How Much Does It Cost to Run an Air Conditioner?

AC running cost? What people really ask is, “how much will my AC cost me per month?” The truth is, it’s a matter of three numbers: wattage the unit consumes, hours the unit operates, and cost per watt of electricity.

None of the three is as simple as it appears. Running power is not the nameplate power. The hours you believe it to run are typically the hours it doesn’t run. Almost every guide fails to mention that the rate on your bill could not necessarily be the rate your air conditioner is charged at.

There are three methods of doing this: the fastest, the most accurate, and the tariff trap which makes most estimates too low.

First, the Unit on Your Bill

The first two terms are routinely confused, and the remainder of this is only comprehensible if they are defined.

kW is the power at a particular time. Consider it speed. kWh is the amount of energy consumed over a period of time. Consider it distance. Consumption of 1 kW for 1 hour is 1 kWh.

The kWh component of your utility bill is the primary cost, typically with a fixed charge, taxes and a tiered, time-of-use charge applied to it. The extras are more significant than people realize and there’s a section about them further down.

Method 1: If You Already Know the Average kW

The fastest calculation, and the least accurate unless you have a good figure for average power.

Cost = Average kW × Hours × Rate

Take a unit averaging 1.5 kW, running 8 hours a day, at 0.20 per kWh:

1.5 × 8 × 0.20 = 2.40 per day
2.40 × 30 = 72.00 per month

Simple enough. The problem is that most people do not know their average kW, and reach for the number printed on the unit instead. That number is the maximum, not the average, which brings us to the second method.

Method 2: Rated Power and a Load Factor

Your AC label shows a rated or maximum input power, perhaps 2.0 kW. The unit almost never draws that continuously. It cycles, and on an inverter model it modulates down once the room reaches setpoint.

So apply a load factor:

Average kW ≈ Rated input kW × Load factor

Reasonable load factors to start from:

A 2.0 kW rated unit at a load factor of 0.6 averages around 1.2 kW. Run that through Method 1 and you have a far more realistic figure than using 2.0 kW directly, which would overstate the bill by two thirds.

Method 3: Working From BTU and Efficiency

Plenty of units advertise cooling capacity rather than input power. If all you have is something like 12,000 BTU/h, which is roughly one ton, you can get to input power through the efficiency rating. If you need to move between BTU/h, kW and tons, our HVAC unit converter handles those conversions.

If the label shows EER or EER2

Input watts ≈ Cooling BTU/h ÷ EER
kW = watts ÷ 1000

A 12,000 BTU/h unit with an EER of 10 draws about 1,200 W, or 1.2 kW.

If your label says EER2 rather than EER, that is the post-2023 US rating and the arithmetic is identical. More on what changed below.

If the label shows COP

Common outside North America, and on many inverter units.

Input kW = Cooling kW ÷ COP

A unit delivering 3.5 kW of cooling at a COP of 3.5 draws about 1.0 kW.

One thing to be careful about before you use that number. EER and COP are measured at rated conditions, so what these formulas give you is input power at full load, not the average draw. Method 1 expects average power, so feeding the rated figure straight in will overstate the bill. Take the 12,000 BTU/h unit above: 1.2 kW rated, but at a load factor of 0.7 it averages 0.84 kW. Over 8 hours a day for a month at 0.20 per kWh, that is 57.60 against 40.32. Skipping the load factor step puts you 43% too high.

Moving between EER and COP

These describe the same thing in different units, which trips people up when comparing equipment sold in different markets. EER is cooling BTU/h per watt of input. COP is watts of cooling per watt of input. Since one watt equals 3.412 BTU/h:

COP = EER ÷ 3.412
EER = COP × 3.412

An EER of 10 is a COP of about 2.93. A COP of 3.5 is an EER of about 11.9. Worth knowing before you conclude that one unit is dramatically better than another when they are actually close.

What about SEER and SEER2

SEER is a seasonal average that folds in mild-weather operation, so it does not convert cleanly to a running-cost figure for any particular day. For estimating a bill, EER, COP, or rated input with a load factor will serve you better.

One thing worth knowing if you are comparing equipment. Since January 2023 the US Department of Energy has rated equipment under a revised test procedure, and the results are named SEER2, EER2 and HSPF2. The new M1 procedure raises external static pressure fivefold, from 0.1 to 0.5 inches of water column, to better represent real ductwork. The equipment did not get worse. The test got harder, and the numbers came down by roughly 5% as a result.

