Data quality statement

The dataset helps track NSW’s performance on improving waste outcomes including reducing waste generation and increasing recycling by diverting tonnes from landfill. The dataset is also a vital resource for industry, community, and government to inform decision making and policy.

Data sources

The primary source for the dataset is data collected under legislation through the NSW Environment Protection Authority’s (EPA’s) Waste and Resource Reporting Portal (WARRP). Scheduled waste facilities are required to provide a waste contribution monthly report (WCMR) or an annual waste report (AWR) under the Protection of the Environment Operations (Waste) Regulation 2014 (‘Waste Regulation’).[1]

Table 1 shows the number and type of facilities that reported against each report type.

Table 1 Number and type of facilities reporting, by report type [2]

Period	Number of facilities reporting by WCMR		Number of facilities reporting by AWR
	Landfills	Recovery facilities	Landfills	Recovery facilities
2015–16	68	120	262	23
2016–17	67	136	251	31
2017–18	67	152	242	40
2018–19	68	165	238	42
2019–20	70	172	232	43
2020–21	71	196	212	64
2021–22	68	190	222	63

Facilities report on tonnes received by:

levy area
waste stream
waste type
source (monthly-reporters only).

Facilities report on tonnes transported by:

levy area (monthly-reporting landfills only)
waste stream
waste type
destination.

All facilities that are subject to paying the waste levy (i.e. those facilities which submit WCMRs) are required to weigh waste in and out of their site using a weighbridge, and most of the key annual-reporting resource recovery facilities also use weighbridges.

Table 2 shows the proportion (by weight) of waste disposed of and recycled that originated from a WCMR report, an AWR report, a non-WARRP source, and data from a different reporting period.

Table 2 Data source for disposal and recycling data

Period	WCMR data		AWR data		Non-WARRP data		Out-of-period data
-	Disposal	Recycling	Disposal	Recycling	Disposal	Recycling	Disposal	Recycling
2015–16	81%	68%	19%	11%	0%	5% [3]	0%	15% [4]
2016–17	79%	70%	21%	22%	0%	8% [5]	0%	<1%
2017–18	80%	71%	20%	20%	0%	9%	0%	0%
2018–19	78%	71%	22%	21%	0%	8% [6]	0%	0%
2019–20	77%	72%	23%	21%	0%	9% [7]	0%	0%
2020-21	75%	70%	25%	20%	0%	10% [8]	0%	0%
2021–22	76%	72%	24%	22%	0%	10% [9]	0%

Where WARRP data was not available, the following secondary sources were used:

recycling data requested from scheduled waste facilities that were not reporting in the WARRP
estimates of recycling by other scheduled waste facilities that were not reporting in the WARRP, using Environment Protection Licence information
IHS Markit export data
consultant reports on plastics recycling, commissioned by the EPA.

Data reliability and validity

The dataset is a reliable and valid estimation of recycling, waste disposal, waste diversion and generation activity within NSW. The following measures and controls are in place to ensure the quality of the dataset:

The WARRP data is subject to auditing by the EPA’s audit team, which checks facility reports against the facility’s weighbridge records. Facilities are identified for auditing based on periodic risk assessments.
The NSW waste regulatory framework actively supports the collection of reliable waste data by mandating reporting requirements, limiting stockpiles to ensure the material is moved through the waste system productively, and mandating the use of weighbridges to accurately measure waste flows for all levy-liable facilities.
It is an offence for facilities to provide false and misleading information under Section 66 (2) of the Protection of the Environment Operations Act 1997 (POEO Act).
The dataset calculations were subject to an extensive internal quality review.
The method and assumptions applied to the dataset have been independently and objectively peer reviewed by the following independent consultants
1. Analytecon
2. Arcadis
3. Blue Environment.

Summary of peer review findings

The independent peer reviews reported that the method:

is fit for purpose to estimate waste generation, recycling, disposal and diversion in NSW
ensures comprehensive coverage of recycling and disposal waste flows in NSW
effectively manages issues of double counting
has been applied consistently across waste types and waste streams.

The reviews made the following recommendations:

ensure transparency – all key assumptions and inclusions/exclusions should be clearly identified
quantify the impact of assumptions to allow users to make their own adjustments as required.

