Why Is My Electric Bill So High in Summer?

Get last summer’s bill and this summer’s bill side by side and find the kilowatt-hours on each, not the dollar total. If the kWh are similar and the dollars jumped, you have a rate problem, not an air conditioner problem. If the kWh jumped too, work out whether the weather explains it before you pay anyone to look at your equipment. This page is about finding which of those three things happened. If your bill is roughly normal and you simply want it lower, that is a different job and a different page.

Before you blame the AC: read the bill

Almost everyone diagnoses a high summer bill by looking at the dollar amount, and the dollar amount is the one number that cannot tell you anything on its own. It is the product of three moving parts — how much electricity you used, what each unit cost, and how many days the billing period covered, and any of the three can move without the other two.

So start with three numbers off the statement itself, not off your memory:

1. Kilowatt-hours used. This is the only pure measure of consumption on the bill. Find it for this period, for the same period last year, and for the month before.
2. Days in the billing period. A 34-day cycle against a 28-day cycle is a 21% difference before anything else happens. Utilities routinely vary the read date by several days.
3. Your effective rate. Divide the total dollars by the total kWh. That single division gives you your real, all-in price per unit, including every fee and tier, and it is the number to compare year over year.

Now the diagnosis writes itself. Same kWh, higher dollars means you paid more per unit. Higher kWh with proportionally higher dollars means you consumed more. Sharply higher kWh with no change in weather or occupancy means something in the house is running when it should not be, and that is the only one of the three that a contractor can fix.

Cause 1 — you actually used more

Cooling load is not linear with temperature. Your air conditioner is fighting the difference between the outdoor temperature and your setpoint, so on a 95°F day with the house at 78°F it is working against a 17-degree gap, and on a 105°F day it is working against a 27-degree gap. That is not ten percent more work; it is closer to sixty percent more, and the run hours climb accordingly. A single ten-day heat wave inside one billing cycle can move a summer bill more than any maintenance item on this site.

The other consumption changes are mundane and easy to forget. Someone home during the day who was not home last year. A window air conditioner added in a back bedroom. A second refrigerator in the garage, which works far harder at 110°F than at 70°F. A pool pump on a longer summer schedule. None of these are HVAC faults and none of them will show up in a technician’s diagnosis, but all of them show up in kWh.

The check that settles it: compare this July to last July, in kWh per day, and then look up how the two months compared for heat. Most utilities publish average temperature or degree-day data alongside your usage history in the online account. If the weather was materially hotter and your kWh rose roughly in proportion, your system is behaving normally and you are looking at a weather bill.

Cause 2 — you paid more per kilowatt-hour

This is the cause most homeowners never consider, and in California it is frequently the largest single factor. EIA’s Electric Power Monthly reported California residential electricity averaging 33.25 cents per kWh in May 2026, against a US residential average of 18.44 cents. The same house, with the same air conditioner and the same habits, costs roughly eighty percent more to cool here than the national average, and none of that is visible in your equipment.

Three separate mechanisms can raise your effective rate without you changing anything:

• Time-of-use pricing. Most California residential customers are now on a time-of-use plan, where electricity used in the late-afternoon and evening peak window costs substantially more than the same electricity used at midday or overnight. Summer is exactly when your air conditioner runs hardest inside that window. Two households with identical consumption can pay meaningfully different bills purely on when they cooled.
• Tiered baselines. Many plans give you an allowance of kWh at a lower price and charge more above it. Summer is when a normal household crosses that line, so the marginal cost of the last several hundred kilowatt-hours is higher than the average cost of the first few hundred. This is why the bill can rise faster than the usage.
• Rate changes. Utility rates are revised, and the revision does not arrive as a letter you remember. This is precisely what the effective-rate division catches: if your dollars-per-kWh moved and your behaviour did not, the rate moved.

The action here is not a repair. It is to open your own utility’s current residential rate page — PG&E, SMUD, SCE and SDG&E all publish them, and confirm which plan you are on, where your peak window sits, and whether a different plan fits your household better. We deliberately do not print specific rates on this page, because they change and a stale number here would be worse than none.

Cause 3 — something is actually wrong

Only now is it worth suspecting the equipment, and the signature to look for is runtime. A faulty air conditioner rarely draws dramatically more current; it runs for far longer to deliver the same cooling. That distinction matters, because it means a fault shows up as hours, not amps.

The system runs and never satisfies

If the unit runs continuously through the afternoon and the house still drifts above setpoint, something is limiting capacity — most often charge, airflow, or a fouled condenser. Our page on an AC that can’t keep up works through the capacity side of this in detail.

A refrigerant leak

A system low on charge cools poorly and runs long. It also damages itself: Copeland’s application engineering bulletin on compressor operation notes that loss of charge causes overheating and repeated cycling of the motor overload protector, and that prolonged operation in that condition can lead to oil pump-out and eventual bearing failure. That bulletin’s stated scope is five- to twelve-ton light commercial equipment, so treat it as the mechanism rather than a residential specification, but the mechanism is the same machine. A leak is not something that stabilises; the charge does not come back. See the signs of a refrigerant leak for what to look and listen for. Refrigerant is EPA-regulated and is certified-technician work, and the A2L refrigerants in new equipment (R-32 and R-454B) are classified as mildly flammable, so a suspected leak is a call, not a project.

A dirty outdoor coil

The condenser rejects heat to outdoor air. Blanket it in cottonwood, dryer lint or shrubbery and it cannot, so head pressure rises, capacity falls and runtime climbs. This is the one genuine efficiency fault that a homeowner can see from the outside without opening anything.

A restricted filter, and what it does not do

A clogged filter starves airflow across the indoor coil. It does not raise the compressor’s current draw or trip a breaker; if anything it lowers amp draw, because there is less heat load on the compressor. What it does is drop the coil temperature until the coil freezes, at which point the system stops cooling entirely. The Department of Energy’s own home cooling guidance puts the energy penalty of a dirty, clogged filter at 5 to 15 percent of an air conditioner’s energy consumption — a real number, and a much smaller one than most maintenance marketing implies. Note that this is the filter figure. It is frequently confused with the roughly 20 percent that ENERGY STAR attributes to duct sealing, which is a completely different measure on a different part of the system.

Short-cycling

The opposite failure: the system starts, runs briefly, and stops, over and over. ENERGY STAR’s analysis of oversized air conditioners notes that efficiency is at its lowest at start-up and climbs over roughly the first ten minutes of runtime, and that short operating times do not allow the system to remove humidity effectively. A short-cycling system therefore spends most of its life in its least efficient minutes. Our page on why an AC short-cycles covers the causes.

A heat pump left on auxiliary heat

If your cooling equipment is a heat pump and the thermostat has been left in emergency or auxiliary heat mode, resistance heat strips can run alongside, or instead of — the efficient compressor cycle. This is a winter problem far more than a summer one, but it is worth ruling out on any heat pump with an unexplained jump, because it is the single most expensive setting on the control.

Working out which of the three you have, in one afternoon

Run these in order. Each step is designed to eliminate a cause, not to fix anything.

1. Normalise the bills. kWh divided by days, this period and the same period last year. Ten minutes, no tools.
2. Divide dollars by kWh for both periods. If that number moved, you have at least a partial rate answer, and you can stop looking for a fault to explain that portion.
3. Check the weather for the two periods in your utility’s usage view. If consumption tracked temperature, cause 1 is doing the work.
4. Time one cooling cycle. On a hot afternoon, note when the outdoor unit starts and when it stops. A healthy system on a design-condition day runs long cycles and satisfies the thermostat. Cycles under about ten minutes point at short-cycling; a unit that simply never stops points at capacity.
5. Look at the filter and the outdoor coil. A filter you cannot see light through is overdue. A condenser packed with debris is visible from three feet away.
6. Only then book a diagnosis. By this point you can tell the technician what you observed rather than asking them to find something, which is a materially better starting position.

What each cause costs to put right

Two of the three causes cost nothing to resolve — a rate problem is fixed by changing plan or shifting usage, and a weather problem is not a problem. Only the third has an invoice attached. The ranges below are contractor-reported market estimates, not published rates; no government body or standards organisation tracks residential HVAC repair pricing, and any site presenting these as sourced figures is overstating what is knowable.

What is not on that table is a whole-system replacement, and that is deliberate. A bill that jumped in one month is a diagnosis question. Whether the system is worth keeping is a separate decision with its own arithmetic — see repair or replace for the framework, and how long an AC lasts for what a realistic remaining life looks like.

What a normal summer cooling bill actually looks like

This is the anchor most homeowners are missing, and it usually changes the conversation. EIA’s Residential Energy Consumption Survey put average household spending on air conditioning at about $265 a year — roughly 12 percent of total home energy expenditure. By climate region it ranged from about $525 a year in the hot, humid Southeast down to roughly $60 a year in the temperate marine region along the West Coast, where nearly half of households did not use air conditioning at all. Those dollar figures come from RECS 2015 data published in 2018, and electricity prices have risen since, so the current national number is higher than $265, but not by a factor of five.

Read your climate, not your state. Coastal California sits near the bottom of that range and barely runs its cooling. The inland Central Valley — Sacramento, Stockton, Modesto, Fresno — does not: those are long, hot inland summers with far more run hours, and a Valley household lands much nearer the national figure than the coastal one. EIA does not publish a Central Valley number and this page will not invent one.

The practical consequence: if cooling is a few hundred dollars a year for a typical household, then a bill that rose by several hundred dollars in a single month is telling you something bigger than a dirty coil. That is the size of a rate change, a billing-period change, or a load that is not the air conditioner at all. It is worth confirming which before you spend money looking for a fault in the equipment.

FAQ