Odd results with STABLE spending policy?

[mountainsoft]

I am using the STABLE spending policy, and have defined our pension and social security payments on the Additional Inputs screen. The Annual Retirement Income on the main screen is left blank.

When I run the simulation, it shows our money runs out in 20 years or so. It appears to completely ignore the pension and SS payments I defined on the Additional Inputs screen.

If I disable the income on the Additional Inputs screen and enter the same total as Annual Retirement Income on the main screen, the money continues the rest of our lives as expected.

What am I doing wrong?

[jimr]

It's tough to say what's happening exactly. My first guess would be an issue with the start and end dates or an issue with the COLA setting on the income entries. The income and expense inputs on the main screen are assumes to begin at retirement (or the age specified for income) and end at the end of the plan. They're also assumed to have a COLA setting of 'track inflation.'

Generally, the best bet in this type of situation is to look at the year-by-year outputs in the detailed view and select the 'show more detail' radio button. This lets you see exactly what the simulation is doing on a year-by-year basis and is the best way to verify that you've entered all the inputs correctly.

[mountainsoft]

Here's what I am entering in FRP. I can't see any obvious errors, but it doesn't seem to be factoring in our pension or SS payments at all. Any idea what I'm doing wrong? It does factor in when using the flexible or conservative spending policy, just not the stable policy.

MAIN SCREEN

Current Age: 53
Retirement Age: 60
Life Expectancy: 100
Inflation - Average: 3%
Inflation Std: 0%
Investment Tax Rate 0%
Income Tax Rate: 10%
Taxable Portfolio: blank
Tax Deferred Portfolio: 73,000
Tax Free: blank
Min withdrawal age: 60
Taxable Savings: blank
Tax Deferred Savings: 6500
Tax Free Savings: Blank
Investing Style: Below Average Risk
Annual Retirement Income: Blank
Retirement Start Age: 60
Annual Spending: 39000
Spending Policy: Stable

ADDITIONAL INPUT CASH FLOWS:

Enabled - Pension, start age 60, end of plan, $27504, 100% taxable, 2.8% cola
Enabled - Social Security, start age 62, end of plan, $4950, 50% taxable, 1% cola
Enabled - Social Security, start age 67, end of plan, $12,249, 50% taxable, 1% cola

[jimr]

I attempted to run the simulation with the values you posted and I think I can see why the results might be confusing.

I believe the additional inputs entries for the pension and ss are being used in the simulation. You can verify this in the detailed view tab in the year-by-year results by looking at the after tax income column. It's a little confusing because these items lose purchasing power to inflation between now and when you retire because they don't track inflation exactly.

Anyhow, I think one key to understanding what's happening is to understand the flexible spending policy. You can read up on it here:
http://www.flexibleretirementplanner.co ... -policies/

If you look at the 'percent of expenses funded' column in the detailed view with the spending policy set to flexible, you'll see that your desired expenses only get funded at approx 80-90% or so after a few years of retirement. This is because the portfolio has a tough time funding the desired withdrawals in the early years of retirement and so the simulation calls for some belt tightening to preserve the portfolio. This feature is disabled with the stable spending policy. With the stable spending policy, the simulation tries to withdraw the desired net spending amount regardless of the impact on the portfolio and likely survivability of the retirement plan.

[mountainsoft]

I think I may be looking at the results wrong. Somehow I was thinking FRP was calculating how my entire income would last through retirement. Am I correct that the graph and success/failure is strictly based on my portfolio (a single traditional IRA in my case)?

In other words, "failure" simply means my IRA runs out before I die, even though we'll have steady pension and SS income after it runs out? If so, that makes more sense to me now.

I'm also a bit confused regarding inflation calculations. If I set my annual spending at 39000, shouldn't that number increase over time to reflect less buying power? In other words, at 3% inflation I would need 39117 to pay for the same thing 39000 paid for the year before? The detailed view shows the same 39000 "expenses to fund" throughout my retirement. I must be misunderstanding how it works.

Finally, can you verify that a traditional IRA should be entered as tax-deferred, instead of taxable or tax free?

Love the program, just trying to understand all the complicated details!

[jimr]

In other words, "failure" simply means my IRA runs out before I die, even though we'll have steady pension and SS income after it runs out? If so, that makes more sense to me now.