These recommendations have been addressed through publishing this Data Quality Statement and the calculation method.

Data consistency

The method and assumptions used to develop this dataset are different to those used in WARR progress reports published prior to 2015–16. This improved methodology means the latest dataset cannot be compared to waste data prior to 2015–16 as this data was collected using an inferior method.

The municipal solid waste (MSW) component of the dataset is not comparable to Local Government WARR Survey datasets due to methodological differences. These differences include:

The Local Government WARR Survey dataset includes only household data (i.e. residential kerbside, clean-up and drop-off) whereas this dataset also includes other council waste.
This dataset likely includes waste generated from interstate and overseas sources, whereas the Local Government WARR Survey should only include waste generated within NSW.
The methods that some councils use to calculate waste disposed and waste recycled can differ from the method used for this dataset. For instance, if a council in the non-levied area (NLA) receives organics at its landfill and uses this waste for landfill capping or other operational purposes, some councils count this as recycled waste, whereas this dataset counts this waste as ‘disposed of’.

This dataset is also not comparable with NSW data in the National Waste Report 2020 or earlier national waste reports because:

the recycling data in the National Waste Report 2018 or earlier was based on the previous NSW methodology
the national waste reports include an estimate of energy recovery from landfill gas whereas this dataset does not
the national waste report method does not include virgin excavated natural material (VENM) or soil in either disposal or recycling, but this dataset does. (Table 3 outlines the effect of counting VENM and soil in the recycled dataset.)

Table 3 Quantity of VENM and soil in Waste Recycled, tonnes per waste stream

Period	MSW	C&I	C&D
2015–16	-	-	1,435,000
2016–17	-	-	1,744,000
2017–18	-	-	1,860,000
2018–19	-	-	2,575,446
2019–20	-	-	2,040,953
2020–21	-	-	2,068,069
2021–22	-	-	2,003,751

Improved data coverage

Identifying and eliminating double-counting in the dataset

Facilities are required to report on waste transported from site under the following classifications:

transported for disposal
transported for recycling to another licensed waste facility in NSW [10]
transported for recycling to an unspecified facility or location (intrastate, interstate or overseas)
transported under a Resource Recovery Order specification.

Waste recycled was calculated as the sum of (c) and (d), and (b) where the receiving facility did not report in the WARRP.

Of the waste reported as being transported for recycling to an unspecified facility or location intrastate, a portion was identified through investigations as actually being transported to another licensed waste facility that reported in the WARRP. This data was excluded from the dataset.

Had the data not been excluded, an additional 2.1 million tonnes would have been included within the 2015–16 dataset, raising the overall diversion rate from 63% to 64%. For 2016–17 and 2017–18, not removing the double-counting would have increased the overall diversion rate from 65% to 67% (2.6 million tonnes were excluded in 2016–17 and 2.7 million tonnes were excluded in 2017–18).

Improving facility reporting to decrease the risk of double-counting has been a key focus for the EPA. In 2021–22, 207,800 tonnes of waste were identified as being double-counted, a reduction from 373,000 in 2020–21, and 438,000 tonnes in 2019–20.

Ensuring comprehensive coverage of recovery facilities

Resource recovery facilities outside the regulated area [11] are not required by legislation to report to the EPA [12]. Additionally, the WARRP did not have full coverage of metals recycling facilities for the reporting periods. To ascertain waste recycled at facilities which did not report in the WARRP, data was requested directly from these facilities when the data was expected to be significant, or informed estimates were made using administrative data, including information from Environment Protection Licences.

Managing imports and exports of waste

It is known that some resource recovery facilities source waste from interstate and overseas, and some disposal facilities source waste from interstate. However, the WARRP reporting templates do not capture the quantity of waste received from interstate and overseas sources, and facilities were likely recording this data as originating from the NLA. Better data capture on interstate and overseas sources may lead to exclusion of this data from future datasets.

Waste transported interstate and overseas for recycling is assumed in the method to be 100% recycled as it is difficult to ascertain what happens to the waste once it leaves the NSW jurisdiction. Waste transported interstate and overseas for recycling is outlined for each waste category in Table 4a and 4b respectively.

