By Alan Silverstein, Fort Collins, Colorado.
Email me at ajs@frii.com.
Last update: January 23, 2021
This webpage contains footnotes linked from my "Thoughts on Retirement Cash Flow" main page. I separated them so the main page doesn't look so long or daunting, and so you can have both open at the same time, say in two separate browser tabs.
I have one main checking account. Nearly every dollar I spend shows up as a withdrawal (debit) from this account sooner or later. For example, this includes monthly credit card payments.
Once a month when balancing my checkbook, I just note the total outgo reported to me by my credit union (or bank). I append that number to a spreadsheet.
Sometimes I modify this debit number in one of two ways:
But beware! Only add back refunds that returned to your checking account. This is subtle and it took me a while to realize it, but refunds into other accounts that are filled out of your checking account, in particular, credit cards, don't need adjusting. The refunds automatically reduce your total spending (that you're tracking, out of the checking account) over time into the other account.
The spreadsheet calculates my monthly and yearly average actual spending based on however many months I've logged since beginning (or last "reset"). The result is a fairly believable (although surprisingly high) dollar figure.
I call this a "top-down" approach, which is simple. Of course you can also use a "bottom-up" method, which is harder, including detailed budgeting and/or expense tracking. Just beware in the latter case that you don't overlook significant "miscellaneous" spending. You might want to do some top-down tracking first as a sanity check.
Note well, don't over-rely on your average spending number until you have at least one year's data, to include repeating periodic expenses. But also note, one or more years of average actual outgo will automatically include any income taxes paid.
Social Security is a big deal for most people. In some cases, like disability or survivorship, it provides value even before you "turn of age." But the rules can be complicated. Here are just a few considerations worth summarizing.
Either way, I occasionally noted my increasing amounts of promised future payments, but discounted them and did not include them in any retirement modeling. "Who knows if I can count on SS to be there?"
As I got within a few years of retirement though, it was easy and worthwhile to expand my spreadsheet a little to see what difference SS might make in the future. Even if like me your retirement investments are considerable, adding SS can still greatly lower your "ratio". If nothing else, it gave me confidence of some extra "cash flow insurance" for later, when I got old enough.
An important part of your retirement income strategy is the decisions you make on how and when you begin taking Social Security distributions. In the case of married couples there are various strategies and benefits coordination that in some cases can significantly increase your overall lifetime benefits. There are additional benefits and strategies available to those that are disabled as well, as those who were previously married for more than 10 years and are currently unmarried.
This advice drives me nuts. The increase should more or less be actuarially level. This means part of it, reasonably enough, represents your deferral of current income. You can think of that as if you started at age 62, but fully reinvested your proceeds in the plan to boost your annuity payment in later years. But the rest of the increase only comes your way if you survive everyone else who did not! "Are you feeling lucky?"
Of course if you live beyond your actuarial life expectancy, the longer you can defer taking benefits, the better you'll do overall.
Now there are some detailed analyses out there (which you can probably find for yourself) saying the deferral is still a good deal (a positive gamble) because the reward rate is too high for today's longer life expectancies, etc. So don't reject deferral just on my say-so... I'm not fond of annuities of any kind, anyway.
Personally I'm likely to grab "the bird in the hand" (no deferral), even if it ends up being partly income-taxable (a good problem to have). Earlier SS payments mean less need to draw upon other resources, allowing them to hopefully keep compounding, similar to if I did pull from them but then reinvested my SS proceeds into them.
Your decision on when and how to take SS benefits shouldn't hinge on taxability, but it's another subtle consideration that might surprise you.
If you need cash flow, why not have banks and other agents mail or deposit your interest and dividends directly to you? After all, you pay income tax on these proceeds each year (1099-INT, 1099-DIV) whether or not you reinvest them.
But only do this if:
Conversely, reinvesting interest and dividends into capital assets like mutual funds can result in much more complicated tax reporting upon sale of the asset, unless your account manager does average cost basis reporting for you -- then it's a wash.
Having a "financial health meter" on your dashboard is really important. (Credit Bob Adams for the metaphor.)
