the IRS is looking to restrict retirement savings
The IRS had little to do with this other than restate what Congress required. Give credit where credit is due.

SECURE 2.0 introduced two notable changes to this system:
mandatory Roth treatment for catch-up contributions by high earners for taxable years beginning after Dec. 31, 2023
optional "super catch-up" contributions for participants ages 60 to 63 for taxable years beginning after Dec. 31, 2024

https://www.hklaw.com/en/insights/publications/2025/05/irs-proposes-key-changes-to-roth-catch-up-contributions
As a practical matter, the executive branch does have limited discretion in carrying out what Congress says, especially in making sure that the law can actually be executed:

Due to concerns that plan sponsors and recordkeepers would be unable to comply with the mandatory Roth catch-up requirement by the original deadline, Notice 2023-62 provided a transition period that delayed the effective date until Jan. 1, 2026 (although, a later effective date may apply for collectively bargained plans).

Even Congress isn't restricting retirement savings; see e.g. rforno's post above. What Congress has always done is to restrain the government's largesse by limiting contributions. That's far and away the larger restriction. And with its new "super catch up" provision, Congress is enabling earners to shelter of another $11K of assets that would otherwise sit in taxable accounts.
Still, not to worry if you're a really high earner (read business partner). Congress continues to give them favorable tax treatment on profit sharing (carried interest) and even on catch up contributions:

No FICA Wages, No Roth Mandate. Participants without FICA wages (e.g., partners who have only self-employment income) are not subject to the Roth requirement.

Stuff like this is another reason I'm glad nearly all of my retirement assets are in taxable accounts .. i don't need to worry about playing these games, planning withdrawls, dealing with RMDs down the road. Keeps things rather simple.

SECURE 2.0 (2022) - Final IRS rules for Catchups for 401k/403b
From 2026, all catchups above the threshold at a single job must go into Roth 401k. So, below the threshold, catchups can go to 401k or Roth 401k.
Threshold used is for income in the prior year & are inflation-adjusted: $145K (2025), $150K (2026 est)
If Roth 401k isn’t available, then one is out of luck – complain to HR.
Catchup for 50+ is $7,500 (2025), $8,000 (2026 est).
Super Catchup for 60-63 is $11,250 (2025, 2026 est). It’s about 150% of the regular catchup amount.
Relevant age is for the DOB in the applicable calendar year.
Estimates will have to be confirmed by IRS later.
403b rules are similar to 401k, but there is also a limited 15-yr catchup.
Catchup for IRA & HSA is $1K. These have their own income limits.
IRS Final Rules, 9/15/25
https://www.irs.gov/newsroom/treasury-irs-issue-final-regulations-on-new-roth-catch-up-rule-other-secure-2point0-act-provisions
(LONG) https://www.federalregister.gov/documents/2025/09/16/2025-17865/catch-up-contributions
Some IRS links in 09/2025 show rules only for 2025, not for 2026.
Other Sources
https://www.wsj.com/personal-finance/retirement/high-earners-age-50-and-older-are-about-to-lose-a-major-401-k-tax-break-75572091?mod=hp_lead_pos11
https://www.fidelity.com/viewpoints/retirement/catch-up-contributions

The 'catch-up' is lost only by those over 50 who earn > $145K (per employer), at least for now.
But it's a horrible signal that the IRS is looking to restrict retirement savings. The govt is so wasteful in its spending and this is simply not an area to be targeted.
There is no excuse for this.

