We are testing a new beta website for gov.scot go to new site

Proposed Plastic Bag Levy - Extended Impact Assessment: Volume 1: Main Report

Listen

4 Life Cycle Assessment

A number of LCAs have been undertaken that compare the environmental impacts of the reusable, plastic, degradable and paper bags typically available in high street shops. The studies have been carried out in the USA, France and Australia (see Appendix 3 for a full list). No studies have been carried out based on data from Scotland or the UK.

We reviewed the studies and identified the French study (carried out by Ecobilan for the retailer Carrefour) as the most relevant to the situation in Scotland (the rationale used for this selection is presented in Appendix 3). We believe that the information available from this study is sufficient to provide a good indication of the likely life-cycle environmental impacts of changing plastic bag usage in Scotland. The Carrefour study (as it will be referred to in this report) is used in the following analysis.

4.1 Stages of the LCA for this Report

The analysis proceeds through the following stages:

  1. Development of scenarios that will influence the numbers and types of bag used.
  2. Quantification of the number of bags of each type (lightweight plastic, reusable plastic, paper, and bin liners) used under each scenario.
  3. Review of the Carrefour study to extract the most relevant data for application in Scotland.
  4. Sensitivity analysis - designed to test the robustness of base case results to plausible variations on the assumptions made.

4.2 Plastic Bag Levy Scenarios

Table 4.1 gives details of the five scenarios investigated for this study, including 'business as usual'.

Table 4.1 Scenarios investigated for this study

Scenario

Summary

Description

0

Current situation

Business as usual

1A

As in the proposed Bill

Based on the introduction of a levy on all lightweight plastic carrier bags including degradable plastic bags, but NOT paper bags.
It includes all distribution points: shops, petrol stations, charity shops, on-street promotional give-aways, etc.

1B

As in the proposed Bill, but excluding small-to-medium enterprises ( SMEs), charities and promotions

Recognises the logistical problems of collecting a levy from all retail outlets. It assesses the extent of the environmental gain for the anticipated large-scale additional effort. The idea is to focus on the larger companies that use the greatest amount of bags and have the resources to enable them to comply more readily with a levy.

2A

As in the proposed Bill + paper bags

Based on applying the levy to all lightweight carrier bags including plastic, degradable plastic and paper.
Includes all distribution points: shops, petrol stations, charity shops, on-street promotional give-aways, etc.
Recognises that the levy is aiming to achieve behavioural change and encourage the use of re-usable bags and not simply a switch to, for example, paper bags.

2B

As in the proposed Bill + paper bags but excluding SMEs, charities and promotions

This scenario is the same as scenario 2A, but excludes SMEs, charities and promotions. Like scenario 1B, it looks at the extent of the environmental benefits without the logistical problems of trying to police and enforce the levy across the board.

4.3 Consumption Data Used to Quantify Environmental Impacts

To understand plastic bag consumption, we used published data to produce consumption figures for the different scenarios in conjunction with data on the impacts on consumers (see Section 5). These figures were derived as follows.

Existing Lightweight Carrier Bag Usage

  • A Defra report stated that 8 billion plastic bags were used in the UK in 2000 [Defra 2003].
  • Other sources [ BBC, WRAP 2005] put this figure at 10 billion per year, from which it has been stated that Scotland's consumption is 1 billion plastic carrier bags per year [Pringle]. This estimate presumes an approximate factor of 10%.
  • There are no actual figures available for the consumption of plastic bags in Scotland. Therefore, we used population statistics [Stats Scot, Stats UK] to scale UK bag consumption data to Scotland. Population statistics show that 8.6% of the UK's population lives in Scotland.
  • Average annual lightweight plastic carrier bag use in Scotland is estimated at 775 million 14.
  • In consultation with the BRC and its members, it was agreed that reusable bag consumption ('bags for life') constitutes an additional 1% 15.
  • There were no statistics available on the level of consumption of paper bags 16. We estimated that paper bag consumption is about 5% of all plastic carrier bag consumption 17.

Consumer Behaviour

In essence, the success of the levy will depend upon consumers' wish to avoid paying the levy and the consequent reduction in the use of plastic carrier bags. If fewer people pay the levy, less revenue will be generated.

