Avoid early withdrawals in early retirement // Roth conversion ladders, IRS rule 72t Substantially Equal Periodic Payments SEPP, early retirement savingswe've learned

Minimizing Early Withdrawals in Early Retirement // Hang On to As Much As You Can for Later

We are planning a two-phase retirement: a more basic lifestyle in phase one (what the Elephant Eaters have dubbed “dirtbag millionaires”), supported by our taxable investments and rental income, followed by a more baller traditional retirement, funded by our more gold-star worthy 401(k)/tax-advantaged funds.

Income Plan Over Time // early retirement and traditional retirement, taxable investments, rental income, tax-deferred investments

One of the things we consistently get questions about is why we’re taking the two-phase approach, especially with well-funded retirement accounts already in place, instead of starting to tap those funds now, and already being able to retire. Of course, what they’re really asking is Isn’t it all just one big pool of money? You can certainly choose to see it that way, but I’ll be crystal clear on where we stand on the question:

No. We don’t think it’s a good idea to see taxable and tax-advantaged funds as one big pool of money.

And we definitely don’t think it makes sense to apply the 4 percent rule to your total balance to figure out what you can spend in your early retirement years, before you can access your tax-advantaged funds without penalty. (We don’t think you should be using the 4 percent rule at all, actually, but that’s another post.)

I know them’s fighting words for some, and I can already hear the imaginary retorts about all the ways someone can access tax-advantaged funds early. “But the Roth conversion ladder!” And “But the substantially equal periodic payments in rule 72t!” Yep, those are things. But we have three big reasons why we aren’t relying on them to fund the early part of our early retirement:

We massively disagree with the conventional wisdom that people naturally spend less as they get older. 

Or even that it’s a fine idea to plan on level spending over your lifetime.

While it’s true that research shows people DO spend less in older age, that research never asks if that’s by choice or if they are forced to spend less because of financial hardship. That’s like asking Irish people during the potato famine if they find themselves eating fewer potatoes, and assuming that they must just be avoiding carbs. Matt at the Resume Gap wrote an excellent post on the question that I wholeheartedly agree with, and he makes the great point that many people in their 70s and beyond simply can no longer afford their pre-retirement lifestyle, which skews the scale.

But in addition to that, it’s still a huge X factor what health care will cost in the future, and not just in the U.S. Most of the countries that provide free health care are also facing demographic challenges that may force greater cost-sharing in the coming years, as more people age out of the workforce and fewer young people enter it to support them. (So don’t get too comfortable, all you folks feeling smug about your better-than-America health care. Or do. Because we’re still totally jealous.)

But we think it’s a mistake to sell out our future selves to escape the workforce a little earlier, and we’re leaving our true retirement money alone so we can both have a less-dirtbag lifestyle and have a bigger safety net for health care expenses and whatever other unknowns might be lurking out there.

Tax rules can and do change, which could spoil your early withdrawal plans.

The backdoor Roth, for example, has been targeted for years as a loophole that needs closing. This applies to the “mega backdoor Roth” too. Given how little people save for retirement, even when they have tax-advantaged vehicles like IRAs and 401(k)s available to them, and especially given the ethnic and racial disparities in savings rates and the ways some say 401(k)s really just help the rich get richer, it’s not at all unthinkable that Congress might rethink our tax laws that pertain to retirement, especially over a long time horizon like many of us have in our early retirements. The current system isn’t actually working, after all, to prepare Americans for retirement, so a responsible Congress (I know, I know) would take a long, hard look at this. For those of us who are the lucky few benefiting from IRAs and 401(k)s, it’s dangerous to be wholly dependent on rules that could change.

Relying on Roth conversions and rule 72t increase the temptation to accidentally sell out future you.

There is nothing inherently wrong with using currently available tax rules to access money you’ve saved, even if the law intended that money to be for your later years. The problem is that relying too heavily on your tax-advantaged funds for early retirement makes it easy to accidentally spend some of the money older you needs later. The money you could convert or withdraw early, tax free, might be money that age 60+ you needs more than age 30/40/50-something you needs it (see the first reason above), but you can’t go back and fix that later on if you’ve already spent the money in your younger years. And by the time you realize that you should have saved that money for later instead of converting it or withdrawing it early, you’ll have a much harder time earning more income than you would have in your 30s/40s/50s.

(And if you say, “But X blogger has more than they need in their future funds, so I think I’ll be fine,” or “But based on historical averages, we’ll have plenty left by the time we hit 60,” don’t fall prey to recency bias. Instead, read this post. And note that everyone currently writing about Roth conversion ladders is still a 30- or 40-something, not a 60- or 70-something who can affirm that they still have enough money left for their later years.)

All of this really boils down to:

Don’t steal from future you. Love future you.

Give future you the gift of no money stress.


More Cautions Against Early Withdrawals

I love that Brandon the Mad Fientist did a detailed, side-by-side analysis of different early withdrawal strategies: the Roth conversion ladder, the substantially equal periodic payments (SEPP) in rule 72t and just sucking it up and paying the 10 percent IRS penalty for early IRA or 401(k) withdrawal. And he found that you could very well be best off paying the penalty vs. all the other complicated strategies over a five-year time horizon, or worst case would only be slightly worse off paying penalties than jumping through the conversion or SEPP hoops. And if you’re penalty-averse and don’t mind running things past a tax accountant or three, your next best option is to go the SEPP route. We’ve filed away the surprising conclusion about the penalty route in our “in case we need another contingency” mental filing cabinet. But as for SEPPs and the Roth conversion plan, we see a few other important downsides worth noting, in addition to all that stuff above about selling out future you:

Forced to sell when you’d rather not — With a SEPP, you must withdraw equal amounts each year, even if your investments are tanking and you’d rather not sell shares. Even if selling shares will create sequence of returns risk that could haunt you long term. And if you’re reliant on a Roth conversion for your cash flow, you might be forced into the same situation, even though there are no tax rules forcing you to sell shares in a down year. Selling shares in a crash vs. sitting tight and living off your multi-year cash cushion could easily be the difference between your money lasting through your lifetime and it coming up short. (Note: this is different from doing a SEPP or Roth conversion just because you want to get those funds out of restricted vehicles and you have room in your tax cap. If you’re just doing that, you could sell at a loss and rebuy at the same low price in an unrestricted vehicle. But we’re talking about when you actually need that money for your cash flow, so can’t rebuy shares.)

Tax (and maybe health care) implications in high earning years — Also with a SEPP, you have to keep withdrawing those equal amounts each year, even if you earn a bunch of money unexpectedly, which could have big tax implications. (We’re not anti-tax, but a lot of these strategies lean heavily on tax minimization.) Worse, if any part of the Affordable Care Act survives, earning just a little over a certain threshold could force you over the ACA subsidy cliff, and cost you far more in health care costs than you gain in income. Impossible to know if any part of that will apply down the road, though.

Do This Instead: The Two-Phase Retirement Plan

Everyone has a different comfort level with this stuff, and ultimately, you should do what lets you sleep best at night, not what some strangers on the internet are doing. But with our belief that we’d all be wise to build our retirement plans around conservative projections and pessimistic forecasts, we think it makes the most sense to build out a two-phase retirement that’s minimally — or at least not wholly — reliant on early withdrawals from your tax-advantaged accounts.

The two biggest reasons why someone would not want to do a two-phased approach are:

  1. The math is more complicated. There’s no easy 25x.
  2. You have a lot in your 401(k) and IRA and don’t want to wait until you have a large taxable balance saved up as well.

Anyone willing to do the hard work of saving for early retirement is up to the challenge of doing a little more math, especially because it means building a way more souped-up spreadsheet that you can brag about to your friends (because we’re all nerds here), and you can live with the fact that we’re talking about delayed gratification.

Related post: How We Calculated Our Numbers for Each Phase of Early Retirement

If going the two-phased route feels like one-more-year syndrome — forcing yourself to work longer than you otherwise might — consider reframing it around these questions:

If your retirement plan doesn’t allow for the possibility that your spending might need to increase significantly in your later years, is it really a solid plan?

Or if you’ll only be able to increase spending when you’re older if the markets are exceptionally kind to you, is it still a solid plan? 

