The Intersection of Legal and Illegal Supply Chains: A Case Study of the Tortoise and Freshwater Turtle Pet Trade in Canada

Location. —

Turtles are both imported to and exported from Canada. Five of Canada’s eight native extant freshwater turtle species are either threatened or endangered (SARA 2023), and some have been targeted by illegal harvesters and traffickers for sale on the global market (e.g., CBC 2018). Non-native turtles are illegally imported into Canada (e.g., 20 live sulcata tortoises [Centrochelys sulcata] [EC Media Relations 2013a] and 205 animals, including African side-neck turtles [subfamily Pelomedusinae] and red-footed tortoises [Chelonoidis carbonarius] [EC Media Relations 2013b]).

Combatting illegal imports and exports is the responsibility of multiple agencies that regulate Canadian trade at different jurisdictional levels. At the federal level, Environment and Climate Change Canada (ECCC)’s Wildlife Enforcement Directorate (WED) enforces the Wild Animal & Plant Protection and Regulation of International & Interprovincial Trade Act (Canada) (WAPPRIITA) and the Species at Risk Act (SARA). With Canada’s ratification of the Convention on International Trade in Endangered Species of Wild Fauna and Flora (CITES), WAPPRIITA prohibits the 1) trade in listed species, except with a permit; 2) import of species that have been exported in contravention of the source country’s domestic laws; and 3) possession of species that were imported in contravention of the above. Most native turtles are SARA-listed, which means their collection and commercialization are restricted. The Canadian Food Inspection Agency (CFIA) also regulates the import of turtles under the Health of Animals Act, which “control[s] diseases and toxic substances that may affect animals and/or be transmitted from animals to humans” (CEC 2017, p. 3). Under this act, the import of live turtles for commercial purposes is prohibited, and a permit is required for imports by zoos, as personal pets, or for scientific and educational purposes (CEC 2017). Restrictions on the commercial import of live turtles came into effect in 1975 in response to the risk of salmonellosis (D’Aoust and Lior 1978). Within Canada, additional provincial and municipal restrictions are in place on the trade and possession of live turtles, which provide varying degrees of protection.

Based on the above federal, provincial, and municipal restrictions, commercial trade in live native turtles and the import of live non-native turtles for commercial purposes are therefore prohibited. The only legal commercial trade in live turtles involves domestic-bred non-native species that, in theory, entered Canada before the restrictions on live turtle imports in 1975 or with a permit. There are no known large-scale breeding farms, so Canadian breeders are thought to be primarily hobbyists, and it has been suggested that low supply is augmented by illegally sourced turtles (CEC 2017). In combination, Canada’s stricter laws and absence of large-scale breeders are thought to drive prices higher than those found in the neighboring United States, which has commercial large-scale breeders and a larger supply of turtles (CEC 2017). This has created a price differential, such that the same turtle species in Canada fetch higher prices than their US counterparts, which is thought to have incentivized illegal trade (CEC 2017).

Data. —

We collected data from federal-level enforcement records and interviews to establish supply chain characteristics. The project was reviewed by Michigan State University’s Institutional Review Board (STUDY00005525) and categorized as Exempt 2ii.

Enforcement Records. —

We pulled federal-level enforcement records from the ECCC WED database using keyword searches for tortoise, turtle, terrapin, and tortue (French for turtle) for the years 2008–2018. We excluded active or open intelligence files and removed files that were not about turtle trade. We de-identified the remaining files and aggregated them to the provincial level. The resulting data set included 439 files, of which we included 155 related to the pet trade and 19 to illegal collection and patrols of protected areas in our data set (excluded files related to the trade in turtle meat and other parts and derivatives). We coded these files for species traded, locations and types of violations, and specific information related to characteristics of trade and enforcement action.

Interviews: Sample Population. —

We used purposive sampling with multiple techniques (Teddlie and Yu 2007) and a combination of snowball and opportunistic sampling to seek participation from actors that may be difficult to identify or engage (Robinson 2014). Our sampling efforts focused on individuals participating in the legal sale of turtles, including direct sales or business-to-business vendors (i.e., pet store managers and owners, wholesalers, online sellers, and breeders). In the ECCC enforcement files (2008–2018), breeders and vendors had the most contact with enforcement officers. As intermediaries (Phelps et al. 2016) (i.e., actors in the middle of the supply chain), breeders and vendors are likely to have an understanding of how animals move from source to end buyer. Based on the enforcement records, we were also aware of vendors shipping turtles across the country and, therefore, suspect that they form a larger network. The majority of the enforcement records (approximately one-third) were in Ontario, so we started sampling there and expanded our search to other provinces using Google searches in French and English. Google searches included exotic pet stores, reptile trade show, vendor, retailer, for sale, sales, rehome, turtle, tortoise, tortue, reptiles, breeders, province names, capital cities, and urban centers with populations over 50,000.

