How much money can you afford to take out of your retirement savings annually?
The table below estimates how long savings will last based on different withdrawal rates and asset allocations. The table was built using methodology similar to that used in the 1997 Trinity Study, well-known research that has helped investors determine a withdrawal rate with an acceptable level of probability that they won't deplete savings too soon.
The table shows how the rate of withdrawal and various portfolio allocations can affect the chance of meeting income needs over a 25-year retirement. A high probability indicates it is more likely you will meet your income needs in retirement, while a low probability indicates you are less likely to do so.
For example, if you had a portfolio with 50% stocks and 50% bonds and you annually withdrew 6%, there would be a 57% chance of meeting your needs. It is assumed that a person retires at year zero and withdraws a required income need each year beginning in year one.
Probability of meeting income needs
Annual Withdrawal Rate
There are a number of factors that can impact whether a portfolio will last through retirement. The table shows how the amount of withdrawal and various portfolio allocations can affect the chance of meeting income needs over a 25-year retirement. It is assumed that a person retires at year zero and withdraws an inflation-adjusted percentage of the
initial portfolio wealth each year beginning in year one. Annual investment expenses were assumed to be 0.79% for stock mutual funds and 0.62% for bond mutual funds. A high probability indicates that an investor is more likely to meet income needs in retirement, while a low probability indicates that an investor is less likely to do so and may face shortfall. The chance of a portfolio running out over a long retirement is less likely as the amount withdrawn decreases and as equities are added. Keep in mind that returns and principal invested in stocks are not guaranteed and they have been more volatile (risky) than bonds.
The image was created using Monte Carlo parametric simulation that estimates the range of possible outcomes based on a set of assumptions including arithmetic mean (return), standard deviation (risk), and correlation for a set of asset classes. The inputs used are historical 1926 through 2011 figures. The risk and return of each asset class, cross-correlation and annual average inflation over this time period follow. Stocks: risk 20.3%, return 11.8%; bonds: risk 5.7%, return 5.5%; correlation –0.01;
inflation: 3.1%. Other investments not considered may have characteristics similar or superior to those being analyzed. The simulation is run 5,000 times, to give 5,000 possible 25-year scenarios. A limitation of this simulation model is that it assumes a constant inflation-adjusted rate of withdrawal, which may not be representative of actual retirement income needs. This type of simulation also assumes that the distribution of returns is normal. Should actual returns not follow this pattern, results may vary.
Government bonds are guaranteed by the full faith and credit of the US government as to the timely payment of principal and interest, while returns and principal invested in stocks are not guaranteed.
Stocks in this example are represented by the Standard & Poor's 90 Index from 1926 through February 1957 and the S&P 500® index thereafter, which is an unmanaged group of securities and considered to be representative of the US stock market in general. Bonds are represented by the five-year US government bond, inflation by the Consumer Price Index and mutual fund expenses are from Morningstar. An investment cannot be made directly in an index. The data assume reinvestment of income and do not account for taxes.
While asset allocation, withdrawal rate and longevity are important factors in retirement income planning, you'll also want to consider: