1.                 When Professor Harry Arthurs introduced the notion of “dependent contractor” to Canadian legal lexicon in his seminal 1965 article[1] – a concept that was adopted by the Ontario Legislature in 1975 – he could not have envisioned that it would apply to couriers using electronic software application (“App”) on a smart phone to deliver a customized meal assigned by an algorithm without any direct communication or direct payment with the customer.  Foresight of technology was unnecessary because he contemplated a classification of worker to fill the void between an employee and an independent contractor that had sufficient elasticity to adapt to new workplaces and innovative modes of service delivery.  Fifty-five years later, Professor Arthurs’ proposal has withstood the test of time and made its way into what is colloquially known as “the gig economy”.

2.                 This is an application for certification filed by Canadian Union of Postal Workers (“CUPW” or “the union”) under the Labour Relations Act, 1995, S.O. 1995, c.1, as amended (the “Act”) seeking to be the exclusive bargaining agent for a group of couriers working in Toronto and Mississauga for Foodora Inc. (“Foodora” or “the employer”).

3.                 The Board conducted an electronic vote commencing on August 9, 2019 and the ballot box was sealed pursuant to the direction of the Board.  The primary issue at this stage in the proceeding is whether the individuals subject to the application are dependent contractors (as argued by the union) or independent contractors (as argued by the employer).  There is also an unfair labour practice complaint filed by CUPW (Board File No. 1376-19-U), which has been adjourned sine die on agreement of the parties.

4.                 The parties tendered an Agreed Statement of Facts that was supplemented by sworn declarations of the witnesses and viva voce evidence.  CUPW called  Houston Gonsalves, Ivan D. Ostos, Brice Sopher and Samuel Tyler, all couriers for Foodora that it considered to be representative witnesses.  CUPW also called evidence from Dylan Boyko, a former dispatcher with Foodora who was a courier at the time of the hearing.  Foodora called Alex Paterson, Head of Rider Management for Canada.  I found all of the witnesses to be credible and forthright.

5.                 As will be apparent throughout this decision, the documentary evidence filed by the parties was very important to the overall analysis. 

6.                 Foodora is a web services company that provides an online marketplace platform connecting consumers to restaurants.  It established itself in the Canadian marketplace when it purchased Hurrier in 2015.  The Board heard no evidence about Hurrier’s operations.  Couriers engaged by Foodora are granted access to the platform through the App.  The App is owned by Foodora.  Couriers use the App to access Foodora’s dispatch system, allowing them to receive delivery opportunities.

7.                 Foodora charges restaurants and customers directly for its services, including accessing the App, marketing, and the services provided by the couriers.  Foodora determines the rates charged to restaurants and customers.  The restaurants and customers of Foodora do not have direct access to the couriers.

8.                 Couriers enter into written agreements with Foodora stipulating the terms and conditions of their engagement.  These agreements expressly state that the couriers are independent contractors.  These contracts are amended by Foodora from time to time.  Mr. Paterson explained that when Foodora purchased the company – Hurrier – it continued using the contract template for couriers.  Over time, Foodora made amendments to those contracts in consultation with its lawyers.  However, it did not revisit the contracts with existing couriers.  The Board was not advised of the differences in the contracts.

9.                 Although Mr. Paterson testified that nothing prevented a courier from proposing his own amendments to the contracts, he conceded that Foodora did not offer to negotiate and that no courier had ever proposed to amend the contract.  After reviewing the contract and considering the evidence of the witnesses, I find that there was no opportunity for a courier to negotiate or amend the contract. 

10.              Foodora recruits couriers through its website and uses an online registration system for retaining couriers.  As part of this process, couriers are required to confirm that they are eligible to work in Canada, that they are over the age of 19, and must indicate their language preference and vehicle type.

11.              Foodora does not conduct hiring interviews.  Couriers may be asked to complete a short screening via telephone or SMS message – a type of text messaging using a smart phone.  Foodora does not require reference checks, criminal background checks or other similar requirements.

12.              Couriers provide most of the tools to perform the work.  These are as follows:

13.              Foodora does not inspect the tools of the couriers other than to ensure the food delivery bag is of sufficient size and condition to transport food.  This is usually done by the courier sending a photo of the food delivery bag to Foodora.  Mr. Tyler testified that after inspection by Foodora, he was asked to purchase a new courier bag.

14.              Foodora does not prohibit couriers from utilizing tools or wearing clothing that are branded with its competitors’ names, logos or similar identifiers while performing services on behalf of Foodora.

15.              Couriers are responsible for repairs and maintenance to their bicycles and vehicles.  Several of the couriers testified that they purchased new tires, brake pads, as well as other parts and paid for various maintenance services.  Couriers also purchase helmets, locks, gloves and rain gear.  Mr. Gonsalves, who drives a vehicle for Foodora in Mississauga, was responsible for maintaining his own insurance as well as regular vehicle maintenance.

16.              Couriers are given an orientation session by Foodora on how to use the App, which usually lasts for 30 minutes.  Foodora does not provide any training on the use of courier-supplied tools or the performance of the work itself beyond the use of the App.

17.              Foodora publishes a “Rider Guide” online, which sets out how to use the App and Foodora’s delivery policies.  In addition to the Rider Guide, Foodora uses different social media platforms to engage with couriers and the public.  Foodora uses the Heymarket platform to communicate with couriers on shift through Foodora’s dispatch using SMS messaging.  In essence, Foodora’s dispatchers can communicate with couriers using a text message software that is logged and tracked.  The Board received many pages of text messages, which will be discussed in greater detail.

18.              Couriers are assigned shifts through the App.  The software portion of the App is referred to as “Rooster”.  Foodora uses an algorithm to create a shift schedule based on anticipated demand.  The lengths of the shifts vary and are set in geographic zones within the City of Toronto and the City of Mississauga.

19.              The available shifts for the following week are released to couriers each Wednesday.  Couriers can select shifts based on three prioritized selection tiers.  Couriers who are ranked in the top 30% of total couriers are provided first priority to select shifts commencing at 10:00 a.m. each Wednesday.  The remaining 70% of couriers are provided access to view and select shifts commencing at 11:00 a.m. each Wednesday.  Couriers who have been idle and have not worked for four consecutive weeks are then provided access to select available shifts at 11:30 a.m. each Wednesday.

20.              Additional shifts are added to the schedule by Foodora on an ad hoc basis based on anticipated customer demand.  Couriers can log into the App to view and select these additional shifts.

21.              Couriers are also free to message dispatch to ask if any additional couriers are needed.  If dispatch determines, based on demand, that additional couriers are needed, a shift is created for the courier at the time.  A courier cannot simply start working for Foodora without the approval of dispatch.

22.              A courier may drop a shift with 24 hours’ notice without consequence.  If a courier wants to drop a shift within 24 hours’ of the start of the shift, the courier may seek to swap the shift with a fellow courier.  Under these circumstances, the shift exchange takes place through the App.  It is not uncommon for couriers to discuss shift exchanges outside the App to ensure someone is willing to take the shift, or attempt to ensure that a friend is able to pick up the shift as soon as the courier releases it back on the App.  However, a direct shift swap between couriers is not permitted.

23.              If a shift swap is not possible, couriers may notify Foodora and provide an explanation for their inability to work the selected shift.  If an explanation is provided to Foodora prior to the start of the shift (e.g. illness), then the courier will not be listed as a “no show”.  A “no show” is one of Foodora’s rider performance metrics, and a courier’s rate of “no shows” will impact the courier’s prioritization rate for shift selection, though it is not determinative of such rate.

24.              Most of the evidence tendered before the Board dealt with couriers who use bicycles to deliver food.  However, Foodora requires couriers working in some geographic zones (such as the west zone in Mississauga) to use a car due to the distance between deliveries.  Mr. Paterson testified that Foodora would consider allowing couriers to use other modes of transportation such as electric bikes (or e-bikes) if the courier could satisfy Foodora that the courier could meet the service standard in a safe and lawful manner.

