Global e-waste targets set for 2023 were missed, with documented collection and recycling at 22.3 percent in 2022

What is happening?

The world generated 62 billion kg of electronic waste in 2022, which is 7.8 kg for every person on the planet. Of that, 13.8 billion kg was documented as formally collected and recycled in an environmentally sound manner. That is 22.3 percent, or 1.7 kg per person. The word documented carries most of the meaning in the figure: it is the mass that compliant systems recorded, which is not the same quantity as the mass that was recycled.

The Global E-waste Monitor 2024, published by ITU and UNITAR in November 2024, accounts for the whole 62 billion kg through four management routes. Formal collection and recycling in an environmentally sound manner takes 13.8 billion kg. A further 16 billion kg is collected and recycled outside formal systems, inside high-income and upper-middle-income countries that already have developed e-waste infrastructure. Another 18 billion kg is handled in low-income and lower-middle-income countries with no developed infrastructure, mostly by the informal sector. The remaining 14 billion kg is disposed of as residual waste, the majority of it landfilled. Those four figures are rounded and sum to 61.8 against a total reported as 62.

Two global targets fell due in 2023 and the same report records that neither was reached. ITU set a target of raising documented collection and recycling to 30 percent by 2023; the measured rate for 2022 was 22.3 percent, and the report states that the world would not be able to meet the target. ITU also set a target of raising the share of countries holding an e-waste policy, law or regulation to 50 percent, which is 97 of 193 countries; as of June 2023 the count was 81, or 42 percent. That leaves 112 countries with no legal instrument of any kind.

In absolute terms the flow is pulling away from the record of it. Between 2010 and 2022 generation rose from 34 to 62 billion kg, an average of 2.3 billion kg a year, while documented collection and recycling rose from 8 to 13.8 billion kg, an average of 0.5 billion kg a year. The report describes the first as outpacing the second by a factor of almost 5. Measured proportionally over the same twelve years the two sit much closer together: 8 to 13.8 is a rise of about 73 percent, against about 82 percent for generation.

Whose problem is this?

The body that set the two targets for 2023 is also one of the two that publish the measurement against them, and the instruments that would deliver those targets sit with national governments instead.

Where does this problem end?

A documented rate and a recycling rate are not the same measurement. The complement of a documentation rate is not a rate of dumping, because a share of what goes undocumented is collected and recycled all the same, and the Monitor puts that share at 16 billion kg for 2022.

What is the state now, and what should it be?

Now

What it should be

The target state here is named by the institution that set it, and it was dated twice for the same year. ITU set a target of 30 percent documented collection and recycling by 2023, and a target of e-waste legislation in 50 percent of countries, meaning 97 of 193, by 2023. The measured rate for 2022 was 22.3 percent and the country count as of June 2023 was 81.

Past 2023 the Monitor publishes scenarios rather than a target, and there are three of them, all measured against generation projected at 82 billion kg for 2030. A business-as-usual scenario, which extends previous growth, gives 20 percent. A progressive scenario gives 38 percent and rests on an assumption that high-income countries reach an 85 percent collection level while other countries reach 10 percent. An aspirational scenario gives 60 percent. The report presents the higher two as reachable if all countries act, so the lowest of the three is the continuation case and not a forecast the report endorses.

How big is it?

Mass. 62 billion kg generated in 2022, at 7.8 kg per person, against 82 billion kg projected for 2030. The ITU launch release of 2024-03-20 states the rise to 2030 as 33 percent.

Composition. 31 billion kg metals, 17 billion kg plastics and 14 billion kg other materials, for 2022.

Metals are the largest single fraction by mass and they are also the fraction that carries the recoverable value, which is why the recovery and value figures below are reported for metals. The two release figures below are not confined to that fraction: one of them is mercury and the other is plastics containing brominated flame retardants.

Where it is generated and where it is recorded. Asia generates almost half the world total, at about 30 billion kg. Per person in 2022, Europe generated 17.6 kg and documented 7.53 kg collected; Oceania generated 16.1 kg and documented 6.66 kg; the Americas generated 14.1 kg and documented 4.2 kg. Some regions document below 1 percent.

Environmental release. Management outside compliant systems releases about 58 thousand kg of mercury a year, and about 45 million kg a year of plastics containing brominated flame retardants.

