Don’t be deceived by the headlines, consumer credit markets are decidedly normal
Credit card balances are typical and financial distress rates are in their normal range
The Federal Reserve Bank of New York releases its quarterly report on Household Debt and Credit tomorrow. Nominal credit card debt will likely set a new record and, if past years are any guide, we anticipate headlines warning about record debt and rising consumer distress.
In this post, we argue that consumer credit markets are more boring than these headlines suggest. Credit card balances are not abnormally high when inflation-adjusted and are below historical averages when compared to incomes. Financial distress rates, while up from rock bottom during the pandemic, are in their normal range. Although there are some pockets of weakness, as long as the labor market remains strong, financial distress and credit delinquencies will likely continue to be in line with historical norms.
Thanks for reading Briefing Book! Subscribe for free to receive new posts and support our work.
Credit card balances are not abnormally high
Headlines warning about record credit card debt typically cite data on nominal – that is, not inflation-adjusted – credit card balances. Adjusting for inflation, credit card balances are fairly typical (see figure). Although they have increased since 2022, aggregate credit card balances are roughly back to their 2019 level and close to their average over 2003 to 2023.
Moreover, compared to incomes, credit card balances are well below their historical average. The figure below shows credit card balances per person as a percentage of mean income. At 6.3%, the percentage is below its historical average (7.3%) and fairly close to the series low (5.2% in 2021). The pattern is similar if we use other income or population measures.1
Finally, the numbers the media cites as credit card debt are more accurately termed card balances. Federal Reserve Bank of Philadelphia data indicates that roughly 30% of credit card balances are held by “transactors” who pay off their full balance every month and are not charged interest. Using credit cards this way is more similar to using a debit card than taking out a loan. Only about 70% of the aggregate balances are held by “revolvers” who do not pay off the full balance and incur interest charges. Indeed, part of the jump in fourth quarter balances will likely be seasonality: a bump in holiday spending, which will inflate overall balances, but will subside in the first quarter as people pay their credit card bills (see the scalloped pattern in the first plot).
Financial distress rates are in their normal range
While generally up from rock bottom levels during the pandemic, measures of financial distress are at or below the midpoints of their historical range. The plot below shows the percentage of balances that are more than 90 days delinquent for different types of consumer loans, with the bars showing the historical range from 2003 through Q3 2023, the vertical dashes showing values from right before the pandemic, and the diamonds showing the most recent value from Q3 2023.
Credit card delinquencies, at 9.4% of balances, are slightly below the midpoint of their historical range (10.4%). They are up from a record low of 7.6% during the pandemic and slightly above their pre-pandemic level (8.4% in Q4 2019).
Delinquencies on mortgages and home equity lines of credit (HELOCs) are near their 2003-2023 lows. The percentage of delinquent mortgage balances is 0.5% versus a low of 0.4%. The percentage of delinquent HELOC balances is 0.7% versus a low of 0.2%. The especially low delinquency rates for housing reflect safeguards put in place after the Housing Crisis that reduced lending to the highest-risk borrowers. They also stem from brisk house price appreciation since the start of the pandemic which has left most borrowers with a considerable cushion between their home price and mortgage balance.
Auto loan delinquencies are at the midpoint of the historical range. The percentage of delinquent balances was rising before the pandemic, peaked at 5.1% in Q1 2020, and has eased to 3.9% based on the most current data (identical to the midpoint at one decimal place). The elevated delinquencies, compared to other loan types, in part reflects pandemic supply chain disruptions that caused large runups in the price of new and used vehicles. That, combined with more aggressive lending and higher interest rates, raised monthly loan payments. Subprime and deep subprime auto loans made in 2022 have been entering default at faster rates than similar loans made in prior years.
Student loan delinquencies are near zero because federal student loans—around 90% of student loans—are not yet being reported delinquent. Federal student loans resumed payments this past fall after a three-year pandemic pause. Delinquencies will start being reported in September 2024.
Buy-now-pay-later delinquencies are not generally reported to the credit bureaus, and we do not know of a source that tracks them in an up-to-date, comprehensive manner. While growing rapidly, the BNPL market is still small. A CFPB report found that BNPL loan volume was $24 billion in 2021 or about 0.5% of credit card purchase volume in that year.2
A strong labor market provides a safeguard against emerging risks
Historically, labor market weakness has been an important determinant of consumer financial distress. Aggregate time series data show a correspondence between jumps in the unemployment rate and elevated delinquency and default. The notable exception was the COVID pandemic when unemployment spiked but reduced consumer spending and government aid reduced financial distress. Research using microdata indicates that cash flow shocks – such as job loss – are much more important than strategic factors in explaining mortgage defaults.
As our colleague Kevin Rinz argued in this newsletter two weeks ago, the labor market is strong and well-positioned to maintain this strength in the year ahead. Absent any unexpected deterioration in the labor market, it is hard to see a broad-based and significant increase in consumer financial distress.
Another encouraging factor is the low rate of mortgage defaults. The mortgage market accounts for 70% of consumer borrowing, and as we learned during the 2008-2009 housing crises, widespread mortgage defaults can drag down the entire economy. Yet post-financial-crisis safeguards that limit risky lending, pandemic forbearance policies, and home price appreciation that provides an equity cushion make it hard to envision any serious downturn in this market where most loans are still at very low interest rates.
One risk factor to watch is the drawdown of pandemic savings. Reduced spending and government aid that kept incomes from falling allowed most people to increase liquid savings and reduce debt in 2020-2021. Research using aggregate savings and individual bank accounts suggests these excess savings are dwindling and will likely run out in 2024. While consumers will have less of a cushion, the impact shouldn’t be overstated: The drawdown will leave the average household no worse off than they were in 2019.
Student loans are another area to watch. Around 10% of student loan balances were delinquent before the pandemic (see figure). While most borrowers have resumed payments without problems and extensive accommodations exist, some are having difficulty. New income-driven repayment plans and better-administered public service loan forgiveness should ease payments for struggling borrowers.
News stories have pointed to rising buy-now-pay-later delinquencies as a warning sign for credit markets. While growing briskly, BNPL is still too small to materially influence credit market aggregates. It is too soon to know whether the BNPL market – with more indebted, lower credit score borrowers – can provide an early indicator of broader issues in credit markets. If BNPL continues to grow, it will be interesting to track the extent to which it substitutes for credit card debt versus expanding total borrowing.
Three cheers for boring
We are the first to admit that an article on the normalcy of consumer credit markets is not optimized for clicks or shares on social media. But we think that precisely because the media has a track record of highlighting bad news, it’s useful to step back and take an inflation-adjusted, historically-informed perspective on the consumer credit market.
Today, consumer credit markets are, frankly, boring. Credit card balances are at normal levels and measures of financial distress, although up from rock bottom during the pandemic, are best thought of as returning to normal after a pandemic-induced increase in financial security.
It is hard to predict upturns in consumer financial distress – Hemingway’s Mike Campbell famously quipped that he went bankrupt “gradually and then suddenly” – and there are pockets of weakness and uncertainty. Still, the strong labor market, with near-record-low unemployment rates for Blacks and Hispanics and especially strong real wage growth for low-income workers, provides an important safeguard against financial distress. Given the large toll of financial distress on households and the economy, we can only hope that the boring times will continue.
The views expressed here are those of the authors and not necessarily those of their employers.
The most recent mean income data is from 2022. We project values for 2023 by conservatively assuming that nominal mean income grew at the same rate as inflation (measured with the CPI-U). If there was real mean income growth, as other sources suggest, the 2023 credit card balances-to-income ratio would be reduced. The share of the population with a credit card has been relatively stable over the last 20 years. If we normalized aggregate credit card balances by cardholders, rather than 16+ year olds, the percentage would shift upwards but the trend would be similar.
The typical BNPL product is offered at the point of sale and allows the consumer to split the purchase into four interest-free installments over a six-week period. BNPL loan volume is a flow that is more comparable to the flow of credit card purchase volume than the stock of credit card balances. The Federal Reserve estimates that total credit card purchase volume was $4.88 trillion in 2021.