Ebook The Effects of Card Level and Bank Level Benefits on Credit Card Interest
Credit card interest rates are higher than other consumer credit interest rates all over the world. Empirical evidence indicates that credit card interest rates are also downward-sticky and show asymmetric response to the changes in the cost of funds. In a seminal paper, Ausubel (1991) showed that although there were about 4,000 banks in the US credit card market, the response of credit card interest rates to the decline in the cost of funds was very slow in the 1983-1987 period. Moreover, he calculated that banks earned three to four times the ordinary banking industry returns from their credit card business. High and sticky interest rates in credit card markets led to a shift of interests from price to non-price competition in these markets. While price competition fails, there is ample empirical and anecdotal evidence that supports the existence of a fierce non-price competition in credit card markets.
Credit card markets in developed countries have been extensively explored. However, very little research has been conducted for developing countries in spite of the recent surge in credit card markets in these economies. Among the very few, Aysan and Muslim (2006) and Aysan and Yildiz (2007) show the failure of price competition in the rapidly growing Turkish credit card market. These studies reveal that the response of credit card interest rates to the decline in the cost funds is economically insignificant. In the current paper, we analyze the nature of non-price competition in the Turkish credit card market, introducing the first evidence of this kind for an emerging market. Analyzing the non-price competition is important and necessary in order to design and implement effective regulations for credit card markets. Central Bank of Turkey has been applying a price ceiling on credit card interest rates since June 2006 in accordance with the recently enacted credit card law. However, the rates still remain exceedingly high compared to other loan rates. Tightening of the price ceiling is on the agenda of the government. Any incorrectly designed regulation may have economy-wide adverse effects since increasing credit card numbers and transaction volumes made credit cards crucial for the functioning of the economy in recent years.
This paper argues that the main reason for extremely high credit card interest rates in Turkey is the low price elasticity of demand for credit cards. Low price elasticity of demand stems from the high switching costs of cardholders. Banks set up and increase switching costs by providing a number of non-price benefits and thus creating captive customers. All the non-price benefits provided to a card holderby her card-issuing bankas part of its product differentiation strategies can be broadly considered as switching cost for the card holder, since she has to forego these benefits if she decides to switch to another card. We divide these non-price benefits offered to credit card customers into two groups: card level benefits that depend on credit card usage, and bank level benefits arising from the quality of general services and characteristics of the issuer bank. Card level benefits of using a credit card include gaining money-points, frequent flyer miles and rewards, being able to pay shopping bill in installments, and taking advantage of the conveniences of online shopping. The quality of general banking services and bank characteristics are important for credit card choice since many cardholders use other services of the issuer banks as well. Anecdotal evidence also suggests that customers obtain credit cards from the banks at which they already have accounts. Using multiple services of a bank increases the cost of switching for cardholders, especially if they are pleased with the quality of these services.
The paper aims to establish the role of switching costs in explaining the high and sticky rates in Turkish credit card market and to identify the instruments through which banks create these costs. An empirical model is developed to examine the relationship between credit card interest rates and card level and bank level measures of non-price benefits to credit card customers. We benefit from the switching cost and bank pricing models of the existing literature. The duopoly model of competition with consumer switching costs in Stango (1999) and the bank pricing models proposed by Neubergen and Zimmerman (1990) and Hannan (1991) provide guidance for choosing the relevant variables in the empirical model. Three groups of explanatory variables are used. The first group encompasses cost variables: the cost of funds, default risk and liquidity risk. In the second group, we use the number of bank branches, capital ratio, and average salaries to account for the bank level non-price benefits of credit cards, which are the quality and characteristics of general banking services. The third category includes card level benefits like money points, travel miles and installments which are proxied by the market shares of issuers. This paper is the first to utilize a recently constructed quarterly panel data set for all 22 issuers in the credit card market in Turkey which spans the period from the last quarter of 2001 to the second quarter of 2006. In line with most switching cost models, fixed and random effect regressions yield a significant and robust positive relationship between switching costs and prices in the credit card market in Turkey, confirming that as the measures of both groups of non-price benefits increase, banks charge higher credit card rates. Our results also support the hypothesis that credit card interest rates adjust to the changes in the cost of funds sluggishly even after controlling for the non-price features. These results are robust to econometric specification and methodology.
The organization of the paper is as follows. In the next section, we examine the non-price competition in the Turkish credit card market. Empirical and theoretical background is presented in Section 3. Section 4 builds the empirical model and explains the data and variables. Results and robustness test are presented in Section 4. Finally, Section 5 concludes.
Download
PDF Ebook The Effects of Card Level and Bank Level Benefits on Credit Card Interest
Posted in :
