Casting a light on Shadow Banking: Near-term growth; long-term cap on bank valuations

Transcription

1 February 26, 2013 China: Banks Equity Research Casting a light on Shadow Banking: Near-term growth; long-term cap on bank valuations Shadow banking out of the shadows In this report we look to address concerns over systemic risks stemming from shadow banking, particularly banks credit exposure to shadow banking products. We estimate shadow banking credit was Rmb23.5tn at the end of 2012 and its net increase contributed 39% to new system wide credit and 27% new M2 last year. We expect continued strong growth in 2013 to Rmb31 tn, combined with total social financing growth of 18% yoy to support c. 8% 2013E GDP growth. Risks: 1) Weaker credit underwriting standards; 2) rising corporate/systemic leverage; 3) liquidity risk; 4) pro-cyclicality, 5) defaults could harm banks balance sheets with possible reluctance to expose retail investors to losses on underlying instruments, which could distort capital allocation. Comparison with the US In the US, shadow banking is largely made up of bonds and securitization. In China, however, products are simpler and there is greater focus on corporate loans. There is also much greater interconnectedness with China banks. Pressure points Inflation and overheating Given China s strong internal and external liquidity, we see limited risk of systemic crisis in the midterm. However, rapidly rising corporate leverage makes corporate EBITDA interest cover/npls more sensitive to CPI inflation. Therefore, overheating, followed by tightening, could cause more collateral damage than in previous cycles. As such, we think it is vital for China to achieve balance in growth and inflation via policies aimed more at consumption and competitiveness. Valuation implications In addition to long-term NPL risks and rising NPL sensitivity to inflation, we believe the opaque interconnections between shadow banking and China banks via wealth management products, informal loan securitization, etc. make it difficult to analyze banks true credit exposure. We think this is likely to increase the discount rates and lower longterm ROE for China banks, capping re-rating. We factor in long-term concerns over shadow banking activities into our target prices, including potential credit exposure via WMP or informal securitization. RELATED RESEARCH Raise estimates but be valuation disciplined; CQRCB/Minsheng A to Neutral; BONB to Buy, 26 Feb Shadow banking trip takeaways: Cyclical risk low; LT risks remain, Jan 22, 2013 Credit watch 5: We estimate 4%-6% total potential NPLs; Buy quality banks, CEB down to Sell, October 11, 2012 CHINA S SHADOW BANKING ACTIVITIES ACCOUNTED FOR 24% OF TOTAL CREDIT AT THE END OF 2012 Source: PBOC, Wind, Gao Hua Securities Research estimates. Ning Ma Goldman Sachs does and seeks to do business with companies +86(10) ning.ma@ghsl.cn Beijing Gao Hua Securities Company Limited Bowei Cheng +86(10) bowei.cheng@ghsl.cn Beijing Gao Hua Securities Company Limited Jessica Wu +86(10) jessica.wu@ghsl.cn Beijing Gao Hua Securities Company Limited covered in its research reports. As a result, investors should be aware that the firm may have a conflict of interest that could affect the objectivity of this report. Investors should consider this report as only a single factor in making their investment decision. For Reg AC certification and other important disclosures, see the Disclosure Appendix, or go to Analysts employed by non-us affiliates are not registered/qualified as research analysts with FINRA in the U.S. The Goldman Sachs Group, Inc. Global Investment Research 30% 25% 20% 15% 10% 5% 0% 24% E 26%

2 Table of contents China shadow banking: likely to grow in 2013E; overheating a risk; bank re-rating capped 4 China s shadow banking system Rmb24 tn in credit extension 8 How did it grow so fast? Demand, regulatory arbitrage, deregulation, competition 10 Shadow banking activities have been key drivers of M2 and credit growth year to date 12 China vs. US: Simpler, but more bank interconnections, higher corporate leverage 16 Key long-term risks: underwriting standards, pro-cyclicality, liquidity, implicit guarantees 22 Inflation/overheating is the key pressure point for systemic risk, as leverage rises 29 WMPs: Likely to remain strong in 2013, despite fundamental risks 35 Appendix 46 Disclosure Appendix 47 Prices in this report are as of the market close of February 19, 2013 unless otherwise indicated. Goldman Sachs Global Investment Research 2

3 Exhibit 1: China H/A share banks trade at average 1.1x 13E reported P/B and 6X 13E P/E 19-Feb Mkt Cap 12M Potential Adj. P/B (X) P/B (X) P/E (X) P/PPOP (X) Div yld (%) EPS growth (%) ROE (%) Target Upside/ Rating Price (US$ bn) Price downside H-shares (HKD) 2012E 2013E 2014E 2013E E 2013E 2014E 2012E 2013E 2014E 2012E 2013E 2014E 2012E 2013E 2014E 2012E 2013E 2014E ICBC (H) 1398.HK Buy % BOC (H) 3988.HK Neutral % CCB (H) 0939.HK Buy % ABC (H) 1288.HK Buy % BoCom (H) 3328.HK Sell % (9.3) CMB (H) 3968.HK Buy* % CNCB (H) 0998.HK Neutral % Minsheng (H) 1988.HK Buy % CQRCB 3618.HK Neutral % H-share average Excluding CMB H-shares A-shares (Rmb) ICBC (A) SS Buy % BOC (A) SS Neutral % CCB (A) SS Buy % ABC (A) SS Buy % BoCom (A) SS Neutral % (9.3) CMB (A) SS Buy % CNCB (A) SS Neutral % Minsheng (A) SS Neutral % SPDB SS Neutral % Industrial SS Buy* % PAB SZ Neutral % Hua Xia SS Sell % BONB SZ Buy % BOBJ SS Neutral % BONJ SS Sell % CEB SS Sell % A-share average Big banks average Shareholding banks average City Bank Average Source: Wind, Gao Hua Securities Research estimates Goldman Sachs Global Investment Research 3