That's right. Failure means that at some point during your retirement, your portfolio balance wasn't large enough to fund the target spending amount.

I'm also a bit confused regarding inflation calculations. If I set my annual spending at 39000, shouldn't that number increase over time to reflect less buying power? In other words, at 3% inflation I would need 39117 to pay for the same thing 39000 paid for the year before? The detailed view shows the same 39000 "expenses to fund" throughout my retirement. I must be misunderstanding how it works.

The detailed view shows all amounts in 'present value' (eg 2017) dollars. So if an amount stays flat from year to year in that view, it means it is exactly keeping up with inflation. If the amount increases from year to year, it is growing faster than inflation. If an amount decreases from year to year, it is growing slower than inflation.

Finally, can you verify that a traditional IRA should be entered as tax-deferred, instead of taxable or tax free?

That's right. Tax free is intended for Roth IRA funds.

[mountainsoft]

The detailed view shows all amounts in 'present value' (eg 2017) dollars. So if an amount stays flat from year to year in that view, it means it is exactly keeping up with inflation.

OK. Got it, thanks.

Another question (sorry) - What probability of success should I be aiming for? Would anything less than 100% be a risky retirement?

Related - On one of my simulations, with stable spending policy, % expenses funded is 100% throughout retirement, and FRP estimates a 143K portfolio balance at 100 years old. So why does the probability of success only say 86%? If it was fully funded throughout retirement with a balance leftover wouldn't that be a 100% success? I'm such a newbie.

Thanks!

[jimr]

What probability of success should I be aiming for? Would anything less than 100% be a risky retirement?

Again, there's no right answer. Using a portfolio of risky assets to fund a retirement income stream always involves some risk, regardless of what probability the planner spits out. The reality is that we can't reliably predict several very important inputs that the planner needs. For example, no one knows what future portfolio returns and standard deviations will look like and even very small changes in these can cause a big difference in a plan's probability of success. It's also impossible to predict future inflation and that can also have a big impact

So every retirement plan has risk. The question about whether to settle for 80% probability of success or 90% or more is a personal choice. The other side of retiring too early and facing a risky retirement is working too long and giving up a chance to enjoy some years before your time is up.

Related - On one of my simulations, with stable spending policy, % expenses funded is 100% throughout retirement, and FRP estimates a 143K portfolio balance at 100 years old. So why does the probability of success only say 86%? If it was fully funded throughout retirement with a balance leftover wouldn't that be a 100% success?

The planner runs 10,000 simulation iterations through your retirement plan. Each pass pulls a new random sequence of portfolio returns that are based on the average return/std deviation that you specify and these largely determine whether that iteration will be a success or not. An 86% probability of success means that around 1400 of those 10,000 passes ended with a depleted portfolio that couldn't fund all the expenses it needed to fund.

That $143k ending portfolio balance is the median value from the 10,000 simulation iterations. That means 50% of the simulation iterations ended with a portfolio value higher and 50% ended with a value that was lower. In the detailed view, you can see the bottom 10% portfolio value (eg 9,000 iterations did better).

Finally, one of the main benefits of tools like the Flexible Retirement Planner is the ability to try different scenarios to understand the contours or nuances of your plan and the impact of various assumptions. As one example, I suspect that the conservative assumption you made about social security keeping up with inflation, while certainly reasonable, has a very large impact on your plan's chances of success. It might be worth trying a scenario where ss tracks inflation, just for comparison. For another example, you could try setting the investing style to custom and increasing/decreasing the portfolio return by a little to see how that impacts the results. The planner comes with a feature called sensitivity analysis that automates some of this and lets you explore a wider range of variations.

[mountainsoft]

I suspect that the conservative assumption you made about social security keeping up with inflation, while certainly reasonable, has a very large impact on your plan's chances of success.

Yes, I am taking a very conservative approach to retirement planning.

I'm only using 75% of our Social Security estimates (from the MySS web site) since there's no guarantee what will happen with SS in the future. And I'm only using 1% for SS cost of living adjustments, even though they may be higher.

I'm also running simulations with lower investment returns than we're likely to get.

I'll play around with the numbers, but would prefer to err on the side of caution. Better to work an extra year than come up short when it's too late to do anything about it.