We invited 83 vendors to participate in an interview, which represented all turtle vendors nationwide that we could identify and contact. Of those, we excluded 12 individuals who were minors or said they did not sell turtles. Seventy-one vendors were eligible for inclusions, and 24 vendors (which included 26 individuals: 10 female, 16 male, median age 40–49) agreed to be interviewed (a 33% response rate).

To assess the validity of our sample, we considered the point at which no new themes were emerging (i.e., saturation) (Copes et al. 2020). Although that point can be difficult to identify, theme saturation may occur within the first 12 interviews (Guest et al. 2006). To further evaluate our sample, we also used information power (Malterud et al. 2016), which accounts for 1) aim, 2) specificity, 3) theory, 4) dialogue, and 5) analysis, for continuous evaluation throughout the process. Sampling also ended when we were no longer provided with new contacts in the snowball sampling process.

Interviews: Interview Format. —

One author (R.B.) conducted all interviews between January and June 2022 in English online (Zoom, Skype, Teams) or over the phone because of COVID restrictions. Interviews were 1-on-1 discussions, which may be more effective than focus groups for socially sensitive topics (Kaplowitz 2000). However, on 2 occasions, pairs of individuals who worked directly together wished to be interviewed simultaneously, so the sample included 26 individuals, representing 24 unique vendors. Interviews lasted, on average, 68 min (38–184 min).

We conducted semistructured interviews, which allowed for requests for clarification and the ability to explore emergent themes (Galletta 2012). Interviews began with open-ended questions about participants’ interest in turtles and their experiences as a vendor, then narrowed to specific questions, followed by additional requests for clarification. We recorded interviews with permission from the participants and transcribed audio recordings in combination with interviewer notes. We redacted mentions of specific names or other identifiable information.

Interviews: Coding. —

We used parallel coding to identify themes in both data sets (Creswell and Plano Clark 2018). The coding process began with a “start list” of codes (Miles et al. 2020) for supply chain characteristics. We coded interviews and enforcement records to contextualize the trade and identify elements related to species, physical environment, and social and relational interactions. Data pulled from enforcement files included summaries of inspections and enforcement actions, actors involved, the types of species involved, locations, and trade routes. In addition, we coded for supply chain and network configurations, actor typology (see Phelps et al. 2016), and legal and illegal activity (e.g., concealment methods for smuggling).

To test for intercoder reliability and agreement, we used a 3-stage process (recommended by Campbell et al. 2013) that began with an initial codebook. We tested the codebook on 2 randomly selected transcripts coded by 2 independent coders. Discrepancies between coders were resolved through a “negotiated agreement” approach, and we subsequently modified the codebook. We then coded 4 additional randomly selected transcripts and made minor revisions to the codebook, achieving a minimum Krippendorff’s alpha (Krippendorff 2004) of 0.67 for each code. All transcripts were then coded using the final codebook in ATLAS.ti v. 22.1.0.

Interviews: Analysis. —

Analysis of the coded data began with tables describing code frequencies, followed by meta-matrices designed to organize, summarize, and compare data around domains of interest (Miles and Huberman 1994; Miles et al. 2020). We used joint displays to systematically compare interviews and enforcement record results (Guetterman et al. 2015; Cresswell and Plano Clark 2018). Through this process, we identified representative quotes and developed preliminary findings (Miles and Huberman 1994; Miles et al. 2020).

Limitations. —

Results build a clearer picture of the supply chain structure but with some key limitations. Results from official records were limited to cases where turtle trade was detected and reported by federal law enforcement. In this study, these enforcement records were from 1 federal agency (ECCC); others, such as CFIA, the Royal Canadian Mounted Police (RCMP), Canada Border Services Agency (CBSA), and provincial- and municipal-level agencies may inspect or intervene at various points along the Canadian supply chain and enforce laws that fall outside ECCC’s mandate. Although some of these activities are captured in the ECCC enforcement data (e.g., CBSA referred cases to ECCC), some incidents could be missing. Law enforcement agencies outside Canada may also directly influence supply chain patterns at the border. Also, because of limitations in accessing recent enforcement records, a lag exists between these records and the interviews.

