Chapter

Chapter 14 Monetary Policy Communication: Frameworks and Market Impact

Editor(s):
Alfred Schipke, Markus Rodlauer, and Longmei Zhang
Published Date:
March 2019
Share
  • ShareShare
Show Summary Details
Author(s)
Michael McMahon, Alfred Schipke and LI Xiang 

The authors thank participants at the 2018 joint People’s Bank of China–IMF conference on China’s Growing Bond Market in a Global Context for helpful comments and feedback. The authors also thank Anil Ari, Tomas Harjes, Ma Jun, Nan Li, Stephen Mulema, Rafael Portillo, Filiz Unsal, Ping Wang, Ting Yan, and People’s Bank of China staff for very helpful comments and suggestions.

Bond markets are an important conduit of monetary policy signals to the economy. Reforms that improve the functioning of bond markets will hence facilitate macroeconomic management effectiveness. Here communication plays an increasingly important role. Good monetary policy communication is not only important to improve the effectiveness of monetary policy in the first place, but by reducing uncertainty it makes bond markets more attractive for investors, further improving monetary transmission.

Central banks increasingly use communication as a key lever of monetary policy. Gone are the days of “never explain, never excuse,” as put by the Governor of the Bank of England, Montagu Norman (1920–44).1 Today, central banks, especially in countries with developed financial systems, communicate regularly. This includes low-frequency communication about their policy frameworks and objectives, as well as high-frequency communication of views on current macro-economic conditions, forecasts related to output and inflation, and the rationale for policy decisions.2 Since the global financial crisis, the implementation of unconventional monetary policies and greater focus on financial sector stability have made communication even more important, especially as it relates to forward guidance.

Compared with other public institutions, central banks are often at the forefront of communication. The People’s Bank of China (PBC) is no different. Over the past few years, it has strengthened its communication and is keenly aware of the heightened global interest in information about the country’s monetary policy. The increase in the number of press conferences and speeches has coincided with sharp stock market adjustments, changes to the exchange rate framework, interest rate liberalization,3 and greater financial market volatility in 2015–16. However, communication has not yet become as potent a policy tool as in many advanced economies (such as at the US Federal Reserve or the Bank of England) and some important emerging markets.

In many respects, China is at a communications crossroads, driven both by domestic and, increasingly, external factors. Financial and external sector liberalization and greater reliance on price- and interest rate–based allocation of resources likewise make communication more important. Given the country’s global footprint and increasing financial sector links, including through different bond and equity connect schemes, this is also true for the international community. As the country moves toward a more flexible exchange rate in conjunction with a monetary policy framework that relies more on short-term interest rates, better communication will significantly improve the effectiveness of monetary policy, reduce excessive volatility, and foster financial sector stability.

This chapter discusses China’s unique institutional setup and provides background information about the PBC’s main communication channels. The chapter reviews the relevant literature and empirically analyzes the effect of the PBC’s communication on financial markets. It concludes with recommended policy actions to increase monetary policy effectiveness and to reduce volatility.

China’s Unique Institutional Setup

Compared with other central banks in advanced economies or emerging markets with modern policy frameworks, China’s central bank communication is more constrained. This is due to its unique institutional arrangements. In particular, China has multiple and overlapping objectives across institutions. The key constraints are as follows:

First, the PBC does not have full decision-making power over money supply targets and interest rate policy, and it has only limited operational independence at the monetary policy instrument level. Thus, the central bank is constrained in the information and forward guidance that it can convey. At the same time, the PBC drafts and executes monetary policy and has some operational independence, such as in setting short-term interest rates through open market operations, short-term liquidity operations, or rates on standing and medium-term lending facilities. However, key decisions need to be approved by the State Council (China’s equivalent to a government cabinet).4 Approval is needed, for example, for changes in the benchmark interest rate, reserve requirements, and the setting of annual monetary aggregates.5 For these key policy instruments, the PBC usually proposes policy plans when key economic indicators such as GDP growth and inflation deviate from the targets set at the beginning of each year by members of the State Council, who then review the plans and make the final decision (HUANG, GE, and WANG 2018). These members also regularly approve the wording of the monetary policy stance (Box 14.1 explains terminology).

Second, monetary policy decisions are the result of consensus-building among various stakeholders. State Council members have a broad range of economic and financial sector objectives, and requests for monetary policy changes can be submitted to the State Council not only by the PBC but also by other ministries or agencies. From the central bank’s perspective, therefore, both the outcome and the timing of important monetary policy decisions are uncertain, limiting the PBC from providing forward guidance. As the implementing agency, the PBC publishes all monetary policy decisions. Moreover, important PBC personnel and budget decisions go beyond the PBC’s remit. The PBC governor is nominated by the premier of the State Council, approved by the National People’s Congress (or its standing committee when the congress is not in session), and finally appointed and dismissed by the country’s president.6 The PBC budget, in turn, is part of the central government’s budget and supervised by the financial department of the State Council.

Third, China’s monetary policy has numerous objectives. According to the Law of the People’s Republic of China on the PBC, “the objective of the monetary policy shall be to maintain stability of the value of the currency and thereby promote economic growth.” The PBC is not an inflation-targeting central bank and its objectives go beyond price stability. As the previous PBC governor stated, “the annual objectives of the PBC mandated by the Chinese government have been maintaining price stability, boosting economic growth, promoting employment, and broadly maintaining balance of payments” (ZHOU 2016a). In addition, the 2017 PBC Work Conference called for monetary policy to balance among economic growth, economic reform, economic structure, household welfare, and financial stability considerations. These multiple objectives often involve trade-offs and reduce transparency.7

As China’s monetary policy framework is increasingly moving from using quantitative targets to a more price-based framework, effective central bank communication is becoming even more important. As summarized in HUANG, GE and WANG (2018), the PBC’s intermediate targets are threefold: quantity-based money supply and bank credit, as well as priced-based market interest rates. Successive waves of interest rate liberalization, which started with money market rates and culminated with the formal elimination of the ceiling on bank deposit rates in 2015, have facilitated the transition toward a modern price-based monetary policy framework. While the process is not complete, the government in 2018 reiterated its commitment to deepen reforms to make both interest rates and the exchange rate more market based (LI 2018).

The emphasis on quantitative targets has declined. Since 1994, China has had quantitative monetary targets. However, reflecting financial innovation and a rapidly changing financial system structure, its intermediate quantitative monetary target, M2, is correlated less and less with inflation and growth (MA 2017). In addition, the M2 outturn has deviated from the target over the past couple of years (Figure 14.1). As a further indication that emphasis on quantitative targets has declined, the 2018 Report on the Work of the Government (Annual Work Report) did not specify a specific target or projection for the monetary aggregate (M2) or for credit aggregates (such as total social financing). Compared to previous Annual Work Reports, this was an important step forward. Instead, the language was more vague: “Our prudent monetary policy will remain neutral, with easing or tightening as appropriate. We need to make sure that the value of M2 money supply, credit, and aggregate financing ensure a reasonable, stable level of liquidity” (LI 2018).

Figure 14.1.Monetary Aggregate M2 Growth, 2010–17

(Percent)

Source: Report on the Work of the Government 2010 through 2017; and CEIC.

Note: M2 = intermediate quantitative monetary target.

Interest rate liberalization and other reforms have allowed the PBC to improve its policy framework. Focus is increasing on short-term money market rates, that is, the 7-day interbank pledged repo rate (DR007).8 In the 2016, third-quarter Monetary Policy Executive Report (MPER), the PBC stated that “DR007 moves around the open market operation 7-day reverse repo rate. The DR007 can better reflect the liquidity condition in the banking system and has an active role to cultivate the market base rate.” Although the PBC has not confirmed DR007 as the policy interest rate yet, it is closely watched by the market. The central bank uses open market operations (OMOs) and the corresponding 7-day OMO repo/ reverse repo rate to signal policy changes. For example, from the end of 2015 until mid-2018, the PBC changed OMO rates on seven occasions, of which three coincided with increases in the US federal funds rate. Combined with interest rates for standing lending facilities and remunerated required or excess reserves (see Table 14.1), the monetary policy framework effectively provides a corridor, that is, an upper and lower bound (Figures 14.2 and 14.3).

Table 14.1People's Bank of China Policy and Asset Rates
Policy ToolsMeaningTenor
Benchmark interest ratePeople's Bank of China sets the benchmark interest rate as reference rates for deposits and lending. It is the official policy rate. It has not changed since October 2015.Less than 6 months, 6 months to 1 year, 1–3 years, 3–5 years, more than 5 years
OMO repo and reverse repoShort-term collateralized loans and borrowing, with a direct effect on interbank liquidity conditions. At present, the 7-day reverse repo is used more frequently in practice, combined with the occasional use of the 14-, 28-, and 63-day reverse repo. Since 2013, the 91- and 182-day repo have not been used; the 21-day repo is rarely used.7-, 14-, 21-, 28-, 91-, 182-day
OMO People's Bank of China bill yieldMoney supply changes by issuing central bank bills. Not in use since November 2013.3-month, 6-month, 1-year, 3-year
Medium-term lending facilityAn instrument to provide medium-term base money to commercial or policy banks that meet the requirements of macroprudential management. Aims to adjust the medium-term funding cost of financial institutions and then the funding cost of the real economy. Created in September 2014.3-month, 6-month, 1-year
Standing lending facilityA liquidity supply channel of the People's Bank of China to meet the temporary liquidity demand of commercial banks and rural cooperative financial institutions. Seen as the upper bound of the interest corridor. Created in early 2013.Overnight, 7-day, 1-month,
Pledged supplemental lendingLong-term and large amounts of financing to support the key areas and weak links of the national economy. Created in April 2014.Long term
Targeted medium-term lending facilityStable source of funding for financial institutions based on growth of their loans for small and private firms.Long term
Short-term liquidity operationsUsed to stabilize market expectations and prevent financial risk when there is a temporary fluctuation in the liquidity of the banking system. Created in January 2013.Very short term
Required reserve ratioShare of deposits that banks are required to hold at the People's Bank of China. Required reserve ratio changes can have long-term and large liquidity implications.
Excessive reserve rateRate on excess reserves. Seen as the lower bound of the interest rate corridor.
Money Market Rate
ShiborShanghai interbank offered rate. The reference rate based on the interest rates at which banks offer to lend unsecured funds to other banks in the Shanghai wholesale money market.Overnight, 1-week, 2-week, 1-month, 3-month, 6-month, 9-month, 1-year
R007The weighted average 7-day repurchase (repo) rate for all financial institutions including all pledged repurchase transactions in the interbank market.7-day
DR007The weighted average 7-day repurchase rate in which a deposit institution uses interest rate bonds as a pledge in the interbank market.7-day
FR007Interbank fixing 7-day repo rate, a benchmark rate based on repo trading rate for the interbank market.7-day
CD rateRate on certificates of deposit issued by depository financial institutions in the interbank market.
Bond Market
Short-term commercial paperUnsecured, short-term debt instrument issued by nonfinancial corporations in the interbank bond market, typically for the financing of accounts receivable and inventories and meeting short-term liabilities.1-month, 3-month, 6-month, 1-year
Medium-term notesDebt financing instruments issued by nonfinancial corporations in the interbank bond market.1-month, 3-month, 6-month, 1-year, 5-year
Treasury bondsBonds issued by the government to raise fiscal funds.3-month, 6-month, 1-year, 3-year, 5-year, 7-year, 10-year, 30-year
Equity Market Rate
Shanghai Stock Exchange composite index returnThe daily return on the exchange's composite index, a stock market index of all stocks that are traded at the Shanghai Stock Exchange.
Source: People's Bank of China.Note: OMO = open market operations.
Source: People's Bank of China.Note: OMO = open market operations.

Figure 14.2.Interest Rate Corridor, 2014–18

(Percent)

Sources: CEIC; and WIND Economic Database (www.wind.com.cn).

Note: DR007 = 7-day (7d) interbank pledged repo rate; Shibor (1w) = 1-week Shanghai interbank offered rate; SLF (7d) = 7-day short-term liquidity facility.

Figure 14.3.PBC Policy and Target Rate, 2014–18

(Percent)

Sources: CEIC; and WIND Economic Database (www.wind.com.cn).

Note: Vertical black bars show recent changes in the US federal funds rate; those in red reflect changes in the People’s Bank of China’s open market operation rates. Benchmark lending rate is six months to one year (6m–1y). bp = basis points; DR007 = 7-day (7d) interbank pledged repo rate; OMO = open market operations; Shibor (1w) = 1-week Shanghai interbank offered rate.

China’s monetary policy framework remains in transition and is currently a hybrid (see MA and GUAN 2018). On the one hand, despite formal interest rate liberalization, banks are still guided by corresponding deposit and lending benchmark rates; changes to these rates go beyond the authority of the PBC and require State Council consent. On the other hand, other policy rates, such as medium-term lending and the pledged supplementary facilities (see Table 14.1), as well as the use of instruments such as changes in required reserves, undermine transparency and complicate communication (Figures 14.4 and 14.5 and Table 14.1).

Figure 14.4.People’s Bank of China Hybrid Monetary Policy Framework, 2014–18

(Percent; right scale, billions of renminbi)

Source: CEIC.

Note: MLF = medium-term lending facility; PBC = People's Bank of China; PSL = pledged supplementary lending; RRR = required reserve ratio; SLF = short-term lending facility.

Figure 14.5.Widening of Interbank-Conventional Deposit Rate, 2014–18

Sources: CEIC; and WIND Economic Database (www.wind.com.cn).

Note: DR007 = 7-day interbank pledged repo rate.

In April 2018, the PBC indicated its intention to move toward a unified interest rate framework. “In some aspects, the interest rates are ‘running on two tracks,’ that is, there are benchmark interest rates for deposits and loans, while the money market rates are fully liberalized. We have eased restrictions on deposit and loan rates, and commercial banks have been offering deposit and loan rates higher or lower than the benchmark interest rate based on their operating conditions. As a matter of fact, the best tactic is for us to gradually unify the two tracks of interest rates, and we are doing just that in the market-based reforms” (PBC 2018a, 2018b).

Evolving Communications Channels

PBC communication, though still evolving, takes place primarily through four main channels (Box 14.1 reviews key terms to describe China’s monetary policy and use of window guidance):

  • Monetary Policy Executive Reports (MPERs): First issued in 2001, the PBC now releases its MPER quarterly. It covers most recent monetary policy decisions; an analysis of output, prices, and money supply developments; information about important sectors; and an assessment of prospects for China’s macroeconomic development. Generally, report coverage is more backward looking, but increasingly it provides useful technical and operational information—often within boxes—and, even more recently, some forward-looking information.

  • Press releases on monetary policy committee meetings: The first monetary policy committee meeting was in 1997 and the meeting now takes place at the end of each quarter. Since 2009, the PBC has published press releases one or two days following the meeting. The actual meeting dates, however, are not announced ahead of time and are only posted on the PBC’s website following the meeting.

  • Speeches and press conferences: Oral communication comprises public speeches by the governor and deputy governors, as well as press conferences.9 Public officials regularly give speeches at public conferences or international central bank governor summits. While the contents center around the topics of the conference and cover the PBC’s policy stance, sometimes speakers talk about future developments in a specific area. Press conferences are usually scheduled after policy announcements to further explain the rationale for decisions.

  • Open market operation notices: To better explain the rationale of open market operations, daily notices have been standard since January 2016.10 Increasingly, these notices provide contextual information by adding phrases such as “given ample liquidity, the PBC has intervened to keep liquidity stable.”11

The PBC is committed to providing more timely information. In 2018, the PBC reiterated that it is committed to “continuously improve the central bank’s credibility and transparency” and would “strengthen its policy interpretation and information disclosure, deliver its policy intentions in a timely way and reasonably guide market expectations” (PBC 2018a). As part of the response to the increasing demand for timely information, the PBC now also communicates regularly through social media, such as Weibo (China’s Twitter equivalent). As Figure 14.6 shows, within only a few years the PBC has already reached a very large following on social media in both absolute and per capita numbers, compared with major central banks.

Figure 14.6.Central Bank Social Media Presence, 2018

(Thousands, left scale; percent, right scale)

Source: Weibo account of the People’s Bank of China; Twitter accounts of the US Federal Reserve, the European Central Bank, the Bank of England, the Bank of Canada, the Bank of Japan, and the Royal Bank of New Zealand; and World Development Indicators.

Literature Review

An increasing body of literature is studying the effect of central bank communication. Existing research focuses on central banks in developed economies, mostly the US Federal Reserve Bank, the European Central Bank, and the Bank of England. Generally speaking, these analyses examine the effects of monetary policy announcements, minutes of monetary policy committee meetings, regularly published reports, interviews and speeches, and congressional testimony or parliamentary committee hearings. An early survey of the literature is Blinder and others (2008). This chapter follows much of this literature and relies on market-based event studies.12 Specifically, the chapter investigates whether PBC communication affects asset prices in China’s financial market in a manner indicating that the communication contains relevant information.

Some analysis codes the communication contents by subjectively assigning values to the texts that are perceived as dovish or hawkish (Jansen and De Haan 2005; Ehrmann and Fratzscher 2007; Berger, Nitsch, and Lybek 2006), or using semantic orientation and text classification to quantify dimensions of the text (Lucca and Trebbi 2009; Hansen and McMahon 2016; Hansen, McMahon, and Prat 2018). These techniques are useful in identifying the direction of communication and can augment the type of analysis undertaken in this chapter. However, two obstacles hamper applying these methods to the Chinese language. First, dictionary methods do not work well for China given the different language structure. Second, communication from the PBC, especially in public speeches, is noncommittal and tends to be carefully drafted. Garcia-Herrero and Girardin (2013) compute subjective “hawkish” and “dovish” scores for previous PBC Governor Xiaochuan Zhou and monetary policy committee member Gang Fan. But such a manual coding approach can only be applied to small amounts of communication, and restricting the analysis coverage may come with the risk of biased results. Hence, this analysis focuses on the fact that PBC officials come out to speak publicly more than on the content of their speech. Further text analysis is an important direction for future studies on PBC communication.

Central bank communication in emerging market economies has only more recently attracted greater attention. For example, Luangaram and Wongwachara (2017) implement text analysis techniques previously applied for advanced economies to analyze communication by the Bank of Thailand. For China, Garcia-Herrero and Girardin (2013) use the hawkish-dovish communication classification described to test the impact of PBC communication on repo-market volatility and trading volume. SHU and Ng (2010) compile indices reflecting the direction and intensity of the PBC’s monetary stance based on the quarterly MPERs and Monetary Policy Committee meeting minutes. SUN (2013) studies the impact of PBC communication on the macroeconomy, including GDP, inflation, and industrial output.

