Comparison · Cost
This is a trick question, and almost every published answer to it is wrong. At the same wattage they all cost the same. Here is what actually varies.
How this page is funded: we earn a commission if you buy through the links here, at no extra cost to you. Picks are chosen before links are attached, which is why every roundup names something to skip. Full disclosure.
There is no cheapest space heater to run. Every resistive electric heater converts essentially 100% of its electricity into heat, so at the same wattage they all cost identically — 27.5 cents an hour for 1,500 W at the U.S. average rate. Savings come from wattage, thermostat accuracy and runtime.
An electric heater has no flue, no exhaust and no combustion. Every watt it draws ends up as heat in the room — including the watts that spun the fan and lit the display, because those become heat too. There is nowhere else for the energy to go. This is about as close to a certainty as consumer-product claims get.
So the marketing claim that a particular element technology is cheaper to run is not a small exaggeration, it is a claim about physics that is not true. It is also why there is no ENERGY STAR category for portable space heaters — a label needs a spread between products to rank, and on this axis there is not one. A ceramic PTC element, an infrared quartz tube, a mica panel and a coil of resistance wire in oil all convert electricity to heat at the same rate, because all of them are converting all of it.
Price does not change this either, which is worth stating because the premium end of the category implies otherwise. A designer machine costing many times a budget tower draws the same 1,500 W from the same outlet and bills you the same 27.5 cents an hour — worked through in our Dyson Hot+Cool review, where the wattage most often quoted online turns out to belong to a different country's model entirely.
| Heater type | Watts | Efficiency | Cost per hour |
|---|---|---|---|
| Ceramic (PTC) tower | 1,500 W | ~100% | 27.5¢ |
| Infrared quartz | 1,500 W | ~100% | 27.5¢ |
| Oil-filled radiator | 1,500 W | ~100% | 27.5¢ |
| Micathermic panel | 1,500 W | ~100% | 27.5¢ |
| Fan-forced | 1,500 W | ~100% | 27.5¢ |
Computed at 18.34¢/kWh (EIA, June 2026). The identical column is the point of the table.
1. Wattage. This is not subtle and it dwarfs everything else. 750 W costs exactly half of 1,500 W, permanently, with no efficiency penalty. If your room only needs 1,000 W, running 1,500 W is buying overshoot.
2. Runtime. The cheapest kilowatt-hour is one you did not use. A timer that removes six hours from every night removes 75% of a nightly bill. Nothing about the element does anything comparable.
3. Thermostat accuracy. A heater that holds 68°F costs less than one overshooting to 73°F, and the gap compounds every night. This is where a digital thermostat with 1°F increments genuinely pays for itself against an unmarked dial.
4. Thermal mass — a real but small effect. An oil-filled radiator keeps radiating after the element cuts out, so it cycles less often and the room drifts rather than dropping. That is a genuine advantage in steadiness. It is not an efficiency advantage: the heat it releases while coasting is heat it already paid for.
Everything above applies to resistive heating, which is what every space heater in this category is. A heat pump is a different machine: it does not make heat, it moves existing heat from outside to inside, so it can deliver two to four units of heat per unit of electricity consumed. That is the only mainstream technology that beats the 1:1 ceiling, and it is why the heat pump comparison reaches a different conclusion from every other page in this section.
The practical corollary of all this is that a cheap heater is not a worse heater in the way that matters — at equal wattage it puts exactly as much heat in the room as an expensive one. Best budget space heaters takes that to its conclusion and itemises what the extra money buys instead, which turns out to be control, silence and a published safety listing.
Ranked for fit, not for hype. Coverage and current draw are computed from each unit’s published wattage. Tap a row for the full write-up, or go straight to the listing.
| # | Product | Best for | Covers | Draw | Price |
|---|---|---|---|---|---|
| 1 |  Lasko 754201The 900 W low setting is the whole point: it is the rare listing that publishes both settings, so you can compute what the cheap mode actually costs instead of guessing. | One person at one desk | 150 sq ft | 12.5 A | $49.99 · Check on Amazon |
| 2 |  De'Longhi TRD40615TNo fan, so no noise and no moving dust — and the sealed oil keeps radiating after the element cuts out, which is why an oil-filled unit feels steadier than a ceramic one at the same wattage. | Silence, and a bedroom | 150 sq ft | 12.5 A | $159.95 · Check on Amazon |
| 3 |  Dreo Solaris 718The rare heater whose published coverage claim survives the arithmetic: 1,500 W is a genuine 300 sq ft at 10 W/sq ft, and Dreo publishes that number rather than implying more. | A whole living room | 150 sq ft | 12.5 A | $129.99 · Check on Amazon |
Coverage computed at 10 W/sq ft, average insulation, 8 ft ceiling. Draw computed as watts ÷ volts. Neither figure is a manufacturer claim.
