## Retirement Withdrawal Calculator

Will you outlive your money given your savings, withdraw ratio and portfolio allocation? This calculator simulates how a retirement nest egg would fare using historical data.

Related to this calculator, check out our Saving for Retirement Calculator and Portfolio Allocation Calculator.

This calculator generates simulation runs for each year of data in our historical dataset (1928 - present) based on what you enter above. It is useful for comparing portfolio allocation outcomes, realistic withdraw rates, and setting a savings goal. It outputs the percent of time the simulated nest egg stayed above water or ran out of money.

Field Summary:

• Length of Retirement - how long do you expect to live after you retire? Age 100? Age 85? To be conservative, enter a higher value.
• Savings at Retirement - how much money you have saved up before retiring and starting to draw on your nest egg.
• Withdraw Amount - how much you plan to withdraw in the first year (this amount does not count social security, pensions, or other income sources, just the amount you plan to take out of your nest egg each year). During the simulations the withdraw amount is adjusted for inflation.
• Withdraw Percent - the percent of your nest egg you plan to withdraw in the first year to live on, see Withdraw Amount notes.
• Goal - how much you would like to have in the end. The calculator displays the percentage of simulations that exceeded the goal. This aims to answer the classic question "Will I die rich or die broke?" .
• Portfolio Strategy - you can pick from the predefined allocations of Stocks, Bonds, and Cash, or enter your own.
• Stocks - percent of funds to put into the US S&P 500 Index.
• Bonds - percent of funds to put into 10 Year US Treasury Bonds, with returns including coupon and price appreciation.
• Cash - percent of funds to put into a 'risk free' investment. The simulator uses the returns of 90 Day US Treasury Bills. A similar rate is attainable with an FDIC insured money market account.

Some insights into the results this tool unearths:

1. You may be surprised to find that an "All Stock" portfolio is risky, but often not as risky in the long run as an "All Cash" portfolio. Why is this? Historically speaking, compared to cash, stocks have done a much better job of a) growing and b) keeping up with inflation. Try switching your portfolio to All Cash and watch how the graph looks like a comb over instead of a mountain.
2. The average is usually higher than the median (mid-point) because the distribution of returns skews to the left. Another way to say this is: high returns are relatively large but infrequent, while lower returns are more common. The bar chart showing the distribution of ending balances is weighted to the left in most setups.
3. Looking at the simulation high and low numbers (which can be mind bogglingly wide), luck plays a role in the individual's outcome. The year in which you retire could make a huge difference, but you won't know until it is too late. For example, the simulations show that people who retired in 1975 were in great shape because of the economic booms in the 80s and 90s, but the people who retired in the mid 60's didn't live long enough to recoup their early losses.
4. Diversification reduces risk but also reduces upside. Portfolios that blend stocks and bonds do a good job of bringing up the low end of the simulation, based on the historical data. This is to a point though... too little stocks in the portfolio and the numbers drop like a rock.
5. Each simulation plods along every year in the way it is programmed to. It always rebalances every year and withdraws the right amount. In reality, life happens, emergencies happen, and investors panic and sell in bad times. This calculator is what they call a 'disciplined investor' in that it doesn't panic during the bad times, nor does it go crazy and buy a diamond studded phone case in the good times.

How the simulations work:

This calculator is NOT a Monte-Carlo simulator in that it does not generate any fake or random data. Instead, this calculator uses historical data and backtests against it. Essentially it replays what happened in each of the years in the dataset given the inputs and then summarizes the results.

For a 30 year retirement period, this calculator will run a simulation from 1928 to 1958, then it will run a simulation from 1929 to 1959, then from 1930 to 1960, and so on. In simulations that go beyond the present year, it will wrap back to 1928 and count up from there. In this sense, the effect of the great depression is factored in for early and late starting years.

Each individual simulation computes returns by stepping through the years (eg 1928, 1929, ... 1958) and performs the following each year:

• Calculates the change in value in the portfolio.
• Adjusts the annual withdraw amount for inflation based on the CPI (consumer price index) for that year.
• Updates the portfolio balance by adding the change in value and subtracting the withdraw amount.
• Rebalances the portfolio.

Regarding the Annual Withdraw Percent:

The famous Trinity Study suggests a 3-4% withdraw rate is a good place to be: "If history is any guide for the future, then withdrawal rates of 3% and 4% are extremely unlikely to exhaust any portfolio of stocks and bonds...". Keep in mind, this calculator and the Trinity Study rely on backtesting, which means historical data is analyzed for a 'best fit'. Yesterday's best fit may turn out to be a very poor fit in the future. Nobody really knows for sure what will happen next. In general terms, a lower withdraw rate means the nest egg with last longer.

