jesurgislac]

Given the tendency Americans have to think they're in a higher income bracket than they actually are (see the many defenders of Bush's tax cuts who honestly think they benefit by them): how many Americans will believe they're not included in the "top third" for whom Sebastian plans to cut out Social Security benefits?

Last time this came up, Sebastian was unable to show any instance where making a benefit means-tested had cut costs and increased take-up among eligible beneficiaries, which he claims are the goals he wants to achieve in making Social Security means-tested. Which irresistibly reminds me of the scientist in Measure of a Man who wanted to take Data to pieces so that he could build more Datas, but who couldn't prove that he even knew how to put Data back together again...

Yes, I really am that much of a Trekkie.

[fdl]

question of the day: to what extent is everyone's payroll taxes staying in the same cohort?

oh, that's not english? let's try again. Sebastian argues, quite persuasively, that a program which transfers funds from poor to rich sucks. but is he right? or is it more the case that the income/wealth brackets of those who pay and those who receive match up pretty well?

after all, it's a little easier to sell a program in which only the rich current payroll taxpayers are paying for funds received by those who don't need it (reserving, for purposes of this post, any disputes about the word "need").

Francis

p.s. and i'm still not satisfied that we have agreement on "need". how much of a lifestyle decline can we impose on those who retire? but i recognize that you can be living beyond your means no matter how much you make and that SocSec is not meant to reward poor decision-making.

[Sebastian Holsclaw]

"after all, it's a little easier to sell a program in which only the rich current payroll taxpayers are paying for funds received by those who don't need it (reserving, for purposes of this post, any disputes about the word "need")."

I don't know how you would go about investigating that. I would intuit that since old people are statistically the richest people, it is unlikely that there are enough working rich people to pay the taxes to cover all the retired rich people. But I don't know what statistics I would use to prove that.

[Jesurgislac]

I would intuit that since old people are statistically the richest people

Perhaps rather than using your intuition, you should check reliable facts? Suggest you might begin with this paper (PDF file) on the distribution of earnings, income, and wealth.

But I don't know what statistics I would use to prove that.

You've yet to use any statistics to prove anything, so far...

[bob mcmanus]

"is it simply an ideological desire to dismantle SS driving this effort?"

Late to this thread. Without taking any theory as my favorite (greed & party building 2 others); and in my radical skeptical way, even thinking the ideological argument should be looked at seriously:

I have seen a case made that the US is in more catastrophic economic shape than most realize. The dollar as reserve currency was a result of the devastation of most of the developed world during WWII. We have been struggling for the last thirty years to retain that advantage as Europe, and then East Asia rebuilt. Now that China and Russia are moving up, and oil is starting to run out, or demand approaching reasonable supply, it will take extraordinary measures, either military imperialism or economic restructuring, to keep the American standard of living, and economic, tech, and military power from declining 30-50% relative to the world in the next generation.

Not that I have much reason to believe Bushco has this stuff in mind. But I do have some. The deficits can be a defense of the dollar, a game of chicken with other nations, for instance. That looks fairly solid.

[CharleyCarp]

I'm not old. I am maybe a little older than some of you all though, and old enough to remember that we already made a deal. I've been paying more than my fair share for SS since the mid-1980s. And if maybe there are sets of assumptions where the deal doesn't pan out for my whole retirement (I'll turn 92 in 2050, but really doubt if I'll make that) those same assumptions don't make the alternative look all that rosy either.

I've got no problem with a means-testing formula, along the lines Sebastion's been promoting, but it's have to be a deal that would stick. And that's the problem. See, I already made a deal, and the people on the other side of the table are spending a whole lot of effort trying to get out of it. Why? Basically, because they wanted to spend all the money I was paying -- since 1984 -- on all manner of crap to which I have been opposed. And they got their way -- in 1984, 1988, sort of in 2000, but definitely in 2004.

If conservatives could be trusted to keep to a deal, I'd have no problem making one. Hell, I voted for the guy who promised to raise my taxes, and not the guy who promised not to. I'd be first in line to support doubling the cap on contributions.

But why on earth would I want to make a deal with someone who has given me no reason on earth to trust that the deal would be honored? Why on earth should I make a deal with someone who spent the extra money I already paid in, under the last deal, on a stupid missle defense system? Or a war to make Iraq safe for Iran? On a prescription drug benefit without volume discounts, for goodness sake?

[Doh]

I've noticed that all of Sebastian's SS comments always start from the premise that the current system is a wealth transfer from poor to rich, but when I suggested that maybe the wealthy don't even get back their contributions under the current system, he (IIRC) told me I was wrong but could not provide any evidence. (Francis, this is an effort to answer your question by looking at net transfers to individual wealthy taxpayers).

It strikes me that if one's argument fundamentally depends on a factual premise for which one has nothing more than an intuition, one should stop making such a big deal about one's views, but then I'm old-fashioned that way.

More succinctly, the problem (if there is one) could be solved better not by means-testing but by lifting the cap on wages to which SS taxes apply. I bet 2/3 of the country isn't even aware there is such a cap.

Finally, I'd note that the idea that we'd save enough money "reforming" SS to "address" medicare is like saying we'd save enough money clipping coupons to buy a helicopter. Ir's highly unlikely, if you mean to preserve either in substantially the form we know now.

[Katherine]

The payroll tax is regressive compared to the income tax, but social security redistributes money downward on net.

Are you talking about means testing when people earn or when they're eligible for benefits? If the former, major problem: D-I-V-O-R-C-E. Social security is a huge safety net for women who either don't work outside the home at all, or who work fewer hours & make less money than their husbands. Which is many, many women.

If you're talking about later--it seems like this disincentivizes private savings in a big way for people anywhere near the cutoff. Which seems like a bad idea.

Since administration costs are so low, I would prefer to make it more progressive on the tax side of things if needed & leave the safety net intact--especially given the uncertainty of Medicare and the general trend of shifting risk from capital to labor.

So I'm not convinced you're right on the merits, and I am convinced it's a horrendous idea politically. The Republican party leadership is NOT interested in redistribution unless you count redistribution up. They prove this at every opportunity.

[Sebastian Holsclaw]

"It strikes me that if one's argument fundamentally depends on a factual premise for which one has nothing more than an intuition, one should stop making such a big deal about one's views, but then I'm old-fashioned that way."

Good freaking God, could we drip the venom a little more? I'm not making 'the earth is flat' levels of assumptions here.

Fact: Retirees have some of the highest wealth percentages in the nation.

Fact: Retirees do not make up the largest percentage of people in the nation when compared to non-retirees.

Logical inference (though certain mathematical possibilities which almost certainly do not apply in this case could make it otherwise) some people who are poorer but working are paying some people who are richer but not working.

Jesurgislac, your reference wouldn't be easy to use for our stated purpose. But if you wanted to do lots of calculations you could try to use table 8. Be sure to divide by number of people in household and be sure to weight for the percentage of people in the age group. But by eyballing it, it seems obvious that the over 65+ group has a per-person wealth greater than everyone except the 51-60 range and greater per-person income than the under 40 group. Which is to say that they have a greater per-person wealth than all but 12% of the population and have greater per-person earnings than at least 40% of the population (pretty impressive for not working). (I'm also a little worried about the statistics because it mentions 'household sample' and I'm not sure I understand if they mean that it is a statistically rigorous slice of the American populace).

Katherine, I'm talking about means testing when they are eligible for benefits. Cutoff point cheating could be deincentivized by starting at or near the cutoff point and ramping down the benefits on a ratio. E.g. if the cutoff is determined to be at the $60,000 level you cutoff at a three dollars for every one above that level.

[ken]

I still don't understand why so many on the left are ok with paying huge amounts to rich people (who not only don't need it, but also live longer and draw benefits longer).

I don't know about others but I am ok with it mainly because SS is subject to what is left of a progressive income tax. It gets added to the top of whatever other taxable income by way of pensions, interest, rent, or dividends a person recieves. So with 'rich people' the government will gets a big chunk of it back. I know this is not perfect but it is better than means testing every retiree monthly to see if they need some asssistence that month. (It is no better if the means test is quarterly or annually, you can see the complications involved)

[Jesurgislac]

Sebastian, you're arguing for a system which will make most of the people I know in the US significantly worse off: you're proposing to means-test Social Security, and this inevitably means that people who are entitled to a benefit don't get it because they can't get through the means-test, and that administering the benefit costs more.

Obviously, this won't affect me directly, but, you know, I like my friends and have no wish to see them suffer a penurious old age. You have a habit of not thinking through on the results of the policies you propose: I think this is yet another example of how you don't want to think about the inevitable long-term results of the policies you tout.

[Sebastian Holsclaw]

Sebastian, you're arguing for a system which will make most of the people I know in the US significantly worse off: you're proposing to means-test Social Security, and this inevitably means that people who are entitled to a benefit don't get it because they can't get through the means-test, and that administering the benefit costs more.

Most of the people you know are in the top 1/3? Good for you. But I'm not feeling sorry for them. Inevitably means people who are entitled can't get it? Why? Are we talking a statistically significant number? There are people who can't get Social Security now because they don't fill out the forms. As for administering the benefit, as we established last time, if I spot you the entire budget of the IRS there is still a savings of at least $15 Billion (with a B not an M). I don't want to hear complaints about the cost of Iraq if you don't think $15 billion every year counts.

"I like my friends and have no wish to see them suffer a penurious old age. You have a habit of not thinking through on the results of the policies you propose"

I don't know how to respond to this. If you think that cutting the benefits to the top 1/3 has anything to do with a penurious old age, you don't have a habit of using the word 'penurious' properly.

The inevitable long term result of the policy I propose is that rich people don't get Social Security money and we save tens of billions of dollars every year while middle class and poor people continue to get Social Security benefits.

[Pete Clay]

Sebastian, could you give us an example of the sort of people on that 1/3 boundary you make so much of? Is the 1/3 defined by total asset ownership or total income or what? By asset ownership do you include homeownership?

And how do you propose to deal with the classic means-testing problem, that people on the boundary have a big incentive to make themselves slightly poorer so they end up better off after the benefit?

[Jesurgislac]

Most of the people you know are in the top 1/3?

Nope. Most of the people I know are in income groups where a large part of their income post-retirement will be Social Security. Under your plan, they'll be worse off: rather than getting the benefit they're entitled to, they'll be required to prove they're poor enough to deserve it. Further, since your plan will inevitably mean that Social Security administration costs will go up, your plan could very well trigger an end to Social Security - which would mean that most of the people I know would be doomed a penurious old age. You may not feel sorry for those people, because you're convinced you'll be rich enough not to need Social Security in retirement and don't plan to have any friends who will be poor enough to need Social Security in retirement, but that's you: it's not me.

As for administering the benefit, as we established last time

"We"? You established nothing last time: you pointed to no concrete instance where you could show that making a benefit means-tested had saved money or increased take-up. Then as now, you wave around airy-fairy figures as if they established something.

The inevitable long term result of the policy I propose is that rich people don't get Social Security money and we save tens of billions of dollars every year while middle class and poor people continue to get Social Security benefits.

In your dreams, Sebastian. You've yet to show how your means-tested system would be different from any other means-tested system of benefits which everyone pays into but only those who can prove themselves poor enough to deserve it get. Until you can show you've figured out how your system will deal with the inevitable problems and consequences that all means-tested benefit systems have run into in the past, you're just playing with dollshouses, ignoring the real world.

[Katherine]

"Katherine, I'm talking about means testing when they are eligible for benefits. Cutoff point cheating could be deincentivized by starting at or near the cutoff point and ramping down the benefits on a ratio. E.g. if the cutoff is determined to be at the $60,000 level you cutoff at a three dollars for every one above that level."

Okeydoke. A few more questions:
1) are assets included? If so I assume there would be some exemptions for a home, etc.? Maybe just liquid assets? If not--it's really wealth that matters more than income, especially at that point in one's life. A given dollar amount in income looks very different if it's the interest on your swiss bank account that you could withdraw in full, versus the amount you get from your teacher's pension.
2) do you realize that benefits and the payroll tax are in effect already capped?
3) Can you look up some numbers as far as what the income would be? 66th percentile of seniors or 66th percentile of the general population?
4) wouldn't keeping benefits the same but lifting the cap on payroll-taxed income do a heckofalot more to finance Medicare? Again, small government for its own sake doesn't necessarily impress me all that much--especially not when you're proposing something that's likely to increase administration costs.

[Sebastian Hoslclaw]

RE administration costs, and especially for Jesurgislac.

For purposes of this discussion I have spotted you the entire budget of the IRS. The IRS currently administers a system which is MUCH more legally complex, involves many more people, and includes adminstering tax issues for corporations with tax situations much more complex than individuals. There is no reason in the world the adminstration cost should approach anywhere near the entire budget of the IRS. Yet for discussion purposes I have talked about it as if such an amount were realistic. And still we get billions of dollars in savings per year. You are assuming that the costs will be three times as much as the whole budget of the IRS. That is $30 billion dollars. You are welcome to continue believing that the adminstration cost objection is valid. You are not using logic to get there, so I can't use logic to refute it. If spotting you the entire budget of the IRS is an 'airy-fairy' figure, its silliness cuts in your favor so whining about it is just ridiculous.

I am looking at census tables now for a useful one which has age, wealth, income and a percentile breakdown.

[Sebastian Holsclaw]

Wow it is amazingly hard to find a good breakdown of wealth and income by age that doesn't just refer to the median income (which is notuseful for purposes of making the cut-off any point other than the median.)

Using the data from this census report, we can get near some of the data. But I'm skeptical of it, because it makes my case look better than I thought it would be. I wonder if the quintile markers are made against the whole population while the assets are made against the 65+ population in that population quintile. (Which the chart gives no way of calculating so far as I can tell). Anyway Table F suggests that those in the 4th quintile over the age of 65 (and obviously more so in the 5th) have almost double the wealth (non-home equity standard even more if you include home equity) of those in categories younger than them and in the lower quintile categories. The fourth quintile and above is .4 of the population while my standard represents a smaller number of people at .33.

But I admit these numbers aren't easy to use nor are they in a format we want, so I'm still looking.

[Sebastian Holsclaw]

BTW, I just want to be explicit about this in case some people disagree. After someone stops working, I think it is fair to expect them to spend their wealth as opposed to an expectation of continuing to amass wealth or having an expectation of maintaining their wealth until they die. If you want to have more money when you die than when you retired, you better have some great investments and/or super-frugal spending habits.

[Sebastian Holsclaw]

For statistical analysis purposes I want to note that if you are looking at household income or household wealth (which is easier to get a hold of than individual numbers) you need to be aware that the number of people in the household drops off dramatically with age (see Jesurgislac's link way above). If you don't take into account the difference between 3.2 people per household and 1.7 people per household, you are going to get a distorted picture of the issue.

[Sebastian Holsclaw]

AHA, here is a useful chart from the AOA. It shows that of people in the 65-74 range (and note that people 74-and-under represent 76.8% of all people over 60) 43.5% (24.8% + 18.7%) of them have incomes of more than $30,000 per year. 43% of them are earning more than $50,000. Since my proposed cut-off is around 33%, I would suggest that the cutoff is probably in the $38,000 to $41,000 zone. The table suggests that a greater percentage of people would receive Social Security under my system if they live past 75 (which is completely intuitive as they spend their wealth) than earlier, though such people represent a much smaller portion of the whole.

I note that the number includes Social Security income, which is annoying but not fatal because these are current retirees who had every right to make their retirement plans on that basis. (Please remember my 30 year lead in period.) I also note that this is household income, and repeat my caution that households are much larger in the younger categories than they are in the older categories (see the census tables cited above).

If anyone comes across income distributions for individuals by age instead of by households, it would be greatly appreciated.