This chapter contributes to the existing literature in the following ways: First, it comprehensively reviews China’s unique institutional setting, which is paramount to understanding its evolving communication channels. Second, in addition to communication tools analyzed elsewhere (that is, quarterly MPERs, quarterly Monetary Policy Committee meeting minutes, as well as speeches and interviews), the study is novel in that it includes OMO notices. As discussed previously, the PBC considers these informative notices an important communication tool and started including them in 2016 (so-called small notes). Third, the chapter analyzes the transmission of communication to different markets.

Empirical Analysis

The empirical event study focuses on whether PBC communications contain news for financial market variables. Consistent with the central bank communication literature, this analysis is interested in the effect of communication on the intermediate target of monetary policy, that is, short-term interest rates. The first step is to construct variables that capture PBC communication events. As summarized in the previous section, the PBC communicates with the market in four main ways: (1) quarterly MPERs; (2) Monetary Policy Committee meeting minutes (minutes); (3) press conferences and speeches by the governor, deputy governors, and the then-chief economist of the PBC Research Bureau (oral); and (4) informative OMO notices. The PBC’s website was searched to collect the date and time at which the MPER, minutes, press conference transcripts, texts of speeches, or OMO notices are posted. This information allows a dummy variable to be created for every trading day13 for each communication format: D(MPER), D(Minutes), D(Oral), and D(OMO Notice), with 1 indicating that the PBC communicates with the market in a certain format and 0 otherwise. An aggregate dummy, D(PBC Communication), was also created, which is 1 if the PBC has utilized at least one of the four communication formats, and 0 otherwise. For oral communication, the analysis is interested in whether the effect of communication from the governor differs from that of others, so it also documents the person who attended the press conference or who delivered the speech. Hence, D(Oral) is disaggregated into D(Oral – PBCGOV) and D(Oral – Others).

Box 14.1.Communication Terminology

Monetary Policy Stance

As do other central banks, the People’s Bank of China uses specific terminology to communicate its policy stance. The signaling, however, is quite infrequent (usually during the Work Conference in December) and requires State Council approval.

Five categories can be identified: (1) moderately loose, (2) moderately tight, (3) tight, (4) prudent, and (5) prudent and neutral.1 While categories (1) to (3) are straightforward, the distinction between “prudent” and “prudent and neutral” is more ambivalent, with the latter assumed to be somewhat tighter than “neutral.” To visualize the potential relationship between the respective monetary policy stances and policy interest rates,2 the following scoring is used: -1 for “moderately loose,” 0 for “prudent,” +0.5 for “prudent and neutral,” +1 for “moderately tight,” and +2 for “tight.” In Figure 14.1.1, the red line shows the monetary policy stance, the blue line shows changes in the benchmark lending rate maturing in six months to one year, and the green line is the 7-day reverse repo open market operations (OMO) rate.

Figure 14.1.1.Monetary Policy Stance and Policy Rates, 2001–17

(Percent)

Source: CEIC.

Note: OMO = open market operations; repo = repurchase.

Changes in language about the monetary stance in Monetary Policy Executive Reports (MPERs) tend to be associated with changes in the benchmark interest rate; a positive change in the score, reflecting a tightening, is associated with an increase in benchmark interest rates (in the following quarter), and vice versa. However, as the figure shows, at times communication lags the actual decision about a benchmark rate change.

That terminology provides the PBC with significant flexibility and is reflected in the fact that during periods of a “prudent” monetary policy stance, the benchmark interest rate has both increased and decreased.

Window Guidance

In addition to written and oral communication disseminated publicly, the PBC provides “guidance” to financial institutions that affect their behavior. This “window” guidance can be related to credit growth, lending to priority sectors, and so on. It can be an effective tool, but it undermines transparency and interest rates as the main signal of monetary policy conditions.

1 See ZHOU (2016a) for a classification. The term “prudent and neutral” was added for the first time at the beginning of 2017.2 The 7-day interbank pledged repo rate, DR007, is one of the important benchmark rates, but has not yet been confirmed as a policy interest rate.

Short-term money market rates are an important focus for the PBC. The analysis therefore examined the effects of communication on the Shanghai interbank offered rate (Shibor), the collateralized interbank repo rate between depository financial institutions (DR007), and the collateralized interbank repo rate between all financial institutions (R007). Moreover, the focus is on the impact of the PBC’s communication on short-term and medium-term commercial paper (STN and MTN), Treasury bonds, and equity markets. Table 14.1 lists the details of these market rates.

To control for other factors that may affect market interest rate movements, the following are included: (1) a dummy variable D(Macro Release) to indicate the release of the main macroeconomic indicators, including the consumer price index (CPI), GDP, Official Purchasing Managers Index and Caixin Purchasing Managers Index (PMI), and foreign trade and foreign exchange reserves; (2) a dummy variable D(CEWC) to indicate the date of the government’s annual Central Economic Work Conference, which determines the direction of economic policy for the following year14; and (3) a dummy variable D(Rate Change) to indicate changes in PBC monetary policy instruments. For the latter, changes in the benchmark interest rate, required reserve ratio, OMO, and other instrument rates—that is, the standing lending liquidity facility (SLF), short-term liquidity operations (SLO), the medium-term lending facility (MLF), and the pledged supplementary lending (PSL)—are included.

The event-study analysis covers trading days during 2013–17. The summary statistics for the explanatory variables are reported in Table 14.1.1, including communication and control variables, and various market rates in Annex Table 14.1.2. On average, the PBC communicated with the market on about 25 percent of all trading days. Speeches or press conferences are the most common format of communication, taking place nearly every six trading days (16.1 percent of days in the sample). OMO information notices are the next most frequent, occurring on 7.8 percent of the days, but they have become much more frequent since February 2017. MPERs represent 1.6 percent; and Monetary Policy Committee meeting minutes, 1.5 percent.15 On days with communication, the average interest rates of the key assets are higher, reflecting more communication toward the end of the sample, when interest rates had risen. The standard deviations of these rates seem to be larger for longer-term assets and smaller for shorter-term assets than on days without communication.

The kernel density of the daily absolute change of the DR007 (Figure 14.7) and 3-month short-term notes (Figure 14.8) for various categories of communication are plotted. These plots show that both communication and control variables indeed have some impact on market prices. Surprisingly, minutes releases are associated with news for repo rates (DR007), but this effect is not reflected in short-term notes. Oral communication, PBC interest rate changes, and OMO notices are associated with relatively higher market reaction for both assets.

Figure 14.7.Kernel Density of Daily Absolute Changes of 7-Day Interbank Pledged Repo Rate (DR007)

Source: Authors’ calculations.

Note: PBC = People's Bank of China.

Figure 14.8.Kernel Density of Daily Absolute Changes of 3-Month Short-Term Notes

Source: Authors’ calculations.

Note: D.STN (3m) = daily changes to 3-month short-term note; PBC = People's Bank of China.

Baseline Results: OLS Analysis of Communication News

First, an ordinary least squares (OLS) analysis is conducted by estimating the following equations:

These regressions, as in other event studies, identify the average reaction of market prices to the different types of communication events; the identifying assumption is that any systematic reaction on communication days is driven by the communication event. The dependent variable is, as for the kernel densities, the daily absolute change in market rates. The advantage of using the absolute value is that the market reaction to the communication irrespective of the direction of that reaction can be measured. The key test concerns the sign of the communication event dummy variable. When communication is associated with news, it would cause a market adjustment and thus positive β coefficient. If, however, a communication event systematically reduced the noise in the market, it would be associated with less market adjustment than on other trading days and the estimated w coefficient would be negative.

The key estimation results, correcting the standard errors using the Huber-White sandwich estimators, are presented visually in Figures 14.9 and 14.10; the full regression results are presented in Annex Tables 14.1.3 and 14.1.4.

Figure 14.9.Estimated Coefficients of Aggregated Communication Events

Source: Authors’ calculations.

Note: CEWC = Central Economic Work Conference; DR007 = 7-day interbank pledged repo rate; MTN (1y) = 1-year medium-term note; PBC = People’s Bank of China; ST CP (3m) = 3-month short-term commercial paper; Treasury (3y) = 3-year Treasury.

Figure 14.10.Estimated Coefficients of Disaggregated Communication Events

Source: Authors’ calculations.

Note: CEWC = Central Economic Work Conference; DR007 = 7-day interbank pledged repurchase rate; Minutes = minutes of Monetary Policy Committee meetings; MPER = Monetary Policy Executive Report; MTN (1y) = 1-year medium-term note; OMO = open market operations; Oral = oral communication; ST CP (3m) = 3-month short-term commercial paper; Treasury (3y) = 3-year Treasury.

Figure 14.10 suggests that PBC communication mainly affects short-term assets, while neither monetary policy instruments nor communication systematically move longer-term markets much, at least in a single day’s time frame. More specifically, the main results are the following:

  • 1. First, as expected, policy instrument changes are associated with market news. However, PBC communication, in general, does not move the market too much, suggesting that the effectiveness of PBC communication is somewhat limited. As reflected in Annex Table 14.1.3, the coefficients of D(Rate Change) are significantly positive for all assets with maturities less than 3 months, while the coefficients of D(PBC Communication) are only significant for the volatility of 1-week Shibor and 1-month short-term notes. Communication slightly calms markets, but does not move the market much.

  • 2. Second, the release of quarterly MPERs tends to reduce the volatility of short-term notes. It acts as an important channel for the PBC to communicate with the market, as it contains a lot of operational details and sometimes forward-looking guidance, though not very often. As Annex Tables 14.1.3 and 14.1.4 show, MPERs are associated with a decline of 0.03 percentage point in the absolute change of 1-month short-term notes and 0.01 percentage point in that of 3-month short-term notes.

