♨

省的不是钱 What you save isn't money

"Insulation pays for itself in seven years" assumes you keep the house at the same temperature afterwards. In New Zealand, mostly, you don't — and that is not a flaw in the plan, it is the point of it.

Most New Zealand houses are underheated. People do not run them at 18 °C and bank the savings; they run them cold because heat is expensive, and when heat gets cheaper they buy more of it.

So a retrofit does not have a payback figure. It has a frontier: same temperature and a smaller bill at one end, same bill and a warmer house at the other, and every mixture in between. Which end you land on is a decision about how you live — and a tool that hands back one number has quietly made it for you.

Nothing is uploaded. The arithmetic runs on the server for one request and nothing is written down.

Where

Somewhere else

The house now

m² °C $/kWh

Heat loss now: 572 W/K — that many watts leak out for every degree the inside is warmer than the outside. Everything a retrofit does shows up as a smaller version of that number and nothing else.

Heated how

After the work

$ $

Fill in that second box and the bands above stop being a guess: a bill you have actually paid measures the house, and the heat loss is worked back out of it.

Two answers, and you have to pick one
Keep it at 16.0 °C save $273 a year — the bill falls from $481 to $208 and the house feels exactly the same
Keep spending $481 reach 20.6 °C up 4.6 °C — the bill does not move and nothing is saved at all

Both are true. Neither is the answer, and the difference between them is not arithmetic — it is a decision about whether you would rather have $273 or 4.6 degrees.

Everything the retrofit can buy

Every point on the curve is reachable after the work. Left is cheaper, up is warmer, and the whole curve sits up and to the left of where you are now — that gap is the retrofit. Which point you end up on is not something the insulation decides.

$0$200$40016°17°18°19°20°21°18 °C — the health line$208 a year at 16.0 °C — saves $273$235 a year at 16.5 °C — saves $246$262 a year at 17.1 °C — saves $219$290 a year at 17.6 °C — saves $191$317 a year at 18.0 °C — saves $164$344 a year at 18.5 °C — saves $137$372 a year at 18.9 °C — saves $109$399 a year at 19.3 °C — saves $82$426 a year at 19.8 °C — saves $55$454 a year at 20.2 °C — saves $27$481 a year at 20.6 °C — saves $0nowwhat you would spend on heating, a year →

The dot is where you are now. Straight left along the bottom arrow is all money; straight up the side arrow is all warmth. Everything between them is on the curve, and the curve is what you bought.

The same thing as a table

Kept atHeating, a yearSavedWarmer by
16.0 °C$208$273+0.0 °C
16.5 °C$235$246+0.5 °C
17.1 °C$262$219+1.1 °C
17.6 °C$290$191+1.6 °C
18.0 °C$317$164+2.0 °C
18.5 °C$344$137+2.5 °C
18.9 °C$372$109+2.9 °C
19.3 °C$399$82+3.3 °C
19.8 °C$426$55+3.8 °C
20.2 °C$454$27+4.2 °C
20.6 °C$481$0+4.6 °C

Right now: $481 a year at 16.0 °C, which is 1,604 kWh of electricity.

The saving and the warmth are the same money spent twice

You can have $273 a year or 4.6 degrees, or any split of the two. You cannot have both, and the reason a payback figure feels wrong to people who live in cold houses is that it silently assumes they will take all of it as money.

You are keeping the house at 16.0 °C, which is below 18

That is not a saving, it is a shortfall. Holding this house at 18 °C would cost $716 a year — $235 more than you spend now. After the work it would cost $316. That number, not the payback, is what the retrofit is actually for: it is the difference between 18 °C being unaffordable and being affordable.

$12,000 pays back in 44 years — if you take it all as money

That figure divides the cost by $273 a year and assumes the house stays exactly as warm as it is now, forever. Take any of it as warmth and the payback gets longer; take all of it and there is no payback at all, because nothing was saved. The number is not wrong. The sentence around it usually is.

What it does not know

Your house. The insulation bands are broad, and two houses in the same band differ more than the bands differ from each other. The accurate route is the one at the bottom of the form: give it a heating bill you have actually paid and it works the heat loss out backwards, because a bill is a measurement of the house and a band is a guess about it.

How much of the house you heat, and when. A steady-state model holds the whole house at the setpoint every hour of the year, and almost nobody does that. The pattern you pick moves every dollar on this page — and it does not move the trade-off between money and warmth at all, which is asserted in the tests rather than hoped for.

Damp. A cold house and a damp house are the same house, and drying it out is most of what people notice after a retrofit. None of that is in the arithmetic here, which only knows about temperature.

Whether the work is worth doing. This computes what a change to the building does to the heat balance. It does not know what your builder will quote, whether the underfloor is accessible, or whether the walls can be done without taking the linings off.

The temperatures are modelled from long-run monthly means, not from a station record, and 18 °C is the World Health Organization's figure for indoor temperature — the same one New Zealand's Healthy Homes standards use for a rental's living room.