The platform behind the curtain: Obfuscated brokerage on retail trading platforms

Introduction

In recent years, retail trading platforms such as Robinhood in the US, Boom in Asia, and eToro in the EU have emerged as popular gateways to global financial markets for lay people. The unprecedented access to assets and services offered by these platforms have driven enthusiasts to praise them as an empowerment of individual investors and a democratisation of finance (McKinsey & Company, 2017; Stonnington, 2021), while critics warn that frictionless trading risks impoverishing ordinary people to the benefit of big capital (Barber et al., 2022; Preda, 2017). As registered financial brokers, investment platforms have obligations to act in their customers’ best interest and only recommend investments that are suitable for them. However, retail trading platforms are economically incentivised to encourage risky trading behaviour and sell complicated derivative products — strategies that tend to be detrimental to people’s wealth (Barber & Odean, 2000). In some instances, the platform acts as the counterpart to its users’ trades and thus benefits directly from their losses. This creates conflicts of interest reminiscent of the kind digital platforms have dealt with since they began to expand their activities from merely connecting sides to offering their own products and services in the market (Khan, 2018; Rieder & Sire, 2014). In their capacity as middlemen, platforms can exploit their position as they broker relationships between user sides, but they risk public backlash, fines, or further regulation when doing so.

It is underexplored how retail trading platforms, as well as platforms more generally, navigate conflicts of interest in their brokerage operations. In this article, we show how eToro obfuscates their various roles as brokers, and we would argue that the combination of obfuscation and brokerage is relevant to understand platform operations across domains. Google’s ad serving operations, for instance, are purposefully opaque and gives Google potential leverage against competitors (Geradin & Katsifis, 2020), while Amazon’s marketplace operations lets them stealthily outcompete sellers on their platform while posing as matchmaker (Khan, 2016). Platform companies across the board are incentivised to shift from matching users in the market to “making the market” themselves instead, moving between connecting users and acting as counterparts to transactions. In this capacity, the broker masks the degree to which it facilitates trade between complementors versus selling its own services directly to users on terms and prices set by the broker. Our case study of obfuscated brokerage in retail trading thus speaks directly to how platforms may exploit their position as brokers to the detriment of their users.

The paper contributes to the discussion of platform power in empirical and theoretical terms. Empirically, we investigate a type of platform, i.e. retail trading platforms, which is underexplored in platformisation scholarship. Retail trading platforms might not be part of the “infrastructural core” in digital society (van Dijck et al., 2018), but they form an integral part of the overall financial ecosystem and are essential gatekeepers for the growing share of ordinary people that invest in and trade financial assets. Retail trading platforms thus occupy an important role in shaping not just people’s hopes and expectations for the future (Beckert, 2016) but also their life chances due to their direct involvement in lay people’s financial matters. Theoretically, we further the discussion of platform power through our conceptualisation of obfuscated brokerage and compound power through cross-promotion. By emphasising obfuscation of underlying operations and conflicts of interest in brokerage activities, we get purchase on how platforms in general translate their privileged position to wield power over their users, promoting their own products and services while obfuscating terms, costs, and the risks that users incur as part of transactions (Khan, 2018).

We begin by presenting retail trading platforms and their history, noting how their brokering activities both depend upon and depart from other types of digital platforms. We then outline our argument for studying platform power as obfuscated and interwoven brokerage and substantiate this through a case study of the retail trading platform eToro. eToro operates in over 140 countries and offers a variety of asset classes (including stocks and crypto) and derivative products, as well as a social network for trading with successful investors. We analyse three different types of brokerage on the platform — financial, informational, and social — and give examples of the ways in which the roles and operations within and across these domains are obfuscated and interwoven to mask underlying conflicts of interest.

Retail trading platforms: Between matchmaking and market-making

In this section, we introduce retail trading platforms and show how they combine the older match-making and intermediating functions of stock brokerage with the recent developments of digital platforms and their operational strategies.

In the financial sector, key parts of the platform model were dominant long before the advent of the internet. Financial exchanges thus functioned as physical and electronic meeting spaces for corporations seeking liquidity and investors looking for opportunities. However, only registered professionals would have direct access to exchanges, whereas ordinary people who wanted access to financial markets, typically called retail investors, would resort to professional stockbrokers that buy and sell stocks and other assets on behalf of customers (for a history, see Roscoe, 2023). However, these brokers would charge high commission fees in addition to pocketing the difference between buy and sell rates (“the spread”). Now many banks and brokers operate retail trading platforms that enable trading financial assets on a smaller budget and with greater ease than previously. These platforms often offer commission-free trading as well as the ability to buy fractions of shares and derivative products such as options, which enable people to invest smaller sums instantaneously through mobile apps. However, this increased access to financial markets comes at a cost. In lieu of commissions, platforms tax trading transactions in other ways and sell derivative products with complex and opaque fee structures (Schwarz et al., in press). On top of this, some platforms sell retail trading data to institutional investors in a process called payment-for-order-flow akin to data brokerage (Crain, 2018). Thus, the retail trading platforms represent a continuation of the age-old broker business but with a further lowering of barriers to access and expansion of speculative services and products.

Retail trading platforms constitute both a centralised gateway and a fragmented sector in finance. In their capacity as gatekeepers for access to financial markets, the platforms, alongside other brokers, constitute societal infrastructures (van Dijck et al., 2019) as they set the conditions for trade for an increasing number of people that invest on their own. Only institutional traders have access to the markets; the rest have to deal with brokers or other types of intermediaries. At the same time, these platforms do not form an “infrastructural core” (van Dijck et al., 2018) for other platforms and complementors. They depend on the networks of cables and standards of stock exchanges (Pardo-Guerra, 2019), as well as the terms and conditions of the institutional investment sector, all of which are usually outside the control of individual platforms. In addition, the retail trading sector is far less concentrated than the prototypical examples of platformised sectors: there is (as of yet) no dominant behemoth in retail trading in the way Alphabet is for search, Meta for social, and Amazon for retail. This means that discussions of concentration and monopolisation of markets — often figuring centrally in discussions of platform power (see e.g. Thomas, 2023) — are of less relevance here. Instead, our main concern is how retail trading platforms utilise their gatekeeping position to manipulate user behaviour.

This comes to the fore in the conflicting roles that retail trading platforms perform. On the one hand, they connect complementors (buyers and sellers of assets) through their matchmaking operations, and, on the other hand, they constitute a complementor in the market themselves (by selling products and services directly), shifting between a multi-sided and one-sided platform models (Evans & Schmalensee, 2016). First and foremost, retail trading platforms operate as financial brokers who give ordinary people access to trading financial assets. In this role, the platforms act as intermediary middlemen connecting traders to “the market”. This basic function, however, tends to mask underlying complexity. Sometimes the market in question is an actual stock exchange, like NASDAQ, but other times platforms rely on yet another type of financial intermediary called market makers, i.e. companies that trade large volumes of stocks and thus provide liquidity in the markets. Additionally, retail trading platforms often sell products directly (“over-the-counter”) such as futures, options, and contracts for differences (CFDs), the latter of which are derivative instruments based on the movement of an underlying asset. When acting in this capacity, platforms are no longer connecting users to a market. Instead, the platform itself makes the market for the products and services by constituting the counterpart in the deal, i.e. either a buyer or seller depending on the direction of the trade. In this way, retail trading platforms constantly shift between matchmaking (connecting) and market-making (dealing). The ability to perform and obfuscate such role-shifts is, we argue here, a general aspect of the power digital platforms exercise when they facilitate interactions between partner sides.

The compounding of platform power through obfuscated brokerage

The match-making and market-making aspects of retail trading platforms dovetails with an influential stream of platform scholarship, which sees platforms as intermediary matchmakers and operators of multisided markets (Evans & Schmalensee, 2016). Together with earlier scholarship dedicated to electronic markets and market makers (Wigand, 1997; Wigand & Benjamin, 2006), this literature focuses on the intermediary aspects of platform operations seen from the perspective of economics and strategic management. As is well-recognised in the more critically oriented literature on platforms, such digital intermediaries are not neutral. Platforms enable transactions and social interaction, but they also exercise power by governing and exploiting interactions between users and platform complementors through the collection, curation, and commodification of data (Crain, 2018; Gillespie, 2010, 2018; Kleis Nielsen & Ganter, 2018; Poell et al., 2021). As platforms deepen their involvement in societal structures, this leads to “conflicts of interests […] played out at various levels” (van Dijck et al., 2018, p. 2), revolving around tensions between public values vs the interests of platform owners and shareholders. More specifically, conflicts of interest between users and platforms may arise when platforms are incentivised to conduct transactions in self-serving ways for financial gain (Rieder & Sire, 2014). For example, when Google directs users toward more profitable search results (Rieder & Sire, 2014), or Amazon favours its own products over competitors in the rankings (Khan, 2018). In this paper, we explore this specific type of conflict of interest inherent to the operations of retail trading platforms.

Our contribution to discussions of platform power lies in the analytical openings resulting from attending to the combination of financial incentives, brokerage, and obfuscation. The sociological perspective on intermediaries as brokers (Gould & Fernandez, 1989; Stovel & Shaw, 2012) allows for closer attention to both the various types of intermediary roles and, especially, the mechanism of appearing neutral while extracting value from users. A broker must project and maintain an image of powerlessness, neutrality, and disinterest. Yet, “the general problem for middlemen brokers […] is that their activities today may undermine their ability to act as a broker in the future, and the temptation to abuse information in the future undermines trust in them today” (Stovel et al., 2011, p. 21327). Put differently: platforms operating as intermediaries need to constantly downplay or disguise their own position, incentives, and illicit alliances to maintain legitimacy. YouTube, for instance, might derive (much) more revenue from the top creators (for a discussion of this, see Ørmen & Gregersen, 2022), but the platform cannot blatantly play favourites and must maintain the illusion of a level playing field to maintain legitimacy in the eyes of the creator class at large. If brokerage power depends upon being seen as a neutral broker who never sides with anyone and has no individual interest in the transactions flowing through the structural centre, such power inheres partly in the ability to maintain the opacity of actual operations. Therefore, platforms need to obfuscate their organisational and transactional relationships (see e.g. Goldstein & Eaton, 2021; Schilke & Rossman, 2018): As platforms seek to project the image of the ideal broker operating as the truly disinterested intermediary, they must work incessantly to obfuscate the (many) conflicts of interest which result from their extractive practices. This combined perspective thus allows for unpacking how platforms navigate conflicts of interest in practice by directing attention to both structures of brokerage and their connections with coordinated mechanisms of obfuscation. In the following, we demonstrate how obfuscation and interweaving of three types of brokerage leads to compound platform power.

eToro as a paradigmatic case of obfuscated brokerage

Our case study of eToro serves as a paradigmatic case (Flyvbjerg, 2006) of obfuscated brokerage as platform power for two core reasons.

First, eToro is a global platform that brokers the sale of regulated as well as unregulated assets and controversial derivative products operating in more than 140 countries and under different jurisdictions (the EU, the UK, the US, Malta, Australia, Seychelles, and Gibraltar), where it offers different services and terms and conditions (eToro, n.d.e). The platform offers a wide range of financial assets including stocks and bonds, cryptocurrencies (more than 70 in total), indices and ETFs (exchange-traded funds), commodities (gold, oil, natural gas, etc.), and currency pairs. On top of this, eToro sells derivative products directly to users such as options and Contract-For-Differences (CFDs), which differ across jurisdictions. For instance, speculative derivative products such as CFDs are illegal in the US but legal in the EU. Crypto as the most popular asset class on the platform (at the time of writing, almost a third of all users hold Bitcoin) remains unregulated. As the primary market for eToro in its current form is the EU (73% of funded accounts are based in Europe, Statista Research Department, 2024), we use the EU operations and legislations as the basis for the analysis. As our analysis shows, eToro has strong financial incentives in trading crypto and selling CFDs, and the complex regulatory framework mixed with this incentive structure provides a strong case for convoluted brokerage activities.

Second, eToro is a first-mover on automated trading for retail clients. The interface allows users to copy all trades in assets by other investors through the CopyTrading feature or portfolios compiled by eToro experts, a service called Smart Portfolio. The CopyTrading features serves as the core in social trading (Gemayel & Preda, 2018), where the platform seeks to build a community around trading (currently at >30 million users) through social media features and ranking mechanisms. When you copy another trader, referred to as a Popular Investor, or one of the curated Smart Portfolios you allocate some of your capital to follow all market moves made by the investor or portfolio. The basics correspond roughly to letting a professional investor manage your portfolio, but it is arguably better seen as a large-scale technological solution for automating so-called mirror-trading, a practice which predates eToro.¹ The novelty lies in the coupling of social network service functionality and a tiered-governance framework (Caplan & Gillespie, 2020) with influencer dynamics of visibility and prominence (Cotter, 2019) to financial trading. This structure connects the social and informational brokerage taking place on digital platforms to the automation of trading.

Our case study identifies and highlights three different kinds of brokerage operations, i.e. financial, informational, and social, as case-specific operations of eToro. At the same time, the practices of obfuscation involved in these operations are analytically generalisable and thus speak both to retail trading platforms and digital platforms more generally.

Methods

To analyse the network of brokerage practices and their connections, our study combines a document analysis (Karppinen & Moe, 2012) of eToro’s website and public communication with a brief critical interface analysis of eToro’s trading app (Burgess, 2021). The process of document collection and analysis was steered by our interest in mapping the roles and operations of the platform and relationships to external regulation of the platform. The key sources were eToro’s official documents on:

• Client Terms and Conditions, eToro Europe, October 2023 (eToro, 2023).
• Best Execution and Order Handling Policy, eToro Europe, September 2022 (eToro, 2022b)
• Conflicts of Interest Policy, eToro (Europe) Ltd, n.d. (eToro, 2022a)
• Key Information Documents, eToro Europe, September 2023 (eToro, n.d.b)
• Fees, eToro Europe, September 2022 (eToro, 2022c).

Important background sources were the EU legal documents that regulate financial markets in the EU, primarily Markets in Financial Instruments Directive, in its current form called MiFID2 (Directive 2014/65/EU). Finally, we consulted journalistic coverage of the platform across territories for context.

Our case study is theory driven, and we use examples to illustrate key instances of the complexity inherent in the various roles and operations of eToro when brokering financial assets, informational goods and services, and social connections. The overall goal is not so much to analyse the reality of the documents (Atkinson & Coffey, 2010), but rather use several sources of data to iteratively map platform operations and roles. The key concepts driving the analysis are brokerage roles, obfuscation of relationships, financial incentives, and conflicts of interest related to potential profits and losses. We focus on the official documents and use the external regulatory texts primarily as background for progressively guiding and steering our mapping. Since most of our materials are documents authored by the platform, we treat them as both highly strategic and potentially unreliable. However, they can serve as useful entry points into the various brokerage roles since the platform is legally required to disclose specific types of information and as background for critically assessing how services and products are presented to the users through the platform interface.

Financial brokerage — This is not an exchange or an asset market

Recall that ordinary non-professional investors cannot deal directly on the exchanges for most financial assets (but not crypto), but have to access markets through brokers. Thus, in its simplest terms, financial brokerage would entail matching buyers with sellers in the various asset markets. Such matchmaking would correspond to the typical understanding of platforms as multi-sided markets, but this is not how financial brokerage works on retail trading platforms like eToro. Instead, the platform operates as an over-the-counter (OTC) broker, where traders are never directly connected to the financial exchanges but instead deal with the platform as a counterpart to all trades (eToro, 2022b). The platform might buy assets in the market on your behalf through other market-makers or on exchanges but it might as well just be the direct seller of their own products and services. The broker thus switches surreptitiously between matchmaking, connecting users to markets, and market-making, where they primarily connect users with the platform itself.

The platform mainly derives revenue from taxing transactions, either through “the spread” (the difference between buy and sell prices), traditional trading fees to crypto transactions (which are thus not commission free), interest and fees in relation to CFD positions where investors borrow money from the platform, as well as fees for both inactivity and withdrawal of assets (eToro, 2022c). Although buyers can be said to own the assets they buy, they cannot retrieve them from the platform. As stated in the terms and conditions:

The eToro trading platform is not an exchange or a market. This means that:

(a) you can only enter into trades and investments with us on the platform, and not third parties;

(b) all trades opened on our platform must be closed on our platform;

(c) all products which you purchase on our platform can only be sold on our platform, and not a third party platform;

(d) you will generally not be able to transfer products into your eToro account, out of your eToro account or to a third party at any time. However, we reserve the right to permit and support this functionality at our discretion, including, for example, the ability for you to transfer certain products between your eToro account and electronic wallets operate by an eToro Europe affiliate; and

(e) our prices will be different from the prices provided by other brokers, the market price, as well as the current prices on any exchanges or trading platforms. (eToro, 2023, p. 5, emphasis added)

An oft-used term for this ownership structure is that users are beneficial owners of the assets while the assets themselves are held by the platform; as such, assets are infrastructurally locked-in and can only be liquidated through (and by) the platform.² Traders thus become subject to changes in the terms and conditions as well as other strategic actions by the platform. You might own the assets in the legal sense, but in practical matters you cannot take ownership of them, and you are legally at the mercy of the platform when it comes to whether assets should be liquidated. As examples, eToro recently delisted selected crypto assets and announced a full stop in trading them, while the rival platform Robinhood announced that it would forcefully liquidate assets in user accounts at a fixed date (Quiroz-Gutierrez, 2023). In sum, OTC brokerage obfuscates matters of ownership and control, making traders’ assets into platform contingent property (Nieborg & Poell, 2018).

At other times, different types of conflict of interest are baked into the transaction itself. This is the case with contract-for-differences (CFDs), a particular type of complex derivative product. CFDs are fundamentally different from traditional assets and their historical roots lie in spread betting (Loussouarn, 2013), which allows short-term betting on either upwards or downwards movements of financial assets without any assets being traded. In the case of modern CFDs, the relationship is typically a bilateral contract between the person taking out the bet and the entity offering the contract, in this case eToro, with no involvement of other market actors and no purchasing of assets. This means that when the trader makes money, the platform loses and vice versa.³

Here, role conflation is central to the obfuscation of actual operations since the platform is no longer offering financial brokerage but instead dealership where the platform sets prices and conditions for its bespoke derivative products. Given the asymmetry of resources related to both financial and juridical matters, this client-dealer relationship is very likely to favour the platform: eToro’s own Key Information Documents (eToro, n.d.b) state that CFDs are “not simple and may be difficult to understand” and rates them at the highest level of risk possible (7 out of 7). eToro is required by EU law to disclose statistics on how risky CFD trading can be for traders, and at the time of writing this was the message on their website (etoro.com): “76% of retail investor accounts lose money when trading CFDs with this provider” (October 27, 2023). Regulation further demands that the platform screens traders based on a risk profile before allowing them to trade CFDs, but a recent lawsuit by the Australian Securities and Investment Commission (ASIC) deemed eToro’s screening efforts “wholly inadequate” and far too broad in the type of investors they gave access to (ASIC, 2023), indicating that the platform encourages and broadly allows the trading of risky derivatives.

This leads us to our final point related to financial brokerage, namely that the incentives to obscure the nature of risky, complex, and costly trades can be identified at the level of the platform interface. eToro arguably applies deceptive design patterns in multiple instances on the platform (Rakovic & Inal, 2023) to hide the real complexity and costs of trading. We will give only one example of such a pattern here, namely the central trading screen dialogue. Here the trader has the option to buy or sell a chosen asset. If the trader chooses to “buy” a stock such as NVDA, the stock ticker for Nvidia Corporation, (Figure 1.A) she undertakes a traditional trade in the market acquiring a number of fractions of the asset at a given price, which are then registered in the trader’s name and protected by EU legislation including rights to compensation (eToro, 2023). The trader only pays through “the spread” for this transaction. However, if the trader buys a cryptocoin such as Bitcoin instead (1.B) she still acquires the underlying asset but is not protected nor compensated if the broker defaults. Furthermore, the price is now the spread and a 1% commission fee, none of which is disclosed in the transaction interface. Furthermore, if the trader changes the “leverage” settings, which are highlighted in blue and take up a far larger share of the screen than the simple buy settings (1.A), then the underlying transaction completely changes. Now (1.C) the trader is not buying the asset but is instead engaging in a CFD trade with leverage. Applying leverage basically means that the trader borrows money from the broker to increase their investment. The consequence is that the trader increases their exposure and can earn or lose much more than they invested in the first place.⁴ Thus, this strategy can lead to quick riches or more likely, following the previously stated abysmal CFD success rate, rapid poverty. Although buying and selling are presented as equal options in the platform interface, they represent radically different actions and relationships with the platform, the first as broker and the latter as dealer. Should the trader choose “sell” (1.D) the trade is carried out as a CFD whether leverage is applied or not. Although eToro is required to notify candidates on the complexity of financial instruments such as derivatives, the platform reduces this complexity in the trading app. On the interface level, risky and complicated options become visually equivalent to more straightforward trading actions by way of commensuration through interface design.

In sum, the platform obfuscates the kind of transaction users engage in, the type of ownership they have, as well as the complexity of financial transactions. The interface might look and feel like a market, but users do not directly interact with any sellers apart from the platform itself nor do they really own the assets they buy. Thus, in many (but not all) of its financial activities, the broker acts more like a seller of products and services than a platform connecting complementors. The power of platforms stems in part from their ability to disguise which role they occupy in relation to users — as connectors or counterparts — at any given moment.

Information brokerage — This is not financial advice

In addition to their provision of various financial services, eToro runs an extensive communication and information operation. Through the platform, users can access financial data and business news curated from an abundance of data sources, including media organisations, relevant to financial markets. In addition, eToro offers an educational universe, i.e. the eToro Academy, where users can take online courses about trading and investment from the most basic steps to more advanced strategies. We see these activities as constituting information brokerage where the platform leverages its privileged access to data and information through targeted distribution of information to its users. Here, the obfuscation largely happens through the organised display of “commensurated metrics” (Espeland & Stevens, 1998), bringing an enormous range of assets and data points into the same simplified scopic regime (extending Knorr Cetina, 2004). The central point of contention for the platform is to deliver the right kind of information to the right users without falling under the strict regulation for providing proper financial consulting.

The central aspect of the platform’s information brokerage is the provision of a plethora of real-time market information via statistics, charts, and analysis tools. A result of ongoing surveillance, digestion, and commensuration this flow of information, much like the professional Bloomberg Terminal (see Figure 2, on the left), enacts the reality of financial markets through past performances and projections of future expectations (Knorr Cetina, 2004). The presentation is shaped by the infrastructures and conventions of the financial world: standardised market information through ticker tracking, statistical operations, and finance-specific visualisations. In the current instantiation, eToro offers more than 50 charts and tools across more than 100 market indicators as well as numerous ways to tweak and edit charts. With these tools it is possible to visualise the performance of individual assets across time and in dimensions based on the exact statistics and charts selected. On the right in Figure 2, a standard “candle stick” graph for the NVDA stock is shown in the eToro app. This type of visualisation gives detailed information about the day-to-day movements of stock prices, which are relevant to day traders (people that trade frequently, typically within the same day) and investors that rely on technical analysis. Both of these trading strategies, however, require extensive knowledge of and professional training in finance to pull off with any success, and even then it is difficult to make money (Barber & Odean, 2000). Increased access to data and analytical tools does not translate to equality of opportunity as retail investors lack the professional training and resources available to institutional investors. Instead, eToro’s scopic regime arguably creates a mere veneer of expertise where retail traders can feel empowered and enticed to adopt active trading strategies. The charts and tools thus obfuscate through scopic simplicity, where the complexity of finance gets boiled down to operating slick drop-down menus in point-and-click interfaces. These commensurated simplified metrics become objectified as indicators of “the market” existing as an objective fact. This hiding in plain sight among bespoke metrics is also arguably a way to deflect accountability of the platform.

In addition, the platform extends information brokerage to its own distinct financial products. Based on large-scale access to market data as well as in-house analysts, eToro has developed a series of curated lists of assets called Smart Portfolios where investors can allocate funds to a portfolio managed by eToro. Such portfolios may consist only of certain types of assets (like stocks or crypto), or they may combine asset classes and derivative products. While portfolios are presented as a smart way to invest “long-term” and diversify one’s investments, Smart Portfolios are very different from more traditional long-term investment strategies. First, they are not like passive index funds and exchange traded funds (ETFS), but “more like the active approach, they are regularly rebalanced and fine-tuned by the eToro investment team” (eToro, n.d.a), which is arguably an attempt by the platform to avoid the responsibilities of funds management as required by EU law (Directive 2011/61/EU). Second, this “fine-tuning” takes place at eToro’s discretion — most portfolios seem to be adjusted on a yearly basis, but the platform reserves the right to change this schedule at any time. Third, when copying the trades in the portfolio and every time the composition of assets is readjusted, the trader incurs all trading costs as normally. In sum, a Smart Portfolio transfers user agency to the platform which then in practice trades on the user’s behalf with limited accountability and oversight and all the fee structures associated with active trading.

As in the previous section on financial brokerage, we would argue that eToro’s role as information broker obfuscates a set of problematic role conflations and resulting conflicts of interest. In financial legislation there is a difference between merely providing information about financial products and services and giving investment advice. The latter is a formalised role that requires certification, demands disclosure of potential conflicts of interest, and entails responsibility of acting in the client’s best interest. To sidestep these requirements, however, many platforms (and financial influencers) frequently state that “this is not financial advice”. In etoro’s case, the disclaimer reads: “Any explanation or information which we give to you as part of a trade, or a copy trade, or about the performance of the trade or copy trade is not intended to be, and should not be considered as advice” (eToro, 2023, p. 5). However, at least within the EU, this is an empty speech act. In the EU legislation for financial services, explicit statements and other kinds of meta-communication do not let advisors off the hook. Instead, any information that is presented as an endorsement of a particular action in the markets addressed to a person in her capacity as an investor and tailored to that person’s situation would count as investment advice (ESMA, 2023). Thus, in its capacity as information broker, eToro walks a regulatory tightrope between distributing (neutral) media services and providing (liable) financial advice.

Social brokerage — This is not a collective endeavour

As mentioned in our initial presentation of the case, the unique selling point for eToro is its positioning as the social trading platform. Figure 3 is a screencap of the landing page of the company’s website at the time of writing, which emphasises the social nature of investing and the resources provided: the many users (30 million) are there to provide “investment ideas”, and the platform itself is both “trusted and friendly”. Elsewhere on the website you are invited to “join the eToro community” which provides access to ”the collective wisdom of millions of investors” because ”investing is social”. The platform has four features to substantiate this element of sociality, i.e. personal profiles, a social news feed, the ability to copy other investors, and the eToro-sanctioned Popular Investor program (eToro, n.d.d). The personal profile, and related mechanics of following others plus a personalised news feed, are obviously shared with dominant social media platforms, although the feed has bespoke trading platform features such as push-notifications relevant to the assets in one’s portfolio. The novel parts are the combination of financial services and social brokerage epitomised in the Popular Investor programme and its implementation of copy trading, which forms an integrated system for managing investment activities across tiers of users.

To attract successful investors and incentivise them to cultivate active trading strategies, eToro operates the Popular Investor program. It is a tiered governance system, known from consumer segmentation and used by contemporary platforms (Caplan & Gillespie, 2020), where successful traders can attract followers and copiers as well as gain remuneration. Popular Investors are thus rewarded for attracting new copiers both in terms of beneficial conditions such as cheaper trading and currency conversion as well as direct monetary compensation in the form of a share of the total assets under management they control (eToro, 2022a). eToro compares this role to traditional asset managers in practical matters, but without the fiduciary duty this role requires in a legal sense (eToro, n.d.c). Popular investors are incentivised to attract as many copy traders as possible and keep them on board, no matter how poorly the portfolio performs. This creates a conflict of interest between the platform and the popular investors against the copiers: eToro benefits from the frequent trading the popular investors drive, the popular investors benefit from cheaper trading expenses removing key downsides of active trading strategies, while the copy investors stand to pay the price in both fees and risks.

The Popular Investor programme is directly integrated with the promotional functions of eToro’s information brokerage, where traders are ranked according to their performance. This ostensible transparency of trading activities on the platform, whereby each trader’s performance is visible to others, constitutes a scopic regime in itself (Gemayel & Preda, 2018), but this one ranks investors instead of assets. The default ranking criteria are number of copiers and portfolio performance the past year. As future gains are notoriously difficult to predict based on earlier performance, the ranking system likely amplifies the rich-get-richer tendencies by giving top investors enhanced exposure. While the platform does not demand that investors pursue a specific strategy, the current top investors ranked by copiers (Figure 4) tend to trade excessively (ranging from about 184 trades per year or 3.9 per week to 1,237 trades per year or 23.3 per week). Again, every time a Popular Investor trades, all copiers follow suit and the platform profits from the trading fees and spreads.

Finally, there is a sense in which the reality of sociality is obscured through promotional efforts. While selling itself as a social trading platform, the actual sociality fostered on the platform can be very thin indeed. First, as copy-trading involves complete displacement of agency to other investors and bots (recall Smart Portfolios, eToro, n.d.a), the relationship is much more like robo-trading or professional wealth management than a community of peers. Second, although the platform places much emphasis on the collective aspect of trading, the risks remain individualised in contrast to investment clubs where people pool resources and decide collectively on which assets to buy and strategies to follow (Harrington, 2008). On top of this, incentive structures, ranking systems, and opaque fee structures are obscured through the marketing material, website promotions, and the platform interface which all promote community and collective endeavour, making it difficult to see that real risks are unevenly distributed in the community of traders, where Popular Investors accrue special benefits and trade under more favourable conditions. Thus, copiers and Popular Investors might be faring the same treacherous seas, but they do so in different boats. Behind all the community-focused marketing material and social media functionalities, this is not a collective endeavour but a strictly individual struggle where the platform itself can be a helper or an adversary, or at least counterpart, in the process.

Concluding discussion

As our analysis has shown, there is power lodged in each of the brokerage types that eToro delivers. The ability to switch between various brokerage roles and thus relationships to end users is a key source of power as illustrated by the case study. Financial brokerage exploits the fact that it is still unfeasible for most individuals to interact directly with the financial markets. This gives retail trading platforms the opportunity to not only connect users to markets to also to make the markets themselves. In the latter position the platforms trade directly with the user, like buying a product from a reseller, instead of finding counterparts in the financial markets. This structural position allows for taxing transactions and exercising pricing power as it sells bespoke financial products and services at costs and conditions set by the platform itself. Thus, in this capacity the broker acts more as a firm than as a platform. This may also lead to obfuscated but direct conflicts of interest between platform and user, where the platform profits from user losses. The platform retains ownership and actual control of assets and obfuscates the complexity of the trades through the simplicity of the parlance and interface design. The information brokerage role circulates packaged information about financial markets. By making market data available through an extensive toolbox for analysis and visualisation, the platform offers a user experience with a veneer of expertise that mimics a professional trading setup designed for active and risky trading instead of passive investment strategies. Information brokerage can create and exploit information asymmetry by targeting users with communications that promote various financial assets, services, and investors aligned with the interests of the platform. The convoluted fee structure by way of spreads can make it costly to compare prices across the retail trading sector (Schwarz et al., in press), and trusting the platform lowers the cost of search for users but may expose them to deals which favour the platform instead of the user. Finally, social brokerage offers power over both individual traders and the copytraders in a manner well-known from operators of multi-sided market structures, especially the model deployed by digital media platforms and their tiered governance regimes designed to manage the creator sides. Popular Investors, much like creators, have limited options besides alignment with the platform, since the commensuration and subsequent valuation of their position is communicated through bespoke metrics which rank-order and make visible specific traders over others. This element of social brokerage both extends privileges to properly aligned Popular Investors while also funnelling regular users towards alignment with the overall business model of the financial operations arm.

These three brokerage functions each embody a specific aspect of platform power, but when they are interwoven the result is a compounding of power. The various types of brokerage — financial, informational, and social — act as levers that can be used strategically to raise the value of the others. This is most visible in the way information brokerage yields promotional power across all activities of the platform, giving preference to trading strategies, products, services, and individuals fundamentally aligned with the interests of the platform. In this way, eToro can e.g. promote trading of cryptocurrencies (which yields better spreads), high frequency trading strategies (which yields an increase in commissions), and Popular Investors aligned with their own interest in risky and frequent trading (which drive trading on the platform overall). However, as the analysis demonstrates, the inherent conflicts of interest need to be obfuscated in brokering the relationship between users and complementors for the platform to maintain legitimacy and profitability. Like all informational regimes this is fundamentally a question of information asymmetry, of controlling invisibility and opacity as well as visibility and apparent transparency.

To summarise, we have analysed in detail how eToro’s brokerage roles and activities are obfuscated and interwoven to mask conflicts of interest and unethical business practices. This double focus speaks directly to larger questions of platform power, market dominance, and regulatory arbitrage and capture, as platform companies constantly seek to expand their integration into new domains, move toward ever more dominant intermediary positions, forge alliances, and avoid regulation. Following Rieder & Sire (2014), conflicts of interest potentially arise whenever platforms engage in various kinds of brokerage, matching sides and making markets for transactions, and this goes for financial products, informational services, as well as physical goods: any platform that poses as a neutral intermediary should be a prime suspect when it comes to self-serving operations. Seen from this angle, platforms are neither markets nor neutral intermediaries but increasingly powerful brokers who must toil to maintain a face of legitimacy by papering over constant conflicts of interest while they seek to further exploit their structural position and connect ever more individuals and organisations to their operational nexus. Obfuscated and interwoven brokerage thus constitutes a core part of the work that platforms do to mask their machinations.