Mostly because that didn't exist fifteen years ago, and neither did the present-day Monte Carlo sites.

"Over forty years, though, those ups and downs will be smoothed to some sort of a long-term performance."
Since we are talking about retirement, forty years should be taken as the upper limit of possible scenarios if one starts at age 60. Perhaps this is the issue as time is constrained. What is the least amount of time a spreadsheet or analysis can be trusted over? Or, maybe it is a period of market cycles?
May you and everyone here live to be 100 years old.
BTW, Warriors take the first game. I hope they go all the way. It's been a long time.

@MJG: Of course a home-built spreadsheet is not going to be able to introduce the year-to-year possible variables of a sophisticated Monte Carlo platform, which obviously takes an entirely different approach, running thousands of possible scenarios and expressing the results as a measurement of probability. Either approach, however, suffers from the fact that the input variables are predicated upon past financial experiences, and neither can anticipate something new coming down the road.
The ss approach fundamentally evaluates a financial situation over a period of time anticipating that over a long period of time some sort of average will be produced. Yes, some years radically up, yes some years radically down. Can the ss identify those specific events in advance? Of course not. Over forty years, though, those ups and downs will be smoothed to some sort of a long-term performance.
By varying the values of the input variable manually, it's not all that hard to establish a performance range from pessimistic to unrealistic, and try to pick a course somewhere down the middle.
Construction of the various scenarios is up to the user of the spreadsheet. Those of us, including you and me, whose parents survived the "great depression", should have the sense to expect major financial fiascos, and condition our input parameters accordingly.
The ss also maintained a cash reserve, completely separate from the other assets, that was designed to compensate for zero income during a four year period. Now obviously, that is an improbable situation. However, over a forty year stretch, that anomaly was smoothed into the overall considerations. You evidently want me to justify my work on some sort of rigid engineering basis. It wasn't engineering at all- it was a very flexible tool that allowed me to see the results over time of a complex financial situation, by adjusting many input variables, with a reasonable chance that the results would give at least some insight to the future. It clearly suggested the spending limits that might be anticipated under a wide variety of circumstances, and for my purposes, succeeded very well. I'll leave the engineering to the designers of our new Bay Bridge section.
I sincerely hope that I have explained what I did to your satisfaction. If not, I give up.
Out.

Hi Dex, Hi Old Joe.
I’ll try to be succinct and reply to both your recent posts.
Dex, sorry I missed your post requesting my input data to the Monte Carlo analysis I did. Regardless, I actually answered your question in my original submittal of the Monte Carlo results that I reported.
For your convenience, I’ll repeat them here. I did 12 simulations in about 5 minutes with a $13K drawdown schedule. I did 6 cases for a 185K initial portfolio value. My baseline was a 7% average annual return with parametric standard deviations of 10%, 14%, and 18%. I repeated the same standard deviations for a 6% annual portfolio return rate. Portfolio survival rates were atrocious.
So I repeated the same return/standard deviation 6 sequence series using a composite portfolio that included all your reserves. For that series, the initial portfolio value was $350K. with the same withdrawal rate. Survival rates improved remarkably, but still never made a high of 80%. That’s a much better outcome, but still represents a not too attractive survival prospect.
I did the analysis using the Monte Carlo tool on the Portfolio Vizualizer.com website. There are better simulators accessible on the web, so don’t interpret my usage as an endorsement. It was available and easy to input.
Old Joe, sorry for the delay. Your response still misses the main thrust of my question. Allow me to rephrase the question.
As you worked down your Spreadsheet on a line-by-line (equivalent to a year-by-year set of entries) basis, annual returns for your various investments were required. What inputs did you make? Did they go up and down each year to reflect the schizophrenic nature of market annual returns?
Your response implies that you used some sort of compounding formula. Specifically, you said: “It started with the amount then available in the cash/bond/equity baskets, and merely projected forward on a compounding basis, year by year, for forty years.”
Your reply suggests that a simple compound return rate was postulated. If so, what was the assumed rate? Was it changed during the 40 year projection period? If so, how were the rate change decisions made? How were the annual investment return variabilities handled?
These are all complex issues because of return’s variability. That’s the specific area whereby Monte Carlo analysis struts its stuff. Random selections are automatically made within the guts of the code that are unbiased and reflect whatever statistics the user wants to explore.
If my question is still too vague, please advise me so and I’ll try again. The guesstimated annual portfolio return on a yearly basis is the crux of the problem. Without multiple Monte Carlo simulations as a tool to model the annual reward variability, portfolio survival prospects are likely to be overstated. A compound rate approach does not capture the wild vicissitudes of market returns.
Thank you for this exchange, but you have not answered my question. Perhaps other MFO members could reframe my question to make it clearer to you.
Best Wishes to both you guys.

