China: Reserve Requirements, 2015–2016

In its second-quarter 2006 Monetary Policy Report, the PBOC explained that it initially used reserve requirements to ensure commercial banks had sufficient liquidity to cover unexpected payments, but that over time it had become a tool for the PBOC to manage banking sector liquidity (PBOC 2006). The PBOC began using required reserves as a policy tool regularly in 2007 (Ma, Xiandong, and Xi 2013).

The PBOC’s crisis-relief RRR cuts were preceded by three cuts in February, April, and June for credit allocation. With the equity market disruptions in August, however, the PBOC cut the RRR by 50 bps for all financial institutions, as it sought to maintain liquidity in the banking system amid the large equity markets disruption (PBOC 2015d). In October, the PBOC cut the RRR once more for liquidity management, providing a 50-bps cut for all financial institutions and a further 50-bps cut for firms with significant agricultural lending (PBOC 2015e). In each case, the PBOC also stated an objective to continue credit flow to preferred sectors.

The PBOC adjusted the benchmark interest rates at the same time as its June, August, and October RRR cuts (PBOC 2015d; PBOC 2015e; PBOC 2015m).  

Besides the RRR, the PBOC managed liquidity through open market operations, including transactions with government deposits, repurchase agreements (repos), and the issuance of new, longer-dated central bank bills (CBBs),^FThe PBOC initially used treasury bonds or securities as sterilization tools, but, in September 2002, it replaced all outstanding securities with CBBs for use in open market operations (Ouyang, Rajan, and Willett 2010). a typically short-term debt instrument designed to absorb liquidity (Wang and Hu 2011).

In 2015, the PBOC complemented, or structured, its open market operations with several new lending facilities, catering to different market participants focused on both real economy lending and banking sector liquidity:

• The medium-term lending facility (MLF), introduced in 2014, supplied banks with three- to 12-month funding, conditional on their use of the funds to channel credit to the PBOC’s preferred sectors, including, among others, small, private enterprises. Participants borrowed at a rate slightly above the PBOC’s repo rate (Jones and Bowman 2019).^FIn 2018, the PBOC introduced a variation of the MLF with a lower interest rate and a longer duration of up to three years, called the targeted medium term-lending facility (People’s Bank of China 2018; Jones and Bowman 2019; Yiping Huang, Ge, and Wang 2020, 50). In 2015, the PBOC conducted a total of 219.48 billion yuan in MLF operations—including 101.45 billion yuan, 51.45 billion yuan, 36 billion yuan, and 30.58 billion yuan in each quarter—with an outstanding balance of 66.58 billion yuan by the end of 2015. In the first half of 2015, the maturity on all MLF operations was three months, but in the second half of the year, the PBOC extended the maturity to six months and cut the interest rate charged by 15 bps in response to market demands for liquidity (PBOC 2015m).
• The pledged supplementary lending facility (PSL), also introduced in 2014, provided three- to five-year funding to policy banks—initially, the China Development Bank and, in 2015, the Agricultural Development Bank of China and the China Export-Import Bank—in direct support of the central government's housing redevelopment projects and other initiatives (Jones and Bowman 2019). In 2015, the PBOC provided a total of 698.1 billion yuan to these three policy banks through the PSL, with an outstanding balance of 1.08 trillion yuan by the end of the year.
• The standing-lending facility (SLF), introduced in 2013, helped market participants meet unexpected short-term liquidity shortfalls, but its use was minimal and often stigmatized (Jones and Bowman 2019). In 2015, the PBOC conducted a total of 334.84 billion yuan in SLF operations, nearly all of which took place in the first quarter of the year. By the end of 2015, the SLF had an outstanding balance of 40 million yuan (PBOC 2015m).

Also in 2013, the PBOC established short-term liquidity operations (SLOs). This monetary policy framework effectively established a corridor to the overnight bank rate, with the seven-day interest rate at the SLF providing the ceiling, and the remuneration rate on excess reserves providing the floor (see Figure 1) (Huang, Ge, and Wang 2020; Ma, Xiandong, and Xi 2013; McMahon et al. 2018).

Figure 1: The PBOC’s Lending Facilities (LHS) and Their Influence on Interest Rates (RHS)

Source: Jones and Bowman 2019.

The Law of the People’s Republic of China on the PBOC (PBOC’s enabling law) permits the central bank to require financial institutions to place deposits with the central bank at a prescribed ratio (Tenth NPC 2003b, Ch. 4, Article 23(1)).

Article 32 of the PBOC law empowers the PBOC to govern the rules and regulations on reserves (Tenth NPC 2003b Article 32(1)). The PBOC needed State Council approval before adjusting bank deposit and lending rates, but the PBOC may have had greater control over changes in the RRR (Fungáčová, Nuutilaine, and Weill 2015; Ma, Xiandong, and Xi 2013).

The PBOC may redefine the reservable deposit base to include different types of deposits in order to adjust the effective reserve requirements without altering headline RRR (Ma, Xiandong, and Xi 2013).

Commercial banks are bound by the Chinese Commercial Bank Law, which stipulates that banks must place deposits with the PBOC and maintain sufficient provision for payment in accordance with the PBOC’s regulations^FChapter VIII, Article 77(3) of the Commercial Bank Law identifies the penalties for noncompliant commercial banks. (Tenth NPC 2003a Ch. III, Article 32).

The PBOC Law states that the goal of China’s monetary policy is to maintain currency stability and facilitate economic growth (Gang 2019; Tenth NPC 2003b, Ch. 1, Article 3). In practice, this meant that PBOC policy was also directed toward other policy goals, such as providing affordable financing to economic sectors or geographical areas (Fungáčová, Nuutilaine, and Weill 2015).

The PBOC administered all changes to the RRR.

Article 7 of the PBOC Law states that the PBOC forms monetary policy under the leadership of the State Council, the highest government body in China,^FHowever, the highest decision-making body in the country is the Standing Committee of the Politburo (Elliott and Yan 2013). See Figure 6 for a hierarchy of the Chinese financial system. and independent from the influence of other government offices or individuals (Tenth NPC 2003b, 2). The State Council determined the functions, responsibilities, composition, and procedures of the PBOC’s Monetary Policy Committee (Tenth NPC 2003b, Ch. 2, Article 12).

According to Article 5 of the PBOC Law, the PBOC submitted all decisions related to China’s money supply, interest rates, and exchange rates, among other issues, to the State Council (Tenth NPC 2003b). In practice, many regulatory and financial sector agencies were involved in setting the exchange rate and interest rate limits, but the State Council had the ultimate decision (Elliott and Yan 2013; Wang and Hu 2011).

The PBOC Law required the central bank to submit work reports to the Standing Committee of the National People’s Congress (NPC) on the conduct of monetary policy and the operations of the financial sector and to record its monetary policy decisions and procedures for the State Council (Tenth NPC 2003b, Ch. 1, Articles 5-6; Ch. 2, Article 12).

The PBOC announced RRR adjustments in press releases, and they typically became effective one week after announcement. In its 2009 Annual Report, the PBOC cited a requirement by the State Council to increase publicity and education around its crisis response and measures to grow the Chinese economy (PBOC 2010).

In its 2015 Annual Report, the PBOC characterized all of its RRR cuts that year as a response to downward economic pressures and as a means “to compensate for the long-term liquidity gap,” alongside open market operations (including reverse repos and SLOs) and lending through the MLF and PSL (PBOC 2015i).

However, the PBOC described its first three RRR cuts of 2015 differently when it announced them. In July, the PBOC characterized its “targeted” February, April, and June cuts as “strengthening the capacity of financial institutions to support the developments of agricultural sector, rural areas and farmers and micro and small enterprises, . . . supporting the key areas and weak sectors in national economy, and providing financial support to popular entrepreneurship and innovation” (PBOC 2015f).

In August and October, the PBOC noted both the credit allocation goal and the liquidity goal in announcing RRR cuts (PBOC 2015d; PBOC 2015e).

In its Monetary Policy Report for the fourth quarter of 2015, the PBOC noted that cutting the RRR sent a strong easing signal, and that such a policy adjustment “may enhance easing expectations, which then will lead to heightened depreciation pressures on the local currency, accelerated capital outflows, and declines in the foreign exchange reserves” (PBOC 2015m, 57).

The PBOC required banks to hold all required reserves as deposits at the PBOC; they could use those reserves for intraday settlements (Ma, Xiandong, and Xi 2013).

Since August 2007, the PBOC required commercial banks to deposit reserves in the form of foreign exchange, which were considered external assets on the PBOC’s balance sheet because they were managed by the State Administration of Foreign Exchange (SAFE), which is part of, but distinct from, the PBOC (Setser and Pandey n.d.).

Shortly after its introduction in 1984, the PBOC’s RRR regime differentiated by deposit type, including corporate deposits, savings deposits, or demand deposits. During the GFC, the PBOC’s reserve requirement covered most deposits, but the PBOC did not specify these deposit types and likely differentiated the RRR based on financial institution type only (Ma, Xiandong, and Xi 2013).^FMa, Xiandong, and Xi (2013) note that the reservable deposit base was expanded in September 2011 to include some margin deposits; this expansion effectively hiked the RRR by 125 bps.