  • 3. Third, oral communication of public speeches, as well as press conferences, appears to calm the market. As Table 14.1.3 shows, oral communication is associated with a decrease in the market news of Shibor 1-week, and 1-month short-term notes. Annex Annex Table 14.1.4 shows that for the PBC’s DR007 target rate, communication events by the governor, either speeches or press conferences, are associated with 0.023 percentage point less market movement. However, oral communication is irregular, and the timing of oral communication is usually determined by market conditions.

Effect on Volatility and the Timing of Communication

Even if there is no systematic effect on the average price in the market, another possibility is that communication events are associated with greater market volatility. OLS regressions do not separately model the volatility of the residuals; the regressions analysis discussed earlier did correct the standard errors on coefficients for heteroscedasticity and auto-correlation. General autoregressive conditional heteroscedastic (GARCH) models can address the possibility that time series models may violate homoscedasticity (constant variance of errors). These models are used extensively in financial econometrics as higher-frequency data tend to have periods of higher and lower volatility.

Specifically, and following Nelson (1991), the volatility effects of communication using an EGARCH methodology (a particular class of GARCH model) is explored. The advantage of an EGARCH model over a GARCH model is that it ensures that the conditional variance is positive and allows for the asymmetric response of the volatility to good and bad news. The EGARCH(1,1) specification features a conditional mean equation in which it is necessary to control for all the relevant variables that drive the mean to minimize the size and variance of residuals:

where ϵt~N(0,σt2) and xt contains any controls in the conditional mean equation. The conditional volatility equation, which measures the drivers of σt2, is:

where D(Commsk)t indicates one of k included communication dummy variables, described previously.

To study the volatility effects while also ensuring that the dependent variables are stationary, the daily change of interest rates as the main dependent variable (∆rt ) is used (earlier, the focus was on market news given by the absolute value of the change in the interest rate). Given that the communication variables are dummy variables that could have a positive or negative effect on the level of the interest rate (depending on the message in the communication), in the mean equation the lag of the dependent variable (∆rt-1) and the change in the 7-day repo rate are controlled for, but not for the communication event dummies.

Instead, the effect of communication events on volatility is allowed. However, as the EGARCH model is a nonlinear model, it does not always have a smooth likelihood function that is easily maximized. To ensure convergence of the estimates, more parsimonious models in this section than were utilized earlier are explored. In particular, the volatility effects of each of the four main communication events as used in equation (14.1) are explored. Then a more disaggregated approach that focuses on the marginal effects (over the average communication event) of D(MPER) and D(Oral – PBCGOV) is explored.

An average PBC communication is associated with higher volatility of short-term market rates (R007). The EGARCH results are presented in Table 14.1.5 and Annex Table 14.1.6. When the effect is disaggregated a bit further, it is found that the association is particularly strong for speeches by the PBC governor. However, one needs to be cautious in interpreting these findings. Especially for ad hoc events (such as an irregular press conference or speech), it could be that the volatility in the market leads the PBC to feel the need to comment. Even if the comment is calming, volatility could remain higher than usual and so the event is associated with high volatility. The EGARCH model attempts to control for this by modeling volatility as dependent on the volatility from the previous trading day as well as other covariates.

The alternative interpretation of these results is that the event causes this higher volatility. When the PBC initiates ad hoc communication, the markets receive not only the content of the communication but also the message that the situation was severe or worrying enough to warrant communication. Regular communication allows markets to focus on the message being delivered since the date for the message is set in advance, as is the case with the Bank of England inflation report or US Federal Open Market Committee meetings.

An advantage of the EGARCH methodology is that it provides an estimate of the conditional variance ( (σ^t2) from each asset market. Figures 14.11 and 14.12 show this estimated conditional variance for the 7-day repo market and the Treasury (3-year) market. These figures, augmented with indicators of speeches by the PBC governor, highlight that it is likely that both explanations are present in the data. In particular, speeches tend to be made irregularly and around episodes of higher volatility (as well as after the liberalization of interest rate markets in 2016). In some cases, the speeches are followed by increased volatility, and in others volatility falls from elevated levels.

Figure 14.11.7-Day Repo Volatility, 2013–17

Source: Authors’ calculations.

Note: D(speech_zhou) = dummy variable for speeches by the People’s Bank of China governor Xiaochuan Zhou; repo = repurchase.

Figure 14.12.Treasury 3-Year Volatility, 2013–17

Source: Authors’ calculations.

Note: D(speech_zhou) = dummy variable for speeches by the People’s Bank of China governor Xiaochuan Zhou.

Use and Impact of OMO Notices

Open market operations are the important way in which the PBC intervenes in the market. The transmission efficiency from OMO interest rates to the targeted market rates largely determines how monetary policy works. Like other central banks, the PBC conducts OMO frequently and publishes OMO notifications on its website. Usually, OMO notices plainly state that the PBC has conducted OMOs at a specific interest rate and amount. The frequency of OMO actions is depicted in Figure 14.13 as the purple triangles. In the sample of 1,249 trading days, the PBC conducted OMO interventions on 639 of them.

Figure 14.13.Market Interventions by the People’s Bank of China, 2013–17

Source: People’s Bank of China.

Note: OMO = open market operations.

Since March 2016, however, the PBC at times has added short sentences about market performance (that is, about liquidity and volatility) to explain the reason behind the OMOs, including information about the injection or withdrawal of liquidity, or even the reason for not intervening in the market.16 These informative OMO notices, compared to the standard ones, are seen as “small notes” passed from the PBC to the market, and are identified as a format of the PBC’s communication. The dummy D(OMO Notice) takes the value of 1 when the PBC publishes an informative notice, including brief rationales for the PBC’s OMO decision. Otherwise the value is 0. There were 98 informative notices in the sample. The first two were in March 2016 and the next in October 2016; they started regularly in February 2017. Many of the notices (68) are issued on days when the PBC does not conduct an OMO action and the notice explains the reasons for the lack of action.

The last type of market intervention by the PBC examined are changes to the OMO interest rate. The dummy D(OMO Rate Change) is defined as equal to 1 if the PBC changed any rates of the repo, reverse repo, and the PBC bill at any tenor. This differs from D(Rate Change) used previously, which documents, in addition to changes in OMO rates, changes in benchmark interest rates, the required reserve ratio, and the other instrument rates such as the short-term liquidity facility, medium-term lending facility, and so on. In the sample, there were 37 trading days on which the PBC changed an OMO rate. All were associated with OMO actions on the same day, while only two occurred with OMO informative notices.

The PBC OMO notices allow us to study the effect of their communication on monetary policy transmission efficiency. Efficient OMO transmission would ensure that interest rates in the market moved relatively in step with the rates engineered by the central bank through market operations. To investigate this, the co-movement of market interest rates (Δrtmkt) is explored with OMO interest rates and the analysis examines how this differs during the period when the PBC was issuing informative notices. The dummy D(Notice Era) is defined, which is 1 during the period after which the PBC was regularly issuing these informative notices, that is, February 2017. The reason to focus on the period of regular notices is that once the market comes to expect and understands such communication, a trading day without a note is informative. The specification used is the following:

Without the interaction term, the estimated coefficient β1 indicates that if the OMO rates increase by 1 percentage point (all rates are represented in percentage points), the money market rates will increase by β1 percentage points; stronger transmission of OMO rates would be captured by larger reaction of market rates, although values over 1 would indicate an overreaction. The estimated coefficient β2, on the interaction term, captures the marginal change in this relationship in the era of the PBC issuing informative notices. A positive and statistically significant β2 coefficient indicates that the PBC s communication from the informative OMO notice has strengthened the transmission efficiency from the OMO rate to the market rate.

The results are presented in Annex Table 14.1.7. The era of issuing informative notices has been associated with a stronger transmission of OMO rates to other market rates. Not unexpectedly, given that the focus is on a 7-day reverse repo interest rate as the OMO rate, the results are strongest at the short end of the yield curve. Nonetheless, there has been a stronger co-movement, even with 1-year Shibor rates, since the OMO notices have been issued.

A first concern is that the informative notice era coincides with the period following the liberalization. The overlap is not perfect; the PBC officially liberalized interest rates in October 2015, whereas the informative notices only became regular in February 2017. Two alternative approaches are considered. First, the correlation on the days on which an informative notice is issued, D(Notice Era), is replaced with D(OMO Notice) in equation (14.5).17 The results, not reported here for the sake of brevity, are similar. A second approach is to estimate the regression in equation (14.5) but limiting the sample to the period after October 2015.

Another concern might be that rather than strengthen the transmission, the notices have simply sped up the pass-through from OMO interest rates to other market interest rates. The regression analysis discussed previously only considers the contemporaneous correlation between market and OMO interest rates. To explore the dynamic effects, one can allow for lags of OMO interest rates to drive market interest rates. To limit the number of interactions, one needs to estimate the following specification, estimated separately for the era before and after informative notices were regularly issued:

Annex Table 14.1.8 presents the relevant estimation results. They suggest that the transmission in the change of the rate is not simply about faster transmission; instead, the transparency from using informative OMO notices appears to have strengthened the transmission channel from PBC intervention to the market. The analysis indicates that understanding the rationale behind the PBC’s change in the OMO rate, or reason for not intervening in markets, tends to help the market understand the policy better and reduces the noise.

Policy Recommendations

The empirical analysis suggests that while the PBC has made a number of important improvements to its communication, the timing for further strengthening communication and undertaking associated institutional changes is propitious.

Institutional changes take time and require resources and political capital. But as China develops its financial system, particularly market-based financing, the need for transparent, clear, timely, and comprehensive communication will continue to increase. Failure to address existing institutional shortcomings may limit the benefits of liberalization and slow economic development.

