The influence of having an environmental orientation on crowdfunding success in fashion industry projects
Bachelor’s Thesis (6013b0505y) Faculty of Economics and Business
BSc Business Administration
Student name: Keven Guglielmina Student number: 11634731
Supervisor: Nazlihan Ugur Coordinator: Michele Piazzai
Statement of Originality
This document is written by Student Keven Guglielmina who declares to take full responsibility for the contents of this document.
I declare that the text and the work presented in this document are original and that no sources other than those mentioned in the text and its references have been used in creating it.
The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.
Table of Contents Abstract 4 Introduction 5 Literature Review 9 Conceptual Framework 15 Methodology 16 Results 26 Discussion 28 References 32 Appendices 37
Abstract
The topic of this paper emerges from an environmental concern about the unsustainable practices in the fashion industry. I investigate the difference in crowdfunding success between projects that have an environmental orientation and those that don’t have. More precisely, I look at the success rate and amounts of funding pledged between 562 environmental and 562
non-environmental crowdfunding campaigns, from the crowdfunding platform Kickstarter. Two hypotheses are drawn. The first is that having an environmental orientation influences the success of its crowdfunding campaign.The second is that having an environmental orientation influences the amount of funding pledged to its crowdfunding campaign. The outcome of this study suggests that projects in fashion do not succeed better in their crowdfunding campaign by having an environmental orientation. However, the analysis of results suggests that having an environmental orientation does have a positive influence on the amounts of funding pledged to a crowdfunding campaign.
Introduction
Each year, outdoor air pollution causes the premature death of 4.2 million people worldwide (Forouzanfar, 2016). The UN climate change conference negotiated the Paris Agreement (COP21), which was signed in 2016, between 196 countries, with the goal of decreasing greenhouse gas emissions and limiting global warming under 2 degrees celsius (United Nations, 2015). Paradoxically, 2016 was the warmest year ever recorded in history (NCEI, 2020). As we just entered the year 2020, we’ve also set a new record for the warmest month of January ever recorded in history (NCEI, 2020).
In 2015, the textile industry was responsible for 1.715 billion metric tons in green-house CO2 emissions (Kerr & Landry, 2017). This accounts for 5% of total worldwide CO2 emissions, in 2015 (IEA, 2020). By 2050, the textile industry could reach for a quarter of our carbon
emissions’ budget of keeping the average temperature below 2 degrees celsius (MacArthur, 2017).
As a matter of fact, it’s not just our air that gets polluted but our water as well, with an average of 8 million metric tons of plastic material entering our oceans every year (Jambeck et al., 2015). If the current state of business in the textile industry doesn’t change, between 2015 and 2050, it will be accountable for an added amount of 22 million metric tons of microplastics in the oceans (MacArthur, 2017).
The fashion industry includes the clothing, footwear, accessories and makeup sectors. It has to take considerable action to be more sustainable for the environment. The current state of business is centered on “fast-fashion” which consists of quick production (usually low quality) while keeping up-to-date with consumers' fast changing tastes (Cachon & Swinney, 2011). The
industry is characterized by products that have a short life, with unpredictable demand, and high consumer impulse to buy (Christopher, Lowson, & Peck, 2004). Following this trend and according to Euromonitor (2016; as cited in MacArthur, 2017), people are buying more clothes, but on average wearing them less. This means an increase in sales of clothes and decrease of their average utilization (how many times a piece of cloth is worn before disposing it). By 2050, total clothing sales are expected to increase by more than 3 times what they are today
(MacArthur, 2017).
There is a growing concern for the environment that the fashion industry keeps having such unsustainable practices. Consumers are more prone to buy eco-friendly clothes if these show high quality, are durable on the long-term and widely available in retail-stores; price is not a significant factor of choosing between sustainable and unsustainable clothes (Jacobs, Petersen, Hörisch, & Battenfeld, 2018; Niinimäki, 2010). Regarding this incremental innovation to a more eco-friendly fashion industry, it would be interesting to see if there is a preference to funding entrepreneurial fashion projects when they care for the environment.
When governmental institutions are not capable of producing change, then social entrepreneurs will bring up the problems and create innovative solutions (Bornstein, 2004). As social
entrepreneurs face social injustice; green entrepreneurs face environmental issues such as climate change (Allen & Malin, 2008). When trying to innovate, it is primordial for an entrepreneur to get all of the necessary resources. Startups that want to get funding for their projects, besides getting loans from banks and investors, are now going towards other unconventional ways to finance their ideas such as crowdfunding (Calic & Mosakowski, 2016).
Crowdfunding enables entrepreneurs to get funding for their project from many people. As small as the funding goal can be, it can give the startup a good financial kick-start.
Kickstarter is the most popular online crowdfunding platform (Alexa.com, 2020;
Cumming, Leboeuf, & Schwienbacher, 2020). As of June 19, 2020, over $5 Billion dollars have been committed and 183,410 projects have been successfully funded (Kickstarter Stats, 2020). From the 488,811 projects ever launched on Kickstarter, 31,838 were from the fashion category. In 2018, Kickstarter collaborated with the Environmental Defense Fund (EDF) to encourage entrepreneurs to “think green” when designing their products (Corcoran & White, 2018). Together, Kickstarter and EDF designed the Kickstarter Environmental Resource Center (KERC) to give insight to entrepreneurs on how they can launch more environmentally
sustainable products (Kickstarter Environmental Resources Center, 2018). This resource center builds upon the concept of the circular economy: “an economic system that replaces the
‘end-of-life’ concept with reducing, alternatively reusing, recycling and recovering materials in production/distribution and consumption processes.” (Kirchherr, Reike, & Hekkert, 2017). The KERC explains different ways projects, mainly the products themselves, can incorporate environmental ideas in them. Through its KERC, the platform passively encourages crowdfunding projects to be conscious about the environment.
There is still no factual data from Kickstarter about how environmental projects actually perform in their crowdfunding, compared to those that do not care about their environmental impact. In other words, it would be interesting to know how the funding for eco-friendly projects in fashion compares to projects in fashion that are not eco-friendly. This would give a
projects. Especially for entrepreneurs that would benefit from knowing if there is a difference in crowdfunding by designing their fashion products environmentally sustainable. Of course, at the same time by keeping in mind that they are being conscious for the well-being of our planet. Consequently, we can summarize our research goal in the form of a question.
Does having an environmental orientation in the fashion industry projects influence crowdfunding success?
This paper contributes in the understanding of the relationship, if there is any, between fashion projects that are eco-friendly and crowdfunding success. I will first look into the literature to better understand crowdfunding, environmental sustainability and look how they behave together. Then, I will hypothesize and test them by analyzing data from Kickstarter. Finally, the results will be explained and what it means for future research.
Literature Review Generalities of Crowdfunding
The concept of crowdfunding comes from crowdsourcing which consists of outsourcing online to people (crowd) important tasks for the conception of a product (Kleemann, Voß, & Rieder, 2008). Crowdfunding is a specific branch of crowdsourcing as it refers to only the funding part of a product or startup. Entrepreneurs in need of money to start their project, create crowdfunding campaigns to attract individuals willing to contribute to the funding of their
projects. On Kickstarter, entrepreneurs who create crowdfunding campaigns are called “creators” and people who fund projects are called “backers”.
In general to incentivize individuals to crowdfund, platforms use one or more of the four models: donation, lending, reward-based and equity (Mollick, 2014). Individuals who believe in
a project and do not expect anything in return make use of the donation system. Otherwise, the lending model makes use of individuals as lenders that expect some rate of return on their investments. The third model is reward-based and individuals expect something in return. For example, reward-based crowdfunding campaigns can promise to give samples of their products or invitations to special events. The fourth and last model, the equity model, gives a small share of the startup when individuals fund a certain amount of money.
According to Belleflamme, Lambert, & Schwienbacher (2014), the equity-based and the reward-based models are the most dominant in the world of crowdfunding.
Kickstarter is an example of a reward-based platform. In addition to giving rewards when backers give certain predefined amounts of money, the platform also lets backers donate money in exchange of nothing. Kickstarter follows the All-Or-Nothing model (Cumming, Leboeuf, & Schwienbacher, 2020). This means that backers pledge money and this money only goes to the creator if the crowdfunding campaign reaches its goal (or more).
Environmental Sustainability
Sustainability is modelled as a triadic relationship composed of three “pillars”: social, environmental and economic (Purvis, Mao, & Robinson, 2019; see Figure 1). According to Purvis, Mao, & Robinson (2019), it is believed that Barbier (1987) defined the starting point of the Three Pillar of Sustainability framework.
The first pillar is about social sustainability which refers to the well-being of people. For example, making sure every person gets access to clean drinkable water, healthy food and shelter. Social sustainability can be summarized as making sure all people thrive as a species (Humans) and do not leave anybody behind.
The second pillar is economical sustainability. It is about maintaining economic prosperity in the world. The financial crisis in 2008 showed that our economical model at that time was not sustainable (Ivashina, & Scharfstein, 2010).
The third pillar is environmental sustainability. It englobes all of our interactions with nature on Earth. It is the most important pillar of sustainability as without environmental sustainability, neither society nor the economy would be possible. Therefore, ensuring and promoting all life on Earth is key to our survival. The covid-19 related sanitary crisis has proven again the tight
interconnection between the three pillars. Society, the environment and the economy, have to be taken into consideration if we want to sustain as a species.
Figure 1
The Three Pillars of Sustainability (taken from Purvis, Mao, & Robinson, 2019)
Each of these pillars is as important for humanity. Without sustaining the environment, we don’t have a home. Without social sustainability, we are doomed to end as a species. Without
In this paper, we will be talking only about the environmental aspect of sustainability. When referring to the UN’s Sustainable Development Goals, environmental sustainability encompasses four goals which are:
Goal 12: Responsible Consumption and Production. Overconsumption is usually bad for the environment because for every product we consume, there is waste creation. Therefore, there is a need for consumers to reuse things, recycle, and buy only what we really need. This also goes for companies that should produce their products while at the same time limiting as much as
possible the production of waste.
Goal 13: Climate Action. It is a scientific fact that our emissions (mainly CO2) are responsible for climate change (Solomon, Plattner, Knutti, & Friedlingstein, 2009). We need to limit our CO2 emissions and reduce air pollution if we want to maintain a livable environment for future generations.
Goal 14: Life Below Water. One of the biggest concerns is our plastic waste in our oceans. Plastics (polymers) are ingested by fish which has drastic consequences for the ecosystem in which they live and for our own health (Halden, 2010). We need to prevent waste from going into our waters.
Goal 15: Life on Land. We need to start caring more about our landscape. Such as: preventing deforestation, limiting our impacts on our fauna and flaura, stopping intensive agriculture (have more sustainable agriculture).
These four goals will be our starting ground for defining crowdfunding projects that are eco-friendly.
Eco-friendly Projects in Crowdfunding
The topic of this paper is about the performance in crowdfunding of fashion projects that are sustainable towards the environment. In the remainder of this paper, both terms
“eco-friendly” and “environmental” will be used with the same meaning. As such, an
environmental crowdfunding campaign refers to a crowdfunding campaign from an eco-friendly project. We want to explore the crowdfunding success of fashion projects that are eco-friendly. To do so, it is important first to specify what determines a projects being categorized as being eco-friendly or not.
By looking at projects on Kickstarter, there are two kinds of eco-friendly orientations: reducing its impact on the environment and promoting the environment.
According to Kickstarter Environmental Resources Center (n.d.), there are numerous ways for a product to reduce its impact on the environment. I will go through the ones that are relevant for fashion projects.
First, a fashion project can reduce its impact on the environment by designing its product to be durable. This means that the product will have a longer lifespan. In contraste, planned
obsolescence is an example of designing products to fail, they have to be replaced more often. This creates more sales, but produces dramatically more waste and pollution. Therefore, products should be designed to last if they care for the environment. Fashion products can use more
durable and sustainable materials and encourage consumers to wash their clothes at lower temperatures and not use a drying machine. This limits the damage to tissues and limits the energy usage.
Second, projects should incentivize consumers to not throw their clothes. Giving the opportunity for easy repairs when clothes are damaged, instead of having to throw them. Also, incentivizing consumers to share or donate their clothes when they don’t use them.
Third, fashion projects should consider producing more sustainably. They can use recycled materials instead of new raw materials. Use natural dyeing techniques and prefer natural fabrics as opposed to polyester which is known to release microplastics in the water (Napper &
Thompson, 2016).
Alternatively, fashion projects can make a positive impact on the environment by
promoting it through reforestation, capturing air pollution, and creating protected sanctuaries for animals. Platforms like Carbon Credit Capital give the opportunity to projects to offset their carbon emissions by buying carbon credits (Carbon Credit Capital, n.d.). A project can
compensate each ton of greenhouse emissions by paying another company to extract it from the air. Also, a project can use a portion of its profit to plant trees or clean the oceans from plastic.
While the first form of eco-friendly measures diminishes the negative impacts on the environment, the second form counters it with positive impacts on the environment. As such, an eco-friendly project can be one or both of the two forms. When accounting for eco-friendly fashion projects, it is important to consider both types of environmental initiatives.
I investigate in this paper the relationship between the success of a crowdfunding campaign in fashion and the project being eco-friendly or not. Few studies have been made on the topic. A paper from Bento, Gianfrate, & Thoni (2019) suggests that around 70% of
sustainable ventures survive the first year after succeeding their crowdfunding campaign. This gives a positive prospective post-funding phase, but accounts for both social and environmental
sustainable projects. Closer to our research question, Hörisch (2015) analyses the success of projects from all the categories on Indiegogo, a similar crowdfunding platform as Kickstarter. He compares 10 projects from the category “environmental” with all the other categories such as Film, Technology, Fashion, etc,... He doesn’t find any significant positive influence of
environmental orientation on the success of crowdfunding. An important limitation is the fact that the sample is way too small (10 environmental projects). Also, an environmental fashion project could prefer being in the fashion category than the environmental category. However, a study Calic, & Mosakowski (2016) contradicts Holrich (2015) by suggesting that an
environmental orientation does have an influence on the success of a crowdfunding campaign. The results of those studies show that there are some opposing views in the literature
surrounding how an environmental orientation of a project influences the success of
crowdfunding campaigns. A limitation of these studies is that they try to measure crowdfunding success for both environmental and social sustainability, in all the industries. My paper will only focus on environmental sustainability in the fashion industry. This will allow an increased accuracy in the results.
Conceptual Framework
This paper investigates the influence of an environmental orientation on crowdfunding success. Coming from the literature, I expect to find a positive influence for a project having an environmental orientation and the success of its crowdfunding campaign. Thus, I have developed a simple conceptual framework (figure 2). I expect to measure a significant positive effect of environmental orientation on crowdfunding success. As such, when a project has an
environmental orientation it will have a significant influence on the success of its crowdfunding campaign.
Figure 2
Conceptual Framework
To analyze the relationship between environmental orientation and crowdfunding success, I have developed two hypotheses (H1 and H2).
Success Rate
H1: Crowdfunding campaigns in fashion that have an environmental orientation are more likely to succeed in meeting their funding goal than non-environmental ones.
The first hypothesis focuses on the success rates of crowdfunding campaigns between eco-friendly projects and those which are not eco-friendly. I will be testing for a [statistically significant] difference in the success rates between the two groups. The success rate is
determined as the number of crowdfunding campaigns that succeeded in the sample divided by the total number of crowdfunding campaigns in the same sample. It is then expressed as a percentage. I expect to find a larger success rate for the environmental sample than the non-environmental one.
Funding
H2: Crowdfunding campaigns in fashion that have an environmental orientation get more funding than non-environmental ones.
The second hypothesis focuses on the difference in the amounts of funding pledged to crowdfunding campaigns, between projects that are eco-friendly and those which are not. I will analyze for a statistically significant difference in the amounts of funding pledged between the two groups. The amount of funding pledged is the amount of money that is promised to be given if the crowdfunding campaign succeeds. As such, I expect to measure a significant increase in the amount of crowdfunding a fashion project gets when it has an environmental orientation.
Methodology Data Collection & Processing
For the purpose of this study, I used WebRobots.io, a web data scraper tool, to gather the data from 211,076 Kickstarter projects. This is about half of all projects ever launched on
Kickstarter (488,188 crowdfunding campaigns). The data extraction was taken on February 13, 2020, by webrobots.io. The dataset from Kickstarter used in this analysis can be downloaded from the website of Web Robots (appendix A).
To clean and process the data, I used R programming language. Its main purpose is for performing statistical analyses. The exact copy of the code to reproduce the analysis can be found in appendix A. However, the analysis part for this paper was performed with IBM's SPSS software. It is a graphical user interface, thus no code, which is more intuitive. This is just a matter of personal preference.
The entire dataset has 211,076 crowdfunding projects. First, I extracted the projects from the fashion category. Then, I removed live/cancelled/suspended projects because the focus of this paper is on those who reached their deadline for funding and got an outcome. Hence, I kept projects that were either successful or failed. Live projects are irrelevant as backers are still
funding the projects. Cancelled projects are those which did not reach their deadline because they were removed by the creator in the middle of the crowdfunding campaign. Too many personal reasons can account for that. For example, the entrepreneur could have found another source of funding. Lastly, suspended projects are those who were removed by Kickstarter itself because the crowdfunding campaign violated Kickstarter’s rules. For example, it is not allowed to pledge for your own crowdfunding campaign (Kickstarter Support, n.d.). After only keeping projects that succeeded and those which failed, I proceeded by checking for missing values in observations. No observations had missing values. However, 1,976 duplicates were removed from the fashion dataset. The cause of having duplicates is because of the method in how the data scraper collects the data. Some projects are collected twice when they belong to two categories. Removing duplicates makes sure all projects in the dataset are unique.
Finally, the fashion dataset contains a total of 9,468 observations (unique crowdfunding campaigns). The sample of fashion projects (9,468) represents a third of the 31,794 projects launched ever in the fashion category (as of June 17, 2020)(Kickstarter Stats, 2020). The
deadline when a project has either succeeded or failed in the fashion sample ranges from January 2011, to February 2020.
The second part consists, using keywords, to separate observations (crowdfunding campaigns) that are eco-friendly and those that are not eco-friendly.
To extract the projects that are eco-friendly I used the keywords: “eco-friendly”,
“environmental”, “sustainable”, “recycle”, “organic”, “waste”, “durable”. The keywords were inspired from a related study from Bento, Gianfrate, & Thoni (2019) which had the keywords: eco-friendly”, “ecological”, “biodegradable” and “recyclable”. However, I added
“environmental”, “organic”, “waste” and “durable”, to improve the reliability and quantity of environmental projects.
The keywords are inclusive in the sense that they also take into account words that are made up of these keywords. For example, the word “environmentally” has “environmental” in it, thus it is taken into account.
I did all the cleaning and categorization of projects using R. Which I ended up with 3 datasets: - A dataset (fashion_env) containing eco-friendly projects has N=562 observations. - A dataset (fashion_not) containing projects that are not eco-friendly has N=8906
observations.
- In addition, a random sample (fashion_not_sample) of N=562 was taken from fashion_not.
The data sample combining 562 environmental oriented projects and 562 non-environmental oriented projects can be accessed using the link provided in appendix A.
Variables
To test for both hypotheses H1 and H2, I have made use of 3 main variables.
First, using a program written in R, I analyzed the names and descriptions of all the projects from our sample. Using the 7 keywords (“eco-friendly”, “environmental”, “sustainable”, “recycle”, “organic”, “waste”, “durable”), I splitted the category “Fashion” into two groups. This allows us to analyze the difference between the two samples based on their environmental
On the one hand, projects which had at least one of the keywords in their name or description were put in the environmental group called fashion_env. All the projects of the environmental group were coded as 1, into the variable is_env.
On the other hand, projects that didn’t include any of the keywords were put in the
not-environmental group called fashion_not. These projects that didn’t appear to be eco-friendly were coded as 0, into the variable is_env.
Therefore, eco-friendly projects are coded as is_env=1 and non-eco-friendly projects are coded as is_env=0.
The second variable (state_binary) is the state of a project. It is important to note that projects that were still live upon the data collection and those which had been cancelled or suspended are excluded from the data. Only crowdfunding campaigns that either succeeded or failed are included. For the analysis, I also coded this variable as binary, named state_binary. Crowdfunding campaigns that succeeded are coded as 1 (state_binary=1) and failures are coded as 0 (state_binary=0).
The third variable (converted_pledged_amount) is the currency converted (in USD) amount of funding pledged for the crowdfunding campaign. Crowdfunding campaigns can be started around the world in their local currency. Therefore, pledged amounts for each
crowdfunding campaign are converted from their local currency to US dollars ($). For the purpose of this analysis, the historical exchange rates are taken on the date of the deadline of each respective crowdfunding campaign. Converted amounts are represented in the variable converted_pledged_amount, which is numerical and continuous. The variable only expresses positive amounts because it is not possible to pledge negative amounts.
Table 1
Quantitative Variables
Variable Description Expressed as
is_env
If the fashion project has an environmental orientation (1) or does not have an environmental orientation (0)
Categorical Binary
(0/1)
state_binary
If the crowdfunding campaign reached its funding goal (pledged > goal) before its deadline.
Success (1) and Failure (0)
Categorical Binary
(0/1)
converted_ple dged_amount
Funding pledged amount converted in USD ($) that was committed to a crowdfunding campaign. Conversion rates are calculated based on the date of the deadline of the crowdfunding campaign.
($; US Dollars) Numerical, continuous
Descriptives - Preliminaries
Before doing any statistical analysis, I looked at the success rate in each group
(environmental and non-environmental). In table 2, the overall fashion sample has a success rate of 62.29 % (not taking into account the environmental orientation). This is way higher than Kickstarter’s official success rate of 28.13%. That is in part because I do not take into account projects that were cancelled or suspended. Also, the sample population of the fashion category is a third of the size of all historical fashion projects ever launched on Kickstarter. Upon this fact,
the success rates presented in table 2 are specific to this study. In the sense that they reflect differences between the two groups (environmental & non-environmental), but do not reflect how Kickstarter counts success rates.
From our environmental sample of 562 eco-friendly projects, there are 378 that succeeded and 184 that failed to meet their goal amount of funding before their deadline. This gives a success rate of 67.26% for the environmental group.
In contrast, the non-environmental sample consists of 8906 crowdfunding campaigns. There are 5520 crowdfunding campaigns that succeeded to meet their goal amount of funding before their deadline, and 3386 that failed to do so. This gives a success rate of 61.98% for the
non-environmental group. Consequently, there is a 5.28 percentage points difference between the two groups. In other words, the difference between the two groups is not as significant as
TABLE 2
Fashion Success Rates
Fashion Success Rates Success Failure
Environmental (N=565) 0.6726 (378 / 562) 0.3274 (184 / 562) Non-Environmental (subsample; N=565) 0.6263 (352 / 562) 0.3737 (210 / 562) Non-Environmental (N=8906) 0.6198 (5520 / 8906) 0.3802 (3386 / 8906) ALL Fashion 0.6229 (5898 / 9468) 0.3771 (3570 / 9468)
To make a fair comparison between the two groups I decided to randomly sample the non-environmental group in a sample of the same size as the environmental group. This initiative comes as the two groups differ greatly in size, their variance are very distinct. The variance for the environmental group of size N=562 is var=1.180E+9. However, the variance for the entire non-environmental of size N=8906 is var=4.847E+9. Subsequently, the variance of the
environmental group is 4 times smaller than the non-environmental group. To get a similar variance, I decided to use a random subsample of the non-environmental group which would limit the large gap between projects. Indeed, the random subsample of the non-environmental group of size N=562 has a variance var=1.034E+9, way closer to the variance of the
environmental sample. Also the success rate for the random subsample of the non-environmental group is 62.63%, which is fairly close to the success ratio of its related larger sample (61.98%).
Now, the two samples have the same size. With this in mind, I checked the descriptives for both groups. For the variable converted_pledged_amount, the two samples follow a
Log-normal distribution (appendix B). Therefore, none of the samples has a normal distribution because they are highly skewed to the right. To confirm that both samples don’t have a normal distribution I used the Kolmogorov-Smirnov and the Shapiro-Wilk tests. Both of the tests are significant, which means an abnormality of the distribution regarding the
converted_pledged_amount variable for both samples (sig=.000).
The descriptives of the crowdfunding amounts for projects in the two samples show that there are visual differences in means and medians. The environmental sample has a
mean(is_env=1)=16,280.03 and median(is_env=1)=7,343.50. The non-environmental sample has a mean(is_env=0)=10,478.77 and median(is_env=0)=1,525.00. The mean for the environmental sample is 1.55 times higher than the non-environmental sample. The median for the
environmental sample is 4.82 times higher than the non-environmental sample.
Also, to make sure both samples have the same variance I performed a Levene’s Test (Appendix C). The null hypothesis is that both samples are homogenous, meaning that their variances are statistically similar (sig>0.05). The result of the Levene’s Test is not significant (sig = .053) which means that the variances are similar between the environmental and the
non-environmental sample. Knowing that both samples share a similar variance helps strengthen the support for the analysis.
Analysis of the Success Rate
The difference in the success rate between the environmental and non-environmental samples is not as important as predicted. The success rate of the environmental sample is 4.63 percentage points higher than the non-environmental sample. As both variables, is_env and state_binary are binary (categorical), I performed a Fisher’s Exact Test on its crosstabulation (Appendix D). The Fisher’s Exact Test checks if two categorical variables are associated between each other. The Fisher’s Exact Test was insignificant. This means that it confirms that is_env and state_binary are not related to each other. Thus, confirms that the environmental orientation of a project does not have an influence in the success of a crowdfunding campaign.
Table 3
Descriptives for the Converted Pledged Amounts Converted Pledged Amounts ENVIRONMENTAL (N=562) NON- ENVIRONMENTAL (subsample; N=562) NON- ENVIRONMENTAL (N=8906) Mean 16,280.03 10,478.77 11,807.84 Variance 1,180,166,263.05 1,033,757,876.26 4,847,075,497.50 Standard Deviation 34,353.55 32,152.11 69,620.94 Minimum 0.00 0.00 0.00 Q1 691.80 220.50 200.00 Median 7,343.50 1,525.00 1,326.50 Q3 18,191.20 8,377.20 7,663.00 Maximum 448,500.00 407,853.00 3,948,547.00
Analysis for the Funding
After performing the preliminary analysis and checking for assumptions, I know that both samples have a similar log-normal distribution and that both samples are statistically similar variances. Now, the interest is in seeing the difference in medians between the two groups. The
choice of analyzing medians and not means is first because of the interest of this paper is about seeing if there are more environmental projects that get a higher funding compared to those that are not environmental. The mean would be a biased way to check that because some projects in one group can have outstanding funding compared to the great majority. Additionally, our samples are highly skewed to the right. None of the samples follow a normal distribution which means medians are a more robust way to analyze the differences in funding between samples. To check for a difference in medians between the two samples I used the Mann-Whitney U Test. It is equivalent to an Independent Samples T Test, but nonparametric. This means that the
Mann-Whitney U Test doesn’t assume the samples to be normally distributed (our case). It just assumes that the data from both samples share the same shape, which is the case for our samples (appendix B). The null hypothesis of the Mann-Whitney U Test is that the medians of the
environmental and non-environmental samples are identical (sig>0.05).
The result of the Mann-Whitney U Test is significant (sig = 0.00) which means that the null hypothesis is rejected (appendix E). Which means that the medians of the two samples are not identical. Therefore, there are more projects in the environmental sample that have a higher pledged amount of funding than those in the non-environmental sample. More precisely, half of the projects in the environmental sample have a pledged amount of 7,343.50 or higher, while half of the projects in the non-environmental samples have only a pledged amount of 1,525.00 or higher.
Results
In the analysis part, two hypotheses have been tested. The first one (H1) is that crowdfunding campaigns from environmental fashion projects are more likely to succeed in
meeting their funding goal than non-environmental ones. When looking at the descriptives, the success rate of environmental crowdfunding campaigns is higher than for non-environmental crowdfunding campaigns. Using samples of the same size, the success rate is 67.26% for the environmental group. Whereas for the non-environmental group, the success rate is 62.63%. This is a 4.63 percentage points difference between the success rates of the two samples (same size). Then, I performed a statistical analysis using Fisher’s Exact Test to check its significance. Fisher’s Exact Test showed an insignificant difference for the success rate between the two groups (appendix D). The environmental orientation of a project does not influence the success of its crowdfunding campaign. Hypothesis H1 is rejected.
The second hypothesis (H2) is that crowdfunding campaigns from environmental fashion projects get more funding than non-environmental ones. I first looked at the statistical
descriptives of both groups (environmental and non-environmental) for the variable converted_pledged_amount. This variable takes the pledged amounts at the end of the
crowdfunding campaign. Those amounts are then converted to US dollars ($) with a conversion rate taken on the day each crowdfunding campaign ended. This allowed for a comparison between projects worldwide without having to worry about multiple different currencies. Projects in the non-environmental group (N=8906) reached up to $3,948,547 dollars. In comparison, projects in the environmental group (N=562) reached up to $448,500 dollars. However, the majority of projects in the environmental group get more funding than in the non-environmental group. Also, the amounts of funding pledged for non-environmental projects vary a lot compared to those in the environmental group which vary less.
To equilibrate the variance in the two groups as understood in the descriptives - preliminaries section above, I decided to randomly sample the non-environmental group. The purpose of having a random subsample from the non-environmental group, with the same size as the environmental group, was to have similar variances between groups. Therefore, both samples had 562 observations each. Indeed, the Levene's Test confirmed that the variances were statistically similar between the two samples (N=562 each)(appendix C). Then, I performed a Mann-Whitney U Test to analyze the differences in medians between the two samples converted pledged amounts (environmental & non-environmental). The Mann-Whitney U Test showed a statistically significant difference between the median of the environmental and the
non-environmental sample (appendix E). Considering that our non-environmental subsample has a higher median than the bigger non-environmental sample (of N=8906), I conclude that the test is relevant for a difference in medians between the two groups (larger samples).
In other words, environmental projects [in general] get more funding than non-environmental ones. Finally, the outcome of the analysis supports the second hypothesis (H2).
Hypothesis H2 is supported.
Discussion
The results of this study show two interesting topics to discuss. First, the outcome of the analysis showed that for crowdfunding campaigns, there is no relation in terms of success if a project is being eco-friendly or not. The first hypothesis was rejected (reject H1), but the second hypothesis was supported (support H2). As such, half of the crowdfunding campaigns of
environmentally oriented projects get $7,343.50 or more of pledged funding. Whereas, half of the crowdfunding campaigns of projects that are not environmentally oriented “only” get
$1,525.00 or more of pledged funding. Nevertheless, a few projects in the non-environmental group perform extraordinarily well with pledged funding amounts going up to $3,948,547, compared to a maximum of $448,500 for environmental ones. However, the analysis shows that there is a larger quantity of eco-friendly projects that get more pledged funding than those which are not eco-friendly.
At first sight, this is quite contradictory that environmentally oriented projects do not
significantly succeed (success/failure variable) better in crowdfunding than non-environmental ones. While, considering that we would randomly take a project that is eco-friendly and one that is not eco-friendly, the one that is eco-friendly would have a higher probability of having more funding pledged for its crowdfunding campaign than the project that is not eco-friendly.
There are a few explanations for these results. First, the fact that H1 is rejected and H2 is supported, can be interpreted as such that people do not blindly fund projects just because they are conscious about the environment. Considering the results of this paper, a project choosing to be eco-friendly does not increase its chance of succeeding in its crowdfunding campaign. However, generally speaking, projects that are being eco-friendly will attract more funding than those which are not. Indeed, it is a generality because there are some exceptions for the
non-environmental group. Some of the non-environmental projects did get outstanding amounts of pledged funding, but they are just a minority (outliers). This goes in line with the idea that crowdfunding campaigns succeed mainly when the quality of the related project is high (Mollick, 2014).
