rental income vs paper portfolio returns vs SWR

[WageSlave]

I didn't want to derail the gamble $50k thread, so I started a new one.  Context:

My FIRE expenses will be around 45k.  I expect to make significantly more than that income-wise (I'm shooting for a 50% savings rate in FIRE in a normal year, with that dropping to no savings at worst in the rare bad years).

Doesn't that imply a 2x safety factor?  If you're using the oft-discussed 4% SWR as your basis, you're actually planning on a 2% SWR, right?

Yes, and no, in that order.  I will not have a SWR, as I will have rental income for my FIRE income.  There will be no withdrawals from a traditional portfolio.

Isn't that kind of a semantic distinction?

I assume by "traditional" portfolio, you mean a "paper" portfolio, i.e. stocks and bonds.  With the traditional portfolios, it's typical to talk about "total" return, meaning both dividends and capital gains.  I've been under the impression that even if one's SWR is less than the dividend return, it's still called a withdrawal rate.  I.e., even if you can entirely fund your retirement from dividends, and never have to sell, it's still called SWR.

So for a rental property portfolio, how is that income different from bond interest or stock dividends?  I guess it's harder (impossible?) to automatically re-invest rental income (compared to the ease at which you can typically DRIP stock dividends or bond fund interest).

Not only that, how do I determine my SWR for a rental?  Based on equity? Total value of the property?  How do I determine that?

My "SWR" will be something silly like 6-12% depending on how you calculate it.

It's more than just a semantic distinction, it's categorically different.

These are honest questions, arebelspy, please don't think I'm trying to call you out or something!

Anyway: I would assume you would base your SWR of a rental the same way you would any investment: by its market value.  Granted, with real estate it's not as easy as going into Google Finance and looking up the current price.  But I would assume you generally have a reasonable idea of the market values of your properties, right?  If they are recently acquired, you could probably just use the purchase price.  For my own rental property, I use the purchase price, though that happened almost 10 years ago (not sure if this is right, it's just easiest :)).

Equity versus property value: that same question would apply to buying paper investments with leverage.  Naturally, your returns (or losses!) always look bigger with leverage.  There are leverage ETFs you can buy; what is their return?

With the exception of market pricing for real estate not being trivial to determine, I guess I don't see how rental income is categorically different than interest on individual bonds (bonds possibly purchased on leverage).

Are the definitions for these things simply done by convention, or am I missing some rationale?

Consider a simple scenario.  I have $1mm cash to invest, and my options are:

• a bond that pays 5% (bought for $1mm cash at face value, no leverage) = $50k/year cash
• a rental property that costs $1mm, and generates $50k/year in rent (paid in cash)

If I live on $25k/year, with either portfolio, isn't my SWR the same?  And exactly what is my SWR?  I would just say it's 2.5%.

I guess another difference is that it's hard or impossible to sell "part" of a rental property, whereas if you're buying index funds, it's relatively easy to sell some shares.  But, there is some nuance there, too: if you own a multi-unit rental, you could sell individual units.  And even if you're dealing in single family properties, as long as you own more than one, you can still sell part of your portfolio by selling individual properties (you have a few huge shares, as opposed to lots of small shares).

Thoughts?

[arebelspy]

In that case, aren't you just setting SWR to ROI?

(I'm assuming by "market value" you actually mean "equity".)

Say I make 75k/yr on properties with market value of $625k.

You'd say then that my SWR is 12%?

I say at that point it's nonsensical to even call it a SWR.  It would just impede understanding, rather than enhance. We all understand what SWR means, and when you then say that you have a 12% SWR you'll just confuse people.

How about "I live on rental income with a 12% ROI"?

I do think it's categorically different than the bond scenario.  It also has other factors that a bond doesn't have (the house is likely appreciating, unlike the bond principal, rents are likely to rise, unlike the bond's annual payout, etc. etc.).

The discussion continued in that thread, and frugaldoc compared it to an annuity, and I responded:

I would consider a rental more like an inflation adjusted annuity (expected income - expenses) where you get your principal back when you sell. Those assets then could be converted to paper and 4% or whatever rule applied.

Yes, it's much more similar to that than a traditional portfolio, IMO.

So if someone had a COLA'd annuity to cover all their expenses, what would their SWR be?

Like I said:

I will not have a SWR, as I will have rental income for my FIRE income.  There will be no withdrawals from a traditional portfolio.

I don't think someone with a COLA'd annuity that covers all their expenses has a SWR.

Rental income is more like owning a business that pays you - it doesn't make sense to talk about the SWR of a business that passes its profits to you, the owner.

[Baylor3217]

What does "the cheat is a millionaire" mean?  Is it from a movie or book?

[TomTX]

In that case, aren't you just setting SWR to ROI?

(I'm assuming by "market value" you actually mean "equity".)

Say I make 75k/yr on properties with market value of $625k.

You'd say then that my SWR is 12%?

I say at that point it's nonsensical to even call it a SWR.  It would just impede understanding, rather than enhance. We all understand what SWR means, and when you then say that you have a 12% SWR you'll just confuse people.

How about "I live on rental income with a 12% ROI"?

Questions:
Is that $75k/year after you have set aside funding for major maintenance items, such as replacing the roof or HVAC?
Have you subtracted an "effective wage" for the time you spend managing the properties/performing repairs/etc?

[arebelspy]

What does "the cheat is a millionaire" mean?  Is it from a movie or book?

http://www.homestarrunner.com/sbemail87.html

Questions:
Is that $75k/year after you have set aside funding for major maintenance items, such as replacing the roof or HVAC?
Have you subtracted an "effective wage" for the time you spend managing the properties/performing repairs/etc?

1) Yes.

2) Time spent is part of that, but it will all be managed by PMs.  I will be paid for the time, so it won't be counted in the ROI, but it is counted in the amount I'll have to spend.  In other words, the time I'll manage the PMs will be paying me (and not part of the rental return), but that pay is lumped in with the rest to come up with that number.  If I take over management at some point, the 75k would shoot higher by about 18k annually.  (But that would obviously be a return on labor, not capital, so again not counted as part of the rental ROI, but a wage for doing "work").

Not sure how it's relevant to the original discussion, but hope that answers those questions.  :)

[Eudo]

I think he's trying to compare apples to apples.

I think you just have to think what a SWR is supposed to be. It's the rate you can pull out cash and still have a reasonable chance of not failing. In general, if you have a ROI that will keep rising with inflation, I think the ROI is a baseline for your safe withdrawal rate. To be conservative, you need to discount it a bit to take into account the fact that the future is uncertain, and your actual experience might not track inflation, etc.

The way I look at it, if you have a diversified portfolio of real estate and you're receiving rent that you expect to roughly track inflation, then your SWR is probably close to (but less than) your ROI (subtracting all maintenance, taxes, allowances for future vacancies, etc). You don't expect a lot of capital appreciation beyond inflation, but you get a predictable cash stream.

If you own stocks, your dividends will probably roughly track inflation (with notable historical exceptions). But dividend rates are low. Additionally, depending on the kind of stocks you own, you probably expect appreciation beyond inflation. But that's much more uncertain, so you keep your SWR fairly low. But the benefit is that you expect to have more growth than a real estate portfolio, so there's a good chance you'll be able to draw down more in the future. Plus you don't have what many consider to be the headache of property management.