It’s not one-more-year syndrome if you’re just truly not ready to retire yet, if your plan is only solid if nothing bad happens when you’re older, or if your contingency plan is to get all your health care abroad. (What happens if you’re in a car accident in the U.S., or you have a sudden aneurysm, and can’t ask the ambulance to just drop you off at a hospital in Thailand? If your whole plan would be sunk then, it’s not a finished plan. Time to keep revising.)

Early withdrawals could still have a place in that finished plan, so long as they don’t put future you in a bad spot. Like if you are aiming to stay in a certain tax bracket each year, and you come in $5,000 under the max some years? That might be a great time if the markets are doing well to convert $5,000 to Roth to pad your cash cushion or otherwise bolster your investments. A few years with small conversions likely won’t make a meaningful difference to your “real” retirement fund, but relying on big conversions year after year, especially when markets are down, will always come at a cost.

Share Your Thoughts!

I know plenty of folks will disagree on this position, so let’s hear it! This discussion is always incredibly informative, and lots of people value getting to read different views to inform their own plans — so don’t hold back. What’s your retirement plan built on, and — more importantly — why? What are we not thinking about in our calculations that you’re accounting for? For folks on board with the two-phase plan, what were your deciding factors? Let’s dig into the details in the comments!

Don't miss a thing! Sign up for the eNewsletter.

Subscribe to get extra content 3 or 4 times a year, with tons of behind-the-scenes info that never appears on the blog.

No spam ever. Unsubscribe any time. Powered by ConvertKit

122 replies »

  1. Very informative. I’ve wondered whether simply paying the penalty works out better, and God knows it’s certainly far less complicated.

    I particularly agree that it’s riskier than it seems to rely upon the tax code staying the same. We’re one big recession away from huge tax changes to all of these accounts. Nobody should be fearful, but it pays to think ahead and keep things more flexible in case the rules change, because that’s possible if not likely, especially for those retiring *really* early.

    • Definitely read that Mad Fientist post linked here — he breaks it all down really well, as usual. And yeah, sweeping tax changes could theoretically come at any time. Sometimes things get grandfathered in and might not affect those already doing something, but we shouldn’t assume that. Especially, as you said, for those on a super early retirement path who have many, many years of relying on very specific tax law to make their plans work.

  2. I think of SEPP withdrawals as Plan H (or am I up to K?) in my contingency hierarchy. It could also be something that could be useful in the 55+ years, as we’d be able to stop again at 59.5.

    I can see why people don’t have a big taxable savings account. I’ve always gone with, “Max out your retirement accounts first.” For couples with 401Ks and under the Roth IRA limit, that’s $47,000 (if my math is correct). It’s a lot of money to put aside before saving even more in yet another account.

    Maybe maxing out your retirement accounts first is a trap. Then again, you have to love the future you as you say, right?

    • Yeah, I think it’s somewhere around that letter for us, too. And by the time we’re 55, we might be looking at doing SEPPs just for tax reasons (but not spending the money — the important distinction in my book). And on the maxing retirement accounts question, I’ll confess that I generally don’t even think about IRAs and Roth IRAs since we’ve been over those limits for a while, so we really just think about the $36Kish for the 401(k) limits for a couple. Anything extra for us goes into taxable because we have no other option, so we’re saving faster there than we might if we were also maxing Roths. But it’s a good point that lots of folks are maxing the tax-advantaged stuff and leaving little to go into taxable — I’d argue that if early retirement is your goal, you have to find a way to balance the two. We’re still maxing our 401(k)s because we can, even though those accounts are more than good, but our situation is definitely not representative.

      • Why not do a backdoor Roth?
        I’ve been over the limit for years as well.
        I even contribute after tax dollars above the $18k so I can convert them to Roth with I’m done.

      • Mainly because we are already well above what we need to have in our traditional retirement funds. We are still maxing our 401(k)s for the tax benefit, but want to keep everything else unrestricted in our taxable funds. ;-)

  3. I lol’d at the “Responsible Congress” line. Only thing they want to be responsible for is lining their own pockets and keeping their seat. Politics aside, I think as the Baby Boomers fade from the majority in politics we will see a drastic change to many facets of our government. After all, the biggest population group now belongs to the Millennials (muahahahaha).

    As for retirement withdrawal strategies, I lean more towards the Roth IRA ladder and rental income strategy. I am HELLA impatient and fattening up my taxable income accounts take time….. so much time. I think the term I’m going for is Barista FI where I’ll be ok living off a fun part time job and whatever other income streams I have going on.

    • I’m still going to hold out hope that a responsible Congress might one day be a thing. You know, like when you guys are in charge! ;-) I totally see the value in the Barista FI approach (love that, btw), so long as little old lady Gwen is still in solid financial shape. :-)

  4. Well, interesting because Roth conversions was on our mind this week too. We just posted about them actually.

    We are planning the 2 phase approach, same as you. However, the Roth conversions will still be a valuable tool for us to lock in taxfree retirement savings, or pay considerably less than we pay now (10% or 15%, vs 33%). Pending tax changes will certainly cause us to revisit these plans, however we aren’t planning on using the converted money during our gap years. I think that is your distinction here? We just want to slide over as much as possible under tax favorable terms. So I think we agree generally with what you are saying but there still may be a place for them for long term tax planning.

    • You’re right — my distinction is entirely around spending. No issue whatsoever if you want to do some conversions and continue holding that money for your traditional retirement years. And totally agree there can be a place for those strategies in your long-term planning, so long as you aren’t selling out future you. ;-)

  5. “Tax rules can and do change, which could spoil your early withdrawal plans.” Replace “early withdrawal” with “life” and I am seeing this now in Trump’s plan to get rid of Public Service Loan Forgiveness. I’d like to think that Congress wouldn’t pass such a plan, but after the health care bill passed the House I have no idea what Congress will or will not do.

    As to early withdrawal, I am lucky enough to have access to a 457(b) instead of a 401(k). They function almost exactly the same except that you can access your 457(b) funds as soon as you leave work, rather than after you hit “retirement age.” Maxing that out every year at the beginning of my career should give me a significant amount of flexibility towards the end of my career.

    • Perfect example! (Though also a super sad and in some cases tragic one.) Things can change at any time, often suddenly. And having that 457(b) is a great asset! I’d still make the same argument not to spend too much of it too early in your life, but having more flexibility with it is certainly a good thing.

  6. We are building in lots of contingencies and making sure we are making conservative assumptions every step of our retirement plan. With a potential 50+ retirement horizon, you can never be too safe. We will be using a 3-3.5% safe withdrawal rate assumption with an extra 30-40% spending cushion for some of the reasons you mentioned above (like higher healthcare costs).

    That said, we’ll use a few loopholes in the tax code to get access to all of our retirement account assets…tax and penalty free. And the SEPP is definitely not in the playbook. I actually just detailed the full plan in a post today if you want to check it out.

    • Lots of contingencies — speaking my language! ;-) If the tax code changes and those loopholes go away, will your plan still be solid, or will you be stuck scrambling? If it’s the former, then you’re most likely in great shape — none of us should want to see our plans crumble because we relied too heavily on loopholes that could close at any time.

      • Exactly! Changes in the tax code are always at risk. I’ll have over a decade of flexibility, so I won’t be scrambling by any means. But I am anxious to see the details in Trumps proposed tax policy to see how it could impact or change the plan.

  7. I really appreciate this post. As someone who has access to an employer 401k, Roth IRA, and taxable brokerage account, it can be difficult to determine how much to contribute to each.

    My current stance has been to fund my taxable account first, with projections to ensure that I have enough to get me from 50-60 without tapping my Roth or 401k. I’ve built in 401k conversions to my Roth, but only as a way to max out the tax bracket and as part of a tax strategy, not because I’ll need the money.

    The future state of taxes has me really worried. I just don’t see a scenario where taxes don’t go up. Having too much in a tax deferred account is a recipe for disaster. I’d rather pay taxes now, while I know how much it is and can stay in the workforce if needed. Getting slammed with a huge tax bill in retirement would be a nightmare, which is a big concern once RMDs begin.

