Monetary stance and favorableness of the monetary policy in the media: the case of Vietnam

Abstract

Purpose

This paper analyzes the effects of the monetary stance on the media's favorable (or otherwise) attitude to the State Bank of Vietnam's (SBV) monetary policy using monthly data from 2011 to 2021. Monetary stance is a multivariate index based on the growth rates of money supply and domestic credit. A large set of articles published in five Vietnam daily newspapers are utilized to construct a view of the media's favorableness to the monetary policy.

Design/methodology/approach

This paper uses hand-collected data from 211 articles published in five newspapers from December 2011 to September 2021 in order to examine the relationship between the monetary stance and the media's favorableness to monetary policy. Following the studies of He and Pauwels (2008) and Xiong (2012), the authors constructed a multivariate stance index to capture most of the important changes in the SBV's monetary policy stance.

Findings

The study's main findings are that a change in monetary stance from easing to neutral/tightening, or from neutral to tightening, is greatly appreciated by the media. The study's findings are robust, especially in terms of alternative measures of the media's favorableness and monetary policy variables.

Research limitations/implications

These findings have important policy implications for implementing SBV's monetary policy.

Originality/value

The main contribution of this paper is that the authors are the first to study the nexus of multivariate monetary stance and the media's favorableness to a central bank's non-inflation-targeting mandate. In particular, the study’s findings confirm that the SBV's multivariate monetary stance affects the media's favorableness, whereas the effect of inflation is statistically insignificant.

Citation

Doan, T.N., Do, D.P. and Luong, D.V. (2023), "Monetary stance and favorableness of the monetary policy in the media: the case of Vietnam", Journal of Asian Business and Economic Studies, Vol. ahead-of-print No. ahead-of-print. https://doi.org/10.1108/JABES-02-2023-0038

1. Introduction

The media's favorableness to central bank operations is a key driver of monetary policy effectiveness (Berger et al., 2011). As financial market players obtain central bank information from the news (and other) media (Hayo and Neuenkirch, 2015), the media's favorableness can enhance the implementation of the central bank policy while simultaneously boosting the interpretation of its operations. However, the concept of media favorableness is much broader than the central bank's information release. By analyzing the central bank's monetary stance, the media can also influence the understanding of the central bank policies and their economic basis with regard to the market and other players, thus affecting the economic performance. Therefore, it is important to examine the linkage between monetary stance and media favorableness.

Berger et al. (2011) are among the first to attempt to examine the driving force of media favorableness to the European Central Bank (henceforth, ECB). With respect to the monetary stance, they recognized the role of an ECB monetary policy surprise, where the interest rate decisions themselves did not affect the favorableness. Similarly, Böhm et al. (2012) showed that any change in the interest rate dampens the Czech National Bank's (henceforth, CNB) media favorableness, and that when the level of interest rates is high, the media views monetary policy decisions more adversely. Both the ECB and CNB have an inflation-targeting objective that is achieved by controlling only a single interest rate. We depart from this literature by investigating the nexus of monetary stance and the media's favorableness to the State Bank of Vietnam (SBV), in which the SBV does not adopt the inflation-targeting strategy and has a multivariate stance. Additionally, examining such a nexus in a transitional economy with an underdeveloped financial market and a low degree of information transparency—namely, Vietnam—may shed new light on the nature of the relationship between media favorableness and the monetary stance.

This study uses hand-collected data from 211 articles published in five newspapers from December 2011 to September 2021 in order to examine the relationship between the monetary stance and the media's favorableness to monetary policy. Following the studies of He and Pauwels (2008) and Xiong (2012), we constructed a multivariate stance index to capture most of the important changes in the SBV's monetary policy stance. Our results show that a change in monetary stance from easing to neutral/tightening, or from neutral to tightening, enhances the favorable attitude of the media to the SBV. After extensive robustness checks, our main findings hold for various measures of media favorableness, as well as for monetary policy variables.

The main contribution of this paper is that we are the first to study the nexus of multivariate monetary stance and the media's favorableness to a central bank's non-inflation-targeting mandate. In particular, our findings confirm that the SBV's multivariate monetary stance affects the media's favorableness, whereas the effect of inflation is statistically insignificant.

The remainder of the paper is organized as follows: Section 2 reviews the literature and develops our hypothesis. Section 3 describes our research methodology, and is followed by Section 4, which includes empirical results, and finally, Section 5 presents our conclusions.

2. Literature review and hypothesis development

2.1 Media favorableness of the monetary policy

The monetary policy plays an essential role in the development of a particular economy. According to Friedman (2000), governments implement monetary policy through central banks to control the daily economic activities of commercial banks, firms and individuals. In other words, both the general public and sophisticated audiences are affected by the monetary policy (Böhm et al., 2012). However, unlike market analysts and researchers, the general public tends to gain information on monetary policy from daily newspapers. Thus, media coverage and favorableness help orient public readers to understand and follow the monetary policy (Hayo and Neuenkirch, 2015). Therefore, effective communication of the monetary policy becomes increasingly important, and the central bank's statements indeed affect media reporting (Segawa, 2021).

Blinder et al. (2008) highlighted two main strands of literature on monetary policy in the media. The first strand focuses on the influences of monetary policy communication on the financial markets, which have drawn the most attention of researchers (Lainé, 2019). Upon analyzing 22 central banks throughout the world, Luangaram and Wongwachara (2017) noted that an informative tone of the central bank's statements is useful to predict the change in policy rate decisions. Meanwhile, by examining the Federal Open Market Committee (FOMC) guidance and comments on the current economic edition, Hansen and McMahon (2016) concluded that the shocks in guidance on the future direction have greater impacts on the financial market than the information on FOMC on the current economic situation. A similar result is observed in the UK, in which the Bank of England's reports on inflation significantly influence the financial market (Reeves and Sawicki, 2007). The second strand of the literature involves the impact of the central banks' communication on the general public's behavior. For example, Hayo and Neuenkirch (2015) documented the fact that financial agents in the UK, Japan, European countries, and the USA rely more on the news media than on the central bank's statements in dealing with changes in the macroeconomic environments. Picault et al. (2022) proved the role of the news media in controlling inflation expectations. Particularly, media favorableness or positive media sentiment toward central bank statements facilitate the long-term inflation expectations. In the Czech Republic, the decisions of the CNB that surprised financial markets were not negatively perceived by the media (Böhm et al., 2012). Moreover, the favorableness and the media coverage became more positive when the CNB changed the interest rate. Berger et al. (2011) found a similar relationship between surprising ECB decisions and media coverage in which favorableness was negatively affected when inflation outweighed the inflation target. Furthermore, in the case of ECB, the favorableness of media reports has positive impacts on explaining those surprising decisions.

2.2 Multivariate-based monetary policy stance

According to Bernanke and Mihov (1998), measuring monetary policy is important for both scientists and policymakers. Initially, researchers focused on the rate of monetary growth, which is presented in the money supply (M1, M2) or the monetary base (Xiong, 2012). For example, Sims (1972) measured the monetary policy stance through monetary supply M1. Nevertheless, a single measure indicator, such as money supply or monetary aggregate, is not an adequate signal of the monetary policy stance (Mishkin, 2007). Romer and Romer (1989) tested the approach proposed by Friedman and Schwartz (1965), which used a narrative approach to study monetary policy. However, the problems of this method are the subjectivity and divergence between the intention and actual policy actions of the monetary authorities (Dotsey and Reid, 1992). While developing this approach, Boschen and Mills (1995) built a scale measuring the FED policy each month. However, this scale is considered to lack monetary analysis (Bernanke and Mihov, 1998).

Because the monetary policy uses a complicated set of monetary tools in practice (Handa, 2008), Bernanke and Mihov (1998) sorted the non-borrowed reserves, total reserves and federal funds rate in a structural vector autoregression (SVAR) model; they defined the policy stance as a linear combination of the policy shock. SVAR has now become the workhorse model for monetary policy analysis in the USA and other developed countries. As the use of SVAR's disruption to capture monetary shocks causes discrepancies among different estimations, the application of this model in policy analysis is questioned. Additionally, in practice, the monetary policy is predicted, specifically by the Taylor rule, which is in contrast to the unpredicted focus of the SVAR. Therefore, the feedback rule derived from the SVAR-based measure could understate the role of monetary policy.

Moreover, the response of the monetary policy stance to macroeconomic variables is estimated to assess the appropriate index to characterize a central bank's behavior. This response is contingent on a monetary policy rule. Taylor (1993) proposes a quantitative study of this framework by regressing the federal funds rate on inflation and the output gap. Therefore, the Taylor rule has become the standard in the analysis of the monetary authorities' reaction in a variety of countries (Woodford, 2001). For example, Eichengreen et al. (1985) studied the Bank of England's discount rate policy under the interwar gold standard. The weekly decision as to whether to increase, cut, or leave the discount rate unchanged is formulated on the nonlinear function of reserve position, the difference between domestic and foreign interest rates, the level of economic activity, and the level of the discount rate. Gerlach (2018) used the European Central Bank's Monthly Bulletin to create the choice variable. The empirical results indicate that monetary policy responds to M3 growth, real economy status, and changes in the exchange rate, except for inflation. Kim et al. (2016) employed three types of constrained ordered choice models and established the decisive role of the output gap and exchange rate in studying the Bank of Korea's interest rate decision-making process. Mgadmi et al. (2021) examined the Taylor rule in the Tunisian context. The result was good, relative to the historical money market rate of the Tunisian central bank. Van Ommeren and Piccillo (2021) used the Taylor rule specification to explain the impact of governor changes in the interest rate setting of central banks of six OECD countries. Applying the Taylor rule, Carvalho et al. (2021) indicated that the OLS estimation could solve the inflation and output gap.

Concerning countries' economic development, Bui and Gábor (2021) revealed that researchers have paid the most attention to the monetary shock of advanced countries, whereas only a few papers have been published on emerging ones. Due to the limitation of the controlling ability of monetary authority (Jawadi et al., 2014), the avoidance of large adjustments in monetary instruments (Friedman, 2000), and the significant dependence of central banks on governments in developing countries' (Jawadi et al., 2014) interest rate and monetary supply are utilized in measuring the monetary policy stance (Bui and Gábor, 2021).

2.3 Monetary stance on media favorableness

2.3.1 Monetary policy and media favorableness

According to Hoggarth (1996), the main goal of monetary policy is to maintain a low and stable inflation rate. This is in agreement with King et al. (2008), in that the main objective of modern monetary policy is to manage inflation expectations. Nevertheless, central banks are unable to control the inflation rate directly. The instruments central banks utilize to maintain low and stable inflation rates include policy rates and reserve requirements.

Bernanke (2004) concluded that through credible media, central banks can indirectly affect the whole economy by setting the policy rate, thus achieving the final goal of monetary policies. Moreover, in terms of readers, which include both experts and non-experts, media communication helps in analyzing and anticipating central banks' future actions (Brand et al., 2006; Ehrmann and Fratzscher, 2005; Jansen and De Haan, 2006). Interestingly, the inflation index is not a significant predictor from the media's perspective despite its impact on the interest rates of an economy (Fogarty, 2005). In contrast, policy rates have attracted attention from the media (Berger et al., 2011). Depending on the changes in the policy rates and interest rates, media favorableness to these factors differs.

In terms of media favorableness, Fogarty (2005) revealed that the actions of central banks can be perceived differently by audiences as a result of different media reports. Indeed, the negative effects of central banks' policies are emphasized more than the neutral or positive impacts on the economy. Based on the work of Mutz (1992), Fogarty (2005) explained the mechanism of the perception of public readers of economic events such that non-economic experts who read newspapers tend to favor the negative side of the actions of central banks.

2.3.2 Effects of monetary stance on media favorableness

Examining the impacts of media on monetary stance, Hayo and Neuenkirch (2015) proposed the “more realistic view” model to determine the role of media coverage and favorableness. Using this model, Picault et al. (2022) concluded that the communications of presidents in ECB meetings on future stances significantly impact media favorableness. Using the hawk and dove communication strategies proposed by Tobback et al. (2017), Picault et al. (2022) observed that media favorableness decreases with dovish communications. Similarly, hawkish communications contribute to the decline of positive media sentiment. However, this study did not firmly remove the media effect. Central bank communications have been proven to be useful in transmitting the actions of central banks in easing, tightening, or maintaining a monetary policy stance (Bennani et al., 2020). However, Darvas (2018) pointed out that when central banks modify the monetary stance in the inflation-targeting economy, by tightening the monetary policy for instance, market participants tend to neglect the central banks' guidance after the systematic forecast. Therefore, upon the application of the model proposed by Hayo and Neuenkirch (2015), the perception of market participants will be affected by media coverage and sentiment. In other words, media favorableness plays an important role in forecasting and guiding markets to follow the actions of central banks. This statement proved true when Claus and Nguyen (2020) examined the expectation of customers who asymmetrically responded to changes in their monetary stance through the orientation of news. This is consistent with Lee et al. (2019), who indicated that the positive and negative tone of the news media reflects the changes in monetary stance linked with the market participants' expectations.

Based on the above discussion, we propose the following hypothesis:

H1.

There is an association between monetary policy stance and media favorableness.

2.4 The moderating role of the volatility of the exchange rate and stock market

In this part, we argue that economic volatility moderates the effect of monetary stance on the media's favorableness. When the exchange rate is more volatile, the central banks under the fixed exchange rate regime tend to actively intervene in the foreign exchange market. Additionally, the central banks make transactions in the open market operations (OMO) to offset their monetary effects, leading to a change in monetary stance. Thus, the high volatility of the exchange rate is associated with frequent changes in the monetary stance. Therefore, fluctuations in the exchange rate are a good reason to describe monetary policy (Böhm et al., 2012). Moreover, Vitale (2003) pointed out that the actual foreign exchange interventions by the central bank raise the monetary policy's transparency and credibility. Fratzscher (2008) demonstrated that communication is an effective tool for nudging the exchange rate in the desired direction, while at the same time, it may even reduce volatility.

Based on the above argument, we pose the following hypothesis:

H2.

The exchange rate volatility negatively moderates the effect of monetary policy stance on media favorableness.

The interactions between the volatility of stock markets and monetary policy have received considerable attention from scholars in economic and financial literature. On the one hand, stock prices are influenced by monetary policy such that a rise in the interest rate negatively affects stock prices. Furthermore, the distribution of information significantly impacts the stock market (Da et al., 2011; Tetlock, 2007). According to the stakeholder theory, media coverage and favorableness crucially affect investors' decision-making, thus affecting the whole stock market (Donaldson and Preston, 1995). While a positive economic outlook boosts the stock market, tightening monetary policy dampens it (Hayo et al., 2012; Su, 2018). According to the critics of the efficient market hypothesis, not all investors can gain enough information and analyze it optimally (Strycharz et al., 2018). Thus, investing decisions are normally affected by emotions, herds and irrational behavior (Nofsinger, 2005; Shiller, 2015; Strycharz et al., 2018). In that context, the media acts as an intermediary in distributing information and shaping investors' decisions in the stock market (Oberlechner and Hocking, 2004), which is significantly important in distributing firms' less accessible information to investors (Strycharz et al., 2018).

On the other hand, stock prices have a certain impact on monetary policy; central banks need to consider the monetary policy when there are stock market fluctuations (Ivrendi and Guloglu, 2012). Indeed, if the media pays more attention to and comments on a particular stock and its traded volume, the price could increase (Dougal et al., 2012; Pinnuck, 2014). As asset prices, such as stock prices, are highly dependent on market expectations and the credibility of the authority, effective communication from central banks can stabilize the stock market.

Based on the above discussion, we raise the following hypothesis:

H3.

The volatility of the stock market negatively moderates the effect of monetary policy stance on media favorableness.

3. Media favorableness

3.1 Data of media coverage

This part outlines the method used to construct the media favorableness index. The SBV issued Circular No. 35/2011/TT-NHNN regulating the disclosure and provision of information. The SBV has started a weekly report—Developments in Money Market and Inter-bank Market—which has been published since December 2011. We collected the media articles based on the issuing date of such reports. We follow Berger et al. (2011) and Böhm et al. (2012) to gather articles published within the following two days of the issuing dates. However, SBV's weekly reports focus on reporting the situation of the previous week. Thus, we used the term “base week” to present the period of time that SBV's reports aim for. As there exists a gap between the issuing date and the base week, we also use the last date of the base week. In other words, the articles will be collected if they are published within the following two days of the issuing dates and within the following two days of the Sunday of the base week. The period spans from December 2011 to September 2021. The last SBV's report in this period was issued on October 1, 2021. Thus, the last issued date of the collected articles would be October 3, 2021. Concentrating on the response of the public to SBV monetary policy decisions during a relatively long span requires some simplifications to minimize the high costs. Hence, we use the method pioneered by McCombs and Shaw (1972) for the agenda-setting function of the mass media to support the gathering of the press articles using pre-defined keywords. We employ branch coding of articles written on monetary policies and published in the columns, such as economic, finance and securities ones, of five newspapers—CaféF, Zingnews, Vnexpress, the SaigonTimes and Thoibaonganhang [1]. The choice of newspapers for the analytical coding is random and obtained from a large set of newspapers with economic/financial columns in Vietnam, without inferring that the chosen one has a higher reputation than the remainder.

Based on Doan et al. (2020) and Deephouse (2000), the information obtained for coding is based on two basic rules: (1) the name of the SBV appears immediately in the title of the article or, (2) the information about the monetary policy mentioned appears at least in five lines in the article; this refers to the cognitive threshold, implying that the information is analytically useful. The evaluation consists of three levels following common practice in media research: 1 – Negative; 0 – Neutral; 1 – Positive. Following the rules suggested by Deephouse (2000), a recording unit is labeled positive when the SBV is appreciated for its monetary policies, negative when the SBV is criticized for its actions, or neutral if the report has no assessment or a balanced one. The authors read and coded full-text versions of all sampled articles. Two colleagues were invited to apply the same coding scheme on a random sample; each one of them worked on half of the whole sample. The two raters agreed up to 90% of record units with the authors, thus implying high intercoder reliability.

3.2 Media favorableness

Our database includes 211 articles published in five newspapers from December 2011 to September 2021. We first distinguish how positively or negatively each report was covered by the five different newspapers. The favorableness indicator for each report and each newspaper is computed as the simple average of the favorableness for all articles published in that newspaper after that report. Given the fact that several time series used in this research are unable to be collected at weekly frequency, we construct monthly favorableness from weekly favorableness as follows:

(1)Favort=∑i=1w(∑m=15Favort,i,m),

where Favort captures the monthly favorableness in month t, while Favort,i,m represents the favorableness of newspaper m in week i of month t. w is the number of weeks observed in month t. As the SBV did not report in some weeks, the value of w is not constant. As done by Janis and Fadner (1965) and Deephouse (2000), we compute the media favorableness as follows:

(2)Favort,i,m={(pt,i,m2−pt,i,mnt,i,m)/(pt,i,m+nt,i,m)2 if pt,i,m>nt,i,m0 if pt,i,m=nt,i,m(pt,i,mnt,i,m−nt,i,m2)/(pt,i,m+nt,i,m)2 if pt,i,m<nt,i,m,

where pm,i,t and nt,i,m are the number of positive and negative recording units for monetary policies published in newspaper m, following the report of week i in month t, respectively. We report the average value of Favor in each newspaper in Figure 1 (Online Appendix). Two points are worth highlighting. First, the favorableness indices based on the issuing date are likely to be higher than those of the last date. Second, the favor index based on the CaféF newspaper is much higher than that of the others. Also, the result of CaféF is unpredicted. According to Nguyen-Pochan (2021), CaféF is considered the source of “news aggregators”. Moreover, CaféF is run by a private media firm, VCCorp. Thus, there is no evidence to prove that the high favorableness index results from a close political relationship. Figure 2 (Online Appendix) displays the distribution of favorableness over time. It can be seen that the favorableness index varies over time, and it tends to receive a positive value post-2018.