I know this can be a 'good problem' for many, but the meta is clear: the IRS restricting 'catchup' contributions for older workers who can afford to do so is one more shift towards the government creating an uncomfortable retirement for many as costs continue to rise everywhere.
(free link)
https://www.wsj.com/personal-finance/retirement/high-earners-age-50-and-older-are-about-to-lose-a-major-401-k-tax-break-75572091?st=o7HbQA&reflink=desktopwebshare_permalink

@msf Excellent. Great information and clarification. Thanks again!
I know that a Roth conversion is a taxable event, of course. But, does it also count as unearned income, as it pertains to LTCG tax treatment. Or is it more of a "unique" event?
I've thought about delaying anywhere from 6 months to 24 months. And using that time to cash out some LTCG positions, perform Roth conversions and/or spend down some (taxable?) accounts.
I am playing with some tax calculators, trying to see what works best. Our spending needs will drop off significantly in 2026. We have been spending on home improvements, automobile upgrades, education, medical/dental, all in preparation for retirement over the past 6-7 years. All of that will be behind us at the end of this year.
We are about 62% tax-deferred, 5% Roth and 33% taxable. Our taxable accounts hold a great deal of LTCG and cash.
A few articles/calculators that I have squirreled away on retirement taxation:
https://www.kiplinger.com/article/retirement/t037-c032-s014-tax-efficient-retirement-withdrawal-strategies.html
https://www.irscalculators.com/tax-calculator
https://www.schwab.com/ira/ira-calculators/roth-ira-conversion

If I am reading it correctly (probably not), my spouse still gets more than her current benefit (claimed at 65 and much lower earnings), when I claim at FRA and she switches to spousal benefits, just not as much as it might've been had she also waited until FRA. We are about the same age. Is that correct interpretation?
Yes.
And the spousal benefit is half of my PIA, regardless of when I claim. Though she cannot switch to spousal, until I start my benefits? In layman's terms her spousal benefit is already fixed/determind before I decide to claim?
Yes. When you claim (as opposed to when you stop working) affects when she can switch to spousal benefits but doesn't affect the amount of those benefits.
Though the longer you work, potentially the larger your PIA becomes, since it is based on your highest 35 years of earnings. And a larger PIA makes her spousal benefits larger. So to maximize spousal benefits, claim early (to start those benefits earlier) but continue working (to increase PIA).
Survivor benefits work the opposite way. The longer you wait before claiming (up to age 70) the larger your own benefits become. Consequently the larger her survivor benefits become (if/when you predecease her).
Having a spouse complicates life :-)
Also, it appears that the SSA is using the terms "FRA" and "normal retirement" interchangeably?
That's certainly what it looks like. Here's an SSA table with a column labeled "Full (normal) Retirement Age". If I'm wrong about the NRA don't shoot me (ouch!).
the specter of cut SS benefits is not unrealistic. So, another unknown variable. And that suggests one should take the money and run.
If the government doesn't reduce benefits, you'll reach the break even point in about 14 years, i.e. at the end of 2039. If nothing is done, the government is predicted to cut benefits by about 1/4 after 2033. So for the last 6 years (2034-2039) you'll be catching up only 3/4 as fast as originally planned.
So rather than taking another six years after 2033 to catch up, you'll need another 8 years (4/3 x 6) to catch up. That makes your break even point the end of 2041 instead of 2039. Whether that extra two years tips the scales is up to you to decide.
The fact that SS is inflation adjusted makes the calculation above simple. It's all in real dollars so you don't have to worry about inflation or depreciating future dollars or present value. It's already baked in.

@msf Thanks for all of that information, most appreciated. I see how muck work you put into it.
I'll clarify the last part first. The anxiety would emanate from wondering the entire time, if I was going to live long enough to break even. lol From wondering if I chose poorly. Not a big issue either way though. More of an afterthought.
Definitely not worried about income or our retirement funds lasting. More about making a "best" choice.
I am now a bit confused about the spousal benefit. If I am reading it correctly (probably not), my spouse still gets more than her current benefit (claimed at 65 and much lower earnings), when I claim at FRA and she switches to spousal benefits, just not as much as it might've been had she also waited until FRA. We are about the same age. Is that correct interpretation?
And the spousal benefit is half of my PIA, regardless of when I claim. Though she cannot switch to spousal, until I start my benefits? In layman's terms her spousal benefit is already fixed/determind before I decide to claim?
Also, it appears that the SSA is using the terms "FRA" and "normal retirement" interchangeably?
I have been vacillating on taking SS at FRA. You have given me something to consider.
@sfnative. A lengthy but valuable read. Thanks.
Based on what has happened thus far, the specter of cut SS benefits is not unrealistic. So, another unknown variable. And that suggests one should take the money and run.

