Do heat pumps work in cold weather? The Scandinavia evidence

Last reviewed: 14 May 2026

In short

Air source heat pumps work in cold weather, and the strongest evidence is the countries that have used them at scale for decades. Norway has the highest per-capita heat pump installation rate in the world — around 60% of households — with design temperatures routinely below −20°C. Sweden and Finland each have ~40% household installation rates in comparably cold climates. By contrast, the UK design external temperature for typical heat-loss calculations is around −3°C (mild by Scandinavian standards). Heat pump efficiency does fall as temperature drops, but a current-generation, properly-sized UK heat pump retains a COP above 2.5 at −7°C ambient and above 2.0 at −15°C — well above the COP of 1.0 that any electric resistance heater operates at. UK fleet performance through the 2022/23, January 2025, February 2025, and January 2026 cold snaps held at design SCOP.

Contents

• The myth and where it came from
• The Scandinavian evidence
• What happens to efficiency as temperature falls
• Cold-climate heat pump technology
• Defrost cycles — normal, not broken
• How UK heat pumps actually performed in recent cold snaps
• Where cold-weather performance can genuinely be a problem
• What this means for your decision
• What this means for homes in Reading

The myth and where it came from

The “heat pumps don’t work in the cold” objection has two structural sources:

Pre-2010 first-generation UK heat pump installs. The early UK heat pump market (~2008–2015) had two compounding problems: poorly-calibrated calculation tools (pre-BS EN 12831-1:2017) that systematically oversized units, and installer inexperience that produced low-quality installs with high flow temperatures and undersized radiators. The resulting installs ran at SCOP of 2.0–2.5 — comparable to old gas boilers in operating cost — and produced widely-circulated stories of “my heat pump didn’t keep up in cold weather.” Those stories have remained in the public memory long after the technical and procedural problems were corrected.

Conflation with air-to-air heat pumps and “AC reverse cycle”. Air conditioners running in reverse-cycle heating mode (typical of the early 2000s wall-mounted “split AC” devices) genuinely did struggle in cold weather — older refrigerants, no modern inverter control, designed for cooling-primary use. The current generation of UK air-to-water heat pumps — using R32 or R290 propane refrigerants with cold-climate compressors — is a different product class. Failure stories from 1990s/early-2000s reverse-cycle ACs are not relevant to current air-to-water installs.

Together, these two sources produced a cultural memory that doesn’t match current product capability. The Scandinavian evidence — which is concurrent and at scale — is the cleanest cross-check.

The Scandinavian evidence

Norway, Sweden, and Finland each made the heat pump shift earlier and more decisively than the UK. The figures:

Norway

Heat pump installation rate ~60% of households — the highest per-capita rate in the world. Norwegian winter design temperatures: −15°C to −30°C depending on region. Oslo design temperature −20°C; Trondheim −20°C; Tromsø −30°C. The Norwegian fleet has been operating at scale since the early 2000s. Air-to-air dominates the market (consistent with Norwegian housing stock — single-dwelling primary heating), with air-to-water and ground source growing.

Field-performance data from the Norwegian Water Resources and Energy Directorate (NVE): air source heat pumps in Norwegian conditions deliver SCOP 2.5–3.0 across the heating year, against design conditions far harsher than UK.

Sweden

Heat pump installation rate ~40% of households. Swedish design conditions: −15°C to −25°C across populated regions. Stockholm −18°C; Göteborg −16°C; Kiruna −33°C. Sweden’s market is more balanced across ground source and air source. Swedish Energy Agency data shows fleet SCOP 3.0–3.5 across air-to-water installs in southern Sweden, with cold-climate-rated units delivering SCOP 2.5+ in northern Sweden.

Finland

Heat pump installation rate ~40% of households. Finnish design conditions: −20°C to −38°C. Helsinki −26°C; Rovaniemi −38°C. Finland’s heat pump fleet is dominated by air source — air-to-air for single-dwellings and air-to-water for apartment block + retrofit — with cold-climate-rated R32 inverters delivering SCOP 2.5–3.2 across Finnish winters.

The headline takeaway: countries with winter design temperatures 15–30°C colder than the UK have heat pump fleets operating successfully at scale across multiple decades. The technical capability gap between Scandinavian conditions and UK conditions is large. UK conditions are well within the explored operating envelope.

What happens to efficiency as temperature falls

A heat pump’s Coefficient of Performance (COP) is the ratio of useful heat output to electrical input. The headline COP figure — typically published at A7/W35 (7°C outdoor air, 35°C flow temperature) — is the test-condition rating. Real COP varies continuously with outdoor air temperature and the flow temperature being delivered.