3.3 Monetary policy stance

The construction of the policy index requires us to categorize the change in each monetary series. We follow most of the literature on the monetary policy stance index to assign value in a set of three choices (1; 0; −1) to represent a tightening change, no change, and an easing change in the policy reactions, respectively (Bernanke, 1990; He and Pauwels, 2008; Xiong, 2012). We filter the M2 and domestic credit growths by employing the Hodrick-Prescott method and by keeping the cyclical components. All the growth rates are calculated as the percentage change during the past 12 months unless otherwise stated. One standard deviation of each series was used as a threshold to classify the fluctuations. We mark tightening (easing) change for any cyclical component that decreases (increases) more than one standard deviation. The rest is treated as unchanged one.

After obtaining the policy change of all variables, we sum them up before constructing the overall monetary policy stance change index, denoted by Stanceyt. Similar to the single index of each policy tool, we assign a value to Stanceyt in a set of three choices (1; 0; −1) to represent a tightening change, no change, and an easing change, respectively, as follows:

(3)Stanceyt={1 if total sum is greater than 00 if total sum is equal to 0−1 if total sum is less than 0,

For further analysis, we create dummy variables, Tightening and Easing, that take the value of one if Stance is equal to 1 and –1, respectively, and 0 otherwise. Figure 3 (Online Appendix) displays the distribution of Stance, Tightening and Easing as well as the growth rate of money supply (M2G) and domestic credit (DCredit) over time. In general, both M2G and Dcredit go together, but in some periods, the two series are diverged. Hence, we use both series to capture these periods. As the SBV keeps its stance for long-enough periods, the monetary stance of SBV's policy persists.

3.4 Model specification

We follow the literature on the determinants of media favorableness (Berger et al., 2011; Böhm et al., 2012; Lyócsa et al., 2019) to specify the model as follows:

(4)Favoryt=α0+λt+β1Stanceyt+β1CONTROLyt+εyt,

where subscripts y and t denote year and month, respectively. Here, λt captures the time-fixed effect and CONTROLit is a set of control variables. To control the effects of economic status, we use inflation (Inflation) and growth rates of the manufacturing product (MPG) and retail sale (RSG). The external sector consists of the trade balance (TradeBalance) and the index of the exchange rate (ERIndex). Table 1 (Online Appendix) reports the correlation matrix, whereas Table 2 (Online Appendix) portrays the statistical summary. As the correlation coefficients are less than 0.8, the multicollinearity problem may not appear in our estimations (Hair et al., 2010). Additionally, the last column of Table 2 (Online Appendix) indicates that all variables are stationary.

4. Empirical results

4.1 Baseline results

Table 3 (Online Appendix) reports the baseline results of the regression on equation (3). While the first column has no fixed effects, we add the month-fixed effects in columns (2)–(4). Concerning the monetary stance index, the coefficient of Stance is positive and statistically significant in all model specifications. More precisely, when the stance index increases by one unit, it improves the favorableness index by 0.4 and 0.43 units in columns (1) and (2), respectively. This finding aligns with that of Böhm et al. (2012) in the sense that when the central bank implements a contractionary policy, the public media tends to perceive monetary policy decisions more positively. Particularly, as reducing money growth or domestic credit is a signal to curb inflation, especially in periods of high inflation, such a policy corresponds to the public's macroeconomic expectations. In addition, given that Vietnam's financial market has yet to be developed, and the banking channel is the most important channel to access external capital, SBV is highly appreciated by the media for the contractionary policy when commercial banks are thought to be capital constrained. This result supports our hypothesis H1.

Turning to other variables, the depreciation of the domestic currency boosts the public preference for monetary policy. Our finding is consistent with Böhm et al. (2012). However, the coefficient of TradeBalance is statistically insignificant. Meanwhile, the change in the headline inflation rate does not affect the favorable perception of monetary policy. Similarly, the effects of manufacturing production growth rate (MPG) and retail sale growth rate (RSG) are mute.

To see the role of each stance dimension, we use Tightening and Easing instead of Stance, and re-regress equation (3). Results reported in columns (3) and (4) indicate that only Easing matters. This suggests that the public media appreciates the expansionary policy.