And the wife, who took her benefits at 65, gets a bigger (spousal) benefit at the same time
FRA = full retirement age, PIA = primary insurance amount
Wife spousal benefit (if less than her own benefit) =
1/2 (or less, if taken before her FRA) x your PIA (independent of when you take benefits).
Spousal benefit is reduced (below 1/2 of PIA) if spouse (here, wife) starts taking benefit before their FRA. But when you take doesn't matter.

If you claim your spousal benefit at FRA, you will receive ½ of your spouse’s PIA regardless of when your spouse claims. I.e. a higher earning spouse claiming his/her reduced benefit at age 62 will not affect a spousal benefit claimed at FRA. A higher earning spouse claiming early would, however, affect Survivor benefits, but that is a topic for a future blog.

https://counterweightpw.com/insights/should-my-spouse-claim-early-understanding-social-security-spousal-benefits

The spousal benefit can be as much as half of the worker's "primary insurance amount," depending on the spouse's age at retirement. If the spouse begins receiving benefits before "normal (or full) retirement age," the spouse will receive a reduced benefit.

https://www.ssa.gov/oact/quickcalc/spouse.html
That links to this SSA definition of PIA (emphasis added):

The "primary insurance amount" (PIA) is the benefit (before rounding down to next lower whole dollar) a person would receive if he/she elects to begin receiving retirement benefits at his/her normal retirement age. At this age, the benefit is neither reduced for early retirement nor increased for delayed retirement.

https://www.ssa.gov/oact/cola/piaformula.html
Finally, please don't rely on AI for help. Here's what Google's AI said (searching for spousal benefit half PIA early retirement):

Your Claiming Age: The spousal benefit calculation uses your PIA at your FRA, not the actual amount you are currently collecting. If you claim your own benefit early (before your FRA), this will result in a lower PIA, and thus a lower maximum spousal benefit.

The first sentence appears correct - your spouse's benefit depends on your PIA (which is calculated as if at FRA). The second sentence is not - if you claim early, that reduces your benefit but not your PIA, which is defined as what your benefit would be at FRA. Thus it does not result in a lower maximum spousal benefit.

My situation seems simple. FRA is 66 and 10 months. My employer will be in no need of my services at almost precisely that time. I also subscribe to the notion that by taking SS, one can let investments grow, as Crash suggested. And the wife, who took her benefits at 65, gets a bigger (spousal) benefit at the same time. My state does not tax retirement benefits, either. And I will be retiring mid-year, so less earned income.
Life expectancy is an unknown. Dad lived to be 96. Mom lived to 81. I think my situation may be closer to my mother's. Split the difference and it would be around 88.
I could delay benefits and use taxable funds, paying lower LTCG. Or make bigger Roth conversions during those years. Splitting it in the middle ( 62 - 70) seems like a decent choice. Maybe not the optimal choice. But waiting would make me anxious, I think.

Do you mean your FRA was 66+ (born between 1955 and 1959 inclusive) or actually 67 (born in 1960 or later)? Those born after 1959 haven't yet reached FRA, so at least one of us is a bit confused.
As yogi's chart above shows, a plurality of people take SS as soon as they are able to (age 62). They may be concerned about losing to the government by dying early, or they may be in need of cash flow.
There's a much smaller spike at age 70. Those people hope/expect to win their bet with the government that they will have a long life. And they are not strapped for cash before then.
The middle spike, not quite as large as the number of people taking benefits as soon as they are eligible, is at FRA (age 66). People might chose that age for a few reasons:
- This maximizes their spousal benefit. If their spouse gets the higher SS payments and dies first, they get that larger income stream. Except that this amount is reduced if they started SS before FRA. So waiting until FRA (but not later) has a benefit.
- This avoids a temporary reduction in SS benefits if they are still working and draw SS before FRA. The reduction is only temporary because the full amount of the reduction is added back (inflation adjusted) to their benefits once they reach FRA. Still, this is a temporary reduction in total cash flow ($1 reduction in SS for every $2 in earnings).
- FRA is perceived as the proper age to start drawing SS. That's when "full" benefits begin. Though what "full" means when there's a sliding scale between ages 62 and 70 is somewhat fuzzy.
ISTM that the reasons for taking benefits at age 62 or age 70 are more compelling than taking benefits at FRA, though selecting FRA does confer some benefits also. As others have said, each person's situation is different.

My full retirement age for social security was 67 and I started taking it, never regretted not waiting until I was 70, so many friends passed away while waiting to draw it at 70. Its a government gamble waiting to 70. The gamble for you that benefits the govt is that you never use it.

So Gretchen Morgenson wound up at NBC. For many years she was a muckraking business reporter at the NYTimes. Mostly excellent, though IMHO she occasionally latched onto something so much that she went over the top. There's some of that here.
TIAA has been under pressure for a couple of decades to retain assets and improve profits. Like many employer plan providers, it began offering retirement plan advice once the Dept of Labor opened up the floodgates in 2006.
Of course TIAA's advisors steer participants into more expensive plans. An irony in the report is that it holds Vanguard out as a model of what should be done, when Vanguard just settled with the SEC for failing to "disclose to clients that its advisors had financial incentives to funnel them into certain managed accounts."
There are no white hats in this industry.
I largely agree with yogi that the article exhibits poor understanding of fixed annuity fees. But Morgenson knows better and does not lack for understanding. Her complaint here is over the top. One does not ask a bank what its expense ratios are on its CDs; the profits are built into the rates it offers. Likewise, one does not ask an insurance company what its expense ratios are on its fixed annuities.
Why not ask Vanguard what the "expense ratios" are of the underlying bonds in its MMF portfolios? Just how much does the Treasury make on its T-bills?
To nitpick, it looks like PTTRX (the only Pimco offering) and RERGX (the only American Funds offering) are only in RI's 457 plan, not the 401(a) plan that's the subject of the article.
https://www.tiaa.org/public/tcm/ri/view-all-investments

This was posted in Facebook TIAA Group and here was my reply:
"My take - 2 links are at the end.
1. Plan fees are $32/yr plus fund ER (if applicable).
2. TIAA Traditional is RCP (flexible). There is no ER as it's run directly from TIAA General Account. RCP rates are published monthly (& here too) & seem good. The guessed ERs in the NBC article are speculations & show poor understanding.
3. Also available are TIAA Stable Value (with lower rates) & Vanguard money-market fund.
4. TDFs available are TIAA RetirePlus Select (ER 3 bps only) & Vanguard TDFs.
5. Other funds available are from American Funds, Pimco, State Street, Vanguard.
6. As for restrictions on in-service withdrawals & loans, neither is available, but that is the decision of Rhode Island. I am surprised that loans aren't allowed. But why blame TIAA? Complain to Rhode Island HR.
IMO, it's a GOOD plan & I hope that TIAA responds to the shoddy NBC piece. If not, you are at the right place to get right information (-:)"
NBC https://www.nbcnews.com/news/us-news/rhode-island-sheriffs-retirement-account-woes-bring-scrutiny-state-run-rcna229290
TIAA Rhode Island DC Plan https://www.tiaa.org/public/tcm/ri/retirement-benefits/plan-405868
https://ybbpersonalfinance.proboards.com/thread/918/tiaa-rhode-island-plan

Ya, it occurs to me that if you're a household of one, and you are a saver rather than a spender by nature, you might claim earlier than Full Retirement Age; take the amount you do NOT spend, and put it back to work by investing, or buy a bond fund or create a CD-ladder, depending upon prevailing rates.

