Is Consumer Loan Growth Creating A Systemic Risk?

Is Consumer Loan Growth Creating A Systemic Risk? Global Structured Finance Research: Darrell Wheeler, New York (1) 212-438-0599; darrell.wheeler@spglobal.com Kirsten R Mccabe, New York 212-438-3196; kirsten.mccabe@spglobal.com Analytical Contact: John Anglim, New York (1) 212-438-2385; john.anglim@spglobal.com Table Of Contents Consumer Debt Isn't An Isolated Figure Evaluating Consumer Loan Sensitivity To Rising Rates Analysis Suggests Multiple Variables Would Have To Deteriorate Consumer Loan Exposures Are Mostly Held In Portfolios As A Whole Loan Credit Risk Current Consumer Leverage Is Manageable, But We Must Watch Growth, Rates, And Other Macroeconomic Factors Related Research WWW.STANDARDANDPOORS.COM/RATINGSDIRECT OCTOBER 26, 2017 1

Is Consumer Loan Growth Creating A Systemic Risk? As the global economic recovery continues into its eighth year, consumer loan balances have continued to rise in many of the world's developed economies. Many would suggest that accelerating levels are somewhat of a callback to 2007, when consumer leverage hit its tipping point. Many factors, such as economic growth and interest rate levels, help determine the consumers' capacity for leverage. However, such a pronounced acceleration in consumer debt growth begs the question of when too much debt poses a systemic risk to the global economy and specifically to structured finance products. Consumer loan exposure across the world's developed countries reached $29.6 trillion at the end of the first quarter of 2017, about $2 trillion greater than the first quarter of 2007 (see chart 1). The U.S. alone accounted for almost half of global consumer debt growth, recently peaking at $14.8 trillion in first-quarter 2017, representing a 10-year increase of $1.4 trillion from first-quarter 2007. Chart 1 WWW.STANDARDANDPOORS.COM/RATINGSDIRECT OCTOBER 26, 2017 2

Consumer Debt Isn't An Isolated Figure Although these debt levels are high, economies and consumer incomes have grown concurrently over the past 10 years along with accelerating debt levels. Thus, in observing the Bank for International Settlements' measure of credit to households and nonprofit institutions serving households (NPISH), as a share of GDP, we see a better picture of loan growth relative to inflation-adjusted economic activity (see chart 2). Since the beginning of the global financial recovery in 2009, consumer debt levels relative to overall GDP have fallen across the U.S., U.K., the eurozone, and Japan. Canadian and Australian markets have been the exception, gradually rising relative to overall GDP at an average annual pace of 2.7% and 1.8%, respectively. China naturally comes to mind when considering expanding consumer debt, with its 11.5% annual growth rate since mid-2009. However, China's ratio of consumer credit to households and NPISH, as a share of GDP, reached just 45.5% in the first quarter of 2017 and has historically trended below that of developed economies. Chart 2 Low global interest rates have helped facilitate consumer debt in the wake of the financial crisis. However, if and when rates begin to accelerate, they can also affect how much debt consumers can endure. To consider the effect of interest rates, we use the household debt service ratio, which reflects the share of income used to pay debt in households, on a WWW.STANDARDANDPOORS.COM/RATINGSDIRECT OCTOBER 26, 2017 3

quarterly basis. Most developed countries have lowered obligations since the start of the recovery, with Australia and Canada having the highest level and least improvement (see chart 3). Chart 3 Chart 3 makes it clear that low rates have been supporting the consumer around the globe. However, it's still unclear how much of an interest rate increase the consumer can withstand before reaching the consumer debt obligations achieved before the 2007 recession. In calibrating our structured finance rating criteria across various consumer products, the unemployment rate is usually the primary driving macroeconomic factor because consumers rarely default while employed. In three recent collateral projection articles, we demonstrated the relationship between unemployment and net loss rates on various consumer loans, under separate recessionary economic scenarios (see Related Research). These scenarios evaluated the unemployment rate along with multiple potential factors that could predict the performance of various loan pools. In these studies, we found that, beyond unemployment, the household debt service ratio is beneficial to our predictions; along with some collateral-specific variables, these regressions can usually predict future losses on consumer loans pretty accurately. Beyond the unemployment and debt service ratio figures, we also considered factors frequently related to the secondary value of the underlying assets (such as used car prices) or how fast revolving debt is accruing. WWW.STANDARDANDPOORS.COM/RATINGSDIRECT OCTOBER 26, 2017 4

Evaluating Consumer Loan Sensitivity To Rising Rates Given that few economists are projecting significant near-term unemployment increases, in considering consumer debt levels, we are left to focus on how rising rates and further debt growth might affect the household debt service ratio, which in the U.S., was over 12% just before the last recession and has fallen to be around 10.% for the past two quarters. Because the U.S. has the largest global consumer exposure, we use it to evaluate loan and rate sensitivity. U.S. debt obligations currently consist of: $9.2 trillion in mortgage debt, which is still well below previous peaks and has an estimated 27-year life; $1.09 trillion in auto loans, which have increased by $300 billion since 2007 as car features and prices have risen. For our analysis, we start with prime loans having an average fixed-rate of 3.93% and subprime having a 16.91% rate, all with an average of 72-month terms; $986 billion in credit card debt, which is in line with previous peak levels, and we adopt the Federal Reserve's assumption of a monthly minimum payment and pay down at a rate of 2%-3% on credit cards balances; $1.4 trillion of student loan debt, which has more than doubled since 2007 and accounts for $800 billion of the post-2007 debt increase. We expect this category to cost 6.8% for 10 years; and $191 billion of other consumer loan debt (which would include unsecured loans), which costs 11.3% and has an average term of 2.2 years. With recent housing rental costs increasing, we thought household rental obligations should be evaluated; however, in looking at the data, we did not see this inflation effect and thus focused on debt-based debt ratios. WWW.STANDARDANDPOORS.COM/RATINGSDIRECT OCTOBER 26, 2017 5

Chart 4 The future costs of these debt obligations depend on the speed at which the rates increase and how often the loan term resets, or in the case of the revolving debt, the total outstanding balance and the resulting minimum payment requirement. In our projections, we increased the short-term rate by 175 basis points over the next 13 quarters to 2020 to match the Federal Reserve's most recent scatter plot of expected rate increases. Using these expected Federal Reserve increases, the household debt service rate only slightly increased to 10.87% by the end of 2018 (see the blue solid line in chart 5) and reached just 11.02% by the end of 2020. This is due partly to longer-term residential mortgages accountting for a large portion of the consumer' debt capacity. In considering mortgage resets, we assumed a 3.1% prepayment rate and that one-seventh of homes turn over annually. Next, to observe conditions that would stress the ratio, we applied a 50-bp rate increase for every planned Federal Reserve bump over the next four years. Even under this stress household debt service ratio only just reaches 11.38% (see the yellow line in chart 5). These muted results suggest that it could be difficult to return to prerecession debt obligation ratios, which ranged from 12.1% to 13.1% in the four years leading up to the recession, simply based on rate increases to existing debt levels; therefore, we started to evaluate potential consumer debt growth. As the U.S. consumer decreased leverage use in the second quarter, we used the five-year average leverage increase for each product in our stress analysis and then applied the two rate scenarios mentioned above. The resulting dotted blue and green lines reached previous debt WWW.STANDARDANDPOORS.COM/RATINGSDIRECT OCTOBER 26, 2017 6

obligation levels starting as early as late 2018, which suggests we will have to continue monitoring overall growth as well as the rate environment. Chart 5 Analysis Suggests Multiple Variables Would Have To Deteriorate In our previous consumer projection articles, we considered other variables depending on the asset type to assess credit risks. Tables 1 and 2 show, for example, the additional variables we use for estimating auto loan and credit card charge-offs, along with our base-case results and those in a 'BBB' stress scenario. The tables demonstrate how far losses could go if a 'BBB' economic downturn occurred. Such risks are sized in the underwriting and rating criteria applied to consumer asset-backed securities (ABS) products and are built into our ratings on these classes (for more information, see "U.S. Credit Card Charge-Offs Will Likely Face An Uphill Climb," published Sept. 6, 2016, and "How Will Auto ABS Pools Perform Through A Credit Cycle?" published March 14, 2017). WWW.STANDARDANDPOORS.COM/RATINGSDIRECT OCTOBER 26, 2017 7

Table 1 How Will Auto ABS Pools Perform Through A Credit Cycle? (Scenario And Forecasts) Unemployment rate (%) Household debt service ratio (%) Manheim Used Vehicle Value Index Prime net loss rate (%) Subprime net loss rate (%) Base-case scenario 2017 4.5 10.2 124.8 0.71 5.66 2018 4.4 10.6 121.5 0.84 6.06 2019 4.4 10.9 117.5 0.97 6.59 2020 4.4 11 115 1.02 6.77 2021 4.4 11 115 1.02 6.77 2022 4.4 11 115 1.04 6.77 'BBB' scenario 2017 4.6 11.3 120.5 0.99 6.96 2018 4.6 13 113.8 1.44 8.9 2019 5.8 12.8 106.7 1.71 9.95 2020 9.3 12.3 113 2.11 11.91 2021 9.6 11.4 120.5 1.89 10.97 2022 8.9 10.7 124.9 1.58 9.6 Table 2 U.S. Credit Card Charge-Offs (Scenario And Forecasts) Unemployment rate (%) Household debt service ratio (%) Revolving credit (bil. $) Personal bankruptcies (000s) Predicted CCQI charge-off rate (%) Predicted FRB charge-off rate (%) Base-case scenario 2017 4.7 10 966 57.95 2.5 3.1 2018 4.5 10 995 57.95 2.9 3.4 2019 4.4 10 1025 57.95 3.3 3.7 2020 4.4 10 1056 57.95 3.6 4 2021 4.4 10 1087 57.95 3.8 4.2 2022 4.4 10 1120 57.95 3.9 4.4 'BBB' scenario 2017 5.1 12.98 987 79.87 2.8 3.5 2018 6.1 12.7 1055 104.58 5 4.9 2019 10 12.33 997 143.12 9.2 9.1 2020 9.9 11.41 907 159.8 8.3 8 2021 9.6 10.7 865 144.08 5.6 5.5 2022 9 10.48 870 118.65 3.5 3.7 CCQI--Credit Card Quality Index. FRB--Federal Reserve Bank. WWW.STANDARDANDPOORS.COM/RATINGSDIRECT OCTOBER 26, 2017 8

Consumer Loan Exposures Are Mostly Held In Portfolios As A Whole Loan Credit Risk Given recent stable conditions, the stresses applied to both the unemployment rate and the household debt service ratio in tables 1 and 2 may seem unrealistic. However, the conditions' accuracy depicts the unexpected deterioration of the previous recession, which should be viewed as a possible outcome. We should also note that while we have only presented analysis of the U.S. household debt ratio and how consumer loans respond to stress, other global markets should respond similarly to consumer debt growth,rate risk, and changes in employment. In fact, on Oct. 24, 2017, our U.K. financial institutions group published "Fast-Growing U.K. Consumer Credit Should Raise Red Flags for Lender," which highlights how today's low yield environment, coupled with the challenges of post-crisis securitization, has left many portfolio lenders exposed to elevated levels of consumer loans. We show a similar concentration trend for U.S. portfolio lenders in table 3. Because the level of outstanding consumer debt has accelerated since 2012, government lending, depository institutions, and insurance companies have picked up significant loan market share, without using securitization. This suggests lenders are comfortable with current consumer credit risks. However, the low securitization utilization in table 3 suggests credit risk, in the form of concentrated portfolios, may be building at various banks, insurance companies, and government and private lenders. Table 3 Consumer Loan Growth And Percentage Securitized Residential mortgages 2001 2004 2007 2010 2013 2016 Second-quarter 2017 Total outstanding balance (bil. $) 5,675 8,292 11,240 10,502 9,946 10,266 10,429 Nonagency MBS (bil. $) 1,014 2,600 1,923 1,239 923 800 Nonagency RMBS securitized (%) 14 25 18 12 9 8 Residential loans held at depository institutions (%) 29 29 27 25 24 24 24 % residential loans held at GSEs plus agency/gse pools 53 46 43 55 60 62 62 Auto loans/leases Total outstanding balance (bil. $) 674 798 801 713 879 1,111 1,093 Auto ABS (bil. $) 167 176 181 116 161 195 201 Securitized (%) 25 22 23 16 18 18 18 Auto loans held at commercial banks (%) 37 38 Credit card debt Total outstanding balance (bil. $) 735 824 1,002 839 858 999 987 Credit Card ABS (bil. $) 268 299 325 217 124 130 133 Securitized (%) 36 36 32 26 15 13 13 Credit card loans held at commercial banks (%) 29 41 31 69 67 68 67 Student loans Total outstanding balance (bil. $) 589 856 1,146 1,407 1,450 Student loan ABS (bil. $) 230 242 230 189 182 Securitized (%) 39 28 20 13 13 MBS--Mortgage-backed securities. RMBS--Residential MBS. GSE--Government-sponsored enterprise. ABS--Asset-backed securities. WWW.STANDARDANDPOORS.COM/RATINGSDIRECT OCTOBER 26, 2017 9

While table 3 shows that securitization utilization is at record lows for almost every consumer loan product, recently originators have started increasing securitization for shorter-term consumer loans, such as autos and credit cards, indicating that they may be becoming more concerned about the risks of rising rates. Securitization for longer-duration residential and commercial mortgages has not increased as those markets have continued to struggle with the securitization challenge of risk retention for longer-term mortgages. Thus, most non-government-sponsored agency mortgage originations have been held by banks or private lenders. Current Consumer Leverage Is Manageable, But We Must Watch Growth, Rates, And Other Macroeconomic Factors Our analysis showed that recent debt levels combined with significant rate increases could raise consumer debt obligations somewhat. But exceeding the previous peak household debt service level would require rate increases that are higher than currently anticipated and/or consumers to continue taking on additional debt. But reaching a peak consumer debt ratio alone wouldn't deteriorate collateral performance because unemployment would also have to increase--an unlikely combination since higher rates usually mean a growing economy. Our previous research has demonstrated that a mix of several economic factors, including leverage, interest rates, personal income, as well as the unemployment rate, would have to progressively worsen to hurt consumer loan performance. For example, unemployment, which is usually the key variable underlying consumer health, would have to increase 8%-10% to duplicate a 'BBB' stress scenario like in the last recession. But these stresses seem unlikely in today's environment, and so we have to focus on rates and global consumer debt levels. Related Research Fast-Growing U.K. Consumer Credit Should Raise Red Flags for Lender, Oct. 24, 2017 How Will Auto ABS Pools Perform Through A Credit Cycle?, March 14, 2017 U.S. Credit Card Charge-Offs Will Likely Face An Uphill Climb, Sept. 6, 2016 The Economic Factors That Affect Housing Prices, Jan. 6, 2016 Only a rating committee may determine a rating action and this report does not constitute a rating action. WWW.STANDARDANDPOORS.COM/RATINGSDIRECT OCTOBER 26, 2017 10

Copyright 2017 by Standard & Poor s Financial Services LLC. All rights reserved. No content (including ratings, credit-related analyses and data, valuations, model, software or other application or output therefrom) or any part thereof (Content) may be modified, reverse engineered, reproduced or distributed in any form by any means, or stored in a database or retrieval system, without the prior written permission of Standard & Poor s Financial Services LLC or its affiliates (collectively, S&P). The Content shall not be used for any unlawful or unauthorized purposes. S&P and any third-party providers, as well as their directors, officers, shareholders, employees or agents (collectively S&P Parties) do not guarantee the accuracy, completeness, timeliness or availability of the Content. S&P Parties are not responsible for any errors or omissions (negligent or otherwise), regardless of the cause, for the results obtained from the use of the Content, or for the security or maintenance of any data input by the user. The Content is provided on an as is basis. S&P PARTIES DISCLAIM ANY AND ALL EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, ANY WARRANTIES OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE OR USE, FREEDOM FROM BUGS, SOFTWARE ERRORS OR DEFECTS, THAT THE CONTENT S FUNCTIONING WILL BE UNINTERRUPTED OR THAT THE CONTENT WILL OPERATE WITH ANY SOFTWARE OR HARDWARE CONFIGURATION. In no event shall S&P Parties be liable to any party for any direct, indirect, incidental, exemplary, compensatory, punitive, special or consequential damages, costs, expenses, legal fees, or losses (including, without limitation, lost income or lost profits and opportunity costs or losses caused by negligence) in connection with any use of the Content even if advised of the possibility of such damages. Credit-related and other analyses, including ratings, and statements in the Content are statements of opinion as of the date they are expressed and not statements of fact. S&P s opinions, analyses and rating acknowledgment decisions (described below) are not recommendations to purchase, hold, or sell any securities or to make any investment decisions, and do not address the suitability of any security. S&P assumes no obligation to update the Content following publication in any form or format. The Content should not be relied on and is not a substitute for the skill, judgment and experience of the user, its management, employees, advisors and/or clients when making investment and other business decisions. S&P does not act as a fiduciary or an investment advisor except where registered as such. While S&P has obtained information from sources it believes to be reliable, S&P does not perform an audit and undertakes no duty of due diligence or independent verification of any information it receives. To the extent that regulatory authorities allow a rating agency to acknowledge in one jurisdiction a rating issued in another jurisdiction for certain regulatory purposes, S&P reserves the right to assign, withdraw or suspend such acknowledgment at any time and in its sole discretion. S&P Parties disclaim any duty whatsoever arising out of the assignment, withdrawal or suspension of an acknowledgment as well as any liability for any damage alleged to have been suffered on account thereof. S&P keeps certain activities of its business units separate from each other in order to preserve the independence and objectivity of their respective activities. As a result, certain business units of S&P may have information that is not available to other S&P business units. S&P has established policies and procedures to maintain the confidentiality of certain non-public information received in connection with each analytical process. S&P may receive compensation for its ratings and certain analyses, normally from issuers or underwriters of securities or from obligors. S&P reserves the right to disseminate its opinions and analyses. S&P's public ratings and analyses are made available on its Web sites, www.standardandpoors.com (free of charge), and www.ratingsdirect.com and www.globalcreditportal.com (subscription), and may be distributed through other means, including via S&P publications and third-party redistributors. Additional information about our ratings fees is available at www.standardandpoors.com/usratingsfees. STANDARD & POOR S, S&P and RATINGSDIRECT are registered trademarks of Standard & Poor s Financial Services LLC. WWW.STANDARDANDPOORS.COM/RATINGSDIRECT OCTOBER 26, 2017 11