Selected Issues

Abstract

Selected Issues

Japanese Business Cycles, External Shocks, and Spillovers1

This chapter develops a quarterly macro-econometric model for Japan, and integrates it within a compact model of the world economy (including the global oil market). A Global Vector Autoregression (GVAR) model is used to evaluate the nature and strength of: economic linkages between globally-systemic countries (including Japan); the size and speed of the international transmission of stress in global financial markets; as well as a global growth slowdown. It also examines outward spillovers from Japan to the rest of the world and how they have changed over time. Spillovers are transmitted across economies via trade, financial, and commodity price linkages. The results show that all regions are more sensitive to developments in China than to output shocks in the Euro Area, the United States, or Japan. Stress in global financial markets can amplify spillovers from growth shocks in systemic economies. While outward spillovers from Japan output shocks have become smaller over time, they remain important for Asia-Pacific economies and the global economy.

A. Introduction

1. A Global Vector Autoregression (GVAR) model is used to determine the size and speed of the transmission of different shocks to/from Japan. We use a dynamic multi-country framework for the analysis—see Cashin et al. (2014), Cashin et al. (2016), Cashin et al. (2017b), Mohaddes and Pesaran (2016), and Mohaddes and Raissi (2018a). The framework comprises 33 country/region-specific models. These individual models are solved in a global setting where core macroeconomic variables of each economy are related to corresponding foreign variables (constructed exclusively to match the international trade pattern of the country under consideration). The model has both real and financial variables: real GDP, inflation, primary balance to GDP ratio, public debt to GDP ratio, real exchange rate, short and long-term interest rates, an index of financial market stress, and the price of oil. This framework can account for various transmission channels, including not only trade relationships but also financial and commodity price linkages—see Dees et al. (2007). All data are quarterly in frequency, for the period 1981Q2 to 2018Q2.

2. The results show that output shocks emanating in globally-systemic economies have important cross-country effects, including for the Japanese economy. Following a one percent decline in China’s GDP, economic activity falls by about 0.3 percent for the median Asian economy (and about 0.25 percent for Japan) and 0.2 percent for the median economy in Europe and the Americas. The corresponding number for the fall in GDP of the median Asian economy in response to a similar growth slowdown in the United States is about 0.2 percent after one year. Adverse spillovers from the Euro Area slowdown are modest (assuming limited financial stress in the Euro Area). These results show that countries are becoming more sensitive to developments in China than to shocks in the Euro Area or the United States, in line with the direction of evolving trade patterns and the emergence of China as a key driver of the global economy. In response to a financial stress shock, real GDP growth slows worldwide (by 0.2 percentage points on average). Finally, while outward spillovers from Japan output shocks have reduced over time, they do have global implications (albeit to a lesser extent than those of China and the United States), and are stronger in its Asia-Pacific geographical proximity.

B. Modelling the Global Economy

3. We employ the GVAR methodology to analyze the international macroeconomic transmission of shocks, with a focus on the Japanese economy. This framework takes into account both the temporal and cross-sectional dimensions of the data; real and financial drivers of economic activity; interlinkages and spillovers that exist between different regions; and the effects of unobserved or observed common factors (e.g. commodity prices and financial risk). This is crucial as the impact of shocks cannot be reduced to a single country but rather involves multiple regions, and this impact may be amplified or dampened depending on the degree of openness of the countries and their trade structure. Before describing the data and our model specification, we provide a short exposition of the GVAR methodology.

The Global VAR Methodology

4. The Global VAR methodology consists of two main steps. First, each country is modeled individually as a small open economy (except for the United States) by estimating country-specific vector error correction models in which domestic variables are related to both country-specific foreign variables and global variables that are common across all countries (such as oil prices and financial market stress). Second, a global model is constructed combining all the estimated country-specific models and linking them with a matrix of predetermined cross-country linkages. More specifically, we consider N + 1 countries in the global economy, indexed by N = 0,1,…,N. Except for the United States, which we label as 0 and take to be the reference country; all other N countries are modelled as small open economies. This set of country-specific models is used to build the GVAR framework.

