Breaking down the Danish study on the environmental impacts of grocery carrier bags

Parkpoom Kometsopha
10 min readApr 18, 2019
Image from:

Last weekend, I came across several viral news articles on Facebook discussing the environmental impacts of grocery carrier bags. They reported that organic cotton bags are much worse than plastic bags and that conventional cotton bags are better than organic cotton ones. All of them had those classic controversial headlines: “Your cotton tote is pretty much the worst replacement for a plastic bag” or “Bad news! Cotton bags have more impacts than plastic bags!!” and so on.

At first, I didn’t want to bother much about them since these articles tend to use invalid references (at least in my experience). However, when I saw that all of their sources came from a study published last year by the government of Denmark — a country that is seriously committed to sustainability — they certainly got my attention.

About the study

This particular study is about Life Cycle Assessment (LCA) of grocery carrier bags in Denmark, which assesses the environmental impacts of various carrier bag types used in the country and later compares the results between them. It considers the whole life-cycle of the products, starting from the raw material acquisition until their end-of-life.

Accordingly, one of its conclusions was that the conventional and organic cotton bags must be used at least 7,100 and 20,000 times respectively in order to meet the environmental performance of LDPE plastic bags. These figures were actually quite surprising, even for me; I know cotton bags need to be reused multiple times before they have lower impacts than plastic bags, but I did not expect the numbers to be THAT much. All of the articles I read so far presented both numbers.

However, was that conclusion accurate? Can cotton bags have such tremendous impacts over plastic bags?

After diving into the research paper, I’d say yes — AND no.

Yes, because the result of the study was presented indeed according to its methodological framework with full data disclosure. The authors also briefly discussed the limitations of their research method as well as the alternative result under a different assumption and scenario.

No, because I believe the framework and the result put the cotton bags at a disadvantage, especially the organic bag, and the authors failed to mention it. It would be difficult for readers without basic knowledge of LCA to make a correct interpretation of the research’s method and the result — hence, these controversial articles.

Breaking down the research

Therefore, in this article, I am going to break down this scientific paper for you in a digestible way so that you are better informed about the study. Additionally, I will also point out issues which the online articles failed to capture, and where the research authors could have engaged in further discussion in order to clarify the assessment result of the organic cotton bag.

Things are about to get pretty nerdy here. I will try to keep the jargon to a minimum, so bear with me.

To begin with, when you want to conduct an LCA study of any product, one of the first necessary steps you need to do is to specify the product’s characteristics. These are, for example, the purpose of use, weight, height, capacity, use frequency or length, and so on. No matter what type of material or shape it has, the studied product(s) must meet the required specifications. Otherwise, it cannot be included in the study.

For this research, the specification of the studied grocery carrier bags was defined as follows:

Carrying one-time grocery shopping with an average volume of 22 liters and with an average weight of 12 kilograms from Danish supermarkets to homes in 2017 with a (newly purchased) carrier bag. The carrier bag is produced in Europe and distributed to Danish supermarkets. After use, the carrier bag is collected by the Danish waste management system

Fourteen common types of carrier bags (in Denmark) were chosen for the study. Most of them were made from various plastics, while the rest were from non-plastic materials like paper, textile (cotton), and composite. To determine the average characteristics of each carrier bag type, the research authors conducted a survey measuring the average weight, volume, thickness, and weight holding capacity. The table below shows the result of the survey.

List of studied products and their average characteristics according to the survey

The problem of the research’s specifications and assumption

Based on the above result, you might notice that not all the carrier bags had their (average) characteristics fit the defined specifications. Some of them either did not have enough average volume and/or weight holding capacity. These are, for instance, the recycled Low-Density Poly Ethylene (LDPE) plastic bag, bioplastic bag, organic cotton bag, etc.

To solve the problem, the research group assumed that the consumer would buy two of those bags instead of one in order to have the same functionality — that is, to do grocery shopping at the volume of 22 liters and/or a weight of 12 kilograms. As a result, the impact assessment result of these bags was presented as the amount of two bags instead of one. (See the table below.)

Reference flow of each carrier bag

This is the first issue for which the articles failed to elaborate.

No matter how probable or logical the assumption might be — although in my opinion, it is unlikely — it puts the bags that were assessed as two bags into a less favorable situation. The organic cotton bag is obviously the most disadvantaged because its volume missed the specification by only 2 liters, even though it can hold weight 5 times heavier than the LDPE bag used as a reference for comparison. (Not to mention the fact that only one organic cotton bag was collected in the survey.)

If the specifications of the bag were changed — let’s say the required volume became 20 liters while the required weight carrying capacity increased to 15 kg — the organic cotton bag would consequently have been assessed using one unit (instead of two) while the referenced LDPE bag would have been assessed using the amount of two bags under the same assumption. Thus, the impact assessment result for the LDPE bag could have been two times higher than the presented result, and the minimum amount of organic cotton bags used in order to meet the same environmental performance of the plastic bag could have been FOUR times less (2x2).

Similarly, the research’s specification also gives an advantage to the conventional cotton bag over the organic one because it meets the requirements within one bag already. As a consequence, the impacts of the conventional cotton bag presented in the result are halved, at a minimum.

Which impact category is really responsible for the high numbers in cotton bags?

Speaking of impacts, this research assessed the environmental impact of the carrier bags across 15 impact categories in total. It also compares the result of each bag with the reference LDPE bag to identify how many times each of them must be used in order to meet the same environmental performance as the reference bag. If the amount of use is positive, the particular bag has a higher impact on that impact category (and less impact, if negative.)

Environmental impact categories and methods used in the assessment

Accordingly, the result shows that the organic cotton carrier bag has much higher impacts than the reference LDPE bag when taking all the environmental impact categories into account — it has to be used at least 20,000 times.

This is where things get complicated.

When the research authors said ‘all categories’, they were actually referring to the highest number of reuse times among those calculated for each impact category. In this case, for both organic and conventional cotton bags, the crazy numbers of reuse times (20,000 & 7100) were given by the ozone depletion impact alone. If we exclude that impact, “the number of reuse times [would only range] from 50 to1400 for the conventional cotton, and from 150 to 3800 for the organic cotton” (see p.18’s footnote in the paper).

For those who never heard of ozone depletion, it is a gradual thinning of the ozone layer in the Earth’s upper atmosphere caused by the release of certain chemical compounds from human-related activities. The loss of the layer would allow an unhealthy amount of ultraviolet (UV) radiation to pass through to the Earth’s surface, increasing health risks such as skin cancer and eye cataracts.

Number of (primary) use for the carrier bags necessary to provide the same environmental performance of the average LDPE carrier bags (which is the reference bag)

What causes ozone depletion in the cotton carrier bag?

Unfortunately, the research authors did not talk about this in their paper (even though they should have). So, allow me to enlighten you.

The major cause for ozone depletion in cotton carrier bags comes from a particular activity in cotton production — and that is the electricity consumption for irrigation. In this research, electricity was assumed to be used in cotton irrigation process because the research author used a dataset of cotton production which has the particular input (‘Market for textile, woven cotton; GLO’ in Ecoinvent v3.4 database).

Accordingly, much of the electricity is assumed to be produced by natural gas. During the transportation of the gas, two fire-suppressing and cooling gases which are known as Halon 1211 & 1311 are used in the gas pipeline distribution system. These gases have very high potential in ozone depletion; their impact is 5–12 times higher than the common ozone-depletion gas called CFC-11. Since a large amount of cotton fibers is needed to produce the cotton bag (250 gram of cotton bag vs. 24 gram of LDPE bag based on the paper), it isn’t that surprising to see the impact of ozone depletion in the cotton bag rate much higher than the reference LDPE bag.

NOTE: We also shouldn’t forget the first issue that I mentioned regarding the large impact figures in the cotton bags presented in the paper, as part of it was heightened by the specifications and assumptions made in this research. This figure could have been more than halved if cotton bags were assessed using the same unit as the LDPE bag (one unit).

But is ozone depletion an issue we should be prioritizing here?

Another important issue which was ignored by the news article and the research paper itself is about the level of significance between different impacts.

Not all impacts of cotton bags require a tremendous amount of use as ozone depletion. For instance, in regard to climate change, the organic cotton bag has to be used only 150 times before it meets the same environmental performance as the LDPE bag. The number is much smaller than the figure in ozone depletion — 20,000 times. It would take me only 1.5 years to beat the LDPE bag using the organic cotton bag in the climate change category — assuming that I do grocery shopping twice a week. And if I were to use the cotton bag on other occasions that would also replace the use of LDPE plastic bags, it would take even less time than that. (But this is not within the scope of the research paper so I will not bring it up.)

Thus, the main question is which impact category should we prioritize over another?

Although there is no real and legitimate answer to the question yet — like seriously, many scientists have been working together to crack this question out — it’s not that difficult for me to convince you that ozone depletion is definitely not the most concerning impact.

In the past (around 1970), ozone depletion was indeed a severe problem when we found out that the immense release of (mostly) CFC-11 gas blew up a large hole in the ozone layer over Antarctica. However, since 1987, when the world adopted the Montreal Protocol, which is an international agreement to phase out from using CFCs, the ozone layer has been recovering.

Planetary boundaries framework (Steffen et al., 2015)

Furthermore, if we look at the study of planetary boundaries which assesses the condition of nine different earth systems that are critical to the preservation of the Earth’s current state — where mankind can live comfortably on the planet (unlike the Ice Age, for example) — we can see that the Earth is being ravaged mostly by the impacts of biodiversity loss and land degradation, imbalanced biogeochemical flows, and climate change. These impacts are currently much more alarming than ozone depletion.

More importantly, the ozone depletion issue from cotton production can be easily mitigated. Since the impact was caused by the use of electricity produced from natural gas in crop irrigation, procuring cotton fibers from sources that do not use electricity, or use electricity from renewable sources, can substantially lower the impact. As a result, the environmental performance of the cotton bags would become much more competitive to the LDPE bag.

On the contrary, there are other impacts from plastic bags which were not mentioned in the study and which are much more difficult to solve — for example, the microplastic leakage and the ingestion of plastic materials by marine life. As far as I know, these impacts cannot yet be assessed in LCA, and it is unfair that the research group did not mention this limitation in their paper even though the impacts are being considered widely by the public today.

And thus concludes my nerdiest writing piece (ever). I hope that the information I provided here has broadened your understanding of the LCA of plastic and cotton bags based on the research from the Danish government.

Last but not least, I would like to say that despite my criticism of the online articles I highlighted, I still share one conclusion with them. Reusable bags, no matter what kind, are indeed better off than the single-use ones. To be sure of that, we all must reuse our bags again and again for as long as possible.



Parkpoom Kometsopha

I am a sustainability specialist and a rock climber. I sometimes write stuff.