Typical manufacturer performance data for a current-generation UK air-to-water heat pump (Mitsubishi Ecodan PUZ-WM, Daikin Altherma 3 M, Vaillant aroTHERM plus — all using R32 refrigerant with inverter modulation):

Outdoor air temperature	COP at W35 (low flow)	COP at W45	COP at W55
+10°C	5.0–5.5	4.0–4.5	3.0–3.5
+7°C	4.5–5.0	3.7–4.2	2.8–3.2
+2°C	3.5–4.0	3.0–3.5	2.4–2.8
−2°C	3.0–3.5	2.6–3.0	2.1–2.4
−7°C	2.5–3.0	2.2–2.6	1.8–2.1
−15°C	2.0–2.5	1.8–2.2	1.5–1.8
−20°C	1.7–2.2	1.5–2.0	1.3–1.6

Three patterns visible:

1. COP falls continuously, but never reaches 1.0 within the heat pump’s operating envelope. Even at −20°C, a current air-to-water heat pump delivers ~1.5 units of heat per 1 unit of electricity — 50% more efficient than a straight electric resistance heater.

2. Lower flow temperature multiplies the cold-weather advantage. A heat pump running at 35°C flow (typical of new-build with underfloor heating) delivers COP 3.0+ at −7°C; the same unit at 55°C flow delivers COP 1.8. This is one reason flow-temperature design discipline and fabric-first design discipline compound — they unlock cold-weather efficiency that high-flow-temperature retrofit surrenders.

3. The drop from +7°C to −7°C is a 40–50% efficiency reduction, not a “the unit stops working” cliff. UK winters spend most heating-season hours at temperatures between +5°C and +10°C; the design-day temperature of −3°C is touched for the coldest 0.4% of hours per year. The seasonal average COP (SCOP) — which weights hours by likelihood — runs at 3.0–4.0 for current UK installs even though peak-cold COP is lower.

Cold-climate heat pump technology

Current generation air source heat pumps for UK and continental European markets use three technologies that extend cold-weather performance:

Inverter compressors with variable modulation. Modern compressors modulate from ~30% to 100% of rated capacity. This means a 7 kW unit can run at 2.5 kW in mild weather (avoiding short-cycling) and 7 kW in cold weather (delivering peak output). Fixed-speed compressors (older technology) ran at single-speed and either short-cycled in mild weather or could not modulate up under cold-weather demand.

Vapour injection compressors (sometimes called “EVI” or “two-stage”). For cold-climate-rated units — typically sold into Scandinavian markets and increasingly available in UK markets — the compressor injects a small amount of intermediate-pressure refrigerant vapour between compression stages. This boosts output at low ambient temperatures by 15–25% versus standard single-stage compression. Mitsubishi Ecodan PUZ-HWM and Daikin Altherma 3 H HT are the UK-market examples.

Modern refrigerants. Current air-to-water heat pumps use R32 (most common) or R290 propane (growing share). Both have lower Global Warming Potential than the older R410A and R134a they replaced. R290 specifically has very low GWP (3 vs R32’s 675). R32 and R290 both retain useful pressure and efficiency at very low outdoor temperatures — the operating envelope is wider than the older refrigerant generation supported.

Defrost cycles — normal, not broken

When the outdoor temperature falls below approximately +5°C and the outdoor air contains moisture (most UK winters), frost accumulates on the outdoor heat exchanger. The frost insulates the exchanger from the air it’s trying to extract heat from, and COP falls progressively until the frost is cleared.

The heat pump clears frost through a defrost cycle: typically a brief reversal of the refrigerant cycle (the outdoor coil becomes the warm coil for 2–10 minutes), during which the frost melts and drains as water. During the cycle, indoor heat output briefly stops; some homeowners notice cooler indoor air for a few minutes. Defrost cycles typically occur 1–8 times per day in cold-with-moisture conditions, using 0.5–1.5 kWh of energy per cycle.

This is expected behaviour, not a fault. Heat pumps are designed to detect frost accumulation and trigger defrost automatically. A unit that didn’t defrost would lose efficiency progressively across each cold-with-moisture period until it could no longer extract useful heat from the air.

Visual cues that defrost is happening: water dripping from the outdoor unit, steam wisps from the bottom as the frost melts, a slight whoosh from the indoor system as the brief reverse-cycle finishes. Some new installs surprise homeowners by these signs — they’re normal operation, not a problem.

How UK heat pumps actually performed in recent cold snaps

UK winters have had multiple sub-zero stretches across the BUS-grant period:

December 2022 / January 2023. Multi-day −5°C to −10°C events across England and Scotland. Octopus Energy’s heat pump fleet (~2,500 instrumented installs at the time) recorded fleet SCOP averaging 3.4 for the season. No widespread fleet failures reported by Octopus, MCS, or DESNZ.

January 2025. Multi-day −7°C event in central England. Octopus’s then-larger fleet (~6,000 instrumented installs) recorded fleet SCOP 3.5 for the season; the cold week’s COP averaged 2.7 across the fleet. Cosy Octopus tariff customers reported running costs comparable to or below gas-equivalent during the cold week, helped by off-peak Cosy pricing.

February 2025. Brief −8°C event in north and east England. Comparable fleet performance.

January 2026. Multi-day −6°C event affecting most of England and Wales. MCS post-event review showed UK fleet SCOP holding at design figures, with no spike in installer call-outs for “heat pump failure” reports.

The pattern across these events: the UK air source heat pump fleet — sized to BS EN 12831-1:2017, designed under MIS 3005-D V3.0, installed under MCS certification — performs to its design SCOP across UK winters. There is no observed “the unit didn’t work” failure pattern at fleet scale.