Interviews had some limitations. It is impossible to know if there is a difference between those who agreed to participate and those who elected not to participate (e.g., those who chose not to participate could be more closely linked to illegal trade). It is also possible that language barriers limited participation. To limit this effect, francophone participants were given the option to speak in French, but they all had fluency in English. Social desirability bias could shape participant responses to questions based on their perceptions of what is socially acceptable (Grimm 2010). We also note that interviews focused on participants’ perception of trade, so responses may not reflect actual behavior. Although it is impossible to fully mitigate this data-validity issue, we used several techniques to minimize it, including asking questions indirectly, i.e., about other people’s behaviors, as opposed to the participant’s own (see also Bergen and Labonté 2020). Participants were provided assurances about confidentiality, and when participants responded to sensitive questions, we requested examples and prompted for more information when appropriate (see also Bergen and Labonté 2020). Participants often provided unprompted information on illegal trade that was consistent with their responses to other questions. We also triangulated responses with enforcement records to further limit bias and found consistency (e.g., interview discussions of smuggling techniques closely matched case files).

Species. —

The combined interview and enforcement data revealed evidence of 58 unique turtle species traded for pets in Canada (Appendix 1). Six more species were uncovered during border inspections, but their presence in domestic trade is unconfirmed (further inquiry was beyond the scope of this paper). According to breeders and enforcement records, 36 species in the Canadian market are domestically captive-bred (although more evidence is needed to confirm if they are all indeed captive-bred). Forty-eight species in Canadian trade are included in CITES, including eight in Appendix 1. Five of the species identified in trade are native to Canada. Of the non-native species, 12 are Critically Endangered and 12 are Endangered on the IUCN Red List (IUCN 2023).

Supply Chain. —

The activities of the different actors (harvesters, intermediaries, and consumers) fell into the SCOR model’s 3 broad categories (following Huan et al. 2004). In Canada, some actors participate in multiple stages of the turtle supply chain, taking on various roles over time (e.g., a breeder who primarily breeds animals may also sell to end buyers; Table 1).

Table 1. Actor types and stage of the supply chain.^a

View Fullscreen

Source. —

The low availability of turtles for sale in Canada, in terms of the volume of animals and diversity of species, emerged as a common theme. Turtles enter the Canadian market in 1 of 3 ways: legal breeding from existing domestic populations, illegal collection from Canadian wilderness, or illegal importation. Here we describe the different actors and activities at each different source.

Source: Breeding. —

Breeders here are defined as those whose animals produce offspring to sell or trade. Breeders do not necessarily have a brick-and-mortar store or regularly sell other pets, nor do they typically sell husbandry equipment. Twelve interview participants actively bred turtles on a relatively small scale to sell in Canada. Based on enforcement records and interview characterizations, we found no evidence of large-scale breeding facilities in Canada like those seen in the United States and other markets. Breeders in the study reported breeding anywhere from 1 to 20 species of turtle, with some specializing in subspecies and color morphs. Some also bred other reptiles. Reported volumes of turtles bred were typically low, though most participants found it difficult to estimate how many offspring they have each year, especially since breeding success varied between years.

Three types of breeders emerged that we coded as professional, hobbyist, and accidental. Professional breeders have multiple turtle species (and, in some cases, other reptiles or taxa) that they intentionally and regularly breed for sale to the public, wholesalers, or pet stores. Hobbyist breeders are pet owners who occasionally breed their turtles and sell the offspring to recoup some of the husbandry costs. Accidental breeders are pet owners whose turtles accidentally produced clutches, which they sell to retailers or could perhaps sell elsewhere.

Some breeders carefully screened their customers to ensure they were capable of properly caring for the animals, including wholesalers and retailers. For example, a participant noted, “I’m very particular about who I sell my turtles to. I won’t just sell them to anyone. I could sell to a whole bunch of wholesalers, but they don’t do any follow-ups or checks or education or whatsoever” [4130] (numbers in the square brackets represent the interview number or enforcement file number). A few noted they were able to be selective about whom they sold to due to the limited supply.

Source: Harvesting. —

Harvesters collect live animals directly from the wild, which is illegal in Canada. Finding some native species, such as spotted turtles (Clemmys guttata), is considered difficult: “they’re not abundant, like the number of hours needed to find one probably defeats the purpose of financial gain if someone’s exporting them, especially since they’re not regulated in other countries [where they are captive-bred]” [7606], but “apparently there are poachers out there who send turtles and tortoises to Asia” [7606].