    I would love it if you wrote another post, elaborating more on this topic. I’m still learning and trying to figure things out so I could use all the help I can get :)

    • I’d be happy to write more on it! What would be most helpful? More on navigating the two-phase planning, or the tax stuff? And yeah, zero issue with doing conversions as a tax strategy, so long as you aren’t relying on that money for cashflow in your pre-60 years, at the expense of your security in later years. ;-)

      • If you could just come up with an entire series on this topic, that would be great :)

        My biggest problem is finding comfort in the projections I’ve created. Right now I’m using the investment spreadsheet on the Charlton’s website (http://wherewebe.com/early_retirement.html) to get an idea of cash flow by year. I’ve added rows to take me to 100 years old and built in lots of assumptions (rate of return, RMDs, Roth conversions, SS, etc).

        I think I have a good grasp on things but it’s hard to know if I’ve overlooked anything. I’ve read a lot about the theories and ideas of funding early retirement but putting it into practice for each person’s specific situation is where it gets sticky.

        If you want to see my file, let me know and I’ll email it to you. It has real numbers so I’m keeping it on the DL, rather than post it to my site ;)

      • Okay, full series coming up! Hahaha. (I don’t think I have THAT much to say on the subject!) It sounds like you have a pretty exhaustive list of factors built in, so it seems like you’re accounting for everything you CAN account for. (Not all the HUGE stuff you can’t, of course… I know that’s what keeps lots of us up and night!) If you want to email it to me, go for it, and I can give you my gut check view on it.

      • Have you written about the pros and cons of investing in traditional IRAs vs. taxable accounts? As someone earlier in my retirement journey, I continue to max out my 401k and Traditional IRA – had not considered allocating some of that to taxable accounts instead.

      • I haven’t, mainly because we’ve been above the IRA threshold for a while, so it’s not part of our calculations. (IRAs will be back in the mix in retirement, though!) I don’t know that I would advise against maxing those accounts, but certainly any additional amounts you’re able to save should go into those taxable accounts to fund your pre-59 1/2 retirement, or at least the first 5 years if you use the Roth ladder.

      • Wow. I had no idea that we haven’t been getting deductions on our IRA contributions. Thank you for opening my eyes to this! (Feeling like a money noob)

      • Did you verify that you didn’t claim them on your taxes? (You could file amended returns and get the difference back!) But glad this is helpful!

      • We’re very hands-off with our tax prep and had no idea there were income limits on IRA deductions. So it’s all fine, but clearly a sign we need to pay more attention to how tax planning affects our investing strategy.

      • It’s true it’s good to know the rules. :-) Though we don’t think about any of that beyond our 401(k)s, mainly because we like the things that taxes pay for (schools! libraries! care for poor people!) and don’t wish to skip out on paying our fair share.

  8. Oh yeah, I’m not about early withdrawals, either. Maybe there are ways around the penalties, but I just don’t think it’s worth it. I agree that it’s much better to save for your golden years in proper retirement accounts, and then live on taxable accounts (like savings or investments) in the interim.

    • Yeah, the penalties are not a huge deal — they’re easy enough to work around. But the loss of security in later years is a VERY big deal!

  9. The reality of SWR depends so much on the age you start withdrawing, risk tolerance in terms of asset allocation and desire to leave a legacy fund or not.

    The classical Trinity study was really geared for more traditional retirees. Dr. ERN described from his in depth analyses on the SWR that if you are on a 60 year timeline in retirement (e.g. reaching FIRE at age 35), have a fairly conservative 75:25 equities:bond mix, are OK with capital depletion (i.e. no legacy fund), you will need to take a 0.5% haircut on your withdrawal rate relative to the retiree who reaches FIRE at age 65 with similar portfolio composition and capital depletion strategy. That’s a significant difference over time and especially in a bear market scenario where usual belt tightening will become really, really tight. Dr. ERN’s numbers start to look even scarier if you go to a 50:50 equities:bonds mix. In that scenario, a 0.75% haircut is necessary to ensure not running out of money in your later years.

    Our taxable portfolio will make up 53% of our total portfolio at FIRE next year, the remainder tax advantaged. My pension is not factored in for this analysis as I will be turning my cash balance pension plan into a life-time annuity to form our “income floor”. It will be 8.5 years until the first 401K (Mr. PIE’s) can be accessed.

    Bottom line, plan for a low withdrawal rate anyway and make sure it is much lower if you are an early retiree. By doing that, you will be just fine looking at your whole portfolio and running the SWR numbers.

    • Thanks for typing out all that additional detail about the Trinity Study — I always think it’s helpful for people to know the details behind the plan that everyone seems to use but no one seems to understand. ;-) I think adjusting your SWR is one thing — and is super important! But assuming level spending over every year of early retirement and traditional retirement potentially another big area for trouble, and that’s something SWR alone doesn’t address. You have a pension that will provide extra security, and other people’s security sources will certainly vary, but it’s important to have some form of backstop to prevent catastrophe in case medical bills, for example, go through the roof and risk bankrupting you.

  10. I totally agree with having unforeseeable costs in retirement and while I initially poo-pooed the IRA conversion, I’ve grown to see some of it’s advantages… IF, you really don’t plan on spending them, just converting them. RMDs can create a huge tax liability down the road. All that being said, a lot can change over the next 50-60 years. I’m a little different in that I still look at it all as one big pool of money; however, I’ll be mechanically accessing my cash flow needs for the next 5 years via conservatively projected taxable investment dividends and cash flow from alternative investments. I honestly haven’t projected out too far in the future since I do think a lot of things can change, some of which may benefit me, other changes will work against me, but I think if the big picture stuff is in place (the one big pool of money, for example), we can still choose this alternative lifestyle.

    Quick comment on the 4% rule – I totally agree there – and sometimes I think that folks forget one of the basic underlying assumptions that the 4% rule was testing whether your investments would last for 30 years. Those extra early retirees could have 60-70 year retirements.

    So where are the fighting words? :) You are so polite with your civil discourse and interesting lines of questioning. When I “discovered” early retirement and the 4% rule, I went through similar thought processes most obsessed with how I would actually access my necessary cash flow while preserving the capital needed to generate that cash flow, thinking how in the heck do I stay retired if I’m selling my stocks all the time to make ends meet? I think it’s cool that we all have different paths and it will be fun and maybe not too stressful to test out our various approaches when the stock market adjusts! I’m not overly confident, for the record, just confident in my ability to be flexible and not afraid to execute one of my several back up options if my investment plan fails.

    • Oh yeah, no issue with converting but not spending, only with the spending. ;-) And yeah, amen on the Trinity Study. Everyone should actually read it before basing their whole retirement plan on it! Mr. PIE just left a great comment that outlines several of the big assumptions in the study, which I have yet to see reflected in anyone’s planning (not that no one is doing that, just that I haven’t seen it). ;-) I think my fighting words are just in saying DON’T RELY ON THE ROTH CONVERSION LADDER! Which is a major tenet for many in FIRE land. And I agree it will be interesting to try out different things — the important point there is that you DO have back-up options (several of them!), and you’re also not reliant on a specific tax rule that could change. Those things are key!

  11. That’s funny… I guess that I must be reading different blogs which aren’t talking too much about RCLs. :) I’ve only latched on to 4 or 5 that I read on a regular basis and only 3 that I read carefully. No offense bloggers, I just follow the low information diet rule.

  12. Very true about tax laws changing. We are dealing with this in Canada. The number one saving vehicle in Canada is the RRSP (Registered Retirement Savings Plan) where you can take part of your income and stash it into some government approved investment where it can accumulate interest tax free. The catch is when you do take it out, you have to declare it as income and pay the taxes. The presumption is you will be in a lower income bracket as a retiree and pay less taxes than you would at the time you stashed it away. However the value of the RRSP has been squeezed on two fronts. If you don’t withdraw it all and spend by age 71, it gets forcibly converted to another higher tax vehicle. Plus we have three forms of government pension benefits in Canada and the RRSP money you withdraw lowers the amount you get in guaranteed income supplement and old age security. These changes came about after I started my RRSP. Plus my husband and I can now split his pension income to allow us to move into a lower tax bracket but if I take out RRSP money then we can’t split as much. So the sacrifices I made as young woman in putting money aside will result in my being penalized eventually. About ten years ago the government came out with Tax Free Savings accounts where you don’t get any up front tax deduction as you do in an RRSP. However if you do take the money out of your tax free savings account, it does not count against your income for pension splitting, guaranteed income supplement and old age security. However our government is now rumbling about how that is unfair because one can be “stinking rich” from tax free savings accounts but still be eligible for additional government funds in old age and pension splitting. There are rumblings about changing that. I have an appointment with my financial advisor next month because I have to decide what to do with my meager RRSPs. If I just let them sit, I will essentially lose them. Welcome to soft socialism Canadian style.

