China’s growth model became increasingly dependent on investment during the first decade of the 2000s, and its footprint in global imports widened substantially. Several economies in China’s supply chain are increasingly exposed to the country’s investment-based growth and face growing risks from a deceleration in investment in China. This chapter quantifies potential global spillovers from an investment slowdown in China, finding that a 1 percentage point slowdown in investment in China is associated with a reduction of global growth of slightly less than one-tenth of a percentage point. The impact is about five times larger than in 2002. Regional supply chain economies and commodity exporters with relatively less diversified economies are most vulnerable to an investment slowdown in China. The spillover effects also register strongly across a range of macroeconomic, trade, and financial variables among G20 trading partners.
Introduction
China’s growth model became increasingly dependent on investment during the first decade of the 2000s. Investment contributed about ½ of China’s GDP growth during this time, with particularly large contributions toward the end of the decade. In part, this investment growth reflects the step increase in infrastructure investment during the 2008–10 stimulus response to the global financial crisis. Investment as a share of GDP increased by close to 6 percentage points during this period (relative to precrisis), reaching 48 percent of GDP in 2010. However, it increasingly appears that other forces, including the ongoing urbanization process, the more recent emphasis on social housing construction, and capacity building in high-end manufacturing and services, are also contributing to investment growth.
Associated with these changes in the profile of investment are important shifts in China’s import basket. As more of its manufacturing gets onshored, the share of machinery imports has been gradually declining (Figure 7.1). At the same time, with China increasingly drawing in larger volumes of minerals and metals, their share of total imports has grown steadily (Figure 7.2).
Mineral Import Volumes
(three-month moving average)
Sources: CEIC Data; and IMF staff estimates.These developments have had a noticeable impact on global trade flows. Major exporters of commodities, capital goods, parts, and components have been sending an increasing fraction of their exports to China during the course of the decade (Figure 7.3). This change in trade flows reflects, to some extent, the fact that supply chains have been increasingly routed through China as the final stage of assembly (see IMF, 2012, for more details).
The importance of exports to China, when assessed relative to trading partner GDP, shows even sharper increases for several economies. This ratio has, on average, quadrupled during the decade (Figure 7.4). Particularly exposed are Asian regional economies such as Taiwan Province of China, Malaysia, and the Republic of Korea—all of which are important exporters of capital goods, parts, and components for final assembly in China.
Assessing Exposures to Investment-Led Growth in China
China’s reliance on investment-based growth raises questions about how the new capacity will be used. Capacity that finds its way onto world markets by way of new exports and that is perceived to be putting downward pressure on global prices would create the potential for retaliatory trade actions that eventually come back to hurt the Chinese economy and slow investment (see Guo, 2011, for more details). Another possibility is that the new capacity remains underutilized, with adverse effects on bank balance sheets and credit conditions (which would make the financing of subsequent investment difficult). A rapid investment slowdown in China under either of these two outcomes will undoubtedly have a global impact, given China’s size and systemic importance.
To get a sense of the potential magnitudes, the spillover from China on trading partner j is measured as,
in which
China Fixed Investment growtht is measured as the annual percentage change of real gross fixed capital formation from the national accounts. This spillover measure varies across countries in a given year based on their export exposure to China and also varies over time based on fluctuations in China’s fixed investment growth. By construction, it only measures the influence of Chinese activity on other economies through the direct trade channel. Indirect trade exposures through vertically integrated intermediate economies are not captured. Another concern with this measure is that it does not reflect financial exposures, which would also have a bearing on growth in trading partners. However, with the comprehensive system of capital controls in place and the dominance of domestic sources of financing, the financial spillover channel is likely to be limited.
The effect of the spillover from China on trading partner growth is estimated using a broad sample of 64 economies exposed to China through the export channel described above. The sample covers 2002–11 (starting with China’s entry into the World Trade Organization) and includes the full set of Organization for Economic Cooperation and Development economies, emerging markets classified under the MSCI index, and key commodity producers. The main specification is
The main coefficient of interest in this regression is β2, which captures the effect of spillovers from China’s investment activity on growth in trading partner j. Additional controls include partner country lagged growth, the annual percentage change in terms of trade (ToTj,t), and macroeconomic volatility (Volatilityj,t measured as the standard deviation of GDP growth calculated over moving five-year windows).
The regression is also estimated using different measures of fixed investment growth in China: overall, manufacturing, and nontradables.1 Manufacturing and nontradables fixed investment are calculated by applying shares from fixed asset investment data (available only from 2003 onward) to the national accounts series on real gross fixed capital formation. This breakdown allows for a comparison of likely effects from a slowdown in investment concentrated in manufacturing versus a deceleration concentrated in nontradables.
Effects of an Investment Slowdown in China
The impact of China’s investment-led growth on trading partners has strengthened as China’s growth model tilts more toward investment and its global footprint on imports widens (Annex 7A, Tables 7A.2–7A.5). Aggregating across all 64 economies (weighted by their purchasing power parity [PPP] shares), the impact on global growth of a 1 percentage point slowdown in investment in China is just under one-tenth of a percentage point. The impact is about five times larger than in 2002 (estimated to be 0.02 percentage point). The most heavily exposed economies are those within the Asian regional supply chain, such as Taiwan Province of China, Korea, and Malaysia (Figure 7.5). The results from Annex 7A, Table 7A.5 (estimated based on the sample years covering the global financial crisis and the stimulus response in China) suggest that if investment growth declines by 1 percentage point in China, GDP growth in Taiwan Province of China, for example, falls by slightly more than nine-tenths of a percentage point. Among the advanced economy exporters of capital goods, Japan suffers a decline of just over one-tenth of a percentage point in response, while growth in Germany declines by a slightly smaller amount.