I've very much enjoyed this thread and the previous thread on social security. Being 61 years old myself and contemplating retirement, I think about all these comments daily.

You post is an example of why I do not think early retirement or even a comfortable retirement is possible for many - the younger the less likely.
Not enough savings, no pension, SS pushed out, investing properly (whatever saving they have).
Then add into the equation that eventually we will have something like a VAT, increasing expenses over the years.

@MJG: I have no idea what exactly you are asking for. The portion of the spreadsheet that dealt with the retirement projections drew data from the main part of the ss that I still use to keep track of financial and fund performance. It started with the amount then available in the cash/bond/equity baskets, and merely projected forward on a compounding basis, year by year, for forty years. It was also possible to change the cash/bond/equity ratios so as to observe how projected performance would vary. I no longer maintain that retirement section, as it required a lot of time and computing horsepower and is now unnecessary. Having survived 2008 in decent shape, I no longer worry about the big picture. As it turns out, my projections were so conservative that to this point income has exceeded spending, and we have yet to draw down anything from the investment pool.
One of the key variables was the amount of income desired annually, or every-other year on an alternating basis. That entry was also automatically increased going forward on a compounded basis, dependent upon the input inflation rate. The ss then calculated income from retirement and SS (reduced by income tax) going forward, and if the desired income exceeded the actual income projected, would then draw down cash, bond, and equity positions to supply the extra amount needed, and then ran the adjusted values forward. Of course, simply reducing the desired income level also had a major impact. By varying the inputs for the principal cash/bond/equity drawdown ratios, it was possible to see which combination gave the most satisfactory results.
The ss retirement section was a major ongoing investment in time, and underwent frequent revisions and modifications as new complexities presented themselves. Now, I've given you all the time on this that I intend to, and certainly more than you deserve given your many nasty comments over the years.
For those unable or unwilling to go to that much trouble, I'm sure that Monte Carlo exercises are much better than nothing.

I've very much enjoyed this thread and the previous thread on social security. Being 61 years old myself and contemplating retirement, I think about all these comments daily.
I once thought early retirement sounded so wonderful and that it seemed by simple compounding on a spreadsheet that it was "easily" doable. 55 was my goal. All the numbers seemed to work. Hell, my portfolio in the 90's, having no idea what I was doing investment wise, was making 10-15% a year. My wife and I would have a million dollars by age 56. Then came the great recession. A wake up call.
I'll offer my thoughts to what I think I've learned while contemplating retirement. This is just my experience - my 2cents FWIW:
- Pay special attention to life expectancy charts. Yes, you likely will live longer then you ever imagined. What do you think those biotech stocks many here invest in are working on - making you live forever, so to speak. LTC and HC will cost even more then today.
- Early retirement in foresight was a pipe dream of the baby boomers. It was based on "things are different now - we are a special generation" - "this isn't your fathers Oldsmobile". Guess what? If it wasn't good for your father or grandfather, it probably isn't good for you. The government made 65 retirement age for a reason - statistically.
- I did much of my spreadsheet work using 8% as a return on retirement investments. Why not. It was the average return of a balanced portfolio. Another pipe dream. Any calculation you make going forward, use a realistic returns number. I would suggest 6% would be on the aggressive side. More realistic, use 5-5.5% in your calculations. Hey, BobC who I highly respect has said the same in many of his posts.
- Monte Carlo software programs are absolutely a great tool to show you where you stand. Are they the beat all-end all? No, of course not. There are way too many variables in life. But I believe in probability. Yes, history repeats itself, especially in arithmetic. Throw real (conservative) numbers into these programs. If the truth hurts - pay attention - it's your future. It's math, not voodoo. The results are a guide. Not the absolute.
- And lastly, I believe fixed annuities have a place for many retirees, very much so. If only in having the piece of mind that you will be able to pay your bills in any economy, it could be worth a look. Blasphemy on this board? Maybe, but I believe a pension, either purchased or through work is very important.
Good luck to all in this journey.

"Yes, that was another adjustable variable of the "horse-drawn hand-executed spreadsheet": both the severity and the length of the downturn."
Yep. It makes sense to run the numbers on a annual basis. Also having exceptionally good years can balance out the issue but that is a longer term scenario. One way is to sweep profits into a conservative investment to hold for that rainy day. That's easy if the good days come before the bad.

