Transcript
CFPB FinEx Webinar:  Insights from the Making Ends Meet Survey
February 22, 2024
Presenter:   Emma Kalish, Office of Research, CFPB
Facilitators:  Joshua Shapiro and Ken McDonnell, CFPB

  
>>CFPB Staff:  Emma, you're still on mute. 

>>Emma Kalish:  Are we good now? 

>>CFPB Staff:  Yes, you are. Yes, you are.

>>Ms. Kalish:  I'm getting a little bit of a echo. All right. I'll keep going. So my name is Emma Kalish. I'm an economist with the Office of Research. I'm going to be talking today about insights from our Making Ends Meet survey. 
 
And before I begin, I have to give you our little disclaimer, which is that this presentation is being made by a CFPB representative on behalf of the Bureau. It does not constitute legal interpretation, guidance, or advice of the Bureau, and any interpretations or views stated by myself are my own and may not represent the Bureau. 
 
So with that, let me give you a little preview of what we're going to talk about today. Today I'm going to quickly introduce the data source and then discuss some findings from our State of Consumer Finances in 2023 report. Then I'll go into detail on some new research from my colleagues on overdraft and nonsufficient funds fees. As Ken said, I'll stop about halfway for questions and then leave time again at the end. 
 
So the Making Ends Meet data is the main data source for this talk. Making Ends Meet is a general financial survey that examines consumers' evolving financial status across a range of measures, and we sent our first sample out in May of 2019 prior to the pandemic. And since then, the Bureau has sent out multiple samples and ways of Making Ends Meet.

Making Ends Meet is a unique dataset compared to other financial well-being surveys because we also have access to consumers' credit data, and that link gives us a broader picture of how consumer credit is changing. For this survey, we contact people via mail, and there's an online option as well, and we result in a sample size of around 2,000 to 3,000 respondents depending on the year. 
  
So just to jump right into our results, I'm going to be showing you a bunch of graphs today, and I'll try to explain what these different graphs are showing because I know it's a little bit confusing just to see tons of bars.
 
So for this graph, the first panel of the graph, those first four bars, show financial well-being over time, and financial well-being is a measure that the Bureau creates from a variety of different survey questions. And what I'd like you to take away here from this first panel is that there was a small spike in financial well-being in 2021Ñthat's the orange barÑduring the pandemic, but since then, those blue and gray bars, financial well-being has returned to the previous level it was at in June of 2019. 
 
What this graph doesn't show is that financial well-being actually slightly improved from 2022 for the lowest-income individuals in our sample but dipped slightly on average for those with the highest income. 
 
he second panel of this graph focuses on a different measure. This comes from a single question asking consumers whether they faced difficulty paying for bills or expenses over the past year. So what we can see from this graph is, again, during the pandemic, in that orange bar, February 2021, the share of consumers who had difficulty fell, but since then it has been rising. And it's almost returned again to pre-pandemic levels. Again, of course, different groups are more likely to experience difficulty when paying bills.
 
So this graph shows the share of consumers of each income group that we look at that said that they had trouble with bills or expenses. So the lowest-income consumers are on the left in light green. Those are consumers that make under $20,000 a year of income, and the higher-income consumers are towards the right. Perhaps unsurprisingly, lower-income consumers are much more likely to have difficulty paying a bill or expense, so 61 percent of those with income less than $20,000, compared to only 12 percent of those with incomes above $125,000. But as we can see from this graph, many low-income consumers don't face financial difficulty, and even some high-income consumers do.
 
This graph looks at the same measure of difficulty paying a bill or expense, but by race and ethnicity instead. And from this, we can see that Black and Hispanic households are more likely to face difficulty paying bills as compared with white.
 
Finally, this graph shows difficulty paying a bill or expense by educational attainment, and again, just going to explain, and then we'll go a little quicker for the next couple. So high school or less is the light green bar on the left, and it moves up all the way up to a four-year college degree, the far right bar, the dark blue. And what we can see from this graph is that while consumers with a four-year degree are less likely to face difficulty compared to others, 22.5 percent, those consumers with some college or who only have a two-year college degree are only very slightly less likely than those without any college degree to face financial difficulty, and those differences are not statistically significant.
 
Most consumers in our sample that face financial difficulty have experienced this difficulty more than once for the past year. So this graph shows the frequency with which consumers who face difficulty report facing that difficulty, and what we find is that over half of consumers who report facing difficulty report it over three times in a year, and a third of consumers face difficulty at least five times.
 
We also asked consumers to think back to the most recent time they had difficulty and tell us what they had difficulty paying for, and what we find is that when people have difficulty, they often have difficulties paying for the most basic household expenses. So those top two categories are utilities and food. Those are basic things that all households are paying for every month.
 
Another thing that we ask consumers is, what would they do if their household lost its main source of income? How long could they cover expenses by doing things like borrowing, using savings, and selling assets? And this graph shows that overall, about half of consumers are only able to cover expenses for two months or less, and this is similar in 2022, the green bars, and 2023, the orange bars. The shift that we see between 2022 and 2023 is that we've seen fewer people saying that they could cover about two months and more people saying that they would only be able to cover about one month.
 
When we look at which types of consumers can only cover one month or less, we see that over 70 percent of consumers with income of $20,000 or less wouldn't be able to cover one month, whereas almost all high-income consumers, 94 percent, would be able to cover at least one month.
 
When we look at the same measure by race, we find that over half of Black and Hispanic consumers would not be able to cover expenses. 
 
And finally, almost two-thirds of those with no college who had a high school degree only could not cover expenses for a full month. Whereas 80 percent of those with a four-year degree can cover at least a month. So there's big differences across these demographic categories.
 
Next, I'm going to talk about access to money, both to savings and credit. So just to tee us up, household income varies from month to month for many households. Households need access to savings and or credit as a financial cushion, and that could come from savings and checking account balances or from available liquidity. And consumers differ in their access to and their use of credit products. Consumers without access to savings or without traditional credit may receive help from friends or family or use alternative services, such as payday loans, when they need access to money.
 
So we asked consumers to tell us if their income varies a lot from month to month, a little from month to month, or mostly stays the same. So this is, again, a graph over time, and the left-most set of bars is the share of consumers who say their income stays about the same each month. The middle set is the consumers that say it varies somewhat from month to month, and the final set is those that say it varies a lot from month to month. And what we find is that the share of consumers who say that their income varies either somewhat or a lot has increased over the last four or five years, and the share that say it varies a lot, in particular, increased last year. 
 
This graph shows the percent that say their income varies either somewhat or a lot, and what we see here is when we look by income, the lowest income consumers see a little bit more variability than others, 41, 42 percent, but variability is certainly not limited to low-income consumers. We see a high degree of variability across the income spectrum.
 
One thing that we know can cushion families and households from shocks and income variation is savings, and this graph shows, again, for 2022 and 2023, the amount of savings in a checking or savings account that households report. The left two sets of bars represent consumers who either don't have a checking and savings account balance at all or who report having $100 or less, and that accounts for 20 percent of consumers. So 20 percent of consumers report having $100 or less in their checking and savings account, and a full 40 percent of consumers have less than $1,000 in their savings and checking accounts, and those patterns are relatively consistent over time between 2022 and 2023.
 
But we also know that checking and savings balances are not the only source of liquidity available to most households. The ability to borrow on a credit card supplements the funds that a household can call upon. So this figure shows the amount of liquidity, which is defined as the credit card limit minus the credit card balance, so how much more credit card limit a household has available to them, available on credit cards specifically, by savings and checking account balance. So the sets of bars, each represent a different amount of the checking and savings balance, and then the height of the bars represents credit card liquidity. And what you can see is that as households have more savings, they also have more liquidity. 
 
So if you look at the time trend of this graph, if you look at the green bar compared to the orange bar compared to the blue bar, what we can see is that for households with at least $500 in their checking and savings accountÑso those households on the right side of the graphÑhave seen significant increases in their credit card borrowing power in the past year. The blue bar is a little bit taller thanÑthe blue and orange bars are a bit taller than the green bar. But for consumers with under $100 in their savings account, which remember makes up over a fifth of consumers, these consumers are also lacking credit card liquidity. So they don't have a lot of savings, and they also don't have a lot of liquidity in their credit cards. 
 
So the next question we ask is, who has credit cards and how are they using them? Most consumers, what we see in this first bar, 83 percent have a credit card. Then looking among consumers that have credit cards, 23 percent of those consumers with a credit card paid a late fee on that credit card at some point in the past year, and 52.5 percent revolved on their credit card the last time they paid it, meaning that they didn't pay their full balance last month and were likely charged interest on that balance.
 
This graph, I know it's a little bit crowded, but it shows the same three measures that I just discussed but by consumer income. So if you just focus on the first set of bars on the left, what we can see is that almost all households with incomes above $50,000Ñso the three blue-gray barsÑhave a credit card. Almost all of those households have a credit card, but under half of those consumers with income of less than $20,000, that light green bar, have a credit card, only 47 percent of those households. 
 
Focusing on late fees and revolvingÑand remember that this is only among consumers that have the credit cardÑwe see that consumers across the income spectrum pay late fees and revolve. Even among the very highest-income consumers, those who have incomes above $125,000, 40 percent of those consumers are revolving. 
 
We also ask consumers about their experiences in getting and applying for credit. This first bar, 41.5 percent of consumers applied for credit in the past year. The second bar, 23.8 percent of consumers didn't even apply for credit because they assumed that they would be rejected. And then finally, among the 41 percent that applied for credit, 36.6 percent of them were turned down for that credit or not granted as much as they asked. 
 
Looking at this, these three same measures by race, what we find is that while consumers of all racial groups are about equally likely to apply for credit, that first set of bars, there are racial differences in the other outcomes. Specifically, white consumers, that light green bar, are almost half as likely to assume that they will be rejected compared to Black and Hispanic consumers. And indeed, we see they are 10 percentage points less likely to be rejected conditional on applying for that credit. 
 
Another way that consumers may make up gaps besides using traditional savings or credit is through informal financial assistance from other households. This graph is going to show householdsÑeither giving, receiving, or bothÑfinancial assistance from friends or family. So about half of households didn't give or receive. That's the tallest set of bars on the left. Then we have about 9.8 percent of consumer households that received assistance only, 31 percent of households that provided assistance only, and then 13 percent of households both gave and received financial assistance over the past year.
 
Most households that give or receive financial assistance do it multiple times throughout the year. Only 13 percent of households that provided assistance only provided it once. Whereas 22 percent provided it two to four times, and 9 percent provided it five or more times. 
 
This graph, again, shows giving and receiving by income. The left set of bars is provided assistance, and what we can see here is that consumers of all income levels are almost equally likely to have provided assistance. Forty percent of consumers who make only 20- to $50,000 provided assistance, and 41 percent, only 1 percentage point higher, of consumers that make $125,000 or more provided assistance.
 
However, when we look at receiving assistance, perhaps unsurprisingly, those consumers in households with $20,000 or less of income, that light green bar, are much more likely to receive assistance than other households, even though they're not statistically significantly less likely to provide assistance than other households.
 
We also see from the provided assistance by race graph on the left that Black households are disproportionately likely to have provided assistance to other households, despite the fact that Black households have disproportionately low incomes on average. We also see that White households are the least likely to both give and receive transfers.
 
Finally, last one before we take a little break, this graph shows the use of alternative financial services that consumers may use to get liquidity:  payday loans, pawn loans, and auto title loans. What we find over time is that the overall use of payday and pawn loans has not changed significantly between 2022 and 2023, even though it declined in the first year of the pandemic. Meanwhile, auto title loans have increased since 2019. That really large jump is more of a statistical anomaly, but overall, we still see an increasing trend in the use of auto title loans.
 
So I'm going to take a pause here for questions, and then I'm going to go into another study that we have on overdraft and nonsufficient funds. So if there's any questions, I'm going to let Josh tell us what to do next. 
 
>>Joshua Shapiro:  Well, Emma, we do have a question here from Sarah. 
 
>> Ms. Kalish:  Great. 
 
>>Mr. Shapiro:  She's asking, are money circles taken into consideration on alternative financial services? 
 
>> Ms. Kalish:  So not in this graph here. We asked specifically about payday, pawn, and auto title, but that would definitely be something that we could ask about in the future. I don't believe it's something that we asked about specifically last year. 
 
>>Mr. Shapiro:  Thank you, Emma.
 
>> Ms. Kalish:  Thank you.
 
>>Mr. Shapiro:  Any other questions from the audience? 
 
[No response.]
 
>>Mr. Shapiro:  And it doesn't look like we have any at the moment, but please feel free to type in any of your questions as they occur to you. We will address them.
 
Emma?
 
>> Ms. Kalish:  Yeah, thank you. So hopefully, that means that all the graphs are super clear, because we're about to look at a bunch more of them. I'm now going to talk about a new report on overdraft and nonsufficient funds fees, and this is not research that I was involved in, but I'm going to be presenting it to you today. This was by my colleagues Jade Elkins, Gio Palloni, and Caroline Ratcliffe. So I'm just going to go over some definitions before we start so that we're all on the same page just about what we're talking about.
 
So when consumers use deposit accounts, checking accounts, savings accounts, when those consumers attempt to withdraw or debit payments for amounts greater than the available funds in their account, there's a couple different things that could happen. They could have an overdraft fee charged if the financial institution covers the transaction and thereby extends credit to the consumer. So if the utility bill comes due and is supposed to be automatically deducted, but the balance isn't high enough and the bill still gets paid, the bank pays the bill and assesses a fee, that would be an overdraft fee.
 
We also have nonsufficient funds fees, which are charged if the financial institution does not cover the transaction, so if the bill is sent back to the utility company unpaid, but the financial institution still assesses the consumer a fee.
 
And of course, not all overdraft or nonsufficient funds transactions are charged a fee. For nonsufficient funds, there is a lot of variation across the transaction types and institutions. And for overdrafts, there are de minimis amounts. There's grace periods and fee waivers. And again, it just depends on the account and the transaction type. But there are a lot of cases in which overdraft and nonsufficient funds fees are charged.
 
And there's been a lot of recent changes in the industry relating to nonsufficient funds fees and overdraft fees, and here's some examples. But generally, many institutions have either eliminated or reduced one or both type of these fees.
 
And that leads to some of the research questions that we were interested in with this report. So what this report covers is the following. How often did consumers incur overdraft and nonsufficient funds fees around 2022? When consumers overdraft their account or are charged a fee, are they expecting that fee, or are they surprised? And how are consumers' overdraft and NSF-related experiences related to socioeconomic and demographic characteristics, credit and debt-related characteristics, and financial well-being?
 
So this graph shows the share of consumers who overdraft and how many times they overdraft. So the left-most bar is the consumers that do not overdraft at all in our sample, and that accounts for 76.4 percent. The blue bar, 14.8 percent of consumers experienced one to three overdraft fees. 5.5 percent of consumers, the black bar, experienced four to ten overdraft fees. And 3.2 percent of consumers experienced more than ten overdraft fees in the last year. 
 
The NSF fees, the categories are a little bit different, but the pattern is the same. NSF fees are slightly less common in our sample than overdraft fees, with 80 percent of consumers experiencing no overdraft fees, 14 percentÑsorryÑNSF feesÑ14 percent experiencing one to three NSF fees, and 5.9 percent experiencing more than three NSF fees in the past year.
 
It's a little bit hard to compare overdraft and NSF fees over time in our data because the way that we asked the question changed as the market changed. But broadly, these fees appear to have fallen during the pandemic from the green bar to the orange bar and then have been increasing again since then, from the orange to the blue to the gray, which finally represents 2023.
 
Now, this figure, I really like this figure, but I find it a little bit confusing. This figure shows whether consumers were surprised by the overdraft, and it divides it up by the number of times they overdrafted. So the leftmost set of bars is for consumers that only overdrafted between one to three times. The far right side is for consumers that overdrafted more than 10 times. And there's two takeaways from this figure. Consumers tend to be surprised by their most recent overdraft. So that's the green bars. Those are consumers that were surprised. Forty-three percent of consumers were surprised overall, and 35 percent of consumers thought it was just possible. So when they got their last overdraft, they were either surprised to get a fee and get that overdraft, or they thought it was possible, but they weren't necessarily sure that they were going to overdraft. Whereas only 22 percent of consumers on their last overdraft actually expected to have that overdraft. However, consumers from households that are charged fewer overdraft fees are notably more likely to report having been surprised. Whereas over half of those consumers with 10-plus overdrafts, that very farthest, most dark bar, are less likely to be surprised because they've had so many overdrafts. They sort of understand what's going to happen at this point.
 
So for the rest of the analyses, we're going to mostly try to group consumers into three main groups, and so there's a no overdraft or NSF group, 74 percent of consumers who did not have any overdrafts or NSF fees in the past year. Then we have this group that we're calling the "occasional overdraft group." That accounts for 8 percent. That's the blue. And those consumers were either charged one to three NSF fees or one to three overdraft fees, but not both. And then the charcoal gray bar is 18 percent of consumers, and that represents consumers that are in the frequent overdraft and NSF. Those consumers were charged either both, one to three overdraft fees and one to three NSF fees, or more than four of either fee type. And that's what we're going to call our frequent overdraft.
 
So this graph is going to show the average credit score over time for each of those three groups, and if we remember the no overdraft group, that's the green group, those consumers have the highest credit scores on average with an average credit score of 744 over the sample period. In the occasional overdraft groupÑthat's the blue groupÑthose consumers have an average credit score of 687, so significantly lower than the no overdraft group, but higher than the frequent overdraft group in the gray, which has an average credit score of 637, about 50 points lower than the occasional overdraft group. So we can see there's like a distinct credit profile for each group.
  
Now this group shows the percentÑor sorryÑthis graph shows the percent of consumers of each of these three types that had no credit left on their credit card. They spent their balance all the way up to their credit limit and, again, showing that measure over time. So what we can see here is that that gray group, the frequent overdraft and NSF fees group, is more likely to have no credit available. But even in that group, just 40 percent of consumers in that frequent category had no credit available, which is considerably higher than the 18 percent of consumers in the no overdraft group that had no credit left. But it still suggests that many consumers who are overdrafting also have access to a potentially cheaper source of credit, which would be charging on their credit card. 
 
This graph shows the use of alternative financial services, payday, title, auto title loans, and pawn loans by those same three groups. So the no overdraft fee group on the left, only 6.5 percent of those consumers used an alternative financial service, compared to 9 percent of those with occasional overdraft fees. And then 25 percent of consumers who frequently overdraft also reported using alternative financial services in the past year. 
 
Okay. Again, a confusing but very worthwhile graph. So this graph is showing the number of times a household had difficulty paying their bills by the overdraft category that they're in. So the left-most set of bars is those consumers that had no overdraft or NSF fees. And that really tall dark green bar says that 75.5 percent of those households that didn't have any overdraft fees never had difficulty paying bills. Moving to the right, the light green bar for that section says 5.9 percent of consumers who had no overdraft fees had difficulty paying bills more than four times, only 5.9 percent. If we move over to the right-most set of bars, those frequent overdraft and NSF fee consumers, we see that only 18.8 percent of those consumers who had frequent overdraft said that they never had financial difficulty. So over 80 percent of frequent overdrafters said they had financial difficulty, and 36 percent of them said that they had financial difficulty paying bills more than four times over the past year. 
 
So similar graph as the previous one, this is looking at how long households could cover expenses if they lost their main income source, so that same question that I talked about way back at the beginning of the presentation. And what we found is that most consumers who have no overdraft or NSF fees are able to cover expenses for a month, two months, or even three or more months, over half of those consumers. When we compare that with consumers that have frequent overdrafts, 40 percent of those frequent overdraft consumers could not even cover two weeks if they lost their main income source, and only 16 percent said that they could cover three or more. So that's a big difference. 
 
So just to sort of sum up some of those results on overdraft and NSF fees, we find that consumers continue to be affected by overdraft and NSF fees despite changes in the market. And overdraft and NSF consumers are more likely to be economically disadvantaged, be low income, have less education, and to be younger. These consumers have less healthy credit profiles, but most consumers, even in that frequent overdraft and NSF group, have some credit card credit available that they could be taking advantage of. 
 
So with that, I would like to thank you. These are the very long links to the two reports that I was discussing this weekÑthis afternoon. We would love it if you would take a look at those, and I look forward to answering your questions. Thank you so much.
 
>>Mr. McDonnell:  Thank you, Emma.
 
>> Ms. Kalish:  Thank you. 
 
>>Mr. McDonnell:  For everyone else, the slide deck will be sent out when we send out the recording, so you'll be able to getÑaccess to those links. Emma, we do have a question here from Nikki [phonetic]. Has any analysis been done on how the frequency of overdraft or NSF fees per person has changed since banks have either reduced fees or eliminated them?
 
>> Ms. Kalish:  Yeah. So this is looking atÑit's not exactly at the number per person. That number actually may be in the report, and I just don't have it in front of me. But I wouldÑyeah, I don't want to tell you the wrong thing, but I would say go take a look at that report because it likely does have some of that information. What we can see is that, you know, we do have this data on the numbers. So yeah, again, I just don't want to tell you the wrong information, but I think it's in that report.
 
>>Mr. McDonnell:  Thank you, Emma.

And we have a question here from Sarah. Any updates on the reduction of the late credit card fees? She believes there was a policy change in place.
 
>> Ms. Kalish:  I don't have any update on that myself, but it would certainly be something that we could look at. I don't want to go try to find that slide at the very beginning, but to look at sort of the trends in late fees, and that is in the report. So actually, my colleague Isaac, who's on the line, might have that answer. I can't dig it up right now. Sorry. Thank you, though. It's a good question and definitely something that this Making Ends Meet data would be able to help us answer.
 
>>Mr. McDonnell:  Thank you, Emma. 
 
Any other questions? 

[No response.]
 
>>Mr. McDonnell:  Okay. For others, let me type in my direct email address. Please feel free to email me if a question pops up in your head later on. Just pop it on over to me, and I will send it on to Emma and her colleagues and be able to get answers to those questions for you.
 
>> Ms. Kalish:  Absolutely. 
 
>?Mr. McDonnell:  If we have no other questions, again, I'd like to conclude here, and thank you all so very much for your time. We do greatly appreciate it, and you will be hearing more from me in the future as we progress on with our FinEx. Thank you very much, everyone. Take care, and Josh, you may close us on out here.
 
>>Mr. Shapiro:  Thank you all. Have a great afternoon.

2