While larger institutional changes will take time, practical actions could improve communication in advance. In fact, these suggestions are better if implemented in advance of or at least in tandem with greater operational independence. For example, the Bank of England first published its Inflation Report in February 1993 when it did not have control over UK monetary policy. When it became operationally independent in June 1997, the central bank’s thinking and analytical tools were already well understood by the market. Lack of independence should hence not be viewed as a reason not to press ahead. As a matter of fact, certain “low-hanging fruit” could be implemented quickly:

  • Making information available in a timely fashion, in one place, and in English: This would go a long way and would also be consistent with the intention to make capital markets more attractive, as well as with China’s renminbi internationalization strategy, and its participation in global forums.18

  • Expanding the PBC’s economic forecasting capacity and publishing forecasts regularly, as well as making information available about the associated framework and models: This would reduce surprises, making monetary policy more predictable. While the PBC’s Research Bureau released economic forecasts in 2015 and 2016, each with a mid-year update, these were discontinued in 2017. Of course, strengthening forecasting would require appropriate resources and expertise; here the PBC could tap into the experience of modern central banks, as well as technical collaboration with the International Monetary Fund.

  • Holding regular press conferences: A regular communication mechanism could reduce information asymmetry between the central bank and the market and allow markets to better interpret decisions and hence reduce uncertainty. As with forecasting, it is useful to begin these efforts even in advance of greater operational independence. Such a process will help the PBC learn how to communicate with the market directly and to build credibility that will serve it well, if it is—as is warranted—to become operationally more independent.19

Conclusions

Central banks are increasingly using communication as a lever of monetary policy and are often at the forefront of communication. The same is true for the PBC, which has taken important steps to improve communication. Given China’s global footprint and stated policies to further liberalize its financial system and continue moving toward price-based monetary policy, further improvements in communication will be critical.

This chapter assesses the impact of the PBC’s communication on financial markets using four types of communication: the quarterly published MPER, the quarterly released Minutes of the Monetary Policy Committee Meeting, press conferences and speeches by governors and deputy governors, and a novel channel, including informative OMO notices. The daily absolute change of several market interest rates in the money and bond markets, as well as equity market prices, are calculated to represent market reaction. They are regressed on PBC communication and control variables to estimate whether PBC communication affects market reaction and volatility.

The empirical analysis highlights that communication can have important benefits and that greater central bank transparency and independence would help further improve the PBC’s effectiveness, including through forward guidance. The introduction of OMO information notices, for example, reduced volatility and improved monetary policy effectiveness. While some institutional changes are likely to take time, some low-hanging fruit could be adopted in the short term. For example, providing timely information in one place (in Chinese and English), expanding PBC forecasting resources and capacity, and holding regular press conferences would not only be helpful for monetary policy, but also increase the attractiveness of China’s capital markets and advance renminbi internationalization.

In 2018, the PBC reiterated its commitment to further strengthen communication and transparency and has already taken additional steps, including through more press conferences and interviews, deeper analysis, and more informative MPERs to guide market expectations. In addition, at the end of 2018, the central bank set up a new working group to translate policy statements and news releases into English to better provide information to international investors (Bloomberg 2018).

Annex 14.1. Statistical Tables
Annex Table 14.1.1.Statistics Summary for Communication and Control Variables
Dummy VariableMeanStandard Deviation
D(CEWC)0.0010.001
D(MPER)0.0150.015
D(Minutes)0.0150.015
D(OMO Notice)0.0380.038
D(Macro Release)0.0880.088
D(Rate Change)0.1160.116
D(Oral – PBC Governor)0.0150.015
D(Oral Communication)0.0590.059
D(PBC Communication)0.1190.119
D(Oral – Other)0.0440.044
Source: Authors’ calculation.Note: CEWC = Central Economic Work Conference; MPER = Monetary Policy Executive Report; OMO = open market operations; PBC = People’s Bank of China.
Source: Authors’ calculation.Note: CEWC = Central Economic Work Conference; MPER = Monetary Policy Executive Report; OMO = open market operations; PBC = People’s Bank of China.
Annex Table 14.1.2.Statistics Summary for Market Rates
Communication=0Communication=1
VariableMeanStandard DeviationMeanStandard Deviation
R0072.6412.6413.0463.046
DR0072.7222.7222.7462.746
Shibor (1d)2.3532.3532.4602.460
Shibor (1w)2.9252.9252.9082.908
MTN (1y)4.0644.0644.1644.164
MTN (3m)4.1384.1384.0804.080
MTN (6m)4.2064.2064.1534.153
ST CP (1m)4.0164.0163.9353.935
ST CP (3m)3.8603.8603.9803.980
ST CP (6m)3.9543.9544.0634.063
Treasury (1y)2.6352.6352.9192.919
Treasury (5y)3.2283.2283.3463.346
Treasury (10y)3.5673.5673.5593.559
SSE Return0.0130.0130.0960.096
Source: Authors’ calculations.Note: DR007 = 7-day interbank pledged repo rate; MTN = medium-term note; R007 = 7-day repo rate for nonfinancial institutions; Shibor = Shanghai interbank offered rate; SSE = sum of squared errors; ST CP = short-term commercial paper; 1d = 1 day; 1w = 1 week; 1m = 1 month; 3m = 3 month; 6m = 6 month; 5y = 5 year; 10y = 10 year.
Source: Authors’ calculations.Note: DR007 = 7-day interbank pledged repo rate; MTN = medium-term note; R007 = 7-day repo rate for nonfinancial institutions; Shibor = Shanghai interbank offered rate; SSE = sum of squared errors; ST CP = short-term commercial paper; 1d = 1 day; 1w = 1 week; 1m = 1 month; 3m = 3 month; 6m = 6 month; 5y = 5 year; 10y = 10 year.
Annex Table 14.1.3.Ordinary Least Squares Estimates for Aggregated Communication
(1) |D.Repo(7d)-DR007|(2) |D.Shibor 1w)|(3) |D.STCP(1m)|(4) |D.STCP(3m)|(5) |D.STCP(6m)|(6) |D.MTN (1y)|(7) |D.Treasury (3y)|(8) |D.Treasury (5y)|(9) |D.SSE Return|
D(CEWQ-0.018

[0.338]
-0.062**

[0.011]
0.053

[0.340]
0.069*

[0.089]
0.052**

[0.044]
0.033

[0.128]
0.0092

[0.280]
-0.0093***

[0.004]
-0.35*

[0.082]
D{Macro Release)-0.011*

[0.076]
-0.026*

[0.051]
-0.014

[0.131]
0.0014

[0.734]
0.00060

[0.863]
0.0015

[0.724]
-0.00046

[0.754]
0.0016

[0.310]
0.17

[0.224]
D{Rate Change)0.088**

[0.014]
0.13**

[0.011]
0.035*

[0.095]
0.021**

[0.039]
0.0100

[0.110]
0.0074

[0.221]
0.00042

[0.837]
0.0010

[0.667]
0.83***

[0.001]
D(PBC Communication)0.0026

[0.682]
-0.058***

[0.000]
-0.020**

[0.023]
-0.0011

[0.797]
-0.0033

[0.261]
0.00069

[0.814]
-0.00011

[0.931]
0.00088

[0.494]
-0.15

[0.147]
Constant0.055***

[0.000]
0.11***

[0.000]
0.079***

[0.000]
0.044***

[0.000]
0.039***

[0.000]
0.036***

[0.000]
0.021***

[0.000]
0.021***

[0.000]
1.32***

[0.000]
R20.0630.0310.0040.0070.0060.0030.0010.0020.019
Source: Authors' calculations.Note: CEWC = Central Economic Work Conference; DR007 = 7-day interbank pledged repo rate; MTN = medium-term note; PBC = People's Bank of China; Shibor = Shanghai interbank offered rate; SSE = sum of squared errors; STCP = short-term commercial paper; 1m = 1 month; 1w= 1 week; 3m = 3 month; 6m = 6 month; 3y = 3 year; 5y = 5 year; 7d = 7 day. The p values appear in brackets.*p<.1;**p<.05;***p<.01.
Source: Authors' calculations.Note: CEWC = Central Economic Work Conference; DR007 = 7-day interbank pledged repo rate; MTN = medium-term note; PBC = People's Bank of China; Shibor = Shanghai interbank offered rate; SSE = sum of squared errors; STCP = short-term commercial paper; 1m = 1 month; 1w= 1 week; 3m = 3 month; 6m = 6 month; 3y = 3 year; 5y = 5 year; 7d = 7 day. The p values appear in brackets.*p<.1;**p<.05;***p<.01.
Annex Table 14.1.4.Ordinary Least Squares Estimates for Communication, by Category
(1) |D.Repo(7d)-DR007|(2) |D.Shibor (1w)|(3) |D.STCP(1m)|(4) |D.STCP(3m)|(5) |D.STCP(6m)|(6) |D.MTN(1y)|(7) |D.Treasury(3y)|(8) |D.Treasury (5y)|(9) |D.SSE Return|
D(CEWQ-0.017