Once a month after balancing my checkbook, I snag and append to a text file (you could use a spreadsheet) just one line the current values of my assets (each rounded to the nearest dollar). Having simplified my holdings, and nowadays with online lookups, it doesn't take me too long. This also allows me to watch and plot the trend, "admire" new high-water marks, etc.
Mostly I use pre-tax current values, but for some assets like real estate, I deduct, say, 6% for approximate sales costs to be a little more accurate.
Another method: One friend taught an account manager (like Vanguard) website how to access his "outside accounts" to automatically update the total value of his portfolio -- excluding assets like a house that can't be updated this way.
Whatever works for you, the goal is to obtain a simple "dashboard view" that's also easy enough to update that you do in fact review it periodically.
Note: If it's true that humans hate losses more than we love equivalent gains, the more often you look, the more likely you are to see a loss instead of a gain. That's why some advisors suggest not even checking your net worth more than annually. But you can feel free to track it as often as you like, once you are aware of this cognitive bias and don't let it bother you.
If your ratio (including provision for income taxes) is lower than your average rate of investment return, your net worth should continue to nominally increase even if/as you withdraw from your assets for living expenses. But your real buying power might decrease with inflation.
If however your ratio plus the average (or year-by-year) rate of inflation (whichever measure you believe, such as core or chained CPI) is lower than your average (or year-by-year) total investment returns, then your real net worth (buying power) should continue to increase.
While saving and planning for retirement I found it useful to informally watch the changes in my ratio from month to month. I saw it mostly dropping as I continued to work and invest, and as my net worth increased. (Apparently faster than my spending needs.) Plus I gathered more monthly datapoints on how much I was actually spending, refining the average.
When my ratio consistently went below 3% -- before even factoring in future Social Security benefits -- I had a pretty good feeling it was OK to stop working, without having to make any other big lifestyle changes. This was especially pleasant when reading advice to simplify your life and cut your costs soon after retiring.
So now my wife and I are both retired, starting at age 57. At least we're calling it retirement, not planning to look for work in the foreseeable future. Of course our backup plan is to return to earning more money somehow. (Update at age 66: So far, so good, we haven't felt any need to earn more money! And to my surprise, my own net worth has actually increased since retiring, not just in nominal terms, but even in real terms, post-inflation. I'm sure that can't last.)
Bear in mind the differences between:
-- which I won't belabor here, you can look them up.
Every year you might have "non-discretionary" sources of "new" (taxable) passive income, including interest and dividends from non-qualified assets. Also you might sell appreciated capital assets, or withdraw (distribute) from qualified retirement funds -- if over 70.5 (or later, 72.0), the rules force you to withdraw and pay tax on RMDs (required minimum distributions) from traditional IRAs, 401ks, etc. So it's likely you will have AGI most years, and perhaps net taxable income, even if you don't have earned income.
By the way, remember to file and pay quarterly estimated taxes (Form 1040-ES) if you expect taxable income exceeding the "safe harbor rules" without sufficient pre-withholding. (If you don't know what I'm talking about, you should definitely go read up on the subject!)
Note that Social Security payments can optionally be subject to withholding, even "excessive" withholding to cover non-SS income streams, as one way to avoid doing quarterly payments.
Certain kinds of taxable income (TI) are called "preferential income", including long-term capital gains, capital gain distributions, and qualified dividends. I abbreviate them all as "PTI". They are taxed at much lower rates than ordinary taxable income (OTI). The tax brackets are the same (you can see this if you Google around), but the percentages are lower.
For example in 2013, people married-filing-jointly pay 10% on OTI up to $17,850, and 15% on TI above that up to $72,500, but 0% on PTI up to $72,500. So in simple terms: If you have a choice, you should "always" take PTI before OTI, to pay less income tax this year.
Now some sources of income are essentially non-discretionary, like interest or dividends from holdings, or IRA RMDs when you're 70.5 (or later, 72.0) or older. However for other sources you can choose if/which way to go. For example, you might decide to earn some (or more) money by working (more) to increase your OTI, or sell short-term (OTI) versus long-term (PTI) capital assets, or perhaps just spend non-taxable (already-taxed) assets like CD balances when they mature.
It's very complicated trying to figure out the tax effects of various options you might have, especially when you consider the current tax year (liquidate certain assets now) versus later tax years (keeping your options open by holding onto certain assets). I can't offer much more specific advice other than to be aware of the OTI/PTI issue, and fold it into your planning.
Once you hit age 59.5 you can think of your traditional IRA (or 401k) distributions (for spending) or Roth conversions (for continued investing) as very similar operations. Both are just taxable withdrawals. The difference is whether you use the funds to refill your short term buckets or, if you can afford to wait, to instead "just" reduce your future RMDs at age 70.5 (or later, 72.0), that is, doing multi-year income-tax leveling) by shifting the money from your traditional IRA to Roth IRA.
I'm learning that some people over age 59.5 focus on conversions first (of course doing them late in the tax year to optimize/minimize tax consequences), but then treat their Roth IRAs at least partially as mere savings accounts. If you need more spending money, you can withdraw it at any time tax-free and penalty-free (even the 5-year clock on conversions goes away when you're old enough) for the "cost" of another 1099-R tax statement. And if you don't need it, any amount you can leave untouched in the Roth grows tax-free.
Also note that converting rather than distributing the same traditional IRA dollars gives you more options later for a do-over (partial or complete recharacterizing) if desirable for various reasons, such as reducing your MAGI below some threshold such as loss of ACA premium tax credit. And Roth conversions can also help you diversify your choices in future years for obtaining money from pre-tax versus post-tax sources (to optimize your tax efficiency).
But of course if you keep your Roth dollars, at least the portion you don't mind pulling back out soon, in volatile assets, you aren't really treating them as short-term (safe) assets, are you? This suggests that at least your Roth conversions should go into bond funds (or equivalent).
Years ago I bought into the notion that upon reaching retirement age most of your assets should be less volatile, say in bond rather than stock funds. I recently came to realize that this is a "spacewise" allocation, whereas a friend does what I now consider to be a "timewise" allocation. He keeps in bond (or REIT) funds only what he thinks he'll need for the next 10 years (including allowance for some emergency spending), and the rest of his portfolio in equities. Every time the S&P500 hits a new high, he considers transferring out some "winnings" from the latter (equities) to the former (bonds/REITS). (This is a kind of market-driven timing of portfolio rebalancing.)
Whenever you have enough money saved to feel comfortable about retiring, your time horizon early in the game is hopefully much longer than 10 years. In which case the timewise model suggests you should be less conservative, say 30%/70% bonds/stocks rather than say 60%/40% (depending on your total assets and your net spending needs after perennial income streams). Odds are you'll do better over your lifetime this way than by being more conservative per the spacewise allocation model. Hmm... I'm still digesting this one.
Whenever you need to liquidate some assets in retirement to create cash flow, how do you review and study your resources to make a good decision? Below is an example of a spreadsheet I cobbled up for our own use (presented here in simple text form), with fake numbers and account types. Also it contains fewer lines than the actual assets my wife and I have between us, but the beauty of this method is that the data still "fits on one screen."
$K TAX% $K_T LIQ TYPE
15 0 0 high cash and bank accounts
31 31 10 high Fidelity non-retirement (% based on lookup)
15 100 15 med Schwab 401K (special case for us due to 55+)
10 0 0 med Fidelity Roth IRA (can withdraw principal before 59.5)
22 100 22 lo Fidelity traditional IRA
93 38 35 lo rental house (after 6% sales costs)
---- ----
186 82 104 tax-free diff
15% estimated future fed + state taxes
(ignoring exemptions, deductions, etc)
174 estimated post-tax total value
Meanings of fields:
So when we need to draw money (cash flow) out of our assets, we can see at a glance what's available, how easy it is to access, and the main tax consequences of converting it. (But not the ordinary versus preferential taxable income tradeoff discussed elsewhere.)
One thing I intend to do every year is to estimate our taxable income from "non-discretionary sources", and ensure we pull enough more to at least use up that year's exemptions and standard deduction (or itemized deduction if you're itemizing).
For your consideration here's a simple way to think about year-to-year changes to your net worth in retirement:
(ratio < returns) => increasing nominal net worth ((ratio + inflation) < returns) => increasing real net worth
By ratio I mean your annual net rate of spending, including income taxes but after deducting your offsetting perennial income, divided by your total net worth. By returns I mean your annual total percentage return across your portfolio. By inflation I mean the annual rate.
I had an epiphany that's not really new or profound, but which took me a while to articulate briefly. If in a given year you spend, say, 2.8% of your net worth (including paying taxes) after any offsetting perennial income streams (like interest, dividends, pensions including SS, annuities, rents after expenses, and royalties), you need total investment returns from your portfolio of at least the same 2.8% to appear to break even.
However to really break even, to have the same (or more) buying power remaining at the end of the year as you had at the start, your returns must also cover inflation. (By whichever measure you prefer to use, such as core or chained CPI; see for example this website or this website.) In 2013 for example you might have needed 2.8% + 1.5% = 4.3% total pre-tax returns to break even, and more to grow your real net worth.
Of course many more detailed variables underlie this simple narrative, which affect your present and future ratios, inflation (national average and your own personal experience based on your "vector" of purchases), and your portfolio returns (including asset allocation and expense ratios). Your break-even probably varies up and down a lot between one year and the next. But you can use this model to judge your chances of "never running out of money" or even increasing your estate over many years, as you juggle your net spending needs against your asset allocation choices.
Note that it's fine to have your real net worth decrease in retirement, so long as you can do whatever you need or want to do, while not outliving your savings. The big questions are how much you'll need in the years ahead and how many of them you have left! But a growing real net worth "now" can reduce some worries about the long term future.
The idea is for you to have 6-24 months (different sources vary) of readily available cash for living expenses or emergencies in some non-volatile form. Of course many people still park some of their short-term money in CDs (accepting early-withdrawal penalty risk) and/or short-term high-quality bonds (accepting some volatility).
In principle the reason for having these buckets is so you don't have to execute "sell" orders frequently, and you aren't forced to liquidate assets during a market downturn. The concept is that you refill your short-term buckets any time you get around to it and/or when it feels right. But in practice this often involves potentially taxable sales of longer-term assets. So I think it's wisest to only do the analysis and transactions late in each tax year when you can assess the impact on your income taxes and (if applicable) your ACA PTC (premium tax credit).
Ironically this means you have less than ideal flexibility to avoid selling during market downturns. Then again, you're not (in theory) supposed to be able to time the markets anyway, or even to make a practice of trying to do so! Never mind that withdrawing on a somewhat periodic schedule causes you to do "inverse dollar cost averaging."
Also every time you refill your short-term buckets and then start spending from them, you drop the total below your intended target. But you don't want to mess around with refilling them too often either, say more than once a year. What to do about this quandary?
I think you actually must establish two goals: How full to replenish your short term buckets annually, and how low you are comfortable letting them get between refills. Say for example, refill annually to the two-year level, and worry if they get below the one-year level. If your planning is good enough the latter shouldn't happen, but if large unexpected expenses materialize you can always refill early or, equivalently, liquidate the extra dollars "out of phase" to cover the unusual event.
After you figure out how much spending money you need on average, you still know that over the long term, say the next 10 years, you're likely to have at least one unforeseen event where you need a significant additional lump sum. This might be to pay cash for a new car, cover some large medical or dental expenses, bail out a relative, etc. And in these cases you might not be able to wait until the end of the tax year to minimize tax consequences.
This is a good reason for having some relatively non-volatile assets (such as bond funds) in non-retirement assets (no- or low-tax situations for capital gains) and/or Roth IRAs, beyond your normal short-term buckets.
About Roths in particular: I used to think withdrawing from them at all was taboo; they should be your "bottom dollars." But now it seems to me that post-retirement, Roth IRAs can serve as "tax shock absorbers" similar to how bond funds are often touted as "volatility shock absorbers" for your portfolio. Just so long as you don't end up draining them too fast compared to your life expectancy.
Websites:
A short list of especially memorable books about personal finance (saving, investing, and retiring) that I've read and learned a lot from:
I haven't read these, but an acquaintance recommended them: