Content uploaded by Paul Belleflamme
Author content
All content in this area was uploaded by Paul Belleflamme on Dec 05, 2017
Content may be subject to copyright.
No. 108, 4th quarter 2017
No. 108, 4th quarter 2017
Dossier
The Sharing Economy:
Myths and Realities
Edited by
Anders HENTEN, Denis LESCOP,
Jean Paul SIMON & Bruno SORIA
Foreword
by Jacques MOULIN, Executive Director of Publication ............................... 5
Introduction
by the Editors ................................................................................................ 9
Papers
Lobbying as Rhetorical Framing in the "Sharing Economy": a case
study on the limits and crisis of the Evidence Based Policy Paradigm
Cristiano CODAGNONE .............................................................................. 15
Platforms at the Heart of New Form of Labour
Patrice FLICHY ............................................................................................ 45
The Rise and Fall of Take Eat Easy, or Why Markets are not Easy to
Take in the Sharing Economy
Paul BELLEFLAMME & Nicolas NEYSEN .................................................. 59
More generous for small favour? Exploring the Role of Monetary and
Pro-Social Incentives of Daily Ride Sharing Using a Field Experiment
in Rural Île-de-France
Dianzhuo ZHU.............................................................................................. 77
Interviews
Frédéric MAZZELLA, BlaBlaCar
Conducted by Yves GASSOT ...................................................................... 99
Guy STANDING, SOAS University of London
Conducted by Anders HENTEN & Iwona WINDEKILDE ........................... 105
Extra paper
Network Sharing and 5G in Europe: The Potential Benefits of Using
SDN or NFV
J. Scott MARCUS & Gabor MOLNAR ....................................................... 113
Features
Regulation and Competition
Universal Basic Internet as a Freemium Business Model to Connect
the Next Billion
Steve ESSELAAR, Steve SONG & Christoph STORK ............................. 139
Strategic Implications of Embedded SIM for the Competitive
Landscape in the Mobile Telecommunications Consumer Segment
Dr. Christian WERNICK & Dr. Christin Isabel GRIES ................................ 157
Author biographies ............................................................................. 171
DigiWorld Economic Journal, No. 108, 4th Q. 2017, p. 59. www.idate.org
The Rise and Fall of Take Eat Easy,
or Why Markets are not Easy to Take
in the Sharing Economy
Paul BELLEFLAMME
Aix-Marseille Univ., CNRS, EHESS, Centrale Marseille, AMSE;
KEDGE Business School and CESIfo, France
Nicolas NEYSEN
HEC Management School, Liège University; Jack Welch College of
Business, Sacred Heart University, Belgium
Abstract: We examine the reasons behind the rise and fall of Take Eat Easy, a Belgian
home food delivery platform, in order to understand better the competitive dynamics that
are at play in the sharing economy. We start by describing the home food delivery market
in Europe and by summarizing the key milestones in the short existence of Take Eat Easy.
We propose next a brief review of what the economic literature on digital platforms
teaches us about what they do and how they compete. On that basis, we analyse more
deeply the strategic choices taken by the leadership of Take Eat Easy, and their
implications. Finally, we combine the two previous analyses to question the sustainability
of the current business model of platforms in the sharing economy.
Key words: Digital, platform, start-up, sharing economy
he rapid development of the mobile Web over the last decade has
led policy makers and scholars to investigate the impact of this
dimension of ICTs on the economy. More specifically, a great deal
of attention has been paid to the way digital-enabled innovations
affect existing market configurations and competition rules. While in some
cases the digital revolution has completely disrupted entire industries – think
about the retail industry with Amazon, the telecommunication industry with
Skype, the electronic device industry with Apple, etc. – in other situations, it
has triggered new market opportunities in more restricted areas of existing
industries. This is for instance what happened in the context of what is today
T
60 No. 108, 4th Q. 2017
called the "sharing economy", i.e., a structured system organising peer-to-
peer transactions, which were previously happening in a decentralized and
disorganised way.
The following analysis focuses on assessing the success factors of
players competing against one another in the sharing economy. Most of the
descriptions of the sharing economy are centred on a handful of success
stories (or supposed so), such as those of Uber or Airbnb for instance. Yet, it
is also worth sharing stories of half-successes or of failures, as there is at
least as much to be learned from them. This is precisely our objective in this
article. By examining the reasons behind the "rise and fall" of Take Eat Easy,
a Belgian home food delivery platform, we aim at understanding better the
competitive dynamics that are at play in the sharing economy.
The home food delivery sector remains a young and volatile market in
Belgium, which is a very small playground compared to other European
countries like the UK or Germany. As we will show in the following lines, the
geographical criterion plays a central role in the end result, even if most
people tend to think that with the Internet, a strategy can get rid of any
physical location consideration.
This article begins by offering a view of the home food delivery market. It
also includes a summary of the key milestones in the short existence of
Take Eat Easy. In the second section, we propose a brief review of what the
economic literature on digital platforms teaches us about their roles, their
strategies, and the competition among them and with other forms of
businesses. This review provides us with the necessary tools for a deeper
analysis of the strategic choices taken by the leadership of the start-up, and
their implications. We also base our analysis on numbers and figures
provided by Take Eat Easy, as well as views directly expressed by one of its
cofounders who had been interviewed twice by one of the authors. Finally, in
Section 4, we combine the two previous analyses to question the
sustainability of the current business model of platforms in the sharing
economy.
Background
In this section, we start by describing the broad context of the European
home food delivery market, before taking a closer look at the Belgian scene,
with a particular focus on the first-mover, namely Take Eat Easy.
Paul BELLEFLAMME & Nicolas NEYSEN 61
The home food delivery market
The home food delivery market is a typical urban phenomenon. In 2015,
this market was worth €83 billion worldwide and represented four per cent of
the food sold through restaurants and fast-food chains (HIRSCHBERG et al.,
2016). For decades, pizzerias have been almost the only restaurants to
deliver food to hungry customers living in city centres. And most of them had
their own pool of mopeds to ensure fast delivery. In recent years however,
the home food delivery market has expanded to other types of restaurants
and segments.
Digital technologies have actually largely contributed to the
transformation of this particular niche in the wider food industry. Here,
platforms usually organise the delivery of meals by acting as intermediaries
between restaurants, customers, and couriers. Platforms expand the market
by providing delivery solutions to restaurants that do not have their own
logistics. A recent study conducted by McKinsey shows that the above-
mentioned traditional form of delivery (i.e., restaurants bringing the food to
the door) has still a 90 per cent market share, with almost three-quarters of
orders placed by phone (HIRSCHBERG et al., 2016). The remaining 10%
represent orders facilitated via community-based online services like
Delivery Hero, Deliveroo, UberEATS or Amazon Restaurants.
These home food delivery platforms are thus new players in this local
ecosystem. Since they operate as intermediaries, their value proposition is
threefold: for restaurants, a costless and riskless extension of their customer
base; for couriers, a source of complementary revenues; for customers, the
convenience of having great food delivered to their doorstep. Hence,
platforms create enough value for each of these groups so as to capture a
slice of it.
Most of these platforms share the same business model. To optimize the
interaction, i.e., maximizing the probability of having lots of individual
successful matches between restaurants and customers, platform operators
rely on algorithms and delivery planning automation mechanisms, which are
made possible by a clever use of digital technologies, such as smartphones
and geolocation. To monetize their services, platforms generally charge
restaurants a share of the overall bill. For instance, if someone orders pizzas
via the platform for a total cost of 55€, the restaurant agrees to leave a
commission of 20% to the intermediary who earns 11€ on this side. On the
other side, the platform can also ask customers a fixed amount per delivery.
62 No. 108, 4th Q. 2017
It might also be a variable amount based on the distance covered by the
courier, just like the way taxis set fees.
The rise and fall of Take Eat Easy
Launched in Brussels in the summer 2013, Take Eat Easy was among
the very first to enter the home food delivery market in Belgium by focusing
on the premium segment. Indeed, their mission was to enable quality
restaurants to provide a reliable delivery service to their clients. Very quickly,
the start-up expanded to other cities. The entry on the international market
started with Paris in October 2014, followed one year later by Berlin, London
and Madrid. At that time, the goal was clearly to become the leading platform
in the European home food delivery market.
To fuel its rapid growth, Take Eat Easy raised €6 million in a first venture
capital funding round in April 2015, and a second one in August 2015 which
brought an additional €10 million investment to sustain growth. Between
August 2015 and July 2016, the young company increased the size of its
team from 10 to 160 employees; it also entered new geographies to end up
with 20 European cities. Moreover, they grew their community from 30,000
to 350,000 customers and concluded partnerships with 3,200 restaurants.
Active customers ordered once every three weeks on average and
represented one third of the total customer base.
Just after having reached a landmark in its history by achieving 1 million
deliveries, Take Eat Easy in July 2016 abruptly announced its intention to
apply for judicial restructuring and to cease trading. The reason for this is
that despite rapid growth, the start-up was making no profits, as announced
on social media in 2016: "Contribution margin has been steadily improving
and is now positive across the group, though not yet high enough to cover
our fixed costs." (see ROOSE, 2016). To understand why revenues did not
cover the costs, it is necessary to dive into the business model.
On each order, Take Eat Easy charged a commission ranging between
25 and 30 %. A fixed delivery fee, €2.5, was also charged to the customer.
The start-up paid the courier with a fixed amount per hour. The number of
deliveries per courier and per hour was the most important metric to
optimize. Aligning demand and supply in time and space is not so easy to
achieve. There are multiple parameters to take into account like last position
of the courier, preparation time (and unexpected delays in preparation),
potential order cancellation, weather conditions, distance to be covered, etc.
Paul BELLEFLAMME & Nicolas NEYSEN 63
As noted by McKinsey experts: "The biggest disadvantage [of home delivery
platforms] is the need to invest in a delivery fleet and drivers, which can
quickly turn into a cash drain if the operators cannot achieve a high rate of
utilization" (HIRSCHBERG et al., 2016). As a consequence, a low courier
utilization could rapidly imply a negative contribution margin.
Finally, despite multiple attempts to convince investors, the start-up did
not succeed in the completion of a third round of funding. Moreover, one of
their historical funding partners decided to aggressively invest in Foodora, a
direct competitor of Take Eat Easy. On the other hand, Deliveroo, another
direct competitor, succeeded in closing a round of funding. In this context,
the management of the start-up considered that it was impossible to
continue operating the business as usual (ROOSE, 2016).
A microeconomic analysis of peer-to-peer platforms
In this section, we use recent advances in the theory of industrial
organisation to understand which type of companies Take Eat Easy and
other similar platforms are
1
. We start by defining peer-to-peer platforms as
multisided platforms; we then detail the external effects that these platforms
seek to internalize; finally, we examine why the emergence of these
platforms is disruptive for incumbent firms.
Peer-to-peer platforms as multisided platforms
Take Eat Easy belongs to the category of "sharing economy platforms" or
"peer-to-peer platforms". In this category, we find companies (mainly for-
profit) that operate a specific kind of marketplaces, which are called "peer-to-
peer" (or P2P for short); in other words, these companies facilitate
exchanges among "peers". The term "peer" is used to highlight the fact that,
in many cases, any user of these platforms can choose to operate either on
the demand or on the supply side of the platform. For instance, on Airbnb,
you can be a host one day or a guest the next day; similarly, on Take Eat
easy, couriers can order a meal through the platform once their shift is over.
Thus, there is no a priori definition of the respective role of any participant.
1
For a textbook coverage of the industrial organisation literature on multisided platforms, see
BELLEFLAMME & PEITZ (2015, Part IX).
64 No. 108, 4th Q. 2017
P2P marketplaces are organised through digital platforms that can be
seen as "multisided platforms". EVANS (2011) notes that a business
opportunity emerges for a multisided platform when distinct groups of
economic agents (the "sides" of the platform) wish to interact but fail to do so
because of too high transaction costs; the platform can then create value by
intermediating between the groups so as to internalize part or all of the
external effects resulting from the interaction among the groups.
P2P marketplaces are just one particular category of multisided
platforms. Other well-documented categories are the following: exchanges
help "buyers" and "sellers" search for feasible contracts and for the best
prices (e.g., eBay, Booking.com, edX...); hardware & software systems allow
applications developers and end users to interact (e.g., Mac OS, Android,
PlayStation...); matchmakers help members of one group to find the right
"match" within another group (e.g., Alibaba, Monster, Tinder...);
crowdfunding platforms allow entrepreneurs to raise funds from a "crowd" of
investors (e.g., Kickstarter, Indiegogo, LendingClub...), transaction systems
provide a method for payment to buyers and sellers that are willing to use it
(Visa, Bitcoin, PayPal...).
As already mentioned, the main function of multisided platforms is to
internalize the various external effects generated by the interaction between
the groups that they serve. They do so by making appropriate decisions
about prices, design and governance rules. A prerequisite for making these
decisions is to have a clear view of the various external effects at play. We
now perform this exercise for the case of P2P platforms.
External effects on peer-to-peer platforms
What are the external effects that are present on P2P platforms? Most
platforms facilitate the interaction between two main groups, which can be
generically referred to as "producers" and "consumers". Let us first identify
the external effects that exist across groups. The goal is to estimate how the
members of one group value an increased participation by the members of
the other group. Clearly, the effects are positive in both directions: if more
producers register with the platform, consumers will be more likely to find the
product or service that matches their needs or tastes; similarly, a larger
Paul BELLEFLAMME & Nicolas NEYSEN 65
number of consumers visiting the platform expands the potential demand for
each producer. Such effects are also called (indirect) network effects
2
.
A larger number of participants, and thus of transactions, may also allow
the platform to enhance the quality of its intermediation services.
Participants and transactions do indeed generate massive flows of data,
which platforms can use to improve the relevance of the algorithms on which
are based their reputation or recommendation systems, or to optimize the
logistics of their operations as in the case of Take Eat Easy. As the co-
founder said (see ROOSE, 2016), "to optimize courier utilization, our
Technology & Operations team focused their efforts on automatic
optimization and dispatching of orders, analytics-based capacity planning,
and deep integration with the restaurants' operations"
3
.
External effects may also arise within groups. In the group of consumers,
word-of-mouth generates positive effects: the more consumers are satisfied
with the services of a platform, the more they talk about it, which contributes
to attract new customers. Naturally, the reverse effect may happen if some
form of congestion is reached (for instance because the infrastructure of the
platform is insufficient to accommodate an increasing mass of transactions).
Another form of external effects may result from the way the platform
monetizes its services. As an illustration, take the surge pricing mechanism
that Uber applies. With this mechanism, the price of a ride depends on the
state of supply and demand at any point in time. As a result, any customer
will be negatively affected if many other customers want to use Uber
services at the exact same time (and if supply takes some time to adjust).
Within the group of producers, external effects may also go in both
directions. The competition among producers generates a negative effect:
for a given number of consumers, the more producers there are on the
platform, the lower the expected profit that each of them can make.
Conversely, a larger number of producers may contribute to sustain the
development of useful third-party services, to the benefit of each producer
(e.g., companies that propose Airbnb hosts to handle a series of services on
their behalf, such as listings, guest arrivals and departures, cleaning
services, etc.). Figure 1 summarizes the various external effects.
2
For a definition of the various forms of network effects, as well as an analysis of their
consequences, see BELLEFLAMME & PEITZ (2017).
3
Similarly, SHEAD (2017) explains that Deliveroo managed to reduce its delivery times by 20%
with a new algorithm using machine learning.
66 No. 108, 4th Q. 2017
Figure 1 - Sources and directions of external effects on a P2P platform
"Uberization" or the disruptive entry of P2P platforms
The emergence of P2P platforms disrupts the working of numerous
industries. Industries such as ride-hailing (with Uber, Lyft or Blablacar) or
short-term accommodation (with Airbnb) come immediately to mind. But
other sectors are also affected, e.g., delivery (with Instacart, Postmates),
food (with MenuNextDoor), on-demand job (with TaskRabbit, Handy) and
even the banking sector (with Lending Club, Funding Circle).
The term "uberization" has been coined to describe the disruption caused
by the entry of these startups. Behind this term, one finds a new form of
competition that is first exerted through the type of organisation (platform vs.
integrated firm), and then through the prices and quality of the product and
services (as is usual in any market).
The choice of organising the company as a platform instead of a
vertically integrated structure entails a number of important effects in terms
of costs, quality and prices; entrant platforms may turn these effects into
competitive advantages with respect to conventional incumbents. As far as
costs are concerned, the main difference comes from the fact that these
startups do not produce anything by themselves: their activity consists in
matching independent producers with customers. As a result, their cost
structure is radically different from that of the conventional firms they
compete with.
Paul BELLEFLAMME & Nicolas NEYSEN 67
Conventional firms usually have large production costs, which they have
endeavoured to reduce by developing their activities. That is, they have
benefited from economies of scale and/or scope and thereby, they have
erected barriers to entry. Yet, as TANEJA (2015) explains, these entry
barriers are largely inoperative against platform businesses that face
intermediation costs instead of production costs. For instance, an integrated
hotel group like AccorHotels managed, through its sheer size, to keep
entrants at bay as long as these entrants shared its business model.
However, AccorHotels was powerless against the entry of Airbnb, which
does not own a single hotel room. Moreover, those costs differences are
sometimes amplified by the fact that entrant platforms elude (at least for a
while) the regulations that constrain the operations of incumbent firms. This
explains why, in the ears of incumbent firms, uberization often rhymes with
unfair competition.
The platform mode of organisation also impacts on the quality of the
products and services offered to the consumers. Because they do not
produce anything by themselves, platforms are much more flexible than
conventional firms (which have their hands tied by their previous production
choices and past investments). Platforms can thus focus on products and
services that match the consumers' tastes, and adapt them fast whenever
necessary. Moreover, through a clever use of digital technologies, platforms
make the products and services they offer easier and more convenient to
use.
Finally, in terms of prices, platforms usually set asymmetric price
structures because they need to incentivize the participation of one group to
make sure that the other group will also participate (the so-called "chicken-
and-egg problem"). This may induce platforms to reduce the price they
charge consumers. The reason is that the additional revenue gained from
one consumer does not stem only from the price that this consumer pays; it
also includes the positive external effect that this consumer exerts on the
producer side. In other words, the platform may find it optimal to reduce
(even to zero) the margin it achieves on the consumer side because it can
more than recoup this loss by increasing its margin on the producer side.
Such a policy is perfectly rational for a platform that internalizes the positive
external effects that exist across the groups that it serves. Yet, from the point
of view of integrated incumbent firms, there is there another reason to cry
foul.
68 No. 108, 4th Q. 2017
Is the platform business model sustainable?
It cannot be denied that by adopting a platform model, a number of
startups managed to gain significant market shares in a variety of sectors, as
attested by the current success of Uber, Airbnb and others. Yet, it is
legitimate to wonder to which extent these successful entries can be
sustained in the long run. P2P platforms do indeed face three major threats:
(i) the reaction of incumbent firms; (ii) the competition coming from other
P2P platforms; and (iii) the "growth for growth's sake" fallacy. We detail
these three threats in turn, before hypothesizing what is in store for P2P
platforms in the near future.
Incumbent firms' reaction
As their market shares were shrinking, incumbent firms deployed new
strategies to respond to the entry of P2P platforms into their market.
MATZLER et al. (2015) describe the main strategies that have been
observed so far. First, incumbent firms can sell the use of their product
instead of selling the product itself (like Daimler AG did with its car sharing
service car2go). Second, they can support their customers in their attempts
to resell (as Ikea Group did by launching a platform where customers can
resell used Ikea goods). A third option is to take advantage of unused
resources and capacities (e.g., by sharing unused office space with the
collaboration of a platform such as LiquidSpace). Fourth, incumbent firms
can strengthen their offering by providing repair and maintenance services
(as Best Buy did by acquiring Geek Squad, specialised in the repair of
computers). Fifth, they can align with collaborative consumption to target
new customers (as Pepsi did when it partnered with Task Rabbit to launch a
new soft drink). Finally, it is possible to embrace the sharing economy to
define new business models (at the image of the platform Kuhleasing.ch
created by Swiss farmers for leasing cows).
Incumbent firms have also lobbied regulators to make them set a level
playing field in their industry, either by extending existing rules to entrant
platforms, or by applying new adjusted rules to old and new players alike.
Paul BELLEFLAMME & Nicolas NEYSEN 69
Competition among P2P platforms
More often than not, it is not one but several P2P platforms that try to
enter a particular market more or less at the same time. For instance, Take
Eat Easy and Deliveroo were launched almost simultaneously.
Competition among P2P platforms is peculiar in that it concerns the
various sides of the platform at the same time: platforms compete indeed to
attract both producers and consumers, knowing that the former group will
not join if the latter group does not, and vice versa. Competition is thus
exacerbated by the presence of indirect network effects: by attracting, say,
one additional producer, a platform not only attracts additional consumers
but, often, it also reduces the rival platforms' ability to do the same (insofar
as the producer that it has attracted is not be able to join any other platform).
As highlighted by the co-founder of Take Eat Easy
4
: "Even if we launched
our activities at the same time, Deliveroo rapidly got ahead of us because
their start-up capital was much bigger than ours and they also had this
advantage of having selected a much more profitable city where to start,
London."
Note that this effect is somehow attenuated when platforms are not
exclusive, i.e., when producers and/or consumers have the possibility to use
several platforms at the same time, a situation known as "multihoming" in
the economics literature
5
. In that case, having one producer more on one
platform does not necessarily mean that there is one producer less on
another platform. VERMEULEN (2017) gives the example of taxi drivers in
Singapore who have at least two smartphones in their car: if a ride comes on
one phone, they accept it and turn the other phone off; in a similar way,
riders have several apps on their phone, which allow them to compare
competing offers in terms of price and quality. As this example shows, even
if the reinforcing nature of network effects is somehow reduced in the case
of multihoming, competition remains extremely fierce on both sides of the
platforms (partly because digital technologies, on which these platforms rely,
make it easier for consumers and producers to compare competing
offerings). In the specific case of the home food delivery market, a priori
open for multihoming, studies have shown that the reality is much more
contrasted: "Once customers sign up [on online food-delivery platforms],
4
We translate here a number of quotes taken from Adrien ROOSE'S interview in
BELLEFLAMME (2017).
5
In contrast, one talks of "singlehoming" when users are restricted to use a single platform.
70 No. 108, 4th Q. 2017
80 per cent never or rarely leave for another platform, creating a strong
winner-take-all dynamic, in which the reward goes to the player who can
sign up the most customers in the shortest amount of time" (HIRSCHBERG
et al., 2016).
As any participant is very valuable for each competing platform, one
understands that in such a context, competition may quickly lead to a
dominant position: as soon as one platform takes a significant lead with
respect to its rivals, its advantage is likely to feed itself because of network
effects (a platform that registers more consumers, attracts more producers,
which in turn attract more consumers, and so on and so forth). As noted
above, the likely outcome of competition among P2P platforms is thus a
"winner-takes-most" situation, with one winner and a couple of "losers", who
take refuge in niche segments or in geographically limited markets.
This scenario is clearly the one that can be observed in the home food
delivery market. "A large platform is more effective than a small one, thanks,
among others, to economies of scale and reputation effects. Logically, the
big platform becomes always bigger and the small one always smaller. Let's
be honest, in our industry, in the end, there is no room for more than one
intermediary", the co-founder of Take Eat Easy said.
As SHAPIRO & VARIAN (1999) explained about twenty years ago, this
competition "for the market" (to be opposed to the usual competition "in the
market") may turn out to be very profitable for the winner, but it is also very
risky (as it is extremely hard to predict a priori who the winner will be).
The "growth for growth's sake" fallacy
Whether a P2P platform fights against a conventional incumbent or
another platform, its main short-term strategy consists in… growing. The
platform needs to reach what is commonly known as a "critical mass" of
users, i.e., a sufficient size from which growth starts feeding itself thanks to
the self-reinforcing power of network effects. Yet, as noted by HAGIU &
ROTHMAN (2016), it is dangerous to try to grow fast at any cost. When
focusing on the number of participants, the platform may neglect the quality
of the intermediation service that it provides; it may then face the risk of
being superseded by an entrant platform, which would have learned from the
mistakes of its predecessor and which would be able to propose
transactions that producers and consumers would find more valuable. As
Paul BELLEFLAMME & Nicolas NEYSEN 71
evidence, the authors refer to the case of Uber, which supplanted Lyft by
copying its P2P ride-hailing model.
Likewise, in the case of Take Eat Easy, it seems that the management
did not pay enough attention to the local market conditions to drive the
international expansion. As noted by Adrien ROOSE, "we should have
adopted a more military strategy, which means going into battle only where
we knew we could end up as market leader. From the start, it should have
been clear for us that cities like London or Paris were no more within reach.
So, one piece of advice that I would give is not to make growth an end in
itself."
The rapid growth strategy is also very costly in the short run. Actually, it is
a bet on the future: the platform accepts to make losses now to attract
participants, hoping that it will recoup this investment once it will have
reached a dominant position. To win this bet, the platform needs to convince
funders that its growth strategy leads to a self-fulfilling prophecy. The
challenge for funders is to bet on the right horse. "Three years after we
created Take Eat Easy, investors didn't believe anymore we could beat
Deliveroo and decided to largely support our nearest competitor. They
thought Deliveroo was at that time best positioned to win the bid. Today,
however, it seems that the entry of UberEATS is reshuffling the cards…", the
co-founder of Take Eat Easy said.
Investors must hope that the final victory, if it happens, will not be a
pyrrhic one. In this respect, the current state of Uber's balance sheet causes
legitimate worries to its investors
6
.
Towards an "Uberization 2.0"?
Take Eat Easy is not the only example of startups that did not manage to
achieve success in spite of a promising start. For instance, Homejoy (an on-
demand home cleaning platform) shut down in July 2015 after two years of
"torrid growth"; according to FARR (2015), the reasons for Homejoy's
collapse were the following: "mounting losses, poor customer retention, a
costly international expansion, run-of-the-mill execution problems, technical
6
Uber lost about $3 billion for all of 2016 (see SOMERVILLE, 2017). Yet, the company keeps
afloat thanks to its incredible capacity to raise funds (about $15 billion in total since it started in
2009; see SORKIN, 2016).
72 No. 108, 4th Q. 2017
glitches and the steady leak of its best workers to direct employment
arrangements with its own (now former) clients." Other examples are Spoon
Rocket (an on-demand pre-made meal delivery service) and Washio (an on-
demand laundry service), which shut down respectively in March and in
August 2016. Even though the failure rate of startups is notoriously high, it
appears that it is even higher among the sharing economy startups. As
GRISWOLD (2016) nicely summarizes it: "These companies—Uber
included—have also learned that even "asset-light" businesses can be cash-
intensive to run. At launch, many had easy access to money from venture
capitalists and relied on expensive subsidies to attract both customers and
workers. Since the fourth quarter of last year, an abrupt decline in funding
has made that model untenable for all but the richest on-demand startups."
What is even more worrying is that the few companies that manage to
survive do not seem to have won the battle yet, as exemplified by Uber's
current difficulties. Hence this nagging question: how to break even? Driving
competitors out of the market is not enough (although it obviously helps); it is
also necessary to block the entry of new competitors and/or the forceful
return of incumbents, which manage to adjust their business model.
To increase their profitability, P2P platforms only have two options:
reduce their costs and/or increase their prices. But none of these strategies
shows much promise. As far as cost reductions are concerned, possibilities
are limited: for most platforms, logistics has already been optimized and it
seems hard to extract more favourable terms from independent producers
7
.
In terms of price reductions, there seems to be even less room for action, as
consumers of P2P marketplaces turn out to be particularly price sensitive.
For instance, OWYANG & SAMUEL (2015) report that 68% of their
respondents (51,078 American and Canadian users of P2P platforms)
indicate low prices as the main reason for using sharing services
8
. It is thus
quite likely that raising prices would make a significant share of consumers
look for substitutable services
9
. And if consumers leave a platform, so will
7
Worse, costs may well increase in the near future: sharing economy producers increasingly
protest—or even file lawsuits—to be treated as employees rather than independent contractors.
For instance, in August 2017, French Deliveroo couriers protested over pay changes
(http://bit.ly/2hvM2EN).
8
"Convenience" is the only more popular reason (quoted by 78% of the respondents).
9
As KLINE (2016) explains, commenting the reasons behind the bankruptcy of Spoon Rocket:
“raising prices would have flown in the face of the company's mission of "sub-10 minute delivery
of sub-$10 meals." Without additional funding or higher prices, there was nowhere to go but
down.”
Paul BELLEFLAMME & Nicolas NEYSEN 73
producers because of the network effects that link the two groups. This
would trigger a vicious circle, which may further put the profitability of the
platform at risk.
According to Adrien ROOSE, this high price sensitivity is also a reason
why major players show a significant advantage: "Although Deliveroo has
reached a critical mass of users, Amazon Restaurants or UberEats can
easily push them out of the market because such giants have enough
reserves to fund a price war. They can reduce the delivery price, reduce
their commission, and better pay couriers… even if they aren't profitable.
Eventually, it is the company with the deepest pockets that wins the market",
he said.
The previous elements suggest that the "pure" platform business model
is hard to sustain in the long run and may thus disappear as such. This
disappearance seems already on its way, as a number of companies start
adopting new "hybrid" business models, whereby they combine vertical
integration for some of their operations and intermediation services for other
operations. Moves along these lines have been recently observed, either
from new platforms or from incumbent firms. For instance, in 2015,
AccorHotels transformed its AccorHotels.com distribution platform into a
marketplace open to a selection of independent hotels (see HotelNewsNow,
2015).
A similar move was shared by Adrien ROOSE about Deliveroo: "Being
confronted to big players, Deliveroo tries to find alternative ways to survive in
the marketplace. They are now investing in professional kitchen space,
which they put at the disposal of their best restaurants' chefs, so as to
expand their offering in areas where demand exceeded supply" (see also
O'HEAR, 2016).
Conclusion
After very a promising start and three years of regular growth, Take Eat
Easy realised that it would never be the winner that "takes" the home food
delivery market and decided that it was better to leave this market
altogether. As this story illustrates quite well the current evolution of the
(commercial) "sharing economy", we found that it was interesting to share it,
so as to help scholars and practitioners understand better the dynamics of
competition in this fast-moving arena.
74 No. 108, 4th Q. 2017
The lessons we can draw from our analysis are threefold.
First, platforms in the sharing economy operate amidst various
sources of positive feedback: the participations of producers and of
consumers reinforce one another; a higher intensity of interaction among
producers and consumers may allow platforms to enhance the quality of
their services (through big data and machine learning); by funding
preferentially the growth of those companies that they believe will prevail,
venture capitalists fuel self-fulfilling prophecies.
Second, these positive feedback effects give rise to a form of cut-
throat competition, where the big gets bigger while the small gets smaller,
leading eventually to dominant positions where a few (or unique) platforms
take most of the market; hence, for those companies that do not manage to
"take the market", the only option is to leave it, which explains why the
failure rate of start-ups is higher in the sharing economy than in other
sectors.
Finally, winning a market is by no means a guarantee for long-lasting
success in the sharing economy: efficient strategies still have to be found to
clear the big losses incurred to grow faster than the competitors and,
eventually, to make a profit; incumbent firms may react forcefully after
having been, temporarily, disrupted; there are always new entrants out
there, ready to propose improved services. The "pure" platform business
model appears thus as shaky, and it is no surprise that "hybrid" business
models emerge, which try to leverage the advantages of both vertical
integration and platform intermediation.
Even if the previous lessons sound like a word of warning, we believe
that startupers should not shy away from the sharing economy and peer-to-
peer platforms. While writing this article, we came across a number of
news
10
that should convince anyone that the sharing economy is a vibrant
sector where it is still worth investing money and energy. First, after
Transport for London decided (on September 22, 2017) to strip Uber of its
licence to operate in London, more than half a million people signed a
petition, showing strong support for Uber in particular and for ride-hailing
platforms in general. Second, Deliveroo managed (on September 24, 2017)
to raise $385 million in new funding with three objectives in mind: open more
delivery-only kitchens, further improve its logistics, and expand into new
cities and countries. Finally, to prove that there is a life beyond a failure in
10
See, respectively, http://bit.ly/2xZ1Vyu, http://tcrn.ch/2fpB8U4, and http://tcrn.ch/2wq1GIN.
Paul BELLEFLAMME & Nicolas NEYSEN 75
the sharing economy, two of the co-founders of the defunct Take Eat Easy
announced (late August, 2017) that they were on the verge of launching a
new startup, dubbed Cowboy, which aims to build and commercialize a new
electronic bicycle, which would "better previous e-bikes in terms of pricing,
product design, and technology." This may be the ground for another study
of, hopefully this time, a success story.
References
BELLEFLAMME, P. (2017): Les plateformes de l'économie collaborative :
fonctionnement en enjeux, In Decrop, A. (Ed) La consommation collaborative. Enjeux
et défis de la nouvelle société du partage. DeBoeck Supérieur, pp. 221-242.
BELLEFLAMME, P. & PEITZ, M.:
- (2015): Industrial Organization: Markets and Strategies (2nd edition), Cambridge:
Cambridge University Press.
- (2017): Platforms and network effects, Forthcoming in CORCHON, L. & MARINI, M.
(Eds), Handbook of Game Theory and Industrial Organization. Edward Elgar.
EVANS, D.S. (2011): Platform economics: Essays on multi-sided businesses,
Competition Policy International.
http://www.marketplatforms.com/wp-content/uploads/Downloads/Platform-Economics-Essays-
on-Multi-Sided-Businesses.pdf
FARR, C. (2015): Why Homejoy failed. Wired, October 26.
https://www.wired.com/2015/10/why-homejoy-failed/
GRISWOLD, A. (2016): "All Washed Up, The Ultimate Symbol of the Uber-for-X
Bubble is out of Business", Quartz, August 30.
https://qz.com/769935/the-ultimate-symbol-of-the-uber-for-x-bubble-is-out-of-business/
HAGIU, A. & ROTHMAN, S. (2016): Network Effects Aren't Enough, Harvard
Business Review 94 (4), 65–71.
HIRSCHBERG, C., RAJKO, A., SCHUMACHER, T. & WRULICH, M. (2006): "The
changing market for food delivery", McKinsey Quarterly, November.
http://www.mckinsey.com/industries/high-tech/our-insights/the-changing-market-for-food-
delivery
HotelNewsNow (2015): "Accor rebrands, opens distribution platform", June 3.
http://www.hotelnewsnow.com/Articles/26372/Accor-rebrands-opens-distribution-platform
KLINE, K. (2016): "How This Startup Blew $13.5 Million and Ended in Bankruptcy",
Inc., April 5.
https://www.inc.com/kenny-kline/how-spoonrocket-blew-135-million-and-ended-in-
bankruptcy.html
76 No. 108, 4th Q. 2017
MATLZER, K., VEIDER, V. & KATHAN, W. (2015): "Adapting to the sharing
economy", MIT Sloan Management Review 56 (2).
O'HEAR, S. (2016): "Deliveroo begins investing in kitchen space to bring more
restaurants to your area, TechCrunch", April 28.
https://techcrunch.com/2016/04/28/soul-kitchen/
OWYANG, J. & SAMUEL, A. (2015): The New Rules of the Collaborative Economy.
VisionCritical.
https://www.visioncritical.com/resources/new-rules-collaborative-economy/
ROOSE, A. (2016): "From 0 to 1,000,000 to ?",Medium, July 26.
https://medium.com/@adrienroose/from-0-to-1-000-000-to-ecb4e2f863c7#.6sw0tvfco
SHAPIRO, C. & VARIAN, H. (1999): "Information Rules: A Strategic Guide to the
Network Economy", Boston: Harvard Business School Press.
SHEAD, S. (2017): "Deliveroo is using an algorithm called 'Frank' to cut food delivery
times by 20%", Business Insider, July 26.
http://uk.businessinsider.com/deliveroo-uses-frank-algorithm-to-cut-delivery-times-by-20-2017-7
SOMERVILLE, H. (2017): "Uber Second-Quarter Bookings Increase, Loss Narrows",
Reuters, August 23.
https://www.reuters.com/article/us-uber-profitability-results/uber-second-quarter-bookings-
increase-loss-narrows-idUSKCN1B32FW
SORKIN, A.R. (2016): "Why Uber Keeps Raising Billions", The New York Times,
June 20.
https://www.nytimes.com/2016/06/21/business/dealbook/why-uber-keeps-raising-billions.html
TANEJA, H. (2015): "Why Startups Are More Successful than Ever at Unbundling
Incumbents", Harvard Business Review, June 18.
VERMEULEN, F. (2017): "What So Many Strategists Get Wrong About Digital
Disruption", Harvard Business Review, January 3.