Simulations with a high withdrawal rate can cause the balance to go negative. Once a simulation's balance goes to zero subsequent returns from investments have no effect.  However, the annual withdrawal amount continues to increase with inflation. A negative ending balance means borrowing was required to cover the withdrawals.

Notes on Inflation:

The numbers this calculator outputs are not inflation adjusted, they are nominal values. The numbers don't translate to actual purchasing power in the starting year of the simulation. However, this calculator does adjust the withdraw amount by the CPI each year of the simulation. For example, given a 30 year retirement and an initial withdraw amount of \$50,000, the simulation starting in 1975 would increase the withdraw amount all the way to \$181,440 in 2005 (in the final year of that simulation run) based on the change in CPI.

Historical Data Used:

The data this calculator uses can be found here.

Past Performance Does Not Indicate Future Results:

Again, this calculator does backtesting. Past performance does not guarantee nor indicate future results. These results are based on simulated or hypothetical performance results that have certain inherent limitations. Unlike the results shown in an actual performance record, these results do not represent actual trading. Also, because these trades have not actually been executed, these results may have under-or over-compensated for the impact, if any, of certain market factors, such as lack of liquidity. Simulated or hypothetical trading programs in general are also subject to the fact that they are designed with the benefit of hindsight. No representation is being made that any account will or is likely to achieve profits or losses similar to these being shown.

DISCLAIMER: This calculator is provided for educational purposes and should not be considered financial or investment advice. We have checked the equations and code used and we think they are right. However, we offer you no guarantee of accuracy. If you find a bug please let us know so we can fix it for you!

To share this calculator by URL or embed it on your own website see the 'Share / Embed' button above next to the results bar after hitting Submit.

michael Murray Jan 22 2020 18:21 UTC

Thank you for implementing my goal suggestion options of stay above zero, keep initial balance, custom. it looks really great and I love the feature. you guys are great.

Vanguard Junkie Jan 22 2020 18:28 UTC

It is interesting that if you start with a sufficient nest egg (\$2M), using a 3.8% - 4% withdrawal rate the odds of staying above zero and maintaining the initial balance are pretty close. Setting myself up to avoid going broke means I'll probably be leaving my kids a BIG chunk of money when I kick the bucket.

Stephen Jan 26 2021 19:59 UTC

Great calculator. Using to show my mom what asset allocation she should have in retirement. Thanks.

Simon May 20 2021 19:35 UTC

Excellent, the one thing missing is the ability to “add money” after a set year, for example you may withdraw the same amount every year but after year 10 a set final pension scheme (set annual amount given automatic) may kick in therefore reducing the amount you still have to withdraw

Tim Jun 6 2021 21:39 UTC

I am not terribly knowledgable about tax implications for retirement or retirement calculators. I don’t understand how tax assumptions are utilized with this calculator. This calculator yields results that are wildly more optimistic than other calculators I’ve played with, one possible explanation is how taxes are handled? Thank you for any clarification….

Jun 7 2021 17:38 UTC

Tim - correct, tax implications are ignored in this calculator.

So if you are taking out \$60k per year, that is pre tax money.

Tax issues get complicated fast in terms of Roth accounts (not taxed), required minimum withdrawals (which force a person to take money out, possibly more than they need), long term capital gains on assets held in taxable accounts, etc etc... That is well beyond what this free calculator can offer.

"wildly more optimistic than other calculators I’ve played with" -> let us know what those might be so we can compare.

Sep 14 2021 13:09 UTC

It would be helpful to have a simpler explanation of the data or general geographic mix. I ran an all stock solution for 1 year with a 10% withdrawal rate. I wanted to get a sense of short term risk. Every simulation was positive. I find it hard to believe that there was never a negative year in a globally diversified stock portfolio. The average growth over 1 year was more than 10%. That also seems high? I know the calculator was designed for long term evaluation, but should it make a difference? Would it ever be possible to add in or exclude certain countries data sets? i.e. just run US returns to help show the benefits of geographic diversity?

Sep 14 2021 16:08 UTC

Willy - an all stock portfolio is 100% S&P 500. There are no international stocks used in this calculator.

