Calculation of footprint for goods and services

Calculate a personal footprint for goods and services consumption, from shopping and maintenance habits. This can be used to nudge people towards more sustainable consumption practices.

Calculations

Income and expenditures

The consumption module calculates the climate emissions produced by goods and services unrelated to food, energy and transport. We calculate the average household’s consumption based on the household’s income, the idea being that the more you earn, the more income you have available to spend on goods and services. Again, this number changes based on the composition of your household. For instance, if there are three children in your household, the money it takes to feed them will mean that less money is allocated to consumption.

We use Statistics Norway’s Survey of Consumer Expenditure and the UK’s Office for National Statistics Family Spending Report to determine the average proportion that a household spends in 50 different consumption categories (based on COICOP). These categories include everything from hospital expenses to sporting equipment, and like the total consumption footprint, these vary based on the composition of your household. If you have no kids, you’re unlikely to spend money on primary school fees, so that gets allocated elsewhere.

The spending amounts are calibrated to the household income, and money spent on energy, transport and food are removed as these categories are considered from a bottom-up perspective. Tax has been removed from the income based on general tax tables, as emissions from tax money are covered by the public emissions. Any donations you make during the year are also removed from your consumption footprint. It’s true that those donations might be used for emissions-intensive activities elsewhere, however those emissions are allocated to the organisation or individual you’re donating to.

We make the assumption that all money you earn is spent in some way. You might put a lot into savings, but our calculator assumes that money will still be spent somewhere down the line on housing, tuition fees, or a different long-term savings goal. We’re currently working on improving our model for calculating the footprint of long term investments like housing.

The date also influences the calculations, and the multipliers used in calculations are updated monthly. We account for inflationary effects, fluctuating prices on certain goods, temporal effects (higher energy prices in winter),  and changing technology over time,  and their impact on your footprint. For example, as manufacturers decarbonize their supply chains, the emissions intensity of their products will decrease. Hence, if you buy the same product five years from now, the footprint of your purchase will be lower in the future.

Input-Out Calculations

We mentioned earlier in this document that household consumption is calculated using a top-down approach, as opposed to the bottom-up approach which is used for food, energy and transport. The top-down approach we use is better known as Input-Output modelling. You can read more details about how these calculations are made in Steen-Olsen et al. (2016). We use separate input-output databases for Norway and the UK. The information regarding the databases and the modelling choices used can be found in the intro to calculator methodology.

This approach doesn’t add up all your spending and calculate your climate emissions based on the combined emissions of that spending. Instead it takes the total emissions associated with each activity in the economy, and then decides what your share of those emissions is based on your income. For example, in Norway the average emissions for clothing are 44g CO2e per NOK spent, so your emissions for clothing are multiplied by the amount your household would normally spend on clothing. The sources for obtaining the average spending patterns are highlighted above.

On top of these consumption calculations, there are several personal habits which can affect the footprint of the user.

General habits related to purchase of goods & services

Quality and repair consumer

The main assumption here is that by purchasing a higher quality product, the user replaces the need to buy a similar product in a relevant time frame. Higher quality products last longer, so the CO2 multipliers are changed accordingly to reflect that. As an example, if you use an object 20% longer than its expected lifetime, you save climate change emissions equal to 20% of production emissions of that object. Analogous to higher quality purchases, we assume that repairing goods gives them a longer life span, thereby reducing the CO2  multiplier associated with goods spending by up to 20%.

Ethical consumer

For this habit it is assumed that the user tends to spend more money on ethical and environmentally friendly purchases. As environmentally friendly purchases are more expensive compared to their counterparts, less money is available to spend on goods & services, thereby reducing the footprint. Additionally, the CO2 multipliers for goods and services are modified to reflect environmentally friendly choices.

Service consumer

For this habit, it is assumed that extra money is spent on services and therefore, less money is available to spend on goods. If more money is spent on services, such as renting a movie online or using a subscription service as opposed to buying DVDs, the emissions related to manufacturing of goods decreases, and overall consumption emissions decrease.

Recycling

This habit gives insight on how much a consumer recycles. We have calculated the maximum amount possible to save by optimal recycling (approximately 100 kg/year), and found the percentage of the average footprint. Then used this to create an across the board value that modifies the CO2 multiplier in all goods and services categories based on the input for this habit.

Habits related to clothing

The IPCC estimates that the global clothing industry is responsible for about 10% of emissions globally. Since clothing has such a significant impact on global emissions and a big influence on a person’s footprint , we decided to include questions targeted specifically at understanding a user’s clothing consumption habits.

Avoided emissions

The concept of avoided emissions is when an action leads to less emissions through e.g. an item not having to be produced. In our model, we attribute avoided (or negative) emissions to a user’s footprint in the case of repairing, donating, renting  or buying used clothing. The avoided emissions are calculated using the assumptions related to replacement rate as discussed in Convert Goods & services. We allocate avoided emissions between the donator and purchaser for donating and buying used. In our model, we give more credit to a person buying used clothing than donating clothes since donating clothes has a lower probability of avoiding emissions. This is based on the work of Farrant et al. (2010), who concluded that only 20% of donated clothing was resold.

Buying high quality clothing

This habit relates to how often a person buys durable, high quality clothing, such as items that have a lifetime guarantee. The footprint per unit of money spent on buying high quality clothing is lower, because the clothing is slightly more expensive to be durable. The total spending on clothing is kept constant, but the person who is buying more quality buys fewer total items and therefore the clothing footprint is lower.

Clothing donations

This habit relates to how often a user donates clothes, with the options never, once per year, once per season or once per month. Each of these options is equated to donating a fraction of the total clothing purchases. Based on the findings of Farrant et al, 2010, we assume that a donated item of clothing has a 60% chance of displacing the purchase of a new piece of clothing. The donor receives 20% of the credit for displaced CO2e of this used clothing purchase. Using the above assumptions, we calculate the avoided emissions due to donated clothing, and subtract it from the total footprint.

Fast fashion clothing purchases

This habit relates to how often a user purchases clothing from fast fashion companies, that is brands that focus on low-cost, high-throughput options. We estimate that 20% of clothing purchases are from fast fashion companies, on average. We further assume that fast fashion options cost half as much as regular fashion, and that the total spending doesn’t depend on the amount of fast fashion purchased. Since the footprint of producing an item of clothing is about  the same for fast and regular fashion options, the multiplier in gCO2e per currency spent will be twice as large  for the fast fashion option. Based on a user’s input of what fraction of their clothing is fast fashion, we can thus adjust the multiplier and the resulting footprint of clothing shopping. 

Washing clothing correctly

This habit relates to whether a user reads the washing instructions for garments and washes them appropriately. Moran et al. (2020) found that the lifetime of a garment can be extended by up to 5% by correct washing; we assume that you buy fewer items of clothing and spend less money if you wash your garments correctly. We further assume that money saved on clothing is respent in service categories, which have a lower impact. This results in a lower footprint in the clothing category.

Clothing shopping amount

This habit relates to how many items of clothing a user buys per year. Data from Laitala et al. (2020) shows that the average person buys about 36 items of clothing per year (the average of the amount reported by survey and that from the import statistics, since they give different numbers). Depending on whether a person buys fewer or more items of clothing than average, the spending is readjusted for all consumption categories.  Buying more items of clothing would equate to a higher spending on clothing and lower spending in other categories, and vice versa.

Clothing as a service

This habit relates to the frequency with which a user rents clothes rather than buying them, and is specified in the number of garments rented per year. We assume that renting clothes is cheaper and that rented clothes have a 60% probability of displacing the purchase of a new item.  As a result, having more rented clothes in your wardrobe leads to a lower spending on clothing, and a lower footprint since rental clothes have a lower CO2 multiplier.  The avoided emissions from buying less clothing as a consequence of renting is also accounted for and subtracted from the total footprint.

Buying used clothing

This habit relates to the number of items of clothes a user buys second hand in a year. We assume that the footprint of secondhand clothing is much lower compared to newly purchased clothing and that second hand clothes are cheaper.  Based on the work of Farrant et al. (2010), we assume that secondhand clothing has a 60% chance of displacing the purchase of a new piece of clothing.  The buyer receives 80% of the credit for displaced CO2e. As a result, having more secondhand clothes in your wardrobe leads to a lower spending on clothing which is offset by increased spending in the less CO2 intensive buying-used category . The avoided emissions from buying less new clothing  is also accounted for and subtracted from the total footprint.

Textile reuse

This habit relates to the frequency with which a user reuses their clothes for other purposes (for e.g. an old t-shirt as a dust cloth), with the options Never, I love new textiles, like an average person, occasionally or I actively try to. Each of these equates to a certain percentage of clothing being reused as textiles. Farrant et al. (2006) concluded that by reusing clothes as textiles, one could reduce spending on textiles by 5%.  We assume that by reusing clothes as textiles, the user buys fewer new textiles, leading to lower spending and emissions in the textile category.

Repairing clothing

This habit relates to the number pieces of clothing  a user repairs in a year. We assume that repairing an item reduces the probability of buying a new one and that less money is spent on clothing as repairing a garment is cheaper than buying a new one. Based on average repair weighted repair price using price data from Repairable and repair frequency data from S. Privett (2018), we calculate the expected spending on repairs based on the user’s input. This is allocated to a clothing repair category, which has a low CO2 intensity. Depending on the number of garments you repair, spending is readjusted from the clothing category to the less CO2 intensive repair category, resulting in a lower footprint. The avoided emissions due to buying less clothes as a consequence of repairing is also accounted for and subtracted from the total footprint.