back to blog homepage

Will I Run Out of Money?

Monte Carlo Stats

Double click on any image to see a full size view

One of the questions I receive the most from clients is whether they will run out of money.  Will their savings be able to sustain their spending during their retirement years.

No one has a crystal ball, but we use a technique called Monte Carlo Simulation.  Monte Carlo portfolio simulation provides a means to test long term expected portfolio growth and portfolio survival during retirement withdrawals, i.e., whether the portfolio can sustain the planned withdrawals during the client’s retirement years.

In the image above, you can see that we use an all equity portfolio that consists of:

45% Large Cap Value Stocks

15% Mid Cap Value Stocks

10% Small Cap Value Stocks

5% Real Estate Stocks

5% Emerging Markets Stocks

10% European Stocks

10% Pacific/Asian Stocks

Our assumptions are that the client will withdraw 5% of the ending market value each year and that inflation will be an average annual 1.5%.

The way the simulation works, the model randomly selects an actual annual historic returns for each investment class above and projects the portfolio balance and distribution amounts over a 30 year period.  It then repeats this multiple times (e.g., 1000), aggregates the results, and gives us a statistical probability that the client’s money will last over the 30 year period of time.

In the example above, the client starts with $300,000 invested in the portfolio allocation as detailed.  He withdraws his 5% at the end of each year, and the portfolio is then rebalanced to the target asset allocation.

The graph below shows you the possible projected balance and portfolio growth pattern based upon these parameters over the 30 year simulation period, with the yellow line showing how the portfolio growth looks at the 75th percentile, the blue line the median portfolio growth, and the orange line showing the 25th percentile.

Monte Carlo Balances

The graph below shows you the possible projected distribution dollar amounts based upon these parameters over the 30 year simulation period, with the yellow line showing how the portfolio growth looks at the 75th percentile, the blue line the median portfolio growth, and the orange line showing the 25th percentile.

Monte Carlo Distributions

The results are that in all cases, there is a 100% probability that the client will be able to sustain a 5% withdrawal rate from their portfolio and that the median market value of the portfolio at the end of 30 years will be $1,025,801 and that the median withdrawal in the 30th year will have grown to a bit over $51,290 in today’s dollars.

Most people, however, do not want to have an all equity portfolio in their retirement years.  Most want a portfolio that is less volatile and that could sustain a higher distribution amount.  The portfolio mix that historically provides the highest return for the least amount of risk is one that is 65% equity and 35% fixed income.  Below, we have adjusted the portfolio above to add 35% fixed income to it, diversified across various fixed asset classes.

29.25% Large Cap Value
9.75% Mid Cap Value
6.50% Small Cap Value
3.25% REIT
3.25% Emerging Markets
6.50% Intl Europe
6.50% Intl Pacific Region
5.00% Long Term Treasuries
10.00% Intermediate Term Treasuries
5.00% Short Term Treasuries
15.00% Corporate Bonds

Additionally, we have increased the distribution amount to 7% of the market value and increased the average annual inflation rate to 3%.  Here are those results:

> Median market value of the portfolio at the end of 30 years will be $623,532

> Median withdrawal in the 30th year will have dropped to a bit over $19,763 in today’s dollars due to the impact of inflation and the lower portfolio growth – in other words, even though you are receiving $47,095 in actual dollars, you will have lost significant purchasing power with this blended allocation such that it will be like having $19,763 today

> There is a 100% probability that the client will be able to sustain a 7% withdrawal rate

When you compare the risk of the two portfolios, in the 100% equity portfolio, your largest down year was a loss of -42.46% compared to a loss of -30.26% for the 65/35 portfolio.  Your volatility for the 100% equity portfolio is a standard deviation of 9.52% compared to the 65/35 portfolio’s standard deviation of 7.82% – the easy way to think of this concept is to ignore the statistics (standard deviation being a measure of the variability of the portfolio returns over the 30 year period – see, ignoring it is better) and view the 100% equity portfolio as 20% more volatile than the 65/35 portfolio.

We can even push the limit and move the portfolio allocation to 50/50.  Your median market value drops to $537,564 but your withdrawals will have dropped to $17,002 in today’s dollars due to the impact of inflation and lower portfolio growth.

Finally, lets look at a 100% fixed income portfolio.  Those are some ugly results.  If we have an expected return of 3% for our fixed income portfolio instead of using historic actual – I don’t think we are likely to ever see those 14% treasury bonds again – we have a median market value at the end of 30 years of $68,391 and a withdrawal of $2,155 in today’s dollars (with the actual withdrawal dropping to $5,147).

What you give up in potential portfolio growth and potential withdrawals with the 100% equity portfolio you get in the ability to sleep better at night knowing that your money is invested in a less risky asset allocation.  However, you have to understand that inflation is a nasty creature and it will eat away at your purchasing power if you do not have adequate growth included in your portfolio.

So what is important to know about this exercise?  Having an equity component in your portfolio is critical to sustaining your purchasing power for the 5% distribution.  The higher your equity percentage, the more risk you assume, but by maintaining your investment allocation you weather the storm (remember we have had several 40%+ stock market crashes in the history of the stock market, with a 90%+ included for good measure) and are able to sustain your lifestyle in retirement.

Now the caveats:  as with any simulation, it is just that, a simulation.  It is not a Nostradamus quality prediction of the future, but rather a probability based upon a mathematical model developed by humans who are not perfect.  However, it is the best tool around at the present time and given that it uses the law of large numbers to analyze the portfolio under a thousand scenarios that have happened in the past, from a statistical standpoint we can rely upon the results.

If you are wondering what the tie-in is with Kashmir and this post, there were rumors long ago that this song referred to Nostradamus predictions.  Just a weird factoid from you humble blogger.

Mark