Three types of harvesters emerged. Specialists intentionally search for and collect animals from the wild. For example, 1 vendor described an incident where a member of a naturalist group was caught collecting turtles from a wild population that the group had been seeking to protect: “We found out somebody we knew had taken them and sold them to somebody we knew in my hometown” [2220] (a behavior the group quickly responded to with peer pressure, but they did not report the incident to the authorities). Another vendor described an incident where someone was caught with “a bunch of Blanding’s turtles he was taking across the border to trade for something in New York State” [7137]. Opportunists find turtles by chance and seize upon a presented opportunity. Some respondents described people finding a turtle outdoors and bringing it home as a pet, or coming into their stores seeking equipment to care for a wild turtle that they found. “I think people think that they rescue turtles. … They’ll find one on the side of the road, they don’t understand that it is kind of migrating to where it needs to go to be safe … so they pick it off the road and they think ‘I rescued this’ and then they put it in a tiny aquarium” [1323]. Some opportunists post found animals for sale online. Protectors collect animals illegally under the guise of conservation. For example, 1 individual claimed that they had collected wild turtle eggs to protect them from potential predators so they could hatch and be released [Enf file 5636], although it is unclear if their actual intent was to release the hatched turtles.

Source: Import (illegal). —

One of the most cited drivers of illegal activity was the price asymmetry between Canada, where there is limited legal supply, and the United States, where one finds large-scale breeders and suppliers. Some express the belief that higher turtle prices in Canada incentivize smuggling from the United States:

I mean, the price of, like, Red-eared sliders [Trachemys scripta elegan], for example, are worth like peanuts. Right? You’re talking about, like, people want animals that are bred so regularly in the U.S., and what happens is that even though there’s so little, there’s just the demand on them is so high. So here they would sell them … they probably get five times the value. [9197]

There were also descriptions of species being sold in Canada that are not typically bred domestically in large numbers, implying that they may have been brought into the country illegally:

But these illegally sourced turtles and tortoises show up online. So, you got to be very careful because a lot of times, like I’ll see on Kijiji, and stuff like that, where there’s just a pile of Red-eared sliders for sale or other species that aren’t being bred so much in Canada, and you kind of know where they’re coming from, like they’re paying, like a buck a Red-eared slider in the States. They’re just hiding them and sneaking them up here. [2298]

Another participant described the suspicious appearance of pancake tortoises Malacochersus tornieri for sale in Canada:

I don’t know anyone that’s breeding those, but you’ll see, you know, occasionally they’ll pop up, or somebody will say, “Oh, I just bought a pancake tortoise” “Oh really cool, where’d you get it?” “Oh, you know, I can’t really tell you,” sort of, hmmm oh ok, … so they’re coming in from somewhere. [7137]

Another vendor described other suspicious activity: “It’s just something that there’s no zoo had them in Canada. So, there’s no way for the breeding population to be here. But, yet these are available. So just makes you kind of wonder” [2298]. Besides the low likelihood that certain species are domestically captive-bred, some participants described animals in poor health as a red flag for smuggled or wild-caught turtles: “it was pretty obvious by their condition and by the fact that a lot of the species weren’t bred in captivity” [7607].

Importers here are described as intermediaries who transport wildlife across the border, whether naively or knowingly. Illegal imports of turtles were identified in the enforcement records (n = 18, alleged violations) and described by interview respondents (n = 13; Table 2). The enforcement files contained cases where importers claimed they did not know the laws (n = 3). One vendor described how someone declared a turtle at customs when they drove across the border. Other importers appear to move small numbers of intentionally hidden animals across the border. It appears some were for personal purposes; for example, a traveler was found with 3 pet Greek tortoises Testudo graeca in his luggage on a flight from Egypt (he claimed that he did not know it was illegal to import them but declared no animals on his customs form) [Enf file 2708]. Turtles may also be imported in a clandestine manner, presumably by those who recognize the restrictions.

Table 2. Techniques used to import turtles into Canada as described in the ECC WED enforcement files and perceived by interviewed vendors.

View Fullscreen

Enforcement records showed animals primarily smuggled from the United States, but some originated outside North America). One vendor said “I think it’s mostly coming up from the States. Some of it comes in from Asia as well, some of the illegal stuff” [5206]. Land (e.g., cars and trucks), mail, small boat, shipping containers, air cargo, and air passenger baggage were among the transportation modes used.

Participating vendors described the intersection between the legal and illegal supply chains, and observed how it is difficult to distinguish legitimately captive-bred (legal) from smuggled (illegal) turtles (Fig. 1). It is hard to prove the point of origin of turtles once in Canada “they’re basically untouchable once they go across the border, you like, can’t do anything” [3381]. So legal vendors could either knowingly or unknowingly sell illegally sourced turtles. There is currently no formal process for documenting offspring of legitimately bred exotic species in Canada. Illegally sourced turtles can easily move through the supply chain, where numerous opportunities exist for laundering, and people assume that if an animal is openly for sale, it must be legal: “I figured if he was there selling openly on Kijiji that everything was above board” [1079].

