Quarterly Snapshot on China's Banking Sector

Quarterly Snapshot on China’s Banking Sector

Fourth Quarter of 2020

March 2021 Contents

 Business Position  Capital and Earnings  Risk Position  Funding & Liquidity  Government Influence  Industry Outlook  Special Topic 1: Relationship between Bond Defaults and Bank Asset Quality  Special Topic 2: Shadow Banking’s Impact on the Banking Sector  Appendix: Related FI Research

China’s robust economic recovery underpins our stable credit outlook for the banking sector

Gross Domestic Product (GDP) 30,000 8

4 20,000

10,000 (4)

0 (8)

(Bil. RMB) (%)

Quarterly Real GDP YoY Growth of Quarterly Real GDP (Right Axis)

Aggregate financing to the real economy (AFRE) maintained high growth in 2020 but as the economy moves beyond COVID, we expect lending growth to moderate and leverage levels to further stabilize

Aggregate Financing to the Real Economy (AFRE) 300 15 2020.09 2020.10 2020.11 2020.12

AFRE 280.05 281.25 283.25 284.83 (Tril. RMB) 250 12.5 . RMB) Tril ( AFRE Growth Rate YoY 13.5 13.7 13.6 13.3 (%)

AFRE AFRE Growth Rate (Right Axis)

Shadow banking activities continued to contract in Q4

AFRE Structure as of end of 2020 YoY Growth Rate of Major AFRE Components Entrusted and 50 Trust Loans 6.1% 40

30 Corporate Bonds 20 9.7% 10 RMB Loans Government 60.2% Bonds 0 16.2% -10

-20 2019Q1 2019Q2 2019Q3 2019Q4 2020Q1 2020Q2 2020Q3 2020Q4 Domestic Equity (%) Financing of Non- RMB Loans Entrusted Loans Trust Loans financial Companies Others 2.9% Corporate Bonds Government Bonds 4.9% Note: Others include foreign currency loans, undiscounted banks’ acceptance bills and other financing instruments. Source: PBOC, collected and adjusted by S&P Global (China) Ratings. Source: PBOC, collected and adjusted by S&P Global (China) Ratings. Copyright © by 2021 S&P Ratings (China) Co., Ltd. All rights reserved. Copyright © by 2021 S&P Ratings (China) Co., Ltd. All rights reserved.

While lending growth remained stable for corporate and household sectors, loans to non-bank financial institutions contracted in Q4

RMB Loan Breakdown as of End of 2020 YoY Growth Rate of RMB Loans Corporate and Public Institution Loans 40 62.8% Non-banking Financial 20 Institution Loans 0.3% 0 Oversea (20) Loans 0.4% (40)

(60) 2019Q1 2019Q2 2019Q3 2019Q4 2020Q1 2020Q2 2020Q3 2020Q4 (%)

Retail Loans Corporate and Public Institution Loans Retail Loans Non-banking Financial Institution Loans 36.6% Oversea Loans

Source: PBOC, collected and adjusted by S&P Global (China) Ratings. Source: PBOC, collected and adjusted by S&P Global (China) Ratings. Copyright © by 2021 S&P Ratings (China) Co., Ltd. All rights reserved. Copyright © by 2021 S&P Ratings (China) Co., Ltd. All rights reserved.

The banking sector’s asset growth moderated in Q4 As of the end of 2020, total assets of the banking sector were 319.7 trillion RMB, up by 10.1% YOY; total liabilities were 293.1 trillion RMB, up 10.2% YOY.

Asset Banking Sector Asset Growth YoY 2019 2020 2020 2020 2020 Growth 30 Q4 Q1 Q2 Q3 Q4 YoY (%) 25 Industry 9.1 10.8 11.0 12.0 11.5 20 average

15 Mega banks 8.2 10.4 10.4 11.3 10.9 10 Joint-stock 5 10.2 12.8 11.8 12.8 11.9 banks 0 City 8.5 8.3 10.0 11.7 10.2 banks (%) 2016-06 2016-09 2016-12 2017-03 2017-06 2017-09 2017-12 2018-03 2018-06 2018-09 2018-12 2019-03 2019-06 2019-09 2019-12 2020-03 2020-06 2020-09 2020-12 Rural Banking Sector Mega Banks Joint-stock Banks financial 7.6 8.2 9.3 10.6 11.6 City Banks Rural Banks institutions

Source: CBIRC, collected and adjusted by S&P Global (China) Ratings. Source: CBIRC, collected and adjusted by S&P Global (China) Ratings. Copyright © by 2021 S&P Ratings (China) Co., Ltd. All rights reserved. Copyright © by 2021 S&P Ratings (China) Co., Ltd. All rights reserved.

Loan demand in the real economy, particularly from small businesses, increased significantly amid COVID; with the pandemic largely under control, overall demand moderated in Q4 Loan Demand Index Published by PBOC 80

(%) Loan Demand Index Loan Demand Index: Manufacturing Loan Demand Index: Infrastructure Loan Demand Index: Large-scale Enterprise Loan Demand Index: Medium-scale Enterprise Loan Demand Index: Small-scale Enterprise

Note: The loan demand index reflects surveyed bankers’ opinions on the change of overall loan demand. Source: PBOC, collected and adjusted by S&P Global (China) Ratings. Copyright © by 2021 S&P Ratings (China) Co., Ltd. All rights reserved.

In 2020, mega banks increased their loans to micro and small enterprises (“MSE”) by almost 50% YoY, as overall loans to MSEs increased by over 30% YoY

Loans to Small and Micro Enterprises (MSE) MSE loan growth 2019 2020 2020 2020 2020 16 compared to Q4 Q1 Q2 Q3 Q4 the previous quarter (%) Industry 3.17 7.60 9.37 7.51 3.43 average 8 Mega banks 3.06 15.19 13.59 10.69 2.45 Joint-stock 3.63 3.35 6.38 8.79 7.02 banks

City banks 4.64 5.66 9.23 6.28 3.81 0 2019-03 2019-06 2019-09 2019-12 2020-03 2020-06 2020-09 2020-12 Rural banks 2.08 5.24 7.01 4.08 2.24 (Tril. RMB) Mega Banks Joint-stock Banks City Banks Rural Banks Note: MSE loans are defined by CBIRC as loans less than RMB 10 million Note: MSE loans are defined by CBIRC as loans less than RMB 10 million Source: PBOC, collected and adjusted by S&P Global (China) Ratings. Source: PBOC, collected and adjusted by S&P Global (China) Ratings. Copyright © by 2021 S&P Ratings (China) Co., Ltd. All rights reserved. Copyright © by 2021 S&P Ratings (China) Co., Ltd. All rights reserved.

New regulations saw growth of property development loans and mortgage loans slow down in Q4

Growth of Mortgage and Real Estate Development  In December 2020, PBOC and CBIRC issued a new rule on controlling banks’ 40,000 20 Loans concentration risk in real estate loans.  The new rule set up different real estate loan concentration limits for different kinds of banks. For example, large banks’ 15 limit for real estate-related loans is 40% of their loan book, within which, the limit for 20,000 mortgage loans is 32.5%. For medium- sized banks, the limits are 27.5% for total 10 real estate loans and 20% for mortgage. In general, the smaller the bank the lower the limit ratio.  Currently, most commercial banks can meet these requirements. The few banks 0 5 that exceed the limits will have a 2019Q1 2019Q2 2019Q3 2019Q4 2020Q1 2020Q2 2020Q3 2020Q4 transitional period ranging from 2 to 4 (Bil.) (%) Mortgage years to bring their real estate portfolio within the limits. Real Estate Development Loans  We believe this new rule will slow down real Growth Rate of Mortgage YOY (Right Axis) estate loan book growth in 2021, and banks Growth Rate of Real Estate Development Loans YOY (Right Axis) need to seek lending growth opportunities in other sectors. Source: PBOC, collected and adjusted by S&P Global (China) Ratings. Copyright © by 2021 S&P Ratings (China) Co., Ltd. All rights reserved.

In 2020 RMB deposits increased rapidly, which was positive for corporates and households’ liquidity management during the COVID period

YoY Growth YoY Growth  In January 2021,banking regulators Deposits Breakdown of from 2019 to from 2018 to issued a new policy on internet-based As of end of 2020 (Bil. RMB) Deposits (%) 2020 2019 retail deposit business. (%) (%)  The new policy tightens the regulation on deposit pricing. It also confines Deposits of Households 92,599 43.56 13.89 13.54 regional banks’ internet-based deposit Deposits of Non- taking activities to their home regions. 66,018 31.06 10.89 5.75 financial Corporates Online deposit business can only be Deposits of conducted on banks’ proprietary on- Government 29,874 14.05 0.64 4.13 line platforms, and not third-party Departments & platforms. Organizations  The policy requires banks to improve Fiscal Deposits 4,477 2.11 9.63 0.74 their asset-liability management and Deposits of Non-bank liquidity management related to 18,311 8.61 6.81 7.28 Financial Institutions online deposits and control their deposit costs within a reasonable Overseas Deposits 1,294 0.61 14.45 4.39 level.  Total Deposits 212,572 100.00 10.21 8.65 This policy further confines regional banks’ activities to their home regions, Source: PBOC, collected and adjusted by S&P Global (China) Ratings. Copyright © by 2021 S&P Ratings (China) Co., Ltd. All rights reserved. and reduces small banks’ potential liquidity risk related to high-cost internet-based retail deposits.

