WHY DOES THE NEGOTIABLE CERTIFICATE OF DEPOSIT MATTER FOR CHINESE BANKING?

Transcription

1 WHY DOES THE NEGOTIABLE CERTIFICATE OF DEPOSIT MATTER FOR CHINESE BANKING? Kerry Liu* Abstract The Negotiable Certificate of Deposit (NCD) is a major financial instrument in China; the value of outstanding Chinese NCDs was approximately RMB8 trillion (equivalent to USD1.2 trillion) in mid This article reviews the most recent developments in the Chinese NCD, including its effect on interest rate pass-through and money creation. Empirical results show that the introduction of the NCD in 2014 influenced the relationship between the policy rate and the lending rate of Chinese commercial banks, and the interest rate pass-through from the policy rate to the lending rate of commercial banks improved subsequently. However, this process has also been accompanied by maturity mismatch, increasing leverage, and decreasing credit ratings for the assets of banks and other financial institutions. Consequently, systemic risk in the Chinese banking system has increased. JEL codes: E52, E58. Keywords: interest rate liberalisation; interest rate pass-through; money creation; Negotiable Certificate of Deposit; People s Bank of China. 1. Introduction The first Negotiable Certificate of Deposit (NCD) was issued in 1961 by the First National City Bank of New York in the United States of America (USA) (Ruebling 1970). In China, the NCD market began on 9 December 2013, when the People s Bank of China (PBoC, China s central bank) issued the Interim Code for the Administration of the Certificate of Deposits 1 (hereafter referred to as the Code). Subsequently, the Chinese NCD market boomed. As of July 2017, the outstanding balance of NCD custody had reached approximately RMB8 trillion (equivalent to USD1.2 trillion). The NCD is now a major financial instrument in China, but to date little research attention has been paid to it and its effects on economic issues such as interest rate pass-through, money creation, and overall systemic risk in the Chinese banking system. This article presents the first review of the impact of NCDs in China. Section 2 provides a brief introduction to the NCD, especially in the USA, and a review of some relevant studies. Section 3 presents some stylised facts about the Chinese NCD market. Section 4 contains an examination of the NCD s effect on interest rate pass-through in China. Section 5 presents a discussion of some policy implications of the NCD. Section 6 concludes the article. 2. The basics of NCDs ACertificate of Deposit (CD) is a time deposit. In the USA, the Federal Deposit Insurance Corporation insures CDs for banks and the National Credit Union Administration does so for credit unions. CDs differ from savings accounts in that they have a specific, fixed term (often one, three, or * Economist at a major international bank in Sydney, Australia. kerry.luke@gmail.com. The author would like to thank an anonymous referee and the Editor for their helpful comments on an earlier version.

2 ECONOMIC AFFAIRS, VOLUME 38, NUMBER 1 97 six months, or between one and five years) and (usually) a fixed interest rate. The bank requires that customers hold the CD until maturity, at which time they can withdraw the principal and the accrued interest. In exchange for the customer depositing the money for an agreed term, institutions usually grant higher interest rates than they do on accounts that customers can withdraw from on demand. A Negotiable Certificate of Deposit (NCD) is a CD with a minimum face value of USD100,000; it is guaranteed by the bank and can usually be sold in a highly liquid secondary market. Financial institutions generally offer NCDs on a short-term basis, meaning the maturity date is one year or less. The development of the NCD in the USA in the early 1960s was closely associated with the process of interest rate liberalisation. Following the bankruptcy of Penn Central Company in 1970, the market became less interested in the commercial papers issued by financial institutions. In response, the Federal Reserve suspended interest rate ceilings on large time CDs with maturities of less than 90 days to enable banks, which were suddenly shut out of the commercial paper market, to raise funds to meet their liquidity needs (Carlson and Wheelock 2015). The Federal Reserve suspended Regulation Q for large time CDs with longer maturities in May 1973 (Cook 1978). 2 Interest rate ceilings were completely lifted in the 1980s. Winningham (1979) examined the NCD market in the USA in the 1960s and 1970s, and estimated that the introduction and spread of the NCD would have a small negative impact on the growth of currency, demand deposits, and M1, a somewhat greater negative impact on the growth of time and savings deposits net of NCD, and a positive impact on the growth rates of M2 and M3. Carlson and Wheelock (2016) studied the NCD market in the USA during the same period, and found that reduced availability of short-term safe public assets, specifically Treasury bills, increased the spreads between the rates on the new instruments and Treasury bill yields, but at the same time tended to boost their market volumes. 3. The Chinese NCD market As stated earlier, the Chinese NCD market began on 9 December According to the Code, the Chinese NCD is available only in the interbank market, not to individual investors. Also, the Chinese NCD is not applied to reserve requirements. NCD interest rates mirror the market rates, the Shanghai Interbank Offered Rate (Shibor 3 ). The main purpose of NCD introduction was to liberalise interest rates, mainly the deposit and lending rates of commercial banks, which remained under regulation until they were fully liberalised on 23 October The Chinese NCD market has boomed since Monthly NCD issuance has increased from around RMB320bn (equivalent to USD52bn) in March to around RMB2 trillion (equivalent to USD296bn) in June The outstanding balance of NCD custody reached around RMB8 trillion (equivalent to USD1.2 trillion) in June The fraction of NCD custody balance over total bond custody balance increased from just 3.5 per cent in March 2015 to 17.2 per cent in June (more detailed information is presented in Figure 1). Some background on the Chinese banking system is a necessary prelude to a discussion of the identity of NCD issuers. At end 2016, Chinese bank assets hit USD33 trillion, which compares with USD31 trillion for the Eurozone, USD16 trillion for the USA and USD7 trillion for Japan. 6 The five largest banks the Industrial and Commercial Bank of China, China Construction Bank, Bank of China, Agricultural Bank of China, and the Bank of Communications are state-owned, and account for around 37 per cent of Chinese banking system assets and deposits as of the second

3 98 K. LIU Figure 1: Amount of Chinese NCD issuance and custody balance, and fraction of NCD custody balance over total bond custody balance by month during March 2015 June Sources: Wind Info; Shanghai Clearing House. quarter of The Chinese government owns the majority of these banks shares; private shareholders own a small fraction. Twelve mid-sized joint-stock commercial banks account for around 18 per cent of Chinese banking system assets and deposits, and small city commercial banks, 133 in all, account for around 12 per cent. There are also over 2,000 rural commercial banks/credit unions, which are very small in terms of size of assets and liability. 7 Finally, there are three policy banks the Agricultural Development Bank of China, China Development Bank, and the Export Import Bank of China. Joint-stock commercial banks and city commercial banks are the dominant players in the Chinese NCD issuing market; state-owned banks and policy banks are involved only marginally. The reasons for this are straightforward. Policy banks do not need to compete for deposits in the market. Stateowned banks have the largest number of branches across China, and they have the greatest capacity to absorb deposits from personal and corporate clients. This places joint-stock commercial banks and city commercial banks at a disadvantage in competition for personal and corporate deposits, so they turn to the NCD market as an alternative source of funding. In 2017 the city commercial banks overtook the joint-stock commercial banks in the NCD issuing market, indicating their increasing reliance on NCDs (more detailed information is presented in Figure 2). The ratio of NCD balance to deposit balance is 13.9 per cent 8 for the city commercial banks, the highest among all types of banks in China. The profile of NCD investors has also changed. Before September 2015, state-owned banks were the main source of NCD funding. This made sense, as state-owned banks can absorb large volume of deposits as a cheap and stable funding source, and accordingly can provide funding to smaller banks. After September 2015, money market funds changed the game, providing per cent of total NCD funding between September 2015 and June 2017 (more detailed information can be found in Figure 3). This major change was the result of a policy shift in August 2015: The China Fund Association issued the Guidelines on the Accounting and Valuation of Securities Investment Fund Participation in NCD (Trial). 9 The Guidelines cleared the way for money market fund

4 ECONOMIC AFFAIRS, VOLUME 38, NUMBER 1 99 Figure 2: NCD issuance by type of bank, per cent. Source: Wind Info. Figure 3: NCD owned by type of institution, per cent. Source: Wind Info. investment in NCDs, which proceeded apace. Chinese money market funds had around RMB3.6 trillion (USD520bn) in net assets in January 2017, making China the second-largest market for such funds globally after the USA (McLoughlin and Meredith 2017). That growth partly reflects the attraction of higher returns on money market funds than bank deposits and easier access through technological innovations. Money market funds and NCDs have played an important role in liberalising interest rates in China, similar to the experience in the USA in the 1970s (International Monetary Fund 2010). The interest rates on NCDs issued by rural commercial banks, which have the lowest level of credit ratings, are the highest. Rates on NCDs issued by city commercial banks are generally the second highest, followed by rates on NCDs issued by the joint-stock commercial banks. The stateowned banks, which have the highest credit ratings, offer the lowest interest rates (more detailed

5 100 K. LIU Figure 4: Issuing interest rates of 3-month NCD by type of bank (per cent) during April 30 June Note: Interest rates from other maturities of NCD and other time periods before April 2017 demonstrate a very similar pattern. information can be found at Figure 4). This interest rate pattern has some important implications, which are further discussed in section The effect of NCDs on interest rate pass-through from the policy rates to the lending rates of commercial banks The data sets employed for empirical analysis in this section are the same as those used in Liu (2017). Accordingly, the lending rate of commercial banks is defined as the annualised quarterly lending rate calculated as quarterly interest income over interest-earning assets. Interest-earning assets include loan balances, deposits with other banks and other financial institutions, debit funds, transactional, held-to-maturity and available-for-sale financial assets, resale financial assets, and investments, but exclude receivable loans. The benchmark lending rate is the quarterly average of the six-month to one-year benchmark lending rate set by the PBoC. The policy rate is the quarterly average of FR007 (seven-day fixedreporate). Other control variables are as follows. The deposit liability ratio is defined as deposits over total liability. The profitability variable is ROE (return on equity). The size variable is the natural log value of total assets. The loan deposit ratio is defined as loans over deposits (it existed to control the prudential regulation requirement of a 75 per cent limit, which was lifted in October 2015). Another variable is the reserve ratio. All data items are from the quarterly financial reports of 19 banks 10 in China ranging from Q3, 2007 to Q2, 2016, obtained from Wind Info. In this section, first, a Chow test is conducted to examine whether the introduction of NCDs starting from 2014 represents a structural break between the relation between the policy rate and the lending rate of commercial banks. Second, a dummy variable (NCD_Dummy) is created to examine the effect of NCDs on interest rate pass-through from the policy rate to the lending rate of commercial banks. (It would also be useful to test whether the volume of NCD issuance influences the interest rate pass-through; however, since Chinese listed banks publish their financial reports only quarterly, the available data sets are too few for a regression analysis. This issue will be addressed in future research, once the number of data items has increased sufficiently.)

6 ECONOMIC AFFAIRS, VOLUME 38, NUMBER Chow test The Chow test is most commonly used to test for the presence of a structural break in a period. A detailed explanation of this test is presented at Appendix A, and the result is presented in Table A1. The Chow test shows the null hypothesis that no breaks exists in Q1, 2014 is rejected within 1 per cent confidence level. Including or excluding the control variables discussed in section 4.2 makes no significant difference. This means that the introduction of NCD in Q1, 2014 significantly influenced the relationship between the policy and lending rates of Chinese commercial banks Multivariate regression A multivariate regression was conducted to examine the effect of NCDs on interest rate pass-through from the policy rate to lending rate of Chinese commercial banks. NCD_Dummy equals 1 for Q1, 2014, otherwise equals 0. The model specification is presented in Appendix B. The regression result is presented in Table A2. The result shows that the regression model works very well. The coefficient of FR007 is positive and significant within a 1 per cent confidence level, indicating that, if the policy rate increases one unit, 0.23 units of policy rate will be passed to the lending rate of commercial banks. The coefficient of interaction term NCD_Dummy* FR007 is positive and significant within a 5 per cent confidence level, showing that this interest rate pass-through improved after the introduction of NCDs. Specifically, an extra 0.05 units of policy rate will be passed to the lending rate of commercial banks. Other variables also perform well. The coefficient of Benchmark Lending Rate is also positive and significant. The performance of the control variables is very similar to the results reported by Liu (2017). 5. Policy implications and discussion The introduction and rapid development of NCDs is part of China s new monetary policy framework. Before 2015, as the result of the appreciating RMB, expectation of further appreciation and continual (and dominant) foreign exchange inflow, the balance sheet of the PBoC had expanded rapidly. After 2015, things changed. Because of huge capital outflow, China s foreign exchange declined from the peak level of around USD4 trillion in June 2014 to just slightly over USD3 trillion in PBoC s balance sheet expansion decelerated, reflecting a change in money creation. As capital outflows drained domestic liquidity, the PBoC began to provide liquidity through reserve ratio cuts in 2015 and open market operations and other re-lending facilities including short-term liquidity operations, standing lending facilities, mid-term lending facilities, pledged supplementary lending and so on in 2016 and 2017 (Liang and Yu 2017). However, the PBoC requires commercial banks to provide collaterals or pledges in order to access these facilities. Because mid-sized and small banks possess few qualified assets suitable for collaterals/pledges, the big banks, mainly the five state-owned banks and some joint-stock commercial banks, obtain most of the new liquidities the PBoC provides. The small banks, mainly city and rural commercial banks, and some mid-sized banks then turn to the issuance of NCDs as a way of obtaining liquidity from big banks, which have abundant funds from the

7 102 K. LIU PBoC. This method of money creation is sharply different from the foreign exchange-centred model that existed before This new model of money creation increases the systemic risk in China s banking system. After the big banks (the five state-owned banks and some joint-stock commercial banks) obtain most of the new liquidities from the PBoC, they can invest these funds in either the real economy or financial markets. In the real economy, EBIT ratios (definedasearningsbeforeinterestandtaxovertotal Assets for sectors including energy, materials, industrials, discretionary consumption goods, nondiscretionary consumption goods, health-care service, and financial services) have been declining since around The only exceptions are the telecommunication service and information technology sectors. 12 Contraction or slowdown in economic activity may have increased the likelihood of borrowers being unable to repay their loans. Under such conditions, the big banks tend to invest in less risky assets with reasonable and sometimes high yields, including NCDs. Mid-sized banks use the funds raised from issuing NCDs to invest in higher-yield NCDs issued by small banks, including city and rural commercial banks. The small banks then invest the funds raised from issuing NCDs in even higher-yield products, including wealth management products, 13 or entrust the funds to brokerages or mutual funds which pursue higher returns still (see Figure 4 for the interest rates of NCDs issued by different types of banks). This whole process is generally accompanied by maturity mismatch, increasing leverage, and decreasing credit ratings for the assets of banks and other financial institutions. Hence, systemic risk in Chinese banking system has increased. Since NCDs only began to attract attention from analysts and regulatory authorities from late 2016, little research has been done on the risks they bring to the Chinese banking system. One major restriction on research is poor data availability. The reporting, collecting and analysis of NCD data is still an ongoing progress of updating and revision. Previous studies have focused on the measurement of systemic risk in the Chinese banking system (Huang et al. 2017), specific risks caused by specific shadow banking activities such as rollover risk from wealth management products for Chinese banks (Acharya et al. 2016), and balance-sheet risks from entrusted lending to small Chinese banks (Chen et al. 2016). Responding to the increasing systemic risk in the Chinese banking system, Chinese regulatory authorities started to strengthen the regulation of the NCD market in early The Notice on Regulating the Interbank Business of Financial Institutions, 14 issued in 2014 by the Chinese Banking Regulatory Commission (CBRC) and other regulatory authorities, stated that the total amount of interbank borrowing, interbank deposit/borrowing, buyback resale/sellback repurchases, and so forth should not exceed one-third of total liabilities. However, in order to promote interest rate liberalisation, the Notice did not define NCDs as interbank liabilities. Under the new policy, from 11 April 2017 the NCD is identified as a factor for the CBRC s consideration. Maturity mismatch and liquidity risk management have become major concerns for the CBRC. Simultaneously, Chinese regulatory authorities including the CBRC and the PBoC need to balance multiple objectives, including promoting interest rate liberalisation, encouraging financial innovation, and strengthening prudential regulation. 6. Conclusions This study provides a novel review of recent developments in Chinese NCDs, including their effects on interest rate pass-through and money creation. While NCDs have existed in China for only a short