With SS benefits to be cut in 2033, unless more money is found to finance it, may also play into your plans for when to take SS
This is discussed in the piece as one of several risks in deferring:
https://www.kitces.com/blog/discount-rate-delaying-social-security-benefits-retirement-planning/#policy-risk
And here:

The possibility that policymakers could reduce benefits after someone has already forgone years of payments to maximize their age-70 benefit triggers what Sandman might call extreme outrage: the outcome is controlled by others (politicians), morally relevant (breaking societal promises), and unfair (changing rules mid-game).

https://www.kitces.com/blog/discount-rate-delaying-social-security-benefits-retirement-planning/#regret-risk
One way to read this piece is to skip to the bottom where two case studies are presented.
One is a case study of a couple with limited resources. They might take SS earlier and make better use of their smaller nest egg in their 60s while they can enjoy it.
They are also somewhat irrational (i.e. human). They would more regret the loss of SS income from dying early than they would enjoy getting extra from SS should they live long and prosper (more on that in Case 2). Especially since they don't anticipate a long life span (based on heredity).
https://www.kitces.com/blog/discount-rate-delaying-social-security-benefits-retirement-planning/#case-study-claiming-early-with-limited-assets
Case 2 is Mr. Spock and spouse. Larger nest egg easier so easier to fund expenses until drawing deferred SS. Long life expectancy, entirely rational. No regrets on what might have been - probabilities are just that.
https://www.kitces.com/blog/discount-rate-delaying-social-security-benefits-retirement-planning/#case-study-claiming-later-with-a-tips-strategy
Some people need to draw SS as soon as it's available because they need the cash flow. Others, like Spock, can make entirely cold, calculated decisions. Many people fall somewhere between the two (closer to case 1).

A Siting from a 2024 Journal of Financial Planning article, Smith and Smith conclude: Our calculations do not support the presumption that the vast majority of people who choose to start their Social Security retirement benefits before age 70 are making a mistake. For example, … with a 4 percent real return, a person has to live to 89 for it to be beneficial to delay the start of benefits from age 67 to 70. However, 77 percent of 67-year-old males die before 89 as do 65 percent of 67-year-old females. Age 70 is not the most financially rewarding age to initiate benefits unless an individual has a low discount rate and/or is confident they will live several years past their life expectancy.
From the Author of the linked article at Kitces' website: Ultimately, the key point is that we need to move beyond simply thinking in terms of portfolio risk when assessing Social Security claiming analysis discount rates. Ideally, we should be thinking more in terms of utility and factoring in all risks, which changes the calculus significantly.

delaying-social-security-benefits-retirement-planning/?

MFO truncated the long post, so the length limit may have been reached. Here is the rest.
INCOME INVESTING/FUNDS. MULTISECTOR BOND funds are riskier but allow exposure to several types of bonds – sovereign, corporates, HYs, EMs. A mix of core and multisector bond funds should be fine for most retail investors. Mentioned are JGIAX, LSBRX, MIAQX, PONAX, TQPAX. (After 12/31/25, LSBRX may have up to 35% in HY and 20% in equity. Hopefully, it will be reclassified as conservative-allocation fund, not as multisector bond fund)
FUNDS. With HY spreads tight (280 bps only), it wasn’t a good time for Vanguard to launch a self-standing active HY ETF VGHY (ER 0.22%, the same as investor VWEHX, not Admiral VWEAX). Vanguard sees it as just filling its ETF line up.
FUNDS. Foreign stocks are outperforming US stocks. Global equity funds as well as several allocation/hybrid funds and TDFs that have foreign stocks are doing well. Death of the old 60-40 turned out to be premature and it’s doing fine now. Mentioned are ETF TGLB; OEFs ABALX, MDLOX, VSMGX.
FUNDS. With lower rates, consider a mix of cyclical stocks (IJR, JRE, XLF; C, CEG, FITB, PNC, VST), defensive stocks (AMLP, XLU, XLV; CPM, CVX, IDA, KO, LLY, PG, PM, UNH, WEC, WELL), US and foreign bonds (HSNIX, PGNPX, VMBS), or allocation/hybrids (none mentioned, but see above). Economic outlook is muddy. Atlanta GDPNow has Q3 at +3.4%, Conference Board at +1-2% (real) is similar.
FUNDS. Where to put CASH? Suggested are T-Bills, money-market funds, ultra-ST bond funds (MINT), ST bind funds (IGSB), ST CDs (Marcus 7-mo 4.15%), online savings account (Barclays 3.9%). (Examples should be takes as “for instance” as there are many other competing choices.)
FUNDS There are lot of mutual-fund-to-ETF conversions. A new crop of ETF classes of funds is also coming (after Vanguard’s patent expired). Reasons are tax-efficiency and lower ERs of ETFs, and much feared frontrunning of active portfolios hasn’t materialized. ETFs also have transparent, semitransparent and nontransparent (almost dead now) structures to suit the needs of sponsors. These conversions may not be good for funds in small and illiquid markets (AFSC is cited as a bad conversion). Several large conversions include DFAC, DFAT, DFAS, DFAX, DFIV, DFUS, DFUV, FELC, JIRE, JMTG. (In related news, SEC will do quick approvals for ETPs based on established commodity futures and that may lead to a surge in new crypto ETPs) (By @lewisbraham at MFO)
(There is also a new SCOREBOARD for funds with weekly and YTD performance data through Thursday from Lipper – Top 25, Bottom 10, Largest 25. Unclear whether this is one-time or a new regular feature.)
RETIREMENT & WELL BEING. If your employer cooperates, try partial retirement first by temporarily reducing the workload or taking a short break (staycation vs vacation). (Unfortunately, some are forced into retirement involuntarily.)