[0.348]
-0.061**

[0.014]
0.054

[0.330]
0.069*

[0.089]
0.052**

[0.043]
0.033

[0.126]
0.0089

[0.293]
-0.0094***

[0.003]
-0.40**

[0.048]
D{Maao Release)-0.011*

[0.073]
-0.027**

[0.045]
-0.014

[0.126]
0.0015

[0.721]
0.00084

[0.810]
0.0016

[0.708]
-0.00032

[0.828]
0.0017

[0.282]
0.19

[0.176]
D{Rate Change)0.089**

[0.015]
0.13**

[0.011]
0.036*

[0.092]
0.021**

[0.050]
0.011*

[0.087]
0.0073

[0.242]
0.00013

[0.949]
0.00098

[0.688]
0.77***

[0.002]
D(MPER)-0.0096

[0.492]
-0.012

[0.637]
-0.030***

[0.000]
-0.011*

[0.057]
-0.011

[0.123]
-0.0068

[0.369]
-0.0023

[0.535]
0.0037

[0.362]
-0.26

[0.191]
D{Minutes)0.028

[0.184]
0.15*

[0.083]
0.032

[0.421]
0.0096

[0.653]
-0.0019

[0.766]
0.013

[0.590]
-0.0021

[0.456]
-0.0082**

[0.017]
0.29

[0.406]
D(OMO Notice)0.015*

[0.065]
-0.088***

[0.000]
-0.0098

[0.334]
-0.00078

[0.863]
0.00012

[0.975]
-0.00074

[0.821]
-0.0024

[0.160]
-0.00062

[0.707]
-0.74***

[0.000]
D(Oral-Zhou)-0.023***

[0.006]
-0.023

[0.263]
-0.0074

[0.577]
-0.00088

[0.911]
0.011

[0.206]
0.0066

[0.423]
0.0022

[0.583]
0.0036

[0.401]
0.19

[0.456]
D{Oral – Other)-0.0029

[0.744]
-0.061***

[0.000]
-0.019*

[0.053]
0.0029

[0.637]
-0.0051

[0.174]
0.0023

[0.544]
0.0017

[0.332]
0.0013

[0.426]
0.13

[0.375]
Constant0.055***

[0.000]
0.10***

[0.000]
0.078***

[0.000]
0.044***

[0.000]
0.039***

[0.000]
0.035***

[0.000]
0.021***

[0.000]
0.021***

[0.000]
1.31***

[0.000]
R20.0710.0410.0040.0080.0080.0050.0030.0070.036
Source: Authors' calculations.Note: CEWC = Central Economic Work Conference; DR007 = 7-day interbank pledged repo rate; MTN = medium-term note; MPER = Monetary Policy Executive Report; R007 = 7-day repo rate for nonfinancial institutions; OMO = open market operations; PBC = People's Bank of China; Shibor = Shanghai interbank offered rate; SSE = sum of squared errors; ST CP = Short-term commercial paper; 1m = 1 month; 1w = 1 week; 3m = 3 month; 6m = 6 month; 3y = 3 year; 5y = 5 year. Thep-values appear in brackets.*p<.1;**p<.05;***p<.01.
Source: Authors' calculations.Note: CEWC = Central Economic Work Conference; DR007 = 7-day interbank pledged repo rate; MTN = medium-term note; MPER = Monetary Policy Executive Report; R007 = 7-day repo rate for nonfinancial institutions; OMO = open market operations; PBC = People's Bank of China; Shibor = Shanghai interbank offered rate; SSE = sum of squared errors; ST CP = Short-term commercial paper; 1m = 1 month; 1w = 1 week; 3m = 3 month; 6m = 6 month; 3y = 3 year; 5y = 5 year. Thep-values appear in brackets.*p<.1;**p<.05;***p<.01.
Annex Table 14.1.5.EGARCH Results of Aggregated Communication
D.R007 (1)D.ST CP(3m) (2)D.MTN(1y) (3)D.Treasury(3y) (4)SSE Return (5)
Conditional Mean
Dependent Variable (t-1)0.14***

(0.000)
0.18***

(0.000)
0.13***

(0.000)
0.07***

(0.023)
0.02

(0.447)
D.Reverse Repo Rate0.38***

(0.024)
-0.00

(0.971)
0.02

(0.772)
-0.02

(0.490)
0.36

(0.873)
Constant0.01***

(0.000)
0.00

(0.492)
-0.00

(0.305)
0.00

(0.969)
0.03

(0.274)
Conditional Volatility
D(CEWC)2.44***

(0.000)
-0.43

(0.422)
-0.26

(0.603)
-0.87

(0.146)
0.46***

(0.009)
D(Macro Release)-0.43***

(0.000)
0.11*

(0.076)
0.04

(0.432)
-0.06

(0.406)
-0.02

(0.514)
D(Rate Change)0.06

(0.302)
0.28***

(0.000)
0.04

(0.461)
0.03

(0.743)
0.03

(0.580)
D(PBC Communication)0.08***

(0.000)
-0.01

(0.873)
-0.01

(0.812)
-0.00

(0.949)
-0.03

(0.251)
Constant-0.08***

(0.000)
-0.36***

(0.000)
-0.35***

(0.000)
-0.74***

(0.000)
0.02*

(0.070)
ARCH
L.earch0.25***

(0.000)
0.17***

(0.000)
0.08***

(0.000)
0.11***

(0.000)
0.01

(0.384)
L.earch0.36***

(0.000)
0.58***

(0.000)
0.30***

(0.000)
0.47***

(0.000)
0.20***

(0.000)
L.egarch0.96***

(0.000)
0.93***

(0.000)
0.94***

(0.000)
0.90***

(0.000)
1.00***

(0.000)
No. of observations12461246124612461159
Source: Authors’ calculations.Note: The table tests the volatility effects of communication using an EGARCH methodology, a particular class of General Autoregressive Conditional Heteroscedastic (GARCH) model, following Nelson (1991) (see glossary). CEWC = Central Economic Work Conference; R007 = 7-day repo rate for nonfinancial institutions; PBC = People’s Bank of China; SSE = sum of squared errors; ST CP = short-term commercial paper. The p-values appear in parentheses.*p < .1; **p < .05; ***p < .01.
Source: Authors’ calculations.Note: The table tests the volatility effects of communication using an EGARCH methodology, a particular class of General Autoregressive Conditional Heteroscedastic (GARCH) model, following Nelson (1991) (see glossary). CEWC = Central Economic Work Conference; R007 = 7-day repo rate for nonfinancial institutions; PBC = People’s Bank of China; SSE = sum of squared errors; ST CP = short-term commercial paper. The p-values appear in parentheses.*p < .1; **p < .05; ***p < .01.
Annex Table 14.1.6.EGARCH Results of Aggregated Communication, by Category and People
D.R007 (1)D.ST CP (3m) (2)D.MTN (1y) (3)D.Treasury (3y) (4)SSE Return (5)
Conditional Mean
Dependent Variable

(t-1)
0.14***

(0.000)
0.13***

(0.000)
0.14***

(0.000)
0.07***

(0.023)
0.03

(0.431)
D.Reverse Repo Rate0.36**

(0.039)
0.00

(0.983)
0.02

(0.798)
-0.02

(0.512)
0.33

(0.882)
Constant0.01***

(0.000)
0.00

(0.225)
-0.00

(0.209)
0.00

(0.992)
0.03

(0.245)
Conditional Volatility
D(CEWC)2.40***

(0.000)
-0.54

(0.354)
-0.13

(0.834)
-0.93

(0.130)
0.37*

(0.055)
D(Macro Release)-0.43***

(0.000)
0.06

(0.414)
0.06

(0.342)
-0.07

(0.342)
-0.03

(0.485)
D(Rate Change)0.06

(0.271)
0.45***

(0.000)
0.13*

(0.057)
0.02

(0.858)
0.03

(0.542)
D(PBC Communication)0.06***

(0.009)
0.09**

(0.026)
-0.03

(0.405)
0.02

(0.745)
-0.02

(0.450)
D(Oral – PBCGOV)0.17*

(0.095)
-0.15

(0.343)
0.74***

(0.000)
-0.24

(0.172)
-0.16**

(0.026)
D(MPER)-0.19

(0.251)
0.33**

(0.028)
0.16

(0.345)
-0.07

(0.751)
-0.12

(0.359)
Constant-0.07***

(0.000)
-0.88***

(0.000)
-0.56***

(0.000)
-0.73***

(0.000)
0.02**

(0.033)
ARCH
L.earch0.24***

(0.000)
0.19***

(0.000)
0.09***

(0.000)
0.11***

(0.000)
0.01

(0.424)
L.earch0.36***

(0.000)
0.77***

(0.000)
0.37***

(0.000)
0.47***

(0.000)
0.20***

(0.000)
L.egarch0.96***

(0.000)
0.84***

(0.000)
0.91***

(0.000)
0.90***

(0.000)
1.00***

(0.000)
No. of observations1,2461,2461,2461,2461,159
Source: Authors’ calculations.Note: The table tests the volatility effects of communication using an EGARCH methodology, a particular class of General Autoregressive Conditional Heteroscedastic (GARCH) model, following Nelson (1991) (see glossary). CEWC = Central Economic Work Conference; R007 = 7-day repo rate for nonfinancial institutions; MPER = Monetary Policy Executive Report; MTN = medium-term note; PBC = People’s Bank of China; repo = repurchase; SSE = sum of squared errors; ST CP = short-term commercial paper. The p-values appear in parentheses.*p < .1; **p < .05; ***p < .01.
Source: Authors’ calculations.Note: The table tests the volatility effects of communication using an EGARCH methodology, a particular class of General Autoregressive Conditional Heteroscedastic (GARCH) model, following Nelson (1991) (see glossary). CEWC = Central Economic Work Conference; R007 = 7-day repo rate for nonfinancial institutions; MPER = Monetary Policy Executive Report; MTN = medium-term note; PBC = People’s Bank of China; repo = repurchase; SSE = sum of squared errors; ST CP = short-term commercial paper. The p-values appear in parentheses.*p < .1; **p < .05; ***p < .01.
Annex Table 14.1.7.Open Market Operations (OMO) Notice and Transmission Efficiency
Variable(1) D.R001(2) D.R007(3) D.DR007(4) D.R014(5) D.Shibor(1d)(6) D.Shibor (1w)(7) D.Shibor (2w)(8) D.Shibor (1m)(9) D.Shibor (3m)(10) D.Shibor (6m)(11) D.Shibor (9m)(12) D.Shibor(1y)
D. Reverse Repo Rate-0.063

