Spending down cash in a stock market downturn - does this actually work?

[GaryA505]

This is something I've been wondering about and haven't found a consensus or answer to. I see a lot of retired people say they hold cash or cash-like liquid assets to spend during a stock market downturn. The idea is to not spend the equity assets, leaving them alone to recover. I know this is may seem like it's a good idea (psychologically at least) but does it actually make any difference? Has anyone done any sort of a simulation on this?

[alex_686]

This is something I've been wondering about and haven't found a consensus or answer to. I see a lot of retired people say they hold cash or cash-like liquid assets to spend during a stock market downturn. The idea is to not spend the equity assets, leaving them alone to recover. I know this is may seem like it's a good idea (psychologically at least) but does it actually make any difference? Has anyone done any sort of a simulation on this?

No. As you have stated it may have psychological benefits.

The problem is that the stock market is more or less a random walk. i.e., there is no mean reversion. That is, that the stock market has gone down doesn't contain any information. Going down doesn't imply it is will go back up to its old level. It is just as likely to fall more as it is to go up.

You can't put together any rational set of market views to hold cash as a buffer or to dollar-cost-average into the market.

[livesoft]

Going down doesn't imply it is will go back up to its old level.

LOL! When has the stock market reached an all-time high, then dropped down never to reach another all-time high? Taking your statement at face value suggests it is totally bogus: Of course the stock market goes back up after going down; otherwise none of us would bother to invest in the stock market.

[gammalaser]

But, if market goes down such that you are below your target equity asset allocation, it makes sense to spend down from non equity part of portfolio to help get back to target AA. But that could be bonds instead of cash.

[nisiprius]

I'm pretty sure it doesn't.

The big problem is that there's no guaranteed upper limit of how long a "downturn" can last, and many offhand statements about the "typical" or "average" or "maximum" length of a bear market are wildly optimistic. The question is: how many year's spending do you hold in cash? I've often seen numbers like "three years."

But secular bear markets can last 7 years or more. Worse yet, 1929-1936 and 1937-1944 were two back-to-back bear markets with only three months in between them... and it is basically a fluke that it just barely recovered. If 1936 had been just slightly less good, 1929-1944 would be a single sixteen-year bear market. And in terms of drawing down a cash bucket, very nearly as bad, because even if 1929-1936 did not fully deplete the cash bucket, three months was not long enough to re-fill it before the next decline hit.

On the other hand, we often think of 2000-2003 (the tech crash) and 2008-2009 (the global financial crisis) as two separate bear markets, but in Morningstar's scoring, inflation-corrected and including dividends, there wasn't quite full recovery in between, so they consider 2000 through 2013 as one single 13-year-long bear market.

So, how much do you hold in cash? Very few people who talk about holding cash to ride through a bear market are holding sixteen years, or thirteen years, or even seven years.

And here's the problem. This strategy may be like certain kinds of gambling system, in that when it works, you are better off than without the system. But only a little better off. While in the case the bear market outlasts the cash buffer, you are worse off than without the system, and a lot worse off. High probability of a small improvement balanced by low probability, but of a large problem.

I have yet to see any "buckets" proposal that says explicitly what, exactly, you are supposed to do if you run out of cash before the bear market is over. It is just implicitly assumed that this won't happen. And that there will only be one bear market--or that there will be enough time between them to fully rebuild the cash bucket before the next one hits.

[alex_686]

LOL! When has the stock market reached an all-time high, then dropped down never to reach another all-time high? Taking your statement at face value suggests it is totally bogus: Of course the stock market goes back up after going down; otherwise none of us would bother to invest in the stock market.

Tell me more - as I glance at a bookshelf laden with many heavy tomes on statistics and financial market history. Tell me how may statement, that past performance isn't indicative of future performance is wrong? It is a far far leap to go from what I said at face value to something which implies that the stock market doesn't have positive expected returns.

Lets say that there is a 25% chance of a down year - which is about the historical average for the US Stock market over the past 100 years. (I don't have the exact figures at my fingertips, can't remember if it is factoring in dividends and inflation, but it is in the ballpark.) If we have a down year in year 1 what is the chance of having a down year in year 2? 25% chance. There is no mean reversion. Results tend to be independent. Maybe. This is debatable. If there is, it is low and happens at extremes. Recovery takes 10ish years. The effect is long enough and weak enough to make my point.

Anyway you slice it the math doesn't support holding cash to ride out a recovery.

[livesoft]

Tell me more -....

You did not specify any time limit in your original statement, so I just assumed "infinite" as the time limit and not 2 years nor 5 years nor even some other time limit. I also note that you wrote "stock market" and not "any given single stock."

Anyways, I think you know that I am a zero cash person.

[jebmke]

When I retired I held both stock funds and bond funds. No cash. When stocks went down in 2008 (I retired Decembe 31, 2007) I was selling bond funds and using some of the proceeds for spending and some to buy more equity funds (re-balance). In that sense, yes, it works. Is that what you intend to do with cash (essentially a "bond" with no duration risk)?

edit: I have no idea whether this makes a difference in the long run as I don't track or forecast performance.