    • It seems like there are drawbacks with every country’s system, and questions about whom it really benefits. In the U.S. we still don’t phase out Social Security benefits, for example, for those who clearly don’t need them. The most we do is tax them more for some people. But I think your larger point is true everywhere: stay flexible and stay informed about changes, because they can have a big impact on various aspects of your plan!

      • The taxing back of money from benefits they paid into (by force) from people who “don’t need it” is always problematic. Generally speaking those who “don’t need it” are always just above the tax bracket you yourself on it, or to quote Margaret Thatcher, sooner or later you run out of rich people to tax. In Canada, we richly reward the indigent and lazy and those lacking foresight to work and save in order to protect those who are truly needy. We punish people who work hard, save and plan ahead, often severely. The idea seems to be to make everyone equally poor and equally dependant on government handouts.

      • I think it’s a matter of perspective. The entire basis for a progressive system of taxation is that people who can afford it pay a larger share to fund services for those who can’t afford it. We apply this principle to essentially every other form of taxation, at least in the U.S., but decided that Social Security should not be run this way, and so folks see it very differently, as an entitlement for all, when it’s truly a semantic difference.

      • In Canada, all our equivalents to social security are taxable. Even if you paid into for years and years and years at the highest possible level, you can have it all taxed back in what we call “clawbacks”. Clawbacks encourage sloth and dependancy. Still, given your debt levels in the USA I don’t see how you can sustain your system as it is. I suspect clawbacks are coming.

      • Ah, that’s interesting. I didn’t know that about the Canadian system. Thanks for the info!

      • Oh and the other thing is I see a difference between entitlements and severely penalizing people who chose to save and be frugal while rewarding those who were profligate. I don’t mind the clawbacks. We have a good pension, we don’t need so much Canada Pension Plan. What I hate is how I did without and saved money as a young woman and I shouldn’t have. I should have spent like a drunk sailor. I am no farther ahead for having done without as a young woman. The new bike for the kids that I did without to put money away should have spent because what I did without is now being used to pay for those who did buy the new bikes.

      • That stuff is so hard to know! Maybe if you’d spent like a drunken sailor, you would have developed totally different habits that would have put you deeply in debt or worse? ;-)

      • A couple of clarifications about some of the things you say below. CPP, which is what you pay into when you work, never gets clawed back. Yes, it is taxable as income when you receive it, but you get a tax credit each year for the amount you pay into it, so it is essentially tax-deferred (i.e. you don’t pay tax when you put it in, but you do when you get it back). The math doesn’t work out perfectly, depending on your tax rate in the years in which you pay into it vs. when you collect, but it is not really that terrible either.

        OAS, which CAN be clawed back, is not something that we contribute to directly from our wages, it comes out of the general tax pool and is available to everyone who meets the residency requirements, whereas CPP is only available if you worked in Canada and paid into it. OAS gets clawed back at a rate of 15 cents on every dollar above an approximately $73,000 threshold (in 2017). That threshold is for an individual, so a household could earn $146,000 before having any of their OAS clawed back. Yes, you will have to pay tax on the income, but it’s literally just income that the government is giving you that you had to do nothing to earn except live in Canada for X number of years.

      • It’s interesting to note that the US is not the only place with complicated rules. Thank you for sharing this!

      • I was referring to how money put in an RRSP gives you a tax free saving account until you take the money back out. When you do take the money back out, you have top pay taxes on the money and on anything earned while it was sheltered. If you have an income of anything over $73,000 not only are you paying income tax on the RRSP withdrawals, but the government clawsback things like OAS. So while you get an pension credit on the OAS it is still clawed back and you do not get any kind of credit on the taxes you pay on your RRSP withdrawals. So if you are right on the edge of the $73000 per year income then it does you very little good to have saved and done without to get the RRSP money put aside. As for the lack of clawbacks on CPP, I know if I had been able to keep that money and invest it myself I would have made a whole lot more than what the government is paying me back. CPP is not kept in separate government account where it gains interest wisely invested. Trudeau has been tapping into the CPP money for things like housing projects in Africa. CPP might as well be in general revenues. It is a tax up front on everyone which then you get some back on as you age, if you are lucky enough to live long enough to apply for it. If you grow to be a ripe old age your forced investment will pay off handsomely, but if you die young the money is just gone. You don’t get to pass it on to your children and grandchildren. So there is no incentive here to save and be wise with money and every incentive to spend like crazy and count on government to support you in your old age.

  13. You and I are in the fortunate position of having plenty of income left over to fund a generous taxable account. Like you guys, we plan to live primarily from the taxable account as long as we can, allowing our investments in retirement plans to grow tax-free as long as possible.

    I do know a number of people who have an overwhelming majority of their retirement assets in tax-deferred accounts. For them, early retirement wouldn’t be possible without Roth conversion ladders, SEPP, etc…

    I expect I’ll be making Roth conversions once our 457(b) is drained, not for spending money, but to decrease or eliminate RMDs. Ideally, I’d like to have little or nothing in tax deferred money by the time I reach 70.5. Of course, that’s 29 years away, so a whole lot can change in the meantime.

    I wholeheartedly agree that you should look at your different accounts separately and I sometimes wonder why the Roth conversion ladder is such a popular tool — Roth is the last money you should touch — but then I realize it’s either that or work until 55+ for most people.


    • You’re right that we’re in an incredibly fortunate position. And the only issue I have with Roth conversions and SEPPs is if people SPEND that money right away, not the conversions for tax purposes themselves. Even for those who have limited ability to fund taxable accounts, I think it’s worthwhile to develop a plan to that includes a stepped-up level of spending (or at least the ability to spend more) after hitting a certain age, say 60 or 65. Just to account for the many uncertainties out there. If the only way to fund an early retirement is by converting or SEPPing (I’m just making up verbs here!), it’s far safer to live on a smaller proportional share in the early retirement years and leave a larger margin for safety in the later years.

  14. I’ve always looked as taxable and tax-deferred as two separate “buckets” as well. Mentally…I envision the tax deferred locked in a bank vault for future use. With a little luck…this money could double every 10 years or so ( depending on market forces of course ). It may be mostly mental accounting…but it works for me as well.

    • That’s a great way of thinking about it! I fully endorse that. :-) After we retire, we’ll start thinking more about managing that tax-deferred buck to avoid the required minimum distribution problem (though, honestly, we’re less concerned about this than most folks — if we end up with more than we need, we’ll happily donate a big chunk to charity and fix that whole “problem”). But that’s a good reason to think about early distributions, not because you’re relying on that money for your ER cash flow! :-)

  15. This is actually the exact strategy I am going to use for my own early retirement. I didn’t know this was a controversial strategy at all. It makes a lot of sense to keep your money in your retirement accounts for as long as possible. This is especially true for Roth IRA accounts. You can enjoy potentially decades of compounded investment gains tax free. I don’t plan to touch any of my retirement accounts until I’m much older. Hopefully in my 60’s.

    • That’s awesome! It’s not really controversial. ;-) There’s just a lot of momentum behind the idea of tapping IRA and 401(k) funds early, for good reason — because more people are able to build up sizable balances quickly in those accounts, compared to non-tax-advantaged taxable investments. But tapping them too much or too early puts our older, less-able-to-earn selves at tremendous risk. So I love that you’re not even thinking about doing that! ;-)

  16. You don’t seem to be counting much on SS for your retirement income, and maybe that is due to your early retirement as it’s calculated on 35 years of earnings, but it will count significantly in mine. I am concerned about taxes on RMDs at 70+, especially as they will likely be excess income. Tax efficient Roth conversions before then may be in order. What impact do you expect from RMDs?