Table 4a Waste transported interstate for recycling, by waste type category [13]

Waste category	Tonnes transported interstate for recycling
-	2016–17	2017–18	2018–19	2019–20	2020-21	2021–22
Glass	59,000	71,000	59,000	124,000	107,000	101,000
Organics	2,000	10,000	26,000	94,000	47,000	88,000
Paper or cardboard	319,000	458,000	429,000	418,000	421,000	391,000
Plastics	4,000	3,000	2,800	19,600	7,500	7,500
Masonry	48,000	18,000	8,000	6,000	-	-
Metals	8,000	9,000	9,000	7,000	7,000	8,000
Other recyclables[13]	663,000	904,000	801,000	327,000	144,000	106,000
Total	1,103,000	1,473,000	1,334,800	995,600	733,500	709,000
Total as a % of total recycling	9%	11%	9%	7%	5%	5%

Table 4b Waste transported overseas for recycling, by waste type category [12]

Waste category	Tonnes transported overseas for recycling
-	2016–17	2017–18	2018–19	2019–20	2020–21	2021–22
Glass	16,000	8,000	1,000	6,000	-	-
Organics	-	-	-	-	-	-
Paper or cardboard	402,000	420,000	476,000	503,000	445,000	377,000
Plastics	58,000	55,000	53,400	39,900	43,000	48,200
Masonry	-	-	-	-	-	-
Metals	336,000	440,000	478,000	531,000	494,000	429,000
Other recyclables[13]	55,000	39,000	93,000	68,000	79,000	75,000
Total	867,000	962,000	1,101,400	1,147,900	1,061,100	929,200
Total as a % of total recycling	7%	7%	8%	8%	7%

Increasing granularity

Generally, increased granularity in the dataset decreases the data quality. Hence, data has been aggregated into waste categories rather than detailed waste types as per previous WARR progress reports. Specific issues are outlined below.

Waste stream

Facilities did not always identify the waste stream for waste received and transported [15]. Table 5 shows the proportion of waste reported as the Unknown waste stream, therefore requiring allocation to the MSW, Commercial and Industrial (C&I) and Construction and Demolition (C&D) waste streams. Further information on the waste stream allocation is provided in the EPA document Calculation Method for Waste Generation, Recycling and Diversion.

Table 5 Proportion (by weight) of received and transported waste reported as Unknown waste stream

Period	% of received waste	% of transported waste
2015–16	3%	30%
2016–17	6%	34%
2017–18	4%	30%
2018–19	4%	16%
2019–20	3%	16%
2020–21	1%	17%
2021–22	2%	19%

Some facilities were identified as having incorrectly reported the waste stream for waste received. Table 6 shows the effect on each waste stream by adjustments made to the reported waste streams for waste received.

Table 6 Effect (% by weight) on waste streams by adjustments to reported waste streams for waste received

Period	MSW	C&I	C&D
2015–16	+8%	-9%	<1%
2016–17	+7%	-6%	>-1%
2017–18	+10%	-9%	>-1%
2018–19	+12%	-11%	>-1%
2019–20	+10%	-10%	<1%
2020–21	+10%	-10%	<1%
2021–22	+12%	-11%	2%

The EPA was unable to find reliable data sources to identify the waste stream source for metal (for MSW and C&I only). This has resulted in some uncertainty around the allocated proportion applied for this waste type. Any significant revisions to this split in the future will likely impact the MSW and C&I recycling rates.

Facility reporting on levy area contained some irregularities, especially the likelihood that waste sourced from interstate and overseas is reported as NLA data. There are likely some errors in the dataset due to facilities reporting levy area as the area in which the facility is located rather than the source of the waste.

Data security

The EPA values the data provided by the NSW waste industry and makes significant efforts to ensure the protection of facility-level data. The method ensures that facility-level data cannot be identified. Facility-level data is protected from disclosure under Section 319 of the POEO Act as it is considered to be commercially sensitive.

[1] Facilities licensed under the Protection of the Environment Operations Act 1997 to manage levy-liable waste facilities (i.e. landfills, waste recycling facilities, waste storage, and waste transfer facilities) must submit WCMRs. Licensed waste facilities that are not required to pay the levy, as well as landfills not located within the regulated area of NSW (i.e. the area which the waste levy applies to), are required to submit an AWR.

