This is (another?) shot at modeling irregular expenses. What I'm really attempting to model here are the major capital expenses (of owning a home).
Past threads I've found have brought out a number of points:
- Measure the past
- A FIRE individual can work (or will do work) in one of various ways and still have some income to address these issues
While I respect these opinions, I believe they miss the central point of trying to build a capital-aware FIRE model. In some cases, they don't work... for someone with a newer home, there isn't capital maintenance to actually measure. I also want to be put in the "go work" situation only if it's NOT a predictable situation. Ignoring that a roof will eventually need to be replaced is not a "black swan" it's "lack of planning".
There is also an apparent "rule" of 1-4% (of home value) expected maintenance cost. Lots of web sites say this, but I've never actually seen the model to know how relevant this is to a frugal mentality (nor have I been a landlord to have that experience). It also ignores the "value of dirt" (is a furnace in San Francisco really 10X one in the Midwest?- I doubt it), so I have low trust in this rule of thumb (but see below).
So the net is, I don't want to be on a path to (or at) FIRE and all of a sudden discover significant expenses that have never been in my model. An effective 10% increase in expense (as an example) would just stink, and with a little bit of homework, shouldn't be a risk I have to take. This is a shot at that "homework".
Here are the rules (and limits/weakness) of the model:
- Does not cover the general-level periodic expenses like sealing a driveway, fence/trim/siding/inside painting, or HVAC/AC Servicing
- Does not cover (for the moment anyway) unusual items (pools, irrigation systems, retaining walls, sandblasting/redoing mortar on bricks)
- Attempts to only cover capital items of significant value (>$500 new), so does not cover Sump Pump, Dish Disposal, Faucets, Broken Windows, Smoke Detectors
- Does not cover contents (Furniture) or non-house capital (Automobile)
- Does not cover non-periodic "black swan" expenses: tree removal, termite damage, water/gas pipe issues, foundation issues, regrading
- Assumes certain items are beyond the necessary modeling life (50+ year life): Kitchen Cabinets, Closet structures, Walls/Structure itself, Electrical Wiring, Showers, Doors, Windows*
*Windows here is probably a weakness, but there seems to be significant variation in expected life on different sites, and I've lived in enough houses with windows so far beyond those expected lives that I don't know what to believe
Special thanks to:
http://www.atdhomeinspection.com/advice/average-product-life/ for many of these product lives. (Many I verified on other sites).
Items without prices are too variable for me to model for others. Fencing depends on type & length, appliances could be debated of new vs craigslist, etc. Others are shown in the mid range of products/costs in the US based on various searches, but you may want to research for costs in your area.
Useful Life - Item (Cost)
25 - Fence
20 - Asphalt Roof ($6.50/sq ft)
20 - Gutters/Downspouts ($5/linear ft)
30 - Siding ($5/sq ft)
20 - Pave Driveway ($6/sq ft)
25 - Deck Replacement ($15/sq ft)
15 - Garage Door
15 - HVAC
11 - Refrigerator
15 - Stove
9 - Dishwasher
10 - Clothes Washer
13 - Clothes Dryer
10 - Water Heater
10 - Carpeting ($4/sq ft)
10 - Laminate/Vinyl ($10/ sq ft)
15 - Refinish hardwood flooring ($3.50/sq ft)
Anyone see any major items I'm missing? Right now I'm thinking of just modeling this as a flat ratio of per year cost and treating that as an annuity cost/sinking fund type model.
Interestingly, this is about 1.5% for my house with new mid-range appliances, so it starts to rationalize the rule of thumb, which is at least moderately comforting that it isn't completely made-up. (I'm in an average COL area.)