Value. USD 91 billion of metals was embedded in the electronic waste generated in 2022, and USD 62 billion of recoverable resources went unaccounted for. About 1 percent of demand for rare earth elements is met by recycled material.

The informal route recovers as well as exposes. Of the 18 billion kg handled in countries with no developed infrastructure, the Monitor records 7 billion kg recovered as viable metals. The same passage describes the methods as burning, leaching and melting, with manual dismantling and selective removal of the valuable components. The recovery and the exposure are the same activity, and a figure that reports only one of them describes half of it.

No affected population is derivable, and the reason is a finding rather than a gap. The World Health Organization states that the number of people working specifically in e-waste recycling is unknown, and offers only the International Labour Organization figure of 16.5 million children working in the industrial sector as a whole in 2020, with waste processing named as one subsector of it. The Monitor gives no global headcount either. Every published quantity in this document is a mass, a mass per person, or a share of mass, and none of them converts into a count of people without a term that no opened source supplies. That is recorded as not derivable rather than estimated.

Under what conditions does it arise?

1. The flow grows on its own schedule. Generation rose from 34 to 62 billion kg between 2010 and 2022, an average of 2.3 billion kg a year, and is projected at 82 billion kg for 2030. Formal capacity has to keep growing to hold a constant share, not merely to keep operating.

2. Recycling that is not recorded does not count toward the measure. 16 billion kg is collected and recycled outside formal systems, inside countries that already have developed infrastructure. Whatever else is true of that mass, it does not appear in the rate against which the targets are read.

3. Disposal is the cheapest route and it is domestic. 14 billion kg is disposed of as residual waste, the majority of it landfilled, and that happens inside the country that generated it.

4. Where infrastructure does not exist, the informal sector performs the recovery. 18 billion kg is handled that way, and 7 billion kg of it comes back as viable metals. Prohibition on its own removes the recovery along with the exposure, and no opened source describes a replacement that is already in place.

5. Movement across borders is a small part of the flow. Uncontrolled transboundary movement is 3.3 billion kg against 62 billion kg generated. That figure is carried into the 2024 report from a 2022 transboundary flows monitor rather than measured for the 2022 data year, so its chain of custody is one step longer than the route tonnages beside it.

What has been tried?

The one binding instrument on this list took effect after the data year of the measurements above. The amendments entered into force on 2025-01-01 and the measured year for the route tonnages is 2022, so no opened source yet reports what changed under them.

What was found?

Why is it still unsolved?

Institutional gap — most countries hold no instrument at all, the two global targets carried a date and nothing else, and among the countries that do legislate a minority set a recycling-rate target.

The first part is coverage. As of June 2023, 81 of 193 countries held an e-waste policy, law or regulation, which leaves 112 with none. The count moved from 78 to 81 across the four years from 2019 to 2023, so the instrument that the targets assume is spreading at slightly under one country a year. Population coverage looks far better than country coverage, at 72 percent, because the countries with the largest populations are among the 81; that number describes how many people live under an instrument and not how many jurisdictions have one.

The second part is that a dated target with no consequence attached to it is a date. Both 2023 targets came due, both were reported as missed in the report co-published by the body that set them, and no opened source identifies anything that follows from missing either. The report continues to measure against the same indicator afterwards, and it publishes scenarios for 2030 rather than a replacement target.

The third part is that the instruments in force mostly do not aim at the measured quantity. Of the 81 countries with an instrument, 48 set a collection-rate target and 37 set a recycling-rate target. The Monitor records that collection rates average 35 percent where legislation carries a collection target, against 22.3 percent worldwide, and then states in the same passage that targets are not a cure on their own and that rates are also higher where systems have been long established. That hedge matters, because maturity and target-setting travel together, and the source declines to separate them.

The fourth part is the one that changes where a remedy would have to act. Most of the mass that leaves the formal system never leaves the country that generated it. About 30 of the roughly 48 billion kg outside formal collection and recycling is either recycled outside formal systems inside countries that already have infrastructure, at 16 billion kg, or landfilled as residual waste, at 14 billion kg. Uncontrolled transboundary movement is 3.3 billion kg, against 62 billion kg generated. Border control is the part of this problem that has a binding instrument and it addresses the smaller share; the larger share is domestic and sits under national instruments that 112 countries do not have.