If anyone has an SSRN membership, the abstract of this paper looks useful. But I haven't seen the actual contents so I don't know for sure.

[Katherine]

I agree that it's reasonable for wealth to decrease, but there's a problem in that you simply don't know how long you'll live. Some of Yglesias' stuff on annuitization in TAPPED is relevant.

Also, if people are spending down their assets before the catastrophic last illness, that will leave Medicare/Medicaid with even more of the bill.

Another reason I have a bias against income cut offs: I'm from the NY area, where a dollar doesn't go nearly as far. However, that is mainly a matter of housing prices, and probably most people in the top third own their homes & have paid off their mortgages by now. Wouldn't necessarily count on it though.

It still seems easier to me to deal with distributional issues on the tax side since administration costs are low and since I have no instrinsic bias against $ going through the government. I also think it's a bit unfair to screw up people's retirement plans. People who are 50 years old and have always counted on social security--to have it half gone by the time they retire just doesn't seem fair. It's a bit late now to start a private pension. It's a reliance issue, to put it in legal terms.

I do think the government has a role to play as far as wealth distribution, but I see social security as primarily an insurance program, not as a program to aid the poor, and it's always been sold that way. Means-testing and benefits cuts might be necessary if the general fund's in bad enough shape and if Medicare continues on the merry path to ruin, but I would prefer to avoid them if possible and address those problems more directly.

[Sebastian Holsclaw]

"Also, if people are spending down their assets before the catastrophic last illness, that will leave Medicare/Medicaid with even more of the bill."

Actually no. If you posit that those assets would (if not used to cover retirement replacement for social security) be used to cover the catastrophic last illness, the net cost to government on that issue is zero. If that money in an amount of X is being spent to replace government costs for Social Security or for Medicare, it doesn't matter which you spend it to replace. X is the same amount in either case.

Hey, I'd love to fix Medicare too. But I don't have great ideas on everything. Medicare is much harder to fix because people have the irrational expectation of getting the very latest technology, with ever better innovation, and without wanting to pay the cost of it. Medicare is most likely to be fixed by setting a non-innovation threshold for Medicare care or by strangling innovation by sucking profit out of the medical system. Or both. If you think telling grandma to sell the four bedroom New York house she owns and lives in alone is tough, try telling people that grandma only gets 1990-level care paid for.

[Katherine]

right, it's the same cost to the government--not less.

[Katherine]

right, I meant it's the same cost to the government when that happens--not a net savings.

the health care system scares me. That Medicare bill last year, that was such horrendous policy. I didn't entirely realize it at the time, too caught up in the equally horrendous energy bill and the foreign policy stuff.

Random statistical question: what have our basic health indicators--infant mortality, life expectancy, that sort of thing--done lately?

[felixrayman]

There should be a rule on this blog against the same poster posting 5 comments in a row.

That's annoying.

[Jesurgislac]

Sebastian: and especially for Jesurgislac.

For purposes of this discussion I have spotted you the entire budget of the IRS.

But you have never said how you will convince anyone to spend the entire budge of the IRS on administering a means-tested version of Social Security. Unless you're envisaging yourself as Dictator with power to tax and spend as you please, your program must have a realistic base, and it is not realistic to suppose that you would be allowed to vastly increase the expenses of administering Social Security for a hypothetical saving. That's the kind of thinking that is permitted to defense departments only.

[Jesurgislac]

Felixrayman: There should be a rule on this blog against the same poster posting 5 comments in a row.

I disagree. All of Seb's comments were ontopic. Probably less annoying than one massive comment comprising them all.

[Doh]

"Good freaking God, could we drip the venom a little more? I'm not making 'the earth is flat' levels of assumptions here."

Sebastian, it wasn't venom so much as frustration. You have been pushing the argument that it is obvious we should cut off SS to the "top" 1/3 of beneficiaries without making any efforts to tie it to the real world, and after a while that gets very frustrating (if not dishonest).

