key: cord-1027905-1hc9s9g8 authors: Funke, Michael; Tsang, Andrew title: The People’s Bank of China’s Response to the Coronavirus Pandemic: A Quantitative Assessment() date: 2020-08-31 journal: Econ Model DOI: 10.1016/j.econmod.2020.08.018 sha: 16ae26efee7fc51b8e37c647c9953afc88625412 doc_id: 1027905 cord_uid: 1hc9s9g8 The People’s Bank of China (PBoC) has implemented numerous measures to cushion the impacts of the COVID-19 health crisis on the Chinese economy. Since the current monetary policy framework features a multi-instrument mix of liquidity tools and pricing signals, we employ a dynamic-factor modelling approach to derive a composite indicator of China’s monetary policy stance. Our quantitative assessment shows that the PBoC’s policy response to the outbreak of the COVID-19 pandemic has been swift and decisive. Specifically, our estimates reveal that the PBoC has implemented novel policy measures to ensure that commercial banks maintain liquidity access and credit provision during the COVID-19 crisis. In December 2019, respiratory illness clusters caused by a novel coronavirus (SARS-COV-2) emerged in Wuhan, the capital of Hubei Province, China. The World Health Organization (WHO) named the disease COVID-19. The pandemic quickly spread from Wuhan to the rest of mainland China: As of July 16, 2020, official statistics show that COVID-19 had caused a cumulative number of 83,622 infections and 4,634 fatalities in China. 1 The need for social distancing and lockdowns has caused substantial collateral economic damage. COVID-19 initially caused a negative supply shock by forcing firms to shut down, which disrupted international supply chains. Moreover, the pandemic, through its negative impact on agent expectations of future income growth, induced a demand-driven recession. Weak aggregate demand, in turn, depressed the incentives of firms to invest. The massive spike in uncertainty added wait-and-see responses by consumers and firms to a shaky world economy, and Against the background of this difficult economic situation, this paper analyzes and quantifies the monetary policy response of the People's Bank of China (PBoC) to the COVID-19 crisis by employing a dynamic-factor modelling framework. Our strategy represents a refinement of the approach laid out by Funke and Tsang (2019) , and includes the PBoC's numerous earmarked (and sometimes arcane) monetary stimulus tools. Given China's prominence in the world economy, Beijing's economic policy course is followed closely. 1 For the classic epidemiological model of epidemic dynamics and its domestic and international spread, see Allen (2017) , Balcan et al. (2010) , Chinazzi et al. (2020) and Wu et al. (2020) . 2 The economic literature on the COVID-19 pandemic is rapidly expanding. For the various facets of the COVID-19 shock and possible policy responses, see Weder di Mauro (2020a, 2020b) ; Alvarez et al. (2020) ; Atkeson (2020) ; Baker et al. (2020) ; Berger et al. (2020) ; Caballero and Simsek (2020) ; Coibion et al. (2020) ; Eichenbaum et al. (2020a Eichenbaum et al. ( , 2020b ; Fornaro and Wolf (2020) ; Guerrieri et al. (2020) ; Krueger et al. (2020) ; and Lewis et al. (2020) . Maliszewska et al. (2020) ; Pindyck (2020) ; and Stock (2020) study the macroeconomic and trade effects and shock transmission of the COVID-19 pandemic. For historical lessons from previous pandemics, see Barro et al. (2020) ; Greenwood et al. (2019) ; Jordà et al. (2020) ; and Velde (2020) . Leiva-Leon et al. (2020) have developed a nowcasting COVID-19 global recession risk indicator. None of these papers focus on China. Several studies have examined the course of China's monetary policy and/or Chinese monetary policy shocks, including Girardin et al. (2017) ; Kamber and Mohanty (2018) ; McMahon et al. (2018) ; and Sun (2015 Sun ( , 2018 . These papers use diverse empirical methodologies to estimate China's monetary policy stance, and obtain results quite consistent with ours. 4 A distinctive contribution of our paper is that we provide evidence for the PBoC's policy response to the pandemic-induced 2020 downturn. The estimated nowcasting indicator of Chinese monetary policy presented here is thus unrivaled in the literature. The paper proceeds as follows. The economic impact of COVID-19 on the Chinese economy and the PBoC's monetary policy response are described in Section 2. In Section 3, we present our dynamicfactor modelling framework. Section 4 reports the estimation results and evaluates China's monetary policy response, and Section 5 presents our conclusions and policy implications. The Chinese government's distancing policies aimed at containing infections and saving lives prevented firms from operating (triggering a supply-side recession) and consumers from consuming (triggering a demand-side recession). In other words, the flattening of the infection curve inevitably steepened the macroeconomic recession curve. While this collateral damage was quite predictable, the extreme speed at which the crisis unfolded was unforeseen. Carmen Reinhart Figures 1 and 2 reveal that the COVID-19 shock quickly cascaded through the economy, morphing into an unparalleled downturn simultaneously impeding demand and supply (slumps of industrial production, retail sales, the purchasing manager indices, the business confidence indices, fixed asset investment and foreign trade in first three months of 2020). In the 2 nd quarter the Chinese economy then started to rebound, albeit rather uneven as the supply recovery is stronger than demand, and investment is stronger than consumption. Retail sales in particular are still rather subdued, as people continue to practice social distancing. The only bright spot is online retail. This pattern is also apparent in GDP growth. Figure 2 shows a historic contraction in the first quarter, with Chinese GDP shrinking 6.8% from the same period a year ago. GDP then reverted to expand 3.2% in the second quarter from a year ago. In the first half however, China's GDP still declined by 1.6% year on year. That is all to say: China's rebound from the pandemic lockdown is impressive, but is not yet back to normal. For policymakers, the biggest surprise has been the depth of slump. The data for January and February 2020 show that real industrial output and nominal retail sales were down by 13.5% and 20.5%, respectively. The official seasonally adjusted purchasing manager index (PMI) and the OECD business confidence index for China fell to its lowest level since the global financial crisis of [2008] [2009] . Within the survey indices, the non-manufacturing sectors suffered the greatest losses. These three expectation-driven indicators suggest extreme stress and illustrate that market participants were increasingly worried that the public health crisis would become a broad-reaching financial crisis. Meanwhile, the seasonally adjusted RWI/ISL Container Throughput Index dropped by 9.9 points in February -the largest monthly decline ever recorded. The driving force was the decline in sharp This highlights the challenges for policymakers posed by the pandemic. While the immediate impact of the public health crisis can be observed, the medium and longer-term effects are difficult to predict. Will China experience a short-lockdown, quick-snapback V-shaped recovery, or will the coronavirus lead to an anemic rebound that looks like a U? While the V-shaped recession has a pointed trough, troughs are more elevated and prolonged in U-shaped recoveries. How much of the economic damage wrought by the temporary shutdown will last even after Chinese firms reopen? There are also potential second and third waves of infection to consider. Domestic and global infections still pose a threat that could well trigger further waves of infection in China that obviates containment measures and lockdowns. This would cause a double-dip, or W-shaped recession and recovery. An ex-ante estimate of the severity and length of the pandemic is challenging to say the least. Amidst the considerable uncertainties, the PBoC took a number of policy measures designed to combat the economic repercussions of the pandemic. The chronological sequence of Chinese monetary policy measures introduced from 31 January -16 July 2020 are summarised in Table 1 . 3.2 PBoC injects RMB 1.2 trillion in liquidity into the banking sector through reverse repo operations, (lowering repo rates by 10 basis points). PBoC injects RMB 500 billion in liquidity into the banking sector through reverse repo operations. PBoC announces plans to support bond issuance by financial institutions for epidemic prevention and control. PBoC announces plans to set up RMB 300 billion in special central bank lending (relending) to provide low-cost funds for banking lending supporting epidemic prevention and control. The central government commits to subsidizing 50% of business interest payments to ensure actual financing costs below 1.6%. PBoC injects RMB 900 billion in liquidity into the banking sector through reverse repo operations. PBoC injects RMB 100 billion in liquidity into the banking sector through reverse repo operations. PBoC injects RMB 100 billion in liquidity into the banking sector through reverse repo operations, as well as RMB 200 billion in medium-term (1-year) liquidity through MLF operations (with 10-bp cut in MLF rate). PBoC lowered 1-year loan prime rate by 10 bps and 5-year loan prime rate by 5 bps. State Council decides to increase the PBoC's relending and rediscount quota by RMB 500 billion for bank lending to support SMEs, as well as lower relending rate by 25 bps to 2.5%. State Council decides to guide financial institutions to issue extra-low-interest loans with a quota of RMB 300 billion for self-employed businesses. At first glance, we see the PBoC has unveiled an unprecedented set of measures intended to ensure China's commercial banks maintain liquidity access and credit provision during the COVID-19 crisis. The chronological sequence in Table 1 further shows that the Chinese monetary policy response was not a one-off reaction, but a successive series of easing actions. Since some of the listed open market policy measures are regular and limited in duration, it is interesting to determine the extra liquidity triggered by the pandemic crisis. The COVID-19 gross liquidity injection of all operations appears to be on the order of RMB 8 trillion for the February-June period. 8 The following section provides a quantitative assessment of the monetary policy measures taken. Judging the overall monetary policy stance in China at any given point in time is difficult due to the vast number of measures which operate through numerous channels. In light of this fact, a dynamic factor model is presented below to gauge the overall stance on monetary policy in China, which the methodology allows us to use a single model with incorporating dimension reduction and variable selection to construct a single and intuitive indicator. In specifying the dynamic factor model, it must be borne in mind that China's monetary policy toolkit has evolved over time. Before diving into the factor modelling approach, we therefore briefly review China's monetary policy reforms in recent years and the resulting contours of the current Chinese monetary policy landscape. In the last decade, the PBoC has upgraded its monetary policy framework to a multiple The amount of the targeted MLF stimulus measures is included in the variable Net OMO withdrawal/total loans (t -1). Table 2 summarizes these five variables and their descriptive statistics. Dynamic factor models are used in applied time series econometrics incorporating unobserved variables. Such models are particularly valuable in nowcasting the state of an economy. As with many useful empirical modelling approaches, factor models are extrapolated from data rather than being deduced from theory. 10 A further development of the model in Funke and Tsang (2019) which includes the novel and specific and earmarked monetary policy measures in the COVID-19 pandemic is used below. The 10 This article is not the venue for a complete review of the factor modelling approach. For worth reading contributions to the theory of dynamic factor models, see Forni et al. (2000 Forni et al. ( , 2004 Forni et al. ( , 2005 and Stock and Watson (2002a, 2002b) . Against the background of possible pandemic-related structural breaks, it is noteworthy that factor models are rather robust to parameter instabilities. See Stock and Watson (2009, pp. 1-57) and Bates et al. (2013) . For comprehensive textbook treatments see Harvey (1989) and Durbin and Koopman (2012) . variable. The number of common factors is determined using the test procedure of Bai and Ng (2002) . The upshot is that a single common factor is appropriate for the data-generating processes. As a result, our dynamic factor model in first differences is specified as follows: where ∆ is the first-difference operator, ‫ݔ‬ ௧ is the unobserved common component at time t, ‫ݕ‬ ሺ݅ = 1, ⋯ ,5ሻ are the five monetary policy instruments, ߚ are the factor loadings, ‫ݑ‬ ௧ ~ iid N(0, ߪ ௨ ଶ ), and ‫ݒ‬ ,௧ ~ iid N(0, ߪ ଶ ). Furthermore, the error terms are orthogonal. The AR(2) lag structure in equations (2) and (3) has been chosen to ensure the iid properties of the residuals. Each monetary policy indicator ‫ݕ‬ ,௧ is demeaned and first-differenced, while the differenced series is assumed as a weakly stationary process that has at least finite second-order moments. In the state-space representation, the measurement equation is written as The estimation procedure consists of a sequence of four steps. First, the maximum likelihood estimation method is employed to estimate the parameters of the dynamic factor model in equations (4) and (5). Second, the current state of the unobserved common factors is obtained by applying the Kalman filter. The Kalman filter is a recursive procedure for computing the optimal estimate of the unobserved state vector ߚ based on the appropriate information set. The algorithm works in a two-step process. In the prediction step, the Kalman filter produces estimates of the current state variables, along with their uncertainties. Once the outcome of the next measurement is observed, these estimates are updated using a weighted average, with more weight being given to estimates with higher certainty. Third, the monetary policy stance time series ( ‫ݔ‬ ௧ ) is calculated by accumulating the estimated series ‫ݔ∆‬ ௧ , assuming the initial value ‫ݔ‬ = 0. Finally, the monetary policy stance time series is rescaled to the range from -2 to +2. 11 The maximum-likelihood estimation results of the dynamic factor models are given in Table 3 policy measures play a major role. One reason for this is the lower interest rate level that has already been achieved. In the period September to December 2008, the 1-year benchmark lending rate dropped from 7.47% to 5.31% (215-bp cut), while the 1-year loan prime rate was reduced by 30-bp from 4.15% to 3.85% in the first half-year 2020. In both crises the RRR was lowered. In the period September to December 2008, the reserve ratios for small and medium-sized banks (large banks) were reduced by 4 (2) percentage points. The RRR cut released about RMB 1,000 billion. Since February 2020, the overall Another feature is the sequential step-by-step approach since the beginning of 2020 which is nicely illustrated by the index. This incremental approach may be due to the declining number of new domestic COVID-19 infections. Although concerns about a second wave of infection persist, the initial assessment is that the economic impact