RDP 2009-04: Price Incentives and Consumer Payment Behaviour 1. Introduction

Over the past decade, payment patterns in Australia have changed substantially. In line with most other countries, electronic means of payment, such as credit cards and debit cards, have grown strongly, while the number of personal cheques written per person has declined significantly (Figure 1). Among electronic payment methods, however, there have been some interesting trends and trend reversals. For example, credit cards enjoyed remarkable rates of growth in the late 1990s, eventually overtaking debit cards as the most common form of non-cash payment. Subsequently, their rate of growth has tailed off and, more recently, there has been a switch back towards debit cards as the most common form of non-cash payment.

Figure 1: Number of Non-cash Payments per Capita

These changes in payment method use over a period of substantial reform provide prima facie evidence that a relationship exists between price incentives and the use of particular payment methods. The aggregate data, however, are not sufficient to make more than tentative conclusions about this relationship. Data on the payments behaviour of individuals offer the prospect of obtaining a more nuanced and accurate view of the influences on payment patterns.

As part of its review of its card payment reforms, the Reserve Bank of Australia collected detailed data on the transaction behaviour of a broadly representative sample of 662 Australians over a two-week period in 2007. These data provide the first comprehensive transaction-level study in Australia of payment methods used by individuals, including, importantly, cash. In this paper, the data collected from this study are used to estimate econometric models of the holding of credit cards, and the use of credit cards, debit cards and cash. Similar to Bounie and François (2006), our dataset allows us to not only control for demographic influences, but also for transaction-related characteristics, such as merchant type and transaction value. Further, of particular interest to this study, our dataset allows us to identify price incentives for credit cards: whether the consumer participates in a loyalty program and whether the consumer typically obtains the benefit of the interest-free period.

We use these data to examine a number of aspects of consumer payments. In particular, we consider what effect loyalty programs have on the choice of payment instruments and examine the patterns of substitution between credit cards, debit cards and cash as a result of these programs. We conduct similar analysis on the effect of access to an interest-free period.

The remainder of the paper is structured as follows. In Section 2, we provide a brief overview of the aggregate trends in the Australian payments system over the past decade, as well as some background on the Reserve Bank's reforms. Section 3 examines previous studies related to our paper. Section 4 discusses the data we use and sketches some features of the data. Section 5 outlines the modelling framework and discusses the results, particularly in relation to price incentives. Section 6 concludes.