The reason why it's important is that once you tie it to the real world, we can start having more of a discussion, i.e., are you referring the "top" by wealth or by income. A person who retires with $1 million in assets (which may put them in high percentile for wealth) may need all of that to provide for an income of $40,000 a year (which would be a much lower percentile for income). Since I (like Katherine) live in a high-cost area of the country, I don't consider $40,000 to be all that wealthy (especially when you consider the increased costs of healthcare for seniors). For purposes of this discussion, I think $40,000 in income is not a point where SS benefits are easily foregone.

All that is preface to thanking you for taking some time to try and dig up the underlying statistical information.

But what is with this statement?

"Not really, hitting the top 1/3 would save at least $20 billion a year. That is a significant portion of the GDP (much more the government's budget)."

I assume there's a typo in there, but I can't figure out what it is.

And you still haven't responded to the point (from Katherine as well) that it appears the whole problem that appears to worry you could be easily resolved by lifting the cap on the wage base for FICA taxes.

(By the way, I still think that wealthy people get less out of the system than they put in, but until I find some facts to back that up, I'll shut up about it.)

[Sebastian Holsclaw]

I haven't seen good figures on the wage cap. I don't think there are enough people earning that much money to raise $20 billion a year, but I could be totally wrong about that.

But you have never said how you will convince anyone to spend the entire budge of the IRS on administering a means-tested version of Social Security. Unless you're envisaging yourself as Dictator with power to tax and spend as you please, your program must have a realistic base, and it is not realistic to suppose that you would be allowed to vastly increase the expenses of administering Social Security for a hypothetical saving.

I don't understand this objection. I am proposing a change to the Social Security system that will save more than $20-25 billion per year (depending on the exact cut-off point. You worry that it will be ridiculously expensive to administer. The entire tax code (which is quite a bit more complicated than my proposal to say the very least) costs right around $10 billion. In reality, the most of the questions important for the cut-off analysis are already answered and adminstered by the IRS and thus would not require much more money to gather the data. But to be hugely generous to your objection I have not assumed the more realistic 1/20 or 1/10 of the cost of the IRS when it comes to administration. But even giving you the entire cost of the IRS there is a huge savings. You now have the objection that I can't force people to pay for that? It comes from the savings. And I don't see how you can call the savings hypothetical. I can't calculate the EXACT amount of the savings until we determine the EXACT location of the cutoff point. But that doesn't give you license to pretend that the number is anywhere near zero. Cutting out the payments to the upper class and upper middle class clearly would save money in the tens of billions of dollars range. That isn't hypothetical. That is simple subtraction. Unless you would care to assert that cutting out payments to the top third or top quarter produces very little savings? But you would need to be explicit about an assertion like that, because I wouldn't assume you would think that.

[Dantheman]

"In reality, the most of the questions important for the cut-off analysis are already answered and adminstered by the IRS and thus would not require much more money to gather the data."

Only if the cut-off is based on income alone. If it is based on wealth, which makes far more sense, a whole different set of data will be needed.

[Sebastian Holsclaw]

And? Even then will it cost more than administering the entire tax code? No. It would involve fewer people and no corporations. The budget items for the census appear to be scattered all over the place, but it seems that running the census program which involves physically sending huge numbers of people door-to-door costs less than $1 billion.

[Doh]

"I don't think there are enough people earning that much money to raise $20 billion a year, but I could be totally wrong about that."

Well, according to the Heritage Foundation ("Web memo #143"), it would be more like $100-120 billion per year. I'm sure there are JCT estimates floating around out there, but that one appeared on the first page of a google search.

Are you saying that your statement that $20 billion is "a significant portion of GDP" is not a typo? Do you have any sense of what GDP is? $20 billion is less than 0.2% of US GDP. It's a rounding error.

[Doh]

P.S. $20 billion is not even a "significant portion" of the federal budget-- it's something like 8%.

[Dantheman]

While determining eligibility for "Social Security 2.0" will involve fewer individuals as recipients than the Internal Revenue Code has individuals as taxpayers, there is far more information required.

Take for example, the asset of a house. For income tax, the primary information needed is purchase price, sales price and any carryover from a prior house. The purchase and sales price are easily determined, and the carryover is likely in the IRS's records. On the other hand, determining the value of a house involves numbers which are less certain, since there was no actual sale of the house. Instead, one estimates the value by comparing it to other nearby houses. However, houses vary considerably based upon information, much of which is not typically in public records (how old are the kitchen fixtures? is there evidence of water damage? how long until the roof will need to be replaced?, etc.). The secondary mortgage market has driven the cost of a terribly inaccurate drive-by appraisal (where none of these factors are considered and all incentives are to reduce the estimated value) to around $200 per house. To do the job correctly by a qualified person is closer to $1,000. Multiplying the latter number by the number of persons eligible for Social Security swamps your $10 billion estimated cost (and even the lower number likely will end up exceeding it).

And that's for just 1 component of determining wealth. Imagine the fun and cost in determining the value of artwork, antiques, etc, which your agency will need to do and which the IRS has no need to get involved in.

[Sebastian Holsclaw]

$20 billion isn't a significant portion of the federal budget? Then why so much whining about the cost of the Iraq war. Practically a rounding error.

8% is quite a bit.

Dan, first I suggested an income based model. The virtue of a wealth-based model would be that it excluded even more people from Social Security payments. So while a wealth based model might be great, it might be too difficult. But worrying about a wealth based model seems like a particularly poor criticism of my suggestion. A wealth based analysis would exclude even more people. Great, if you figure out how to do it write it up. The other option being discussed is excluding no rich people. Complaining that my proposal doesn't exclude enough rich people while arguing that all rich people should get benefits seems rather odd.

BTW, many rich people who aren't going to qualify won't have to be paid for by the administration costs of the system. They won't apply for benefits.

[Dantheman]

"The virtue of a wealth-based model would be that it excluded even more people from Social Security payments."

No (at least to me). The virtue is that it is more accurate in determining who would need and not need the payments.

[Sebastian Holsclaw]

So figure out an accurate way to determine it and tell us about it. But in general people with very high wealth will also have noticeable enough income to push past the threshold while it would be very odd for people with high incomes to have low wealth--especially late in life. So keying it to income is likely to be underinclusive, but as such would not tend to exclude poor people who we want to be able to use the safety net. I would rather err on the side of allowing more poorer people even if we slightly overinclude richer people. If you can find a way to accurately do that on a wealth basis, great. But it isn't really an objection to my argument that I can't do it perfectly. The current system doesn't do it at all.

[Edward]

I've been thinking about this idea that Social Security shouldn't go to the very wealthy and wondering if it can't just be voluntary.

Seriously, there could even be some sort of social recognition for it. At retirement you forego your claim to the money, because you don't need it, and get a certificate or medal from the President thanking you for being a generous American.

[Doh]

Sorry, 8% was a decimal place error (as opposed to a rounding error) -- $20 billion is 0.8% of the $2.4 trillion federal budget.

The Iraq war is more expensive by an order of magnitude, and that is just through the current fiscal year-- not on the "infinite horizon" the Bush Administration is pushing lately (when it suits them).

Sorry if the "whining" bothered you. Next time the Administration wants to squander all that money to test the theories of a few idiots in power (and wreak havoc on the lives of thousands of Americans), I'll be sure to say "thank you sir, may I have another?"

[Sebastian Holsclaw]

Iraq is expected to cost $20+ billion dollars every year for the next 75 years? I had no idea.

[Dantheman]

"So figure out an accurate way to determine it and tell us about it."

No thank you. You're the one saying its better and cheaper than the current system. I'm the one saying the costs of doing it correctly are higher than you think.

"in general people with very high wealth will also have noticeable enough income to push past the threshold"

Not really. Imagine 2 people -- one whose $500,000 house's mortgage was paid off, and is now living off a reverse mortgage or other borrowing against it and thus has no income, versus a person who has no substantial assets, and works at the grocery store to supplement Social Security. Which do you think should have their benefits cut? Which will an income only system cut the benefits of?

