## Box 1. Communication Terminology

## Source details

**Canonical URL:** [Box 1. Communication Terminology](https://www.imf.org/-/media/files/publications/wp/2018/wp18244.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/wp/2018/wp18244.pdf.md)
- [Structured JSON version](/-/media/files/publications/wp/2018/wp18244.pdf.json)

---

### Monetary policy stance and scoring
- The People’s Bank of China (PBC) uses five categories to communicate policy stance: (1) moderately loose, (2) moderately tight, (3) tight, (4) prudent, and (5) prudent and neutral.
- Signaling is infrequent (usually during the Work Conference in December) and requires State Council approval.
- Scoring used to visualize relationship between stance and policy interest rates:
  - -1 for “moderately loose”
  - 0 for “prudent”
  - +0.5 for “prudent and neutral”
  - +1 for “moderately tight”
  - +2 for “tight”
- Observation: changes in language about the monetary stance in 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), but communication can lag the actual decision about a benchmark rate change.
- During a “prudent” stance, the benchmark interest rate has both increased and decreased, reflecting flexibility in the terminology.

### Window guidance
- The PBC provides “window” guidance to financial institutions that can affect credit growth and lending to priority sectors.
- While effective, window guidance undermines transparency and reduces the role of interest rates as the main signal of monetary policy conditions.

### Communication channels, empirical strategy, and sample
- Four main PBC communication formats analyzed:
  - D(MPER): quarterly Monetary Policy Executive Reports
  - D(Minutes): Monetary Policy Committee meeting minutes
  - D(Oral): press conferences and speeches by the governor, deputy governors, and the then-chief economist of the PBC Research Bureau
    - Disaggregated into D(Oral - PBCGOV) and D(Oral - Others)
  - D(OMO Notice): informative OMO notices (so-called “small notes”)
- Aggregate dummy D(PBC Communication) = 1 if at least one of the four formats occurred on a trading day.
- Market rates studied include SHIBOR, DR007, R007, STN and MTN, treasury bonds, and equity markets.
- Control variables included:
  - D(Macro Release): CPI, GDP, Official PMI and Caixin PMI, foreign trade, foreign exchange reserves
  - D(CEWC): date of the Central Economic Work Conference
  - D(Rate Change): changes in benchmark interest rate, required reserve ratio, OMO, SLF, SLO, MLF, PSL
- Sample period: trading days during 2013–17.
- On average, the PBC communicated with the market on about 25 percent of all trading days.
  - Speeches/press conferences: 16.1 percent of days (nearly every six trading days)
  - OMO information notices: 7.8 percent of days (more frequent since February 2017)
  - MPERs: 1.6 percent of days
  - Minutes: 1.5 percent of days
- When post time is between 3:00 pm and 9:00 am the following morning, the date is adjusted to the next trading day.

### Key empirical findings (OLS and event-study)
- General result: policy instrument changes are associated with market news; PBC communication in general does not move the market much, suggesting limited effectiveness of communication.
- D(Rate Change) coefficients are significantly positive for all assets with maturities less than 3 months.
- D(PBC Communication) coefficients are significant only for the volatility of 1-week Shibor and 1-month short-term notes.
- MPERs tend to reduce the volatility of short-term notes:
  - MPERs are associated with a decline of 0.03 percentage points in the absolute change of 1-month short-term notes.
  - MPERs are associated with a decline of 0.01 percentage points in the absolute change of 3-month short-term notes.
- Oral communication (speeches and press conferences) appears to calm the market:
  - Oral communication is associated with a decrease in market news of Shibor 1week and 1-month short-term notes.
  - For the PBC’s DR007 target rate, governor communication (speeches or press conferences) is associated with 0.023 percentage points less market movement.
- Minutes releases are associated with news for repo rates (DR007) but not reflected in short-term notes.
- Oral communication is irregular and often timed by market conditions.

### Volatility and timing (EGARCH analysis)
- EGARCH(1,1) models used to explore volatility effects and asymmetric responses.
- Average PBC communication is associated with higher volatility of short-term market rates (R007).
- Speeches by the PBC governor are particularly strongly associated with higher volatility.
- Two interpretations:
  - Volatility leads the PBC to initiate ad hoc communication (communication follows elevated volatility).
  - Communication itself causes higher volatility in some cases (markets react to the content or its implications).
- EGARCH conditional variance plots for the 7-day repo market and Treasury (3 year) market show:
  - Speeches tend to be made irregularly and around periods of higher volatility (and after the liberalization of interest rate markets in 2016).
  - In some cases speeches are followed by increased volatility; in others volatility falls from elevated levels.

### OMO notices and transmission efficiency
- OMOs conducted frequently; in the sample of 1,249 trading days, the PBC conducted OMO interventions on 639 days.
- Informative OMO notices (“small notes”):
  - First two were in March 2016, another in October 2016; started regularly in February 2017.
  - There were 98 informative notices in the sample.
  - Many notices (68) were issued on days when the PBC did not conduct an OMO action and explained reasons for lack of action.
- D(OMO Rate Change): 1 if the PBC changed any repo, reverse repo, or PBC bill rates at any tenor.
  - There were 37 trading days with an OMO rate change in the sample; all associated with OMO actions on the same day; only 2 coincided with OMO informative notices.
- Empirical specification for transmission (equation (5)):
  - ∆r_t^mkt = α_1 + α_2 D(Notice Era)_t + β_1 ∆r_t^OMO + β_2 D(Notice Era)_t × ∆r_t^OMO + ε_t
  - Here, a positive and statistically significant β_2 indicates stronger transmission during the era of informative notices.
- Findings:
  - The era of issuing informative notices (regularly since February 2017) has been associated with stronger transmission of OMO rates to other market rates, strongest at the short end but present up to 1-year Shibor.
  - Analysis suggests informative OMO notices strengthened, rather than merely sped up, transmission from PBC intervention to market rates.
- Robustness considerations:
  - The Notice Era overlaps imperfectly with the October 2015 interest rate liberalization; alternative specifications (using D(OMO Notice) or restricting sample post-October 2015) yield similar results.
  - Dynamic analysis with lags (equation (6)) supports the conclusion that transparency from informative OMO notices improved transmission efficiency.

### Policy recommendations (short- to medium-term)
- Institutional changes are needed but take time; practical "low-hanging fruit" can be implemented quickly:
  - Make information available in a timely fashion, in one place, and in English.
  - Expand PBC economic forecasting capacity and publish forecasts regularly, including framework and models.
    - Note: PBC Research Bureau released economic forecasts in 2015 and 2016 (with a mid-year update) but these were discontinued in 2017.
  - Hold regular press conferences to reduce information asymmetry and help markets interpret decisions.
- These measures are useful to implement even prior to greater operational independence to build credibility and market understanding.
- Examples: the Bank of England published its Inflation Report in February 1993 prior to operational independence in June 1997.

### Conclusions
- Communication can have important benefits for monetary policy effectiveness; greater central bank transparency and independence would further improve PBC effectiveness, including via forward guidance.
- The introduction of OMO information notices reduced volatility and improved monetary policy effectiveness.
- Short-term practical measures (timely consolidated information in Chinese and English, expanded forecasting, regular press conferences) would aid monetary policy, attract capital market participation, and advance renminbi internationalization.
- In 2018, the PBC reiterated commitment to strengthen communication and transparency and has taken additional steps, including more press conferences/interviews, deeper analysis, and more informative Monetary Policy Executive Reports.

### Selected statistics and regression highlights
- Communication frequency (sample averages):
  - D(CEWC): Mean 0.001 Sd 0.001
  - D(MPER): Mean 0.015 Sd 0.015
  - D(Minutes): Mean 0.015 Sd 0.015
  - D(OMO Notice): Mean 0.038 Sd 0.038
  - D(Macro Release): Mean 0.088 Sd 0.088
  - D(Rate Change): Mean 0.116 Sd 0.116
  - D(Oral – PBC Governor): Mean 0.015 Sd 0.015
  - D(Oral Communication): Mean 0.059 Sd 0.059
  - D(PBC Communication): Mean 0.119 Sd 0.119
  - D(Oral - Other): Mean 0.044 Sd 0.044
- Market-rate means (Communication=0 vs Communication=1) examples:
  - R007: Mean 2.641 Sd 2.641 | Mean 3.046 Sd 3.046
  - DR007: Mean 2.722 Sd 2.722 | Mean 2.746 Sd 2.746
  - SHIBOR(1d): Mean 2.353 Sd 2.353 | Mean 2.460 Sd 2.460
  - SHIBOR(1w): Mean 2.925 Sd 2.925 | Mean 2.908 Sd 2.908
  - MTN(1y): Mean 4.064 Sd 4.064 | Mean 4.164 Sd 4.164
  - Treasury(1y): Mean 2.635 Sd 2.635 | Mean 2.919 Sd 2.919
  - SSE Return: Mean 0.013 Sd 0.013 | Mean 0.096 Sd 0.096
- Selected OLS coefficient highlights (Table 1.3 and 1.4):
  - D(Rate Change): e.g., column (1) 0.088** [0.014]; column (9) 0.83*** [0.001]
  - D(PBC Communication): column (2) -0.058*** [0.000]; column (3) -0.020** [0.023]
  - D(MPER): column (3) -0.030*** [0.000]; column (4) -0.011* [0.057]
  - D(OMO Notice): column (2) -0.088*** [0.000]; column (9) -0.74*** [0.000]
  - D(Oral - Zhou): column (1) -0.023*** [0.006]
  - R-squared examples: 0.063, 0.031, 0.004, 0.007, 0.006, 0.003, 0.001, 0.002, 0.019 (Table 1.3); 0.071, 0.041, 0.004, 0.008, 0.008, 0.005, 0.003, 0.007, 0.036 (Table 1.4)
- Selected EGARCH results (Table 1.5 and 1.6):
  - D(PBC Communication): conditional volatility 0.08*** (D.R007) (0.000) (Table 1.5)
  - D(Oral-PBCGOV): conditional volatility 0.17* (D.R007) (0.095), 0.74*** (D.MTN(1y)) (0.000), -0.16** (SSE Return) (0.026) (Table 1.6)
  - ARCH and persistence: L.earch and |L.earch| coefficients significant; L.egarch coefficients 0.90*** to 1.00*** (p-values (0.000)) indicating persistent volatility dynamics.
- OMO Notice × transmission (selected OLS and dynamic results):
  - D.Reverse Repo Rate×Notice Era coefficients include 1.09*** [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] (Table 1.7)
  - Dynamic results with Notice Era = Yes show large and highly significant contemporaneous coefficients for many maturities, e.g.:
    - D.R001 (Yes) 1.07*** [0.001]
    - D.R007 (Yes) 6.18** [0.039]
    - D.Shibor(1d) (Yes) 0.83*** [0.000]
    - D.Shibor(1w) (Yes) 0.36*** [0.000]
    - D.Shibor(1y) (Yes) 0.20*** [0.000] (Table 1.8)

*Source: wp18244 - Box 1. Communication Terminology (https://www.imf.org/-/media/files/publications/wp/2018/wp18244.pdf)*

### Box 1. Communication Terminology

### Box 1. Communication Terminology

### Monetary Policy Stance
- The People’s Bank of China (PBC) uses five categories to communicate policy stance: (1) moderately loose, (2) moderately tight, (3) tight, (4) prudent, and (5) prudent and neutral.
- Signaling is infrequent (usually during the Work Conference in December) and requires State Council approval.
- Scoring used to visualize relationship between stance and policy interest rates:
  - -1 for “moderately loose”
  - 0 for “prudent”
  - +0.5 for “prudent and neutral”
  - +1 for “moderately tight”
  - +2 for “tight”
- Observation: changes in language about the monetary stance in 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), but communication can lag the actual decision about a benchmark rate change.
- During a “prudent” stance, the benchmark interest rate has both increased and decreased, reflecting flexibility in the terminology.

### Window Guidance
- The PBC provides “window” guidance to financial institutions that can affect credit growth and lending to priority sectors.
- While effective, window guidance undermines transparency and reduces the role of interest rates as the main signal of monetary policy conditions.

### Communication Channels and Empirical Strategy
- Four main PBC communication formats analyzed:
  - D(MPER): quarterly Monetary Policy Executive Reports
  - D(Minutes): Monetary Policy Committee meeting minutes
  - D(Oral): press conferences and speeches by the governor, deputy governors, and the then-chief economist of the PBC Research Bureau
    - Disaggregated into D(Oral - PBCGOV) and D(Oral - Others)
  - D(OMO Notice): informative OMO notices (so-called “small notes”)
- Aggregate dummy D(PBC Communication) = 1 if at least one of the four formats occurred on a trading day.
- Market rates studied include SHIBOR, DR007, R007, STN and MTN, treasury bonds, and equity markets.
- Control variables included:
  - D(Macro Release): CPI, GDP, Official PMI and Caixin PMI, foreign trade, foreign exchange reserves
  - D(CEWC): date of the Central Economic Work Conference
  - D(Rate Change): changes in benchmark interest rate, required reserve ratio, OMO, SLF, SLO, MLF, PSL
- Sample period: trading days during 2013–17.
- On average, the PBC communicated with the market on about 25 percent of all trading days.
  - Speeches/press conferences: 16.1 percent of days (nearly every six trading days)
  - OMO information notices: 7.8 percent of days (more frequent since February 2017)
  - MPERs: 1.6 percent of days
  - Minutes: 1.5 percent of days
- When post time is between 3:00 pm and 9:00 am the following morning, the date is adjusted to the next trading day.

### Key Empirical Findings (OLS and Event-Study Results)
- General result: policy instrument changes are associated with market news; PBC communication in general does not move the market much, suggesting limited effectiveness of communication.
- D(Rate Change) coefficients are significantly positive for all assets with maturities less than 3 months.
- D(PBC Communication) coefficients are significant only for the volatility of 1-week Shibor and 1-month short-term notes.
- MPERs tend to reduce the volatility of short-term notes:
  - MPERs are associated with a decline of 0.03 percentage points in the absolute change of 1-month short-term notes.
  - MPERs are associated with a decline of 0.01 percentage points in the absolute change of 3-month short-term notes.
- Oral communication (speeches and press conferences) appears to calm the market:
  - Oral communication is associated with a decrease in market news of Shibor 1week and 1-month short-term notes.
  - For the PBC’s DR007 target rate, governor communication (speeches or press conferences) is associated with 0.023 percentage points less market movement.
- Minutes releases are associated with news for repo rates (DR007) but not reflected in short-term notes.
- Oral communication is irregular and often timed by market conditions.

### Volatility and Timing (EGARCH Analysis)
- EGARCH(1,1) models used to explore volatility effects and asymmetric responses.
- Average PBC communication is associated with higher volatility of short-term market rates (R007).
- Speeches by the PBC governor are particularly strongly associated with higher volatility.
- Two interpretations:
  - Volatility leads the PBC to initiate ad hoc communication (communication follows elevated volatility).
  - Communication itself causes higher volatility in some cases (markets react to the content or its implications).
- EGARCH conditional variance plots for the 7-day repo market and Treasury (3 year) market show:
  - Speeches tend to be made irregularly and around periods of higher volatility (and after the liberalization of interest rate markets in 2016).
  - In some cases speeches are followed by increased volatility; in others volatility falls from elevated levels.

### OMO Notices and Transmission Efficiency
- OMOs conducted frequently; in the sample of 1,249 trading days, the PBC conducted OMO interventions on 639 days.
- Informative OMO notices (“small notes”):
  - First two were in March 2016, another in October 2016; started regularly in February 2017.
  - There were 98 informative notices in the sample.
  - Many notices (68) were issued on days when the PBC did not conduct an OMO action and explained reasons for lack of action.
- D(OMO Rate Change): 1 if the PBC changed any repo, reverse repo, or PBC bill rates at any tenor.
  - There were 37 trading days with an OMO rate change in the sample; all associated with OMO actions on the same day; only 2 coincided with OMO informative notices.
- Empirical specification for transmission (equation (5)):
  - ∆r_t^mkt = α_1 + α_2 D(Notice Era)_t + β_1 ∆r_t^OMO + β_2 D(Notice Era)_t × ∆r_t^OMO + ε_t
  - Here, a positive and statistically significant β_2 indicates stronger transmission during the era of informative notices.
- Findings:
  - The era of issuing informative notices (regularly since February 2017) has been associated with stronger transmission of OMO rates to other market rates, strongest at the short end but present up to 1-year Shibor.
  - Analysis suggests informative OMO notices strengthened, rather than merely sped up, transmission from PBC intervention to market rates.
- Robustness considerations:
  - The Notice Era overlaps imperfectly with the October 2015 interest rate liberalization; alternative specifications (using D(OMO Notice) or restricting sample post-October 2015) yield similar results.
  - Dynamic analysis with lags (equation (6)) supports the conclusion that transparency from informative OMO notices improved transmission efficiency.

### Policy Recommendations (short- to medium-term)
- Institutional changes are needed but take time; practical "low-hanging fruit" can be implemented quickly:
  - Make information available in a timely fashion, in one place, and in English.
  - Expand PBC economic forecasting capacity and publish forecasts regularly, including framework and models.
    - Note: PBC Research Bureau released economic forecasts in 2015 and 2016 (with a mid-year update) but these were discontinued in 2017.
  - Hold regular press conferences to reduce information asymmetry and help markets interpret decisions.
- These measures are useful to implement even prior to greater operational independence to build credibility and market understanding.
- Examples: the Bank of England published its Inflation Report in February 1993 prior to operational independence in June 1997.

### Conclusions
- Communication can have important benefits for monetary policy effectiveness; greater central bank transparency and independence would further improve PBC effectiveness, including via forward guidance.
- The introduction of OMO information notices reduced volatility and improved monetary policy effectiveness.
- Short-term practical measures (timely consolidated information in Chinese and English, expanded forecasting, regular press conferences) would aid monetary policy, attract capital market participation, and advance renminbi internationalization.
- In 2018, the PBC reiterated commitment to strengthen communication and transparency and has taken additional steps, including more press conferences/interviews, deeper analysis, and more informative Monetary Policy Executive Reports.

*Source: wp18244 - Box 1. Communication Terminology (https://www.imf.org/-/media/files/publications/wp/2018/wp18244.pdf)*

### REFERENCES

### wp18244 - REFERENCES

### Key cited works
- Adrian, Tobias, Douglas Laxton, and Maurice Obstfeld. 2018. “An Overview of Inflation-Forecast Targeting.” In Advancing the Frontiers of Monetary Policy, edited by Adrian, Tobias; Douglas Laxton, and Maurice Obstfeld, Washington, DC: IMF.
- Berger, H., V. Nitsch, and T. Lybek. 2006. “Central Bank Boards Around the World: Why Does Membership Size Differ?” IMF Working Paper No. 2006-2281. International Monetary Fund, Washington.
- Bernanke, B. S. 2007. “Federal Reserve Communications.” Speech at the Cato Institute 25th Annual Monetary Conference, Washington, D.C., November 14. https://www.federalreserve.gov/newsevents/speech/bernanke20071114a.htm.
- Blinder, A. S., M. Ehrmann, M. Fratzscher, J. D. Haan, and D.-J. Jansen. 2008. “Central Bank Communication and Monetary Policy: A Survey of Theory and Evidence.” Journal of Economic Literature 46 (4): 910–45.
- Dincer, N. N. and B. Eichengreen. 2014. “Central Bank Transparency and Independence: Updates and New Measures.” International Journal of Central Banking 10 (1): 189–259.
- Ehrmann, M., and M. Fratzscher. 2007. “The Timing of Central Bank Communication.” European Journal of Political Economy 23 (1): 124–145.
- Gürkaynak, R. S., B. Sack, and E. Swanson. 2005. “Do Actions Speak Louder Than Words? The Response of Asset Prices to Monetary Policy Actions and Statements.” International Journal of Central Banking 1 (1).
- Hansen, S. and M. McMahon. 2016. “Shocking Language: Understanding the Macroeconomic Effects of Central Bank Communication.” Journal of International Economics 99 (S1): S114–S133.
- Hansen, S., M. McMahon, and A. Prat. 2018. “Transparency and Deliberation within the FOMC: A Computational Linguistics Approach.” The Quarterly Journal of Economics 133 (2): 801–870.
- Lucca, D. O., and F. Trebbi. 2009. “Measuring Central Bank Communication: An Automated Approach with Application to FOMC Statements.” NBER Working Paper No. w15367.
- People’s Bank of China (PBC). 2018a. “Information Disclosure and Transparency (in Chinese).” PBC, Beijing. May 23. http://www.pbc.gov.cn/zhengwugongkai/127924/128038/128109/3544192/index.html.
- Zhou, Xiaochuan. 2016. Press Conference on Financial Reform and Development during the 2016 National People's Congress March 12, 2016. http://www.pbc.gov.cn/goutongjiaoliu/113456/113469/3030450/index.html.

### Appendix 1 — Statistics summary (selected)
- Table 1.1. Statistics Summary for Communication and Control Variables
  - D(CEWC): Mean 0.001, Sd 0.001
  - D(MPER): Mean 0.015, Sd 0.015
  - D(Minutes): Mean 0.015, Sd 0.015
  - D(OMO Notice): Mean 0.038, Sd 0.038
  - D(Macro Release): Mean 0.088, Sd 0.088
  - D(Rate Change): Mean 0.116, Sd 0.116
  - D(Oral – PBC Governor): Mean 0.015, Sd 0.015
  - D(Oral Communication): Mean 0.059, Sd 0.059
  - D(PBC Communication): Mean 0.119, Sd 0.119
  - D(Oral - Other): Mean 0.044, Sd 0.044

- Table 1.2. Statistics Summary for Market Rates (Communication=0 vs Communication=1)
  - R007: Mean 2.641 Sd 2.641 | Mean 3.046 Sd 3.046
  - DR007: Mean 2.722 Sd 2.722 | Mean 2.746 Sd 2.746
  - SHIBOR(1d): Mean 2.353 Sd 2.353 | Mean 2.460 Sd 2.460
  - SHIBOR(1w): Mean 2.925 Sd 2.925 | Mean 2.908 Sd 2.908
  - MTN(1y): Mean 4.064 Sd 4.064 | Mean 4.164 Sd 4.164
  - MTN(3m): Mean 4.138 Sd 4.138 | Mean 4.080 Sd 4.080
  - MTN(6m): Mean 4.206 Sd 4.206 | Mean 4.153 Sd 4.153
  - ST CP(1m): Mean 4.016 Sd 4.016 | Mean 3.935 Sd 3.935
  - ST CP(3m): Mean 3.860 Sd 3.860 | Mean 3.980 Sd 3.980
  - ST CP(6m): Mean 3.954 Sd 3.954 | Mean 4.063 Sd 4.063
  - Treasury(1y): Mean 2.635 Sd 2.635 | Mean 2.919 Sd 2.919
  - Treasury(5y): Mean 3.228 Sd 3.228 | Mean 3.346 Sd 3.346
  - Treasury(10y): Mean 3.567 Sd 3.567 | Mean 3.559 Sd 3.559
  - SSE Return: Mean 0.013 Sd 0.013 | Mean 0.096 Sd 0.096

### Regression findings (selected coefficients and significance)
- Table 1.3. OLS Estimates for Aggregated Communication (coefficients with bracketed p-values)
  - D(CEWC):
    - column (2) coefficient -0.062** [0.011]
    - column (4) coefficient 0.069* [0.089]
    - column (5) coefficient 0.052** [0.044]
    - column (8) coefficient -0.0093*** [0.004]
    - column (9) coefficient -0.35* [0.082]
  - D(Macro Release):
    - column (1) -0.011* [0.076]
    - column (2) -0.026* [0.051]
  - D(Rate Change):
    - column (1) 0.088** [0.014]
    - column (2) 0.13** [0.011]
    - column (3) 0.035* [0.095]
    - column (4) 0.021** [0.039]
    - column (9) 0.83*** [0.001]
  - D(PBC Communication):
    - column (2) -0.058*** [0.000]
    - column (3) -0.020** [0.023]

  - Constants (all columns shown): 0.055*** [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]
  - R-squared: 0.063, 0.031, 0.004, 0.007, 0.006, 0.003, 0.001, 0.002, 0.019

- Table 1.4. OLS Estimates for Communication: By Category (selected)
  - D(CEWC):
    - column (2) -0.061** [0.014]
    - column (4) 0.069* [0.089]
    - column (5) 0.052** [0.043]
    - column (8) -0.0094*** [0.003]
    - column (9) -0.40** [0.048]
  - D(MPER):
    - column (3) -0.030*** [0.000]
    - column (4) -0.011* [0.057]
  - D(Minutes):
    - column (2) 0.15* [0.083]
    - column (8) -0.0082** [0.017]
  - D(OMO Notice):
    - column (1) 0.015* [0.065]
    - column (2) -0.088*** [0.000]
    - column (9) -0.74*** [0.000]
  - D(Oral - Zhou):
    - column (1) -0.023*** [0.006]
  - Constants (all columns shown): 0.055*** [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]
  - R-squared: 0.071, 0.041, 0.004, 0.008, 0.008, 0.005, 0.003, 0.007, 0.036

### EGARCH results (selected notable coefficients and p-values)
- Table 1.5. EGARCH Results of Aggregated Communication (columns: D.R007, D.ST CP(3m), D.MTN(1y), D.Treasury(3y), SSE Return)
  - Conditional mean:
    - Dependent Variable(t-1): 0.14*** (D.R007) (0.000), 0.18*** (D.ST CP(3m)) (0.000), 0.13*** (D.MTN(1y)) (0.000), 0.07*** (D.Treasury(3y)) (0.023), 0.02 (SSE Return) (0.447)
    - D.Reverse Repo Rate: 0.38*** (D.R007) (0.024), -0.00 (D.ST CP(3m)) (0.971), 0.02 (D.MTN(1y)) (0.772), -0.02 (D.Treasury(3y)) (0.490), 0.36 (SSE Return) (0.873)
  - Conditional volatility:
    - D(CEWC): 2.44*** (D.R007) (0.000), -0.43 (D.ST CP(3m)) (0.422), -0.26 (D.MTN(1y)) (0.603), -0.87 (D.Treasury(3y)) (0.146), 0.46*** (SSE Return) (0.009)
    - D(Macro Release): -0.43*** (D.R007) (0.000), 0.11* (D.ST CP(3m)) (0.076)
    - D(Rate Change): 0.28*** (D.ST CP(3m)) (0.000)
    - D(PBC Communication): 0.08*** (D.R007) (0.000)
  - ARCH terms:
    - L.earch: 0.25***, 0.17***, 0.08***, 0.11***, 0.01 (with p-values listed as (0.000), (0.000), (0.000), (0.000), (0.384))
    - |L.earch|: 0.36***, 0.58***, 0.30***, 0.47***, 0.20*** (p-values all (0.000))
    - L.egarch: 0.96***, 0.93***, 0.94***, 0.90***, 1.00*** (p-values all (0.000))
  - Observations: 1246, 1246, 1246, 1246, 1159

- Table 1.6. EGARCH Results by Category and People (selected)
  - Conditional volatility:
    - D(CEWC): 2.40*** (D.R007) (0.000), -0.54 (D.ST CP(3m)) (0.354), -0.13 (D.MTN(1y)) (0.834), -0.93 (D.Treasury(3y)) (0.130), 0.37* (SSE Return) (0.055)
    - D(Rate Change): 0.45*** (D.ST CP(3m)) (0.000), 0.13* (D.MTN(1y)) (0.057)
    - D(PBC Communication): 0.06*** (D.R007) (0.009), 0.09** (D.ST CP(3m)) (0.026)
    - D(Oral-PBCGOV): 0.17* (D.R007) (0.095), 0.74*** (D.MTN(1y)) (0.000), -0.16** (SSE Return) (0.026)
  - ARCH terms and observations parallel Table 1.5
  - Observations: 1246, 1246, 1246, 1246, 1159

### OMO Notice and transmission efficiency (selected results)
- Table 1.7. OLS (Notice Era interaction)
  - D.Reverse Repo Rate×Notice Era: coefficients (columns 1–12) include
    - 1.09*** [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]
  - Notice Era main effects (selected):
    - Notice Era coefficient for column (9) 0.0042** [0.011], column (10) 0.0044*** [0.000], column (11) 0.0045*** [0.000], column (12) 0.0047*** [0.000]
  - D.Reverse Repo Rate coefficients (uninteracted) are generally small and not uniformly significant; e.g., column (7) -0.15*** [0.008], column (8) -0.12*** [0.002]

- Table 1.8. OMO Notice and Transmission Efficiency: Dynamic Results (selected dynamics)
  - With Notice Era = Yes, D.Reverse Repo Rate shows large and highly significant contemporaneous coefficients in many specifications, e.g.:
    - D.R001 (Yes) 1.07*** [0.001]
    - D.R007 (Yes) 6.18** [0.039]
    - D.R014 (Yes) 1.37** [0.005]
    - D.Shibor(1d) (Yes) 0.83*** [0.000]
    - D.Shibor(1w) (Yes) 0.36*** [0.000]
    - D.Shibor(3m) (Yes) 0.63*** [0.000]
    - D.Shibor(6m) (Yes) 0.27*** [0.000]
    - D.Shibor(9m) (Yes) 0.31*** [0.000]
    - D.Shibor(1y) (Yes) 0.20*** [0.000]
  - Dynamics include significant lag coefficients (L. D. Reverse Repo Rate, L2., L3.) across many maturities with mixed signs and significance levels (p-values reported in table).

### Model fit and significance notes
- OLS R-squared values (selected): table 1.3 range from 0.001 to 0.063; table 1.4 range from 0.003 to 0.071.
- EGARCH models report L.egarch coefficients close to 1 (0.90*** to 1.00***) with p-values (0.000) indicating persistent volatility dynamics.
- Significance notation used throughout: *p < 0.1, **p < 0.05, ***p < 0.01. P-values are reported in parentheses or square brackets in tables.

*Source: wp18244 - REFERENCES (IMF working paper appendices and reference list).*

---


_Source: https://www.imf.org/-/media/files/publications/wp/2018/wp18244.pdf_