Water plays seem to me the ultimate story stocks. It's an easy story to understand given our droughts, warming climate, and probable competition for a valuable resource; however, the 4 ETFs (PHO +51%; CGW +73%; PIO +52%; and FIW (65%) investing in the water theme have all underperformed SPY (+91%) over the last five years. Granted, several of the holdings of the funds are foreign stocks which have underperformed as a group during that time period. I lost money on Flexible Solutions (FSI), maybe the ultimate sucker's story stock: the company has a product that sits on the surface of reservoirs or swimming pools and retards evaporation. Good stories don't always translate into good returns. Maybe water stocks will have their day in the sun, if the metaphor doesn't pain you.

Hi Old Joe,,
Your reply still does not explain how you made your inputs. I'm still puzzled. I'm not as smart as you claim I think I am.
Please give us a sample input for a couple of years. For example, what would be your estimated equity input for 2016, for 2017, and for 2018? How were they specifically determined on a year-by-year basis?
My replies to your insidious and continuing charges are purely defensive. I merely react; I never initiate. That's a major distinction between our actions.
Best Wishes.

"The markets do look vulnerable here. a big downturn in retirement would be crippling. The recovery rate of the investments in the portfolio could easily double that four years into eight years. Of course, we all adapt our portfolios and our spending habits if such a downturn comes."
@JohnChisum: Yes, that was another adjustable variable of the "horse-drawn hand-executed spreadsheet": both the severity and the length of the downturn. Again, bias towards the Murphy factor was emphasized, even if by hand. I have to tell you, though, that 2008 resembled the worst-case of my projected downturn.

@Dex, Agree on that. Each person develops their own answer. There is not one set figure or style for everyone.
The markets do look vulnerable here. a big downturn in retirement would be crippling. The recovery rate of the investments in the portfolio could easily double that four years into eight years. Of course, we all adapt our portfolios and our spending habits if such a downturn comes. Top Ramen is still inexpensive.

@Dex, The number that jumped out at me was the 4 years of cash for expenses. Isn't the general consensus calling for 18 months to 2 years? 4 years seems high to me and could be a drag on your portfolio.

That depends upon your comfort level - 4 would cover many downturns. Also, not 'cash' near cash - short term investments - you could ladder treasury bonds.

@Dex, The number that jumped out at me was the 4 years of cash for expenses. Isn't the general consensus calling for 18 months to 2 years? 4 years seems high to me and could be a drag on your portfolio.

That depends upon your comfort level - 4 would cover many downturns.

"How were the yearly annual return estimates entered into the various equity/fixed income/cash inputs as a function of Spreadsheet time?"
The answer is given above, if you care to read it. The allocation ratios between cash, bonds, and equity were variable. Each of those components was automatically calculated on a compounding basis via forty year tables. You vary the input ratios and income assumptions for each income class, the tables automatically ran that via compounding for forty years. I again note your snide description of, as you dismissively term it, a "hand-executed Spreadsheet".
With respect to your comment on "initiating this wasteful activity", I suggest that a word count of your continuing rants and diatribes might be illuminating. Mostly Just Gas.
"Best Wishes" to you, and your Trojan Horses too.

@Dex, The number that jumped out at me was the 4 years of cash for expenses. Isn't the general consensus calling for 18 months to 2 years? 4 years seems high to me and could be a drag on your portfolio.
In general, these types of questions are hard to answer due to all the variables and the unknown future expenses. While some here discredit the Monte Carlo calculators, I find them worthwhile as a confirmation tool. In these important decision making processes, it is better to use all the tools available.

Hi Old Joe,
Your contribution is a first-rate submittal that nicely details your applied procedure when making your retirement decision. Thank you for your excellent effort.
That procedure includes many elements that are embedded in a Monte Carlo analysis.
The question that was asked many years ago, and even today remains unanswered is: How were the yearly annual return estimates entered into the various equity/fixed income/cash inputs as a function of Spreadsheet time?
You were making many future uncertain estimates. For example, how were return’s variability handled for each year in your hand-executed Spreadsheet? History matters in determining a portfolio’s survival likelihood.
Monte Carlo does this arduous, challenging task with a random number input routine controlled by various statistical formats.
Old Joe, you continue to assert that I claim some superior intellect. In your recent post you said when referring to yourself: “yes, MJG, an assumption!, indicating of course a lesser intellect than you obviously possess”. Sarcasm that doesn’t work.
That’s just plain wrong! I have never claimed and will never claim a superior intellect. Your false charges are Trojan Horses designed to denigrate and to distract. Please address the substance of my posts instead of personal attacks. Your campaign is failing. Seasoned MFOers recognize who is initiating this wasteful activity.
Best Wishes.