If a levy is introduced and does not include paper bags, it is anticipated that there will be an increased take-up of paper bags as well as 'bags for life'. Our estimate of the take-up of alternative carrier bag options is based on 'assumed percentage reductions' as used in Australian [ DEH] and South African [ FRIDGE] studies.

Our interpretation of consumer behaviour is based on the following assumptions:

  • A levy would be charged at £0.10 per bag on lightweight plastic or paper carrier bags. This would lead to a 90% reduction in demand for each type of carrier bag, based on the experience in the Republic of Ireland.
  • Under scenarios 1A and 1B (in which paper bags are not subject to the levy), it is assumed that of consumers not purchasing a lightweight plastic carrier bag:
  • 30% will not require any type of carrier bag ('no bag');
  • 45% will switch to heavyweight plastic carrier bags (or similar);
  • 25% will switch to paper carrier bags 18.
  • Under scenarios 2A and 2B (which include paper bags in the levy base), it is assumed that of consumers not purchasing a lightweight plastic bag:
  • 42.5% of consumers will not require any type of carrier bag;
  • 57.5% of consumers will switch to heavyweight carrier bags (or similar) 19.
  • Under scenarios 2A and 2B, the estimated reduction in paper bags is assumed to result in a 70% switch to heavyweight carrier bags (or similar).
  • It has been assumed that a typical heavyweight carrier bag is used 20 times before replacement 20. Therefore, the 45% of consumers who choose to switch to a heavyweight carrier bag will purchase five such bags in place of 100 lightweight carrier bags. This gives a 1/20th ratio for calculating the numbers of heavyweight carrier bags used under the levy scenarios.
  • Spending at SMEs has been assumed to account for 30% of total household expenditure 21. In order to exclude SMEs from being subject to the levy, we have simply reduced total expenditure by households on items likely to involve the acquisition of a carrier bag (of any type) by 30%.

Bin Liner Consumption

  • We included bin liner consumption to account for the displacement effect of people switching to or using additional purpose-made bin liners instead of carrier bags in the event of a levy.
  • As no UK or Scotland specific data were available for current bin liner use, Irish data were used and scaled for Scotland along population ratios. An Australian study [ DEH] reports a 77% increase bin liner consumption in the Republic of Ireland, from around 91 million to 161 million , following the introduction of the PlasTax. We have assumed a similar 77% increase in bin liner use for Scotland, i.e. from 118 million/year currently to 208 million/year post-levy 22.
  • We have not included black refuse sacks and disposable nappy sacks as information on the relevant sales volumes was not available. In addition, there were no statistics available for bags made of polypropylene in Scotland. Although retailers felt that a levy would instigate an increase in sales of kitchen swing bin liners, they did not feel that it would alter their sales of black refuse sacks to any great extent [Nolan- ITU Pty Ltd, personal communication].

We combined the assumptions and data discussed above to give the annual bag and bin liner consumption shown in Table 4.2 for the different scenarios.

Table 4.2 Estimated annual carrier bag consumption under the different scenarios23

Total number of bags consumed under each scenario (millions/year) 24

0

1A

1B

2A

2B

Plastic carrier bag ( HDPE, lightweight)

775

78

287

78

287

Plastic reusable bag ( LDPE, heavyweight)

8

23

19

29

23

Paper bag (single use)

39

213

161

4

14

Total bags used

822

314

467

111

324

Bin liners

118

208

181

208

181

It is predicted that:

  • Under scenarios 1A and 2B, there would be a drop in lightweight plastic carrier bag usage of 697 million/year.
  • This decrease would not be so profound if SMEs were excluded (scenarios 1B and 2B) when it would be 488 million/year.
  • If paper bags were not included in the levy, there would be annual increases of 174 million paper bags under scenario 1A and 122 million bags under Scenario 1B.
  • 'Bags for life' would only increase by 11-21 million/year due to them being reused 20 times.
  • Bin liner consumption would increase by 90 million/year if SMEs were included in the levy (scenarios 1A and 2A), or 63 million/year if not (scenarios 1B and 2B).

We combined these data on bag consumption with information on the life-cycle environmental impacts of different types of bags to determine the relative environmental impacts of each scenario in Scotland (Sections 4.5-4.7).

4.4 Relevant Results from the Carrefour LCA

The assumptions and scope of the Carrefour analysis are summarised in Appendix 3.

