Have We Got the Bottle? Implementing a Deposit Refund Scheme in the UK

Have We Got the Bottle?

Implementing a Deposit Refund

Scheme in the UK

A report for the Campaign to Protect Rural England

Authors:

Dominic Hogg

Debbie Fletcher

Tim Elliott

Maxine von Eye

September 2010

Eunomia Research and Consultancy

Eunomia is an environmental consultancy that supports positive change. Our clients, from

both government and the private sector, turn to us for policy development and analysis,

service design and review, technology and market assessment, regulatory compliance and

project management.

Policy Development

Our commitment to environmental issues dates back to well before these concerns were

prominent in the national political debate. Now as then, Eunomia seeks to lead the way in

policy formulation and implementation. We are able to marry perspectives from science,

economics, politics and social science to bring forward practical proposals with the

potential to deliver cost-effective benefits. We are recognised as leaders not only in

understanding the direction of waste and energy policy, but in determining its trajectory

through commissions for Defra, Department of Energy and Climate Change, Renewable

Fuels Agency and the Committee on Climate Change.

Local Government

Our technical skills, commercial experience and local government know-how mean our

clients turn to us when they want to achieve high recycling rates and high levels of resident

satisfaction as cost effectively as possible. In recent years we have worked with over 100

local authorities, supporting them with contract procurements, partnership development,

service efficiency reviews and service design.

Private

We act on behalf of a range of organisations operating in the waste and energy sectors,

including financial institutions, utilities, property companies and technology providers.

Eunomia is recognised as a leader in understanding the direction and trajectory of

environmental policy. We are therefore able to provide key insight to ensure that our clients

are one step ahead and can make key commercial decisions quickly and effectively.

Eunomia Research and Consulting Ltd

62 Queen Square

Bristol

BS1 4JZ

United Kingdom

Tel: +44 (0)117 945 0100

Fax: +44 (0)8717 142 942

www.eunomia.co.uk

Campaign to Protect Rural England

The Campaign to Protect Rural England (CPRE) is a charity that campaigns for a sustainable

future for the English countryside, a vital but undervalued environmental, economic and

social asset to the nation.

We want a beautiful, tranquil and diverse countryside that everyone can value and enjoy. We

promote positive solutions for the long-term future of the countryside, ones which respect

the character of England’s natural and built landscapes.

We aim to:

• Influence land use in town and country for people and nature.

• Protect and enchance beauty, tranquillity and local distinctiveness.

• Increase and harness public and political support for the countryside.

CPRE’s Stop the Drop campaign is working to stop the blight of litter and fly-tipping on our

countryside, cities, waterways, towns and villages.

Campaign to Protect Rural England

128 Southwark Street

London

SE1 0SW

Tel: +44 (0)20 7981 2800

Fax: +44 (0)20 7981 2899

www.cpre.org.uk

Registered charity number: 1089685

CPRE is a company limited by guarantee, registered in England, number 4302973

Disclaimer

Eunomia Research & Consulting has taken due care in the preparation of this report to ensure that

all facts and analysis presented are as accurate as possible within the scope of the project. However

no guarantee is provided in respect of the information presented, and Eunomia Research &

Consulting is not responsible for decisions or actions taken on the basis of the content of this report.

Acknowledgements

Our thanks to Canadean Ltd, TOMRA Systems ASA, Reverse Vending Corporation, Anker Andersen

A/S, Frank LeBlanc at the Department for the Environment in New Brunswick, REPANT ASA, Pasi

Nurminen at Palpa and Bryan Howell (HBH Business Solutions Inc.) for their inputs into the

discussion on costs associated with setting up a deposit refund system and for the provision of data

where requested.

Contents

1.0 Introduction and Background ...................................................................................5

1.1 What are Deposit Refund Schemes ...................................................................... 7

2.0 Deposit Refund Schemes in the UK .........................................................................7

2.1 Scotland ...................................................................................................................14

2.2 Summary View ........................................................................................................15

3.0 Economic Rationale for Deposit Refund Schemes............................................. 15

3.1 Summary View ........................................................................................................17

4.0 Possible Benefits of Deposit Refund Schemes ................................................... 18

4.1 Increasing Recycling ..............................................................................................18

4.2 Effects on Littering .................................................................................................24

4.3 Implications for Transport .....................................................................................31

4.4 Summary View ........................................................................................................32

5.0 Methodology for Cost Benefit Analysis ................................................................. 33

5.1 Summary of Existing Deposit Systems Worldwide ...........................................34

5.2 Mass Flows – Baseline and Scenarios ................................................................35

5.3 UK Deposit Refund System Model.......................................................................38

5.4 Cost Reduction in Existing Waste Collection Systems......................................46

5.5 Environmental Impacts..........................................................................................50

5.6 Cost Benefit Analysis..............................................................................................52

6.0 Results from Cost Benefit Analysis....................................................................... 54

6.1 Complementary System ........................................................................................54

6.2 Parallel System .......................................................................................................65

6.3 Sensitivities..............................................................................................................68

6.4 One-Off Costs...........................................................................................................78

6.5 Single Market Considerations...............................................................................79

6.6 Cross Border Issues – Private Trade in Alcohol bringing Non-Deposit

Containers into the UK .....................................................................................................82

7.0 Summary and Conclusions..................................................................................... 84

A.1.0 Review of Deposit Refund Systems................................................................... 92

A.2.0 Cost Benefit Analysis Model .............................................................................100

A.2.1 Materials to be Included in Deposit Refund System ..................................... 102

A.2.2 Baseline...............................................................................................................103

A.2.3 Scenarios.............................................................................................................113

A.3.0 The Deposit Refund System Model..................................................................119

A.3.1 The Deposit and Return Rates .........................................................................123

A.3.2 Handling, Collection, Logistics, and Processing............................................. 125

A.3.3 On-Going Costs for Central System..................................................................147

A.3.4 Material Revenues .............................................................................................149

A.3.5 Administration Fee.............................................................................................149

A.3.6 Set-Up Costs........................................................................................................150

A.4.0 Additional Cost Modelling .................................................................................154

A.5.0 Environmental Impacts .....................................................................................158

1.0 Introduction and Background

Eunomia Research & Consulting is pleased to present this report to the Campaign to

Protect Rural England (CPRE). The report investigates the environmental and

financial implications of the introduction of a UK-wide deposit refund system (DRS).

In April 2008, CPRE launched its Stop the Drop campaign against litter and fly-

tipping, with the twin aims of getting existing litter picked up and preventing further

litter being dropped. As part of the campaign it worked with Policy Exchange in 2009

to publish Litterbugs: How to deal with the problem of littering

, which detailed a

suite of proposals for addressing litter.

One of the key recommendations of that report was for the introduction of a

national deposit scheme, linked into broader waste and recycling policies, in light of

the research findings that deposit refund schemes (DRSs) significantly reduce litter

and help to promote virtuous cycles of behaviour.

Discussion regarding DRSs is often polarised between the views of ardent

supporters, and those of equally vehement opponents. The available theoretical

literature, however, suggests that such schemes are an efficient means of

increasing recycling rates and reducing litter, though a key issue in moving from

theory to practice is determining the costs of administering and implementing the

scheme.

A review of secondary literature, however, does not shed much further light on the

matter. In one such review, we recently concluded:

Views appear to be polarised regarding the costs of introducing such a

scheme, and few studies from the available literature would appear to enable

one to confidently assert that the approach incurs costs well above the

additional benefits which might be derived. There are, however, indications

that such schemes would need to be designed with great care to ensure costs

were aligned with the hoped-for benefits.

There is, therefore, a pressing need to understand what might be the costs and

benefits of introducing such a scheme, recognising that it would provide an efficient

means to increase recycling and reduce littering.

The aim of this report, therefore, is to investigate the costs and benefits of a UK-wide

DRS and advance the debate on the benefits and disadvantages of DRSs.

Through

bottom-up modelling, we sought to answer the following question:

Policy Exchange and CPRE (2009) Litterbugs: How to deal with the problem of littering, London:

Policy Exchange, 2009.

Eunomia et al. (2009) International Review of Waste Management Policy: Annexes to Main Report,

Report for Department of the Environment, Heritage and Local Government, Ireland, September

2009.

‘How do the benefits of introducing a UK-wide DRS for certain beverage

container packaging compare with the costs of implementation and

operation?’

The report is particularly timely given the devolved administrations active

commitment to achieving zero waste economies and the publication in May 2010 of

the Coalition’s Programme for Government, which states that:

“We will work towards a ‘zero waste’ economy, encourage councils to pay

people to recycle, and work to reduce littering”.

This report details how DRSs can offer the opportunity to reduce littering, to increase

recycling and consequently to increase the amount of waste that is diverted from

landfill and other residual waste treatment options. Individuals are able to return

their empty containers whilst ‘on the go’ and to subsequently recoup their deposit,

with the system thus encouraging them to deal with their waste in a responsible

manner. Hence the twin objectives of reducing littering and increasing recycling can

be met.

Significantly, this study uses logistics modelling to understand how the costs of

household waste collections change when the DRS is put in place. To our

knowledge, no study has carried out this work in a satisfactory manner. It is,

however, crucial for understanding the true costs (net of savings) of introducing a

DRS.

Furthermore, most existing studies only assume one scenario, where the existing

kerbside collection systems remain in place. This study examines the costs and

benefits associated with introducing a DRS in the UK under two scenarios. First, it

models a complementary system, which means beverage containers are no longer

collected at the kerbside. Second, it looks at a parallel system, where the household

kerbside systems for beverage containers target the same range of materials that

are covered by the DRS.

In addition, the report looks in considerable detail at the potential environmental

benefits associated with the increase in material collection, above existing systems,

as well as the potential savings derived from removing deposit-bearing litter from

the environment. It also seeks to understand the associated negative effects – the

disamenity - of litter, though the literature here is somewhat lacking and highlights a

clear need for further research.

We note that this research was undertaken by the consultants on the understanding that the results

may not be favourable. The aim was always to be as objective as possible, despite the orientation of

CPRE. CPRE was also aware that the results might not support its views.

HM Government (2010) The Coalition: Our Programme for Government, available at

http://www.cabinetoffice.gov.uk/media/409088/pfg_coalition.pdf

The closest any study comes to doing this adequately is a study by BDA Group in Australia. The

study most often cited by opponents of deposit schemes is one by BIO Intelligence Service, which

includes no serious attempt to model the change in the cost of the kerbside collection logistics.

1.1 What are Deposit Refund Schemes

DRSs have been defined as follows:

“A deposit-refund system is the surcharge on the price of potentially polluting

products. When pollution is avoided by returning the products or their

residuals, a refund of the surcharge is granted.” OECD, Glossary of Statistical

Terms.

A DRS encourages the return of the materials into an organised reuse, recycling or

treatment / disposal process. The producers typically finance the process through

the payment of an administration fee on each container. Drinks containers are the

most common target of DRSs, though economic theory suggests the schemes could

be applicable to hazardous materials and other waste streams, subject to

transaction costs being minimised. The systems can encourage recycling and / or

reuse where otherwise it is easy to dispose of containers with the residual waste or

for them to be discarded as litter. The same policy mechanism can also be used to

target difficult to dispose of, or hazardous, items to ensure that these do not reach

the residual waste stream. This can be considered a waste prevention policy as it

reduces the hazardousness of materials in the waste stream.

Non-drinks container examples include:

¾ Batteries (Swedish Östhammar example) and car batteries (Germany);

and

¾ Tyres (Maine, USA).

Finally, some countries, such as Sweden, make use of vehicle scrapping charges,

which discourages the dumping of vehicle bodies in rural areas and ensures that

cars are returned to registered scrapping destinations at the end of their life.

Appendix A.1.0 details a number of countries and states which have made use of

deposit schemes.

2.0 Deposit Refund Schemes in the UK

Some readers will be old enough to remember that in the 1970’s, in the UK, one

often paid a deposit on bottles of fizzy drinks and beer. When the drink was finished,

one would return empty bottles to the store, or even have them collected from the

front door as part of the milkman service, in order to retrieve the deposit. The

system led to high return rates for glass bottles, which were typically washed for

refilling. The bottles were designed for re-use many times over.

Since the broad demise of these schemes in the UK (though they are still applied in

some notable cases, for example the A. G. Barr scheme in Scotland), there have

been various studies which have looked at the use of deposit refunds here.

http://stats.oecd.org/glossary/detail.asp?ID=594

http://www.eeb.org/activities/waste/EEB-mini-brief-deposit-schemes-for-Batteries-March2004.pdf

The (sometimes temporary) scrapping charges which have become popular across nations in the

context of the current economic decline have their precedent in the more permanent schemes which

some countries employ to ensure that end-of-life vehicles are returned to an appropriate recycler.

This Chapter critically reviews the way in which previous Governments have

appraised the potential for using DRSs in the UK. It highlights the fact that particular

views have been formed, even in the absence of the evidence which would be

required to enable one to develop such views on an objective basis.

One of the initial review studies was undertaken by ERL (now ERM) in 1992.

The

study’s remit was to look at the general applicability of economic instruments in the

field of waste management. Deposit refunds scored poorly in that they were not

considered applicable to the bulk of waste being managed.

In the second half of the 1990s, the UK introduced its own mechanism for seeking

to ensure compliance with the obligations placed upon EU Member States through

the Packaging Directive. The mechanism which eventually emerged, following a

protracted gestation, was based upon tradable ‘compliance notes’, though opinion is

somewhat divided as to whether the tradability of these compliance notes – the

Packaging Recovery Notes (PRNs) and the Packaging Export Recovery Note (PERNs)

- was intended at the initial design stage.

Since 1997, the existing system – which leads to periodic setting of new targets,

associated fluctuations in the value of the PRNs and PERNs in line with the relative

tightness of their supply relative to their demand, and periodic anxiety regarding

whether and if targets might be met - has been the focus of policy-related activity in

the UK. This is despite the fact that the system has a number of shortcomings.

Perhaps the most notable of these is that the bulk of the cost of compliance does

not fall upon those whom it should most obviously fall i.e. the producers and

consumers of ‘packaged’ products. Instead, taxpayers shoulder much of the burden

of compliance, so that there is no relationship between packaging consumption and

the contribution made to meeting the costs of compliance. There are obvious

alternatives which would readily rectify the situation, such as taking the means of

financing compliance outside the tax system and placing it upon consumption, so

effectively aligning the system with the well-established polluter-pays principle.

Notwithstanding the continuation of the existing policy (albeit with periodic

revisions), there have been a number of occasions where alternatives were closely

ERL (1992) Economic Instruments and Recovery of Resources from Wastes. HMSO, London

Environmental Resources Management (then ERL).

One of the authors of this report was involved in work for what was then the Department for the

Environment, Transport and the Regions (DETR) to explore the nature of the changes which might

usefully be made to capitalise on the de facto tradability of PRNs and PERNs (see ECOTEC (1998)

Packaging Regulations: gathering Evidence on PRNs and the Impact of the regulations on Local

Authority recycling, Final Report to DETR, June 1998), the point being that the scheme did not

obviously exhibit some of the fundamental prerequisites which might be expected of a rational

scheme of tradable ‘evidence’.

As discussed below, the Packaging Strategy claims that an independent group reviewed

alternatives and effectively determined that ‘variations’ on the existing scheme were superior. As we

discuss below, we have not been able to find the evidence referred to in the Packaging Strategy.

considered. In the Strategy Unit’s review of waste strategy in 2002, it was suggested

that:

Action is needed in the following five areas to put the right long term

economic and regulatory framework in place: […]

[…] new measures to encourage reuse, such as deposit-refund schemes and

designing civic amenity sites for re-use;

The document included a number of recommendations, one of which was as follows:

Recommendation 3: Defra and WRAP should consider the options for

increasing incentives for the re-use of goods. More work is needed to assess

the preferred means for different products and to establish where the impact

on the waste stream would be greatest.

Defra responded to the Strategy Unit’s recommendations in 2003. It noted, in

response to the recommendation regarding re-use:

The Strategy Unit has suggested deposit refund schemes as one way of

encouraging individuals to participate actively in reuse and recycling. Defra is

commissioning a joint study, with other interested departments, to update

previous work on the benefits and costs of such schemes with a view to

identifying the contribution they could make to increased reuse and recycling.

The study that was subsequently commissioned was undertaken by Oakdene Hollins

in 2004.

The findings of this study were somewhat divided, according to what the

DRS was trying to achieve. The study noted that using DRS as an effective

instrument for promoting re-use in the UK would probably not deliver the required

payback, given that the majority of packaging is already one-way in the UK, and

hence there would be significant expense associated with setting up the refillables

processing infrastructure, and producers would be likely to continue using one-way

packaging anyway.

However, on the positive side, the study reported that introduction of a DRS in the

UK on one-way packaging would generate an increase in recycling and a reduction in

litter, and that, based on their approach of calculating the expected costs associated

with achieving recycling targets using the existing policy mix, introducing a DRS was

found to generate a surplus for every option. The study noted that the overall costs

of running such a system would be highly dependent on the return rate of

containers, and that the ‘unclaimed’ deposits were the principle cause of the overall

surplus. Final positive notes from the study included the following:

Strategy Unit (2002) Waste Not, Want Not: A Strategy for Tackling the Waste Problem in England,

December 2002.

Defra (2003) Government Response to Strategy Unit Report ‘Waste Not, Want Not’, HMSO:

February 2003.

Oakdene Hollins (2004) Deposit Return Systems for Packaging: Applying International Experience

to the UK, report prepared for Defra, December 2004

A DRS targeted on light-weight containers, cans and plastic bottles, would

integrate well with the existing policy mix that provides incentives for Local

Authorities to collect heavy and/or biodegradable wastes.

Establishing a new social habit of returning containers to redeem deposits

may create a platform on which other more sustainable patterns of

consumption could be built in the future.

This study was subjected to a Peer Review by Perchards. This review made some

legitimate comments of a critical nature regarding the report. It picked apart the

distinction between European and US-type systems, and made cogent remarks

about the operations of DRSs for refillables. At the same time, it made some critical

remarks which seem slightly unfair towards the study, notably around the authors’

views regarding the effects of a DRS on the PRN market.

In conclusion, the authors noted:

We conclude that the report does not meet the project specification. We

cannot see how it could be used to aid policy formulation unless it was

completely rewritten.

This is a strong recommendation, and one which seems somewhat uncharitable. The

Peer Review report itself is not without flaws, some of them amounting to basic

misunderstandings. It is usually the nature of a peer review to seek to improve the

quality of a report in progress rather than to completely cast it aside. On the Oakene

Hollins recommendation that a DRS be applied to selected non-refillable beverage

containers, they state:

the information provided in the report to support this conclusion is muddled

and disjointed. Although they have recommended this option, the report

lacks a thorough discussion of how such a DRS might be structured, how it

would operate, and how it would be funded. We challenge the proposal that it

should be funded by unredeemed deposits since this would create an

economic incentive to the operators not to achieve a high return rate.

The latter comment is instructive given that elsewhere, the peer reviewers suggest

that at the specified level of deposit, the return rate proposed by Oakdene Hollins

might be too high. It is inconsistent to suggest on the one hand that the suggested

return rate is too high, yet on the other to criticise the consultants for suggesting

that a DRS might be funded through unclaimed deposits. As we highlight in the work

below, the likelihood that system costs are covered by the revenue from unclaimed

deposits becomes high at the low return rates which Perchards seem to suggest

should be used at the consultants’ proposed level of deposit.

A more recent review was carried out by ERM.

Although this later study does not

fully reference the Oakdene Hollins work undertaken following the Strategy Unit

review, it states, in describing the background to the project, that:

Perchards (2005) Deposit Return Systems for Packaging Applying International Experience to the

UK, Peer Review of a Study by Oakdene Hollins Ltd., Report to Defra 14 March 2005.

Defra commissioned a report on deposit schemes in 2004. Its basic findings

suggested that a deposit system would be problematic. However, there has

been renewed interest in the possible use of deposit systems and the use of

reverse vending as a collection method for beverage containers.

There is no indication as to why the previous study found DRSs to be ‘problematic’.

In passing, we note that the ERM study continued:

In particular, the Campaign to Protect Rural England (CPRE) supports the

introduction of a deposit system as a means to reduce litter. CPRE has

recently launched a campaign with this aim in mind.

The ERM study set out to answer a series of questions. The approach to providing

answers to these was based largely upon gathering opinion from organisations

involved in the packaging chain (see Jane Kennedy’s written statement in May

2009, quoted later in this section). There was no attempt to understand the cost

implications of the introduction of such a scheme, nor any serious attempt to

quantify potential benefits. These were considered outside the study’s scope.

Consequently, the evidence required in order to answer a number of key questions

was never provided. Somewhat surprisingly, this did not stop the authors from

offering answers to these questions, even in the absence of the necessary evidence

to support them. Furthermore, these unsubstantiated views were taken forward in

the subsequent Government thinking of the time.

ERM made the following closing comment in their study:

It is not disputed that a deposit scheme would increase recycling, but

alternative schemes could achieve the same or better results at a lower cost.

This is not demonstrated in their work, and as such can only be regarded as

speculation. The study includes no figures relating to the costs of intensifying

kerbside collection systems relative to the costs of implementing a DRS. It also

reaches the view that deposit schemes do usually achieve higher recycling rates

than other schemes, so the final comment regarding the achievement of ‘the same

or better results’ actually contradicts views expressed elsewhere in the report.

The closest the study comes to providing evidence in respect of costs is a single

paragraph containing no figures:

Providing financial support for the expansion of council collection schemes

offers a relatively cost-effective means of continuing the trend of improving

rates. Furthermore, the investment would cover a much wider range of

materials – not only certain beverage containers, but other items of

packaging, and non-packaging materials as well.

Notwithstanding the absence of evidence, the same message is restated in Defra’s

Packaging Strategy of 2009:

ERM (2008) Review of Packaging Deposits System for the UK, Final Report to Defra, December

2008.

ERM (2008) Review of Packaging Deposits System for the UK, Final Report to Defra, December

2008.

In July 2008, Defra commissioned consultants ERM to study the feasibility of

setting up a deposit scheme for drinks containers in the UK. The study

examined whether deposit schemes were likely to increase recycling and to

reduce litter. It considered the role of reverse vending, the possible impact of

deposits on existing collection systems, the issues that would need to be

considered in setting up a deposit system, and possible alternatives to

deposits. It looked at the experience of deposits in four countries – Denmark,

Germany, the Netherlands and Sweden – to draw lessons for a possible UK

scheme. The study concluded that deposit systems are likely to increase

recycling but that other measures may achieve the same goals more cheaply.

An effective deposit and return scheme could divert materials from existing

arrangements such as bottle banks or kerbside collections, which have been

developed, for the most part, with public funding.

Under a different take-back model, retailers could introduce their own bring

banks for a wide range of packaging materials. Tesco and Sainsbury’s have

both trialed this in various formats. On the positive side, this could be seen as

the logical extension of producer responsibility, relieving local authorities of

their duty to collect packaging waste. On the negative side, the ability to

provide this infrastructure is limited by space, and accessibility for

householders is likely to be an issue. Also, this would deprive local authorities

of a source of revenue by taking valuable recyclates away. Were this system

to replace local authority collections, it is likely that the amount of packaging

collected for recycling would drop, as the effort of getting the recyclates to

the collection points would be more onerous for the public than kerbside

collections.

Yet another model would be for retailers and manufacturers to complement

local authority provisions by setting up bring banks for items that may not

otherwise be widely collected by local authorities. This has been done by the

carton industry body ACE UK (Alliance for Beverage Cartons and the

Environment) to ensure that cartons, such as those made by Tetra Pak are

widely collected across the UK. Deposit and bring systems which involve

human interaction tend to result in high quality materials. The design of

unsupervised bring banks makes a significant difference to their use and

contamination.

Unless take-back provisions are made mandatory, and cover the range of

materials which is currently being collected by local authorities, local

authority collections would need to continue in parallel. Overall, the operation

of two parallel systems is likely to increase costs to consumers and

taxpayers, with uncertain results on the quantities collected.

Some of this is based on supposition, whilst other statements present only half the

picture. The statements to the effect that local authorities would lose a source of

revenue, and that the operation of two parallel systems would likely increase costs,

are both statements which reflect only part of the overall picture. They make no

Defra (2009) Making the Most of Packaging: A Strategy for a Low-carbon Economy, June 2009

reference to the possible savings on logistics, and on sorting of local authority

collected materials, which would occur if a deposit scheme was introduced. They

also fail to take into account any additional benefits the scheme could provide.

The Packaging Strategy discussed deposit schemes as an alternative means to fund

packaging recycling. It noted:

In the second half of 2008, the independent Packaging Recycling Action

Group (PRAG) set up a working group to look at funding mechanisms. It

compared variations on the last three options above [ie. including deposit

refunds] against a set of criteria including, among other things, the likely

effectiveness of each option in terms of increasing quality and quantity, value

for money, ease of implementation and visibility to consumers and local

authorities. Variations on the current system consistently scored higher than

the alternatives.

There is no report arising from the PRAG working group referenced in the Packaging

Strategy. We have not been able to track down any document which arrived at this

view, and indeed, our attempts to do so suggest that the view attributed to the PRAG

– that variations on the current system offered the best way forward – does not

accurately represent the views expressed by the PRAG.

A briefing from the Industry Council for Packaging and the Environment (INCPEN)

may be referencing the same group when it states:

A group advising the UK government concluded recently that a deposit

system could cost between £1 billion and £7 billion to establish, depending

on how the system was set up.

The basis for this estimate is not made clear, and again, there is no study

referenced. The breadth of the range in the cost estimate suggests that the analysis

was a cursory one.

Similar lines have been taken by Ministers in Written Answers to Parliamentary

Questions. The then Environment Minister responded as follows to a written

question on 13 May 2009:

Mr. Drew: To ask the Secretary of State for Environment, Food and Rural

Affairs (1) what steps he has considered to increase levels of plastic and

glass bottle recycling; if he will bring forward proposals to promote bottle

return schemes in shops; and if he will make a statement; [273465]

(2) what measures he has considered to increase levels of plastic and glass

bottle recycling; what consideration he has given to tax incentives for (a)

companies who recycle, (b) companies who do not and (c) the promotion of

bottle return schemes in shops; and if he will make a statement. [273380]

Incpen (2008) Mandatory Deposits on Packaging, May 2008.

Written Answers to Questions, Wednesday 13 May 2009 Column 757W

http://www.publications.parliament.uk/pa/cm200809/cmhansrd/cm090513/text/90513w0001.ht

m#09051353000016

Jane Kennedy: […] In December 2008 DEFRA published a report into

packaging deposit systems and the role they might play in increasing

recovery and recycling of single use drink containers (plastic, aluminium and

glass) in the UK. The report was completed in consultation with a range of

industry stakeholders and reviewed deposit systems in four other EU member

states to assess the implications of introducing such a system in the UK.

The report concluded that while deposit schemes would increase recycling,

alternative schemes could achieve the same or better results at a lower cost,

as the relative cost of introducing a deposit scheme system was high. For

example, the deposit scheme operating in Germany costs three times as

much per container as a household collection system.

However, the Government are keeping an open mind in regard to deposit

schemes and ‘reverse vending’, where vending machines take used bottles

and cans for recycling and usually give a reward such as supermarket loyalty

points or vouchers. A number of reverse vending systems are being set up by

major retailers and the performance of these systems is being monitored.

It is not possible to speak about what is or is not ‘relatively cost effective’ without

knowing what the relative costs are. We have seen no information from any study

which would suggest that work to understand the relative costs has actually been

undertaken.

2.1 Scotland

The Climate Change (Scotland) Bill contains powers to introduce deposit and return

schemes in Scotland. In a consultation on legislative measures to implement a zero

waste policy, the Scottish Government recommended this in 2008.

However, this recommendation was diluted by the time of the consultation on the

Zero Waste Plan:

As outlined throughout the draft Plan, the Government, working with

partners, is already taking steps to improve recycling facilities across

Scotland and plans to take further steps.

For example, section 3.6 of the draft Plan notes that the Scottish Government

is convening a round table to consider what more could be done to establish

recycling zones in public places right across Scotland. If voluntary measures

to increase recycling should not succeed, then the Scottish Government

would consider if the regulatory route should be followed.

In the final version of the plan, there is no mention of DRSs, though there is mention

of a study to review options for ‘extended producer responsibility and “take-back”

schemes in Scotland.’

Scottish Government (2008) Consultation on Legislative Measures to Implement Zero Waste, July

2008, http://www.scotland.gov.uk/Resource/Doc/1056/0063943.pdf

Scottish Government (2009) Scotland’s Zero Waste Plan: Consultation, Annex N - Packaging,

August 2009, http://www.scotland.gov.uk/Resource/Doc/282143/0085295.pdf

2.2 Summary View

The apparently established line that alternatives to deposit refunds could achieve

the same or better results at lower cost than a deposit scheme remains to be clearly

demonstrated. Although it is suggested, in the previous Minister’s response, that the

past Government retained an open mind on this matter, the ease with which its

mind seems to have been closed contrasts with this avowed openness. The view of

ERM, and the opinions subsequently expressed by Defra and its Ministers, appears

to lack any firm basis.

There is a clear need for some objective assessment of what the relative costs, and

benefits might be for DRSs. This study is intended to fill this gap with regard to the

UK.

3.0 Economic Rationale for Deposit Refund

Schemes

DRSs are a particular form of product tax/recycling subsidy. In such programmes,

also known as ‘bottle bill’ programmes, consumers pay a deposit (tax) on a

container at the time of purchase. This should, in theory, be set at the extra social

cost of improper disposal over the net recycling cost (assuming there is already an

Advance Disposal Fee (ADF) on the manufacturer equal to the net recycling cost).

This means that if the product is improperly disposed of, that individual pays the

external cost of improper disposal by foregoing the refund, which would typically be

set equal to the initial deposit. As with many environmental policy instruments, the

theory is somewhat simpler than the practice. As we discuss later in the report, for

example, the basis for valuing the social cost of improper disposal is not as strong

as one would like.

The distinction between DRSs on the one hand, and ADFs coupled with a household

recycling refund on the other, is in the re-collection of the product at the end of its

useful life. Deposit schemes generally involve a separate collection path, rather than

being collected as part of the municipal recycling system.

Several theoretical studies have argued that a deposit/refund is the best policy in

the presence of illegal disposal.

Palmer et al modelled paper, glass, plastic,

aluminium, and steel. They found a substantial difference in the intervention levels

necessary to achieve reductions in disposal with the various policies. A $45/ton

deposit /refund would reduce all wastes by 10%. Alternatively, the government

Scottish Government (2010) Scotland’s Zero Waste Plan, June 2010,

http://www.scotland.gov.uk/Resource/Doc/314168/0099749.pdf

T. Dinan (1993) Economic Efficiency Effects of Alternative Policies for Reducing Waste Disposal,

Journal of Environmental Economics and Management 25: 242–56; D. Fullerton and T. C. Kinnemann

(1995), Garbage Recycling and Illicit Burning or Dumping, Journal of Environmental Economics and

Management, 29 (1); Peter S. Menell (1990) Beyond the Throwaway Society: An Incentive Approach

to Regulating Municipal Solid Waste, Ecology Law Quarterly, vol. 17, pp. 655-739; Hilary Sigman

(1995) A Comparison of Public Policies for Lead Recycling, RAND Journal of Economics, vol. 26, no. 3

(Autumn), pp. 452-478.

could obtain a comparable reduction using an ADF of $85/ton or a recycling subsidy

of $98/ton.

A key point is that the deposit/refund creates incentives for both recycling and

source reduction, whereas an ADF or a recycling subsidy takes advantage of only

recycling or source reduction in isolation.

However, it is important to note that the

theoretical studies, in abstracting from the real world situation, have not taken into

account all the potential costs of administering such schemes. In fact they are not

precisely modelling existing ‘bottle bill’ programmes, but rather a more generalised

version of deposits and refunds, applied ‘upstream’ on manufacturers and recyclers.

In a further study, Palmer and Walls accept that in practice there could be

significant administrative costs associated with refunding deposits, which could

reduce the efficiency of the approach.

This issue is discussed by Palmer et al with

numerical estimates of the effects of administrative costs on the overall efficiency of

deposit refunds relative to product taxes and recycling subsidies.

Viewing their

results alongside empirical evidence from Ackerman et al., they suggest that

administrative costs may be of the same order as the cost savings from using a

deposit/refund.

Due to such considerations, Palmer et al., Fullerton and

Kinnaman, and Palmer and Walls all argue that deposit refunds should be imposed

upstream on producers rather than on final consumers to minimise administration

and transaction costs.

Most theoretical studies recommend DRSs as economically efficient mechanisms to

increase rates of recycling.

This includes UK-based reviews, such as that

undertaken by Turner et al (1996) which stated:

K. Palmer, H. Sigman and M. Walls (1997) The Cost of Reducing Municipal Solid Waste, Journal of

Environmental Economics and Management 33, 128-50.

Palmer and Walls (1997) Optimal Policies for Solid Waste Disposal Taxes, Subsidies and

Standards. Journal of Public Economics 65(8): 193-205.

K. Palmer, H. Sigman and M. Walls (1997) The Cost of Reducing Municipal Solid Waste, Journal of

Environmental Economics and Management 33, 128-50.

Frank Ackerman, Dmitri Cavander, John Stutz, and Brian Zuckerman (1995) Preliminary Analysis:

The Costs and Benefits of Bottle Bills, Draft report to U.S. EPA/Office of Solid Waste and Emergency

Response, Boston, Mass.: Tellus Institute.

D. Fullerton and T. C. Kinnemann (1995), Garbage Recycling and Illicit Burning or Dumping, Journal

of Environmental Economics and Management, 29 (1); Palmer et al (1997); Palmer and Walls

(1997).

See, for example, Dinan, T.M. (1993) Economic Efficiency Effects of Alternative Policies for

Reducing Waste Disposal, Journal of Environmental Economics and Management, 25: 242-256.;

Fullerton, D. and Kinnaman, T. (1995) Garbage, Recycling and Illicit Burning or Dumping, Journal of

Environment Economics and Management, 29: 78-91; Pearce, D.W. and R.K. Turner (1993) Market-

based approaches to solid waste management, Resources, Conservation and Recycling 8: 63-90.

Porter, R.C. (1978) A Social Benefit Cost Analysis of Mandatory Deposits on Beverage Containers,

Journal of Environmental Economics and Management, 5: 351-375. Sigman, H. (1995) A Comparison

of Public Policies for Lead Recycling, Rand Journal of Economics 26: 452-478; Thomas Skinner and

Don Fullerton (1999), The Economics of Residential Solid Waste Management, NBER Working Paper

7326 http://www.nber.org/papers/w7326

; K. Palmer and M. Walls (1999) Extended Product

Responsibility: An Economic Assessment of Alternative Policies, Discussion Paper 99-12, January

The environmental economics literature has analysed the relative merits, in

economic efficiency terms, of a number of economic instruments and

regulatory approaches to the solid waste disposal and recycling balance issue

(Dinan, 1993; Fullerton and Kinnaman, 1995; Sigman, 1995). The general

finding was that policies focusing only on input use or on waste outputs

cannot generate the optimal (economically efficient) balance between

recycling, disposal and production output. Thus recycling subsidies directed at

input use cannot generate the efficient amount of waste disposal unless

coupled with a tax or subsidy on consumption. Primary product taxes need to

be coupled with both an output tax and a tax on other production inputs to be

economically efficient. A regulatory measure such as a recycling content

standard also cannot generate the efficient level of output and waste disposal

unless it is augmented by taxes on other inputs to production, together with

either a tax or a subsidy on the final product. Acquiring the detailed firm-

specific information necessary in order to set the efficient levels of taxes and

standard is clearly not a practicable proposition for public policy makers.

This same body of analysis finds that the deposit-refund instrument (in which

the product tax and the refund are equal to the marginal social cost of waste

disposal) is an efficient mechanism and is equivalent to taxing disposal (for

nonreturners) but without the attendant illegal disposal problems. The

efficiency advantage of the deposit-refund instrument will in practice be

reduced the higher the administration and consumer inconvenience costs

involved. (Porter, 1978; Pearce and Turner, 1993). […]

The waste management policy area is one in which a greater use of economic

instruments does seem warranted. The combination of landfill tax and

recycling credits to be introduced in the UK is in a rough and ready way a step

in this direction. This policy reorientation is however only partial and

efficiency gains still remain untapped, given the fact that the landfill tax does

not reflect the full social costs of waste disposal in the UK. Economic analysis

finds the deposit refund instrument to be highly rated in economic efficiency

terms and also has applications in some hazardous waste problems.

The theoretical case in favour of deposit refunds appears, therefore, to be quite

compelling.

3.1 Summary View

One could be forgiven for being surprised – given the generally supportive view in

respect of the theoretical literature – that deposit schemes are not more widely

applied than they already are. However, the theoretical studies tend to be

considered abstract from reality, in that they do not fully consider the costs of

1999, Washington DC: Resources for the Future; Don Fullerton and Amy Raub (2003) Economic

Analysis of Solid Waste Management Policies, in OECD (2004) Addressing the Economics of Waste,

Paris: OECD.

R. Kerry Turner, J. Powell, A. Craighill (1996) Green Taxes, Waste Management And Political

Economy, CSERGE Working Paper WM 96-03.

administering such schemes in real terms. Once again, this highlights the need for

studies which consider the range of possible costs and benefits which may be

associated with DRSs.

4.0 Possible Benefits of Deposit Refund

Schemes

DRSs are reported, in the literature, to have a range of possible environmental

benefits. The key ones mentioned in the literature are:

1. Increasing the recycling of containers covered by deposits (for refill or

recycling);

2. Reducing the extent of littering;

3. Increasing the use of / reducing the extent of decline in the use of refillables;

and

4. Avoiding harmful chemicals being mobilised in the environment (usually not

in beverage schemes, eg. lead acid batteries, or pesticides).

In addition, there are likely to be some effects on the efficiency of logistics with

regard to both kerbside collections and the DRSs. Significantly, this study uses

logistics modelling to understand how the costs of household waste collections

change when the DRS is put in place, as well as to establish the key drivers in terms

of the logistics costs of the DRS itself. To our knowledge, no study has carried out

this work in a satisfactory manner.

In the discussion that follows, we consider the literature in respect of the first two

issues. The third, regarding refillables, is not a key aspect of the scheme we propose

(see below).

The fourth does not obviously apply to packaging, though there is

some evidence to suggest that the effects of packaging on the marine environment

do indeed give rise to such concerns (touched upon below in the consideration of

effects on littering).

4.1 Increasing Recycling

Ideally, one has some indication of ‘before’ and ‘after’ performance, controlling for

other variables. To some extent, this is made difficult by the absence of usable data.

Surprisingly few studies actually take this approach.

Some data allows for comparison of performance in areas with and without

deposits. In the US, in 1999, the recycling performance of states with and without

deposits in place is shown in

Figure 4-1. The recycling rates, and the number of

containers recovered per capita, were far higher in the deposit states. However, this

could simply be evidence of the absence of adequate collection infrastructure in the

no-deposit states, so it cannot be considered as robust evidence.

For a discussion regarding the refillables issue, see Eunomia et al. (2009) International Review of

Waste Management Policy: Annexes to Main Report, Report for Department of the Environment,

Heritage and Local Government, Ireland, September 2009.

Figure 4-1: Performance of US States With and Without Deposits, 1999

Recycling rate (%) and per capita containers

recovered in the US (1999)

490

191

Deposit states (10) Non-deposit states (40)

Recycling rate Containers per capita recovered

Of some interest is the performance of deposit schemes in the context of wider

recycling systems. In Sweden, for example, the recycling rate for all plastic

packaging increased from 17% to 30% between 2003 and 2005 (44% in 2006). In

the same period, recycling rates for Polyethylene terephthalate (PET) plastics under

the deposit scheme were 77% to 82% (85% in 2007).

Once again, this, in and of itself, might not prove much. The components of plastic

packaging are many and varied, and PET bottles are readily recyclable. Perhaps

more telling, however, is the performance in respect of metals. Recycling rates for

all metal packaging were around 65% in 2004-2005, but the recycling rate for

aluminium under the DRS was 85% to 86% in the years 2002 to 2007. The return

rate for glass bottles is 99% on 33cl bottles and 90% on 50cl bottles.

In Denmark,

return rates in 2007 were 84% for cans, 93% for plastic bottles and 91% for glass

bottles.

Similarly, in Germany, recycling rates in 2005 were 50%, 85%, 76% and 79% for

plastics, tinplate, aluminium and glass respectively. The reported return rates under

the deposit scheme are 95-99%.

35,36

Figure 4-2 shows collection rates achieved in 2002 by international deposit

schemes. This shows that very few countries see low rates of return, with some

jurisdictions achieving close to 100% return rates. As would be expected under

economic theory, deposit scheme return rates increase as the deposit increases,

http://www.sverigesbryggerier.se/eng/1-emballage/1-index.html, accessed January 2009.

ERM (2008) Review of Packaging Deposit Systems for the UK, Report for DEFRA, December 2008,

accessed from http://randd.defra.gov.uk/Document.aspx?Document=WR1203_7722_FRP.pdf

Wolfgang Ringel (2008) The German Deposit System on One Way Beverage Packaging,

Presentation to the first Global Deposit Summit, Berlin 2008.

Data from the DPG (Deutsche Pfandsystem GmbH (System Operator)) in March 2010 puts

the 2009/10 return rate for PET bottles at 98.5%.

with higher deposits leading to an enhanced incentive (see Figure 4-3). Figures for

Denmark are shown in Figure 4-4.

Figure 4-2: Collection Rates for Non-refillable Containers in Deposit Systems, 2002

Note: Figures based on data collected from system operators, data from 2002

Source: Wolfgang Ringel (2008) Introduction on Deposit Refund Systems, Scottish Government Litter

Summit, Edinburgh 26

November 2008.

Figure 4-3: Relationship Between Level of Deposit and Return Rate

100

0,0 2,0 4,0 6,0 8,0 10,0 12,0 14,0 16,0

Deposit (€ cents per container)

Collection rate (% )

Source: System operators, Container Recycling Institute, Data from 1997-2002

Collection rate in deposit systems for non-refillable containers (%)

59 59

70 70 70

77 77 77

83 83 83

84 84 84

93 93

95 95

100

California

Prin ce

dward

New

oun

dland

Massachusetts

Connec

cut

Maine

New Yo rk

th A

ustr

alia

Brunswick

tar

Quebec

lber

enm

ark

Nova Scotia

Saska

hewan

sh Columb

Oregon

Swed

Vermon t

Iowa

way

Finland

anitoba

Michigan

Figure 4-4: Return Percentages of One Way & Refillable Beverages in Denmark

Source: Christian Fischer (2008) Producer Responsibility Schemes Versus Deposits and Taxes- Danish

Experiences, PRO Europe Congress, 15 May 2008

Taiwan differs from the European context, as the deposit refund scheme started

without any other producer responsibility systems in place. The scheme therefore

acted in isolation to increase recycling, first of PET and later with a number of other

materials. Taiwan now claims a 100% PET recycling rate, using its DRS.

Some have suggested that it is not the case that recycling rates are higher under

deposit schemes. However, those who suggest this usually do so on the basis of

reviewing recycling rates for all packaging. For example, the European Organisation

for Packaging and the Environment (EUROPEN) argues:

There are no compensating benefits with regard to an overall improvement in

recycling performance. The Perchards report showed that overall recycling

rates in Member States with deposit systems are not higher than those of

comparable EU countries where there are no special arrangements for

beverage containers.

Deposits, however, do not apply to all packaging. The Perchards report itself

states:

It is certainly true that deposit systems for non-refillable beverage containers

can achieve higher recycling rates for the beverage containers affected than

when these containers are handled through general recycling systems.

However European experience shows that deposit systems do not achieve a

EUROPEN (2007) Economic Instruments in Packaging and Packaging Waste Policy, Brussels:

EUROPEN.

G. Bevington (2008) A Deposit and Refund Scheme in Ireland, Report commissioned by Repak Ltd.,

September 2008.

higher recycling rate for all packaging of a given material, because beverage

containers represent too small a proportion of the total tonnage of that

packaging material.

Drinks containers typically represent only about 10% of all packaging and the

recycling rate for beverage containers in general recycling systems is likely to

be higher than the recycling rate for all packaging of the same materials.

The report then alludes to the performance of Belgium in respect of the recycling of

all packaging even though this is clearly not a good comparator for reasons which

the previous extract makes clear (the targeted materials – beverage containers –

are a relatively small fraction of all packaging). In particular, the largest fraction of

the packaging stream is always paper and card, which is also an easy, and relatively

low cost, material to recycle. Consequently, in most countries, the packaging

recycling rate will be heavily influenced by capture of a material that is irrelevant to

any sensible discussion regarding DRSs.

Perchards responded to a similar criticism in a previous report, where they had

earlier suggested that the recycling rates achieved in deposit schemes were no

greater than those achieved in other countries through presenting targets related to

all packaging.

They responded:

We decided to make a comparison based on overall recycling rates achieved

because the Packaging and Packaging Waste Directive repealed the Directive

on containers of liquids for human consumption, reflecting that the scope of

policy has broadened out to all packaging. A further expansion and/or

restructuring of EU policy is now under consideration which may result in

targets for all products of specific materials. Thus, specific arrangements just

for beverage containers go against the trend in EU policy in this area.

GRA has challenged our line of argument, saying that the fact that deposit

systems handle only a small amount of packaging is no reason not to have a

deposit system. GRA used a medical analogy to illustrate argument – if you

have a medicine that can cure 10% of the patients, but not the other 90%, is

there a reason not to use the medicine for the 10%? However, this is not an

exact analogy, because a medicine does exist for a significant proportion of

the other 90% of packaging, namely selective collection.

Several other stakeholders have also challenged the basis of our comparison,

arguing that a clearer picture would emerge from a comparison based on

recycling rates for household packaging waste alone, or, even better, of

beverage containers between deposit states and Member States relying on

packaging recovery systems. Unfortunately, though, we were not able to

obtain data which would have enabled us to pursue this suggestion.

This argument fails to counter the possibility that it might be possible to design a

system which targets the 10% through one system and targets the remaining 90%

through another. No one advocating DRSs is necessarily arguing against ‘selective

collection’ of, for example, cardboard or wood. Equally, several successful selective

Perchards (2005) Study on the Progress of the Implementation and Impact of Directive 94/62/EC

on the Functioning of the Internal Market, Final Report to the European Commission, May 2005.

collection systems for packaging – notably the Belgian one - achieve high rates of

recycling without targeting non-bottle plastics from households.

So, the question remains open as to whether a complementary system or, when

correctly modelled, a parallel system of selective collection and DRS might be

superior to either one acting independently. The rather obvious point is that the

costs of an existing recycling system are unlikely to remain the same when

substantial quantities of generally low density beverage packaging are no longer

being collected. This question has only really been examined in one study as far as

we are aware (see later in this Section). Generally, the question regarding costs has

not been properly answered for the simple reason that the question which really

matters has not been interrogated in sufficient depth. This research attempts to

shed light upon what the implied changes in costs and benefits might actually be.

This is not to deny the possibility of high recycling rates of packaging being achieved

without DRSs. Other EU countries, such as Belgium, have achieved impressive

recycling performance without them. Based on its Fost Plus managed packaging

collection system, Belgium recycled 67% of plastic bottles in 2007 (comprising both

Polyethylene terephthalate (PET) and High-density polyethylene (HDPE)) and 97.5%

of metal packaging (steel and aluminium cans).

Belgium has a producer

responsibility scheme in place which is fully funded by obligated industry. It also sets

targets well above those prevailing in the UK at present, and also has near-universal

implementation of so-called ‘pay as you throw’ schemes at the household level, a

policy which the Coalition Government has clearly set itself against. One might still

argue, even in this case, that there might be room for improvement through use of a

deposit scheme where plastic bottles are concerned.

In the UK, Alupro, the aluminium industry’s trade body, says 98% of English

households have kerbside collections of aluminium cans, but capture rates can be

anywhere between 30% and 70%.

The ‘cans-only’ recycling rate was estimated to

be 52% in 2008.

Therefore, even with a ‘free to the consumer’ system (in terms of

marginal cost), and very widespread coverage, the capture rate is still much less

than is seen in countries with a DRS. This may be partly a reflection of the fact that

35% of aluminium cans are consumed away from home, in the workplace, and at

sports, leisure and travel locations, according to Alupro. Consequently, this waste

stream would be one for which DRSs may be well suited, not least since such

containers would subsequently then be less likely to arise as litter, with individuals

encouraged to recycle whilst ‘on the go’ in order to claim back their deposits.

For plastic bottles, according to RECOUP, with 18.1 million (72%) of the UK’s

26 million households receiving kerbside collections (and bring schemes also in

place), the recycling rate in 2009 was 39%. If the same average performance was

maintained, the recycling rate would increase to 55% under universal coverage by

Fost Plus (2007) Annual Report, http://www.fostplus.be/files/EN/8/GB_AR.pdf

Ends Report (2009) Defra Report Rejects the case for Bottle Deposits, January 2009

http://www.endsreport.com/index.cfm?action=report.article&articleID=20119&q=deposit%20refund

&boolean_mode=all

Alupro website, http://www.alupro.org.uk/facts%20and%20figures.htm, accessed May 2009.

kerbside services. The 55% figure appears to be well below what is routinely

achieved by DRSs, though it must be stated that the RECOUP data covers all plastic

bottles, whereas deposit schemes do not always cover 100% of all plastic bottles

(see Section 5.2 below).

4.2 Effects on Littering

There is evidence to suggest that deposit refund policies can reduce litter and even

reduce the number of lacerations caused by glass in the environment.

Several one-

way deposit systems were implemented with the clear objective of reducing littering

(eg. Sweden, British Columbia, California, Michigan and others, with Hawaii a more

recent example of this trend). The potential for DRSs to be effective in reducing

littering has an intuitively plausible rationale - if the deposit is significant and the

consumer does decide to litter, the possibility exists that someone else will pick up

the container to redeem the deposit. INCPEN suggests that this can worsen the litter

situation in some cases. They make the statement that:

Perversely, a deposit can contribute to the litter problem. There have been

reports of homeless people emptying litter bins to obtain deposit containers,

leaving other items on the street.

There is no evidence offered to support this view.

The Container Recycling Institute suggested significant reductions in littering

following the introduction of deposits in some US states (see Figure 4-5). The effects

on used beverage containers (UBCs) and on total litter are shown as being between

70-80% and 30-40%, respectively. It must be said, however, that all studies of this

nature suffer in terms of the lack of clarity about the metric used to measure the

contribution of beverage containers to total litter. It is not clear what the most

relevant indicator should be (counts, volume, hazardousness, etc.) partly because no

systematic studies have been carried out, to our knowledge, to understand the

contribution of different attributes of litter to the disamenity suffered by those who

experience litter. There is also the significant matter of cost to be considered since

clearing litter costs an ever-increasing amount of public money. Indeed, recent

figures from the Department of Communities and Local Government show that the

amount spent by local authorities in England on cleaning up litter and street

cleansing rose by almost £100M in the last year to £858M in 2009/10.

M. Douglas Baker, MD, Sally E. Moore, and Paul H. Wise, MD, PhD, MPH, "The Impact of 'Bottle Bill'

Legislation on the Incidence of Lacerations in Childhood", American Journal of Public Health, October

1986.

Incpen (2008) Mandatory Deposits on Packaging, May 2008.

http://communities.gov.uk/publications/corporate/statistics/financialstatistics202010

Figure 4-5: Reduction in Littering in US States Linked to Deposit Schemes

Reduction of littering in 6 US states after the introduction of container

deposit systems.

100

UBC litter reduced Total litter reduced

NY OR VT ME MI IA

70%~80% reduction

30%~40% reduction

• Deposit on one-way

containers reduces

UBC littering by 70-

80%

• Deposit on one-way

containers also

reduces the total

amount of littering by

30-40%

Source: Container Recycling Institute, USA

The Policy Exchange and CPRE report, Litterbugs,

cited a study suggesting litter in

New York State declined by 30% in the wake of the use of a DRS.

Over the past 25

years, according to official figures, the New York State Returnable Container Act

1983:

1. Reduced container litter by 70-80% and roadside litter by 70%;

2. Achieved redemption rates between 65-80%.

Where counter-arguments to the ‘litter reduction’ effect are put forward, these very

rarely challenge the likely reality of this effect. Indeed, the counter-arguments tend

to adopt the view that this effect is not significant because beverage containers

constitute only a small proportion of litter.

Even if one accepts the argument that this might be true, implicit in the counter-

argument appears to be an assumption that if litter ‘is there’, then the amount of it

is not a matter of any importance, or more specifically, that the reduction in the

quantity of beverage packaging in litter is of no significance. Yet none of the

literature actually offers any evidence to support this implied claim. The validity of

the implied claim is also affected by the nature of the assumption (as highlighted

above) concerning the metric used to measure ‘litter’. What the right metric might

Policy Exchange and CPRE (2009) Litterbugs: How to Deal with the Problem of Littering, London:

Policy Exchange, 2009.

New York Public Interest Research Group, www.nypirg.org/enviro/bottlebill/myths.html ; Bottle

Bill Resource Guide, www.bottlebill.org/legislation/usa/newyork.html

Kruman J, Bottle Bill at 25, New York State Conservationist, August 2007, New York State

Department of Environmental Conservation, www.dec.ny.gov/chemical/8500.html

New York State Department of Environmental Conservation Beverage Container Deposit And

Redemption Statistics: October 2004 - September 2005, 2006.

be has not, as discussed above, been given adequate consideration by either

advocates, or detractors, of the effects of deposit schemes.

Generally, the argument tends to be that beverage containers are a small fraction of

litter, and that therefore, eliminating this would not solve the litter problem. A

limitation of this argument is that it assumes that the relevant indicator regarding

litter is the measure used in the surveys referred to, typically, ‘counts’ of litter.

It could easily be argued that the disamenity effect of litter is as much a function of

its volume, and possibly its potential to persist, than simply the number of items (ie.

the counts). Given the relative insignificance – in volume terms – of chewing gum

and cigarette ends, it could be considered that beverage containers actually

contribute significantly to litter-related disamenity due to their contribution to the

visibility of litter.