[0.179]
-0.0097

[0.897]
-0.0038

[0.962]
-0.11

[0.130]
-0.045*

[0.091]
0.027

[0.772]
-0.15***

[0.008]
-0.12***

[0.002]
-0.0069

[0.582]
0.0037

[0.292]
0.0019

[0.512]
0.0021

[0.455]
Notice Era0.0071

[0.573]
0.0089

[0.714]
0.0034

[0.659]
0.017

[0.456]
0.0038

[0.730]
0.0023

[0.792]
0.0050

[0.557]
0.0053

[0.414]
0.0042**

[0.011]
0.0044***

[0.000]
0.0045***

[0.000]
0.0047***

[0.000]
D. Reverse Repo Rate x Notice Era1.09***

[0.001]
6.12**

[0.011]
1.36**

[0.040]
0.84***

[0.003]
0.51***

[0.000]
0.33***

[0.001]
0.38***

[0.000]
0.74***

[0.000]
0.27***

[0.000]
0.30***

[0.000]
0.20***

[0.000]
0.19***

[0.000]
Constant-0.0017

[0.860]
-0.0010

[0.913]
-0.0020

[0.678]
-0.00023

[0.981]
-0.0014

[0.895]
-0.0012

[0.888]
-0.0011

[0.895]
-0.0011

[0.861]
-0.0000023

[0.999]
-0.00023

[0.634]
-0.00049

[0.245]
-0.00061

[0.115]
R20.0000.0040.0030.0010.0000.0000.0010.0010.0030.0190.0230.029
QuantileOLSOLSOLSOLSOLSOLSOLSOLSOLSOLSOLSOLS
Source: Authors' calculations.Note: DR007 = 7-day interbank pledged repo rate; OLS = ordinary least squares; R001 = 1-day repo rate for nonfinancial institutions; R007 = 7-day repo rate for nonfinancial institutions; R014 =14-day repo rate for nonfinancial institutions; Shibor = Shanghai interbank offered rate: Id = 1 day; 1w= 1 week; 2w = 2 week; 1m = 1 month; 3m = 3 month; 6m = 6 month; 9m = 9 month; 1y = 1 year. The p-values appear in brackets.*p < 0.1; **p < 0.05; ***p < 0.01.
Source: Authors' calculations.Note: DR007 = 7-day interbank pledged repo rate; OLS = ordinary least squares; R001 = 1-day repo rate for nonfinancial institutions; R007 = 7-day repo rate for nonfinancial institutions; R014 =14-day repo rate for nonfinancial institutions; Shibor = Shanghai interbank offered rate: Id = 1 day; 1w= 1 week; 2w = 2 week; 1m = 1 month; 3m = 3 month; 6m = 6 month; 9m = 9 month; 1y = 1 year. The p-values appear in brackets.*p < 0.1; **p < 0.05; ***p < 0.01.
Annex Table 14.1.8.Open Market Operations Notice and Transmission Efficiency: Dynamic Results
Variables(1) D.R001(2) D.R001(3) D.R007(4) D.R007(5) D.DR007(6) D.DR007(7) D.R014(8) D.R014(9) D.Shibor (1d)(10) D.Shibor (1d)(11) D.Shibor (1w)(12) D.Shibor (1w)
Notice EraNoYesNoYesNoYesNoYesNoYesNoYes
D. Reverse Repo Rate-0.058

[0.245]
1.07***

[0.001]
-0.0016

[0.984]
6.18**

[0.011]
-0.026

[0.732]
1.37**

[0.039]
-0.084

[0.296]
0.83***

[0.005]
-0.038

[0.199]
0.48***

[0.000]
0.031

[0.744]
0.36***

[0.000]
L. D. Reverse Repo Rate0.11

[0.181]
1.13

[0.240]
0.12

[0.351]
-2.10***

[0.000]
0.024

[0.860]
0.52

[0.471]
0.12**

[0.033]
0.53

[0.436]
0.12

[0.191]
0.84

[0.158]
0.088

[0.371]
0.12

[0.298]
L2.D. Reverse Repo Rate-0.17**

[0.043]
0.67***

[0.000]
-0.16

[0.463]
2.10***

[0.004]
0.26***

[0.007]
-0.28*

[0.061]
0.026

[0.870]
0.70***

[0.008]
-0.14*

[0.065]
0.074

[0.350]
-0.24

[0.105]
-0.00019

[0.999]
L3.D. Reverse Repo Rate-0.081

[0.633]
1.65

[0.188]
-0.12

[0.222]
5.40

[0.122]
0.12**

[0.035]
0.67

[0.259]
-0.33

[0.216]
6.18***

[0.000]
-0.096

[0.481]
0.11*

[0.086]
-0.061

[0.633]
0.089

[0.188]
Constant-0.0018

[0.849]
0.0016

[0.845]
-0.0012

[0.899]
0.0019

[0.933]
-0.00064

[0.895]
0.00047

[0.941]
-0.00041

[0.966]
0.0081

[0.690]
-0.0015

[0.887]
0.0013

[0.696]
-0.0014

[0.868]
0.00086

[0.466]
R20.0010.0310.0010.0500.0080.0200.0030.0420.0010.0330.0020.030
Variable(13) D.Shibor (2w)(14) D.Shibor (2w)(15) D.Shibor (1m)(16) D.Shibor (1m)(17) D.Shibor (3m)(18) D.Shibor (3m)(19) D.Shibor (6m)(20) D.Shibor (6m)(21) D.Shibor (9m)(22) D.Shibor (9m)(23) D.Shibor (1y)(24) D.Shibor (1y)
Notice EraNoYesNoYesNoYesNoYesNoYesNoYes
D. Reverse Repo Rate-0.13*

[0.065]
0.24***

[0.000]
-0.12***

[0.004]
0.63***

[0.000]
-0.0072

[0.561]
0.27***

[0.000]
0.0036

[0.302]
0.31***

[0.000]
0.0020

[0.500]
0.21***

[0.000]
0.0022

[0.450]
0.20***

[0.000]
L. D. Reverse Repo Rate0.11**

[0.012]
-0.045

[0.763]
-0.33**

[0.013]
0.34**

[0.012]
-0.0053

[0.540]
0.15***

[0.000]
0.0013

[0.438]
0.24***

[0.000]
-0.000027

[0.983]
0.26***

[0.000]
0.00062

[0.528]
0.23***

[0.000]
L2. D. Reverse Repo Rate-0.12

[0.268]
0.19***

[0.004]
-0.30**

[0.036]
0.32**

[0.018]
0.013

[0.235]
0.16***

[0.000]
0.0059

[0.268]
0.12***

[0.000]
0.0044

[0.392]
0.21***

[0.005]
0.0026

[0.263]
0.18***

[0.003]
L3. D. Reverse Repo Rate-0.32

[0.320]
0.20**

[0.027]
-0.079

[0.342]
0.37***

[0.000]
0.0055

[0.449]
0.32*

[0.064]
0.0018

[0.221]
0.12***

[0.002]
-0.0013

[0.621]
0.16***

[0.004]
-0.00057

[0.525]
0.14***

[0.010]
Constant-0.0014

[0.864]
0.0035***

[0.000]
-0.0019

[0.767]
0.0031*

[0.085]
0.000016

[0.990]
0.0035***

[0.001]
-0.00022

[0.649]
0.0037***

[0.000]
-0.00049

[0.249]
0.0033***

[0.000]
-0.00061

[0.118]
0.0035***

[0.000]
R20.0040.0630.0100.0750.0000.0690.0000.1400.0000.3300.0000.256
Source: Authors' calculations.Note: DR007 = 7-day interbank pledged repo rate; OLS = ordinary least squares; R001 = 1-day repo rate for nonfinancial institutions; R007= 7-day repo rate for nonfinancial institutions; R014=14-day repo rate for nonfinancial institutions; Shibor = Shanghai interbank offered rate: Id = 1 day; 1w= 1 week; 2w = 2 week; 1m= 1 month; 3m = 3 month; 6m = 6 month; 9m = 9 month; ly = 1 year.Thep-values appear in brackets.*p<.1;**p<.05;***p<.01.
Source: Authors' calculations.Note: DR007 = 7-day interbank pledged repo rate; OLS = ordinary least squares; R001 = 1-day repo rate for nonfinancial institutions; R007= 7-day repo rate for nonfinancial institutions; R014=14-day repo rate for nonfinancial institutions; Shibor = Shanghai interbank offered rate: Id = 1 day; 1w= 1 week; 2w = 2 week; 1m= 1 month; 3m = 3 month; 6m = 6 month; 9m = 9 month; ly = 1 year.Thep-values appear in brackets.*p<.1;**p<.05;***p<.01.
References

    AdrianT.D.Laxton andM.Obstfeld. 2018. “An Overview of Inflation-Forecast Targeting.” In Advancing the Frontiers of Monetary Policy edited by T.AdrianD.Laxton andM.Obstfeld. Washington, DC: International Monetary Fund.