25.              The ordering and delivery process is as follows.  A customer who is signed up for Foodora’s service on the App is able to place an order through the App with a restaurant that is signed up with Foodora.  That restaurant receives the order (on a separate handheld device provided by Foodora) through the App.  The App then sends the delivery opportunity to the closest available courier (using an algorithm that takes into consideration certain metrics including proximity).  The courier can either accept or decline the order.  I will have more to say about Foodora’s changing practice when a courier declines an order.  If the courier accepts the order, the courier proceeds to the restaurant to pick the order up.  The courier is advised the approximate time the order will be ready and the time promised for delivery to the customer.  Once the order is ready, the courier takes the order to the customer’s location using the most optimal route as determined by the App.

26.              There are variations to the process.  A large order or corporate order might be manually assigned to the courier by dispatch.  A courier may arrive at the restaurant to discover the order is delayed or that a second order has been added by dispatch (the latter incident is referred to as stacking).

27.              There is no specific training for the courier, nor are there prescribed standards other than the requirements of the delivery bag.  There is no cash in the transaction as the customer pays online through the App, and Foodora, acting as the intermediary, transacts directly with the restaurant.

28.              Issues that arise during delivery (e.g. a customer does not answer the door) may be resolved directly by the courier or may be reported to Foodora’s dispatch.  Although the parties agreed that Foodora encourages couriers to troubleshoot and resolve issues directly, and provides support through dispatch, there are limited options for a courier.  The courier cannot alter the price to offer a discount for an issue, nor can the courier provide a coupon for a future delivery.  On the other hand, couriers may contact the customer directly to inquire about a substitution at the same price if an issue arises at a restaurant.  Any complaints by customers or restaurants may be expressed directly to couriers or to Foodora’s customer support by the courier.

29.              Where a delivery cannot be completed because a customer is not home or because of an incorrect address, couriers will often first attempt to contact the customer directly through the information made available on the App.  If the courier is unable to contact the customer, the courier is expected to direct the issue to dispatch.  If the customer cannot be reached, dispatch will ask that the courier keep the order with them and continue to their next order.  If the customer is subsequently located, dispatch will inform the courier and the courier will complete delivery of the earlier order prior to accepting their next order.

30.              Foodora determines which courier is offered orders based on its algorithm that takes into consideration the GPS location of the courier, the restaurant and the customer.

31.              There are circumstances where multiple orders are placed at the same restaurant or nearby restaurants by customers who are in proximity.  In these circumstances, Foodora will determine that the most efficient way to deliver the orders is for the assignment to be given to the same courier.  This is referred to as “stacks” or “doubles” or “triples”.

32.              Delivering “stacks” is more lucrative for the courier because the courier is paid double for the same kilometers.  While this is most often determined by Foodora without consulting the courier, there was at least one occasion where Mr. Tyler suggested to dispatch that “stacking” would be better.

33.              Through the App, Foodora determines the most efficient route for the courier to take to complete the delivery.  However, the evidence was that the courier could change the route, and with permission of the dispatcher, could re-arrange the order of delivery if the courier was “stacked”.

34.              According to the Agreed Facts, couriers are contractually obligated to provide services in an efficient, effective, competent and professional manner.  Should a courier fail to meet such standards, whether through complaints, no shows, late logins for selected shifts, or similar issues, Foodora retains discretion to take appropriate action.  Such action may include a discussion with the courier, changing a courier’s scheduling priority or available shift zones for scheduling, or deactivation of their engagement.

35.              Foodora has the discretion to activate or deactivate couriers, in accordance with the terms of their agreement.  This occurs in several ways.  If a courier does not sign up for a shift over a substantial period – the evidence was anywhere from four to six weeks – the courier is deactivated, but not before being given advanced notice.  If a courier responds and indicates that they intend to resume services in the future, their account may not be deactivated.  Furthermore, if a courier’s account has been deactivated for non-use and the courier wishes to resume providing services, the courier may notify Foodora of such and their account may be reactivated at Foodora’s discretion.  Mr. Paterson also testified that the list of couriers is reviewed on an annual basis to ensure there has been activity.  Couriers are deactivated if they were not active during the winter months.  This allows Foodora to ensure that there are enough shifts in the summer months when it is less busy.

36.              Compensation is determined by Foodora in accordance with the contract.

37.              Couriers typically receive $4.50 per order, plus $1.00 per km between pickup and drop off, as well as any tips or applicable incentives.  Couriers may request and negotiate adjustments to kilometer calculations to reflect route challenges or changes, changes to drop off locations, and similar issues.  This might occur where there are parades, large outdoor events, or severe weather conditions.

38.              In order to incentivize couriers in designated zones with lower customer demand, Foodora offers a base level guarantee of compensation equal to $16.00 per hour, subject to the courier meeting established performance metrics (e.g. completing shift, commencing shift on time, completing all deliveries offered while on shift).  In such circumstances, Foodora conducts a weekly reconciliation of actual compensation received by the courier (inclusive of tips) against the hours of work performed by the courier at a rate of $16.00 per hour.  Foodora then provides a “top up” payment to ensure that the courier has received compensation equal to a rate of $16.00 per hour.  If a courier fails to meet the required performance metrics, then the courier receives their normal rate without any “top up” payment.

39.              Mr. Gonsalves testified that he was permitted to decline orders, but that this would disentitle him from the guarantee of $16.00 per hour.

40.              Occasionally, Foodora offers “boosts” whereby a courier can make more money per delivery should they pick up a shift during peak periods.  These premiums are used to try to ensure that Foodora has enough couriers to cover all shifts.  The amount and timing of these boosts are solely determined by Foodora.

41.              Foodora makes no deductions from courier compensation for tax, employment insurance or Canada Pension Plan purposes.  Payment to couriers is processed and transferred through a third-party payment system, VersaPay, on a weekly basis.  Couriers are required to maintain a VersaPay account, and payments are accessed by couriers through such a third party.

42.              Couriers are not enrolled in any Foodora-sponsored group health benefits.  Couriers access WSIB through Foodora.  If a courier is injured, they are requested to report their injury to Foodora, however, not all couriers do this.  There was no evidence tendered by the parties about access to WSIB benefits or claims that have been made.

43.              Spills are tracked by Foodora via “Spill Reports”, which couriers are requested to fill out where an order they are tasked with has spilled.  Foodora uses this data to advocate for better containers with restaurants with whom it has a contractual relationship.  The parties stipulated the following fact:

44.              When an order is not ready for pick up at the expected time, Foodora provides a $5.00 payment to the courier for each period of twenty minutes the courier waits past the agreed restaurant pick up time.  Foodora charges restaurants a penalty for excessive wait times as a disincentive for future delays.  The $5.00 payment provided to couriers is a portion of the penalty imposed by Foodora against the restaurant in question.  Couriers are provided with the $5.00 compensation regardless of whether the restaurant at issue pays any penalty.  Dispatch must approve the payment of any compensation and couriers may be required to provide information proving that they are entitled to the compensation.

45.              Foodora employs employees in a variety of positions, including account managers and customer service employees to perform a host of duties including managing relationships with restaurants, marketing and supporting courier functions through dispatch.  Foodora has a team of employees tasked with recruiting new restaurants and a team of employees who are responsible for maintaining relationships with the existing roster of restaurants.

46.              The Board heard brief evidence about a pilot project in Ottawa whereby couriers deliver alcohol to customers.  The Board was advised that these couriers are treated as employees of Foodora.

47.              Foodora operates office hours for its Toronto office, during which couriers can attend to ask questions or pick up Foodora branded material.  The evidence was that Foodora created the office hours to provide enough support to couriers at specified times.  However, couriers are free to come to the office at any time.

48.              The parties narrowed the evidence of the couriers to four couriers as it was impractical and unnecessary to call numerous couriers as witnesses.  Through these witnesses the Board heard some evidence about their dependency on Foodora for income and economic wellbeing.

49.              All four couriers who testified acknowledged that they worked for other courier delivery services and performed other services.  It was also apparent from a chart prepared by Foodora that the four witnesses worked on average more shifts for Foodora than other couriers.

50.              Mr. Gonsalves worked approximately five shifts per week and earned $9,595.24 from December 31, 2018 to July 31, 2019.  He also worked as a courier for Skip-the-Dishes (a competitor of Foodora) (“Skip”) and delivering newspapers for Metroland Media Group.  He earned $7,405.31 from Skip in 2018 and $2,007.66 from January 1, 2019 to July 31, 2019.  He earned $640.72 from May 9, 2019 to August 12, 2019 delivering newspapers.

51.              Mr. Ostos also delivers for DoorDash and UberEats (both competitors of Foodora), in addition to working as a musician.  As of July 31, 2019, he had only worked one shift for DoorDash (earning $28.16).  From April 2018 to the end of August 2018, Mr. Ostos earned $1,1718.15 from UberEats and was unable to work the rest of the year due to an injury.  From January 1, 2019 to July 31, 2019, he earned $3,677.21 from UberEats.  During this same period, he earned $7,063.58 from Foodora, and earned $19,290.56 from January 1, 2018 to September 2018.

52.              Mr. Sopher also delivers for UberEats and earns some income (approximately $10,000 in each of 2018 and 2019) from his work as a musician/DJ.  From March 15, 2019 to August 4, 2019, he earned $4,361.11 from UberEats.  From Foodora, he earned $16,499 in 2018 and $5,134.19 from January 1 to August 4, 2019.

53.              Mr. Tyler also delivers for Skip and UberEats as well as working at a café.  From January 1, 2019 to August 5, 2019, he earned $1,910.89 from UberEats and had not worked for Skip by July 31, 2019.  In 2018, he earned $8,469.47 from the café and $3,160.90 between January 1, 2019 and July 31, 2019.  From Foodora, he earned $20,190.96 between January 1, 2019 and August 4, 2019.

54.              The four representative witnesses all earned an honorarium of $20.00 per hour from CUPW for time spent during the organizing campaign.

55.              Couriers can take a break at any time in their shift by advising dispatch that the courier needs a break.  Upon request and approval, the courier is taken offline by dispatch.  The Board was shown numerous examples of couriers asking for (and always being granted) breaks for issues with a bike or personal issues.  Although some witnesses such as Mr. Tyler suggested that he has been denied a break request, I was not shown any example of the communications with dispatch where this occurred.  There does not appear to be a limit on the number of breaks or the duration of breaks taken by couriers.  However, in order to be on break and not subject to delivery assignments, the dispatcher must approve the break and put the courier offline.

56.              Couriers were also put on break by dispatch if the courier declined an order.  While the couriers believed that this was a punitive measure to deter declining orders, Mr. Paterson testified that there was an operational reason for placing the courier on break.  He explained that by placing the courier on break, it removed the courier from the queue and therefore avoided a situation where the system reassigned the same order to the courier.  At some point, Foodora reduced the mandatory break period from five minutes to one minute.

57.              There were numerous examples of couriers asking dispatch to be reassigned or re-dispatched for various reasons that included both personal reasons (e.g. personal preference) or operational (e.g. courier felt it was more efficient to not end shift or delivery in a particular area)

58.              As explained earlier, couriers select shifts to work based on the shift schedule provided by Foodora.  Once a courier signs up for a shift, the courier is expected to work the shift or find a substitute.  If the courier wishes to start early, start late, leave early or work late on the shift, the courier must secure permission from dispatch.

59.              The Board was shown numerous examples where the couriers made such requests to dispatch and were provided with a response to their request.  It seems apparent that the response from dispatch was based on demand, anticipated or actual, at the time the request was made.  Similarly, there were times when the dispatcher invited the courier to work longer than the stipulated shift and the courier was free to accept or decline without repercussion.

60.              On the occasion when the courier sought to end their shift early, dispatch almost always granted the request.  While dispatch might encourage the courier to stay, it was clear that the courier’s request would almost always be granted.  However, Mr. Tyler explained that he rarely asked to leave early or to create a shift for him unless he knew the circumstances allowed for such a request.

61.              Dispatchers record problems with couriers in a Rider Feedback Form that was also referred to as a Strike Log.  The Rider Management Team reviews the Strike Log and follows up with the courier when deemed necessary.  Such issues include refusal to accept an order, refusal or repeated failure to indicate on the App that a delivery was complete, being late for shift, or engaging in inappropriate communications.

62.              The expectations for how and when dispatchers record strikes are set out in the Logistics Guide.  The Logistics Guide was described by Mr. Boyko as a “training guide” or “best practices guide” that is available online for dispatchers.  Mr. Boyko authored parts of the Logistics Guide and explained that it was approved by his superiors.

63.              There are three types of strikes: Low, Medium and High.  I will briefly set out the types of strikes that are available for dispatchers to issue against couriers.  I will go into greater detail when I explain how this demonstrates the level of control Foodora has over the couriers.

64.              A low strike is to be issued in the following circumstances: App mis-clicks (a mistake by the courier) that are not immediately reported to dispatch; unreasonable re-dispatches (when the courier asks for an order to be reassigned); leaving in the middle of a shift without asking; poor food handling; purposely delaying order; poor conduct with dispatch; signing in as the wrong vehicle type; wrong order delivered/picked up.

65.              Couriers are advised to issue medium strikes in “…situations that are severe enough that is clearly a courier not following procedure and negatively impacting [an] order”.  This applies when blame can be attributed to the courier, not the restaurant or customer.  These situations are listed as follows: “Not following a reasonable unreachable procedure and severely impacting delivery; refusing order after pick-up; disappear mid-shift in understaff/key zone with zero explanation; obvious pickup of incorrect food; inappropriate communication with customer; spillage (obvious negligence); attempting to work without adequate equipment; refusing orders/to move locations within a guarantee zone; signing in then immediately being unable to work; refusing to accept or decline an order”.

66.              The Logistics Guide states that High Strikes are “…reserved for situations which require immediate follow up from Rider Management and will result in some form of RM feedback to the courier”.  These are described as follows: criminal issues (e.g. assault, thefts); threatening dispatch; severe negligence and refusal to do due diligence; harassment; and refusing an order after pickup – repeat occurrence.  The Guide further directs a dispatcher to assign a high strike to the courier when the following circumstances are present “…took food, never gave to customer; disappearing mid-shift for closing shift or high-leverage hours; GPS spoofing to avoid receiving orders; completely unexplained delays of 30+ minutes or more on an order; running multiple delivery platforms where the quality of Foodora orders is negatively impacted.”

67.              Couriers who received strikes jeopardized their priority status and thus their access to preferred shifts.

68.              There were times (albeit few) when dispatch unilaterally ended a courier’s shift before it was scheduled to end.  Mr. Tyler’s declaration indicates that this occurred to him.

69.              There was extensive, albeit inconsistent, evidence about dual apping.  Dual apping occurs when a courier is working for Foodora and simultaneously logs in with UberEats, for example, to make deliveries.  The courier is essentially juggling both delivery services at the same time.

70.              For some time, Foodora prohibited dual apping.  However, in spring 2019 it changed its position to allow dual apping provided it did not interfere with the service delivery standards of the couriers on shift.

71.              As was apparent from the evidence, it is to the couriers’ advantage to dual app since the courier, while on shift with Foodora, can easily log on to UberEats to see what that service is paying for deliveries.  The Board heard evidence that other delivery services occasionally paid more than Foodora, and thus made deliveries for the other service more attractive.

72.              Mr. Tyler testified that he was unaware that he was permitted to engage in dual apping.  He acknowledged that he engaged in the practice, but was careful not to alert dispatch.  It was his understanding that dispatch monitored his movements and would be able to detect a travel pattern that suggested that he was delivering for another service.  There was one example when he told dispatch he was checking the UberEats App, but he explained that he did so to check his internet connection.  Mr. Tyler’s evidence was that he was uncomfortable telling Foodora that he was working for both Foodora and UberEats at the same time.

73.              Mr. Ostos acknowledged that he engaged in dual apping, but kept this practice secret from Foodora out of fear of consequences.  During cross-examination, Mr. Ostos candidly acknowledged if a restaurant preparing a Foodora order was delayed for 20 minutes or more (for which he earned a $5.00 payment), he would occasionally make deliveries for UberEats during the waiting period.  He was presented with a comparison of Foodora and UberEats delivery records showing this practice.  Mr. Ostos testified that if he was caught by Foodora he believed that he would be disciplined.  However, as an experienced courier, he understood what he could deliver for UberEats in a short timeframe and avoid detection from Foodora.

74.              The Board is tasked with determining whether Foodora couriers are dependent contractors within the meaning of the Act.

75.              In their submissions, the parties made reference to the following authorities: Abdo Contracting Company Ltd., 1977 CanLII 422 (ON LRB); Nelson Crushed Stone, 1977 CanLII 409 (ON LRB); Indusmin Limited, 1977 CanLII 520 (ON LRB); Canadian Fuel Marketers Group, Inc., 1978 CanLII 593 (ON LRB); Blue Line Taxi Co. Limited, 1979 CanLII 950 (ON LRB); A. Cupido Haulage Limited, 1980 CanLII 748 (ON LRB); Niagara Veteran Taxi, 1981 CanLII 958 (ON LRB); Algonquin Tavern, 1981 CanLII 812 (ON LRB); Journal LeDroit, 1985 CanLII 1047 (ON LRB); Cradleship Creche of Metropolitan Toronto, 1986 CanLII 1397 (ON LRB); Hamilton Yellow Cab Company Limited, 1987 CanLII 3130 (ON LRB); Carpino Carpentry Ltd., 1991 CanLII 6081 (ON LRB); Diamond Taxicab Association (Toronto) Limited, 1992 CanLII 6786 (ON LRB); Huntsville District Memorial Hospital (Algonquin Health Services), 1998 CanLII 18261 (ON LRB); Rizzo & Rizzo Shoes Ltd. (Re), 1998 CanLII 837 (SCC); Toronto Star Newspapers Ltd., 2001 CanLII 6537 (ON LRB); Ian Dejordan c.o.b. IDM Refinishing, 2003 CanLII 6526 (ON LRB); Excel Forest Products Ltd., 2004 CanLII 3224 (ON LRB); Gateway Delivery Ltd., 2004 CanLII 33775 (ON LRB); Torbear Contracting Inc., 2005 CanLII 35124 (ON LRB); Greater Essex County District School Board, 2010 CanLII 47130 (ON LRB); Hamilton Cab, 2011.CanLII 7282 (ON LRB); Blue Line Transportation Ltd., 2012 CanLII 38608 (ON LRB); Royal Tek Stucco Ltd., 2012 CanLII 4761 (ON LRB); Timothy Ronald Spicer o/a T&R Construction, 2017 CanLII 14550 (ON LRB); Gates of Humber Ridge Inc., 2017 CanLII 60996 (ON LRB); Superior Sand, Gravel & Supplies Ltd., 1978 CanLII 467 (ON LRB); Sherman Sand and Gravel Ltd., 1978 CanLII 563 (ON LRB); Egg Films Inc, 2012 NSLB 120 (CanLII); Egg Films Inc. v. Nova Scotia (Labour Board), 2014 NSCA 33 (CanLII); Peter Barnacle, Michael Lynk & Roderick Wood, Employment Law in Canada, 4th ed, vol 1, loose-leaf (Markham, Ont: LexisNexis Canada, 2005); Sopinka, Lederman and Bryant, The Law of Evidence in Canada, 2nd ed (LexisNexis Canada Inc., 1999); Canada Post Corporation v Canadian Union of Postal Workers, 2016 CanLII 79621 (CA LA); Canada Post Corporation v. Canadian Union of Postal Workers, 2019 ONCA 476 (CanLII); Craftwood Construction Co. Ltd., 1980 CanLII 940; The Citizen, 1985 CanLII 990; Atway Transport Inc., 1989 CanLII 3245; Ajax/Pickering News Advertiser, 1993 CanLII 7838; Thurston v Ontario (Children's Lawyer), 2019 ONCA 640 (CanLII).

76.              As I observed at the outset of this decision, the origin of “dependent contractor” in Canadian legal scholarship can be traced to a scholarly article written by Professor Harry Arthurs (see footnote 1) where he proposed the category of “dependent contractor” as an intermediary between, at one end of the spectrum, an employee, and at the other end, the independent contractor.  Professor Arthurs argued that dependent contractors ought to have access to the collective bargaining regime.

77.              Ultimately, the concept of dependent contractor was adopted and defined by the Ontario Legislature in 1975, in what is now section 1(1) of the Act.  That section reads as follows:

78.              It is clear from this definition that individuals may be entitled to collective bargaining even if they are not employed under a contract of employment – that is, even if they would not be considered employees at common law.

79.              The Act stipulates that an employee under the Act includes a dependent contractor:

80.              The Board’s historical analysis of dependent contractor status was set out in Toronto Star, supra.  It is not necessary to reproduce that jurisprudence in this decision.  Suffice it to say that the Board has examined iterations of work relationships to determine whether the individuals look more like employees or independent contractors.  While the industries have varied (e.g. trucking, couriers, taxi drivers, construction), the essential question for the Board has always been: do these individuals more closely resemble the relationship of an employee or that of an independent contractor?  This is a comparative analysis that focuses on a range of factors in the labour relations context.  It has always been, and continues to be, a factual determination.

81.              The seminal case, which was the focus of the parties’ arguments is Algonquin Tavern, supra.  In that case, the Board had to determine whether burlesque entertainers were dependent contractors.  After reviewing the various frameworks used by the courts and tribunals in Canada and the United States to make such a determination, the Board identified relevant factors that were useful to the analysis, none of which were determinative and all of which had surfaced, at some point, in the Canadian and American jurisprudence.  The Board has consistently returned to these factors when examining whether an individual or group of individuals are acting as dependent contractors or independent contractors.

82.              Over the years, the Board has observed that the factors set out in Algonquin Tavern, supra often overlap.  No single factor is determinative and not all factors are applicable in every case.  In this respect, it is very much a fact-based inquiry.

83.              As the parties focused their arguments on these factors, I find it helpful to analyze the evidence of this case along this framework.

84.              As the Board has said in previous decisions, it is inconsistent with an employment relationship if the individual is able to fulfill the work obligation with someone else’s labour and skill.

85.              Couriers do not use substitutes.  The evidence of the representative witnesses was consistent in that substitutes were not used, nor were they permitted.

86.              The evidence of Mr. Paterson was that Foodora required couriers to sign up with Foodora through the App for safety and security purposes.  He explained that the company was concerned about releasing addresses, buzz codes, and other personal information of its customers to individuals who were strangers to Foodora.  He also explained that when dealing with the safety of its couriers, Foodora needed to know who was on the road and making deliveries for the company in case there were accidents or incidents on the road.  Mr. Paterson tried to temper this evidence by suggesting there could be an occasion where he might approve the use of the substitute.  However, it was uncontested that this had never occurred, and Mr. Paterson could not be specific about circumstances when this might occur.

87.              The contract prepared by Foodora does not contemplate the use of substitutes.  Moreover, Foodora’s business is not structured in a way that allows a courier to subcontract the delivery services to substitutes.  The use of the App is personalized to the registered user.

88.              There were two emails entered as exhibits that terminated a courier’s services for using substitutes.  In an email dated May 9, 2017, Mr. Paterson informed a courier that “…It’s inappropriate to share accounts to setup a person for deliveries who has not first been screened and approved by our recruiting team.”  In an email dated May 10, 2019, a Rider Fleet Manager terminated a courier’s services for, among other things, account fraud.  When the courier asked for clarification, the manager wrote, “After completing an order, you sent an inappropriate text message to the customer.  The number used to text message the customer was connected to another account under the name of Dorian Pergjoka.”

89.              Another element to the use of substitutes is the ability to give away shifts.  Foodora does not allow direct swaps of shifts.  According to the Rider Guide, if a courier provides more than 24 hours’ notice that the individual cannot make a shift, that shift goes into a shift pool for other couriers to accept.  If the courier requests a shift swap with less than 24 hours’ notice and that shift is not picked up by another courier, it is treated by Foodora as a “no show”.  The courier also has an option of providing reasons to the manager with a request to not count the missed shift as a “no show”.  The manager has the discretion to grant the request.  I will explain the consequences of a “no show” in greater detail later.

90.              Several conclusions flow from the evidence about the prohibition against using substitutes.  First, it is apparent that the identity of the courier is important to Foodora’s business.  For reasons of safety and security of its customers and couriers, Foodora needs to know who is delivering food to its customers.  Second, Foodora has a system of controls – scheduling, restrictions on shift swaps, processes for requesting absences – to ensure that it knows which couriers are delivering to its customers, the precise time of those deliveries, and any reason for disruption.  The deliveries are not mere economic units that can be passed on to friends, subordinates or family members (as was the case in Toronto Star, supra).  Rather, the reliable and timely delivery of the food is so important to Foodora’s business that it must closely monitor the actions of the couriers and cannot risk losing such control through the use of substitutes.

91.              The restrictions on the use of substitutes strongly resembles an employment relationship rather than an independent contractor relationship.  Unlike a plumber or electrician, for example, who might use a helper or a substitute to lighten the workload or increase profitability, the courier is limited by Foodora to his own skill and labour, much like an employee is limited in a traditional employment relationship.  Thus, this factor favours a conclusion that Foodora couriers are working as dependent contractors.

92.              The statutory definition stipulates that a person can be a dependent contractor even if he owns the tools used to perform the work.  Typically, in construction cases, individuals who have been found to be dependent contractors have owned such tools as a van.  Thus, the ownership of tools is not a significant factor in the analysis, although it is a factor to be considered.

93.              Foodora couriers provide some tools without any input by Foodora: bicycles, helmets, and in some cases, cars.  The courier is responsible for the maintenance and repair of these tools.

94.              Several tools are provided by the courier, but under specific instruction by Foodora: a delivery bag that must meet specific dimensions; a smart phone that is GPS enabled with a data plan.  Foodora argued that, along with the maintenance and repair costs, these tools constituted an investment by the couriers into the business.

95.              An important tool provided by Foodora is the App (and the software and algorithms that support the App), which is regularly updated by Foodora.

96.              Foodora argued that the Board has not prioritized the value of tools in other cases.  I am not persuaded by this point.  First, there was no evidence about whether the couriers purchased the bicycle and smart phone for the sole purpose of making deliveries.  It seems highly unlikely that these items were purchased exclusively for Foodora work or courier work in general.  But, I need not draw any conclusion about those “tools” since the most significant tool is the App.  The Board cannot ignore the significant disparity in the importance of the tools to the delivery of food.  Just as the Board would not treat a shovel brought by the employee to the job site as equivalent to the backhoe provided by the contractor, the Board cannot treat the App as an equivalent to the bicycle and smart phone.

97.              The software developed and owned by Foodora in the form of an App is the lynchpin in the process to deliver food.  It is the mechanism that allows a customer to place an order, for the restaurant to receive the order, and for an algorithm to assign the delivery to the courier.  Through the App, Foodora controls payment by the customer, makes payment to the restaurant, and calculates the amount earned (including tips) by the courier.

98.              The App also allows Foodora (not the courier) to generate customer lists and information; an inventory of restaurant customers; and goodwill and brand recognition.  The App as a tool enables Foodora to continually develop and grow its business.  This is not available to the courier who performs the service as determined by Foodora through the App.

99.              While the courier must invest in some of these tools by deciding how much to spend on a bicycle or car, the investment need for the App is the single most important part of the system.  If Foodora makes a decision about the App – whether it is to make an improvement or find an efficiency – that decision can directly impact the profit/loss of the enterprise.  While the tools used to make deliveries are supplied by both the courier and Foodora, the importance of the App cannot be ignored.  It is the single most important part of the delivery process and is a tool owned and controlled by Foodora.  If the App was software that could be licensed or sold to the courier (e.g. accounting software, a website) for the courier’s entrepreneurial activity, the analysis might lead to a different conclusion.  But as the evidence made clear, the App is exclusively owned and developed by Foodora.  In this respect, the courier more closely resembles an employee who is permitted to use the company’s software than an independent contractor.

100.           In Algonquin Tavern, supra, the Board described this factor as follows:

101.           When looking at entrepreneurial activity, the Board has said this is a qualitative analysis, not a quantitative measurement (See Comfort Guard Services, supra at para 14).  That is, the focus is on the opportunity or chance of profit/loss by virtue of the individual’s entrepreneurial acumen.  It is not a measurement of how much revenue is made relative to other factors.

102.           Within the Foodora App, couriers can make more money by working harder.  That is, an efficient courier delivering food at a fast pace at key times and thus increasing the volume of deliveries will make more money than a courier who does not work as hard.  But, this is not entrepreneurial activity.  It is no different than an effective salesperson who is paid commission or a person who works more than one job or a worker who produces more on a piece-work compensation scheme.

103.           Outside of the App, there was evidence of couriers performing services for more than one App, referred to as dual apping.  This arose in two contexts.  First, while on shift for Foodora, a courier might make deliveries for UberEats.  Second, a courier might make two deliveries for different companies at the same time.  In either circumstance, by virtue of the courier’s hard work the courier might increase his earnings.

104.           This is not entrepreneurial activity.  This is not “…acumen, sensitivity to the needs of the market, astute investment, innovation, or risk taking”.  It is hard work.  And hard work must not be mistaken for entrepreneurial activity.  The courier is confined to the rules and restrictions imposed by Foodora and is only permitted to increase his earnings subject to Foodora’s rules.  For example, Mr. Ostos testified that he could deliver for UberEats while on shift with Foodora so long as it did not impede the Foodora delivery.  Mr. Sopher was taken to numerous examples where he asked Foodora to place him on break when the UberEats records indicate that he continued to make deliveries, the implication being that he was dishonest with Foodora in order to take advantage of the surge rates offered by UberEats at peak times.  Mr. Paterson acknowledged that Foodora allowed its couriers to dual app provided it did not impact Foodora’s delivery standards.

105.           Important to this analysis are the restrictions imposed by Foodora on the courier’s ability to make a profit.  A courier cannot advertise or promote his service, skill or ability.  A courier cannot develop individual relationships with customers or restaurants in a way that allows for individual selection or request of that particular courier.  A courier could not offer a coupon, discount or incentive to a customer in hopes of gaining loyalty.  Other than a tip, a courier cannot be paid any more money than Foodora allows and the entire transaction (other than when a customer gives a cash tip) occurs through the App.  The reality is that a courier cannot improve their chance to make profit through the customary entrepreneurial tools.

106.           The existence of Foodora’s team dedicated to marketing to new restaurants as well as a dedicated team focused on existing relationships with over 4000 restaurants is the type of entrepreneurial activity that couriers have no access to nor are they permitted to engage in.

107.           The risk of loss for the courier is minimal since the compensation scheme, as determined by Foodora, entitles the courier to be paid regardless of issues with the restaurant, the customer or the delivery.  Foodora bears the risk that if the restaurant makes a mistake or the customer cannot be located.  Foodora bears the ultimate risk that when it determines to make changes to the App (e.g. it often pushes out updates); changes its delivery rules (e.g. reducing the time it places couriers on a break following a declined order); or introduces new delivery options (e.g. it recently introduced new deliveries of alcohol in certain regions), that it will result in a loss or profit.  It is Foodora that decides whether to introduce such changes based on its own profit/loss strategy and assessment of the marketplace. In circumstances where there is a spill or accident, Foodora arranges for a new delivery and the courier is still paid, occasionally a prorated amount for the portion of the courier’s travel.  The courier is a mere cog in the wheel that is powered by Foodora.

108.           At most, the risk of loss arises because of the couriers’ expenses with respect to gas (if the courier operates a vehicle), maintenance, cellular data plan or smart phone.  While it is conceivable that a courier might incur more expenses than they make delivering food for Foodora, such risk is remote in the circumstances and was certainly not made out in the evidence.

109.           The limitations on entrepreneurial activity are similar to the facts in Toronto Star, supra.  In that case, the newspaper courier could deliver papers for a competitor or simultaneously deliver flyers for compensation while delivering the Toronto Star newspaper.  The Board found that the newspaper couriers had no entrepreneurial activity:

110.           The Board went on to say that working multiple jobs was not evidence of entrepreneurial activity.  To the contrary, it lends itself more favourably to a conclusion that there is economic dependence on different employers.  The Board stated as follows:

111.           Foodora couriers are able to make more money if they work harder, either through doing more Foodora deliveries or dual apping.  But, this is akin to working multiple part-time or casual jobs where the employee decides the most desirable place to work at a particular time.  If a salesperson opts to work on a Saturday because commissions will be higher, it is not considered entrepreneurial activity.  If a bartender wants to work at night because there are more tips, it would not influence the classification of the bartender as an employee.  Foodora couriers do not have the opportunity to increase their compensation through anything other than their labour and skill.  A close examination of this factor supports the conclusion that couriers are dependent contractors.

112.           When examining this factor, the Board considers whether the individual sells his services to multiple purchasers that are diverse in nature.  If the individual has a long-standing relationship with one or a limited number of purchasers, he is more likely to be considered a dependent contractor.  As explained by the Board in Algonquin Tavern, supra, the factor favours the conclusion of dependent contractor if the circumstances or contractual relationship limit the individual’s opportunity to utilize his skill with other purchasers or where his primary customer is given a priority.

113.           The evidence was clear that when a Foodora courier accepts a shift, he is expected to give priority to Foodora.  A courier cannot sit dormant for the entire shift.  A courier cannot repeatedly decline orders.  A courier cannot simply disappear for the shift or not show up.  This was the evidence of the representative witnesses but also matched by Foodora’s Logistics Guide.  Although couriers may deliver for other courier services (or have other jobs), it is Foodora’s policy that its service standards must be given priority.  A failure to do so will result in a strike against the courier or, as evident with Darren Whiteley, termination of services.

114.           It is also not accurate to say that couriers “sell” their services to other purchasers.  Couriers might perform deliveries for compensation, but there is not a packaged skill set or knowledge that is for sale.  This is not an akin to a skilled trade performing work on an ad hoc basis when the customer needs the service.  As already explained, a courier is not able to sell courier services directly to the customer or restaurant.  This factor leans more in favour of an employment relationship rather than an independent contractor relationship.

115.           The level of dependence is to be measured from what flows from the terms and conditions of the relationship between the parties (Blue Line Transportation Ltd. supra; The Citizen, supra).  In The Citizen, supra, the Board explained:

116.           Flowing from the statutory definition of dependent contractor, the Board’s analysis is not directed at the level of dependence on the industry as a whole.  Rather, the inquiry is about whether the individual is economically dependent on another person.  Chairman Donald Carter described the analysis in Adbo Contracting, supra, as follows:

117.           Moreover, as stated in IDM Refinishing, supra, at paragraph 30 the statutory definition does not require an absolute standard of economic dependence, but rather one “more closely resembling the relationship of an employee than that of an independent contractor”.

118.           Foodora spent considerable time arguing about the significance of the financial earnings of the witnesses and the fact that the representative witnesses worked significantly more than other couriers.  Mr. Paterson prepared a document showing, among other things, that the union’s witnesses had worked more frequently and earned more money than the average and median number of couriers in a 30-week period in 2019.  Foodora relied on this document to argue that couriers had greater independence.

119.           It further argued that the evidence of the witnesses’ work for other employers demonstrates that the majority of their income was not earned from Foodora.  On this basis, it argues that there is no economic dependence on Foodora.

120.           The Board accepts that the frequency of work and earnings of the union’s witnesses were on the higher end of the scale.  The evidence was that Mr. Sopher earned $4,955.34, Mr. Gonsalves earned $9,558.44, Mr. Ostos earned $7,007.59 and Mr. Tyler earned $18,409.83 in the 30-week period in 2019.  These witnesses earned a weekly average of $439.34.  During this same period, the average earnings for all couriers who worked in that period was $245.07[2].  The union did not shy away from this evidence as very little can be gleaned from these statistics.

121.           As explained in Toronto Star, supra, the Board has not confined its analysis to simply a numerical measurement of dependence.  The focus has been contextual, taking into consideration the terms and conditions of the relationship and the duration of the relationship.  The Board explained as follows:

122.           The portability of the App (couriers are not physically linked to a worksite) and the nature of the business allows couriers flexibility in the performance of the work.  The evidence showed that couriers are able to make deliveries for more than one company provided they do not compromise Foodora’s service standards.  However, much of the couriers’ work is controlled by the App using an algorithm developed, owned and controlled by Foodora for the sole purpose of advancing Foodora’s business interests.

123.           Foodora argues that couriers have little to no dependence on Foodora as couriers are free to (and often do) work for other food courier companies.  It points to its drop-in orientation session, lack of specific on-the-job training, and the numerous ways couriers can choose to work for Foodora as examples of the independence of the couriers.  However, this fails to take into account that Foodora has a network of incentives and prohibitions to steer and control the behaviour of the couriers.  The union pointed to three specific examples.

124.           First, Foodora controls the structure of shifts, when shifts are offered, how many people can work, the length of shifts, and the geographical zones.  Access to the scheduled shifts is based on the courier’s rating as determined by Foodora.  Access to ad hoc shifts (at either the request of the courier or Foodora) is closely controlled by Foodora to ensure that service delivery needs are met and also that there is sufficient work for the scheduled couriers.  There were also examples where a dispatcher contacted the courier if the courier was late signing in for his shift.

125.           Second, once a courier accepts a shift, Foodora controls how that shift can be swapped, returned back to the scheduling system or removed from the courier’s obligation without penalty.  There is no need to repeat what I have already described about Foodora’s rules in the Rider Guide.  It is also possible for a dispatcher to end a courier’s shift if the courier engages in misconduct during his shift.  Foodora must also grant permission (through a dispatcher) to a courier who wants to work an unscheduled shift or stay later.  Such permission is often granted, but is dependent on customer demand and service levels.  Couriers are not free to work whenever they want to.

126.           Foodora also “thins” its list of couriers on an annual basis by removing couriers who have not performed services for Foodora.  Mr. Paterson explained that a separate “thinning” process occurs when couriers who have not registered for a shift for an eight-week period are de-registered from the App.  Thus, a courier is required to take shifts in order to remain registered with the App.  Unlike an independent contractor who may move from client to client, the courier must maintain an ongoing relationship with Foodora.

127.           Finally, once a courier is assigned an order, he is expected to accept it and make the delivery.  Foodora tendered evidence that its previous practice of tracking the rate of declines by a courier has ended.  While that may be true, it is indisputable that couriers on shift are expected to accept deliveries on the App when they are assigned.   The business of Foodora would suffer, if not fail, if there were widespread declines of orders.  A courier could not simply decline all orders for an entire shift or transfer orders to another courier.  To the contrary, the courier is expected to make a request if he wants a break or to leave early.  The dispatch logs presented before the Board contain many such examples.  While Foodora argues that such requests are so routinely granted that it is tantamount to a notification as opposed to a request, this does not change the fact that a courier is required to make a request to dispatch for all interruptions of service.

128.           Other restrictions on mobility exist.  A courier who wishes to utilize the Guarantee Zone must meet the stipulated criteria.  The Rider Feedback Report is replete with strikes against couriers for being unresponsive (otherwise referred to as “ghosting”), refusing orders after accepting them, making a late delivery, or failing to indicate on the App that the delivery was completed.  Dispatch can also close a zone at any time, thus limiting a courier’s ability to work in that area.  Mr. Boyko testified that this occurs during parades, road closures or large events, such as a Toronto Raptors game.

129.           The fact that couriers had other sources of income is not necessarily indicative of economic independence.  It is not uncommon for individuals to have multiple part-time jobs.  Individuals may work at one part-time job more frequently than another.  This does not deprive them of their employment status, nor does it suggest that they are economically independent.  In the Board’s view, the evidence of the couriers reflects the common challenges faced by workers with multiple part-time jobs.  The only difference is that couriers are on-call for work through sophisticated technology and utilize their downtime to work a second job.  The technology allows Foodora to call upon the courier when customer demand is present in real time.

130.           There is a risk in placing too much emphasis on measuring economic dependence by way of a numerical threshold as urged by the employer (e.g. the percentage of work performed by an individual for a particular entity versus other entities).  This is especially true for employees who work multiple part-time jobs.  While some cases may lend themselves to that analysis (see for example Blue Line Transportation, supra where taxis could not be used for any other income), it will not be appropriate in sectors where services are performed on a part-time basis.  Certainly, the statutory definition that governs the Board’s analysis does not call for a numerical measurement.

131.           Foodora filed a recent decision of the Ontario Court Appeal in Thurston, supra.  Although Foodora made no submissions about this case, I will briefly address why it is distinguishable from the instant matter as Foodora made reference to the case in its opening statement.  The case arose in the context of a wrongful dismissal action filed by a lawyer who had provided legal services to the Office of the Children’s Lawyer for over 13 years.  When her retainer was not renewed, she brought a claim alleging that she was a dependent contractor and entitled to 20 months’ notice of termination.  The Court applied the traditional common law test for dependent contractor and found that the Plaintiff had only earned 39.9% of earnings from the Defendant and therefore could not be considered dependent at law.

132.           The decision in Thurston, supra does not impact the Board’s analysis for several reasons.  The primary reason is that the Board is confined to the statutory definition of dependent contractor, which does not place an emphasis on an economic measurement of dependency with a numerical threshold.  The analysis in Thurston, supra was with respect to the common law tests for determining dependent contractor status.  Additionally, for more than three decades, the Board’s jurisprudence has examined the nature and context of the relationship, whether that relationship arises in the construction sector, the trucking industry, the taxi service or with couriers.  In each of these cases, evidence was tendered about levels of control, entrepreneurial activity and independence with a focus on the fundamental question: does the individual look more like an employee or an independent contractor.  Finally, Foodora’s submissions were focused on the application of the Board’s jurisprudence with specific attention to the factors in Algonquin Tavern, supra.  Again, other than filing a copy of the case, Thurston, supra was not mentioned as a supporting authority for Foodora’s case.

133.           The Board was referred to numerous cases where dependent contractors had multiple sources of income: Toronto Star, supra; Cradleship Creche, supra; Journal Le Droit, supra; Huntsville District Memorial Hospital, supra.  This is particularly common in the construction industry where skilled trades will perform work for more than one company: Carpino Carpentry, supra; Royal Tek Stucco, supra; and Gates of Humber, supra.   In these cases, the Board found that the individuals had no real ability to generate their own customers or line of business.  The same conclusion applies to couriers working for Foodora (see also Toronto Star, supra at para 103).

134.           The duration of the relationship is part of the analysis with a longer relationship demonstrating a stronger case of economic dependence.  As a relatively new company, it is understandable that Foodora’s couriers would not have a long tenure.  Mr. Gonsalves has been a courier since November 2018; Mr. Ostos since 2016; Mr. Sopher since 2015; and Mr. Tyler since 2019.  On a continuing and weekly basis, they signed up for shifts or picked up shifts.  While I accept the employer’s point that the representative witnesses disproportionately worked more frequently than other couriers, their tenure and regular service for Foodora looks more like an employment relationship than an independent contractor.

135.           The Board was advised in closing submissions that of the 1191 couriers (including those that were challenged), 23% of the couriers had more than 19 months of service as of January, 2018.  While this certainly indicates a high degree of turnover, it also indicates a substantial number of couriers have tenure with Foodora.  On its own, it is difficult to draw a conclusion.  But, when considered with Foodora’s insistence that its couriers remain active, and Foodora’s practice of “thinning” its list, the tenure of the couriers resembles a part-time or casual workforce.

136.           The evidence must be considered in the context of the work environment.  These couriers do not report to a specific worksite.  They do not have a single supervisor.  They may never know their work colleagues.  There is no opportunity for job promotion, skill enhancement, or development.  The traditional factors that support job tenure are not present in this work environment.  This says nothing about the couriers’ commitment to Foodora when performing the work.  But, it does help explain why tenure may not be as important a factor in this type of work environment.

137.           After carefully considering the evidence with respect to mobility and independence, the Board is convinced that this factor weighs in favour of the conclusion that the couriers are dependent contractors.  There is a complex system of incentives and restrictions that limit the choices of the courier.  It has the hallmarks of the type of employment relationship the Board often sees in the form of an on-call employee or elect-to-work employee.

138.           The ability to negotiate or alter fees is indicative of independent contractor status.  The corollary is also true – the inability to alter fees or the presence of a uniform fee structure suggests employee status.  However, where the services and fees are standardized, or the market is competitive, this factor may be neutral.

139.           All Foodora couriers share the same terms and conditions as determined by Foodora subject to Foodora’s practice of allowing longer service couriers to continue work under the terms and conditions of their pre-existing contracts.  This practice has no impact on the uniform fee schedule charged by Foodora to its customers.  Although Mr. Paterson testified that it is conceivable for a courier to propose a change to the contract, it has never occurred and there does not appear to be an opportunity to negotiate with Foodora.  The couriers who testified described a similar scenario of being presented with the terms and conditions as determined by Foodora with the requirement that the courier sign the contract before being allowed to register on the App.

140.           The rates are not standardized in the market.  UberEats pays a different fee structure than Foodora that is based on supply and demand.  While Foodora’s fees have some variance, its fee structure for delays and corporate orders is prescribed by Foodora.  Mr. Boyko testified that dispatch may manually adjust compensation for travel if there are road closures.  But a courier is unable to unilaterally adjust those rates.

141.           A Foodora courier has no independent opportunity to vary his rate.  A courier cannot, for example, charge more because of poor weather or less because of good road conditions.  A courier cannot vary the fee depending on demand from customers or supply of couriers at a particular time.  Moreover, there is no ability to negotiate with the customer or the restaurant to vary the rate.  Even when a courier is delivering a corporate order, the gratuity is prescribed by Foodora.

142.           The inability to vary the fee charged by the courier makes the courier more like an employee who receives a standard wage rate (or piece rate) rather than an independent contractor who has the ability to vary his fees to suit his needs or the environment.

143.           It bears repeating that the Board’s experience in applying these factors is that there is considerable overlap and to some extent interdependence among the factors.  With respect to integration, the Board, in Algonquin Tavern, supra, explained the following at paragraph 64:

144.           Couriers are heavily, if not entirely, integrated, into Foodora’s business.  Foodora’s revenue depends entirely on the reliable and timely delivery service of the couriers.  In turn, the couriers rely solely on Foodora’s App – an instrument that facilitates relationships, as well as payment, with customers and restaurants.  With the exception of a layer of supervisors and the dispatchers (who might be supervisors, but that is not a conclusion I need to make), couriers are the service side of the business.

145.           The level of integration is even greater than the Board found in Toronto Star, supra, which was described as follows:

146.           In Toronto Star, supra, couriers could identify themselves to the customer and give a Christmas Calendar, provided by the Toronto Star, with their name and phone number.  This does not exist with Foodora couriers who have little or no interaction with the customer other than to deliver the food.  It would only be by coincidence that a courier might encounter the customer on more than one occasion.

147.           There is no opportunity, nor reason, for the Foodora courier to develop any type of relationship with the customer or restaurant.  In every practical sense, Foodora ensures the relationship is between itself, the customer and the restaurant.  The courier is a cog in the economic wheel – an integrated component to the financial transaction.  This is a relationship that is more often seen with employees rather than independent contractors.

148.           Foodora argued that the Board should be concerned with how the couriers identify themselves.  It points to an exchange between Mr. Tyler and dispatch when Mr. Tyler identified himself as an “independent contractor” in an attempt to enforce his rights to decline an order.  Foodora relies on Journal LeDroit, supra, at para 21 where the Board said that the parties’ perspectives are useful.  It also referred to Ajax/Pickering News Advertiser, supra, where the Board noted that the couriers had treated themselves for tax purposes as independent contractors.

149.           A single incident by a courier sending an SMS message to a dispatcher in the heat of a moment to assert independence is not persuasive evidence of the couriers’ perspective about their status.  It sheds no light on how couriers perceive their relationship with Foodora or how they declare their relationship with Foodora for tax purposes or other purposes.  It could just as easily be interpreted as nothing more than Mr. Tyler reminding dispatch what he has been told by Foodora, that he is an independent contractor.  It is not persuasive evidence of his status nor his perception of status.

150.           The Board concludes that when examining the level of integration, the couriers more closely resemble employees rather than independent contractors.

151.           The parties agreed that this factor was non-existent in this matter as there was no specific degree of specialization.  The factor is neutral.

152.           As I have explained throughout this decision, Foodora has implemented numerous controls on the generation and flow of work, whether it be from developing relationships with restaurants, to the exclusive utilization of the App, to the scheduling and control of the couriers’ work.

153.           In addition to tracking and reporting issues with couriers, and investigating issues, the use of Global Positioning System (“GPS”) technology is an additional layer of control.  The dispatch logs reveal questions from dispatchers to couriers about their location.  It would be unwieldly to go through all of these examples.  A few will suffice.  On January 6, 2019 Mr. Gonsalves was told by the dispatcher that he was going the wrong way and “Return to St. Clair to complete your order!”.  On January 8, 2019 he was told to “…go back to the zone please”.  On January 26, 2019, Mr. Gonsalves received the following message: “your GPS hasn’t moved for a while, where are you right now?”

154.           The GPS tracking was not restricted to Mr. Gonsalves.  The following is an exchange with Mr. Sopher and a dispatcher on July 7, 2018:

155.           I accept the evidence of Mr. Paterson that interactions between dispatchers and couriers are minimal, perhaps as low as 5% or even lower.  I also accept the evidence that dispatchers are not actively monitoring the GPS coordinates of couriers.  But, the focus is not on the frequency of exercising control.  Rather, it is about the right and ability of the company to control how the work is performed that lends more favourably to a conclusion that the individuals are dependent contractors.  The evidence, as described under the various factors, shows that Foodora couriers might work independently, but always within the parameters unilaterally established by Foodora and under the watchful eye of dispatch.  Mr. Boyko testified that dispatchers can monitor the location of the courier and send a message to the courier if there is an issue with their location.  Mr. Boyko said that his personal rule as a dispatcher was to only reach out to a courier if he saw no activity for 10 minutes.

156.           Moreover, the advancement of technology – algorithms, GPS, automated alerts, SMS communications – allows Foodora to control the operation with minimal human interaction.  This does not mean Foodora does not closely supervise the couriers.  It is not as though Foodora sends the courier out to make deliveries and hopes the courier reaches the customer.  To the contrary, the sophisticated technological advancements permit Foodora to closely monitor every move of the courier to ensure its service standards are met.

157.           Such level of control is apparent from reviewing the Rider Feedback Log where dispatchers reported to Rider Management about issues with couriers.  These reports were often described as strikes (as described earlier, strikes are issued for undesirable behaviour that is detailed in the Logistics Guide) and recorded when a courier inappropriately declined an order, was unresponsive, was late without explanation, or could not be tracked by dispatch (described by dispatch as MIA). 

158.           A review of the strike list in the Logistics Guide illustrates the extent of control Foodora has over the couriers.  Dispatchers are told to issue a low strike if a courier “mis-clicks” on the App and does not immediately report it to Dispatch.  But, Dispatch is also told not to issue a strike if the courier reports the “mis-click immediately” as “…all damage can be mitigated, save for a quick text or call to the customer affected”.  Another low strike for “Wrong order delivered/picked up” illustrates the same level of control, where the instructions are as follows:

159.           At its simplest, the dispatcher has the discretion to issue low level strikes against a courier if the courier engages in undesirable behaviour.  Whether the objective is to curb the behaviour through punishment or communication is irrelevant since it clearly illustrates that Foodora is monitoring the conduct of couriers and following up with the courier when it feels necessary.

160.           The evidence about what constitutes a medium strike further illustrates the persistent monitoring of courier behaviour and enforcement of Foodora expectations.  For example, despite Foodora’s argument that couriers are free to not work, a medium strike is issued for disappearing in understaffed zones or when a courier’s negligence causes a spill.  It is apparent that GPS monitoring is involved in this supervision based on the following description [reproduced verbatim]:

161.           This level of monitoring and supervision is what is commonly seen in an employment relationship whereby supervisors are told about the types of behaviour that warrant employee discipline.

162.           Much like progressive discipline in an employment relationship, the Logistics Guide calls for escalation of strikes for more serious behaviour.  The type of listed behaviours supports the union’s contention that couriers have very little choice once on the shift.  While the courier might be able to decline an order, the courier must be responsive in a way that conforms with Foodora’s expectations.  Otherwise the courier will be issued a high strike for repeatedly refusing an order after pick up, disappearing during a high-leverage shift, or GPS spoofing to avoid receiving orders.  The point being that once on a shift, a courier is expected to work.  This is consistent with the evidence tendered by the representative witnesses.

163.           While Mr. Paterson testified that there was minimal follow up with couriers, and often times no follow up, the evidence presented by the union showed examples of severe consequences.  In an email dated October 23, 2017, a courier’s services were terminated by an email from Oliver Wheller, Courier Growth Manager for “purposeful order delays as a result of you working with another delivery company”.  Similarly, in an email dated May 28, 2019, William DeSouza, Rider Fleet Manager, terminated a courier’s services for “excessive delays” and “consistent

164.           The frequency of interactions with dispatchers is undoubtedly limited because the software (e.g. the App) is so effective at monitoring the delivery process.  In as much as Mr. Paterson testified to the minimal interactions with dispatchers, it also demonstrated that Foodora controls the delivery process through the App with minimal human interaction.  It is an automated system guided by an algorithm. 

165.           These two documents – the Logistics Guide and the Rider Guide - set out how the services are to be performed, Foodora’s expectations of the couriers, and the consequences of non-compliance with those expectations.  When read in the context of the dispatch communications, the Rider Feedback Report, and the evidence of the couriers, these two documents are the type of policy manuals that apply in the traditional employment context.  Put simply, the two documents explain how Foodora expects to operate its business.

166.           When examining the evidence under this factor, the relationship between couriers and Foodora more closely resembles an employment relationship.

167.           The Board has no comparative evidence to measure the magnitude of the contract amount.  The sums paid to the couriers are obviously significant enough to motivate them to perform services.  At times, couriers are motivated to work for other delivery services or not work at all.  It really cannot be said that the magnitude of the contract amount is a relevant factor in the Board’s analysis.  The parties said as much in their submissions.

168.           The terms and manner of payment are more closely aligned with what one might see in an employment relationship.  The couriers are paid on a weekly basis by way of direct deposit to their bank account based on the previous weeks’ earnings.  Through an online system, couriers can access the details of their payment.

169.         However, it is conceivable that such an arrangement might also exist with an independent contractor.  As such, the Board finds this factor to be neutral to the overall analysis.

170.           The parties did not make submissions with respect to this factor.  That is likely because their submissions substantially covered this factor under other headings.

171.           The Board has carefully reviewed the evidence called by the parties using the factors historically considered by the Board from Algonquin Tavern, supra in the interpretation of the statutory definition of dependent contractor.  The couriers are selected by Foodora and required to deliver food on the terms and conditions determined by Foodora in accordance with Foodora’s standards.  In a very real sense, the couriers work for Foodora, and not themselves.

172.           This is the Board’s first decision with respect to workers in what has been described by the parties and the media as “the gig economy”.  However, the services performed by Foodora couriers are nothing new to the Board and in many ways are similar to the circumstances of the Board’s older cases.  This is not the Board’s first case examining the relationship of couriers.  The Board has been tasked with the same questions about dependent contractors in various sectors including transportation and construction.  Such cases have always been fact-based inquiries that require a balancing of factors.  This case is no different in many respects.

173.           For the foregoing reasons, the Board finds that Foodora couriers are dependent contractors and must be treated as such under the Act.  As the evidence bears out, couriers more closely resemble employees than independent contractors.

174.           There remain outstanding issues between these parties about the eligible voters on the list.  The matter is referred to the Manager, Field Services.

175.           I remain seized.