Repair or replace a washing machine? What lifecycle research can tell us

Research question

Under what conditions does extending a household washing machine's service through repair reduce lifecycle greenhouse-gas emissions relative to replacement?

How this review was prepared

On 20 October 2026 we used targeted, AI-assisted public web searches combining washing-machine repair, replacement, durability, lifecycle assessment and operating lifespan. We prioritised identifiable primary publications in publisher, JRC and university repositories. Selection was purposive and English-language. Commercial calculators and unsupported savings claims were excluded; other-appliance results were not treated as washing-machine findings.

Full text was checked for the open 2018 laundry study and relevant sections of the final 2016 JRC report. The 2014, 2019 and 2022 items were appraised through original abstract records. Full-text access to the 2022 paper failed, so no detailed numerical result from it is used.

This review was not preregistered. It had no exhaustive database search, independent duplicate screening or formal risk-of-bias scoring. Study estimates were not pooled. Source selection and access limits constrain the synthesis.

Start with the same washing service

A useful comparison asks how a household will obtain the washing service it needs over the same future period. Comparing one old machine with one new machine is insufficient when they provide different years of service, different capacity or different numbers of washes. The durability framework developed by Ardente and Mathieux explicitly treats lifetime extension and the alternative replacement product together.1

For this review, the relevant question is a decision made now: repair the existing appliance or obtain an alternative. Its original manufacturing emissions have already occurred. Keeping it does not erase those emissions, and purchasing a replacement does not make the new manufacturing stage disappear. An assessment needs a stated way of allocating production across the service provided. This is our interpretation of the comparison problem, rather than a new LCA result.

What a historical model illustrates

The final 2016 JRC report provides a bounded refurbishment example. Its reuse situation 3 describes a machine completing an assumed 12.5-year first life and supplying a six-year second life after refurbishment. With unchanged operating efficiency and a replacement consuming 10% less energy, the model reports 5.0% lower lifecycle climate impact.2

Chapter 2 excludes same-owner repairs before the end of first use. This second-life reuse scenario is not an estimate for an ordinary in-home repair. Its end-of-life scope excludes later disposal and recycling credits.2

The percentage belongs to this historical scenario. It is not an eServUK repair result, a typical repair saving or a prediction for today's electricity supply. The comparator, service period and refurbishment assumptions must stay attached to it.

Operating emissions depend on how the machine is used

Shahmohammadi and colleagues modelled laundry washing using consumer behaviour, measured energy use and country electricity mixes. Their functional unit was one wash cycle. Importantly, the calculation excluded manufacturing the appliance, so it cannot independently answer whether buying a new machine is preferable.3

For a real comparison, programme choice, wash frequency and the electricity emissions factor should be recorded rather than hidden behind a single average. A household that rarely washes and one using the machine intensively are different cases. Today's decision also concerns future electricity supply. Applying an old grid factor unchanged across a long future period is an assumption that should be visible.

Repair needs to deliver additional useful service

The 2019 repair-centre study found that repair outcomes differed across faults. Door-related and some simpler interventions had different observed repair rates from bearings, drums, tubs or electronics. These observations concern equipment taken to a particular service operator; they do not establish a universal probability that a proposed repair will last.5

The practical implication is to ask what the diagnosis supports. A modest repair that restores dependable service and a major intervention followed by early failure should not receive the same assumed benefit. Parts, additional journeys, testing and repeat work belong in the repair pathway. An honest assessment should describe uncertain remaining service as a range or scenario, not a guarantee.

Efficiency improvements change the comparison

Alejandre and colleagues examine appliance replacement through scenarios involving both manufacturing and use-stage improvements.4 Their approach reinforces the need to consider a replacement's production burden alongside its operating performance. We have not adopted a replacement age or numerical efficiency threshold from the paper because its detailed model was not fully accessible during this review.

For readers, the useful distinction is between a verified operating difference and an attractive label. The relevant candidate should be identified, with comparable programme data and capacity. Buying a larger machine, changing washing frequency or using different settings can alter the service being compared. Those changes need their own explanation rather than being credited automatically to replacement.

A transparent question set for the decision

The following is our editorial decision framework, drawn from the issues above. It is a way to organise a conversation, not a validated emissions calculator:

1. What is the fault, and what work and parts does the repair actually require? Ask the appropriate professional to explain the diagnosis and any unresolved limitation.

2. What further service is reasonably expected? Distinguish a supported assessment from an optimistic assumption, and consider what happens if the item fails again sooner.

3. What is the genuine alternative? Identify the replacement's capacity, production information where available, transport, setup and likely service period.

4. How will either option be used? Record a comparable quantity of washing, relevant programme energy and the electricity context. Include material differences in water or consumables when the chosen assessment covers them.

5. What has been omitted? Travel, repeat repairs, disposal assumptions and a change in household behaviour can matter. State uncertain inputs and test whether a different reasonable value would reverse the conclusion.

CO₂, CO₂e and the scope of a climate claim

CO₂ is one greenhouse gas. CO₂e expresses greenhouse gases on a common climate-impact scale using a stated metric and time horizon.6 A lifecycle climate result should retain the original study's unit and assessment basis. A figure reported as kg CO₂e must not be relabelled as a direct measurement of CO₂ released during a repair.

Climate performance is also one part of the decision. Material demand, water, other environmental effects, affordability and safe operation are separate questions. A favourable result for one impact category is not proof of improvement in every category. This article addresses the climate comparison; it does not certify an appliance as safe or recommend continued use of a faulty item.

Evidence limitations and what would improve the answer

The selected literature combines models, an operating-emissions study and repair observations. These are complementary forms of evidence rather than equivalent measurements. The JRC durability work and later repair-centre paper may share underlying research and should not be counted as independent confirmations simply because their publication dates differ.

Much of the detailed modelling is historical and European rather than specific to a current UK household. Three additional papers were assessed through their original abstracts or records, limiting the depth of appraisal. There is no pooled effect, formal certainty rating or claim that the search found every relevant study.

Stronger future work would pair a documented fault and actual repair inventory with model-specific replacement data, a clearly defined amount of washing and plausible future electricity scenarios. Recording sensitivity to extra lifespan and repeat repairs would help readers see where a conclusion is robust and where the available evidence cannot decide.

What readers can take from this review

The reviewed evidence supports asking a better question than whether repair or replacement is always greener. Compare a specific repair and a specific alternative over equivalent service, with the assumptions exposed. Seek a clear diagnosis and understand the likely additional use before accepting a carbon-saving claim.

eServUK's repair-first approach is a reason to examine repair seriously. It does not supply the missing measurements or justify a standard emissions credit for every job. Where information is incomplete, reporting that uncertainty is more useful than publishing a precise-looking saving.

Disclosures

This article reports a secondary narrative synthesis, not new experimental data. eServUK has a commercial interest in repair services. Source studies may have their own funding and interests; the laundry study uses detergent-industry data and includes industry-affiliated authors. No external academic peer review of this article is claimed.