Cash and Accrual Accounting in a Randomized Field Experiment

How to cite

Muhn, Maximilian; Breuer, Matthias; Bonetti, Pietro. 2026. "Cash and Accrual Accounting in a Randomized Field Experiment." AEA RCT Registry. AEARCTR-0018516. https://doi.org/10.1257/rct.18516-1.0

Download: BibTeX · RIS — for reference managers without a browser translator. Zotero and most others pick up this page's citation metadata directly.

Maximilian Muhn* , Matthias Breuer , Pietro Bonetti

* Primary investigator

Registry ID
AEARCTR-0018516
DOI
https://doi.org/10.1257/rct.18516-1.0
First registered
Last updated
Status
In development
Trial dates
2026-06-06 – 2027-12-31
Countries
Private
JEL codes
G21, M41, O16, G32, D83, C93
Registry record
http://www.socialscienceregistry.org/trials/18516

Abstract

We intend to examine the decision usefulness of cash- and accrual-based performance reports in a large-scale randomized field experiment. Specifically, we intent to investigate the effects of providing bank-branch managers with periodic reports featuring cash- or accrual-based summary measures of their branch’s loan-portfolio performance on loan-portfolio characteristics (size and riskiness).

Intervention

The intervention introduces periodic branch-level performance reports into an existing cloud-based accounting system. The reports summarize recent lending activity at the branch level and provide comparisons with prior months.

Experimental design

We conduct a randomized field experiment with bank branches using a cloud-based accounting system. The experiment is designed to generate random variation in branch-level reporting and in the accounting basis of the reports.

Design details

Randomization method
Branches are randomly assigned by computer to one of three conditions: cash-based report, accrual-based report, or control. Randomization is conducted within banks where feasible, including stratified randomization, and tries to minimize differences in pre-treatment covariates.
Randomization unit
Bank branch.
Clusters
Depends on the number of enrolled institutions (and their number of branches). Currently planning with three (initial institutions) and a sample of up to 800 branches. Any additional institutions will increase the sample size and we strive for the largest size possible.
Observations
Same, only more granular if loan-level data can be extracted (currently aggregated to branch-level obs); can be observed over time.
Arms
Planning 1/3, 1/3 and 1/3.
Primary outcomes
Branch-level loan-portfolio outcomes.
Secondary outcomes
Branch managers’ information acquisition and engagement with the system.

Cited by

No citation data available for this registration yet.

Keywords