    • It’s true — we don’t expect much from Social Security. In part because, as you said, we’ll have a lot of low earning years factored in. And in part because we expect there to be reforms to the system that (honestly, we hope) reduce benefits for high net worth individuals in an effort to increase its solvency. And we fully expect to be high net worth individuals at the point when we’d qualify! ;-) I think doing Roth conversions for tax purposes is one thing, and is fine if you don’t spend all that money. My concern is primarily with spending that money in your early retirement years and leaving yourself in bad shape in your later years. We’ll write more about RMDs next year after we’re retired, but in our case (which we understand is almost certainly not broadly applicable), if we’re in an RMD situation, we’ll donate the overage to the charitable causes of our choice, solving both the high tax “problem” and our desire to give back to the extent possible. :-D

  17. Hmm…

    If you’re really never going back to work I think you SHOULD think of this as one big pot of money. Why? Taxes. You’re better off placing some holdings in tax advantaged accounts (REITs, Bonds, TIPS, etc) and keeping the more tax efficient holdings in your taxable account.

    I hear you on keeping the retirement assets separate for your retirement years. But…you’re getting ready to retire! The more money you can keep from Uncle Sam the more you’ll have to keep for Your Next Life.

    • I’ve seen plenty of advice that would disagree with you on those particular investment vehicles, simply because the returns don’t match what you need to earn to hit projected goals, but I understand your rationale. That said, we are not interested in skipping out on taxes. We believe in schools and roads and many important things taxes fund. Our tax rate will already decline dramatically when we quit, and we are okay paying our fair share in traditional retirement if our investments happen to do well. Or we can always donate the overages to charity if we want to decrease our taxes while also doing good. Regardless of whether you agree with me (I know many folks here don’t and that’s fine!), I’d at least recommend putting your own future security ahead of whatever distaste you may feel for paying Uncle Sam. ;-)

  18. Definitely agree with you ONL, I’ve been trying to boost my taxable accounts and generate a few passive income streams so I wouldn’t have to withdraw from my tax-advantaged accounts until I have to retire.

    If I have enough to get by and don’t need it, then I won’t use it and don’t have to go through the processes of SEPP or other loopholes to withdraw from them.

    I may also just work a few extra years to boost up the taxable account because I enjoy my job. If I didn’t, then I’d leave once I’m financially independent.

    • That’s all super smart of you! I didn’t even address other passive streams here like rental properties, but those are certainly preferable in my mind to cleaning out those tax-advantaged funds early! Another alternative to working longer could be to work in a much more limited capacity for a few years, even if you can’t save much. If you can just leave your investments untouched for a few years, while shortening the amount of time they provide for you, you’ll increase your chances of success while still decreasing your overall work volume. Just in case you hadn’t already considered that. ;-)

  19. Thanks for the mention. However, while I agree with the sentiments of this article, our approach will look quite different on a couple of levels.

    We plan to continue to make some income indefinitely and not draw down consistently on any of our investments for quite a while.

    Instead, we’ll plan on pulling the cord with well under 25X our current expenses, and focus on living a life that involves some continued work, but much more balance.

    I get your sentiment that there is a lot of uncertainty in traditional retirement. Our approach is therefore not really worry about planning a traditional retirement, and instead focus on building a lifestyle we will be happy with much sooner, while actually giving us more options and thus more security in the long run.

    • I didn’t mean to suggest that you guys have the same financial plan as us! Just the dirtbaggier early retirement component. ;-) I think you guys talking about earning some side income in ER helped us think that through and realize that we want that as well, at least as a hedge against sequence risk, and perhaps also to maintain a bridge back to real employment, should we need it. (Plus, obviously I am a busy body and need to be working on things. So might as well get paid.) ;-)

      • Glad we had some influence on you! I think that we have a pretty similar approach to risk as you, and to me traditional retirement either involves too much risk OR building a large buffer which would take more time than we are willing to give it. And also agree, I just can’t imagine life where I am completely unproductive for 40+ years, so why not just incorporate some productive work into the plans?

      • We lucked into the large buffer, just because we discovered FIRE a little later, and one of us (ahem, not me) was a good boy scout about maxing out his 401(k) early and often. But all the same, I wish to keep that buffer there, not blow in in the intervening years! So might as well work some paid employment (or more likely self-employment) into the equation.

  20. Our total/target number =1 big pool of money, but how we use each one is a totally different strategy.

    Our latest forecasts showed that if we can keep building costs down, and work until 2020 – since we both like our jobs and current schedules, this shouldn’t be a big deal – then we can make it to 60 without touching any 401k money. That’s assuming no extra income from anywhere, low market growth, high inflation, yada, yada, yada. For us, that seems reasonable, and if we get to 57-58 and it looks like we may have to tap the 401k’s we’re not super put out by it.

    Mainly because our projections are assuming zero side income forever, which we don’t see as a likely scenario. But how do you model, random, yet to be created side income? You got me. I just throw small amounts of side income randomly starting and stopping into cfiresim and see how it plays out, lol. I do the same with various outflows of cash too representing yet to be known amounts healthcare might cost… Ugh…

    Therefore, we see this current projection of ours as really conservative, because if one or both of us brought in even $10k/year, that’s like 20% of our yearly spend. That makes a ‘Yuge (lol) difference in projected outcomes. Since we’re both looking at having some sort of side gig – even knowing it won’t pay close to what we currently make we’re good with that.

    I also cringe a little when seeing people using the RCL or SEPP as planning tools to make their ER work. That just seems a bit too risky for me. We’re glad they might be available if needed, but I don’t want my plan to depend on having to use them.

    • That DOES sound reasonable! I didn’t say this in the post, but part of why “one big pool” doesn’t work is that you might be at an overall total that works, but the money isn’t structured to be especially helpful. Like if we count home equity, we’re waaaaay above our original total number, but in the most unhelpful way possible. The breakouts matter a lot, too. As for modeling future hypothetical side income, you got me! But let me know if you figure it out. ;-) And I’m glad to hear you’re not reliant on a RCL or SEPP plan! Though I do think Physician on FIRE’s earlier point about having a large taxable savings bucket being a privilege for high earners is apropos to you guys, too. Not a bad thing, just worth noting that it might not be possible for everyone. ;-)

      • I agree with his point definitely. Had we been in different positions in life, I’d be maxing out our 401k’s and rolling over as much as I can because that’s the way I’d have to go about it.

        Fortunately, we are in a pretty nice position, so we don’t have to rely on that route which makes it a little less scary for me anyway. Although, I’m sure if our plan did involve rolling 401k’s into IRA’s then we’d have all of the same conservative estimates and contingencies and what not factored in so I’d be comfortable with that as well.

        It’s all in dealing with what hand life gives you and what you make of it. There are so many ways to get to FIRE that it’s kind of mindboggling when you start thinking about all of them. No one way is any more right or wrong than another, just a different path to the same goal. :)

  21. 4am comments? Do FI bloggers/commenters not sleep?!
    I would like to see Brandon update his article for nondeductible Roth contributions. I suspect the conversion ladder is more attractive for high earners who exceed the MAGI limit for deductible IRA contributions.

    • I think the desire to leave behind sleep deprivation is at least 98% of my motivation to retire early. ;-) hahaha. And I bet Brandon would be open to that suggestion, or running the comparison with different tax bracket scenarios — you should suggest it to him!

  22. I would never, never, never convert my regular IRA to a Roth! A future tax law change could cause you to pay taxes on the Roth. Oh, they would never do that. They have ways to not break that promise. How about Roth withdrawals above a certain income level would be taxed. Then you have voluntarily double taxed yourself.

    As for withdrawal strategies, I favor selling taxable investments first. That really lowers your tax rate, as much of the value of the sale is not taxed as it is ‘return of your principle’ and the remaining capital gains have a favorable tax rate. Then let the tax deferred IRA grow without a tax bite until withdrawals.

    However, geeks (like me) could modify that somewhat to withdraw from taxable and tax deferred accounts based on keeping the marginal tax rate down to a reasonable amount, saving some of the taxable funds for later when reduction of total tax is important. I’m not there yet as I just retired (early).

    • I definitely see the appeal of the Roth conversion generally — being able to control WHEN you take the tax hit, and in which bracket you fall when you do that. (And my two cents is that there wouldn’t be a double, future tax hit. They get one bite at the apple.) It’s the spending down of the funds for later that concerns me, when we all have a million variables hanging over our heads down the line!


