Exploring Check Use by Businesses and Consumers

Federal Reserve Bank of Atlanta Talk About Payments Webinar

Exploring Check Use by Businesses and Consumers

Transcript

Jean Roark: Hello, and welcome to the Federal Reserve Bank of Atlanta's Talk About Payments webinar. Our topic today is the 2018 Check Sample Survey findings. I'm Jean Roark from the Federal Reserve Bank of St. Louis, and I'll be your moderator. Before turning the call over to the speakers, I'd like to run through some logistics on slide two.

If you haven't joined us through the webinar yet, click the link you received after registering. For the best webinar experience, use the FAQ document, which can be found using the Materials button in the webinar player page—but I'll highlight a few important notes for you.

You can listen to the audio through your PC speakers or through the phone. If you use the phone option, slides will not sync with audio unless you change just one setting. You can do that by selecting the gray gear located on the upper right corner of the slide window, just above the presentation. From there, you should see a few options in the media chooser, and you can select the phone option. You'll only want to select that phone option if you're actually listening to the audio through your phone.

You can expand the size of the slides. To do that, you can use the Maximize button in the upper right corner of the slide window, located on the webinar player page. If you'd like a PDF version of today's presentation, you can access it using the Materials button.

We'll be taking your questions today, and you can submit them anytime during our call. If you've joined us via the webinar, you can use the Ask Question button on the player page, or feel free to email us at rapid@stls.frb.org. Either way, we'll get your questions queued up for our presenters.

And with all the logistics out of the way, it's now my pleasure to turn the call over to Catherine Thaliath to kick us off today.

Catherine Thaliath: Thank you, Jean. Good afternoon, everyone. Thank you all for joining us. To kick off the webinar, we wanted to share with you some fast facts from this year's Check Sample Survey—and of course we'll be delving into this in more detail later on in the presentation. So to start: For checks paid, just over half of checks were written by consumers, and businesses were the recipients of two-thirds. By value, businesses made up three-fourths of checks. For checks returned, consumers accounted for two-thirds by number, and by value, businesses [accounted] for two-thirds.

And finally, when looking at return reasons, two-thirds of checks were returned due to insufficient funds by number, and about half by value. Checks returned due to possible fraud was a small fraction of overall returns sampled, by number.

So before we continue on, I would like to do a quick introduction. My name is Catherine Thaliath, and here with me today is Nancy Donahue. We're with the Retail Payments Risk Forum at the Federal Reserve Bank of Atlanta.

Nancy Donahue: Good afternoon, everyone. Last housekeeping item before we jump into what you all came to hear: I would like to mention that the views expressed in this presentation are our own and do not necessarily reflect the views of the Federal Reserve Bank of Atlanta or the Federal Reserve System.

We are going to kick off the webinar with a couple of rapid-fire questions to get everyone thinking about their own check use. You should see a polling question pop up on your screen in just a couple of seconds. Okay, you should see the polling question now: In a given month, how many checks do you write? One to three, four to six, seven to nine, 10 or more, or "Who writes checks these days?"

So Catherine, how many checks do you write each month?

Thaliath: Honestly, I rarely write checks. I think the last time I wrote a check was probably sometime last year. What about you, Nancy?

Donahue: Other than the checks that are created on my behalf by my online bill pay, I rarely write any checks.

So we're going to give it just a few more seconds to give everybody a chance to respond. I think we have just about all of the results; just a couple more seconds. So let's go ahead and close out the poll and see the results.

Wow! Fifty-seven percent in the "one to three" range, and then 28 percent said, "Who writes checks?"

What do you think about that, Catherine?

Thaliath: Yes, my answer would be E for that as well: "Who writes checks these days?"

But that's pretty helpful, and this is actually very consistent with the results of the Survey of Consumer Payment Choice. As you can see on the slide, there's a downward trend in the number of check payments made by consumers in a typical month from 2009 to 2018, with the average being 3.4 check payments per month in 2018. This is down from about eight check payments made per month on average in 2009.