View Figure

• Download as Powerpoint
• Download Figure

The extent of illegal trade is inherently difficult to assess. In the opinion of some interviewed vendors, there are low levels of illegally sourced animals in Canadian trade: “I think there really isn’t that much supply, but I don’t think that many people are doing that [smuggling] where it’d be a huge issue anyway. I think you would see it flooded if that was the case, like where people, tons of people were doing it. Like, you’d see tons of shit on Kijiji or whatever, like Craigslist kind of thing” [6722]. However, others perceived it as being more prevalent: “I would say people are definitely smuggling in” [1079]; “I think, though, also the smugglers are supplying more the wholesalers—a couple of them. There was one, in particular, I’m not going to mention names” [4130]. Thirty percent of vendors participating in this study described purchasing turtles (at least once) that they suspected had been smuggled into the country. Some knowingly acquired illegally sourced turtles that were already in Canada and expressed the belief that it is not illegal to purchase these animals once they are in the country. Others described learning that past purchases had likely been illegally sourced when their supplier was later caught smuggling turtles.

Make: Time and Place Utility. —

Time and place utility is added as turtles are moved closer to the consumer in the supply chain. Value is added to turtles by vendors positioning them in the market where they are available for sale to other vendors or end customers; positioning turtles at a point of sale, such as online markets, pet expos, and pet stores, creates time and place utility (Table 3). Pet stores and specialty reptile stores may sell other animals (e.g., lizards, spiders), equipment, and food. Reptile and pet expos, often pet- and reptile-specific events, occur multiple times per year, often at convention centers. Online sales go through websites (e.g., specialty reptile websites, breeder, or pet store websites), social media, or online sale platforms. Online platforms were key points of information exchange, where sales were described as largely occurring on store or breeder websites, Kijiji (an online classified ad website that allows seller anonymity) or, sometimes, Facebook forums. Online platforms were also used by people seeking a particular animal.

Table 3. Sales locations for live turtles in Canada.

View Fullscreen

Appendix 1. Species in trade in Canada (2008–2022) with CITES and IUCN Red List as of July 2023.

View Fullscreen

Make: Intermediary Activity. —

Intermediaries, such as wholesalers and pet stores (retailers), specialize in delivering animals to customers (Table 3; Fig. 1). Wholesalers were described as generally selling a variety of animals, such as aquarium fish, other reptiles (in addition to turtles), arachnids, and amphibians, to pet retailers across the country. Some are also turtle breeders. Retailers (including franchise and privately owned pet and aquarium stores) sell turtles, other pets, and equipment directly to end buyers in-store, at expos, or online. Many of the pet retailers interviewed focused on “specialty” or “exotic” pets (for example, some had birds, but many focused on snakes, lizards, turtles, amphibians, and in some cases, arachnids). One vendor classified retailers into multiple categories: “I think there’s like, kind of two classes, the ones who really care about the hobby and the animals and want to do it right. And then there’s the other ones that are trying to get it for the quick buck, and they see or hear what some people are making off stuff, and they’re more just the greed” [2298].

To source animals, retailers either breed their own animals or purchase them from breeders, wholesalers, or other individuals who have them in their collection. Vendors described 2 types of turtle suppliers: trusted and ad hoc. Those who used trusted suppliers always purchased from a limited set of wholesalers or breeders whom they felt they could trust to reliably deliver quality animals. Other vendors took more of an ad hoc approach and purchased animals based on supply regardless of their relationship with the supplier. These included one-off purchases from wholesalers, professionals, hobbyists, accidental breeders, previously unknown people who walked into their store selling animals, or online vendors (e.g., Kijiji). Although some vendors clearly preferred one strategy, others used both trusted and ad hoc suppliers.

Deliver. —

The final stage in the IWT supply chain is to deliver animals to their buyers. Without demand and the associated buyers, these supply chains need not exist. Our respondents described numerous aspects related to how animals were transported, what demand looked like, and how they selected whom to sell to in the form of varied types of buyers.

Deliver: Final Transit Considerations. —

Vendors (breeders, wholesalers and retailers) openly distributed turtles across the country: “We ship across Canada, from B.C. [British Columbia] or the Maritimes or Saskatchewan, yeah everywhere” [5524]. Turtles were shipped by air cargo (n = 4) or couriers, which can provide next-day delivery. Many vendors (n = 11) described using specialty services that facilitate courier shipments of reptiles through logistics support and packaging: “They check the weather and everything, and then we put heat pads in a box, and it’s an overnight shipping. It comes to the customer the day after. … I’ve never had any trouble, really no disease, no death, nothing” [5301]. Other vendors used courier services directly (n = 2): “I have my own FedEx live harmless reptile license, so I just kind of cut out the middleman, and I ship using FedEx, much cheaper” [7607]. Although most vendors reported overall success with shipping animals, incidents have happened when heat packs failed, or packages were lost for extended intervals, and the turtle being transported died. A few vendors located long distances from a transportation hub commented that they consolidate multiple shipment orders. “You try, and you know, amalgamate where, you know, if I have to drive to the airport … I want to make sure like three or four or five shipments get out at the same time” [1042].