  23. Nice post.

    I like the reference to people like me that might be too optimistic about future health care given our current situation. Evan when it sparkles doubt on our pre-mature plan (you deserve a star for this – I do believe in having a plan that is under siege to emerge as a better plan)

    And in fact, having no conversion option from my pension saving (except a 33pct withdrawal penalty – no thank you) I should in fact build a 2 phaze plan. Not yet ready for that… lazy me. i will let the idea run around in my head for a while…

    • Thanks! And wow, that’s a nice set of compliments, coming from you — I know you have spent a lot of time thinking about these questions! :-)

  24. I have long doubted the validity of the 4% swr. In this market, a lot of models are predicting low future return and I personally believe in it.

    On the spending side of things I completely agree with you. Indeed, I personally am worried that a lot of people are retiring immaturely only to end up in shortfalls, minimum wages, and tears. A lot of the contingency plans don’t work because they fail to account for the fact that people only need jobs to supplement income when they are not available.

    What is interesting is the income/investing side. Knowing that the expected return might be low, what can one do differently. I personally occasionally see one off opportunities but to find these in a scale that makes enough sense to the bottom line is itself a full time job and indeed a highly skillful one that the truly successful people can make a ton of money. So I don’t know what can be done and this is what holds me back.

    • :::clap clap clap clap::: Yes, exactly. No one needs a job to make up for portfolio shortfalls when the economy is booming. (The 2008 recession drilled that lesson into us big time: https://ournextlife.com/2016/06/29/back-to-work/) On the investing side, I wouldn’t pretend to begin to give advice there. I know we find the most comfort with indexing, because we aren’t trying to beat anything except inflation. And we keep our retirement budget well padded so that we can cut it back by 40% if need be. Plus we have a gajillion contingencies. It’s not a perfect formula, but it lets us sleep at night! ;-)

      • Glad you have written about the states of the economy. I have seen too many articles pushing for higher and higher swr.

      • We think that’s so important to talk about! Worst thing that happens if you’re too conservative is you end up with too much money. But if you’re too aggressive? Catastrophe.

  25. I think you’re mixing two different ideas here that individually have may have their merits, but your overall conclusion is really muddy. “Quotes” are heavily paraphrased.

    1st idea: “You should be conservative with your retirement projections.” Great!
    – “4% rule might be aggressive.” Sure.
    – “Older people might want to spend more than they are able too.” Possible.
    – “Unknown healthcare costs and implications.” Absolutely!

    2nd idea: “Early withdrawal strategies add extra risk to a retirement plan.” Sketchy at best…
    – “Roth ladders and 72t withdrawal rules may change.” Certainly possible, but there is always the 10% penalty method you mentioned. Easy solution is to be 10% more conservative with retirement accounts, but it shouldn’t stop you from taking them into account.
    – “Money you convert/withdraw early may serve you better in the future.” This one in particular doesn’t make any sense. You addressed it a little in comment responses by saying you meant converting/withdrawing to equal spending, but any early retirement plan should have the spending levels fairly locked down (with some flexibility of course). If you’re not spending money out of the retirement accounts, then you’re spending the money out of taxable accounts instead and impacting your future self in the same way. I think you’re once again encouraging people to be more conservative, but mixing it with withdrawal strategies doesn’t line up.
    – “Forced to sell when you’d rather not”. Wrong for the reasons you mentioned in comment responses. Moving money out of retirement accounts doesn’t have to mean selling. Even if something like a SEPP withdrawal requires moving into cash, it’s simple to go and buy the same funds in a taxable account with that money the same day and end up in the exact same investment situation as if you left it alone. There is no additional sequence of return risk with the early withdrawal strategies.
    – “Tax implications in high earning years”. If you “earn a bunch of money unexpectedly”, then paying some taxes on the SEPP withdrawals should be the least of your problems. The unexpected income already puts you ahead of your projections at that point, it can be used to fund spending! Paying some taxes in this fortunate situation of earning a bunch of unexpected money is a silly thing to worry about.
    – “The Mad Fientist said Roth ladders and SEPP may be worse.” The article you linked says the exact opposite! The conclusion was that it’s far better to utilize pre-tax Traditional accounts and pay the penalty than invest in Roth or taxable instead, but using either the Roth Ladder OR SEPP payments came out ahead of paying the penalty. This is most obvious by looking at the final table with the age of account depletion, but the whole article might be worth a re-read ;)

    In general, I agree with the sentiment that people should be conservative with their retirement projections because of several unknowns. Unfortunately, this has very little if anything to do with utilizing Roth ladders, SEPP payments, or anything else to access pre-tax money before age 59! Your personal situation appears to have the right balance of taxable vs. retirement accounts in which you can take a comfortable early retirement by only using the taxable before age 60, but I’d avoid trying to fit other people’s situation into that same box. Your 2-phase retirement plan can be executed just as well by starting out with a lower withdrawal percentage (like 2%) up until an older age where it’s then possible to expand spending if everything has gone according to plan and you desire to spend more.

    My point is that doesn’t matter if that early conservative withdrawal strategy comes from taxable or retirement accounts using various early withdrawal methods, the overall impact to success rate and being able to increase spending later on remains the same.

    Keeping the retirement accounts out of your early retirement planning may work great for you guys, but I think telling others they should do the same will lead them to the wrong conclusions. For example, someone reading this post that already has a conservative plan, but has most of their money (90%+) in retirement accounts should feel just as comfortable retiring as someone with the same amount of money and same spending plan who happens to have over half their investments in taxable accounts. The big number across all accounts and spending plan (specifically as a % of that big number) are what matters, withdrawal strategies are tangential at best.

    Having typed this all out, I probably should have just written a post on my blog lol. Maybe I’ll do that as well. Cheers :)

    • I agree with Noah here. Your preferred 2 bucket approach is a mental accounting trick, but doesn’t actually benefit your future self. You benefit by having a large (enough) portfolio, not because you plan to spend some now and some later. Spending from your tax advantaged accounts doesn’t cheat future you, unless you didn’t save enough money to begin with. But that’s a saving problem, not an account allocation problem.

      • On a purely technical level, you’re right. But there’s a reason why most people have success with paying themselves first vs. having the discipline to hold back on spending and save a set amount at the end of each month. How we behave with our money and how we structure it matters hugely in our success. The post was already too long to go into all that, but that’s a big part of this framework, too — by making the “phase 2” money off-limits, we’re less likely to touch it, which puts us in a better position come our true retirement years, when we will be less able to side hustle or do whatever else we might be inclined to do if we realize we undersaved. So, sure. Of course it’s an accounting trick. But so is the debt snowball, and paying yourself first, and tax deferral for that matter. ;-)

    • Since we sidebarred about this, I am not going to try to go through this point by point. I think we fundamentally disagree on some key stuff, which is fair, but I do think you missed that this is all about RELYING on conversions for CASH FLOW, as stated in the post. The SEPP conversion into something taxable that you mentioned could make sense as an overall allocation strategy, but if you’re doing that annual cashout to live on, then you’re not just rolling it into something else. So that’s where the risk comes in — potentially having to sell, for cash flow, when the markets are down. If you’re into doing Roth conversions and SEPPs for fun, more power to you. This isn’t fundamentally about allocation. This is about how you fund your retirement, as well as how much you reserve for your later years.

      Some people love optimizing every bit of their finances, and those are folks who will want to run simulations and see how they can get their taxes as low as possible, their returns as high as possible, etc. That’s not everyone, though. Plenty of people (evidenced by lots of comments here) don’t want to have to mess with that stuff, and so this is another way to think about options. Plenty of others (and I’d include us in that category) *need* some of those mental tricks to be financially successful. We save a ton because we never see any of that money in our checking account. We hide it from ourselves. So call this an accounting trick if you wish, but there’s something important and powerful here about just hiding money from yourself in the early retirement years so you don’t spend it ahead of when you really need it.

      • I didn’t miss that part, my point is that it doesn’t matter what you are relying on for cash flow, the risk remains the same.

        SEPP conversions, Roth conversions, or even pulling from a taxable account all have the risk of needing cash flow and selling funds in a down market. There is no additional risk when that money comes from retirement accounts versus a brokerage account, at some point you will have to sell in a down market.

        Solid withdrawal plans account for this by having a safe withdrawal rate well below the average returns, but the source of the funds does not matter.

      • FYI, I’ve updated the post to make some of this stuff clearer, and to correct some things that needed factual correcting.

        Re: the risk, I am primarily focused on the SEPP, as I state in the post, because with Roth conversions or taxable accounts, if the markets are down, you could simply opt not to take any distributions at that point and live off your cash cushion. With SEPP, you have no choice but to take the withdrawal each year unless you want to pay a massive penalty and years of back taxes, and even if you just sell and then immediately buy back the same shares, your cost basis is now lower which will have future capital gains implications, perhaps even short term capital gains, depending on when you need that money. Being forced to sell at a bad time IS a potential risk of the SEPP, and limits your flexibility in a way the other approaches do not. You’re right that if you NEED cash flow and don’t have a cash cushion to fall back on, then all the sources of funds are equally risky. But I’m assuming that anyone proceeding with a well thought-out FI plan has a cash cushion for exactly this reason.

    • Thank you for writing all that out! You’re completely right. I was fretting about having to correct all those errors and presumptions in the post and then I saw that you’d already done it. Well said.

  26. I have suddenly figured out an analogy to describe the retirement/taxable account situation. Isn’t it like APPL, which has cash overseas and cash inside the nation. For any given spending, which pile of cash should it use? Should it repatriate the cash or not? The answer is whichever is more economical.

  27. In one of the comments above you wrote that you don’t expect to collect much from social security due to many year of low earnings. Have you seen this: http://rootofgood.com/early-retirement-social-security/ ?

    Also, I think I agree with a few of the recent comments. Not sure why it matters where your retirement savings are (tax deferred vs taxable) as long as there’s enough to cover expenses. Who cares if one is spending from Roth conversions if one’s tax advantaged accounts are more than substantial? Interested to read your responses.

    Just signed up for your newsletter. Last thing I need is one more thing coming to my email but I very much enjoy and appreciate your writing and perspective. Imagine, recognizing that taxes actually pay for things our society needs and being willing to pay one’s fair share. Years ago I made some snide comment about paying too many damn taxes and a close friend responded that she doesn’t mind paying taxes as they build roads and pay for needed services, etc. What a concept. I’ll never forget having that gripe blown out of the water. Although I’d bet you’re not too keen on paying the taxes that support some of the wars we’ve been involved in….

    Thanks for your thoughtful blog.

    • I have seen Justin’s post on Social Security. I also think a lot about SS is likely to change in the intervening years, and there’s a strong likelihood that high net worth households will be excluded or limited in what they collect. (And I kinda think it should be that way — that’s how we view every other form of taxes except Social Security!) And I think, in a technical sense, as long as you save amply for traditional retirement, it’s true that it doesn’t matter where that money lives, with the possible caveat of the SEPP and being forced to sell in down markets, etc. However, there is a strong cognitive bias to leave things alone that are untouchable for now, vs. things that are fair game for cash flow now. It’s the same reason most experts advise you keep your emergency fund at a bank other than your regular one. And that’s a big part of my recommendation here, though I didn’t directly state it (too many words already!). Of course, everyone’s different, and some people are disciplined enough to leave the future money alone even if it’s in the same set of accounts!

      I’m also delighted you joined the e-newsletter list! It means a lot that you’re welcome to get another thing in your inbox. :-) And I’m extra glad you appreciate my perspective on taxes. As you well know, it’s so easy to complain about them without actually stopping to think what would happen if everyone practiced tax minimization. You’re right that we don’t support everything they fund (can’t imagine anyone does!), but we’d sure rather be in a society where we benefit collectively from taxes than the alternative! ;-)

      • If social security gets means tested I believe it will be based on income rather than wealth. After all, someone with a million dollar net worth with a $400k paid for house and $600k portfolio sounds “wealthy” but at a 4% withdrawal rate they are only going to have $24k a year as a couple to live off of.

        Also it will be an accounting nightmare for the government to value, track, and audit every American’s net worth for this purpose. Limiting by income would be much easier. Given that fact, you could have an awful lot of assets in Roth’s, real estate, etc, and still show a relatively low enough income to still qualify for a means tested social security. Most early retirees likely don’t need to fear this in my opinion.

      • I think you are most certainly right, and true wealth means testing is both too invasive to be acceptable, and too cumbersome for the government to undertake. That said, we were lucky to get an early start on our retirement savings, so shouldn’t need any supplement in true retirement. If we happen to get it, great. But we will be fine without it. And given how risk-averse I am, I am much more comfortable with a plan that doesn’t rely on something that could be reconfigured or scaled back at any point in the future! :-)

    • I’ve also read Justin’s post on early retirement and social security. That perspective seems about right, though it matters little to me with my 35+ years of contributing. It is wrong that DC politicians continue to seek ways to avoid paying these entitlement obligations, no matter the size of the benefit. Means testing? They already are doing it with up to 85% of SS being taxable, which goes to the IRS, not back into Social Security to keep it solvent.

      I wonder how fewer wars we’d engage in if we actually had to raise taxes to pay for them. A MOAB and a bunch of cruise missiles, well those will stimulate the economy! They were probably nearing the end of their shelf lives and needed to be used somewhere. Oh, the next few generations can pay for them.

      I don’t mind paying fair taxes for needed things in our society, but I hate seeing them wasted.

      • Yeah, it does seem nuts that taxed SS benefits don’t feed back into Social Security! And totally with you on that war spending. >:-(

  28. I agree with your thought process. I don’t want to rely on my tax advantage accounts for our early retirement. Instead, they are more of my retirement emergency fund. If something goes wrong (ex. Health wise), I still have funds to rely on.

    One question though. What is your issue with Roth conversions? Do you not like them because of the assumption they will be withdrawn and used before 59 1/2? Converting 401k to Roth when you have the right opportunity is a solid tax strategy for your future years, even if you have no plans to withdraw them before phase 2.

    • I like how you’re thinking of it as your retirement emergency fund. My issue is entirely with spending the tax advantaged funds early, not with actually doing the conversions for tax purposes and then sitting on the money for later.

  29. I agree, a few years back we had to withdraw all of our Roth principle as I noticed we didn’t have enough funds in taxable accounts.

    Since we’ll be in the 0% tax bracket for cap gains when we ER, it just makes no sense to have an over funded Roth.

    I also don’t have a problem with 10% penalty withdrawals if done modestly. Our plan is to withdraw 28k per year from a tIRA, which will incur 2800 in penalty tax, and offset that with 2k in child tax credits. The rest of our finding will come from our taxable accounts early on. Later on if we run out of taxable investments, we’ll do a mix of setting up rule 72t as well as penalty withdrawals.

    I’d also like to point out that the 4% rule is one man’s opinion and not a rule at all. It’s based on flawed logic where Depression returns, which were negative, are disaccoatiated with the increase in purchase power during that era due to deflation.

    • Don’t worry, I am not going to defend the 4% rule, as you know. ;-) In drawing down your IRA and considering some 72t withdrawals in early retirement, are you guys comfortable that you’re leaving an ample supply of funds for your later years?

      • Oh yes, most definitely. I use our reasonable historical return of 9.7% (SP500 with dividends reinvested) and our account value grows each year even with our withdrawals. I think it’s important to realize that the value today is not static and accounts will grow quite a bit in future years. There is no reason to think returns won’t continue well into the future. 9.7% is pretty conservative (1965-2016) when you consider that there is a 50% chance that future returns will be even higher than that, with dividends reinvested.

        Now, before we ask questions about why I am not taking inflation into account, I do. I see these as two separate equations though. My spending per year will inflate by 2%, and our investments will increase, on average, 9.7% per year for our equity portion. We will also have two years of living expenses socked away into a high grade corporate bond fund, in order to “smooth out” the declines in the market and not touch our equity in a bear market.

        I am certain this will be a winning strategy. It’s true, you can’t die with your money. Well, you can, but someone else will get to enjoy it. There is no need to plan for a worst case scenario. If that was the case I’d buy all kinds of insurance I don’t currently have instead of self insuring.

        Have you factored into account the risk of dying with too much money? How are you planning for that?

      • I still think that’s way too optimistic, and a lot of very smart people who think about this stuff all day every day think we’ll be lucky to get 5% over the long-term future. But none of that matters — what matters is that you have a plan that you’re comfortable with. No need to convince anyone else that it’s a good plan so long as you’ve thought it through and are comfortable with it.