of the pandemic has been a sharp, but short, recession. It will be followed by a V-shaped recovery with a return to normal in the second half of 2020 and growth accelerating in 2021. The supply-side recovery shown in Figure 13 It is noteworthy that China's fiscal policy response to the pandemic has so far been relatively restrained by international standards. In contrast, many advanced economies have implemented substantial fiscal measures according to the principle "there is no time to lose". Corporate bailouts have been a core element. However, it must be borne in mind that China may not need such unprecedented fiscal parachutes as they are already built into the system. The predominant share of corporate loans go to state-owned firms and hence already enjoy implicit government guarantees. See the IMF policy action tracker at https://www.imf.org/en/Topics/imf-and-covid19/Policy-Responses-to-COVID-19. Understanding the COVID-19 policy reaction of China is critical. The sheer size of the Chinese market and China's integration into the global supply chains has made it a significant driver of global growth. For this reason, the economic policy response in China and the country's economic development play a key role in the global economic recovery. 14 Against this background, this paper analyses the PBoC's monetary policy response to counter the COVID-19 pandemic and quantifies it using a dynamic factor model. J o u r n a l P r e -p r o o f A Primer on Stochastic Epidemic Models: Formulation, Numerical Simulation, and Analysis A Simple Planning Problem for COVID-19 Lockdown What will be the economic impact of COVID-19 in the US? Rough estimates of disease scenarios Determining the Number of Factors in Approximate Factor Models COVID-Induced Economic Uncertainty Modeling the Spatial Spread of Infectious Diseases: The Global Epidemic and Mobility Computational Model Economics in the time of COVID-19, CEPR Press VoxEU.org eBook Mitigating the COVID Economic Crisis: Act Fast and Do Whatever It Takes The Coronavirus and the Great Influenza Pandemic: Lessons from the 'Spanish Flu' for the Coronavirus's Potential Effects on Mortality and Economic Activity Consistent Factor Estimation in Dynamic Factor Models with Structural Instability An SEIR Infectious Disease Model with Testing and Conditional Quarantine A Model of Asset Price Spirals and Aggregate Demand Amplification of a 'Covid-19' Shock To Guide or Not to Guide? Quantitative Monetary Policy Tools and Macroeconomic Dynamics in China The Effect of Travel Restrictions on the Spread of the 2019 Novel Coronavirus (COVID-19) Outbreak The Cost of the Covid-19 Crisis: Lockdowns, Macroeconomic Expectations, and Consumer Spending The Transition of China to Sustainable Growth -Implications for the Global Economy and the Euro Area Time Series Analysis by State Space Methods The Macroeconomics of Epidemics The Macroeconomics of Testing and Quarantining Covid-19 Coronavirus and Macroeconomic Policy The Generalized Dynamic Factor Model: Identification and Estimation The Generalised Dynamic Factor Model: Consistency and Rates The Generalized Dynamic Factor Model: One Sided Estimation and Forecasting The Direction and Intensity of China's Monetary Policy Conduct: A Dynamic Factor Modelling Approach From the Middle Kingdom to the United Kingdom: Spillovers from China China's Evolving Monetary Policy Rule: From Inflation-Accommodating to Anti-Inflation Policy An Equilibrium Model of the African HIV/AIDS Epidemic Macroeconomic Implications of COVID-19: Can Negative Supply Shocks Cause Demand Shortages? Forecasting People's Republic of China -Selected Issues Longer-Run Economic Consequences of Pandemics Do Interest Rates Play a Major Role in Monetary Policy Transmission in China? Projecting the Transmission Dynamics of SARS-CoV-2 Through the Postpandemic Period Macroeconomic Dynamics and Reallocation in an Epidemic Real-Time Weakness of the Global Economy: A First Assessment of the Coronavirus Crisis U.S. Economic Activity During the Early Weeks of the SARS-COV-2 Outbreak Rebalancing China's Economy: Domestic and International Implications The Potential Impact of COVID-19 on GDP and Trade: A Preliminary Assessment The Case for Rules in the Conduct of Monetary Policy: A Concrete Example China's Monetary Policy Communication: Frameworks, Impact, and Recommendations COVID-19 and the Welfare Effects of Reducing Contagion This Time Is Different: Eight Centuries of Financial Folly Data Gaps and the Policy Response to the Novel Coronavirus A Probability Model of the Coincident Economic Indicators Forecasting Using Principal Components from a Large Number of Predictors Macroeconomic Forecasting Using Diffusion Indexes Forecasting in Dynamic Factor Models Subject to Structural Instability What Measures Chinese Monetary Policy? A Narrative Indicator of Monetary Conditions in China Evaluating McCallum rule as a policy guideline for China What Happened to the US Economy During the 1918 Influenza Pandemic? A View Through High-Frequency Data Nowcasting and Forecasting the Potential Domestic and International Spread of the 2019-nCoV Outbreak Originating in Wuhan, China: A Modelling Study J o u r n a l P r e -p r o o f