[Hyperchicken]

It's one of the many fallacies that Bogleheads are as prone to as anyone else.

"What if the market is down and I need to sell?"

At its core this is another instance of loss aversion cognitive bias, I believe.

[Badinvestor]

After the Russian Revolution, as I understand it, the Russian stock market didn't do too well. However, the ruble didn't do too well either. Lenin deliberately set out to make it worthless. Spending down cash thus didn't work very well for Russian stockholders. Fleeing the jurisdiction and earning one's living by working -- like the stereotypical Russian nobleman who became a waiter in a Paris cafe -- was more effective.

[alex_686]

Tell me more -....

You did not specify any time limit in your original statement, so I just assumed "infinite" as the time limit and not 2 years nor 5 years nor even some other time limit. I also note that you wrote "stock market" and not "any given single stock."

Anyways, I think you know that I am a zero cash person.

I chose year-by-year because that is how most IPS are set up.

I didn't specify a time limit nor single stock/stock market because it doesn't matter. I have seen this thing studied both on a intra-day and multi-decade scale. Both in theory and in practice.

Holding cash to ride out the dip assumes that there are dips and the length of dips can be predicted. Neither are true. Dips are actually discontinuities but that may be a bit technical and pedantic. In any event, that there is some form of mean reversion. It either isn't there or is so weak that it isn't actionable.

As a mental exercise, try to put together a rational set of market expectations where it makes sense to withdrawal cash if the stock market goes down. Can you do it without making assumptions on how the future market will do based on prior results? i.e., can you do it without resorting to market timing.

[livesoft]

"What if the market is down and I need to sell?"

Today the stock market is down and I needed to sell something in order to pay upcoming September bills. So in my taxable account I sold some shares of VEU (down about 1.6% today) to raise the cash. But at nearly the same time, I sold some BND shares in my 401(k) which are up today and bought some VEU-equivalent shares with the money. So if I hadn't told folks the details, it would look like I sold bonds today at a gain in order to pay my upcoming September bills.

See also: https://www.bogleheads.org/wiki/Placing ... ed_account

Of course, I cannot predict the future and do not know if BND, VEU, and other things will be lower later in the day, tomorrow, next month, or whenever.

[MrDrinkingWater]

This is something I've been wondering about and haven't found a consensus or answer to. I see a lot of retired people say they hold cash or cash-like liquid assets to spend during a stock market downturn. The idea is to not spend the equity assets, leaving them alone to recover. I know this is may seem like it's a good idea (psychologically at least) but does it actually make any difference? Has anyone done any sort of a simulation on this?

For those who don't have enough of a pension payout and/or Social Security benefits and/or periodic dividend payouts which all add up to being more than their living expenses, when there is a stock market downturn, the decision becomes (1) can I cut back enough on expenses to not sell any stocks right now, or (2) do I go back to work, or (3) what is the most beneficial or least worst asset to sell to meet the expenses not covered by income?

For some people, keeping six months to five years of living expenses in cash, often just earning a low interest rate, is what they do so that they have somewhere from six months to five years to (1) not cut back on expenses, (2) not go back to work, or (3) not sell other assets at a less beneficial time for those other assets to be sold.

For those who choose a five-year cash cushion, I think they are making a risk-based judgment that most bear markets apparently last fewer than five years in a row, and the next bear market will not start again too soon, or be too long of a bear market. They are okay with having five years of living expenses usually making a low-interest rate return, rather than higher returns (at higher risks) that intermediate bonds or a broad-market low-cost stock index fund might alternatively provide.

For those who choose a shorter-duration six-month cash cushion, that might reflect what their behavior was during their working years of keeping six months of living expenses on hand, mostly in the hope that a new job could be found within six months after losing a job.

[Hacksawdave]

This is something I've been wondering about and haven't found a consensus or answer to. I see a lot of retired people say they hold cash or cash-like liquid assets to spend during a stock market downturn. The idea is to not spend the equity assets, leaving them alone to recover. I know this is may seem like it's a good idea (psychologically at least) but does it actually make any difference? Has anyone done any sort of a simulation on this?

Not a simulation, but from the last go-round as a recent retiree, yes. From 8/1/2021 to 03/1/2023, I started with $76,525 in cash and finished with $23,211 left. This was supplemented by a few automatic distributions from income producing funds of $41,759. I covered all 18 months with the income distributions and cash totaling $95,073.

All equity funds were reinvested along with Roth and 401k account holdings. That was an 18-month period where not a single stock or bond share was sold. I still had some cash left to buy CA long-term municipals in December of 2023 before rates started falling further thus increasing the NAV price. Balance-wise I fully recovered one year later on 3/1/2024, even with a few bond funds still under.