[Sebastian Holsclaw]

Why wouldn't the income from a reverse mortgage count as income?

[Doh]

"Iraq is expected to cost $20+ billion dollars every year for the next 75 years?"

Show me where I said that, please.

All I said was that Iraq would have costs beyond the end of this fiscal year, which I think is pretty indisputable. (As an aside, I think probably those costs will exceed the PV of $20 billion a year for 75 years, but that's just an "intuition").

Sebastian, for a while I thought you were in training to be David Brooks-- throwing out these posts that seem to tackle big issues but without any regard to facts, so that when you drill down on them there's nothing there. I admit that I frequent blogs to avoid that sort of punditry. In a couple posts recently, however, you seem to have graduated to putting inaccurate words in people's mouths. For myself I really think I'm going to have to classify your posts under "Not yet a Buddha."

[Dantheman]

"Why wouldn't the income from a reverse mortgage count as income?"

It is not according to current tax law, nor what most accountants understand income to be. You are trading 1 asset (the net value of the house) for another (cash).

To think of this another way, many people in the last few years refinanced their house and borrowed more than the loan they paid off, taking equity out. This is not income for tax purposes, even though they got cash in hand.

[Sebastian Holsclaw]

Doh,

"The Iraq war is more expensive by an order of magnitude, and that is just through the current fiscal year-- not on the "infinite horizon" the Bush Administration is pushing lately (when it suits them)."

Implication of bolded phrase was that we are analyzing costs at an order of magnitude greater than social security on a recurring basis. Did I misinfer?

If I did, what were you trying to say?

On reverse mortgages, I am rather surprised to learn that the don't count as income for income tax purposes. I see that is true through googling around, but I think it might be more complicated. "You are trading 1 asset (the net value of the house) for another (cash)." Trading stock for a gain in cash hits you with captial gains taxes. If my $50,000 house appreciates to $500,000 can I really get a reverse mortgage for the full amount and pay no taxes? Seems shocking to me, I suspect we are missing something in the equation. Though I fully expect you could define reverse mortgages payments as income for social security accounting purposes and get around the problem anyway.

I forgot to deal with part two of your hypothetical, "versus a person who has no substantial assets, and works at the grocery store to supplement Social Security. Which do you think should have their benefits cut? Which will an income only system cut the benefits of(f)?"

What grocery-store job is going to put you in the upper third of income? Management? And if you are earning more than $45,000 why should we pay your Social Security? Furthermore, haven't we already discussed reducing benefits on a sliding scale toward the termination point with a 2-1 or 3-1 dollar earned to dollar reduced ratio to reduce the incentive to cheat?

So the answer to your question (if we define reverse mortgage receipts as non-income for such purposes) is apparently 'neither'. Which, if you think the reverse mortgage problem is unsolveable, would be a great reason to object to my plan--apparently everyone would be eligible for benefits. Somehow we avoid that in other social programs.

[Jadegold]

Let's just face the fact Sebastian's plan has about as much chance of being enacted as Barry Goldwater's corpse being exhumed and installed as the Grand Vizier of Baghdad.

Soooo...anybody catch the appointed President's press conference today? You know, the one where he refused to negotiate with himself on the SS issue? It seems his staff has gone on early XMas vacation, not having prepped him better.

“I know what you’re trying to get me to do. You’re trying to get me to answer ‘Why this,’ ‘why that,’ to take positions -- don’t bother to ask me.”

Now, that's leadership!

[Dantheman]

"Furthermore, haven't we already discussed reducing benefits on a sliding scale toward the termination point with a 2-1 or 3-1 dollar earned to dollar reduced ratio to reduce the incentive to cheat?"

If all you are suggesting is that above a threshold, benefits are reduced on a sliding scale, I believe it was already enacted into law, and then repealed due to unpopularity (although my memory may be wrong). I would not strongly oppose it. I have been under the impression you wanted a threshold and an absolute cut-off above that point.

[Edward]

spambot bastard!!!!!

[Alicia]

Parts wind parts car my coupon, off back become face.