|
...under certain elements create obstacles to subsequent entry and allow incumbents to earn rents even when entry occurs. These elements include certain capital investments, such as building a clientele when switching costs are present, learning by doing, or economies of scale.
The extant empirical literature exploring the relationship between the order of entry and a firm's market presence and performance has focussed on differentiated products where innovation is central to product development. In contrast, this paper focuses on a service industry. In particular, within the context of the banking industry, this paper investigates whether early movers have an advantage over later movers by exploring whether the order of entry is related to the degree of market dominance. Various factors could give rise to such an advantage in banking. Both the anecdotal and empirical evidence suggest that consumer switching costs are significant in banking, with some evidence on supply-side factors such as economies of scale as well. The banking industry provides a unique opportunity to study such a phenomenon, with the availability of decades of data that allow us to determine the date of entry and exit of thousands of firms into hundreds of local markets, with information on their market shares, as well as other firm characteristics. Moreover, the data provide substantial variation in dates of entry into a given market, as it is common in banking markets to find banks that entered several decades ago coexist with banks that entered only recently.
While there is no direct research evidence indicating an early-mover advantage in banking, an abundance of anecdotal evidence suggests that, at least in the short run and for large banks, a dynamic disadvantage exists for later entrants. In particular, established banks that expand into new markets appear to perform worse than their competitors, losing deposits, loans, and profits. (1) Some of the reasons put forth for this poor performance include lack of personal contact and loss of personal relationship for the client, "cookie cutter" products that are not tailored to the individual customer's needs, and the banker's lack of knowledge about the local community. Conversely, incumbent banks tend to accumulate proprietary information about their customers and their local communities. Indeed, an early-mover advantage may be more likely in banking than in other industries due to the importance of relationships.
In order to illustrate the advantage of an incumbent, we use a variation of the von Stackelberg two-stage, two-firm model based on Spence (1977) and Dixit (1979), which reinterpret the basic model as a reduced-form derived from short-run product--market competition under capacity constraints. While our analytical framework is quite simple, it embodies the main aspects of sequential entry under sunk costs. In particular, it illustrates how early movers can make their investment decisions strategically and fare better than later movers, by "pre-empting" the market through a higher capital investment, thereby limiting the erosion of rent due to entry. In this model, the incumbent or early mover, who decides how much capital to invest by taking into account how the entrant will react to its choice of capital, ends up with higher capital investment and subsequent profits than the entrant. This captures the incumbent's early-mover advantage. In our setup, capital investment is interpreted, without loss of generality, as building a clientele, with incumbents developing larger clienteles than entrants. In terms of the banking data, the model implies that incumbents should enjoy larger deposit market shares than later entrants.
Our unique data set, which covers the period 1972-2002 and over 10,000 bank entries, allows us to identify the entry and exit date (if any) of all banking firms in urban markets--defined at the level of the metropolitan statistical area--that occurred throughout the three decades, as well as the incumbent firms as of our starting date in 1972. At each point of time, we divide banks into four groups, based on how long ago they have entered the market, ranging from the last 5 years to over 20 years ago. The regression analysis focuses on the period 1992-2002, in order to have banks fall into each category. The data include 318 markets and more than 8,000 bank entries (of the 10,655 entries within 1972-2002, 5,381 occur within 1992-2002; of the other 5,274 occurring within 1972-91, 3,115 of them survive into the 1990s).
A test of an early-mover advantage is useful for many reasons. First, it may help in understanding the type of strategic competition that takes place in the industry. Second, measuring the magnitude of any early-mover advantage is a way to gauge how significant barriers to entry are in the industry, indirectly measuring the factors giving rise to them, such as switching costs and other consumer transaction costs on the demand side, and cost advantages from learning by doing, decreasing average costs and capital accumulation on the supply side. Moreover, banking markets make a great laboratory for the study of an early-mover advantage. Unlike some previous work, which relies on surveys, polls, and various assumptions in order to distinguish pioneer firms, identifying incumbents and later entrants is straightforward in the case of banking markets given the nature of bank entry and the available banking data. In addition, the relevant geographic market in the banking industry, which tends to be local and defined at the metropolitan level, provides a wealth of cross-market variation.
Our results show that banks that enter markets early enjoy larger deposit market shares, after controlling for firm, market, and time effects. The later a bank enters, the smaller is its market share relative to early entrants. While all early entrants have an advantage over later entrants, it is particularly the 1972 incumbents (those firms present at the beginning of our sample that survive into the 1990s) that have the largest market share advantage. While entrants face a disadvantage, they can ameliorate such an effect if they enter by merger as opposed to opening a branch or through a de novo charter. Similarly, geographically diversified entrants face a lower disadvantage relative to more locally limited entrants. Moreover, we find that early entrants appear to have such hold in the market by strategically investing in larger branch networks.
In addition, we are able to distinguish the early-mover advantage from alternative stories. In particular, a learning model, where firms face a self-reinforcing productivity shock every period and discover their type over time, implies an equilibrium where the size distribution of firms is increasing in the age of firm cohorts. Similarly, under a scenario of imperfect capital markets, such that most of the firm's ability to invest and therefore grow are derived from internally generated funds, the firm's assets will be correlated with the firm's age. In order to rule out these alternative stories, we explore whether earlier entrants have larger shares of the market compared to later entrants, after surviving in the market for the same number of years. That is, we explicitly account for the order in which entry has occurred by holding the number of years since entry constant across all banks, and therefore address the potential survivorship bias in our sample. Even when we control for the number of years in the market, we find that a bank's share decreases by 0.1 percentage points for a change in its order of entry from nth to (n + 1)th. Moreover, each additional year in the market increases a bank's deposit market share by 0.01 percentage points, regardless of the order in which the bank entered the market. Thus, this test is particularly powerful as it allows us to separate the effects of market tenure from those of the order of entry. The test indicates that while market tenure increases a bank's market share, the later a bank enters a market, the lower its market share relative to early entrants.
We also offer another test for whether our earlier results are related to a strategic advantage of early entry. If access to capital markets and/or firm learning are the main factors determining a bank's entry and growth in a market, we should find that multi-market banks that are incumbents in some markets but entrants in others do not depict large differences in market shares across these markets. If the bank enters a market and wants to grow to the extent of the incumbent, it should be able to do so. However, this is not what we find. On the contrary, larger, multi-market banks do not achieve in new markets the large market shares that they have in markets where they are incumbents. That is, even these bank entrants do worse relative to incumbents. A bank with "deep pockets" can enter a market and build a large branch network to drive out smaller banks if it wants to--that is, if it believes it would be profitable to do so as it would be able to attract other banks' customers. The evidence in this paper shows that there is a fraction of consumers that will stay with the incumbent regardless, and that is why even a large bank with access to capital and accumulated knowledge might not become as large in markets where it enters later, and branch out in these as much. In fact, compared to the overall sample, the results indicate that the difference between these banks' market shares when they are entrants and those when they reach maturity in the market are larger. This suggests that larger, multi-market entrants grow faster than smaller, single-market entrants.
Moreover, we find that high-growth markets show a smaller difference between late and early movers. These results support the prediction from consumer switching costs models that higher population growth markets should exhibit less of an early-mover advantage. Indeed, they suggest that consumer switching costs are likely an important factor behind the documented market share difference. In low-growth markets, the number of new consumers is low, and therefore entrants should face greater barriers to their market growth.
There are at least two important policy implications of the paper's main finding that entrants cannot compete on an equal footing with incumbents. The first is that a large entrant might not signify as much of a threat to an established community bank. This may help explain why so many thousands of community banks survive when many had predicted their demise--the large banks merging across the nation cannot enter their markets and just take their local market shares away from them. Second, for antitrust analysis, potential entry might not be as effective as a competitive force in a market, especially one with low-population turnover. Thus, regulators, who define potential entry as a "mitigating factor" for the possible anti-competitive effects of a merger, should adjust for the fact that entrants may not be able to fully compete head to head with incumbents. We discuss both of these implications in the paper.
The paper is organized as follows. Section 1 briefly reviews the literature on the advantage of early movers, while Section 2 provides a discussion of how such an advantage might arise in banking as well as the prior empirical evidence on this issue. Section 3 describes the empirical framework. Section 4 presents our empirical results, and Section 5 concludes.