In January 2015, the PBOC redefined the coverage for reserve requirements to include deposits for securities transactions and settlements, deposits by banks for non-financing purposes, deposits by SPVs, deposits by other financial institutions, and deposits by overseas financial institutions.^FPreviously, these deposits had been recorded under the interbank account item in the PBOC’s statistical coverage. However, the PBOC applied a reserve requirement of zero to these deposit types (PBOC 2014f). The PBOC’s public statements in 2015 did not address whether the RRR cuts applied to these deposit types.

Since the PBOC began adjusting the RRR more regularly in 2007, the PBOC used a “time-point” assessment. Chinese banks generally relied on deposit funding. Thus, given the low remuneration rate on required reserves, a high RRR imposed a large opportunity cost^FBecause a policy-rate hike, intended to be contractionary, might have attracted foreign capital inflows and offset the monetary tightening, an increase in unremunerated (or partially remunerated) reserves effectively “taxed” banks by raising the lender’s cost of capital and the cost of credit in the economy, and may have widened the loan-deposit rate spread (within the state-mandated deposit rate cap) (Ma, Xiandong, and Xi 2013). on banks. Under the time-point framework, banks had to meet the RRR on a daily basis (based on a 10-day rolling average of the stock of deposits) and therefore tended to hold excess reserves as a buffer, averaging 1.5% of deposits in 2011 (Ma, Xiandong, and Xi 2013; Turner, Tan, and Sadeghian 2012).

In September 2015, the PBOC revised the RRR assessment from a time-point framework to an averaging method. Under this method, the PBOC applied the RRR to the average end-of-day reserve balances between the fifth day of a 10-day period and the fourth day of the next 10-day period, divided by the deposit balance at the end of the previous 10-day period (PBOC 2015l).

The PBOC also said that the reserves on any given day could be up to 100 bps lower than the RRR. The PBOC said this feature would provide “an important buffer mechanism. During the assessment period, the reserves of a financial institution may be less than the requirement at some points, as long as the overall average reaches the reserve requirement.” The PBOC noted that the level of bank reserves and money market rates were “stable with a slight decrease” throughout the transition to the averaging method. It said that a small number of banks had slightly overdrawn on their reserves as “an effective response to meet short-term fluctuations. Such behavior has promoted market stability at the macro level” (PBOC 2015l, 11–12).

In 2011, the PBOC formally introduced dynamically differentiated reserve requirements. Once the requirements were implemented, the PBOC frequently recalibrated the “parameters” of the adjustment mechanism to take new factors into account, and generally computed a firm’s applicable RRR based on its capital adequacy, presence in agricultural lending and rural provinces, and internal and risk controls. In 2015, the PBOC considered five factors: 1) the recipients of “agro-supporting” and small-business loans, 2) the capital adequacy ratio, 3) internal and risk controls, 4) the launch of new branch offices, and 5) regional development (PBOC 2015l).

Required reserves generally applied to all Chinese-domiciled financial institutions. The PBOC Law defines “financial institutions” as Chinese-domiciled policy banks, commercial banks, urban credit cooperative, rural credit cooperatives, and other deposit-taking institutions (Tenth NPC 2003b, Ch. 8, Article 52). Four of the five RRR cuts that the PBOC made in 2015 were targeted at specific institutions, largely involved in lending to rural areas and the agricultural sector, or for the purposes of financing leases or car purchases. Institutions had to meet certain criteria on solvency, which the PBOC did not make public; the PBOC noted that the four targeted RRR cuts applied to more than 98% of all financial institutions (PBOC 2015m).

On February 5, the PBOC’s Monetary Policy Committee met and, under the orders of the central government, cut benchmark interest rates and the RRR for small and medium-size institutions by 50 bps (PBOC 2015a).

The MPC met four more times in 2015, and cut the RRR for institutions on each occasion: April 20, June 28, August 25, and October 24 (PBOC 2015b; PBOC 2015c; PBOC 2015d; PBOC 2015e).

RRR adjustments took effect roughly one week after they were announced. The PBOC said this gave financial institutions a sufficient amount of time to adjust their reserves (PBOC 2006).

The PBOC held the RRR steady through 2013 and 2014, with a headline RRR for large institutions of 20%. In 2015, the PBOC adjusted the RRR on five occasions, including four “universal cuts,” applying to all depository institutions (for a cumulative 250 bps), and five “targeted” cuts, which reduced the RRR another 50-650 bps from the headline level (PBOC 2015m). The PBOC maintained these RRR levels through 2016.