[2] Some facilities provide both types of reports (i.e. landfills located outside the regulated area that receive waste from both the regulated area and the non-regulated area). This table does not include facilities that provided nil returns (i.e. no waste was received or transported within the period).

[3] Of the non-WARRP data, 48% was used to calculate metals recycling (only a limited number of metals recyclers and processors were reporting in the WARRP in 2015–16), 31% was used to calculate organics recycling (composters in the non-levied area are not required to report under the Waste Regulation) and 15% was used to calculate plastics recycling (only a limited number of plastics recyclers and processors were reporting in the WARRP in 2015–16).

[4] Of the out-of-period data (data collected outside of the 2015-16 FY but used to fill a data gap), 63% was used to calculate metals recycling (only a limited number of metals recyclers and processors were reporting in the WARRP in 2015–16), and 22% was used to calculate organics recycling (some data was not available for a major composter for the 2015–16 period).

[5] Of the non-WARRP data, 65% was used to calculate metals recycling (only a limited number of metals recyclers and processors were reporting in the WARRP in 2016–17), 24% was used to calculate organics recycling (composters in the non-levied area are not required to report under the Waste Regulation) and 8% was used to calculate plastics recycling (only a limited number of metals recyclers and processors were reporting in the WARRP in 2016–17).

[6] Of the non-WARRP data, 70% was used to calculate metals recycling (only a limited number of metals recyclers and processors were reporting in the WARRP in 2018–19), 10% was used to calculate organics recycling (composters in the non-levied area are not required to report under the Waste Regulation) and 13% was used to calculate plastics recycling (only a limited number of plastics recyclers and processors were reporting in the WARRP in 2018–19).

[7] Of the non-WARRP data, 77% was used to calculate metals recycling (only a limited number of metals recyclers and processors were reporting in the WARRP in 2019–20), 7% was used to calculate organics recycling (composters in the non-levied area are not required to report under the Waste Regulation) and 7% was used to calculate plastics recycling (only a limited number of plastics recyclers and processors were reporting in the WARRP in 2019–20).

[8] Of the non-WARRP data, 79% was used to calculate metals recycling (only a limited number of metals recyclers and processors were reporting in the WARRP in 2020–21), 11% was used to calculate organics recycling (composters in the non-levied area are not required to report under the Waste Regulation) and 6% was used to calculate plastics recycling (only a limited number of plastics recyclers and processors were reporting in the WARRP in 2020–21).

[9] Of the non-WARRP data, 77% was used to calculate metals recycling (only a limited number of metals recyclers and processors were reporting in the WARRP in 2021–22), 11% was used to calculate organics recycling (composters in the non-levied area are not required to report under the Waste Regulation) and 8% was used to calculate plastics recycling (only a limited number of plastics recyclers and processors were reporting in the WARRP in 2021–22).

[10] For WCMRs only, facilities had to specify the destination facility.

[11] The regulated area refers to the regions that are subject to the waste levy and comprises the Sydney metropolitan area, the Illawarra and Hunter regions, the central and north coast local government areas (LGAs) to the Queensland border as well as the Blue Mountains and Wollondilly LGAs.

[12] This had the largest impact on organics recycling data, as several large composters operate outside the regulated area. Several small Material Recovery Facilities (MRFs) also operate outside the regulated area to process commingled recyclables, however it was assumed that the majority of this recycling data is captured at an end processor reporting in the WARRP (e.g. a paper mill) or at co-located council landfills (which are required to report in the WARRP).

[13] This data may differ from other published figures on waste exported from NSW. These discrepancies are primarily due to differences in the waste types which are included in each waste type category and differences in the definition of ‘waste’.

[14] The data for ‘other recyclables’ includes mixed waste reported by facilities to be transported interstate for recycling. In 2016–17, it was 535,000 tonnes, 842,000 tonnes in 2017–18, 739,849 tonnes in 2018–19, 269,614 tonnes in 2019–20 and 100,515 tonnes in 2020–21.

[15] While the recycling data was based on transported waste, the received waste streams were used to reallocate the Unknown waste stream for waste transported and in calculation of Waste Disposed.