If you run a 1 year retirement with a 10% withdrawal rate, starting with \$1M, you'd compound \$1M by the average 1 year return from 1928 - today, and that results in an average balance of \$1.01M, a low of \$470k, and a high of \$1.43M. So the range of being 100% in the SP 500 for 1 year is very wide!

Sep 14 2021 16:22 UTC

Thanks, I see it now.

Sep 18 2021 00:59 UTC

Have you considered adding a field where users can enter the fee % they are paying on their investments? Most investors can't purchase the index without paying a fee. It may be interesting to see the impact of a 1% fee?

Sep 18 2021 14:25 UTC

Correct, fees would eat into returns. In the past, before low cost index funds existed, it would have been harder to get access to the entire SP 500 without incurring transaction fees, etc. So the data is skewed up slightly.

We will put this on the list, but I'm not sure how much it would change the numbers. Most index funds today are no-load and have very low fees. For example the Vanguard 500 Index Fund Admiral Shares (VFIAX) has an expense ratio of 0.04%.

Karl Nov 6 2021 13:41 UTC

This is very helpful. Also impressed by the response time. Thank you for sharing this. Quick question. How do I think about negative ending balances?

Nov 6 2021 19:46 UTC

@karl -> How do I think about negative ending balances?

Well it means the person went broke before the end of the simulation. In other words their money did not last them. Need to either reduce the withdrawal rate, pick a different asset allocation, or start with more money.

Brett Nov 17 2021 17:47 UTC

First, thank you so much for providing this calculator...I'm an engineer that wound up focusing on statistical analysis so this is prized tool!

The "Distribution of Ending Balances" appears to be too high.

I plugged in a 40 year retirement, with three million in savings, a 3.6% withdrawal rate, and a growth portfolio and I got an ending balance of mean = 40.23 million and median = 33.00.

I created a little calculator in Excel to with 70% stocks at 10.09% and 30% bonds at 4.31% minus the 3.6% annual withdrawal over forty years and I got an ending balance of 3.13 million.

Any guidance would be greatly appreciated.

Thank you, Brett

Nov 17 2021 19:25 UTC

@Brett -> Thanks for your comment.

The best way to explain it is to have you look at the linked spreadsheet below. Notice how the Withdrawal starts at 3.6% but then only increases with inflation. That may be one key difference. Other than that, see how the formula works for the stock/bond return, and how the balance carries forward. It is doing automatic re-balancing each year.

Let me know if you have additional questions!

Peter Nov 26 2021 19:41 UTC

Nice calculator for what it does. It would be much, much better if it allowed for the “go-go” “slow-go” and “no-go” stages. Also feeding in minimum Canadian RRIF withdrawal limits (once we get to 90 years old we must take out at least 25% of the portfolio, so that has a huge impact) would help a lot too.

In my case I want 4 years of \$x then 6 years of \$x minus mortgage, then 6 years of \$x minus mortgage minus government payments, 14 years of 80% of that (slow-go years) then 6 years of 80% of that for no-go years, then die at 90. I’ve been using the Financial Goals calculator at portfoliovisualizer.com for this, but the more calculators I can use makes me happier with the decision to retire at age 60.

Nov 26 2021 20:09 UTC

@peter -> Required minimum withdrawals, staged withdrawals, tax considerations etc get really involved (and beyond the scope of this calculator).

You might try our Income Spending Simulator It uses a fixed return vs historical back testing which this tool does. You can model just about anything in there.

JB Nov 28 2021 17:40 UTC

Not sure what this spreadsheet is trying to show. (hope I'm not missing something). It is based on 40 years of static returns...is that reasonable? What about negative returns and the affect on portfolio balance.? Think about the three consecutive years of negative returns for the years 2000,2001,2002...and 2008 where some portfolio allocations returned -38%.
Historically, the S&P has had 19 years of negative returns or 26.8% of the time. So it would seem realistic to build a random rate of return in any portfolio projection. Incidentally, Excel has a function specifically for this: NORM.INV(RAND(),ArithmeticAverage,StandardDeviation).

Nov 29 2021 18:57 UTC

@JB -> Not sure what this spreadsheet is trying to show.

We were helping another person with the math in their own spreadsheet since their version wasn't matching up with what our calculator was showing.

Correct, 40 years of a static return is not reasonable, but on average, the compound adjusted rate of return for stocks can be estimated and plugged in. You just have to ignore the volatility component, but the final numbers work out the same.

This calculator does back testing using actual data (it is not a monte carlo simulation and does not generate data).