GVAR Literature and Spillovers

5. The GVAR is a modelling framework of the world economy designed to explicitly model economic and financial interdependencies across markets and countries at national and international levels. The framework was originally proposed by Pesaran et al. (2004) and further developed by Dees et al. (2007). It links individual country-specific models in a coherent manner to form a global modelling framework by using time series, panel data, and factor analysis techniques. It has been used in bank stress testing; the analysis of China’s growing importance for the rest of the world economy (Cashin et al. (2016, 2017b)); the international macroeconomic transmission of weather shocks (Cashin et al. 2017a); the impact of commodity price shocks (see Mohaddes and Pesaran (2016) for the global macroeconomic consequences of country-specific oil-supply shocks and Cashin et al. 2014 for the differential effects of demand- and supply-driven commodity price shocks); other real and financial shocks, as well as in forecasting applications. For an extensive survey of developments in GVAR modelling, see Chudik and Pesaran (2016).

6. The issue of spillovers from financial and growth shocks in systemic economies has received extensive attention in the literature. Important papers include coverage in the IMF’s Spillover Reports (IMF 2012, 2014a) and Regional Economic Outlooks for Asia and the Pacific (IMF 2014b, 2015, 2016). Cashin et al. (2017b) and Dizioli and others (2016) find that a slowdown and rebalancing in China can have significant spillovers to ASEAN-5 economies, particularly those with higher trade exposures to China and commodity exporters. They also show that a surge in global financial market volatility could translate into a fall in world economic growth of around 0.3 percentage points after one year. Cashin et al. (2016) and Kose et al. (2017) study spillovers from the U.S. growth slowdown. Dees et al. (2007) explore the international linkages of the Euro Area. This chapter contributes to the literature in several dimensions. First, the price of oil in the model is determined in international markets, thereby being affected by both global demand and supply conditions. An index of stress in global financial markets is added to the model, to account for the financial contagion channel of impact. Second, the chapter investigates the extent to which the global impact of a slowdown in Japan has changed over the past three decades. Third, the estimation sample of the GVAR model is extended from 2013Q1 to 2018Q2.

Model Specification

7. The GVAR model includes 33 economies, which together cover more than 90 percent of world GDP (see Table 1). We obtain quarterly data (for the period 1979Q1–2016Q4) from Mohaddes and Raissi (2018b) on oil prices, domestic macroeconomic variables (treated as endogenous) and country-specific foreign variables (taken to be weakly exogenous) for the 33 economies, and update this data until 2018Q2.

Table 1.

Japan: Countries in the GVAR Model

article image
Source: Mohaddes and Raissi (2018b).

8. The modeling framework includes an index of financial stress (FSIt) in advanced economies. The FSIt for advanced countries is constructed by Cardarelli et al. (2009) as an average of the following indicators: the beta of banking sector stocks; TED spread; the slope of the yield curve; corporate bond spreads; stock market returns; time-varying stock return volatility; and time-varying effective exchange rate volatility. Such an index facilitates the identification of large shifts in asset prices (stock and bond market returns); an abrupt increase in risk/uncertainty (stock and foreign exchange volatility); liquidity tightening (TED spreads); and the health of the banking system (the beta of banking sector stocks and the yield curve). We model FSIt as a common variable, in other words, it is included as a weakly exogenous variable in each of the 33 country/region-specific models, but we allow for feedback effects from any of the macro variables to FSIt.