4 China shadow banking: likely to grow in 2013E; overheating a risk; bank re-rating capped Over the past two years there has been a rapid increase in corporate bond issuance, trust product sales, and wealth management products (WMP), otherwise known as shadow banking activities. These products are often not well understood by the market and have led to concerns on the systemic risks involved, particularly on banks credit exposure to these products, as they are not required to disclose their exposure or make capital and provision charges. In this report, we address the following: Size and impact We estimate c. Rmb24tn in credit was extended in China s through shadow banking activities in FY12 (Exhibit 3), accounting for 24% of total credit to corporates and consumers, and 45% of 2012E GDP. Shadow banking activities made up 39% of the increase in total incremental new credit and 27% in incremental new M2 in FY12. Drivers of growth The rapid growth in shadow banking activities has been driven by strong demand for credit from corporates and rising returns from depositors due to regulated low deposit rates. China regulators also encourage non-bank credit growth to diversify credit concentration away from banks. In addition, regulatory requirements in the regulated banking sector (e.g., loan/deposit ratio, capital adequacy ratio, and credit quota) are driving increasing innovation in financial products given the structural decline in forex capital inflows. China vs. US shadow banking We find that China s shadow banking products are simpler direct credit or informal securitization products, have greater interconnections with banks, are smaller in size but grow more rapidly, contribute to fast-rising corporate leverage, and may have higher implicit bank guarantees given their retail investor focus. Key risks: Rising corporate and financial system leverage; implicit guarantees distort efficient capital allocation Risks to China s shadow banking activities include: (1) relatively weak credit underwriting standards vs. bank loans; (2) fast-rising leverage in the non-bank corporate and financial system; (3) higher liquidity risks vs. bank loans, pro-cyclical growth that could dampen the counter-cyclical macro policies; and (4) investors perception of the implicit guarantees provided by banks and government on trust/bond/wmp, and related moral hazard issues. If these risks are not addressed properly, we believe the systemic risks to China s banks will gradually increase. Investors perception of implicit guarantees (i.e., the bail-out of corporate bonds/trust/wmp by the Chinese government and banks) will likely impact their investing behavior, thus distorting capital allocation. Lastly, regarding the short-term risk, a sharp slowdown in shadow banking activities due to either severe regulatory tightening or a number of credit events triggered by macro tightening could negatively impact M2, new credit growth, and GDP growth. Goldman Sachs Global Investment Research 4

5 Near-term systemic risk is low, but high inflation/overheating could be a key pressure point for system distress Given China s strong internal and external liquidity and still relatively strong mid-term GDP growth outlook, we believe the risk of a systemic crisis in the mid-term is low. However, the rapid rise of corporate leverage could cause corporate EBITDA interest cover/npl formation to be more sensitive to high CPI inflation (e.g. 3%+), due to rising leverage and borrowing costs. Potential overheating, followed by tightening, could cause more collateral damage than in previous cycles. We believe it is vital for China to achieve the fine balance between growth and inflation achieved less by investment stimulus policies and more via policies to boost consumption and/or the competitiveness of corporate China. WMP/trust products to continue their growth in 2013E, despite potential volatility and tighter regulations We expect that growth in WMP/trust products and other shadow banking activities will maintain its strong growth in 2013E albeit slower off a high base and with potential volatility given the government s policies to modestly grow total social financing (TSF) and develop corporate bond markets, favorable banking sector liquidity, and recovery in GDP/corporate earnings. We forecast 2013E new TSF volume to reach Rmb16.9tn, up 7% yoy, with the total TSF balance of growing at 18% yoy vs. 20% in This should be sufficient to support c. 8% GDP growth as the TSF balance growth rate is higher than our EC teams nominal GDP growth estimate of 12.6%. We do not expect regulators to stop or limit the current WMP growth trend, but we believe they will likely improve current disclosure requirements, asset matching, risk management centralization, and investor selection/protection. This could potentially increase short-term volatility, but WMP growth should continue as part of the long-term interest rate deregulation process. The statement by the Central Economic Working Conference of the Communist Party held on December suggests China will maintain modest growth in TSF in 2013, which includes many shadow banking activities. In a worst-case scenario, if regulators require banks to move off-balance sheet WMP credit exposure back to their balance sheet and take the same capital charges, we estimate this would reduce their CAR by a manageable 34bp in 2013E. Raised banks earnings estimates on cyclical macro recovery, but remain valuation disciplined; prefer quality banks We have published our new estimate and rating changes in a separate report published today (February 26 th ) Estimates up but keep valuation discipline; CQRCB/Minsheng A to Neutral; BONB to Buy. Post the recent sharp valuation rebound of China bank shares driven by cyclical macro recovery, we believe the next leg of performance will come from consensus earning upgrade around the FY12/1Q13 result announcement. China banks H/A shares still look reasonably valued in our opinion with current prices implying 6.2%/5.7% 2013 NPLs vs. our 4%-6% estimates. (Please see the end of the report for more detailed discussion of valuation for China banks.) However, we believe the long-term structural issues related to shadow banking are likely to cap valuation re-rating potential of China banks: Rising corporate leverage increases long-term NPL risks, and raises the sensitivity of NPLs to inflation Shadow banking s opaque interconnections with China banks make investors difficult to analyze the true credit exposure of WMP, informal loan securitization in interbank assets, etc. Thus, it is likely to lead to high discount rates for China banks versus global peers. Goldman Sachs Global Investment Research 5

