What do Scots toss?

What are people most likely to litter? This is a surprisingly tricky question to answer, as the data on litter is poor. However, if we want to have effective policies to reduce littering, it’s important to understand the behaviours we need to change, and the scale of the impact they could achieve.

While we have some helpful studies on the composition of litter, most of the associated discussion is about the number of items versus the volume of items in each snapshot. As I’ve discussed elsewhere, focussing attention on the number of litter items tends to mean counting (and re-counting) small, hard to remove litter items like cigarette butts and chewing gum, while looking at volume highlights the contribution of larger items like beverage containers to the litter problem.

This is useful, up to a point: we know, for example, that in midwinter in Scotland, when outdoor consumption of drinks is probably at its lowest, beverage containers account for 6.4% of litter (by item count) But what does this actually mean in terms of consumer behaviour?

Fit to drop

To answer this, we need to raise a question that is rarely asked:  what can the available information tell us about the rate of littering of specific types of item? By bringing together various sources we can begin to get an idea.

A good starting point is a Zero Waste Scotland (ZWS) report from 2013 which presents the composition (by weight) of litter in Scotland. This identifies the following proportions:

• Plastic bottles – 8.6%
• Packaging glass – 9.1%
• Metal cans – 4.0%

The ZWS report also puts a figure on the annual tonnage of litter dropped on the ground and subsequently cleared by local authorities in Scotland each year – at least 15,000 tonnes. This is a conservative figure as it excludes:

• litter dropped, and then cleared, on other public land (e.g. hospitals, schools and the transport network) or private land (e.g. stadiums and shopping centres);
• litter that is not picked up, and that either (a) accumulates over a long period of time – although, in due course, much of this might eventually be cleared – or (b) ends up being washed into water courses and ultimately to the sea; and
• litter that is correctly discarded in litter bins.

However, if we take 15,000 tonnes as an estimate of litter dropped and cleared each year in Scotland, the following tonnages can be calculated for specific littered items:

• Plastic bottles – 1,290 tonnes
• Packaging glass – 1,365 tonnes
• Metal cans – 600 tonnes

Weights and measures

To understand what this means in terms of an item-specific littering rate we need to know what is placed on the market in Scotland annually. Eunomia’s 2015 report for ZWS on the key design features and feasibility of a potential Scottish Deposit Refund Scheme (DRS), and the appendix to the report, provide such figures:

• Plastic bottles – 39,000 tonnes
• Glass bottles – 165,000 tonnes
• Metal cans – 14,100 tonnes

On this basis, the following item specific littering rates, i.e. proportion of all items purchased that are littered, can be calculated:

• Plastic bottles – 3.3%
• Glass bottles – 0.8%
• Metal cans – 4.3%

However, alternative figures were put forward by Valpak for the number of items placed on the market each year. These figures were lower, at:

• Plastic bottles – 36,000 tonnes
• Glass bottles – 127,000 tonnes
• Metal cans – 9,000 tonnes

This, of course, would imply higher littering rates of:

• Plastic bottles – 3.6%
• Glass bottles – 1.1%
• Metal cans – 6.7%

We can further refine our estimates if we note that plastic bottles of two different polymer types are placed on the market. PET bottles are widely used for water and fizzy drinks, while HDPE is more commonly used for milk and household products like shampoo. While the litter composition data does not distinguish by type of plastic, HDPE bottles are much less likely to be littered given that their contents are predominantly consumed at home.

Figures in the appendix to Eunomia’s DRS report show that approximately 60% by weight of the plastic bottles placed on the market is PET. Using a lower figure of circa 21,500 tonnes (60% of 36,000 tonnes) of PET placed on the market, and assuming that almost all littered plastic bottles are PET, the implied littering rate for PET increases to 6%.

Litter bearers

Of course, this is all based on a very conservative estimate of the amount littered each year in Scotland of just 15,000 tonnes. I would imagine that the true figure is likely to be at least several thousand tonnes higher, meaning that the actual littering rates would be greater still.

The wrong kind of deposit… We need to find ways to tackle frequently littered items. Photo courtesy of Chris Sherrington.

While better data is needed to understand more completely where the actual littering rates lie, for me, the relative ordering for the different item types feels about right:

• Littering of glass, by weight, is relatively low, reflecting the propensity for fewer ‘on the go’ beverages to be served in large, heavy glass containers; rather such containers tend to be used for beverages consumed at home or in pubs or restaurants.
• Plastic bottles are widely used in ‘on the go’ consumption, but they can be refilled, and so may be kept for reuse; or failing that, they can be compressed, and securely closed with a screw cap to prevent any dregs spilling out. That makes them relatively convenient to put in a bag or pocket and take home or to the nearest litter bin.
• Cans, however, have no such potential for a second use, and cannot be resealed. Putting them in a bag to take home is thus a less attractive prospect, and the incentive to just be rid of them is probably greater than for plastic bottles, which fits with the higher apparent rate of littering.

Rate of return

So what does this mean in terms of the number of items littered – not captured in litter bins, but genuinely littered. The appendix to Eunomia’s DRS feasibility report indicates that:

• 744 million PET bottles are placed on the market in Scotland each year. Applying a deliberately conservative assumption of just 3% being littered (less than I calculate above) this would be over 60,000 littered PET bottles per day, every day.
• 148 million ferrous cans and 526 million aluminium cans are placed on the market in Scotland each year. Applying a littering rate of 4% (again a conservative figure, lower than any calculated above) would mean over 70,000 cans littered in Scotland daily.

While each type of container may have its own specific characteristics that can affect the propensity for the item to be littered, evidence from overseas indicates that such behaviour can be changed dramatically with the introduction of a deposit refund scheme. The promise of a refund attached to the used beverage container brings about a significant shift in perception of the item from something with a negative, to a positive value. A 2005 peer review of a DRS study for Defra, which, in general terms, was not supportive of introducing a DRS, highlighted examples from the United States where reductions in beverage container litter in excess of 80% occurred once a DRS was implemented.

Assuming, in line with findings from the US, that a Scottish DRS reduces littering by 80%, then it would mean that  50,000 fewer PET bottles and 60,000 fewer drinks cans would be littered each day.

As well as a cleaner environment, this would bring carbon savings. The best result we can expect for a littered plastic bottle is that it is picked up and enters the residual waste stream, much of which is now destined for incineration. Recycling an extra 50,000 PET bottles per day instead of incinerating them would save in the order of 1,000 tonnes of CO₂ equivalent each year, according to the Scottish carbon metric.

The data on litter may be poor, but that needn’t prevent us from making good decisions as to how it should be tackled.