9. Given the importance of oil shocks for the global economy, the modeling framework also includes nominal oil prices in U.S. dollars in the country-specific models. We allow for oil prices to be determined in global commodity markets rather in the U.S. model alone (as is more standard in the literature), given that oil prices are also affected by, for instance, any disruptions to oil supply in the Middle East. Therefore, in contrast to the existing GVAR literature, we model the oil price equation separately and then introduce the price of oil as a weakly exogenous variable in all countries, thereby allowing both demand and supply conditions to influence the oil price directly.2

C. Spillover Analysis

10. This section analyzes inward spillovers to Japan from: (i) macroeconomic shocks in other systemic economies (China, the Euro Area, and the United States), (ii) a potential global growth slowdown, and (iii) stress in global financial markets. It also studies outward spillovers from Japan to other Asian countries and globally.

uA07fig01

Responses of Output to a Negative GDP Shock in China

(In percent)

Citation: IMF Staff Country Reports 2020, 040; 10.5089/9781513529424.002.A007

Source: Author’s estimates.Note: Figure depicts annual percent change in output of a group of countries associated with a 1 percent decline in GDP of China. Symbols — and Δ denote the median and average responses across countries in each group. The boxes show the 25th–75th percentile responses and the whiskers show the minimum and maximum responses.

China Slowdown

11. China’s real GDP growth is slowing— from an average of about 10 percent over the period 1980–2013 to an average of 6½ percent between 2017 and 2019. These developments, together with market concerns about the future performance of the Chinese economy (also affected by trade tensions with the United States), are resulting in spillovers to other economies (especially to countries in the Asia-Pacific region) through trade links, weaker commodity prices, and financial linkages. The results of the GVAR model show that a one percent negative GDP shock in China translates into lower economic growth globally (text chart). Countries with large trade exposures to China are most vulnerable to a slowdown in this country. The effects on the GDP of Asia and Pacific countries are generally large (with the median effect being 0.3 percent after one year) owing to strong trade linkages with China. Countries in Europe and North America also suffer a decline in economic output when China’s growth slows—by about 0.2 percent each after one year (because some are included in the supply chain of China). Turning to South American, Middle East and African countries, following a negative GDP shock in China, the median output of these countries falls by about 0.2 and 0.4 percent, respectively, largely owing to weaker commodity prices. These findings are somewhat to be expected given the emergence of China as a key driver of the global economy in recent decades. Following a negative China GDP shock, real median output in Japan falls by about 0.25 percent after one year.

United States Slowdown

12. As a result of the dominance of the United States in the global economy, any slowdown in this country can bring about negative spillovers to other economies. Lower commodity prices are one channel through which a negative U.S. shock affects countries, conveying a negative impact on growth prospects of commodity-exporters (e.g. Middle East and Africa). The median effects of a one percent negative U.S. output shock for output in North American countries are the largest (in absolute values)—median output falling by about 0.65 percent owing to strong trade links between Mexico, Canada, and the United States. Furthermore, the continuing dominance of U.S. debt and equity markets, backed by the strong global role of the U.S. dollar, plays an important role in spillovers to other countries. The results of the GVAR model show that the influence of the U.S. on other economies remains larger than direct trade ties would suggest, owing to third-market effects together with increased financial integration that tends to foster the international transmission of business cycles (text chart). For instance, following a negative U.S. GDP shock, real median output in Europe and Asia falls by about 0.2 percent after one year. Similarly, following a negative U.S. GDP shock, real median output in Japan falls by about 0.2 percent after one year.

uA07fig02

Responses of Output to a Negative GDP Shock in United States

(In percent)

Citation: IMF Staff Country Reports 2020, 040; 10.5089/9781513529424.002.A007

Source: Author’s estimates.Note: Figure depicts annual percent change in output of a group of countries associated with a 1 percent decline in GDP of the United States. Symbols — and Δ denote the median and average responses across countries in each group. The boxes show the 25th–75th percentile responses and the whiskers show the minimum and maximum responses.

Euro Area Slowdown

13. The adverse impact on output of a one percent negative GDP shock in the Euro Area is most significant for other European countries. Median output in European countries falls by 0.65 percent after one year following the shock (text chart). Beyond Europe, growth spillovers vary from country to country. High dependencies are observed for South American countries, with annual output elasticity of about 0.3 operating via trade and commodity-price channels. For other countries/regions, the impact is modest (assuming limited financial stress in the Euro Area). Following a negative Euro Area GDP shock, real median output in Japan falls by about 0.2 percent after one year.

uA07fig03

Responses of Output to a Negative GDP Shock in Euro Area

(In percent)

Citation: IMF Staff Country Reports 2020, 040; 10.5089/9781513529424.002.A007

Source: Author’s estimates.Note: Figure depicts annual percent change in output of a group of countries associated with a 1 percent decline in GDP of Euro Area. Symbols — and Δ denote the median and average responses across countries in each group. The boxes show the 25th–75th percentile responses and the whiskers show the minimum and maximum responses.

Global Growth Slowdown

14. Trade policy uncertainty, geopolitical tensions, and idiosyncratic stress in key emerging market economies weighed on global growth in the second half of 2019. While accommodative global financial conditions have continued, raising growth prospects for 2020, growth forecasts in key emerging market economies remained subdued. Should downside risks to global growth materialize, and be accompanied by rapidly deteriorating financial sentiment, the results from the GVAR model suggest that countries in Asia-Pacific or Middle East and Africa would be disproportionately affected, owing to third-market effects (text chart). Specifically, following a one percent fall in global GDP, the median country in the Asia-Pacific region could experience a reduction in GDP of about 1.5 percent after one year. The median impact is even larger in the Middle East and Africa region given a demand-driven reduction in global commodity prices. Similar to the rest of Asia-Pacific region, following a negative global GDP shock, real median output in Japan falls by about 1.3 percent after one year.

uA07fig04

Responses of Output to a Negative Global GDP Shock

(In percent)

Citation: IMF Staff Country Reports 2020, 040; 10.5089/9781513529424.002.A007

Source: Author’s estimates.Note: Figure depicts annual percent change in output of a group of countries associated with a 1 percent decline in Global GDP. Symbols — and A denote the median and average responses across countries in each group. The boxes show the 25th–75th percentile responses and the whiskers show the minimum and maximum responses.

Stress in Global Financial Markets

15. Stress in global financial markets could emanate from an increase in risk premiums in reaction to a decline in investor sentiment triggered by a deteriorating outlook, or weak policy frameworks amidst concerns about debt sustainability in some advanced economies. Such shocks could lead to higher interest rates, exchange rate volatility, corrections in stretched asset valuations (for example, equity and real estate), and sudden international financial flow reversals. These developments would strain leveraged companies, households, and sovereigns; worsen bank balance sheets and profitability; and damage the public finances of advanced and emerging market economies. Cashin et al. (2017b) discuss the implications of financial stress in advanced economies, and point to potentially sizable spillovers from such shocks.

uA07fig05

Responses of Output to Potential Stress in Global Financial Markets

(In percent)

Citation: IMF Staff Country Reports 2020, 040; 10.5089/9781513529424.002.A007

Source: Author’s estimates.Note: Figure depicts annual percent change in output of a group of countries associated with a one-standard deviation positive shock to FSI. Symbols — and Δ denote the median and average responses across countries in each group. The boxes show the 25th–75th percentile responses and the whiskers show the minimum and maximum responses.

16. The GVAR model results indicate that in response to potential stress in advanced economies’ financial markets, global economic growth decelerates in the short-run. More specifically, a one standard deviation shock to the financial stress index (FSI)3 translates into slower global economic activity—with world output falling by around 0.2 percent below the pre-shock level on average over the first year (text chart). The growth impact on different countries/regions depends on the magnitude and duration of the FSI shock; countries’ economic fundamentals; and the size of safe-haven flows. For instance, model estimates show that the impact is greatest in South America. Growth spillovers to the median Asian country is about 0.1 percent. Following a negative shock to the financial stress index, real median output in Japan falls by about 0.2 percent after one year.

uA07fig06

Responses of Output to a Negative GDP Shock in Japan (2014–16 Weights)

(In percent)

Citation: IMF Staff Country Reports 2020, 040; 10.5089/9781513529424.002.A007

Source: Author’s estimates.Note: Figure depicts annual percent change in output of a group of countries associated with a 1 percent decline in GDP of Japan. Symbols — and Δ denote the median and average responses across countries in each group. The boxes show the 25th–75th percentile responses and the whiskers show the minimum and maximum responses.

Outward Spillovers from Japan

17. Outward spillovers from Japan to other Asian economies and the rest of the world are important. The results from the GVAR model indicate that output shocks in Japan matter, particularly for its immediate neighborhood, but also have global implications (albeit to a lesser extent than those of China, the United States, and the Euro Area). A one percent decline in the GDP of Japan (using 2014–16 trade weights) generates relatively large output losses in other Asian countries, corresponding to around 0.2 percent after one year (text chart). This indicates that the influence of output shocks emanating from Japan remains important for countries in the Asia-Pacific region. However, a Japanese slowdown only has a modest negative effect in other non-Asian regions (with a median effect of about 0.1 percent). To investigate whether the global impact of a negative output shock in Japan has changed over the past three decades, we re-estimated the GVAR model using trade weights averaged over 1984 to 1986 (text chart). Comparing these results with those obtained from our original specification (using 2014–16 trade weights), the results from the GVAR model indicate that the impact of the Japan output shock has declined significantly over the past thirty years. Following a negative Japan GDP shock (using 1984–86 trade weights), real median output declines in other Asian countries by about 0.4 percent after one year, with output losses in other regions of around 0.3 percent after one year. This finding suggests that the influence of negative Japan output shocks on the global economy has become smaller in recent decades.

uA07fig07

Responses of Output to a Negative GDP Shock in Japan (1984–86 Weights)

(In percent)

Citation: IMF Staff Country Reports 2020, 040; 10.5089/9781513529424.002.A007

Source: Author’s estimates.Note: Figure depicts annual percent change in output of a group of countries associated with a 1 percent decline in GDP of Japan. Symbols — and Δ denote the median and average responses across countries in each group. The boxes show the 25th–75th percentile responses and the whiskers show the minimum and maximum responses.

References

  • Cardarelli, R., S. Elekdag, and S. Lall, 2009, “Financial Stress, Downturns, and Recoveries.” IMF Working Paper 09/100, International Monetary Fund, Washington, DC.

    • Search Google Scholar
    • Export Citation
  • Cashin, P., K. Mohaddes, M. Raissi, and M. Raissi, 2014, “The Differential Effects of Oil Demand and Supply Shocks on the Global Economy,” Energy Economics, Vol. 44, pp. 11334.

    • Search Google Scholar
    • Export Citation
  • Cashin, P., K. Mohaddes, and M. Raissi, 2016, “The Global Impact of the Systemic Economies and MENA Business Cycles,” In I.A. Elbadawi and H. Selim (eds.), Understanding and Avoiding the Oil Curse in Resource-Rich Arab Economies, Cambridge University Press: Cambridge, pp. 1643.

    • Search Google Scholar
    • Export Citation
  • Cashin, P., K. Mohaddes, and M. Raissi, 2017a, “Fair Weather or Foul? The Macroeconomic Effects of El Niño,” Journal of International Economics, Vol. 106, pp. 3754.

    • Search Google Scholar
    • Export Citation
  • Cashin, P., K. Mohaddes, and M. Raissi, 2017b, “China’s Slowdown and Global Financial Market Volatility: Is World Growth Losing Out?Emerging Markets Review, Vol. 31, pp. 16475.

    • Search Google Scholar
    • Export Citation
  • Chudik, A. and M.H. Pesaran, 2016, “Theory and Practice of GVAR Modeling,” Journal of Economic Surveys, Vol. 30, pp. 16597.

  • Dees, S., F. di Mauro, M.H. Pesaran, and L.V. Smith, 2007, “Exploring the International Linkages of the Euro Area: A Global VAR Analysis,” Journal of Applied Econometrics, Vol. 22, pp. 138.

    • Search Google Scholar
    • Export Citation
  • Dizioli, A., J. Guajardo, V. Klyuev, and M. Raissi, 2016, “Spillovers from China’s Growth Slowdown and Rebalancing to the ASEAN-5 Economies,” IMF Working Paper 16/170, International Monetary Fund, Washington, DC.

    • Search Google Scholar
    • Export Citation
  • International Monetary Fund, 2012, 2012 Spillover Report. July, Washington, DC.

  • International Monetary Fund, 2014a, IMF Multilateral Policy Issues Report: 2014 Spillover Report. June, Washington, DC.

  • International Monetary Fund, 2014b, Regional Economic Outlook: Asia and the Pacific. Washington, DC, April.

  • International Monetary Fund, 2015, Regional Economic Outlook: Asia and the Pacific. Washington, DC, May.

  • International Monetary Fund, 2016, Regional Economic Outlook: Asia and the Pacific. Washington, DC, May.

  • International Monetary Fund, 2019, World Economic Outlook: Global Manufacturing Downturn, Rising Trade Barriers. Washington DC, October.

    • Search Google Scholar
    • Export Citation
  • Kose, A., C. Lakatos, F. Ohnsorge, and M. Stocker, 2017, “The Global Role of the U.S. Economy: Linkages, Policies and Spillovers,” Policy Research Working Paper 7962, World Bank: Washington, DC.

    • Search Google Scholar
    • Export Citation
  • Mohaddes, K. and M.H. Pesaran, 2016, “Country-Specific Oil Supply Shocks and the Global Economy: A Counterfactual Analysis,” Energy Economics, Vol. 59, pp. 38299.

    • Search Google Scholar
    • Export Citation
  • Mohaddes, K. and M. Raissi, 2018a, “The U.S. Oil Supply Revolution and the Global Economy,” Empirical Economics, pp. 132.

  • Mohaddes, K. and M. Raissi, 2018b, “Compilation, Revision and Updating of the Global VAR (GVAR) Database, 1979Q2-2016Q4,” University of Cambridge: Faculty of Economics (mimeo).

    • Search Google Scholar
    • Export Citation
  • Pesaran, M. H., T. Schuermann, and S. Weiner, 2004, “Modelling Regional Interdependencies Using a Global Error-Correcting Macroeconometric Model,” Journal of Business and Economics Statistics, Vol. 22, pp. 12962.

    • Search Google Scholar
    • Export Citation
1

Prepared by Paul Cashin (APD), Kamiar Mohaddes (University of Cambridge), and Mehdi Raissi (FAD).

3

A one standard deviation positive shock to FSI in advanced economies is two-thirds of the shock that occurred during the European sovereign debt crisis, and one-tenth of the shock that occurred during the Global Financial Crisis.

Japan: Selected Issues
Author: International Monetary Fund. Asia and Pacific Dept
  • View in gallery

    Responses of Output to a Negative GDP Shock in China

    (In percent)

  • View in gallery

    Responses of Output to a Negative GDP Shock in United States

    (In percent)

  • View in gallery

    Responses of Output to a Negative GDP Shock in Euro Area

    (In percent)

  • View in gallery

    Responses of Output to a Negative Global GDP Shock

    (In percent)

  • View in gallery

    Responses of Output to Potential Stress in Global Financial Markets

    (In percent)

  • View in gallery

    Responses of Output to a Negative GDP Shock in Japan (2014–16 Weights)

    (In percent)

  • View in gallery

    Responses of Output to a Negative GDP Shock in Japan (1984–86 Weights)

    (In percent)