Replication Folder for E-money: The Role of Payment Networks and the Transaction Fee
Description
This replication folder contains the relevant programs and calculations/data used in the paper mentioned above. These codes and calculations/data were used to produce the paper's figures and tables. One may use them to replicate the paper's findings. MATLAB 2025a or higher (with Econometrics and Optimization toolboxes) is required to run these codes. This paper develops a micro-founded monetary model in which a profit-maximizing monopolist payment network endogenously determines transaction fees alongside agents' portfolio choices between cash and e-money, which is more secure than cash. There are two equilibrium regimes with e-money: a mixed equilibrium, where both monies circulate, and a pure e-money equilibrium. Threat of entry from a competing network constrains the monopolist's fee, significantly shrinking the monetary policies and parameter values that support the pure e-money equilibrium. Calibrated to 2024 U.S. data, the model yields a non-trivial monetary policy pass-through: a one percentage point increase in the nominal interest rate raises the equilibrium transaction fee by 0.13 - 0.35 percentage points. Despite competitive pressure from a potential entrant, e-money equilibria consistently deliver welfare losses relative to pure cash, driven by a holdup problem and the distortionary proportional fee. Replicators are requested to read the README file first. Data sharing: No primary data was used. Secondary data is public; however, for replicators' convenience, it is included in the relevant folders/subfolders (described below) of the replication folder. Analysis data: All analysis data has been provided in the replication package. Calculation of average transaction fee for card payments and transaction fee for Bankgirot is included in relevant spreadsheets as explained below. Non-public data: No non-public data has been used. Helcim's data is from its website (open to the public). However, screenshots have been provided in the relevant folder. eMarketer data is from the company's website (open to the public). Code sharing: All codes used in this project are available in the replication folder. All programs are cleaned, well-commented, and well-formatted. They should produce the exact same figures, tables, or calibration as in the paper if executed on MATLAB 2025a or higher according to the instructions provided in the ## Calibration and List of Figures, Tables and Programs section of the README file.
Files
Steps to reproduce
Replicators are requested to read the README file. The README file includes instructions that guide replicators to the relevant folder/subfolder, programs, spreadsheets, information, and data.
Institutions
- Krea UniversityAndhra Pradesh, Sri City Pvt Ltd-SEZ