Andrew Haughwout, Donghoon Lee, Daniel Mangrum, Jonathan McCarthy, Davide Melcangi, Joelle Scally, and Wilbert van der Klaauw

The strength of consumer spending so far this year has surprised most private forecasters. In this post, we examine the factors behind this strength and the implications for consumption in the coming quarters. First, we revisit the measurement of “excess savings” that households have accumulated since 2020, finding that the estimates of remaining excess savings are very sensitive to assumptions about measurement, estimation period, and trend type, which renders them less useful. We thus broaden the discussion to other aspects of the household balance sheet. Using data from the New York Fed’s Consumer Credit Panel, we calculate the additional cash flows made available for consumption as a result of households’ adjustments to their debt holdings. To detect signs of stress in household financial positions, we examine recent trends in delinquencies and find the evidence to be mixed, suggesting that certain stresses have emerged for some households. In contrast, we find that the New York Fed’s Survey of Consumer Expectations still points to a solid outlook for consumer spending.

Surprising Consumption Strength

Real personal consumption expenditures (PCE) have been remarkably sturdy since the onset of the pandemic, to the surprise of many analysts over much of this period. The surprises have been especially notable over the first half of this year, as real PCE growth has held up in the face of ongoing monetary policy tightening and this spring’s banking system stress. 

To quantify the extent of these surprises, we chart errors in the six-month-ahead Blue Chip Consensus forecasts of quarterly real PCE growth (measured at an annual rate). A positive forecast error (shown in the blue bars) at time t means that the realized growth rate (shown in red) was above the forecasts from six months earlier.

Consumption Has Surprised to the Upside, Especially in 2023

Sources: Bureau of Economic Analysis; Blue Chip Economic Indicators.
Notes: The red line shows the real (inflation-adjusted) quarterly annualized rate of growth for personal consumption expenditures (PCE). The blue bars show the difference between realized PCE growth and the six-month-ahead Blue Chip Consensus forecasts, with positive values indicating that consumption growth was higher than expected.

In the first half of 2021, real PCE grew much faster than predicted, likely due to an unexpectedly fast rollout of vaccines and a larger-than-expected fiscal stimulus–in terms of both magnitude and multiplier effect on consumption. Throughout 2022, however, consumption was weaker than forecasted, probably due to a combination of higher-than-expected inflation, a larger effect on disposable income from the unwinding of pandemic-related fiscal support, and a faster-than-expected tightening of financial conditions.

But in 2023, we’ve seen upside surprises once again, particularly for the first quarter. Moreover, the most recent Blue Chip Consensus forecast for consumption growth in 2023:Q3 is higher than what was expected six months ago. This shift has occurred because many forecasters in the past few months have abandoned their projections of recession and negative consumption growth. We now attempt to understand these forecast errors.

Excess Savings

We begin our discussion with excess savings, which has received a lot of attention from economists and the business press. The idea is that large fiscal transfers and reduced consumption opportunities during the pandemic led households to save more than they otherwise would have done and now those savings may be available to support consumption. There is tremendous uncertainty, however, about how much excess savings still remain in the household sector.

While analysts generally agree that excess savings reached high levels over the course of 2021, significant differences about their recent level have developed; for example, see Aladangady et al., de Soyres et al., and Abdelrahman and Oliveira (Higgins and Klitgaard study excess savings in the international context). The differences in estimates for the United States are attributable to technical factors like the assumed pre-pandemic trend, and different views about whether the savings rate or gross household saving (in dollars) is the appropriate way to think about any excess.

As we move further beyond the pandemic, measuring excess savings becomes increasingly fraught, since it relies heavily on assumptions about behavior in the absence of the pandemic. Consequently, in thinking about the recent resilience of consumption and the implications for the future, a broader assessment of households’ financial positions now seems a more important consideration than excess savings in isolation. In the remainder of the post, we focus on an important element of such an assessment: the role of debt in supporting households’ capacity to sustain consumption.

Household Debt

In addition to savings, households have relatively illiquid assets (like housing) and liabilities (like mortgages and credit card debts) on their balance sheets. The pandemic period featured forbearances on several types of debt, along with large fiscal transfers and very low interest rates, leading to significant improvements in household cash flows. For example, about 14 million households refinanced their mortgages, reducing their mortgage bill by $30 billion per year through 2021. The red line in the next chart shows that the cumulative savings from these lower payments stood at about $120 billion as of 2023:Q2, with recent quarters bringing declines as newer mortgages carry higher balances and higher interest rates.

In addition to these savings, homeowners withdrew unusually large amounts of home equity, primarily in the form of cash-out refinances during the period of low rates. These funds, shown in the blue line below, are also available for consumption and amount to $280 billion in 2023:Q2.

Equity Extraction and Mortgage Refinances Contributed to Liquid Funds Available for Consumption

Source: New York Fed Consumer Credit Panel / Equifax

Other forms of household debt also supported consumption. Payments on student debt, which competes with auto loans to be the second largest household sector liability, have largely been in forbearance since the early stages of the pandemic. Payments on federal student loans prior to the payment moratorium totaled about $70 billion per year, meaning that through 2023:Q2 about $260 billion was left in the household sector; see the blue line in our next chart. By comparison, auto loans (red line) have made relatively small contributions to the funds available for consumption, while some of the funds that households saved have been reflected in reduced credit card balances (gold line).

Credit Card Paydowns Offset Student Loan Forbearance

Source: New York Fed Consumer Credit Panel / Equifax.

In total, mortgages—through equity extraction and lower interest payments—have provided about $400 billion of the excess savings since 2019, and nonmortgage debt has added about $110 billion as the positive cash flow from student loans is partly offset by the negative cash flow of credit cards. Of course, reduced credit card balances position households well for future consumption: since reduced balances typically mean that more credit is available for future use.

Other Indicators of Households’ Financial Health

These positive cash flows from debt suggest that the household sector is in a strong position. Other indicators also support this assessment. Debt delinquencies are generally low, led by remarkably low mortgage delinquencies (shown in gold in the next chart). Auto loan and credit card delinquencies, on the other hand, have risen fairly sharply from their troughs during the pandemic and are now back to their 2019 levels. A key question going forward is whether these delinquency rates will level off or continue to rise. A further increase in delinquencies would indicate that, for at least some households, cash flow has become insufficient to support their financial obligations.

Will Delinquency Rates Continue to Rise?

Source: New York Fed Consumer Credit Panel / Equifax.

As a second set of indicators, we use data from the New York Fed’s Survey of Consumer Expectations to assess households’ near-term expectations regarding their spending, debt delinquency, household income, and earnings growth. Median year-ahead expected spending growth has retreated somewhat from its high 2022 levels, but its current reading of 5.3 percent and six-month average of 5.4 percent remain well above its pre-pandemic level in February 2020 of 3.1 percent.

The same pattern is true for median expected household income growth and median expected earnings growth, which have averaged 3.2 percent and 2.9 percent, respectively, in recent months—well above their six-month averages going into the pandemic (2.7 percent and 2.4 percent, respectively). Consistent with these findings, the median probability of missing a debt payment over the next three months has been relatively low and stable over the past six months at an average of 11.3 percent, compared to a six-month average of 12.2 percent going into the pandemic. 

What’s Next?

Overall, households report solid and stable expectations for spending growth, consistent with our evidence on the strength and liquidity of household balance sheets, including relatively low delinquencies. Of course, the period of very low interest rates that supported many of these developments is decidedly over, at least for now, suggesting that household finances will likely tighten further in the coming months. Additionally, the resumption of student loan payments could have substantial negative effects on vulnerable households. We will return to this important issue in our accompanying post.

How to cite this post:
Andrew Haughwout, Donghoon Lee, Daniel Mangrum, Jonathan McCarthy, Davide Melcangi, Joelle Scally, and Wilbert van der Klaauw, “An Update on the Health of the U.S. Consumer,” Federal Reserve Bank of New York Liberty Street Economics, October 18, 2023, https://libertystreeteconomics.newyorkfed.org/2023/10/an-update-on-the-health-of-the-u-s-consumer/.

recession stimulus monetary policy fed federal reserve mortgages recession interest rates consumer spending gold

Read More

Raji Chakrabarti, Daniel Mangrum, Sasha Thomas, and Wilbert van der Klaauw

After forty-three months of forbearance, the pause on federal student loan payments has ended. Originally enacted at the onset of the COVID-19 pandemic in March 2020, the administrative forbearance and interest waiver lasted until September 1, 2023, and borrowers’ monthly payments resumed this month. As discussed in an accompanying post, the pause on student loan payments afforded borrowers over $260 billion in waived payments throughout the pandemic, supporting borrowers’ consumption and savings over the last three years. In this post, we analyze responses of student loan borrowers to special questions in the August 2023 SCE Household Spending Survey designed to gauge the expected impact of the payment resumption on future spending growth, the risk of credit delinquency for borrowers, and the economy at large. The findings suggest that the payment resumption will have a relatively small overall effect on consumption, on the order of a 0.1 percentage point reduction in aggregate spending from August levels, and a (delayed) return of student loan delinquency rates back to pre-pandemic levels. Across groups, we see little variation in spending responses but find that low-income borrowers, female borrowers, those with less than a bachelor’s degree, and those who were not in repayment before the pandemic expect the highest likelihood of missed student loan payments.

The SCE Household Spending Survey is fielded every four months as a rotating module of the Survey of Consumer Expectations (SCE), which itself is a monthly, nationally representative internet-based survey of a rotating panel of household heads conducted by the Federal Reserve Bank of New York since June 2013. Here, we focus on responses by about 1,000 respondents to a special set of questions added to the August 2023 survey. Of these respondents, 225 reported having outstanding student loans, of which a subset of 151 respondents indicated that their federal student loans were previously “paused” but will be entering repayment in October. The remaining group includes those whose payments were never paused or those who are enrolled in school full-time and not resuming repayment. We asked those borrowers entering repayment how they plan to afford their looming monthly student loan payments and how their probability of missing student and non-student-loan payments will change due to the payment resumption.

We begin by briefly discussing our sample. An overwhelming majority of our sample of student loan borrowers held federal loans (with 74 percent reporting they hold federal loans only and 20 percent reporting they hold both federal and private loans). Of the 151 respondents who will be entering repayment, 71 percent were making monthly payments prior to the payment pause; roughly half of the borrowers in repayment were in a standard (ten-year) repayment plan (36 percent) and half were in an income-driven repayment (IDR) plan (35 percent). About 23 percent of our sample entering repayment were in deferment or forbearance prior to the pandemic, most under in-school deferment. Around 6 percent of ­borrowers were not actively making payments despite payments being required.

Expectations for Income-Driven Repayment Enrollment

We began by asking borrowers if they would enter the standard ten-year repayment plan (the default option) or enroll in an IDR plan. The Biden Administration recently debuted a new IDR plan, the Saving on a Valuable Education (SAVE) plan, that lowered payments for low-income borrowers and has already enrolled over four million borrowers (as of September 5). Our survey results suggest that the appealing terms of the SAVE plan for low-income borrowers will likely increase enrollment in IDR plans. Of those borrowers who were previously in a standard repayment plan, 20 percent expect to enroll in an IDR plan, and 84 percent of those who were previously in an IDR plan expect to remain enrolled in IDR—results that taken together would represent a modest uptick in IDR enrollment among the more seasoned borrowers. Meanwhile, borrowers who were not in repayment prior to the pandemic overwhelmingly favor IDR over the standard payment, with 78 percent of first-time repayers stating an intent to enroll in IDR. As shown by the flows in the chart below, we estimate the IDR enrollment among those in repayment would increase from 50 percent pre-pandemic to 58 percent after payments resume.

The SAVE Plan Will Likely Drive New Interest in Income-Driven Repayment (IDR) for Student Loan Borrowers

Expectations for Changes to Monthly Spending

Next, we turn to borrower’s expectations for changes in monthly spending (separate from student loan payments) due to the resumption of payments. More specifically, we ask borrowers, “When student loan payments resume from October, how do you expect that the payment resumption will affect your average monthly spending in the three months starting with October 2023?” On average, borrowers expect to reduce consumption by around $56 per month from their average monthly spending reported in August. If we scale this monthly decline up to the 28 million borrowers with federally-managed loans currently in forbearance, this would suggest nearly a $1.6 billion decline in monthly spending, or 0.1 percentage point of August 2023 personal consumption expenditures (PCE). For context, average monthly student loan payments for federally-managed loans was around $6 billion prior to the pandemic.

In the chart below, we plot the average reported change in expected October spending for paused borrowers as a share of their August reported average monthly spending. Most groups report relatively small expected reductions in spending while some groups report higher expected future spending despite the resumption of payments (survey panelists without student loans also report higher future spending). These relatively modest consumption declines, although not statistically different from zero, could be because borrowers already began adjusting consumption prior to August or because borrowers plan to reduce and/or deplete savings to make payments. They are also likely to reflect the large share expecting to enroll in the more generous IDR program. Due to our relatively small sample size, 95 percent confidence bands are wide across groups; however, the point estimates with the largest differences are between those with at least a bachelor’s degree (who expect larger spending reductions) and those with less than a bachelor’s degree, potentially reflecting differences in average outstanding student loan balances and payment sizes.

Paused Student Loan Borrowers Only Expect Modest Consumption Declines from August Spending when Payments Resume

Expectations for Missing Student Loan Payments

We also asked paused student loan borrowers about the expected probability (“percent chance”) they would miss a student loan payment or a non-student-debt payment in the three months following the payment resumption. Overall, paused borrowers reported an average probability of missing a student loan debt payment of 22.6 percent. Note that this statistic may overstate expected hardship and payment difficulty. Recent guidance from the U.S. Department of Education informs student loan servicers to not report missed payments to credit bureaus. As such, borrowers may be more likely to voluntarily miss payments while consequences are less severe.

In the chart below, we compare the self-reported probability of missing a student loan payment across several groups, finding stark and statistically significant differences across gender and income. Female respondents reported more than twice the probability of missing a student loan payment at 28.9 percent compared to 12.5 percent for males. Additionally, borrowers with household income lower than $60,000 reported an average probability of missing a payment of nearly 39 percent, compared to 14.3 percent for those with household income above $60,000. Although the estimates are not statistically different, non-white borrowers reported a higher average likelihood of missing a payment than white non-Hispanic borrowers and those without a college degree reported a higher likelihood than those with a degree. Lastly, we see a large difference in expectations for missed payments between borrowers who were in repayment prior to the pause and those who are entering repayment for the first time, with first-time repayers expecting more than twice the likelihood of missed student loan payments.

Expectations for Missed Student Loan Payments Are High, but Similar to Pre-Pandemic Levels

But how do expectations for missed payments compare to payment delinquency before the payment pause? While we do not have an apples-to-apples, pre-pandemic comparison for expectations of student loan missed payments, we can compare this probability with the borrower delinquency rate from our 2022 Student Loan Update, based on credit report data. As shown in the update, in the fourth quarter of 2019, roughly 15 percent of all student loan borrowers were either ninety or more days delinquent or in default. However, the denominator on this delinquency rate includes borrowers not in repayment, a category of borrowers we exclude from this SCE survey sample. Removing the 15.4 million borrowers reported by the Department of Education as not in repayment (that is, in school, grace, deferment, or forbearance) suggests a pre-pandemic delinquency rate of 23 percent (for those in repayment)—a rate quite similar to the self-reported average probability of missing a student loan payment in the SCE survey of 22.6 percent.

Expectations for Missing Non-Student-Loan Payments

Lastly, we asked student loan borrowers to report the expected increase in the likelihood of missing a non-student-debt monthly obligation (such as a mortgage, credit card, or auto loan payment) due to student loan payments restarting. On average, borrowers reported a 11.8 percent increase in the likelihood of missing a non-student debt payment owing to the student debt payment resumption. Borrowers across groups were also much more similar in their expectations of missing payments for other obligations than for student loans, with no evidence of statistical difference between groups. Interestingly, female respondents reported a lower probability of missing a non-student-loan payment than male respondents (although not statistically different). That female respondents report a far higher likelihood of missing student loan payments than males suggests female borrowers may be more likely than male borrowers to prioritize their non-student-loan obligations ahead of student debt if they face difficulties fulfilling all debt obligations.

Conclusion

Consumer spending has been surprisingly strong so far in 2023. However, there is considerable concern about the strength of headwinds stemming from the resumption of student loan payments, with some economic forecasters predicting it could lower consumption growth by as much as 0.8 percentage point. There are also concerns about rising delinquencies as payments resume, perhaps to levels higher than before the pandemic. Our findings here based on expectations survey responses suggest only modest reductions in spending for borrowers entering repayment (of approximately 0.1 percentage point of August PCE) and likelihood of missed student loan payments roughly in line with pre-pandemic levels. One reason for these relatively small effects is that potentially many borrowers already made changes to their savings and consumption decisions after learning that payments would certainly resume in October. The chart below shows some evidence for this hypothesis. Here, we plot the daily deposits at the U.S. Treasury by the Department of Education, of which the overwhelming majority are federal student loan payments. We see that deposits increased after the U.S. Supreme Court decision reversing the broad student loan forgiveness program and continued to rise up until the end of the zero percent interest waiver. This pattern seems consistent with some borrowers electing to make bulk payments against their loans after learning that their loans would not be forgiven and before interest resumed.

Total Daily Education Department Deposits at U.S. Treasury

Another likely reason behind the less-than-dire forecast as the payment pause ends is the strength still apparent in the health of the U.S. consumer. Several policy changes by the White House and Department of Education bode well, too. A large take-up of the new SAVE plan would reduce monthly payments and waive unpaid interest for low-income student loan borrowers, and a one-year “on ramp” for borrowers will ignore missed payments for credit reporting purposes. In addition, more than $127 billion in federal student loans across over 3.6 million borrowers was cancelled or forgiven during the pandemic payment pause. While these factors will make the resumption of payments more smooth than otherwise, and lessen the expected decline in consumption growth, some student loan borrowers will surely struggle managing their debt obligations just as before the pandemic forbearance. Nevertheless, we expect the potential spillover to the broader economy to be limited, and we will continue to monitor developments in the coming months.

Chart data

How to cite this post:
Raji Chakrabarti, Daniel Mangrum, Sasha Thomas, and Wilbert van der Klaauw, “Borrower Expectations for the Return of Student Loan Repayment,” Federal Reserve Bank of New York Liberty Street Economics, October 18, 2023, https://libertystreeteconomics.newyorkfed.org/2023/10/borrower-expectations-for-the-return-of-student-loan-repayment/.

default fed federal reserve white house consumer spending

Read More

Historic price fractal, strong technicals and Bitcoin ETF approval euphoria could keep BTC prices elevated for the rest of 2023.

Bitcoin (BTC) is up over 70% so far in 2023, helped by a banking crisis and hopes about a spot BTC exchange-traded fund (ETF) approval in the United States.

So, will Bitcoin price continue to climb for the rest of 2023? Cointelegraph takes a look at its chances and possible BTC price targets.

Related: BTC price models hint at $130K target after 2024 Bitcoin halving

2020 Bitcoin price fractal eyes $50K

Bitcoin’s ongoing market trend looks very similar to its price action during the 2017 to 2020 period. Therefore, it can repeat a key breakout moment (the greened area) for a decisive bull run, argues popular BTC analyst Stockmoney Lizards.

The 2020 breakout preceded a Bitcoin bull run to its record high of $69,000 in November 2021. If a similar breakout occurs, the analyst sees BTC’s price climbing to the $45,000-$50,000 range by year’s end.

Unlike Bitcoin’s 2020 price rally, however, Bitcoin is currently facing headwinds from the United States Federal Reserve's tightening, resulting in lower liquidity in the market.

Bitcoin's price is down 40% since April 2022 when the Federal Reserve's balance sheet reached its peak.

Standard Chartered also sees Bitcoin at $50K

Standard Chartered also predicts Bitcoin will reach $50,000 by the year's end.

Geoff Kendrick, the global head of research and chief strategist at Standard Chartered, argues that increased miners’ profitability will reduce their need to sell BTC, leading to a lower supply against potential rising demand.

“Increased miner profitability per BTC (Bitcoin) mined means they can sell less while maintaining cash inflows, reducing net BTC supply and pushing BTC prices higher,” Kendrick stated in a report.

Interestingly, the number of Bitcoin held by miners surged during the Ordinals hype in May 2023, stabilizing since.

Bitcoin price technical pattern hints at $32,000

Meanwhile, one Bitcoin technical setup puts its year-end target near $32,000, also a recent price peak. 

Notably, BTC has entered the breakout stage of its prevailing bump-and-run reversal (BARR) Bottom pattern. The BARR Bottom pattern typically resolves after the price breaks above its descending trendline resistance and rises by as much as the pattern’s maximum height, as shown below.

The same BARR Bottom pattern played out accurately in Dogecoin’s case in June 2022. BTC price will be up another 12.75% by the end of 2023 if this pattern plays out as intended.

This article does not contain investment advice or recommendations. Every investment and trading move involves risk, and readers should conduct their own research when making a decision.

bitcoin btc etf

Read More