[arebelspy]

Yeah, I think you're just trying to cram a situation that doesn't really apply into a preconceived box.

But if you'd rather say someone has a 12% SWR just to be able to use the term "SWR" instead of "ROI," I'm okay with it.

[MDM]

From http://www.bogleheads.org/wiki/Safe_withdrawal_rates (highlighting added): "A safe withdrawal rate [SWR] is defined as the quantity of money, expressed as a percentage of the initial investment, which can be withdrawn per year for a given quantity of time, including adjustments for inflation, and not lead to portfolio failure; failure being defined as a 95% probability of depletion to zero at any time within the specified period."

SWR is a statistical thing, perhaps best used in the denominator with spending in the numerator to determine a savings target for FI. 

Your actual withdrawal rate (AWR or simply WR) is whatever you actually do remove from your investments.  This implies your investments are liquid enough that you can convert them to cash to pay for your expenses. 

E.g., if you had 10 $100K rental properties that were all vacant, you could sell one, invest $70K in the market, and use $30K for expenses and have a 3% WR that year.  If your rentals are providing more cash flow than you need for expenses, you can invest the extra and your WR is negative.

[Eudo]

Fair enough. We can define things however we want, as long as we're all clear on what we mean.

I'd argue it's a useful concept to have a word for, though. I mean if you know what your costs per year are going to be going forward and want to know how much you need in a lump sum to support that lifestyle, you take your costs and multiply by a factor. In my book, the number you multiply by is 1/SWR.

If you really believe you can earn 12% a year forever after inflation, then you need 8.3 times your annual living costs invested.

The reason it's so low is because your ROI is so high. So the real question is why is your ROI so high (higher than the returns from a typical safe purely passive investment). The answer is presumably because it's not truly passive. It's what many others would call a job. Nothing wrong there. But if you wanted to calculate the true SWR for a diversified portfolio of real estate you'd want to assume you were truly passive and we're doing no work, which is why I think TomTX wanted to subtract out all costs including what it would cost to hire a management company. That typically brings the high real estate ROIs down to more reasonable levels.

[arebelspy]

From http://www.bogleheads.org/wiki/Safe_withdrawal_rates (highlighting added): "A safe withdrawal rate [SWR] is defined as the quantity of money, expressed as a percentage of the initial investment, which can be withdrawn per year for a given quantity of time, including adjustments for inflation, and not lead to portfolio failure; failure being defined as a 95% probability of depletion to zero at any time within the specified period."

(Emphasis original.)

Thanks for illustrating my point - that description doesn't fit with rental income.

Let's say I buy a property for 100k, putting 25k down and get a 30 year mortgage.  It rents for 1500/mo.  30 years later rents have doubled (so now 3000/mo), the house is paid off.  I net 18k annual.  My initial investment was 25k.   Would you say my SWR is 72% then?  Of course not.

We already have a term for it.  ROI.  Why do we need to also call it SWR?  They are two completely different things.

[skyrefuge]

Rental income is more like owning a business that pays you....a business that passes its profits to you, the owner.

Um, you know you just described stock ownership there too, right?  :-)

I think it's quite sensible and useful to apply the concept of SWR to any portfolio of assets. The FIRE simulators don't traditionally include real estate as one of the assets you can select, but that's mainly because it's hard to include the historical data, not because real estate is so "categorically different" that it can't be included (stocks are "categorically different" from bonds, after all, and they're both included as options).

Playing with the simulators shows that the safe withdrawal rate varies depending on your asset allocation. So if a portfolio that consists of 100% directly-owned real estate can produce a SWR that's also different than any portfolio of stocks and bonds, there's nothing weird or unexpected about that. I don't think that's a reason to say "SWR + real estate makes no sense".

Rather, placing it in the SWR context raises interesting and valuable questions: if the SWR from a real estate portfolio can really be 3x that from a stocks/bonds portfolio, why is that? Is it a clue the SWR is being overestimated, and an indication that the plan is riskier than believed? If not, then it's a really strong argument in favor of directly owning real estate: "if you invest in real estate, you can safely retire with 2/3rds less assets than if you invest in stocks/bonds!"

The simulators show us that a higher allocation to stocks improves the SWR (over longer terms), so it would be pretty useful to know that real estate does an even better job of raising the SWR.

Perhaps with your original quote in this thread, you're already allowing for the possibility (consciously or subconsciously) that your SWR is lower than your ROI? By spending only 50% of your ROI (your WR), are you protecting yourself from ROI volatility in the future? Similar to how 40% growth in the stock market "protects" you against future 30% drops, and produces an SWR far lower than 40%? If so, then that 50% of ROI is what you consider your SWR. If not, if you think you'd be fine spending 100% of your rental income and are simply saving half of it for unrelated purposes, then your SWR is equal to your ROI.

[WageSlave]

I think, especially after reading MDM's post, that I use the term "SWR" too loosely.  I've always looked at it like a black box: you have an investment portfolio worth X, and you take out Y per year to cover expenses.  Here the investment portfolio could be whatever (rentals, stocks, bonds, tulips), and the cash flow comes either directly from rents/interest payments or selling assets.

Let me ask this question: what if you had half rentals and have paper portfolio?

In that case, aren't you just setting SWR to ROI?

Not necessarily.  What if your ROI is much greater than the amount you need to live?  Say your net rental income (after maintenance, fees, taxes, etc) is $50k/year, but you only need $25k/year to live.  Clearly ROI isn't equal to SWR.  What do you call the $25k number in relation to your portfolio?

Say I make 75k/yr on properties with market value of $625k.

You'd say then that my SWR is 12%?

Yes, but only if you spend the full $75k.

But if you spend less than that, then the remainder basically goes back into your portfolio as an allocation (cash or whatever you re-invest into, e.g. capital improvements on existing property, more properties, or traditional paper investments).

I do think it's categorically different than the bond scenario.  It also has other factors that a bond doesn't have (the house is likely appreciating, unlike the bond principal, rents are likely to rise, unlike the bond's annual payout, etc. etc.).

You can theoretically get closer to real estate-like parameters with something like a TIPS ladder.

I don't think someone with a COLA'd annuity that covers all their expenses has a SWR.

Agreed, or even a COLA'd pension for that matter... or (for non-early retirees) social security, if it covers all expenses.

Rental income is more like owning a business that pays you - it doesn't make sense to talk about the SWR of a business that passes its profits to you, the owner.

Owning (dividend-paying) stocks is (partial) ownership in a business that pays you.

I think I see where you're coming from: let's say I buy a McDonald's franchise, but hire a manager to do all the real work.  I just sign off on "big" decisions, but get a cut of the profits.  This is not unlike a rental situation, agreed?  So if I get enough franchise profits to entirely cover my expenses, I don't really have a "SWR".

But let's say I go in with a group of people, form something like an LLC, and the LLC buys a franchise.  Again, we hire a manager to oversee all the operational work.  But profits from this franchise are paid to the LLC, which in turn are divided up among the LLC partners.  Let's say I do this several dozen times, each with a different group of people.  In aggregate, all these partnerships pay enough to cover my expenses.  Is it a SWR or no?  But from my perspective, this arrangement is starting to look more and more like holding a handful of (dividend paying) stocks.  Where is the category distinction?

Where I'm coming from: my ideal FIRE plan is that my "SWR" is low enough that I never actually have to sell any paper investments.  In other words, I want the cash flow from dividends and interest to cover all expenses (with a little extra for cushion).  From a block-box perspective, I think this approach is similar to yours: I have an underlying investment portfolio that should keep pace with inflation and generates some cash.  So would you say that "SWR" isn't an appropriate term for me, either?

MDM posted while I was composing this...

Your actual withdrawal rate (AWR or simply WR) is whatever you actually do remove from your investments.  This implies your investments are liquid enough that you can convert them to cash to pay for your expenses. 

E.g., if you had 10 $100K rental properties that were all vacant, you could sell one, invest $70K in the market, and use $30K for expenses and have a 3% WR that year.  If your rentals are providing more cash flow than you need for expenses, you can invest the extra and your WR is negative.

So, yeah, I've been using "SWR" when I actually mean "[A]WR".

I added the emphasis above to ask the question, what counts as "removing"?  :)  In a stock portfolio, if you use dividend payments to cover your living expenses, is that considered "removing"?  In my mind, anything that inhibits the natural growth of an investment would be removing.  So, not doing dividend re-investment would fall into this category: clearly, all things being equal, the DRIP'ing stock portfolio will grow more that the one that doesn't DRIP.  Same goes for a rental: the rental income could be similarly re-invested, either as down payment for more rentals, or capital improvements on existing rentals, etc.

I'd argue it's a useful concept to have a word for, though. I mean if you know what your costs per year are going to be going forward and want to know how much you need in a lump sum to support that lifestyle, you take your costs and multiply by a factor. In my book, the number you multiply by is 1/SWR.

This is kind of what I was getting at: what portfolio value are you shooting for in relation to your expenses (i.e. "what's your number?").  But maybe, you don't look at it that way?  Maybe you're coming at it from a purely cash flow perspective?

Still seems like a semantic issue to me.  :)

[arebelspy]

Rather, placing it in the SWR context raises interesting and valuable questions: if the SWR from a real estate portfolio can really be 3x that from a stocks/bonds portfolio, why is that? Is it a clue the SWR is being overestimated, and an indication that the plan is riskier than believed? If not, then it's a really strong argument in favor of directly owning real estate: "if you invest in real estates, you can safely retire with 2/3rds less assets than if you invest in stocks/bonds!"

IMO one major factor is because there is a lot less sequence of return risk.

Let me ask this question: what if you had half rentals and have paper portfolio?

Then you subtract the rental income from your expenses, and the remainder of your expenses divide it by your paper portfolio to get your SWR.

In that case, aren't you just setting SWR to ROI?

Not necessarily.  What if your ROI is much greater than the amount you need to live?  Say your net rental income (after maintenance, fees, taxes, etc) is $50k/year, but you only need $25k/year to live.  Clearly ROI isn't equal to SWR.  What do you call the $25k number in relation to your portfolio?

Your withdrawl rate.

I think I see where you're coming from: let's say I buy a McDonald's franchise, but hire a manager to do all the real work.  I just sign off on "big" decisions, but get a cut of the profits.  This is not unlike a rental situation, agreed?  So if I get enough franchise profits to entirely cover my expenses, I don't really have a "SWR".

But let's say I go in with a group of people, form something like an LLC, and the LLC buys a franchise.  Again, we hire a manager to oversee all the operational work.  But profits from this franchise are paid to the LLC, which in turn are divided up among the LLC partners.  Let's say I do this several dozen times, each with a different group of people.  In aggregate, all these partnerships pay enough to cover my expenses.  Is it a SWR or no?  But from my perspective, this arrangement is starting to look more and more like holding a handful of (dividend paying) stocks.  Where is the category distinction?

Again it has to do with sequence of return risk.  The reason why the Trinity study has a SWR of 4% when stocks have typically returned 7 real is due to that.  If the income from your franchise is stable, and will remain so, you can withdraw more, due to the high return having less volatility.  Ditto real estate.

Where I'm coming from: my ideal FIRE plan is that my "SWR" is low enough that I never actually have to sell any paper investments.  In other words, I want the cash flow from dividends and interest to cover all expenses (with a little extra for cushion).  From a block-box perspective, I think this approach is similar to yours: I have an underlying investment portfolio that should keep pace with inflation and generates some cash.  So would you say that "SWR" isn't an appropriate term for me, either?

If the dividends are stable, and you can fairly confidently project they will be so and cover your income even in a 2008 situation, and it increases at or above inflation, yeah, I wouldn't be sweating over what exactly your SWR is.

So, yeah, I've been using "SWR" when I actually mean "[A]WR".

Right.

I'd argue it's a useful concept to have a word for, though. I mean if you know what your costs per year are going to be going forward and want to know how much you need in a lump sum to support that lifestyle, you take your costs and multiply by a factor. In my book, the number you multiply by is 1/SWR.

This is kind of what I was getting at: what portfolio value are you shooting for in relation to your expenses (i.e. "what's your number?").  But maybe, you don't look at it that way?  Maybe you're coming at it from a purely cash flow perspective?

Still seems like a semantic issue to me.  :)

No, the number is irrelevant to me, if I FIRE with 2MM or $4 (or a negative net worth).  If I get stabilized cash flow that will remain for a long time, I'm good with that.

[skyrefuge]

Would you say my SWR is 72% then?

No, but not because we just throw up our hands due to the large value of the number.

SWR doesn't change over time. It's a number that is calculated (estimated) at the start of retirement, and does not change. The increase in rent and paying off of the mortgage are all factors that are projected (or taken from historical data) and baked into that original SWR calculation.

[arebelspy]

Would you say my SWR is 72% then?

No, but not because we just throw up our hands due to the large value of the number.

SWR doesn't change over time. It's a number that is calculated (estimated) at the start of retirement, and does not change. The increase in rent and paying off of the mortgage are all factors that are projected (or taken from historical data) and baked into that original SWR calculation.

Right.  So I'm working.  I buy a rental.  I spend everything (live paycheck to paycheck).  30 years later I retire when it's paid off, living on 18k annually from the rental.  (See above scenario for more detail).  What is my SWR that I calculate now at the start of retirement that does not change?  The BH definition has it as my cash flow (that rises with inflation) over initial investment.  There's obviously time value of money issues in there creating problems.

[beltim]

It seems to me that the major difference is the control of how profits are used.  Skyrefuge questions a 12% return on investment for real estate because that's three times the safe withdrawal rate of a diversified portfolio of stocks and bonds.  But that's comparing two different things.  The economic equivalence should be that the profitability of the companies in the stock index and the profitability of the real estate are the same. 

Stocks are not typically optimized for constant inflation-adjusted earnings.  Most companies seek to grow profits much more quickly than that, which exposes you to risk.  Because you have to sell at market prices, there's sequence of return risk that arebelspy mentioned.  If you could live off dividends, you could have a swr up to the dividend yield because you'd never touch the capital.  Unfortunately for this strategy, stocks pay out a much smaller percentage of profits as dividends than they used to: http://en.wikipedia.org/wiki/Dividend_payout_ratio

[WageSlave]

From http://www.bogleheads.org/wiki/Safe_withdrawal_rates (highlighting added): "A safe withdrawal rate [SWR] is defined as the quantity of money, expressed as a percentage of the initial investment, which can be withdrawn per year for a given quantity of time, including adjustments for inflation, and not lead to portfolio failure; failure being defined as a 95% probability of depletion to zero at any time within the specified period."

(Emphasis original.)

Thanks for illustrating my point - that description doesn't fit with rental income.

Let's say I buy a property for 100k, putting 25k down and get a 30 year mortgage.  It rents for 1500/mo.  30 years later rents have doubled (so now 3000/mo), the house is paid off.  I net 18k annual.  My initial investment was 25k.   Would you say my SWR is 72% then?  Of course not.

I think the only reason it doesn't fit is due to the use of leverage (the mortgage).  You would also have the same dilemma if you bought a stock portfolio on margin.  I suspect most MMM/Boglehead types aren't explicitly using leverage to buy stocks, but are doing so implicitly via the mortgage on their primary residence, i.e. buying more paper investments instead of paying down the mortgage.  The use of implicit leverage for buying paper investments is not uncommon, yet I doubt any of these people think about the problem you present.  Rather, the cost of their portfolio is the market price, and the ROI is based on that.  Going off that rational, I would use the "transaction price" as the basis for ROI on the rentals.  Obviously, it will be lower than the ROI based only on the down payment amount.  Over 30 years, however, you will have spent the transaction price plus interest on the property.  But what about inflation?  Maybe a compromise is to apply some kind of inflation (and maybe tax deduction) discount model to your 30 years' worth of payments, and use that as your "true" cost basis for calculating ROI.

IMO one major factor is because there is a lot less sequence of return risk.
[ ... ]
Again it has to do with sequence of return risk.  The reason why the Trinity study has a SWR of 4% when stocks have typically returned 7 real is due to that.  If the income from your franchise is stable, and will remain so, you can withdraw more, due to the high return having less volatility.  Ditto real estate.

Ditto TIPS ladder?

Anyway, high returns + no sequence of return risk = badass.  :)

Right.  So I'm working.  I buy a rental.  I spend everything (live paycheck to paycheck).  30 years later I retire when it's paid off, living on 18k annually from the rental.  (See above scenario for more detail).  What is my SWR that I calculate now at the start of retirement that does not change?  The BH definition has it as my cash flow (that rises with inflation) over initial investment.  There's obviously time value of money issues in there creating problems.

Not sure about the BH definition, but how is it any different if the only thing that changes in your scenario is that instead of buying a rental over 30 years, you buy a TIPS ladder?

What if your ROI is much greater than the amount you need to live?  Say your net rental income (after maintenance, fees, taxes, etc) is $50k/year, but you only need $25k/year to live.  Clearly ROI isn't equal to SWR.  What do you call the $25k number in relation to your portfolio?

Your withdrawl rate.

OK, now I'm happy.  :)

[South african]

Hi.  This is my first comment on this forum even though I have been a Mr mm reader for a few years. I have considered if listed REIT investments is not the ideal investment for early retirees. These are my reasons:
1) It pays out rental income from a diversified property holding (office, commercial and Retail) without any of the hassles that comes with owning direct property.
2)The larger companies is easily traded therefore it is liquid.
3) the starting rental yield is usually higher than dividends
4) the rental income usually increases by inflation if you hold a diversified holding of REIT.
5) the rental income is usually a lot more stable than the share price at which these companies trade.

Therefore if you cover your living expenses with rental income from REIT then you kind of have met your SWR requirements. If the retal yield is 5% then you need to invest less than the normal 4% and 3% theory of SWR because you never need to sell any assets to fund your expenses if you live off the rental income from the ReIt alone.  It is a special kind of dividend investor. 

Of course it would help to have a safety margin if rentals do stagnate or vacancies increase, but the impact of these events on rental income is not as severe as market crashes on a sell assets to live scenario.

Anyways just wanted to throw my thoughts into this forum and see if my reasoning is sound.

[2527]

I think the concept of SWR really only makes sense within the context of a portfolio of stocks, bonds, or mutual funds of stocks and bonds.  The thinking being:  in addition to dividends, you can sell as certain amount of the underlying investment to achieve your SWR, because 1) it is possible, and 2) the underlying investment is in general, over time, increasing in value.

With a physical real estate investment or business, you get, presumable, and income stream net of expenses and provisions for maintenance and repairs.  You can't sell 1-2% of an apartment building, or laundromat, or McDonald's franchise every year.

[skyrefuge]

What is my SWR that I calculate now at the start of retirement that does not change?  The BH definition has it as my cash flow (that rises with inflation) over initial investment.

No, in the BH definition "initial investment" refers to the value of the investment at the initiation of retirement, not at the initial purchase of the asset. In other words, it's the current (at-retirement) value of your portfolio, which in this case would simply be the current market value of your property. There is no relation to the past, just as cost basis has no part in the SWR calculation for a stock portfolio.

Next, the amount of cash flow generated by your property is actually unrelated to SWR. Your Safe Withdrawal Amount (SWA) might be less than or greater than your cash flow. It's simply the maximum amount that you can keep spending each year into the future (inflation-adjusted) without depleting your portfolio.

If that SWA is less than your cash flow, then you'll end up with unspent cash each year. Note that this is not "extra" cash. It likely must be used in some productive way (purchasing more assets), because the SWR calculation you made at the start of retirement assumed that it would be used productively (perhaps history showed that the underlying assets did not keep up with inflation, and thus additional asset acquisition was needed). If it was truly "extra" cash that didn't need to be reinvested, then your SWA/SWR would have been higher to begin with.

If that SWA is more than your cash flow, then you'll have to do some tricks (reverse mortgage?) to increase your cash flow to pay your expenses, since you can't just "sell a few shares" of your property like you can with stocks. Alternatively, you could just spend only your cash flow, but then your withdrawal amount would be even safer than necessary, and you would be leaving asset appreciation on the table.

In short, your SWR is a rough, generalized projection, not tied to a particular house, stock, or bond, so it isn't calculated based on specific transactions you have made.  According to cFIREsim, the 30-year, 95%-success SWR for a portfolio of 100% stocks is 3.8%. For 100% bonds it's 2.6%. For 100% gold it's 1.8%.

What is it for real estate? I don't know. You probably have a better idea than I do. I would guess that it's at least as high as the 3.8% for stocks (with, as you note, the lower sequence of returns risk helping a lot).

So the point would be, if someone can determine a rough SWR for a portfolio of 100% real estate, then that would give someone a starting point to say "ok, my expenses are $20k per year, so I need to accumulate $X in real estate assets before I can retire". And it could also be useful to landlords already in retirement who find themselves with a cash flow much greater than the SWR suggests. It could be an indication that some of that cash needs to be saved in order to survive an upcoming real estate crash.

[arebelspy]

I agree with everything you said, skyrefuge, so I'm skipping over that, and just want to address the last paragraph:

So the point would be, if someone can determine a rough SWR for a portfolio of 100% real estate, then that would give someone a starting point to say "ok, my expenses are $20k per year, so I need to accumulate $X in real estate assets before I can retire". And it could also be useful to landlords already in retirement who find themselves with a cash flow much greater than the SWR suggests. It could be an indication that some of that cash needs to be saved in order to survive an upcoming real estate crash.

To me this is the tail wagging the dog.  Figuring out an SWR for a portfolio of real estate, then calculating how much you need to save to hit that level of total portfolio based on your expenses is backwards.  Instead figure out how much cash flow you need to cover your expenses and how many properties (and what they'll cost) to get that.  Then you can calculate the SWR, if you want, though it won't really mean anything.

It amounts to the same thing, we're just coming at it from different sides, but it makes sense to me to calculate how much cash flow you need, rather than try to calculate some mystery SWR that no one is sure about.  :)

[skyrefuge]

Instead figure out how much cash flow you need to cover your expenses and how many properties (and what they'll cost) to get that.

That works, as long as the cash flow from your real estate and your SWA are very close to each other. But if your cash flow is only 1% of your market value and your SWR is 5% (implying that real estate is more like "growth stocks", and the majority of the return comes from appreciation rather than dividends), then doing it your way would cause the retiree to accumulate far more (unleveraged) assets than they actually need, thus unnecessarily delaying their retirement.

This is the same issue that people have who want to take a stock-dividends/bond-interest-only approach for retirement income. In today's environment, a total-return approach is necessary in order to generate an income that approaches your SWR, and people who refuse to sell assets will be saving more (and working longer) than they need to. A large part of the FIRE philosophy is to not continue wasting time working when you already have "enough", so I think this is an important point.

Conversely, if your cash flow is something high, like 15% of market value and your SWR is 5%, then your method would have people spending all their cash when it actually needs to be reinvested to prevent ruin.

Presumably you (and other real estate investors) have already made the determination that cash flow and SWA end up fairly close to each other, so that's why it seems like a bunch of unnecessary calculations from your perspective. But I bet if someone figured out a reasonable way to get historical real estate returns into cFIREsim, you'd be one of the first to try it out. :-)

[WageSlave]

I think the concept of SWR really only makes sense within the context of a portfolio of stocks, bonds, or mutual funds of stocks and bonds.  The thinking being:  in addition to dividends, you can sell as certain amount of the underlying investment to achieve your SWR, because 1) it is possible, and 2) the underlying investment is in general, over time, increasing in value.

With a physical real estate investment or business, you get, presumable, and income stream net of expenses and provisions for maintenance and repairs.  You can't sell 1-2% of an apartment building, or laundromat, or McDonald's franchise every year

But what if you own 100 apartment buildings/laundromats/franchises?  Then you could sell a small percentage of your portfolio.  Ok, so most of us won't have the means to buy 100s of rentals or franchises.   But what if we form partnerships and buy partial stakes in 100s of rentals/small businesses?  Then we could sell our partial ownership stakes, and easily do tiny percentages every year.  That sounds just like stocks, only it's private equity instead of public.

Or, let's give up the partial ownership thing: what if we invest in something like ATMs or vending machines?  You could easily have a portfolio of hundreds of these, especially if you built it up over 10 years or so.

[WageSlave]

Presumably you ... have already made the determination that cash flow and SWA end up fairly close to each other, so that's why it seems like a bunch of unnecessary calculations from your perspective...

But he's looking at cash flow that's twice his expenses:

My FIRE expenses will be around 45k.  I expect to make significantly more than that income-wise (I'm shooting for a 50% savings rate in FIRE in a normal year, with that dropping to no savings at worst in the rare bad years).

Doesn't that imply a 2x safety factor?  If you're using the oft-discussed 4% SWR as your basis, you're actually planning on a 2% SWR, right?

Yes, and no, in that order.  I will not have a SWR, as I will have rental income for my FIRE income.  There will be no withdrawals from a traditional portfolio.

So the spent amount isn't an SWR, but still a WR:

In that case, aren't you just setting SWR to ROI?

Not necessarily.  What if your ROI is much greater than the amount you need to live?  Say your net rental income (after maintenance, fees, taxes, etc) is $50k/year, but you only need $25k/year to live.  Clearly ROI isn't equal to SWR.  What do you call the $25k number in relation to your portfolio?

Your withdrawl rate.

ARS, once you FIRE, what will you do with your surplus cash flow?

This is the same issue that people have who want to take a stock-dividends/bond-interest-only approach for retirement income. In today's environment, a total-return approach is necessary in order to generate an income that approaches your SWR, and people who refuse to sell assets will be saving more (and working longer) than they need to. A large part of the FIRE philosophy is to not continue wasting time working when you already have "enough", so I think this is an important point.

But we all define "enough" differently: ARS is looking for 2x cash flow with his rental strategy (1x for spending and 1x for safety (I assume)).  I, as somebody looking at a dividend/interest-only approach, am looking for roughly 1x cash flow for expenses, but the other 1x (maybe better) for safety.

Going by the numbers ARS has suggested in this and other threads, his strategy will likely be a much quicker and more efficient way to achieve FIRE... and in fact, every time I read his posts I want to go out and get a big rental portfolio!

Changing gears on the topic a bit...

But I bet if someone figured out a reasonable way to get historical real estate returns into cFIREsim, you'd be one of the first to try it out. :-)

In today's big data world, I'd think historical real estate returns would be relatively easy to produce.  I'm going to speculate here, though: I suspect that the typical numbers would suggest that real estate isn't necessarily any better than traditional investments; on average, I bet they're not too dissimilar from stocks and bonds.  Doesn't mean you can't do better than the average though.  That's where I always end up when I start thinking about real estate: it's trivial to find mediocre deals, but to find the good deals requires some knowledge, patience and persistence.  IOW, I suspect ARS's rental performance is above average.  If nothing else, the impression I get from reading his posts is that he's put a reasonable amount of time into learning about the rental market; I don't think his rentals just fell into his lap.  ARS, would you agree?  Any "luck" you've had is the kind of luck that comes from hard work and dedication?  "Fortune favors the prepared" and all that?

Plus, from the perspective of funding a retirement: if building a rental portfolio was as easy as a BH-style index fund portfolio, and the rental returns were higher and didn't pose sequence-of-returns risk---why wouldn't everyone do it?  It's not like rentals are a new thing; it's probably the "second" oldest profession.  ;)  Certainly, there are risks associated with renting, and if you don't hire a PM, it starts to look more like a job than a passive investment.

Compared to "traditional" investment portfolios, are rentals...

• providing roughly the same returns?
• riskier, therefore demanding a higher returns?
• part of a much less efficient market?

Interestingly enough (ironically?) though I'm in a traditional portfolio, I'm still using the same thought process as ARS to derive my number, that is: what portfolio value X do I need to produce Y cash flow?  He happens to define X in terms of rental properties, and I define it in terms of index fund shares.  I call it an SWR even though it happens to be exactly the same amount of dividends/interest I'm looking for.  Is it a 4% SWR with a 2x safety margin, or simply a 2% SWR?

[2527]

One thing to keep in mind, from my dad's experience, and our experience with his estate.  All things being equal, stocks are easier to manage when you are winding down your life, and they are easier to handle in an estate. 

Also, an index fund is always staying in the "current" economy.  A piece of real estate may be in a neighborhood that goes downhill.   

[brewer12345]

You know, all of these quibbles would be moot if you would just put all your money in an equity indexed annuity...

[beltim]

You know, all of these quibbles would be moot if you would just put all your money in an equity indexed annuity...

The pinnacle of wealth management strategies?

[brewer12345]

You know, all of these quibbles would be moot if you would just put all your money in an equity indexed annuity...

The pinnacle of wealth management strategies?

So they tell me.  No downside, equity linked upside, what's not to like?

[beltim]

You know, all of these quibbles would be moot if you would just put all your money in an equity indexed annuity...

The pinnacle of wealth management strategies?

So they tell me.  No downside, equity linked upside, what's not to like?

No downside?  Your head must be in the clouds.

I'm himalay-rious.

[arebelspy]

Instead figure out how much cash flow you need to cover your expenses and how many properties (and what they'll cost) to get that.

That works, as long as the cash flow from your real estate and your SWA are very close to each other. But if your cash flow is only 1% of your market value and your SWR is 5% (implying that real estate is more like "growth stocks", and the majority of the return comes from appreciation rather than dividends), then doing it your way would cause the retiree to accumulate far more (unleveraged) assets than they actually need, thus unnecessarily delaying their retirement.

This is the same issue that people have who want to take a stock-dividends/bond-interest-only approach for retirement income. In today's environment, a total-return approach is necessary in order to generate an income that approaches your SWR, and people who refuse to sell assets will be saving more (and working longer) than they need to. A large part of the FIRE philosophy is to not continue wasting time working when you already have "enough", so I think this is an important point.

Conversely, if your cash flow is something high, like 15% of market value and your SWR is 5%, then your method would have people spending all their cash when it actually needs to be reinvested to prevent ruin.

Interesting. 

I can see the former scenario, but not the latter.  How would you determine that SWR?

[arebelspy]

So the spent amount isn't an SWR, but still a WR

Correct.  Because there's no way to know what the SWR is.  My WR will be about half my ROI.  But ask me in about 50 years if it was also a SWR.

ARS, once you FIRE, what will you do with your surplus cash flow?

Invest it.

But I bet if someone figured out a reasonable way to get historical real estate returns into cFIREsim, you'd be one of the first to try it out. :-)

In today's big data world, I'd think historical real estate returns would be relatively easy to produce.  I'm going to speculate here, though: I suspect that the typical numbers would suggest that real estate isn't necessarily any better than traditional investments; on average, I bet they're not too dissimilar from stocks and bonds.  Doesn't mean you can't do better than the average though.  That's where I always end up when I start thinking about real estate: it's trivial to find mediocre deals, but to find the good deals requires some knowledge, patience and persistence.  IOW, I suspect ARS's rental performance is above average.  If nothing else, the impression I get from reading his posts is that he's put a reasonable amount of time into learning about the rental market; I don't think his rentals just fell into his lap.  ARS, would you agree?  Any "luck" you've had is the kind of luck that comes from hard work and dedication?  "Fortune favors the prepared" and all that?

Absolutely agree, and well put. 

I think the real estate market isn't nearly as efficient as the stock market, thus the opportunity to find "deals" that may not exist (or if they do, to a value investor, are to a much smaller extent). The ability to take advantage of this takes lots of knowledge and hard work, however.

Plus, from the perspective of funding a retirement: if building a rental portfolio was as easy as a BH-style index fund portfolio, and the rental returns were higher and didn't pose sequence-of-returns risk---why wouldn't everyone do it?  It's not like rentals are a new thing; it's probably the "second" oldest profession.  ;)  Certainly, there are risks associated with renting, and if you don't hire a PM, it starts to look more like a job than a passive investment.

Compared to "traditional" investment portfolios, are rentals...

• providing roughly the same returns?
• riskier, therefore demanding a higher returns?
• part of a much less efficient market?

I think it goes back to the above answer - I'd expect most properties not bought right to underperform the market.  But a deal should outperform it.  A good one should be less volitile as well.

Again, it goes back to knowledge. If you don't know what the fuck you're doing, you can get in a lot of trouble with rentals, and get in way over your head.  People who don't know what they're doing buy at bad prices.  They get bad tenants.  They get bad contractors.  Bad property managers.  Etc. 

Since so many people dive in without education (ESPECIALLY since so many do it unwillingly via accidental landlording, and they don't go into it preparing and learning about it), there are enough horror stories out there to scare many people away.  Plenty of people won't go into it because they don't want to unclog toilets, when I never have, and never will. 

Interestingly enough (ironically?) though I'm in a traditional portfolio, I'm still using the same thought process as ARS to derive my number, that is: what portfolio value X do I need to produce Y cash flow?  He happens to define X in terms of rental properties, and I define it in terms of index fund shares.  I call it an SWR even though it happens to be exactly the same amount of dividends/interest I'm looking for.  Is it a 4% SWR with a 2x safety margin, or simply a 2% SWR?

This goes back to skyrefuges "enough" though - I think waiting until 2% dividends give you enough cash flow to retire leaves you with way more than is necessary.  But you may want it, in order to sleep at night.  And that's what it's all about - what someone is comfortable with.

[arebelspy]

You know, all of these quibbles would be moot if you would just put all your money in an equity indexed annuity...

Hah.  I laughed. 

Then BELTIM with the DOUBLE WHAMMY PUNS.

The pinnacle of wealth management strategies?

No downside?  Your head must be in the clouds.

I'm himalay-rious.

I'm not even going to bother with any more ("You're tops" type) that would just pale in comparison.

I tip my hat to you, sir.

[Shor]

If you consider the traditional portfolio, you have two types of appreciation:
1. Value. The underlying asset is worth more.
2. Dividend / Interest. A portion of the asset is split off and given over to you. In some cases, like a company this doesn't detract from the value of the underlying asset as teh company itself produces profit over time.

The implication in the SWR is that the expenses > Dividend output of your assets.
If you have so much in assets that your dividends stays sustainably higher than your expense, you would be able to pay everything without detracting from your "principal" amount. The goal of the SWR figure was to drawn down the portfolio and if the amount ended up above $0 it was a success. The 4% figure was the measure where 95% of all cases ended up above 0, even if it was a pittance. If you literally kept all of your principal then you supposedly retired way too late, and could have retired much earlier and still end up with a positive result at the end of your life cycle.

Some people might want to leave assets to their progeny, others might not want to see their life's work torn down piece by piece. Hell, I would feel a failure if I left the next generation with nothing to help pursue their dream.

This is the difference between the paper idea of SWR and the Actual Withdrawl Rate which people are comfortable with executing.

[arebelspy]

If you consider the traditional portfolio, you have two types of appreciation:
1. Value. The underlying asset is worth more.
2. Dividend / Interest. A portion of the asset is split off and given over to you. In some cases, like a company this doesn't detract from the value of the underlying asset as teh company itself produces profit over time.

The implication in the SWR is that the expenses > Dividend output of your assets.
If you have so much in assets that your dividends stays sustainably higher than your expense, you would be able to pay everything without detracting from your "principal" amount. The goal of the SWR figure was to drawn down the portfolio and if the amount ended up above $0 it was a success. The 4% figure was the measure where 95% of all cases ended up above 0, even if it was a pittance. If you literally kept all of your principal then you supposedly retired way too late, and could have retired much earlier and still end up with a positive result at the end of your life cycle.

Some people might want to leave assets to their progeny, others might not want to see their life's work torn down piece by piece. Hell, I would feel a failure if I left the next generation with nothing to help pursue their dream.

This is the difference between the paper idea of SWR and the Actual Withdrawl Rate which people are comfortable with executing.

Most of the time an SWR of 4% left more than what was started with, not the very low amounts you're implying.  Does that happen sometimes and be counted a success?  Sure.  But in general (historically) the market grows above 4%, so your 4% covers your spending plus inflation and often even grows above that.

[Nords]

I've never seen so much vocabulary tortured for so little purpose, even in a nuclear reactor plant manual.

But I'm glad that Brewer could make me laugh about it!

[MidWestLove]

The piece where rental feels so different to me vs paper assets like stocks/bonds is that in rental you carry the actual liability. If something happens inside , the person can always come back to you with claim that you 'knew or should have known' ... imagine this scenario as a stock owner , at most you are on the hook for is the value of the stock. legal liability, maintenance, responsibility to respond to complain (or be summoned to court) , deal with police if something is happening on your property - all very different from truly passive ownership and there is a reason why you are paid more for that as a landlord. it is business ownership and not a passive investment.

[beltim]

I tip my hat to you, sir.

My pleasure!  Thanks to Brewer for setting me up

[foobar]

Thats what insurance is for:) You can outsource all of that property management stuff. The question then is can you still make 7% on your investment if you do that. You could also make a good argument that 80k of rental income is worth less than 80k of QDivs.

The piece where rental feels so different to me vs paper assets like stocks/bonds is that in rental you carry the actual liability. If something happens inside , the person can always come back to you with claim that you 'knew or should have known' ... imagine this scenario as a stock owner , at most you are on the hook for is the value of the stock. legal liability, maintenance, responsibility to respond to complain (or be summoned to court) , deal with police if something is happening on your property - all very different from truly passive ownership and there is a reason why you are paid more for that as a landlord. it is business ownership and not a passive investment.

[skyrefuge]

But he's looking at cash flow that's twice his expenses:
So the spent amount isn't an SWR, but still a WR:

Really, a "WR" is only the money that you actually spend. So ARS's WR will be approximately half of his ROI, not the entire ROI. The other half of his cash flow isn't being "withdrawn" from his portfolio, it's simply changing the asset allocation of his portfolio (or expanding it if he's putting it into more real estate).

Whether a particular WR is more or less than the SWR cannot be determined by cash flow amount; the two things are unrelated. Though in ARS's case, where his WR is substantially less than his cash flow, it would be pretty shocking to learn 50 years from now that his WR was greater than his SWR.

But we all define "enough" differently:

Though in this case where we're talking about SWR, we're actually defining it quite specifically. It's the amount you need to accumulate to show a 95% historical success rate over an N-year period for your given portfolio. With your dividend/interest-only approach, you will likely accumulate far more assets than you need, and thus, your SWR (the amount you could safely withdraw) will be higher than your actual WR. (Note that the SWR research does not use dividends/interest as a factor; it just looks at total return.) That's fine, it's just a conservative approach, and people will be able to say, with more objectivity than usual for such words, that you have more than "enough".

I'm going to speculate here, though: I suspect that the typical numbers would suggest that real estate isn't necessarily any better than traditional investments; on average, I bet they're not too dissimilar from stocks and bonds.  Doesn't mean you can't do better than the average though.

Yeah, this is the interesting question. Is it easier to outperform the market in real estate by being a "good property picker" than it is to outperform the market in stocks by being a "good stock picker" (where it's known to be almost impossible)? I think it probably is, though I imagine there has been far less research into this question in the real estate domain than there has been in the stock market domain. Despite that strong evidence, tons of people still wrongly believe they can outperform the market in stocks over the long term, so I bet even more people think they can outperform the real estate market; hopefully they're more correct about their skills.

Conversely, if your cash flow is something high, like 15% of market value and your SWR is 5%, then your method would have people spending all their cash when it actually needs to be reinvested to prevent ruin.

Interesting. 

I can see the former scenario, but not the latter.  How would you determine that SWR?

The SWR would be determined the way that they all are, by computing yearly balances based on historical data. The question is, what investment would produce historical data like that, where the generated SWR is lower than the cash flow rate? It's probably not real estate, but I could see something like CDs fitting that profile. Imagine you have a $1m 1-year CD with a nominal interest rate of 5%. Every year, you spend the $50k of interest from it, and then roll the principal into another $1m 1-year CD @ 5%. So your cash flow rate will be 5%, but your SWR will something less than that. That's because after the first year, 5% of $1m is no longer an inflation-adjusted $50k, and that would start you on the death spiral of eating into your principal every year to generate the income you need. To spend at the actual SWR (maybe 2%?) you would need to spend only $20k in that first year, and then roll the remaining $30k back into the next year's CD on top of your existing $1m principal.

[arebelspy]

I'm going to speculate here, though: I suspect that the typical numbers would suggest that real estate isn't necessarily any better than traditional investments; on average, I bet they're not too dissimilar from stocks and bonds.  Doesn't mean you can't do better than the average though.

Yeah, this is the interesting question. Is it easier to outperform the market in real estate by being a "good property picker" than it is to outperform the market in stocks by being a "good stock picker" (where it's known to be almost impossible)? I think it probably is, though I imagine there has been far less research into this question in the real estate domain than there has been in the stock market domain. Despite that strong evidence, tons of people still wrongly believe they can outperform the market in stocks over the long term, so I bet even more people think they can outperform the real estate market; hopefully they're more correct about their skills.

I'd have guessed there was way more people who think they can outperform stocks than who invest with intent in real estate.  Briefly in 2004-2006 many people jumped in, but they weren't thinking to outperform it, just ride the wave up.

Most small landlords are accidental, most large are companies built around the business model.  There exists niches for real estate investors.

In any case, like I said, I think the real estate market is much more inefficient.  The stock market used to be more inefficient (back in Graham's day, perhaps).  The speed at which real estate moves, lack of information (though more is becoming available electronically, there's still a lot that is not), etc. all helps a savvy investor pick and choose carefully.

I'm an index funder, but not in real estate.

Conversely, if your cash flow is something high, like 15% of market value and your SWR is 5%, then your method would have people spending all their cash when it actually needs to be reinvested to prevent ruin.

Interesting. 

I can see the former scenario, but not the latter.  How would you determine that SWR?

The SWR would be determined the way that they all are, by computing yearly balances based on historical data. The question is, what investment would produce historical data like that, where the generated SWR is lower than the cash flow rate? It's probably not real estate, but I could see something like CDs fitting that profile. Imagine you have a $1m 1-year CD with a nominal interest rate of 5%. Every year, you spend the $50k of interest from it, and then roll the principal into another $1m 1-year CD @ 5%. So your cash flow rate will be 5%, but your SWR will something less than that. That's because after the first year, 5% of $1m is no longer an inflation-adjusted $50k, and that would start you on the death spiral of eating into your principal every year to generate the income you need. To spend at the actual SWR (maybe 2%?) you would need to spend only $20k in that first year, and then roll the remaining $30k back into the next year's CD on top of your existing $1m principal.

Gotcha.  So theoretically if your rents don't increase with inflation, nor do your property values, your SWR is less than your ROI.  If they increase with inflation (or rather, with your personal spending rate) or higher, it's equal.  If they increase faster, your SWR is greater than your ROI (i.e. you could have been selling a property every once in awhile and still had assets increasing).

That's one reason (the other big one being diversification) why I want a surplus to invest annually (likely in index funds) - I fully expect rents and home prices to track inflation (or rather, wage increases), but would like to have my WR less than my ROI so I can reinvest and hopefully make sure it turns out my WR is a SWR.

Thanks for helping me clarify the "why" a little in my mind!

[WageSlave]

Again, it goes back to knowledge. If you don't know what the fuck you're doing, you can get in a lot of trouble with rentals, and get in way over your head.  People who don't know what they're doing buy at bad prices.  They get bad tenants.  They get bad contractors.  Bad property managers.  Etc. 

Since so many people dive in without education (ESPECIALLY since so many do it unwillingly via accidental landlording, and they don't go into it preparing and learning about it), there are enough horror stories out there to scare many people away.  Plenty of people won't go into it because they don't want to unclog toilets, when I never have, and never will. 

Changing gears again: ARS, have you ever considered consulting, or "renting" your expertise?  Or partnering with an angel investor type?  Might be some opportunities for you to pick up a little extra cash and make the whole "wager $50k?" problem moot.  ;)

Though in this case where we're talking about SWR, we're actually defining it quite specifically. It's the amount you need to accumulate to show a 95% historical success rate over an N-year period for your given portfolio. With your dividend/interest-only approach, you will likely accumulate far more assets than you need, and thus, your SWR (the amount you could safely withdraw) will be higher than your actual WR. (Note that the SWR research does not use dividends/interest as a factor; it just looks at total return.) That's fine, it's just a conservative approach, and people will be able to say, with more objectivity than usual for such words, that you have more than "enough".

(Emphasis mine.)

This is an honest question, not part of the pedantry that inspired this thread: is the 95% success factor generally considered part of the formal SWR definition?  That's kind of where my admittedly ultra-conservative approach starts: when I first played with tools like FIRECalc, I only looked for 100% success scenarios.

In any case, like I said, I think the real estate market is much more inefficient.  The stock market used to be more inefficient (back in Graham's day, perhaps).  The speed at which real estate moves, lack of information (though more is becoming available electronically, there's still a lot that is not), etc. all helps a savvy investor pick and choose carefully.

So I recently bought a house (primary residence).  I forgot how long the closing process is; I was stuck inside a title company's little conference room for about half a day (along with a couple lawyers).  I had to sign a thousand papers, fax this, get my wife to sign a few things, verify that, etc etc.  It's just this elaborate process.  And then I see all the line items on the HUD-1 that spell out the "friction" for real estate transactions.  It's the first house I've closed on since working for a high frequency trading company, and so only then did it occur to me how huge a difference there is in trading real estate versus financial products.  My house basically cost nothing relative to the total value of trades going on in electronic marketplaces---in a matter of seconds.  The contrast is astounding.

There is enough information coming out of the electronic exchanges to allow for profitable trades to be predicted algorithmically... is that amount of quality data readily available on real estate?  It would be neat (and very profitable) if one could construct a program to filter all real estate listings through a program and flag ones that were highly like to be good buys (for whatever purpose, flipping, renting, teardown, etc).

[skyrefuge]

This is an honest question, not part of the pedantry that inspired this thread: is the 95% success factor generally considered part of the formal SWR definition?  That's kind of where my admittedly ultra-conservative approach starts: when I first played with tools like FIRECalc, I only looked for 100% success scenarios.

I guess it might be less formal than I made it seem. In 1998, three dudes published what became known as the Trinity Study, and many people cite that as the main inspiration for the "4% rule" that became widespread in the world of personal finance. In that study, the success rate withdrawing 4% from a 50/50 stocks/bonds portfolio over 30 years was 95%. Since "4%" became the rule widely endorsed by financial advisers, that implies a widespread belief that 95% is "good enough"; if everyone figured 100% success was necessary, we might have settled on 3% instead of 4%. (on the other hand, maybe "4%" really originated with Bengen's earlier work, which came up with 4.15% for 100% safety, because he used slightly different indexes than the Trinity Study. On the other other hand, cFIREsim, which uses more history, gives 90% success for a 4% withdrawal).

Anyway, for an interesting discussion on what success rates we should shoot for, see William Bernstein's essay. The main takeaway: "history teaches us that depriving ourselves to boost our 40-year success probability much beyond 80% is a fool’s errand, since all you are doing is increasing the probability of failure for political, economic, and military reasons relative to the failure of banal financial planning."

[arebelspy]

Changing gears again: ARS, have you ever considered consulting, or "renting" your expertise?  Or partnering with an angel investor type?  Might be some opportunities for you to pick up a little extra cash and make the whole "wager $50k?" problem moot.  ;)

Not really.  Most people who sell real estate info are scams, and it's just too much of a turnoff.  People have approached me about investing together, and I've done some joint venture partnerships, but I'm not going to go out of my way to look for stuff like that.

So I recently bought a house (primary residence).  I forgot how long the closing process is; I was stuck inside a title company's little conference room for about half a day (along with a couple lawyers).  I had to sign a thousand papers, fax this, get my wife to sign a few things, verify that, etc etc.  It's just this elaborate process.  And then I see all the line items on the HUD-1 that spell out the "friction" for real estate transactions.  It's the first house I've closed on since working for a high frequency trading company, and so only then did it occur to me how huge a difference there is in trading real estate versus financial products.  My house basically cost nothing relative to the total value of trades going on in electronic marketplaces---in a matter of seconds.  The contrast is astounding.

There is enough information coming out of the electronic exchanges to allow for profitable trades to be predicted algorithmically... is that amount of quality data readily available on real estate?  It would be neat (and very profitable) if one could construct a program to filter all real estate listings through a program and flag ones that were highly like to be good buys (for whatever purpose, flipping, renting, teardown, etc).

Right, the amount of friction involved in the market used to also be higher - you'd call a broker (literally) and he'd tell you what to buy.  It's quite different now with stocks.  Real estate is still similar to that (but replace broker with Realtor).

There is limited information on real estate.  More if you have MLS access.  But the problem is that most "deals" aren't found on the MLS, where most of the information is.  But yes, you can filter and use algorithms for that, then jump on the rare deal when they pop up (or when a market is below where you judge it should be - usually due to fear and lack of information on other people's part - jump on the best of the many deals).  You have to know what you're doing though, your criteria, etc.

Real estate is a lot of fun though.  :)

[Mazzinator]

I just wanted to say thank you to all of you for this thread. I never cared about the terminology but needed help in wrapping my head around it overall. I learned/realized earlier that it would "costs less" in buying rental properties for an income stream vs buying index funds/using the 4% swr, but i couldn't explain or "calculate" how or why.

Anyways, i just wanted to say thanks and let you know that this discussion actually helped someone!!!

[arebelspy]

I just wanted to say thank you to all of you for this thread. I never cared about the terminology but needed help in wrapping my head around it overall. I learned/realized earlier that it would "costs less" in buying rental properties for an income stream vs buying index funds/using the 4% swr, but i couldn't explain or "calculate" how or why.

Anyways, i just wanted to say thanks and let you know that this discussion actually helped someone!!!

Possibly (and probably, if done right). But make sure you don't withdraw too much (i.e. your whole rental income) without saving for capital expenses, law of large numbers things happening at once aka murphy's law, etc.  Be smart about it (thus the idea that WR may be higher than SWR, even if WR = ROI).  :)