4.2 Further analysis

In this part, we use the favorableness index based on the issuing date (Favor_Iss). The result in column (1) of Table 4 (Online Appendix) is consistent with our baseline. However, it can be seen that the effects of Stance on Favor_Iss are much lower than those on Favor_Last. One possible reason for this is that when the issuing date is far behind the last date of the base week, the link between monetary stance and media favorableness weakens. Next, we replace Stance with policy rate (PolicyRate), growth rates of money supply (M2G) and foreign reserves (ReservesG). The result reported in column (2) of Table 4 (Online Appendix) shows that the coefficients of M2G and PolicyRate are negative and statistically significant, which reinforces our main finding.

Next, we consider the moderating role of the exchange rate and stock exchange index. Volatility is the ratio of cyclical components obtained from the Hodrick-Prescott filter to its level. In each measure, we first add the volatility into equation (1), and then add its interaction term with the monetary policy stance. The estimation results are displayed in Table 5. It can be seen that the volatility of exchange negatively affects media favorableness, whereas that of the stock exchange rate index is insignificant. However, only the interaction term is significant in column (4) of Table 5, while the main coefficients are not; thus, there seems to be an amplification between Stance and Volatility, and easing seems to set off more volatility and influence favorableness mostly through this channel. Another reason is that there is a marginal effect on specific values. This means that if the stock exchange rate is volatile at the same time as the changes in the monetary stance, the media favorableness is negatively impacted. However, there is no evidence for the moderating role of the volatility of the stock exchange index in the effects of the monetary stance on favorableness. Put them together, volatility is not shown to be moderating the effect of the monetary stance on favorableness.

5. Conclusion

In this paper, we have examined the effect of the monetary stance on the variability of the favorableness of the SBV's monetary policy decisions in the media in the period 2011–2021. We use a rich set of articles published in the five most relevant Vietnamese daily newspapers. We construct a multivariate stance index based on the growth rates of M2 and domestic credit by using the Hodrick-Prescott filter. We find that monetary stance affects the SBV's media favorableness in the sense that a change from easing to neutral/tightening or from neutral to tightening is highly appreciated by the general public. Meanwhile, the depreciation of the Vietnamese dong receives positive reporting, while the effects of inflation and the growth rate of manufacturing production and retail sale are statistically insignificant.

Our findings suggest several policy implications. Changes in monetary stance and domestic currency's depreciation are key drivers of media favorableness, whereas inflation plays no role. The SBV should not be too smooth in managing the growth rate of money supply and domestic credit. Additionally, the SBV can depreciate the Vietnam dong to appeal to the general public.

Table 5

Moderating role of volatility

	(1)	(2)	(3)	(4)
	Volatility_ERIndex		Volatility_StockIndex
Variables	Favor_Last	Favor_Last	Favor_Last	Favor_Last
D.Inflation	0.08	0.09	0.15	0.13
	(0.134)	(0.136)	(0.138)	(0.131)
MPG	0.01	0.01	0.01	0.01
	(0.021)	(0.021)	(0.025)	(0.023)
RSG	0.01	0.01	−0.00	0.00
	(0.013)	(0.013)	(0.014)	(0.013)
TradeBalance	−0.13	−0.13	−0.09	−0.10
	(0.131)	(0.129)	(0.143)	(0.131)
ERIndex	0.13***	0.13***	0.11***	0.10***
	(0.020)	(0.019)	(0.019)	(0.019)
Stance	0.41**	0.41**	0.39*	0.30
	(0.183)	(0.187)	(0.200)	(0.203)
Volatility	−0.31***	−0.26**	0.01	−0.01
	(0.116)	(0.130)	(0.012)	(0.011)
Stance* Volatility		0.12		−0.04***
		(0.176)		(0.015)
Constant	−18.11***	−17.34***	−14.98***	−13.87***
	(2.808)	(2.645)	(2.626)	(2.605)
Observations	115	115	115	115
R-squared	0.441	0.444	0.406	0.464
Month FE	YES	YES	YES	YES

Note(s): Robust standard errors in parentheses

***p < 0.01, **p < 0.05, *p < 0.1

Source(s): Authors' own work

Note

1. The title of the newspapers in Vietnamese are, respectively, Café F, Zingnews, Vnexpress, Kinh tế Sài Gòn, and và Thời báo Ngân hàng.

The supplementary online appendix

The supplementary material for this article can be found online.