The Carrefour study considered four types of carrier bag:

  • HDPE bags made from virgin polymer (lightweight plastic carrier bags).
  • Reusable LDPE bags made from virgin polymer ('bags for life').
  • Paper bags made from recycled fibres.
  • Biodegradable starch-based bags.

We have not considered biodegradable starch-based bags in the analysis of the Scottish situation because they are not thought to be used in any great numbers. Numbers for plastic bioerodable bags (made from HDPE polymer with trace degradant additives) are used at a few outlets, but considerably more conventional HDPE bags are used. We have assumed that the environmental life-cycle impacts of bioerodable bags are comparable to conventional plastic bags as they are both made from HDPE, albeit with a small addition of degradation-promoting compounds. The consumption of bioerodable bags is included within the consumption of lightweight plastic bags.

The Carrefour study examined energy, resource use and pollutant emissions over the whole lifecycle of the bags, i.e. it included production of the raw materials, manufacture of the bags, transport of the bags to the retailer, and disposal at the bags' end-of-life. For plastic bags, for example, the lifecycle begins with extraction and refining of oil and the production of plastic, pigments ink and glue.

In the Carrefour study, the lightweight plastic bags are manufactured in Malaysia, Spain and France, and the heavyweight 'bags for life' are manufactured in France. Paper bags made from recycled paper are produced in Italy for Carrefour. It has been assumed that the bags are produced from old newspapers/magazines.

The Carrefour study examined both incineration and landfilling of bags at the end of their life. For the base case, we selected data that reflect landfilling of the bags as a large proportion of all waste is sent to landfill in Scotland 25. However, we have also performed a sensitivity analysis that considers an alternative waste management strategy (see below).

The Carrefour study assessed the environmental impact of the energy use, resource use, waste generation and pollutant emissions from the lifecycle of each type of bag by examining their contribution to eight environmental indicators (see Appendix 3). Table 4.3 shows the environmental indicator score for each of the different types of bags, relative to the lightweight plastic bag, for the base case with all material sent to landfill at the end of the lifecycle.

The lightweight plastic bag has been given a score of 1 in all categories as a reference point. A score greater than 1 indicates that another bag ('bag for life' or paper) makes more contribution to the environmental problem than a lightweight plastic bag when normalised against the volume of shopping carried. A score of less than 1 indicates that it makes less of a contribution, i.e. it has less environmental impact than a lightweight plastic bag.

The indicators take account of emissions which occur over the whole lifecycle. They can therefore occur in different locations depending on where different parts of the lifecycle are located. For global environmental problems such as climate change, the location of the emission is not important in assessing the potential environmental impact. For other regional or local environmental impacts, however, it can be significant. For example, the impact of eutrophication of a water body will depend on the water characteristics. This is a well-known limitation of lifecycle impact assessment methodology: LCA quantifies the potential risk of environmental damage rather than actual harm.

Table 4.3 Environmental impacts of different types of carrier bag relative to a lightweight plastic carrier bag26

Indicator of environmental impact

HDPE bag (lightweight)

Reusable LDPE bag (used 2x)

Reusable LDPE bag (used 4x)

Reusable LDPE bag (used 20x)

Paper bag (single use)

Consumption of non-renewable primary energy

1.0

1.4

0.7

0.1

1.1

Consumption of water

1.0

1.3

0.6

0.1

4.0

Climate change (emission of greenhouse gases)

1.0

1.3

0.6

0.1

3.3

Acid rain (atmospheric acidification)

1.0

1.5

0.7

0.1

1.9

Air quality (ground level ozone formation)

1.0

0.7

0.3

0.1

1.3

Eutrophication of water bodies

1.0

1.4

0.7

0.1

14.0

Solid waste production

1.0

1.4

0.7

0.1

2.7

Risk of litter 27

1.0

0.4

0.4

0.4

0.2

There are two key stages in the overall production process as laid out in the LCA:

  1. Winning the raw materials from nature ( e.g. drilling for and then refining crude oil) and converting them into commodities ( e.g. polyethene granules).
  2. Manufacturing the bags themselves from these commodities.

The Carrefour study concluded that, for all bags, the main environmental impacts come from the first of these stages, i.e. the extraction and production of the materials (polyethene and paper) that are then used to make bags. The second stage ( i.e. the manufacture of the bags themselves) is generally of less importance though not negligible. The study found that transport contributed very little to the environmental impacts. The end-of-life phase also makes a significant contribution to some indicators - most notably, the production of solid waste.