Where cold-weather performance can genuinely be a problem

Four conditions where a UK heat pump install does materially struggle in cold weather:

1. Undersized installs, where the calculated load was wrong (pre-2017 calculation tool, or non-MCS install) and the heat pump cannot meet design-day load. The system falls back to immersion backup; the homeowner sees bill spikes and (in extreme cases) cold rooms. This is a sizing / install-quality problem, not a heat pump capability problem. (Our sizing-matters guide covers this.)

2. Pre-2015 installs running on first-generation refrigerants (R410A) and fixed-speed compressors. These units have meaningfully poorer cold-weather COP than current units. They remain in the field but are an aging cohort; new installs since 2018 use the current refrigerant + inverter generation.

3. Hybrid systems where the heat pump is sized below 100% of design-day load (per MIS 3005-D V3.0 hybrid rule allowing 55% of load at 55°C flow at design conditions). In a cold snap, the gas boiler covers the gap. The homeowner doesn’t see a heat pump “failure” — the hybrid is designed to use gas for peak cold demand — but the heat pump’s contribution to cold-weather demand is, by design, lower than it would be in a heat-pump-only install.

4. Properties with extreme heat loss (uninsulated solid wall + single glazing + leaky envelope), where the design-day calculated load is so high that even a correctly-sized heat pump struggles to maintain setpoint without significant fabric-first investment. These properties exist in UK retrofit but are a small minority with well-known fabric-first paths.

None of these is a “heat pumps don’t work in cold weather” issue — each is a specific design, sizing, or fabric issue with a documented remedy.

What this means for your decision

The “do heat pumps work in cold weather?” question typically arises from a different decision-relevant question: “will my heat pump keep me warm and at reasonable cost on a UK cold day?”

The decision-relevant answer:

• A current-generation, MCS-installed, BS EN 12831-1:2017-sized, MIS 3005-D V3.0-designed air-to-water heat pump in a UK property will heat the property to setpoint across the heating year, including all UK design-day conditions — provided the survey is accurate and the property’s fabric is reasonable.
• Running costs in cold weather are higher than mild weather — COP falls, the unit runs more hours, electricity consumption rises. Heat-pump-specific tariffs (Cosy Octopus successors; EOn Heat Pump) help materially because they provide cheap off-peak electricity for cylinder reheat and partial space heating preheating.
• Defrost cycles are normal, not a fault.
• The horror stories typically trace to specific install problems — oversizing, undersizing, high flow temperature, undersized radiators — rather than fundamental heat-pump capability limits.

What this means for homes in Reading

Reading’s design external temperature is −3°C (99.6% percentile, Heathrow weather reference). This is mild by Scandinavian standards — Stockholm (−18°C), Oslo (−20°C), and Helsinki (−26°C) are all 15–23°C colder than Reading’s design day, and heat pumps work at scale in all three. The cold-weather capability question is comfortably settled by the Scandinavian evidence; what determines whether your Reading install delivers the design performance is the design itself, not the climate.

For Reading homeowners, the cold-weather operating envelope is best understood through three checks:

1. Was the survey carried out to BS EN 12831-1:2017 + MIS 3005-D V3.0, using −3°C as the design external temperature? This is the methodology question. A compliant survey produces a heat pump sized to meet your home’s load at Reading’s design day. Pre-2017 methods over-stated the calculated load and led to oversized installs; current methods produce accurate sizing.

2. Was a heat-pump-specific tariff considered in the running-cost arithmetic? Cosy Octopus tariff was discontinued in March 2026, but a range of heat-pump-friendly tariffs (Octopus Tracker, EOn Heat Pump, etc.) provide cheap off-peak electricity that materially improves cold-weather running cost. Standard tariffs are workable but produce poorer cold-week running-cost figures.

3. Was fabric quality assessed alongside the heat pump design? A heat pump install on a property with poor fabric runs at low SCOP and high cold-weather running cost. The fabric-first principle applies most strongly to Reading’s Victorian and Edwardian terrace stock in central Reading and lower Caversham. (Our fabric-first guide covers the priority order.)

Cold weather is not the limiting factor for heat pumps in Reading. Sizing, flow temperature, fabric, and radiator adequacy are the variables that matter. Get those right and the cold-weather performance follows the same pattern as the Norwegian, Swedish, and Finnish fleets — which is to say: it works.

Related guides

• Heat pump defrost cycles — why your heat pump isn’t broken when it stops in winter (A2.6 — publishes cadence)
• Heat-pump-specific tariffs — Cosy Octopus successors, EOn Heat Pump, and how tariff choice affects cold-weather running cost (A11.5 — publishes cadence)
• SCOP, COP and HSPF explained — the efficiency metrics on which cold-weather performance is measured
• Why heat pump sizing matters — sizing errors are the actual root of most cold-weather complaints

---

Get a quote for an MCS-compliant heat pump install in Reading — sized to UK design conditions, designed for cold-weather performance, installed to current standards. | Call us on 0118 [number] | Email [address]

---