As stated by other posters, nobody knows how long the downward trend could last. The preparations called for a 1973-74 bear market, and 18 months was the exact length of both. The only difference was stocks declined less than in 73-74, but bond losses made up for the lower equity number.

[Johm221122]

If I was going to use cash reserves as my plan for retirement withdrawals in worse/ bad case scenario it would be part of the plan. I would cut spending or not take inflation adjustments also (not necessarily the first down year but it would definitely be second year)

[GaryA505]

So, not to derail my own thread, but let's put a little twist on this "spend cash in a market downturn" strategy. Rather than equities with a cash buffer, let's say you have two retirees:

Retiree A: Starts with a portfolio of 60% VTSAX total US stock and 40% VBTLX total US bond. Annual (or monthly) withdrawals are taken from whichever part is above it's allocation (please assume they do not remain balanced at 60/40 over time). So, this retiree can avoid selling stocks during a big downtown, but does not rebalance so does not buy equities at low prices during such a downtown.

Retiree B: Holds only Vanguard Balanced VBIAX (60/40). Withdrawals have to come from this one fund, of course, and it rebalances itself.

Which one will come out ahead?

[Wiggums]

We rebalance per our IPS. Rebalancing during a crash, in the past, has put you back into the green faster. Rebalancing too ofter may add to your tax liability.

Our holdings are quite old. The oldest being 39+ years old. So we sell from the portfolio when we need money. The market can be high or low.

During the pandemic market crash, we rebalanced and started adding to our portfolio since we were stuck at home and had extra cash.

[alex_686]

Which one will come out ahead?

Define "ahead".

The second portfolio would be the more efficient portfolio. i.e., return per unit of risk.

The first portfolio would be less efficient. So a greater chance of failure. It would also tend to skew towards equities over time, so higher expected return and a higher. So a higher expected terminus value and a higher expected chance of failure.

The first portfolio is also harder to model. A 60/40 portfolio would be roughly in the same spot in a up-down market as a down-up market. This ad-hoc model is path dependent.

[btraven]

The opportunity cost of holding cash for more or less a lifetime has to be considered. This opportunity cost buys the market timing opportunity of not selling a security in a "dip". However, we don't really know that the dip is a dip except in the long term, so it is market timing.

However, we have forgone the opportunity that cash (tens or hundreds of thousands) would be better invested over a lifetime in the market with the long term expectation of gain. Isn't that more BH-y? I am ignoring short term cash flow and tax considerations

[life_force_prana]

So, how much do you hold in cash? Very few people who talk about holding cash to ride through a bear market are holding sixteen years, or thirteen years, or even seven years.

And here's the problem. This strategy may be like certain kinds of gambling system, in that when it works, you are better off than without the system. But only a little better off. While in the case the bear market outlasts the cash buffer, you are worse off than without the system, and a lot worse off. High probability of a small improvement balanced by low probability, but of a large problem.

I have yet to see any "buckets" proposal that says explicitly what, exactly, you are supposed to do if you run out of cash before the bear market is over. It is just implicitly assumed that this won't happen. And that there will only be one bear market--or that there will be enough time between them to fully rebuild the cash bucket before the next one hits.

What if someone plans to hold 20-25 years of expenses in "cash-like" instruments like I/EE bonds, TIPS Ladders, TBill Ladders and Stable Value Funds? (BTW 2 out of those 4 are inflation protected so no loss of opportunity cost or value)

I do not understand why all these topics about holding cash get bashed as some sort of bucket strategy that suffers from bucket re-filling problems, or cash gets defined in a limited way.

I see 3 (paper) asset types: equity funds (to use for spending or refill cash in upmarket), bond funds (to rebalance against stock funds), and cash as defined above.

What am I missing with this line of thinking?

[brightlightstonight]

So, how much do you hold in cash? Very few people who talk about holding cash to ride through a bear market are holding sixteen years, or thirteen years, or even seven years.

And here's the problem. This strategy may be like certain kinds of gambling system, in that when it works, you are better off than without the system. But only a little better off. While in the case the bear market outlasts the cash buffer, you are worse off than without the system, and a lot worse off. High probability of a small improvement balanced by low probability, but of a large problem.

I have yet to see any "buckets" proposal that says explicitly what, exactly, you are supposed to do if you run out of cash before the bear market is over. It is just implicitly assumed that this won't happen. And that there will only be one bear market--or that there will be enough time between them to fully rebuild the cash bucket before the next one hits.

What if someone plans to hold 20-25 years of expenses in "cash-like" instruments like I/EE bonds, TIPS Ladders, TBill Ladders and Stable Value Funds? (BTW 2 out of those 4 are inflation protected so no loss of opportunity cost or value)

I do not understand why all these topics about holding cash get bashed as some sort of bucket strategy that suffers from bucket re-filling problems, or cash gets defined in a limited way.

I see 3 (paper) asset types: equity funds (to use for spending or refill cash in upmarket), bond funds (to rebalance against stock funds), and cash as defined above.

What am I missing with this line of thinking?