Top pick · One person at one desk
The 900 W low setting is the whole point: it is the rare listing that publishes both settings, so you can compute what the cheap mode actually costs instead of guessing.
| Published specification | |
|---|---|
| Rated power | 1,500 W high / 900 W low, at 120 V |
| Thermostat | Adjustable dial; 2 heat settings plus fan-only |
| Size | 9.2 x 7 x 6 in, 2.93 lb |
| Cord | 6 ft (Lasko's published figure) |
| Safety | Tip-over switch, overheat protection, cool-touch housing |
| Warranty | 3 years |
| Heat type | Ceramic (PTC) |
| Maker’s coverage claim | 100 sq ft |
| Safety listing | ETL listed, per Lasko's published specification — one of the few in this roundup that names a mark rather than leaving it to the rating plate. |
Who it is for
A desk, a small office, a dorm room — somewhere you want heat on a person rather than in a room.
What it is not for
A nursery, and not unattended use. It publishes a tip-over switch and overheat protection, but no child lock — and the USFA's advice to switch a heater off when you leave the room applies to every unit on this site, listed or not.
2 · Silence, and a bedroom
No fan, so no noise and no moving dust — and the sealed oil keeps radiating after the element cuts out, which is why an oil-filled unit feels steadier than a ceramic one at the same wattage.
| Published specification | |
|---|---|
| Rated power | 1,500 W at 120 V |
| Controls | Mechanical, with a programmable 24-hour timer |
| Fluid | Permanently sealed diathermic oil — never needs refilling |
| Maker's claim | Up to 75% more radiant surface area |
| Mobility | Front and rear handles, snap-on wheels |
| Heat type | Oil-filled |
| Maker’s coverage claim | None published — the coverage figure above is ours |
| Safety listing | No listing mark published in the listing text; verify the UL or ETL mark on the rating plate. |
Who it is for
A bedroom, a nursery at a safe distance, or anywhere the fan noise of a ceramic heater is the actual problem you are trying to solve.
What it is not for
Fast heat, and not a room you only occupy for twenty minutes. It also should never be tucked beside a bed or curtain — hot surface, 3-foot rule.
3 · A whole living room
The rare heater whose published coverage claim survives the arithmetic: 1,500 W is a genuine 300 sq ft at 10 W/sq ft, and Dreo publishes that number rather than implying more.
| Published specification | |
|---|---|
| Rated power | 1,500 W at 120 V |
| Maker's coverage claim | 300 sq ft |
| Thermostat | 41–95°F digital, 1°F increments |
| Oscillation | 30° / 60° / 90° / 120° |
| Dimensions | 7.9" D x 7.9" W x 29.7" H, 8 lb |
| Safety | Tip-over cutoff, overheat cutoff, child lock, thermal cutoff |
| Heat type | Ceramic (PTC) |
| Maker’s coverage claim | 300 sq ft |
| Safety listing | ETL listed (stated by Dreo on the product page) |
Who it is for
A living room, bedroom or den at or under 300 sq ft with average insulation and an 8-foot ceiling, where you want to set a temperature and stop thinking about it.
What it is not for
A garage, an uninsulated basement, or anything over about 300 sq ft. It is not under-powered — 1,500 W is the plug-in ceiling — the room is simply too big for one outlet.
None of them is cheaper than another at the same wattage. All electric resistance heaters are about 100% efficient. The cheapest heater to run is the one with the lowest wattage that still holds your room, on a timer.
Not per hour — 1,500 W is 1,500 W. They can use slightly less over an evening because the oil keeps radiating after the element cuts out, so the thermostat cycles less often. The effect is real but modest.
Per kilowatt-hour consumed, no. They can result in lower consumption because radiant heat makes a person feel warm without raising the whole room's air temperature, so you may run them at a lower setting or for less time.
Use the lowest wattage setting that holds the room, put it on a timer, set an accurate thermostat rather than a dial, close the door, and turn the whole-house system down while the heater runs.
The only comparison where a space heater loses on physics: 1 unit of heat per unit of power versus 2–4.
All electric heaters are 100% efficient. These are picked on the three things that actually cut the bill.
Cost per hour for any wattage at any rate — the calculator, the formula and the full grid.
1,500 W in the US, not the 2,250 W UK coverage quotes — and what that changes about the room it heats.
Where the 750 / 1000 / 1500 W tiers come from, and why 1500 W is a hard ceiling on a U.S. outlet.
Per-BTU arithmetic for gas, oil and electric — and the one condition that decides the answer.
Silent and slow against fast and audible — and what the thermal-mass argument is really worth.
There is no ENERGY STAR category for portable heaters — and the reason tells you what to check instead.
A $23 heater makes the same heat as a $460 one. Here is what the money actually buys, and what it does not.
1,500 W is a cap, not a feature. Identical heat, identical cost — so what is left to choose on?