I also want to note that the most recent report, which was released in June of this year, shows that on average around three checks were made in a typical month in 2019—so a little less than what the average was in 2018. But I specifically wanted to call out the 2018 number, as that corresponds to the reference here of the CSS. And, of course, if you want more information, you can find the Survey of Consumer Payment Choice on the Atlanta Fed's website.

Nancy, do you want to take us through our next rapid-fire question?

Donahue: Sure thing. You should see another poll appear on your screen in just a moment. You should see the question now.

We're curious: Have your check writing habits changed as a result of the coronavirus pandemic? Yes, I write more checks; yes, I write fewer checks; or no, I write the same amount or no checks at all. What say you?

So Catherine, I'm guessing your check writing habits haven't changed this year.

Thaliath: That's right, yes. For me the answer is C—I write the same amount of checks, or no checks.

Donahue: Mine has stayed unchanged as well. It sounds like our check habits are maybe less than average, too. What do you think?

Thaliath: Yes, that's right. I'd have to agree, since—as we just talked about—the average was around three checks per month in 2019. Since neither one of us really writes checks, I would have to agree: Our check writing habits are less than average.

Donahue: Less than average. It looks like everyone has responded, so let's go ahead and close out the poll and see what everyone said. So, "unchanged." They're all consistent with us, letter C.

Thaliath: That's right. As you saw from the graph earlier, the average number of check payments made in a typical month has declined over the years. This graph from the Federal Reserve Payments Study shows an overview of what the payments landscape looked like in 2018. And as you can see in red there, out of all the noncash payment types, checks were the only payment type that had a negative compound annual growth rate, which was negative 7.2 percent.

And also looking at the other payment types, cards grew the fastest by both number and value—and more specifically, debit and credit card payments grew faster than any other payment type.

Continuing on the topic of check trends, here we see the number of check payments over the years from 2000 to 2018. As you see from the graph, after a slowdown from 2012 to 2015, the rate of decline in checks actually returned to the previous compound annual growth rate of negative 7.2 percent. I also want to note that over the 18 years of the Federal Reserve Payments Study, checks have nearly fallen to one-third of the original total number of checks written, from around 42 billion in 2000 to 14.5 billion by 2018, which equates to a 5.8 percent decline by number, and a 2.4 percent decline by value.

Taking a closer look at 2018 noncash payments' proportions, these bar graphs show the makeup of the overall payments landscape by number and value in 2018 according to the Federal Reserve Payments Study. By number, there were twice as many ACH payments as checks, although combined ACH and checks represented only one-quarter of all payments. I also want to point out that ACH includes both debits and credits, and by 2018 the number of ACH debits totaled more than the number of checks.

By value checks were the second highest portion, whereas by number checks made up the lowest portion. So this goes to show that although checks are not the first payment choice for everyday transactions, they're still being used occasionally for transactions that are high-dollar amounts.

Turning back to the results of the Check Sample Survey, these bar charts show a breakdown of the shares by number and value of the checks paid and checks returned, also broken down by payer and payee. When looking at shares by number for both checks paid and checks returned, the primary payer was consumers; however, that trend is flipped when looking at shares by value, as the primary payer is businesses. The main driver of the large value of business checks are checks written to other businesses, and we'll get into this in more detail later on in the webinar.

So as part of our analysis, we also included the median value of checks in addition to the average value, and the reason we looked at median value is because these are less affected by extreme outliers, since half of the checks are below the median and the other half are above the median. So when looking at median value regardless of payer, checks to businesses were less than $175 in value for both checks paid and checks returned; and the median value of returned checks written by businesses was the greatest at $504.

And once again, all these results we're sharing with you today can be found in greater detail in the downloadable Excel data tables in the report, which is on the Atlanta Fed website.

So now that we've shared with you an overview of the CSS results, before we dive deeper, let's talk a little bit about what exactly the CSS is. The Check Sample Survey, which we call the CSS, has been conducted since 2001 and examines check usage by businesses and consumers in the U.S. From 2001 to 2016, the CSS relied chiefly—and sometimes even exclusively—on large commercial banks to provide the check sample use in the review.