Shipping was occasionally unnecessary when the vendor could deliver the turtle in person, or when the vendor preferred to arrange private transport because of their concerns about the reliability of third-party shipping. “We try to reduce the chance of shipping as best we can” [8755] said one. Another vendor commented that they use their own trucks because they “don’t really believe in utilizing systems that aren’t 100% animal savvy. You can tell me all day long you’ll get a box there within 24 hours and I’ll tell you all day long, there’s too many variables to risk the animal” [9547].

Demand. —

For an IWT supply chain to exist, demand must either exceed or match supply. Our respondents noted that turtles sold very quickly, in some cases before they even hatched: “I already have people on my waiting list for 2022 babies. I have one species that’s pretty much sold out. Like basically the minute they’re hatched, they are pretty much sold out; I have deposits on them” [4130]. This vendor also described a rise in demand for turtles that corresponded to the COVID-19 pandemic: “But I didn’t produce enough to supply the demand, especially since COVID hit, the demand for turtles tripled” [4130]. Although the majority of sellers described turtles selling quickly, a few noted that some turtles would remain unsold for months. Some vendors felt that the number of potential buyers for turtles is small, and therefore it is possible to saturate the market. Turtles require a significant amount of space, and the equipment can be expensive, so people may keep only 1 or a few. Those wanting rare species may only get 1, meaning there are low amounts of repeat business for live animals.

Buyer Selection. —

Like some breeders described above, many vendors claimed to be selective about their customers. They described being hesitant to sell to owners they thought would be ill-equipped to care for the turtle. Some vendors described spending time educating customers about animal husbandry. Many described a family-like connection to their customers. They built relationships to foster repeat customers who return to purchase food supplies and other animals. Some vendors described caring deeply for the welfare of the turtles and were willing to accept returns from former customers in the event they were unable to provide proper care. Such relationships appeared more common between small family-owned vendors and their customers. A few vendors perceived a lack of staff experience at chain stores that they believe translated to lower standards for animal husbandry and less screening of customers than independent stores.

End Buyers. —

The types of end buyers described by vendors varied greatly: “It’s really random, random [sic], we’ve got customers that will shock you” [1323] said 1 vendor, who has experienced everything from “grandmas” to people with “face tattoos”. One vendor said, “I’ve sold tortoises to like a 21-year-old and to an 80-year-old guy, it’s all over the place. It’s very hard to be like, this is the demographic” [3381]. Despite the range of customers, a few patterns emerged. One vendor described 2 common customer types:

[T]here’s two types of clients. First type is they’re there for the kids … they just want a turtle that’s cheap, easy to care for the kids … and we try to dissuade them to buy one of those types … because there’s no such thing as a cheap turtle. Even if the turtle is cheap the equipment to take care of it, it’s going to be very expensive. … And the other type of client are really passionate about a particular species that are looking for that one, and we’re going to try to help them find the particular tortoise they are looking for. [5524]

Parents feeling nostalgia for their own childhood seek turtles for their children: “Usually what I find the most for these turtle buyers are people who had little turtles when they were kids, and now they have kids, and they want to do that for their kids” [2298]. These parents typically look for inexpensive turtles (e.g., Red-eared sliders); as another vendor described, “[the customers say] ‘I want to buy a turtle for my child; I want them to have that experience.’ You’re probably buying the cheapest turtles, you’re probably not buying the $800 red cheek mud, you’re going to buy the $200 red-eared” [1042].

Conversely, some turtle buyers are more committed to long-term turtle husbandry. Vendors described committed buyers and how they typically had a preference for either turtles or tortoises. Since turtles are generally aquatic, they require quite different care from terrestrial tortoises. “They’re very different, they’re not as, like anything alike … their care is completely different” [5206]. Whether they preferred turtles or tortoises, committed buyers were perceived to take more time to do research on care than people looking for an inexpensive turtle as a pet: “people come in, and they ask for specific species, they know, they’re very well prepared. It’s very extreme, the difference actually” [5206]. Committed buyers were described as keeping only 1 or 2 turtles, particularly if they lived with limited space in urban centers. However, those in rural areas “have more land, tend to keep groups … where it gets very cold in the wintertime, people build some pretty crazy things on, like, in barns or in basements” [5206]. Some of the committed buyers were also reported to have written considerations for their turtles in their wills.

