Technological disruption poses challenges and opportunities to businesses which have traditionally dominated financial services, with the rate of change and number of challengers increasing. These challenger businesses are often referred to as ‘fintech’.
Fintech business models generally involve the significant use of technology and data analytics which often replace or automate functions historically undertaken by people. This different business model can introduce complex transfer pricing (TP) issues related to intangibles into financial service businesses, which are traditionally focused on people and the provision of capital.
Typically, a start-up or technology company carves out and competes in one area, rather than seeking to offer the full-scale of services that many financial services firms offer. They heavily utilise technology and automation, including the performance of front-office functions to try to gain a competitive edge. This goes beyond the use of technology by many financial service incumbents, which historically is utilised as an enabler of traditional business models.
In responding to these challenges, financial services firms can face internal hurdles to deploy and develop their own technology; for example, the need to show high returns for technology investments to internal stakeholders, as well as legacy system constraints. Given this, acquisitions can also form an important part of their response.
Start-ups, on the other hand, tend to be agile and can create offerings that are highly tailored to specific markets and users. With a heavy focus on aspects such as user experience and interface, fintech business models appeal to an increasingly tech savvy customer base. In addition, tech giants are building large financial services-related patent portfolios, and are increasingly entering regulated areas. Companies like Apple and Google can leverage their extensive reach and user base and their sophisticated in-house development capabilities to offer services that are easily accessible through their products and service ecosystems.
Intangibles in financial services
Analysis of the value related to intangible assets, including the pricing of any intercompany transactions, is a complex area of TP and one that requires special attention from businesses.
However, first it would be helpful to provide examples of what is meant by technology and intangibles in financial services and the issues that tend to be particularly important for TP purposes. (The term ‘intangible’, as defined by the OECD Transfer Pricing Guidelines (OECD TP Guidelines), refers to something which is not a physical asset or a financial asset, which is capable of being owned or controlled for use in commercial activities, and whose use or transfer would be compensated had it occurred in a transaction between independent parties in comparable circumstances).
Here it is possible to make a basic distinction into two categories:
- Technologies that are ‘infrastructure’ such as middle and back-office systems that are focused on supporting existing operations, functions and processes. From a TP perspective, because these are generally supportive in nature, as opposed to profit driving, businesses will normally be focused on costs and where these are borne, and whether they should be charged out within the group (with an appropriate mark-up), as opposed to considering whether some form of super-profit remuneration should be due.
- The second category is where the technology is not supportive in nature, but rather is front and center of the business. Perhaps the technology is capable of making decisions independently of constant human oversight (for example, deciding whether to offer a credit card to a customer). This is where there can be a clear ‘value-add’ provided by the technology. That is, the intangible property contributes to profits above those expected in the industry or for particular people functions, and therefore drives residual profit.
The remainder of this section will focus on this latter category of technology.
Of course, the underlying functionality of both can include many of the common elements of fintech business, such as automation and data analytics. But it is the purpose of the technology and its role in the business, and its contribution to profit driving activities, that should drive the TP outcome.
How should businesses go about seeking to price intercompany transactions (such as royalties) associated with these types of ‘value-add’ technologies? This is important for two reasons: one, to determine whether there should be a charge at all for the intangible property, regardless of the quantum, and two, if it is established that there should be a charge, how much should it be and how can it be supported from a TP perspective?
This can be difficult. An initial consideration is that given that the tech-related intangibles in these types of fintech businesses will often be new and innovative, it is usually difficult to find market comparables from the usual sources, such as third-party databases. Hence, it may not be possible to reliably apply the comparable uncontrolled price (CUP) method.
Alternatives can include the profit split method or corporate finance methodologies such as discounted cash flow. However, the challenge here faced by many groups is finding quantifiable evidence such as cash flows that directly demonstrate the value generated by the technology for the business. This is, of course, perhaps not surprising, as it is typically difficult to separate income or cash flows associated with the intangible property from the overall revenues of the business. Depending on the inherent nature and the functionality of the technology, businesses can therefore consider both the direct and indirect economic benefits derived from an intangible to ascertain its value for TP purposes.
Alternatively, businesses can consider data available from acquisitions. For example, if the group has made an acquisition and a portion of the purchase price was allocated to technology intangibles, this could provide an indication of the total value of technology in the group vis-à-vis other parts of the business.
Finally, businesses should not forget the hard-to-value intangibles (HTVI) material in the OECD TP Guidelines. HTVI may include those which, per paragraph 6.190 of the OECD TP Guidelines, are “expected to be exploited in a manner that is novel at the time of the transfer and the absence of a track record of development or exploitation of similar intangibles makes projections highly uncertain”. As this would describe many fintech business, businesses should monitor any intangibles that meet the HTVI criteria over time, and consider what this could mean for their TP.
Imagine an insurance carrier that has a platform that provides flexible insurance coverage based on the evolving needs and behaviors of the policyholder. The system uses telematics which the carrier uses to assess how risky the policyholder is based on tracking the behavior of the policyholder (e.g. driving at high or low speeds, accelerating aggressively or sedately, etc.).
The technology here is crucial, but it can also require human input in terms of the programming of the parameters (for example, how much the premium will go up based on behavior that is deemed to be riskier). From a TP perspective, the difficulty is then determining how to remunerate the technology system and the human activities that developed the system and provide input into it in terms of ongoing actuarial and underwriting activities. Would ‘routine’ remuneration, such as a cost plus, be appropriate, or should there be a split of residual profits that are attributable to the technology and associated human functions?
However, this assumes that it is possible to measure these residual profits. Businesses should look to consider any economic benefits, direct or indirect, derived from the intangible, as they undergo pricing exercises. For example, in an insurance context:
- Direct benefits could be an increase in premiums or brokerage commissions, or making individual contracts more profitable; and
- Indirect benefits could be creating new opportunities, accessing new markets, saving the time of staff by enhancing underwriting capabilities which could ultimately lead to cost savings, or being able to pursue opportunities and generate more revenues.
Example: High frequency trading
As a relatively mature business, high frequency trading can be used as an example of how some of these issues can apply in a practical scenario. Broadly, high frequency trading involves the use of algorithmic software to place high volume trades based on inputs including market data, financial data, event data, etc. Software is housed on hardware which has a very low latency connection to the relevant exchange (latency refers to the time it takes for the data to be transferred from its source to its destination). Elements of the hardware (e.g. the low latency connection to the exchange, the server equipment) may be provided by third party service providers (i.e. co-location services).
When considering an appropriate TP policy, the following characteristics would be important:
- The server equipment (e.g. high speed, secure, etc.);
- Location of the server (e.g. close proximity to the exchange);
- Provision of market data flow (e.g. the collection of market data relating to the relevant exchange); and
- Software (e.g. development of trading algorithms).
The key challenge is determining an appropriate return for the technology undertaking front office activity which is traditionally performed by people. In doing so, consideration of whether the hardware (and more specifically the location of the hardware in proximity to the exchange) should be rewarded with a non-routine return.
There is no OECD guidance which specifically considers servers in a high frequency trading context. However, comments in a 2001 non-consensus OECD paper entitled ‘Attribution of profit to A permanent establishment involved in electronic commerce transactions’ note that an appropriate remuneration for a server would generally be a routine return (albeit in a marketing server context).
In addition, the existence of third-party co-location service providers in the market could be used to support a realistic alternatives approach to rewarding the server with a routine return. Under this approach, the return generated by third party co-location service providers could be used as a comparable to determine the routine return for the server (although data availability may be an issue)
An argument could be made that low latency and proximity to the exchange are critical to the algorithm used in the software, and to the overall high frequency trading model, such that the server (in location) should receive a non-routine return. The merits of this argument would depend on the specific facts. For example, if there were rights associated with the location (e.g. the right to host a server proximate to the exchange), this could fall within the definition of an intangible asset under the OECD TP Guidelines, and non-routine value could be attributed to it.
On the other hand, without such rights it may be argued that the choice of location is not something capable of being owned, and may more appropriately be considered a location advantage and comparability factor, rather than an intangible asset deriving a non-routine return.
These factors and arguments should be considered by businesses in the context of their specific fact scenarios, as well as any local country guidance from relevant jurisdictions.
It is worth noting that, in addition to the TP points described above, any permanent establishment implications of owning or leasing servers should be considered in the context of local country law (including any exceptions for such activities) and applicable treaties.
Example: Investment management
The investment management sector is one of the last financial services sectors expected to be impacted by fintech, and most changes are anticipated to be in data analytics and automation of asset allocation, which are called ‘robo advisors’.
The ultimate goal of the introduction of the robo advisor is to improve customer service with as little human interaction as possible. The robo advisor performs the functions of investment advisors in the traditional asset management business, through a digital platform. These functions can include:
- Survey and collect data and information from clients on their financial status and their investment goals;
- Evaluate the client’s data and develop an investment strategy; and
- Invest client assets based on the client’s investment goal and the client’s risk profile.
The key function of the robo advisor is the analysis of client data with algorithms, and automatic investment of client assets within the investment strategy.
Robo advisors can also offer clients a low cost alternative compared to traditional advisors, with significantly smaller advisor fees and can most of the time offer the service with no minimum account balance.
The use of technology in place of people raises the question of how TP policies related to robo advisors differ from that of traditional advisors. Since robo advisors are replacing some of the functions of an investment advisor, who are essentially remote, from a TP perspective it is vital to understand and evaluate the modified value chain. This includes consideration of issues that were commonly less relevant in the financial services industry, such as the role of technology, brand intangibles, as well as data and related synergies. Hybrid models, where a client can receive joint advice from a robo and a traditional investment advisor, are being explored by some firms and will require further consideration both from a traditional and new value chain perspective based the functions performed by each advisor.
In a typical value chain analysis for investment management, the location of the investment advisor is a key factor for TP purposes. For robo advisors, as is the case for many other fintech business models, the location of management and control of the development, enhancement, maintenance, protection and exploitation (DEMPE) functions for intangibles is a key consideration.
In addition, analysing functions of the robo advisor related to data management and client service is important in a similar way to analysing front, middle and back-office functions in a traditional business model. For example, client onboarding and administration, typically a front office function in traditional business models can be automated in the robo advisor business model. Similarly, front office functions such as marketing and client acquisition may also be replaced with digital marketing and advertising.
As demonstrated by these examples, the replacement of the human interface of traditional investment advisors with technology and data requires a reassessment of TP policies and analysis.
Fintech business models can present unique challenges from a TP perspective. For financial service incumbents, consideration of the value attributed to intangibles that perform front office functions may present a departure from pricing which historically focused on key people and the provision of capital. As fintech businesses expand, either through organic growth or acquisition, tax and TP should be considered and managed at an early stage in order to align future outcomes with the business’s business and regulatory objectives.
Finally, while this article considers the TP aspects of intangibles used in common fintech business models, an important point going forward is likely to be the value attributed to the data that the technology leverages. Data is generally not considered to be an intangible asset (e.g. on the basis that it does not fall within the definition of an intangible asset under the OECD TP Guidelines). However, as tax authorities may seek to attribute value to market data due to the importance of data to fintech business models, it will be important to monitor developments in this area.
T: +44 20 7007 5350
Jeremy Brown is a director in Deloitte UK’s financial services TP team with 13 years of experience in TP.
Jeremy’s chief focus is on advising multinational clients across the insurance, asset management and banking sectors on TP documentation, planning, economic analysis and engagement with tax authorities. He has considerable experience managing large, complex TP projects covering various types of transactions, including revenue/profit splits, licensing of IP, financing, intra-group reinsurance and services.
Jeremy holds a degree in economics and is qualified as a chartered financial analyst (CFA).
T: +81 080 4183 7349
Luke is a TP senior manager in the Deloitte Tokyo office with 13 years of TP and international tax experience in Japan and Australia.
Luke works with Japanese and foreign multinational clients with a focus on financial services and life sciences. He has considerable experience assisting clients to plan, implement and support their corporate tax positions including complex pricing policies involving profit splits and IP transactions, PE profit attribution and intra-group financing. He supports his clients in tax controversy through audit defence and APAs, as well as compliance and documentation.
Luke holds undergraduate degrees in science and law, and a master of laws. He is qualified as an Australian lawyer.
T: +49 69 7569 57232
Anodri Suchdeve is a senior manager at Deloitte in Frankfurt, Germany.
Anodri started her career at Morgan Stanley and worked in the corporate tax department in New York, and financial control group in Frankfurt, primarily focused on controlling, TP, financial and regulatory reporting and thereafter the operational TP team in New York. After serving many years at Morgan Stanley, she joined Mitsubishi UFJ Financial Group in New York, and led the operational TP group as the Vice President while continuously streamlining operations, systems and especially alleviating and harmonising challenges between business units and related parties.
Anodri advises a number of financial services clients on various TP matters, including operational TP topics.
© 2021 Euromoney Institutional Investor PLC. For help please see our FAQ.