For example, a 2008 survey by ENCAMS for INCPEN highlighted that there were

44,040 counts of cigarette butts as compared with 582 counts of beverage

containers (201 counts of ‘soft drinks cans, 188 soft drink plastic bottles, and 90

alcohol cans, 44 alcohol glass bottles, 43 drinks cartons, 9 alcohol cartons, 4

alcohol plastic bottles and 3 soft drink glass bottles (see Table 4-1)). An article by

The 44,040

butts would occupy, therefore, 22.02 litres. By contrast, the bottles and cans would

occupy around 163 litres in their uncompacted form. In other words, though

accounting for around 1.5% of the counts as compare with cigarette butts, they

would occupy around seven times the volume in their uncompacted form. The

plastic bottles alone (which are less readily compacted than, for example, cans)

would occupy more than three times the volume of the cigarette butts at an average

size of 330ml (which could be an underestimate). This highlights the fact that if

count data is a poor proxy for perceived impact of litter, and if volume is a more

appropriate one, then beverage packaging is a significant contributor to litter, this

being disproportionately large relative to its prevalence in surveys based only on

‘counts’.

Table 4-1: Raw Data from ENCAMS 2008 Survey for INCPEN

Type of Litter Count Type of Litter Count

Gum Staining 175,690

Drinks Cartons 43

Cigarette Ends 44,040 Comm. Warehouse Packaging 40

Sweet Wrapper 916 Other Paper Litter 38

Cigarette Related 366 Post Office Elastic Bands 34

Other Litter 335 Drinking Straws 29

Discarded Food & Drink 220 Commercial Industry

Packaging

Soft Drink Cans 201 Commercial Office Packaging 24

Plastic Soft Drink Bottles 188 ATM Receipt 17

Kathleen M. Register (2000) Cigarette Butts as Litter—Toxic as Well as Ugly, "Underwater

Naturalist", Bulletin of the American Littoral Society, Volume 25, Number 2, August 2000,

http://www.longwood.edu/cleanva/ciglitterarticle.htm

Fast Food Packaging 188 Lottery Related 16

Soft Drink Bottle Tops 183 Commercial Packaging 16

Snack Packaging 168 Match Boxes 14

Matches 135 Commercial Flyers 11

3D Gum 118 Other Travel Tickets 9

Tissue 93 Cartons for Alcohol 9

Alcohol Drinks Cans 90 Construction other Materials 8

Till Receipts 85 Warehousing other Materials 6

Other Materials 79 Commercial Office other

Materials

Carrier Bags 78 Paper Bags 4

Travel Tickets 67 Plastic Bottles for Alcohol 4

Alcoholic Bottle Tops 60 Commercial Food Other 4

Drink Cups 52 Commercial Construction Pack 4

Lolly/Ice Cream Related 51 Batteries 3

Gum Wrappers 50 Commercial other Retail 3

Discarded Newspaper 47 Soft Drink Glass Bottles 3

Glass Alcohol Bottles 44 Telephone Related 2

TOTAL 223,915

Source: ENCAMS (2009) Litter Composition Survey of England, Aug-Oct 2008, report

for Incpen, March 2009.

In their review of the Oakdene Hollins report for Defra, Perchards were critical of the

study’s examination of the effect of deposit refunds on litter even though the

principal objective of the study was, as noted by Perchards to “ascertain either of

these approaches [ie. DRS for reuse and/or for recycling of packaging] will confer

positive benefit over and above current policy approach to managing” packaging

waste.” It is difficult to understand why this would not be a legitimate area of inquiry

given the overriding objective, but Perchards state:

‘litter abatement was not referred to in the Defra specification, and we

wonder why it was discussed so thoroughly’.

Interestingly, they themselves enter the debate around litter, highlighting, as they do

so, work from the US which demonstrated that DRSs have, on average, the effect of

eliminating 81% of all deposit-related litter in the US.

They also show results for litter reduction in the US, drawing on the work of Syrek,

which highlights the difference in the effect of DRSs when the effect on litter is

considered from the perspective of items, or from the perspective of volume. Studies

relying on counts indicate that DRSs may have a small impact if large numbers of

other items are present. On a volume basis, however, DRSs seem to achieve, in the

figures highlighted by Perchards, a reduction of the order of 33-38% in total litter

(the figures can exceed 30% on an item basis too).

Perchards (2005) Deposit Return Systems for Packaging Applying International Experience to the

UK, Peer Review of a Study by Oakdene Hollins Ltd., Report to Defra 14 March 2005

These figures hardly look like the basis for an argument not to examine the effects

of DRSs on litter. Rather, it looks like a very good one for doing so, the more so since

the author upon whose work they rely has noted:

‘Litter is usually considered to be first and foremost a visual form of pollution

where the larger items are more visible to pedestrians and doubly so to

motorists. However, the primary problem with including the small items is

they bias the results towards the less visible components of litter.

It should also be noted that the US systems discussed were not achieving the same

return rates as many European systems. The effect on litter reduction might be

expected to be higher in systems where the return rates are higher.

Perchards go on to cite a statement from the same US author who uses, as

representative of the cost of removing one beverage container from litter through

DRSs, the cost of handling 164 containers through the DRS scheme. The author

notes

Note that this analysis is concerned only with litter reduction and ignores any

impacts such a program might have on waste reduction or materials and

energy conservation

One might have thought that given that something else was happening with 163

other containers (being collected and recycled), this approach to analysing the costs

of litter reduction might have been deemed completely and utterly inappropriate, yet

this is given as credible evidence that DRSs are an expensive form of litter reduction.

The same claim is made, incidentally, by INCPEN:

Recent work by Syrek in 2003 shows that under US conditions, deposits are

by far the most expensive way of eliminating an item of litter. He also points

out that unlike the 1970s, when a relatively large percentage of containers

ended up as litter, recent surveys show that even in non-deposit states less

than 0.3% of all containers now end up as litter.

This is not credible research. It is rather like assuming that the cost of a meal for

one person at a wedding are equal to the whole cost of the wedding meal, when

there are 163 other people at the same wedding

In jurisdictions such as Hawaii, where the prevalence of beverage containers in litter

has been a motivation for the introduction of a DRS, the problem also extends to

pollution of the marine environment. One report from the State of Hawaii shows how

beverage containers have changed in terms of their prevalence in litter (debris) over

time.

The data are shown in Table 4-2 and Table 4-3.

Steven Stein and Daniel Syrek (2005) New Jersey Litter Survey: 2004, A Baseline Survey of Litter at

94 Street and Highway Locations, Report for the New Jersey Clean Communities Council, January 28,

2005. http://www.njclean.org/2004-New-Jersey-Litter-Report.pdf

INCPEN (2008) Mandatory Deposits on Packaging,

http://www.incpen.org/pages/data/MANDATORY%20DEPOSITS%20May%202008.pdf

State Of Hawaii Department Of Health (2008) Pursuant To Sections 342g-102.5(H), 342g-114.5(B),

And 342g-123, Hawaii Revised Statutes, Requiring The Department Of Health To Give A Report On

Table 4-2: Counts of Debris Found During Cleanup in Hawaii since the Start of the

Deposit Refund Scheme in October 2002

Beverage Container

Type

2003 2004 2005 2006 2007

Glass Bottles 7,687

11,362

7,194

5,759

5,008

Plastic Bottles 5,246

5,215

3,824

4,799

2,965

Metal Cans 4,946

6,894

3,518

3,959

2,932

Total 17,879

23,471

14,430

14,517

10,905

Source: State Of Hawaii Department Of Health (2008) Pursuant To Sections 342g-102.5(H), 342g-

114.5(B), And 342g-123, Hawaii Revised Statutes, Requiring The Department Of Health To Give A

Report On The Activities Of The Deposit Beverage Container Program, Report To The Twenty-Fifth

Legislature State Of Hawaii 2009, November 2008

Table 4-3: Percentage of Total Debris Collected During Cleanup

Beverage Bottles & Cans 2003 2004 2005 2006 2007

Glass, Metal, & Plastic 15.9%

14.5%

12.3%

8.7%

6.7%

Source: State Of Hawaii Department Of Health (2008) Pursuant To Sections 342g-102.5(H), 342g-

114.5(B), And 342g-123, Hawaii Revised Statutes, Requiring The Department Of Health To Give A

Report On The Activities Of The Deposit Beverage Container Program, Report To The Twenty-Fifth

Legislature State Of Hawaii 2009, November 2008

The report notes:

While there appears to be a downward trend in the number of bottles and

cans found at beaches, beverage containers, along with associated caps and

lids, continue to be a large portion of beach litter. This is why it is important to

continue to place a deposit on beverage containers to decrease the

temptation to litter and increase the incentive to recycle.

An interesting feature of the Hawaii data is that it shows the problem is not simply

one of terrestrial litter. Indeed, beverage containers appear to be (relatively) more

problematic in underwater cleanups (see Table 4-4).

Regarding plastics in particular, a UNEP report notes the prevalence of plastic

bottles, caps and bags among the key forms of marine litter giving rise to

increasingly serious problems at sea. Evidently, in the marine environment, it is the

The Activities Of The Deposit Beverage Container Program, Report To The Twenty-Fifth Legislature

State Of Hawaii 2009, November 2008.

longevity and potential harm caused by plastics that makes them of particular

concern.

Table 4-4: Top Five Debris Items Collected During the 2007 Cleanup

Land Cleanups Only Number of

Debris Items

Percent of

Total

Collected

1. Cigarettes & Filters 72,053 44.7%

2. Caps & Lids 21,210 13.1%

3. Food Wrappers and Containers 16,554 10.3%

4. Beverage Containers (glass, metal, plastic) 10,505 6.5%

5. Cups, Plates, and Utensils 7,331 4.5%

Underwater Cleanups Only Number of

Debris Items

Percent of

Total

Collected

1. Fishing Line 1081 54%

2. Beverage Containers (glass, metal) 393 19.6%

3. Cigarettes, Filters, & Cigar Tips 248 12.3%

4. Food Wrappers and Containers 55 2.7%

5. Caps & Lids 39 1.9%

Source: State Of Hawaii Department Of Health (2008) Pursuant To Sections 342g-102.5(H), 342g-

114.5(B), And 342g-123, Hawaii Revised Statutes, Requiring The Department Of Health To Give A

Report On The Activities Of The Deposit Beverage Container Program, Report To The Twenty-Fifth

Legislature State Of Hawaii 2009, November 2008

A study undertaken in Australia suggested that deposit schemes were likely to be

the most effective policy option for reducing litter amongst those considered for

improving recycling:

Ljubomir Jeftic, Seba Sheavly, and Ellik Adler (2009) Marine Litter: A Global Challenge, Report for

UNEP, April 2009,

http://www.unep.org/regionalseas/marinelitter/publications/docs/Marine_Litter_A_Global_Challeng

e.pdf

BDA Group (2009) Beverage Container Investigation, Report for the EPHC beverage Container

Working Group, March 2009.

A national CDS [container deposit scheme] is expected to provide the

greatest reduction in overall litter levels, with the potential to provide a 6%

reduction in the total national litter count and a 19% reduction in the total

national litter volume.

Finally, there is another way in which removal of used beverage containers from

litter could contribute to cleaner streets. Given that beverage containers are

relatively voluminous items, their removal from litter bins would leave more room

for other waste. The CPRE’s Litterbugs report confirms that 91% of the public believe

that increasing the number of bins is the most effective way of reducing litter.

useful parallel approach might be to free up space in existing bins. The report cites

the New York bottle bill as reducing container litter by 70-80%. Clean-up costs, as

well as landfill costs, were reduced. The scheme enjoys solid public support (84% of

voters in 2004) and so has been extended in 2009 to cover non-carbonated drinks,

which make up 27% of beverage sales.

4.3 Implications for Transport

Perchards suggested that introducing a deposit scheme in Ireland would be

unfavourable from an environmental perspective, partly owing to the duplication in

logistics:

Deposit containers would be transported separately for recycling from other

packaging. That would mean additional trucks, with increased energy and

carbon impacts. One set of trucks would collect containers from retailers and

another would transport packaging waste collected from the existing bring

and kerbside system. Deposit containers would have to be kept separate

from other packaging. It may sometimes be possible to collect both together,

as deposit packs would have to be in sealed containers, but collection

contracts would be awarded separately for each stream, so this arrangement

would often not be possible.

This argument is unconvincing and shows little comprehension of the factors

affecting collection logistics. Currently, in Ireland, as in the UK, much of the

collection of dry recyclables makes use of kerbside collections. The removal of low

bulk density beverage containers (plastic and cans) from the collected waste stream

would have the effect of freeing up volume in containers, and on recycling vehicles,

improving the logistics for the collection of other materials using the existing

collection schemes.

Given:

¾ That the logistics of picking up relatively large quantities per pick-up from

specific collection points would be favourable under a deposit scheme;

A. Lewis, P. Turton and T. Sweetman (2009) Litterbugs: How to Deal with the Problem of Littering,

Report for CPRE, March 2009.

G. Bevington (2008) A Deposit and Refund Scheme in Ireland, Report commissioned by Repak Ltd.,

September 2008.

¾ The absence of a need for sorting at a dedicated sorting facility (with its own

energy demand); and

¾ The likely very low loss rates in the recycling process because of the pre-

segregated nature of the material,

then the argument made would appear to amount to pure speculation, based upon

an extraordinarily unlikely scenario in which there is no effect on the logistics of the

pre-existing system, even when low bulk density materials are removed from that

collection.

Furthermore, if the recovery of containers through a deposit scheme is very high,

arguably, there is limited need for a ‘duplicate’ scheme. Indeed, this closely

approximates to the Danish approach. A considerable part of the literature speaks

of the possible duplication in cost of running parallel schemes. The commentary

overlooks the point that when captures are very high from deposit schemes, then

there is very little duplication, and kerbside schemes can concentrate on optimising

the logistics of collecting the remaining materials, such as paper and card.

4.4 Summary View

The evidence suggests that DRSs are likely to increase the capture of the targeted

materials for recycling. This is unsurprising as the deposit gives the purchaser an

incentive to take the material back to an appropriate location in order to generate a

refund.

The captures achieved appear to be influenced by the level of the deposit applied.

This is in line with what is expected by economic theory.

The schemes also appear to influence the prevalence of litter. It is true that deposit

schemes do not affect littering of items such as cigarette butts or chewing gum,

both of which are prevalent in terms of counts, but the contribution of beverage

packaging to the volume of litter appears to be disproportionately large relative to

its prevalence as revealed in surveys based on counts. Hence, DRSs affect the

volume of litter in a manner which is disproportionate to the prevalence of beverage

packaging in litter as recorded through ‘counts’. There are good reasons to believe

that the volume of litter (not the ‘count’) is what gives rise to the associated

disamenity.

The suggestion that transport impacts are significantly worsened when DRSs are

introduced is not based upon detailed analysis. There is a requirement to

understand how the logistics of the DRS reduce the transport movements in the

kerbside scheme (for recyclables as well as refuse). Only when these effects are

understood can the net effect be quantified.

5.0 Methodology for Cost Benefit Analysis

In order to examine the potential costs and benefits associated with the introduction

of a DRS in the UK, the following key steps were employed:

1) Review of existing DRSs worldwide;

2) Formulation of high-level design of a DRS to be modelled for the UK (including

range of materials);

3) Establishment of baseline tonnages of waste collected at the kerbside, through

bring sites, as commercial waste, via on-the-go recycling and from street

sweepings, and also the total number of units placed on the market. In order to

maximise the potential impact of introducing a DRS, we have modelled a system

that covers the following beverage container materials:

A) Plastic bottles made from PET;

B) Metal cans, both steel and aluminium; and

C) Glass beverage containers.

The modelled system targets non-refillable containers, because the market for

refillables in the UK is much smaller than for non-refillables and there will

typically already be systems of collection for re-use of refillables e.g. glass milk

bottles.

4) Determination of tonnages that would be diverted from each of these waste

flows into the DRS in:

A) a complementary system; and

B) a parallel system.

5) Establishment of the costs and revenues for both the complementary and

parallel DRSs;

6) Determination of changes in costs relative to the baseline achieved by changing

the flow of deposit-bearing waste into kerbside collections, bring sites,

commercial waste stream, on-the-go recycling and street sweepings, and the

DRS;

7) Determination of key environmental impacts (benefits and disbenefits)

associated with complementary and parallel systems in relation to the baseline.

To include the disamenity associated with litter;

Few systems cover, for example, cartons such as tetrapak. One of the reasons for this relates to

the shape of the containers. Advances in reverse vending machine technology are expected to make

it possible to include beverage cartons in future schemes.

CRR (2009) Policy Study: Refillables – Evaluation of Market Opportunity in the UK, Centre for

Remanufacturing and Reuse, August 2009, available at

http://www.remanufacturing.org.uk/pdf/story/1p317.pdf?-

session=RemanSession:42F9475818a2d30D7AXwp1883067

8) Estimation of the impact of the scheme on the flows of revenues through the

Packaging Recovery Note/ Packaging Export Recovery Note (PRN/PERN)

system; and

9) Pulling all figures together to produce a cost benefit analysis for the introduction

of a DRS in the UK.

The key inputs and assumptions used for each of these steps are outlined in the

remainder of this section.

5.1 Summary of Existing Deposit Systems Worldwide

In order to determine the high-level configuration of the DRS model for the UK, we

were able to draw on a substantial amount of existing literature regarding current

DRSs in place worldwide. This literature included the following studies:

¾ BIO Intelligence Service (2005) Environmental- and Cost Efficiency of

Household Packaging Waste Collection Systems: Impact of a Deposit System

on an Existing Multimaterial Kerbside Selective Collection System, report

written for Apeal.

¾ ERM (2008) Review of Packaging Deposits System for the UK, Final Report

produced for Defra, December 2008.

¾ Ernst & Young (2009) Assessment of Results on the Reuse and Recycling of

Packaging in Europe, report produced for the French Agency for Environment

and Energy Management (ADEME), March 2009.

¾ Eunomia et al. (2009) International Review of Waste Management Policy:

Annexes to Main Report, Report for Department of the Environment, Heritage

and Local Government, Ireland, September 2009.

¾ EUROPEN (2007) Economic Instruments in Packaging and Packaging Waste

Policy, Brussels: EUROPEN.

¾ Oakdene Hollins (2004) Deposit Return Systems for Packaging: Applying

International Experience to the UK, report for Defra, December 2004

¾ Oakdene Hollins (2008) Refillable Glass Beverage Container Systems in the

UK, Report for WRAP, 26 June 2008.

¾ Perchards (2007) Study on Factual Implementation of a Nationwide Take-

back System in Germany After 1 May 2006, Final Report, 14 February 2007.

¾ Robert C. Anderson (2004) International Experience with Economic Incentives

for Protecting the Environment, Report for US EPA, Nov 2004;

¾ Covec Report (2008) Potential Impacts of the Waste Minimisation (Solids)

Bill: Update Report, Report for Packaging Council of New Zealand, May 2008,

http://www.pca.org.au/uploads/00548.pdf ;

¾ N. Tojo, T. Lindhqvist and G. Davis (2001) OECD Seminar on Extended

Producer Responsibility, EPR: Programme Implementation and Assessment,

Seminar for the OECD, December 2001

¾ RDC-Environment & Pira International (2003) Evaluation of costs and

benefits for the achievement of reuse and recycling targets for the different

packaging materials in the frame of the packaging and packaging waste

directive, Report for the European Commission, March 2003

¾ D. O’Connor (1996) Applying Economic Instruments in Developing Countries:

From Theory to Implementation, OECD Development Centre, May 1996

A summary of the international DRSs currently in place can be found in Appendix

A.1.0. From the literature available, and predominantly based on previous work

undertaken by Eunomia, we were looking to extract the key lessons learned from

existing systems in order to determine a preferred high-level model for the UK.

outline of the chosen system is given in Section 5.3.

5.2 Mass Flows – Baseline and Scenarios

The first step in the development of the quantitative model was to decide which

beverage containers could be covered by a deposit scheme.

The materials we have included in the DRS are one-way (non-refillable) beverage

containers. The following containers were considered to be relevant:

1) Plastic bottles made from PET eg. containers for fizzy drinks, mineral water,

squash. The recycling symbol on these products is:

2) Metal cans, both steel and aluminium eg. containers for fizzy soft drinks,

alcoholic beverages, energy drinks etc.

3) Glass beverage containers eg. beer bottles, wine bottles, soft drink bottles etc.

Although there is, strictly speaking, no reason why, in theory, other containers or

packaging could not be collected in these systems, the model has been designed

around beverage containers for the following key reasons:

¾ More investment in technology would be required in order to enable

recognition in reverse vending machines (RVMs)/counting centres for other

types and, importantly, shapes of containers/packaging.

¾ Beverage containers are more likely than other types of food-based

containers to be consumed away from home and thus end up as litter.

¾ Hygiene issues – particularly in association with plastic milk bottles, but also

for other food-based containers; hygiene issues associated with milk bottles

Eunomia et al. (2009) International Review of Waste Management Policy: Annexes to Main Report,

Report for Department of the Environment, Heritage and Local Government, Ireland, September

2009.

http://www.bottlebill.org/about/benefits/curbside.htm

have been stated as a reason for not including high-density polyethylene

(HDPE) in existing DRSs.

The modelled system targets non-refillable beverage containers, because the

market for refillables in the UK is much smaller than for non-refillables and there will

typically already be systems of collection for re-use of refillables, eg. glass milk

bottles.

For the purposes of this modelling, we have thus considered refillables as

‘exempt’ from the deposit scheme. If the system we are proposing were to be

implemented, then we would suggest that exemptions for refillables should be

supported only where return rates exceed a minimum level (probably in excess of

75%). This minimum return rate would ensure that any producers seeking to switch

to refillable beverage containers in order to circumvent the deposit scheme would

still need to meet relatively high return rates for these containers, and would thus

also contribute to the improvement in the management of beverage containers

across the UK. Focusing on non-refillables in the DRS will exploit the potential for

increased recycling rates, lead to an increase in the quality of material collected for

recycling through the deposit mechanism and reduce litter levels.

The second step in building the cost benefit analysis model was to consider the

current material flows in the UK, where the waste arises and how much of the waste

is sent for recycling compared to how much requires disposal. Figure 5-1 indicates

the possible material flows in our container universe (before the DRS).

In this study we have assumed the baseline year is around 2015; the landfill tax

escalator will have increased to £80 per tonne by 2014/15, and by this time, it is

also likely that fully comprehensive kerbside collection services will have been rolled

out to all households in the UK.

In order to provide direct comparison, the ensuing

changes to the baseline as a result of the introduction of a DRS have also been

modelled based on the same year.

Relevant data sources covering all of the elements shown in Figure 5-1 were used to

extract information relating to the containers we are considering. Many of the data

sources do not provide disaggregation down to the level required, so some

assumptions were required to break down material fractions into the following

elements:

¾ Glass Bottles ≤500 ml

¾ Glass Bottles >500 ml

¾ PET Bottles ≤500 ml

¾ PET Bottles >500 ml

ERM (2008) Review of Packaging Deposits System for the UK, Final Report produced for Defra,

December 2008.

CRR (2009) Policy Study: Refillables – Evaluation of Market Opportunity in the UK, Centre for

Remanufacturing and Reuse, August 2009, available at

http://www.remanufacturing.org.uk/pdf/story/1p317.pdf?-

session=RemanSession:42F9475818a2d30D7AXwp1883067

‘Fully comprehensive’ means a system that would collect all of the containers within the scope of

this study.

¾ Cans (Ferrous)

¾ Cans (Aluminium)

All the data sources and assumptions are set out in Appendix A.2.0.

Figure 5-1: Possible Container Material Flows (Pre-Deposit Refund System)

Source: Eunomia

The final step was to model the quantitative impacts of introducing a DRS. In order

to show some of the variation in the overall costs of the system in relation to

different potential configurations that might occur in the UK, two scenarios were

considered as follows:

Complementary – where no beverage containers are collected at the

kerbside ie. the DRS is complementary to the existing kerbside schemes;

and

Parallel – where the household kerbside systems for beverage containers

operate in parallel to the DRS.

Some general principles relating to each of the scenarios were considered. These

principles centred on the likely return rates that might be expected in the

complementary and parallel DRSs. The modelling then established what containers

would be left in the remaining current waste collection routes (kerbside collections,

bring sites/Household Waste Recycling Centres (HWRCs), commercial waste

collections, on-the-go recycling, street sweeping, left in the environment), and the

subsequent costs and benefits associated with both the DRS and other waste

collection routes. The approach taken was to calculate the change in material flows

that is brought about by the introduction of a DRS and, from this, to calculate the

change in cost (rather than building up the total costs of managing all beverage

containers in the UK both with, and without, a DRS). Nonetheless, for some

elements of the system, changes could only be calculated through understanding

absolute costs both before and after the introduction of the deposit scheme, and

then subtracting the one from the other (eg. kerbside collection costs). The

assumptions relating to the mass flows in the system are fully considered in

Appendix A.2.2.

5.3 UK Deposit Refund System Model

The various stakeholders involved in operating a DRS are likely to include:

¾ A government body authorising the system and associated finances, and

setting recycling targets for the various materials;

¾ A central organisation owned and run (within the constraints set by the

authorising body) by, for example, non-governmental organisations (NGOs),

industry bodies, producers, breweries and retailers;

¾ The manufacturers of containers, producers and importers of beverages and

industries that ‘fill’ the containers;

¾ Any retailers which sell beverages in the UK;

¾ All consumers which purchase beverages in the UK; and

¾ Businesses and organisations involved with the collection, sorting and

reprocessing of waste containers.

Within the DRS, various stakeholders are involved in the material flows of beverages

(pre and post-consumption), deposit payments, other finances and sales or

container return data. An overview of the key elements, material and finance flows,

in the UK’s DRS model developed for this study is given in Figure 5-2.

Figure 5-2: Deposit Refund Model

Source: Eunomia

The system developed for this study is based on similar principles to the systems

which exist in Denmark (Dansk Retursystem) and other Scandinavian countries

(Norsk Resirk, Returpack and Palpa), and in a number of provinces within Canada

(ENCORP Atlantic Ltd, ENCORP Pacific Inc), although the details reflect the UK’s

structure of retailing. The operation of the system is outlined in the following points:

¾ As beverages are produced and sold to wholesalers, or directly to retailers,

producers send sales data to a central system along with a payment

matching the total value of the deposits on all items sold. The cost of the

deposits is then paid back to the producers, by wholesalers or retailers, upon

sale. The same happens as wholesalers sell items to retailers. Producers also

pay an administration fee to cover the remaining costs of the system. This is

set each year to reflect market prices of recyclate, amongst other factors;

¾ When the consumer purchases a beverage they pay the deposit to the

retailer, so the retailers are also reimbursed the total value of deposits;

¾ As consumers return empty containers to stores or other take-back centres,

the deposit is paid to them by the retailer. This puts the retailer out of pocket,

so the retailer then sends the returns data to the central system, which then

reimburses the retailer for those returned containers for which a deposit has

been paid out to the consumer. Thus the circle of deposit payments is closed.

As the return rate for containers is not 100%, the unclaimed deposits result

in revenue being retained by the system, which can be used to fund its

operation.

¾ In addition to the deposit, the central system pays a handling fee to the

retailer for each returned container, the intention being to compensate the

retailer for loss of space (storage requirements) and time (in processing the

deposit and taking back the containers). Handling fees are reviewed and

adjusted each year;

¾ Returned empty containers are collected in a number of ways. Automated

systems of collection use reverse vending machines or automated counting

machines. Manual collection is also possible. in this instance the retailer

accepts the container, over the counter, and stores it in bags or crates within

the store/outlet for transport;

¾ Where the containers are collected via an automated machine, the sorted

(and predominantly compacted) material can be transported directly to a

recycler, with material revenues being paid back into the central system.

This differs to the typical systems employed in countries such as Sweden and Canada, where

collections occur at a small number of redemption centres rather than at every retail outlet. We

believe that in order to maximize return rates and to remove the need for consumers to travel

individually make their way to redemption centres to return their containers, a denser network of

collection points would be more appropriate for the UK, and would eliminate additional

environmental impacts which might arise from making ‘dedicated journeys’ to redemption centres.

Thus we have modeled the system based on a higher number of collection points via both automated

and manual methods of collection, similar to systems used in Norway and Denmark.

Material revenues will also be paid on those containers that are collected

manually, though this material will first have to be transported to a dedicated

centre for counting, sorting and compacting, before it can be hauled on to a

recycling facility. These costs are met by the central system;

¾ The central system is the focal point for the flow of information regarding

container sales and finance for the whole DRS. A significant one-off cost will

be required to initially set up the DRS, including all the necessary

administrative support, which we have modeled as being met by ‘one-off’

producer and retailer joining fees. There will also be on-going costs

associated with administering the system which are covered as part of the

producer administration fee paid on each unit that is placed on the market.

The overall administration fee payable by the producers/ importers is

calculated as the balance of income from material revenues and unclaimed

deposits against the costs of collection, transport, processing, admin and

handling fees. In other words, the administration fee guarantees the DRS is

‘cost neutral’ overall.

It is worth noting that the system modeled here differs to that which exists in

Germany, where the organisation that manages the deposit refund scheme, the

DPG, only has an ‘over-seeing’ capacity. The system in Germany is much less

centralised, with retailers able to set up their own systems of collection and

processing, and payments moving directly between the producer and retailer, rather

than going through a central system.

Given the array of problems that have been

highlighted in association with the German system, that this system has been

recognised as an expensive scheme and that the scheme was originally partly set up

to try to encourage the refillables market, we chose to model a central system for

the UK and learnt from the issues that have occurred in the German system.

68,69

It is also worth noting the recent communication from the European Commission,

which states several safeguards that need to be respected in relation to how a DRS

should be designed in order to ensure a fair, open and transparent system,

including:

1) A countrywide system (which could be run either via a non-government

organisation (NGO), a government body or via the producers/distributors

concerned, and which may consist of more than one system operator so long as

the systems are compatible with each other). This will:

A) Ensure a sufficient number of return points for consumers to encourage

participation in the system.

Ernst & Young (2009) Assessment of Results on the Reuse and Recycling of Packaging in Europe,

report produced for the French Agency for Environment and Energy Management (ADEME), March

2009.

Perchards (2007) Study on Factual Implementation of a Nationwide Take-back System in

Germany After 1 May 2006, Final Report, 14 February 2007.

G. Bevington (2008) A Deposit and Refund Scheme in Ireland, Report commissioned by Repak

Ltd., September 2008.

B) Avoid ‘island solutions’ – a retailer-owned patchwork of different return

systems which are not compatible and which often force additional costs on

suppliers to adapt packaging to the requirements of the specific retailer.

2) A system which is open to all economic participators in the sector concerned –

including imported products under non-discriminatory conditions. This will avoid

creating an unjustified barrier to trade or distorting competition.

3) A system which ensures that there is no discrimination between those products

that are exempt and those that are subject to a deposit and that any

differentiation is based on objective criteria, ie. in principle, focus on material

and not on content of beverages as it is the former which drives the

environmental performance of the system.

Point 1 is addressed by the use of a central system approach to the UK deposit

refund model, as discussed above. In addition, we have modeled the UK system as

requiring a collection point at almost all retail outlets that sell beverage containers,

in order to ensure a sufficient number of return points for consumers and to remove

the need for consumers to travel individually to redemption centres to return

containers. In order to give the retailer a choice in how returned containers are

subsequently collected, and to make the return easier for larger stores to which

most containers would be likely to be returned, we have also modeled each retail

outlet as either using an automated system of collection (eg. reverse vending

machine or automated counting centre) or a manual collection, where the retailer

takes back the container over the counter and stores the containers in bags/crates

within, or at the front of the store/outlet for onward transport. This approach is

similar to the collections offered in countries such as Norway and Denmark.

In addressing Point 2 and Point 3, we have included all non-refillable beverage

containers placed on the UK market by both domestic and international producers in

our modelling, and have targeted beverage containers according to material (see

Section 5.2).

Finally, it should also be noted that the material revenues used in the DRS varied

from those used in the kerbside collection system. Given the higher quality material

that can be collected in the DRS and the greater amount of material that will be

handled by one central system, the values used for PET, steel and aluminium

collected through this system are higher than those collected at the kerbside.

the other hand, we have assumed that the value of glass will be the same,

irrespective of whether collected via a deposit scheme or at the kerbside. This is

highly favourable to the kerbside schemes, recognising that many commingled

systems will not produce glass of a quality suitable for re-melt. Details of the

material revenues used in the modelling can be found in Appendix

A.3.4.

One of the crucial elements in the deposit model is the setting of the deposit itself.

The overall deposit rates modelled are given in Table 5-1, based on achieving a 90%

return rate.

Both sets of values are based on a combination of Letsrecycle figures

(http://www.letsrecycle.com/

) and our own market knowledge.

Table 5-1: Proposed Deposits for Containers in UK Deposit Refund System

Container size

% of UK market

Deposit

≤ 500ml

65%

£0.15

> 500ml

35%

£0.30

Overall

£0.20

The choice to model two deposit rates for different container sizes was based on the

majority of existing deposit schemes worldwide. The choice of deposit was

calculated in two key stages. Firstly, the return rate was plotted as a function of the

deposit across existing schemes (see Figure 5-3), in order to establish what deposit

would be required to reach a 90% deposit rate. The deposits were converted from

the local currency of the DRS to GB Pounds (GBP) using OECD Purchasing Power

Parities (PPP) from 2008 to give a better estimate of the value of the deposit than

simply using the current exchange rate.

Figure 5-3 illustrates that, in order to

achieve a return rate of 90%, the deposit should be set at around 20p per container.

Secondly, UK market research shows that approximately 65% of potential deposit

refund containers sold have volumes less than or equal to 500ml, while 35% have

volumes greater than 500ml.

Using this split, we set the deposit at 15p for small

containers and 30p for large containers.

In respect of the relationship between deposits and return rates, Perchards

questioned Oakdene Hollins’ assumption regarding return rates:

OH [Oakdene Hollins] claim that “international experience shows that a

deposit of 5p per container should result in a 70% return rate.” They do not

explain why a 12p deposit on refillable containers in the UK 25 years ago

produced a return rate which was of the same order of magnitude.

There is a big difference in the return rates achieved when sales are typically

made by the crate – for beer or bottled waters, in countries where there is

strong product loyalty often associated with local manufacture. Sales by the

crate have never been the custom in the UK, where supermarkets did not

stock soft drinks or beer until the advent of non-refillable bottles and cans

and changes in the licensing laws.

If what is being claimed is that the only jurisdictions which achieve high return rates

are those where the majority of sales are in crates, then this is not true. Our

approach assumes that the principle motivation driving returns is an economic one.

OECD (2010) Purchasing Power Parities (PPP), Accessed May 2010,

http://www.oecd.org/department/0,3355,en_2649_34357_1_1_1_1_1,00.html

Personal communication with Canadean®, May 2010, http://www.canadean.com/

Figure 5-3: Return Rates as a Function of Deposits in PPP-Adjusted GB Pounds.

y = 0.0529Ln(x) + 0.9876

40%

50%

60%

70%

80%

90%

100%

110%

0.00 0.05 0.10 0.15 0.20 0.25 0.30

Deposit in PPP adjusted GBP

Return Rate

Denmark Norway Finland USA Israel Australia Log. (All)

Source: Eunomia

Elsewhere, Perchards note:

Swedish law mandates a 90% return rate for aluminium beverage cans but

the deposit set by the Swedish deposit system is only 50 öre (3p). The return

rate each year is stable and 85% was achieved in 2003. This suggests that,

although the level of the deposit can affect the return rate achieved, other

factors are also significant, for instance

• the habit of return not being lost

• ease of return and maybe consumer incentives (reverse vending

machines)

• buying by the crate rather than buying individual bottles, so there is

one single high-value transaction.

It is reasonable to assume that other factors will play a role in determining levels of

take back, notably the ease of return, but it is unreasonable to suggest that the

magnitude of the incentive is of limited relative significance. It is also not

reasonable to assume that a ‘habit of return’ cannot be re-established once lost.

Such ‘habits’ have to be established in order to be, subsequently, lost, and their

establishment and demise tends to track the existence or otherwise of DRSs. The

Swedish economist, Thomas Sterner, made interesting observations of this nature.

Regarding aluminium cans, he notes:

Recycling rates have passed 90 percent and it appears that the increase in

nominal deposits from 25 to 50 ore in 1987 has played a role. The fact that

the real value of deposits has fallen since then does no, however, seem to

have reduced collection. Maybe there have been changes in attitude and

information or ease of deposition. [….] Perhaps deposits are important just as

signals to change behaviour, which then has its own inertia.

Regarding wine and liquor bottles, in the same country (Sweden), he notes:

The nominal deposits rose from just 15 ore to 2.00 SSEK/bottle. In real

terms, the increase has been more than 100 percent since 1970. It appears

that the recycling rate goes up each time the nominal deposit rises.

This highlights the fact that the way people respond to deposits is likely to be

complex, but there seems little reason to believe there will be no price effect. There

appears, in any case, to be no basis for an assumption that a ‘habit’ cannot,

therefore, be established anew. The system we have proposed makes returns

convenient so should support such habit reformation. These habits do not exist

independently of DRSs, but rather, the DRSs tend to generate them (and as Sterner

suggests, these habits may be characterised by some ‘inertia’).

Other key elements involved in modelling the DRS include determination of the

handling fee payable to the retailers. This is based on acquiring an understanding of

the retail landscape across the UK, the likely proportion of retailers that would use

automated machines compared to manual take-back, the cost to the retailers of the

take-back process, and the costs of transport, containers and counting centres

required to deliver the collected containers for re-processing. In addition, the on-

going administrative costs associated with running the central system were

determined, as were the one-off costs associated with initial set up of the system.

It is important to note that the DRS in the UK will be different from those in other

countries because:

a) There are very few DRSs left operating in the UK (especially for alcoholic

beverages, beer bottles etc), so most containers are one-way and will be eligible

for inclusion in the system;

b) Modern behavioural attitudes appear to place a premium on waste collection

activities which make minimal demands on personal time - thus drop-off ought

to be quick and locations easily accessible;

c) There is a relatively high population density; and

d) The historic nature of retail outlets has led to a structure which is essentially

characterised by large numbers of small outlets operating in a decentralised

Thomas Sterner (1999) Waste Management and Recycling, in T. Sterner (ed.) (1999) The Market

and the Environment: the Effectiveness of Market-based Policy Instruments for Environmental

Reform, Cheltenham: Edward Elgar

manner (clearly many have now been replaced by larger supermarkets, but a

considerable number remain).

All of these points mean that the system must have the ability to collect 1) large

quantities of glass, especially from pubs and bars; 2) a high proportion of containers

from a large number of dispersed outlets, and on a frequent basis; and 3) ensure

that take back is possible through easily accessible locations, with minimal take

back time.

Following establishment of the retail landscape, the handling fee was calculated by

ensuring the following elements were included in the cost calculations:

¾ RVMs (reverse vending machines);

¾ Retail Space Infringements;

¾ Labour;

• Pickup / Unloading;

• Take Back;

¾ Transport;

¾ Counting; and

¾ Containers.

Figure 5-4 summarises the key components modelled in the DRS in the UK. Full

details of the step-by-step assumptions used in relation to the development of the

DRS are described in Appendix A.3.0.

5.4 Cost Reduction in Existing Waste Collection Systems

One of the key elements missing in the majority of existing studies on DRSs (see

Section 5.1) is the reduction in costs associated with fewer containers having to be

collected through all the routes set out in Figure 5-4. Even where this reduction has

been modelled in previous studies, it has been modelled rather poorly. This oversight

will result in an over-statement of the potential cost to society associated with the

introduction of a deposit refund scheme. Therefore, one of the key components of

this study was the inclusion of all relevant costs, most importantly the change in

costs from household kerbside collection systems.

Eunomia’s proprietary waste collection model, Hermes, has been used to investigate

the effect of implementing a DRS in the UK on kerbside collection schemes. Hermes

is a sophisticated spreadsheet-based tool that allows a wide range of local authority

specific and collection scheme specific variables to be modelled. The optimisation of

these variables allows us to build scenarios to reflect local circumstances accurately.

The main outputs of the model are recycling performance and cost.

Eunomia is confident that Hermes is as reliable a tool as any of its kind. It has been

used to model systems for authorities that collectively manage around 25% of the

UK’s total municipal waste. It is used to support contract procurement advice and

contract dispute resolution by building ‘shadow’ bids against which contractors’

tender submissions can be tested. Hermes has also been used in the context of

studies of relevance to national policy, and to undertake a cost benefit analysis for

the kerbside collection of food waste across the UK.

See Appendix A.2.0 for more

details.

Costs for all the other services that would be impacted by the introduction of a DRS

were also modelled. Appendix A.4.0 provides full details for each of the following

cost elements:

¾ Collection of containers through bring sites;

¾ Collection of containers through Household Waste Recycling Centres

(HWRCs) – both recycling and disposal;

¾ Commercial waste recycling / refuse collection; and

¾ Collection of containers from on-street litter bins and through street

sweeping.

Finally, it is worth considering the implications of a DRS for the existing policy

regarding packaging recycling. The UK’s Producer Responsibility (PR) scheme is

based upon a system of tradable compliance credits. The value of these credits

drops when the supply of the compliance credits far exceeds demand. For example,

targets for paper and card recycling are easily met at present, so the value of

Packaging Recovery Notes (PRNs) for paper and wood are around £2 per tonne. The

materials for which the level of performance is closer to what is required by the

targets are those which would be covered by a DRS (plastics, glass, steel and

aluminium).

Eunomia (2007) Dealing with Food Waste in the UK, report for WRAP, March 2007

Figure 5-4: Key Components Modelled in the Deposit Refund System in the UK

It would be expected, therefore, that the deposit scheme would reduce the price of

PRNs against counterfactual levels. It is difficult to know what this price reduction

might be. The value of PRN and PERN sales in past years has been as high as £100

million, though it has more commonly been of the order of £50 million (see Table

5-2). It seems not unreasonable to imagine that industry would see the cost of PRNs

fall by something in the order of £30 million as a result of the DRS, owing to the

increase in recycling. This seems particularly likely as the Government considers

increased packaging recycling targets.

If targets were not to increase so significantly, it might be that the DRS could lead to

a diminished need for the existing PRN mechanism. The key material would

probably be plastics, for which the potential for significant increases in packaging

recycling relates to a range of materials, of which the beverage container is only

one. Disbanding the existing policy might have significant benefits for companies in

terms of the efforts they make to demonstrate compliance. It would also eliminate

the costs of administering the policy at the UK level. These cost reductions may be

significant for companies across the UK, and for the public sector.

It should be noted that similar points were made by Oakdene Hollins in their work

for Defra, where there is discussion of the DRS overwhelming the existing PRN

system. Perchards, though, were critical of Oakdene Hollins’ assessment of the

effect of a DRS on the (then) existing UK packaging system, in particular on the price

of PRNs. We find the Perchards comments demonstrative of a lack of understanding

of the workings of the existing market mechanism. The following paragraph

highlights the extent of the confused views held by the reviewers:

PRNs are a market-based instrument. We think that much more thought is

needed to the impact on the PRN mechanism of a legally mandated DRS,

which is a command-and-control mechanism. One of the objectives of the

PRN system is to provide funding for collection, the idea being that PRN

revenues will allocate this efficiently. A deposit system would interfere with

this – its expense would make it unlikely to survive the competition with

cheaper collection systems in the PRN system. PRNs would need to be

revised to take account of it.

Quite what this means, perhaps only the authors know. However, most

commentators, including the OECD, for example, would argue that a DRS is no more

‘command and control’ than the PRN mechanism, driven, as that is, by the year to

year setting of recycling targets (which then influence the price of the PRNs). The

argument that a DRS would not survive because of its expense is to make light of

the fact that at the end of life, a consumer would have the option of either placing

the material in a kerbside container and receiving no deposit, or returning the

material to a vendor and recouping the deposit. It seems reasonable to assume that

if returns are convenient, consumers will choose the latter route.

Defra (2010) Implementing the Packaging Strategy: Recovery and Recycling Targets, Funding

Transparency and Technical Changes: A Consultation on Proposed Changes to the Producer

Responsibility Obligations (Packaging Waste) Regulations 2007 (as amended), March 2010.

In concluding, they state:

A more thorough discussion is needed of the possible interplay between a

DRS and the existing PRN system. We found it hard to understand how OH

concluded that PRN prices for the materials handled through a DRS would fall

to zero. The pack types that they recommend should be handled through a

DRS (plastic bottles, alu cans and possible some glass) represent only a

proportion of the packaging of those materials. The PRN system is a market

mechanism which, in our view, would need to be adapted if a DRS for

selected non-refillable containers were introduced.

This exhibits a complete lack of appreciation of the fact that the value of PRNs

reflects the imbalance, at the margin, of the supply of PRNs and the demand on the

part of obligated companies / compliance schemes for this evidence. If the

additional material collected by a DRS implies that packaging recycling targets in a

given year will be easily exceeded, then the value of PRNs would indeed fall close to

zero, just as it did in 2003 when the Government opted not to increase targets from

2002 levels. The point here is that the market for PRNs is tightest for the same

materials typically targeted by a DRS.

5.5 Environmental Impacts

Environmental impacts associated with the introduction of a DRS will occur from the

following processes:

1) Changes in the recycling of beverage containers;

2) Changes in the disposal of beverage containers; and

3) Changes in emissions associated with the collection and transportation of

containers to recyclers.

Regarding environmental benefits associated with diverting beverage container

waste from disposal, it should be noted that we have assumed 25% of the UK’s

waste will be managed though thermal facilities in the future, with the remainder

going to landfill (see Appendix

A.5.1.4). This is particularly important in the context

of PET, as avoidance of the disposal of plastics via thermal treatment would lead to

a significant reduction in greenhouse gas (GHG) emissions, whereas avoidance of

the disposal of plastics via landfill would have little overall impact on emissions.

The two main elements considered in modelling the environmental impacts

associated with the collection, recycling and disposal of beverage containers were a)

GHG emissions and b) air quality impacts. The approach to valuing these two

components is set out in Appendix A.5.0.

Table 5-2: Expenditure on PRNs and PERNs, 1999 - 2005

1999 - 2001 2003 2004 2005

PRN PERN PRN PERN PRN PERN PRN PERN

Paper £69,400,000 £3,900,000 £22,245,337

£1,744,263

£12,915,351 £3,938,957 £17,373,786 £12,447,897

Compost £4,889

Glass £16,600,000 £700,000 £8,223,179 £959,118 £10,600,571 £2,561,816 £15,642,454 £5,997,046

Steel £10,500,000 £4,400,000 £1,229,585 £1,245,237

£2,460,142 £2,181,782 £9,006,337 £18,631,570

Aluminium £800,000 £100,000 £432,628 £61,052 £779,113 £759,497 £1,859,214 £1,224,424

Plastics £15,100,000 £2,700,000 £1,634,515 £615,845 £1,792,668 £1,755,251 £6,960,361 £8,866,465

Wood £8,600,000 £5,322,252 £0 £5,254,424 £8,391,940 £0

Recovery

Clinical £7,855 £5,529

EfW £2,168 £7,897

EFW (MSW) £5,352,903 £480,622

RDF £102,663 £72,382

Recovery Total £5,465,589 £566,430 £677,973

Totals £121,000,000

£12,000,000

£44,553,085

£4,625,515

£34,368,699 £11,197,303 £59,916,954 £47,167,402

PRN + PPERN £133,000,000

£49,178,600

£45,566,002 £107,084,356

Source: Defra

Furthermore, there is a negative environmental impact, or disamenity, associated

with uncollected litter. A study by Cambridge Economic Associates indicates that the

average household would be willing to pay £25 per annum to live in a

neighbourhood where the streets are kept clean.

Unfortunately, however, this

value does not cover the potential willingness to pay to remove litter from rural

areas, and, as far as we are aware, there are no studies attempting to place a value

on the disamenity experienced in such circumstances in the UK.

The only significant study of this nature of which we are aware was carried out in

Australia by Pricewaterhouse Coopers. This indicated that households are willing to

pay, on average, AUS $4.15 per 1% reduction in litter. The quantification of

‘reduction’ is not clear, but if, in line with the work of Stein and Syrek, we take the

view that size (volume) is a proxy for visual impact, and that visual impact is what

residents most notice, then we might assume that households interpreted this in

terms of volume reduction.

Assuming this to be the case, then if one also assumes:

• Beverage cans occupying 25% by volume of litter (which may be

conservative);

and

• 80% reduction in beverage-related litter as a result of a DRS

then the effective reduction in litter volume would be equivalent to 20% of the total.

Using the Pricewaterhouse Coopers figures, converted to UK exchange rates (we

have used the rate UK£ 1 = AUS $1.73), the value of this would be £48 per

household.

This gives a net figure, across 26 million households in the UK, of £1,248 million per

annum.

5.6 Cost Benefit Analysis

The principle objective of the cost benefit analysis was to identify the net cost or

benefit to society from the introduction of a DRS in the UK. Conventionally, this net

cost or benefit is represented as follows:

Cambridge Economic Associates et al (2010) Developmental Work to Value the Impact of

Regeneration, Technical Report: Environmental Quality and Amenity, May 2010

Steven Stein and Daniel Syrek (2005) New Jersey Litter Survey: 2004, A Baseline Survey of Litter at

94 Street and Highway Locations, Report for the New Jersey Clean Communities Council, January 28,

2005. http://www.njclean.org/2004-New-Jersey-Litter-Report.pdf

The analysis in Section 4.2 above suggests that the effect of DRSs on the volume reduction in litter

in the US may be well above this figure, and in systems which are achieving lower return rates than

the one modelled here. In addition, the analysis of the composition of litter in the UK is at least

suggestive of a relatively high proportion of the volume being occupied by beverage containers.

These figures are typical of the levels of reduction reported under DRSs for beverage containers

(see, for example, Perchards (2005) Deposit Return Systems for Packaging Applying International

Experience to the UK, Peer Review of a Study by Oakdene Hollins Ltd., Report to Defra 14 March

2005).

In this study, the environmental costs comprise the impacts of GHG emissions and

other air emissions, expressed in monetary terms. The main sources of

environmental impacts are given in the previous section.

In conventional cost-benefit analysis, the approach taken would be to strip out taxes

and subsidies from Government and to cost all environmental impacts. This

approach ensures that double counting is avoided. One of the reasons (though not

the only one) for doing this is that if the analysis includes both the financial costs,

inclusive of taxes and subsidies, and the environmental impacts of dealing with

waste, there is a risk of double counting where the taxes and subsidies are

themselves designed to reflect environmental impacts.

In practice, however, this tends to lead to figures which are unfamiliar to those

engaged in the industry, and which do not reflect the incentives which they face

within the market place. The most obvious example, perhaps, is with landfilling.

Here, the tax stands at £48 per tonne, rising to £80 per tonne by 2014/15. The

impacts of landfilling cans, however, are not expected to be anywhere close to £80

per tonne (the principal externalities of landfill relate to methane generation, and

cans do not biodegrade in landfills). However, actors in the market still need to pay

the tax.

In our approach, therefore, so as to ensure that the financial figures represent the

actual costs to operators, but recognising also the potential for double counting, the

approach we have taken is:

1. To model costs as they would be ‘in the market’

2. To include in the environmental analysis only those impacts which are either

a. not covered by existing policies or

b. addressed by policies, but where the impact exceeds the level of

internalisation implied by the policies which are in place (we have

included only the element of the environmental impact that is not

internalised by the policy).

This approach has the merit of reflecting ‘actual costs’ in the market, but also

ensures there is no double counting of the environmental impacts associated with

different ways of managing waste (which would occur if the impacts under 2b were

considered even where impacts are already internalised in market prices by policy).

In this study, we have attempted to build as comprehensive a model as possible in

terms of covering all the costs and benefits associated with the collection and

management of wastes. These include capital costs, operating costs, labour costs,

opportunity cost of lost retail or storage space, disposal costs, material revenues

and deposits unclaimed or ‘lost’ by the consumer. Full details of each stage of the

modelling, including a detailed methodology, data sources, and individual cost

elements are presented in Appendix A.2.0 to Appendix A.5.0. Figure 5-5 provides a

Cost or Benefit = Financial Costs + Environmental Costs

summary of the key elements that have been considered in calculating the overall

cost or benefit of introducing a DRS in the UK.

Figure 5-5: Cost Benefit Analysis Model - Overview

Source: Eunomia

6.0 Results from Cost Benefit Analysis

Following the methodology described in the previous section, the overall costs and

benefits associated with the introduction of a DRS in the UK have been calculated,

the results of which are presented in this section. As noted before, this will form the

basis of answering the research question:

‘How do the benefits of introducing a UK-wide DRS for certain beverage

container packaging compare with the costs of implementation and

operation?’

The costs in this analysis have all been measured relative to a baseline. This

baseline encompasses a situation from 2015 onwards, where full kerbside

collection services have been rolled out to all households in the UK. The overall costs

therefore represent a change to rather than an absolute cost of service provision.

However, the costs associated with the DRS will represent the annualised running

costs of the system, as there is no such system operating in the baseline.

To explain how the total costs for each scenario are derived, the DRS costs will be

described first, followed by the additional financial impacts, the monetised

environmental costs, and finally, the sum of these elements. The costs and benefits

associated with the complementary DRS will be described first, followed by the key

differences between this system and the parallel DRS. Some of the key factors

affecting the results are then considered in a series of sensitivity analyses.

6.1 Complementary System

The complementary system is one where the DRS complements rather than

competes against existing household collection services. This is enacted by changing

the service specification of the existing kerbside systems in such a way that the

householder is obliged not to place the relevant beverage containers in their

recycling box, bag or bin. Note that because the system might still collect glass jars,

metal food cans or other plastic packaging, the “mis-use” of this system would not

compromise the recycling system. Indeed, operators would have an incentive to

ensure that deposit container “contaminants” are segregated from the remainder of

the material so as to obtain the deposit revenues from them, though the nature of

the collection scheme might determine how straightforward or difficult this might

be.

Under this scenario, therefore, all beverage container waste being collected from

household recycling services is eliminated from the model. It is, however, assumed

that 5% of containers still end up in household refuse, as some people will not

change their behaviour patterns as a consequence of paying the deposit and will

deem the additional effort required to return any containers too significant. A small

decrease has also been modelled in the change in management of containers

though other routes, so that the overall return rate of containers in the DRS is

calculated as 90% (see Appendix A.4.0). This return rate is based upon the value of

the deposit placed on the container and similar return rates from other countries

(see Appendix A.3.1).

The share of annual costs and revenues involved in the operation of the DRS are

shown in Figure 6-1. The administration fee payable by the producer for each unit

placed on the market is calculated in order to bridge the outstanding imbalance

between the costs and revenues in the central system. Hence the overall system is

one in which all ‘net costs’ are recovered ie. the system costs equal the system

revenues.

Figure 6-1 shows the overall proportions of revenues and costs in the system. It does

not, however, provide any indication of who would be paying for which element. As

such, Figure 6-2 illustrates the flows of money throughout the DRS for each of the

key stakeholders. It is important to note that throughout this cost benefit analysis,

and in the sensitivity analysis that follows, a positive figure represents a ‘benefit’

and a negative figure represents a ‘cost’ to the stakeholder in question or for the

overall system.

The key points to draw from

Figure 6-2 are:

¾ Handling costs for the retailers are estimated to be around £576 million per

annum. We have modelled that these costs would be compensated by the

central system through a per unit administration fee of 4p for retailers with

Reverse Vending Machines (RVMs), and 1p for those without. Collection and

counting costs, financed by the central system, are likely to be around £337

million per year. Hence the retailers are compensated for all their costs, so

the net cost to them is zero;

¾ The consumers who do not, or cannot, return the containers they purchase

will lose the deposits they have paid – this is signified by the grey box in

Figure 6-2. At an overall 90% return rate, consumers would forfeit a total of

£491 million of unclaimed deposits. In our model, this revenue helps fund

the operation of the system. For this reason, it is important to have

complementary targets in place to ensure that the system is not designed so

as to ‘deliberately underperform’, and enable full funding through unclaimed

deposits. With targets in place, the system would likely be designed with a

deposit set at a rate designed to deliver the desired performance level,

consistent with the level of infrastructure provision. The provision of many

easily-reached return points should minimise the level of unclaimed deposits

as long as the deposit it set at a reasonable level.

Figure 6-1: Revenues and Costs in Operation of Deposit Refund System –

Complementary Scenario

¾ The outstanding imbalance between the costs and revenues (including

unclaimed deposits) in the central system of around £212 million are

recovered through the producer administration fees. The handling fee

payments to the retailers and the administration costs of the central system

are offset by the revenue generated from producer administration fees,

income from material sales and unclaimed deposits;

¾ The producer administration fee equates to around 0.7p per container placed

on the market; given that the producer will be likely to pass at least some of

this cost onto the consumer, this fee could be described as the ‘cost’ to

society of implementing the DRS. However, even if the producers decided to

pass 100% of this cost through to the consumer there would probably be very

little change in terms of volume of sales given that the additional cost per

unit is relatively low;

¾ The system implies a net cost to the producers i.e. producers are effectively

paying for the collection of beverage containers that they place on the

market.

The costs of the system are thus born by those who are responsible for the

generation of the waste – the producer and the consumer.

Figure 6-2: Cash Flows in the Deposit Refund System – Complementary Scenario,

£millions 2010 Real Terms

Pay deposit for

each sold container

= £5,257M

Purchase beverages:

pay deposit = £5,257M

Purchase beverages:

pay deposit = £5,257M

Pay deposit for

returned cans

= £4,766M

(Retailer cost – space and

labour in handling deposit-

bearing containers = £576M)

Income from material sales

= £210M

(Includes cost of administering

system, collection and transport

logistics = £337M)

+ Revenues from lost

deposits (= £491M)

Pay admin fee

per container

= £212M

Pay deposit to retailer for each

returned container = £4,766M

Pay handling fee

to retailer = £576M

Balance

= £zero

Balance

= £zero

Balance

= -£212M

Balance

= -£491M

Note positive figures imply savings

The shift in the management of beverage containers from the baseline situation to

the complementary scenario results in a reduction in the requirement to collect

waste from a number of routes. A material reduction in the amount of waste

requiring collection will thus lead to a change in the overall costs of service provision

for each of these routes. Table 6-1 shows the financial savings that accrue from the

introduction of the DRS under the complementary scenario.

It is clear from Table 6-1 that local authorities would save a significant amount of

money on their annual waste collection costs as a result of the introduction of a

DRS. In effect, the burden of waste management shifts from the ‘taxpayer’ to the

individual person (consumer), and the industry (producer/ importer) that is

responsible for generating the waste beverage containers. Thus, the ‘polluter pays’

principle is reinforced by the introduction of the deposit refund policy. Commercial

enterprises that are currently responsible for managing their employees’ wastes

should also see some financial benefit. The reduction in costs of administering the

PRN system also provides significant savings to businesses.

Table 6-1: Financial Savings from Existing Waste Collection Activities, £millions

Service

Local

Authority/

Taxpayer

Commercial

Enterprises

Change in household recycling collection costs

£129M

Change in bring site costs

£3M

Change in HWRC costs

£1M

Change in litter collection costs

£27M

Saving from reduced expenditure on PRN

system

£30M

Additional cost to retail outlets outside of

-£19M

Change in commercial waste collection costs

£36M

Balance

£159M

£47M

Total Saving

£206M

Source: Eunomia

In addition to the financial costs associated with the running of the system, there

will also be environmental impacts that should be monetised and included in the

overall analysis of costs and benefits. The environmental impacts associated with

the introduction of the DRS and the resultant reduction in the collection and disposal

of containers via other waste management routes is summarised in

Table 6-2 and

Figure 6-3. The assumptions that underpin these calculations are presented in

Appendix A.5.0.

Table 6-2: Monetised Environmental Impacts, £millions

Environmental Impact Monetised Value

Recycling (GHG + AQ) £88M

Disposal (GHG + AQ) £6M

Transport -£25M

Disamenity of litter £1,248M

Total £1,317M

Source: Eunomia

Figure 6-3: Annual Monetised Environmental Impacts (Complementary System)

£millions 2010 Real Terms

Tangible benefits accrue from the reduction in GHG emissions and air pollutants as

a result of increased recycling and reduced disposal (net = £94M). Costs also occur

from the net additional emissions from vehicles used to collect and process empty

containers. It should be noted, however, that not all transportation impacts are

negative, as there is avoided transportation from a reduction in waste collected at

the kerbside and from commercial premises. Vehicles are subsequently able to

travel further distances before they reach capacity and need to offload or ‘tip’ the

collected waste, which results in a reduction in the number of longer journeys made

in order to ‘tip’ the material and hence in the number of overall vehicles required.

The overall net benefit of these tangible environmental impacts is £69M.

The figure for disamenity is as calculated in Section 5.5, based upon work by

Pricewaterhouse Coopers. We illustrate overall results with and without this figure,

given the absence of UK-specific studies, and the influence on results.

The greenhouse gas benefits of a particular policy, or system, are keenly pored over

by policy makers. We therefore present the quantity of GHG emissions saved, above

the baseline situation, in Figure 6-4.

Figure 6-4: Greenhouse Gas Savings from the Introduction of a Deposit Refund

System, thousand tonnes CO

equivalent

Note that the emissions are presented from a ‘global’ perspective, as the location of

marginal recycling activities is unclear, both in the UK and abroad. In essence, the

value of these benefits, as reported under the UK’s domestic emissions inventory,

would change depending upon the location of the primary and secondary materials

production.

To put this saving into a policy context, the UK Marginal Abatement Cost Curve

(MACC) analysis, undertaken in 2008, estimated that by 2022 the waste sector

could reduce its emissions of carbon dioxide equivalent by around 9 million tonnes,

with the majority of savings (7.5 to 8 million tonnes) coming from biowaste

treatment and a shift from landfill to other residual processes. Therefore, the

savings generated from any level of additional dry recycling, up to the maximum of

607 thousand tonnes, would significantly help achieve the abatement required

under statutory greenhouse gas targets, set out by the Committee on Climate

Change.

Putting it All Together - The Whole Picture

The overall costs and benefits to society from the introduction of a DRS in place of

kerbside collections of containers are summarised in

Table 6-3 and Figure 6-5.

It should be noted that Table 6-3 and Figure 6-5 illustrate the net benefit once the

DRS is up and running. It does not include the £84 million one-off costs that would

be associated with the initial setting up of the DRS, as these would only be incurred

over the first year or two of the system. From society’s perspective, depending on the

pay-back period, these costs will be covered within the first few years of

implementation, and would be met by fees payable by producers and retailers as

they join up to the scheme. One-off costs are discussed in more detail in Section 6.4.

This is particularly significant since our baseline assumes that kerbside collection systems of a

relatively high quality exist for all households at the time the DRS is introduced.

Based on the cost benefit analysis presented in Table 6-3, the key messages from

the overall cost benefit associated with the introduction of a complementary DRS in

the UK are:

¾ DRS costs, from society’s perspective, are around £703 million per annum.

Not all these costs, however, are subsequently met by producers of

beverages. This is because the system’s finances are effectively bolstered by

unclaimed deposits (to the tune of £491 million). Producers, therefore, would

pay only £212 million. The unclaimed deposit is important to represent as a

cost, because if the return rate does approach 100%, then the producers

would have to pay an amount which approximates to the total cost of the

system in the absence of unclaimed deposits (i.e. £0.7 billion);

¾ The cost to producers (net of unclaimed deposits) of operating the DRS is

roughly equivalent to the savings produced from a) a reduction in the

collection of beverage containers by local authorities and commercial

enterprises, and b) a reduction in costs for operating the PRN system.

Therefore the financial costs, net of savings, are close to zero. Other studies

have generally focused upon the lost revenue to local authorities where

materials are no longer available for collection. Quite apart from the fact that

it is known that not all local authorities are securing the full benefits

associated with this revenue stream, the studies have failed to appreciate

that local authorities will save far more in terms of operating logistics than

they lose in terms of material revenues. This is true irrespective of whether

the DRS operates in a complementary or parallel fashion for the simple

reason that a significant proportion of low density packaging materials no

longer have to be collected though kerbside recycling / refuse systems;

¾ Local authorities around the UK are expected to save around £159 million

per year in avoided waste management costs. This is a saving of around £7

per household per annum. In other words, for an average waste collection

authority of 50,000 households, the financial saving in real terms would be

around £360,000 per annum. This is a valued means of saving public sector

costs at a time when cuts are being made to reduce the deficit, and

consideration is being given to the transfer of services to the private sector;

¾ The environmental benefit is significantly different with and without the

disamenity associated with littering. We have included both figures in order

to show the ‘worst-case’ scenario cost benefit analysis, where individuals

place zero value on the removal of litter from their environment, separately

from perhaps a more likely scenario, where individuals are willing to pay

somewhere in the region of £48 per household per annum to remove litter

from the environment. There will, without question, be some disamenity

value that arises through littering, and the analysis presented here shows

that the reduction of containers in the environment could generate a

significant benefit to society;

¾ In our model, a minority of consumers forego £491 million in deposits. There

are interesting questions as to what might lead people to forego deposits in

systems where the aim is to make returns as convenient as possible. There

will be cases where it may be difficult for some individuals to make returns,

though for such individuals, there may be others who can gain the returns for

them (or simply ensure no additional deposit is paid on a ‘replacement’

purchase). For others, the calculus may be more simple in that the marginal

income to be gained is too insignificant to warrant any (additional) effort to

regain the deposit, though here, the prospect remains that others will choose

to benefit from the deposit. Nonetheless, this does increase the outlay from

consumers, and is considered as a cost of operating the system, though the

effect is to defray the costs born by the producers;

¾ More research into the disamenity associated with littering would be

desirable in order to firm up the figures in Table 6-3 and give greater strength

to determining the argument for or against the introduction of a DRS in the

UK. Under the current analysis, however, the overall considerations appear

favourable for introducing a complementary system where one incorporates

the disamenity associated with litter; and

¾ For society as a whole, there is a net cost of £428 million where no allowance

is made for the benefits generated from reduced littering. Once this is

factored in, however, the position changes quite dramatically. Indeed, society

derives a net benefit of £820 million. Indeed, the system implies a

benefit:cost ratio of the order 2:1.

Another way of looking at these results is that the benefits to society will be positive

as long as the disamenity associated with the reduction in litter is valued in excess

of £428 million, or around £16.46 per household.

The costs of running the DRS can effectively be summarised as the cost to

producers plus the revenue foregone by consumers in the form of unclaimed

deposits. This is just over £700 million.

Table 6-3: Overall Costs and Benefits, £millions – Complementary Scenario

Cost or Benefit (-ve

is a cost), in

£millions

Financial Effects

Deposit Refund System (to Producers)

-£212M

Collection and Treatment/Disposal (to Local Authorities)

£159M

Change in Cost of PRNs (conservative estimate)

£30M

Collection and Treatment/Disposal (to Commerce)

£17M

Consumers (unclaimed deposits)

-£491M

Net Financial Costs

-£497M

Environmental Effects

without disamenity

£69M

with disamenity

£1,317M

Total Benefit to Society

without disamenity

-£428M

with disamenity

+£820M

Source: Eunomia

Note consumer costs are the unclaimed deposits that the consumer ‘loses’ as a result of not

returning their containers in order to collect their deposit

Figure 6-5: Summary of Annual Ongoing Costs and Benefits from the Introduction of a ‘Complementary’ Deposit Refund System,

£millions 2010 Real Terms

Note: Positive figures indicate benefits

, negative figures indicate costs

6.2 Parallel System

Under the parallel system, the kerbside system remains in operation for beverage

containers alongside the DRS. In our modelling, this has the assumed effect that

some people will continue placing containers in their household recycling or refuse

collection, even though they have paid the deposit, on the premise that the

convenience factor outweighs the financial loss of the deposit. With this in mind the

overall return rate for the system was set at 80%, 10% less than where no return

can be made at the kerbside. Figure 6-6 provides an overview of the costs

associated with the parallel DRS, based on an 80% return rate, and Table 6-4

summarises the overall costs and benefits derived under the parallel scenario.

Figure 6-6: Overview of Parallel Deposit Refund System Costs

80% overall return rate, £millions 2010 Real Terms

The figures presented in Table 6-4 and Figure 6-6 indicate that the net benefits to

society associated with introducing a DRS alongside the existing kerbside collection

are similar to those under the complementary scenario. The main differences are

discussed below:

¾ Given that the return rate of containers placed on the market has been

assumed to be 10% less than under the complementary scenario, ie. 80%,

the value of the deposit (15p or 30p depending on the size of container)

results in a significant revenue being generated from lost or ‘unclaimed’

deposits in the parallel system. Consequently, the unclaimed deposits and

material revenues negate the need for an administration fee to be paid by

producers (see Figure 6-7) and generate a surplus revenue of £249 million

for the central system. As is the case in some Nordic countries, this revenue

could be used to fund environmental projects, as well as to invest in further

optimisation of the DRS infrastructure where applicable (for example, to

increase infrastructure to achieve higher return rates);

Table 6-4: Overall Costs and Benefits, £millions – Parallel Scenario

Cost or Benefit (-ve is

a cost), in £millions

Financial Effects

Deposit Refund System (to Producers)

£249M

Collection and Treatment/Disposal (to Local Authorities)

£143M

Change in Cost of PRNs (conservative estimate)

£30M

Collection and Treatment/Disposal (to Commerce)

£15M

Consumers (unclaimed deposits)

-£944M

Net Financial Costs

-£508M

Environmental Effects

without disamenity

£65M

with disamenity

£1,313M

Total Benefit to Society

without disamenity

-£443M

with disamenity

+£805M

Source: Eunomia

*Note consumer costs are the unclaimed deposits that the consumer ‘loses’ as a result of not

returning their containers in order to collect their deposit.

http://www.dansk-retursystem.dk/content/us/importers_and_producers/deposits_and_fees

¾ The corollary of this is that consumers actually incur a greater cost, in the

form of unclaimed deposits, than under the complementary system.

Effectively, what is being modelled is an assumption regarding the response

of consumers to the relative ease with which containers can be dealt with

through the kerbside system and the DRS. This assumption is clearly open to

question. It may well be that a sufficiently well resourced infrastructure would

imply that the differentials in performance may be smaller, especially if the

deposit is seen as ‘worth having’, or if the scheme is considered ‘worth

supporting’; and

¾ Given the lower amount of material being collected in the DRS, the overall

logistics costs were calculated as £15 million lower in the parallel system,

and the material revenues generated were also lower by the order of £22

million. In addition, given that there will be an increase in the number of

containers collected at the kerbside in the parallel system, there is a

resultant decrease in the overall savings available to local authorities of £15

million in comparison to the complementary system.

It should be noted that this modelling does not take into consideration the fact that,

given the size of the deposit, the local authority itself would be likely to implement

measures to separate out the beverage containers in the kerbside scheme, and

claim back the deposits as an income stream. On a per container basis, a collection

authority can gain significantly higher revenue from the redemption of the deposit

than from the sale of the material only.

Therefore, it is quite possible that the

authority would go to some effort to ensure that people place containers in the

recycling system rather than in refuse, and to either separate them out in kerbside

sort collections, or install an automated counting machine at a transfer station in

order to redeem the deposits from the central system. Consequently, the figures for

the system could converge with those calculated under the complementary

scenario, with principal differences being related to the cost of installing relevant

separation equipment.

We would thus suggest that both systems could be considered feasible systems for

the UK. However, given that the operational costs are similar between the parallel

and complementary scenarios, and they are likely to converge in any case, it seems

sensible to allow householders to continue to use existing kerbside collection

systems if they so desire. This might also be a more sensible approach given the

existence of containers which bear no deposit (imported by consumers from

abroad).

The overall costs of the DRS are, as with the parallel scheme, given by the cost to

producers plus the revenue foregone by consumers in the form of unclaimed

deposits. This is just under £700 million in this case. The key change between the

parallel and complementary systems, driven by the assumptions regarding return

rates, is the distribution of these costs between producers and consumers. Other

things being equal (notably, the magnitude of the deposit), a scheme with a higher

The average revenue per container is 1.7p. This is significantly less than the 15p or 30p deposit.

return rate will lead to reduced revenues in the form of unclaimed deposits and

higher costs to producers.

Since DRS costs to producers are lower with lower return rates, it would appear

sensible to introduce target recycling rates for these materials to encourage higher

return rates from the system, and reduce the incentive for poor system design /

inadequate infrastructure. This would also result in the convergence of the

complementary and parallel systems, in terms of return rate. The effect of this is to

lower the revenue generated from unclaimed deposits, thus leading to slightly

higher administrative fees, but with the ultimate outcome that greater

environmental benefits are delivered.

Figure 6-7: Revenue Source under Parallel System

Source: Eunomia

6.3 Sensitivities

In order to understand the robustness of the results and conclusions presented

above, we have undertaken a series of sensitivity analyses. The results from these

analyses are detailed in this Section. Given the discussion above in relation to the

parallel system, all of the sensitivities presented relate to the results under the

complementary scenario.

The sensitivity analyses are first undertaken as discrete elements, in order to explore

the relationship between each key variable and the overall cost benefit associated

with introducing the deposit scheme. Thus each sensitivity is run under the

assumption of ‘all other things being equal’. Testing of the overall results, using

multivariable analysis, is described at the end of the section, in order to identify

those variables that have the most significant influence on the results that have

been obtained in this study.

6.3.1 Change in Automated vs Manual Take-back

As part of the sensitivity analysis, we explored what might happen if more retail

outlets were able to install RVMs that allowed them to clear and compact containers

on site, rather than requiring more regular manual collections. It is worth noting that,

although as modelled here, the cost of a consumer friendly machine destined for the

supermarket forecourt is around £17,000, a simple RVM designed more for smaller

shops and back-rooms/ storage areas might cost as little as £4,500 to £8,500. If a

greater number of retail outlets were to install these simple RVMs to streamline the

take-back process, then the collection logistics would be significantly optimised due

to the higher bulk densities of the material and the elimination of counting centres.

If the setup of the system is changed so that at least 80% of retailers can install on-

site automated takeback systems, then 160,000 machines would be required in

addition to the existing 40,000 modelled in the central case. In both cases the net

benefit changes to a cost, when disamenity is not considered, but it is still a benefit

when litter disamenity is included.

This shows that despite the savings from optimised collections, and less down-

stream processing, the number and capital cost of RVMs play a significant role in

the financial costs of the system.

Table 6-5: Overall Costs and Benefits, £millions – Take-back Methodology Sensitivity

Central

Case

€5,000

Machine

€10,000

Machine

RVM Costs £209

£226 £419

Collection Costs £323

£252 £252

Total Benefit to Society

without disamenity

-£428

-£511 -£721

with disamenity

+£1.2B

+£737 +£527

Source: Eunomia

6.3.2 Change in Collection Logistics

Under the central complementary scenario, it has been assumed that most of the

larger stores will be able to utilise backhauling (see Appendix A.3.2.6).

If this is not

the case, or conversely if distribution companies are able to backhaul from much

smaller premises, the overall collection requirements of the system will change

significantly. In the central scenario, we have modelled that around 37% of retail

outlets will be able to utilise backhauling. This is based on estimates of the

Note that backhauling refers to the return trip that is made by a truck after delivering a load to a

specified destination. This return trip, on which the truck would otherwise be empty, is used, where

possible, to transport items back to where the truck journey commenced from.

proportion of each retail category able to backhaul. The key assumptions in the

setting of these conditions were:

¾ All supermarkets are of a large enough size to backhaul;

¾ 70% of medium sized stores would be large enough to backhaul;

¾ 20% of convenience stores will be serviced by large-scale distribution

companies which will backhaul;

¾ Half of pubs (whose main trade is beverages) will be supplied by a

distribution company large enough to backhaul. In practice many pubs are

supplied by a small number of large suppliers or breweries, so in reality the

potential for backhauling using existing collection logistics could be more

substantial than estimated; and

¾ The potential for backhauling is diminished when considering other catering

sectors due to the lower volumes of containers per outlet.

37% of retail outlets equates to around 50% of the total number of containers in the

system, and containers are the single most significant cost in the dedicated

collection round logistics.

Table 6-6 shows the change in cost benefit if these assumptions are altered.

Table 6-6: Overall Costs and Benefits, £millions – Collection Logistics Sensitivity

Central

Case - 37%

Outlets

Backhauling

10% Outlets

Backhauling

80% Outlets

Backhauling

Collection Costs £323

£423 £227

Total Benefit to Society

without disamenity

-£428

-£534 -£326

with disamenity

+£1.2B

+£714 +£922

Source: Eunomia

It can be seen from Table 6-6 that the results are sensitive to the input assumptions

regarding backhauling. Shifting between low and high values for backhauling

changes costs, relative to the baseline, by +/-£100 million or so. Given the large

number of supermarkets in the UK retail landscape, and their likely desire to want to

backhaul where reverse vending machines are installed in their outlets, it seems

unlikely that less than 10% of retail outlets would be engaged in backhauling.

Personal communication with Tesco during the TOMRA RVM trial, 2009.

6.3.3 Change in Environmental Assumptions

The way in which emissions to the environment are valued and the price that is used

will both have a significant bearing on the results obtained. Details of the way in

which carbon and air quality valuations have been undertaken are discussed in

Appendix A.5.0. Two sets of damage costs have been considered in this study - the

UK Government’s Interdepartmental Group on Costs & Benefits (IGCB) damage costs

and the Clean Air for Europe (Café) damage costs.

These datasets use slightly

different values in their assessment of damage costs. The Café data set was used

for the cost benefit analysis presented here in the central case. Table 6-7 illustrates

that when using the UK Government’s Interdepartmental Group on Costs & Benefits

(IGCB) damage costs as opposed to the Clean Air for Europe (Café) damage costs,

there is a reduction in environmental benefit of around £21 million in air quality

benefits.

Table 6-7: Overall Costs and Benefits, £million – Change in Environmental

Assumptions

Central

Air Quality –

Café to IGCB

Damage

Costs

Increase in

additional

journeys by

consumer –

10% to 50%

Average

Emissions

Standard

Euro 5 to

Euro 6

Environmental Costs £70

£49

-£8 £87

Total Benefit to Society

without disamenity -£428

-£449

-£506 -£411

with disamenity +£1.2B

+£799

+£742 +£837

Source: Eunomia

Furthermore, in assuming an increase from 10% to 50% in the number of additional

(dedicated) trips that the consumer might make solely to drop off their deposit-

bearing containers, there would be a reduction in environmental benefit of

£78million. Finally, if we assume that the vehicles required in the deposit refund

collection logistics are of Euro 6 emissions standard (a more stringent standard

which the UK’s vehicle fleet will have to meet in the future) rather than Euro 5, there

is an overall reduction in vehicle emissions, generating an additional £17 millon of

environmental benefit in comparison to the central scenario.

In addition to these impacts, and as discussed previously, the disamenity associated

with littering is also uncertain. Hence all results have been presented both with and

Damage costs are a way of converting an environmental impact into a financial impact, by looking

at the financial costs to society of some form of environmental impact. In this case, for air quality

impacts, this is done through assessing the cost to society of people’s ill health that results from air

pollution. In the case of climate change, this is done by calculating the financial cost to society of

putting in place measures to reduce climate change impacts associated with greenhouse gases.

without the potential benefit to society associated with the removal of litter from the

environment.

Table 6-7 shows the change in environmental costs associated with each of the key

changes in the environmental assumptions. When considering total benefit less

litter disamenity, the increase in dedicated journeys by the consumer does take the

overall benefit to society over the tipping point – all other things being equal.

However, when disamenity is considered, none of the sensitivities challenge the

robustness of the overall conclusions.

6.3.4 Potential Switch in Manufacture Away from Deposit-Bearing Containers

Though not forming part of the scope of this study, it is perhaps worth briefly

discussing the potential switch in choice of container material in order to avoid the

DRS. In the recent study undertaken by ERM on DRSs, it was noted that the over-

riding issue in the manufacturer’s choice of which material to use to contain

beverages was centred on consumer acceptance, rather than on avoidance of being

in a deposit scheme.

However, the study did note one example where the

manufacturer had changed their material choice specifically because of the DRS,

with one producer in Sweden switching to a bottle made from plastic that was not

part of the scheme. Subsequent adjustments to the legislation brought that

particular product back into the DRS. However, this might not be as easy if, for

example, the producer were to turn to cartons or pouches as alternative

containment methods.

Policy makers should, therefore, be aware that some equivalent form of extended

producer responsibility for those beverage containers falling outside the DRS might

be necessary in order to avoid the shift to materials that are not so easy to include in

DRSs for technical reasons. A clear problem with the existing producer responsibility

system is that the costs of complying with obligations do not fall fully upon the

obligated companies. A considerable part of the cost of meeting targets for

packaging waste recycling is met through central government grant and Council Tax,

so that there is very little incentive for packaging waste producers to be mindful of

the recyclability of the retail packaging they place on the market.

A range of instruments could be used to ensure that the incentives for product

switching are reduced. Simply put, it would make sense for producers to be obliged

to achieve high rates of recycling of packaging and to meet the costs of doing so (as

under the DRS). Alternative mechanisms such as packaging levies could be used to

influence the materials chosen by manufacturers. Denmark effectively discharges

its packaging waste obligations through a combination of taxes and deposit refunds,

with the tax rates varying depending upon whether the packaging falls within or

outside a deposit scheme.

ERM (2008) Review of Packaging Deposits System for the UK, Final Report produced for Defra,

December 2008.

Eunomia (2009), International Review of Waste Management Policy: Annexes to Main Report,

Report for the Department of the Environment, Heritage and Local Government, Ireland, p.316-321

6.3.5 Multi-Variant Analysis in Cost and Environmental Performance

A significant number of variables are included in the cost benefit model. To test all

inputs and possible outcomes would require significant time, and would not

necessarily provide any further understanding of the fundamental financial impacts

associated with the introduction of a DRS in the UK. However, as discussed

previously, to simply consider a range of discrete scenarios makes it difficult to

obtain an overall idea of the ‘worst’ or ‘best’ case outcomes.

A simulation tool called Crystal Ball® was thus used to perform Monte Carlo analysis

on the key inputs, and record the range of results. The goal of Monte Carlo analysis

is to determine how random variation, lack of knowledge, or error affects the

sensitivity, performance or reliability of the system that is being modelled. ‘Monte

Carlo’ simulation is categorised as a sampling method, because the inputs are

randomly generated from probability distributions to simulate the process of

sampling from an actual population. The key variables tested in the model are given

below (larger variations have been given to assumptions which are less certain):

¾ Deposit Refund System Costs:

• Material Revenues +/- 20%

• Total Transport Cost +/- 50%

• Total Container Costs +/- 50%

• Total Counting Centre Costs +/- 20%

• RVM Unit Costs +/- 30%

• Labour Costs +/- 50%

• Retail Floorspace Rateable Value +/- 50%

• Deposit +/- 20%

• Central System Administration Cost + 200%/ - 50%

¾ Existing Waste Collection Costs

• Bring Site Costs +/- 80%

• HWRC Costs +/- 80%

• Litter Collection Costs +/- 80%

• Commercial Collection Costs +/- 80%

¾ Environmental Costs

• Unit Damage Costs for Air Emissions and GHGs +/- 50%

• Littering Disamenity Benefits £10 to £50 per household

• Transport Damages – Additional Miles +/- 80%

All inputs are varied linearly between the limits given.

¾ Mass Flows:

• Return Rates 85% to 95%

Figure 6-8 to Figure 6-12 show the outputs from the simulation. They represent a

probability distribution based upon the inputs described above. The certainty of the

result falling between the upper and lower bounds (the blue area) is shown at the

bottom of the chart.

Figure 6-8: Monte Carlo Analysis – Net Costs for DRS

Source: Eunomia

Figure 6-8 shows that the costs of the DRS system have an 80% likelihood of lying

between £587 and £817 million. These costs are, as was highlighted earlier, made

up of the costs to producers and the unclaimed deposits from consumers. These two

components are shown in Figure 6-9 and Figure 6-10. Producer costs show an 80%

likelihood of lying between £67 and £443 million, whilst unclaimed deposits are

80% likely to lie between £303 and £605 million.

Figure 6-9: Net Costs to Producers

Figure 6-10: Cost to Consumers (Unclaimed Deposits)

Perhaps the most important figures are those for the net benefit to society. In the

case where no benefit from disamenity is included, Figure 6-11 shows that the costs

are 80% likely to lie between £311 and £537 million.

Figure 6-11: Monte Carlo Analysis – Net Benefit to Society (without Littering

Disamenity)

Source: Eunomia

Once disamenity is included, the picture changes radically. As Figure 6-12 shows,

there is an 86% likelihood that benefits will exceed costs. In the central 80%

likelihood interval, the net benefits to society range from -£43 million to

+£769 million. The most likely outcome is a net benefit to society of £365 million.

The important conclusion therefore, when including the disamenity associated with

litter, is that there is a strong likelihood that the benefits will exceed the costs.

The final presentation, in

Figure 6-13, shows how the figures appear if one adopts

the view that unclaimed deposits are ‘voluntarily’ foregone, and if one excludes

them from the costs of the DRS. This shows that under this assumption, the net

benefits are always in excess of zero. The benefit figures are 80% likely to rest

between £376 and £1,244 million. Strictly speaking, this ignores costs which are

incurred by consumers, but it might be argued that these costs simply represent

income that the consumers have chosen to forego. The validity of this line of

argument clearly depends upon the adequacy (convenience) of the infrastructure for

beverage container returns. Furthermore, if schemes are highly successful in

generating returns, then clearly, the unclaimed deposits fall close to zero and

Figure

6-13 appears very much as Figure 6-12 above.

Figure 6-12: Monte Carlo Analysis – Net Benefit to Society (with Littering

Disamenity)

Source: Eunomia

Figure 6-13: Monte Carlo Analysis – Total Benefit to Society (with Littering

Disamenity), and Excluding Costs Incurred by Consumers as Unclaimed Deposits

The key sensitivities in the model can also be extracted from within the simulation

software. In examining the implementation of a DRS in the UK, it is important to

note that the five most influential variables affecting the value of total benefit to

society are:

1) The value of disamenity per household associated with littering (65%

contribution to variance);

2) The return rate of containers into the system (14% contribution to variance);

3) The deposit applied to the containers;

4) The space costs of retail floorspace; and

5) The total costs of transporting containers during collection and processing.

This indicates that more research into the disamenity associated with littering,

potential return rates and deposit value would help in providing greater robustness

to the results.

6.4 One-Off Costs

Little detailed and sourced information appears to be available on the initial set up

costs that would be required for a DRS in the UK. We therefore constructed the costs

that we believe would be associated with setting up this type of system, based

primarily on what tasks would be required and when (provided by TOMRA), and the

associated number of days that would be required for each task.

A breakdown of

the key tasks involved and the resource and capital costs that we suggest would be

involved in developing and implementing the system are given in Table A-27 in

Appendix A.3.6.

Based on the modelling, a total cost of £32 million would be required to set up the

central DRS, plus an additional £1.25 million for the producers to change their

labelling, and an additional £51 million for the retailers to adapt their store areas to

accommodate the new system requirements. In addition, significant investment

would be required in order to purchase infrastructure such as RVMs and counting

centres. It is unlikely, however, that these would be purchased outright; rather the

typical approach in existing systems seems to be to finance this infrastructure over a

number of years (modelled at around 5 years for RVMs). We have thus incorporated

these annualised costs into the overall ongoing logistics costs of the system.

There is a dearth of detailed information on the overall set-up costs of existing DRSs.

We could find no credible sources on the detailed calculation of joining fees for

either producers or retailers which would be required to cover these one-off costs.

Joining fees vary across existing deposit schemes and can also vary from year to

year (to compensate for changes in return rates and material values, and hence the

subsequent cost of the overall DRS). For example, in Finland, the producer can

currently opt to pay either a one-off lifetime joining fee of £6,698, or an annual

TOMRA (2001), Zentrale Organization Einweg Pfand Deutschland: Business Model Development

Guide.

joining fee of £1,498 over a 5 year period, and must also pay a per product

additional fee of around £300 in both circumstances.

In Denmark, the joining fee

for producers is currently set at £238 per annum, and retailers also pay an annual

fee of £59 to make them eligible to receive handling fee payments.

In Norway, a

one-off joining fee of £3,307 is currently charged for each producer, plus an

additional £551 per product.

For the purposes of this high level modelling, we have thus not attempted to split

the one-off costs into joining fees per producer or per retailer. However, based on

fees in existing deposit schemes, given the size of the one-off costs presented here

and based on the number of producers and retailers, it can be expected that the

average fee for producers might be somewhere around £3,000 to £5,000 (to be

paid as a one-off or over several years) with a per product fee in the region of £300,

and an average retailer fee of around £100 to £200.

A number of key decisions

would require further consideration beyond this study in order to determine how the

one-off costs of the system would be covered, including the following:

¾ Should both the producer and the retailer be charged a joining fee?

¾ If so, how should the one-off costs of the central system be split between the

producer and the retailer?

¾ Should the joining fee be a one-off membership, or an ongoing annual fee?

¾ Should a per product fee be charged on top of a more general fee in order to

reflect the size of producer?

¾ Should the retailer joining fee vary according to annual turnover or some

other equivalent measure?

6.5 Single Market Considerations

Care has to be taken in designing DRSs in Europe such that they are proportionate

in their effect, and do not effectively become trade barriers, or obstacles to the free

movement of goods within the EU Single Market, in ways which are disproportionate

relative to the environmental outcome being sought.

One of the concerns within Europe has been the distributional impact across

domestic and foreign producers of beverages. Several documents highlight the

potential for DRSs to place foreign producers at a disadvantage relative to domestic

producers, the argument being that there is a protectionist slant to DRSs, and that

they fragment the Single Market. In May 2009, the European Commission issued a

http://www.palpa.fi/english,exchange rate at 0.8813

http://www.dansk-retursystem.dk/content/, exchange rate at 0.1189

http://www.resirk.no/Calculator-83.aspx, exchange rate at 0.1102

Based on approximately 1,000 producers and 185,000 retailers; if retailers were, for example, to

cover 70% of set-up costs, this would equate to joining fee of £119 per retail outlet. Remaining costs

would be then be covered by producer one-off fees of £3500 plus a per product fee of £300, based

on an average of 20 products per producer (Coke has approximately 300 products on the market,

whereas smaller firms might have less than 10).

Communication on Beverage packaging, deposit systems and the free movement of

goods:

While regulatory steering measures taken at Member State level in order to

introduce systems for the reuse of beverage packaging may serve

environmental goals, they also have the potential to divide the internal

market. For market operators engaged in activities in several Member States

these systems often make it more difficult to take advantage of business

opportunities within the internal market. Instead of selling the same product

in the same packaging in different markets, they are required to adapt their

packaging to the requirements of each individual Member State, which

usually leads to additional costs.[…]

Past experience and ongoing cases show that the adoption of unilateral

measures in different Member States still poses problems. In particular,

infringement procedures in the beverage sector have shown that national

measures can lead to distortions of competition and, in some cases, to the

partitioning of the internal market, which runs counter to the internal market

aim of Directive 94/62/EC.

Whilst there is clearly an issue associated with DRSs, there are other ways in which

the Single Market remains fragmented, partly owing to the degree to which national

authorities still exercise control over the workings of the domestic market. In many

respects, the principle of subsidiarity (which tends to give Member States

considerable freedom in the way they transpose EU legislation into domestic

legislation) tends to limit the extent to which a Single Market can ever be said to

exist. Notwithstanding the objectives to see this ideal realised in practice, this can

only really happen if Member States simply abrogate complete responsibility for

policy making to the European Parliament. This seems an unlikely outcome, and not

just in the near future. Consequently, the fragmentation of the internal market

remains a fact of life.

We note in passing that the UK has a strict packaging code developed and

implemented relatively recently by the Food Standards Agency (FSA). This requires

UK specific packaging to be produced. Therefore, there would be little additional

burden to producers for developing UK specific packaging, and logistics following the

implementation of a DRS, if this was already the case.

The Communication is presented as a list of ‘do’s’ and ‘don’ts’ for Member States to

assist them in the design of their policies in such a way that problems in respect of

the internal market are minimised or eliminated. It is important to note, however,

that what the Communication very clearly avoids saying is that DRSs are not legal. It

merely highlights the fact that the design of such systems should be carefully

considered, particularly with regard to their effect on the workings of the internal

market. Moreover, it recognises that such schemes can be justified even where they

may be construed as barriers to trade:

European Commission (2009) Communication from the Commission on Beverage packaging,

deposit systems and the free movement of goods, C(2009) 3447 final Brussels 8

May 2009.

The fact that they would qualify as a trade barrier, however, does not prevent

these national provisions from being justified on grounds relating to the

protection of the environment. According to the Court of Justice, a deposit

and return system may increase the proportion of empty packaging returned,

and at the same time lead to a more targeted sorting of packaging waste.

Moreover, it may help prevent littering, as it gives consumers an incentive to

return empty packaging. Finally, insofar as those national provisions

encourage the producers or distributors concerned to have recourse to

reusable packaging, they contribute towards a general reduction in the

amount of waste disposed of, which is a general goal in environmental policy.

In practice, this means that Member States are allowed to introduce

mandatory deposit systems if, on the basis of the individual Member State's

discretion, this is considered necessary for environmental reasons.

It goes on to note, however, the need for a balanced and proportionate approach:

If a Member State opts for a mandatory deposit and return system, it must

nevertheless observe certain requirements in order to ensure that a fair

balance is struck between environmental objectives and internal market

needs. In view of the additional burden on imported products, such systems

must take note of the specific situation and must use means which do not go

beyond what is necessary for the purpose envisaged.

It then describes approaches considered to be appropriate in the case where such

schemes are introduced. This advice appears sound in the light of the considerable

opposition which such schemes have engendered in respect to internal market

effects.

Our analysis highlights the fact that particularly if the environmental disbenefits of

littering are concerned, the overall change in costs are justified by the environmental

benefits. In this respect, the approach could not be said to be disproportionate, and

ought, therefore, to have much to recommend it. Much depends, of course, upon the

efficiency with which the system is designed and operated.

The OECD reports that DRSs can create barriers to trade under the following

circumstances:

¾ if the initial deposits are high compared to the value of the goods;

¾ if foreign producers see that the costs of participating in a co-operative

retrieval and recycling scheme are out of proportion to their market share;

¾ if non-refillable containers are an important condition for the competitiveness

of imports;

¾ if they are applied only to certain types of containers or packaging which are

primarily used for imported products; or

¾ if they are applied in a fashion which is discriminatory or which unduly

favours domestic products.

OECD (1993) Applying Economic Instruments to Packaging Waste: Practical Issues for Product

Charges and Deposit Refund Systems, Paris: OECD.

The OECD concerns reflect those which might be of concern to the World Trade

Organisation. They principally concern the desire to ensure that DRSs are not

disguised measures to protect national producers. Again, there seems no reason to

believe that would be the case in our proposed design, as UK producers are no less

likely to use the targeted materials than foreign producers.

The system we have proposed does not appear to fall foul of any of the OECD

concerns. Some might argue that for some products, the deposit will be quite high

relative to the value of the goods. This is a somewhat subjective matter, but the

deposit levels here do not seem out of step with other systems.

Another concern in Europe has been the need to ensure sufficient infrastructural

provision to cope with increased / altered collection and reprocessing requirements.

This would also seem to be covered in our proposed design.

In their review of the Oakdene Hollins study, Perchards suggest that the inclusion of

wine bottles may be problematic:

We would warn against the inclusion of wine bottles in a DRS. Participation

in a DRS usually requires special marking requirements (a machine readable

logo or bar code that an RVM can identify) and may result in higher

compliance costs. Because the UK produces hardly any wine, these burdens

would fall solely on importers. This would be vehemently challenged as a

barrier to trade – particularly by the French wine producers.

It should be noted that no mandatory European deposit system includes

wines or spirits bottles.

In our model, we have included wine and spirits bottles. The aim of the system was

to be as comprehensive as possible, and it could reasonably be argued that to

exclude wines and spirits would be more discriminatory than their inclusion. The

wording of the OECD above is instructive in that it argues that the measure would

constitute a barrier to trade only if it affected products which are primarily imported.

If all policy measures were conceived in such a way that for any product which was

principally imported, exemptions were to be applied, then all policy would become

ridiculously complex, with imports effectively considered as beyond the remit of

domestic policy. Furthermore, wine bottles (and also spirit bottles) are included in

mandatory deposit schemes in Iceland (a country in the European Economic Area

(EEA)) and Israel, and South Australia and Finland are to introduce this capability

soon.

6.6 Cross Border Issues – Private Trade in Alcohol bringing Non-

Deposit Containers into the UK

We note this issue because a) the scale of cross border trade is significant and b) it

could be claimed by opponents of deposit systems to be a critical issue. The key

European Commission (2009) Communication from the Commission on Beverage packaging,

deposit systems and the free movement of goods, C(2009) 3447 final Brussels 8

May 2009.

Communication with Hans F. Lauszus (Anker-Andersen).

driver is the relative price of beverages. This price may fluctuate from region to

region based upon local conditions, but the two most significant factors on relative

pricing between countries are 1) exchange rates and 2) excise duty on alcohol.

Significant differentials in price were found in a 2001 EU customs report between

UK and France.

Assuming the relative costs of manufacturing and transport are not significant (or,

equivalently, that consumers buy in bulk to minimise these) it is clear that one of the

key drivers on the price of beer is duty. One can see why consumers cross the border

from the UK to France to purchase alcohol, much of it beer and wine.

The customs report also noted that:

“we see that the UK loses most revenue [alcohol duty], at €400 million per

annum”

“There is significant smuggling in the UK, particularly on beer”

Additionally, the effect of exchange rates will also impact on the relative price of all

beverages, not just those which attract an excise duty.

It is clearly possible that, notwithstanding the fact that the deposit is a temporary

payment (and is returned when the can is returned) that some consumers might

perceive it as more beneficial to shop in other countries, such as France or Ireland.

PRO-Europe claims:

“Consumers tend to try to avoid paying deposits by shifting to deposit free

products. This includes shopping in stores across borders where mandatory

deposits are not applied. Consequently, retailers in the border region are

faced with tremendous loses due to ‘customer migration’.”

Source: PRO-EUROPE Position Paper Mandatory Deposit Systems

However, it seems that cross-border purchases between the UK and France already

occur, and in the case of a deposit system in the UK consumers will be more

motivated by the existing ‘without deposit’ differentials in price (i.e. alcohol duty),

rather than the deposit itself (which will be less than the average difference in price

of a can of lager, for example).

This private trade would cause a problem if, for example, containers purchased in

France were not able to be accepted by the DRS return mechanisms in the UK. If

RVMs or automated counting centres did not hold relevant European Article Number

(EAN) codes, the systems would reject the containers and consumers would not be

able to return them in this manner. This problem is easily overcome, even without

the parallel operation of existing kerbside collection systems; if existing kerbside

schemes remained consumers would simply return the empty containers in these

Customs Associates Ltd (2001) Study on the competition between alcoholic drinks, Final Report

February 2001,

http://ec.europa.eu/taxation_customs/resources/documents/study_comp_between_alcoholdrinks_

en.pdf

Pro-Europe (n/a) PRO EUROPE Comments on: Mandatory Deposit Systems for One-Way Packaging,

http://www.pro-e.org/files/08-11_Position_Paper_Mandatory_Deposit_RBV01.pdf

schemes if they were rejected (or for travellers into the UK existing bring or ‘on-the-

go’ recycling services could be utilised).

However, to ensure that the consumer does not lose confidence in the system, and

become irritated when containers are rejected, memory cards in RVMs should be of

appropriate size to enable them to hold all relevant EAN codes from France and

Ireland prior to construction (a minimal incremental cost). This would mean that the

system would recognise the container and it would be accepted, and transferred to

reprocessors. Clearly no deposit would be paid back to the consumer, and no admin

fee would be paid by the French, for example, beer producer, but the central system

would benefit from the sale of the material. This would be especially beneficial in

the case of aluminium cans.

7.0 Summary and Conclusions

In this report we have investigated the environmental and financial implications of

the introduction of a UK-wide DRS. The research question that we were aiming to

answer through this study was:

‘How do the benefits of introducing a UK-wide DRS for certain beverage

container packaging compare with the costs of implementation and

operation?’

In modelling a potential deposit refund model for the UK, we were able to examine

closely the costs and revenues that might be involved in the implementation of a

DRS. Based on existing examples, we calculated that a deposit of 15p and 30p

would be required for beverage containers of ≤500ml and >500ml respectively in

order to achieve a return rate in the region of 90% for the glass bottles, cans and

PET bottles that we included in the DRS. The majority of the cost calculations for the

system centred firstly on how retailers would take back the returned containers

(automatic machine or manual) and the associated compensation that they would

thus require, and secondly on the subsequent collection, counting and transport of

those containers to re-processors. Hence a significant part of the modelling was

based on building up a picture of the retail landscape across the UK. On-going

administration costs for the system were also factored into the modelling.

Based on the complementary scheme as being the central scenario (where beverage

containers are only collected in the DRS and not in kerbside recycling), we

calculated that the DRS on its own could cost somewhere in the region of £700

million per annum. With an assumed 90% return rate, the costs are distributed

across producers, in the form of a 0.7p administration fee on each container placed

on the market, and consumers, to the extent that they choose to forego the possible

income from the deposit. Most of the administration fee would be expected to be

passed onto consumers, though even if the producers decided to pass 100% of this

cost, there would probably be very little change in terms of volume of sales given

that the additional cost per unit is relatively low, and the demand elasticity is not

especially high.

In the parallel scenario, we assumed a return rate at 10% less than in the

complementary scenario, with the rationale being that convenience would prevail if

the kerbside system was still in operation. Due to the value of the deposit,

significant revenue was thus generated from lost or ‘unclaimed’ deposits which,

combined with slightly lower overall DRS costs, generated revenue from unclaimed

deposits which exceeds the cost of the DRS.

Given the size of the deposits (relative to the revenue from material sales), however,

the return rate would be likely to converge on that of the complementary system, be

it due to individuals or the local authority itself extracting containers from the

kerbside collections in order to redeem the deposit. Consequently, the costs falling

on producers might move closer to those suggested under the complementary

system, with local authorities / waste companies realising some of the value of the

unclaimed deposits foregone by consumers.

Since both systems could be considered feasible systems for the UK, then given that

the operational costs are similar between the parallel and complementary

scenarios, and they are likely to converge in any case, it seems sensible to allow

householders to continue to use existing kerbside collection systems if they so

desire.

Importantly, we also included the resultant savings that would be achieved in other

waste management routes, particularly at the kerbside, as a result of the

introduction of a DRS. The removal of containers altogether from the kerbside

collection system, and a slight reduction in containers at bring sites, HWRCs, in

street sweepings and from on-the-go recycling results in a saving of around

£159 million per year for local authorities in avoided waste management costs. This

is a saving of around £7 per household per annum. The additional producer

administration fee costs for the DRS are more or less offset by the savings derived

from other waste management routes. In effect, the overall cost is shifted

specifically onto producers and consumers rather than the population as a whole.

We also examined the environmental impacts associated with the introduction of a

DRS, taking into account the positive impacts associated with increased recycling

and reduced disposal of beverage containers, and the negative impacts associated

with increased transportation required by consumers in returning containers to

collection points, and in the collection and transport of containers from the retail

outlet to the counting centres and beyond. The overall balance of environmental

impacts associated with the introduction of a DRS is a benefit of around £69 million.

The additional recycling from the introduction of a DRS could save up to 607 kt CO

equivalent per annum, which would also significantly help achieve the abatement

required under statutory GHG targets, set out by the Committee on Climate

Change.

100

Finally, we have also taken the first steps towards trying to ascertain the potential

environmental disamenity associated with litter in the environment, and the

potential financial benefit that consequently results from a reduction in beverage

containers present in the environment due to the deposit refund policy. We reasoned

that this benefit could be somewhere in the region of £1.2 billion per annum.

100

This is particularly significant since our baseline assumes that kerbside collection systems of a

relatively high quality exist for all households at the time the DRS is introduced.

We also attempted to construct the one-off costs that would be associated with the

set-up of a DRS in the UK. Based on the modelling, a total cost of £32 million would

be required to set up the central DRS, plus an additional £1.25 million for the

producers to change their labelling, and an additional £51 million for the retailers to

adapt their store areas to accommodate the new system requirements. These one-

off costs are certainly not insignificant amounts; however, given the large number of

producers and retailers involved in the UK market, it should be possible to split the

costs sensibly in order to ensure that the subsequent joining fees are both

reasonable and manageable for both producers and retailers.

The combined overall cost benefit analysis indicates that, even with the additional

costs incurred in the running of the DRS, there is a high likelihood of a significant

net benefit to society. The influence of the reduction in disamenity associated with

litter appears to be particularly strong. Although there is some uncertainty regarding

the magnitude of this, the suggestion is that if households experience a level of

disamenity of the order £16 or so for the removal of 80% or so of beverage related

litter, then the system makes sense from the perspective of society.

To conclude, the modelling indicates that the introduction of a DRS in the UK is:

a) Likely to cost around £84 million per annum to set up if well designed;

b) Likely to cost around £700 million per annum to run (net of revenues);

c) Unlikely to introduce very significant costs to producers. Even at 90% return

rates, in our modelling, the unclaimed deposits fund around 70% of system

costs;

d) Likely to generate savings to local authorities (and hence, to reduce the burden

of taxation) by around £160 million;

e) Likely to deliver strong environmental benefits in terms of:

i. reduced greenhouse gas emissions and air pollutants, mainly from increased

recycling, in the region of £69 million; and

ii. additional benefits associated with the reduction in the disamenity

associated with litter, potentially in the region of £1.2 billion.

Therefore, the case for the introduction of a DRS appears fairly compelling. Even

where we use Monte Carlo analysis to vary the benefits associated with reduced

litter (with the reduced disamenity varying between £10 and £50 per household on

a random basis), the likelihood of net benefits accruing is 86%. This may be

conservative given that our central estimate for disamenity is of the order £48 per

household per year.

With this in mind, a number of key recommendations have been developed for UK

policy makers. These are presented below.

A DRS can arise as a consequence of a decision to implement a mandatory scheme,

or as a response, from industry, to high recycling targets. Defra has argued (albeit,

we suggest, on limited evidence) that there are alternative schemes which can

achieve the same outcomes as DRSs at a lower cost. The evidence in support of this

view is thin, with only Belgium achieving recycling rates approaching the levels

achieved in DRSs (and then, not for PET bottles, for example). Belgium has a

producer responsibility scheme in place which is fully funded by obligated industry. It

also sets targets well above those prevailing in the UK at present, and also has near-

universal implementation of so-called pay as you throw schemes at the household

level, a policy which the Coalition Government has clearly set itself against. Finally, it

is not clear how levels of beverage container litter compare between, for example,

Belgium and those countries where DRSs generate high return rates.

Recommendation 1: The UK Government should introduce a deposit

refund system

Even if other systems could meet the recycling rates achieved by DRSs, there is

scant evidence that they can achieve the same benefits in respect of litter reduction.

The environmental benefits associated with litter reduction are dominant in this

analysis. Therefore, it is clear the Government must consider a DRS from the

perspective of achieving high return rates for recyclable beverage containers, and

significantly addressing beverage container litter.

The research carried out in this report suggests a DRS can:

• Increase recycling rates of beverage containers through rewarding returns;

• Reduce litter by generating an incentive to ‘not throw’ away’;

• Generate environmental gains both in terms of reduced litter and reduced GHG

emissions

In addition, a DRS is a significant mechanism which can be used to deal with

beverage containers (and other packaging) which will reduce the costs to central

government, local authorities and taxpayers of dealing with packaging. This is a

particularly relevant factor in the current economic climate.

DRSs are not always mandatory. They can arise from the setting of high targets for

producers to meet, with DRSs becoming the means to meet those targets. There

are, however, difficulties to be overcome in setting litter reduction targets, hence our

recommendation in favour of a DRS. Even so, in order to ensure that the DRS is

established with convenient infrastructure for returns, it makes sense to set targets

for recycling of beverage containers.

Recommendation 2: High targets should be set for the recycling of all

beverage containers, irrespective of material type

Targets are a pre-requisite for a well-functioning DRS since, in their absence,

schemes may be designed which are inconvenient, deliver low recycling rates, and

lead to high levels of unclaimed deposits, which may even become a source of

revenue.