The overall conclusion from the Carrefour study was that reusable plastic bags (so-called 'bags for life') are more sustainable than all types of lightweight carrier bags (plastic, paper, or degradable) if used four times or more (columns 4 and 5 in Table 4.3), offering the greatest environmental benefits over the full lifecycle of any bags used.

Figure 4.1 summarises these findings. Paper carrier bags have a bigger environmental impact than lightweight plastic bags in all categories apart from risk of litter. Paper bags have a particularly high impact on the environment in terms of 28:

  • Eutrophication of water bodies (rivers, lakes, etc.) due to pollutants released to water during the manufacture of the paper.
  • Water consumption.
  • Greenhouse gas emissions
  • Production of solid waste.

Figure 4.1 Summary of the environmental impacts of different carrier bags from the Carrefour LCA

Figure 4.1 Summary of the environmental impacts of different carrier bags from the Carrefour LCA

4.5 Applying the Results to Scotland

We used data from Table 4.2 on plastic bag and bin liner consumption in conjunction with the relative environmental impact scores in Table 4.3 to assess the relative environmental impacts of the four levy scenarios compared with the current situation (scenario 0, 'business as usual'). We used the assumption from the Carrefour study that a reusable bag is reused 20 times 29.

To allow an assessment of the predicted change in bin liner consumption, it was assumed that the lifecycle impact of manufacturing bin liners is the same as for HDPE carrier bags per unit weight 30. This is an approximation, which may overestimate the environmental impact of bin liners, and hence underestimate the benefits of the four levy scenarios. More details about the calculations are given in Appendix 3.

The results of the base case comparison are shown in Figure 4.2. The base case applies the results from the Carrefour study (Table 4.3) directly to the bag use data in Table 4.2. This implicitly accepts the use of French data on bag weights and volumes. The results give the percentage change in the environmental impact score for each of the levy scenarios compared with the current situation (scenario 0). In all scenarios where the levy is applied, consumption of non-renewable energy, atmospheric acidification, the formation of ground level ozone and the risk of litter fall considerably compared with the current situation.

In scenarios 1A and 1B where paper bags are exempt from the levy, the impacts are greater than the current situation for the consumption of water and eutrophication. However, they are approximately equivalent for the emission of greenhouse gases and the production of solid waste. This is due to a trade-off between the impacts from the additional paper bags consumed and the environmental benefits from the reduction in the use of lightweight plastic bags. The overall environmental impact from scenarios 1A and 1B is therefore predicted to remain very similar to today's situation. This is because the benefits of reducing plastic carrier bag use are displaced by the increased use of paper bags.

It is only in scenarios 2A and 2B, where the levy is applied to paper as well as plastic carrier bags, that consumption of water, emission of greenhouse gases, eutrophication of water bodies and production of solid waste are significantly reduced. This is because paper bags have a high score in these environmental categories relative to plastic bags (see Table 4.3 and Table A3.1 in Appendix 3).

In all cases, the environmental benefits increase (and environmental impacts reduce) when SMEs are included in the levy.

Fig 4.2 Change in environmental indicators due to a levy

These environmental effects will occur at different locations around the globe depending on where the raw materials are derived, where the bags are manufactured and how far they have to travel. The bulk of plastic bags for the Scottish market are made in the Far East and imported, whereas Scotland has a considerable paper bag manufacturing sector. Furthermore, some of the effects ( e.g. ground level ozone formation) are more localised and some are regional ( e.g. the consumption of water and emission of acidic gases), while others such as climate change resulting from fossil fuel combustion are global problems.

While we believe these broad messages about relative environmental impacts are applicable to the Scottish situation, there are differences between France and Scotland that mean that specific environmental impacts will differ. This is due to inherent France-specific assumptions in the original LCA work such as the characteristics and usage of bags, and to differences in the environmental impacts of manufacturing and waste disposal in the two countries. In particular, we note the following differences between the assumptions made in the French LCA and the situation in Scotland:

  • The Carrefour study assumed that plastic bags weigh 6g as opposed to 8g in Scotland.
  • The French study states that the paper checkout bags used by Carrefour weigh 52g. Paper checkout bags 31 in Scotland weigh 51g [ CBC]. In the LCA base case, the Carrefour value was taken as representative for Scotland as it was assumed that checkout bags would be more affected by a levy, in terms of numbers and nationwide coverage, than boutique paper carriers with handles. In the sensitivity analyses (see below), the test used the average weight of 99g for all types of paper bags. 32
  • The Carrefour study assumed that a plastic bag has a volume of only 14 litres while a paper bag has a volume of 20.5 litres. This means fewer paper bags are required for the same amount of shopping. For Scotland, however, we would expect no significant difference on average in the volume of shopping carried in the two types of bag. One reason for this is the tendency for 'double bagging', where customers use two paper bags instead of one because they are concerned that a single paper bag may rip open.
  • The Carrefour study takes for its base case an average waste management scenario for France, i.e. 45% of paper bags being recycled, 25% being incinerated and 26% landfilled. For the base case in this study, we used one of the Carrefour sensitivity analyses in which all waste is sent to landfill; this is much closer to the current Scottish position where 88% of waste is landfilled 33 [ SEPA].

Various sensitivity analyses are presented in Appendix 3 to demonstrate the robustness of results against these factors. These analyses are:

  • Sensitivity analysis 1: Assume paper bags weigh 99g instead of 52g.
  • Sensitivity analysis 2: Assume on average that paper and plastic bags are used to carry the same volume of shopping.
  • Sensitivity analysis 3: Assume lightweight plastic bags weigh 8g instead of 6g.
  • Sensitivity analysis 4: Combined effects of sensitivity analyses 2 and 3.
  • Sensitivity analysis 5: Assume the same split across recycling, incineration and landfill as in France.

The main results of the sensitivity analyses are:

  • Repeating the analysis using a higher bag weight or 'effective' volume of paper bags led to a significant worsening in the performance of scenarios 1A and 1B for all categories except for 'risk of litter'. The categories of solid waste generation and acid rain, for which a small benefit was originally recorded under the base LCA (Carrefour, 100% of end-of-life bags landfilled), became a disbenefit (to a lesser extent for acid rain). The effect on solid waste generation is driven by the greater weight of paper bags compared with plastic bags (this feeds directly through to waste generation at the end of the lifecycle) and by the waste produced during paper production.
  • Such effects are counteracted to a large degree by the assumption that lightweight plastic bags in Scotland are 8g compared to 6g in France.
  • The assumptions on alternative waste management strategies (sensitivity analysis 5) have little effect on the results.
  • The results for scenarios 1A and 1B are affected significantly by the sensitivities explored. This is as a result of encouraging people to switch from plastic bags to paper. Whereas, the results for scenarios 2A and 2B, where paper bags are also subject to the levy, show little change. In all cases studied and for all environmental indicators, scenarios 2A and 2B improved on the business as usual case by between 30% and 70%. The most restrictive scenario (2A, where all outlets including SMEs and charities are subject to the levy) shows a uniform improvement over scenario 2B of around 16% relative to business as usual.

It is important to recognise that the scores from the LCA represent potential risk and not actual environmental damage. Quantification of actual damage would require an impact pathway assessment that traces emissions from source to exposure to the quantification of impacts from specific industrial and waste management facilities. Such analysis is outside the scope of this report. It is noted, however, that some categories of effect are much more site-sensitive than others. For example, eutrophication of water bodies is only a problem where effluents are discharged untreated to a nutrient-sensitive water body. Climate change impacts, in contrast, are not sensitive to the site of the greenhouse gas release.

4.6 Displacement of Plastics in Scotland

In this section, we calculate the changes in tonnages of materials consumed in the scenarios based on the bag numbers data from Table 4.2 and the unit weights 34 for bags given in Table 4.4.

Table 4.4 Unit bag weights used in this study

Weight (grams per unit)

Lightweight plastic carrier bags

8

Paper bags

51

Heavyweight plastic carrier bags

47

Bin liners

15

Table 4.5 shows the estimated changes in the weight of carrier bags (tonnes) used across Scotland in scenario 1A compared with the current pre-levy situation (scenario 0). Note that paper bags are not subject to the levy in scenario 1A.

Table 4.5 Change in annual consumption of materials for scenario 1A*

Bag

Pre-levy consumption (tonnes)

Expected post-levy consumption (tonnes)

Expected absolute change 35 (tonnes)

Expected % change

Lightweight plastic carrier bags

6,200

620

-5,580

-90%

Heavyweight plastic bags; 'bags for life'

364

1,102

+738

+203%

Bin liners

1,764

3,122

+1,358

+77%

Total for polyethene

8,328

4,844

-3,484

-42%

Total for paper

1,976

10,869

+8,893

+450%

* Numbers have been rounded so may not add up exactly. Negative numbers mean less material used and positive numbers mean more material is used.