This year's study includes checks processed by the Federal Reserve in 2018, and unlike in prior studies, the 2018 report includes returned checks. Including these returned checks gave us an opportunity to consider why items are returned, and also includes those items that are potentially fraudulent.

So how was the CSS conducted? The first step in conducting the CSS was the sample selection process. In this stage, we selected a predetermined number of forward items. These are items that are sent through the Fed for clearing between two banks—and we'll get into that in more detail later. And we also selected a predetermined number of return files on each business day from February 2018 to January 2019 (so that way we have a full 12-month period). At the end of the 12-month period, a random sample of checks were then selected for analysis.

I know everyone is anxious to hear the results, so I don't want to spend too much time on the methodology. But I do encourage you to go to the Atlanta Fed website, where you can read in greater detail how the CSS was conducted.

Continuing on with the methodology, the next phase was the check interrogation phase. At a very high level during this phase, each check is reviewed up to three times by three different investigators, who answer both objective and subjective questions about the content of the check through long and short survey questionnaires. The only time a third investigator examines the check is if the answers to the long survey and short survey don't match. Again, you can see a detailed explanation of this process in the technical appendix, and we've also posted the survey forms in appendix D of the report.

And finally, this slide shows the model that we used to categorize each of the approximately 64,000 checks that were examined and ultimately framed the results. I do want to bring your attention to this slide specifically, as this final process is key to understanding the purpose and counterparty relationship of each check.

Starting with the Purpose category, checks categorized as "bill" or "POS" were payments made to a business by either consumers or other businesses. You can see that counterparty relationship in the column to the right as noted by B2B and C2B. Really, the main difference between the two is that bill payments did not occur at the point of sale, and POS payments did. If we weren't certain if the payment was to a business and it occurred at a POS or elsewhere, then these are categorized as "indeterminate."

The fourth category is "income," and we categorized all business-to-consumer payments as income. Examples of these include payroll, benefits, rebates, and bill payments to small businesses that cannot be distinguished from consumers. An example of this would be a small business owner that's using a consumer checking account to manage the business's finances.

And finally, "casual" payments imply informal or irregular usage, and this reflects the wide variety of uses in this broad group. They are from consumers to other consumers. Examples of this include informal payments to family, friends, and acquaintances, for gifts, babysitting, lawn mowing, reimbursements, et cetera. Once again, you can find more details in the technical appendix of the report.

So now that we have the methodology and key terms out of the way, let's dive a little deeper into the CSS results, beginning with checks paid. I touched on this a little bit earlier, but checks paid are also known as forward checks. These are checks returned, collected, and paid as checks. So this survey includes remotely created checks, which Nancy will get into later, and excluded checks converted and collected through ACH.

This chart shows shares by counterparty, and we see that nearly 60 percent by number of forward checks were for bill payments, whether written by a business or a consumer. Payments at the POS accounted for just 4 percent of processed checks. B2C payments—all of which, as I mentioned before, are income payments—made up 20 percent of check payments, and C2C payments, all of which are classified as casual, were the lowest category at about 14 percent. It's interesting to note here that B2C payments accounted for 26 percent of the shares by number, with the majority being bill payments.

And although many businesses still use checks to pay vendors and merchants, I think it'll be interesting to see if there's a decline in check usage by businesses as they continue to adopt new accounts payable fintech solutions.

This graph shows shares by value, and bill payment checks represented three-fourths of checks. C2C payments by value only represented 6.5 percent, which is less than half the share by number and is the lowest category compared to all the others by number and value. As C2C platforms are becoming more popular, it's not surprising that individuals—including myself—are writing fewer checks and opting for P2P solutions such as Venmo and Cash App to pay other individuals.

And finally, these graphs show the median value for businesses and consumers as the payer. As you saw in the previous slide, C2C is the lowest category when looking at shares by both number and value, but when looking at median value, C2C payments—or casual payments—is the highest, with the consumer being the payer.

Now I'll turn it over to Nancy to discuss checks returned.

Donahue: Thank you, Catherine. As we mentioned before, returns are new to the CSS this year. When we say returns—return items, checks returned, et cetera—we are referring to checks that were returned by the paying bank to the depositing bank for some reason. The 2018 CSS examined return items across 30 reasons to understand the predominant reasons items are returned, and to what extent possible fraud might play a role.

