“Digital banking impact on Turkish deposit banks performance”

The technological developments in the banking sector have significant implications for banks and are dramatically changing the way retail banks conduct their business. Banks can invest in digital banking (DB) services either to acquire a strategic advan- tage or because doing so has become a strategic necessity. This study is organized to examine if DB service channels have any positive or negative impact on Turkish de- posit banks’ performance. With this aim in mind, in the first stage of the proposed DEA model, physical assets are used. Then, in the second stage, DB service channels are added to see if they have any impact on banks’ performance. The results show that the banks are investing in DB services just to keep the competition as it is. In other words, they invest in DB services as a strategic necessity. DB services do not provide any strategic advantage to any banks in terms of financial performance or efficiency since the banks are already efficient. Investing in DB only helped to preserve their strategic po-sitions. The Turkish deposit banking industry is very competitive and very profitable, and it is necessary to invest in DB services just to keep the competition as it is. This study is organized to examine if DB service channels have any positive or negative impact on Turkish deposit banks’ performance. To the end, in the first stage of the proposed DEA model, physical assets were just used. Then, in the second stage, DB service channels were added to the model to evaluate if they have any impact on banks’ performance. In both stages all banks are efficient except two, Isbank and YKB. In other words, with or without DB service channels the five banks are efficient and two banks are not efficient. It can be concluded that the five efficient banks have competitive advantage in terms of physical and DB service channels. These five banks invest enough in DB services to keep their high performance. In other words, if they had not invested in DB as they did, their efficiency would be affected negatively and in the second stage they could not be efficient. The banks are investing in DB services just to keep the competition position as it is. It can be concluded that they invest in DB services as a strategic necessity. DB services do not provide any strategic advantage to any banks in terms of financial performance or efficiency.

The technological developments in the banking sector, digital banking (DB) in particular, have significant implications for banks and are dramatically changing the way retail banks are conducting their business. Over the last decade DB has had a major impact on customer interfaces. The speed of change has increased because of the introduction of new technologies and evolution of customer needs. Telephone, Internet and mobile banking have become major ways of delivering multi and even omni-digital channel DB services to customers, a shift that is challenging traditional banking services (Cortiñas, Chocarro, & Villanueva, 2010). DB enables customers to conduct banking transactions anytime and anywhere, faster and with lower fees, therefore it is more attractive for customers compared to traditional banking services (Sayar & Wolfe, 2007). Despite the fact that DB has important and valuable advantages for customers, they have embraced DB services to different degrees. Nevertheless, more and more basic banking transactions are shifting from physical channels to digital channels, leading to a major transformation of banks' strategic positions. Most banking institutions invest in IT to improve delivery of financial services on digital channels to keep pace with global competition. customer base, expanding the market share, decreasing the dependence on traditional banking services and branches, and responding quickly and more accurately to the customer's constantly changing needs and expectations.
As customers' behaviors and needs change and expectations increase, preserving the current ones and gaining new ones at the same time as increasing profitability and decreasing costs becomes key, especially in a highly competitive and almost zero (even negative) interest rate global environment. In this regard, DB enables banks to improve services for changing customer needs, minimize costs by reducing physical transactions with customers in branches, reduce the gap between customer expectations and delivered services (Japparova & Rupeika-Apoga, 2017), boost customer loyalty and satisfaction and generate revenue from different consumer segments.
The adoption of information technology (IT) in the banking sector has significantly changed the banking structure from the traditional banking system to the digital banking system. Advances in IT have been the driving force of DB services for banks over the years. Advances in IT can affect the firms in two ways. First, by investing in IT, firms can extend their business models, improve their business processes, efficiency and effectiveness, and increase customer satisfaction. In this way they can acquire competitive and strategic advantage by investing in IT. In this first way firms invest in IT deliberately and proactively to gain strategic and competitive advantage. Goh and Kauffman (2013) define this view as the strategic advantage perspective. Second, firms are forced to invest in IT by their competitors. As technology becomes pervasive and more accessible, sustaining any strategic and competitive advantage becomes a challenge. While competitors move rapidly to invest in IT, which enables them to gain a competitive advantage, some of the firms can face sustained disadvantages in changing environments. Goh and Kauffman (2013) define this view as strategic necessity. Since the market conditions force the firms to invest in IT, in this case the firms are passive and reactive to the environmental conditions. They have to invest in IT because the market conditions force them to. If they don't invest in IT as their competitors do, they can lose their market share, their current customer base and their opportunity to gain new customers. In the first alternative, the firms who invest in IT define the market conditions. If they invest in the appropriate technologies, which give them strategic advantage, they are the winners in terms of customer satisfaction, market share and financial performance. Whatever the reason behind the IT investment, in both cases, banks are trying either to gain the strategic advantage or strategically to sustain their position by investing in DB services. But what would be the impact on the performance or efficiency of the bank? Is it always profitable? Is it always efficient? In this study, DB services impact on banks' performance and efficiency is analyzed. Whatever the reason for adopting DB services, it is essential to look at the impact on bank performance and efficiency. There is a large volume of research into bank performance but much less on the impact of DB services on performance. Despite the importance of measuring bank performance based on the DB strategy, there is not enough empirical research on this issue. Thus, this paper offers a new perspective on measuring bank performance by using Data Envelopment Analysis (DEA) in terms of DB services and is providing a new insight into this issue in terms of theoretical and practical results.

1.
The use of technology in DB and its impact on meeting customer needs, increasing operational efficiency and financial performance can be understood by taking into account different factors such as: customers' perception about user friendliness and easy use, user interface quality, and Internet and mobile banking service quality (Mbama & Ezepue, 2018).
Some research shows that adoption of online banking technologies is a significant strategic choice for banks' competitive position, since a wider range of online banking services plays a crucial role to influence the financial performance of a bank by providing more profit than for those with a limited online access ( Goh and Kauffman (2013) argue that there are two reasons for IT investments: strategic advantage and strategic necessity. In their case of the US commercial banking industry, strategic necessity affects significantly IT investments and is more critical than strategic advantage. Their research indicates that (Goh & Kauffman, 2013): • Banks that have IT investments in Internet banking likely have better performance.
• IT investments are affected by bank transaction cost and consumer deposits.
• IT investments that materialized to gain strategic advantage seem to have been diminishing over the years, whereas IT investments as a strategic necessity seem to have been increasing. It is vital to understand the channel preferences of the customers. Some of them prefer using a single channel. They use only one channel at a time, e.g. Branch or ATM or Internet banking. Others prefer a multi-channel approach. Some of them use more than one channel, e.g. Internet banking, call center, ATM and branches, etc. Therefore, banks are challenged to integrate all banking services into an omni-channel which is a multi-channel approach that seeks to provide the customer with seamless banking services whether the customer is banking from a PC or mobile device, or ATM, or in a branch, so that customers experience the same level of service regardless of how they are interacting with their banks. Thus, understanding this behavior and integrating all banking services channels consistently not only provide a strategic advantage but also are a competitive necessity for banks to understand customer cross-channel transaction behavior, provide a more robust and consistent customer experience, and manage channels effectively (Liu, 2016

2.
The aim of this article is to analyze the effects of digital banking services on banks' performance and efficiency by using DEA as an analytical tool. DEA can be employed to analyze relative efficiency of organizations and/or parts of organizations that are similar in terms of their resources and their results. Multiple inputs and multiple outputs can be utilized for efficiency calculation.
In this regard seven deposit banks in Turkey have been chosen. The deposit banks' second hand data and annual reports were obtained from their web sites and from the Turkish Banks Association (TBA) web site. The seven deposit banks and their main variables are shown in Table 1. As shown, these seven deposit banks make up 70% of assets, 74% of deposits and 73% of net profits of the total banking industry in Turkey.

3.
The output variables are related to the banks' service and revenue, while the input variables measure the banks' operating costs. In order to evaluate deposit banks' performance, used models are shown in Figure 1 and Figure 2. In the first model, three inputs that relate to costs and physical banking: total assets ( (Thanassoulis, 1999), and number of POSs, are included in calculations as inputs. Obtained first and second stage results are then analyzed to measure to what extent DB service channels affect banks' performance. Therefore, it has been evaluated how the DB services have impact on the bank's efficiency scores. Efficiency calculations are made by both MaxDEA and DEA-Solver LV software. Statistics on input and output data is given in Table 2.

Banking industry in Turkey
The banking industry is a major part of the financial system in Turkey, accounting for 82% of total assets. Deposit banks have 91% of all bank employees and 90% of all banks total assets as of December 2017. There are 33 deposit banks in Turkey, nine of them are privately owned, three state-owned, one bank is under the deposit insurance fund and 20 foreign banks, according to the Turkish Banks Association (TBA).

Analysis of DEA scores
Firstly, calculations are made for the first stage model by using banks' physical assets, number of employees, number of branches and total assets of 2017 as input; and asset growth rate, total deposits, total credits, and net profit of 2017 as output. Then, DB service channels, number of POSs, number of credit cards, number of ATMs and digital banking market share are included in the model as inputs, and efficiency scores are recalculated for the second stage model. Therefore, adding DB service channels to the model allows to determine how and to what extent DB service channels affect banks' efficiency scores.
There are two traditional DEA models; first one is input oriented, second one is output oriented. The input oriented model aims to minimize inputs while keeping the same outputs level, on the other hand, the output oriented model aims to maximize outputs while keeping the same inputs level. In this study, the output oriented model is the most appropriate one since the main aim of the bank is to maximize deposits and credits and so profit (Kahveci, 2011(Kahveci, , 2012. Although, the Constant Returns to Scale (CRS) model, suggested by Charnes, Cooper, and Rhodes (1978), is used for technical efficiency, the Variable Returns to Scale (VRS) model, suggested by Banker, Charnes, and Cooper (1984), is used for pure technical efficiency. An organization's performance defined by technical efficiency is described by maximizing the produced level of outputs at the given input level (Farrell, 1957). The technical efficiency (CRS) score for a Decision Making Unit (DMU) shows relative performance of particular DMU compared to all other DMUs in that particular sample. However, scale efficiency (SE) expresses whether an organization is operating at its optimal size. The relation between technical efficiency and pure technical efficiency is defined by the equation below (Kahveci, 2011(Kahveci, , 2012Ulucan, 2002).

CRS
. VRS SE (1) All the results for the first and second stage models are given in Table 4. When the results are analyzed, almost all the banks are efficient except Türkiye İş Bankası A.Ş. (Isbank) and Yapı ve Kredi Bankası A.Ş. (YKB) in both stages. It is interpreted that they are not efficient in terms of physical service channels and digital service channels. Although both banks are not efficient, they have over 0.9 in This study is organized to examine if DB service channels have any positive or negative impact on Turkish deposit banks' performance. To the end, in the first stage of the proposed DEA model, physical assets were just used. Then, in the second stage, DB service channels were added to the model to evaluate if they have any impact on banks' performance. In both stages all banks are efficient except two, Isbank and YKB. In other words, with or without DB service channels the five banks are efficient and two banks are not efficient. It can be concluded that the five efficient banks have competitive advantage in terms of physical and DB service channels. These five banks invest enough in DB services to keep their high performance. In other words, if they had not invested in DB as they did, their efficiency would be affected negatively and in the second stage they could not be efficient. The banks are investing in DB services just to keep the competition position as it is. It can be concluded that they invest in DB services as a strategic necessity. DB services do not provide any strategic advantage to any banks in terms of financial performance or efficiency. By investing in DB as they did, they have preserved their strategic advantage. Although YKB and Isbank are not efficient in either stage, they have a high score of over 0.9. They can make some improvements by arranging their assets to their outputs. They have to focus on both physical and digital service channels and to transform their resources to the desired results. In terms of scale efficiency, YKB has to look into the right scale in accordance with its inputs and outputs in both stages whereas Isbank does not have scale efficiency in the first stage without DB services, but in the second stage, with DB services it does have scale efficiency. The other five banks also have scale efficiency, so they do not need any scale arrangements.
Banks could invest in IT for DB services with two main concerns. The first one is saving costs and the second one is satisfying customer experiences and expectations. A successful transformation process should be both cost saving and satisfactory for customers. Either focusing solely on cost saving rather than customer satisfaction or solely on customer satisfaction rather than cost saving could be disastrous for the banks. In the first case, banks can invest in cost saving technologies that do not meet customer needs. In the second case, they can invest in customer satisfactory technologies that are not profitable or are costly. In both cases it results in non-efficient investments and the financial performance of banks can be negatively affected. In Turkish deposit banks' case, the two stages of DEA scores show that DB service channels do not have any negative or positive impact on banks' performance and efficiency. But, overall, the Turkish deposit banks examined in this research are highly efficient in terms of physical channels and DB channels since they are efficient in both stages. If the banks continue to invest in DB services in the same way as in the past they will keep their position and their efficiency. Isbank and YKB can increase efficiency by arranging DB services.
The Turkish deposit banking industry is very competitive and it is necessary to invest in DB services just to keep the competition as it is. It could be concluded that in the Turkish case investing in DB services is just a strategic necessity since the competition is fierce. The banking industry is already profitable and all banks in this study have a good amount of profit. That is why almost all of them are efficient in both stages. When we look at the capital/profit ratio, the average ratio of all seven banks is 14%, whereas YKB and Isbank have a 12% capital/profit ratio, lower than the other five banks. Thus, this also explains why those two banks are not efficient compared to the others.
Despite the importance of measuring bank performance based on the DB strategy, there is not enough empirical research on this issue. Thus, this paper gives a new perspective on measuring bank performance by using DEA in terms of DB services. For further research, with more detailed DB data, banks' past performance could be compared and how well developed their DB performance was over time could be analyzed. Besides, Turkish banks' DB applications and strategies can be compared with other countries' banks, thus international comparisons could also be made by using the suggested DEA model.