"Generally, a bottoms up approach is better i.e. budget, net worth, pension, SS etc."
First, the planning and accumulation of data: I designed a worksheet which divided our expenses into various categories, including those which were optional and those which were absolutely required. I didn't buy into the "oh, once you're retired you're expenses will be lower" theme which was so popular at the time.
I did the same thing as my budget shows.
Once a month we tabulated all of our expenses, and filled out that worksheet. After three or four years we had a pretty good idea of our spending patterns. Assuming (yes, MJG, an assumption!, indicating of course a lesser intellect than you obviously possess) that that pattern was a reasonable model for the future, I then spent many, many hours designing a spreadsheet that could consider our net worth, lack of debt, anticipated incomes from pensions, investments and SS, ongoing expenses, and special expenses.
See the link above for the people who retired early on $500K

Basically, we both did the same thing and came to the same conclusion.

"Generally, a bottoms up approach is better i.e. budget, net worth, pension, SS etc."
@Dex: You've done it now. I prescribed something similar many years ago, based upon my personal experience. Since it didn't involve elaborate financial engineering and theories, but rather common sense based upon life experience, our would-be master of the financial universe regarded that as inherently inferior to his vaunted Monte Carlo approach. He is under the misapprehension that I denigrate Monte Carlo, which is in fact inaccurate. Because he is evidently incapable of contemplating that anything he believes may not be factual, it has led to the present acrimonious exchange of opinion. You may be in danger of joining me in that category.
First, the planning and accumulation of data: I designed a worksheet which divided our expenses into various categories, including those which were optional and those which were absolutely required. I didn't buy into the "oh, once you're retired you're expenses will be lower" theme which was so popular at the time.
Once a month we tabulated all of our expenses, and filled out that worksheet. After three or four years we had a pretty good idea of our spending patterns. Assuming (yes, MJG, an assumption!, indicating of course a lesser intellect than you obviously possess) that that pattern was a reasonable model for the future, I then spent many, many hours designing a spreadsheet that could consider our net worth, lack of debt, anticipated incomes from pensions, investments and SS, ongoing expenses, and special expenses.
It allowed the entry and consideration of variables for inflation, investment income according to various asset allocation percentages (cash/bonds/equity), special expenses, and major financial disasters, a la 2008, which could be inserted at any desired year. It allowed for two expense modes: one allowed variable optional expenses each year (travel and so forth), and an alternative mode which held the expenses to the mandatory for one year, but allowed for additional optional expenses the second year, and so forth alternating every other year. Being a pessimist by nature, every facet of the calculations was designed to utilize the least favorable case. For instance, it assumed that cash would generate less interest than actual inflation, that bond income would merely break even, and that equity income would be in a variable input range from 3 to 5% above inflation.
For income source and expenses it generated a series of compounding tables which ran out for forty years, starting at age 60, which would have taken us to 100 years of age. From all of the subsidiary tables (each of which, using compounding calculations, took about one page of the spreadsheet) it generated a master table showing the remaining investment assets year-by-year.
It was prescient, in that 2008 struck at the least favorable possible time in our case: a few years after we retired. Because of the deliberate pessimistic bias of the calculations, we've weathered that disaster, and are still in excellent financial condition.
If you want a solid substantial approach based upon known variables, rather than Monte Carlo assumptions, I can recommend the spreadsheet approach, based upon personal experience.
Regards- OJ

Hi Dex,
It appears that you have made a decision that is comfortable for your circumstances. Good!
I wish you well and hope that your plan is successful. You were very wise to include safety factors in your wealth component breakout with the inclusion of “near cash” and “contingency” funding.
I surely do not want to rain on your parade, but I did a few exploratory Monte Carlo simulations for a 30-year retirement period to test the robustness of your analysis that included Social Security benefits. Warning red flags went flying.
I did the analyses using your postulated 7% annual return. I examined return standard deviations that ranged between 10% to 18%. Results were disastrous for the 185K investment portfolio by itself. Projected survival rates were at lower mud levels. There is an extremely high probability that you will need to heavily dip into your near cash and contingency components.
Even assuming a $350,000 dollar portfolio earning a 7% annual average return rate, the survival odds never reach an 80% level. That’s a risky route by my compass.
Given the meager market returns in recent years, an assumed 7% annual return rate seems a bit bushytailed. I briefly examined a 6% annual average return; not surprisingly, the estimated survival likelihoods went even further South.
Obviously, you will follow your analytical results and your gut feelings. I proffer only a caution, Take a little time to more fully evaluate your circumstances. You might want to reassess your analysis including portfolio return’s variability as a major element in that rework.
Whatever action you finally implement, I do wish you a relaxed, happy, and trouble-free retirement that permits you to complete any bucket list that inspires you.
Best Wishes.