By number, consumers wrote two-thirds of returned checks, and they wrote over half of them to businesses. Overall, businesses received three-quarters of returned checks by both number and value. For both consumer and business payees, "bill" was the largest category by number. "Income"—which, as Catherine explained, is business-to-consumer—followed closely behind "bill" payments.

By value, two-thirds of unpaid checks were written by businesses. Overall, half of checks returned were written by businesses to other businesses (in other words, B2B), followed by C2B at 23 percent. Like it did with number, overall "bill" dominated at nearly 70 percent of the total value of checks returned. Checks written at the point of sale represented only 2.5 percent by value, which is reflective of the trends in POS payment choices as well.

Shares by payer and payee differed from the patterns in forward checks, or checks paid, in that checks written by consumers were more prevalent in the sample of checks returned than in the sample of checks paid.

So here we see the median value of checks returned by counterparty and purpose. What stands out most to me in this slide is the median value of casual checks, which is nearly double that of bill and over five times greater than POS. While consumers have adopted new means of exchanging money with each other using various P2P platforms, the high median value for both forward and returns suggests that for larger-dollar transactions, consumers continue to use checks. This may be for budgeting and expense tracking purposes, but it's also important to point out that the consumer category likely includes small businesses that cannot be distinguished from consumers, which may influence the medians that you see here.

As you know, checks are returned by the paying bank for many reasons, and as mentioned before these results reflect 30 different return reasons. The downloadable Excel data tables that Catherine mentioned, that are posted on the Atlanta Fed website, include proportions for each of those 30 return reasons. The CSS further summarize them into three broad categories: nonsufficient funds, or NSF; stop payment and similar; and possible fraud. We have included an "all other" category here for completeness as well.

In addition to nonsufficient funds, the NSF category also includes items that were returned for uncollected funds. We chose to include uncollected funds in this manner because the outcome for the payee is the same: a check that they deposited was returned and charged back to their account. The category "stop payment and similar" includes administrative returns such as stale-dated items, closed account, unable to locate, and frozen/blocked account.

Lastly, the possible fraud grouping, which we will get into in a couple of slides, includes refer to maker (which led the category).

Okay, Catherine, I think we have another rapid-fire question for everybody before we move on to the next topic, right?

Thaliath: That's right. So once again, everyone, you should see a polling question appearing on your screen here shortly. All right, so there it goes. Moving into our next section on possible fraud, we'd like to hear whether anyone in the audience today has been a victim of check fraud in the last five years: A, yes; or B, no.

Thankfully, I've never been a victim of check fraud because I write so few checks in the first place.

Donahue: I don't ever recall having experienced check fraud per se, but I do recall a time when people would steal boxes of checks out of your mailbox, which usually meant you had to close your account and reopen a new one.

Thaliath: Oh, wow.

Donahue: It looks like we have just a few more people that are responding, so we'll just give them a couple more seconds to record their answer.

There we go: 91 percent. That's pretty overwhelming. Wow.

Moving on to our fraud section: So the "possible fraud" grouping encompass nine return reasons, as shown here, of which "refer to maker" compose the greatest share by both number and value. So "refer to maker" broadly means that the payee of the check should contact the maker to find out why it was returned. We chose to include "refer to maker" in this category based on discussions with large DIs that indicated use of this return reason when an item was thought to be potentially fraudulent but was not adequately attributable to one of the other reasons. It's unknown, at least to us, the extent to which this practice is employed across the larger population of financial institutions.

So I would just like to take a minute to say that we welcome your input and feedback on this question as we seek greater precision in our research efforts. Our email addresses can be found at the end of this presentation, and your insights on this topic or on anything that we discuss today are much appreciated.

So by value, the story is much the same—although it is worth noting that "duplicate presentment," which ranked below "altered/fictitious" by number, is second highest by value here.

Now that we know the return reasons that make up the "possible fraud" category, what does it look like by counterparty and purpose? By number, 60 percent of possible fraud checks were drawn on business accounts, and consumers were the recipients at the same rate. And while income checks were most prevalent at 43 percent, as you see here in purple, bill payments written by businesses and consumers compose the next largest share at 36 percent. Consistent with possible fraud checks overall, the primary reasons for B2C checks were "refer to maker" and "altered/fictitious." These two reasons accounted for nearly 75 percent of B2C checks by number.

Not shown on this slide, "refer to maker" was also the primary return reason for bill checks. However, "duplicate presentment" was more prevalent than "altered/fictitious," as was the case with income. "Refer to maker" and "duplicate presentment" accounted for over 85 percent of bill checks by number.

Checks drawn on business accounts made up 82 percent of potentially fraudulent checks by value; other businesses were the recipients of nearly 56 percent of those checks, which were almost exclusively bill payments. Income checks represented the second largest category at over 26 percent, and then point of sale was a miniscule share at less than half a percent.

So possible fraud, which is represented in blue on this slide, was largely consistent with the shares of checks returned (except for B2C and C2B). Income checks written to consumers represented the larger share of possible fraud checks by both number and value compared to the overall sample of checks returned. The inverse is true for C2B, with consumer checks written to businesses representing a considerably larger share of overall returns compared to that of possible fraud.

This slide compares the median value of checks returned, and possible fraud checks returned, where the payer is a business. In each of the purpose categories, possible fraud median was considerably lower, with the exception of B2C (which was nearly the same). Also, the medians for possible fraud, B2B bill and B2C income checks, were nearly equal.

Lastly, where the payer is a consumer, we again see consistency in the median value of three of the four purpose categories, with the exception of casual payments. The median for casual items was nearly three times that of the next highest category, which was bill, and over seven times that of POS.

So now that everyone has heard our thoughts on check fraud, what's your reaction? Catherine is going to take us through the next rapid-fire question.

Thaliath: Absolutely. In a moment, you should see another polling question appear on your screen; there it goes. The question is: Does your organization use the "refer to maker" reason to return potentially fraudulent checks when another more appropriate reason isn't present? A, yes; B, no; or C, unsure.

Donahue: We have a large number of depository institutions joining us on the webinar today, so I am curious to see how everyone responds to this question.

Thaliath: Yes; me, too. And this will be very helpful as we consider our future research.

Donahue: We're getting quite a few responses.

Thaliath: That's right. All right, I think we're just about ready to close out the poll. Let's see the results.

Donahue: Interesting; kind of a split decision there.

Thaliath: That's right. Still, very helpful. Thank you all for your feedback. Nancy, do you want to walk us through remotely created checks?

Donahue: Absolutely. The 2018 CSS report includes a separate appendix on remotely created checks that discusses our approach to these items. Once again, you can access this information on the Atlanta Fed website, using the link found at the end of this presentation. And I believe, just in case they haven't figured it out yet, someone posed a question about if we would provide the presentation at the end. If you click on the Materials button, you can download the slide deck. Those links are present in the deck at the end, as I said.

So what defines a "remotely created check"? Broadly, RCCs are checks issued on the payer's behalf; they are authorized by the payer and drawn on the payer's account, but do not bear the payer's signature. Reg CC defines an RCC as a check that is not created by the paying bank, and that does not bear a signature applied, or purported to be applied, by the person on whose account the check is drawn. Reg CC further acknowledges the difficulty in distinguishing between checks that are created by the payee or its agent from other checks, such as checks created by a customer's bill payment service.

We chose to categorize RCCs in three ways: checks with the number "6" in position 44 of the MICR line; checks with the number "6" and some type of signature authorization, other than the applied signature of the payer; and then checks with only a signature authorization reference. Some of the signature authorization references that we found in our research included: "no signature required"; "signature on file"; and phrases such as "authorized by drawer," "authorized by the depositor or the payer," "preauthorized payment," and "this is a bill payment draft."

Of the approximately 65,000 items that we examined, approximately 1,400 forward and 1,200 return checks contained one or more of these three characteristics. So in the absence of consistent distinguishing features, it was not possible for us to exclude from the CSS sample every item that may have been created by the paying bank. Some of those may be included in these results, and so users of this data should bear this in mind.

As you can see from the charts on your screen, by number and value, checks with a signature authorization only were the predominant characteristic in both the forward and return samples. And as you would likely expect, consumer payers to businesses were most prevalent for both RCCs paid and returned, by number and value, all of which were bills. What this means is that the consumer calls up to pay a bill, gives their account number and routing number, and the business creates a check on their behalf without their signature.

The median value of a C2C RCC return was over four times that of RCCs paid. A couple of thoughts, just to give you context to this number: First off, it is possible that this category might include small business payees, as we've mentioned before, which are sometimes indistinguishable from consumers and have similar payments, choices and behaviors. Secondly, this category represented a very small number of checks—approximately 50—so it is possible that the presence of some large value items influenced the median. For B2B, the trend is reversed and RCCs paid is more than RCCs returned by nearly four times.

And lastly, going back to the possible fraud discussion: By number, RCCs made up only 10 percent of items and 2 percent by value, and the median value of a possible fraud RCC return is $118.

We have received several questions from folks since the report released concerning trends or comparisons to previous results, and if you've gotten a chance before now to read the actual report you may recall that we recommended against such comparisons. I'd like to take a moment to explain how the 2018 CSS differs from prior studies, and why we discourage comparisons to previous results.

The table shown here on this slide reflects some of the key differences between this year's CSS and that of past years. These data are not comparable to other similar studies or data sets that may exist, including previously released CSS figures, because the sampling came from an exclusive data set (meaning the Fed image sample). Almost any data set of which we're aware has some exclusivity issue, and therefore built-in bias of some type. It's hard to adjust for that, which is why we've advised against comparing—not only in the paper, but we do so again here as well.

In particular, previous CSS studies we conducted were heavily weighted toward large banks, with lots of corporate customers and on-us volume. The most recent sample that we employed for the 2018 study includes many more small bank depositors, but doesn't have any on-us volume. Both included interbank checks, referred to here in the slide as transit items (meaning the check was exchanged between two banks for clearing).

And while we think that this current set is a broader mix of checks, overall it's still not fully representative of all checks, and the excluded items could meaningfully change the outcomes and conclusions. In framing the previous CSS versus the current CSS, you could reasonably argue that both are wrong or that both are right; the unarguable point is, each has their own set of limitations.

So for those reasons we think it's wise to stay away from comparing or trending these data against other data, because each represents something different and the differences aren't trivial. But as our research evolves, and to the extent that we can repeat the study or perform similar studies, we'll strive to build out our trends and work to make comparisons that are sensible.

And now we would like to open the floor to your questions that you might have. You can submit questions two ways: using the Ask Question button in the webinar tool, or via email at rapid@stls.frb.org. So let's see if there's anything in the queue so far. I think we have quite a few.

Starting out our first question, Catherine, I'll throw this one to you: Is the Fed sample a national sample?

Thaliath: Great question. Yes, it is a nationwide sample in that the convenient sample is drawn from the universe of checks processed by the Fed (which is approximately one-third of all forward and one-half of all return items). And this includes banks of all types and sizes, in all locations in the U.S. However, I do want to emphasize that that does not represent the full population of checks because it does exclude on-us items and checks settled directly between two banks. Great question.

Donahue: Yes. Are there plans to conduct additional analysis on the types of banks in the Fed sample?

Thaliath: Sure, I'll take that question. There are plans and discussions under way for further analysis. Right now, what we know is that there are approximately 8,000 unique routing numbers captured in our sampling selection. We're also equally interested in understanding the makeup as it relates to the type of bank, size of bank, and geographic location of the bank. Ideally, we'd like to have something forward-looking, too, as this will be beneficial to us to understand what that composition is in order to interpret future results as well.

Donahue: And when will the next CSS be conducted?

Thaliath: Great question. Historically, it's been conducted every three years, making 2021 the next in the cycle—so this would be published in 2022. We're also considering interests, both internally and externally, in increasing the frequency of this research. In fact, that's actually one of our polling questions that we'd love to hear from you guys. And ultimately, whatever is decided will be influenced by the status of the COVID pandemic at the time.

Donahue: We have a number of questions, and we'll try to get through as many as time will allow today. If we're not able to answer them all in the webinar, we will try to distribute answers to everyone following the webinar. So, first question (I'll take this one): Does "business payments" include government payments?

The short answer to that is "yes." Because governments have the same payment choices and options available to them as businesses, they are included in our statistics and results in the business category.

So, next question: Why wouldn't a check returned due to the account being closed be considered possible fraud?

That's a great question, and I'm sure the bankers in our audience have lots of thoughts on this, too. I could see that in two different ways: It could be that someone got ahold of some old checks and is committing fraud by writing those checks, knowing that the account is closed. But it could also just be a timing thing, or an accident that someone accidentally used an old checkbook—so not intentional. So I think you can make the argument both ways on that.

What is meant by "indeterminate?"  Catherine, would you like to answer that?

Thaliath: Yes. "Indeterminate" is when we were not able to determine if the check was made to a business via a point of sale system or not.

Donahue: Right. So, next question: Why did you change the sampling method this time?

We had, for a number of years, gone to a handful of large financial institutions to contribute checks to the study. For various reasons—primarily sensitivity around SPII [sensitive, personally identifiable information]—it was getting more difficult to collect checks to conduct the study, and so we actually performed a parallel study internally in 2015 to test the feasibility of using Fed images. For those reasons, primarily, we moved forward with using the Fed images exclusively for the study.

We have another question: How should banks use this data for future planning, if there is a new baseline without any apples-to-apples trending? 

So as we said, moving forward in our research, it is our goal to return to providing those comparisons once we have apples-to-apples samples.

I think we might have time... let me go through the questions real quick....

Someone asked: "You should include age ranges for how many checks per month." That is a great question, and I would encourage you to check out the survey, the Diary of Consumer Payments—we have a link to that at the end of our slide presentation—that provides views such as that as well. But that is, in terms of the CSS, out of scope since we are collecting our data from actual check images, so we would not have access to that demographic information.

So I think, Catherine, our listeners gave us some great food for thought with the rapid-fire responses and questions today. I think we have one more ask of them before wrapping up.

Thaliath: That's right. We definitely got some very helpful insights from our audience today, so thank you once again for your feedback.

So closing things out we just have one final polling question for you, and you should see that show up on your screen shortly. There goes the question. The question is: The CSS has been conducted on a three-year cycle since its inception in 2001. Should we update check data such as the CSS on a more, or less, frequent schedule? A, annually if possible; B, every two years is sufficient; C, every three years is sufficient; and D, it's not that necessary to update this information.

And while we wait on the polling results, I wanted to give you a little bit of background about the Retail Payments Risk Forum. Our primary mission is to identify risk in existing and emerging retail payments, and we do this through research of payments, products and systems, collaborating with industry participants, regulators, law enforcement and others in the Federal Reserve System, and also convening payments providers and parties integral to establishing new products, laws and regulations, policies, and standards that affect and shape retail payments.

And I think we're just about ready to close out the poll here.

Donahue: So kind of a split decision again, but great feedback from everyone. Thank you.

Thaliath: Absolutely. Thank you. And of course, if you'd like more detailed information on the research that we discussed here today you can go to the Materials section, which Jean and Nancy mentioned earlier. There you can download the PDF version of today's presentation, and you'll be able to click on those links that you see here on the slides.

Donahue: Once again, thank you for your interest in and support of the Atlanta Fed's payments research. We hope that you found today's session useful, informative, and applicable to your organization. We ask that you take a few minutes to complete the participant survey, so that we can continue to provide you with content that is timely and relevant.

You can respond to the survey in two ways: using the survey link on your screen, or by email (which will be sent shortly following the webinar). Either way, your thoughts and comments will make their way to us directly, and you only need to respond once.

And I would like to just say one more time, because we did have several questions: You can download today's presentation by clicking the Materials button here. Also, a recording of today's webinar, and the materials, will be posted on the Atlanta Fed website under the Retail Payments Risk Forum.

Thank you, once again—and have a great day, everyone.

Thaliath: Thank you, everyone.