I agree that the performance of a few years in cash and a few years of a TIPS ladder is unlikely to be wildly different. But I also draw a distinction between years-of-cash or a TIPS ladder set up for a very specific reason, like bridging to social security, and years-of-cash forever for fear of a downturn.

Market timing is unlikely to work; lifecycle timing is much more plausible. I don't know when the next downturn will start or end; I do know when I will turn 70 and be ready to take SS (with one not-so-small caveat).

[life_force_prana]

So, how much do you hold in cash? Very few people who talk about holding cash to ride through a bear market are holding sixteen years, or thirteen years, or even seven years.

And here's the problem. This strategy may be like certain kinds of gambling system, in that when it works, you are better off than without the system. But only a little better off. While in the case the bear market outlasts the cash buffer, you are worse off than without the system, and a lot worse off. High probability of a small improvement balanced by low probability, but of a large problem.

I have yet to see any "buckets" proposal that says explicitly what, exactly, you are supposed to do if you run out of cash before the bear market is over. It is just implicitly assumed that this won't happen. And that there will only be one bear market--or that there will be enough time between them to fully rebuild the cash bucket before the next one hits.

What if someone plans to hold 20-25 years of expenses in "cash-like" instruments like I/EE bonds, TIPS Ladders, TBill Ladders and Stable Value Funds? (BTW 2 out of those 4 are inflation protected so no loss of opportunity cost or value)

I do not understand why all these topics about holding cash get bashed as some sort of bucket strategy that suffers from bucket re-filling problems, or cash gets defined in a limited way.

I see 3 (paper) asset types: equity funds (to use for spending or refill cash in upmarket), bond funds (to rebalance against stock funds), and cash as defined above.

What am I missing with this line of thinking?

I agree that the performance of a few years in cash and a few years of a TIPS ladder is unlikely to be wildly different. But I also draw a distinction between years-of-cash or a TIPS ladder set up for a very specific reason, like bridging to social security, and years-of-cash forever for fear of a downturn.

Market timing is unlikely to work; lifecycle timing is much more plausible. I don't know when the next downturn will start or end; I do know when I will turn 70 and be ready to take SS (with one not-so-small caveat).

I am referring to rebalancing to target AA so not sure that is market timing. No need to know if/when downturn/upturn starts/end.

I see the bridge to SS TIPS ladder concept. I would define 20-25 years of cash as amount needed for residual living expenses (after factoring in any fixed income like pensions, annuity, SS, rental income, etc.) so in that sense its not just till one starts claiming SS.

[brightlightstonight]

So, how much do you hold in cash? Very few people who talk about holding cash to ride through a bear market are holding sixteen years, or thirteen years, or even seven years.

And here's the problem. This strategy may be like certain kinds of gambling system, in that when it works, you are better off than without the system. But only a little better off. While in the case the bear market outlasts the cash buffer, you are worse off than without the system, and a lot worse off. High probability of a small improvement balanced by low probability, but of a large problem.

I have yet to see any "buckets" proposal that says explicitly what, exactly, you are supposed to do if you run out of cash before the bear market is over. It is just implicitly assumed that this won't happen. And that there will only be one bear market--or that there will be enough time between them to fully rebuild the cash bucket before the next one hits.

What if someone plans to hold 20-25 years of expenses in "cash-like" instruments like I/EE bonds, TIPS Ladders, TBill Ladders and Stable Value Funds? (BTW 2 out of those 4 are inflation protected so no loss of opportunity cost or value)

I do not understand why all these topics about holding cash get bashed as some sort of bucket strategy that suffers from bucket re-filling problems, or cash gets defined in a limited way.

I see 3 (paper) asset types: equity funds (to use for spending or refill cash in upmarket), bond funds (to rebalance against stock funds), and cash as defined above.

What am I missing with this line of thinking?

I agree that the performance of a few years in cash and a few years of a TIPS ladder is unlikely to be wildly different. But I also draw a distinction between years-of-cash or a TIPS ladder set up for a very specific reason, like bridging to social security, and years-of-cash forever for fear of a downturn.

Market timing is unlikely to work; lifecycle timing is much more plausible. I don't know when the next downturn will start or end; I do know when I will turn 70 and be ready to take SS (with one not-so-small caveat).

I am referring to rebalancing to target AA so not sure that is market timing. No need to know if/when downturn/upturn starts/end.

Rebalancing to a target AA isn't market timing.

I think I see what you were getting at with your first message: nisiprius wrote

Very few people who talk about holding cash to ride through a bear market are holding sixteen years, or thirteen years, or even seven years.

But one could argue that a 20 or 30 year TIPS ladder is a (well-designed) version of "holding cash to ride through a bear market".

[typical.investor]

This is something I've been wondering about and haven't found a consensus or answer to. I see a lot of retired people say they hold cash or cash-like liquid assets to spend during a stock market downturn. The idea is to not spend the equity assets, leaving them alone to recover. I know this is may seem like it's a good idea (psychologically at least) but does it actually make any difference? Has anyone done any sort of a simulation on this?

