Payment Convergence in the EMU

The primary goal of this research is to analyze shifts in European payment behaviors, focusing on the impact of the Single Payment Area (SPA) within the European Monetary Union (EMU). The study examines how factors like financial opening, regulations, and technological innovation influence the demand for various payment instruments: cash, cards, checks, credit transfers, and direct debits. A model of conditional convergence is developed and tested using instrumental variables on annual panel data. The research specifically aims to measure the convergence of payment behaviors across EMU countries, assessing the significance of major integration steps. A key finding is the convergence in demand across payment methods, with the notable exception of checks.

The study's novel approach lies in its inclusion of regulations and financial opening as key explanatory variables influencing payment instrument demand changes within the EMU. The researchers posit that the observed alterations in payment behaviors result from the combined effect of the SPA (unifying national retail payment markets), technological advancements, and socio-macroeconomic factors. The creation of a comprehensive legal framework by the European Commission and the integration of payment structures by banking associations are believed to harmonize banking practices and promote convergence in payment behaviors. The study emphasizes that more open and internationally harmonized payment markets allow for quicker adoption of foreign technologies and payment habits. Therefore, the research aims to determine the degree of this convergence process by highlighting the influences of financial opening, banking integration, and the harmonization of regulations within the European Union.

The study acknowledges that while the SPA aims for secure and efficient payment transactions of any size across the EU, similar to national-level efficiency, this has not been fully achieved. The introduction of the Euro in 2002 did not automatically resolve issues such as longer processing times and higher costs for cross-border payments compared to domestic transactions. Significant obstacles remain in accessing financial services across EU borders. Since the 1990s, various legal measures (regulations, directives, recommendations) and banking agreements have attempted to remove these barriers, with the creation of the European Payment Council (EPC) and the establishment of a common legal framework by the European Commission as notable examples. These measures focused on eliminating the 'border effect' for credit institutions and ensuring consumers have access to payment services with a competitive quality-to-price ratio across the EU. However, the success of the SPA will also depend on consumer willingness to adopt new payment methods and transparency in pricing of banking services.

The document presents case studies of various European countries to illustrate diverse national payment habits and the impact of regulations. In Finland, despite high electronic payment usage, cash remains significant, with factors like the decline in ATMs and bank policies encouraging card usage playing a role. France exhibits high check usage due to a regulated guarantee scheme and banking policies. The UK and Ireland show increased cash and check use, partly due to a lack of investment in electronic payment infrastructure. Germany showcases a preference for cash, linked to historical factors and high minimum transaction values for credit cards. The study uses these national examples to highlight the differences in payment preferences and infrastructure across the EU and how these relate to the effectiveness of regulations aimed at creating the SPA. A key example is Regulation (EC) 2560/2001, which aimed to equalize prices for cross-border payments, excluding checks.

The research examines specific regulatory initiatives within the EMU and their impact on payment behavior. The start of the second stage of the EMU (January 1994) under the Maastricht Treaty increased the liberalization of capital movements and payments, potentially boosting cross-border payment instruments. The third stage (January 1999), involving the Euro's introduction and the TARGET payment system, enhanced financial information transparency and payment system integration, potentially influencing foreign supply on domestic payment instrument demand. However, the study notes that existing regulations primarily target cross-border payments and don't fully address harmonization across jurisdictions. Variations in payment order revocability and other legal aspects across EU countries illustrate the need for further harmonization to leverage the full benefits of the SPA. Regulation on cross-border payments also addressed the problem of high fees for non-resident bank accounts held abroad. The study notes the importance of standardization and accessibility of payment instruments in promoting convergence.

This study uniquely incorporates 'financial opening' as a key explanatory variable to understand shifts in payment instrument demand. Unlike previous research, it considers the integration of payment systems, measured by the total SWIFT (Society for Worldwide Interbank Financial Telecommunication) messages sent and received. While acknowledging that SWIFT message volume isn't a perfect measure of financial opening, the researchers chose it due to data availability. The rationale is that a high degree of integration leads to harmonized payment system regulations, potentially encouraging the convergence of payment behaviors. The study hypothesizes a positive correlation between financial opening and the demand for payment methods that facilitate cross-border transactions, reflecting increased international business and the use of international payment methods due to the establishment of bank branches in foreign countries.

The study examines the impact of regulations on payment instrument demand, highlighting specific regulatory initiatives. The introduction of the single currency (Euro) in 1999 (REG99) is analyzed, along with the second stage of the EMU in 1994 (REG94) which marked the full liberalization of capital movements and payments. The Directive 97/5/EC on cross-border credit transfers (REG_CTD) is also considered. The researchers anticipate a positive correlation between these regulatory variables and the demand for payment methods offering faster, cheaper, and more reliable cross-border transactions. They expect REG94 to increase the use of cash, cards, and credit transfers, while REG_CTD should boost credit transfers by ensuring efficient processing. REG99 is hypothesized to increase cash usage due to lower exchange fees and, indirectly, card usage via ATM withdrawals abroad. However, the research acknowledges that these regulations and results are preliminary, with further harmonization needed to enhance price transparency and banking competition.

Technological factors are incorporated into the model, focusing on the availability of Automated Teller Machines (ATMs) and Electronic Funds Transfers at Point Of Sale (EFTPOS) terminals. These variables measure technology's influence on cash and card usage. The study distinguishes itself from previous works (GL and HPV) by not using these technological indicators as explanatory variables for other payment instruments due to data limitations. ATMs are expected to increase cash usage by reducing transaction costs, potentially leading to lower average cash balances and increased card use for withdrawals. EFTPOS terminals are expected to primarily increase card use rather than cash use. The exclusion of technological indicators for other payment methods due to data limitations reflects a practical constraint on the study's scope.

The study includes macroeconomic variables—interest rates and private consumption—to assess their influence on payment instrument demand. Interest rates are found to have a significantly positive coefficient for all payment instruments in value terms, generally aligning with the Baumol-Tobin (BM) theory suggesting an inverse relationship between interest rates and cash holdings. A rise in interest rates is expected to reduce cash usage and increase cashless payments. However, the observed positive relationship between cash demand and interest rates is unexpected and attributed to relatively low interest rates reducing the opportunity cost of holding cash in deposit accounts. Private consumption shows a significant negative coefficient only for credit transfers in volume, indicating that increased private consumption (affecting consumer and business spending) may impact other retail sectors more.

The study delves into the concept of convergence, a crucial issue in empirical growth literature. It discusses different econometric methodologies used to assess convergence, referencing the neo-classical growth model of Solow (1956) and the work of Barro and Sala-i-Martin (1992, 1995) on β-convergence (where growth rate is negatively correlated with the initial level of the variable). Both absolute (unconditional) and relative (conditional) β-convergence are considered. The study acknowledges criticisms of β-convergence tests, such as susceptibility to Galton's fallacy and the issue of non-stationarity in time series. It explains how the use of dynamic panel data addresses these problems, allowing control for unobservable country-specific effects and handling the correlation between growth rate and initial variable level. The study also introduces σ-convergence, which analyzes the evolution of dispersion within the sample using standard deviation, highlighting Quah's (1993) observation that β-convergence is necessary but not sufficient for σ-convergence.

The research employs a dynamic panel data model with fixed effects to analyze convergence in payment instrument demand. This model accounts for two sources of persistence: autocorrelation from the lagged dependent variable and individual effects representing country heterogeneity. The study explains why ordinary least squares (OLS) and standard fixed effects estimators are unsuitable due to autocorrelation bias and inconsistency. Instead, it utilizes the Arellano-Bond (1991) generalized method of moments (GMM) estimator. This GMM approach leverages instruments based on orthogonality conditions between lagged dependent variable values and errors. The choice of instruments is carefully discussed, considering strictly exogenous and predetermined variables. The study justifies its decision to avoid using too many lags as instruments to prevent poor performance in small samples due to over-identification restrictions. The lagged dependent variable is included to capture the persistence of past payment behaviors, helping to manage multicollinearity while preserving information.

The study presents the results of its convergence tests, showing evidence of both conditional β-convergence and σ-convergence in volume and value for all payment instruments except checks. This aligns with the expectation that payment instrument demand would follow the European harmonization efforts initiated by the SPA. The divergence for checks is attributed to their primarily national nature, different legal rules across countries, lack of cross-border infrastructure, and high charges. The results also shed light on competition and substitution between payment methods, revealing limited competition except for a two-sided competition between cards and checks, and cards and credit transfers. The lack of competition between cash and cards is deemed surprising given their substitutability and linked to the existence of minimum transaction thresholds for card payments. The study also observes that the convergence process is faster for cash, attributed to its flexibility and fewer technological and administrative constraints compared to other cashless methods. The impact of habit on payment behavior, while present for some instruments, is less significant than the influence of other variables.

Analysis of substitution effects reveals surprisingly limited competition between different payment instruments. While some expected competitive relationships exist (e.g., between card and check usage, and card and credit transfer usage), the overall picture suggests less intense rivalry than might be anticipated. The study observes two-sided competition between cards and checks (in terms of volume), and between cards and credit transfers (also in volume). However, a non-reciprocal competition exists between credit transfers and checks, indicating that the decline in check use is largely driven by credit transfer competition. The competition between cards and credit transfers is unexpected, considering cards are mainly used for face-to-face transactions, and credit transfers for remote payments. This finding is linked to the high rate of electronic payment adoption in Scandinavian countries. These findings contrast with previous research (HPV) that indicated competition between all payment methods except debit cards.

The study notes an unexpected lack of competition between cash and card payments, despite their close substitutability. This result is attributed to the existence of a 'de minimis' threshold for card payments at Point of Sale (POS) terminals. This minimum transaction value for card payments shifts the competitive dynamic, favoring cash for low-value transactions. This minimum amount threshold may explain why the shift from cash to newer electronic payment methods has been slower than predicted. However, the study does find a positive, albeit low (0.12), elasticity of demand between cash and cards, a result similar to that found by Guariglia and Loke (GL). For other payment instruments, the study finds insignificant coefficients, indicating a lack of measurable competitive relationships.

The analysis explores the competition between checks, direct debits, and credit transfers, emphasizing the distinct characteristics of these instruments. The absence of competition between direct debits and checks is explained by the differing initiation points of the funds transfer. Direct debits are initiated by the beneficiary, while credit transfers are initiated by the debtor directly to their payment service provider without involving the beneficiary. This difference in payment initiation signifies a lack of direct substitution between these two payment methods. Similarly, the lack of competition between direct debits and credit transfers is explained by the fact that credit transfers are initiated directly by the debtor. The study notes that these instruments offer unique functionalities, limiting their direct substitutability for certain transactions. This explanation is useful in interpreting the study's findings and offering insights into the dynamics of payment instrument selection beyond simple competition.

The study observes that the convergence process is significantly faster for cash payments than for other instruments, which is attributed to cash's flexibility and fewer technological and administrative constraints. Unlike cashless instruments requiring specific technologies (EFTPOS, giro ATMs, internet connections), cash transactions are less constrained, resulting in lower transaction costs and increased convenience that fuels its usage. Regarding the persistence of payment habits, the study contrasts its findings with those of GL. While GL found a significant positive impact of habit on all payment instruments, this study shows that habit significantly influences only cash (in volume and value), checks (in volume), and direct debits (in value). Even then, the impact of habit is relatively low compared to other variables, indicating a low degree of persistence and suggesting that other factors are more influential in shaping current payment choices.

The study incorporates interest rates as a macroeconomic variable to examine their impact on payment instrument demand. The results show a significantly positive coefficient for interest rates across all payment instruments in terms of value. This generally aligns with the Baumol-Tobin (BM) theory, which posits an inverse relationship between interest rates and cash holdings. Higher interest rates are expected to decrease cash usage and increase cashless payments, as the opportunity cost of holding cash increases. However, the study observes an unexpected positive relationship between cash demand and interest rates. This counterintuitive finding is attributed to the relatively low interest rates observed during the study period (1990-2001), which reduced the lost interest earnings from holding cash in deposit accounts, potentially making holding cash more attractive even with slightly higher interest rates.

The study also includes private consumption as a macroeconomic variable. The results reveal a significant negative coefficient for private consumption only on the volume of credit transfers. This finding suggests that increased private consumption, which includes both consumer and business spending, might have implications for other sectors of the retail industry. The study notes that growth in household final consumption expenditure is typically directed towards consumable goods and services (leisure, travel, entertainment, healthcare) that are often paid for using cash, cards, or checks. However, the coefficients for these payment methods are not significant, limiting the study's ability to draw firm conclusions regarding the specific mechanisms at play. This suggests further exploration is needed to fully interpret the impact of private consumption changes on the payment method choice.

The study's most significant finding is the convergence of payment behaviors within the European Monetary Union (EMU). This convergence, observed across various payment instruments except checks, is not primarily attributed to shifts in consumer preferences but rather to the dynamics of product standardization at a European level. The research suggests that the observed convergence reflects the broader trend of retail banks converging through standardized products and services, rather than a fundamental change in consumer payment habits. This standardization is driven by both the banks' intrinsic motivation to leverage competitive opportunities arising from banking integration and regulatory incentives to create a common legal framework for retail payments. The study's conclusion challenges the conventional view that payment habits are slow to change, suggesting that changes can be accelerated by the convergence of retail banking markets and legislative changes.

The observed convergence in payment behaviors is strongly linked to the standardization of products and services within the retail banking market. The study highlights that consumers often lack a deep understanding of the complexities of the retail market, including the intricacies of bank fee structures and contract terms. This lack of understanding, combined with high transaction costs and switching costs (due to non-portability of account numbers), hinders consumer mobility and limits their ability to readily switch banks in search of more attractive retail payment services. Therefore, the convergence is attributed to standardization stemming from both providers' competitive motives and regulators' efforts to establish a harmonized legal environment for payments. The ongoing efforts by regulators and the banking industry to eliminate non-market distortions (price controls, taxes, subsidies) suggest that the convergence trend will intensify with the full achievement of the Single Euro Payment Area (SEPA), diminishing the influence of cultural differences on payment preferences.

The study's findings challenge the notion that payment habits are inherently resistant to change. By incorporating novel explanatory variables (regulations, financial opening, banking integration), the research demonstrates that legislative modifications and market reconfigurations can significantly impact payment behaviors, potentially at a faster rate than previously assumed. This suggests that future research should focus on analyzing the interplay of these factors to gain a more nuanced understanding of the dynamics shaping payment choices. The observed convergence suggests that policy efforts to promote standardization and reduce non-market distortions within the retail banking sector can play a vital role in accelerating the harmonization of payment behaviors across the EU and achieving the full potential of the SEPA. Future studies could further explore the detailed aspects of price transparency and their impact on the competitive landscape of financial services.