Contents

The issuance of hybrid bonds and slower capital consumption led to a mild increase of capital adequacy ratios in Q4

Reported Capital Adequacy Ratio (CAR) 22 CAR 2019 2020 2020 2020 2020 (%) Q4 Q1 Q2 Q3 Q4 20 Industry average 14.64 14.53 14.21 14.41 14.70 18

Mega banks 16.31 16.14 15.92 16.25 16.49 16

Joint-stock 14 13.42 13.44 12.92 13.30 13.60 banks 12 City banks 12.70 12.65 12.56 12.44 12.99 10 (%) Rural banks 13.13 12.81 12.23 12.11 12.37

Foreign bank 18.40 18.43 18.76 18.56 18.32 Industry Average Mega Banks subsidiaries Joint-stock Banks City Banks Rural Banks Foreign Bank Subsidiaries Source: CBIRC, collected and adjusted by S&P Global (China) Ratings. Source: CBIRC, collected and adjusted by S&P Global (China) Ratings. Copyright © by 2021 S&P Ratings (China) Co., Ltd. All rights reserved. Copyright © by 2021 S&P Ratings (China) Co., Ltd. All rights reserved.

The issuance of perpetual bonds led to an increase of tier-1 capital/total capital ratio in Q4

Reported Capital Adequacy Ratios (CAR) Capital 2019 2020 2020 2020 2020 15 quality Q4 Q1 Q2 Q3 Q4 metrics (%)

Core tier-1 12 capital / Total 74.59 74.88 73.68 72.45 72.93 capital

Tier-1 capital 81.63 82.17 81.70 80.99 81.90 / Total capital 9 (%) Tier-2 capital 18.37 17.83 18.30 19.01 18.10 / Total capital Total CAR Tier 1 CAR Core Tier 1 CAR

Hybrid bonds have been very important instruments for banks to maintain stable capitalization, and regional banks’ hybrid bond issuance increased in 2020

Bank Hybrid Bond Issuance by Type of Bonds Bank Hybrid Bond Issuance by Type of Banks 1,400 1,400

700 700

0 0 (Bil. 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 RMB) (Bil. RMB) Subordinated Bond Tier-2 Capital Bond Perpetual Bond Mega Banks Joint-Stock Banks City Banks Rural Banks Source: Wind, collected and adjusted by S&P Global (China) Ratings Copyright © by 2021 S&P Ratings (China) Co., Ltd. All rights reserved. Foreign Bank Subsidiaries Private Banks Source: Wind, collected and adjusted by S&P Global (China) Ratings Copyright © by 2021 S&P Ratings (China) Co., Ltd. All rights reserved. 15

Despite the principal write-down of Baoshang Bank’s hybrid bonds, the hybrid bond market remains open to regional banks Hybrid bond Issuance of Mega & Joint-stock Banks 6.50 300 6.00 5.50 200 5.00 . RMB) 4.50 (%) Bil

( 100 4.00 3.50 0 3.00

Issuance Amount Average Coupon Rate(Right-Axis) Hybrid bond Issuance of Regional Banks 350 6.50 300 6.00 250 5.50 200 5.00 . RMB) 150 4.50 (%) Bil

( 100 4.00 50 3.50 0 3.00

Issuance Amount Average Coupon Rate(Right-Axis)

The issuance of perpetual bonds with common equity conversion features is an attempt to mitigate investors’ concerns over hybrid bond recovery rates Hypothetical Example of Capital Instrument Notching  Previously, hybrid bonds issued by Chinese banks Possible Rating Outcome typically had principal write-down features rather than a common equity conversion feature. As a result, SACP aspc investors of Baoshang Bank’s tier-2 capital bond Government Support +3 Notches suffered a total loss without any residual claim due to the write-down. ICR AAspc

 Recently, the regulator approved the issuance of hybrid Senior unsecured bond AAspc bonds with common equity conversion features for two Perpetual Bond BBB regional banks, Chouzhou Bank and Ningbo Commerce spc Bank. Tier-2 Capital bond( With Government Aspc+ Support)  In January 2021, the first perpetual bond with a Tier-2 Capital

common equity conversion feature was issued in China’s bond( Without BBBspc+ interbank market by Ningbo Commerce Bank. The bond Government Support) Note: We typically don’t assume perpetual bonds will receive government support. In was 500 million RMB, with a coupon rate of 4.80%. contrast, government support may or may not be available to tier-2 capital bond. Case by case evaluation for specific bonds needs to be carried out in such cases. Therefore, the final notching adjustment may or may not incorporate extraordinary government support. Source: S&P Global (China) Ratings. Copyright © by 2021 S&P Ratings (China) Co., Ltd. All rights reserved. 17

Assuming the migration rate of special mention loans (“SML”) equals 30%, banks with a loan loss reserve /(NPLs + SMLs) ratio lower than 60% may be under capital erosion pressure

Loan Loss Reserve /(NPLs + SMLs) of Major Regional Banks as of End of 2019 40 Eastern China Northern China Central China Southern China Northeast China Southwest China Northwest China

Number Regional Banks of 0 <10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 160 170 180 190 200 >200 Loan Loss Reserve /(NPLs + SMLs) (%)

Eastern China Northern China Central China Southern China Northeast China Southwest China Northwest China 6,000 5,000 RMB) 4,000 bil 3,000 2,000 1,000 Asset Size( 0 <10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 160 170 180 190 200 >200 Loan Loss Reserve /(NPLs + SMLs) (%) Sources: Wind, S&P Global (China) Ratings. Copyright ©2021 by S&P Ratings (China) Co., Ltd. All rights reserved. 18

Results of stress testing conducted by PBOC show that some regional banks are facing capital pressure caused by bad debts

 According to the “China Financial Stability Report 2020”, 1,520 small and mid-sized banks participating in the testing had an overall NPL ratio of 2.99% and an overall capital adequacy ratio of 12.76% as of the end of the first quarter of 2020. The capital adequacy ratio and reserve coverage ratio could still hold up to 10.5% and 100% respectively if the overall NPL ratio rises by 4.55 percentage points to 7.54%.

 If the NPL ratio rises 100%, 200% and 400% respectively, the overall capital adequacy ratio would fall to 11.54%, 9.51% and 5.16% respectively. 589, 786 and 977 banks would fail the stress testing respectively, accounting for 23.55%, 40.30% and 62.56% of the small and mid-sized banks tested by assets.

 The stress testing of the PBOC pays special attention to the risks associated with customer concentration, off- balance sheet business, local government related debts, real estate loans, etc.

 As of the end of the first quarter of 2020, the average NPL ratio of small and mid-sized banks was 2.99%. Assuming an annualized migration rate of SMLs of 30%, a YoY loan growth rate of 10%, and an annual write-off ratio of 1%, the NPL ratio as of the end of the first quarter of 2021 would still remain at the 2.9% level, which is way below the 7.54% level, indicating controllable risk.

Net interest margins remained stable in Q4

Net Interest Margin (NIM) NIM 2019 2020 2020 2020 2020 3.50 (%) Q4 Q1 Q2 Q3 Q4

3.00 Industry average 2.20 2.10 2.09 2.09 2.10

2.50 Mega banks 2.12 2.04 2.03 2.03 2.05

Joint-stock 2.12 2.09 2.08 2.09 2.07 2.00 banks

City banks 2.09 2.00 2.00 1.99 2.00 1.50 (%) Rural banks 2.81 2.44 2.42 2.43 2.49

Industry Average Mega Banks Foreign bank 1.78 1.70 1.67 1.64 1.61 Joint-stock Banks City Banks subsidiaries

In our view, since NIMs were relatively stable, the main reason for falling earnings was increased credit cost in Q4

Return on Assets (ROA) 1.80 ROA 2019 2020 2020 2020 2020 (%) Q4 Q1 Q2 Q3 Q4

1.40 Industry average 1.02 0.98 0.83 0.80 0.77

1.00 Mega banks 1.07 1.02 0.88 0.88 0.89

0.60 Joint-stock banks 1.01 0.99 0.81 0.81 0.75

0.20 City banks 0.89 0.81 0.74 0.66 0.55

(%)

2014-03 2014-07 2014-11 2015-03 2015-07 2015-11 2016-03 2016-07 2016-11 2017-03 2017-07 2017-11 2018-03 2018-07 2018-11 2019-03 2019-07 2019-11 2020-03 2020-07 2020-11 Rural banks 1.07 0.98 0.76 0.72 0.62 Industry Average Mega Banks Joint-stock Banks City Banks Foreign bank 0.64 0.86 0.73 0.57 0.47 Rural Banks Foreign Bank Subsidiaries subsidiaries Source: CBIRC, collected and adjusted by S&P Global (China) Ratings. Copyright © by 2021 S&P Ratings (China) Co., Ltd. All rights reserved. Source: CBIRC, collected and adjusted by S&P Global (China) Ratings. Copyright © by 2021 S&P Ratings (China) Co., Ltd. All rights reserved.

Affected by rising credit cost and the seasonable increase of cost-to- income ratio, the annualized ROE of commercial banks dropped from 10.05% in Q3 to 9.48% in Q4 Cost-to-income Ratio and Return on Equity (ROE) of the Commercial Banking Industry 35 25

30 20 25

20 15

15 10 10 5 5

0 0 (%) (%) 2014-03 2014-06 2014-09 2014-12 2015-03 2015-06 2015-09 2015-12 2016-03 2016-06 2016-09 2016-12 2017-03 2017-06 2017-09 2017-12 2018-03 2018-06 2018-09 2018-12 2019-03 2019-06 2019-09 2019-12 2020-03 2020-06 2020-09 2020-12

Cost-to-income Ratio Return on Equity (Right Axis)

Contents

Sector-wide non-performing loan (NPL) and special mention loan (SML) ratios both dropped in Q4

Loan 2019 2020 2020 2020 2020Q NPL and SML Amounts in China’s Banking Classification Q4 Q1 Q2 Q3 4 Sector (%) 4,500 Normal Loans 95.23 95.12 95.31 95.38 95.59 4,000 Special Mention 2.91 2.97 2.75 2.66 2.57 3,500 Loans (SML) Non-performing 1.86 1.91 1.94 1.96 1.84 3,000 Loans (NPL) - Substandard 2,500 0.78 0.84 0.92 0.94 0.87 Loans

2,000 - Doubtful 0.78 0.78 0.76 0.74 0.72 Loans 1,500 (Bil.) - Loss Loans 0.30 0.29 0.27 0.28 0.25

Special Mention Loans Nonperforming Loans Total 100 100 100 100 100

Regional banks’ NPL ratios decreased due to the bad debt write-off in Q4

NPL Ratio 2019 2020 2020 2020 2020 Non-performing Loan (NPL) Ratios (%) Q4 Q1 Q2 Q3 Q4 4.50 Industry 1.86 1.91 1.94 1.96 1.84 average 3.50

Mega banks 1.38 1.39 1.45 1.50 1.52 2.50

Joint-stock 1.64 1.64 1.63 1.63 1.50 1.50 banks

City banks 2.32 2.45 2.30 2.28 1.81 0.50 (%) Rural banks 3.90 4.09 4.22 4.17 3.88

Industry Average Mega Banks Foreign bank Joint-stock Banks City Banks 0.67 0.71 0.69 0.67 0.58 subsidiaries Rural Banks Foreign Bank Subsidiaries

The SML migration rate of state-owned mega banks increased marginally during the COVID period, while that of joint-stock banks declined

Average Migration Rate of SMLs 60

0 2015Y 2016Y 2017Y 2018Y 2019Y 2020H1 (%) Mega Banks Joint-stock Banks

Note 1: The annual data sample of joint-stock banks does not include Guangfa Bank and Hengfeng Bank. The data as of the end of June 2020 does not include Guangfa Bank, Hengfeng Bank and Bohai Bank. Note 2: Migration rate of SMLs= downward migration amount of SMLs at the beginning of the period/(SMLs at the beginning of the period - decrease of SMLs within the period) * 100%. Source: Wind, collected and adjusted by S&P Global (China) Ratings. Copyright © by 2021 S&P Ratings (China) Co., Ltd. All rights reserved.

Joint stock banks’ declining SML migration rate is related to changes made to their asset credit risk classification policy Migration Rate of SMLs of Mega Banks Migration Rate of SMLs of Major Joint-stock

90 90 Banks 80 80 70 70 60 60 50 50 40 40 30 30 20 20 10 10 0 0 2015Y 2016Y 2017Y 2018Y 2019Y 2020H1 2015Y 2016Y 2017Y 2018Y 2019Y 2020H1 CITIC Bk CMB CMBC CIB SPDB (%) (%) ICBC CCB ABC BOC PSBC BoCom PAB CEB HXB CZB CBHB

Note 1: Migration rate of SMLs= downward migration amount of SMLs at the beginning of the period/(SMLs at the beginning of the period - decrease of SMLs within the period) * 100%. Note 2: CMB-China Merchants Bank, CIB-Industrial Bank, SPDB-Shanghai Pudong Development Bank, CITIC Bk-China CITIC Bank, CMBC-China Minsheng Banking Corp., CEB-China Everbright Bank, PAB-Ping An Bank, HXB-Hua Xia Bank, CGB-China Guangfa Bank, CZB-China Zheshang Bank, CBHB-China Bohai Bank, HFB-Hengfeng Bank. Source: Wind, collected and adjusted by S&P Global (China) Ratings. Copyright © by 2021 S&P Ratings (China) Co., Ltd. All rights reserved.

Regional banks’ SML migration rate has been a lot more volatile than that of state-owned mega banks Migration Rate of SMLs of City Banks Migration Rate of SMLs of Rural Banks

90 90 80 80 70 70 60 60 50 50 40 40 30 30 20 20 10 10 0 0 2015Y 2016Y 2017Y 2018Y 2019Y 2020H1 2015Y 2016Y 2017Y 2018Y 2019Y 2020H1 (%) (%) Bank of Beijing Bank of Shanghai Bank of Nanjing Chongqing Rural Qingdao Rural Zijin Rural Bank of Ningbo Bank of Hangzhou Bank of Chongqing Changshu Rural Wuxi Rural Jiangyin Rural

Note: Migration rate of SMLs= downward migration amount of SMLs at the beginning of the period/(SMLs at the beginning of the period - decrease of SMLs within the period) * 100%. Source: Wind, collected and adjusted by S&P Global (China) Ratings. Copyright © by 2021 S&P Ratings (China) Co., Ltd. All rights reserved.

Reserve coverage ratios of city banks and foreign bank subsidiaries increased in Q4

Reserve Coverage Ratios Reserve 400 2019 2020 2020 2020 2020 Coverage Ratio Q4 Q1 Q2 Q3 Q4 350 (%)

300 Industry 186.08 183.20 182.40 179.89 184.47 250 average

200 Mega banks 234.33 231.70 227.97 221.18 215.03

150 Joint-stock 192.97 199.89 204.33 199.10 196.90 banks 100

City banks 153.96 149.89 152.83 154.80 189.77

(%) Rural banks 128.16 121.76 118.14 118.62 122.19 Industry Average Mega Banks Foreign bank Joint-stock Banks City Banks 313.90 299.33 312.09 322.49 367.87 Rural Banks Foreign Bank Subsidiaries subsidiaries

A write-off of legacy bad debt helped banks clean up their balance sheets to weather through possible NPL increases in 2021 In 2020, the banking industry disposed of 3.02 trillion RMB in bad debts. To put it into perspective, that is 112% of the total NPL balance of 2.7 trillion RMB recorded at the end of 2020. In 2020, 1.22 trillion RMB in bad debts were written off in China.

Quarterly Loan Writing-off in China 500 450 400 350 300 250 200 150 100 50 0

(Bil. RMB)

Contents

Inter-bank market interest rates have recovered to pre-COVID levels

6.00 Issuance Interest Rate and Amount of 3-Month Negotiable Certificates 300 of Deposits (“NCD”) Issued by Banks 5.00 250

4.00 200 . RMB) bil 3.00 150

Interest Rate (%) Rate Interest 2.00 100 Issue Amount ( Amount Issue

1.00 50

0.00 0 2018-01 2018-04 2018-07 2018-10 2019-01 2019-04 2019-07 2019-10 2020-01 2020-04 2020-07 2020-10 Mega Banks Issue Amount Joint-stock Banks Issue Amount City Banks Issue Amount Rural Banks Issue Amount Mega Banks Issue Rate Joint-stock Banks Issue Rate City Banks Issue Rate Rural Banks Issue Rate

Note: The data include all 3M NCDs that are issued from Jan 1, 2018 to Dec 31, 2020. Source: Wind, collected and adjusted by S&P Global (China) Ratings. Copyright ©2021 by S&P Ratings (China) Co., Ltd. All rights reserved.

The quarterly monetary policy sentiment index suggests execution of monetary policy is normalizing to pre-COVID levels

Quarterly Monetary Policy Sentiment Index Published by PBOC 80.00

70.00

60.00

50.00 （ % ）

40.00

30.00

20.00

Note： the policy sentiment index has a scale of 0 to 100%. When the index is higher than 50%, it means the surveyed bankers believe the monetary policy is expansionary; when the index is lower than 50%, its means the surveyed bankers believe the monetary policy environment is tightening. Source: PBOC, collected and adjusted by S&P Global (China) Ratings. Copyright © by 2021 S&P Ratings (China) Co., Ltd. All rights reserved. 33

Regional banks’ liquidity profiles have been improving, liquidity risk is overall controllable Results of Liquidity Stress Testing  According to the China Financial Stability Report 2020 issued Conducted by the PBOC by the Central Bank, the Central Bank selected 1, 550 banking 100 institutions for stress testing. Among these 1, 550 tested banks,

94.90% and 91.87% of them passed mild and severe liquidity 80 stress scenarios, up by 2.59 percentage points and 5.45

percentage points compared to the previous year’s results. 60  In our view, banks run by counterparties present a major liquidity risk for high-risk regional banks. Therefore, high

Pass Rate (%) Rate Pass 40 dependence on wholesale funding is an important sign of

liquidity vulnerability if the bank also suffers capital shortage 20 and severe bad debt pressure.  Due to the tightening regulation of Internet deposits, some 0 small and medium-sized banks that rely on internet deposits Mildly adverse scenario Severely adverse scenario for funding will face greater pressure on liquidity in the short 2019 2020 Source: China Financial Stability Report 2020, Collected by S&P Ratings (China) term and may increase their use of wholesale funds. Copyright ©2021 by S&P Ratings (China) Co., Ltd. All rights reserved.

We expect higher issuance of negotiable certificates of deposits (NCD) in 2021 Among the over 300 banks which have issued their NCD issuance plan, 46% plan to increase their issuance in 2021, 39% plan to maintain the same level of issuance, and 15% to reduce their issuance. The planned 2021 NCD issuance amount is an increase of 13.6% compared to 2020, representing an increase from 14.3 trillion RMB to 16.3 trillion RMB.

Increase in the planned issuance of NCDs  We believe the normalization of the 120 monetary policy environment is one 100 important reason for increased NCD

80 issuance is 2021.  In addition, tightening regulations on 60 high-yield structured deposits and 40 deposits outside home regions for 20 regional banks also increased these Number of Banks banks’ short-term need for wholesale 0 funding. Given the poor stickiness of these deposit products, we believe the regulatory move to control their

Percentage Increase in the Planned Issuance of NCDs in 2021 growth is positive for small banks’ liquidity management in the long term. Note 1: Data as of Feb 22th 2021. Note 2: In addition, there are 20 banks who didn’t plan NCD issuance in 2020, and plan to issue in 2021. In addition, 157 banks haven’t published their NCD plan for 2021 yet. Source: Wind, collected and adjusted by S&P Global (China) Ratings. Copyright ©2021 by S&P Ratings (China) Co., Ltd. All rights reserved. 35

Regional banks’ wholesale funding costs show clear differentiation

Medians of Coupon Rate of 3M-NCDs of Regional Banks in 2020 3.4

2.6

2.2

1.8

1.4 Issuance Median Coupon (%) Rate

City Banks in the Region Rural Banks in the Region

Note: The Issuance coupon rate median is the median of coupon rate of 3 month-NCDs issued by city banks and rural banks in different provincial areas during 2020. Source: Wind, collected and adjusted by S&P Global (China) Ratings. Copyright ©2021 by S&P Ratings (China) Co., Ltd. All rights reserved. 36

Market concerns over regional banks’ asset quality has been made apparent in banks’ wholesale funding costs

Average Issuance Rate of 3M NCDs of Regional Banks in 2020 vs Regional NPL Ratio as of End of 2019

3.5 Shaa…

Guizhou Qinghai Jilin Hainan Guangdong Heilongjiang 3 Anhui Hebei Liaoning Hubei Gansu Yunnan Zhejiang Shandong Jiangxi Guangxi Beijing Henan According to PBOC’s stress testing, Fujian Inner Mongolia Shanxi Tianjin regional banks as a whole can 2.5 Hunan Sichuan Shanghai maintain a capital ratio of 10.5% Xinjiang Jiangsu and reserve coverage ratio of 100% Tibet as long as the NPL ratio is below Chongqing

Average Issue Rate of 3M 3M in NCD 2020 (%) of Issue Rate Average 7.54%.

2 0 1 2 3 4 5 6 7 8 9 NPL Ratio as of End of 2019 (%)

Note: NCD issuance rate is the median of all 3M NCD issued in 2020 by city and rural banks within each provincial area. Dotted lines indicate the average levels of the related ratios. Sources: PBOC, Wind, collected and adjusted by S&P Global (China) Ratings. Copyright ©2021 by S&P Ratings (China) Co., Ltd. All rights reserved. 37

Contents

The government and regulator have encouraged regional banks to continue providing credit support to MSEs in 2021

In January 2021, PBOC announced plans to continue providing favorable policies for regional banks to support their efforts in MSE lending.  For MSE loans of regional banks extended for over six months, the central bank provides these banks with a incentive of 1% of the extended loan principal amount through monetary policy tools. The maximum amount of the incentive is decided by the State Council.  The central bank gives qualified regional banks cheaper funding through monetary policy tools for any new loans with a term of over 6 months lent by such banks to MSEs. Such measures will continue in the first quarter of 2021. 40% of the qualified lending will be eligible for this monetary policy tool. Regional banks with a central bank rating in level 1-5 can qualify for this policy.

Systemic impact is an important consideration for government support decisions According to the Financial Stability Report 2020 published by the central bank, when it comes to making decisions on providing support to troubled banks, it tries to balance overall financial stability with moral hazard control. The following three factors are considered to come up with the best solution possible. Degree of risk: according to the central bank, the first step is to decide whether a troubled bank is having a liquidity problem or an insolvency problem. If a bank has liquidity difficulties but is still solvent, the deposit insurance fund or the central bank can provide liquidity support and the bank can recover with government support. However, for banks that have become insolvent, in principle, market discipline should be implemented and it may exit the market. Systemic impact: according to the central bank, it analyzes the troubled banks’ systemic impact based on their sizes, complexities, coverage and linkage with other financial institutions. For a bank with systemic importance, even if it has become insolvent, it may be difficult to put it into bankruptcy process. A more “gentle” approach may need to be adopted to avoid triggering systemic risk. Other constraining factors: according to the central bank, there are three other major factors. The first is information risk. Some banks in trouble may have hidden their true risk from their regulator. The information risk becomes particularly dangerous in urgent situations. The second factor is market sentiment. When the market is in a panic mode, it becomes more difficult to restore a bank’s capital and viability because it becomes more difficult to attract new strategic investors. In addition, market sentiment also affects the degree of externality of any bank risk incident. The third factor is the efforts made by the regional governments. Risk resolution efforts tend to be more effective when regional government really steps up and takes responsibility.

Regional governments play a key role in bailing-out and restructuring local small and mid-sized banks

As small regional banks face an increasingly tough competitive environment and financial performance challenges, there were a few small bank mergers in 2020, which typically required coordination by regional governments. We believe these merged banks now have better business franchise as they enjoy improved economies of scale. But the sustainability of such new banks typically depends on efforts to clean-up legacy bad debts and injection of new capital, where local governments play a crucial role.

 In November 2020, Bank of Sichuan was set up after a merger between Panzhihua City Commercial Bank and Liangshan Prefecture Commercial Bank. These two city banks in Sichuan Province suffered serious bad debt problems in recent years prior to the merger. The newly established Sichuan Bank had a paid-in capital of 30 billion RMB, with contributions from new strategic investors. Legacy bad debts were sold off before the merger.

 Shanxi Provincial Government plans to merge five city banks in the province, including a merger of Datong Bank, Jincheng Bank, Changzhi Bank, Jinzhong Bank, and Yangquan Bank into one bank. The provincial government will issue an earmarked bond to inject capital into this newly merged bank.

 In January 2021, Liaoning Provincial Government announced its plan to establish a new provincial-level city bank, merging 12 local city banks (with approximately 1.2 trillion RMB in total assets as of the end of 2019). To inject new capital into the bank, the government will introduce Liaoning Provincial Financial Holding Group, other strong national corporates and Deposit Insurance Fund Management Company as strategic investors. Currently, there are 15 city banks in Liaoning Province, among which, Shengjing Bank, Jinzhou Bank and Bank of Dalian are not included in the merger plan.

Several regional banks in Liaoning have serious bad debt problems, so we believe the merger can significantly improve their overall creditworthiness

The 12 regional banks in Liaoning to be merged account for about 1/3 of total city bank assets in the province. Some have significant SML and NPL levels, far above the industry average level. As of the end of 2020, the accumulative bond default rate of Liaoning-based corporates was 10.6%; the regional NPL ratio was 4.6% as of the end of 2019.

Asset Comparison of Banks to be Merged and Other SML+NPL Ratios of Major City Banks in Liaoning Regional Banks in Liaoning As of End of 2019 40

30 Banks to be merged 20

Shengjing 10 Bank 0 Jinzhou Bank (%)

Bank of Dalian

Banks to be merged Banks not included in the merger Note 1: Jinzhou Bank’s SML+NPL ratio dropped from 23% as of end of 2019 to 7.8% as of the end of June 2020 Note: This chart doesn’t include Tieling Bank and Benxi Bank thanks to bad debt cleanup efforts. because of lack of data. Note 2：Orient Asset Management Company is the controlling shareholder of Bank of Dalian. In 2015, the AMC Sources: Wind, collected and adjusted by S&P Global (China) Ratings. helped clean up billions of legacy bad debts of the bank through capital injection. Copyright ©2021 by S&P Ratings (China) Co., Ltd. All rights reserved. Sources: Wind, collected and adjusted by S&P Global (China) Ratings. Copyright ©2021 by S&P Ratings (China) Co., Ltd. All rights reserved. 42

Local government bonds have become a new source for regional bank capital injections  In July 2020, the State Council decided to allow local governments to use special bonds earmarked to support small and mid-sized banks’ capital. Therefore, the Ministry of Finance decided on a quota of 200 billion RMB for such special local government bonds, which is to be divided among 18 regions (Tianjin, Hebei, Zhejiang, Shandong, Guangdong, Inner Mongolia, Liaoning, Jilin, Heilongjiang, Shanxi, Jiangxi, Henan, Hubei, Guangxi, Sichuan, Yunnan, Shaanxi and Gansu).

 In December 2020, Guangdong Government became the first government to issue the special bond (“20 Guangdong 97”) to support small and mid-sized banks, which raised 10 billion RMB, with a coupon interest rate of 3.52%. The money raised by the bond will be used to inject capital to the Guangdong Yunan Rural Credit Cooperative, Guangdong Puning Rural Commercial Bank, Guangdong Jiedong Rural Commercial Bank, and Guangdong Luoding Rural Commercial Bank with amounts of 4 billion RMB, 3.7 billion RMB, 1.4 billion RMB and 0.9 billion RMB respectively.

 Zhejiang, Shanxi, Guangxi and Inner Mongolia have issued special bonds earmarked to inject capital into local banks. Banks identified as beneficiaries include Bank of Wenzhou (5 billion RMB), a new Shanxi provincial bank (15.3 billion RMB) , 21 small banking institutions in Guangxi (11.8 billion RMB), Bank of Inner Mongolia (5.5 billion RMB), and Bank of Ordos (3 billion RMB).

 Sichuan Province plans to issue 11.4 billion RMB of special bonds to inject capital into small and medium-sized banks.

Ministry of Finance has given regional banks more time to implement new accounting standards for financial instruments and set up transitional period for regulatory capital calculation under the New Accounting Standards

On December 30, 2020, the Ministry of Finance, together with CBIRC, announced a new plan for the implementation of the New Accounting Standards of Financial Instruments (the Chinese version of IFRS 9).  Option to Postpone the Implementation of the New Accounting Standard for One Year Banks not listed on the stock market that may find it difficult to implement the New Accounting Standards of Financial Instruments may postpone its implementation to January 1, 2022.  Transition Arrangement for Regulatory Capital Calculation For banks not listed on the stock market, within five years of the implementation of the New Accounting Standards, they may choose to add a certain percentage of the core tier-1 capital deducted due to the implementation of the expected credit loss approach back to its core tier-1 capital. The “add back ratio” for the first and second year is 100%, 75% for the third year, 50% for the fourth year, and 25% for the fifth year. Although the regulatory capital transition arrangement doesn’t actually improve the economic capital of the banks, it reduces the chance of a severe drop of reported capital adequacy ratios by the implementation of New Accounting Standards, thus gaining banks more time to improve their actual capitalization.

The regulator has looked to improve corporate governance of small and mid-sized banks, which is essential for their long-term viability Distribution of Corporate Governance Assessment Results of Banks and Insurance  In August 2020, CBIRC issued a three-year action plan for improving Companies corporate governance in the banking and insurance industry. In order to strengthen market discipline, improve supervision, and regulate 1000 shareholder behavior, in July and December 2020, CBIRC published lists

of financial institution shareholders with serious compliance infractions. 800  CBIRC stated that it will adhere to the three bottom lines of "long-term

stability", "transparency and integrity", and "fairness and 600 reasonableness" to effectively improve capital quality and corporate governance transparency through strict supervision, improve the level of compliance and internal control, and enhance the ability to prevent and 400 resolve financial risks.

 CBIRC recently issued its results of financial institution corporate 200 governance assessment results covering 1,605 commercial banks and

187 insurers. About 22% of the institutions failed the assessment (rated 0 below C). A B C D E Note: A is the best result, and E is the worst. Assessment below C means failure to pass. Sources: CBIRC, collected and adjusted by S&P Global (China) Ratings. Copyright ©2021 by S&P Ratings (China) Co., Ltd. All rights reserved.

The central bank and CBIRC have been issuing new laws, policies and regulations to improve the regulation of banks

 In October 2020, PBOC published a comment-seeking draft of the new Commercial Bank Law, which added new contents related to corporate governance, business scope, interest rate pricing, risk resolution, central bank’s Macro Prudential Assessment (MPA) based on the industry and policy development in recent years.  In December 2020, PBOC and CBIRC issued a regulation on the assessment of systemically important banks to identify domestic systemically important banks (D-SIBs). The D-SIB identification process will be a combination of quantitative scoring and qualitative judgement by the regulator. The biggest 30 banks will qualify for the preliminary assessment. The final list will be decided by the Financial Stability and Development Committee of the State Council.  In January 2021, CBIRC issued a comment-seeking draft of commercial bank liability quality management regulations. These new regulations are to enhance banks’ liability and liquidity management, and have better regulation on deposit interest pricing and internet-based deposit. We believe these new regulations are positive for the stability of the deposit base of regional banks in the long term.  In January 2021, CBIRC issued a new document on the risk resolution of village banks. China has over 1,600 village banks, and some of them have shown significant credit risk. This document is to encourage the founding banks of village banks to lead the risk resolution. The methods to save those high-risk village banks include new capital injection from the founding banks, merge and acquisition, and attracting new strategic investors. This document also offers rewards to founding banks who are successful in reducing the risk of their village banks.

Contents

Industry Outlook Factor Outlook Key points  We expect moderate loan demand after the economy enters the post-COVID stage, and we expect banks’ balance sheet growth to moderate in 2021. Business  Driven by regulators’ efforts, we expect a lower pace of real estate-related loans (including mortgage and property Stable Position development loans) growth.  Due to tightening regulation on internet-based deposits, some small banks may have more difficulty attracting deposits in 2021, but we expect better control of deposit cost thanks to regulatory oversight.  As loan demand growth moderates to a normal level, the capital consumption of lending activities will slow down. Capital &  Some regional banks have capital shortage problems due to high bad debts. Stable Earnings  As asset quality pressure continues, credit costs will remain high, which we believe may be the dominating factor affecting profitability in 2021.

Stable  We expect more MSE NPLs caused by COVID as forbearance measures phase out in 2021. In addition to MSEs, the Risk with defaulted large enterprises, such as HNA Group, also weigh on the asset quality of banks. Position downward  Thanks to the sound macro economy, we expect overall asset quality of the banking sector to remain stable, but we pressure also expect some acute bad debt problems among certain regional banks, mainly because of legacy bad debts.

 The central bank states it plans to maintain the continuity, stability and sustainability of its monetary policy, which will be prudent monetary policy with proper flexibility and moderation. Therefore, we expect reasonably good liquidity Funding & systemwide. Stable Liquidity  Some regional banks that are over-reliant on internet-based deposits may need to increase their use of wholesale funding as deposit taking becomes more difficult under tightening regulations, but we don’t expect an acute liquidity shortage. The tightening regulation on internet-based deposits is positive for banks’ liquidity risk management in the long term.

Industry Outlook (cont.)

Factor Outlook Key Points

Stand-alone Stable with  Although provisioning costs will remain high in 2021, the stand-alone credit quality of mega banks and leading Credit downward joint-stock banks will remain stable. Quality pressure  Due to legacy bad debts and asset quality pressure form COVID, some regional banks will come under greater asset quality and capital pressure in 2021.  We expect the government’s supportive attitude toward the banking industry to remain unchanged.  Provincial governments have an important role in stabilizing the credit quality of small and medium – sized Government banks. Stable  The regulator has issued several new regulations in recent months, particularly on the liability side of the Influence business, continuously improving its regulation on banks.  The government is firm on preventing any “relapse” of shadow banking activities. We believe the regulatory position is credit positive for the banking sector. Issuer Credit Stable  Although some regional banks will show more stress in 2021, due to government support and strong regulation, Quality we expect overall stable credit quality of the industry.

Contents  Business Position  Capital and Earnings  Risk Position  Funding & Liquidity  Government Influence  Industry Outlook  Special Topic 1: Relationship between Bond Defaults and Bank Asset Quality  Special Topic 2: Shadow Banking’s Impact on the Banking Sector  Appendix: Related FI Research

Bond Defaults Reflect Asset Quality Pressure in Different Regions In recent years, corporate bond defaults have been increasing. Regional banks typically have substantial exposure to those defaulted issuers. Therefore, in our view, the bond market default performance is another way to help differentiate the asset quality of regional banks. Number of Cumulative Bond Ratio of Number of Defaulted Bond Number of Region Bonds Default Rate (%) Defaulted Issuers (%) Amount (bil.) Defaulted Issuers Defaulted Hunan 0.09 0.42 0.80 2 1 Shaanxi 0.14 1.60 0.85 2 2 Yunnan 0.24 0.92 0.99 1 1 Chongqing 0.25 2.11 2.33 5 4 Guangxi 0.29 1.02 1.00 2 1 Xinjiang 0.43 2.13 1.05 2 2 Sichuan 0.56 1.66 5.51 11 5 Tianjin 0.61 5.30 6.59 9 8 Guangdong 0.65 3.07 25.67 33 16 Jiangsu 0.69 1.51 21.65 36 12 Hubei 0.81 1.73 6.59 9 4 Beijing 0.88 4.10 107.83 90 22 Zhejiang 1.22 1.59 26.05 39 10 Fujian 1.31 3.61 15.98 15 7

Bond Defaults Reflect Asset Quality Pressure in Different Regions (Cont.) Cumulative Defaulted Bond Number of Ratio of Number of Number of Region Bond Default Amount Bonds Defaulted Issuers (%) Defaulted Issuers Rate (%) (bil.) Defaulted Shanghai 1.32 2.92 43.42 31 10 Shandong 1.50 3.41 24.35 42 14 Jilin 1.54 7.69 2.94 5 4 Gansu 1.65 5.00 2.66 6 2 Henan 1.70 5.77 10.14 19 9 Anhui 2.50 2.76 16.42 25 6 Shanxi 3.39 2.30 22.69 19 2 Inner Mongolia 4.53 10.00 4.38 7 5 Hebei 4.76 5.10 24.00 21 5 Heilongjiang 5.92 9.38 5.72 10 3 Ningxia 7.48 15.00 2.40 4 3 Qinghai 9.12 5.26 6.17 3 1 Liaoning 10.61 9.09 38.09 47 9 Hainan 11.82 36.36 11.11 12 8 Total 1.13 2.84 437.37 507 176

Relationship between Bond Defaults and Regional NPL Ratios

Regional Cumulative Bond Default Rates as of End of 2020 vs Regional NPL Ratios as of End of 2019

12 Hainan

Liaoning 10 Qinghai 8

6 Hubei Heilongjiang Jiangsu 2020 (%) Hebei Shanghai Inner Mongolia 4 Anhui Guangdong Shanxi Beijing 2 Hunan Fujian Henan Gansu Jilin Sichuan Shandong Xinjiang Tianjin Cumulative Bond Default Rates as of End of of End as of Rates Bond Default Cumulative 0 Zhejiang Yunnan 0 1Chongqing 2 Guangxi3 4 5 6 7 8 9 Shaanxi NPL Ratio as of End of 2019 (%)

Note: The numbers may not be 100% comparable due to different approaches taken by each region when calculating the ratio. The dotted line indicates the average of the ratio. Sources: Regional Financial Institution Performance Reports of different regions published by PBOC, Wind, collected and adjusted by S&P Global (China) Ratings. Copyright ©2021 by S&P Ratings (China) Co., Ltd. All rights reserved.

High-risk large enterprises put significant pressure on banks’ asset quality

NPLs of Banking Institutions  According to China Financial Stability Report 2020 4000 published by the central bank, there are about 30,000 large 3000

business entities in China. The combined assets of large 2000 . RMB)

enterprises totaled 119.6 trillion RMB, accounting for bil ( 22.9% of the total enterprise assets in China. 1000  As of the end of 2019, 575 large enterprises nationwide had 0 significant financial troubles. Among them, 460 enterprises 2011 2012 2013 2014 2015 2016 2017 2018 2019 NPL volumes of Overall China Banking Institutions reported severe liquidity strains, 120 already defaulted on Non-Performing Assets from Large Enterprises Nationwide

bonds, 27 enterprises’ equity was frozen by courts, and 67 Sources: PBOC, collected and adjusted by S&P Global (China) Ratings. filed for bankruptcy or restructuring. The total financing of Copyright ©2021 by S&P Ratings (China) Co., Ltd. All rights reserved. these 575 enterprises amounted to 3.88 trillion RMB, Distressed Large Enterprise Debt Breakdown by among which 646.2 billion RMB had been classified as non- Sector Mining and Others, 5.16% performing assets, accounting for only 16.7% of the total Quarrying, financing for those enterprises. We believe that the 5.56% Construction, distressed large enterprises will continue to put great 7.72% pressure on banks’ asset quality in 2021. Real Estate, Manufacturing,  The distressed large enterprises are mostly in the 8.46% 33.63%

manufacturing sector (241), wholesale and retail sector Leasing and (87), and transportation, warehousing and postal service Business Services, 12.06% sector (51), with their respective debts standing at 1.3 Transportation, Warehousing and Postal trillion RMB, 493.2 billion RMB and 570.4 billion RMB. Wholesale and Retail, 12.71% Service, 14.70% Sources: China Financial Stability Report 2020, collected and adjusted by S&P Global (China) Ratings. Copyright ©2021 by S&P Ratings (China) Co., Ltd. All rights reserved. 54

Highly-leveraged large enterprises whose bonds defaulted in 2020 weigh on the asset quality of banks

 In 2020, 224 bonds defaulted (worth 238.6 billion RMB) involving 73 borrowers. According to our preliminary estimate, Distribution of Liability Size of 73 Enterprises the total liabilities of those enterprises were around 4.7 trillion with Bond Defaults in 2020 RMB. We believe the majority of those liabilities (above 70%) 40 were debts borrowed from banks and bond markets, putting 35 stress on the asset quality of financial institutions.  The majority of these defaulted corporates have a total liability 30 lower than 50 billion RMB. 25  Some large enterprises defaulted in 2020. The top 20 defaulted 20 borrowers accounted for 76% of the total liabilities of the 73 defaulters. 15  HNA Group is the largest borrower that defaulted in 2020. Number of Entities Numberof 10 According to its half year financial report in 2019, it borrowed loans worth 373.2 billion RMB from banks. 5  Assuming the HNA Group’s loans were not classified as NPL by 0 banks by the end of 2019, if we add HNA Group’s borrowing into <25 50 75 100 125 150 175 200 225 250 275 300 >300 bad debts, the actual bad debt of the banking industry would Total liabilities (Bil. RMB) be up by over 10% by the end of 2019, and the NPL ratio will increase from the reported 1.86% to 2.15%. Sources: Wind, collected and adjusted by S&P Global (China) Ratings. Copyright ©2021 by S&P Ratings (China) Co., Ltd. All rights reserved.

Large corporates that defaulted in 2020 typically had significant bank borrowing The government has paid great attention to risks posed by highly-leveraged large enterprises. In January 2021, CBIRC stated it will coordinate with regional governments to solve risk caused by defaulted large enterprises.

Total Enterprise whose bonds defaulted in Total liabilities Enterprise whose bonds defaulted in No. No. liabilities 2020 (Million RMB) 2020 (Million RMB) Yongcheng Coal and Power Holding Group Co. 11 134,394.96 Ltd. 1 HNA Group Co., Ltd. 706,726.40 12 Huachen Automotive Group Holdings Co., Ltd. 132,843.62 2 Peking University Founder Group Co.,Ltd. 302,951.44 13 Cefc Shanghai International Group Limited 127,881.45 3 Haihang Capital Investment Co., Ltd. 260,020.89 14 Hainan Airlines Holding Co.,Ltd. 127,701.24 4 Tianjin Bohai Leasing Co., Ltd. 238,837.92 15 Macrolink Holdings Co., Ltd. 90,086.48 5 Bohai Leasing Co.,Ltd. 224,976.23 16 Wintime Energy Co., Ltd. 78,218.38 6 Tsinghua Unigroup Co., Ltd. 202,938.42 17 Kangmei Industrial Investment Holdings Co., Ltd. 72,014.20 7 Tahoe Group Co., Ltd 197,374.64 18 Fujian Fu Sheng Group Co., Ltd. 70,663.87 8 Tianjin Real Estate Group Co., Ltd. Nanjing Fullshare Industrialholding Group Co., 197,342.61 19 61,406.28 Ltd. 9 Citic Guoan Group Co., Ltd. 163,863.19 20 Zhongrong Xinda Group Co., Ltd. 61,019.93 10 Dongxu Group Co., Ltd. 154,023.13 Sources: Wind, collected and adjusted by S&P Global (China) Ratings. Copyright ©2021 by S&P Ratings (China) Co., Ltd. All rights reserved.

Creditors’ Committees are important for bad debt recovery of large borrowers. The final recovery rate is uncertain considering the high complexity of highly leveraged defaulted large enterprises  In December 2020, CBIRC, NDRC, PBOC, and CSRC jointly issued a In February 2021, China Fortune Land new regulation on the work process of FI creditors’ committees. The Development Co Ltd. held its first creditors’ work process rule is built on the creditors’ committee regulation committee meeting. The participants issued by CBRC in 2016 and made further improvement based on include PBOC, CBIRC, CSRC, Hebei the new development in recent years. This new rule will help government, financial institution creditors, improve the corporate resolution and exit mechanism, protect the and the chairman and senior managment rights of financial institutions, and crack down on willful defaulting. team of the company. The meeting urged the banks not to withdraw their loans from the This rule covers the role and responsibility of creditors’ committee, borrower to stabilize its funding structure. makes it clear that the bond custodians can also organize creditors’ As of the end of June 2020, China Fortune committee, set up the boundaries of the power of the committee Land Development Co Ltd. had borrowed and support financial institutions to protect their rights in case of loans worth 115.8 billion RMB and issued willful defaulters. bonds worth 55.5 billion RMB.

Contents

The regulator defines shadow banking as financial intermediary services outside the traditional bank lending system, typically carried out by NBFIs

 China’s shadow banking business is closely related to the banking Size and Growth of Shadow Banking Products sector, and banks have been main players in shadow banking. The banking sector accounts for about 90% of the total assets of the 120 80 financial industry. 100  In terms of funding, banks have been the major provider of funding 60 80 for shadow banking products. For example, over 50% of asset . RMB) Tril management products privately placed and over 40% of trust ( 60 40 (%) products were funded by banks. In terms of assets, the majority of 40 shadow banking clients have been bank clients. 20  Driven by demand from banks, from 2008 to 2016, shadow banking 20 grew over 20% year on year. In early 2017, the size of shadow 0 0 banking peaked at 100.4 trillion RMB. In comparison, the banking 2010 2011 2012 2013 2014 2015 2016 sector’s total assets were 182 trillion RMB and the total loan size Assets Management Products of Insurance Companies Private Equity of the banking sector was 87 trillion RMB as of the end of 2016. Subsidiaries of Fund Companies Fund Companies The size of shadow banking business was larger than the loan Assets Management Products of Brokers book, and grew at a much higher pace before 2017. Trust Assets Wealth Management Products of Banks YoY Growth Rate of Shadow Banking Products Size (Right Axis) YoY Growth Rate of Total Assets of Banking Industry (Right Axis) Sources: China Shadow Banking Report by CBIRC, collected and adjusted by S&P Global (China) Ratings. Copyright © by 2021 S&P Ratings (China) Co., Ltd. All rights reserved 59

Before 2017, banks’ capital consumption and reserve requirements for holding shadow banking products were much looser compared to loans with similar risk profiles, creating regulatory arbitrage opportunities

Major problems with aggressive growth of shadow Size and Growth of Non-standard Investment Portfolio banking before 2017: of Commercial Banks  Increased leverage to a dangerous level. Macro leverage 25 100 increased from 200% in June 2013 to 239% at the end of 2016.  Light regulation. Unlike tightly regulated loan lending, shadow 20 80 banking was lightly regulated before 2017, providing 15 60 opportunities for regulatory arbitrage. Banks could bypass . RMB) regulatory requirements on capital, reserves, lending limits, and (%)

Tril 10 40 credit risk classification by holding shadow banking products. (  Lack of transparency. Information disclosure on such products 5 20 was much less transparent than loan lending business, offering banks opportunities to hide risk. 0 0 2011 2012 2013 2014 2015 2016 2017  Highly contagious risk. Banks took on each others’ risk through Balance YoY Growth Rate (Right Axis) selling and buying shadow banking products between each Sources: China Shadow Banking Report by CBIRC, collected and adjusted by S&P Global (China) other. Ratings. Copyright © by 2021 S&P Ratings (China) Co., Ltd. All rights reserved

Before the cleanup, shadow banking distorted the real capital level of individual banks and the banking sector as a whole

 When banks repackaged their loans into “trust products” through trust companies, or “wealth management products” (WMPs) through banks’ wealth management business units, and put those assets into investment portfolios instead of their loan book, they could bypass loan lending regulations, including capital and reserve requirements, industry risk limits, or single-name concentration risk limits, as well as loan credit risk classification and disclosure procedures.

 After loan assets were changed into on-balance sheet investment assets or off-balance sheet WMPs, the assets’ credit risks were not actually mitigated or diversified, but the on-balance sheet investment assets and off-balance sheet WMPs were not subject to the same capital or reserve requirements as loans.

 According to CBIRC, from 2013 to 2017, banks held increasingly large amounts of loan-like WMPs, and reduced their regulatory capital requirements by reducing their risk-weighed asset (“RWA”) by at least 6 trillion RMB by holding loans as WMPs. To put it into perspective, as of the end of 2016, the total capital of the banking sector was 14.74 trillion RMB and the total RWA was 111.03 trillion RMB . If we add the 6 trillion RMB RWA back, the actual capital adequacy ratio would be about 12.60% instead of the reported 13.28%.

Before the cleanup, the complex structure of securitization also helped disguise real risk

 Prior to 2017, banks could also circumvent regulations through Risk Weights of Securitization Products in securitization. After loan assets were securitized, banks typically RWA Calculation chose to hold the senior tranches (which have a low risk weight Re- Securitization of 20%, while a corporate loan has a risk weight of 100%), and Long-term Credit Rating Securitization (%) put the equity tranche (which have very high risk weights) into (%) off-balance sheet wealth management products which were not AAA to AA- 20 40 included in RWA calculation. Because of banks’ implicit guarantee on WMPs, the actual credit risk of the underlying A+ to A- 50 100 assets remained with the banks, while regulatory capital requirements were reduced dramatically due to the BBB+ to BBB- 100 225 securitization process. BB+ to BB- 350 650  Banks also engaged in re-securitization of their securitized B+ and Below B+ or Not 1250 1250 loans, making the credit risk even more complex to understand Rated or monitor. Sources: China Shadow Banking Report by CBIRC, collected and adjusted by S&P Global (China) Ratings.  The size of ABS products increased by over 30 times from 2012 Copyright © by 2021 S&P Ratings (China) Co., Ltd. All rights reserved to 2016.

Before the cleanup, asset quality information of some banks was distorted because shadow banking products were not subject to loan credit risk classification (NPLs + SMLs) / Gross Loans of Bank of Jinzhou  In April 2016, the IMF pointed out that Chinese 30 NPLs SMLs enterprises with total loans worth 1.3 trillion USD could not cover interest payment with their revenue, 20 accounting for 14% of the total debts of listed Chinese enterprises. 14% was a much higher ratio than the NPL Loan (%) Loan 10 ratio reported by banks. As of the end of March 2016,

the reported banking sector NPL ratio was 1.75%. (NPLs + SMLs) / Gross 0 2014 2015 2016 2017 2018 2019 2020Q2  Liquidity risk was also increased due to the creation of asset pools with significant asset-liability mismatch. NPL Ratio of Hengfeng Bank For example, before 2017, most WMPs had a term of 30 less than 3 months, and over 90% of them had a term less than 1 year, while the underlying assets typically 20 had a term of 3 to 4 years. Such liquidity risk was also 10

very contagious since banks were holding each other’s (%) Ratio NPL shadow banking products. 0 2014 2015 2016 2018 2019

The shadow banking cleanup from 2017 to 2019 has been effective in terms of stabilizing the financial sector

 The government approached the cleanup in a cautious and steady manner and avoided massive market disruption.

 The scale of shadow banking dropped to 84.8 trillion RMB as of the end of 2019, down by 16% compared to the peak in early 2017. The ratio of shadow banking/GDP decreased from 123% as of the end of 2016 to 86% as of the end of 2019, representing a drop of 37 percentage points.

 More importantly, after the cleanup, shadow banking has become more transparent, and the product structure has become easier to understand, with more capital and reserve buffers put in place for shadow banking businesses.

Major measures taken during the cleanup:  Crack down on compliance infractions;  Crack down on fake capital injection and hidden shareholding structure;  Crack down on inappropriate intervention by biggest shareholders, and improve internal control of financial institutions; and  Make regulatory rules consistent, and improve regulatory process and coverage to minimize regulatory arbitrage activities.

The regulator plans to ensure that future shadow banking activities are properly regulated and prevented from becoming a tool for regulatory arbitrage The regulator plans to work on the following aspects: China's Shadow Banking Size As of End of 2019 Trillion No. Products  Improve regulatory monitoring to reduce regulatory arbitrage through RMB new ways and prevent relapse of aggressive high-risk shadow banking 1 SPV Investments Bought from other Banks -- activities. Full coverage and consistency of regulations are important 2 Wealth Management Products of Banks 23.40 to prevent regulatory arbitrage or regulatory blind spots. All shadow Inc: Interbank Wealth Management 0.84 banking activities should be subject to regulation. The regulatory rules Non-standardized Product Investments 1.41 on credit risk classification, risk weights, capital calculation, and Standardized Asset Management Investments 9.16 credit loss provisioning should be further improved for shadow Others 11.99 banking. 3 Entrust Loans 11.44 4 Trust Company Products 17.94  Separate different businesses to prevent contagion risk. The priority is 5 Non-stock Public Offering Funds 13.47 to set up firewalls between private funds and mutual funds, between 6 Asset Management Products of Brokers 18.23 on-balance sheet and off-balance sheet businesses of banks, and 7 Asset Management Products of Insurers 2.76 between asset management for clients and proprietary investment 8 Non-equity Private Equity Funds 4.00 business. 9 Securitization 3.83 Guaranteed Fixed Income Products of Financial Guarantee 10 2.70  Approach mixed financial operations in a prudent manner. Commercial Companies banks have to be run prudently, insurers should focus on their role in 11 Leasing Assets of Financial Leasing Companies 2.18 risk redistribution and management, securities companies should 12 Loan of Small Loan Lending Companies 0.93 guide money into value investment, and trust companies and asset 13 P2P Loans of Online Lending 0.49 managers are not supposed to provide implicit guarantee. 14 Commercial Factoring of Commercial Factoring Companies 0.10 15 Loans of Pawnshop 0.10 16 Nesting -16.77 17 Total 84.80 Sources: China Shadow Banking Report by CBIRC, collected and adjusted by S&P Global (China) Ratings. Copyright © by 2021 S&P Ratings (China) Co., Ltd. All rights reserved 65

We believe the negative impact of shadow banking on the banking sector’s asset quality and capital has been substantially reduced but not completely eliminated

 So far, the perception of implicit guarantees has been reduced but not eliminated. We expect off-balance sheet WMPs issued by banks to continue weighing on bank capital in the near future. More time is needed to completely eliminate the hidden capital consumption by WMPs.  Legacy bad debts accumulated from the heydays of shadow banking continue to weigh on banks’ asset quality and capital. More time is needed to clean up the legacy bad debts and improve capitalization. HNA Group filed for bankruptcy restructuring in early 2021. It is a typical example of aggressive conglomerate expansion during the peak of shadow banking. HNA Group had about 500 billion RMB loans and bonds outstanding, the impairment of which would take a toll on the banking sector’s earnings.  Although the levels of capital and reserve of loan-like non-standard assets have improved significantly, there is still room for further increase. The ratio of reserve/total loan-like assets increased from 0.62% in early 2017 to 1.93% as of the end of 2019. In comparison, the ratio of reserve/gross loans was 3.46% as of the end of 2019.  After the shadow banking cleanup, the banking sector was in a better position to cope with the unprecedented COVID shock.  More credit was pumped into the real economy to help people and businesses get through the pandemic in 2020, but the regulator has been firm on controlling shadow banking activities, and we haven’t seen any “relapse” of aggressive shadow banking activities.  Shadow banking activities are not going to be eliminated. It will continue to be an important part of China’s financial system. But we expect it to be better regulated.

Contents

Related Issuer Research

Credit Rating Report: JPMorgan Chase Bank (China) Company Limited, Feb 24, 2021

Credit Rating Report: China Construction Bank Corporation, Jan 19, 2021

Credit Rating Report: HSBC Bank (China) Company Limited, Jan 12, 2021

Credit Rating Report: Citibank (China) Company Limited, Oct 20, 2020

Credit Rating Report: Postal Savings Bank of China Co., Ltd., Oct 9, 2020

Credit Rating Report: Shanghai Rural Commercial Bank Co., Ltd., Jan 14, 2020

Related Industry Research

Baoshang Bank Hybrid Bond Write-Down Demonstrates Loss-Absorbing Features of Hybrid Bonds in China, Nov 16, 2020 New Challenges Arise for Credit Loss Provisioning Under IFRS 9 Amid COVID-19, Aug 12, 2020

Regulator’s Disclosure of Non-compliant Shareholders is Positive for Banks’ Corporate Governance, Jul 13, 2020

Risk Differentiation Exists Between Banks’ Hybrid Capital Instruments and Senior Bonds, Jul 1, 2020

Robust Economic Recovery Supported Mega Banks’ Financial Performance in Q3, Nov 10, 2020

Quarterly Snapshot on China’s Banking Sector | Third Quarter of 2020, Nov 25, 2020

Chinese Banks’ Q2 Performance Remained Stable but COVID-19 Credit Cost Continues to be a Concern, 20 Aug, 2020

Quarterly Snapshot on China’s Banking Sector (Second Quarter of 2020), 20 Aug, 2020

Government Support Mitigates COVID-19's Pressure On Chinese Banks` Creditworthiness, 20 May, 2020

Quarterly Snapshot on China’s Banking Sector: First Quarter of 2020, May 20, 2020

Related Industry Research

Government Support Mitigates COVID-19's Pressure On Chinese Banks` Creditworthiness, 20 May, 2020

Foreign Bank Subsidiaries Maintain Sound Credit Quality Despite COVID-19 | A Study on the Creditworthiness of Foreign Bank Subsidiaries in China, 23 Oct, 2020

Joint-Stock Banks Capable of Sustaining Lending Growth Under Earning Pressure: A Study on the Credit Quality of Chinese Joint-stock Banks Amid COVID-19, 15 June 2020

Mega Banks: Strong Enough to Withstand COVID-19, Apr 24, 2020

City Banks: Indispensable Financiers of China’s SMEs, Mar 24, 2020

The Role of Rural Banks in China’s County Economy – A Study into the Creditworthiness of China’s Rural Banks, Mar 24, 2020

Criteria and Criteria-Related Commentaries

Methodology: Financial Institutions

Commentary: Understanding S&P Global (China) Ratings Financial Institutions Methodology

Commentary: Understanding S&P Global (China) Ratings Approach To Support

General Considerations On Rating Modifiers And Relative Ranking

Commentary: Understanding S&P Global (China) Ratings General Considerations on Rating Modifiers and Relative Ranking Methodology

S&P Global (China) Ratings -Panda Bond Methodology

Analytical Contact

• Zheng Li • Yifu Wang, CFA, CPA • Ying Li, CFA, FRM • Beijing • Beijing • Beijing • +86-10-6516-6067 • +86-10-6516-6064 • +86-10-6516-6061 • [email protected] • [email protected] • [email protected]

• Longtai Chen • Xuefei Zou, CPA • Beijing • Beijing • +86-10-6516-6065 • +86-10-6516-6070 • [email protected] • [email protected]

Copyright © 2021 by S&P Ratings (China) Co., Ltd. All rights reserved. S&P Ratings (China) Co., Ltd. (“S&P Ratings”) owns the copyright and/or other related intellectual property rights of the abovementioned content (including ratings, credit-related analyses and data, valuations, model, software or other application or output therefrom) or any part thereof (Content). No 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 S&P Ratings. The Content shall not be used for any unlawful or unauthorized purposes. S&P Ratings 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 Ratings' opinions, analyses, forecasts and rating acknowledgment decisions (described below) are not and should not be viewed as 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 Ratings 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 Ratings does not act as a fiduciary or an investment advisor except where registered as such. While S&P Ratings has obtained information from sources it believes to be reliable, S&P Ratings does not perform an audit and undertakes no duty of due diligence or independent verification of any information it receives. Rating-related publications may be published for a variety of reasons that are not necessarily dependent on action by rating committees, including, but not limited to, the publication of a periodic update on a credit rating and related analyses. A RATING ISSUED BY S&P RATINGS IS ASSIGNED ON A RATING SCALE SPECIFICALLY FOR USE IN CHINA, AND IS S&P RATINGS' OPINION OF AN OBLIGOR’S OVERALL CREDITWORTHINESS OR CAPACITY TO MEET SPECIFIC FINANCIAL OBLIGATIONS, RELATIVE TO THAT OF OTHER ISSUERS AND ISSUSES WITHIN CHINA ONLY AND PROVIDES A RANK ORDERING OF CREDIT RISK WITHIN CHINA. AN S&P RATINGS' RATING IS NOT A GLOBAL SCALE RATING, AND IS NOT AND SHOULD NOT BE VIEWED, RELIED UPON, OR REPRESENTED AS SUCH. S&P PARTIES ARE NOT RESPONSIBLE FOR ANY LOSSES CAUSED BY USES OF S&P RATINGS' RATINGS IN MANNERS CONTRARY TO THIS PARAGRAPH. 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 Ratings reserves the right to assign, withdraw or suspend such acknowledgement at any time and in its sole discretion. S&P Ratings disclaims 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 Ratings 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 Ratings may have information that is not available to other S&P Ratings business units. S&P Ratings has established policies and procedures to maintain the confidentiality of certain non-public information received in connection with each analytical process. S&P Ratings may receive compensation for its ratings and certain analyses, normally from issuers or underwriters of securities or from obligors. S&P Ratings reserves the right to disseminate its opinions and analyses. S&P Ratings' public ratings and analyses are made available on its Web site www.spgchinaratings.cn, and may be distributed through other means, including via S&P Ratings' publications and third-party redistributors.