Take On Payments

In Shakespeare's "A Midsummer Night's Dream," Helena said, "Though she be but little, she is fierce," in reference to the power of her romantic foe, Hermia. In today's pop culture, this quote can be found on T-shirts, coffee mugs, inspirational wall hangings, and social media memes touting women's power. But it has a broader meaning to me, one that says small voices are every bit as important as large ones.

In the payments industry, I think of the small voices as being the smaller financial institutions—which are crucial to the success of the Federal Reserve Payments Study, contributing a great deal to the study findings. The study, which estimates the number and value of noncash payments made by U.S. consumers and businesses as well as the data around payments fraud, is intended to inform policymakers, the industry, and the public about aggregate trends in the nation's payments system. Most recently, this work culminated in a benchmark report on U.S. payments fraud from 2012 to 2016.

One important component of the study is to collect data on checks, ACH, wire transfers, cards, cash withdrawals and deposits, third-party fraud, and related information from a nationally representative sample of commercial banks, savings institutions, and credit unions, from the largest to the smallest. So what exactly is meant by a "nationally representative sample"?

In a nutshell, for our estimates to be representative of national payment volumes, we have to account for all sources of volume. If we include only the largest institutions or leave out some segments, the estimates can be biased, either too large or too small. Even though much of payments volume is concentrated in the largest institutions, it is impossible to know how much so without having a good estimate for all segments of the banking population. Past surveys have shown that the segments can exhibit very different trends from study to study. For example, from 1995 to 2000, total checks at large commercial banks fell, while total checks at credit unions and savings institutions grew. (Read more about that in this report from the Federal Reserve Board of Governors.) Without the information from credit unions, the decline in checks would have appeared larger than it actually was.

Study participants are selected from among U.S. commercial banks, savings institutions, and credit unions. According to reports filed with the Federal Reserve in 2015, there were approximately 10,600 of these depository institutions (DI) in the United States that met the criteria (see the table). Using Call Report data filed with the Federal Reserve, a sample frame of slightly under 3,800 institutions was determined to be representative of the entire population of U.S. financial institutions. Each institution type is further grouped according to deposit size.

*For commercial banks and savings institutions, this is the sum of public checkable deposits (or checking account balances) and money market deposit accounts. For credit unions, this reflects public checkable deposits only.

Source: Table adapted from Geoffrey Gerdes and Xuemei Liu. "Improving Response Quality with Planned Missing Data: An Application to a Survey of Banks," in The Econometrics of Complex Survey Data: Theory and Applications (Advances in Econometrics, volume 39), ed. Kim P. Huynh, David T. Jacho-Chavez, and Gautam Tripathi. Available April 1, 2019.

As the table shows, financial institutions in each category with the lowest maximum deposit size comprise approximately 46 percent of the total number of U.S. institutions. Of this group, consisting of more than 4,800 DIs, just under 700 were invited to participate in the study, or approximately 18 percent of the total sample.

Take, for example, credit unions with a maximum deposit size of $11.2 million. In 2016, there were approximately 3,100 institutions in this category, and 242 were invited to participate in the study to represent that segment. Similarly, 96 savings institutions with a maximum deposit size of $46.3 million were selected to represent the overall segment of just under 300 institutions.

Grouping institutions in this way improves the quality of results, as the institutions within each category share many similar characteristics. The smaller institutions have a unique voice and experience that the larger DIs cannot represent. To develop a true and accurate national picture of the payments landscape, it is important that all voices be heard.

I hope your takeaway from this post is that the contributions of all financial institutions—large and small—are important to the accuracy and representativeness of the data that the Federal Reserve Payment Study reports. And although study participants may sometimes think their institutions are small fish in a big pond, their survey contributions serve as the voice of their peers, and in the collective, that whisper becomes a mighty voice.

  By Nancy Donahue, project manager in the Retail Payments Risk Forum  at the Atlanta Fed