Blockchain and Bank Lending Behavior: A Theoretical Analysis

Abstract

The application of blockchain technology in the financial industry has become the focus of global attention; thus, the traditional credit business of commercial banks faces unprecedented opportunities and challenges. Based on the theoretical analysis, this study systematically examines the dynamic impact of blockchain on the lending behavior of commercial banks. The results indicate that, in the case of interest rate marketization, the blockchain expands bank credit through the mechanism of “reducing management costs”; in the case that interest rates have not yet been marketized, the blockchain can also increase bank lending through the channel of “correcting the price of funds” by narrowing the difference between the market equilibrium and the bank deposit interest rate. The results are of great significance for a deeper understanding of the impact and mechanism of blockchain on bank credit.

JEL codes: G21; O31; O33

Introduction

Blockchain is one of the most important and innovative financial technologies developed in recent years, and is widely concerned with capital markets and financial institutions (Andoni et al., 2019; Casino et al., 2019; Schinckus, 2020). It provides a secure and trusted infrastructure for transactions between unfamiliar parties without central authority by establishing a distributed ledger (Nakamoto, 2008). In the recent years, blockchain has gradually expanded its technical base to support various businesses, such as voting systems (Mosley et al., 2022), lease contracts (Hewa et al., 2021), insurance (Chondrogiannis et al., 2022), government services (Shahaab et al., 2023), and the application of blockchain in the financial sector is particularly worthy of attention.

A rich wealth of literature has studied the applications of blockchain in financial and economic fields in recent years (Javaid et al., 2022). For example, Yermack (2017) evaluated the potential impact of blockchain technology on corporate governance from the perspectives of managers, institutional investors, minority shareholders, and auditors. Raskin and Yermack (2018) believed that central banks may use this technology to launch their own digital currency. Dai and Vasarhelyi (2017) examined the operation of the blockchain to achieve a real-time, verifiable, and transparent accounting ecosystem. Wang et al. (2019) studied the mechanism of blockchain to solve the financing difficulties of SMEs. Cong et al. (2019) modeled the endogenous growth of a blockchain ecosystem and provided an estimation framework for cryptocurrency and crypto tokens. Bai et al. (2022) developed a conceptual framework for blockchain and supply chains by shedding light on the challenges of blockchain adoption.

With the continuous application of blockchain in financial fields, the world’s systematically important banks actively participate in blockchain technology application innovation in various ways (Orji et al., 2020; Patel et al., 2022). One method is to join blockchain alliances. For example, more than 40 major banking groups, including the Bank of America, Citigroup, HSBC, Credit Suisse, and others have joined the alliance and are committed to the development and application of blockchain in the financial sector, as well as industry standards and agreement formulas. At the same time, BNP Paribas and Societe Generale also work with the European Settlement Bank and Euronext to establish a blockchain clearing system for the European continent. The second method involves establishing cooperation with professional blockchain technology companies. For example, Ripple provides business innovation support to several internationally renowned banks through distributed ledger technology. Moreover, some large banks are actively exploring blockchain applications independently. Banks such as the UBS, Citigroup, Deutsche Bank, and Barclays Bank have established blockchain laboratories.

Theoretically, commercial banks attach importance to blockchain technology for the following reasons. First, it helps banks make credit arrangements that are more efficient and reasonable. Fanning and Centers (2016) believed that with the distributed storage of the blockchain, all operations, and capital flows of the debtor and guarantor can be recorded on the blockchain. Therefore, banks can always query user transaction records and statuses. Second, it reduces the operating costs by increasing the level of automation. Guo and Liang (2016) analyzed the shortcomings of traditional banks in terms of operational processes, operational efficiency, information acquisition, and information security. They pointed out that blockchain can build large and low-cost networks. Blockchain technology can realize asset digitization and point-to-point value transfer, thereby rebuilding the financial infrastructure, which will greatly improve the efficiency of the process of clearing and settling financial assets after the transaction. Third, information asymmetry can be alleviated using a blockchain. Wang et al. (2019) examined the mechanism of credit easing problems caused by information asymmetry after the introduction of blockchain and found that low-risk SMEs can obtain bank loans even without providing collateral if they borrow from blockchain technology. Therefore, blockchain technology can enable more SMEs to obtain loans, which will alleviate the financing difficulties of SMEs to a certain extent, as well as enable banks to more effectively identify the authenticity of lenders’ information, thus effectively preventing various risks brought about by information asymmetry. However, the new financial format of blockchain may also have negative impacts on traditional commercial banking businesses (Peters & Panayi, 2016). For example, the decentralization of blockchains has caused banks to face the risk of being replaced. The book record system created using blockchain technology can distribute transaction records in different places, and the book numbers can be changed in time, guaranteed to be complete, and not falsified. The application of this technology means that commercial banks are less important as payment and credit intermediaries.

Therefore, the question of whether and how blockchain affects bank lending behavior remains ambiguous (Murinde et al., 2022). Blockchain technology can significantly improve the efficiency of bank operations and alleviate the problem of information asymmetry, thus helping banks expand credit. However, it has also brought shocks to commercial banks, such as the risk of being replaced, which is not conducive to bank credit expansion (Wang et al., 2019). Accordingly, the purpose of this study is to explain the mechanism of blockchain on bank lending behavior by conducting a theoretical model containing blockchain constraints that characterize the dynamic impact of blockchain on the credit volume of commercial banks.

The main contributions of this paper lie in two aspects. First, it is among the first few studies to theoretically discuss the role of blockchain in expanding bank loans under different interest rate marketization process, emphasizing the role of blockchain in banking scenario. Second, this paper is an enrichment and supplement to the existing bank lending theory by considering recently prosperous financial technology. This study finds that the application of blockchain can expand bank credit through the mechanism of “reducing management costs” and the channel of “correcting the price of funds,” enriching the literature on bank credit determination.

The remainder of this paper is organized as follows. Section 2 constructs theoretical models to analyze the blockchain mechanism in bank lending. Section 3 presents a discussion. Section 4 concludes the study.

Theoretical Model

Situation 1: The Application of Blockchain Can Reduce Management Costs

To illustrate the effect of blockchain on the response to bank lending, we follow Dell’ Ariccia et al. (2017) and Cheng and Qu (2020), and present a model of a representative bank that pursues profit maximization (modified from Peek and Rosengren [1996] and Kishan and Opiela [2000]). This bank is assumed to have two assets: required reserves (R) and loans (L); and two liabilities, demand deposits (D) and capital (E). The balance sheet constraint requires

$$\mathrm{R}+\mathrm{L}=\mathrm{D}+\mathrm{E}.$$

(1)

The deposit reserve ratio is ρ, and the commercial bank does not hold excess deposit reserves.

$$\mathrm{R}=\rho \mathrm{D},0<\rho <1.$$

(2)

We first assume that the deposit interest rate has been under marketization. Thus, the scale of commercial bank deposits is inversely related to the market interest rate (e.g., the federal funds rate [r_ff]),

$$\mathrm{D}={\mathrm{D}}_0-{\mathrm{D}}_1{\mathrm{r}}_{\mathrm{ff}},{\mathrm{D}}_1>0.$$

(3)

Assuming that the loan interest rate has also been under marketization, the scale of commercial bank loans is positively correlated with the extent to which the credit market equilibrium interest rate ( $r_L^{*}$ ) is higher than the bank loan pricing (r_L):

$$\mathrm{L}={\mathrm{L}}_0+{\mathrm{L}}_1(r_L^{*}-{\mathrm{r}}_{\mathrm{L}}),{\mathrm{L}}_1>0.$$

(4)

The capital scale of commercial banks is negatively correlated with the distance between the capital market equilibrium rate and the bank equity capital return rate:

$$\mathrm{E}={\mathrm{E}}_0+{\mathrm{E}}_1(r_E^{*}-{\mathrm{r}}_{\mathrm{E}}),{\mathrm{E}}_1<0.$$

(5)

The mean market rates are assumed to be directly related to the federal funds rate (r_FF) with fixed spreads given by f₀ and g₀:

$$r_E^{*}={\mathrm{a}}_0+{\mathrm{r}}_{\mathrm{ff}.}$$

(6)

$$r_L^{*}={\mathrm{g}}_0+{\mathrm{r}}_{\mathrm{ff}.}$$

(7)

We assume that representative commercial banks pay management and service costs in their daily operations. Following Kopecky and VanHoose (2004), representative banks manage their balance sheet items using a quadratic cost function. Constants w and v represent the unit management cost coefficients of deposits and loans, respectively.

$$C={\left(\frac{w}{2}\right)}^{2}D+{\left(\frac{v}{2}\right)}^{2}L.$$

(8)

Suppose the bank applies the blockchain technique. It is assumed that the construction of a blockchain platform may help commercial banks upgrade their business processes, speed up data processing, and reduce the service costs invested in verifying the authenticity of data. Hence, we assume that blockchain application can improve management efficiency and reduce the marginal cost of bank units.

$$w=w\left(\mathrm{bc}\right),\frac{\partial w\left(\mathrm{bc}\right)}{\partial \mathrm{bc}}<0.$$

(9)

$$v=v\left(\mathrm{bc}\right),\frac{\partial v\left(\mathrm{bc}\right)}{\partial \mathrm{bc}}<0.$$

(10)

Banks are assumed to maximize profits ( $\pi$ ),

$$\pi =r_{L}L-r_{D}D-r_{E}E-C.$$

(11)

Profits include interest income on loans (r_LL) minus the interest paid on demand deposits (r_DD), the costs of managing capital (r_EE), and daily operations (C).

According to the above assumptions, the objective function and constraints of representative commercial banks are as follows:

$$\mathrm{Max}\pi =r_{L}L-r_{\mathrm{ff}}D-r_{E}E-C$$

S.t

$$\begin{array}{c}\{\begin{array}{c}R+L=D+E\hfill \\ \multicolumn{1}{c}{R=\rho *D,0<\rho <1}\\ \multicolumn{1}{c}{L=L_0+L_1({r_L}^{*}-r_L),L_1>0}\\ \multicolumn{1}{c}{D=D_0-D_1*r_{\mathrm{ff}},D_1>0}\\ \multicolumn{1}{c}{E=E_0+E_1({r_E}^{*}-r_E),E_1<0}\\ \multicolumn{1}{c}{{r_E}^{*}=a_0+\varphi *r_{\mathrm{ff}}}\\ \multicolumn{1}{c}{{r_L}^{*}=g_0+\varphi *r_{\mathrm{ff}}}\end{array}\\ c=\frac{w}{2}*D^2+\frac{V}{2}*L^2\end{array}$$

Equation (11) is maximized with respect to L after converting E, D, R, C, r_L, and r_E into a function of L, and we obtain that:

$$\{\begin{array}{c}E=(\rho -1)(D_0-D_1*r_{\mathrm{ff}})+L\hfill \\ \multicolumn{1}{c}{D=D_0-D_1*r_{\mathrm{ff}}}\\ \multicolumn{1}{c}{R=\rho *D=\rho *(D_0-D_1*r_{\mathrm{ff}})}\\ \multicolumn{1}{c}{c=\frac{w}{2}*{(D_0-D_1*r_{\mathrm{ff}})}^2+\frac{V}{2}*L^2}\\ \multicolumn{1}{c}{r_L={r_L}^{*}-\frac{L-L_0}{L_1}=g_0+\varphi *r_{\mathrm{ff}}-\frac{L-L_0}{L_1}}\\ \multicolumn{1}{c}{r_E={r_E}^{*}-\frac{E-E_0}{E_1}=a_0+\varphi *r_{\mathrm{ff}}-\frac{(\rho -1)(D_0-D_1*r_{\mathrm{ff}})+L-E_0}{E_1}}\end{array}$$

Therefore, the objective function (11) can be rewritten as follows,

$$\begin{array}{c}max\pi =(g_0+\varphi *r_{\mathrm{ff}}-\frac{L-L_0}{L_1})*L-r_{\mathrm{ff}}*(D_0-D_1*r_{\mathrm{ff}})\hfill \\ \multicolumn{1}{c}{-(a_0+\varphi *r_{\mathrm{ff}}-\frac{(\rho -1)(D_0-D_1*r_{\mathrm{ff}})+L-E_0}{E_1})*}\\ ((\rho -1)(D_0-D_1*r_{\mathrm{ff}})+L)-\frac{w}{2}*{(D_0-D_1*r_{\mathrm{ff}})}^2-\frac{V}{2}*L^2\hfill \end{array}$$

(12)

The first-order necessary conditions are used to solve for L,

$$\begin{array}{c}\multicolumn{1}{c}{g_0+\varphi *r_{\mathrm{ff}}-\frac{2L}{L_1}+\frac{L_0}{L_1}-v*L}\\ \multicolumn{1}{c}{-(a_0+\varphi *r_{\mathrm{ff}}-\frac{2(\rho -1)(D_0-D_1*r_{\mathrm{ff}})+2L-E_0}{E_1})=0}\end{array}$$

(13)

To solve for Equation (11), we can get the optimal L,

$$L^{*}=\frac{{\mathrm{\Delta }}_1+{\mathrm{\Delta }}_2(D_0-D_1*r_{\mathrm{ff}})}{{\mathrm{\Delta }}_3+v*{\mathrm{\Delta }}_4}$$

(14)

Among them, ${\mathrm{\Delta }}_1=g_0{E}_1{L}_1+E_1{L}_0-E_0{L}_1-a_0{L}_1<0,{\mathrm{\Delta }}_2=2L_1(\rho -1)<0,{\mathrm{\Delta }}_3=-2(L_1-E_1)<0,{\mathrm{\Delta }}_4=L_1{E}_1<0.$

Testable hypotheses can be derived by taking the derivatives of L with blockchain (bc):

$$\frac{\partial L}{\partial v}*\frac{\partial v}{\partial \mathrm{bc}}=\frac{{\mathrm{\Delta }}_4*({\mathrm{\Delta }}_1+{\mathrm{\Delta }}_2(D_0-D_1*r_{\mathrm{ff}}))}{-{({\mathrm{\Delta }}_3+v*{\mathrm{\Delta }}_4)}^2}*\frac{\partial v}{\partial \mathrm{bc}}>0.$$

(15)

Equation (15) implies that the application of blockchain has increased the scale of commercial credit through the “reduction of management fee” channels. Alternatively, the new technology of the booming blockchain has reduced the management costs of commercial banks, thereby increasing bank credit.

Situation 2: The Application of Blockchain Can Correct the Price of Funds

In the previous section, we assume that the deposit interest rate is under marketization. Now, we impose constraints on the deposit market. Empirical research of many scholars shows that when the deposit interest rate is under long-term control, the real interest rate is lower than the equilibrium interest rate. Therefore, the higher the market equilibrium interest rate is than the deposit benchmark interest rate, the lower the scale of deposits accepted by commercial banks. Hence, the constraints of the deposit market now change to:

$$D=D_0+{D_1}^{\prime }(r_D-r_{\mathrm{ff}}),{D_1}^{\prime }>0$$

(16)

We assume that non-tampering and transparency in blockchain data can promote interest rate marketization, thus correcting the funds prices of commercial banks (Nautz & Schmidt, 2009).

$$f\left(\mathrm{bc}\right)=r_D-r_{\mathrm{ff}},\frac{\partial f(\mathrm{bc})}{\partial \mathrm{bc}}<0,$$

(17)

Holding other conditions unchanged, the objective function and constraints of representative commercial banks are as follows:

$$\mathrm{Max}\pi =r_{L}L-r_{D}D-r_{E}E-C$$

S.t

$$\{\begin{array}{c}R+L=D+E\hfill \\ \multicolumn{1}{c}{R=\rho *D,0<\rho <1}\\ \multicolumn{1}{c}{L=L_0+L_1({r_L}^{*}-r_L),L_1>0}\\ \multicolumn{1}{c}{D=D_0+{D_1}^{\prime }(r_D-r_{\mathrm{ff}}),{D_1}^{\prime }>0}\\ \multicolumn{1}{c}{E=E_0+E_1({r_E}^{*}-r_E),E_1<0}\\ \multicolumn{1}{c}{{r_E}^{*}=a_0+\varphi *r_{\mathrm{ff}}}\\ \multicolumn{1}{c}{{r_L}^{*}=g_0+\varphi *r_{\mathrm{ff}}}\end{array}$$

$$\begin{array}{c}c=\frac{w}{2}*D^2+\frac{V}{2}*L^2\\ \mathrm{f}\left(\mathrm{bc}\right)={\mathrm{r}}_{\mathrm{D}}-{\mathrm{r}}_{\mathrm{ff}},\frac{\partial f(\mathrm{bc})}{\partial \mathrm{bc}}<0\end{array}$$

By eliminating other variables in the expression of L, the objective function can be rewritten as follows:

$$\begin{array}{c}max\pi =(g_0+\varphi *r_{\mathrm{ff}}-\frac{L-L_0}{L_1})*L-(f+r_{\mathrm{ff}})*(D_0+{D_1}^{\prime }*f)\hfill \\ \multicolumn{1}{c}{-(a_0+\varphi *r_{\mathrm{ff}}-\frac{(\rho -1)(D_0+{D_1}^{\prime }*f)+L-E_0}{E_1})*}\\ ((\rho -1)(D_0+{D_1}^{\prime }*f))-\frac{w}{2}*{(D_0+{D_1}^{\prime }*f)}^2-\frac{V}{2}*L^2\hfill \end{array}$$

(18)

The first-order necessary conditions are used to solve for L,

$$\begin{array}{c}\multicolumn{1}{c}{g_0+\varphi *r_{\mathrm{ff}}-\frac{2L}{L_1}+\frac{L_0}{L_1}-V*L}\\ \multicolumn{1}{c}{-(a_0+\varphi *r_{\mathrm{ff}}-\frac{2(\rho -1)(D_0+{D_1}^{\prime }*f)+2L-E_0}{E_1})=0}\end{array}$$

(19)

To solve for Equation (19), we can get the optimal L,

$$L^{*}=\frac{{\mathrm{\Delta }}_1+{\mathrm{\Delta }}_2(D_0+D_1^{\prime }*f)}{{\mathrm{\Delta }}_3+v*{\mathrm{\Delta }}_4}$$

(20)

Among them, ${\mathrm{\Delta }}_1=g_0{E}_1{L}_1+E_1{L}_0-E_0{L}_1-a_0{L}_1<0,{\mathrm{\Delta }}_2=2L_1(\rho -1)<0,{\mathrm{\Delta }}_3=-2(L_1-E_1)<0,{\mathrm{\Delta }}_4=L_1{E}_1<0$ .

The testable hypotheses can then be derived by taking the derivatives of L with the blockchain (bc):

$$\frac{\partial L}{\partial f}*\frac{\partial f}{\partial \mathrm{bc}}=\frac{{\mathrm{\Delta }}_2*D_1^{\prime }*f}{{\mathrm{\Delta }}_3+v*{\mathrm{\Delta }}_4}*\frac{\partial f}{\partial \mathrm{bc}}>0.$$

(21)

Equation (21) indicates that the blockchain has increased the lending of commercial banks through the channel of “correcting price of funds.” Alternatively, with blockchain, the market price mechanism has been rationalized, and the difference between the benchmark deposit and the market equilibrium interest rate has been reduced, thereby increasing commercial bank lending behavior.

By considering the two effects together, therefore,

$$\frac{\partial L}{\partial \mathrm{bc}}=\frac{\partial L}{\partial f}*\frac{\partial f}{\partial \mathrm{bc}}+\frac{\partial L}{\partial v}*\frac{\partial v}{\partial \mathrm{bc}}>0.$$

(22)

Equation (22) implies that, under the two channels of blockchain that reduce management costs and correct the price of funds, the bank’s borrowing level can be improved.

Discussions

Next, we analyze and discuss the implications derived from the conclusions of the theoretical model. Information asymmetry between enterprises and banks is the main reason for restricting bank credit expansion (Stiglitz & Weiss, 1981). Therefore, banks go through a series of procedures when issuing loans, including pre-lending investigations, risk assessments, loan approvals, post-loan management, and liability identification (Rose & Hudgins, 2012). This process often takes a relatively long time and costs a lot of money, increasing the operating costs of the bank. For example, it is difficult for banks to obtain real information on SMEs, and the cost of bank pre-lending investigations is high. Moreover, banks need to check the status and value of the collateral, requiring a large amount of manpower and material resources.

Blockchain is an emerging decentralized architecture and distributed computing paradigm made up of chains of blocks (Nakamoto, 2008). It relies on program algorithms to automatically record credit-related information and store it on each computer in the blockchain network, which is transparent, non-tamperable, and low in cost (Schinckus, 2020). Specifically, each distributed node can encapsulate the transaction data received over a period of time into a time-stamped data block and link to the current main blockchain through a specific hash algorithm and Merkle tree data structure. Each block is connected to form the longest main chain from the creation to the current block, thereby recording the complete history of the data and providing data traceability and positioning functions. To simplify, blockchain technology can be compared with a shared ledger, which has the technical characteristics of consensus, encryption, and non-tampering. Thus, transactions of assets, funds, transactions, etc., of individuals or businesses using blockchains can be accurately recorded on the block.

Commercial banks can store and share customer credit information in this organization in an encrypted form. When a customer applies for a loan, the lending institution can directly complete the credit information by directly retrieving the corresponding information data of the blockchain after obtaining authorization rather than asking the central bank for credit information inquiry. Simultaneously, obtaining data from the blockchain ensures authenticity and real-time acquisition of the acquired information. Banks can also build asset management systems using blockchain technology, and use this system for post-lending verification and management (Wang et al., 2021). Through this system, the authenticity of the collateral is verified, and the transfer of collateral can be monitored in real time, thereby greatly reducing operating costs. Additionally, smart contracts for blockchain technology can increase the efficiency of commercial banking operations. It is often necessary to involve multiple trading entities in complex trading projects. Multiparty communication and verification often result in information asymmetry. Smart contracts in the blockchain can be made more efficient and standardized by reducing manual intervention by automating the operation of rules or protocols. Therefore, the application of the blockchain can also replace the large number of account registration and settlement functions performed by the back-office departments in the bank, thereby saving the time and cost of manual recording and operation. Hence, in the credit business, banks use the blockchain to realize real-time information storage, verification, and tracking to realize the intelligent operation of credit business, which not only reduces the mistakes of manual transactions but also greatly improves business efficiency, thus helping to expand credit business in banks.

Moreover, with blockchain, banks can collect and evaluate the credits of all participants and control them in real time. From the perspective of market risk, information asymmetry and asset mismatches in the intermediary market not only lead to losses but also disrupt the market pricing mechanism. The non-tampering and transparency of the data in the blockchain can make the reaction price of the whole market more realistic in response to the demand for funds, change the pricing mechanism of funds and the supply and demand model of funds, and break the pattern of high spreads enjoyed by commercial banks, thus forming a more realistic market interest rate that is conducive to promoting the marketization process, thereby increasing bank credit.

Conclusions

Blockchain is a booming technology, and its application has undergone profound changes on a global scale; in particular, its application in the financial industry has become the focus of global attention both in academia and in practice. The explosive development of blockchain in finance has rapidly changed the original financial ecology, and thus, commercial banks’ credit businesses face unprecedented opportunities and challenges. In this context, we examine the relationship between blockchain and bank lending behavior, which has great theoretical and practical significance.

This study introduces the blockchain financial constraint hypothesis into the theoretical model following Dell’ Ariccia et al. (2017) and Cheng and Qu (2020), and examines the dynamic impact of blockchain on the credit volume of commercial banks. Our theoretical model shows that, in the case of interest rate liberalization, blockchain can expand bank lending by reducing operating costs and improving operational efficiency. When the deposit interest rate is not market oriented, the blockchain can also increase bank lending by promoting the interest rate marketization process through price transparency.

This study has several significant policy implications. Regulatory authorities should keep pace with times, pay attention to the advantages of blockchain, encourage financial innovation, and provide more support to commercial banks using blockchain technology. Countries that have not yet fully realized market interest rates should steadily promote financial system reform, gradually complete interest rate liberalization, encourage blockchain technology involved in traditional financial business to complement each other, and promote financial services to better serve the real economy.

Nevertheless, more work needs to be done in the future. First, the building of the theoretical model is static and does not combine the current and future decisions of banks. Second, the model fail to integrate the environmental aspects such as supply chain analysis. Third, it is also necessary to carry out a qualitative analysis in the future to test the theoretical model if data is available.