The last type of end buyers that emerged were collectors who had multiple species and sought new and rare species to add to their collection: “The collectors will start with the beginner one and go, ‘well it’s nice to be a little bit rare, something a little bit more difficult, something a little bigger, something you know maybe aquatic’, and so they’ll just start with the basic to get [started] and then they’ll fall in love” [4130]. Some become repeat customers, “there’s a [name of profession removed] actually in [name of city removed] that usually contacts me once or twice a year to see what I have available” [1079].

Supply Chain Characteristics. —

The Canadian restrictions on the import of live turtles for commercial purposes mean that legal supply for pet turtles is reliant on domestic breeders. Our data suggest that in comparison to the United States, Canada appears to have relatively few breeders. Evidence of captive breeding is missing for more than one-third of the non-native species currently being traded. Further research is needed to determine if any of these species are actually being captive-bred in Canada or if they are likely to have been wild-caught or captive-bred abroad and smuggled in. As found in the CEC report (2017), demand for turtles in Canada (including rare species) and higher prices than in neighboring countries were seen by participants in this study as an incentive for smuggling.

Many of the actors in the supply chain aligned closely with those described by Phelps and colleagues’ (2016) framework, except for one: people who harvest turtle eggs under the guise of conservation, with the stated intent to protect them from predators by headstarting in captivity, a practice used by conservationists for species recovery (e.g., for wood turtles in Ontario; Mullin et al. 2020). This narrative fits into Shukhova and MacMillan’s (2020) typology of lifesavers who keep animals because they want to save them. It highlights protection of wild populations as a possible motivation (or potentially an excuse) for illegal collection.

Most study participants could not provide a singular description of end buyers, instead describing them as existing along a spectrum. However, two typologies consistently emerged that appear to align with the concept of new experience seekers (e.g., parents looking for turtles for their children) and collectors looking for unique pets (Shukhov and MacMillan 2020). Inexpensive Red-eared sliders, not bred extensively in Canada, can be (illegally) sourced from the United States to meet the demand for inexpensive children’s pets. In contrast, other buyers are willing to pay higher prices for specific or rare species to expand their collection. Further theoretical inquiry and research are needed to identify social marketing benchmarks (e.g., segmentation) for the development of evidence-based demand reduction strategies (Greenfield and Veríssimo 2019) that will effectively target the various actors, particularly on the demand side of the supply chain.

Our study also revealed several vendor types. Brick-and-mortar stores and expos are major points of sale, but online sales emerged as an ongoing theme consistent with the trade in turtles (Mandimbihasian et al. 2020; Bielby et al. 2023) and other taxa abroad (Lavorgna 2014; Stringham et al. 2021; Marshall et al. 2022). Although some sales occurred through vendor websites, Kijiji was consistently mentioned for turtle sales in Canada. However, scam ads from foreign countries occasionally appear, as they also do in other online wildlife markets (e.g., Indonesia; Morgan and Chng 2018) and can be problematic. Wildlife trade assessments often survey online markets to identify and quantify the species in trade (Stringham et al., 2020), so fraudulent ads could undermine these investigations and lead to inaccurate estimates of trade volume and the rarity of species involved; however, we note that overall, trade is likely underestimated (Hamers et al. 2023).

Intersection of Legal and Illegal Supply Chains. —

The sale of Canadian-bred non-native turtle species in Canada is legal (except in some provinces and municipalities). Yet our study reveals ongoing detected and undetected illegal imports intersecting with the legal supply chain. The scale of the problem is difficult to assess. A few study participants described the level of illegal activity as low, but most vendors were aware of illegal trade, and almost one-third openly described having purchased animals that they suspected were likely sourced illegally. However, identifying the origin of turtles (e.g., if they are captive-bred in Canada) is a challenge because no formal mechanism exists for tracking this information, or a way to easily tell the difference between a Canadian-born and non-Canadian-born turtle. Therefore, once within the borders of Canada, it can be extremely difficult for law enforcement to conclusively identify smuggled animals. Except in cases where a species was known not to exist in captivity in Canada before the implementation of the restriction on turtle imports in the 1970s, all non-native species in-country at any point can, in theory, be represented as legitimate. At the time of writing, the only point at which an illegal-sourced non-native turtle can be conclusively designated as such is during the act of crossing a border into Canada (or unless there is strong evidence that there are no captive breeding populations in Canada).