6 Moreover, there are increasing signs of inflation pressure and rising property prices, and therefore a more complicated macro backdrop in FY13 in finding balance between growth and inflation than the straightforward monetary easing situation in 3Q12/4Q12. Although at this point, we do not expect China regulators to adopt harsh tightening measures, as inflation is still low at below regulators comfort zone (CPI was 2% in January, and our ECS team forecasts CPI to gradually trend up to 3.4% in 4Q13),, we believe the broad based cyclical recovery driven share price rebound appear to be over. As such, we prefer banks with ability to achieve higher than peers earning growth, and with high ROE and PPOP earning power to absorb NPL risks if macro/npl starts to deteriorate in an overheating environment. We believe our top picks including ICBC, CCB, ABC, CMB, Minsheng (H), Industrial Bank and BONB have strong franchise, CAR, ROE and earning power, and strong ability to execute growth through their strong deposit franchise, ability to grow SME loans to offset NIM pressure, and strong execution capability. We developed a score table to rank banks in the following criteria (Exhibit 2): Exhibit 2: Buy rated ICBC, CCB, ABC, CMB H/A shares; Industrial, Minsheng (H), Bank of Ningbo A-shares rank highly in the fundamental stock picking scores 2013 ICBC CCB ABC BOC BoCom CMB CNCB Minsheng CQRCB Industrial SPDB PAB Hua xia CEB BOBJ BONB BONJ 1. Ability to absorb impact from lower NIM, rising NPL Low loan/deposit ratio High SME exposure Off-balance sheet guarantee/loans (2011) Ability to execute growth High CAR and return on capital Core tier I CAR PPOP/RWA Internal capital generation- RWA growth Reasonable valuation and dividend yield Implied NPL formation rates that do not match banks' risk profile Dividend yield Total score Note: Score 1 means better than sector average, while score -1 or -2 means worse than sector average. Higher score means better fundamentals while lower score refers to weaker fundamentals. Source: Gao Hua Securities Research estimates. Goldman Sachs Global Investment Research 6

7 We have factored into our 12-month target prices the long-term concerns over shadow banking activities, such as potential underreported credits within WMP or informal securitization: Our 12-month target prices use conservative long-term ROE assumptions on banks leverage/car, NIM, and credit cost assumptions to reflect concerns of under-reported leverage. Our long-term ROE assumptions factor in 10%/9.5% core tier I CAR for big/smaller banks under Basel III capital requirement (vs. minimum 8.5%/7.5% requirement by 2019), and average 112bp credit costs (vs. average 53bp in 2013E). We also use adjusted BVPS to arrive at our target prices, based on average 6.3% total potential NPL ratios for China banks (vs. 1% NPL ratio that we expect the banks to report in 2013), which reduces 2013E BVPS by an average 5.5%/8.9% for H-share banks and A-share banks. We also incorporate the credits within banks WMP into adjusted BVPS calculation. See our sector report, Estimates up but keep valuation discipline; CQRCB/Minsheng A to Neutral; BONB to Buy (Feb 26, 2013). Exhibit 3: China banks current valuation is pricing in an average 2013E NPL ratios for H/A-share banks of 6.2%/5.7%; our Buy rated banks generally have high implied NPL ratios suggested by current share price vs. their fundamentally above-peer asset quality China banks share prices implied NPL ratios E NPL ratio Implied new formation H-share ave Formation covered by general provision Buy-rated stocks A-share ave Source: Datastream, Gao Hua Securities Research estimates. Goldman Sachs Global Investment Research 7

8 China s shadow banking system Rmb24 tn in credit extension The definition of shadow banking varies among international organizations and communities: In the Economic Premise No. 88 published by the World Bank, Shadow banking comprises a set of activities, markets, contracts, and institutions that operate partially (or fully) outside the traditional commercial banking sector, and, as such, are either lightly regulated or not regulated at all. According to the Global Shadow Banking Monitoring report 2012 published by the Financial Stability Board (FSB) in November, 2012, the shadow banking system can broadly be described as credit intermediation involving entities and activities outside the regular banking system. These activities typically involve part or full of the maturity, liquidity and credit transformation. In this report, we focus our analysis of China s shadow banking activities on credit creation outside formal banking channels. We define China s formal banking system to include: banks, group financing companies owned by industrial groups, and financial leasing companies under the supervision of the CBRC and the PBOC. These are deposit-taking institutions whose loans and deposits have been incorporated into the PBOC s credit quota. We estimate Rmb24tn in total credit in the following shadow banking channels in 2012, +35% yoy over 2011 Trust loan products (excluding trusts based on capital markets): Rmb5.4tn, +51% yoy Trust loan products are corporate credit products structured by CBRC-regulated trust companies with a fixed yield, and usually sold to retail investors and distributed by banks or trust companies. Trust companies act as trustee for the retail investors in terms of structuring the products, monitoring the lenders and their projects, and resolving NPLs. Entrust loans (unadjusted for intra group loans): Rmb5.7tn, +29% yoy Entrust loans are loans that corporates lend to each other, using banks as a book entry entity. The PBOC collects such data from banks. Corporate bonds: Rmb6.9tn, +46% yoy Informal loans: Rmb4.6tn, -3% yoy Informal loans are typically loans that individuals lend to corporate/individuals without financial institutions as intermediaries or as book entry entities. Loans securitized through brokers asset management channels: up to around Rmb1tn in 2012 Loans securitized through banks WMPs: Banks WMPs had total assets of around Rmb7.1tn in 2012, +42% vs (Exhibit 4). We estimate around 25% of WMPs or Rmb1.8tn are credit and trust products, as well as around 10% WMP or Rmb710bn in corporate bonds have been packaged in these products. However, most of the banks WMP credit exposure have been incorporated into trust loans, entrust loans, and corporate bond data, as banks need to use a trust product to securitize these loans into WMPs. Please also see Exhibit 44 in the Appendix (page 46) for a detailed breakdown of banking credits and shadow banking credits. Goldman Sachs Global Investment Research 8