No. As you have stated it may have psychological benefits.

I argue that it can have the benefit tangible benefit of lowering Sequence of Return Risk (SORR).

The problem is that the stock market is more or less a random walk. i.e., there is no mean reversion. That is, that the stock market has gone down doesn't contain any information. Going down doesn't imply it is will go back up to its old level. It is just as likely to fall more as it is to go up.

Sure ok. But as Kitces notes looking at 10 year returns (1926-2017):

Starting valuations clearly matter, and they matter a lot. Higher starting values mean that not only are future expected returns lower, but the best outcomes are lower and the worst outcomes are worse.

You can't put together any rational set of market views to hold cash as a buffer or to dollar-cost-average into the market.

I think that SORR (Sequence of Return Risk) actually does build an arguement for holding some less volatile assets from which to spend. It doesn't necessarily have to be cash but a cash-like liquid assets such as bonds or money market funds or stable value funds could help mitigate the risk. Some put annuities in there too.

See https://www.whitecoatinvestor.com/4-met ... urns-risk/ or https://www.kitces.com/blog/understandi ... d-decades/

So you have to find a way to not run out of money. Holding too much low returning cash won't help. Neither will spending out of volatile assets in downturn.

[Robin1234]

On the other hand, we often think of 2000-2003 (the tech crash) and 2008-2009 (the global financial crisis) as two separate bear markets, but in Morningstar's scoring, inflation-corrected and including dividends, there wasn't quite full recovery in between, so they consider 2000 through 2013 as one single 13-year-long bear market.

So, how much do you hold in cash? Very few people who talk about holding cash to ride through a bear market are holding sixteen years, or thirteen years, or even seven years.

The SPY peak in 8/2000 and 2/2013 match up. With an intervening peak in 10/2007. So pretty much most of the time one was behind and only saw lasting upside post 2/2013

Below is a picture of 2000-2013 SPY
https://postimg.cc/HjcLdN9c

[rkhusky]

Holding cash in case of a market downturn has the problem of writing down a rule for when to start spending from cash. And also writing down a rule for when to start replenishing the cash bucket. And also the problem of actually following your rules, because there is usually the temptation to modify the rules when the going gets tough.

[Lawrence of Suburbia]

Some retirees with modest portfolios (say, 25x) and not-great S.S. stipends, simply can't afford multi-year cash cushions, sadly. You kinda need the whole 60/40 or whatever the A.A. is.

For us, it's cling to the 4% rule and hope for the best!

[celia]

This is something I've been wondering about and haven't found a consensus or answer to. I see a lot of retired people say they hold cash or cash-like liquid assets to spend during a stock market downturn. The idea is to not spend the equity assets, leaving them alone to recover. I know this is may seem like it's a good idea (psychologically at least) but does it actually make any difference? Has anyone done any sort of a simulation on this?

A lot of "retired people" have a steady income like SS or pensions or RMDs (if they spent a lifetime saving). They don't necessarily have to save cash, since more of it will be coming in.

However, if you decide to sell some stock, the loss is then realized and sold stock cannot return to it's previous value in your portfolio (because you no longer own it!).

[Wanderingwheelz]

This is something I've been wondering about and haven't found a consensus or answer to. I see a lot of retired people say they hold cash or cash-like liquid assets to spend during a stock market downturn. The idea is to not spend the equity assets, leaving them alone to recover. I know this is may seem like it's a good idea (psychologically at least) but does it actually make any difference? Has anyone done any sort of a simulation on this?

No. As you have stated it may have psychological benefits.

The problem is that the stock market is more or less a random walk. i.e., there is no mean reversion. That is, that the stock market has gone down doesn't contain any information. Going down doesn't imply it is will go back up to its old level. It is just as likely to fall more as it is to go up.

You can't put together any rational set of market views to hold cash as a buffer or to dollar-cost-average into the market.

History says otherwise. The broad stock market is more likely to go up after it drops than to go down more.

[DoctorE]

Are we talking portfolio under the water high mark of retirement or stock market down from all time highs?

Way I see it it's a sequence of return issue... once you get beyond that, why does one have to spend from cash when there's some income coming in.

Assuming one is retired and no more DCA, probably has some SS, probably has some dividends coming in, probably has some bond coupons coming in. Their portfolio might not be at all time highs but it could still be higher (before or even after inflation) than what they contributed to it. So in that case, do you consider the portfolio to be in a downturn if the stock market is down?

That's the issue I have with this concept, and I hope someone can explain it better to me...

[rossington]

So, not to derail my own thread, but let's put a little twist on this "spend cash in a market downturn" strategy. Rather than equities with a cash buffer, let's say you have two retirees:

Retiree A: Starts with a portfolio of 60% VTSAX total US stock and 40% VBTLX total US bond. Annual (or monthly) withdrawals are taken from whichever part is above it's allocation (please assume they do not remain balanced at 60/40 over time). So, this retiree can avoid selling stocks during a big downtown, but does not rebalance so does not buy equities at low prices during such a downtown.