Several related articles prompted a return of this ad-hoc series only after a week.
LINK1 LINK2
UP & DOWN WALL STREET. Investment-grade CORPORATES (US & foreign) are now seen safer than respective sovereign debts. Global deficits, spending and debt servicing are rising. The assumption of risk-free Treasuries when US continues to print money is becoming shaky. Corporate spreads are tight because the baseline Treasury yields are rising, and corporate yields are lower due to high demand. MSFT AAA 10-yr yield (the only other genuine AAA is JNJ; ignoring tons of artificial AAA securitized credits) is now below 10-yr Treasuries. Some French corporates also have yields below French sovereigns. This phenomenon of negative spreads will become more commonplace. Many corporations are taking advantage of low spreads to issue new debt. US Aggregate Bond Index (AGG, BND; the so-called US total bond market although it’s only investment-grade) has 50%+ Treasuries (vs 21% in 2007). Just as SP500 has become concentrated in Magnificent 7, the (so-called) US total bond market has become concentrated in Treasuries. Options for low/no Treasuries include (true) total bond market (IUSB), corporates (LQD), etc. Central banks have also diversified and now hold more gold than Treasuries.
STREETWISE. QUARTERLY or SEMIANNUAL reporting? It won’t matter if that’s made optional and then the market decides. Formal reporting requirement goes back to 1934 (regulators want to do something after a major crash) and quarterly reporting has been required since 1970 (there was a minor market hiccup then, but did that stop the big dot. com bubble and crash?). So, now there are 3 unaudited 10-Qs and 1 audited 10-K; companies can delay them with permission. President TRUMP made a similar proposal in 2018, but SEC then did nothing beyond holding public hearings – this time may be different (agency heads have rolled for lesser offenses).
Nasdaq is favoring semiannual reporting citing savings. Of course, private-equity/credit saves LOT of money by infrequent or no reporting, and did you hear that it may be going into your 401k? US companies spend 37% on R&D vs only 17% for European companies – will that change by reporting frequency? In Europe, 50% of companies report quarterly, 50% semiannually, so what? Most people may instinctively react negatively to reduced investor information, but how many really dig through 10-Qs and 10-Ks? Reg FD forbids sharing of nonpublic information with selected groups, but some of that goes on and more of that may happen with infrequent reporting. Analysts would still be far off the marks as they are in their annual estimates, sometimes revised several times during the year. A study by Brown U suggested a split system – large companies reporting quarterly, small companies reporting semiannually. Journalist/humorist HOUGH recommends that the best would be no reporting at all, let companies save lot of money and let FOMO and mojo rule (-:)