For Scotland, there would be a saving of 5,580 tonnes of polyethene from 90% fewer lightweight plastic carrier bags being used. This has to be balanced, however, against the increase in 'bags for life' and bin liners - a total of 2,096 tonnes. Taken together, these data show an estimated net decrease of 3,484 tonnes of polyethene consumed per year in Scotland. Paper bag usage would increase under this scenario by 8,893 tonnes per year.

The summary information for all four levy scenarios is summarised in Table 4.6.

Table 4.6 Change in annual consumption of materials for all four levy scenarios across Scotland

1A: Proposed levy

1B: Proposed levy excluding SMEs

2A: Proposed levy + paper bags

2B: Proposed levy + paper bags excluding SMEs

Decrease in polyethene consumption (tonnes)*

-3,484

-2,439

-3,214

-2,250

Change in paper consumption (tonnes)*

+8,893

+6,225

-1,779

-1,245

Net change (tonnes)

+5,409

+3,786

-4,993

-3,495

* Does not account for black refuse sacks or nappy bags.

In summary, it is predicted that polyethene amounts would reduce across all four levy scenarios, but that paper amounts would increase in scenarios 1A and 1B and decrease in scenarios 2A and 2B.

If paper carrier bags are not subject to the levy (as in scenarios 1A and 1B), the total tonnage of carrier bags used actually increases. This is because shoppers will switch from the relatively lighter plastic carrier bags to the much heavier paper carriers. Where paper is included in the levy, both show a decrease in the overall tonnage of waste material (paper and plastic) needing disposal. Scenario 2A, where paper and all businesses are levied, shows the best overall reductions (4,993 tonnes) relative to the situation today. Scenario 1A performs worst - waste actually increases by 5,409 tonnes per year.

4.7 Conclusions on Lifecycle Impacts

This study has used an existing published lifecycle study from France to gain an indication of the relative lifecycle environmental impacts of different types of bag. This has then been combined with estimates of changes in bag use under four levy scenarios to examine the resulting changes in environmental impacts from bag usage.

Using the Carrefour study introduces an element of uncertainty into the results owing to national differences between Scotland and France affecting the lifecycle, i.e. the way in which electricity is generated, the amount of transport required and final disposal methods.

However, based on the results of our various sensitivity analyses, we believe the pattern of environmental impacts described in the Carrefour study will be similar to those in Scotland. It is our view that the results described above are sufficiently relevant to Scotland to serve as a useful guide to decision-making on policies concerning carrier bags. However, for the reasons presented above , the findings in this report cannot be used for a precise quantification of environmental impacts. This would require a full lifecycle analysis based on the Scottish situation, which is outside the scope of this study.

The main conclusions from our analysis are:

  • The analysis shows that there would be an environmental benefit for some of the indicators depending on what consumers choose to use were a levy to be introduced.
  • More specifically, the biggest environmental improvement is seen in scenarios 2A and 2B where paper bags are included in the levy. These occur for all environmental indicators
  • In scenarios where paper bags are excluded, the environmental benefits of reduced plastic bag usage are negated for some indicators by the impacts of increased paper bag usage. This is because a paper bag has a more adverse impact than a plastic bag for most of the environmental issues considered. Areas where paper bags score particularly badly include water consumption, atmospheric acidification (which can have effects on human health, sensitive ecosystems, forest decline and acidification of lakes) and eutrophication of water bodies (which can lead to growth of algae and depletion of oxygen).
  • Heavyweight, reusable plastic bags (the so-called 'bags for life') are more sustainable than all types of lightweight plastic carrier bags if used four times or more. They give the greatest environmental benefits over the full lifecycle.
  • Paper bags are anywhere between six to ten times heavier than lightweight plastic carrier bags and, as such, require more transport and its associated costs. They would also take up more room in a landfill if they were not recycled.
  • The analysis demonstrates that SMEs and paper bags should be included to maximise the potential environmental benefit of the levy. The inclusion of paper bags in the levy makes a greater contribution to maximising environmental benefits than inclusion of SMEs.