Retiree B: Holds only Vanguard Balanced VBIAX (60/40). Withdrawals have to come from this one fund, of course, and it rebalances itself.

Which one will come out ahead?

Hello GaryA505,
What if either portfolio was large enough so that the dividends/interest were significant enough to reduce or even eliminate any withdrawals during the downturn no matter how long it could last? Not to mention SS, pension etc.

[20cm]

Variations on this thread come up from time to time. They're always missing a rule set for how big the cash buffer is at the start and for when and how much it gets refilled. You need to define a specific set of rules you will be executing to perform any kind of simulation or backtest.

It feels intuitively helpful to "not sell stocks when they are down" but the cash doesn't last forever, so what prevents becoming a forced seller of stocks when they are even further down? If the answer is an even bigger cash buffer that only gets refilled when stocks regain the prior all time high, what does that do to the portfolio's total return over the long term?

.

[dbr]

Right off the bat "I won't have to sell stocks when they are down." is nonsense. "when they are down" is not defined.

But there is an actionable plan that comes in a first necessary part and a second optional part:

1. (Required) Maintain your asset allocation. While withdrawing from your portfolio also sell bonds and buy stocks if stocks are below target allocation, and vice-versa.

2. (Optional) If your plan is conservative in rate withdrawal nothing else is needed. If you really want to worry then spend less when the portfolio is "down." You will have to find an algorithm to tell you whether or not the portfolio is "down" and by how much. A start might be something like Variable Percentage Withdrawal. Gut feeling may also suggest when to spend less.

Also, there is no clear proof buckets systems are helpful.

[brightlightstonight]

Right off the bat "I won't have to sell stocks when they are down." is nonsense. "when they are down" is not defined.

But there is an actionable plan that comes in a first necessary part and a second optional part:

1. (Required) Maintain your asset allocation. While withdrawing from your portfolio also sell bonds and buy stocks if stocks are below target allocation, and vice-versa.

2. (Optional) If your plan is conservative in rate withdrawal nothing else is needed. If you really want to worry then spend less when the portfolio is "down." You will have to find an algorithm to tell you whether or not the portfolio is "down" and by how much. A start might be something like Variable Percentage Withdrawal. Gut feeling may also suggest when to spend less.

Also, there is no clear proof buckets systems are helpful.

I agree entirely with your first sentence and your last, but... if we're talking about retirement and decumulation, maintaining a fixed asset allocation is not the only strategy. Rising equity glidepaths or McClung's Prime Harvesting (megathread at viewtopic.php?t=192105) are alternatives which are rule-based and can be backtested.

Prime Harvesting is essentially "spending cash as long as the market is down" (though it's bonds, not cash). But you have to tolerate the possibility of getting to 100% stocks in retirement, which is kinda terrifying

[jsapiandante]

Right off the bat "I won't have to sell stocks when they are down." is nonsense. "when they are down" is not defined.

But there is an actionable plan that comes in a first necessary part and a second optional part:

1. (Required) Maintain your asset allocation. While withdrawing from your portfolio also sell bonds and buy stocks if stocks are below target allocation, and vice-versa.

2. (Optional) If your plan is conservative in rate withdrawal nothing else is needed. If you really want to worry then spend less when the portfolio is "down." You will have to find an algorithm to tell you whether or not the portfolio is "down" and by how much. A start might be something like Variable Percentage Withdrawal. Gut feeling may also suggest when to spend less.

Also, there is no clear proof buckets systems are helpful.

+1. I always wonder how people who keep 2-5 years in cash and separate from their allocation would do if there’s a prolonged market down turn of 2-5 years or longer. Once the market has recovered and you’ve spent all your cash, do you just not replenish the cash back up to your desired amount? Because if you had to, you’d be pulling more than you would need from your asset allocation to fill that bucket back up? I would suspect it would take several years to do so.

It’s easier to just pick an allocation and adjust accordingly in down/up markets to maintain. I always feel that the ability to be flexible with spending is the great equalizer.

[GaryA505]

Right off the bat "I won't have to sell stocks when they are down." is nonsense. "when they are down" is not defined.

But there is an actionable plan that comes in a first necessary part and a second optional part:

1. (Required) Maintain your asset allocation. While withdrawing from your portfolio also sell bonds and buy stocks if stocks are below target allocation, and vice-versa.

2. (Optional) If your plan is conservative in rate withdrawal nothing else is needed. If you really want to worry then spend less when the portfolio is "down." You will have to find an algorithm to tell you whether or not the portfolio is "down" and by how much. A start might be something like Variable Percentage Withdrawal. Gut feeling may also suggest when to spend less.

Also, there is no clear proof buckets systems are helpful.

I agree entirely with your first sentence and your last, but... if we're talking about retirement and decumulation, maintaining a fixed asset allocation is not the only strategy. Rising equity glidepaths or McClung's Prime Harvesting (megathread at viewtopic.php?t=192105) are alternatives which are rule-based and can be backtested.

Prime Harvesting is essentially "spending cash as long as the market is down" (though it's bonds, not cash). But you have to tolerate the possibility of getting to 100% stocks in retirement, which is kinda terrifying

OK, so how about this:
50% stock
50% bonds
Spend from whichever is higher.
No balancing other than what happens due to withdrawals.
Allocation will drift.

I believe this is a variation of McClung's Prime Harvesting.

[bonesly]

You did not specify any time limit in your original statement, so I just assumed "infinite" as the time limit and not 2 years nor 5 years nor even some other time limit.

Why assume infinite when the OP topic is about cash buckets in a down market (and thus how long it take to refill said bucket when the market "recovers enough")?

As an aside, the Trinity study on safe withdrawal rates did not assume any cash bucket strategy, so I agree with @alex_686 and @nispirius that use of such a strategy is a psychological comfort (probably at the cost of a cash-drag on performance).

[alex_686]

I argue that it can have the benefit tangible benefit of lowering Sequence of Return Risk (SORR). ... Sure ok. But as Kitces notes looking at 10 year returns (1926-2017): ... I think that SORR (Sequence of Return Risk) actually does build an arguement for holding some less volatile assets from which to spend. It doesn't necessarily have to be cash but a cash-like liquid assets such as bonds or money market funds or stable value funds could help mitigate the risk. Some put annuities in there too. ... So you have to find a way to not run out of money. Holding too much low returning cash won't help. Neither will spending out of volatile assets in downturn.

I think you post is off base and I am of the strong opinion that you are wrong, but it is a fascinating subject. On that, here is a CFA Institute conference on this subject.

https://rpc.cfainstitute.org/en/researc ... emium-2021

Do starting valuations matter? Yes, I do believe that. Lots of nuances here. They indicate future returns. So what? There are two interoperations.

The first is that they indicate the current economic structure. For example, currently the S&P 500 P/E ratio is around in the mid 20s. That is the new and stable valuation forward. The is no mean reversion. That means the real returns going forward will be around 4%. Starting valuations matter, that is the starting value. You may not like the starting value but that is it.

If this interpretation is true the logical course would be to invest now.

The second is that there is mean reversion. Maybe there will be a full reversion, with the S&P P/E ratio going into the mid-teens. Maybe there will be a half reversion, with the S&P P/E going to the 20s. Maybe the reversion will be quick, occurring over 3 years. Maybe slow, over 10 years.

If the second interpretation is true what should you do? Well, you could overweight bonds until the market corrects. Except that market signals and mean reversion are weak. Historically, this has not been a winning strategy for larger institutional accounts who have tried this. They have held more bonds and got lower returns.

But this is all window dressing, because you are arguing about SORR. Risk metrics are about the same regardless of valuations. The odds of a down market are about the same if the P/E ratio is historically low, medium, or high.

So investing only during favorable valuations doesn't solve this problem. Waiting to invest only during favorable valuations means more time in bonds and probably lower returns.

The real answer for SoRR is to accept lower returns and invest in more bonds. Or, in my case, investing along the Low Volatility and Quality factors.

[alex_686]

History says otherwise. The broad stock market is more likely to go up after it drops than to go down more.

Tell me more. What statistical tests and methodology are you using?

[Triple digit golfer]

Why not keep the AA steady? That is, if you're goal is 50/50 and a market downturn brings you to 46/54, spend from the 54 until you're back at 50/50.

This will have you spending cash or bonds when equities are down, and vice versa, thus always selling high, at least relative to your other holdings.

[GaryA505]

Why not keep the AA steady? That is, if you're goal is 50/50 and a market downturn brings you to 46/54, spend from the 54 until you're back at 50/50.

This will have you spending cash or bonds when equities are down, and vice versa, thus always selling high, at least relative to your other holdings.

That was my question a couple of posts above. You may never get back to 50/50 if your spending isn't high enough, but maybe that's OK.

[Claudia Whitten]

This is why I think a low allocation to stocks makes sense in retirement. I have enough cash and cash-like instruments to last the rest of my life. I also have 30 percent in equities. If the market is up, the 30 percent will help my portfolio beat inflation. If it's down, my low percentage in equities won't cause me to starve or panic-sell. I certainly wouldn't gamble my retirement on the hope that my remaining years on this earth will see a rising stock market.

In retirement, the goal is asset preservation, for me, not accumulation or outperformance, and the primary benchmark against which I measure success is inflation, not what the S&P or some other index does. I don't see large allocations to risk assets as being conducive to good sleep or a worry-free retirement, personally.

https://seekingalpha.com/article/303663 ... r-retirees

[Triple digit golfer]

Why not keep the AA steady? That is, if you're goal is 50/50 and a market downturn brings you to 46/54, spend from the 54 until you're back at 50/50.

