When the best laid plans ...

[jleonard99]

Suppose that I use OWL to develop a retirement plan. Given my assets and household financial plan and all that stuff I fiddle around and finally settle on a set of policies that make me comfortable: spending levels, when to take social security, roth conversion strategy, asset allocations, etc. and pressure test it against a bunch of different possible rate path futures and longevity futures. I'm all comfy that we won't run out of money and can still accomplish our life goals. Nice and tidy!

It's January first .. a new year ... I complete the annual refresh of my plan (update rates, taxes, remaining assets, etc.) ... all good ... still on track ... and my plan and I agree that I'll withdraw $X from my retirement accounts for this year's expenses.

.. and then .. I don't ... life happens and I need to widthdraw something different than planned, like 1.3X.

How do I model this unexpected expense to evaluate it's potential impacts? That is, how do I model taking more (or less) during the first year (or current year if you do this annually) of the plan while holding all else constant?

My first guess is to use the big-ticket item category to include the additional first year expense and go from there.

Is there a better workflow?

[mdlacasse]

This is like budgeting. When individuals are putting their assets in a retirement plan, they should always leave apart the safety net, the one we are told to keep all our life: 6 months of salary set aside in a safe investment account in case some emergency happens. There is no reason to eliminate this need at retirement. Your safety net should be excluded from your plan, and act as a sludge fund to smooth your spending needs.

This is what most financial advisors will tell you: Keep your safety net excluded from your retirement plan as it continues to serve the same purpose as before.

[jleonard99]

I agree - but cashflow at the beginning of a retirement plan might be sketchy as one transitions from a steady paycheck to paying oneself from their retirement pool. What if I need to restore my safety next funds by taking more from my retirement pool because that's the only source I've got available? From an OWL modeling perspective, do I show this extra, one-time need as a big ticket item in the first year? or is there a better way?

[sengsational]

Speaking just for how I'd do it, I probably wouldn't attempt to model it at all, except that next year, when it came time for actionable events, the effects of the one-time situation, now in the past, would be reflected in the model inputs.

I guess I'm concluding that unless there's some action that would be taken based on running the model, I wouldn't bother modeling the event.

But let's say you want your plan to be robust enough to survive a big bump. It seems like putting a big ticket expenditure in your plan would work. Your available spend would be lower, and if you stuck to that spending level, your encounter with the big expenditure could come and go and your spending level would be unchanged.

[mdlacasse]

Indeed, what we always need to keep in mind is that retirement planning is not planning for 30 years. It is to explore the sensitivities and adjust your actionable items every time needed. And all these actionable items are in the very near future, this year.

This is why the 4% rule and all fixed rules are not that useful. Individuals will react to their environment. If you need a new car, you might slow down on your restaurant outings or vacations. If the market does wonders, you might consider giving gifts to your family or your favorite charity. Or buy that new car earlier.

The plan will never be perfect because of all the uncertainty involved. But we have 30 years to make it right ;-)