101

We also note that these targets should apply even to materials and

container types which are not so easily included as part of a DRS, to prevent

producers switching between container types which are, and are not, subject to

targets.

Our study has suggested rates of deposit expected to generate return rates in the

order of 90%. We suggest, therefore, a target rate of 85%, in the first instance, is set

for the targeted beverage containers. Sanctions should be considered in the event of

non-compliance.

In principle, were a DRS to be introduced, whether of mandatory nature, or in

response to the existence of targets, we would recommend the following:

Recommendation 3: A central system should be established to

administrate the deposit refund system.

We would recommend one central system, potentially owned by various

stakeholders, such as industry groups, NGOs and retailers, which would operate to

meet the recycling targets specified by the Government and administrate the DRS

(see Recommendations 1 and 2). This is a similar system to the Scandinavian

approach. The exact nature of the central system would probably reflect the way in

which the DRS emerges (mandatory, or more voluntary in nature), with discussions

to be had about the amount of outsourcing of various functions that would need to

be undertaken.

Recommendation 4: PET bottles, glass bottles and aluminium and steel

cans should be covered by a UK-wide deposit refund system.

We recommend that the following beverage containers should be covered within a

UK-wide DRS:

• PET bottles

• Glass bottles

• Aluminium and steel cans

The option to include other plastic bottles should also be considered. The decision to

include beverage cartons, in light of technological developments being made, should

also be considered prior to full implementation of the scheme. In any case, as

suggested above, such containers should also be made the subject of high recycling

targets.

101

Although in this case, it seems likely that consumers would ‘internalise’ lost deposits in the price

of the beverage, more so than where the deposit can easily be recouped. This might, in turn, depress

demand for the beverage itself.

In order to meet the targets mentioned above (see Recommendation 2), we would

suggest that the level of the deposit (ultimately to be determined by the central

system) should be in the region of:

• 15p for containers ≤500ml; and

• 30p for containers >500ml.

These are the average values used in our modelling – it is recognised that schemes

may differentiate these by material.

Recommendation 5: In order to deal effectively with imported beverage

containers, the deposit refund scheme should operate in parallel with

kerbside recycling services and/or the deposit refund scheme should be

designed to accept containers from France and Ireland.

A large volume of containers crosses the border with France, and to some extent the

Republic of Ireland, as a result of private trade in alcoholic beverages. Neither

country has a deposit system.

We have examined both ‘parallel’ and ‘complementary’ systems in this report, with a

parallel scheme running alongside existing kerbside collections and a

complementary scheme replacing the provision for recycling certain materials at the

kerbside. We have noted that a parallel DRS may increase unclaimed deposits, but

that we expect matters to converge, with some extraction of deposit-bearing

containers from the kerbside recycling system. Given the minimal effect on financial

flows other than the unclaimed deposits, we think it would make sense to still

operate the kerbside recycling service. This enables those consumers who import

beverage containers which bear no deposit to still recycle their containers.

An alternative approach would be to design the system to accept containers with

European Article Number (EAN) codes from France and the Republic of Ireland to

overcome the problem. This would allow reverse vending machines to accept

containers which bear no deposit. Clearly, in these cases, no deposit would be

redeemed, but it seems important for the system to accept containers from abroad,

to ensure consumer confidence is not lost (and especially if the existence of the DRS

leads some local authorities to cease collecting the targeted beverages over the

medium- to long-term).

Recommendation 6: A timescale of introducing a deposit refund

scheme by 2015 should be considered.

Whether in setting recycling targets, or in the context of introducing a mandatory

scheme, UK Government should consider the time scales for the implementation of

a DRS. Four to five years appears to be an appropriate time to allow for

infrastructure development and communication with all stakeholders. In addition,

this would allow for some transitional issues to be considered. For example, local

authorities’ collection schemes will be affected by the implementation of a DRS. In

order to realise the financial benefits to authorities from the DRS’s operation, time

to consider contractual positions and service design would appear appropriate.

Recommendation 7: Further research into the disamenity of litter

should be commissioned by the Coalition Government.

The Coalition Government has set its sights on reducing litter, yet there is no UK-

based study, to our knowledge, that allows us to estimate the negative effects – the

disamenity – of litter. The costs of litter and street cleaning are now a major part of

local authorities’ waste management budgets. They also appear to be costs which

are spent in seeking to address an environmental issue which is consistently cited

by residents as being a priority. There is a need for more research to be undertaken

regarding the disamenity associated with littering, in order to strengthen the

evidence base for estimating the impacts upon litter of a DRS in the UK. Indeed,

given the Coalition Government’s determination to address litter, it would seem

appropriate to consider the economic benefits which might be derived from the

clean-up of litter, if only to understand the level of resource which should be

committed to addressing the matter.

APPENDICES

A.1.0 Review of Deposit Refund Systems

Table 7-1: Experience with Deposit Refund Schemes in Other Countries / States

Country

System

Year of Intro

Containers Covered

Capture

Rate*

Deposit

Redemption

Site

Driver

Reference

Austria

Law to make deposit

regulatory

1992

PET bottles (non-

refillables excluded)

30% PET

60% Cans

$0.40

Government

www.BottleBill.or

Belgium

Ecotaxes Act of

1993

Containers taxed $0.52 per

litre unless they have

deposit.

1993

Beer, soda and soft

drinks containers

$0.12 <50 cl

$0.24 >50 cl

Government

www.BottleBill.or

Croatia

Deposit-return plus

‘incentive fee’ to be paid by

producer if 50% refill isn’t

met (5% paid still, if target

is met).

2005

Glass, PET and metal

containers for beer, soft

drinks, water, wine and

spirits.

Government

EUROPEN Report

2007

102

Denmark

Packaging Law. All beer

and soft drinks must be

sold in refillable bottles.

Metal banned until 2002.

Regulatory deposit for

imported glass/plastic

containers. Ecotax also.

1989

(amended

1991)

Beer and soft drinks

containers. Deposits on

some wine and spirit

bottles dependent on

retailer.

99.5 % (beer

and soft

drinks

containers

only)

€0.13 Type A –

Cans, plastic and

glass bottles <0.5 l

€0.20 Type B –

Plastic bottles @0.5

€0.40 Type C –

Cans, plastic and

glass bottles >0.5 l

Government

www.BottleBill.or

g Ernst & Young

2009

103

Estonia

Deposit-return

2004

Beer, low alcohol drinks,

carbonated/ non-

carbonated soft drinks,

water, juice, cider and

perry.

Glass 1.0 kroon

(refill and NRB)

Metal and PET < 0.5

0.5 kroons;

PET>0.5l 1kroon

Retailers

Government

EUROPEN Report

2007;

http://www.eesti

pandipakend.ee/

eng/epp/emblem

Finland

ax on beverage containers

Exemption from tax only if

1970s

(amended

One-way beer and soft

drink containers

Glass bottles

99%

Non-refillables:

€0.15 cans

8,000 sites

Government

http://www.palpa

.fi/retail-

102

EUROPEN (2007) Economic Instruments in Packaging and Packaging Waste Policy, Brussels: EUROPEN.

103

Ernst & Young (2009) Assessment of Results on the Reuse and Recycling of Packaging in Europe, report produced for the French Agency for

Environment and Energy Management (ADEME), March 2009.

Country

System

Year of Intro

Containers Covered

Capture

Rate*

Deposit

Redemption

Site

Driver

Reference

part of refillable deposit

scheme.

1990)

Cans 86% €0.10 plastic

bottles <0.35 l

€0.20 plastic

bottles 0.35-1 l

€0.40 plastic

bottles >1 l

Tax

$0.24 beer $0.47

plastic $0.71 glass

trade/recycling-

systems

Germany

Einwegpfand Deposit on

one-way a standard

amount, deposit on

refillables manufacturer

dependent, not legally

specified, though tend to

be similar.

2003

Not containers for wine,

fruit juice or spirits

Quota-

Glass 90%

Alu. 90%

Plastic 80%

For one-way:

≤1.5 l €0.25

>1.5 l €0.50

Manufacturer

Hungary

Tax linked to market share

quotas.

2005

Beer, low-alcohol drinks,

wine, mineral water,

carbonated and non-

carbonated soft drinks.

Quota-

Beer 67%

Low alcohol

28% Wine

20%

Iceland

Tax on non-refillable

containers.

2008

Non-refillable glass,

steel, aluminium and

plastic.

Kiribati

Special Fund Act 2004

2004

Aluminium cans and PET

drinks bottles

$0.05 ($0.04

returned)

Kaoki Mange

operating

centres.

www.BottleBill.or

Malta

Deposit Return

System

Previous ban of non-glass

beverage containers, lifted

Mexico

Higher tax on non-refillable

bottles and cans.

Fed. States of

Micronesia

Kosrae Recycling

Program

(Deposit-return)

1991

(amended

2006)

Currently only aluminium

cans, but glass and

plastic expected to be

added soon.

20,000 cans

per day

$0.06 ($0.05 back)

Kosrae Island

Resource

Management

Authority

(KIRMA) sites

www.BottleBill.or

Netherlands

Agreement deposit

1993

Soft drinks and water in

one-way and refillable

glass and PET containers

Refillable

glass 98%

Refillable

PET and glass:

$0.16 <0.5 l

$0.72>0.5 l

Industry

www.BottleBill.or

Country

System

Year of Intro

Containers Covered

Capture

Rate*

Deposit

Redemption

Site

Driver

Reference

PET 99%

Norway

Deposit on containers and

tax dependent on return

rate.

Refillables only exempt if

95% return rate is

achieved. Retailers (on site

>25m²) selling non-

refillables, must also sell

similar products in

refillable.

1994

Most drinks excluding

milk, vegetable juices

and water

Wine/ spirits

60%

Beer 98%

Soft drinks

98%

$0.16 <0.5 l

$0.40 >0.5 l (+Tax

inversely

proportional to

return rate, but if

above 95%, no tax)

Over 9000

establishments

in the country,

plus 3000

deposit

machines

where receipt is

given

Tax is

government

driven, but

recycling fee in

place is retailer

driven

http://www.resirk

.no/Introduction-

64.aspx

Peru

Deposit on some bottles

620ml size beer bottles

Portugal

Fillers must ensure quotas

met

Retailers must sell

refillables for all non-

refillables sold.

Quotas-

Beer 80%

Wine (with certain

exceptions) 65% Soft

drinks 30%

South Africa

Deposit return system,

voluntary i.e. manufacturer

driven, not Government.

Around 1948

Approx. 75% beer, 45%

soft drinks and some

wine and spirits bottles

Between 8-15% of

product cost

(or 0.5-1% if

wine/spirit)

Manufacturer

Spain

Return

overall 87%

Reuse beer

57%

http://www.cerve

ceros.org/

Sweden

Law requires rate of 90%

recycling of aluminium

cans, or complete ban.

Industry implemented

deposit system to avoid

this. PET introduced later

as well.

deposit

Deposit on

one-way

containers-

1984 for

cans.

1994 for PET

(refillables

already in

place)

Aluminium cans and PET

law. Deposit now on

most beverage

containers.

Recovery

rate of 80-

90% on one

way

containers

Voluntary Cans

$0.07

Refillable PET $0.56

One-way PET $0.14-

0.24

Law

government

driven.

Standard bottle

and deposit

brewer/bottler

driven.

www.BottleBill.or

g; Ernst & Young

(2009)

103

Switzerland

Deposits required on all

refillable drinks containers

except cans, which have a

voluntary tax of $0.04.

1990

All above a certain

weight (currently all!)

Refillable

glass 95-

98%

Refillable

PET 70%

Ref. glass

$0.16<0.6 l

$0.40 >0.6 l

Ref and one-way

PET $0.40>1.5 l

Government

www.BottleBill.or

Country

System

Year of Intro

Containers Covered

Capture

Rate*

Deposit

Redemption

Site

Driver

Reference

South Australia

Container Deposit

Legislation- deposit

required on almost all

drinks containers, with

onus on manufacturer/

wholesaler to ensure

convenient system in place

for deposit of container/

refunds for customers.

1975

(integrated

into

Environment

Protection

Act in 1993)

Most included except

wine (unless in plastic

bottle), milk, pure fruit

juice or flavoured milk

>1l.

85% non-

refillable

glass

84% cans

74% PET

$0.10 if refillable to

retailer (rare) $0.05

if refillable to

collection depot

(99.9% done this

way)

Mostly

collection

depots, though

some store

refillables.

Government

legislation with

manufacturer/

wholesaler

responsibility

www.BottleBill.or

Canada- Alberta

All containers sold in

Alberta (including imports)

must be registered through

the Beverage Container

Management Board

(BCMB).

1972

All beverage containers

regulatory except milk,

which is under a

voluntary scheme

Glass (AB

Beer) 96%

Glass (import

beer) 92%

Alu (beer)

89%

Alu (soft)

79%

Overall 78%

$0.05 <1 l

$0.20 >1 l Beer

$0.10

215

independent

depots and 78

retail outlets

(for beer

bottles and

cans only)

Initially

government,

until 1997

when it was

turned over to

private sector

www.BottleBill.or

Canada- British

Columbia

All containers must be

refillable, and none

collected can be landfilled

or incinerated.

Beer separate system,

though still under

legislation.

1970

All beverage containers

except milk, soya milk,

infant formulas, dietary

or meal supplements, or

other milk substitutes.

81.3%

Non-alcoholic $0.05

<1 l

$0.10> 1 l

Alcoholic (not incl.

beer)

$0.10 < 1 l

$0.20 >1 l Beer

$1.2 per dozen

Depots or

retailers (all

retailers

obliged to take

back as much

as they sell).

Beer back to

retailer.

Industry

www.BottleBill.or

Canada-

Manitoba

Beverage producers given

option of setting up

deposit-return system, or

adding a 2 cent per

container levy. Only beer

producers choose the

former.

1995

Beer containers only

Refillable

beer 95.5%

Dom beer

74%

Glass 34%

Overall

residential

31%

$0.10

Retailer

Opportunity

government

driven,

implementati-

on producer

driven

www.BottleBill.or

Canada- New

Brunswick

Deposits paid on all

containers (bar milk), but

1992

(revised

All except milk

Refillable

beer 96%

<0.5 l $0.10

>0.5 l $0.20

89 depots

around the

Industry

www.BottleBill.or

g; Encorp Atlantic

Country

System

Year of Intro

Containers Covered

Capture

Rate*

Deposit

Redemption

Site

Driver

Reference

whilst full paid back on

refillables, only half paid

back on non-refillables.

1999)

Dom beer

75%

Non-alcoholic

75%

province.

Inc

(unpublished)

104

Canada-

Newfoundland

Half-back system, with

manufacturers prohibited

from selling containers

other than recyclable or

refillable for selected

products. Beer operated

separately, run by brewers.

Only have to refund when

customer buying (1 for 1),

otherwise negotiable.

1997

Beverage containers

smaller than 5l,

excluding milk, dietary

supplements and

medicine.

Refillable

beer 95%

Domestic

beer 55%

Non-alcoholic $0.08

($0.04 back)

Alcoholic (excluding

beer) $0.20

($0.10 back)

Beer varies -full

refund when same

number of beer

bought as empties

returned.

‘Green Depots’

run as

businesses.

Beer returned

to certain retail

outlets.

Government,

but brewers for

beer system.

www.BottleBill.or

Canada-

Northwest

Territory

Deposit-return system, with

additional handling

charges for different

products/ materials in

container.

2005

All beverage containers

except milk.

Very new

system, so

no certain

figures yet.

Approx. 72%

Wine or spirit $0.25

Other $0.10

Plus additional

$0.05-0.10 handling

fee

18 government

depots or 26

community

depots.

www.BottleBill.or

Canada- Nova

Scotia

Half-back deposit system.

Full refund on refillables,

half on non-refillables.

All beverage containers

except milk.

Refillable

beer 96%

Dom. beer

70%

Non-alcoholic $0.10

Alc. refillable

<1 l $0.10

>1 l $0.20

Alc. non-refillable

<0.5 l $0.10 >0.5 l

$0.20

83 province-

wide depots.

RRFB-Resource

Recovery Fund

Board

Government

and industry.

www.BottleBill.or

Canada- Ontario

Deposit-return system on

alcoholic drinks containers

only.

Use of ‘Industry Standard

Bottle’.

Alcoholic drinks

containers

Refillable

‘industry

standard

bottles’ beer

97%

Containers up to

630 ml, or metal

containers up to 1 l

$0.10 Over those

sizes $0.20

Beer store only

Brewers

www.BottleBill.or

Canada- Prince

Edward Island

Non-refillable drinks

containers for beer or soft

drinks banned since 1977.

1977 ban,

1984 deposit

Soft drinks and alcoholic

drinks. Wine may be

included.

Refillable

beer 96%

Wine/ spirit

Non-al

<0.5 l $0.15

0.5 l-1 l $0.30

Mainly retailers

(inc.

supermarkets

www.BottleBill.or

104

ENCORP Atlantic Ltd (unpublished), Accounting for Success: EnSys – A Materials Management System to Support the Recovery of Used Beverage

Containers in the Province of New Brunswick, May 2008.

Country

System

Year of Intro

Containers Covered

Capture

Rate*

Deposit

Redemption

Site

Driver

Reference

Wine may have half-back

system in place.

59%

Soft 98%

>1 l $0.70 Alc.

$1.20 per dozen, or

).07 each

and

convenience

stores), also 15

depots

Canada- Quebec

Return-to-retail deposit

system, with industry

required to fund kerbside

collection for containers

not part of the system.

All beer and soft drinks

containers (not juice,

water and iced tea)

Refillable

beer 98%

Dom. beer

76%

Soft drinks and beer

cans

$0.05

Beer bottles $0.10

Beer bottles and

soft drinks >450ml

$0.20

Retailers

(including

depanneurs -

small

convenience

stores not

usually

included in

Canada).

www.BottleBill.or

Canada-

Saskatchewan

Deposit-return system plus

environmental handling

charge (EHC) for non-

refillable containers, for

recycling, and beer bottle

deposit system for

refillables.

1973- Litter

Control

Regulations

(unclear,

appears the

deposit

system

introduced to

this in 1998)

All beverage containers

apart from milk (under

voluntary system).

Refillable

beer 92%

Dom. beer

cans 95%

Alu. cans

95%

Glass 83%

Overall 86%

Deposits vary widely

for diff. materials

and sizes

Non-ref. glass

$0.40-1.00

Metal cans

$0.10-0.20

Beer bottles

can only

receive full

refund if

returned to 10

specific sites,

but can be

returned for

less at retailers.

Other returns at

71 SARCON

site

Government

www.BottleBill.or

http://www.sarcs

arcan.ca/sarcan/

faqs.php

Canada-Yukon

No kerbside collection.

Deposit-return system, with

‘recycling club’ for children

offering ‘prizes’ as well as

refund if certain numbers

reached. Refillables not

charged recycling fund fee,

all others are.

1998

All beverage containers

except milk.

Refillable

bottles 103%

Non-refill.

bottles 113%

(?)

Liquor

containers

<200ml 99%

1L 90%

>1L 79%

(includes

refillables)

D=deposit,

R=refund

Liquor ref. D=$0.10

R=$0.10

Liquor non

<0.5 l D=$0.15

R=$0.10

>0.5 l D=$0.35

R=$0.25

22 depots or

four Liquor

Commission

outlets

Government

www.BottleBill.or

USA-California

California Beverage

Container Recycling and

Litter Reduction Act

1987

(Expanded

2000 to

Non-refillable drinks

containers, inc. beer,

spirits, carbonated, fruit

Alu 73%

Glass 58%

PET 46%

Under 24oz $0.05

Over 24oz $0.10

Redemption

centres (not

retailers)

www.BottleBill.or

Country

System

Year of Intro

Containers Covered

Capture

Rate*

Deposit

Redemption

Site

Driver

Reference

Deposit-return system on

non-refillable containers

include all

non-

carbonated

and non-

alcoholic

drinks

excluding

milk.)

drinks and some

vegetable juices. Not

milk.

HDPE 51%

Overall 61%

USA-Connecticut

Beverage Container

Deposit and Redemption

Law

Deposit-return system.

1980

Beer, malt, soft drinks

and mineral water.

Not recorded.

In 2004 CRI

estimated

recycling rate

to be similar

Massachuset

ts of 69%

$0.05

Redemption

centres, or

retailers (but

only for brands

/products they

sell).

www.BottleBill.or

http://www.cga.c

t.gov/2005/rpt/2

005-R-0836.htm

USA- Delaware

Beverage Container

Legislation

Deposit-return system

1982

Wholesale

1983 Retail

All non-aluminium beer,

malt, carbonated,

mineral water and soda

water containers less

than 2 quarts (approx.

1.9l).

Not recorded. $0.05

Retail stores,

but only for

brands they

sell.

www.BottleBill.or

USA-Hawaii

Deposit Beverage

Container Law Deposit-

return system

2002

All beverage containers

excluding milk and dairy

derived products, except

tea and coffee or liquor

containers.

72% for

2008

$0.05

Redemption

centres or

retailers (if not

within 2miles

of red. centre in

highly pop.

areas, or if

under 5,000sq

ft of retail

space

Government

www.BottleBill.or

USA-Iowa

Beverage Container

Deposit Law

Deposit-return system.

Deposit containers banned

from landfill in 1990.

1979

Beer, soft drinks, soda

water, mineral water,

wine, liquor and wine

coolers.

93% Not less than $0.05

Redemption

centres or

retailers (who

can refuse if

they have an

agreement with

former).

www.BottleBill.or

USA- Maine

Refillable Beverage 1978

Beer, soft drink, wine Not recorded. Wine and liquor Redemption

www.BottleBill.or

Country

System

Year of Intro

Containers Covered

Capture

Rate*

Deposit

Redemption

Site

Driver

Reference

Maine

Container Law Deposit-

return system

cooler, mineral water.

Expanded to include

wine, liquor, water and

non-alcoholic drinks in

1989.

$0.15 Other $0.05

centres or

retailers (who

can refuse if

they have an

agreement with

former).

USA-

Massachusetts

Beverage Container

Recovery Law

Deposit-return system

1983

Beer, soft drinks and

carbonated water.

69% $0.05

Any retail

establishment

that sells the

container.

www.BottleBill.or

USA- Michigan

Michigan Beverage

Container Act

Deposit-return system

1978

Beer, soft drinks,

carbonated and mineral

water. Wine coolers and

canned cocktails in

1988.

97% $0.10

Retail stores

wwiw.BottleBill.or

USA- New York

New York State Refillable

Container Law

Deposit-return system

1983

Beer and other malt

drinks, carbonated soft

drinks, wine coolers,

mineral and soda

waters.

Soft drink

62%

Beer 77%

Wine coolers

65% Overall

70%

Minimum of $0.05

Retail stores

and

redemption

centres.

www.BottleBill.or

USA- Oregon

The Beverage Container

Act

Deposit-return system

Only US deposit law with no

handling fee.

1972

Beer, malt, carbonated

soft drinks, mineral and

soda water and (as of

2009) water and

flavoured water. Bottles

and cans under 3L

Overall 84%

Standardized refill

bottles $0.02

Non-standardized

and non-refillable

$0.05

Retail stores.

www.BottleBill.or

USA- Vermont

Beverage Container Law

Deposit-return system

1973

Beer, soft drinks, malt,

soda and mineral water,

mixed wine and liquor

(added 1987).

Overall 90-

95%

Liquor above 50ml

$0.15

Other $0.05

Retail stores

and

redemption

centres.

www.BottleBill.or

Source: data based on report by Oakdene Hollins (2008) Refillable Glass Beverage Container Systems in the UK, Report for WRAP, 26 June 2008.

*The report notes that capture rate includes containers returned for recycling as well as refilling. Separate figures were not so readily available. Unless

specifically listed as something else, the monetary unit is American dollars. The only exception is Canada where the Canadian dollar is used.

Percentages given for US capture rates are taken from various sources, often telephone conversations by the Bottle Bill researchers. For more detailed

references see www.BottleBill.org

100

A.2.0 Cost Benefit Analysis Model

The cost benefit analysis (CBA) model has been developed by Eunomia as a bespoke

model that also utilises Eunomia’s existing municipal waste collection model,

Hermes.

105

The overall structure of the model is given in Figure A-7-1.The key elements are:

1) A waste baseline for each of the key materials, which will include modelling of

kerbside collection of household waste;

2) Scenario waste flow modelling;

3) Deposit refund (DR) system modelling;

4) Environmental impacts of recycling and disposal (including the disamenity of

litter); and

5) Final results calculations.

The remainder of this section first provides details on the materials that we have

included in scope for the deposit refund system, as these will form the focus of the

mass flow modelling. It then examines the waste mass flow assumptions used in

order to model the baseline, followed by the key changes that are subsequently

made to the waste mass flows as a result of introducing a deposit refund model in

the UK.

105

Hermes has been used in projects for WRAP, Defra, Welsh Assembly Government, and by many

local authorities around the UK. It has also been used in developing shadow bids in procurement

processes for waste collection services.

101

Figure A-7-1: Cost Benefit Analysis Model Schematic

Mass Flow Baseline

Scenario Waste

Flows

Monetised

Envi ron ment al

Impacts (per tonne)

Costs – Bring/

Commercial/ Litter

(per tonne)

Deposit Refund Costs

Reduct ion in

tonnes arising

at kerbside

Costs and Benefits

Change in Tonnages

102

A.2.1 Materials to be Included in Deposit Refund System

The materials that we have included in the deposit refund system are one-way (non-

refillable) beverage containers as follows:

1) Plastic bottles made from PET (Polyethylene Terepthalate) e.g. fizzy drinks,

mineral water, squash bottles. The recycling symbol on these products is:

2) Metal cans, both steel and aluminium e.g. fizzy soft drinks, beer cans, energy

drinks etc.

3) Glass beverage containers e.g. beer bottles, wine bottles, soft drink bottles etc.

Although there is, strictly speaking, no reason why, in theory, other containers or

packaging could not be collected in these systems, the model has been designed

around beverage containers for the following key reasons:

¾ Beverage containers are more likely than other types of food-based

containers to be consumed away from home and thus end up as litter;

106

¾ More investment in technology would be required in order to enable

recognition in reverse vending machines (RVMs)/counting centres for other

types and, importantly, shapes of containers/packaging;

¾ It enables industry-specific modelling, reducing the number of stakeholders

and facilitating easier management of the system; and

¾ Hygiene issues, in particular with regard to plastic milk bottles and other

food-based containers, have been given as a reason for not including high-

density polyethylene (HDPE) in existing deposit refund systems.

107

The modelled system targets non-refillable containers, because the market for

refillables in the UK is much smaller than that for non-refillables, and because a

deposit refund system would encourage the capture of non-refillables which are

purchased away from home as well as those consumed in the household. Some

deposit refund systems (DRSs) do remain in the UK, but they are the exception

rather than the rule, and target the smaller market of refillable glass bottles (e.g. A.

G. Barr scheme in Scotland, milk rounds across the UK), rather than the growing

market of disposable containers.

108

Targeting non-refillables exploits the potential

106

http://www.bottlebill.org/about/benefits/curbside.htm

107

ERM (2008) Review of Packaging Deposits System for the UK, Final Report produced for Defra,

December 2008.