9 Exhibit 4: China shadow banking credits reached Rmb24 tn in FY12, representing 24% of total corporate and consumer financing Credits from banks and shadow banking activities in China Base Low High RMB bn E 2013E 2013E Banks' loans and off-balance acceptance 12,913 15,533 16,974 18,597 21,676 26,148 30,055 39,601 48,614 56,045 64,148 73,308 71,326 74,931 As % of total corporate and consumer financing 86% 84% 81% 81% 80% 79% 76% 75% 70% 68% 65% 62% 73% 70% As % of total corporate financing 92% 92% 89% 89% 88% 87% 83% 83% 79% 76% 73% 71% 73% 70% As % of total GDP 107% 114% 106% 101% 100% 98% 96% 116% 121% 118% 124% 125% 122% 128% Yoy growth rate 15% 20% 9% 10% 17% 21% 15% 32% 23% 15% 14% 14% 11% 17% Shadow banking credits 1,055 1,362 2,001 2,208 2,969 3,878 5,992 7,885 12,973 17,405 23,560 30,571 26,609 32,406 As % of total corporate and consumer financing 7% 7% 10% 10% 11% 12% 15% 15% 19% 21% 24% 26% 27% 30% As % of total corporate financing 8% 8% 11% 11% 12% 13% 17% 17% 21% 24% 27% 29% 27% 30% As % of total GDP 9% 10% 13% 12% 14% 15% 19% 23% 32% 37% 45% 52% 45% 55% Yoy growth rate 23% 29% 47% 10% 34% 31% 55% 32% 65% 34% 35% 30% 13% 38% Informal loans + small corporate loans/pawn 964 1,163 1,485 1,431 1,673 1,951 2,885 2,576 3,961 4,679 4,561 5,198 4,789 5,473 As % of total corporate financing 7% 7% 8% 7% 7% 6% 8.0% 5.4% 6.4% 6.4% 5.2% 5.0% 4.9% 5.1% As % of total GDP 8% 9% 9% 8% 8% 7% 9.2% 7.6% 9.9% 9.9% 8.8% 8.9% 8.2% 9.3% Yoy growth rate 17% 21% 28% -4% 17% 17% 48% -11% 54% 18% -3% 14% 5% 20% Trust loans ,444 3,592 5,424 6,812 5,858 7,051 As % of total corporate financing 0% 1% 1% 2% 4% 5% 6% 7% 6% 7% Yoy growth rate 106% 84% 155% 47% 51% 26% 8% 30% Entrust loans (un-adjusted for intragroup loans) ,300 1,979 3,109 4,409 5,689 7,089 6,542 7,395 As % of total corporate financing 0% 0% 2% 2% 2% 3% 4% 4% 5% 6% 6% 7% 7% 7% Yoy growth rate 19% 43% 56% 39% 52% 57% 42% 29% 25% 15% 30% Corporate bonds ,286 2,371 3,459 4,725 6,887 9,472 7,920 9,986 As % of total corporate financing 1% 1% 1% 2% 2% 2% 4% 5% 6% 6% 8% 9% 8% 9% Yoy growth rate 117% 62% 40% 116% 72% 20% 75% 84% 46% 37% 46% 38% 15% 45% Credits securitized in brokers' AM business 1,000 2,000 1,500 2,500 As % of total corporate financing 1.1% 2% 2% 2% Yoy growth rate 100% 50% 150% Memo Bank WMPs' total AUM 467 1, ,015 2,500 5,000 7,100 8,000 7,700 8,500 Credit assets plus trust products ,250 1,775 2,000 1,925 2,125 Corp bonds As % of total GDP 2% 5% 3% 3% 6% 11% 14% 15% 15% 16% Yoy growth rate 184% -28% 6% 146% 100% 42% 60% 54% 70% Note: We believe the loans, corporate bonds and trust loans in banks WMPs have been largely captured in each category of the shadow banking activities, as these credits need to use trust products or entrust products to be packaged. Source: PBOC, CBRC, Wind, China Trust Association, Gao Hua Securities Research estimates. Goldman Sachs Global Investment Research 9

10 How did it grow so fast? Demand, regulatory arbitrage, deregulation, competition We believe the rapid growth of shadow banking activities in China can attribute to: 1) Strong loan demand from corporates due to relatively low real borrowing rates vs. nominal GDP growth rates especially from private companies and SMEs that could not be satisfied by the formal banking system due to loan quotas or sector loan regulations. For instance: Property developers, private mining or manufacturers, and local government-owned financing vehicles, as banks are either limited by regulators on the total amount of loans to give to these sectors, or have insufficient credit quota to serve their credit demand. Big corporate or government-backed enterprises that issue bonds for less expensive financing than bank loans with similar maturities. 2) Regulators continuing to promote direct financing instead of indirect financing through banks, or total social financing (TSF). We expect regulators especially PBOC to continue to promote direct financing through non-banking financing channels instead of indirect financing through banks, as they feel this may help diversify the credit risks away from commercial banks. However, we believe this argument is debatable as increasing corporate debt through non-bank channels would affect banks if there were to be NPLs from such corporates; 3) Strong household demand to diversify financial assets away from Rmb deposits, for which interest rates are still regulated. Banks short-term WMPs typically offer 3.5%+ yield (c. 4%+ for non-principal guaranteed products), vs. c. 2.8% for 3-month deposits. 4) Incentives to securitize or transfer loans outside the formal banking system to avoid the following limits and increase leverage: Credit quota: We forecast Rmb8.8tn in FY13E. 75% loan/deposit (L/D) ratio cap: For example, in order to issue a Rmb100 loan, banks would have to prepare Rmb130 in deposits to meet the L/D ratio cap. In contrast, moving loans to a trust or WMP would mean 100% L/D ratio rather than 75%. Capital adequacy ratio (CAR): For instance, loans offered in trust products and WMPs do not need to book capital charges. 5) The competition among brokers, trust companies and insurance asset management business to help banks securitize or transfer loans outside the formal banking system : Moreover, we expect continued informal loan securitization through emerging channels such as brokers asset management business. These emerging channels directly compete with traditional channels such as trust products, banks WMP products and interbank entrusted payment products. Now brokers charge 5bp-10bp to cooperate with banks to securitize loans, while trust companies still charge 20bp-30bp; and as such, brokers asset under management increased sharply to Rmb1.5 tn in 4Q12 from about Rmb300 bn in 3Q12. And if trust products were to slow down due to regulatory tightening, the brokers asset management companies may replace trust companies in helping banks securitize loans. Goldman Sachs Global Investment Research 10

11 6) Forex capital inflows have declined since 2011 and have slowed M2 and deposits creation, which makes it difficult for banks to comply with 75% L/D rule without informal loan securitization. Given the structural slowdown in M2 and sector deposit growth, in light of the potentially lower volumes of forex capital inflows, we believe the 75% L/D ratio cap and 20% required reserve ratio (RRR) will make it increasingly difficult for China banks to manage liquidity. We estimate the new capital inflow per year in contributed to new M2 creation of 18%-56%. However, capital inflows have declined since 2011, given low market expectations for Rmb appreciation amid a slowdown in China exports due to low global demand and a rise in domestic inflation and labor costs. In our view, China banks will likely continue to be incentivized to securitize, sell loans, and/or classify loans as an off-balance sheet item if capital inflows remain low and China keeps the 75% L/D ratio cap. These new loans have already created M2/deposits but are not subject to the L/D ratio cap and loan quota. Exhibit 5: We expect capital inflows to decline in 2012E/2013E; shadow banking activities (interbank, bond, WMP) will likely contribute to new M2 growth Rmb bn E 1Q13E 2Q13E 3Q13E 4Q13E New M2 in total 3,288 4,140 5,010 5,784 7,172 13,506 11,563 12,575 12,260 12,700 4,466 3,240 1,910 3,084 Rmb new loans 1,920 1,649 3,066 3,634 4,170 9,629 7,951 6,875 8,200 8,800 2,816 2,640 1,760 1,584 as % of total 58% 40% 61% 63% 58% 71% 69% 55% 67% 69% 63% 81% 92% 51% FX inflows 1,774 1,862 2,777 2,940 4,005 2,468 3,268 2, as % of total 54% 45% 55% 51% 56% 18% 28% 22% 3% 7% 7% 5% 8% 10% Fiscal deposits decline ,100 as % of total -3% -4% -6% -12% -1% -3% -3% -1% 2% 4% -3% -11% -5% 36% Interbank, bond, WMP , ,997 3,503 2,500 1, as % of total -9% 19% -11% -2% -13% 14% 6% 24% 29% 20% 34% 25% 5% 3% M2 balance 25,411 29,551 34,560 40,344 47,517 61,022 72,585 85,160 97, , , , , ,120 M2 yoy growth 14.6% 17.6% 16.9% 16.7% 17.8% 27.7% 19.7% 13.6% 13.8% 13.0% 13.8% 13.6% 13.4% 13.0% Source: PBOC, Gao Hua Securities Research estimates. Goldman Sachs Global Investment Research 11

12 Shadow banking activities have been key drivers of M2 and credit growth year to date New credit contribution: c.39% contribution to total debt financing, especially FAI financing Based on the PBOC s disclosed TSF data, we estimate new credit extended through trust loans, entrust loans, corporate bond issuance, and bank acceptance account for 33%, 39% and 39% of total debt financing in FY11/FY12/FY13E, respectively (Exhibit 6). We include bank acceptance data instead of bill discount loans. We find it is difficult to track bill discount loan securitization because bill discount loans are viewed as interbank assets given banks guarantees of these trade bills, and thus they could easily be packaged as part of banks and brokers WMP, as well as interbank assets. In addition, the rapid growth in corporate long-term loans, bond and trust products in FY12 and January 2013 at 46% and 134% yoy (at Rmb5.3tn and Rmb764bn respectively), has been a key source of fixed asset investment financing, which accelerated fixed asset investments and helped stabilize the economy. That said, we believe the TSF data issued by the PBOC underestimate some credits made through the shadow banking system, e.g.: 1) trust products in the TSF do not include equity trust products, and TSF data do not include loans securitized through brokers asset management products, etc; and 2) It also does not account for the fluctuation in informal loans. We believe informal loans could shrink in 2012 vs. 2011, given the small- and medium-sized enterprise (SME) bankruptcies in Wenzhou City, Yangtze River delta, Zhengzhou City, and others. Exhibit 6: New credits extended from trusts, entrust loans, bank acceptance, and corporate bond issuance account for 33%, 39% and 39% of TSF s debt financing in 2011, 2012E, and 2013E, respectively g Rmb bn E Jan-13 Rmb loans 2,770 1,920 1,649 3,066 3,634 4,170 9,629 7,951 7,470 8,200 8,800 1,070 Forex loans , Entrusted loans ,130 1,300 1,280 1, Trust loans ,290 1, Bank acceptances 202 (24) ,331 1,030 1,050 1, Corporate bonds issue ,301 1,201 1,370 2,250 2, Non financial institution equities Others Total financing(new addition) 3,423 2,434 2,009 3,866 5,929 5,839 14,154 14,300 12,830 15,760 16,888 2,540 Total debt financing 3,317 2,327 1,935 3,653 5,336 5,419 13,447 13,414 11,943 14,986 15,988 2,462 L-t loan+trust loans+corp bond 1,992 2,726 6,731 6,011 3,652 5, yoy (%) 37% 147% -11% -39% 46% 134% Medium term to long term loans 1,589 1,985 4,992 4,424 2,081 1, Total financing(balance) 17,486 19,762 21,875 25,723 31,673 37,519 51,665 65,963 78,324 94, ,971 88,384 yoy (%) 24% 13% 11% 18% 23% 18% 38% 28% 19% 20% 18% Total debt financing(balance) 17,275 19,444 21,482 25,117 30,475 35,901 49,340 62,751 73,866 88, ,840 91,303 Shadow banking as % of total debt financing 9.6% 12.5% 10.9% 13.4% 26.4% 22.1% 21.5% 37.6% 32.7% 39.2% 38.7% 49.3% Source: PBOC, Gao Hua Securities Research estimates. Goldman Sachs Global Investment Research 12

13 Rapid shadow banking growth in 2013E support macro recovery, if not acceleration Despite the potential modest slowdown of new trust volume in 2013E due to strengthened due diligence, distribution and regulatory requirements, we forecast the total social financing (TSF) balance to grow 18%, after our meetings with several regulators, trust companies and banks WMP product managers in January As such, we expect the continued and firm macro cyclical recovery in FY13E with limited downside risks to our 8.2% GDP growth outlook per our ECS team forecasts, as the TSF balance growth rates are higher than our 13% nominal GDP growth rate estimate. Exhibit 7 shows that TSF balance growth rates have a 1-year lead over nominal GDP growth since Moreover, we estimate 2013E new TSF volume will account for 29% of GDP, stable vs. 30% in 2012 (Exhibit 8). For details, please see our banking sector report, titled Shadow banking trip takeaways: Cyclical risk low; LT risks remain, Jan 22, Exhibit 7: TSF balance growth rates have a 1-year lead over nominal GDP growth since % 35% 30% 25% 20% 15% 10% 5% 0% TSF balance growth Nominal GDP growth Exhibit 8: We estimate 2013E new TSF volume will account for 29% of GDP, stable vs. 30% in 2012 (Rmb bn) 18,000 16,000 14,000 12,000 10,000 8,000 6,000 4,000 2,000 - (2,000) 25% 15% Rmb loans 11% 18% 22% 19% 42% 36% 27% Total financing(new addition) As % of GDP 45% 40% 35% 30% 30% 29% 25% 20% 15% 10% 5% 0% Source: Wind, PBOC, Gao Hua Securities Research estimates Source: Wind, PBOC, Gao Hua Securities Research estimates New M2 contribution in FY12: around Rmb3.5tn, or 27% total incremental new M2 We estimate c. Rmb3.5tn in new M2 and Rmb deposits were generated by shadow banking activities in 2012, representing 27% of total M2. Contribution to M2 growth ytd likely came from new Rmb loans, capital inflows/outflows, an increase/decrease in fiscal deposits, and non-bank or shadow banking activities (Exhibits 9-10). We see three ways that shadow banking activities could impact M2 creation: Banks purchase corporate bonds in the primary corporate bond issuance market: This is similar to banks issuing loans to corporates. However, it differs in that the loan is booked as a securities investment rather than as a loan in the banks balance sheets. We estimate investments in corporate bonds by banks own balance sheet or WMPs accounted for around 52% of corporate bonds issuance in 1H12, partly due to the small size of money market funds in China (Exhibit 11). Goldman Sachs Global Investment Research 13

14 Exhibit 9: 2012 new M2 generation from interbank /bond /WMP totaled Rmb3.5tn in FY12s Exhibit 10: Dec. monthly M2 movement: interbank and WMP together contributed around Rmb1.3tn to M2 Rmb bn Breakdown of FY12 M2 growth Rmb bn Breakdown of Dec12 M2 growth 100,000 98,000 96,000 94,000 92,000 90,000 88,000 86,000 84,000 82,000 80,000 78,000-85,160 Dec11 M2 8,200 Loan growth FX inflows Fiscal deposits inflows 3,368 Interbank, bond, WMP 97,420 Dec12 M2 98,000 97,500 97,000 96,500 96,000 95,500 95,000 94,500 94,000 93,500 93,000-94,480 Nov12 M2 454 Loan growth 135 FX inflows 1191 Fiscal deposits inflows 1160 Interbank, bond, WMP 97,420 Dec12 M2 Source: PBOC, Gao Hua Securities Research. Source: PBOC, Gao Hua Securities Research. Exhibit 11: Commercial banks account for the 51.5% investment in corporate bonds, short-term and mid-term corporate paper Investor distribution of corporate bonds, 1H % Rural banks 10.1% City banks 7.6% Foreign banks 0.3% 24.4% Nationwide banks 33.5% 13.2% 3.2% Banks Brokers Insurers Asset mgmt. firms 0.0% Individuals 7.7% Others Source: Chinabond.com, Gao Hua Securities Research estimates. Goldman Sachs Global Investment Research 14

15 Loan securitization through trusts, WMPs, and interbank channels: Banks lend to corporates that generate new M2. At the same time, banks may use trust products to securitize these loans or discounted bills into their WMPs. They may also book these loans through interbank assets (i.e., interbank entrust payment products, whereby bank A asks bank B to issue a loan to a corporate, while bank B books the loan as an interbank asset to bank A). Month-end and quarter-end WMP fluctuation: Banks typically allow a large amount of WMPs to mature before month-end or quarter-end, so that these funds flow back to the banks balance sheets as deposits to meet the L/D ratio requirement. Thus, M2 increases, although there is no new M2 creation from these activities. After the month-/quarter-end, banks typically start issuing WMPs to attract or retain their deposit customers, and as such, deposits flow to WMPs and M2 decreases. c.13% M2 growth in 2013E, largely dependent on continued growth in shadow banking activities We forecast around 13% yoy M2 growth in 2013E, vs. 14% in 2012, based on our projection of the new Rmb loans, capital inflows, and shadow banking securitized new loans (Exhibit 5). We view this as largely a neutral monetary stance, rather than a tightening stance, given our GS Global ECS Research team s forecast for real GDP growth of 8.2%. Our detailed M2 contribution forecasts (Exhibit 5) include: New Rmb loans of Rmb8.8tn, up from Rmb8.2tn in 2012E Foreign exchange inflows of Rmb900bn in 2013E, a yoy increase of Rmb540bn Fiscal deposit release of Rmb500bn, or a slightly higher fiscal deficit in 2013E vs. 2012E Shadow banking activities including bond, WMP, and interbank activities will continue to contribute Rmb2.5tn in 2013E Key risk to our forecasts: Whether China regulators could significantly tighten or slow shadow banking growth in We expect some potential tightening in risk management of shadow banking activities. This may lead to some volatility in growth; however, growth should persist given the pro-growth economic policies, stabilization of corporate profitability and corporate loan asset quality to limit corporate defaults, and the PBOC s M2 growth target of c.13% yoy. Moreover, the statement by the Central Economic Working Conference of the Communist Party held on December suggests China will maintain a modest growth in TSF in 2013, which includes many shadow banking activities. Goldman Sachs Global Investment Research 15

16 China vs. US: Simpler, but more bank interconnections, higher corporate leverage We compare and contrast the shadow banking activities in China and the US: 1) Simpler products in China: Most are either debt instruments or the first layer of loan securitization China s shadow banking activities may be viewed as the first-layer securitization of loans. For instance, trust loan products could be viewed as privately placed bonds, while banks may sell, transfer, or securitize loans through WMPs. In contrast, the US has well developed non-bank financing and capital markets, including corporate bonds, mortgage-backed securities (MBS), and other products such as structured investment vehicles (SIV), collateral debt obligations (CDO), which are more complex products based on securitized asset-backed securities (ABS)/MBS or corporate loan portfolio. 2) Greater interconnection with China commercial banks: bridge loans, liquidity support, regulatory arbitrage There is a stronger interconnection between China s shadow banking activities and the commercial banking system, including: Shadow banking usually provides bridging loans to corporates so they can make bank loan payments, as China banks have a strict regulatory requirement of not allowing corporates to revolve any of their working capital loans, i.e., all corporate loans have to be repaid before borrowing them again even for trade finance purposes. As such, China corporates have to manage their liquidity more strictly than their overseas counterparts. Many corporates seek short-term bridge loans from informal loan lenders or entrust loan lenders to meet their short-term liquidity needs. Many of these activities may be used by banks to arbitrage loan quota, L/D ratio, and capital requirement. For instance, banks can sell or transfer loans to their WMPs by using a trust loan structure. In this way, banks can serve customers loan needs without having to meet the credit quota, L/D, or capital requirement. Moreover, banks may act as liquidity providers to WMPs and trust products in the event of a default. Given China s social stabilization policy, there may be implicit guarantee to banks WMPs, corporate bonds or trust products borrowed by the stateowned enterprises (SOE) or big private companies. This means that some banks may be asked by the government to provide liquidity or to bail out a defaulted corporate borrower. Exhibit 13 summarizes the connections between China s shadow banking activities and official banking activities. In contrast, the underlying assets for US ABS, MBS, SIVs, and CDOs originate from banks or specialized finance corporations. These products may be bought by institutional investors and mutual funds. There are fewer social concerns in the event of an underlying asset default in the US, and there is no need for banks to satisfy a loan quota and L/D ratio limit. Goldman Sachs Global Investment Research 16

17 3) Size: China s shadow banking activities are much smaller than the US, but have grown rapidly Exhibit 12 shows that China s shadow banking activities are smaller than the US if we measure the underlying credits behind these two systems. China: The size of shadow banking credit exposure has grown rapidly since Shadow banking credits in 2012 accounted for 24% of the total credit balance to corporate and consumers, vs. 15% in 2009 (Exhibits 4 and 14). US: Despite the decline in activity since the global financial crisis, US shadow banking is now similar in size to US commercial banks, as measured by liabilities vs. US commercial banks (Exhibit 15). 4) Credit exposure more skewed to corporate leverage vs. consumer leverage in the US China s shadow banking is mainly exposed to corporate credit and has little exposure to household credit, while a significant amount of US securitization and related structure products are based on mortgages or MBS, or other consumer credits such as credit card receivables. Most of the credit exposure of shadow banking activities such as trust, entrust loans, WMPs and corporate bonds, and the majority of informal loans in China are extended to corporates. Thus, the rapid growth of these products will likely increase already-high corporate leverage. 5) Investor base: The majority of trust/wmp investors are retail investors in China, vs. institutional investors in US The high proportion of retail investors in China also highlights the risk that banks may be required to provide liquidity support to troubled corporate borrowers in the event of a credit default to achieve social stability goals. Other risks include reputational risk, or bank staff mis-selling WMPs to retail customers who are not suitable for buying such products. China is similar to the US during the 1980s: Interest deregulation accelerated MMF/bond growth, as the backdrop for disintermediation, shadow banking activity growth In the US, interest rate deregulation in the 1980s triggered MMF growth and individual investors moved their deposits out of the banking system into money market funds (MMF), thus raising banks L/D ratios. In 2011, assets under management at MMFs in China totaled less than Rmb1tn (or US$170bn), vs. US$2.5tn in the US and US$8.8tn for the entire mutual fund market in the US. We believe China banks WMPs are similar to China MMFs in terms of the underlying investment assets, and China banks manufacture, manage, and distribute these products themselves. To some extent, the majority of profits generated by WMPs are retained within the China banks. In contrast, profits generated by MMFs in the US move out of the banks. As such, we view China s WMP growth as a way for its banks to mitigate the impact of increasing disintermediation trend of bank deposits moving to high yield WMP or MMFs, and as a result of deposit interest rate deregulation in China which started in June Goldman Sachs Global Investment Research 17

18 Exhibit 12: China shadow banking activities are relatively simple loans or loan securitization, vs. the US Diagram of shadow banking activities in China vs. US, and the regulators Source of Funds (deposits) Credit Intermediation Use of funds (borrowing) Regulators NPL loss absorption Traditional commercial banks (intermediated) Households/ Corporations Shadow Banking activities Deposit taking banks, group finance companies and financial leasing companies, take deposits, originate and hold loans Households/ Corporations CBRC/PBOC on CAR, L/D, RRR, NPLs, lending standards, etc Banks recover/restruture NPLs, and use profits to absorb losses 1. Non-intermediated Individuals Corporates Informal loans Entrusted loans Corporates lend to each other Banks work as book entry entities Individuals/ Corporates Corporates 1. No regulators on specifc activities 2. Interest rate cap at 4x of PBOC's benchmark rates Lenders bear the losses. Lenders bear the losses. But banks may be hurt given the cross-lending of corporate 2. Intermediated Qualified individual investors/ Corporates Individuals/ Corporates Trust products Trust companies structure/package loans, euqity investment, mezzanine debts Banks WMP Banks package and securitize loans Corporates Corporates CBRC has some capital requirements (50X+ leverage on AUM/equity) and product marketing CBRC has some product marketing regulations Legally trust companies and banks are not liable for NPL loss for trust and WMPs, but in practice, they may be asked to pay customers' loss after recovering the assets given social stability concerns Individuals/ Corporates Brokers' WMP Brokers package and securitize loans, usually under banks' request Corporates CSRC has some capital requirements (50X+ leverage on AUM/equity) and product marketing regulations For banks' loan securitization through brokers' WMP, brokers expect banks to provide implicit liquidity support Individuals/ Corporates Corporate bonds Brokers/banks as underwriters to help corporate issue bonds Corporates Prior approvals from NDRC, CSRC, PBOC; disclosure requriement Investors should bear the loss though some local government tends to bail-out local SOE debts US shadow banking (multiple intermediaries) Households/ Corporations/ Institutions/ Securities Lenders/ Pension funds MMF purchases CP, ABCP, repo, etc. ABS intermediatio n ABS issuance Loan warehousing Loan origination Household/ Business Borrowing SEC regulates securitization Investors bear the loss NOTE: MMF is money market mutual fund; CP is commercial paper; agreement, and ABS is asset-backed securities. ABCP is asset-backed CP, repo is repurchase Source: Federal Reserve Bank of Dallas, Gao Hua Securities Research. Goldman Sachs Global Investment Research 18

19 Exhibit 13: China shadow banking activities: most activities have an interconnection with banks As of end-2012 Bank WMPs Total assets: RMB 7.1tn Total credits: RMB 1.8tn Supervised by CBRC 1.shift deposit back and forth 2.securitize loans as a channel to banks 3.generate fee income for banks Informal loans Total size: About RMB 4.5tn 1.work as bridge loans of SMEs/property firms to repay/rollover bank loans OFFICIAL BANKING SYSTEM Commercial banks Banks, group financing companies, financial leasing firms Assets: RMB 131tn loans: RMB 67tn Entrusted loans Total credits: RMB 5.7tn No regulation limit on capital Firms lend to each other 1.not account for credit quota 2.work as bridge loans of corporates to repay/rollover bank loans Corp bonds (incl. LGP bonds) Total size: RMB 7tn 1. about 52% bonds bought by banks and their WMPs 2.A significant part underwritten by banks Trust companies Total AUM: RMB 7tn Total credits: RMB 5.3tn Supervised by CBRC 1. banks distribute trust and make fee income 2. Securitize loans as a channel to banks 1. securitize loans as a new channel to banks Brokers asset mgmt. Total credits: RMB 1tn Capital regulation similar to trusts Source: Gao Hua Securities Research estimates. Goldman Sachs Global Investment Research 19

20 Exhibit 14: China s shadow banking credits accounted for 24% of total credit in 2012 Shadow banking as % of total corporate and consumer debts Exhibit 15: US shadow banking liabilities exceeded commercial banking activities since 1995 and are similar in size to commercial banks in 2012 The size of commercial banks and shadow banking liabilities 30% 25% 26% 24% 20% 15% 10% 5% 0% E Note: US shadow banking liabilities equal the sum of the following liability items from the flow of funds accounts: CP, repos, borrowed securities, agency- and GSE-backed securities, mortgage pools, ABS, and MMFs, minus commercial banks federal funds (excess reserves), repos, and CP. Abstract from the Staff paper by the Federal Reserve Bank of Dallas, published in November 2012, titled understanding the risks inherent in Shadow banking. Source: PBOC, Wind, Gao Hua Securities Research estimates. Exhibit 16: Compared with other developing countries, China s shadow banking system is relatively small vs. Thailand, Philippines Share in total financial sector assets in 2010 Source: Flow of Funds Accounts of the United States, the Federal Reserve Bank of Dallas. Exhibit 17: Shadow banking s share in total financial sector has grown in the past few years Share in total financial sector assets in 2010 Note: For China, informal lending is not captured in this chart. Abstract from the paper of World Bank Economic premise published in November 2012, titled Chasing the Shadows: How Significant Is Shadow Banking in Emerging Markets? Source: FSAPs, Central bank, other regulatory agency reports, World Bank. Note: For China, informal lending is not captured in this chart, which cause under estimation of shadow banking s share, especially in 08/09/10. Abstract from the paper of World Bank Economic premise published in November 2012, titled Chasing the Shadows: How Significant Is Shadow Banking in Emerging Markets? Source: FSAPs, Central bank, other regulatory agency reports, World Bank. Goldman Sachs Global Investment Research 20