The fifth part is the indicator itself. The measured quantity is documentation of environmentally sound handling, and 16 billion kg is collected and recycled without entering it. A country can improve collection and recycling and move the rate very little, and a rate can fall while the recycled mass rises, because the denominator is growing at 2.3 billion kg a year.

What observation would mean it is solved?

Candidates — (a) documented collection and recycling rises as a share of generation, sustained across more than one edition of the measurement (b) the count of countries holding an e-waste instrument rises materially above 81 of 193, and the count setting a recycling-rate target rises above 37 (c) the mass handled without developed infrastructure falls from 18 billion kg (d) generation per person falls from 7.8 kg.

(a) alone can move for the wrong reason. The rate is a ratio with a denominator that grew 2.3 billion kg a year over twelve years, so it can rise because generation slowed rather than because collection improved. It can also rise because more of the same recycling got recorded, which is a real improvement in knowledge and no improvement in handling.

(b) is the one this failure type points at, since 112 countries have no instrument and only 37 aim at a recycling rate. It is also the slowest, because the count moved by three countries in four years.

(c) is the observation that would matter most for health, and it is the one with no measurement of people attached to it. The mass is published; the workforce exposed to it is not counted by any opened source.

(d) is the only candidate that reduces the problem rather than redistributing it, and no opened source names a target for it.

Read together, (a) reports the record, (b) builds the machinery, (c) moves the harm and (d) shrinks the flow. Any one of them can improve while the other three do not.

What is it connected to?

Fills with researcha related problem is recorded here as `global-plastic-recycling-stall`. Adjacent to this document are repair and product-lifetime rules, which shorten the flow before it becomes waste, and the recovery of critical and rare earth metals, where the Monitor puts the recycled contribution at about 1 percent of demand. Relation type and evidence grade were not established, and the text of the European repair directive was not among the opened sources.

What these sources do not say

• How many people work in informal e-waste processing. The World Health Organization states the number specifically in e-waste recycling is unknown, and offers the International Labour Organization figure of 16.5 million children in the industrial sector as a whole in 2020, with waste processing as one subsector. The Monitor gives tonnages by route and no headcount. This is an absence in the record rather than an omission in the reporting.
• How many people live near informal processing sites. No opened source gives a figure, so the documented health outcomes cannot be attached to an exposed population.
• What the resolution that set the 2023 targets actually says. Both targets and both misses are recorded in the Monitor. The ITU e-waste backgrounder page references standard L.1031 and does not carry the numeric targets, and the resolution text was not among the opened sources, so the legal character of the targets is not established here.
• How the two published growth figures fit together. The second edition gives average annual growth of 2.3 billion kg and the March 2024 launch material gives 2.6 million tonnes a year. No opened source reconciles them.
• What changed after the Basel amendments took effect. They entered into force on 2025-01-01 and the measured year for the route tonnages is 2022, so no opened source measures the flow under the new entries.
• Whether collection targets cause higher collection rates. The Monitor gives the 35 percent average and immediately states that targets are not a cure on their own and that rates are also higher where systems are long established. It does not separate the two effects, and neither does any other opened source.
• Route shares by named country. The four routes are reported by income group. No opened source breaks them down country by country, which is why this document carries no geographic breakdown.
• What the informal recovery would be replaced by. The Monitor records the methods and the 7 billion kg of metals recovered, and no opened source describes a formal system standing ready to take that mass in the countries where it happens.

See the evidence

One primary source was read as full text. The Global E-waste Monitor 2024, Edition 2, was downloaded and read directly, and every figure graded high above comes from it. The two launch releases and the publisher pages were read as pages rather than as the report, and figures taken only from them are graded medium; where a figure appears in both the report and a release, the report is the row that carries it. The World Health Organization fact sheet and the Secretariat page of the Basel Convention were opened directly. Two rows carry no URL and state why, and neither of them supports a claim in the body that is not carried by another row. Where the sources disagree the disagreement is left visible: the second edition and the March 2024 launch material give different figures for average annual growth, and this document cites the edition it read. The report is published under a non-commercial Creative Commons licence, so facts and figures are taken from it and no expression, table layout or image is carried over. This is a Path A output, a research-based definition, so observation_refs is empty and provenance_mode: press-derived.

People affected

Estimated range Not derivable

The reason and what is missing are listed under “What is missing” below