Impact on Trading Partner Growth of Investment Slowdown in China
Source: IMF staff estimates.Note: Impact of a 1 percentage point slowdown.Among commodity exporters, the impact of a slowdown in investment growth in China is likely to be largest on mineral ore exporters with relatively less diversified economies and a higher concentration of exports to China. In response to a 1 percentage point slowdown in investment growth in China, the estimated effect on Chile’s growth is a reduction of close to two-fifths of a percentage point (Figure 7.6). By contrast, the larger commodity exporters with more diversified economies, such as Australia and Brazil, suffer relatively smaller declines in growth.
Impact on Commodity Exporters of Investment Slowdown in China
Source: IMF staff estimates.Note: Impact of a 1 percentage point slowdown.A sectoral decomposition of China’s overall fixed investment in manufacturing and nontradables shows that the magnitude of spillovers from a slowdown in manufacturing nontradables fixed investment (NFI) are broadly similar to the effects from a slowdown in overall NFI (Table 7A.5). The impact associated with a slowdown concentrated in nontradables is considerably smaller. The impact on Taiwan Province of China’s growth is about three-fourths of a percentage point compared with slightly greater than nine-tenths of a percentage point for a generalized investment slowdown (Figure 7.7). Similarly, Chile’s growth declines by about a third of a percentage point in response to a slowdown concentrated in tertiary sector investment in China (Figure 7.8), compared with two-fifths of a percentage point in the broader investment slowdown described above.
Impact on Trading Partner Growth of a Slowdown in Nontradables Investment in China
Source: IMF staff estimates.Note: Impact of a 1 percentage point slowdown.Impact on Commodity Exporters’ Growth of a Slowdown in Nontradables Investment in China
Source: IMF staff estimates.Note: Impact of a 1 percentage point slowdown.The results also suggest that China’s manufacturing investment reflects the influence of the global business cycle, but nontradables investment has a spillover impact above and beyond the effect of global growth (Annex 7A, Table A7.6). Once a control for global growth excluding China is added to the regression, the spillover effect via manufacturing fixed investment in China is no longer significant.
Implications of a Hand-Off to Consumption
If the capacity currently being installed in China is absorbed domestically (which would require consumption to accelerate in response to the structural reforms envisaged in the 12th Five-Year Plan), a smooth hand-off from investment- to consumption-led growth can be achieved. China’s growth would moderate into the medium term as outlined in the rebalancing scenario in IMF (2011).
The benefits of such an outcome for consumer goods exporters are, however, likely to be small. China’s share in global consumer goods imports has increased at a slower pace than its share in global consumption since the late 1990s. It currently plays a small role as an importer of consumer goods, accounting for only 2 percent of global consumer goods imports (see IMF, 2012, for more details).
The panel regression approach using the broad sample of 64 economies confirms that this is the case. The low import intensity of consumption in China suggests that the direct spillover effect from consumption growth on trading partner growth is negligible. A similar exercise to the one outlined above, but that instead quantifies potential spillovers from consumption growth in China, shows that the effects on trading partner growth are insignificant (Annex 7A, Table 7A.7).
Effects of an Investment Slowdown on G20 Macroeconomic Indicators
A complementary approach uses a factor-augmented vector autoregression (FAVAR) to gauge the domestic and global spillovers of a slowdown in China’s fixed asset investment (FAI). The FAVAR framework is extended into a two-region model that allows China to interact with the other G20 economies. The analysis captures the feedback from China to the rest of the world, and vice versa, over time. It also captures the spillover effect among the rest of the G20 economies from a specific event originated in China.
Market participants monitor hundreds of economic variables in their decision-making processes, which provides motivation for conditioning the analysis of their decisions on a rich information set. The FAVAR framework extracts information from the rich data set to gauge the impact of particular forces that may not be directly observable. These “forces” are treated as latent common components that are interrelated, and their impacts on economic variables are traced through impulse response functions. Accounting for unobserved variables provides a better chance of avoiding findings based on spurious association.
Briefly, the model is a stable FAVAR in growth (except for balances and interest rates) with five common factors for each region (China and the rest of the G20 economies) and China’s FAI.2 The model uses one lag. The Cholesky factor from the residual covariance matrix is used to orthogonalize the impulses, thereby imposing an ordering of the variables in the VAR and treating investment as exogenous in the period of shock. The results are robust to reordering within factor groups.
The data set is a balanced panel of 390 monthly time series from the G20 stretching from January 2000 to September 2011, with 68 China-specific variables and 322 from the rest of the G20. The sample contains at least one full cycle of investment in China. It starts right before China’s entry into the WTO and spans the time when the country became increasingly integrated with the world economy.
Because the model is in growth, the experiment assumes an exogenous, temporary, one standard deviation growth shock to China’s FAI. The shock dampens within three quarters and dissipates fully after about 40 months. Specifically, this is a one-time 15 percentage point (seasonally adjusted, annualized) drop in FAI growth that largely reverts to trend growth within seven to eight months.3 Whereas this is a temporary, negative growth shock, the decline in FAI level is permanent. The shock is approximately equivalent to a 2½ percent drop from the real FAI baseline level in 12 months. The analysis does not assume any policy response beyond that already in the sample. Peak impacts in each 24-month period are reported with standard error bands in Figures 7.9–7.13. Impacts on levels 12 months after the shock, in percentage below baseline, are also derived and reported for comparison in Annex 7B, Tables 7B.1 and 7B.2.
Peak Impact on Industrial Production
(seasonally adjusted annual rate)
Source: IMF staff estimations.Note: Impact of a 1 standard deviation shock.* Canada’s economic activity is represented by monthly real GDP index, all industries.Peak Impact on Real GDP
(seasonally adjusted annual rate)
Source: IMF staff estimations.Note: Impact of a 1 standard deviation shock.Peak Impact on Exports to China
(seasonally adjusted annual rate)
Source: IMF staff estimations.Note: Impact of a 1 standard deviation shock.Peak Impact on Stock Market Indexes
(seasonally adjusted annual rate)
Source: Author’s estimations.Note: Impact of a 1 standard deviation shock.Peak Impact on World Metal and Rubber Prices
(seasonally adjusted annual rate)
Source: Author’s estimations.Note: Impact of a 1 standard deviation shock.A temporary shock to China’s FAI growth would reverberate around the world, with the spillover impacts on G20 economies dissipating after about five to eight quarters (Figures 7.9 and 7.10). In this exercise, the approximate impact on GDP growth would vary with the size of the industrial-production-to-GDP ratio in each economy.4 The implied peak impact on PPP-weighted G20 GDP growth is −0.2 percentage point, which translates to about 0.1 percent below baseline at 12 months after the shock originated in China (see Annex 7B, Table 7B.1). Overall, capital goods manufacturers that have sizable direct exposure to China (through exports to China as a percentage of own GDP) and are highly integrated with the rest of the G20 (therefore sharing adverse feedback from a negative shock in China with other trading partners such as Germany and Japan) would see more of the impact on economic activity. One year out, the impact is also sizable for Canada. The impact on Indonesia’s output is not statistically significant over the entire period, likely because coal exports to China became important only toward the end of the first decade of the 2000s.
The results also show that global trade activity would decline (total exports and total imports for every G20 economy would weaken), which suggests that economies that derive significant benefit from global trade expansion and have deeper links via supply chain countries over the past decade, such as Germany and Japan, should be more hard hit in the second round (see also Chapter 8). The impact on Korea’s GDP peaks within the first two quarters and fades away more quickly, consistent with Korea’s large direct exposure to China, but second-round effects through supply chain countries are smaller than for Japan and Germany.
The slowdown in growth of exports to China for Brazil, India, and Korea mirrors the impacts on their industrial production growth (Figure 7.11). For the United Kingdom, however, from which exports to China slow the most, the exports to China are not an important component of final demand, and the impact on economic activity looks to be moderate.5 Brazil, whose exports to China are agricultural and heavy in mineral commodities, would also experience nonnegligible spillover effects on export growth. Australia’s relatively large direct exposure to China suggests a substantial direct impact, but other forces (e.g., the Australian dollar exchange rate behaves as a shock absorber) seem to blunt the effect on Australia’s industrial production, which accounts for about 20 percent of GDP. Nevertheless, other indicators, such as employment growth and total import growth (not shown here), point to a slowdown in Australia’s economic activity. Overall trade expansion with China would also slow as global and China demand growth weaken.
The spillover effects are also captured in asset prices (Figure 7.12). Specifically, the impact on the stock market indexes in G20 economies would be as large as 5–5½ percentage points in India and Brazil and between 4 and 4½ percentage points in the euro area, Germany, and Japan—and would last for as long as four to five quarters.
Even as nonfuel primary commodity price inflation—especially metal price inflation—retreats, the impact on global inflation appears to be almost negligible. A global growth slowdown, initiated by a temporary investment growth slowdown in China, would lead to a drop in iron ore, aluminum, copper, lead, nickel, and zinc price growth of as much as 3–9 percentage points (Figure 7.13). This is equivalent to a decline in price levels of about 2–5½ percent below baseline levels, one year out (see Annex 7B, Table 7B.2). How crude oil prices would be affected in this exercise is unclear (the impulse responses show a drop in crude price growth, with peak at about three quarters after impact, but the responses are not statistically significant).
The model suggests that the role of China’s investment drive in boosting construction-related metal prices between 2008 and 2011 has been significant. Annex 7B, Table 7B.4 reports the extent of China’s contribution to metal price growth during 2008–11, with the counterfactual (no investment drive) scenario assuming China’s real gross fixed capital formation had grown at the same pace as real GDP, so that the investment-to-GDP ratio was maintained at its end-2007 level.
Conclusion
A rapid investment slowdown in China is likely to have large spillover effects on a number of China’s trading partners. At the macro level, each 1 percentage point deceleration in China’s investment growth is estimated to subtract between one-half and nine-tenths of a percentage point from GDP growth in regional supply chain economies such as Taiwan Province of China, Korea, and Malaysia. Major commodity producers with relatively large exposures to China, such as Chile and Saudi Arabia, are also likely to suffer substantial growth declines in response to an investment deceleration in China.
The spillover effects from an investment slowdown in China also register strongly across a range of macroeconomic, trade, and financial variables among G20 trading partners as well as world commodity prices. Within this group, a decline in China FAI would have a substantial impact on capital goods manufacturing economies with relatively sizable exports to China (as a percentage of own GDP) and that are highly integrated with the rest of the G20, such as Germany and Japan. For economies that rely less on China’s demand, such as the United Kingdom and India, the spillover effects on industrial production and aggregate output are moderate. Important commodities exporters, such as Canada and Brazil, would experience nonnegligible spillover effects on export growth, which would translate into somewhat significant output losses and slowdowns in overall economic activity. Worsened global growth prospects would also be reflected in commodity prices. One year after the shock, commodity prices, especially metal prices, could fall by as much as 0.8–2.2 percent from baseline levels for every 1 percent drop in China’s FAI.
Annex 7A. Regression Results and Contributions to Growth from Exports to China, Select Economies
Regression Results
Summary Statistics for Analysis of Investment Spillovers, 2002–11
Summary Statistics for Analysis of Investment Spillovers, 2002–11
| Mean | Standard Deviation | Minimum | Maximum | |
|---|---|---|---|---|
| Year-over-year percent change | ||||
| China fixed investment | 13.5 | 3.7 | 9.7 | 23.5 |
| China manufacturing fixed investment | 16.6 | 2.9 | 11.0 | 20.6 |
| China nontradables fixed investment | 11.4 | 6.2 | 5.6 | 26.8 |
| Exports to China/GDP, percent | ||||
| Australia | 2.4 | 1.4 | 1.1 | 5.0 |
| Brazil | 1.0 | 0.4 | 0.5 | 1.8 |
| Chile | 4.7 | 2.2 | 1.8 | 8.0 |
| Germany | 1.3 | 0.4 | 0.7 | 2.0 |
| Japan | 2.1 | 0.6 | 1.0 | 2.8 |
| Korea | 8.3 | 2.6 | 4.1 | 12.0 |
| Malaysia | 8.6 | 3.2 | 5.2 | 16.6 |
| Taiwan Province of China | 12.9 | 4.9 | 3.3 | 18.0 |
| United States | 0.4 | 0.2 | 0.2 | 0.7 |
Summary Statistics for Analysis of Investment Spillovers, 2002–11
| Mean | Standard Deviation | Minimum | Maximum | |
|---|---|---|---|---|
| Year-over-year percent change | ||||
| China fixed investment | 13.5 | 3.7 | 9.7 | 23.5 |
| China manufacturing fixed investment | 16.6 | 2.9 | 11.0 | 20.6 |
| China nontradables fixed investment | 11.4 | 6.2 | 5.6 | 26.8 |
| Exports to China/GDP, percent | ||||
| Australia | 2.4 | 1.4 | 1.1 | 5.0 |
| Brazil | 1.0 | 0.4 | 0.5 | 1.8 |
| Chile | 4.7 | 2.2 | 1.8 | 8.0 |
| Germany | 1.3 | 0.4 | 0.7 | 2.0 |
| Japan | 2.1 | 0.6 | 1.0 | 2.8 |
| Korea | 8.3 | 2.6 | 4.1 | 12.0 |
| Malaysia | 8.6 | 3.2 | 5.2 | 16.6 |
| Taiwan Province of China | 12.9 | 4.9 | 3.3 | 18.0 |
| United States | 0.4 | 0.2 | 0.2 | 0.7 |
Fixed Effects Regression for WTO Period, 2002–11
Significant at 1 percent level.
Significant at 5 percent level.
Significant at 10 percent level.
Fixed Effects Regression for WTO Period, 2002–11
| Total Investment (1) |
Manufacturing (2) |
Nontradables (3) |
|
|---|---|---|---|
| China spillover effect | 0.0128*** | 0.0381*** | 0.0255*** |
| (0.00418) | (0.0106) | (0.00561) | |
| Terms of trade | 9.69e-06*** | 0.000589 | 0.000260 |
| (1.85e-06) | (0.00303) | (0.00306) | |
| Volatility of growth | –0.424 | –0.771*** | –0.854*** |
| (0.271) | (0.231) | (0.247) | |
| SAMPLE YEARS | 2002–11 | 2002–11 | 2002–11 |
| Number of countries | 64 | 64 | 64 |
| Observations | 640 | 448 | 448 |
| R-squared | 0.03 | 0.13 | 0.14 |
Significant at 1 percent level.
Significant at 5 percent level.
Significant at 10 percent level.
Fixed Effects Regression for WTO Period, 2002–11
| Total Investment (1) |
Manufacturing (2) |
Nontradables (3) |
|
|---|---|---|---|
| China spillover effect | 0.0128*** | 0.0381*** | 0.0255*** |
| (0.00418) | (0.0106) | (0.00561) | |
| Terms of trade | 9.69e-06*** | 0.000589 | 0.000260 |
| (1.85e-06) | (0.00303) | (0.00306) | |
| Volatility of growth | –0.424 | –0.771*** | –0.854*** |
| (0.271) | (0.231) | (0.247) | |
| SAMPLE YEARS | 2002–11 | 2002–11 | 2002–11 |
| Number of countries | 64 | 64 | 64 |
| Observations | 640 | 448 | 448 |
| R-squared | 0.03 | 0.13 | 0.14 |
Significant at 1 percent level.
Significant at 5 percent level.
Significant at 10 percent level.
Fixed Effects Regression for Global Crisis and Stimulus Period, 2008–11
Significant at 1 percent level.
Significant at 5 percent level.
Significant at 10 percent level.
Fixed Effects Regression for Global Crisis and Stimulus Period, 2008–11
| Total Investment (1) |
Manufacturing (2) |
Nontradables (3) |
|
|---|---|---|---|
| China spillover effect | 0.0741*** | 0.0901*** | 0.0561*** |
| (0.0105) | (0.0201) | (0.00747) | |
| Terms of trade | –0.00414 | –0.00159 | –0.00393 |
| (0.00433) | (0.00428) | (0.00433) | |
| Volatility of growth | –0.828*** | –0.566*** | –0.897*** |
| (0.146) | (0.184) | (0.141) | |
| SAMPLE YEARS | 2008–11 | 2008–11 | 2008–11 |
| Number of countries | 64 | 64 | 64 |
| Observations | 256 | 256 | 256 |
| R-squared | 0.2 | 0.14 | 0.21 |
Significant at 1 percent level.
Significant at 5 percent level.
Significant at 10 percent level.
Fixed Effects Regression for Global Crisis and Stimulus Period, 2008–11
| Total Investment (1) |
Manufacturing (2) |
Nontradables (3) |
|
|---|---|---|---|
| China spillover effect | 0.0741*** | 0.0901*** | 0.0561*** |
| (0.0105) | (0.0201) | (0.00747) | |
| Terms of trade | –0.00414 | –0.00159 | –0.00393 |
| (0.00433) | (0.00428) | (0.00433) | |
| Volatility of growth | –0.828*** | –0.566*** | –0.897*** |
| (0.146) | (0.184) | (0.141) | |
| SAMPLE YEARS | 2008–11 | 2008–11 | 2008–11 |
| Number of countries | 64 | 64 | 64 |
| Observations | 256 | 256 | 256 |
| R-squared | 0.2 | 0.14 | 0.21 |
Significant at 1 percent level.
Significant at 5 percent level.
Significant at 10 percent level.
Generalized Method of Moments Regression for WTO period, 2002–11
Significant at 1 percent level.
Significant at 5 percent level.
Significant at 10 percent level.
Generalized Method of Moments Regression for WTO period, 2002–11
| Total Investment | Manufacturing | Nontradables | |
|---|---|---|---|
| (1) | (2) | (3) | |
| Lagged GDP growth | 0.230*** | –0.127 | –0.0751 |
| (0.0527) | (0.0886) | (0.0950) | |
| China spillover effect | 0.0332*** | 0.0457*** | 0.0367*** |
| (0.00840) | (0.0132) | (0.00718) | |
| Terms of trade | –2.50e-06 | –1.91e-06 | –0.000655 |
| (1.27e-05) | (0.00321) | (0.00332) | |
| Volatility of growth | –0.299 | –1.312*** | –1.407*** |
| (0.278) | (0.289) | (0.263) | |
| SAMPLE YEARS | 2002–11 | 2002–11 | 2002–11 |
| Number of countries | 64 | 64 | 64 |
| Observations | 640 | 384 | 384 |
| Arellano-Bond test of no second-order auto-correlation in first-differenced errors (p-value) | 0.22 | 0.08 | 0.15 |
Significant at 1 percent level.
Significant at 5 percent level.
Significant at 10 percent level.
Generalized Method of Moments Regression for WTO period, 2002–11
| Total Investment | Manufacturing | Nontradables | |
|---|---|---|---|
| (1) | (2) | (3) | |
| Lagged GDP growth | 0.230*** | –0.127 | –0.0751 |
| (0.0527) | (0.0886) | (0.0950) | |
| China spillover effect | 0.0332*** | 0.0457*** | 0.0367*** |
| (0.00840) | (0.0132) | (0.00718) | |
| Terms of trade | –2.50e-06 | –1.91e-06 | –0.000655 |
| (1.27e-05) | (0.00321) | (0.00332) | |
| Volatility of growth | –0.299 | –1.312*** | –1.407*** |
| (0.278) | (0.289) | (0.263) | |
| SAMPLE YEARS | 2002–11 | 2002–11 | 2002–11 |
| Number of countries | 64 | 64 | 64 |
| Observations | 640 | 384 | 384 |
| Arellano-Bond test of no second-order auto-correlation in first-differenced errors (p-value) | 0.22 | 0.08 | 0.15 |
Significant at 1 percent level.
Significant at 5 percent level.
Significant at 10 percent level.
Generalized Method of Moments Regression for Global Crisis and Stimulus Period, 2008–11
Significant at 1 percent level.
Significant at 5 percent level.
Significant at 10 percent level.
Generalized Method of Moments Regression for Global Crisis and Stimulus Period, 2008–11
| Total Investment (1) |
Manufacturing (2) |
Nontradables (3) |
|
|---|---|---|---|
| Lagged GDP growth | –0.130 | –0.241** | –0.122 |
| (0.139) | (0.113) | (0.139) | |
| China spillover effect | 0.0543*** | 0.0511*** | 0.0434*** |
| (0.0103) | (0.0116) | (0.00797) | |
| Terms of trade | –0.00685 | –0.00632 | –0.00684 |
| (0.00430) | (0.00394) | (0.00424) | |
| Volatility of growth | –1.973*** | –2.006*** | –2.001*** |
| (0.423) | (0.378) | (0.416) | |
| SAMPLE YEARS | 2008–11 | 2008–11 | 2008–11 |
| Number of countries | 64 | 64 | 64 |
| Observations | 256 | 256 | 256 |
| Arellano-Bond test of no second-order autocorrelation in first-differenced errors (p-value) | 0.11 | 0.28 | 0.12 |
Significant at 1 percent level.
Significant at 5 percent level.
Significant at 10 percent level.
Generalized Method of Moments Regression for Global Crisis and Stimulus Period, 2008–11
| Total Investment (1) |
Manufacturing (2) |
Nontradables (3) |
|
|---|---|---|---|
| Lagged GDP growth | –0.130 | –0.241** | –0.122 |
| (0.139) | (0.113) | (0.139) | |
| China spillover effect | 0.0543*** | 0.0511*** | 0.0434*** |
| (0.0103) | (0.0116) | (0.00797) | |
| Terms of trade | –0.00685 | –0.00632 | –0.00684 |
| (0.00430) | (0.00394) | (0.00424) | |
| Volatility of growth | –1.973*** | –2.006*** | –2.001*** |
| (0.423) | (0.378) | (0.416) | |
| SAMPLE YEARS | 2008–11 | 2008–11 | 2008–11 |
| Number of countries | 64 | 64 | 64 |
| Observations | 256 | 256 | 256 |
| Arellano-Bond test of no second-order autocorrelation in first-differenced errors (p-value) | 0.11 | 0.28 | 0.12 |
Significant at 1 percent level.
Significant at 5 percent level.
Significant at 10 percent level.
Generalized Method of Moments Robustness Check for Global Crisis and Stimulus Period, 2008–11
Significant at 1 percent level.
Significant at 5 percent level.
Significant at 10 percent level.
Generalized Method of Moments Robustness Check for Global Crisis and Stimulus Period, 2008–11
| Total Investment (1) |
Manufacturing (2) |
Nontradables (3) |
|
|---|---|---|---|
| Lagged GDP growth | –0.0805 | –0.0412 | –0.0772 |
| (0.0803) | (0.0949) | (0.0809) | |
| China spillover effect | 0.0250*** | 0.00889 | 0.0211*** |
| (0.00815) | (0.00973) | (0.00608) | |
| Terms of trade | –0.000646 | –0.00199 | –0.000725 |
| (0.00297) | (0.00280) | (0.00297) | |
| Volatility of growth | –1.363*** | –1.119*** | –1.388*** |
| (0.276) | (0.332) | (0.274) | |
| World growth ex China | 0.696*** | 0.875*** | 0.684*** |
| (0.0920) | (0.109) | (0.0916) | |
| SAMPLE YEARS | 2008–11 | 2008–11 | 2008–11 |
| Number of countries | 64 | 64 | 64 |
| Observations | 256 | 256 | 256 |
| Arellano-Bond test of no second-order autocorrelation in first-differenced errors (p-value) | 0.43 | 0.05 | 0.49 |
Significant at 1 percent level.
Significant at 5 percent level.
Significant at 10 percent level.
Generalized Method of Moments Robustness Check for Global Crisis and Stimulus Period, 2008–11
| Total Investment (1) |
Manufacturing (2) |
Nontradables (3) |
|
|---|---|---|---|
| Lagged GDP growth | –0.0805 | –0.0412 | –0.0772 |
| (0.0803) | (0.0949) | (0.0809) | |
| China spillover effect | 0.0250*** | 0.00889 | 0.0211*** |
| (0.00815) | (0.00973) | (0.00608) | |
| Terms of trade | –0.000646 | –0.00199 | –0.000725 |
| (0.00297) | (0.00280) | (0.00297) | |
| Volatility of growth | –1.363*** | –1.119*** | –1.388*** |
| (0.276) | (0.332) | (0.274) | |
| World growth ex China | 0.696*** | 0.875*** | 0.684*** |
| (0.0920) | (0.109) | (0.0916) | |
| SAMPLE YEARS | 2008–11 | 2008–11 | 2008–11 |
| Number of countries | 64 | 64 | 64 |
| Observations | 256 | 256 | 256 |
| Arellano-Bond test of no second-order autocorrelation in first-differenced errors (p-value) | 0.43 | 0.05 | 0.49 |
Significant at 1 percent level.
Significant at 5 percent level.
Significant at 10 percent level.
Spillover Effects of Consumption Growth in China
Significant at 1 percent level.
Significant at 5 percent level.
Significant at 10 percent level.
Spillover Effects of Consumption Growth in China
| (1) | (2) | |
|---|---|---|
| Lagged GDP growth | 0.198*** | –0.240* |
| (0.0541) | (0.131) | |
| China spillover effect (consumption) | –0.0110 | –0.0319 |
| (0.0145) | (0.0362) | |
| Terms of trade | 5.13e-06 | –0.00472 |
| (1.49e-05) | (0.00427) | |
| Volatility of growth | –0.345 | –2.059*** |
| (0.285) | (0.430) | |
| SAMPLE YEARS | 2002–11 | 2008–11 |
| Number of countries | 64 | 64 |
| Observations | 640 | 256 |
| Arellano-Bond test of no second-order autocorrelation in first-differenced errors (p-value) | 0.77 | 0.45 |
Significant at 1 percent level.
Significant at 5 percent level.
Significant at 10 percent level.
Spillover Effects of Consumption Growth in China
| (1) | (2) | |
|---|---|---|
| Lagged GDP growth | 0.198*** | –0.240* |
| (0.0541) | (0.131) | |
| China spillover effect (consumption) | –0.0110 | –0.0319 |
| (0.0145) | (0.0362) | |
| Terms of trade | 5.13e-06 | –0.00472 |
| (1.49e-05) | (0.00427) | |
| Volatility of growth | –0.345 | –2.059*** |
| (0.285) | (0.430) | |
| SAMPLE YEARS | 2002–11 | 2008–11 |
| Number of countries | 64 | 64 |
| Observations | 640 | 256 |
| Arellano-Bond test of no second-order autocorrelation in first-differenced errors (p-value) | 0.77 | 0.45 |
Significant at 1 percent level.
Significant at 5 percent level.
Significant at 10 percent level.
A decomposition of exports by product type for large capital goods exporters and commodity producers shows that the contribution to growth generated by exports to China has increased sharply during the period of the financial crisis and the stimulus response in China (Figure 7A.1).
Contribution to Growth of Exports to China, Period Average
Sources: IMF, Direction of Trade Statistics; and IMF staff estimates.In contrast to the cross-country regression in the main text, this calculation is a straight bilateral accounting exercise and does not provide a causal effect of specific spillovers from China’s investment activity. It does, however, confirm the result from the cross-country exercise of the growing influence of China on trading partner growth. The calculation also shows that despite accounts of the rise of luxury goods exports (such as high-end passenger cars) to China from Japan and Germany, the fraction of growth they account for in these source economies is still relatively small. Finally, with regard to large commodity exporters, the contribution to growth from mineral exports to China has more than doubled during 2008–11 compared with 2001–07. For Australia, they accounted for slightly less than one-half of growth in the later interval. For Brazil, the accounting exercise confirms that the economy appears to be well diversified and exports to China account for a relatively small fraction of overall growth even during the period of infrastructure expansion in China (Figure 7A.2).
Contribution to Growth of Mineral Exports to China, Period Average
Sources: IMF, Direction of Trade Statistics; and IMF staff estimates.Annex 7B
Impacts on Economic Activity Indicators One Year After a 1-Percent Exogenous Decline in China’s Real Total Fixed Asset Investments
(percent below baseline level)
Estimates for Australia and Indonesia are not statistically significant.
Canada’s economic activity is represented by monthly real GDP index, all industries.
Impacts on Economic Activity Indicators One Year After a 1-Percent Exogenous Decline in China’s Real Total Fixed Asset Investments
(percent below baseline level)
| World Indicators | Industrial Production | Real GDP |
|---|---|---|
| Argentina | 0.54 | 0.11 |
| Australia* | 0.02 | 0.00 |
| Brazil | 0.25 | 0.05 |
| Canada** | n.a. | 0.06 |
| China | 0.12 | 0.10 |
| France | 0.17 | 0.02 |
| Germany | 0.61 | 0.11 |
| India | 0.28 | 0.05 |
| Indonesia* | 0.15 | 0.05 |
| Italy | 0.46 | 0.08 |
| Japan | 0.55 | 0.12 |
| Mexico | 0.34 | 0.09 |
| Russian Federation | 0.25 | 0.05 |
| Saudi Arabia | 0.09 | 0.02 |
| South Africa | 0.30 | 0.05 |
| Korea, Rep. of | 0.14 | 0.04 |
| Turkey | 0.45 | 0.09 |
| United Kingdom | 0.13 | 0.02 |
| United States | 0.21 | 0.03 |
| European Union | 0.19 | 0.03 |
| PPP-weighted average | 0.06 | |
Estimates for Australia and Indonesia are not statistically significant.
Canada’s economic activity is represented by monthly real GDP index, all industries.
Impacts on Economic Activity Indicators One Year After a 1-Percent Exogenous Decline in China’s Real Total Fixed Asset Investments
(percent below baseline level)
| World Indicators | Industrial Production | Real GDP |
|---|---|---|
| Argentina | 0.54 | 0.11 |
| Australia* | 0.02 | 0.00 |
| Brazil | 0.25 | 0.05 |
| Canada** | n.a. | 0.06 |
| China | 0.12 | 0.10 |
| France | 0.17 | 0.02 |
| Germany | 0.61 | 0.11 |
| India | 0.28 | 0.05 |
| Indonesia* | 0.15 | 0.05 |
| Italy | 0.46 | 0.08 |
| Japan | 0.55 | 0.12 |
| Mexico | 0.34 | 0.09 |
| Russian Federation | 0.25 | 0.05 |
| Saudi Arabia | 0.09 | 0.02 |
| South Africa | 0.30 | 0.05 |
| Korea, Rep. of | 0.14 | 0.04 |
| Turkey | 0.45 | 0.09 |
| United Kingdom | 0.13 | 0.02 |
| United States | 0.21 | 0.03 |
| European Union | 0.19 | 0.03 |
| PPP-weighted average | 0.06 | |
Estimates for Australia and Indonesia are not statistically significant.
Canada’s economic activity is represented by monthly real GDP index, all industries.
Impacts on Selected Commodity Prices One Year After a 1 Percent Exogenous Decline in China’s Real Total Fixed Asset Investment
(percent below baseline level, year-over-year)
Impacts on Selected Commodity Prices One Year After a 1 Percent Exogenous Decline in China’s Real Total Fixed Asset Investment
(percent below baseline level, year-over-year)
| World Prices: | |
|---|---|
| Metals | 1.3 |
| Nonfuel primary commodities | 0.7 |
| Zinc | 2.2 |
| Nickel | 1.8 |
| Lead | 1.8 |
| Copper | 1.6 |
| Iron ore | 0.8 |
| Aluminum | 1.0 |
| Rubber | 0.6 |
| Silver | 0.6 |
| Gold | 0.2 |
Impacts on Selected Commodity Prices One Year After a 1 Percent Exogenous Decline in China’s Real Total Fixed Asset Investment
(percent below baseline level, year-over-year)
| World Prices: | |
|---|---|
| Metals | 1.3 |
| Nonfuel primary commodities | 0.7 |
| Zinc | 2.2 |
| Nickel | 1.8 |
| Lead | 1.8 |
| Copper | 1.6 |
| Iron ore | 0.8 |
| Aluminum | 1.0 |
| Rubber | 0.6 |
| Silver | 0.6 |
| Gold | 0.2 |
Impacts on Trade indicators One Year After a 1 Percent Exogenous Decline in China’s Real Total Fixed Asset Investment
(percent below baseline level)
Import-weighted.
Export-weighted.
Impacts on Trade indicators One Year After a 1 Percent Exogenous Decline in China’s Real Total Fixed Asset Investment
(percent below baseline level)
| Trade Indicators | Import Value | Export Value |
|---|---|---|
| Argentina | 2.24 | 0.35 |
| Australia | 0.75 | 0.13 |
| Brazil | 0.98 | 0.58 |
| Canada | 0.91 | 0.87 |
| China | 0.74 | 0.74 |
| France | 0.69 | 0.85 |
| Germany | 0.74 | 0.85 |
| India | 0.42 | 0.79 |
| Indonesia | 0.48 | 0.77 |
| Italy | 1.01 | 1.15 |
| Japan | 0.87 | 0.66 |
| Mexico | 0.90 | 0.94 |
| Russian Federation | 0.85 | 0.56 |
| Saudi Arabia | 0.44 | 0.95 |
| South Africa | 0.68 | 0.14 |
| Korea, Rep. of | 0.65 | 0.74 |
| Turkey | 0.93 | 0.52 |
| United Kingdom | 0.93 | 0.90 |
| United States | 0.92 | 0.58 |
| European Union | 0.83 | 0.90 |
| Weighted average | 0.82* | 0.76** |
Import-weighted.
Export-weighted.
Impacts on Trade indicators One Year After a 1 Percent Exogenous Decline in China’s Real Total Fixed Asset Investment
(percent below baseline level)
| Trade Indicators | Import Value | Export Value |
|---|---|---|
| Argentina | 2.24 | 0.35 |
| Australia | 0.75 | 0.13 |
| Brazil | 0.98 | 0.58 |
| Canada | 0.91 | 0.87 |
| China | 0.74 | 0.74 |
| France | 0.69 | 0.85 |
| Germany | 0.74 | 0.85 |
| India | 0.42 | 0.79 |
| Indonesia | 0.48 | 0.77 |
| Italy | 1.01 | 1.15 |
| Japan | 0.87 | 0.66 |
| Mexico | 0.90 | 0.94 |
| Russian Federation | 0.85 | 0.56 |
| Saudi Arabia | 0.44 | 0.95 |
| South Africa | 0.68 | 0.14 |
| Korea, Rep. of | 0.65 | 0.74 |
| Turkey | 0.93 | 0.52 |
| United Kingdom | 0.93 | 0.90 |
| United States | 0.92 | 0.58 |
| European Union | 0.83 | 0.90 |
| Weighted average | 0.82* | 0.76** |
Import-weighted.
Export-weighted.
Impact on Metal Prices of China’s Investment, 2008–11
Impact on Metal Prices of China’s Investment, 2008–11
| Model’s Implied Difference from Counterfactual, in percent | Actual Change, in percent | Counterfactual Change (without China’s Investment Drive), in percent | |
|---|---|---|---|
| (A) | (B) | (B–A) | |
| Zinc | 75.1 | 16.5 | –58.6 |
| Nickel | 60.6 | 8.4 | –52.2 |
| Lead | 59.3 | 14.7 | –44.7 |
| Copper | 52.6 | 26.7 | –25.9 |
| Aluminum | 33.4 | –6.9 | –40.3 |
| Iron ore | 24.3 | 172.6 | 148.3 |
| Silver | 20.3 | 135.1 | 114.8 |
| Rubber | 18.4 | 84.3 | 65.9 |
| Gold | 4.0 | 79.9 | 75.9 |
Impact on Metal Prices of China’s Investment, 2008–11
| Model’s Implied Difference from Counterfactual, in percent | Actual Change, in percent | Counterfactual Change (without China’s Investment Drive), in percent | |
|---|---|---|---|
| (A) | (B) | (B–A) | |
| Zinc | 75.1 | 16.5 | –58.6 |
| Nickel | 60.6 | 8.4 | –52.2 |
| Lead | 59.3 | 14.7 | –44.7 |
| Copper | 52.6 | 26.7 | –25.9 |
| Aluminum | 33.4 | –6.9 | –40.3 |
| Iron ore | 24.3 | 172.6 | 148.3 |
| Silver | 20.3 | 135.1 | 114.8 |
| Rubber | 18.4 | 84.3 | 65.9 |
| Gold | 4.0 | 79.9 | 75.9 |
References
Ahuja, A., and A. Myrvoda, 2012, “The Spillover Effects of a Downturn in China’s Real Estate Investment,” Working Paper 12/266, (Washington: International Monetary Fund).
Guo, Kai, 2011, “Factor Pricing, Overcapacity, and Sustainability Risks,” People’s Republic of China Spillover Report—Selected Issues (unpublished; Washington: International Monetary Fund).
International Monetary Fund, 2011, “People’s Republic of China: 2011 Article IV Consultation,” Country Report 11/192 (Washington).
International Monetary Fund, 2012, “Is China Rebalancing? Implications for Asia,” in Asia and Pacific Regional Economic Outlook, April (Washington).
The nontradables sector is defined to include utilities, construction, transport and storage, information technology, wholesale and retail trade, catering, banking and insurance, real estate, leasing and commercial services, education, health care, sport and entertainment, and public administration.
More detailed descriptions of the model and estimation strategy can be found in Ahuja and Myrvoda (2012).
A one standard deviation shock is equivalent to 1.2 percentage points in month-over-month, seasonally adjusted, growth rates.
Industrial production is defined differently from country to country. The Organization for Economic Cooperation and Development definition includes production in mining, manufacturing, and public utilities (electricity, gas, and water), but excludes construction.
Exports to China are mostly in machinery, equipment, and industrial supplies for the United Kingdom and mineral commodities and primary metal products for India. Canada’s exports to China are more diversified in minerals and manufactured commodities.
