How should we "save" Social Security?

I’m not against SS paying a more generous benefit. We should increase taxes that fund it, and the benefits that it pays out. I realize projections are not certainties. But all projections I have seen show that “fixing” SS is a technically easy thing to do.

Medicare is the more difficult item.

No we don’t. We can add a cap. I believe that the curve flattens and there are diminishing returns at a certain point currently anyway.

Your other point about a reduction in income tax as a result of increased SS tax is a good one that I hadn’t considered. I have never seen that discussed when talking about raising the cap.

The SS benefit formula currently has 3 tiers: 90% of your former wage, 32% of your former wage, and 15% of your former wage up to the SS-taxed maximum. Which for 2023 is $160,200.

So much like income taxes in reverse, the relatively poor get a real good deal and the very highly paid get a less-good deal. Someone who was always in the lowest wage bracket would receive a benefit of 90% of their pre-retirement wages, whereas someone who earned exactly the SS max every year would get ~25% (a blend of the 90%, 32%, and 15% brackets) of their pre-retirement wages.

Under current law, someone who’s wages are more than the SS maximum each year pays no extra tax on those extra wages and and gets no extra benefit as well.

Changing that so they pay more tax but get no more benefit changes SS from an admittedly re-distributionist, but still you-pay-in/you-get-out system to one that is simply “A tax on me given to you”. In other words, welfare.

Which is a dirty word in the USA. There will be a very different sort of pushback if the SS wage cap for tax is raised with no SS benefit increase to go along.


As to my second point, I suspect you misunderstand what I meant. I did not say that very well.

Various commentators look at people like my exemplar $3.5M executive and think that if we eliminated the SS wage tax cap then $3.5M times the SS tax rate would flow into the system. Nope. It’d be $500K times the SS tax rate coming in. Compared to $160.2K times the SS tax rate under 2023 law. So the bump in SS-taxable value would be ~$340K, not $3.34M.

As most high earners / rich people / fatcats derive rather little of their income from wages, any projection that ignores this difference is at risk of overstating the gains to be made by a very significant factor. Said another way, in our highly unequal society, just looking at gross household incomes is a very bad proxy for gross household wages.

SS is and always has been all about wages and only wages. Both at the taxing end and at the benefit end. It (partially) replaces your wage paycheck, not your savings / investment earnings. Be the former or the latter great or small.

Of course one could simply change the system by deciding all income of all sorts is subject to SS tax. But once again that is a significant ideological change that will bring out a very different sort of opposition. That can of worms can be opened. But it would be a political / policy mistake to assume it doesn’t exist then be surprised when the worms spring forth.

I do think politically, (morally?) if you triple the amount I’m paying (as an example taxing everything from $160,000 in wages to $500,000 in wages) you should pay me something more. Certainly not triple, but something noticeable.

Everyone making a million or more kicks in more. A lot more.

Hear me out, first five million are exempt, but from then forward, every million in income means you kick in $100,000. Which goes entirely to social programs. You don’t get more benefits, BECAUSE YOU DONT NEED THEM! You don’t get a tax write off or deduction.

Tons of things that shouldn’t be, are means tested every day. Time to start doing it to the 1%, not just the welfare mom, or immigrant.

Please correct me if I’m wrong, but I’ve read that many (most?) people, between their and their employer’s portion, end up paying more into SS than they get back in benefits.

Given a 35 year time frame, this seems insane and like the biggest problem with SS. If I put $5400/yr into Excel’s FV with a rate of 5% for 35 years, I get $487,729 on total deposits of $189,000.

Now, I realize most of us start at a low salary and hopefully work up to a good one, and that math is beyond me. But it seems to me that the biggest problem with SS is that the input money is just stuffed in the mattress.

Am I missing something? (Besides the fact that there are those who think there’s already a massive surplus of capital sloshing around the system)

If you live long enough you’ll probably get more back than you put in. If you die before you start to collect, you get nothing. You probably have to live into your 80s to get more back. I’m sure someone has done the math. It’s beyond me.

SS is ridiculously easy to fix (and I’ve expressed my opinions previously.) The ONLY impediment is finding the political will, and admitting the reality, that it IS an income redistribution program.

As with so many aspects of our economy, the wealthiest - who profit the most from our system, scream the loudest about how they would be harmed by any changed.

Using the existing formula up to the new maximum (or up to infinity if the maximum were scrapped altogether) would have that result:

I can live with that.

But I’m pushing back on this:

Most people aren’t putting in $5400/year for 35 years. For example, 35 years ago (1988) the average wage reported to SocSec was around $19,000 and the tax rate (employer + employee) was 12.12%, for a total contribution around $2300.

Even that is probably overstating the contribution because an early-career worker probably isn’t getting paid anything close to average wages. Putting in a third to half as much in the early years is going to significantly decrease the compounding and hence the final total.

Meanwhile, the average retired worker is currently collecting $1828/month, or a little shy of $22K annually. The average person retiring at 65 can expect to live around 20 years (19 for men, 22 for the ladies), so even if there is never another COLA they can expect to receive ~$440K, and SocSec benefits are inflation-adjusted so most will get rather more.

Finally, SocSec comes with disability, survivors, and dependents benefits: currently, around 7.8 million disabled workers, 5.8 million survivors, and 3.8 million dependents of retirees and the disabled receive their own benefit checks each month. The cost of that insurance would have to be figured separately if SocSec wasn’t available.

In other words, most people get a lot MORE out of SocSec than they contribute.

Fair enough, and I’m not trying to assign homework (but I really don’t know how to calculate this), but I’d like to know:
Someone starts work at 20, making let’s say $40k/yr, works to 62, and ends at $75k/yr. That seems like a reasonable average worker, to me. For this, let’s ignore inflation because hopefully it cancels out. I think we can assume equal annual increases to go from 40 to 75.

What is that person’s total pay-in, and what would that person have if those same contributions were in a retirement account earning 5% / year (which should be pretty conservative)?

Maybe- you have to account for survivor’s benefits. When I was young there was a family where the breadwinner father died when the kids were young. Because of the number of kids and the rules in effect at the time, by the time the youngest kid was no longer eligible for benefits , Mom was old enough eligible for spousal benefits. That family collected far more than Dad ever contributed and that’s not nearly the only situation - people get disabled in their 30s and 40s and between their benefit and the dependents’ benefits , they get paid more than they contributed. My grandfather started collecting SS disability in his 50s and lived to be 89 - he definitely got back more than he contributed. And for some people to get back more than they put in others must get back less. But that’s the nature of insurance ( which is at least part of what SS s ) Nobody thinks there’s anything wrong with paying for homeowner’s insurance for 35 years and getting nothing back because there weren’t any fires/thefts/lawsuits.

Excellent point. Thanks for the correction

For starters, somebody beginning at $40K and retiring at $75K is already above average; median wages in 2022 for fulltime workers aged 20 to 24 were $36,712, rising to $63,648 for those aged 45 to 54 before falling back slightly for older workers.

Also, if you retire at 62, you’re cutting your SocSec benefit by 30% for retiring five years before your full retirement age.

For purposes of this calculation, are we assuming this someone never had a period of unemployment, unpaid leave, etc.? Any such period (a woman taking unpaid or only partially paid maternity leave, e.g.) would have a permanent effect on compounded interest, but need not affect SocSec very much since SocSec counts only the 35 highest years of earnings anyway.

To what age are we assuming this person lives? Somebody retiring at 62 and dying at 70 has a different financial profile than somebody retiring at 65 and living to 100.

Is five percent really a conservative figure? The average annual return of the stock market, for example, is only about six and a half percent after inflation, and being fully or even mostly invested in stocks as you approach retirement isn’t usually considered conservative.

Well, yes; that’s how it generally works when you put money in decades ago and take it out now.

The scenario under discussion was someone starting work at age 20 and continuing to age 62. Long-term investment in the stock market for most of that period, shifting to less volatile investments toward the end, is a standard strategy.

Sure, but those less volatile investments have not tended to have yields anything close to five percent in recent years. The average annual yield on bonds, for example, has been under two percent for most of the past decade, so it doesn’t make a lot of sense to assume you’re going to still be earning that high yield after you’ve shifted out of the stock market.

Median or average? Anyway, it doesn’t matter. Are you asking because you are offering to do the math (in which case, use whatever assumptions you like)?

I chose that age because it’s the earliest you can retire (w/benefits). But since what I’m interested in is how much the value of the contribution would be if it were invested, it doesn’t really matter what the benefit is. We could compare at 67 if we prefer.

Jesus. It seems unnecessarily complicated to get into that. If one isn’t working, one isn’t paying into SS, one isn’t paying into our hypothetical alternative investment, and one isn’t accruing SS benefits. Although, money already invested WOULD continue to compound (or not, depending on the market) while the SS benefit would be static or declining.

But that has no bearing on what the value of the investment is at 62 (or 67, or 700)

I believe I said let’s ignore inflation for this. And when you say stock market, what do you mean? Total market? S&P? For the S&P, I’ve never heard anyone use such a low number, especially with reinvested dividends, which I think should be the assumption in a retirement savings scenario.

It’s not clear to me if you’re trying to refine the hypothetical or just poke holes in it, but the fact remains that the money paid into SS could be invested in another way, and we could make a reasonable effort of what that would look like.

You still “get”. Just not a lot more. Same with taxes- the poor get more out of their taxes that rich people do, in many cases.

Yes, we should take off the cap, and still increase the amounts paid out- but in a smaller amount.

He wasn’t talking about “means testing”.

Yeah, that and Unemployment should not be taxed.

He included pensions.

I do not like means testing. A slow age raise is fine. Getting rid of the cut off will save SocSec for a long time.

Yep. Make it 15% up until say $200K, then 10% up to 1M then 5% after that.

Sorta, but remember the dollar they put in 35 years ago is worth more.

Not to mention, that doesn’t count all those who put in, but take nothing out , like those who die early, or undocumented aliens.

It isn’t a “an income redistribution program.” Sure, like many things, some people get more out than pay in. Unemployment insurance for example. But others get nothing out.

Yes. Of course.

Those 1988 dollars were worth more than 2.5 times as much, maybe nearly 3X now.

So that $2300 was really more like $5700, which is MORE than your $5400.

Or nothing at all.

Yep.

Which is why the standard advice is to put money in the stock market, let it ride the ups and downs over the years, and shift it to bonds and such as retirement approaches and stability becomes more important than yields. With that strategy, five percent over the entire period is indeed a conservative estimate.