An Examination of the Performance and Prospects for the Future of Internet-Primary Banks

Article excerpt

Abstract We study Internet-primary banks for differences in performance versus newly chartered traditional banks from 1996 through 2003. Internet-primary banks are larger, with lower net interest margins and loan losses, and higher wage and physical capital rates than newly chartered traditional banks. Univariate ROE is lower, but profit efficiency is significantly higher on average for Internet-primary banks compared to all newly chartered banks as well as those that survive through the sample period. In multivariate models, Internet-primary banks continue to have significantly higher average profit efficiency, and particularly so when of sufficient size.

Keywords Banking * Profit Efficiency * Internet Banking

JEL codes G21 * G20

1 Introduction

The Internet has altered the way in which many industries conduct business. Banking is no exception as technology and innovative thinking have changed the design and delivery of banking services (e.g., Egland et al. 1998). There are two general business models to provide Internet banking: "bricks and clicks" and "Internet-primary" banks. The "bricks and clicks" model utilizes traditional brick-and-mortar offices, supplemented with the Internet, similar to a firm with a physical market presence, such as Barnes and Noble, also operating a website where products can be purchased. Most banks using the Internet have chosen this model to provide increased customer service or product-enhancement rather than for primary product delivery. In contrast, the "Internet-primary" bank model is an electronic, virtual bank having little or even no branching system (bricks) and relying on others' physical networks to meet the majority of customers' physical transaction needs.

The basic premises behind the Internet-primary model itself are questioned by many (see De Young 2001, for a more complete discussion). The main cost advantage would be the absence of investment in brick-and-mortar locations and the ongoing personnel expense of staffing those branches. There are potentially large economies of scale on the revenue side because bank customers are not limited geographically and average computer system costs decline rapidly as customers are added. Offsetting these possible advantages would be greater technological investment along with higher salaries for more qualified technical employees as compared to the average bank teller.

We examine Internet-primary bank performance using profit efficiency rather than accounting ratios. Profit efficiency allows for a better understanding of the underlying economics of using the Internet as the predominant product delivery platform. Profit efficiency indicates how well management produces outputs for a given input mix along with other market characteristics and is measured as the distance from the best-practice frontier. This method avoids the well-known limitations of accounting performance - which is particularly important across different business strategies. If Internet banks are more profit efficient than traditional banks, those that survive the typical poor performance of start-up banks as well as the challenges of on-line product delivery could be positioned to be formidable competitors. Additionally, our analysis enhances the understanding of Internet-primary bank expansion through exploiting the technology-based scale economies as identified by De Young (2005) and Delgado et al. (2004).

On average, we find Internet-primary banks are larger, with lower net interest margins, ROE, and loan-loss ratios, and higher wage and physical capital rates than newly chartered traditional banks.1 Lower accounting performance is typical in prior findings (e.g., Furst et al. (2002); De Young (2005)) and we confirm that these results remain for our sample which spans a longer time period. In contrast to our own accounting performance univariate results as well as De Young (2005) and Delgado et al. (2004), Internet-primary banks exhibit significantly higher performance when measured as profit efficiency compared to all newly chartered brick-and-mortar banks in both the total sample and a survivor sample. …