Implications
Entrepreneurs in the fashion industry should consider designing their products more environmentally sustainable. Of course, in terms of environmental benefit, but also if they are planning to launch a crowdfunding campaign for their projects. It seems that projects that are being eco-friendly and publicly share about their environmental initiatives, have a higher chance of getting more funding than if they choose not to be conscious of the environment. On its own, a fashion project being eco-friendly does not mean it will succeed in its crowdfunding campaign, but will get more funding compared to non-environmental projects. This goes in hand with Allison, Davis, Webb, & Short (2017), who suggests that inexperienced crowdfunders invest for projects which personally inspire them (identity), compared to critical crowdfunders that will focus on the quality of the project. This also supports the suggestion of Mollick (2014), that the success of a crowdfunding campaign relates to the quality of the project.
Based on the analysis and the outcomes of this paper, further in depth research can be undertaken on the topic. For example, by investigating the demographics of the people who fund projects that are sustainable for the environment. Also, search if people’s main motive to fund eco-friendly projects is because of the environmental orientation or because of other factors. Lastly, it would be interesting to see if for other industries the outcome of this study would be similar. Depending on the outcome, it could generalize the results of this study to a broader audience.
Limitations
The analysis relies on using keywords to select crowdfunding campaigns of projects that were environmentally oriented or not. As not much similar research has been done, the selection
of the keywords to be used was not systemic. Keywords were selected on the basis of their performance to include environmental projects. It is possible that environmental projects were not being included in the environmental sample. Even though most of them were checked for consistency, there is a margin of error to consider. An idea for future research to select projects based on their environmental orientation would be to use machine learning.
For future studies, it would be suggested to add control variables to add robustness in the analysis. Adding more demographic variables could also show new gaps in the topic.
Conclusion
This paper explored with an environmental perspective the success rate and pledged amounts of crowdfunding campaigns in the fashion industry. The fact that in 2050, the fashion industry will grow by 3 times in market sales as of today, is increasingly worrying for the environmental sustainability of our planet (MacArthur, 2017). I took a financial perspective for entrepreneurs, to see if projects perform better in crowdfunding when they are explicitly conscious about its negative impacts on the environment and act towards making a more sustainable planet. Two hypotheses were being tested in the analysis. First, H1 tested if
crowdfunding campaigns in fashion that are environmentally focused are more likely to succeed in meeting their funding goal than non-environmental ones. H1 proved to be rejected, as there was no significant difference in the success rate between environmental and non-environmental crowdfunding campaigns. Second, H2 tested if crowdfunding campaigns in fashion that are environmentally focused get more funding than non-environmental ones. There was support for H2, as most environmental projects performed better in terms of crowdfunding pledged amounts, than most non-environmental ones. This shows that people are more prone to give to an
eco-friendly project, but does not alone explain the success of a crowdfunding campaign. Further research has to be done on the topic to explain this relationship in more detail.
References
Alexa.com (2020). Kickstarter.com Competitive Analysis, Marketing Mix and Traffic. Amazon. Retrieved on May 13, 2020, from https://www.alexa.com/siteinfo/kickstarter.com
Allen, J. C., & Malin, S. (2008). Green entrepreneurship: a method for managing natural resources?. Society and natural resources, 21(9), 828-844.
Allison, T. H., Davis, B. C., Webb, J. W., & Short, J. C. (2017). Persuasion in crowdfunding: An elaboration likelihood model of crowdfunding performance. Journal of Business
Venturing, 32(6), 707-725.
Barbier, E. B. (1987). The Concept of Sustainable Economic Development. Environmental Conservation, 14(2), 101–110.
Belleflamme, P., Lambert, T., & Schwienbacher, A. (2014). Crowdfunding: Tapping the Right Crowd. Journal of Business Venturing, 29(5), 585–609.
Bento, N., Gianfrate, G., & Thoni, M. H. (2019). Crowdfunding for sustainability ventures. Journal of Cleaner Production, 237, 117751.
Bornstein, D. (2004). How to Change the World: Social Entrepreneurs and the Power of New Ideas.
Brundtland, G. H., & Khalid, M. (1987). Our common future. Earth and Us, 29–31.
http://www.unece.org/fileadmin/DAM/ie/se/pp/EnCom15/28Nov/SustDev/HELD_SustD ev_UNECE_EnComm15_2006_c.pdf
Butticè, V., Colombo, M. G., Fumagalli, E., & Orsenigo, C. (2019). Green oriented
crowdfunding campaigns: Their characteristics and diffusion in different institutional settings. Technological Forecasting and Social Change, 141, 85-97.
Cachon, G. P., & Swinney, R. (2011). The value of fast fashion: Quick response, enhanced design, and strategic consumer behavior. Management science, 57(4), 778-795. Carbon Credit Capital (n.d.). Retrieved June 10, 2020, from
https://carboncreditcapital.com/carbon-credits/
Calic, G., & Mosakowski, E. (2016). Kicking off social entrepreneurship: How a sustainability orientation influences crowdfunding success. Journal of Management Studies, 53(5), 738-767. https://onlinelibrary.wiley.com/doi/pdf/10.1111/joms.12201
Corcoran, H. & White, D. (2018, November 27). New Features on Kickstarter Encourage Creators to Think Green. Medium. Retrieved from
https://medium.com/the-fourth-wave/new-features-on-kickstarter-encourage-creators-to-think-gr een-b17a05f41bab
Christopher, M., Lowson, R., & Peck, H. (2004). CREATING AGILE SUPPLY CHAINS IN THE FASHION INDUSTRY. International Journal of Retail & Distribution
Management, 32(8), 367–376.
