The "How much I need to retire" thread

[TexasHusker]

The short answer is, it depends.

But you don't have to reinvent the wheel to answer the question. Use the Monte Carlo simulation, put in all of your parameters, and voila! There's your answer!

https://www.portfoliovisualizer.com/...rlo-simulation

Or if you want something a bit less complicated, there's this tool:

https://www.firecalc.com/

[corn18]

Those are my 2 goto tools. But they have their limitations.

Monte Carlo sims

Any Monte Carlo sim trying to model market returns has a big problem: market returns are not normal. They do not fit any curve unless you do unnatural things with the sim. That results in the best results being much better than historical and the worst results being much worse (this has come to be known as the "tail"). The mean seems to do pretty well. What does this mean to the future retiree? If you are chasing a 100% success rate in a Monte Carlo sim, you will have to save (aka work) a lot longer to hit that number. Or have more rentals or haircut joints to produce more income. So, I use my Monte Carlo sim to see what I need to get a 95% success rate. While that sounds scary that 5% of my 1,319 runs failed, the 95% success rate seems to match a 100% success rate using historical data. And all Monte Carlo sims have this issue. Fidelity, the Flexible Retirement Planner, Portfolio Visualizer... all of them. There have been doctoral thesis written on the subject with no good solution.

Historical Modeling

I also have a historical model in my spreadsheet. It uses data from the Trinity study (the original 4% rule) through 2020. The problem with all historical sims is the oft cited argument that you cannot use past performance to predict the future. While I agree with the premise, I disagree that the approach is not useful. Most of the people/posters that poo poo the historical models are not anywhere close to retirement and they don't have to make a decision soon or now whether they have enough to retire. Easy to throw stones at the guys/gals in the arena. Or they are the type that insist that you should just save enough to have 50x expenses (2% withdrawal rate) and don't worry yourself with historical returns. That's fine if you are rich or you want to work for the rest of your life and die 3 weeks after you retire. I could have easily hit 50x @ age 60 or so had I kept working. But I am just fine with 4% and retiring now @ 55. The other issue with historical modeling is what historical data to use. Firecalc is good because it gives you the option to pick what feeds the model. The Flexible Retirement Planner lets you pick your asset allocation, but not adjust types of equity or bonds. This can really effect the outcome.

So what do I use? I use them all. Once I hit 95% in the historical models (about 90% in the Monte Carlo models), I was ready to pull the plug. Certainly a lot riskier than a typical bogelehead or someone on early-retirement forums. I ended up at 100% on both models with room to spare, so it was easy to make the transition. I was going to retire either way, but 100% is nice.

And I am retired. I do nothing. No two hours a week running my rentals. No one hour a week caring for my businesses. I am responsible for nothing. I get a military pension check every month. I am retired.

[disneysteve]

I needed to hear this.

I mainly look at FireCalc and keep playing with it to make sure we're at 100% but I know in my head that we don't truly need to be at 100% to be "ready". If we're at 98 or 95 that's close enough. If there's a down year or things seem a bit tight, there's enough wiggle room in the budget that we can adjust accordingly.

Another factor that is often overlooked is sequence of spending. Studies always show that retirees spend more in the early years of retirement than they do later. I think it's after 70 that the spending tends to start slowing down. My mom is 90 and barely spends anything outside of necessities. She's not dining out and taking cruises and such anymore. She hasn't even driven a car for several years. I'm sure her spending today is a fraction of what it was 20 years ago. But the calculators tend to assume your spending will be steady and increase annually with inflation.

[Nutria]

This seems... flawed. You can't forecast how much you need for years of retirement without knowing how much you spend right now.

Take a copy of your current budget, and modify for retirement (removing items obsoleted by retirement, adding new categories, and modifying existing ones). Now you've got a monthly retirement budget.

Then take your investment portfolio, subtract 12x the monthly retirement budget, and increase the remainder by whatever growth percent you decide on.

Repeat that in a spreadsheet every year (remembering to also increase the budget every year at an inflation rate).

That'll tell you how many years your portfolio will last. (Of course, if you portfolio is big enough, then the annual growth will be larger than your yearly expenses and so it'll last in perpetuity.)

[TexasHusker]

The other problem is the timing of the withdrawals. Let's say you entered the year 2000 with a cool $2 million, invested it in an S&P 500 fund, and decided to withdraw 4 percent. The next decade you would have gotten slaughtered. You would have recovered some of it this past decade. But if the current decade shapes up like the 2000-2009 decade, you're dead broke in 8 more years. I admit I am not Excel expert, so someone jump in if I get a formula wrong.

Bottom line: A lot of people don't realize that the S&P 500 has lost value in 4 of the last 9 decades. That's 44 percent of the time. 56% of the time, you're a winner! I suppose that is somewhat better than the 47.5% odds at the Blackjack table if you are a practiced player. So flip a coin, place your bets, and hope for the best if you go that route.

			$2,000,000	starting amount
2000	-10.14%		$1,717,200
2001	-13.04%		$1,413,277
2002	-23.37%		$1,002,994
2003	26.38%		$1,187,584
2004	8.99%		$1,214,348
2005	3.00%		$1,170,778
2006	13.62%		$1,250,238
2007	3.53%		$1,214,372
2008	-38.49%		$666,960
2009	23.45%		$743,362
2010	12.78%		$758,364
2011	0.00%		$678,364
2012	13.41%		$689,333
2013	29.60%		$813,375
2014	11.39%		$826,018
2015	-0.73%		$739,988
2016	9.54%		$730,583
2017	19.42%		$792,463
2018	-6.24%		$663,013
2019	28.88%		$774,491
2020	16.26%		$820,423
2021	13.08%		$847,735
2022	-10.14%		$681,774
2023	-13.04%		$512,871
2024	-23.37%		$313,013
2025	26.38%		$315,586
2026	8.99%		$263,957
2027	3.00%		$191,876
2028	13.62%		$138,009
2029	3.53%		$62,881
2030	-38.49%		-$41,322
2009	23.45%		-$131,012

[Jluke]

^^^
your last row says 2009

[TexasHusker]

^^^
your last row says 2009

Oh s***

[corn18]

This is where things go sideways fast. My models use total returns (cap gains + dividends) and are in real dollars not nominal. I have no idea what data you used for your example. But I have total real returns for stocks and bonds going back to 1891 in my model. I could run any scenario you like if you are serious. Your threads usually turn into nonsense.

BTW, a 100% S&P 500 portfolio fails terribly with a 4% real withdrawal rate. You would have to be an idiot to retire with that portfolio.

[TexasHusker]

Your threads usually turn into nonsense.

Corn, the thing I like about you, is that you’re always so coy.

[Nutria]

The other problem is the timing of the withdrawals. Let's say you entered the year 2000 with a cool $2 million, invested it in an S&P 500 fund, and decided to withdraw 4 percent. The next decade you would have gotten slaughtered. You would have recovered some of it this past decade. But if the current decade shapes up like the 2000-2009 decade, you're dead broke in 8 more years. I admit I am not Excel expert, so someone jump in if I get a formula wrong.

It looks like you're broke in 30 years.

[TexasHusker]

It looks like you're broke in 30 years.

correct

[corn18]

One of my favorite charts.

[Nutria]

My calculations show that -- after 20 years -- you'd be richer than at the start.
My data comes from https://dqydj.com/sp-500-return-calculator/ and includes dividends.

Year	Start Bal	BOY Withdrawal	S&P Adjustment	EOY Balance
2000	$2,000,000	$80,000	-5.756%	$1,809,485
2001	$1,809,485	$80,000	-12.821%	$1,507,748
2002	$1,507,748	$80,000	-20.187%	$1,139,529
2003	$1,139,529	$80,000	22.263%	$1,295,412
2004	$1,295,412	$80,000	12.816%	$1,371,179
2005	$1,371,179	$80,000	7.086%	$1,382,672
2006	$1,382,672	$80,000	14.267%	$1,488,524
2007	$1,488,524	$80,000	6.313%	$1,497,444
2008	$1,497,444	$80,000	-39.233%	$861,338
2009	$861,338	$80,000	30.030%	$1,015,974
2010	$1,015,974	$80,000	14.017%	$1,067,169
2011	$1,067,169	$80,000	2.105%	$1,007,949
2012	$1,007,949	$80,000	16.799%	$1,083,835
2013	$1,083,835	$80,000	29.715%	$1,302,125
2014	$1,302,125	$80,000	15.863%	$1,415,991
2015	$1,415,991	$80,000	2.038%	$1,363,218
2016	$1,363,218	$80,000	11.731%	$1,433,752
2017	$1,433,752	$80,000	20.912%	$1,636,849
2018	$1,636,849	$80,000	-1.821%	$1,528,499
2019	$1,528,499	$80,000	26.153%	$1,827,325
2020	$1,827,325	$80,000	18.499%	$2,070,563

If BOY withdrawals increase every year by 3%, you still have $481K left over.

Year	Start Bal	BOY Withdrawal	S&P Adjustment	EOY Balance
2000	$2,000,000	$80,000	-5.756%	$1,809,485
2001	$1,809,485	$82,400	-12.821%	$1,505,655
2002	$1,505,655	$84,872	-20.187%	$1,133,970
2003	$1,133,970	$87,418	22.263%	$1,279,546
2004	$1,279,546	$90,041	12.816%	$1,341,952
2005	$1,341,952	$92,742	7.086%	$1,337,729
2006	$1,337,729	$95,524	14.267%	$1,419,430
2007	$1,419,430	$98,390	6.313%	$1,404,437
2008	$1,404,437	$101,342	-39.233%	$791,852
2009	$791,852	$104,382	30.030%	$893,917
2010	$893,917	$107,513	14.017%	$896,634
2011	$896,634	$110,739	2.105%	$802,438
2012	$802,438	$114,061	16.799%	$804,018
2013	$804,018	$117,483	29.715%	$890,539
2014	$890,539	$121,007	15.863%	$891,603
2015	$891,603	$124,637	2.038%	$782,596
2016	$782,596	$128,377	11.731%	$730,966
2017	$730,966	$132,228	20.912%	$723,946
2018	$723,946	$136,195	-1.821%	$577,048
2019	$577,048	$140,280	26.153%	$550,995
2020	$550,995	$144,489	18.499%	$481,706

[disneysteve]

My calculations show that -- after 20 years -- you'd be richer than at the start.

If BOY withdrawals increase every year by 3%, you still have $481K left over.

And that assumes a 100% S&P 500 stock portfolio which certainly isn't how a retiree should be allocated. I don't really think anyone of any age should be allocated that way.

The whole point of the Trinity study and 4% withdrawal rate is to maximize your odds of not outliving your money. It accounts for crashes and bear markets but also used a 60/40 portfolio if I'm not mistaken. It might have been 50/50. I'm not sure. A quick search didn't answer that question. I'm sure someone knows.

[Nutria]

And that assumes a 100% S&P 500 stock portfolio which certainly isn't how a retiree should be allocated. I don't really think anyone of any age should be allocated that way.

The whole point of the Trinity study

And my point was that TH's spreadsheet is flawed, not to state that 100% S&P 500 is a Good Idea.