1. EARLY-MOVER ADVANTAGE: THEORY AND EVIDENCE
An early-mover advantage might arise under various demand and supply conditions. 2 On the demand side, factors such as switching costs, network externalities, and buyer inertia due to quality uncertainty and/or habit formation can result in an early mover having an advantage over later entrants. The supply-related factors include setup and sunk costs, scale economies, supply chain, and learning by doing.
All of these factors have been studied under various setups. (3) However, the sequential incumbent entrant games that explicitly model the first-mover advantage are few in the literature. One insightful paper in this respect is that by Schmalensee (1982) who studies the long-lived advantages to pioneering brands when buyers face imperfect information about product quality. In a two-period model, two brands enter sequentially, and while being identical, the second brand has a disadvantage because customers have already tried the first brand in the first period and found that it is satisfactory.
Within the economics literature, empirical work on the advantage of early movers, which has mostly been interpreted as a market share advantage, has focused on the pharmaceutical industry (Grabowski and Vernon 1992, Hurwitz and Caves 1988, Gorecki 1986), with some other applications such as those to financial innovation (Tufano 1989) and Internet search engines (Gandal 2001). Grabowski and Vernon (1992) analyze 18 drugs experiencing generic competition upon patent expiration by exploring price and market share patterns of incumbents and entrants, generally finding, as the earlier literature, that pioneers maintain premium price positions after entry while their market shares erode over time. Tufano (1989) finds a strong relationship between product innovation and investment banks' market shares using a database of 58 financial innovations. Gandal (2001), based on a period of 1 year, uses a discrete choice logit model to find that early entrants in the Internet search engine market enjoy a consumer utility premium.
There is a significant body of evidence from marketing research on pioneer and early-entrant advantages for non-financial firms. This work, however, is based on surveys or data that are restrictive and present problems related to the levels of industry aggregation and distinguishing the order of entry, among others. (4) Examples are Robinson and Fornell (1985) on consumer goods and Robinson (1988) for industrial goods.
2. THE EARLY-MOVER ADVANTAGE AND THE BANKING INDUSTRY
In the banking industry, demand-driven factors are likely to be important in giving rise to an early-mover advantage. (5) Both the anecdotal and empirical evidence suggest that consumer switching costs are significant. The evidence on supply-side factors, such as economies of scale, is mixed, showing both economies and diseconomies of scale in bank products.
Switching costs appear to be prevalent in the use of banking services. (6) Anecdotal evidence suggests that depositors find it costly to close an account with their current bank to open an account in another bank. Time is invested in doing so, funds may be tied in the process, and the new service might require some specific investment in learning to use it. Other switching costs might include uncertainty over the quality of service, such as branch service quality, product availability, and even how long it takes to get through the phone system to a customer representative. (7) Customer inertia is likely to be such that in order for a consumer to switch banks, at least one of the following should occur: current service deteriorates relative to expected new service at another bank and this deterioration is enough to cover switching costs; large discount offered by another bank; some other large expected gain from switching (e.g., new technology that allows new products that the current bank is not offering yet, such as Internet access).
The empirical evidence suggests that customers stay with the same bank for a long time largely due to the cost of switching banks. Based on survey data, Kiser (2002a) finds that the average household stays with the same bank for 10 years, while the most frequently cited motivation for changing banks is a household relocation. Moreover, she...
NOTE: All illustrations and photos
have been removed from this article.
More articles from Journal of Money, Credit & Banking
Monetary and fiscal policy switching., June 01, 2007
Precautionary balances and the velocity of circulation of money., June 01, 2007
Monetary policy rules in a New Keynesian Euro area model., June 01, 2007
Unique symptoms of Japanese stagnation: an equity market perspective., June 01, 2007
Understanding the large negative impact of oil shocks., June 01, 2007
Looking for additional articles?
Search our database of over 3 million articles.
Looking for more in-depth information on this industry?
Search our complete database of Industry & Market reports by text, subject, publication
name or publication date.
About Goliath
Whether you're looking for sales prospects, competitive information, company
analysis or best practices in managing your organization,
Goliath can help you meet your business needs.
Our extensive business information databases empower business
professionals with both the breadth and depth of credible,
authoritative information they need to support their business
goals. Whether it be strategic planning, sales prospecting,
company research or defining management best practices -
Goliath is your leading source for accurate information.
|