    BergerH.V.Nitsch andT.Lybek. 2006. “Central Bank Boards around the World: Why Does Membership Size Differ?IMF Working Paper 06/281International Monetary FundWashington, DC.

    BernankeB. S.2007. “Federal Reserve Communications.” Speech at the Cato Institute 25th Annual Monetary Conference Washington DC November 14. https://www.federalreserve.gov/newsevents/speech/bernanke20071114a.htm.

    BlinderA. S. and others.2008. “Central Bank Communication and Monetary Policy: A Survey of Theory and Evidence.” Journal of Economic Literature46 (4): 91045.

    Bloomberg. 2018. “China’s Quiet Central Bank Now Wants to Talk to Global Investors.” December 4. https://www.bloomberg.com/news/articles/2018-12-04/china-s-quiet-central-bank-now-wants-to-talk-to-global-investors.

    DincerN. N. andB.Eichengreen. 2014. “Central Bank Transparency and Independence: Updates and New Measures.” International Journal of Central Banking10 (1): 189259.

    EhrmannM. andM.Fratzscher. 2007. “The Timing of Central Bank Communication.” European Journal of Political Economy23 (1): 12445.

    Garcia-HerreroA.and E.Girardin. 2013. “China’s Monetary Policy Communication: Money Markets Not Only Listen, They Also Understand.” HKIMR Working Paper 02/2013 Hong Kong Institute for Monetary Research. http://www.pbc.gov.cn/zhengwugongkai/127924/128038/128109/3544192/index.html.

    GürkaynakR. S.B.Sack andE.Swanson. 2005. “Do Actions Speak Louder Than Words? The Response of Asset Prices to Monetary Policy Actions and Statements.” International Journal of Central Banking1 (1): 5593.

    HansenS. andM.McMahon. 2016. “Shocking Language: Understanding the Macroeconomic Effects of Central Bank Communication.” Journal of International Economics99 (S1): S114S133.

    HansenS. andA.Prat. 2018. “Transparency and Deliberation within the FOMC: A Computational Linguistics Approach.” Quarterly Journal of Economics133 (2): 80170.

    HansenS.M.McMahon andM.TONG. 2018. “The Long-Run Information Effect of Central Bank Communication.” Working Paper. Oxford University and Bank of England.

    HUANGY.T.GE andC.WANG. 2018. “Monetary Policy Framework and Transmission Mechanisms.” Chapter 3 in the Handbook of China’s Financial System edited by M.AmstadG.Sun andW.Xiong.

    International Monetary Fund (IMF). 2018. “People’s Republic of China Staff Report for the 2018 Article IV Consultation.” Washington, DC. https://www.imf.org/en/Publications/CR/Issues/2018/07/25/Peoples-Republic-of-China-2018-Article-IV-Consultation-Press-Release-Staff-Report-Staff-46121.

    JansenD. J. andJ. D.Haan. 2005. “Talking Heads: The Effects of ECB Statements on the Euro-Dollar Exchange Rate.” Journal of International Money and Finance24 (2): 34361.

    KohnD. L. andB. P.Sack. 2003. “Central Bank Talk: Does It Matter and Why?Divisions of Research and Statistics and Monetary Affairs Federal Reserve BoardWashington, DC.

    LIK.2010–2018. Report on the Work of the Government. National People’s CongressBeijing.

    LuangaramP. andW.Wongwachara. 2017. “More than Words: A Textual Analysis of Monetary Policy Communication.” PIER Discussion Paper 54Puey Ungphakorn Institute for Economic ResearchBangkok.

    LuccaD. O. andF.Trebbi. 2009. “Measuring Central Bank Communication: An Automated Approach with Application to FOMC Statements.” NBER Working Paper 15367National Bureau of Economic ResearchCambridge, MA.

    MAJ. 2017. “Interest Rate Transmission in a New Monetary Policy Framework.” In Modernizing China: Investing in Soft Infrastructure edited by W. R.LamM.Rodlauer andA.Schipke. Washington, DC: International Monetary Fund.

    MAJ. andT.GUAN. 2018. “Interest Rate Liberalization and Reform of China’s Monetary Policy Framework.” China Financial Press (in Chinese).

    NelsonD. B.1991. “Conditional Heteroskedasticity in Asset Returns: A New Approach.” Econometrica59 (2): 34770.

    People’s Bank of China. 2018a. “Information Disclosure and Transparency” (in Chinese). Beijing. May 23. http://www.pbc.gov.cn/zhengwugongkai/127924/128038/128109/3544192/index.html.

    People’s Bank of China. 2018b. “Q&A of Governor Yi Gang at Sub-forum ‘Monetary Policy Normalization’ of BOAO Forum for Asia Annual Conference.” http://www.pbc.gov.cn/english/130724/3523967/index.html.

    ReevesR. andM.Sawicki. 2007. “Do Financial Markets React to Bank of England Communication?European Journal of Political Economy23 (1): 20727.

    SHUC. andB.Ng. 2010. “Monetary Stance and Policy Objectives in China: A Narrative Approach.” HKMA China Economic Issues1 (10): 140.

    SUNR. 2013. “Does Monetary Policy Matter in China? A Narrative Approach.” China Economic Review26 (C): 5674.

    TobbackE.S.Nardelli andD.Martens. 2017. “Between Hawks and Doves: Measuring Central Bank Communication.” ECB Working Paper 2085European Central BankFrankfurt.

    ZHOUX. 2016a. Press Conference on Financial Reform and Development during the 2016 National People’s Congress March 12. http://www.pbc.gov.cn/goutongjiaoliu/113456/113469/3030450/index.html.

    ZHOUX.2016b. “Managing Multi-Objective Monetary Policy: From the Perspective of Transitioning Chinese Economy.” The 2016 Michel Camdessus Central Banking Lecture June 24International Monetary FundWashington, DC.

On principles and practices of inflation-forecast targeting in advanced economies, see Adrian, Lax- ton, and Obstfeld (2018).

On May 11, 2015, the PBC expanded the upper limit of the deposit rate floating range from 1.3 times to 1.5 times the benchmark rate; on August 26, 2015, the central bank completely liberalized the upper limit of deposit rates with fixed terms longer than one year; and on October 24, 2015, it cancelled the deposit interest rate upper limit of commercial banks and rural cooperative financial institutions. On August 11, 2015, the PBC announced that the CNY/USD (Chinese yuan/US dollar) rate should refer to the closing rate of the previous day. On December 11 that year, the CFETS CNY index was first released by the China Foreign Exchange Trade System (CFETS) to reinforce the reference to the basket of currencies. Since February 2016, the CNY/USD fixing rate mechanism consisting of the closing rate and the currency basket has been forming gradually and the transparency of the mechanism has increased.

See Article 2 of the General Rules in the People’s Bank of China Law.

See Article 5 of the General Rules in the People’s Bank of China Law.

Deputy governors are appointed and dismissed by China’s premier.

On transparency, see Dincer and Eichengreen (2014).

Compared to another 7-day repo rate, which covers banks and nonfinancial institutions (R007), DR007 is the weighted average of participating banks.

Both governors and relevant PBC staff are available to speak to journalists. The PBC’s governor gave 63 percent of 86 speeches delivered by the governor or deputy governors between 2007 and the end of 2017. Out of 140 press conferences during the same period, the governor took questions or gave interviews in about 20 percent of them. Oral communication surged in 2015–16, when important policy changes took place, such as interest rate liberalization, changes in the exchange rate regime, and a sharp adjustment in equity prices. Communication tends to be more ex post, with a focus on explanations rather than guiding expectations, reflecting—among other things—China’s unique institutional setup and the PBC’s lack of operational independence.

The PBC started conducting weekly open market operations in 2003 and has held biweekly operations since 2004.

From 2017 to June 30, 2018, the PBC published 179 such notes, 122 of them on days with open market operations.

A prominent paper in the event-study analysis of communication is Gürkaynak, Sack, and Swanson (2005), which focuses on Federal Reserve monetary policy announcements. Kohn and Sack (2003) examine the effects of a broader set of Federal Reserve communications. Reeves and Sawicki (2007) carry out similar analysis for the Bank of England Inflation Report. More recently, Hansen, McMahon, and TONG (2018) show that central bank communication on risks and uncertainty is an important source of information for long-maturity yields. See also Tobback, Nardelli, and Martens (2017).

When the post time was between 3:00 pm and 9:00 am the following morning, the date was adjusted to the next trading day.

There are other regularly held meetings at the central level, such as the Politburo meetings, that may also have implications for monetary policy. By adding a dummy variable to indicate the date of the Politburo meetings with economic issues discussed, it is found that the Politburo meetings do not significantly affect financial market rates and this additional variable does not affect the results of PBC communication (results are available upon request). Therefore, Politburo meetings in the following analysis are not controlled for in order to keep a concise specification.

Since 2018, the PBC has taken additional measures to further strengthen policy transparency and communication by responding to public concerns in a timely manner and guiding market expectation. For instance, to better explain the rationale of OMOs, the PBC released 85 notices during the 123 trading days in the first half of 2018. In addition, the PBC has strengthened traditional channels, such as holding more press conferences (conducting 17 press conferences or interviews during the first six months of 2018), and providing deeper analysis and more informative MPERs.

The oldest notification dates back to January 2004. The PBC conducted OMOs weekly before May 2004, then adjusted the frequency to biweekly until January 2016. From 2016 to 2017, the PBC published OMO notifications daily, even for days without OMOs. Informative OMO notices started in March 2016.

The alternative equation is the following:

In January 2019, the PBC launched a new English webpage. See http://www.pbc.gov.cn/en/3688006/index.html.

    Other Resources Citing This Publication