8 ECONOMIC AFFAIRS, VOLUME 38, NUMBER period, the development of the Chinese NCD market has been extraordinarily rapid. Just as in the USA, the initial purpose of introducing NCDs was to facilitate interest rate liberalisation. The Chow test reported in this article shows that the introduction of NCDs in 2014 influenced the relationship between the policy and lending rates of Chinese commercial banks. Moreover, multivariate analysis showed that the interest rate pass-through from the policy rate to the lending rate of commercial banks improved after NCD introduction. More specifically, an extra 0.05 unit of policy rate will be passed to the lending rate of commercial banks for the change of 1 unit of policy rate. However, while interest rates have been liberalised, and money creation has changed from being foreign exchangecentred before 2015 to being heavily NCD-focused from 2016, this process is accompanied by maturity mismatch, increasing leverage, and decreasing credit ratings for the assets of banks and other financial institutions. Systemic risk has increased in the Chinese banking system. In response, Chinese authorities began to strengthen regulation of NCDs in early Appendix A Chow test The Chow test is a test of whether the true coefficients in two linear regressions on different data sets are equal. In econometrics, it is most commonly used to test for the presence of a structural break in a period which can be assumed to be known a priori (for instance, a major event) (Chow 1960). Table A1: Result of Chow breakpoint test: Q1, 2014 Chow breakpoint test: Q1, 2014 Null hypothesis: No breaks at specified breakpoints Varying regressors: All equation variables Equation sample: F-statistic Prob. F (2,455) Log likelihood ratio Prob. Chi-Square (2) Wald statistic Prob. Chi-Square (2) Appendix B Multivariate regression The econometric model specification is presented as follows: Lending Rate i;t ¼ δ 0 þ δ 1 FR007 i;t þ δ 2 Benchmark Lending Rate i;t þ δ 3 NCD Dummy FR007 þ δ 4 Deposit Liability Ratio i;t þ δ 5 Loan Deposit Ratio i;t þ δ 6 Reserve Ratio i;t þ δ 7 ROE i;t þ δ 8 LnðsizeÞ i;t þ fixed effects þ ε The basic form of this model assumes that the standard errors have the same variance across all observations. Since the panel data sets have multiple observations for the same firm, the standard errors are possibly correlated across firms and/or years. Heteroscedasticity-consistent standard errors should be applied. In this study, the Huber White standard errors are applied.

9 104 K. LIU Table A2: Regression results for the sample period Q3, 2007 Q2, 2016, with consideration of the NCD effect. Variable Coefficient Prob. (%) C FR007 (Policy rate) Benchmark Lending Rate NCD Dummy*FR Deposit liability ratio Loan deposit ratio Reserve ratio Return on equity Ln(size) Number of observations 459 Adjusted R-squared Notes: Cross-section fixed dummy variables are included and panel least squares are applied with White standard errors. Dependent variable: Lending Rate of Commercial Banks. Notes 1. (in Chinese) (accessed 30 August 2017). 2. Regulation Q was a Federal Reserve regulation which prohibits the payment of interest on demand deposits, and imposes interest rate ceilings on other types of saving accounts including time deposits (Cook 1978). Regulation Q was repealed in July 2011 (Board of Governors of the Federal Reserve System 2011). 3. Shibor is a daily reference rate based on the interest rates at which banks offer to lend unsecured funds to other banks in the Shanghai wholesale money market. They are calculated from rates quoted by the 18 banks with the highest credit ratings. 4. The Shanghai Clearing House (an interbank clearing house set up for the settlement and clearing of financial products does not report the amount of NCD issuance in the period December 2013 to February Wind Info. Wind Info ( is the mostly widely used Chinese economic and financial data and information provider. It serves more than 90% of the financial firms in the Chinese market, and 75% of the Qualified Foreign Institutional Investors in China. 6. Wind Info, European Central Bank, Federal Reserve System, China Banking Regulatory Commission, and Bank of Japan. 7. Wind Info, China Banking Regulatory Commission. 8. Wind Info, as of the second quarter of Asset Management Association of China. (in Chinese) (accessed 30 August 2017). 10. They are the Agricultural Bank of China, Bank of China, Beijing Bank, Bank of Communications, China Construction Bank, CITIC Bank, China Merchant Bank, Overbright Bank, Guiyang Bank, Huaxia Bank, Industrial and Commercial Bank of China, Industrial Bank, Jiangyin Bank, Minsheng Bank, Nanjing Bank, Ningbo Bank, Pudong Development Bank, Pingan Bank and Wuxi Bank. 11. Wind Info. 12. Wind Info. 13. Wealth management products in China are investments that offer fixed rates of return well above regulated interest rates for deposits, and are often used to fund investments in sectors where bank credit is restricted. They are typically actively managed by banks, with other firms commonly used as channels, but few are recorded on banks balance sheets. See Perry and Weltewitz (2015) store/761d be2a5eaaf387eaefb56/ (in Chinese) (accessed 30 August 2017). References Acharya, V., J. Qian and Z. Yang (2016) In the Shadow of Banks: Wealth Management Products and Issuing Banks Risk in China _all.pdf (accessed 9 December 2017). Board of Governors of the Federal Reserve System (2011) Press Release: Federal Reserve issues final rule to repeal Regulation Q, which prohibited the payment of interest on demand deposits. 14 July. gov/newsevents/pressreleases/bcreg a.htm (accessed 12 December 2017).

10 ECONOMIC AFFAIRS, VOLUME 38, NUMBER Carlson, M. and D. C. Wheelock (2015) The Lender of Last Resort: Lessons from the Fed s First 100 Years, ino. Humpage (ed.), Current Federal Reserve Policy under the Lens of Economic History. Cambridge: Cambridge University Press Carlson, M. and D. C. Wheelock (2016) Near-Money Premiums, Monetary Policy, and the Integration of Money Markets: Lessons from Deregulation, Journal of Financial Intermediation. In press. jfi Chen, K., J. Ren and T. Zha (2016) What We Learn from China s Rising Shadow Banking: Exploring the Nexus of Monetary Tightening and Banks Role in Entrusted Lending. NBER Working Paper No Cambridge, MA: National Bureau of Economic Research. Chow, G. C. (1960) Tests of Equality Between Sets of Coefficients in Two Linear Regressions, Econometrica 28(3), Cook, T. (1978) Regulation Q and the Behavior of Savings and Small Time Deposits at Commercial Banks and Thrift Institutions, Economic Review of Federal Reserve Bank of Richmond, November/December, Huang, Q., J. D. Haan and B. Scholtens (2017) Analysing Systemic Risk in the Chinese Banking System, Pacific Economic Review. International Monetary Fund (2010) Global Financial Stability Report: Sovereigns, Funding, and Systematic Liquidity. Washington, DC: International Monetary Fund. Liang, H. and X. Yu (2017) China s Monetary Policy Framework: Toward a New Normal. China International Capital Corporation D8EB5B53855CE54F0FAD85015AE4C5EA.pdf (accessed 9 December 2017). Liu, K. (2017) China s Interest Rate Pass-through to Commercial Banks Before and After Interest Rate Liberalisation, Economic Affairs 37(2), McLoughlin, K. and J. Meredith (2017) The Rise of Chinese Money Market Funds, Reserve Bank of Australia Bulletin, MarchQuarter, Perry, E. and F. Weltewitz (2015) Wealth Management Products in China, Reserve Bank of Australia Bulletin, June Quarter, Ruebling, C. (1970) The Administration of Regulation Q, Federal Reserve Bank of St. Louis Review, February, Winningham, S. (1979) The Effect of Money Market Certificates of Deposit on the Monetary Aggregates and their Components, Federal Reserve Bank of Kansas City Economic Review 64(7),