Is the Consumer Stressing Out?

Viewpoints

February 07, 2018

“I’m tapped out Marv. American Express got a hit man lookin’ for me!”

A classic line from the 1987 movie “Wall Street” continues to age like a fine wine. We would postulate that it may describe the current state of overlevered U.S. consumers, particularly those categorized as subprime. Once again, consumer excess, aided by easy Fed policies that have anchored interest rates at or near its Zero Interest Rate Policy (ZIRP) for almost a decade, have facilitated the accumulation of $3.8 trillion in consumer non-mortgage debt. Consumer debt now accounts for a record 29% share of real disposable income, while the personal savings rate has plummeted to 2.4%, its lowest level since 2005. Consumers are stretched to the hilt to live the high life. Eventually consumers, like Bud Fox in the movie, must earn more or else debt collectors show up.

Currently, consumer non-mortgage debt includes $1.0 trillion of credit card/revolving loans, $1.3 trillion of auto/other loans and $1.5 trillion of student loans. Credit card and auto loans, in particular, have been experiencing accelerating delinquencies. As a result, finance companies/banks have been increasing bad debt provisioning to build balance sheet reserves due to expectations of rising defaults. Exhibit 1 illustrates the highest net charge off rates (NCOs) in years. All major credit card companies and banks are experiencing increasing credit costs from the trough of late 2015.

Exhibit 1: Net Charge Off Rate for Credit Card Companies

Source: TCW, GS Inc.

Higher Industry Net Charge Off and Delinquency Rates

Exhibit 2: Net Charge-Off Rate on Credit Card Loans, All Commercial Banks

Source: FRED

Exhibit 3: Net Charge Off Rate on Credit Card Loans, (Banks Not in Top 100 by Assets)

Source: FRED

The Federal Reserve (FRED) graph in Exhibit 2 above illustrates rising NCOs for the entire U.S. banking universe. NCOs increased from a trough in 4Q15 at 2.9%, which coincidentally was the same quarter the Fed executed its maiden interest rate hike of this cycle. The larger U.S. banks that dominate credit card issuance have focused on prime and super prime consumers post the Great Financial Crisis (GFC), and have enjoyed a prolonged period of low charge off rates concurrent with the Fed’s almost decade long ZIRP. However, since 2015 the Fed has progressively raised interest rates from ZIRP while NCOs at the larger banks have started to rise, albeit off a low base. NCOs were 3.6% at 3Q17, closing in on a 4% rate, a level that matched the end of previous business expansions in 2000 and 2008.

Interestingly, smaller banks (those not in the Top 100 by asset size) are experiencing far more rapid charge off rate deterioration at 7.9% (See Exhibit 3). Is this a precursor to larger banks experiencing much higher loss trends as well or just anomalous? Time will tell. Could the larger bank NCO rise be just another midcycle phenomenon pulled forward by Fed hikes such as in the mid ‘90s? It is possible, but far from certain in our view, given the current accumulation of overall leverage in the financial system expressed by Bank Credit/GDP at 63%, levels last seen in 2008. (See Exhibit 4)

Exhibit 4: U.S. Bank Credit (Loans + Securities)/GDP

Source: FRED

Exhibit 5: Net % of U.S. Banks Tightening Credit Card Lending Standards

Source: FRED

Banks Tighten the “Credit Box”

Rising interest costs and lower credit standards due to stiff competition have resulted in higher credit costs.

Even unsecured personal loans have seen rising credit costs, as FinTechs and non-traditional sources of credit have consolidated loans for consumers. Consumers consolidated borrowing but turned around and re-levered with new refinancing capital instead of prudently de-levering.

Banks have responded by materially tightening their “credit box” or underwriting standards per the Fed’s SLOO Survey. (See Exhibit 5 above). Recently, nearly 10% of banks reported tightening credit card lending standards, ironically the same level that ended the last two business expansions. Will this just be another spurious correlation? It remains to be seen, but at the very least it sends a shot across the bow to the consensus thesis of unshackled consumer strength, further amplified by the recent Tax Cuts and Jobs Act (“TCJA”). A corollary is banks surveyed by the Fed also tightened standards on auto loans & breached the 10% level, which is consistent with action on credit card loans. (See Exhibit 6)

Exhibit 6: Net % of U.S. Banks Tightening Auto Loans Lending Standards

Source: FRED

Growth Math vs. Credit Normalization

Credit card loans have seen substantial growth over the last few years, and not surprisingly credit costs started to rise in 2016. Most of the increase was initially attributed to “growth math” by card-issuing company executives. This is a seasoning phenomenon that occurs when loan growth rates increase and weaker borrowers default. As newer and weaker front-book pools become a larger share of portfolios, the overall NCO rate rises, and would then be expected to fall after the seasoning effect. Recently lenders have been tightening standards, and high growth pool NCO rates are expected to taper. The problem with the growth math thesis is that now even seasoned back-book vintages are posting rising NCO rates. This is a result of normalization of credit which is less idiosyncratic and more a function of economic factors and consumer stress.

Credit Card Assessed Interest Increasing

Credit card assessed interest is the annualized ratio of total finance charges (including late fees) to the total average daily balances against which the charges were assessed, as defined by the Federal Reserve. In 3Q17 the assessed interest charge jumped to 15%, levels similar to previous business cycle peaks. (See Exhibit 7) This jump may be a result of a maxing out process of the consumer as late fees, cash advance and over-limit fees increase and customers tend to borrow more on their credit cards as a last resort. The higher assessed rate could also encapsulate an incremental mix shift within weaker subprime and private label cards. The spread between the assessed rate and stated average APR rate, which largely reflects Fed rate hikes, is widening. Industry credit card limits of $3.5 trillion are close to the previous 2008 peak while utilization rates are still low. (See Exhibit 8)

A widening gap may be indicative of late-cycle dynamics, and potential mean reversion suggests that stressed consumers may start to increase card utilization soon.

Exhibit 7: Credit Card Assessed Interest Rate vs. Stated APR Rate

Source: FRED

Exhibit 8: Credit Card Utilization Rates (LHS) vs. Card Limits in $Trillions (RHS)

Source: NY Fed Consumer Panel/Equifax

Credit Card Loan Binge to Fund Bitcoin Bubble

The recent Bitcoin mania has now spilled into the card space as well. According to LendEDU, a student loan refinancer, 18% of Bitcoin buyers used a credit card, and of those 22% said they didn’t pay off their credit card balances after their purchase. Almost 90% of those buyers planned to pay off balances by selling their Bitcoin, suggesting overly optimistic assumptions of continued price appreciation. The number of Google searches for “Buy Bitcoin with Credit Card” surged in 4Q17, given the exponential Bitcoin price action up +1,400% in 2017. It isn’t unlikely that consumers are also trying to circumvent cash advance limits. They may be buying crypto currencies on their credit cards and then draining cash from bitcoin ATMs for consumption spending. (See Exhibits 9 and 10)

Exhibit 9: Credit Card Loan Growth (YoY Change)

Source: FRED

Exhibit 10: Buy Bitcoin with Credit Card

Source: Google

Subprime Auto Delinquencies Soaring

According to the NY Fed and Equifax, subprime auto lending (<620 FICO score) is a $300 billion market accounting for 27% of total motor vehicle loans. Auto finance companies possess a dominant 70% share of the subprime space. Post the GFC, banks have remained cautious focusing on higher credit quality consumers to grow their share of 760+ FICO customers. (See Exhibit 11). Auto finance companies, on the other hand, have been aggressively weakening underwriting standards as shown in Exhibit 11, including for subprime borrowers. Subprime delinquency rates among auto finance companies have been skyrocketing close to 10%, putting the rate in line with 2009. (See Exhibit 12).

Exhibit 11: Outstanding Loan Balances by Origination Credit Score

Source: New York Fed Consumer Credit Panel/Equifax
Note: Credit Score is Equifax Risk Score 3.0

Exhibit 12: Transition into 90+ Days Delinquent by Origination Credit Score

Source: New York Fed Consumer Credit Panel/Equifax
Note: Credit Score is Equifax Risk Score 3.0

Why are Consumer Credit Costs Rising in a Benign Economic Environment?

Consumer credit costs have been rising in an economic environment characterized by tepid economic growth over the last two years. Full year 2017 Real YoY GDP growth was +2.3%, accelerating from +1.5% in 2016, while unemployment is at trough levels of 4.1%. Several potential drivers include:

Decelerating Jobs and Weak Wage Growth

Employment growth has decelerated over the last few years, wages have been growing at a subdued 2-2.5% rate, while the costs of rent, healthcare, food and other living items have been rising at a faster rate. (See Exhibits 13 and 14). As a result, the savings rate has been declining with new spending financed with more credit borrowing. In fact, the personal savings rate has been surprisingly poor this entire cycle. The CEO of Assurant, Inc. (insurer of mobile devices) recently stated in 2017, in talking about the U.S. consumer: “The reality is half of Americans can’t afford to write a $500 check.” This speaks to the great wealth inequity in the U.S. exacerbated by the Fed’s ZIRP and QE policy.

Exhibit 13: U.S. Employment Growth (YoY Change)

Source: FRED

Exhibit 14: Avg. Hourly Earnings for Production and Non-Supervisory Employees (YoY Change)

Source: FRED

Consumer Credit and its Share of Real Disposable Income are at Record Highs

Consumer credit is running at $3.8 trillion surpassing its 2008 peak by +45%, and accounting for a record 29% share of consumer real disposable income and 19% of nominal GDP. (See Exhibits 15 and 16)

Exhibit 15: Total Consumer Credit Owned & Securitized ($in Billions)

Source: FRED

Exhibit 16: Consumer Credit/GDP (LHS) vs. Consumer Credit/ Real Disposable Income (RHS)

Source: FRED

Fully Loaded Financial Obligation Ratio Trending Higher

The U.S. Consumer Financial Obligations ratio has fallen since the GFC, given lower mortgage balances and lower interest rates. However, upon including healthcare costs the Financial Obligation Ratio increases dramatically making it more difficult to service debt. (See Exhibit 17)

Exhibit 17: Financial Obligations Ratio (FOR)

Source: Federal Reserve, Bureau of Economic Analysis, Morgan Stanley Research estimates
* Household Financial Obligations as a percent of Disposable Personal Income, Percent, Seasonally Adjusted

Higher Interest Rates = Higher Charge Offs

The Fed rate hike cycle, begun in 4Q15, has been increasing consumer debt service costs while measured wage growth persists. As a result, NCOs have risen and the Fed continues to increase interest rates at a rapid pace. (See Exhibit 18). We believe that the rate of change of the Effective Fed Fund’s Rate off a low absolute base is an important determinant of weaker consumer credit metrics, as most consumer debt is keyed off the front end of the yield curve.

Peak Spread between Consumer Confidence, Real PCE and Savings -> Late Cycle Phenomenon?

Consumer confidence is currently soaring to levels not seen since close to previous peaks in financial and economic markets. This may be due to the fact that household net worth to disposable income is at an all-time high of 673% or 2.9 standard deviations above its mean since 1951.

The gap between household net worth and the tangible income economy continues to widen to unprecedented levels (Exhibit 19). The last time we saw this movie the argument was similar to today’s: The equity market was acting as savings for the consumer in the Dot.com Bubble, while the Housing Bubble obliged almost a decade later, abrogating the need for thrift. History hasn’t been kind to this kind of bubble logic. Today, the discrepency within aggregate household wealth has never been more skewed. The majority of households don’t have the wherewithal to participate in the Fed’s wealth effect, and as such bear the brunt of financial stress.

Exhibit 18: Effective Fed Funds Rate vs. U.S. Bank Net Charge Off Rate

Source: FRED

Exhibit 19: Wealth Economy Has Decoupled From Tangible Economy

Source: TCW, FRED

What is disconcerting is that this cycle peak in wealth and confidence is being accompanied by a plummeting savings rate. (See Exhibits 20 and 21). We view this as akin to consumer confidence out-punting its coverage in football parlance. Historically, when consumer confidence is robust in the face of a plummeting savings rate, personal consumption expenditures (70% of GDP) typically roll over thereafter. Given the weak savings rate and jump in leverage the consumer seems vulnerable to further interest rate hikes.

Exhibit 20: OECD Consumer Confidence vs. Personal Savings Rate Confidence Outpunting Coverage --> PCE Declines?

Source: FRED

Exhibit 21: Peak Spread Between Real Personal Consumption Expenditures (PCE) YoY Growth & Personal Savings Rate

Source: FRED

Could Credit Costs Accelerate Substantially?

Jobless claims recently printed at a 45-year low of +216K and have been lower than the key level of +300k for the last 43 months. Claims are now 1.6 standard deviations below their historic mean since 1967. There is a 67% correlation between claims and card NCOs. An OLS regression between NCOs and claims show a strong 45% co-efficient of determination with very strong t-statistics. (See Exhibit 22). If claims mean revert at some point, consumer NCOs should increase, possibly dramatically, given tight historical correlations.

Exhibit 22: Credit Card Net Charge Offs vs. Initial Jobless Claims Strong 67% Correlation 45% of NCO Changes are Explained by Jobless Claims

Source: TCW, FRED

Conclusion

TCW believes there is significant potential that consumer NCOs and credit stress may continue to increase, promulgated by a levered consumer, tightening lending standards, stretched financial obligations ratio, and an aggressive Fed rate hike cycle – but not necessarily in a straight line. In addition, the Fed’s draining of bank reserves as part of its balance sheet reduction plans should make consumer access to liquidity incrementally more difficult over time, as overall monetary aggregates correspondingly decelerate.

At the very least the consensus party lines that “Subprime credit costs are contained” because “This time is different” should be met with skepticism. We believe the aforementioned adages may not age as well as “I’m tapped out Marv. American Express got a hit man lookin’ for me” as we move through the twilight of this seasoned and levered credit cycle.

 

Media+Attachments

Legal Disclosures


This material is for general information purposes only and does not constitute an offer to sell, or a solicitation of an offer to buy, any security. TCW, its officers, directors, employees or clients may have positions in securities or investments mentioned in this publication, which positions may change at any time, without notice. While the information and statistical data contained herein are based on sources believed to be reliable, we do not represent that it is accurate and should not be relied on as such or be the basis for an investment decision. The information contained herein may include preliminary information and/or "forward-looking statements." Due to numerous factors, actual events may differ substantially from those presented. TCW assumes no duty to update any forward-looking statements or opinions in this document. Any opinions expressed herein are current only as of the time made and are subject to change without notice. Past performance is no guarantee of future results. © 2018 TCW