The practical consequence: never compare a SEER2 figure against an older SEER figure and conclude the newer unit is less efficient. Compare like with like, or you will reach the wrong answer about equipment that is genuinely better.

Typical Power Ranges as a Sanity Check

Ballpark full-load figures. Actual draw is often lower on inverter units once the space is at temperature.

The ranges are wide because efficiency, installation quality, outdoor temperature and building fabric all move the number. If your calculation lands far outside these bands, check your inputs before believing the result.

Why Inverter Units Behave Differently

A fixed-speed unit is either on at full power or off. An inverter unit draws more at startup to pull the space down quickly, then settles to a much lower steady draw to hold temperature.

That makes running cost far more sensitive to how the system is used and how the building performs:

• Setpoint, which decides how hard the unit has to work to maintain conditions
• Outdoor temperature and solar gain through glazing
• Insulation and air tightness
• Humidity, since removing moisture is real cooling work rather than wasted energy, and it consumes capacity that never shows up as a temperature change
• Door openings and occupancy

Which is why two homes with identical 1.5 ton units can produce bills that look nothing alike.

Same Unit, Three Different Bills

Take a 2.0 kW rated unit, 8 hours a day, 30 days, at 0.20 per kWh. Only the load factor changes.

Same equipment, same hours, nearly double the cost. The building matters more than the unit. If you are sizing rather than costing, our heat load calculator works out what capacity the space actually needs, and oversizing is one of the errors covered in our guide to common HVAC design mistakes.

The Tariff Trap: Your Average Rate Is Probably the Wrong Number

This is where most running-cost estimates go wrong, including the ones produced by online calculators.

If you are on a flat tariff, the rate on your bill is the rate your AC pays, and everything above holds. But on a tiered tariff, the first block of consumption is cheap and everything above a threshold costs more. Air conditioning is almost always the load that pushes you over that threshold, which means the AC's electricity is charged at the top tier, not the average.

Work through it. Say the first 300 kWh cost 0.12, everything above costs 0.28, and your home uses 250 kWh a month before any cooling. Add an AC averaging 1.4 kW for 8 hours a day, which is 336 kWh a month.

Total consumption: 250 + 336 = 586 kWh
Bill: (300 × 0.12) + (286 × 0.28) = 36.00 + 80.08 = 116.08
Bill without AC: 250 × 0.12 = 30.00
True cost of the AC: 116.08 − 30.00 = 86.08

That works out to an effective 0.256 per kWh for the air conditioning, even though dividing the whole bill by total consumption gives an average of 0.198.

Use that average figure and you would estimate 336 × 0.198, or about 66.56. The real cost is 86.08. You would be under by roughly 20 a month, close to a quarter of the actual figure.

If you are on a tiered tariff, use your highest applicable tier rate for AC estimates, not your average rate. On time-of-use pricing, use the peak-period rate, because cooling demand peaks in the afternoon at exactly the hours the utility charges most for.

Finding Your Real Average kW

Estimates are fine, but if you want the actual number for your installation:

• Compare kWh totals on your bill or utility portal across similar weeks with and without heavy cooling use. The difference is your AC
• If you have a smart meter with an hourly usage view, the cooling load usually stands out clearly against the baseline
• For plug-in window or portable units, a plug-in energy monitor gives a direct reading. Respect the monitor's rating and do not daisy-chain adaptors

For a hardwired split system, leave measurement to a qualified technician. Taking readings at the isolator or indoor unit means working on live circuits, and it is not worth the risk to confirm a number you can estimate closely enough with the methods above.

What Actually Lowers the Bill

In rough order of impact for effort:

• Raise the setpoint. Even one degree cuts run time noticeably, and the effect compounds across a season. Setting it far lower than you want does not cool the room any faster, because a fixed-speed compressor only has one output — it simply runs for longer than it needs to
• Clean or replace filters. A blocked filter makes the fan and compressor work harder for less airflow
• Cut solar gain with blinds, curtains, or reflective film on the worst-facing glazing
• Seal air leaks around doors and windows, since every bit of infiltration is load you are paying to remove
• Keep the indoor unit's intake and return path clear. Furniture in front of a return is a surprisingly common cause of poor performance
• Run a ceiling fan alongside the AC, which improves comfort at a higher setpoint

Frequently Asked Questions