In February 2015, the PBOC announced two differentiated cuts to the headline RRR. First, the PBOC announced a 50-bps cut for all financial institutions to 19.5%, without stating a purpose. Second, it announced an additional 50 bps cut for city commercial banks and noncounty-level rural commercial banks, to 19.5%, in order to “support structural adjustments” and to increase lending to small and micro enterprises, the agricultural sector, rural areas, and major water conservancy projects. The PBOC also cut the RRR for the Agricultural Development Bank of China by an additional 400 bps (PBOC 2015a).

In April 2015, the PBOC cut the RRR for the same reasons and further differentiated by institution type. The PBOC cut the headline for all institutions by 100 bps. Based on the results of “dynamic evaluations of financial institutions,” the PBOC offered an additional 100-bps cut for rural institutions, banks, credit cooperatives, financial leasing companies, and auto financing companies, to 11.5%; an additional 50-bps cut for state-owned enterprises with significant lending to the agricultural sector, to around 18.5%; and an additional 200 bps for the Agricultural Development Bank of China, to around 10.5% (PBOC 2015b; PBOC 2015k). The preferential cuts to the Agricultural Development Bank of China, on top of the headline cuts totaling 150 bps, brought its applicable RRR down from 18% in early 2015 to 10.5% in April (PBOC 2015k).

After the April 2015 cuts, the level of reserves in the banking system reached a record high, and the PBOC forecasted excess reserves of USD 475 billion by June. At the time, the PBOC said liquidity in the banking system remained adequate, and that it was unnecessary to provide extra liquidity by cutting the RRR for all financial institutions. It said its June 2015 RRR cuts were targeted toward support for “the development of the real economy” and “to promote structural adjustment” (PBOC 2015f).

By August, however, the PBOC announced another 50-bps RRR cut for all financial institutions following the large equity markets disruption (see Figures 3 and 4). The PBOC also announced a further 50-bps reduction for rural institutions, banks, and credit cooperatives, to 10.5%; and an additional 300-bps cut for financial-leasing and auto-financing companies, to around 13.5% (PBOC 2015d). In October, the PBOC cut the RRR once more for the purpose of liquidity management, implementing a 50-bps cut for all financial institutions and a further 50-bps cut for firms with significant agricultural lending (PBOC 2015e).

The PBOC did not attach additional conditions or restrictions to its RRR adjustments.

China’s monetary policy consisted of quantity-based tools—including required reserves, as well as “window guidance,” credit quotas, and other administrative measures—and price-based tools, such as benchmark rates (and other policy interest rates) and open market operations (Fungáčová, Nuutilaine, and Weill 2015). PBOC Governor Yi Gang stated that the central bank managed liquidity with open market operations and required reserves and adjusted the money supply with central bank lending (Gang 2019). Given China’s exchange rate regime and interest rate policy, the PBOC’s monetary-policy framework relied more on RRR adjustments than open market operations (Ma, Xiandong, and Xi 2013). Often, the PBOC coordinated benchmark interest rate cuts and decreased CBB issuance with RRR cuts (PBOC 2008).

Historically, deposit rates operated in a corridor with the baseline lending rate acting as the floor and the baseline deposit rate constituting a ceiling (Prasad and Zhang 2014, 195). Since 2015, the PBOC targeted the seven-day repo rate (“DR007”), between an interest rate corridor with a ceiling defined by the SLF (or MLF) interest rate and the floor by the remuneration rate on excess reserves (see Figure 4) (Huang, Ge, and Wang 2020). Throughout 2015, the PBOC continued efforts to formally liberalize interest rates, a process started in 2003 (see Figure 8 in Appendix A) (McMahon et al. 2018; Prasad and Zhang 2014).^FIn May 2015, the PBOC expanded the upper limit of the floating range for the deposit rate from 1.3 to 1.5 times the benchmark rate. In August 2015, the upper limit of deposit rates was fully liberalized for fixed terms longer than one year. In October, the PBOC eliminated the upper limit of the deposit interest rate for commercial banks and rural credit cooperatives and began focusing on short-term interest rates, such as the Shanghai Interbank Offered Rate (SHIBOR) (McMahon et al. 2018; Yiping Huang, Ge, and Wang 2020).

The PBOC eased the RRR beginning on February 5, 2015, and made its last cut on October 24, 2015. The PBOC did not preannounce an end date to the easing, nor state publicly the criteria under which it would discontinue its easing.