Cumming, D. J., Leboeuf, G., & Schwienbacher, A. (2020). Crowdfunding Models: Keep-It-All vs. All-Or-Nothing. Financial Management, 49(2), 331–360.
Elkington, J. (1994). Towards the Sustainable Corporation: Win-Win-Win Business Strategies for Sustainable Development. California Management Review, 36(2), 90–100.
https://doi.org/10.2307/41165746
Euromonitor (2016). Apparel & Footwear in 2016: Trends, Developments and prospects. Euromonitor International.
Forouzanfar, M. H., Afshin, A., Alexander, L. T., Anderson, H. R., Bhutta, Z. A., Biryukov, S., ... & Cohen, A. J. (2016). Global, regional, and national comparative risk assessment of 79 behavioural, environmental and occupational, and metabolic risks or clusters of risks, 1990–2015: a systematic analysis for the Global Burden of Disease Study 2015. The lancet, 388(10053), 1659-1724.
Halden, R. U. (2010). Plastics and health risks. Annual Review of Public Health, 31(1), 179–194. Hörisch, J. (2015). Crowdfunding for environmental ventures: an empirical analysis of the
influence of environmental orientation on the success of crowdfunding initiatives. Journal of cleaner production, 107, 636-645.
IEA (2020). Global Energy Review 2020, page 17. Retrieved June 12, 2020, from https://webstore.iea.org/global-energy-review-2020
Ivashina, V., & Scharfstein, D. S. (2010). Bank Lending During the Financial Crisis of 2008. Journal of Financial Economics, 97(3), 319–338.
Jacobs, K., Petersen, L., Hörisch, J., & Battenfeld, D. (2018). Green thinking but thoughtless buying? An empirical extension of the value-attitude-behaviour hierarchy in sustainable clothing. Journal of cleaner production, 203, 1155-1169.
Kerr, J., & Landry, J. (2017). Pulse of the fashion industry. Global Fashion Agenda. Kickstarter Environmental Resources Center (2018). Retrieved on May 7, 2020, from
https://www.kickstarter.com/environment
Kickstarter Stats (2020, June 19). Retrieved on June 19, 2020, from https://www.kickstarter.com/help/stats
Kickstarter Support (n.d.). Retrieved on June 10, 2020, from https://help.kickstarter.com/hc/en-us
Kirchherr, J., Reike, D., & Hekkert, M. (2017). Conceptualizing the circular economy: An analysis of 114 definitions. Resources Conservation and Recycling, 127, 221–232. Jambeck, J. R., Geyer, R., Wilcox, C., Siegler, T. R., Perryman, M., Andrady, A., ... & Law, K.
L. (2015). Plastic waste inputs from land into the ocean. Science, 347(6223), 768-771. Kleemann, F., Voß, G. G., & Rieder, K. (2008). Un(der)paid Innovators: The Commercial
Utiliza- tion of Consumer Work through Crowdsourcing. Science, Technology & Innovation Studies, 4(1), 5–26.
MacArthur, F. E. (2017). A new textiles economy: redesigning fashion’s future. Retrieved March 22nd, 2020, from
https://www.ellenmacarthurfoundation.org/assets/downloads/publications/A-New-Textile s-Economy_Full-Report.pdf
McNeill, L., & Moore, R. (2015). Sustainable fashion consumption and the fast fashion conundrum: fashionable consumers and attitudes to sustainability in clothing choice. International Journal of Consumer Studies, 39(3), 212–222.
https://doi.org/10.1111/IJCS.12169
Mollick, E. (2014). The dynamics of crowdfunding: An exploratory study. Journal of business venturing, 29(1), 1-16.
Napper, I. E., & Thompson, R. C. (2016). Release of synthetic microplastic plastic fibres from domestic washing machines: Effects of fabric type and washing conditions. Marine pollution bulletin, 112(1-2), 39-45.
Niinimäki, K. (2010). Eco-clothing, consumer identity and ideology. Sustainable development, 18(3), 150-162.
NOAA National Centers for Environmental information (2020, April). Climate at a Glance: Global Rankings. Retrieved on May 13, 2020, from https://www.ncdc.noaa.gov/cag/ Purvis, B., Mao, Y., & Robinson, D. (2019). Three pillars of sustainability: in search of
conceptual origins. Sustainability Science, 14(3), 681–695.
Solomon, S., Plattner, G.-K., Knutti, R., & Friedlingstein, P. (2009). Irreversible climate change due to carbon dioxide emissions. Proceedings of the National Academy of Sciences of the United States of America, 106(6), 1704–1709.
United Nations (2015). Paris Agreement. Retrieved May 13, 2020, from
https://treaties.un.org/pages/ViewDetails.aspx?src=TREATY&mtdsg_no=XXVII-7-d&c hapter=27&clang=_en
Appendices Appendix A
Link of the entire dataset of projects from Kickstarter (13/02/2020):
https://s3.amazonaws.com/weruns/forfun/Kickstarter/Kickstarter_2020-02-13T03_20_04_893Z. zip
Link to the code in R used to clean and code the data:
https://drive.google.com/file/d/1KUgQFGR8nyCRg5V2nizZAAl0hUM8xo-d/view?usp=sharing
Link to the data sample used for the analysis:
