Author Topic: Tell me about your inflation adjustments  (Read 6117 times)

Eric

  • Magnum Stache
  • ******
  • Posts: 4057
  • Location: On my bike
Tell me about your inflation adjustments
« on: October 20, 2017, 03:04:42 PM »
I'm curious as to how this works in real life.  Of course we're all aware of the Trinity Study methodology of withdrawing a set amount increasing by inflation each year.  Are you following this method?  If so, do you simply look up CPI stats each year?  And which CPI?  Or do you track your local prices and adjust accordingly?  Or maybe you haven't taken an inflation adjustment yet.

I realize that with this raging bull market it's probably not super important, but at some point it may be.  Tell me about your plans and how you make (or plan to make) your inflation adjustments.

secondcor521

  • Walrus Stache
  • *******
  • Posts: 5503
  • Age: 54
  • Location: Boise, Idaho
  • Big cattle, no hat.
    • Age of Eon - Overwatch player videos
Re: Tell me about your inflation adjustments
« Reply #1 on: October 20, 2017, 05:09:22 PM »
I've only been FIREd about two years.

In real life, I don't worry about it.  Any inflation I have experienced so far has been outweighed by other much larger factors on the expense side of the ledger.  Deciding to send my two younger kids to expensive private high schools and lower taxes both outweigh what inflation has theoretically done to me by a factor of 35 and 15, respectively.

My written plan is to follow the Trinity study methodology and do what was long ago called the payout period reset or "retire again and again" model.  Meaning that every year I will take the maximum of last year's withdrawal plus inflation or 4% of my current balance.

Since I don't actually worry about it, I don't look up CPI stats, but if I did I would use CPI-U.

Some things I do to address inflation indirectly:

1.  Have a high stock allocation.  My IPS calls for 90% until after the next downturn and recovery, at which point I will likely go to 100%.

2.  Have low expenses relative to my SWR.  Currently I am at about 1.95% gross WR and allow myself up to 4%.  So if inflation is bad for a while that'd be unpleasant but survivable.

3.  Continuously minimize expenses.  This is a lot of different things, including #2 above.  But it also includes DIY painting my house, comparison shopping on expensive stuff, etc.

SwordGuy

  • Walrus Stache
  • *******
  • Posts: 8955
  • Location: Fayetteville, NC
Re: Tell me about your inflation adjustments
« Reply #2 on: October 20, 2017, 05:28:42 PM »
We fire this coming spring, but inflation is definitely something we're planning for.  I remember inflation when I was younger back in the 1970s.  It was nasty.

Our plan has been to diversify our earnings into five different income streams.

1) Social Security:     44.6% of targeted FI expenses.               
2) Passive Farm Income: 26.6%.
3) Rental Income:       12.8% climbing to 44.8% over 4 years post FIRE.
4) Stock/Bond:          60.9% at 3.5% SWR, more over time at historic average growth rate.
5) Pension:              2.0% (every little bit helps)


Social Security will somewhat track inflation, lagging about a year, though nothing seems to keep up with medical cost increases.
Farm income should track inflation just fine.
Rental income should track inflation with a year or two lag, depending upon lease terms and whether the tenants are good quality.
The stock portion of the portfolio, by far the largest, should track inflation.   The bonds aren't there for growth anyway.  I'll get rid of much of them as time allows.
No idea whether the pension will stay up with inflation or not.


John Doe

  • 5 O'Clock Shadow
  • *
  • Posts: 81
Re: Tell me about your inflation adjustments
« Reply #3 on: October 21, 2017, 05:56:40 PM »
I don’t really worry about inflation too much as the majority of our investments are in dividend paying stocks that have a history of annually raising their payouts and the increases are generally higher than today’s inflation expectations.

Frankies Girl

  • Magnum Stache
  • ******
  • Posts: 3899
  • Age: 86
  • Location: The oubliette.
  • Ghouls Just Wanna Have Funds!
Re: Tell me about your inflation adjustments
« Reply #4 on: October 21, 2017, 07:03:35 PM »
Is it weird if I don't calculate any of that stuff any more?

It's been almost 3 years since I FIREd, over 1 year for the husband and we're fine. We're actually more than fine. I think last time I figured our percentage for expenses it was barely 2% of my total portfolio (not the net worth).

So anything inflation-wise is really tiny and doesn't move the needle much at all.

I mean, I'm peeved to note food prices rising, and I note some other items I used to by at X amount are now regularly priced at X+Y... but it doesn't figure into me actually changing much of anything because I usually shop sales/clearances and just won't buy multiples of something that is higher priced.

Eric

  • Magnum Stache
  • ******
  • Posts: 4057
  • Location: On my bike
Re: Tell me about your inflation adjustments
« Reply #5 on: October 24, 2017, 12:11:18 PM »
So no one actually calculates inflation?  Or has a mechanism to adjust for it?  That's a bit surprising to me, considering that inflation adjustments seem to be a cornerstone of just about every type of withdrawal plan.

devan 11

  • 5 O'Clock Shadow
  • *
  • Posts: 54
  • Location: Iowa, USA
Re: Tell me about your inflation adjustments
« Reply #6 on: October 24, 2017, 11:28:02 PM »
I wanted to stick with a 1.8% withdrawal for living expenses and use a Roth ladder to convert to Roths with the rest of our lower tax tier money.  I decided that our home needed some renovations and repairs.  We spent some of the 'stache on those for this year, but that is about ended.  Our 'stache is up above CPI-e inflation even after some significant spending.  We use cpi-e as our base rate to make sure that when the time comes to need it, we won't have already fallen behind.  Once our goal of converting is completed, we will be freed to spend a 3.5% or so per year.  At 65, we may up it. To 4%.
We have some other income sources to get us by, too.

As long as we are gaining faster than our cpi-e adjusted stache is climbing, i'm comfortable about our spending.  So rather than give ourselves a raise, I maintain an adjusted rising stache.

AdrianC

  • Handlebar Stache
  • *****
  • Posts: 1211
  • Location: Cincinnati
Re: Tell me about your inflation adjustments
« Reply #7 on: October 25, 2017, 08:15:37 AM »
I figure things are going to cost what they cost. I'm going to withdraw* what we need and want to spend. If it get's much above 5% of remaining stash I'm going to be concerned.

I have planned for medical insurance inflation at a much higher rate than typical CPI adjustments. We also have college savings for the kids in separate 529 accounts that we don't count as part of our stash. They'll get to use that for their education. If it's not enough they'll make up the rest. That's the deal.

* Though we haven't started to withdraw yet. My consulting business is still throwing off more than enough residual income.

devan 11

  • 5 O'Clock Shadow
  • *
  • Posts: 54
  • Location: Iowa, USA
Re: Tell me about your inflation adjustments
« Reply #8 on: October 25, 2017, 01:00:53 PM »
I wanted to stick with a 1.8% withdrawal for living expenses and use a Roth ladder to convert to Roths with the rest of our lower tax tier money.  I decided that our home needed some renovations and repairs.  We spent some of the 'stache on those for this year, but that is about ended.  Our 'stache is up above CPI-e inflation even after some significant spending.  We use cpi-e as our base rate to make sure that when the time comes to need it, we won't have already fallen behind.  Once our goal of converting is completed, we will be freed to spend a 3.5% or so per year.  At 65, we may up it. To 4%.
We have some other income sources to get us by, too.

As long as we are gaining faster than our cpi-e adjusted stache is climbing, i'm comfortable about our spending.  So rather than give ourselves a raise, I maintain an adjusted rising stache.

Retire-Canada

  • Walrus Stache
  • *******
  • Posts: 8684
Re: Tell me about your inflation adjustments
« Reply #9 on: October 25, 2017, 03:05:07 PM »
So no one actually calculates inflation?  Or has a mechanism to adjust for it?  That's a bit surprising to me, considering that inflation adjustments seem to be a cornerstone of just about every type of withdrawal plan.

That doesn't surprise me. People are not spending a fixed amount each year like robots and I bet most people set a high theoretical spending level for a safety margin so they'd have a fair while after FIREing before inflation caught up with their actual spending. Add some ongoing cost optimization and inflation effects are moot for many years.

Eric

  • Magnum Stache
  • ******
  • Posts: 4057
  • Location: On my bike
Re: Tell me about your inflation adjustments
« Reply #10 on: October 25, 2017, 04:45:28 PM »
So no one actually calculates inflation?  Or has a mechanism to adjust for it?  That's a bit surprising to me, considering that inflation adjustments seem to be a cornerstone of just about every type of withdrawal plan.

That doesn't surprise me. People are not spending a fixed amount each year like robots and I bet most people set a high theoretical spending level for a safety margin so they'd have a fair while after FIREing before inflation caught up with their actual spending. Add some ongoing cost optimization and inflation effects are moot for many years.

I'm not surprised that people aren't spending the exact same thing every year.  I'm surprised that no one seems to even have a mechanism to record and assess inflation.  I was expecting to see a few different plans, but so far I'm not really seeing anything resembling a plan at all.  Of course it doesn't matter much if inflation continues to be low, but that could certainly change.  Or having a 2% or lower WR could certainly play into it too, but I'm definitely not working that long.

I was attempting to poll people to figure out my own plan, but it appears that I'll need to do all of my own research on this issue.  Dammit. 

Even if I don't use it to make adjustments to yearly spending, at the very least I'll want to know how inflation has affected my initial portfolio value after 5, 10, or 20 years.  Even low inflation sneaks up on you on a long time frame.  I certainly don't want to be stuck viewing my portfolio in 2017 dollars.  And I think this is something that's inherently hard for humans to grasp.  Ask any old person about what things should cost.
« Last Edit: October 25, 2017, 04:47:10 PM by Eric »

SwordGuy

  • Walrus Stache
  • *******
  • Posts: 8955
  • Location: Fayetteville, NC
Re: Tell me about your inflation adjustments
« Reply #11 on: October 25, 2017, 05:33:21 PM »
So no one actually calculates inflation?  Or has a mechanism to adjust for it?  That's a bit surprising to me, considering that inflation adjustments seem to be a cornerstone of just about every type of withdrawal plan.

If my sources of income won't grow at the inflation rate then I had better calculate for inflation.

But if my income sources will adjust for inflation, there's not that much need to do so.

In my case, social security, farm income, stock income and rental income will adjust (with a lag period of about a year) to inflation.   That means I'll lose a bit of purchasing power due to the lag, but then again I'll gain a bit due to having a fixed rate mortgage.   So why bother?



Eric

  • Magnum Stache
  • ******
  • Posts: 4057
  • Location: On my bike
Re: Tell me about your inflation adjustments
« Reply #12 on: October 25, 2017, 05:51:06 PM »
So no one actually calculates inflation?  Or has a mechanism to adjust for it?  That's a bit surprising to me, considering that inflation adjustments seem to be a cornerstone of just about every type of withdrawal plan.

If my sources of income won't grow at the inflation rate then I had better calculate for inflation.

But if my income sources will adjust for inflation, there's not that much need to do so.

In my case, social security, farm income, stock income and rental income will adjust (with a lag period of about a year) to inflation.   That means I'll lose a bit of purchasing power due to the lag, but then again I'll gain a bit due to having a fixed rate mortgage.   So why bother?

Because none of those listed sources are guaranteed to keep up with inflation?  And even if they did, don't you want to set your "income" based on the actual inflation and not just a guess at what inflation might be?  In your original post you mentioned remembering the inflation of the 70s.  You realize that stock returns didn't even come close to keeping up with inflation during the same time period, right?

And as I mentioned, in the current environment the wait and see approach is certainly not going to be a detriment.  But since I'm working out my soon to be implemented plan, I was hoping to see how others actually planned for it.  I'm glad it works for you, but I'll need more record keeping than simply trusting that my income will keep up without actually tracking it.

Retire-Canada

  • Walrus Stache
  • *******
  • Posts: 8684
Re: Tell me about your inflation adjustments
« Reply #13 on: October 25, 2017, 07:11:51 PM »
I'm surprised that no one seems to even have a mechanism to record and assess inflation.

If I ever wanted to know what inflation was I could:

1. consult the Bank of Canada inflation calculator for a general answer
2. review my FIRE budget and see how individual items have changed in price over time to determine my personal inflation rate

That said I have no plans to adjust anything around FIRE on an annual basis relating to inflation.

aceyou

  • Handlebar Stache
  • *****
  • Posts: 1669
  • Age: 40
    • Life is Good - Aceyou's Journal
Re: Tell me about your inflation adjustments
« Reply #14 on: October 25, 2017, 07:36:39 PM »
I think one reason people don't track inflation is because it isn't very actionable.  People make spending decisions on an item-by-item basis, not based on what the average price of all goods/services is. 

If the price of housing and new cars rises dramatically, the country will probably experience inflation.  But if I own my house outright and switch used cars every decade and am FIRED, that's probably not affecting MY bottom line.  And since most 30-year periods from the Trinity study leave people with far more money than they even started, most people probably never reach any point of concern even if inflation does effect their bottom line a bit. 

onewayfamily

  • 5 O'Clock Shadow
  • *
  • Posts: 92
Re: Tell me about your inflation adjustments
« Reply #15 on: October 26, 2017, 05:56:51 AM »
Yeah similar to what Retire-Canada posted - if you're worried about inflation maybe just check what it's been every 2-3 years during FIRE, or track your own expenses for 3-12 months again to confirm your budget is still in line with your plans and what you consider safe.

I believe arebelspy is one of our resident inflation experts - you might want to search for some of his posts or maybe he'll find time in his busy FIREd schedule to post here on this thread :-)

Playing with Fire UK

  • Magnum Stache
  • ******
  • Posts: 3449
Re: Tell me about your inflation adjustments
« Reply #16 on: October 26, 2017, 06:46:10 AM »
I'm (pre-FIRE) tracking my personal inflation level. I figure that's more relevant than any national inflation figures.

Eric

  • Magnum Stache
  • ******
  • Posts: 4057
  • Location: On my bike
Re: Tell me about your inflation adjustments
« Reply #17 on: October 26, 2017, 10:59:06 AM »
I think one reason people don't track inflation is because it isn't very actionable.  People make spending decisions on an item-by-item basis, not based on what the average price of all goods/services is. 


I'm thinking more of a long term effect.  What got me started on this path was I downloaded all my yearly salary info from the SS office, and then used that to calculate my SS payment using the formula(s) in this Kitces article.  I was pretty shocked when I saw that my first "real" job's original $32k salary was the equivalent of $50k in today's dollars.  That's a ~60% inflation increase in 15 years.  And I've barely noticed.

I took that shock an applied it to how I view portfolio values and how I would judge my current path.  For example, if I retire with $1MM and 15 years from now, I had $900k, I'd would still considered that to be pretty decent.  However, even though I've barely noticed inflation in the last 15 years, if I apply that same amount to my portfolio, I'm actually only sitting at ~$540K in 2017 dollars.  So that's a pretty big difference between having a portfolio that's reduced by 10% vs one reduced by half.

I realize that a lot of this is a "well duh" realization, but my overall point  is that humans are really bad at recognizing inflation.  All the numbers in every retirement study are real numbers.  And yet, I barely recognize inflation in my daily life.  I want to make sure that I have a mechanism that will allow me to actually notice and not wake up in 15 years and realize that I have a lot less money than I thought.

Eric

  • Magnum Stache
  • ******
  • Posts: 4057
  • Location: On my bike
Re: Tell me about your inflation adjustments
« Reply #18 on: October 26, 2017, 11:02:42 AM »
Yeah similar to what Retire-Canada posted - if you're worried about inflation maybe just check what it's been every 2-3 years during FIRE, or track your own expenses for 3-12 months again to confirm your budget is still in line with your plans and what you consider safe.

I certainly plan to continue to track my spending, basically forever.  A big issue for me is that I have a variable spending plan, so it's not as simple as just comparing current spending to past spending to judge inflation, since my spending will likely be all over the map (literally, as I plan to travel full time) and that will have nothing to do with inflation.  In addition, the inflation I do experience will not be tied to the inflation that my investments are experiencing. (although this is somewhat mitigated by currency exchange from USD)
« Last Edit: October 26, 2017, 11:05:32 AM by Eric »

wannabe-stache

  • Stubble
  • **
  • Posts: 154
Re: Tell me about your inflation adjustments
« Reply #19 on: October 26, 2017, 11:05:54 AM »
I think one reason people don't track inflation is because it isn't very actionable.  People make spending decisions on an item-by-item basis, not based on what the average price of all goods/services is. 

If the price of housing and new cars rises dramatically, the country will probably experience inflation.  But if I own my house outright and switch used cars every decade and am FIRED, that's probably not affecting MY bottom line.  And since most 30-year periods from the Trinity study leave people with far more money than they even started, most people probably never reach any point of concern even if inflation does effect their bottom line a bit.

^ This.  If I was pulling the FIRE trigger at an aggressive SWR and had no plans to go back to work, I might be more concerned but I don't know actions i would take other than cutting back on expenses.
To another post above this, rental income, stocks and other items do tend to trade in sympathy with inflation. It's not "guaranteed" to be a one for one correlation but even if inflation is up 3% and your rental income goes up 2%, is that because it "didn't keep pace with inflation" or perhaps owner failed to make maintenance capex, better units came online in the market, etc?

Rubyvroom

  • Bristles
  • ***
  • Posts: 368
  • Location: Minnesota
Re: Tell me about your inflation adjustments
« Reply #20 on: October 26, 2017, 11:22:54 AM »
It sounds like I'm different than most on here in that I throw in a 3% annual increase on most line items to look at a medium-term forecast that takes us out a few years past the planned FIRE date, which we're hoping will be 2020. Granted, we are actually still tightening up our spending so our expenses have gone ~down~ the past few years, but I still forecast that some things outside of my control will go up.

For example, I know that Comcast/Xfinity (those bastards) increase our internet each year, our water bill always goes up due to rate increases, etc. I don't want to get so incredibly detailed that I track the increases for each line item or peg it to actual inflation averages, so I just do a blanket 3% across the board. I obviously don't increase things like mortgage (though insurance/taxes could certainly go up), or any other items that are totally within my control. But yes, I do absolutely account for inflation. If I want to FIRE at $40,000 expenses, I want to know that I've built a strong enough portfolio that in 5 years I can still support ~$45,000 in spend, and so on. That and I like spreadsheeting, so maybe (from the sounds of it) I'm just over-complicating things for funzies.

Financial.Velociraptor

  • Handlebar Stache
  • *****
  • Posts: 2148
  • Age: 51
  • Location: Houston TX
  • Devour your prey raptors!
    • Living Universe Foundation
Re: Tell me about your inflation adjustments
« Reply #21 on: October 26, 2017, 01:26:27 PM »
Posting from FinCon17!

I support myself from income earned via writing options.  The inflation rate will tend to be reflected in the prevailing interest rate and thus show up in my premium income.  This way, I expect my income to grow dynamically with inflation changes.

RyanAtTanagra

  • Handlebar Stache
  • *****
  • Posts: 1316
  • Location: Sierra Mountains
Re: Tell me about your inflation adjustments
« Reply #22 on: October 26, 2017, 02:33:33 PM »
I think one reason people don't track inflation is because it isn't very actionable.  People make spending decisions on an item-by-item basis, not based on what the average price of all goods/services is. 

If the price of housing and new cars rises dramatically, the country will probably experience inflation.  But if I own my house outright and switch used cars every decade and am FIRED, that's probably not affecting MY bottom line.  And since most 30-year periods from the Trinity study leave people with far more money than they even started, most people probably never reach any point of concern even if inflation does effect their bottom line a bit.

I was going to post something similar.  Overall inflation is irrelevant, it's your PERSONAL inflation you need to keep track of, which you already know you'll be doing, it's your yearly spending.  If you spent 40k last year and 50k this year you had 25% inflation.  You definitely want to account for that.  But if you spent 40k last year and this year made some life changes or efficiency improvements and spent only 35k, but inflation was 3%, which of those is more relevant to your stash in the long term?  I'd say the first.

If overall inflation is effecting you, you'll see it in your yearly spending and you will account for it as you go when you do an annual comparison of spending vs stash size.

electriceagle

  • Pencil Stache
  • ****
  • Posts: 521
Re: Tell me about your inflation adjustments
« Reply #23 on: November 05, 2017, 05:01:12 AM »
My written plan is to follow the Trinity study methodology and do what was long ago called the payout period reset or "retire again and again" model.  Meaning that every year I will take the maximum of last year's withdrawal plus inflation or 4% of my current balance.

Isn't this risky? My understanding is that each starting year has an n% chance of failing to provide the required 4% for the 30 subsequent years. We don't really know what "n" is, but it is non-zero.

Each time you reset, you take the aforementioned n% risk. Reset twice, and your risk is 3n. That has to add up. (note: actually a little less than 3n, but I don't feel like thinking mathy right now.)

Put another way, the 4% rule relies on having unspent excess profits in up years to make up for the down years. If you start to spend the excess profits by resetting, you might not have enough for the down years.

In answer to the original poster's question, I think that a lot of people look at their expenses and decide whether they need an inflation increase. If they don't, the amount goes into the risk reduction pot instead of the spending pot.

Villanelle

  • Walrus Stache
  • *******
  • Posts: 6657
Re: Tell me about your inflation adjustments
« Reply #24 on: November 05, 2017, 05:12:55 AM »
A significant part of our retirement income will come from Husbands inflation-adjusted pension (military).  I'm not sure how we will address the rest.  I was hoping someone would post a great answer, because I've had this question for a while.  We will likely set things up so that our minimum spend is about the same as DH's pension, so that we can always cut way back and withdraw little or nothing if we need to.  But I'd like to have a way to define for ourselves what "need to" means. 

Retire-Canada

  • Walrus Stache
  • *******
  • Posts: 8684
Re: Tell me about your inflation adjustments
« Reply #25 on: November 05, 2017, 06:46:24 AM »
Isn't this risky? My understanding is that each starting year has an n% chance of failing to provide the required 4% for the 30 subsequent years. We don't really know what "n" is, but it is non-zero.

Each time you reset, you take the aforementioned n% risk. Reset twice, and your risk is 3n. That has to add up. (note: actually a little less than 3n, but I don't feel like thinking mathy right now.)

Put another way, the 4% rule relies on having unspent excess profits in up years to make up for the down years. If you start to spend the excess profits by resetting, you might not have enough for the down years.

You are essentially starting the clock on a poor early sequence of return risk [one of the two main portfolio killers] fresh every time you do that. If you don't do that and get past say 10yrs or so of retirement with solid returns you can start to feel confident you've avoid one of the main risks.

The other risk is high inflation out running the returns on your portfolio.

secondcor521

  • Walrus Stache
  • *******
  • Posts: 5503
  • Age: 54
  • Location: Boise, Idaho
  • Big cattle, no hat.
    • Age of Eon - Overwatch player videos
Re: Tell me about your inflation adjustments
« Reply #26 on: November 05, 2017, 10:29:01 AM »
Isn't this risky? My understanding is that each starting year has an n% chance of failing to provide the required 4% for the 30 subsequent years. We don't really know what "n" is, but it is non-zero.

Each time you reset, you take the aforementioned n% risk. Reset twice, and your risk is 3n. That has to add up. (note: actually a little less than 3n, but I don't feel like thinking mathy right now.)

Put another way, the 4% rule relies on having unspent excess profits in up years to make up for the down years. If you start to spend the excess profits by resetting, you might not have enough for the down years.

You are essentially starting the clock on a poor early sequence of return risk [one of the two main portfolio killers] fresh every time you do that. If you don't do that and get past say 10yrs or so of retirement with solid returns you can start to feel confident you've avoid one of the main risks.

The other risk is high inflation out running the returns on your portfolio.

Good questions.

First, some background reading:  http://www.retireearlyhomepage.com/popr.html.

The n% that is mentioned is somewhere between 5% and 0%, depending on what parameters and assumptions one makes.  With the assumptions I make - the main ones being flexibility rather than blind adherence and having other backups - I view the risk as 0%.

If you accept the assumptions implicit in the 4% rule as commonly stated, the risk of running out, if there is any at all, would not increase additively.  What would increase is the likelihood of coming closer to running out - i.e., ending up with a low portfolio balance at some time during the 30 year period.  Stated another way, there would be an increase in net overall spending, with the potential for a concern (but not actual risk) about running out later.  To see why this is the case, consider trying to explain why the risk is greater for a person who retired last year and saw their portfolio grow and now has $1M and wants to continue their 30 year retirement with a $40K withdrawal rate, and their friend, who has $1M and wants to start a 30 year retirement with a $40K withdrawal rate.

I know that "sequence of returns risk" is one of the latest things to be (re)discovered by the 4% blogosphere and researchers.  But this risk is either (a) a non-existent risk if one accepts the assumptions implicit in the 4% rule as commonly stated - which I do - it is already baked in and represented by the existence of relatively lower ending portfolio distributions, or (b) an alternate expression of the concern that the future will be worse than the past.  In the latter case, as previously alluded to, I have about a dozen other backups and contingency plans that will work in the non-killer-asteroid case.  In the killer asteroid case, I also have a very simple backup plan, which is to be thankful I got to spend some retired time before getting killed by an asteroid.

As far as inflation goes, I am allocated 90% to stocks, which I believe will serve as adequate protection against high inflation.  (I may be wrong here, I've only thought about this logically and rely on commonly held opinions and haven't done any actual research.)

Overall, another thing that makes me feel safe - and renders this conversation mostly moot in practical terms for me -
 is that my net withdrawal rate is currently under 1%.  What I originally wrote was confusing now that I re-read it, but I don't completely follow my written plan.

Retire-Canada

  • Walrus Stache
  • *******
  • Posts: 8684
Re: Tell me about your inflation adjustments
« Reply #27 on: November 05, 2017, 10:40:42 AM »
Overall, another thing that makes me feel safe - and renders this conversation mostly moot in practical terms for me -
 is that my net withdrawal rate is currently under 1%.  What I originally wrote was confusing now that I re-read it, but I don't completely follow my written plan.

At 1%WR we can stop talking about this. ;)

Paul der Krake

  • Walrus Stache
  • *******
  • Posts: 5854
  • Age: 16
  • Location: UTC-10:00
Re: Tell me about your inflation adjustments
« Reply #28 on: November 05, 2017, 12:05:38 PM »
Inflation matters in the aggregate, at the population scale. Individual households needs and expenses are too volatile to draw any meaningful conclusions on the effect of inflation on their finances.

Nords

  • Magnum Stache
  • ******
  • Posts: 3421
  • Age: 63
  • Location: Oahu
    • Military Retirement & Financial Independence blog
Re: Tell me about your inflation adjustments
« Reply #29 on: November 11, 2017, 01:58:17 PM »
I'm curious as to how this works in real life.  Of course we're all aware of the Trinity Study methodology of withdrawing a set amount increasing by inflation each year.  Are you following this method?  If so, do you simply look up CPI stats each year?  And which CPI?  Or do you track your local prices and adjust accordingly?  Or maybe you haven't taken an inflation adjustment yet.

I realize that with this raging bull market it's probably not super important, but at some point it may be.  Tell me about your plans and how you make (or plan to make) your inflation adjustments.
So no one actually calculates inflation?  Or has a mechanism to adjust for it?  That's a bit surprising to me, considering that inflation adjustments seem to be a cornerstone of just about every type of withdrawal plan.

That doesn't surprise me. People are not spending a fixed amount each year like robots and I bet most people set a high theoretical spending level for a safety margin so they'd have a fair while after FIREing before inflation caught up with their actual spending. Add some ongoing cost optimization and inflation effects are moot for many years.

I'm not surprised that people aren't spending the exact same thing every year.  I'm surprised that no one seems to even have a mechanism to record and assess inflation.  I was expecting to see a few different plans, but so far I'm not really seeing anything resembling a plan at all.  Of course it doesn't matter much if inflation continues to be low, but that could certainly change.  Or having a 2% or lower WR could certainly play into it too, but I'm definitely not working that long.

I was attempting to poll people to figure out my own plan, but it appears that I'll need to do all of my own research on this issue.  Dammit. 

Even if I don't use it to make adjustments to yearly spending, at the very least I'll want to know how inflation has affected my initial portfolio value after 5, 10, or 20 years.  Even low inflation sneaks up on you on a long time frame.  I certainly don't want to be stuck viewing my portfolio in 2017 dollars.  And I think this is something that's inherently hard for humans to grasp.  Ask any old person about what things should cost.
There's two sides to this issue, and a potential third issue is that you might possibly be overthinking it.

The first issue:  applying national numbers to individuals.  Actuaries can predict the length of a cohort's lifespan, but nobody can predict when an individual will die.  In the same metaphor, the CPI has nothing to do with your personal inflation.

We started our FI 15 years ago with a retirement budget, and the first thing that happened was all of our expenses dropped.  Our budget was based on history (with a conservative buffer) so it was higher than it needed to be.  Of course our commuting expenses disappeared, and our driving expenses went down.  In FI we took a victory lap but then overhauled all of our utility bills, insurance policies, and other non-discretionary expenses.  We also changed some lifestyle like cooking more meals and eating less takeout. 

We kept tracking our expenses (and updating the budget) but nothing turned into an inflationary budget-buster.  Some years our expenses would be way up for travel (because we planned more travel) and other years we'd have a lumpy expense (like a home renovation) but overall our spending was less than inflation.  (Our airline spending actually dropped because of travel hacking and having the time for slow travel during shoulder seasons.)  Our electric bill could have grown with inflation but we bought a photovoltaic array and produce more than we use.  (It had a five-year payback and is now "free electricity" for life.)  Our phone bill could have grown but we ditched our landline for a smartphone.

The only bills that have reliably gone up have been property taxes, cable TV, and water/sewer.  The first two haven't risen enough to make a difference (because Hawaii property taxes are low and so is our cable bill) and the third is caused by replacing Oahu's rotting sewage infrastructure.

Around 2015 (which included five months of overseas travel) we stopped tracking our spending.  I still track our assets in Personal Capital, and I can look at our credit-card spending (we rarely use cash).  Our spending has not risen with inflation, although it's gone up a bit since 2015.  Our assets have risen faster than inflation.

Humans are not robots.  Even if you withdraw 4% + CPI for a few years, you'll probably slow that down when your checking account balance keeps rising.  If you run into a recession then you might even cut spending, although technically the studies say that's not necessary.  Variable spending will save your ass(ets) and keep your portfolio from failing.

The second issue:  asset allocation.  If you're concerned about inflation, then invest in assets which can keep up with inflation. 

Because of my military pension, we have a >90% asset allocation to equities.  Our investment portfolio has actually grown faster than inflation (and faster than my pension COLA) despite starting our retirement in 2002 and riding out both that recession and 2008-09.  In other words, no brilliant market timing took place or was even necessary.  I can see that in another 5-10 years the dividends will cover the expense gap between my pension and our spending, and we'll no longer be touching the principle.

Not everyone wants an aggressive asset allocation, but there are plenty of blue-chip equity dividend funds whose dividends grow faster than inflation.  Bond asset allocations can be invested in I bonds (although the investor limits mean that it takes a while) or TIPS.  Retirees can buy a bare-bones inflation-adjusted annuity, or start with a SPIA to get through the first decade or two.  The SPIA will erode with inflation but after two decades the sequence-of-returns risk will be darn near zero.

The possible third issue:  you might be rethinking parameters that are already handled by the Trinity Study and the 4% SWR.  Both of those account for historical inflation.  If you insist on doing your own research dammit then you could replicate a lot of that with a Monte Carlo algorithm.  (Monte Carlo tends to be more conservative than historical data, but you can synthesize a lot more data runs with MC.)  Even despite inflation and robotic spending of the 4% SWR, at least 80% of the time your portfolio survived for at least 30 years.

You could watch your expenses and inflation for the first decade of FI's sequence-of-returns risk.  However you could also compare your portfolio withdrawal rate over that decade, and you'll probably see that the %% is dropping as the portfolio grows faster than your spending (and faster than your personal CPI).  By the time you get past the first decade, you may have a withdrawal rate that's below 3%-- and all of the research says that's sustainable for at least 50 years. 

Here's a net-worth chart indexed to our FI=100%.  Today, 18 years later, we're well over 200% and growing.
http://the-military-guide.com/hey-nords-hows-net-worth/
« Last Edit: November 11, 2017, 02:02:12 PM by Nords »

AdrianC

  • Handlebar Stache
  • *****
  • Posts: 1211
  • Location: Cincinnati
Re: Tell me about your inflation adjustments
« Reply #30 on: November 11, 2017, 05:53:10 PM »
The only bills that have reliably gone up have been property taxes, cable TV, and water/sewer.
No healthcare inflation?

Nords

  • Magnum Stache
  • ******
  • Posts: 3421
  • Age: 63
  • Location: Oahu
    • Military Retirement & Financial Independence blog
Re: Tell me about your inflation adjustments
« Reply #31 on: November 11, 2017, 06:08:35 PM »
The only bills that have reliably gone up have been property taxes, cable TV, and water/sewer.
No healthcare inflation?
Ah, good point, I stand corrected.

I don't usually bring up healthcare here (I'm a military retiree on Tricare Prime).  DoD has indexed those "enrollment fees" to the CPI.  Before that they were flat for 15 years of Congressional gridlock.

If healthcare is over 30% of a FI budget then yeah, that'd be problematic. 

In anticipation of other questions, our vehicle insurance has been largely flat for nearly 20 years because we don't carry collision or comprehensive.

Again, everyone has to assess their own personal rate of inflation and also has to choose an asset allocation which will at least attempt to keep up with inflation.  It has to be analyzed in the context of the whole budget for your individual situation, not based on national averages or individual categories. 


gerardc

  • Pencil Stache
  • ****
  • Posts: 767
  • Age: 40
  • Location: SF bay area
Re: Tell me about your inflation adjustments
« Reply #32 on: November 11, 2017, 06:52:28 PM »
Isn't this risky? My understanding is that each starting year has an n% chance of failing to provide the required 4% for the 30 subsequent years. We don't really know what "n" is, but it is non-zero.

Each time you reset, you take the aforementioned n% risk. Reset twice, and your risk is 3n. That has to add up. (note: actually a little less than 3n, but I don't feel like thinking mathy right now.)

Put another way, the 4% rule relies on having unspent excess profits in up years to make up for the down years. If you start to spend the excess profits by resetting, you might not have enough for the down years.

You are essentially starting the clock on a poor early sequence of return risk [one of the two main portfolio killers] fresh every time you do that. If you don't do that and get past say 10yrs or so of retirement with solid returns you can start to feel confident you've avoid one of the main risks.

The other risk is high inflation out running the returns on your portfolio.

Good questions.

First, some background reading:  http://www.retireearlyhomepage.com/popr.html.

The n% that is mentioned is somewhere between 5% and 0%, depending on what parameters and assumptions one makes.  With the assumptions I make - the main ones being flexibility rather than blind adherence and having other backups - I view the risk as 0%.

If you accept the assumptions implicit in the 4% rule as commonly stated, the risk of running out, if there is any at all, would not increase additively.  What would increase is the likelihood of coming closer to running out - i.e., ending up with a low portfolio balance at some time during the 30 year period.  Stated another way, there would be an increase in net overall spending, with the potential for a concern (but not actual risk) about running out later.  To see why this is the case, consider trying to explain why the risk is greater for a person who retired last year and saw their portfolio grow and now has $1M and wants to continue their 30 year retirement with a $40K withdrawal rate, and their friend, who has $1M and wants to start a 30 year retirement with a $40K withdrawal rate.

I know that "sequence of returns risk" is one of the latest things to be (re)discovered by the 4% blogosphere and researchers.  But this risk is either (a) a non-existent risk if one accepts the assumptions implicit in the 4% rule as commonly stated - which I do - it is already baked in and represented by the existence of relatively lower ending portfolio distributions, or (b) an alternate expression of the concern that the future will be worse than the past.  In the latter case, as previously alluded to, I have about a dozen other backups and contingency plans that will work in the non-killer-asteroid case.  In the killer asteroid case, I also have a very simple backup plan, which is to be thankful I got to spend some retired time before getting killed by an asteroid.

As far as inflation goes, I am allocated 90% to stocks, which I believe will serve as adequate protection against high inflation.  (I may be wrong here, I've only thought about this logically and rely on commonly held opinions and haven't done any actual research.)

Overall, another thing that makes me feel safe - and renders this conversation mostly moot in practical terms for me -
 is that my net withdrawal rate is currently under 1%.  What I originally wrote was confusing now that I re-read it, but I don't completely follow my written plan.


Ah, this is a great example where intuition can be misleading. Restarting the clock on your 4% WR is extremely risky.

I wrote some code to run a Trinity-study like simulation of this withdrawal strategy. For a 50-year period, 90% stock / 10% bond allocation, and 4% initial WR, I get 95.7% success rate with only inflation adjustments, but 38.3% by taking the max of (inflation adjustment; 4% of the current portfolio value), which is horrible, and worries me because many people on this forum have similar strategies which biases the success rate (also see my previous thread on the effect of target setting).

Now, we can do a little better by mandating that we can only reset the WR if the value has been maintained for at least N years. Here's the success rate in function of N:

1 year: 38.3%
2 years: 72.3%
3 years: 92.6%
4 years: 92.6%
5 years: 93.6%
6 years: 95.7%
7 years: 95.7%
8 years: 95.7%

So if your portfolio has maintained a minimum value for at least 6 years, I'd say you are relatively safe to increase your WR, but doing this every year is very risky.

Also, I urge you guys to learn to code; it took me literally 10 minutes to have a complete solution and get a better idea of this effect. Otherwise, you can rack your brain for hours and spew wall of text BS all you want, but you'll never know for sure.

secondcor521

  • Walrus Stache
  • *******
  • Posts: 5503
  • Age: 54
  • Location: Boise, Idaho
  • Big cattle, no hat.
    • Age of Eon - Overwatch player videos
Re: Tell me about your inflation adjustments
« Reply #33 on: November 11, 2017, 07:26:45 PM »
Isn't this risky? My understanding is that each starting year has an n% chance of failing to provide the required 4% for the 30 subsequent years. We don't really know what "n" is, but it is non-zero.

Each time you reset, you take the aforementioned n% risk. Reset twice, and your risk is 3n. That has to add up. (note: actually a little less than 3n, but I don't feel like thinking mathy right now.)

Put another way, the 4% rule relies on having unspent excess profits in up years to make up for the down years. If you start to spend the excess profits by resetting, you might not have enough for the down years.

You are essentially starting the clock on a poor early sequence of return risk [one of the two main portfolio killers] fresh every time you do that. If you don't do that and get past say 10yrs or so of retirement with solid returns you can start to feel confident you've avoid one of the main risks.

The other risk is high inflation out running the returns on your portfolio.

Good questions.

First, some background reading:  http://www.retireearlyhomepage.com/popr.html.

The n% that is mentioned is somewhere between 5% and 0%, depending on what parameters and assumptions one makes.  With the assumptions I make - the main ones being flexibility rather than blind adherence and having other backups - I view the risk as 0%.

If you accept the assumptions implicit in the 4% rule as commonly stated, the risk of running out, if there is any at all, would not increase additively.  What would increase is the likelihood of coming closer to running out - i.e., ending up with a low portfolio balance at some time during the 30 year period.  Stated another way, there would be an increase in net overall spending, with the potential for a concern (but not actual risk) about running out later.  To see why this is the case, consider trying to explain why the risk is greater for a person who retired last year and saw their portfolio grow and now has $1M and wants to continue their 30 year retirement with a $40K withdrawal rate, and their friend, who has $1M and wants to start a 30 year retirement with a $40K withdrawal rate.

I know that "sequence of returns risk" is one of the latest things to be (re)discovered by the 4% blogosphere and researchers.  But this risk is either (a) a non-existent risk if one accepts the assumptions implicit in the 4% rule as commonly stated - which I do - it is already baked in and represented by the existence of relatively lower ending portfolio distributions, or (b) an alternate expression of the concern that the future will be worse than the past.  In the latter case, as previously alluded to, I have about a dozen other backups and contingency plans that will work in the non-killer-asteroid case.  In the killer asteroid case, I also have a very simple backup plan, which is to be thankful I got to spend some retired time before getting killed by an asteroid.

As far as inflation goes, I am allocated 90% to stocks, which I believe will serve as adequate protection against high inflation.  (I may be wrong here, I've only thought about this logically and rely on commonly held opinions and haven't done any actual research.)

Overall, another thing that makes me feel safe - and renders this conversation mostly moot in practical terms for me -
 is that my net withdrawal rate is currently under 1%.  What I originally wrote was confusing now that I re-read it, but I don't completely follow my written plan.


Ah, this is a great example where intuition can be misleading. Restarting the clock on your 4% WR is extremely risky.

I wrote some code to run a Trinity-study like simulation of this withdrawal strategy. For a 50-year period, 90% stock / 10% bond allocation, and 4% initial WR, I get 95.7% success rate with only inflation adjustments, but 38.3% by taking the max of (inflation adjustment; 4% of the current portfolio value), which is horrible, and worries me because many people on this forum have similar strategies which biases the success rate (also see my previous thread on the effect of target setting).

Now, we can do a little better by mandating that we can only reset the WR if the value has been maintained for at least N years. Here's the success rate in function of N:

1 year: 38.3%
2 years: 72.3%
3 years: 92.6%
4 years: 92.6%
5 years: 93.6%
6 years: 95.7%
7 years: 95.7%
8 years: 95.7%

So if your portfolio has maintained a minimum value for at least 6 years, I'd say you are relatively safe to increase your WR, but doing this every year is very risky.

Also, I urge you guys to learn to code; it took me literally 10 minutes to have a complete solution and get a better idea of this effect. Otherwise, you can rack your brain for hours and spew wall of text BS all you want, but you'll never know for sure.

Questions for you, in the spirit of seeking to understand:

1.  Did you read the web page I linked?
2.  Did you try answering the thought experiment underlined in my post?

And general responses:

1.  When do you think that I used intuition?  Rereading what I wrote, I believe I was using logic, research...and widely held and respected opinions in the case of my second to last paragraph.
2.  Without seeing your code and data, I'm not sure that what you modeled and what I described are in fact the same thing.  Also, I don't know you or your coding skills so I'm not very confident that the results of your 10 minute exercise should be relied on more than logic, the research from John Greaney, and my reading, thinking about, and discussion of the subject for the past 25 years.
3.  As a side note, I have a BS in Computer Science and have coded off and on for about 40 years.
4.  You missed, or chose to ignore that my actual WR is under 1%.

FWIW, I read your other thread when you wrote it and again just now and didn't find your logic or data there persuasive either.  On this subject, I think you're a pessimist and I'm an optimist, and I think it is quite possible that we both are biased in our views.
« Last Edit: November 11, 2017, 07:35:22 PM by secondcor521 »

gerardc

  • Pencil Stache
  • ****
  • Posts: 767
  • Age: 40
  • Location: SF bay area
Re: Tell me about your inflation adjustments
« Reply #34 on: November 11, 2017, 10:12:04 PM »
Questions for you, in the spirit of seeking to understand:

1.  Did you read the web page I linked?
2.  Did you try answering the thought experiment underlined in my post?

And general responses:

1.  When do you think that I used intuition?  Rereading what I wrote, I believe I was using logic, research...and widely held and respected opinions in the case of my second to last paragraph.
2.  Without seeing your code and data, I'm not sure that what you modeled and what I described are in fact the same thing.  Also, I don't know you or your coding skills so I'm not very confident that the results of your 10 minute exercise should be relied on more than logic, the research from John Greaney, and my reading, thinking about, and discussion of the subject for the past 25 years.
3.  As a side note, I have a BS in Computer Science and have coded off and on for about 40 years.
4.  You missed, or chose to ignore that my actual WR is under 1%.

FWIW, I read your other thread when you wrote it and again just now and didn't find your logic or data there persuasive either.  On this subject, I think you're a pessimist and I'm an optimist, and I think it is quite possible that we both are biased in our views.

1. I skimmed it, but couldn't find success rates for the reset method, only the effect on withdrawal amounts and terminal values, so it couldn't directly answer the question I was interested in, i.e. risk of failure.
2. No

2. Feel free examine my code, you very well might find some error:
Code: [Select]
import json
import numpy as np

START_YEAR = 1871
END_YEAR = 2015
PERIOD = 50
STOCK_FRACTION = 0.9
BONDS_FRACTION = 0.1

# From https://github.com/boknows/cFIREsim-open/blob/master/js/marketData.js
market_data = json.load(file('market.json'))
market_data = {int(year): properties for year, properties in market_data.iteritems()}

def Simulate(start_value, start_year, end_year, initial_withdrawal, reset_years):
  value = start_value
  die = None
  all_values = []
  withdrawal = initial_withdrawal
  for year in xrange(start_year, end_year):
    all_values.append(value)
    if reset_years and len(all_values) >= reset_years:
      maintained_value = min(all_values[-reset_years:])
      new_withdrawal = maintained_value * initial_withdrawal
      if new_withdrawal > withdrawal:
        withdrawal = new_withdrawal
        #print 'new withdrawal: %.1f' % (100 * new_withdrawal)

    properties = market_data[year]
    next_properties = market_data[year + 1]
   
    value -= withdrawal
    if value > 0:
      stock_gains = (1. + properties['dividends']) * (1. + properties['growth'])
      bonds_gains = 1. + properties['fixed_income']
      value *= STOCK_FRACTION * stock_gains + BONDS_FRACTION * bonds_gains
    value *= float(properties['cpi']) / next_properties['cpi']

    if value <= 0 and die is None:
      die = year
    #print year, value
  return {'end_value': value, 'die': die}

def SuccessRate(period, withdrawal, reset_years):
  successes = []
  for start_year in xrange(START_YEAR, END_YEAR - period):
    drawdown_results = Simulate(1.0, start_year, start_year + period, withdrawal, reset_years)
    die_year = drawdown_results['die']
    success = die_year is None
    successes.append(success)
  return {'success_rate': np.mean(successes)}

4. I chose to ignore WR < 1% because nothing interesting would come off of analysing success rates in that regime (obviously it would stay 100%).

I'm not a pessimist nor an optimist, I'm just an accuracy-ist in my estimation of the probabilities of success in those admittedly limited models, that's all. I'm aware flexibility trumps all, and that more accuracy isn't necessarily useful since there is so much uncertainty unaccounted for by the model anyway, plus the limited historical data; I still find that identifying effects like these helps to refine our intuition about the risk of different withdrawal strategies, which can often be wrong.

secondcor521

  • Walrus Stache
  • *******
  • Posts: 5503
  • Age: 54
  • Location: Boise, Idaho
  • Big cattle, no hat.
    • Age of Eon - Overwatch player videos
Re: Tell me about your inflation adjustments
« Reply #35 on: November 12, 2017, 11:11:42 AM »
^ Thanks for the reply.  I appreciate it.

Villanelle

  • Walrus Stache
  • *******
  • Posts: 6657
Re: Tell me about your inflation adjustments
« Reply #36 on: November 12, 2017, 10:43:14 PM »
The only bills that have reliably gone up have been property taxes, cable TV, and water/sewer.
No healthcare inflation?
Ah, good point, I stand corrected.

I don't usually bring up healthcare here (I'm a military retiree on Tricare Prime).  DoD has indexed those "enrollment fees" to the CPI.  Before that they were flat for 15 years of Congressional gridlock.

If healthcare is over 30% of a FI budget then yeah, that'd be problematic. 

In anticipation of other questions, our vehicle insurance has been largely flat for nearly 20 years because we don't carry collision or comprehensive.

Again, everyone has to assess their own personal rate of inflation and also has to choose an asset allocation which will at least attempt to keep up with inflation.  It has to be analyzed in the context of the whole budget for your individual situation, not based on national averages or individual categories.

I mostly agree with the notion that actual inflation is irrelevant and personal choices are going to determine everything.

Im just not sure where that leaves us (husband and me) as far as setting a maximum withdraw amount.  If inflation is %XX, I don't care.  But if my personal spending dictates I withdraw %Y, where do I draw the line at which I must either cut spending or find income?  We are a bit aware from RE, but that's the piece I can't figure out. I'd feel like I'm going to want to know that if I'm wanting/needing to withdraw more than %z.z (or some other number) then I am going to need to find some other solution. 

4% over and over (re-retire every year) is questionable for me.  Beyond my comfort level. But I know myself enough to know I'm going to want some sort of limit or limiting calculation, too. 

Nords

  • Magnum Stache
  • ******
  • Posts: 3421
  • Age: 63
  • Location: Oahu
    • Military Retirement & Financial Independence blog
Re: Tell me about your inflation adjustments
« Reply #37 on: November 12, 2017, 11:03:58 PM »
Im just not sure where that leaves us (husband and me) as far as setting a maximum withdraw amount.  If inflation is %XX, I don't care.  But if my personal spending dictates I withdraw %Y, where do I draw the line at which I must either cut spending or find income?  We are a bit aware from RE, but that's the piece I can't figure out. I'd feel like I'm going to want to know that if I'm wanting/needing to withdraw more than %z.z (or some other number) then I am going to need to find some other solution. 

4% over and over (re-retire every year) is questionable for me.  Beyond my comfort level. But I know myself enough to know I'm going to want some sort of limit or limiting calculation, too.
There's plenty of ways to complicate the variable spending tactics.

You could start with the relatively simple 4%/95% system espoused by Bob Clyatt:  withdraw 4% of the portfolio every year.  It's not adjusted for inflation. Take 4% of the portfolio's balance at the beginning of your current year no matter whether the markets are up or down.  If the next 4% withdrawal is less than 95% of the previous year's withdrawal then you could take 95% of the previous year's withdrawal.
http://www.retireearlyhomepage.com/clyatt.html

Researcher Wade Pfau has 10 other ways to handle the issue:
https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2579123

Retire-Canada

  • Walrus Stache
  • *******
  • Posts: 8684
Re: Tell me about your inflation adjustments
« Reply #38 on: November 13, 2017, 07:30:38 AM »
I mostly agree with the notion that actual inflation is irrelevant and personal choices are going to determine everything.

Im just not sure where that leaves us (husband and me) as far as setting a maximum withdraw amount.  If inflation is %XX, I don't care.  But if my personal spending dictates I withdraw %Y, where do I draw the line at which I must either cut spending or find income?  We are a bit aware from RE, but that's the piece I can't figure out. I'd feel like I'm going to want to know that if I'm wanting/needing to withdraw more than %z.z (or some other number) then I am going to need to find some other solution. 

4% over and over (re-retire every year) is questionable for me.  Beyond my comfort level. But I know myself enough to know I'm going to want some sort of limit or limiting calculation, too.

If you need a formula just use 4% of original stash amount + a bump for CPI every year. Yes your personal inflation may be different than CPI, but it gives you a published number to work with. If your personal inflation is lower than CPI you don't need to spend your limit every year. If your personal inflation is higher than CPI that lets you know you need to look at your investment status and determine if you can take out more than what CPI indicates. It also gives you a signal that maybe your spending needs tweaking.

If you get lucky with a good early sequence of returns worrying about small inflation increases may become pointless once your 'stash doubles.

AdrianC

  • Handlebar Stache
  • *****
  • Posts: 1211
  • Location: Cincinnati
Re: Tell me about your inflation adjustments
« Reply #39 on: November 13, 2017, 02:27:52 PM »
Ah, this is a great example where intuition can be misleading. Restarting the clock on your 4% WR is extremely risky.

I wrote some code to run a Trinity-study like simulation of this withdrawal strategy. For a 50-year period, 90% stock / 10% bond allocation, and 4% initial WR, I get 95.7% success rate with only inflation adjustments, but 38.3% by taking the max of (inflation adjustment; 4% of the current portfolio value), which is horrible, and worries me because many people on this forum have similar strategies which biases the success rate

Very interesting. Each time you reset to 4% you are restarting the 95.7% success rate. Not sure if this is valid, but I notice that 0.957^22 = 0.38, i.e. 22 resets (which is a lot of resets, but believable over 50 years).

3% usually comes up at a 100% success rate. Does that mean restarting the clock on your 3% WR is no problem?

gerardc

  • Pencil Stache
  • ****
  • Posts: 767
  • Age: 40
  • Location: SF bay area
Re: Tell me about your inflation adjustments
« Reply #40 on: November 13, 2017, 05:49:05 PM »
Very interesting. Each time you reset to 4% you are restarting the 95.7% success rate. Not sure if this is valid, but I notice that 0.957^22 = 0.38, i.e. 22 resets (which is a lot of resets, but believable over 50 years).

I calculated that at 4% WR, the average number of resets over a 50-year period is surprisingly only 5.9 (std = 3.9), so I think there is a compounding effect of the resets occurring at market peaks, i.e. very dangerous times to be resetting... Think about it, the only way you'd stop resetting under that strategy, is if there is a long sequence of inflation-adjusted losses, tapping into the sequence of return risks.

For 50-year periods, there are only a handful of starting years leading to failure with constant 4% WR: 1906, 1907, 1929, 1964, 1966, 1969. Most 50-year periods include one of these failure years reasonably early, and chances are you'll reset right before in most cases.

So, basically, a reset is only dangerous if you see a crash soon after and you keep withdrawing the new amount... not the brightest idea... if you revert back quickly, you'll be fine. This is similar to the strategy of waiting for N years of sustained stash amount before resetting, except now you're allowed to reset right away but backtrack if things go haywire.


3% usually comes up at a 100% success rate. Does that mean restarting the clock on your 3% WR is no problem?

Yes, it has to. The only exception I found was playing with borderline 100% success rate, like 3.6% WR, where success rate goes down to 91.5% with resets, but that's only because some recent starting years are skipped (due to not having at least 50 years of data) but fail before the 50 years are up, so the reset method will catch that failure... that's just a limitation of the model.
« Last Edit: November 13, 2017, 05:51:48 PM by gerardc »