108

See http://www.agbarr.co.uk/agbarr/newsite/ces_general.nsf/wpg/corporate_responsibility-

courtauld_commitment_2!OpenDocument

103

for increased recycling rates, a reduction in litter levels and an increase in the quality

of material that is collected for recycling through the deposit mechanism.

A.2.2 Baseline

The first step in building the cost benefit analysis model was to consider the

material flows in the UK, where the waste arises and how much of the waste is sent

for recycling compared to how much ends up requiring disposal. Figure A-7-2

indicates the possible material flows in our container universe (before the DR

System).

Figure A-7-2: Possible Container Material Flows (Pre-DRS)

The second step was to consider which year to model as the baseline year. In this

study we have assumed the baseline year is around 2015; the landfill tax escalator

will have increased to £80 per tonne by 2014/15, and by this time, it is also likely

that fully comprehensive kerbside collection services will have been rolled out to all

104

households in the UK.

109

In order to provide direct comparison, the ensuing changes

to the baseline as a result of the introduction of a deposit refund system have also

been modelled based on the same year.

A.2.2.1 Household Kerbside Collection Modelling

Eunomia’s proprietary waste collection model, Hermes, has been used to investigate

the effect of implementing a deposit refund system in the UK on kerbside collection

schemes. Hermes is a sophisticated spreadsheet-based tool that allows a wide

range of authority specific and collection scheme specific variables to be modelled.

The optimisation of these variables allows us to build scenarios to accurately reflect

local circumstances. The main outputs of the model are recycling performance and

cost.

Eunomia is confident that Hermes is as reliable a tool as any of its kind. It has been

used to model systems for authorities that collectively manage around 25% of the

UK’s total municipal waste. It is used to support contract procurement advice and

contract dispute resolution by building ‘shadow’ bids against which contractors’

tender submissions can be tested. Hermes has also been used in the context of

studies of relevance to national policy, and to undertake a cost benefit analysis for

the kerbside collection of food waste across the UK.

110

The first step in building up a model of UK-wide kerbside collection schemes was to

develop a number of scenarios based on combinations of urban/rural classifications

and dry recycling collection systems currently employed across the UK. Classification

of local authorities across the United Kingdom into rural and urban varies from

nation to nation. For England, Wales and Northern Ireland, we were able to obtain

already-classified data on local authorities based on a broadly similar six-fold

classification as follows:

111,112,113

• Major Urban (districts with either 100,000 people or 50 percent of their

population in urban areas with a population of more than 750,000).

• Large Urban (districts with either 50,000 people or 50 percent of their

population in one of 17 urban areas with a population between 250,000 and

750,000).

109

Fully comprehensive means a system that would collect all of the containers in scope in this

study.

110

Eunomia (2007) Dealing with Food Waste in the UK, report for WRAP, March 2007

111

Defra (2008) Classification of Local Authority Districts and Unitary Authorities in England: An

Introductory Guide, Updated September 2008

http://www.defra.gov.uk/evidence/statistics/rural/documents/rural-defn/LAClassifications-

introguide.pdf

112

ONS (2005) Rural and Urban Area Definition for Middle Layer Super Output Areas, available at

http://www.ons.gov.uk/about-statistics/geography/products/area-classifications/rural-urban-

definition-and-la-classification/rural-urban-definition/index.html

113

NISRA (2005) Urban Rural Classification 2005, provided by NINIS, May 2010

105

• Other Urban (districts with fewer than 37,000 people or less than 26 percent

of their population in rural settlements and larger market towns).

• Significant Rural (districts with more than 37,000 people or more than 26

percent of their population in rural settlements and larger market towns).

• Rural-50 (districts with at least 50 percent but less than 80 percent of their

population in rural settlements and larger market towns).

• Rural-80 (districts with at least 80 percent of their population in rural

settlements and larger market towns).

For Scotland, the data available was not already classified on an overall local

authority basis; data was, however, available on the percentage of the population in

each local authority that can be found in each of six different urban/rural categories

given above.

114

We therefore used these percentages to classify each authority into

one of the six listed categories.

These classifications were then grouped to create a three-fold classification for all

authorities:

• Urban (Major Urban and Large Urban);

• Towns (Other Urban and Significant Rural); and

• Rural (Rural-50 and Rural-80).

In addition to the urban/rural classification, the number of households in each

district and the recycling collection scheme were collated for each district in the UK.

The dry recycling kerbside collection schemes were categorised into three types as

follows:

• Fortnightly single stream commingled (where the material is collected

unsorted and subsequently sorted at a Materials Recovery Facility (MRF)),

referred to as ‘Commingled’;

• Fortnightly two-stream (residents are asked to separate material into two

different collection containers, typically fibres in one (i.e. paper, card) and

containers in the other (i.e. cans, plastics, sometimes glass), with some

subsequent further separation required at a MRF), referred to as ‘Two-

stream’;

• Weekly sorting of materials at the kerbside, referred to as ‘Kerbside Sort’.

Alongside each of the dry recycling schemes, we modelled fortnightly residual waste

collections and weekly food waste collections (with the latter modelled as a being

collected on a separate vehicle for the commingled and two-stream systems, and as

being collected at the same time as the dry recyclables on the same vehicle in the

kerbside sort system). Garden waste collections were excluded from the analysis,

given that garden waste is typically collected in a separate pass and that the

114

Scottish Executive (2004) Urban Rural Classification 2003-2004, available at

http://www.scotland.gov.uk/Resource/Doc/47251/0028898.pdf

106

logistics for this waste stream would not subsequently be affected by any change in

beverage containers in the kerbside system.

This data was based primarily on our extensive knowledge of kerbside systems

across the United Kingdom, with any missing information provided by the Online

Recycling Information System (ORIS) sourced from WRAP.

115

The three rural/urban classifications were combined with the three collection

systems to create nine baseline categories. A summary of the number of authorities

in each of the nine categories, the average number of households and the average

population are provided in Table A-2, Table A-3 and Table A-4 respectively.

For the purposes of modelling in Hermes, the overall averages given in Table A-3 and

Table A-4 for the number of households and population were subsequently used for

the basic configuration of the three “districts” to be modelled, named Urban, Towns

and Rural. For each of these three districts, the three collection systems were then

modelled separately.

Table A-2: Number of Authorities in Our Nine Categories.

Urban Towns Rural Total

Commingled 60 48

155

Kerbside Sort 61 58

182

Two-Stream 29 23

Total 150 129

124

403

Table A-3: Average Number of Households in Each Authority for Each Category,

Rounded for Use in Modelling.

Urban Towns Rural Overall Average

Commingled 91,200

51,600

50,300

66,600

Kerbside Sort 79,500

51,800

46,900

59,400

Two-Stream 105,700

66,800

49,700

80,200

Overall Average 89,300

54,400

48,500

115

WRAP (2010) Recycling Information, Accessed May 2010.

http://www.wrap.org.uk/local_authorities/research_guidance/online_recycling_information_system

_oris/mapping.html

107

Table A-4: Average Population in Each Authority for Each Category, Rounded for Use

in Modelling.

Urban

Towns

Rural

Overall Average

Commingled 179,000

121,000

110,000

138,000

Kerbside Sort 185,000

115,000

113,000

136,000

Two-Stream 247,000

151,000

109,000

182,000

Overall Average 195,000

123,000

111,000

One of the most important inputs to the kerbside modelling is the quantity of

material collected for recycling. For this model, the tonnages were based upon the

WRAP report on kerbside recycling performance in England.

116

This report gives

recycling in kg per household per year for the six-fold rural/urban classification

above, as well as for the different recycling systems. The results for the six

rural/urban classifications were averaged in pairs to create three classifications.

The results printed in blue in Table A-5, Table A-6 and Table A-7 are the upper

quartile data for captures of the three materials (cans, glass and plastics) which

would be covered by the deposit refund system. We used the upper quartile because

we assume that recycling systems in 2015 would be performing better than the

median in 2007/8. Using a mathematical analysis technique (least squares

method), we were able to calculate the tonnage data for the nine scenarios given in

each table based on the data provided by WRAP.

Table A-5: Cans Dry Recycling Performance in kg/hhld/yr Calculated using Least

Squares from WRAP Kerbside Recycling Performance Data

Urban Towns Rural WRAP Data

Commingled 9.2 9.9

10.4

Kerbside Sort 12.2 12.8

13.4

Two-Stream 10.3 9.7

9.7

WRAP data 10.5 11

11.5

116

WRAP in association with Icaro Consulting (2009) Analysis of kerbside dry recycling performance

in England 2007/8, Summary report for WRAP, December 2009

108

Table A-6: Glass Dry Recycling Performance in kg/hhld/yr Calculated using Least

Squares from WRAP Kerbside Recycling Performance Data

Urban Towns Rural WRAP data

Commingled 49.4 53.2

58.4

Kerbside Sort 42.3 57.3

63.4

Two-Stream 45.5 43.6

45.1

WRAP Data 49.5 53

Table A-7: Plastic Dry Recycling Performance in kg/hhld/yr Calculated using Least

Squares from WRAP Kerbside Recycling Performance Data

Urban Towns Rural WRAP data

Commingled 11.8 12.2

12.2

Kerbside Sort 10.4 11.4

11.4

Two-Stream 10.4 9.9

9.7

WRAP Data 11 11.5

11.5

In addition the model also accounts for the tonnages of paper, card and food waste

being collected in the kerbside recycling systems; these were kept constant

throughout all modelling and therefore do not affect the results, but were simply

used to calibrate baseline performance based on known authorities.

In order to then drill down further into the actual tonnages of material that would

feed into the deposit refund model, we made the following additional assumptions

(in % by weight). These were based around detailed household compositions and our

knowledge of operational collection systems; furthermore they were validated when

outputs were sense checked against existing sources of packaging waste generation

and recycling, and found to be comparable:

¾ 75% of glass captured is beverage (as opposed to other) container glass;

¾ 50% of the plastic captured at the kerbside (mainly comprised of bottles) is

PET, with 50% being HDPE;

¾ 20% of cans are aluminium, most of which are beverage containers;

¾ 80% of cans are ferrous metals, with 30% of these being beverage

containers. The remaining ferrous cans being food containers, etc.

The total quantity of waste left in the residual stream is calculated as:

109

Other important assumptions for the Hermes modelling were the participation rate

(the percentage of households that participate in the recycling scheme at least once

every two weeks) and the set-out rate (the percentage of households that set out a

box in any given week, which will be less than the participation rate because not

everyone is setting out recyclables at every opportunity). For consistency across all

three types of collection scheme, both of these values were kept constant for all

options modelled, and were set at high-performing values of 85% and 72%

respectively for the dry recycling, and 100% and 90% for residual waste collection.

However, in reality, the set out and participation rates would be likely to be closer for

fortnightly commingled collections than for weekly kerbside sort.

A.2.2.2 Bring Sites / Household Waste Recycling Centres (HWRCs)

WasteDataFlow (WDF) was interrogated and data for all waste collection authorities

(WCAs) and Unitary authorities was compiled for the whole of the UK. The relevant

questions in WDF are:

The following assumptions, based upon the rationale described above, were

required to estimate the quantities of beverage containers currently being recycled

through these sources:

¾ All of ‘brown glass’ is beverage containers;

¾ All of ‘green glass’ is beverage containers;

¾ 35% of ‘clear glass’ is beverage containers (the remaining 65% is jars and a

small quantity of other clear glass such as broken pint glasses);

¾ 75% of ‘mixed glass’ is beverage containers;

¾ 50% of Plastics are PET beverage containers;

¾ 80% of mixed cans are ferrous metals / 20% are aluminium; and

¾ Commingled materials are disaggregated using the proportions of source

segregated materials captured for recycling.

See Table A-9 for estimates of the tonnages of beverage containers from bring sites

and HWRCs.

Waste in Refuse = Total Waste Collected at Kerbside

–

Waste Captured for

Question 16: Civic Amenity (CA) sites: Tonnes of material collected for

recycling/reuse at CA Sites operated by local authority or its contractors

Question 17: Bring sites: Tonnes of material collected for recycling/reuse at

bring sites operated by local authority or its contractors

110

A.2.2.3 Commercial Wastes

The definition of commercial wastes in this study includes all waste from non-

household sources. This includes beverage containers deposited in refuse or

recycling schemes from commercial or industrial enterprises.

Estimates of the total quantity of glass, dense plastics and metals in the commercial

waste stream for recycling or requiring disposal were taken from the recent landfill

bans cost benefit analysis.

117

Assumptions were then made to estimate the

proportion of beverage containers in each waste fraction, e.g. glass bottles in ‘glass.’

These estimates were then adjusted where required to ensure that a reasonable and

logical baseline was achieved.

See Table A-9 for estimates of the tonnages of beverage containers from

businesses.

A.2.2.4 On-the-Go Recycling and Street Sweepings

WasteDataFlow was interrogated and data for all waste collection authorities

(WCAs) and Unitary authorities for on-the-go recycling was compiled for the whole of

the UK. The relevant question in WDF is:

It should be noted that a caveat is included with the tonnages reported in WDF for

street recycling. There appeared to be large variations in the data reported between

authorities. Upon investigation, it was determined that the reporting of waste

collected in on-street recycling bins is not rigorous, and there are improvements to

be made in this area of reporting.

The same assumptions used for bring sites and HWRCs to estimate the quantity of

beverage containers captured for recycling (Appendix A.2.2.2) were also applied to

the tonnages reported for on-the-go recycling.

WDF was also interrogated to gather information on the total quantity of waste

collected from street sweepings for disposal (ie. litter that is collected from the

environment for disposal, at the expense of the taxpayer). This far outstrips the total

quantity of waste collected through on-the-go recycling bins. The relevant question is

given below:

The total quantity of street sweepings collected for disposal was disaggregated, on a

material basis, using the composition obtained by AEA Technology from the Welsh

117

Eunomia (2010) Landfill Bans Feasibility Research, Final Report for WRAP, March 2010,

http://www.wrap.org.uk/downloads/FINAL_Landfill_Bans_Feasibility_Research.f5cf24f9.8796.pdf

Question 34: Street recycling: Tonnes of material collected for recycling at street

recycling bins.

Question 23: Details of other wastes collected for disposal (residual waste not

collected for recycling) - Collected household waste: Street Cleaning

111

Assembly Government study.

118

Additional assumptions were required to estimate

the likely quantities of beverage containers in each waste fraction. These

assumptions are as follows:

¾ 75% of the 'Packaging glass' fraction are Glass Bottles;

¾ 50% of the 'Dense plastic bottles' fraction are PET Bottles ;

¾ 20% of the 'Ferrous food and beverage cans' fraction are Steel Beverage

Cans; and

¾ 60% of the ‘Non-ferrous food and beverage cans’ fraction are Aluminium

Beverage Cans.

119

See Table A-9 for estimates of the tonnages of beverage containers from on-the-go

recycling and street sweepings.

A.2.2.5 Total Products Placed on the Market / Total Waste Arisings / Containers

Remaining in Environment

In order to estimate the quantity of beverage containers left in the environment as

litter, we have combined a top down approach with the ‘bottom up’ approaches

considered in the Sections above. Our simple approach estimates that:

*note there will also be a small amount of re-use for other purposes (eg. home-brewing, plant pots, re-

using plastic bottles). The amount of re-use is unknown, but we would expect it to be relatively small

compared to the recycling and disposal figures used in this study.

The figure for ‘Total Containers Placed on Market’ is the top down approach,

calculated from known estimates for the total number of containers placed on the

market and the average weight of a container. Canadean® (beverage industry

information specialists) supplied us with data pertaining to the quantities of

different beverages, by container type, placed on the UK market in 2009.

The weights of beverage containers were determined through actual measurements

of sample products, from existing studies, and from websites such as The

Environmental Register of Packaging PYR Ltd.

120,121

The average weights for

different container types used in the study are given in Table A-21. This data allowed

118

AEA Technology, MEL Research, Waste Research and WRc (2003) The Composition of Municipal

Solid Waste in Wales, Report to the Welsh Assembly Government, December 2003

119

More food containers (soup tins etc) are made from ferrous metals, hence why the proportions of

beverage containers are higher for the non-ferrous fraction.

120

WRAP (2008) Bulk Shipping of Wine and its Implications for Product Quality, Final Report:

GlassRite: Wine, May 2008,

http://www.wrap.org.uk/downloads/Bulk_shipping_wine_quality_May_08.1be9881a.5386.pdf

121

PYR (n/a) Packaging Weight Units, Accessed 1

May 2010,

http://www.pyr.fi/eng/forms/packaging-date-questionaire/packaging-weight-units.html

Containers Remaining in the

Environment as Litter

Total Containers Placed on Market –

Containers Captured for Recycling and

Treatment / Disposal*

112

us to calculate the total weight of containers placed on the UK market every year. By

subtracting the tonnages of beverage container waste collected for recycling and

disposal from the total weight of containers placed on the UK market, the resultant

figure indicates the tonnage of beverage containers that would remain in the

environment as litter. The overall number and tonnages of containers placed on the

market and the amount of containers found in the environment are given in Table A-

Table A-21: Average Weight per Container Type

Container Kgs

Soft/Beer & Cider Bottles 0.3

Wine bottles 0.5

PET Bottles 0.033

Ferrous Cans 0.035

Aluminium Cans 0.017

Source: Eunomia

The weight and number of containers predicted by our bottom up analysis was

greater than that estimated using the Canadean® data. This is due to the fact that

the Canadean data does not include imports from ‘private trade’ with other EU

countries (such as France and Ireland). The volume of beer, and wine, for example is

considered to be significant (see Section 6.6 for more detail).

A.2.2.6 Summary Baseline Figures

Table A-9 shows the mass flow baseline upon which subsequent calculations were

undertaken on the costs and benefits associated with the introduction of a deposit

system. Due to the high-level nature of this study, a full analysis of the ranges and

uncertainties in the modelling could not be accomplished. However, we believe the

estimates provided in Table A-9 to be reasonable, being, as they are, based on

reasoned argument, and rationalised to the greatest extent possible. Furthermore,

the tonnages were cross-checked with packaging data available from various

sources.

122,123

The figures were within acceptable error margins, especially when

considering data in the waste sector (often of low quality).

From the figures provided in Table A-9, it can be seen that:

¾ A significant quantity of containers will be sold in the UK every year (around

28 billion);

122

David Davies Associates (2009) PackFlow 2012: UK Compliance with the European Packaging &

Packaging Waste Directive, Volume 1: Summary Report & Recommendations, November 2009.

123

Advisory Committee on Packaging (2008) Packaging in Perspective, November 2008.

113

¾ The implied commercial recycling rate is high. Note, however, that the

baseline is modelled based on 2015, when the landfill tax has escalated to

£80 per tonne, which would promote the uptake of commercial recycling

services above and beyond the levels seen today; and

¾ A significant quantity of waste is left in the environment every year (~300

million bottles / cans etc). It is important to re-iterate the limitations of this

study in estimating the amount of beverage container litter that is present in

the environment. Unfortunately, we could find no studies or research that

have previously tried to estimate this figure with which to compare our

estimates. Our modelling does make an estimate of the potential disamenity

benefit possible from the removal of beverage container litter from the

environment.

The overall recovery rate for the containers in scope in this study, under the baseline

system (pre-DRS), is calculated at 68% (this is significantly higher than what is

currently being achieved, being based around a baseline where it is assumed that a

comprehensive set of kerbside recycling services has been rolled out across all local

authorities). There is, therefore, scope to increase the environmental benefits

associated with greater recovery of these materials.

A.2.3 Scenarios

This section describes the two central scenarios that have been modelled for the

introduction of a deposit refund system in the UK. These scenarios are as follows:

The scenarios result in changes in mass flow compared to the baseline described in

Appendix

A.2.2. To determine the magnitude of the change, we estimated the likely

situation following implementation of the DR system, and then calculated the

difference compared to the baseline.

Note that we have assumed the total quantity of containers placed on the market

remains constant across the scenarios. In reality the quantity may fall slightly as the

price of the container increases, though discussion with Canadean

suggests that an

increase in cost of 1-2p (ie. the admin charge calculated from the deposit refund

model) would be unlikely to instigate any significant changes in demand, with

demand being primarily driven by other factors such as the weather (determining

how thirsty we are) and promotional deals.

124

Demand is generally believed to be

relatively inelastic.

124

Personal communication with Canadean®.

1. Complementary – where no beverage containers are collected at the

kerbside i.e. the deposit refund system is complementary to the

existing kerbside schemes; and

2. Parallel – where the household kerbside systems for beverage

containers operate in parallel to the deposit refund system.

114

In both cases the mass flows were adjusted so that the overall return rate for the

deposit refund system was set at reasonable levels. The rationale for the likely

return rates follows.

115

Table A-9: Mass Flow Baseline for CBA Modelling

No. of

Containers

(millions)

Tonnages (thousand tonnes)

hhld Kerbside Bring HWRCs Commercial Litter Products

Placed on

Market

Placed

Market

Recycling

Refuse

Recycling

Refuse

Recycling

Refuse

Recycling

Refuse Environment

Glass

Bottles

5,905 2,288

1,002

312

234

505

68 23

PET

Bottles

8,846 314

164

37 3

Cans

(Fe.)

5,717 200

0.1

18 2

Cans

(Al.)

7,271 124

0.1

26 1

Total 27,740 2,926

1,152

605

249

562

114

148 29

116

We start with the maximum return rate likely from a standalone system (scenario 1

– the ‘complementary’ system). Through looking at existing systems, it is estimated

that a likely return rate for all beverage containers would be 90%.

125,126,127,128

This

is by no means the maximum (or minimum) value, but likely to be a reasonable

estimate, especially if containers cannot be collected at the kerbside.

For the parallel system (scenario 2), consumers whose behaviour is not influenced

by the loss of deposits will continue to leave containers within the DR system in

kerbside boxes or in their residual waste bin. It could be argued that those for whom

the deposit is valuable might remove the containers from kerbside boxes and bins.

This could also include kerbside collection crews / MRF operators. However, the

capture via these methods will be unlikely to be as high as through the deposit

refund system itself. We therefore assume that the overall return rate for containers

in the system will be less than the ‘complementary’ scenario, and we estimate the

return rate might be 80% under the ‘parallel’ scenario.

The general approach to determining the waste flows for the two scenarios is given

in the sections below. This is followed by summary tables showing, in quantitative

terms, the effects of the two scenarios.

A.2.3.1 Scenario 1 (Complementary System) Waste Flow Principles

The general principles used to estimate the likely future waste flows as a result of a

complementary deposit refund system are:

¾ There is a 100% reduction in the quantity of beverage containers collected for

recycling through household kerbside collection systems;

¾ The quantity of containers in refuse will not be zero, as some people simply

will not take the containers back to a collection point. We have assumed 5%

of containers remain in household refuse (clearly this is an unknown factor,

no data supports this figures specifically. But if return rates of >90% are

possible, and some containers, of which the householder is the greatest

generator, are disposed of, 5% disposal in household refuse seems a

reasonable estimate. The main point being that this activity would occur so it

is important to make some estimation of if in the modelling work); and

¾ It is assumed that the same percentage reduction is applied to the number of

containers collected in each waste management route (other than at the

kerbside) in order to achieve an overall return rate in the deposit refund

system of 90%.

125

http://www.dansk-retursystem.dk/

126

http://www.palpa.fi/

127

http://www.resirk.no/Frontpage-63.aspx

128

Eunomia et al. (2009) International Review of Waste Management Policy: Annexes to Main

Report, Report for Department of the Environment, Heritage and Local Government, Ireland,

September 2009.

117

A.2.3.2 Scenario 2 (Parallel System) Waste Flow Principles

The general principles used to estimate the likely future waste flows as a result of a

parallel deposit refund system are:

¾ There is an 80% reduction in the quantity of beverage containers collected

through household kerbside collection systems

¾ It is assumed that the same percentage reduction is applied to the number of

containers collected in each waste management route (other than at the

kerbside) in order to achieve an overall return rate in the deposit refund

system of 80%.

In reality, it might be that there is a larger reduction in the number of containers

collected through particular waste management routes as opposed to others. For

instance, it would be easier for an individual to pick containers out of litter bins or

the environment than from bring banks or commercial waste routes. Hence it might

be expected that fewer beverage containers would be found in litter bins and the

environment than in bring banks or commercial waste routes. However, given the

lack of evidence to support this theory, we have modelled the same reduction in

beverage containers for each management route.

A.2.3.3 Summary Scenario Waste Flows

Table A-23 shows the change in waste mass flows as a result of the implementation

of a complementary or parallel deposit refund system in the UK.

As illustrated in Table A-23 the magnitude of the shift from existing collection routes

increases under the complementary scenario – this is due to the fact that the

possibility of collection from household kerbside collection schemes is removed.

118

Table A-23: Change in Mass Flows Resulting From Introduction of Complementary and Parallel Deposit Refund Systems

No. of

Containers

Tonnages (thousand tonnes)

hhld Kerbside

Bring

HWRCs

Commercial

Litter

Products

Placed on

Market

otal

Arisings

Recyclin

Refuse

Recyclin

Refuse

Recycling

Refuse

Recyclin

Refuse

Env.

via DR

System

Complementary Deposit System (Scenario 2)

Glass

Bottles

0 0

-805

-249

-183

-26 -4

-394

-69

-13

-53

-18 1,814

PET Bottles

0 0

-57

-146

-4

-3 -1

-12

-11

-1

-29

-2 265

Cans (Fe.)

0 0

-40

-74

-6

-2 0

-25

-6

-2

-14

-2 170

Cans (Al.)

0 0

-24

-39

-1

-1 0

-8

-3

-20

-1 98

Parallel Deposit System (Scenario 1)

Glass

Bottles

0 0

-1,002

-246

-192

-27 -4

-414

-73

-14

-56

-19 2,047

PET Bottles

0 0

-71

-152

-4

-3 -1

-12

-11

-1

-30

-3 288

Cans (Fe.)

0 0

-45

-77

-6

-2 0

-26

-6

-2

-15

-2 181

Cans (Al.)

0 0

-34

-42

-1

-1 0

-8

-3

-21

-1 112

119

A.3.0 The Deposit Refund System Model

The various stakeholders in an operating deposit refund system are:

¾ A government body authorising the system and associated finances, and

setting recycling targets for the various materials;

¾ A central organisation owned and run (within the constraints set by the

authorising body) by, for example, non-governmental organisations (NGOs),

industry bodies, producers, breweries and retailers;

¾ The manufacturers of containers, producers of beverages and industries that

‘fill’ the containers;

¾ Any retailer which sell beverages in the UK;

¾ All consumers which purchase beverages in the UK; and

¾ Businesses and organisations involved with the collection, sorting and

reprocessing of waste containers.

Various stakeholders are involved in the material flows of beverages (pre and post-

consumption), deposit payments, other finances and sales or container return data.

An overview of the key elements, material and finance flows, in the UK’s deposit

refund system model developed for this study is given in Figure A-7-3.

The system developed for this study is based on similar principles (though the

details reflect the UK’s structure of retailing) to the systems which exist in Denmark

(Dansk Retursystem) the Scandinavian countries (Norsk Resirk, Returpack and

Palpa), and in a number of provinces within Canada (ENCORP Atlantic Ltd, ENCORP

Pacific Inc). The operation of the system is described in the following points: