Recent Shifts in Determinants of Internet Banking Adopters: Empirical Evidence from 2001 and 2004 SCF

Beginning of article

The purpose was to investigate whether there were shifts in the factors affecting the probability of adopting Internet banking. The technology acceptance model, the theory of planned behavior, and previous research were used to develop a model. Data were drawn from the 2001 and 2004 Survey of Consumer Finances. Three shifts were identified. In 2004, households adopted Internet banking regardless of income while income was a determinant in 2001. Second, the length of the planning horizon was not significant in 2004 suggesting that household's use Internet banking was a convenience and not a result of long-term planning. Third, occupation as a proxy for social norm influenced the probability of adopting Internet banking.

Introduction

Security First Network Bank (SFNB) was founded in October, 1995 in Atlanta, Georgia. This institution claimed to be the first ever 'pure' Internet bank (Cramer and Harrell, 1996). Even though it has only been a decade since SFNB was established, no respectable financial institution in the world today functions without extending its offer via Internet. Advancement in information technology, which enabled implementation of digital services, has rapidly gained recognition of financial institutions in the last years. Accessing banks online has become a regular practice of 35% of U.S. households, and it is a forecast that 55 million households will use online banking services by 2010 (Bruene, 2005).

The phenomenon of rapid advancement in implementation of information technology by banks has attracted a significant amount of researchers' attention. Numerous studies, employing a variety of methodologies addressed the adoption of Internet banking. Kim, Widdows and Yilmazer (2005) investigated the determinants of Internet banking adoption based on individuals' benefits and costs. Their study used the 2001 Survey of Consumer Finances (SCF). A similar study by Gerrard and Cunningham (2003) addressed the economic benefits of Internet banking adoption among customers in Singapore. Chang (2003) found that 25% of the population in Korea adopted electronic banking, which means, that in some countries Internet banking has become a widespread service. Some researchers have also addressed the adoption of Internet banking based on the diffusion of innovation and technology acceptance model, or theory of planed behavior (Chang, 2003; Lee, Lee, and Eastwood, 2003; Chau and Lai, 2003). Jun and Cal (2001) addressed the problem of determinants of Internet banking adoption incorporating perceived service quality as a relevant factor explaining adoption. The problem has also been explored from the supply side perspective (Gopalakrishnan, Wischnevsky, and Damanpour, 2003; Tan and Teo, 2000) and in the qualitative studies of adoption of Internet banking among corporate customers (Rotchanakitumnuai and Speece, 2003).

While the previous studies answer the questions about determinants of consumers' adoption of Internet banking, no one has investigated the shifts in the determinants of Internet banking adoption with data from the U.S. Some researchers admit that due to the turbulent nature of the electronic services market, the determinants of adoption of online services still need attention (Kolodinsky, 2004). Providers of online banking services could benefit from learning how these determinants have evolved over recent years. This knowledge could help to adjust market offers by targeting the most promising groups of potential customers. Bank managers understand why they should encourage customers to utilize the electronic banking channel. Their benefits are described by Chang (2002), who compared the average costs of money transfer transactions. For traditional checking transaction the cost is estimated at approximately $1, for ATM or telephone banking it varies between $.30 and $.50, while the cost of Internet transaction is less than $.01.

The purpose of this research is to determine the dynamics of change in the determinants of adoption of online banking services. The study will use the 2001 and 2004 Surveys of Consumer Finances (SCF) to identify the shifts in the explanatory power of determinants of Internet banking adopters. The online banking services, which are the focus of the study, are all kinds of services offered by commercial banks, saving or loan banks and credit unions.

Review of literature

Electronic banking is defined as the use of computers and telecommunications to enable banking transactions, usually without human interaction. A person with a personal computer can make transactions, either via direct connection or by accessing the bank's web site. Internet banking is defined as making transactions, payments, etc., over the Internet, usually through the bank's website (Encyclopedia Britannica Online, 2006). In this research, Internet banking and on-line banking are synonymous. Electronic banking is a broader term that includes Internet banking, but also other ways of enabling banking transactions such as telephone banking, m-banking, etc. The term e-banking, often used in other studies will not be used in this paper, as many discrepancies exist in the definitions.

The adoption of Internet banking channels is conditional on ownership of computer and access to Internet. It is assumed that the likelihood of owning a personal computer will have a significant effect on the adoption of Internet banking. Yin, DeVaney and Stahura (2005) investigated the determinants of U.S. household expenditures on computers and found several demographic and socioeconomic factors to be significant predictors of probability of computer purchase, e.g. age, gender, education and income. Indirectly, these factors are expected to affect the probability of Internet banking adoption. A cost-benefit study by Kim, et al. (2005) found that age and education were significant predictors of Internet banking adoption. Their analysis showed that younger and better educated customers were more likely to adopt Internet banking and that the probability of adopting Internet banking decreases with age for all levels of education. Similar findings were reached by studies based on the technology diffusion model (e.g. Chang, 2003). Age, education and income factors, however, are expected to have less effect on technology acceptance as the technology matures (Rogers, 2003).

The descriptive studies reported by the U.S. Bureau of Census show patterns of change in the computer ownership and Internet access for U.S. households. Statistics show that 62 percent of households owned a computer in 2003, and 55 percent of households had Internet access from home. In 2001, about 56 percent of households owned a computer and 50 percent had Internet access. The same trends occurred throughout the last decade. These trends are identifiable regardless of the age of the householder. For example, among households with heads over 65, 28 percent had a computer and 23 percent had Internet access in 2001. In 2003, among older households, 34 percent owned a computer and 29 percent had Internet access. There has been a small increase in the percentages of computer owners for all education groups. For example in 2001, 78 percent of the households who were headed by college graduate had a home computer. In 2003 this percentage increased to 82 percent.

Gender and race may have a significant effect on the adoption of Internet banking. Some of …