  30. Nice post. I actually worry about if I might end up with too much in pre tax money. The one thing that keeps me wanting to put money in the 401K is the thought that my wife and I will most likely move to a state like Florida where there is no state/retirement income tax. That would be officially avoiding state income tax on all that money. Assuming the laws don’t change. How does that play into your thoughts on pre and post tax money. My father in law moved from CT to FL and he said it felt like getting a 8% raise.

    • I think that’s a totally legit concern, though less of one for most early retirees, since many of us don’t have a ton of taxable income anyway!

      • agree. BTW, I like your two phase plan. Makes really good sense to me.

      • Thanks! We’ve changed a ton about our approach over time, but the two-phase model is the one thing that has stuck. ;-)

  31. We plan to do Roth conversions once our income is lower because if we left it all in 401k & traditional IRA for 30+ years the growth would make the RMDs large enough that all our social security income would be taxable and likely throw us in a higher tax bracket based on current laws. That was my one criticism of Mad FIentist article is that it didn’t appear to take taxes on RMDs, especially as it relates to making SSI taxable and a potentially higher tax bracket, into account. Perhaps that was in the full spreadsheet (I didn’t review it). I haven’t seen many tackle the combination of early retirement, RMDs and SSI in much later years and am happy to see you are talking about what comes after early retirement. We are also on the PoF plan to minimize what is in the pretax accounts by our late 70’s.

    Because of the college funding rules of what a parent “asset” is (which appears to exclude retirement accounts and home equity at many schools) we aren’t planning much in taxable accounts until the youngest is in college to maximize school funding opportunities. Instead we plan to fund our dirtbag millionaire years with part time fun jobs to cover annual spending, and tapping the Roth only if something highly unusual happens.

    • That seems like a well thought-out plan! I have work, for what it’s worth, that some private schools do look at parents’ retirement accounts, maybe more as an indicator than as direct funding, but schools that only use the FAFSA don’t look at that stuff for sure. (We’re both state school grads, though, so are big fans of staying public!)

      I’ll confess that we haven’t yet thought much about RMDs, though I’m sure will tackle them down the road. I know we come at all of this from a fundamentally different place than many folks in that we don’t see high taxes as a problem, we see them as something that will only happen if we’re incredibly fortunate. And our general thinking has always been that if our income is too high one year and a bunch will get eaten by taxes, then we’ll just donate the difference to charity to lower the taxes and do more good. But definitely a good topic to address down the road!

      • I’d agree taxes aren’t a big problem because you only owe income tax if you have income so a nice “problem” to have. On the other hand we are also not above structuring things in our favor as the tax code allows. It allows us to donate directly to classrooms at our local school rather than feed the bureaucracy at all the levels between the tax collector and classroom and hope that after the trickle down there is still something left for classroom enhancements. It’s harder to get IRS-approved contributions documents from classroom donations without going through district but going through the district means they take a cut and limit choice of approved vendors who generally cost more. Bureaucrats.

        I have heard that about some private schools as well. It may or may not work out but no harm trying, right?

      • It’s admirable that you all are so dedicated to making contributions that benefit your local school! I know I’ve heard that school districts would always prefer more funding from tax dollars because that’s easier to anticipate, rather than direct gifts which are all over the map, but if your goal is to improve things directly in the classroom, then I think your impulse is the right one and is more likely to achieve what you are aiming for. Good on you! :-D

  32. People do 72t or the ladder because they don’t have much saved in taxed accounts. I don’t think anyone thinks they should spend their tax-deferred accounts before they spend their taxed accounts. Others do the ladder mostly to use-up the 0% tax bracket and prevent or reduce RMDs. Not everyone has the same proportion of taxed/IRA/Roth so there is no single answer to the problem.
    I stopped working (maybe retired) 17 months ago and will probably do the ladder if I have room to use the 0% bracket. I have rental income and some taxed stash so I don’t know for sure but my plan is similar to yours.

    • My big picture thesis is not to screw future you by assuming level spending over the years, and I’ve seen a number of folks advocating for using the ladder to spend more of their money during their younger years. That’s the real problem, not truly the way you choose to take distributions. Congrats on (maybe) retiring! ;-)

  33. We don’t have the same two stage approach exactly like yours but I think our plan is more like yours than the majority of people in this FIRE community. We don’t put every last dollar in our tax qualified accounts as I want access to money in those first 5-10 years without having to worry about conversion rules and as you mentioned, increasing taxes and potentially falling off the subsidy cliff.

    Our plan for the first 5-10 years is to use our rental income to cover a large portion of our expenses, work part time (growing my husbands side business a little), and supplementing all of that with our taxable accounts. I have been playing with our numbers to see if we can hold off starting our Roth conversion ladder for 5 years and if we have enough to make things work. The calculations as you mention are not quite as easy as the 25x and 4% rules but I like some of the flexibility of our plan. I don’t want to have to worry about playing the Roth conversion game if our income creeps close to the subsidy cliff. Glad we are not the only ones relying completely on the conversion ladder.

    • Your note to me underlines something that I didn’t explicitly state in the post, but which is definitely true: NOT relying on Roth conversions or 72t distributions makes your plan inherently more flexible. So, sure, you might still do some converting, but you have other income sources built in, and can do more to tweak your income to fit into health care subsidy rules (if they stay in place!), tax situations, etc. Sounds like a well thought out plan!

  34. Great post to get a discussion going. We have what could be considered a ‘two bucket approach’. However, about 4 years ago I started to think of all our accounts as one, mainly for the tax advantages. The US tax code adds alot of complexity. As some have mentioned RMDs will likely be important and I don’t know how to plan for future changes, so I plan for today’s tax code and SS (however I reduce my SS estimate to 75%). There are two great sources that show how beneficial it can be to draw down retirement accounts and then delay SS.

    1) Michael Kitces: “Tax-Efficient Spending Strategies From Retirement Portfolios”

    2) Dana Anspach “Control Your Retirement Destiny: Achieving Financial Security Before The Big Transition”

    Individual situations can really affect this alot. I have a spreadsheet that shows planned Roth conversions before age 59.5, then drawdown from 403b from 59.5 to 70 to reduce the effect of the 403b on RMDs and SS taxes. Then SS at 70. There are also planned Capital Gains or Loss harvesting in taxable account. If market goes down then I’ll be able to roll more IRA into Roth. Cap Gains harvesting and Roth rollovers and tax efficiency are a major reason to think of your accounts as one. You have equities in taxable, and fixed income in tax deferred accounts. It is a bit complicated but check out the Kitces article that shows it has huge effects.

    I do agree that 4% is questionable given the current Shiller PE ratio and people should be conservative and have a buffer.

    • You’ve clearly done a lot of calculations, which is awesome. Have you also factored health care into all of this? Obviously health care is in huge flux right now, but if any income-based-premiums stay in place, it might be highly disadvantageous to do things like tax loss harvesting and backdoor Roth conversions. You might save some money from being taxed, but at the cost of higher premiums that more than offset those tax savings. Curious if you’ve thought any of that through!

      • Yes, great point. Root of Good has a nice article on this “Don’t Fall Off The Affordable Care Act Subsidy Cliffs”. The current plan is to roll over just enough each year from 403b to Roth (not backdoor) to generate income at just under 200% FPL. The conversions will take advantage of low income after FIRE and will be close to free. With this plan ACA would cost about $4,000 for a couple in WA state. Re: tax loss harvesting, if we do that, we will simply roll more over that year from 403b into Roth to get to just under 200% FPL for ACA. Of course ACA repeal has to some degree thrown a wrench in this plan, so now we are just budgeting $8,000 a year for health insurance based on a what I think a worst case scenario for a high deductible plan will be. This is all much more complicated than it needs to be because of US tax code, but the benefits of harvesting gains and then losses is just too good to pass up!

      • Ha, yes I know that RoG post well and have linked to it at least a dozen times. Guessing you haven’t found my health care posts yet! ;-) Worth considering: tax loss harvesting up to a cap might make sense in one year, but then you’ve locked in a lower cost basis, which ups your capital gains in future years and may inadvertently push you over a cap down the road or limit what you can take out, thus constraining your spending. Something to consider. And it’s good you’re budgeting more for health care now in light of the uncertainty. Let’s hope $8K is enough!