To conceptualize the intersection between legal and illegal turtle trade in Canada we consider different patterns, which we represented as a capital letter “Y” for non-native species and a line for native species (Fig. 2). For non-native species, one finds two streams of animals, legal on 1 side and illegal on the other. Once illegally sourced animals enter Canada, the two populations can mix (and the branches of the letter Y intersect), after which all can be represented as legal. This intersection, especially for non-native turtle species that are known to have been in-country before the restrictions came into effect, encourages trafficking (i.e., if a turtle is transported across a border undetected, it becomes saleable in the legal turtle market). Even experts have difficulty distinguishing legitimately sourced animals from illegally sourced animals laundered into legal trade (Nijman and Shepherd 2015b; Brandis et al. 2018; Leupen and Shepherd 2018). Native species cannot be legally harvested or traded, so even on apparently “legal” platforms (e.g., open sales online), the trade remains illegal. This can be represented by a line where the market changes, but the status does not. This could also occur with non-native species where evidence is conclusive that they have not been legally imported before or since the 1970s restrictions. Theoretically, legal and illegal supply chains could also intersect in an “X” pattern. In this case, some illegal animals are distinguishable (e.g., through DNA or stable isotope analysis), while others are not.

View Figure

• Download as Powerpoint
• Download Figure

Implications for Research and Theory. —

Applying supply chain models to the problem of illegal live turtle trade can promote an understanding of the creation of time and place utility and, in turn, of the factors and contexts that help move illegally sourced animals into legal markets by concealing their origins. By integrating illegal markets within such models, researchers can establish connections with theories from other disciplines, such as criminology. For instance, opportunity theories of crime, which explore the intersection of motivated offenders and suitable targets in the absence of capable guardianship (e.g., law enforcement) in specific spatial and temporal contexts (Cohen and Felson 1979), can be integrated with the supply chain model. Additionally, other areas of criminology, such as research on white-collar crime (Chan and Gibbs 2019), criminal capital (McCarthy and Hagan 1995; Loughran et al. 2013), and informal guardianship (Viollaz et al. 2022), may provide theories and frameworks to help inform our understanding of the underlying mechanisms of illegal turtle supply chains that collide with legal. This interdisciplinary approach, drawing on knowledge from conservation sciences, supply chain management, criminology, and other social sciences (Gore et al. 2023), presents an opportunity to gain crucial insights into the vulnerabilities of legal trade to illegal activity and the patterns that may unfold (e.g., the X and Y presented above). By leveraging diverse perspectives, researchers can develop comprehensive strategies for combating illegal wildlife trade to better understand the dynamics of illegal supply chains.

Implications for Practice. —

Legal trade is often presumed to be sustainable, but concerns are ongoing that it can mask illegal or unsustainable trade (Hughes et al. 2023), with examples of this occurring in the Canadian turtle trade. As reported by our study subjects, legal and illegal turtles are moving through Canadian supply chains because not enough has been done to stop the trade. Pursuing process improvements (e.g., inter- and intraorganizational processes, improved training, cross-sector collaboration; Duensing et al. 2023), developing new policy measures, regulating trade in captive-bred animals, and identifying captive-breeding operations and marking systems (Davies et al. 2022) are several approaches we suggest investigating to address these issues.

Process improvements could be made at the business level. Vendors in this study represented themselves as taking time and care to select responsible buyers for their turtles, and some vendors had strategies for purchasing from trusted suppliers. Although we recognize the possibility that these statements could be shaped by social desirability bias and that they may, in practice, take less care, it is nonetheless worth considering outreach to the industry to encourage vendors and buyers to be more discerning about their supply chain and focus on ensuring the animals they sell are legitimately sourced. Fostering cross-sector collaboration (Duensing et al. 2023), such as engaging couriers and online sellers, may also provide avenues for disrupting the illegal trade. Davies and colleagues (2022) also suggested forensic tools to distinguish captive-bred from wild-caught animals, for example, isotope markers (van Schingen et al. 2016; Brandis et al. 2018; Andersson et al. 2021).

Our findings improve our understanding of the nature and structure of the Canadian live turtle supply chain, giving context for identifying opportunities for other interventions. Under the current federal laws, it is difficult for federal law enforcement to target trade in animals that have been illegally sourced but are already in the country. Although some parts of Canada have provincial and municipal regulations on the possession of certain taxa, it is worth exploring the development of legal frameworks that would specifically target the possession of illegally sourced animals. Additionally, Situational Crime Prevention techniques and tools designed to reduce opportunities for wildlife trade (Lemieux 2014; Glusek et al. 2022) could be used at key points along the supply chain to make it more difficult to trade illegally sourced animals.

Conclusions. —

In this paper, we present data describing the overlap between legal and illegal supply chains in Canada’s turtle trade. Once in-country, a turtle (legal or illegal) from a species known to exist in Canada prior to import restrictions in the 1970s can be nearly irrefutably claimed to be legal. Addressing this issue will require a collaborative effort that draws on expertise from different fields and agencies to develop a strategy to effectively tackle the complex intersection between legal and illegal supply chains.