research paper on banking and finance pdf

• Open access
• Published: 18 June 2021

Financial technology and the future of banking

• Daniel Broby   ORCID: orcid.org/0000-0001-5482-0766 1  

Financial Innovation volume  7 , Article number:  47 ( 2021 ) Cite this article

48k Accesses

62 Citations

5 Altmetric

Metrics details

This paper presents an analytical framework that describes the business model of banks. It draws on the classical theory of banking and the literature on digital transformation. It provides an explanation for existing trends and, by extending the theory of the banking firm, it illustrates how financial intermediation will be impacted by innovative financial technology applications. It further reviews the options that established banks will have to consider in order to mitigate the threat to their profitability. Deposit taking and lending are considered in the context of the challenge made from shadow banking and the all-digital banks. The paper contributes to an understanding of the future of banking, providing a framework for scholarly empirical investigation. In the discussion, four possible strategies are proposed for market participants, (1) customer retention, (2) customer acquisition, (3) banking as a service and (4) social media payment platforms. It is concluded that, in an increasingly digital world, trust will remain at the core of banking. That said, liquidity transformation will still have an important role to play. The nature of banking and financial services, however, will change dramatically.

Introduction

The bank of the future will have several different manifestations. This paper extends theory to explain the impact of financial technology and the Internet on the nature of banking. It provides an analytical framework for academic investigation, highlighting the trends that are shaping scholarly research into these dynamics. To do this, it re-examines the nature of financial intermediation and transactions. It explains how digital banking will be structurally, as well as physically, different from the banks described in the literature to date. It does this by extending the contribution of Klein ( 1971 ), on the theory of the banking firm. It presents suggested strategies for incumbent, and challenger banks, and how banking as a service and social media payment will reshape the competitive landscape.

The banking industry has been evolving since Banca Monte dei Paschi di Siena opened its doors in 1472. Its leveraged business model has proved very scalable over time, but it is now facing new challenges. Firstly, its book to capital ratios, as documented by Berger et al ( 1995 ), have been consistently falling since 1840. This trend continues as competition has increased. In the past decade, the industry has experienced declines in profitability as measured by return on tangible equity. This is partly the result of falling leverage and fee income and partly due to the net interest margin (connected to traditional lending activity). These trends accelerated following the 2008 financial crisis. At the same time, technology has made banks more competitive. Advances in digital technology are changing the very nature of banking. Banks are now distributing services via mobile technology. A prolonged period of very low interest rates is also having an impact. To sustain their profitability, Brei et al. ( 2020 ) note that many banks have increased their emphasis on fee-generating services.

As Fama ( 1980 ) explains, a bank is an intermediary. The Internet is, however, changing the way financial service providers conduct their role. It is fundamentally changing the nature of the banking. This in turn is changing the nature of banking services, and the way those services are delivered. As a consequence, in order to compete in the changing digital landscape, banks have to adapt. The banks of the future, both incumbents and challengers, need to address liquidity transformation, data, trust, competition, and the digitalization of financial services. Against this backdrop, incumbent banks are focused on reinventing themselves. The challenger banks are, however, starting with a blank canvas. The research questions that these dynamics pose need to be investigated within the context of the theory of banking, hence the need to revise the existing analytical framework.

Banks perform payment and transfer functions for an economy. The Internet can now facilitate and even perform these functions. It is changing the way that transactions are recorded on ledgers and is facilitating both public and private digital currencies. In the past, banks operated in a world of information asymmetry between themselves and their borrowers (clients), but this is changing. This differential gave one bank an advantage over another due to its knowledge about its clients. The digital transformation that financial technology brings reduces this advantage, as this information can be digitally analyzed.

Even the nature of deposits is being transformed. Banks in the future will have to accept deposits and process transactions made in digital form, either Central Bank Digital Currencies (CBDC) or cryptocurrencies. This presents a number of issues: (1) it changes the way financial services will be delivered, (2) it requires a discussion on resilience, security and competition in payments, (3) it provides a building block for better cross border money transfers and (4) it raises the question of private and public issuance of money. Braggion et al ( 2018 ) consider whether these represent a threat to financial stability.

The academic study of banking began with Edgeworth ( 1888 ). He postulated that it is based on probability. In this respect, the nature of the business model depends on the probability that a bank will not be called upon to meet all its liabilities at the same time. This allows banks to lend more than they have in deposits. Because of the resultant mismatch between long term assets and short-term liabilities, a bank’s capital structure is very sensitive to liquidity trade-offs. This is explained by Diamond and Rajan ( 2000 ). They explain that this makes a bank a’relationship lender’. In effect, they suggest a bank is an intermediary that has borrowed from other investors.

Diamond and Rajan ( 2000 ) argue a lender can negotiate repayment obligations and that a bank benefits from its knowledge of the customer. As shall be shown, the new generation of digital challenger banks do not have the same tradeoffs or knowledge of the customer. They operate more like a broker providing a platform for banking services. This suggests that there will be more than one type of bank in the future and several different payment protocols. It also suggests that banks will have to data mine customer information to improve their understanding of a client’s financial needs.

The key focus of Diamond and Rajan ( 2000 ), however, was to position a traditional bank is an intermediary. Gurley and Shaw ( 1956 ) describe how the customer relationship means a bank can borrow funds by way of deposits (liabilities) and subsequently use them to lend or invest (assets). In facilitating this mediation, they provide a service whereby they store money and provide a mechanism to transmit money. With improvements in financial technology, however, money can be stored digitally, lenders and investors can source funds directly over the internet, and money transfer can be done digitally.

A review of financial technology and banking literature is provided by Thakor ( 2020 ). He highlights that financial service companies are now being provided by non-deposit taking contenders. This paper addresses one of the four research questions raised by his review, namely how theories of financial intermediation can be modified to accommodate banks, shadow banks, and non-intermediated solutions.

To be a bank, an entity must be authorized to accept retail deposits. A challenger bank is, therefore, still a bank in the traditional sense. It does not, however, have the costs of a branch network. A peer-to-peer lender, meanwhile, does not have a deposit base and therefore acts more like a broker. This leads to the issue that this paper addresses, namely how the banks of the future will conduct their intermediation.

In order to understand what the bank of the future will look like, it is necessary to understand the nature of the aforementioned intermediation, and the way it is changing. In this respect, there are two key types of intermediation. These are (1) quantitative asset transformation and, (2) brokerage. The latter is a common model adopted by challenger banks. Figure  1 depicts how these two types of financial intermediation match savers with borrowers. To avoid nuanced distinction between these two types of intermediation, it is common to classify banks by the services they perform. These can be grouped as either private, investment, or commercial banking. The service sub-groupings include payments, settlements, fund management, trading, treasury management, brokerage, and other agency services.

How banks act as intermediaries between lenders and borrowers. This function call also be conducted by intermediaries as brokers, for example by shadow banks. Disintermediation occurs over the internet where peer-to-peer lenders match savers to lenders

Financial technology has the ability to disintermediate the banking sector. The competitive pressures this results in will shape the banks of the future. The channels that will facilitate this are shown in Fig.  2 , namely the Internet and/or mobile devices. Challengers can participate in this by, (1) directly matching borrows with savers over the Internet and, (2) distributing white labels products. The later enables banking as a service and avoids the aforementioned liquidity mismatch.

The strategic options banks have to match lenders with borrowers. The traditional and challenger banks are in the same space, competing for business. The distributed banks use the traditional and challenger banks to white label banking services. These banks compete with payment platforms on social media. The Internet heralds an era of banking as a service

There are also physical changes that are being made in the delivery of services. Bricks and mortar branches are in decline. Mobile banking, or m-banking as Liu et al ( 2020 ) describe it, is an increasingly important distribution channel. Robotics are increasingly being used to automate customer interaction. As explained by Vishnu et al ( 2017 ), these improve efficiency and the quality of execution. They allow for increased oversight and can be built on legacy systems as well as from a blank canvas. Application programming interfaces (APIs) are bringing the same type of functionality to m-banking. They can be used to authorize third party use of banking data. How banks evolve over time is important because, according to the OECD, the activity in the financial sector represents between 20 and 30 percent of developed countries Gross Domestic Product.

In summary, financial technology has evolved to a level where online banks and banking as a service are challenging incumbents and the nature of banking mediation. Banking is rapidly transforming because of changes in such technology. At the same time, the solving of the double spending problem, whereby digital money can be cryptographically protected, has led to the possibility that paper money will become redundant at some point in the future. A theoretical framework is required to understand this evolving landscape. This is discussed next.

The theory of the banking firm: a revision

In financial theory, as eloquently explained by Fama ( 1980 ), banking provides an accounting system for transactions and a portfolio system for the storage of assets. That will not change for the banks of the future. Fama ( 1980 ) explains that their activities, in an unregulated state, fulfil the Modigliani–Miller ( 1959 ) theorem of the irrelevance of the financing decision. In practice, traditional banks compete for deposits through the interest rate they offer. This makes the transactional element dependent on the resulting debits and credits that they process, essentially making banks into bookkeeping entities fulfilling the intermediation function. Since this is done in response to competitive forces, the general equilibrium is a passive one. As such, the banking business model is vulnerable to disruption, particularly by innovation in financial technology.

A bank is an idiosyncratic corporate entity due to its ability to generate credit by leveraging its balance sheet. That balance sheet has assets on one side and liabilities on the other, like any corporate entity. The assets consist of cash, lending, financial and fixed assets. On the other side of the balance sheet are its liabilities, deposits, and debt. In this respect, a bank’s equity and its liabilities are its source of funds, and its assets are its use of funds. This is explained by Klein ( 1971 ), who notes that a bank’s equity W , borrowed funds and its deposits B is equal to its total funds F . This is the same for incumbents and challengers. This can be depicted algebraically if we let incumbents be represented by Φ and challengers represented by Γ:

Klein ( 1971 ) further explains that a bank’s equity is therefore made up of its share capital and unimpaired reserves. The latter are held by a bank to protect the bank’s deposit clients. This part is also mandated by regulation, so as to protect customers and indeed the entire banking system from systemic failure. These protective measures include other prudential requirements to hold cash reserves or other liquid assets. As shall be shown, banking services can be performed over the Internet without these protections. Banking as a service, as this phenomenon known, is expected to increase in the future. This will change the nature of the protection available to clients. It will change the way banks transform assets, explained next.

A bank’s deposits are said to be a function of the proportion of total funds obtained through the issuance of the ith deposit type and its total funds F , represented by α i . Where deposits, represented by Bs , are made in the form of Bs (i  =  1 *s n) , they generate a rate of interest. It follows that Si Bs  =  B . As such,

Therefor it can be said that,

The importance of Eq. 3 is that the balance sheet can be leveraged by the issuance of loans. It should be noted, however, that not all loans are returned to the bank in whole or part. Non-performing loans reduce the asset side of a bank’s balance sheet and act as a constraint on capital, and therefore new lending. Clearly, this is not the case with banking as a service. In that model, loans are brokered. That said, with the traditional model, an advantage of financial technology is that it facilitates the data mining of clients’ accounts. Lending can therefore be more targeted to borrowers that are more likely to repay, thereby reducing non-performing loans. Pari passu, the incumbent bank of the future will therefore have a higher risk-adjusted return on capital. In practice, however, banking as a service will bring greater competition from challengers and possible further erosion of margins. Alternatively, some banks will proactively engage in partnerships and acquisitions to maintain their customer base and address the competition.

A bank must have reserves to meet the demand of customers demanding their deposits back. The amount of these reserves is a key function of banking regulation. The Basel Committee on Banking Supervision mandates a requirement to hold various tiers of capital, so that banks have sufficient reserves to protect depositors. The Committee also imposes a framework for mitigating excessive liquidity risk and maturity transformation, through a set Liquidity Coverage Ratio and Net Stable Funding Ratio.

Recent revisions of theory, because of financial technology advances, have altered our understanding of banking intermediation. This will impact the competitive landscape and therefor shape the nature of the bank of the future. In this respect, the threat to incumbent banks comes from peer-to-peer Internet lending platforms. These perform the brokerage function of financial intermediation without the use of the aforementioned banking balance sheet. Unlike regulated deposit takers, such lending platforms do not create assets and do not perform risk and asset transformation. That said, they are reliant on investors who do not always behave in a counter cyclical way.

Financial technology in banking is not new. It has been used to facilitate electronic markets since the 1980’s. Thakor ( 2020 ) refers to three waves of application of financial innovation in banking. The advent of institutional futures markets and the changing nature of financial contracts fundamentally changed the role of banks. In response to this, academics extended the concept of a bank into an entity that either fulfills the aforementioned functions of a broker or a qualitative asset transformer. In this respect, they connect the providers and users of capital without changing the nature of the transformation of the various claims to that capital. This transformation can be in the form risk transfer or the application of leverage. The nature of trading of financial assets, however, is changing. Price discovery can now be done over the Internet and that is moving liquidity from central marketplaces (like the stock exchange) to decentralized ones.

Alongside these trends, in considering what the bank of the future will look like, it is necessary to understand the unregulated lending market that competes with traditional banks. In this part of the lending market, there has been a rise in shadow banks. The literature on these entities is covered by Adrian and Ashcraft ( 2016 ). Shadow banks have taken substantial market share from the traditional banks. They fulfil the brokerage function of banks, but regulators have only partial oversight of their risk transformation or leverage. The rise of shadow banks has been facilitated by financial technology and the originate to distribute model documented by Bord and Santos ( 2012 ). They use alternative trading systems that function as electronic communication networks. These facilitate dark pools of liquidity whereby buyers and sellers of bonds and securities trade off-exchange. Since the credit crisis of 2008, total broker dealer assets have diverged from banking assets. This illustrates the changed lending environment.

In the disintermediated market, banking as a service providers must rely on their equity and what access to funding they can attract from their online network. Without this they are unable to drive lending growth. To explain this, let I represent the online network. Extending Klein ( 1971 ), further let Ψ represent banking as a service and their total funds by F . This state is depicted as,

Theoretically, it can be shown that,

Shadow banks, and those disintermediators who bypass the banking system, have an advantage in a world where technology is ubiquitous. This becomes more apparent when costs are considered. Buchak et al. ( 2018 ) point out that shadow banks finance their originations almost entirely through securitization and what they term the originate to distribute business model. Diversifying risk in this way is good for individual banks, as banking risks can be transferred away from traditional banking balance sheets to institutional balance sheets. That said, the rise of securitization has introduced systemic risk into the banking sector.

Thus, we can see that the nature of banking capital is changing and at the same time technology is replacing labor. Let A denote the number of transactions per account at a period in time, and C denote the total cost per account per time period of providing the services of the payment mechanism. Klein ( 1971 ) points out that, if capital and labor are assumed to be part of the traditional banking model, it can be observed that,

It can therefore be observed that the total service charge per account at a period in time, represented by S, has a linear and proportional relationship to bank account activity. This is another variable that financial technology can impact. According to Klein ( 1971 ) this can be summed up in the following way,

where d is the basic bank decision variable, the service charge per transaction. Once again, in an automated and digital environment, financial technology greatly reduces d for the challenger banks. Swankie and Broby ( 2019 ) examine the impact of Artificial Intelligence on the evaluation of banking risk and conclude that it improves such variables.

Meanwhile, the traditional banking model can be expressed as a product of the number of accounts, M , and the average size of an account, N . This suggests a banks implicit yield is it rate of interest on deposits adjusted by its operating loss in each time period. This yield is generated by payment and loan services. Let R 1 depict this. These can be expressed as a fraction of total demand deposits. This is depicted by Klein ( 1971 ), if one assumes activity per account is constant, as,

As a result, whether a bank is structured with traditional labor overheads or built digitally, is extremely relevant to its profitability. The capital and labor of tradition banks, depicted as Φ i , is greater than online networks, depicted as I i . As such, the later have an advantage. This can be shown as,

What Klein (1972) failed to highlight is that the banking inherently involves leverage. Diamond and Dybving (1983) show that leverage makes bank susceptible to run on their liquidity. The literature divides these between adverse shock events, as explained by Bernanke et al ( 1996 ) or moral hazard events as explained by Demirgu¨¸c-Kunt and Detragiache ( 2002 ). This leverage builds on the balance sheet mismatch of short-term assets with long term liabilities. As such, capital and liquidity are intrinsically linked to viability and solvency.

The way capital and liquidity are managed is through credit and default management. This is done at a bank level and a supervisory level. The Basel Committee on Banking Supervision applies capital and leverage ratios, and central banks manage interest rates and other counter-cyclical measures. The various iterations of the prudential regulation of banks have moved the microeconomic theory of banking from the modeling of risk to the modeling of imperfect information. As mentioned, shadow and disintermediated services do not fall under this form or prudential regulation.

The relationship between leverage and insolvency risk crucially depends on the degree of banks total funds F and their liability structure L . In this respect, the liability structure of traditional banks is also greater than online networks which do not have the same level of available funds, depicted as,

Diamond and Dybvig ( 1983 ) observe that this liability structure is intimately tied to a traditional bank’s assets. In this respect, a bank’s ability to finance its lending at low cost and its ability to achieve repayment are key to its avoidance of insolvency. Online networks and/or brokers do not have to finance their lending, simply source it. Similarly, as brokers they do not face capital loss in the event of a default. This disintermediates the bank through the use of a peer-to-peer environment. These lenders and borrowers are introduced in digital way over the internet. Regulators have taken notice and the digital broker advantage might not last forever. As a result, the future may well see greater cooperation between these competing parties. This also because banks have valuable operational experience compared to new entrants.

It should also be observed that bank lending is either secured or unsecured. Interest on an unsecured loan is typically higher than the interest on a secured loan. In this respect, incumbent banks have an advantage as their closeness to the customer allows them to better understand the security of the assets. Berger et al ( 2005 ) further differentiate lending into transaction lending, relationship lending and credit scoring.

The evolution of the business model in a digital world

As has been demonstrated, the bank of the future in its various manifestations will be a consequence of the evolution of the current banking business model. There has been considerable scholarly investigation into the uniqueness of this business model, but less so on its changing nature. Song and Thakor ( 2010 ) are helpful in this respect and suggest that there are three aspects to this evolution, namely competition, complementary and co-evolution. Although liquidity transformation is evolving, it remains central to a bank’s role.

All the dynamics mentioned are relevant to the economy. There is considerable evidence, as outlined by Levine ( 2001 ), that market liberalization has a causal impact on economic growth. The impact of technology on productivity should prove positive and enhance the functioning of the domestic financial system. Indeed, market liberalization has already reshaped banking by increasing competition. New fee based ancillary financial services have become widespread, as has the proprietorial use of balance sheets. Risk has been securitized and even packaged into trade-able products.

Challenger banks are developing in a complementary way with the incumbents. The latter have an advantage over new entrants because they have information on their customers. The liquidity insurance model, proposed by Diamond and Dybvig ( 1983 ), explains how such banks have informational advantages over exchange markets. That said, financial technology changes these dynamics. It if facilitating the processing of financial data by third parties, explained in greater detail in the section on Open Banking.

At the same time, financial technology is facilitating banking as a service. This is where financial services are delivered by a broker over the Internet without resort to the balance sheet. This includes roboadvisory asset management, peer to peer lending, and crowd funding. Its growth will be facilitated by Open Banking as it becomes more geographically adopted. Figure  3 illustrates how these business models are disintermediating the traditional banking role and matching burrowers and savers.

The traditional view of banks ecosystem between savers and borrowers, atop the Internet which is matching savers and borrowers directly in a peer-to-peer way. The Klein ( 1971 ) theory of the banking firm does not incorporate the mirrored dynamics, and as such needs to be extended to reflect the digital innovation that impacts both borrowers and severs in a peer-to-peer environment

Meanwhile, the banking sector is co-evolving alongside a shadow banking phenomenon. Lenders and borrowers are interacting, but outside of the banking sector. This is a concern for central banks and banking regulators, as the lending is taking place in an unregulated environment. Shadow banking has grown because of financial technology, market liberalization and excess liquidity in the asset management ecosystem. Pozsar and Singh ( 2011 ) detail the non-bank/bank intersection of shadow banking. They point out that shadow banking results in reverse maturity transformation. Incumbent banks have blurred the distinction between their use of traditional (M2) liabilities and market-based shadow banking (non-M2) liabilities. This impacts the inter-generational transfers that enable a bank to achieve interest rate smoothing.

Securitization has transformed the risk in the banking sector, transferring it to asset management institutions. These include structured investment vehicles, securities lenders, asset backed commercial paper investors, credit focused hedge and money market funds. This in turn has led to greater systemic risk, the result of the nature of the non-traded liabilities of securitized pooling arrangements. This increased risk manifested itself in the 2008 credit crisis.

Commercial pressures are also shaping the banking industry. The drive for cost efficiency has made incumbent banks address their personally costs. Bank branches have been closed as technology has evolved. Branches make it easier to withdraw or transfer deposits and challenger banks are not as easily able to attract new deposits. The banking sector is therefore looking for new point of customer contact, such as supermarkets, post offices and social media platforms. These structural issues are occurring at the same time as the retail high street is also evolving. Banks have had an aggressive roll out of automated telling machines and a reduction in branches and headcount. Online digital transactions have now become the norm in most developed countries.

The financing of banks is also evolving. Traditional banks have tended to fund illiquid assets with short term and unstable liquid liabilities. This is one of the key contributors to the rise to the credit crisis of 2008. The provision of liquidity as a last resort is central to the asset transformation process. In this respect, the banking sector experienced a shock in 2008 in what is termed the credit crisis. The aforementioned liquidity mismatch resulted in the system not being able to absorb all the risks associated with subprime lending. Central banks had to resort to quantitative easing as a result of the failure of overnight funding mechanisms. The image of the entire banking sector was tarnished, and the banks of the future will have to address this.

The future must learn from the mistakes of the past. The structural weakness of the banking business model cannot be solved. That said, the latest Basel rules introduce further risk mitigation, improved leverage ratios and increased levels of capital reserve. Another lesson of the credit crisis was that there should be greater emphasis on risk culture, governance, and oversight. The independence and performance of the board, the experience and the skill set of senior management are now a greater focus of regulators. Internal controls and data analysis are increasingly more robust and efficient, with a greater focus on a banks stable funding ratio.

Meanwhile, the very nature of money is changing. A digital wallet for crypto-currencies fulfills much the same storage and transmission functions of a bank; and crypto-currencies are increasing being used for payment. Meanwhile, in Sweden, stores have the right to refuse cash and the majority of transactions are card based. This move to credit and debit cards, and the solving of the double spending problem, whereby digital money can be crypto-graphically protected, has led to the possibility that paper money could be replaced at some point in the future. Whether this might be by replacement by a CBDC, or decentralized digital offering, is of secondary importance to the requirement of banks to adapt. Whether accommodating crytpo-currencies or CBDC’s, Kou et al. ( 2021 ) recommend that banks keep focused on alternative payment and money transferring technologies.

Central banks also have to adapt. To limit disintermediation, they have to ensure that the economic design of their sponsored digital currencies focus on access for banks, interest payment relative to bank policy rate, banking holding limits and convertibility with bank deposits. All these developments have implications for banks, particularly in respect of funding, the secure storage of deposits and how digital currency interacts with traditional fiat money.

Open banking

Against the backdrop of all these trends and changes, a new dynamic is shaping the future of the banking sector. This is termed Open Banking, already briefly mentioned. This new way of handling banking data protocols introduces a secure way to give financial service companies consensual access to a bank’s customer financial information. Figure  4 illustrates how this works. Although a fairly simple concept, the implications are important for the banking industry. Essentially, a bank customer gives a regulated API permission to securely access his/her banking website. That is then used by a banking as a service entity to make direct payments and/or download financial data in order to provide a solution. It heralds an era of customer centric banking.

How Open Banking operates. The customer generates data by using his bank account. A third party provider is authorized to access that data through an API request. The bank confirms digitally that the customer has authorized the exchange of data and then fulfills the request

Open Banking was a response to the documented inertia around individual’s willingness to change bank accounts. Following the Retail Banking Review in the UK, this was addressed by lawmakers through the European Union’s Payment Services Directive II. The legislation was designed to make it easier to change banks by allowing customers to delegate authority to transfer their financial data to other parties. As a result of this, a whole host of data centric applications were conceived. Open banking adds further momentum to reshaping the future of banking.

Open Banking has a number of quite revolutionary implications. It was started so customers could change banks easily, but it resulted in some secondary considerations which are going to change the future of banking itself. It gives a clear view of bank financing. It allows aggregation of finances in one place. It also allows can give access to attractive offerings by allowing price comparisons. Open Banking API’s build a secure online financial marketplace based on data. They also allow access to a larger market in a faster way but the third-party providers for the new entrants. Open Banking allows developers to build single solutions on an API addressing very specific problems, like for example, a cash flow based credit rating.

Romānova et al. ( 2018 ) undertook a questionnaire on the Payment Services Directive II. The results suggest that Open Banking will promote competitiveness, innovation, and new product development. The initiative is associated with low costs and customer satisfaction, but that some concerns about security, privacy and risk are present. These can be mitigated, to some extent, by secure protocols and layered permission access.

Discussion: strategic options

Faced with these disruptive trends, there are four strategic options for market participants to con- sider. There are (1) a defensive customer retention strategy for incumbents, (2) an aggressive customer acquisition strategy for challenger banks (3) a banking as a service strategy for new entrants, and (4) a payments strategy for social media platforms.

Each of these strategies has to be conducted in a competitive marketplace for money demand by potential customers. Figure  5 illustrates where the first three strategies lie on the tradeoff between money demand and interest rates. The payment strategy can’t be modeled based on the supply of money. In the figure, the market settles at a rate L 2 . The incumbent banks have the capacity to meet the largest supply of these loans. The challenger banks have a constrained function but due to a lower cost base can gain excess rent through higher rates of interest. The peer-to-peer bank as a service brokers must settle for the market rate and a constrained supply offering.

The money demand M by lenders on the y axis. Interest rates on the y axis are labeled as r I and r II . The challenger banks are represented by the line labeled Γ. They have a price and technology advantage and so can lend at higher interest rates. The brokers are represented by the line labeled Ω. They are price takers, accepting the interest rate determined by the market. The same is true for the incumbents, represented by the line labeled Φ but they have a greater market share due to their customer relationships. Note that payments strategy for social media platforms is not shown on this figure as it is not affected by interest rates

Figure  5 illustrates that having a niche strategy is not counterproductive. Liu et al ( 2020 ) found that banks performing niche activities exhibit higher profitability and have lower risk. The syndication market now means that a bank making a loan does not have to be the entity that services it. This means banks in the future can better shape their risk profile and manage their lending books accordingly.

An interesting question for central banks is what the future Deposit Supply function will look like. If all three forms: open banking, traditional banking and challenger banks develop together, will the bank of the future have the same Deposit Supply function? The Klein ( 1971 ) general formulation assumes that deposits are increasing functions of implicit and explicit yields. As such, the very nature of central bank directed monetary policy may have to be revisited, as alluded to in the earlier discussion on digital money.

The client retention strategy (incumbents)

The competitive pressures suggest that incumbent banks need to focus on customer retention. Reichheld and Kenny ( 1990 ) found that the best way to do this was to focus on the retention of branch deposit customers. Obviously, another way is to provide a unique digital experience that matches the challengers.

Incumbent banks have a competitive advantage based on the information they have about their customers. Allen ( 1990 ) argues that where risk aversion is observable, information markets are viable. In other words, both bank and customer benefit from this. The strategic issue for them, therefore, becomes the retention of these customers when faced with greater competition.

Open Banking changes the dynamics of the banking information advantage. Borgogno and Colangelo ( 2020 ) suggest that the access to account (XS2A) rule that it introduced will increase competition and reduce information asymmetry. XS2A requires banks to grant access to bank account data to authorized third payment service providers.

The incumbent banks have a high-cost base and legacy IT systems. This makes it harder for them to migrate to a digital world. There are, however, also benefits from financial technology for the incumbents. These include reduced cost and greater efficiency. Financial technology can also now support platforms that allow incumbent banks to sell NPL’s. These platforms do not require the ownership of assets, they act as consolidators. The use of technology to monitor the transactions make the processing cost efficient. The unique selling point of such platforms is their centralized point of contact which results in a reduction in information asymmetry.

Incumbent banks must adapt a number of areas they got to adapt in terms of their liquidity transformation. They have to adapt the way they handle data. They must get customers to trust them in a digital world and the way that they trust them in a bricks and mortar world. It is no coincidence. When you go into a bank branch that is a great big solid building great big facade and so forth that is done deliberately so that you trust that bank with your deposit.

The risk of having rising non-performing loans needs to be managed, so customer retention should be selective. One of the puzzles in banking is why customers are regularly denied credit, rather than simply being charged a higher price for it. This credit rationing is often alleviated by collateral, but finance theory suggests value is based on the discounted sum of future cash flows. As such, it is conceivable that the bank of the future will use financial technology to provide innovative credit allocation solutions. That said, the dual risks of moral hazard and information asymmetries from the adoption of such solutions must be addressed.

Customer retention is especially important as bank competition is intensifying, as is the digitalization of financial services. Customer retention requires innovation, and that innovation has been moving at a very fast rate. Until now, banks have traditionally been hesitant about technology. More recently, mergers and acquisitions have increased quite substantially, initiated by a need to address actual or perceived weaknesses in financial technology.

The client acquisition strategy (challengers)

As intermediaries, the challenger banks are the same as incumbent banks, but designed from the outset to be digital. This gives them a cost and efficiency advantage. Anagnostopoulos ( 2018 ) suggests that the difference between challenger and traditional banks is that the former address its customers problems more directly. The challenge for such banks is customer acquisition.

Open Banking is a major advantage to challenger banks as it facilitates the changing of accounts. There is widespread dissatisfaction with many incumbent banks. Open Banking makes it easier to change accounts and also easier to get a transaction history on the client.

Customer acquisition can be improved by building trust in a brand. Historically, a bank was physically built in a very robust manner, hence the heavy architecture and grand banking halls. This was done deliberately to engender a sense of confidence in the deposit taking institution. Pure internet banks are not able to do this. As such, they must employ different strategies to convey stability. To do this, some communicate their sustainability credentials, whilst others use generational values-based advertising. Customer acquisition in a banking context is traditionally done by offering more attractive rates of interest. This is illustrated in Fig.  5 by the intersect of traditional banks with the market rate of interest, depicted where the line Γ crosses L 2 . As a result of the relationship with banking yield, teaser rates and introductory rates are common. A customer acquisition strategy has risks, as consumers with good credit can game different challenger banks by frequently changing accounts.

Most customer acquisition, however, is done based on superior service offering. The functionality of challenger banking accounts is often superior to incumbents, largely because the latter are built on legacy databases that have inter-operability issues. Having an open platform of services is a popular customer acquisition technique. The unrestricted provision of third-party products is viewed more favorably than a restricted range of products.

The banking as a service strategy (new entrants)

Banking from a customer’s perspective is the provision of a service. Customers don’t care about the maturity transformation of banking balance sheets. Banking as a service can be performed without recourse to these balance sheets. Banking products are brokered, mostly by new entrants, to individuals as services that can be subscribed to or paid on a fee basis.

There are a number banking as a service solutions including pre-paid and credit cards, lending and leasing. The banking as a service brokers are effectively those that are aggregating services from others using open banking to enable banking as a service.

The rise of banking as a service needs to be understood as these compete directly with traditional banks. As explained, some of these do this through peer-to-peer lending over the internet, others by matching borrows and sellers, conducting mediation as a loan broker. Such entities do not transform assets and do not have banking licenses. They do not have a branch network and often don not have access to deposits. This means that they have no insurance protection and can be subject to interest rate controls.

The new genre of financial technology, banking as a service provider, conduct financial services transformation without access to central bank liquidity. In a distributed digital asset world, the assets are stored on a distributed ledger rather than a traditional banking ledger. Financial technology has automated credit evaluation, savings, investments, insurance, trading, banking payments and risk management. These banking as a service offering are only as secure as the technology on which they are built.

The social media payment strategy (disintermediators and disruptors)

An intermediation bank is a conceptual idea, one created solely on a social networking site. Social media has developed a market for online goods and services. Williams ( 2018 ) estimates that there are 2.46 billion social media users. These all make and receive payments of some kind. They demand security and functionality. Importantly, they have often more clients than most banks. As such, a strategy to monetize the payments infrastructure makes sense.

All social media platforms are rich repositories of data. Such platforms are used to buy and sell things and that requires payments. Some platforms are considering evolving their own digital payment, cutting out the banks as middlemen. These include Facebook’s Diem (formerly Libra), a digital currency, and similar developments at some of the biggest technology companies. The risk with social media payment platform is that there is systemic counter-party protection. Regulators need to address this. One way to do this would be to extend payment service insurance to such platforms.

Social media as a platform moves the payment relationship from a transaction to a customer experience. The ability to use consumer desires in combination with financial data has the potential to deliver a number of new revenue opportunities. These will compete directly with the banks of the future. This will have implications for (1) the money supply, (2) the market share of traditional banks and, (3) the services that payment providers offer.

Further research

Several recommendations for research derive from both the impact of disintermediation and the four proposed strategies that will shape banking in the future. The recommendations and suggestions are based on the mentioned papers and the conclusions drawn from them.

As discussed, the nature of intermediation is changing, and this has implications for the pricing of risk. The role of interest rates in banking will have to be further reviewed. In a decentralized world based on crypto currencies the central banks do not have the same control over the money supply, This suggest the quantity theory of money and the liquidity preference theory need to be revisited. As explained, the Internet reduces much of the friction costs of intermediation. Researchers should ask how this will impact maturity transformation. It is also fair to ask whether at some point in the future there will just be one big bank. This question has already been addressed in the literature but the Internet facilities the possibility. Diamond ( 1984 ) and Ramakrishnan and Thakor ( 1984 ) suggested the answer was due to diversification and its impact on reducing monitoring costs.

Attention should be given by academics to the changing nature of banking risk. How should regulators, for example, address the moral hazard posed by challenger banks with weak balance sheets? What about deposit insurance? Should it be priced to include unregulated entities? Also, what criteria do borrowers use to choose non-banking intermediaries? The changing risk environment also poses two interesting practical questions. What will an online bank run look like, and how can it be averted? How can you establish trust in digital services?

There are also research questions related to the nature of competition. What, for example, will be the nature of cross border competition in a decentralized world? Is the credit rationing that generates competition a static or dynamic phenomena online? What is the value of combining consumer utility with banking services?

Financial intermediaries, like banks, thrive in a world of deficits and surpluses supported by information asymmetries and disconnectedness. The connectivity of the internet changes this dynamic. In this respect, the view of Schumpeter ( 1911 ) on the role of financial intermediaries needs revisiting. Lenders and borrows can be connected peer to peer via the internet.

All the dynamics mentioned change the nature of moral hazard. This needs further investigation. There has been much scholarly research on the intrinsic riskiness of the mismatch between banking assets and liabilities. This mismatch not only results in potential insolvency for a single bank but potentially for the whole system. There has, for example, been much debate on the whether a bank can be too big to fail. As a result of the riskiness of the banking model, the banks of the future will be just a liable to fail as the banks of the past.

This paper presented a revision of the theory of banking in a digital world. In this respect, it built on the work of Klein ( 1971 ). It provided an overview of the changing nature of banking intermediation, a result of the Internet and new digital business models. It presented the traditional academic view of banking and how it is evolving. It showed how this is adapted to explain digital driven disintermediation.

It was shown that the banking industry is facing several documented challenges. Risk is being taken of balance sheet, securitized, and brokered. Financial technology is digitalizing service delivery. At the same time, the very nature of intermediation is being changed due to digital currency. It is argued that the bank of the future not only has to face these competitive issues, but that technology will enhance the delivery of banking services and reduce the cost of their delivery.

The paper further presented the importance of the Open Banking revolution and how that facilitates banking as a service. Open Banking is increasing client churn and driving banking as a service. That in turn is changing the way products are delivered.

Four strategies were proposed to navigate the evolving competitive landscape. These are for incumbents to address customer retention; for challengers to peruse a low-cost digital experience; for niche players to provide banking as a service; and for social media platforms to develop payment platforms. In all these scenarios, the banks of the future will have to have digital strategies for both payments and service delivery.

It was shown that both incumbents and challengers are dependent on capital availability and borrowers credit concerns. Nothing has changed in that respect. The risks remain credit and default risk. What is clear, however, is the bank has become intrinsically linked with technology. The Internet is changing the nature of mediation. It is allowing peer to peer matching of borrowers and savers. It is facilitating new payment protocols and digital currencies. Banks need to evolve and adapt to accommodate these. Most of these questions are empirical in nature. The aim of this paper, however, was to demonstrate that an understanding of the banking model is a prerequisite to understanding how to address these and how to develop hypotheses connected with them.

In conclusion, financial technology is changing the future of banking and the way banks intermediate. It is facilitating digital money and the online transmission of financial assets. It is making banks more customer enteric and more competitive. Scholarly investigation into banking has to adapt. That said, whatever the future, trust will remain at the core of banking. Similarly, deposits and lending will continue to attract regulatory oversight.

Availability of data and materials

Diagrams are my own and the code to reproduce them is available in the supplied Latex files.

Adrian T, Ashcraft AB (2016) Shadow banking: a review of the literature. In: Banking crises. Palgrave Macmillan, London, pp 282–315

Allen F (1990) The market for information and the origin of financial intermediation. J Financ Intermed 1(1):3–30

Article   Google Scholar  

Anagnostopoulos I (2018) Fintech and regtech: impact on regulators and banks. J Econ Bus 100:7–25

Berger AN, Herring RJ, Szegö GP (1995) The role of capital in financial institutions. J Bank Finance 19(3–4):393–430

Berger AN, Miller NH, Petersen MA, Rajan RG, Stein JC (2005) Does function follow organizational form? Evidence from the lending practices of large and small banks. J Financ Econ 76(2):237–269

Bernanke B, Gertler M, Gilchrist S (1996) The financial accelerator and the flight to quality. The review of economics and statistics, pp1–15

Bord V, Santos JC (2012) The rise of the originate-to-distribute model and the role of banks in financial intermediation. Federal Reserve Bank N Y Econ Policy Rev 18(2):21–34

Google Scholar  

Borgogno O, Colangelo G (2020) Data, innovation and competition in finance: the case of the access to account rule. Eur Bus Law Rev 31(4)

Braggion F, Manconi A, Zhu H (2018) Is Fintech a threat to financial stability? Evidence from peer-to-Peer lending in China, November 10

Brei M, Borio C, Gambacorta L (2020) Bank intermediation activity in a low-interest-rate environment. Econ Notes 49(2):12164

Buchak G, Matvos G, Piskorski T, Seru A (2018) Fintech, regulatory arbitrage, and the rise of shadow banks. J Financ Econ 130(3):453–483

Demirgüç-Kunt A, Detragiache E (2002) Does deposit insurance increase banking system stability? An empirical investigation. J Monet Econ 49(7):1373–1406

Diamond DW (1984) Financial intermediation and delegated monitoring. Rev Econ Stud 51(3):393–414

Diamond DW, Dybvig PH (1983) Bank runs, deposit insurance, and liquidity. J Polit Econ 91(3):401–419

Diamond DW, Rajan RG (2000) A theory of bank capital. J Finance 55(6):2431–2465

Edgeworth FY (1888) The mathematical theory of banking. J Roy Stat Soc 51(1):113–127

Fama EF (1980) Banking in the theory of finance. J Monet Econ 6(1):39–57

Gurley JG, Shaw ES (1956) Financial intermediaries and the saving-investment process. J Finance 11(2):257–276

Klein MA (1971) A theory of the banking firm. J Money Credit Bank 3(2):205–218

Kou G, Akdeniz ÖO, Dinçer H, Yüksel S (2021) Fintech investments in European banks: a hybrid IT2 fuzzy multidimensional decision-making approach. Financ Innov 7(1):1–28

Levine R (2001) International financial liberalization and economic growth. Rev Interna Tional Econ 9(4):688–702

Liu FH, Norden L, Spargoli F (2020) Does uniqueness in banking matter? J Bank Finance 120:105941

Pozsar Z, Singh M (2011) The nonbank-bank nexus and the shadow banking system. IMF working papers, pp 1–18

Ramakrishnan RT, Thakor AV (1984) Information reliability and a theory of financial intermediation. Rev Econ Stud 51(3):415–432

Reichheld FF, Kenny DW (1990) The hidden advantages of customer retention. J Retail Bank 12(4):19–24

Romānova I, Grima S, Spiteri J, Kudinska M (2018) The payment services directive 2 and competitiveness: the perspective of European Fintech companies. Eur Res Stud J 21(2):5–24

Modigliani F, Miller MH (1959) The cost of capital, corporation finance, and the theory of investment: reply. Am Econ Rev 49(4):655–669

Schumpeter J (1911) The theory of economic development. Harvard Econ Stud XLVI

Song F, Thakor AV (2010) Financial system architecture and the co-evolution of banks and capital markets. Econ J 120(547):1021–1055

Swankie GDB, Broby D (2019) Examining the impact of artificial intelligence on the evaluation of banking risk. Centre for Financial Regulation and Innovation, white paper

Thakor AV (2020) Fintech and banking: What do we know? J Financ Intermed 41:100833

Vishnu S, Agochiya V, Palkar R (2017) Data-centered dependencies and opportunities for robotics process automation in banking. J Financ Transf 45(1):68–76

Williams MD (2018) Social commerce and the mobile platform: payment and security perceptions of potential users. Comput Hum Behav 115:105557

Download references

Acknowledgements

There are no acknowldgements.

There was no funding associated with this paper.

Author information

Authors and affiliations.

Centre for Financial Regulation and Innovation, Strathclyde Business School, Glasgow, UK

Daniel Broby

You can also search for this author in PubMed   Google Scholar

Contributions

The author confirms the contribution is original and his own. All authors read and approved the final manuscript.

Corresponding author

Correspondence to Daniel Broby .

Ethics declarations

Competing interests.

I declare I have no competing interests.

Additional information

Publisher's note.

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/ .

Reprints and permissions

About this article

Cite this article.

Broby, D. Financial technology and the future of banking. Financ Innov 7 , 47 (2021). https://doi.org/10.1186/s40854-021-00264-y

Download citation

Received : 21 January 2021

Accepted : 09 June 2021

Published : 18 June 2021

DOI : https://doi.org/10.1186/s40854-021-00264-y

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

• Cryptocurrencies
• P2P Lending
• Intermediation
• Digital Payments

JEL Classifications

• Open access
• Published: 08 June 2020

Deep learning in finance and banking: A literature review and classification

• Jian Huang 1 ,
• Junyi Chai   ORCID: orcid.org/0000-0003-1560-845X 2 &
• Stella Cho 2  

Frontiers of Business Research in China volume  14 , Article number:  13 ( 2020 ) Cite this article

68k Accesses

114 Citations

70 Altmetric

Metrics details

Deep learning has been widely applied in computer vision, natural language processing, and audio-visual recognition. The overwhelming success of deep learning as a data processing technique has sparked the interest of the research community. Given the proliferation of Fintech in recent years, the use of deep learning in finance and banking services has become prevalent. However, a detailed survey of the applications of deep learning in finance and banking is lacking in the existing literature. This study surveys and analyzes the literature on the application of deep learning models in the key finance and banking domains to provide a systematic evaluation of the model preprocessing, input data, and model evaluation. Finally, we discuss three aspects that could affect the outcomes of financial deep learning models. This study provides academics and practitioners with insight and direction on the state-of-the-art of the application of deep learning models in finance and banking.

Introduction

Deep learning (DL) is an advanced technique of machine learning (ML) based on artificial neural network (NN) algorithms. As a promising branch of artificial intelligence, DL has attracted great attention in recent years. Compared with conventional ML techniques such as support vector machine (SVM) and k-nearest neighbors (kNN), DL possesses advantages of the unsupervised feature learning, a strong capability of generalization, and a robust training power for big data. Currently, DL has been applied comprehensively in classification and prediction tasks, computer visions, image processing, and audio-visual recognition (Chai and Li 2019 ). Although DL was developed in the field of computer science, its applications have penetrated diversified fields such as medicine, neuroscience, physics and astronomy, finance and banking (F&B), and operations management (Chai et al. 2013 ; Chai and Ngai 2020 ). The existing literature lacks a good overview of DL applications in F&B fields. This study attempts to bridge this gap.

While DL is the focus of computer vision (e.g., Elad and Aharon 2006 ; Guo et al. 2016 ) and natural language processing (e.g., Collobert et al. 2011 ) in the mainstream, DL applications in F&B are developing rapidly. Shravan and Vadlamani (2016) investigated the tools of text mining for F&B domains. They examined the representative ML algorithms, including SVM, kNN, genetic algorithm (GA), and AdaBoost. Butaru et al. ( 2016 ) compared performances of DL algorithms, including random forests, decision trees, and regularized logistic regression. They found that random forests gained the highest classification accuracy in the delinquency status.

Cavalcante et al. ( 2016 ) summarized the literature published from 2009 to 2015. They analyzed DL models, including multi-layer perceptron (MLP) (a fast library for approximate nearest neighbors), Chebyshev functional link artificial NN, and adaptive weighting NN. Although the study constructed a prediction framework in financial trading, some notable DL techniques such as long short-term memory (LSTM) and reinforcement learning (RL) models are neglect. Thus, the framework cannot ascertain the optimal model in a specific condition.

The reviews of the existing literature are either incomplete or outdated. However, our study provides a comprehensive and state-of-the-art review that could capture the relationships between typical DL models and various F&B domains. We identified critical conditions to limit our collection of articles. We employed academic databases in Science Direct, Springer-Link Journal, IEEE Xplore, Emerald, JSTOR, ProQuest Database, EBSCOhost Research Databases, Academic Search Premier, World Scientific Net, and Google Scholar to search for articles. We used two groups of keywords for our search. One group is related to the DL, including “deep learning,” “neural network,” “convolutional neural networks” (CNN), “recurrent neural network” (RNN), “LSTM,” and “RL.” The other group is related to finance, including “finance,” “market risk,” “stock risk,” “credit risk,” “stock market,” and “banking.” It is important to conduct cross searches between computer-science-related and finance-related literature. Our survey exclusively focuses on the financial application of DL models rather than other DL models like SVM, kNN, or random forest. The time range of our review was set between 2014 and 2018. In this stage, we collected more than 150 articles after cross-searching. We carefully reviewd each article and considered whether it is worthy of entering our pool of articles for review. We removed the articles if they are not from reputable journals or top professional conferences. Moreover, articles were discarded if the details of financial DL models presented were not clarified. Thus, 40 articles were selected for this review eventually.

This study contributes to the literature in the following ways. First, we systematically review the state-of-the-art applications of DL in F&B fields. Second, we summarize multiple DL models regarding specified F&B domains and identify the optimal DL model of various application scenarios. Our analyses rely on the data processing methods of DL models, including preprocessing, input data, and evaluation rules. Third, our review attempts to bridge the technological and application levels of DL and F&B, respectively. We recognize the features of various DL models and highlight their feasibility toward different F&B domains. The penetration of DL into F&B is an emerging trend. Researchers and financial analysts should know the feasibilities of particular DL models toward a specified financial domain. They usually face difficulties due to the lack of connections between core financial domains and numerous DL models. This study will fill this literature gap and guide financial analysts.

The rest of this paper is organized as follows. Section 2 provides a background of DL techniques. Section 3 introduces our research framework and methodology. Section 4 analyzes the established DL models. Section 5 analyzes key methods of data processing, including data preprocessing and data inputs. Section 6 captures appeared criteria for evaluating the performance of DL models. Section 7 provides a general comparison of DL models against identified F&B domains. Section 8 discusses the influencing factors in the performance of financial DL models. Section 9 concludes and outlines the scope for promising future studies.

Background of deep learning

Regarding DL, the term “deep” presents the multiple layers that exist in the network. The history of DL can be traced back to stochastic gradient descent in 1952, which is employed for an optimization problem. The bottleneck of DL at that time was the limit of computer hardware, as it was very time-consuming for computers to process the data. Today, DL is booming with the developments of graphics processing units (GPUs), dataset storage and processing, distributed systems, and software such as Tensor Flow. This section briefly reviews the basic concept of DL, including NN and deep neural network (DNN). All of these models have greatly contributed to the applications in F&B.

The basic structure of NN can be illustrated as Y  =  F ( X T w  +  c ) regarding the independent (input) variables X , the weight terms w , and the constant terms c . Y is the dependent variable and X is formed as an n  ×  m matrix for the number of training sample n and the number of input variables m . To apply this structure in finance, Y can be considered as the price of next term, the credit risk level of clients, or the return rate of a portfolio. F is an activation function that is unique and different from regression models. F is usually formulated as sigmoid functions and tanh functions. Other functions can also be used, including ReLU functions, identity functions, binary step functions, ArcTan functions, ArcSinh functions, ISRU functions, ISRLU functions, and SQNL functions. If we combine several perceptrons in each layer and add a hidden layer from Z 1 to Z 4 in the middle, we term a single layer as a neural network, where the input layers are the X s , and the output layers are the Y s . In finance, Y can be considered as the stock price. Moreover, multiple Y s are also applicable; for instance, fund managers often care about future prices and fluctuations. Figure  1 illustrates the basic structure.

The structure of NN

Based on the basic structure of NN shown in Fig.  1 , traditional networks include DNN, backpropagation (BP), MLP, and feedforward neural network (FNN). Using these models can ignore the order of data and the significance of time. As shown in Fig.  2 , RNN has a new NN structure that can address the issues of long-term dependence and the order between input variables. As financial data in time series are very common, uncovering hidden correlations is critical in the real world. RNN can be better at solving this problem, as compared to other moving average (MA) methods that have been frequently adopted before. A detailed structure of RNN for a sequence over time is shown in Part B of the Appendix (see Fig. 7 in Appendix ).

The abstract structure of RNN

Although RNN can resolve the issue of time-series order, the issue of long-term dependencies remains. It is difficult to find the optimal weight for long-term data. LSTM, as a type of RNN, added a gated cell to overcome long-term dependencies by combining different activation functions (e.g., sigmoid or tanh). Given that LSTM is frequently used for forecasting in the finance literature, we extract LSTM from RNN models and name other structures of standard RNN as RNN(O).

As we focus on the application rather than theoretical DL aspect, this study will not consider other popular DL algorithms, including CNN and RL, as well as Latent variable models such as variational autoencoders and generative adversarial network. Table 6 in Appendix shows a legend note to explain the abbreviations used in this paper. We summarize the relationship between commonly used DL models in Fig.  3 .

Relationships of reviewed DL models for F&B domains

Research framework and methodology

Our research framework is illustrated in Fig.  4 . We combine qualitative and quantitative analyses of the articles in this study. Based on our review, we recognize and identify seven core F&B domains, as shown in Fig.  5 . To connect the DL side and the F&B side, we present our review on the application of the DL model in seven F&B domains in Section 4. It is crucial to analyze the feasibility of a DL model toward particular domains. To do so, we provide summarizations in three key aspects, including data preprocessing, data inputs, and evaluation rules, according to our collection of articles. Finally, we determine optimal DL models regarding the identified domains. We further discuss two common issues in using DL models for F&B: overfitting and sustainability.

The research framework of this study

The identified domains of F&B for DL applications

Figure  5 shows that the application domains can be divided into two major areas: (1) banking and credit risk and (2) financial market investment. The former contains two domains: credit risk prediction and macroeconomic prediction. The latter contains financial prediction, trading, and portfolio management. Prediction tasks are crucial, as emphasized by Cavalcante et al. ( 2016 ). We study this domain from three aspects of prediction, including exchange rate, stock market, and oil price. We illustrate this structure of application domains in F&B.

Figure  6 shows a statistic in the listed F&B domains. We illustrate the domains of financial applications on the X-axis and count the number of articles on the Y-axis. Note that a reviewed article could cover more than one domain in this figure; thus, the sum of the counts (45) is larger than the size of our review pool (40 articles). As shown in Fig.  6 , stock marketing prediction and trading dominate the listed domains, followed by exchange rate prediction. Moreover, we found two articles on banking credit risk and two articles on portfolio management. Price prediction and macroeconomic prediction are two potential topics that deserve more studies.

A count of articles over seven identified F&B domains

Application of DL models in F&B domains

Based on our review, six types of DL models are reported. They are FNN, CNN, RNN, RL, deep belief networks (DBN), and restricted Boltzmann machine (RBM). Regarding FNN, several papers use the alternative terms of backpropagation artificial neural network (ANN), FNN, MLP, and DNN. They have an identical structure. Regarding RNN, one of its well-known models in the time-series analysis is called LSTM. Nearly half of the reviewed articles apply FNN as the primary DL technique. Nine articles apply LSTM, followed by eight articles for RL, and six articles for RNN. Minor ones that are applied in F&B include CNN, DBM, and RBM. We count the number of articles that use various DL models in seven F&B domains, as shown in Table  1 . FNN is the principal model used in exchange rate, price, and macroeconomic predictions, as well as banking default risk and credit. LSTM and FNN are two kinds of popular models for stock market prediction. Differently, RL and FNN are frequently used regarding stock trading. FNN, RL, and simple RNN can be conducted in portfolio management. FNN is the primary model in macroeconomic and banking risk prediction. CNN, LSTM, and RL are emerging research approaches in banking risk prediction. The detailed statistics that contain specific articles can be found in Table 5 in Appendix .

Exchange rate prediction

Shen et al. ( 2015 ) construct an improved DBN model by including RBM and find that their model outperforms the random walk algorithm, auto-regressive-moving-average (ARMA), and FNN with fewer errors. Zheng et al. ( 2017 ) examine the performance of DBN and find that the DBN model estimates the exchange rate better than FNN model does. They find that a small number of layer nodes engender a more significant effect on DBN.

Several scholars believe that a hybrid model should have better performance. Ravi et al. ( 2017 ) contribute a hybrid model by using MLP (FNN), chaos theory, and multi-objective evolutionary algorithms. Their Chaos+MLP + NSGA-II model Footnote 1 has a mean squared error (MSE) with 2.16E-08 that is very low. Several articles point out that only a complicated neural network like CNN can gain higher accuracy. For example, Galeshchuk and Mukherjee ( 2017 ) conduct experiments and claim that a single hidden layer NN or SVM performs worse than a simple model like moving average (MA). However, they find that CNN could achieve higher classification accuracy in predicting the direction of the change of exchange rate because of successive layers of DNN.

Stock market prediction

In stock market prediction, some studies suggest that market news may influence the stock price and DL model, such as using a magic filter to extract useful information for price prediction. Matsubara et al. ( 2018 ) extract information from the news and propose a deep neural generative model to predict the movement of the stock price. This model combines DNN and a generative model. It suggests that this hybrid approach outperforms SVM and MLP.

Minh et al. ( 2017 ) develop a novel framework with two streams combining the gated recurrent unit network and the Stock2vec. It employs a word embedding and sentiment training system on financial news and the Harvard IV-4 dataset. They use the historical price and news-based signals from the model to predict the S&P500 and VN-index price directions. Their model shows that the two-stream gated recurrent unit is better than the gated recurrent unit or the LSTM. Jiang et al. ( 2018 ) establish a recurrent NN that extracts the interaction between the inner-domain and cross-domain of financial information. They prove that their model outperforms the simple RNN and MLP in the currency and stock market. Krausa and Feuerriegel ( 2017 ) propose that they can transform financial disclosure into a decision through the DL model. After training and testing, they point out that LSTM works better than the RNN and conventional ML methods such as ridge regression, Lasso, elastic net, random forest, SVR, AdaBoost, and gradient boosting. They further pre-train words embeddings with transfer learning (Krausa and Feuerriegel 2017 ). They conclude that better performance comes from LSTM with word embeddings. In the sentiment analysis, Sohangir et al. ( 2018 ) compares LSTM, doc2vec, and CNN to evaluate the stock opinions on the StockTwits. They conclude that CNN is the optimal model to predict the sentiment of authors. This result may be further applied to predict the stock market trend.

Data preprocessing is conducted to input data into the NN. Researchers may apply numeric unsupervised methods of feature extraction, including principal component analysis, autoencoder, RBM, and kNN. These methods can reduce the computational complexity and prevent overfitting. After the input of high-frequency transaction data, Chen et al. ( 2018b ) establish a DL model with an autoencoder and an RBM. They compare their model with backpropagation FNN, extreme learning machine, and radial basis FNN. They claim that their model can better predict the Chinese stock market. Chong et al. ( 2017 ) apply the principal component analysis (PCA) and RBM with high-frequency data of the South Korean market. They find that their model can explain the residual of the autoregressive model. The DL model can thus extract additional information and improve prediction performance. More so, Singh and Srivastava ( 2017 ) describe a model involving 2-directional and 2-dimensional (2D 2 ) PCA and DNN. Their model outperforms 2D 2 with radial basis FNN and RNN.

For time-series data, sometimes it is difficult to judge the weight of long-term and short-term data. The LSTM model is just for resolving this problem in financial prediction. The literature has attempted to prove that LSTM models are applicable and outperform conventional FNN models. Yan and Ouyang ( 2017 ) apply LSTM to challenge the MLP, SVM, and kNN in predicting a static and dynamic trend. After a wavelet decomposition and a reconstruction of the financial time series, their model can be used to predict a long-term dynamic trend. Baek and Kim ( 2018 ) apply LSTM not only in predicting the price of S&P500 and KOSPI200 but also in preventing overfitting. Kim and Won ( 2018 ) apply LSTM in the prediction of stock price volatility. They propose a hybrid model that combines LSTM with three generalized autoregressive conditional heteroscedasticity (GARCH)-type models. Hernandez and Abad ( 2018 ) argue that RBM is inappropriate for dynamic data modeling in the time-series analysis because it cannot retain memory. They apply a modified RBM model called p -RBM that can retain the memory of p past states. This model is used in predicting market directions of the NASDAQ-100 index. Compared with vector autoregression (VAR) and LSTM, notwithstanding, they find that LSTM is better because it can uncover the hidden structure within the non-linear data while VAR and p -RBM cannot capture the non-linearity in data.

CNN was established to predict the price with a complicated structure. Making the best use of historical price, Dingli and Fournier ( 2017 ) develop a new CNN model. This model can predict next month’s price. Their results cannot surpass other comparable models, such as logistic regression (LR) and SVM. Tadaaki ( 2018 ) applies the financial ratio and converts them into a “grayscale image” in the CNN model. The results reveal that CNN is more efficient than decision trees (DT), SVM, linear discriminant analysis, MLP, and AdaBoost. To predict the stock direction, Gunduz et al. ( 2017 ) establish a CNN model with a so-called specially ordered feature set whose classifier outperforms either CNN or LR.

Stock trading

Many studies adopt the conventional FNN model and try to set up a profitable trading system. Sezer et al. ( 2017 ) combine GA with MLP. Chen et al. ( 2017 ) adopt a double-layer NN and discover that its accuracy is better than ARMA-GARCH and single-layer NN. Hsu et al. ( 2018 ) equip the Black-Scholes model and a three-layer fully-connected feedforward network to estimate the bid-ask spread of option price. They argue that this novel model is better than the conventional Black-Scholes model with lower RMSE. Krauss et al. ( 2017 ) apply DNN, gradient-boosted-trees, and random forests in statistical arbitrage. They argue that their returns outperform the market index S&P500.

Several studies report that RNN and its derivate models are potential. Deng et al. ( 2017 ) extend the fuzzy learning into the RNN model. After comparing their model to different DL models like CNN, RNN, and LSTM, they claim that their model is the optimal one. Fischer and Krauss ( 2017 ) and Bao et al. ( 2017 ) argue that LSTM can create an optimal trading system. Fischer and Krauss ( 2017 ) claim that their model has a daily return of 0.46 and a sharp ratio of 5.8 prior to the transaction cost. Given the transaction cost, however, LSTM’s profitability fluctuated around zero after 2010. Bao et al. ( 2017 ) advance Fischer and Krauss’s ( 2017 ) work and propose a novel DL model (i.e., WSAEs-LSTM model). It uses wavelet transforms to eliminate noise, stacked autoencoders (SAEs) to predict stock price, and LSTM to predict the close price. The result shows that their model outperforms other models such as WLSTM, Footnote 2 LSTM, and RNN in predictive accuracy and profitability.

RL is popular recently despite its complexity. We find that five studies apply this model. Chen et al. ( 2018a ) propose an agent-based RL system to mimic 80% professional trading strategies. Feuerriegel and Prendinger ( 2016 ) convert the news sentiment into the signal in the trading system, although their daily returns and abnormal returns are nearly zero. Chakraborty ( 2019 ) cast the general financial market fluctuation into a stochastic control problem and explore the power of two RL models, including Q-learning Footnote 3 and state-action-reward-state-action (SARSA) algorithm. Both models can enhance profitability (e.g., 9.76% for Q-learning and 8.52% for SARSA). They outperform the buy-and-hold strategy. Footnote 4 Zhang and Maringer ( 2015 ) conduct a hybrid model called GA, with recurrent RL. GA is used to select an optimal combination of technical indicators, fundamental indicators, and volatility indicators. The out-of-sample trading performance is improved due to a significantly positive Sharpe ratio. Martinez-Miranda et al. ( 2016 ) create a new topic of trading. It uses a market manipulation scanner model rather than a trading system. They use RL to model spoofing-and-pinging trading. This study reveals that their model just works on the bull market. Jeong and Kim ( 2018 ) propose a model called deep Q-network that is constructed by RL, DNN, and transfer learning. They use transfer learning to solve the overfitting issue incurred as a result of insufficient data. They argue that the profit yields in this system increase by four times the amount in S&P500, five times in KOSPI, six times in EuroStoxx50, and 12 times in HIS.

Banking default risk and credit

Most articles in this domain focus on FNN applications. Rönnqvist and Sarlin ( 2017 ) propose a model for detecting relevant discussions in texting and extracting natural language descriptions of events. They convert the news into a signal of the bank-distress report. In their back-test, their model reflects the distressing financial event of the 2007–2008 period.

Zhu et al. ( 2018 ) propose a hybrid CNN model with a feature selection algorithm. Their model outperforms LR and random forest in consumer credit scoring. Wang et al. ( 2019 ) consider that online operation data can be used to predict consumer credit scores. They thus convert each kind of event into a word and apply the Event2vec model to transform the word into a vector in the LSTM network. The probability of default yields higher accuracy than other models. Jurgovsky et al. ( 2018 ) employs the LSTM to detect credit card fraud and find that LSTM can enhance detection accuracy.

Han et al. ( 2018 ) report a method that adopts RL to assess the credit risk. They claim that high-dimensional partial differential equations (PDEs) can be reformulated by using backward stochastic differential equations. NN approximates the gradient of the unknown solution. This model can be applied to F&B risk evaluation after considering all elements such as participating agents, assets, and resources, simultaneously.

Portfolio management

Song et al. ( 2017 ) establish a model after combining ListNet and RankNet to make a portfolio. They take a long position for the top 25% stocks and hold the short position for the bottom 25% stocks weekly. The ListNetlong-short model is the optimal one, which can achieve a return of 9.56%. Almahdi and Yang ( 2017 ) establish a better portfolio with a combination of RNN and RL. The result shows that the proposed trading system respond to transaction cost effects efficiently and outperform hedge fund benchmarks consistently.

Macroeconomic prediction

Sevim et al. ( 2014 ) develops a model with a back-propagation learning algorithm to predict the financial crises up to a year before it happened. This model contains three-layer perceptrons (i.e., MLP) and can achieve an accuracy rate of approximately 95%, which is superior to DT and LR. Chatzis et al. ( 2018 ) examine multiple models such as classification tree, SVM, random forests, DNN, and extreme gradient boosting to predict the market crisis. The results show that crises encourage persistence. Furthermore, using DNN increases the classification accuracy that makes global warning systems more efficient.

Price prediction

For price prediction, Sehgal and Pandey ( 2015 ) review ANN, SVM, wavelet, GA, and hybrid systems. They separate the time-series models into stochastic models, AI-based models, and regression models to predict oil prices. They reveal that researchers prevalently use MLP for price prediction.

Data preprocessing and data input

Data preprocessing.

Data preprocessing is conducted to denoise before data training of DL. This section summarizes the methods of data preprocessing. Multiple preprocessing techniques discussed in Part 4 include the principal component analysis (Chong et al. 2017 ), SVM (Gunduz et al. 2017 ), autoencoder, and RBM (Chen et al. 2018b ). There are several additional techniques of feature selection as follows.

Relief: The relief algorithm (Zhu et al. 2018 ) is a simple approach to weigh the importance of the feature. Based on NN algorithms, relief repeats the process for n times and divides each final weight vector by n . Thus, the weight vectors are the relevance vectors, and features are selected if their relevance is larger than the threshold τ .

Wavelet transforms: Wavelet transforms are used to fix the noise feature of the financial time series before feeding into a DL network. It is a widely used technique for filtering and mining single-dimensional signals (Bao et al. 2017 ).

Chi-square: Chi-square selection is commonly used in ML to measure the dependence between a feature and a class label. The representative usage is by Gunduz et al. ( 2017 ).

Random forest: Random forest algorithm is a two-stage process that contains random feature selection and bagging. The representative usage is by Fischer and Krauss ( 2017 ).

Data inputs

Data inputs are an important criterion for judging whether a DL model is feasible for particular F&B domains. This section summarizes the method of data inputs that have been adopted in the literature. Based on our review, five types of input data in the F&B domain can be presented. Table  2 provides a detailed summary of the input variable in F&B domains.

History price: The daily exchange rate can be considered as history price. The price can be the high, low, open, and close price of the stock. Related articles include Bao et al. ( 2017 ), Chen et al. ( 2017 ), Singh and Srivastava ( 2017 ), and Yan and Ouyang ( 2017 ).

Technical index: Technical indexes include MA, exponential MA, MA convergence divergence, and relative strength index. Related articles include Bao et al. ( 2017 ), Chen et al. ( 2017 ), Gunduz et al. ( 2017 ), Sezer et al. ( 2017 ), Singh and Srivastava ( 2017 ), and Yan and Ouyang ( 2017 ).

Financial news: Financial news covers financial message, sentiment shock score, and sentiment trend score. Related articles include Feuerriegel and Prendinger ( 2016 ), Krausa and Feuerriegel ( 2017 ), Minh et al. ( 2017 ), and Song et al. ( 2017 ).

Financial report data: Financial report data can account for items in the financial balance sheet or the financial report data (e.g., return on equity, return on assets, price to earnings ratio, and debt to equity ratio). Zhang and Maringer ( 2015 ) is a representative study on the subject.

Macroeconomic data: This kind of data includes macroeconomic variables. It may affect elements of the financial market, such as exchange rate, interest rate, overnight interest rate, and gross foreign exchange reserves of the central bank. Representative articles include Bao et al. ( 2017 ), Kim and Won ( 2018 ), and Sevim et al. ( 2014 ).

Stochastic data: Chakraborty ( 2019 ) provides a representative implementation.

Evaluation rules

It is critical to judge whether an adopted DL model works well in a particular financial domain. We, thus, need to consider evaluation systems of criteria for gauging the performance of a DL model. This section summarizes the evaluation rules of F&B-oriented DL models. Based on our review, three evaluation rules dominate: the error term, the accuracy index, and the financial index. Table  3 provides a detailed summary. The evaluation rules can be boiled down to the following categories.

Error term: Suppose Y t  +  i and F t  +  i are the real data and the prediction data, respectively, where m is the total number. The following is a summary of the functional formula commonly employed for evaluating DL models.

Mean Absolute Error (MAE): \( {\sum}_{i=1}^m\frac{\left|{Y}_{t+i}-{F}_{t+i}\right|}{m} \) ;

Mean Absolute Percent Error (MAPE): \( \frac{100}{m}{\sum}_{i=1}^m\frac{\left|{Y}_{t+i}-{F}_{t+i}\right|}{Y_{t+i}} \) ;

Mean Squared Error (MSE): \( {\sum}_{i=1}^m\frac{{\left({Y}_{t+i}-{F}_{t+i}\right)}^2}{m} \) ;

Root Mean Squared Error (RMSE): \( \sqrt{\sum_{i=1}^m\frac{{\left({Y}_{t+i}-{F}_{t+i}\right)}^2}{m}} \) ;

Normalized Mean Square Error (NMSE): \( \frac{1}{m}\frac{\sum {\left({Y}_{t+i}-{F}_{t+i}\right)}^2}{\mathit{\operatorname{var}}\left({Y}_{t+i}\right)} \) .

Accuracy index: According to Matsubara et al. ( 2018 ), we use TP, TN, FP, and FN to represent the number of true positives, true negatives, false positives, and false negatives, respectively, in a confusion matrix for classification evaluation. Based on our review, we summarize the accuracy indexes as follows.

Directional Predictive Accuracy (DPA): \( \frac{1}{N}{\sum}_{t=1}^N{D}_t \) , if ( Y t  + 1  −  Y t ) × ( F t  + 1  −  Y t ) ≥ 0, D t  = 1, otherwise, D t  = 0;

Actual Correlation Coefficient (ACC): \( \frac{TP+ TN}{TP+ FP+ FN+ TN} \) ;

Matthews Correlation Coefficient (MCC): \( \frac{TP\times TN- FP\times FN}{\sqrt{\left( TP+ FP\right)\left( TP+ FN\right)\left( TN+ FP\right)\left( TN+ FN\right)}} \) .

Financial index: Financial indexes involve total return, Sharp ratio, abnormal return, annualized return, annualized number of transaction, percentage of success, average profit percent per transaction, average transaction length, maximum profit percentage in the transaction, maximum loss percentage in the transaction, maximum capital, and minimum capital.

For the prediction by regressing the numeric dependent variables (e.g., exchange rate prediction or stock market prediction), evaluation rules are mostly error terms. For the prediction by classification in the category data (e.g., direction prediction on oil price), the accuracy indexes are widely conducted. For stock trading and portfolio management, financial indexes are the final evaluation rules.

General comparisons of DL models

This study identifies the most efficient DL model in each identified F&B domain. Table  4 illustrates our comparisons of the error terms in the pool of reviewed articles. Note that “A > B” means that the performance of model A is better than that of model B. “A + B” indicates the hybridization of multiple DL models.

At this point, we have summarized three methods of data processing in DL models against seven specified F&B domains, including data preprocessing, data inputs, and evaluation rules. Apart from the technical level of DL, we find the following:

NN has advantages in handling cross-sectional data;

RNN and LSTM are more feasible in handling time series data;

CNN has advantages in handling the data with multicollinearity.

Apart from application domains, we can induce the following viewpoints. Cross-sectional data usually appear in exchange rate prediction, price prediction, and macroeconomic prediction, for which NN could be the most feasible model. Time series data usually appear in stock market prediction, for which LSTM and RNN are the best options. Regarding stock trading, a feasible DL model requires the capabilities of decision and self-learning, for which RL can be the best. Moreover, CNN is more suitable for the multivariable environment of any F&B domains. As shown in the statistics of the Appendix , the frequency of using corresponding DL models corresponds to our analysis above. Selecting proper DL models according to the particular needs of financial analysis is usually challenging and crucial. This study provides several recommendations.

We summarize emerging DL models in F&B domains. Nevertheless, can these models refuse the efficient market hypothesis (EMH)? Footnote 5 According to the EMH, the financial market has its own discipline. There is no long-term technical tool that could outperform an efficient market. If so, using DL models may not be practical in long-term trading as it requires further experimental tests. However, why do most of the reviewed articles argue that their DL models of trading outperform the market returns? This argument has challenged the EMH. A possible explanation is that many DL algorithms are still challenging to apply in the real-world market. The DL models may raise trading opportunities to gain abnormal returns in the short-term. In the long run, however, many algorithms may lose their superiority, whereas EMH still works as more traders recognize the arbitrage gap offered by these DL models.

This section discusses three aspects that could affect the outcomes of DL models in finance.

Training and validation of data processing

The size of the training set.

The optimal way to improve the performance of models is by enhancing the size of the training data. Bootstrap can be used for data resampling, and generative adversarial network (GAN) can extend the data features. However, both can recognize numerical parts of features. Sometimes, the sample set is not diverse enough; thus, it loses its representativeness. Expanding the data size could make the model more unstable. The current literature reported diversified sizes of training sets. The requirements of data size in the training stage could vary by different F&B tasks.

The number of input factors

Input variables are independent variables. Based on our review, multi-factor models normally perform better than single-factor models in the case that the additional input factors are effective. In the time-series data model, long-term data have less prediction errors than that for a short period. The number of input factors depends on the employment of the DL structure and the specific environment of F&B tasks.

The quality of data

Several methods can be used to improve the data quality, including data cleaning (e.g., dealing with missing data), data normalization (e.g., taking the logarithm, calculating the changes of variables, and calculating the t -value of variables), feature selection (e.g., Chi-square test), and dimensionality reduction (e.g., PCA). Financial DL models require that the input variables should be interpretable in economics. When inputting the data, researchers should clarify the effective variables and noise. Several financial features, such as technical indexes, are likely to be created and added into the model.

Selection on structures of DL models

DL model selection should depend on problem domains and cases in finance. NN is suitable for processing cross-sectional data. LSTM and other RNNs are optimal choices for time-series data in prediction tasks. CNN can settle the multicollinearity issue through data compression. Latent variable models like GAN can be better for dimension reduction and clustering. RL is applicable in the cases with judgments like portfolio management and trading. The return levels and outcomes on RL can be affected significantly by environment (observation) definitions, situation probability transfer matrix, and actions.

The setting of objective functions and the convexity of evaluation rules

Objective function selection affects training processes and expected outcomes. For predictions on stock price, low MAE merely reflects the effectiveness of applied models in training; however, it may fail in predicting future directions. Therefore, it is vital for additional evaluation rules for F&B. Moreover, it can be more convenient to resolve the objective functions if they are convex.

The influence of overfitting (underfitting)

Overfitting (underfitting) commonly happens in using DL models, which is clearly unfavorable. A generated model performs perfectly in one case but usually cannot replicate good performance with the same model and identical coefficients. To solve this problem, we have to trade off the bias against variances. Bias posits that researchers prefer to keep it small to illustrate the superiority of their models. Generally, a deeper (i.e., more layered) NN model or neurons can reduce errors. However, it is more time-consuming and could reduce the feasibility of applied DL models.

One solution is to establish validation sets and testing sets for deciding the numbers of layers and neurons. After setting optimal coefficients in the validation set (Chong et al. 2017 ; Sevim et al. 2014 ), the result in the testing sets reveals the level of errors that could mitigate the effect of overfitting. One can input more samples of financial data to check the stability of the model’s performance. This method is known as the early stopping. It stops training more layers in the network once the testing result has achieved an optimal level.

Moreover, regularization is another approach to conquer the overfitting. Chong et al. ( 2017 ) introduces a constant term for the objective function and eventually reduces the variates of the result. Dropout is also a simple method to address overfitting. It reduces the dimensions and layers of the network (Minh et al. 2017 ; Wang et al. 2019 ). Finally, the data cleaning process (Baek and Kim 2018 ; Bao et al. 2017 ), to an extent, could mitigate the impact of overfitting.

Financial models

The sustainability of the model.

According to our reviews, the literature focus on evaluating the performance of historical data. However, crucial problems remain. Given that prediction is always complicated, the problem of how to justify the robustness of the used DL models in the future remains. More so, whether a DL model could survive in dynamic environments must be considered.

The following solutions could be considered. First, one can divide the data into two groups according to the time range; performance can subsequently be checked (e.g., using the data for the first 3 years to predict the performance of the fourth year). Second, the feature selection can be used in the data preprocessing, which could improve the sustainability of models in the long run. Third, stochastic data can be generated for each input variable by fixing them with a confidence interval, after which a simulation to examine the robustness of all possible future situations is conducted.

The popularity of the model

Whether a DL model is effective for trading is subject to the popularity of the model in the financial market. If traders in the same market conduct an identical model with limited information, they may run identical results and adopt the same trading strategy accordingly. Thus, they may lose money because their strategy could sell at a lower price after buying at a higher.

Conclusion and future works

Concluding remarks.

This paper provides a comprehensive survey of the literature on the application of DL in F&B. We carefully review 40 articles refined from a collection of 150 articles published between 2014 and 2018. The review and refinement are based on a scientific selection of academic databases. This paper first recognizes seven core F&B domains and establish the relationships between the domains and their frequently-used DL models. We review the details of each article under our framework. Importantly, we analyze the optimal models toward particular domains and make recommendations according to the feasibility of various DL models. Thus, we summarize three important aspects, including data preprocessing, data inputs, and evaluation rules. We further analyze the unfavorable impacts of overfitting and sustainability when applying DL models and provide several possible solutions. This study contributes to the literature by presenting a valuable accumulation of knowledge on related studies and providing useful recommendations for financial analysts and researchers.

Future works

Future studies can be conducted from the DL technical and F&B application perspectives. Regarding the perspective of DL techniques, training DL model for F&B is usually time-consuming. However, effective training could greatly enhance accuracy by reducing errors. Most of the functions can be simulated with considerable weights in complicated networks. First, one of the future works should focus on data preprocessing, such as data cleaning, to reduce the negative effect of data noise in the subsequent stage of data training. Second, further studies on how to construct layers of networks in the DL model are required, particularly when considering a reduction of the unfavorable effects of overfitting and underfitting. According to our review, the comparisons between the discussed DL models do not hinge on an identical source of input data, which renders these comparisons useless. Third, more testing regarding F&B-oriented DL models would be beneficial.

In addition to the penetration of DL techniques in F&B fields, more structures of DL models should be explored. From the perspective of F&B applications, the following problems need further research to investigate desirable solutions. In the case of financial planning, can a DL algorithm transfer asset recommendations to clients according to risk preferences? In the case of corporate finance, how can a DL algorithm benefit capital structure management and, thus, maximize the values of corporations? How can managers utilize DL technical tools to gauge the investment environment and financial data? How can they use such tools to optimize cash balances and cash inflow and outflow? Until recently, DL models like RL and generative adversarial networks are rarely used. More investigations on constructing DL structures for F&B regarding preferences would be beneficial. Finally, the developments of professional F&B software and system platforms that implement DL techniques are highly desirable.

Availability of data and materials

Not applicable.

In the model, NSGA stands for non-dominated sorting genetic algorithm.

A combination of Wavelet transforms (WT) and long-short term memory (LSTM) is called WLSTM in Bao et al. ( 2017 ).

Q-learning is a model-free reinforcement learning algorithm.

Buy-and-hold is a passive investment strategy in which an investor buys stocks (or ETFs) and holds them for a long period regardless of fluctuations in the market.

EMH was developed from a Ph.D. dissertation by economist Eugene Fama in the 1960s. It says that at any given time, stock prices reflect all available information and trade at exactly their fair value at all times. It is impossible to consistently choose stocks that will beat the returns of the overall stock market. Therefore, this hypothesis implies that the pursuit of market-beating performance is more about chance than it is about researching and selecting the right stocks.

Almahdi, S., & Yang, S. Y. (2017). An adaptive portfolio trading system: A risk-return portfolio optimization using recurrent reinforcement learning with expected maximum drawdown. Expert Systems with Applications, 87 , 267–279.

Article   Google Scholar  

Baek, Y., & Kim, H. Y. (2018). ModAugNet: A new forecasting framework for stock market index value with an overfitting prevention LSTM module and a prediction LSTM module. Expert Systems with Applications, 113 , 457–480.

Bao, W., Yue, J., & Rao, Y. (2017). A deep learning framework for financial time series using stacked autoencoders and long-short-term memory. PLoS One, 12 (7), e0180944.

Butaru, F., Chen, Q., Clark, B., Das, S., Lo, A. W., & Siddique, A. (2016). Risk and risk management in the credit card industry. Journal of Banking & Finance, 72 , 218–239.

Cavalcante, R. C., Brasileiro, R. C., Souza, V. L. F., Nobrega, J. P., & Oliveira, A. L. I. (2016). Computational intelligence and financial markets: A survey and future directions. Expert System with Application, 55 , 194–211.

Chai, J. Y., & Li, A. M. (2019). Deep learning in natural language processing: A state-of-the-art survey. In The proceeding of the 2019 international conference on machine learning and cybernetics (pp. 535–540). Japan: Kobe.

Google Scholar  

Chai, J. Y., Liu, J. N. K., & Ngai, E. W. T. (2013). Application of decision-making techniques in supplier selection: A systematic review of literature. Expert Systems with Applications, 40 (10), 3872–3885.

Chai, J. Y., & Ngai, E. W. T. (2020). Decision-making techniques in supplier selection: Recent accomplishments and what lies ahead. Expert Systems with Applications, 140 , 112903. https://doi.org/10.1016/j.eswa.2019.112903 .

Chakraborty, S. (2019). Deep reinforcement learning in financial markets Retrieved from https://arxiv.org/pdf/1907.04373.pdf . Accessed 04 Apr 2020.

Chatzis, S. P., Siakoulis, V., Petropoulos, A., Stavroulakis, E., & Vlachogiannakis, E. (2018). Forecasting stock market crisis events using deep and statistical machine learning techniques. Expert Systems with Applications, 112 , 353–371.

Chen, C. T., Chen, A. P., & Huang, S. H. (2018a). Cloning strategies from trading records using agent-based reinforcement learning algorithm. In The proceeding of IEEE international conference on agents (pp. 34–37).

Chen, H., Xiao, K., Sun, J., & Wu, S. (2017). A double-layer neural network framework for high-frequency forecasting. ACM Transactions on Management Information Systems, 7 (4), 11.

Chen, L., Qiao, Z., Wang, M., Wang, C., Du, R., & Stanley, H. E. (2018b). Which artificial intelligence algorithm better predicts the Chinese stock market? IEEE Access, 6 , 48625–48633.

Chong, E., Han, C., & Park, F. C. (2017). Deep learning networks for stock market analysis and prediction: Methodology, data representations, and case studies. Expert Systems with Applications, 83 , 187–205.

Collobert, R., Weston, J., Bottou, L., Karlen, M., Kavukcuoglu, K., & Kuksa, P. (2011). Natural language processing (almost) from scratch. Journal of Machine Learning Research, 12 , 2493–2537.

Deng, Y., Bao, F., Kong, Y., Ren, Z., & Dai, Q. (2017). Deep direct reinforcement learning for financial signal representation and trading. IEEE Transactions on Neural Networks and Learning Systems, 28 (3), 653–664.

Dingli, A., & Fournier, K. S. (2017). Financial time series forecasting—A machine learning approach. International Journal of Machine Learning and Computing, 4 , 11–27.

Elad, M., & Aharon, M. (2006). Image denoising via sparse and redundant representations over learned dictionaries. IEEE Transactions on Image Processing, 15 (12), 3736–3745.

Feuerriegel, S., & Prendinger, H. (2016). News-based trading strategies. Decision Support Systems, 90 , 65–74.

Fischer, T., & Krauss, C. (2017). Deep learning with long short-term memory networks for financial market predictions. European Journal of Operational Research, 270 (2), 654–669.

Galeshchuk, S., & Mukherjee, S. (2017). Deep networks for predicting the direction of change in foreign exchange rates. Intelligent Systems in Accounting, Finance and Maangement, 24 (4), 100–110.

Gunduz, H., Yaslan, Y., & Cataltepe, Z. (2017). Intraday prediction of Borsa Istanbul using convolutional neural networks and feature correlations. Knowledge-Based Systems, 137 , 138–148.

Guo, Y., Liu, Y., Oerlemans, A., Lao, S., Wu, S., & Lew, M. S. (2016). Deep learning for visual understanding: A review. Neurocomputing, 187 , 27–48.

Han, J., Jentzen, A., & Weinan, E. (2018). Solving high-dimensional partial differential equations using deep learning. The proceedings of the National Academy of Sciences of the United States of America (PNAS) ; 8505–10).

Hernandez, J., & Abad, A. G. (2018). Learning from multivariate discrete sequential data using a restricted Boltzmann machine model. In The proceeding of IEEE 1st Colombian conference on applications in computational intelligence (ColCACI) (pp. 1–6).

Hsu, P. Y., Chou, C., Huang, S. H., & Chen, A. P. (2018). A market making quotation strategy based on dual deep learning agents for option pricing and bid-ask spread estimation.   The proceeding of IEEE international conference on agents (pp. 99–104).

Jeong, G., & Kim, H. Y. (2018). Improving financial trading decisions using deep Q-learning: Predicting the number of shares, action strategies and transfer learning. Expert Systems with Applications, 117 , 125–138.

Jiang, X., Pan, S., Jiang, J., & Long, G. (2018). Cross-domain deep learning approach for multiple financial market predictions. The proceeding of international joint conference on neural networks (pp. 1–8).

Jurgovsky, J., Granitzer, M., Ziegler, K., Calabretto, S., Portier, P. E., Guelton, L. H., & Caelen, O. (2018). Sequence classification for credit-card fraud detection. Expert Systems with Applications, 100 , 234–245.

Kim, H. Y., & Won, C. H. (2018). Forecasting the volatility of stock price index: A hybrid model integrating LSTM with multiple GARCH-type models. Expert Systems with Applications, 103 , 25–37.

Krausa, M., & Feuerriegel, S. (2017). Decision support from financial disclosures with deep neural networks and transfer learning Retrieved from https://arxiv.org/pdf/1710.03954.pdf Accessed 04 Apr 2020.

Book   Google Scholar  

Krauss, C., Do, X. A., & Huck, N. (2017). Deep neural networks, gradient-boosted trees, random forests: Statistical arbitrage on the S&P500. European Journal of Operational Research, 259 (2), 689–702.

Martinez-Miranda, E., McBurney, P., & Howard, M. J. W. (2016). Learning unfair trading: A market manipulation analysis from the reinforcement learning perspective. In The proceeding of 2016 IEEE conference on evolving and adaptive intelligent systems (EAIS) (pp. 103–109).

Chapter   Google Scholar  

Matsubara, T., Akita, R., & Uehara, K. (2018). Stock price prediction by deep neural generative model of news articles. IEICE Transactions on Information and Systems, 4 , 901–908.

Minh, D. L., Sadeghi-Niaraki, A., Huy, H. D., Min, K., & Moon, H. (2017). Deep learning approach for short-term stock trends prediction based on two-stream gated recurrent unit network. IEEE Access, 6 , 55392–55404.

Ravi, V., Pradeepkumar, D., & Deb, K. (2017). Financial time series prediction using hybrids of chaos theory, multi-layer perceptron and multi-objective evolutionary algorithms. Swarm and Evolutionary Computation, 36 , 136–149.

Rönnqvist, S., & Sarlin, P. (2017). Bank distress in the news describing events through deep learning. Neurocomputing, 264 (15), 57–70.

Sehgal, N., & Pandey, K. K. (2015). Artificial intelligence methods for oil price forecasting: A review and evaluation. Energy System, 6 , 479–506.

Sevim, C., Oztekin, A., Bali, O., Gumus, S., & Guresen, E. (2014). Developing an early warning system to predict currency crises. European Journal of Operational Research, 237 (3), 1095–1104.

Sezer, O. B., Ozbayoglu, M., & Gogdu, E. (2017). A deep neural-network-based stock trading system based on evolutionary optimized technical analysis parameters. Procedia Computer Science, 114 , 473–480.

Shen, F., Chao, J., & Zhao, J. (2015). Forecasting exchange rate using deep belief networks and conjugate gradient method. Neurocomputing, 167 , 243–253.

Singh, R., & Srivastava, S. (2017). Stock prediction using deep learning. Multimedia Tools Application, 76 , 18569–18584.

Sohangir, S., Wang, D., Pomeranets, A., & Khoshgoftaar, T. M. (2018). Big data: Deep learning for financial sentiment analysis. Journal of Big Data, 5 (3), 1–25.

Song, Q., Liu, A., & Yang, S. Y. (2017). Stock portfolio selection using learning-to-rank algorithms with news sentiment. Neurocomputing, 264 , 20–28.

Tadaaki, H. (2018). Bankruptcy prediction using imaged financial ratios and convolutional neural networks. Expert Systems with Applications, 117 , 287–299.

Wang, C., Han, D., Liu, Q., & Luo, S. (2019). A deep learning approach for credit scoring of peer-to-peer lending using attention mechanism LSTM. IEEE Access, 7 , 2161–2167.

Yan, H., & Ouyang, H. (2017). Financial time series prediction based on deep learning. Wireless Personal Communications, 102 , 683–700.

Zhang, J., & Maringer, D. (2015). Using a genetic algorithm to improve recurrent reinforcement learning for equity trading. Computational Economics, 47 , 551–567.

Zheng, J., Fu, X., & Zhang, G. (2017). Research on exchange rate forecasting based on a deep belief network. Neural Computing and Application, 31 , 573–582.

Zhu, B., Yang, W., Wang, H., & Yuan, Y. (2018). A hybrid deep learning model for consumer credit scoring. In The proceeding of international conference on artificial intelligence and big data (pp. 205–208).

Download references

Acknowledgments

The constructive comments of the editor and three anonymous reviewers on an earlier version of this paper are greatly appreciated. The authors are indebted to seminar participants at 2019 China Accounting and Financial Innovation Form at Zhuhai for insightful discussions. The corresponding author thanks the financial supports from BNU-HKBU United International College Research Grant under Grant R202026.

BNU-HKBU United International College Research Grant under Grant R202026.

Author information

Authors and affiliations.

Department of Mathematics, The Hong Kong University of Science and Technology, Hong Kong, China

Division of Business and Management, BNU-HKBU United International College, Zhuhai, China

Junyi Chai & Stella Cho

You can also search for this author in PubMed   Google Scholar

Contributions

JH carried out the collections and analyses of the literature, participated in the design of this study and preliminarily drafted the manuscript. JC initiated the idea and research project, identified the research gap and motivations, carried out the collections and analyses of the literature, participated in the design of this study, helped to draft the manuscript and proofread the manuscript. SC participated in the design of the study and the analysis of the literature, helped to draft the manuscript and proofread the manuscript. The authors read and approved the final manuscript.

Corresponding author

Correspondence to Junyi Chai .

Ethics declarations

Competing interests.

The authors declare that they have no competing interests.

Additional information

Publisher’s note.

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Part A. Summary of publications in DL and F&B domains

Part b. detailed structure of standard rnn.

The abstract structure of RNN for a sequence cross over time can be extended, as shown in Fig. 7 in Appendix , which presents the inputs as X , the outputs as Y , the weights as w , and the Tanh functions.

The detailed structure of RNN

Part C. List of abbreviations

Rights and permissions.

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/ .

Reprints and permissions

About this article

Cite this article.

Huang, J., Chai, J. & Cho, S. Deep learning in finance and banking: A literature review and classification. Front. Bus. Res. China 14 , 13 (2020). https://doi.org/10.1186/s11782-020-00082-6

Download citation

Received : 02 September 2019

Accepted : 30 April 2020

Published : 08 June 2020

DOI : https://doi.org/10.1186/s11782-020-00082-6

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

• Literature review
• Deep learning

Research Topics & Ideas: Finance

PS – This is just the start…

We know it’s exciting to run through a list of research topics, but please keep in mind that this list is just a starting point . To develop a suitable education-related research topic, you’ll need to identify a clear and convincing research gap , and a viable plan of action to fill that gap.

Overview: Finance Research Topics

• Corporate finance topics
• Investment banking topics
• Private equity & VC
• Asset management
• Hedge funds
• Financial planning & advisory
• Quantitative finance
• Treasury management
• Financial technology (FinTech)
• Commercial banking
• International finance

Corporate Finance

These research topic ideas explore a breadth of issues ranging from the examination of capital structure to the exploration of financial strategies in mergers and acquisitions.

• Evaluating the impact of capital structure on firm performance across different industries
• Assessing the effectiveness of financial management practices in emerging markets
• A comparative analysis of the cost of capital and financial structure in multinational corporations across different regulatory environments
• Examining how integrating sustainability and CSR initiatives affect a corporation’s financial performance and brand reputation
• Analysing how rigorous financial analysis informs strategic decisions and contributes to corporate growth
• Examining the relationship between corporate governance structures and financial performance
• A comparative analysis of financing strategies among mergers and acquisitions
• Evaluating the importance of financial transparency and its impact on investor relations and trust
• Investigating the role of financial flexibility in strategic investment decisions during economic downturns
• Investigating how different dividend policies affect shareholder value and the firm’s financial performance 

Investment Banking

The list below presents a series of research topics exploring the multifaceted dimensions of investment banking, with a particular focus on its evolution following the 2008 financial crisis.

• Analysing the evolution and impact of regulatory frameworks in investment banking post-2008 financial crisis
• Investigating the challenges and opportunities associated with cross-border M&As facilitated by investment banks.
• Evaluating the role of investment banks in facilitating mergers and acquisitions in emerging markets
• Analysing the transformation brought about by digital technologies in the delivery of investment banking services and its effects on efficiency and client satisfaction.
• Evaluating the role of investment banks in promoting sustainable finance and the integration of Environmental, Social, and Governance (ESG) criteria in investment decisions.
• Assessing the impact of technology on the efficiency and effectiveness of investment banking services
• Examining the effectiveness of investment banks in pricing and marketing IPOs, and the subsequent performance of these IPOs in the stock market.
• A comparative analysis of different risk management strategies employed by investment banks
• Examining the relationship between investment banking fees and corporate performance
• A comparative analysis of competitive strategies employed by leading investment banks and their impact on market share and profitability

Private Equity & Venture Capital (VC)

These research topic ideas are centred on venture capital and private equity investments, with a focus on their impact on technological startups, emerging technologies, and broader economic ecosystems.

• Investigating the determinants of successful venture capital investments in tech startups
• Analysing the trends and outcomes of venture capital funding in emerging technologies such as artificial intelligence, blockchain, or clean energy
• Assessing the performance and return on investment of different exit strategies employed by venture capital firms
• Assessing the impact of private equity investments on the financial performance of SMEs
• Analysing the role of venture capital in fostering innovation and entrepreneurship
• Evaluating the exit strategies of private equity firms: A comparative analysis
• Exploring the ethical considerations in private equity and venture capital financing
• Investigating how private equity ownership influences operational efficiency and overall business performance
• Evaluating the effectiveness of corporate governance structures in companies backed by private equity investments
• Examining how the regulatory environment in different regions affects the operations, investments and performance of private equity and venture capital firms

Need a helping hand?

Asset Management

This list includes a range of research topic ideas focused on asset management, probing into the effectiveness of various strategies, the integration of technology, and the alignment with ethical principles among other key dimensions.

• Analysing the effectiveness of different asset allocation strategies in diverse economic environments
• Analysing the methodologies and effectiveness of performance attribution in asset management firms
• Assessing the impact of environmental, social, and governance (ESG) criteria on fund performance
• Examining the role of robo-advisors in modern asset management
• Evaluating how advancements in technology are reshaping portfolio management strategies within asset management firms
• Evaluating the performance persistence of mutual funds and hedge funds
• Investigating the long-term performance of portfolios managed with ethical or socially responsible investing principles
• Investigating the behavioural biases in individual and institutional investment decisions
• Examining the asset allocation strategies employed by pension funds and their impact on long-term fund performance
• Assessing the operational efficiency of asset management firms and its correlation with fund performance

Hedge Funds

Here we explore research topics related to hedge fund operations and strategies, including their implications on corporate governance, financial market stability, and regulatory compliance among other critical facets.

• Assessing the impact of hedge fund activism on corporate governance and financial performance
• Analysing the effectiveness and implications of market-neutral strategies employed by hedge funds
• Investigating how different fee structures impact the performance and investor attraction to hedge funds
• Evaluating the contribution of hedge funds to financial market liquidity and the implications for market stability
• Analysing the risk-return profile of hedge fund strategies during financial crises
• Evaluating the influence of regulatory changes on hedge fund operations and performance
• Examining the level of transparency and disclosure practices in the hedge fund industry and its impact on investor trust and regulatory compliance
• Assessing the contribution of hedge funds to systemic risk in financial markets, and the effectiveness of regulatory measures in mitigating such risks
• Examining the role of hedge funds in financial market stability
• Investigating the determinants of hedge fund success: A comparative analysis

Financial Planning and Advisory

This list explores various research topic ideas related to financial planning, focusing on the effects of financial literacy, the adoption of digital tools, taxation policies, and the role of financial advisors.

• Evaluating the impact of financial literacy on individual financial planning effectiveness
• Analysing how different taxation policies influence financial planning strategies among individuals and businesses
• Evaluating the effectiveness and user adoption of digital tools in modern financial planning practices
• Investigating the adequacy of long-term financial planning strategies in ensuring retirement security
• Assessing the role of financial education in shaping financial planning behaviour among different demographic groups
• Examining the impact of psychological biases on financial planning and decision-making, and strategies to mitigate these biases
• Assessing the behavioural factors influencing financial planning decisions
• Examining the role of financial advisors in managing retirement savings
• A comparative analysis of traditional versus robo-advisory in financial planning
• Investigating the ethics of financial advisory practices

The following list delves into research topics within the insurance sector, touching on the technological transformations, regulatory shifts, and evolving consumer behaviours among other pivotal aspects.

• Analysing the impact of technology adoption on insurance pricing and risk management
• Analysing the influence of Insurtech innovations on the competitive dynamics and consumer choices in insurance markets
• Investigating the factors affecting consumer behaviour in insurance product selection and the role of digital channels in influencing decisions
• Assessing the effect of regulatory changes on insurance product offerings
• Examining the determinants of insurance penetration in emerging markets
• Evaluating the operational efficiency of claims management processes in insurance companies and its impact on customer satisfaction
• Examining the evolution and effectiveness of risk assessment models used in insurance underwriting and their impact on pricing and coverage
• Evaluating the role of insurance in financial stability and economic development
• Investigating the impact of climate change on insurance models and products
• Exploring the challenges and opportunities in underwriting cyber insurance in the face of evolving cyber threats and regulations

Quantitative Finance

These topic ideas span the development of asset pricing models, evaluation of machine learning algorithms, and the exploration of ethical implications among other pivotal areas.

• Developing and testing new quantitative models for asset pricing
• Analysing the effectiveness and limitations of machine learning algorithms in predicting financial market movements
• Assessing the effectiveness of various risk management techniques in quantitative finance
• Evaluating the advancements in portfolio optimisation techniques and their impact on risk-adjusted returns
• Evaluating the impact of high-frequency trading on market efficiency and stability
• Investigating the influence of algorithmic trading strategies on market efficiency and liquidity
• Examining the risk parity approach in asset allocation and its effectiveness in different market conditions
• Examining the application of machine learning and artificial intelligence in quantitative financial analysis
• Investigating the ethical implications of quantitative financial innovations
• Assessing the profitability and market impact of statistical arbitrage strategies considering different market microstructures

Treasury Management

The following topic ideas explore treasury management, focusing on modernisation through technological advancements, the impact on firm liquidity, and the intertwined relationship with corporate governance among other crucial areas.

• Analysing the impact of treasury management practices on firm liquidity and profitability
• Analysing the role of automation in enhancing operational efficiency and strategic decision-making in treasury management
• Evaluating the effectiveness of various cash management strategies in multinational corporations
• Investigating the potential of blockchain technology in streamlining treasury operations and enhancing transparency
• Examining the role of treasury management in mitigating financial risks
• Evaluating the accuracy and effectiveness of various cash flow forecasting techniques employed in treasury management
• Assessing the impact of technological advancements on treasury management operations
• Examining the effectiveness of different foreign exchange risk management strategies employed by treasury managers in multinational corporations
• Assessing the impact of regulatory compliance requirements on the operational and strategic aspects of treasury management
• Investigating the relationship between treasury management and corporate governance

Financial Technology (FinTech)

The following research topic ideas explore the transformative potential of blockchain, the rise of open banking, and the burgeoning landscape of peer-to-peer lending among other focal areas.

• Evaluating the impact of blockchain technology on financial services
• Investigating the implications of open banking on consumer data privacy and financial services competition
• Assessing the role of FinTech in financial inclusion in emerging markets
• Analysing the role of peer-to-peer lending platforms in promoting financial inclusion and their impact on traditional banking systems
• Examining the cybersecurity challenges faced by FinTech firms and the regulatory measures to ensure data protection and financial stability
• Examining the regulatory challenges and opportunities in the FinTech ecosystem
• Assessing the impact of artificial intelligence on the delivery of financial services, customer experience, and operational efficiency within FinTech firms
• Analysing the adoption and impact of cryptocurrencies on traditional financial systems
• Investigating the determinants of success for FinTech startups

Commercial Banking

These topic ideas span commercial banking, encompassing digital transformation, support for small and medium-sized enterprises (SMEs), and the evolving regulatory and competitive landscape among other key themes.

• Assessing the impact of digital transformation on commercial banking services and competitiveness
• Analysing the impact of digital transformation on customer experience and operational efficiency in commercial banking
• Evaluating the role of commercial banks in supporting small and medium-sized enterprises (SMEs)
• Investigating the effectiveness of credit risk management practices and their impact on bank profitability and financial stability
• Examining the relationship between commercial banking practices and financial stability
• Evaluating the implications of open banking frameworks on the competitive landscape and service innovation in commercial banking
• Assessing how regulatory changes affect lending practices and risk appetite of commercial banks
• Examining how commercial banks are adapting their strategies in response to competition from FinTech firms and changing consumer preferences
• Analysing the impact of regulatory compliance on commercial banking operations
• Investigating the determinants of customer satisfaction and loyalty in commercial banking

International Finance

The folowing research topic ideas are centred around international finance and global economic dynamics, delving into aspects like exchange rate fluctuations, international financial regulations, and the role of international financial institutions among other pivotal areas.

• Analysing the determinants of exchange rate fluctuations and their impact on international trade
• Analysing the influence of global trade agreements on international financial flows and foreign direct investments
• Evaluating the effectiveness of international portfolio diversification strategies in mitigating risks and enhancing returns
• Evaluating the role of international financial institutions in global financial stability
• Investigating the role and implications of offshore financial centres on international financial stability and regulatory harmonisation
• Examining the impact of global financial crises on emerging market economies
• Examining the challenges and regulatory frameworks associated with cross-border banking operations
• Assessing the effectiveness of international financial regulations
• Investigating the challenges and opportunities of cross-border mergers and acquisitions

Choosing A Research Topic

These finance-related research topic ideas are starting points to guide your thinking. They are intentionally very broad and open-ended. By engaging with the currently literature in your field of interest, you’ll be able to narrow down your focus to a specific research gap .

When choosing a topic , you’ll need to take into account its originality, relevance, feasibility, and the resources you have at your disposal. Make sure to align your interest and expertise in the subject with your university program’s specific requirements. Always consult your academic advisor to ensure that your chosen topic not only meets the academic criteria but also provides a valuable contribution to the field. 

Find The Perfect Research Topic

How To Choose A Research Topic: 5 Key Criteria

How To Choose A Research Topic Step-By-Step Tutorial With Examples + Free Topic...

Research Topics & Ideas: Automation & Robotics

Research Topics & Ideas: Robotics 50 Topic Ideas To Kickstart Your Research...

Research Topics & Ideas: Sociology

Research Topics & Ideas: Sociology 50 Topic Ideas To Kickstart Your Research...

Research Topics & Ideas: Public Health & Epidemiology

Research Topics & Ideas: Public Health 50 Topic Ideas To Kickstart Your Research...

Research Topics & Ideas: Neuroscience

Research Topics & Ideas: Neuroscience 50 Topic Ideas To Kickstart Your Research...

📄 FREE TEMPLATES

Research Topic Ideation

Proposal Writing

Literature Review

Methodology & Analysis

Academic Writing

Referencing & Citing

Apps, Tools & Tricks

The Grad Coach Podcast

thank you for suggest those topic, I want to ask you about the subjects related to the fintech, can i measure it and how?

Please guide me on selecting research titles

I am doing financial engineering. , can you please help me choose a dissertation topic?

I’m studying Banking and finance (MBA) please guide me on to choose a good research topic.

I am studying finance (MBA) please guide me to choose a good research topic.

I’m studying Master in Islamic Banking and Finance.

Can you suggest a good research topic. Please

I’m doing Masters in Islamic Banking and Finance. Would you kindly suggest a good research topic. Please

Submit a Comment Cancel reply

Your email address will not be published. Required fields are marked *

Save my name, email, and website in this browser for the next time I comment.

Submit Comment

• Print Friendly

Academia.edu no longer supports Internet Explorer.

To browse Academia.edu and the wider internet faster and more securely, please take a few seconds to  upgrade your browser .

Enter the email address you signed up with and we'll email you a reset link.

• We're Hiring!
• Help Center

Banking and finance Project Topics And Research Materials In Nigeria

How to access Banking and finance Project Topics And Research Materials In Nigeria

Related Papers

Projectclue Projectclue

How to get undergraduate Accounting Project Topics And Research Materials In Nigeria with recent research works

Academic Leadership Online Journal

ONODUGO CHRIS

The Nigerian banking sector acts as the lubricant of economic growth and development through effective financial services that impact every sector of the economy. The banking sector has been the hub and heartbeat of the Nigerian economy. It has played prominent roles in sustaining the development of the Nigerian economy through its financial obligations and performing their traditional roles of financing capital projects, mobilizing savings through a network of branches for investments. Thus, they help in capital formation. Other roles performed by the banks in the economy include financing trade, agriculture, industry, consumer activities and also help in the implementation of monetary policies. Despite having so many sectors in the economy that depend on banking, banks in Nigeria though diversified are yet to realize their full potentials even with the range of specialized actors tapping specific niches. There has been considerable innovation and diversification in the business of major commercial banks. Some of them have engaged in the areas of consumer credit, credit cards, merchant banking, leasing, mutual funds, Academic Leadership (Online Journal) ISSN: 1533-7812 Vol-21-Issue-09-September-2020 P a g e | 29 etc. This study will, therefore, examine the recent developments and trends of the Nigerian banking industry.

How to Get Undergraduate Education Project Topics And Research Materials In Nigeria

valuefronteiraonline.com

olusegun ademola

Gideon Emmanuel

Banking is an integral and core part of any economy, without which economic activities might be virtually impossible because of the strategic role of banks in not only safe-keeping of money and other vital documents of trade, but as facilitators of economic interests of sorts in the economy. However so important and strategic the roles of banks are in the economy of a nation, these have somewhat been far-fetched in developing countries, especially in Africa and most particularly, Nigeria the focus of this paper. Banking in Nigeria has started somewhat early enough, but has not grown vibrant enough to ginger the much desired growth and vibrancy of the Nigerian economy. The paper seeks to trace the evolution and genealogy of banking activities in Nigeria, its growth through different stages and through the years up until the present. The paper is thus divided n four parts starting with the introduction, to historical development, the evolution of banking in Nigeria, and the conclusion...

Tunde Ogunbado

Islamic Banking in Nigeria got approval from the Central Bank in 2011, after several years of attempts by different groups and organisations. A frame work released in the same year spell out guidelines for the establishment, operations, shari’ah governance and supervision of Islamic banks which can be standalone full-fledged institutions, subsidiaries or widows in this paper and attempts is made to discuss “Islamic Banking as a viable alternative to Conventional Banking in Nigeria. Methodology relied on biography appraisal of such studies that ate related to the study. The paper begin with a short background of Islamic Banking in Nigeria and follows with the basis for the introduction of Central Bank of Nigeria and Non interest financial institutions, Central Bank of Nigeria model, different between Islamic banking, challenges and prospect of Islamic banking and ends with suggestion that there should be a sharia supervisory board for any Islamic bank and that board should consists of trust worthy scholars who are highly qualified to issue fatawa on financial transactions.

Abdullahi Abubakar Lamido

Abstract With a predominant Muslim population, Northern Nigeria hosts the majority of the country’s population. It is blessed with various economic potentials, including vast arable land most of which is cultivable. Despite various initiatives by the Federal and State Governments as well as local and international NGOs to alleviate poverty and engender economic development, the vast majority of the populace of the region lives below the poverty line. This paper explores some of the specific roles Islamic banks can play in addressing the challenge of poverty and underdevelopment in the region. The paper highlights on the strategic role of Islamic banks’ unique model of project-based funding and profit and loss sharing, and through their veritable products such as Salam, Bay’ bi thamanAjiletc, Islamic banks have the potential of boosting agriculture and human development in both the rural and urban communities of Northern Nigeria. Key Words: Economic Development, Islamic Banking, Financing, Northern Nigeria

International Journal of economics, Business and Human Behaviour Editor

Beijing Law Review

Enya Nwocha

This paper titled “Banking Law and Economic Development in Nigeria: Contributions and Constraints of the Banks and Other Financial Institutions Act” came against the background of the current instability and uncertainty in the Nigerian banking sector and the need to strengthen the regulatory regime for optimal performance of the sector. The Banks and Other Financial Institutions Act is the principal legislation regulating banking in Nigeria and has a general application to all banks. This paper has therefore examined the strengths and weaknesses of this Act and measured its contributions to economic development in the country as well as its inherent constraints. The paper has found, among other things that the Act has contributed immensely to economic growth in the country through a multi-dimensional sectoral approach. However, a number of defects in the Act still need to be rectified. The paper has recommended the easing of over-regulation by striking out certain self-contradictory...

Loading Preview

Sorry, preview is currently unavailable. You can download the paper by clicking the button above.

RELATED PAPERS

Dunsin Ogunsemoyin

Ismail Radwan

Sylvester Brown

Kehinde Alehile

Admir Korac

Ugabi Igbaji

Journal of Emerging Economies and Islamic Research

Nuruddeen Abba Abdullahi

Habila Magit

CHAMPION RESEARCH PRESS

AFINOTAN OJITOBOME

Shehu Usman Rano Aliyu

Daniel Ayegbeni

Victor Atoi

American Journal of Social and Management Sciences

Yekini Lawal

Oyintola Isiaka I Amusa

Mohammad Dabiri , Ismail Olaoye

GATR Journals

kubebeka leonard

DALIS T. DABWOR

Publisher ijmra.us UGC Approved

Olusegun Olowe

Oluseun Paseda

Ikechukwu Acha

abdulfatah suleiman

Edinam Eton

Journal of Economics and Sustainable Development

David O Ugwunta

RELATED TOPICS

•   We're Hiring!
•   Help Center
• Find new research papers in:
• Health Sciences
• Earth Sciences
• Cognitive Science
• Mathematics
• Computer Science
• Academia ©2024