This will have you spending cash or bonds when equities are down, and vice versa, thus always selling high, at least relative to your other holdings.

That was my question a couple of posts above. You may never get back to 50/50 if your spending isn't high enough, but maybe that's OK.

You will if you have rebalancing bands. Let's say you have a 5% band. If market drops to 46/54, you spend from the 54. If it drops to 45/55, you rebalance to 50/50, and spend evenly from there.

[dbr]

You did not specify any time limit in your original statement, so I just assumed "infinite" as the time limit and not 2 years nor 5 years nor even some other time limit.

Why assume infinite when the OP topic is about cash buckets in a down market (and thus how long it take to refill said bucket when the market "recovers enough")?

As an aside, the Trinity study on safe withdrawal rates did not assume any cash bucket strategy, so I agree with @alex_686 and @nispirius that use of such a strategy is a psychological comfort (probably at the cost of a cash-drag on performance).

I tend to look at schemes to "not sell when stocks are down" as being psychological on the face of it due to absence of actual analytical sense to it. For myself investments are sufficiently volatile by nature, stocks or bonds, that the plan assumes money is withdrawn lots of times when things are "down." It is just how it works and a non-problem.

[GaryA505]

Why not keep the AA steady? That is, if you're goal is 50/50 and a market downturn brings you to 46/54, spend from the 54 until you're back at 50/50.

This will have you spending cash or bonds when equities are down, and vice versa, thus always selling high, at least relative to your other holdings.

That was my question a couple of posts above. You may never get back to 50/50 if your spending isn't high enough, but maybe that's OK.

You will if you have rebalancing bands. Let's say you have a 5% band. If market drops to 46/54, you spend from the 54. If it drops to 45/55, you rebalance to 50/50, and spend evenly from there.

True, but you don't really need the rebalancing. Just let it drift!

[dbr]

A comment to add to this is that if an investor wants to isolate his withdrawals for spending from the volatility of investments there are tools to do that. Two of those would be to buy an SPIA or to set up a long TIPS ladder. But again showing that either has a clear cut substantive effect on outcomes is not easy.

[brightlightstonight]

This is why I think a low allocation to stocks makes sense in retirement. I have enough cash and cash-like instruments to last the rest of my life. I also have 30 percent in equities. If the market is up, the 30 percent will help my portfolio beat inflation. If it's down, my low percentage in equities won't cause me to starve or panic-sell. I certainly wouldn't gamble my retirement on the hope that my remaining years on this earth will see a rising stock market.

In retirement, the goal is asset preservation, for me, not accumulation or outperformance, and the primary benchmark against which I measure success is inflation, not what the S&P or some other index does. I don't see large allocations to risk assets as being conducive to good sleep or a worry-free retirement, personally.

https://seekingalpha.com/article/303663 ... r-retirees

A low allocation to stocks trades one form of risk (stock volatility) for another form of risk (longevity risk).

A sufficiently large portfolio (and SS/pensions/etc.) mitigates both risks. If you have sufficient assets for that to work, that's great. But then again, if you have sufficient assets any plan "works" (from a strictly numeric perspective), so choosing the one that lets you sleep better is an excellent tradeoff.

[Random Poster]

Why not keep the AA steady? That is, if you're goal is 50/50 and a market downturn brings you to 46/54, spend from the 54 until you're back at 50/50.

This will have you spending cash or bonds when equities are down, and vice versa, thus always selling high, at least relative to your other holdings.

Did that plan work when both stocks and bonds were down?

There seems to be some belief that when stocks (or bonds) decrease, bonds (or stocks) increase.

But that belief should have been eviscerated over the past few years.

[dbr]

Why not keep the AA steady? That is, if you're goal is 50/50 and a market downturn brings you to 46/54, spend from the 54 until you're back at 50/50.

This will have you spending cash or bonds when equities are down, and vice versa, thus always selling high, at least relative to your other holdings.

Did that plan work when both stocks and bonds were down?

There seems to be some belief that when stocks (or bonds) decrease, bonds (or stocks) increase.

But that belief should have been eviscerated over the past few years.

Yes, it works. It is normal for both stocks and bonds to sometimes both be "down - whatever that means" or "up." The process rolls along and does what it does. The results are what you see in any withdrawal rate model, such as, for example, the historical results here:

https://engaging-data.com/visualizing-4-rule/

Sometimes one is spending from a portfolio that as a whole is "down" or is "up" Since withdrawal rate is the driver of success in these models one can refer to spending models that adjust spending according to portfolio value*. Asset allocation is not a driver of success in these models so messing around with asset allocation doesn't do much.

*see VPW etc., or even in old models like FireCalc the Bernicke and Cliatt rules.

[Wanderingwheelz]

History says otherwise. The broad stock market is more likely to go up after it drops than to go down more.

Tell me more. What statistical tests and methodology are you using?

I’m using common sense. Would anyone risk their capital in the stock market if it didn’t have an upwards bias?