Firecalc, income and retiring in the current year

[tb001]

I know this probably doesn’t matter in the bigger picture, but I want to make sure I’m using firecalc correctly...

I’m trying to figure out if we’re good to pull the trigger this year if we need to. Note that we’ll keep going omy if possible, but it’s not completely out of the question that both DH and I are unemployed as of the end of 2019.

I’ve left the retirement date in firecalc at 2019 and input our current investment portfolio. We likely have a windfall coming to us in the spring that, added to our current portfolio, will put us within 300k of our target for a 95% success rate.

We expect to make approximately 300k this year from our jobs, but this equals our current annual spend. I’m using the manual spend function and have 300k entered as our year one spend, but I’m not sure how to account for the 300k in income, since it’s not getting added to the portfolio, if that makes sense.

I’m not sure how firecalc looks at current year retirement. Is it the start of the year, so the 300k in income is actually offsetting the 300k anticipated spend?

And on a related note, if we hit a bear mkt run, am I correct in assuming that once we do, we’re still at a 95% success rate, albeit more subject to SOR risk, since we’ve essentially hit our number? Our income pretty quickly gets dwarfed by investment returns/hits, so it’s frustrating to feel like we are there in April, but maybe not in December! Or if mkt takes a downturn before the windfall hits!

 

[footenote]

What's the "windfall" ? How "likely" is it? (90%? 80%?)

Please post more details: for starts, your forecast annual post-retirement spend and portfolio without the windfall.

 

[tb001]

What's the "windfall" ? How "likely" is it? (90%? 80%?)

Please post more details: for starts, your forecast annual post-retirement spend and portfolio without the windfall.

Am I missing something—not sure why any of that additional detail is necessary?

I’m assuming the windfall is 100%. Without it, we wouldn’t be considering RE, so any other scenario doesn’t matter.

Spend post retirement is highly variable, so I’ve used the manual spend option and entered it by year, then solved for portfolio $ with a 95% success rate. Obviously, if my spend estimates are wrong, we have a problem, but I think they’re sufficiently padded.

 

[ERD50]

windfall noun
wind·​fall | \ ˈwin(d)-ˌfȯl
\
Definition of windfall

1 : something (such as a tree or fruit) blown down by the wind
2 : an unexpected, unearned, or sudden gain or advantage

I assume def #2? Usually there are some unknowns with a 'windfall'. That's why it was asked, "are you sure?".

Does your retirement count on some source that you do not already own?

-ERD50

 

[tb001]

I assume def #2? Usually there are some unknowns with a 'windfall'. That's why it was asked, "are you sure?".

Does your retirement count on some source that you do not already own?

-ERD50

Yes, #2, though you could argue about the earned part. Coming from a separation agreement that is being negotiated. At this point, I am ~90% sure this will be coming. It would be a small fraction of our invested assets, most of which is invested in vanguard index funds, but enough to put us within striking distance.

 

[COcheesehead]

I am probably not the only one that finds the OP kind of confusing.
Can we start over with clearer details? The prose is hard to decipher.

 

[Katsmeow]

One option would be to simply base this on what you think your portfolio will be at the start of next year and base your spending on what you will be spending beginning next year.

That way you don't have to try to deal with this year in Firecalc at all.

 

[tb001]

I am probably not the only one that finds the OP kind of confusing.
Can we start over with clearer details? The prose is hard to decipher.

Sigh, I know--I really did try to make it simple!

OK, let's go with a hypothetical to make the numbers easy. I ran Firecalc using a manual spend model and use the 'Investigate' tab to determine starting portfolio value needed to hit 95% success is $6.5M, assuming we retire in 2019. Our first yr spend is 300K. Current investment portfolio *today* is $5.7M. Anticipate adding $500K to the portfolio early in 2019 as part of a separation agreement. Earned income, after taxes for 2019 is 300K, so equivalent to our first yr spend.

So we have a $6.2M portfolio, including the separation agreement, but haven't hit that 'magic' $5M number in our bank account to get to 95% success. But we have income to cover that first year 2019 spend...

 

[tb001]

One option would be to simply base this on what you think your portfolio will be at the start of next year and base your spending on what you will be spending beginning next year.

That way you don't have to try to deal with this year in Firecalc at all.

I think this may be part of the answer--but if I run for 39 yrs instead of 40, and investigate starting portfolio needed, it's not 300K less

 

[tb001]

OK, after running the analysis slightly differently, I'm starting to wonder if there's a bug in FireCalc.

The portfolio for a 95% success rate using 'Investigate starting portfolio' function says we need 300K more than if I just enter a starting portfolio into firecalc and look at the success rate of the portfolio. Strange...

 

[ERD50]

OK, after running the analysis slightly differently, I'm starting to wonder if there's a bug in FireCalc.

The portfolio for a 95% success rate using 'Investigate starting portfolio' function says we need 300K more than if I just enter a starting portfolio into firecalc and look at the success rate of the portfolio. Strange...

Success rates get quantized. You sometimes need to move quite a bit to add or subtract a success or failure. But when you calculate for a specific success rate, it can calculate a specific dollar amount.

There probably is no actual exact 95% success rate, it gets quantized to the nearest level. If you put that extra 300K in, you will probably still get ~95% success. There is a range of inputs that will produce 95%.

To get better answers, include a link to your inputs. You can find that on the reults page.

-ERD50

 

[tb001]

Success rates get quantized. You sometimes need to move quite a bit to add or subtract a success or failure. But when you calculate for a specific success rate, it can calculate a specific dollar amount.

There probably is no actual exact 95% success rate, it gets quantized to the nearest level. If you put that extra 300K in, you will probably still get ~95% success. There is a range of inputs that will produce 95%.

To get better answers, include a link to your inputs. You can find that on the reults page.

-ERD50

Thanks ERD, I think that's exactly what's happening. I've been using the "investigate portfolio" function as a number to target, but it looks like we may hit that *magic* 95% success rate a lot sooner than I expected!

 

[Bir48die]

Whew. Thank goodness

 

[Perryinva]

Not sure why 95% is considered a magic go/no go. All you have to do is model on a later include date, and results can change dramatically. Many people start at 1939, as 1871 thru 1939 bears little to no resemblance to the current market machinations. It sort of makes Firecalc an historical plus Monte Carlo sim.

 

[audreyh1]

Not sure why 95% is considered a magic go/no go. All you have to do is model on a later include date, and results can change dramatically. Many people start at 1939, as 1871 thru 1939 bears little to no resemblance to the current market machinations. It sort of makes Firecalc an historical plus Monte Carlo sim.

Who starts at 1939? Not many I suspect.

I definitely keep the earlier data as it gives a wider set of tough situations, making outcomes about 5% worse for the models I’ve run, comparing post say 1925 outcomes to those prior.

I’m modeling %remaining portfolio, so there isn’t a failure % per se but rather how far the portfolio can drop in real terms and thus income.

I think why people accept 95% is that they realize they can adjust well ahead if their current withdrawal rate rises dramatically. That compared to working several more years to reach 100%.

 

[Perryinva]

I think you missed my point. The percent success is strictly the number of successful runs divided by the total runs. If I run from 1871 vs 1939, I have the exact same number of failures, but from a larger pool, so the success rate goes up, not down. My point is the percentage success is less meaningful than the number of failures. I look for 4 or less failures, personally. The percentage means little to me, for the exact reason you listed for %.