Empirical Evidence on Growth Spillovers from China to New Zealand
Abstract
EconStor is a publication server for scholarly economic literature, provided as a non-commercial public service by the ZBW.
Full text
Osborn, Denise R; Vehbi, Tugrul Working Paper Empirical Evidence on Growth Spillovers from China to New Zealand New Zealand Treasury Working Paper, No. 13/17 Provided in Cooperation with: The Treasury, New Zealand Government Suggested Citation: Osborn, Denise R; Vehbi, Tugrul (2013) : Empirical Evidence on Growth Spillovers from China to New Zealand, New Zealand Treasury Working Paper, No. 13/17, ISBN 978-0-478-40361-9, New Zealand Government, The Treasury, Wellington This Version is available at: https://hdl.handle.net/10419/205641 Standard-Nutzungsbedingungen: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may be saved and copied for your personal and scholarly purposes. You are not to copy documents for public or commercial purposes, to exhibit the documents publicly, to make them publicly available on the internet, or to distribute or otherwise use the documents in public. If the documents have been made available under an Open Content Licence (especially Creative Commons Licences), you may exercise further usage rights as specified in the indicated licence. https://creativecommons.org/licenses/by/4.0/
Empirical Evidence on Growth Spillovers from China to New Zealand Denise R Osborn and Tugrul Vehbi New Zealand Treasury Working Paper 13/17 July 2013
NZ TREASURY WORKING PAPER 13/17 Empirical Evidence on Growth Spillovers from China to New Zealand MONTH / YEAR July 2013 AUTHORS Denise R Osborn Economics, School of Social Sciences University of Manchester Manchester M13 9PL United Kingdom Tugrul Vehbi New Zealand Treasury PO Box 3724 Wellington New Zealand Email Telephone [email protected] 64-4-917 6149 ISBN (O NLINE ) 978-0-478-40361-9 URL Treasury website at July 2013: http://www.treasury.govt.nz/publications/research-policy/wp/2013/13-17 Persistent URL: http://purl.oclc.org/nzt/p-1580 ACKNOWLEDGEMENTS The authors would like to thank Enzo Cassino, Patrick Conway, Mardi Dungey, Ozer Karagedikli, Peter Mawson, David Oxley and Michael Ryan for their helpful comments and suggestions. NZ TREASURY New Zealand Treasury PO Box 3724 Wellington 6008 NEW ZEALAND Email Telephone Website [email protected] 64-4-472 2733 www.treasury.govt.nz DISCLAIMER The views, opinions, findings, and conclusions or recommendations expressed in this Working Paper are strictly those of the author(s). They do not necessarily reflect the views of the New Zealand Treasury or the New Zealand Government. The New Zealand Treasury and the New Zealand Government take no responsibility for any errors or omissions in, or for the correctness of, the information contained in these working papers. The paper is presented not as policy, but with a view to inform and stimulate wider debate.
WP 13/17 | EMPIRICAL EVIDENCE ON GROWTH SPILLOVERS FROM CHINA TO NEW ZEALAND i Abstract This paper provides a quantitative analysis of the impact on New Zealand of economic growth in China through the framework of an econometric model. The analysis compares the roles of China and the US both for growth in New Zealand and also for world commodity prices, the latter being important for New Zealand as an exporter of primary products. Finally, in the light of the increasing role of China in the world economy over the last two to three decades, the paper also investigates whether spillover effects from China to New Zealand have changed over this period. Using models estimated from the mid- 1980s to 2011, we find that growth spillovers from China are important for New Zealand, with estimates of the accumulated increase in domestic GDP from a one percent increase in output growth in China being in the range of around 0.2 to 0.4 percent. It is striking that growth spillovers are substantially greater from the US than from China, despite the latter's increasing importance in the world economy. Both domestic and foreign shocks have been important drivers of real exchange rate fluctuations, while the contribution of the latter has been relatively more important. The time-varying estimates provide some evidence of time-variation, with the greatest impact from China applying for about a decade from the mid-1990s, but also being relatively large in the latter part of our sample period. JEL CLASSIFICATION C32, E32, F43, F44 KEYWORDS Structural VAR, growth spillovers, commodity prices
WP 13/17 | EMPIRICAL EVIDENCE ON GROWTH SPILLOVERS FROM CHINA TO NEW ZEALAND ii Table of Contents Abstract ............................................................................................................................... i Executive Summary ........................................................................................................... 1 1Introduction .............................................................................................................. 2 2Methodology ............................................................................................................. 4 3Data and Preliminary Analysis ................................................................................ 7 3.1Growth rate correlations ................................................................................................. 7 3.2Commodity prices ........................................................................................................... 8 4SVAR Results ......................................................................................................... 10 4.1Baseline model .............................................................................................................10 4.2Transmission via commodity prices ..............................................................................15 4.3Historical decomposition ...............................................................................................17 4.4Further commodity price analysis .................................................................................19 5Time-Varying Estimation ....................................................................................... 22 6Conclusion .............................................................................................................. 24 References ....................................................................................................................... 25 Appendix .......................................................................................................................... 27 List of Tables Table 1: Estimated cumulative responses of real commodity prices to US and China GDP shocks ...............................................................................................................................................21 Table 2: Estimated cumulative domestic GDP responses to China and US shocks in models with aggregate and sub-group commodity price indices ...................................................................21 Appendix Table 1 – Data description ................................................................................................27 List of Figures Figure 1: Correlations of New Zealand GDP growth with China, USA and Australia ......................... 8 Figure 2: Real commodity prices and New Zealand terms of trade .................................................... 9 Figure 3: SVAR Responses to a China GDP shock .........................................................................11 Figure 4: Responses to a US GDP shock .........................................................................................12 Figure 5: Responses to Australia GDP shock ...................................................................................13 Figure 6: Accumulated GDP responses to growth shocks in the US and China ..............................14 Figure 7: Responses to China GDP shock with commodity price channel .......................................16 Figure 8: Responses to US GDP shock with commodity price channel ...........................................17 Figure 9: Contributions of foreign shocks to New Zealand's economic growth ................................18 Figure 10: Contributions of foreign shocks to New Zealand's real exchange rate ...........................19 Figure 11: Time-varying four-quarter cumulative impact of China GDP shock ................................23 Appendix Figure 1 – Schematic description of the extended model .................................................28
WP 13/17 | EMPIRICAL EVIDENCE ON GROWTH SPILLOVERS FROM CHINA TO NEW ZEALAND 1 Executive Summary This paper provides a quantitative analysis of the impact on New Zealand of economic growth in China through the framework of an econometric model. It is the third of three working papers looking at the impact of China on the New Zealand economy. The analysis compares the roles of China and the US both for growth in New Zealand and also for world commodity prices. Using models estimated from the mid-1980s to 2011, growth spillovers from China are important for New Zealand, with estimates of the accumulated increase in domestic GDP from a one percent increase in output in China being in the range of 0.2 to 0.4 percent. It is striking that growth spillovers are substantially greater from the US than from China, despite the latter's increasing importance in the world economy. The impact of China on global commodity prices has been steadily increasing over time, with growth in China having strongest effects on dairy and aluminium price inflation. US demand, on the other hand, is generally more important than China in driving global commodity prices. Both domestic and foreign shocks are important drivers of real exchange rate fluctuations, the contribution of the latter (particularly commodity prices) is relatively more important. Alongside the broader descriptive analysis by Bowman and Conway (2013a, 2013b), these results highlight the growing importance of China for the contemporary New Zealand economy.
WP 13/17 | EMPIRICAL EVIDENCE ON GROWTH SPILLOVERS FROM CHINA TO NEW ZEALAND 2 Empirical Evidence on Growth Spillovers from China to New Zealand 1 Introduction China is now the second largest economy in the world at current exchange rates and, according to the International Monetary Fund, is forecast to overtake the United States later in the next decade in purchasing power parity (PPP) terms. Indeed China's economic performance over the last two decades has been remarkable, accounting for 3.9% of world gross domestic product (GDP) in 1990 but 15.0% in 2012.1These figures would be striking at any time, but in the context of faltering global growth since the onset of the 2008 financial crisis, they underline the importance of continued growth in China for the world economic outlook. China is now the second largest market for New Zealand's product exports, behind only Australia. In discussing this, Bowman and Conway (2013a, 2013b) point particularly to the increased industrialisation and urbanisation of China. This, alongside increased per capita income and a changing diet, has led to rising demand for primary products from New Zealand, especially dairy, meat and forestry products. These effects are reinforced because Australia has also received substantial benefits from China's growth, with these most evident in its mining sector.2 It is to be anticipated, therefore, that over recent years New Zealand has received positive spillovers from growth in China not only by increasing exports to that country, but also through effects on its long-term major trading partner, namely Australia. This paper provides a quantitative analysis of the impact on New Zealand of economic growth in China through the framework of an econometric model. This complements the more descriptive analysis of Bowman and Conway (2013a, 2013b) by providing estimates of, for example, the effect on New Zealand of a one percent increase (or decrease) in China's output growth, comparing this with estimated effects from growth in the US. The analysis also compares the roles of China and the US for world commodity prices, which are important for New Zealand as an exporter of primary products. Finally, in the light of the increasing role of China in the world economy over the last two to three decades, the 1 The projections in the database accompanying International Monetary Fund (2013) show China accounting for 18.3% of world GDP (based on a purchasing power parity valuation) in 2017, compared with 17.9% for the US. The 1990 and 2012 figures for China are also from this source, which shows the US share of world GDP share declining from 24.7% to 18.9% over the same period. 2 Plumb, Kent and Bishop (2012) discuss the implications for Australia of strong growth in Asia. Specifically, they use recent data to illustrate the current position in relation to the three phases of adjustment predicted by macroeconomic theory for a small resource-rich open economy that is subject to a commodity price boom.
WP 13/17 | EMPIRICAL EVIDENCE ON GROWTH SPILLOVERS FROM CHINA TO NEW ZEALAND 3 paper also investigates whether spillover effects from China to New Zealand have changed over this period. A large literature exists on the role of the US in the international economy, with that country widely assumed to drive world economic growth. However, despite it now being the world's second largest economy, surprisingly few studies are currently available focusing on the role of China and, further, their results are somewhat ambiguous. In particular, Arora and Vamvakidis (2011) find large growth spillovers from China to both the rest of Asia and the world (38 and 172 countries, respectively) over the last two decades. On the other hand, although Sato, Zhang and McAleer (2011) detect evidence that the impact of China on other East Asian economies has increased since 1978, they conclude that growth in China plays a small role for that region relative to the US. Sun (2011) focuses on effects in New Zealand and Australia, studying the role of growth in 'emerging Asia' (including China) versus that in the US. Although Sun (2011) finds that 'emerging Asia' plays a stronger role than the US for Australia in the decade from 2000, she detects no impact from that region on New Zealand over the same period. This leads her to conclude that growth in 'emerging Asia' affects New Zealand only indirectly through its impact on Australia. Nevertheless, this is a surprising finding in the context of the different products exported by these two antipodean countries to China and that country's role as the second largest market for New Zealand exports. The small group of papers mentioned in the preceding paragraph use broadly similar modelling methodologies based on vector autoregressive (VAR) models. Nevertheless, there are important differences across their approaches, as discussed in the next section. The present paper also employs a VAR modelling framework, but imposes restrictions in order to capture adequately the effect of a dominant world economy (China and/or the US, as appropriate) on New Zealand. Since Australia is New Zealand's largest trading partner, and in the light of the results of Sun (2011), our analysis also examines the role played by that country. Although previous VAR analyses for New Zealand, such as Buckle, Kim, Kirkham, McLellan and Sharma (2007) or Dungey and Fry (2009), employ a larger range of domestic variables than in the current study, more parsimonious specifications of the domestic sector are employed here to facilitate more detailed examination of international spillovers. In particular, those studies employ a single 'international' or 'foreign' economy, whereas we investigate separate effects for China, the US and Australia on New Zealand. The paper is organised as follows. Section 2 discusses methodological issues in relation to previous studies and also the nature of the VAR analysis of this paper. This is followed by a discussion of the data employed, including some preliminary analysis, in Section 3. Our principal results are contained in Section 4, which examines spillovers to New Zealand from China and the US through a range of VAR models, focusing particularly on a specification that includes commodity prices. The following section then focuses on changes over time, with conclusions drawn in Section 6.
WP 13/17 | EMPIRICAL EVIDENCE ON GROWTH SPILLOVERS FROM CHINA TO NEW ZEALAND 4 2 Methodology In line with other empirical studies of cross-country growth spillovers, this paper employs a VAR methodology, which explicitly allows for dynamic interactions between variables. The main tool employed is the so-called structural vector autoregressive (SVAR) model, which in our context is used to specify that a small country (specifically New Zealand) can be affected by, but does not influence, growth in a large country (China or the US, say). Representing output growth for the foreign country and New Zealand in a particular period t (typically a quarter) by yF,t and yNZ,t, respectively, the form of the SVAR3 often used in this context can be written as , , , , , , , , , (1) in which yNZ,t depends on contemporaneous yF,t (through the term , in the second equation), but not vice versa, while the lagged variables capture cross-country and internal growth dynamics. Additional lags, beyond one period, can easily be accommodated in this system and are also discussed below. Each equation is a dynamic regression model and, provided that the SVAR captures all the dynamics of growth, the disturbances uF,t and uNZ,t are uncorrelated over time. Further, through the imposition of the assumption that contemporaneous causality runs from yF,t to yNZ,t, these disturbances are also mutually uncorrelated. Since they cannot be predicted by the system, uF,t and uNZ,t are often referred to as 'shocks'. The VAR then permits estimation of the effects over time of a shock (such as a 1% increase) in output growth in the large foreign country on growth in the small country. The form of the SVAR given in (1) embeds economic information through the causality assumption that the small open economy will not have contemporaneous feedback to the large economy4. An assumption of this type is made by virtually all papers concerned with international growth spillovers to small countries, including Sun (2011). However, we prefer to impose a stronger assumption. Although the contemporaneous causality is assumed to apply only from the large to the small economy, the SVAR model of (1) treats the two countries in a symmetric way in terms of potential feedbacks or spillovers over time. However, it is a priori implausible that a small country like New Zealand will affect growth in China or the US in any way. This implies that the restriction b12 = 0 (with corresponding zero restrictions also on any further lags) should be imposed in (1), so that the system becomes , , , , , , , , (2) and output growth in the large economy is influenced by its own past, but not yNZ,t-1. Restrictions of this form, often referred to as exogeneity restrictions, were popularised by Cushman and Zha (1997) in the context of modelling Canada and the US. Such exogeneity restrictions are imposed in the New Zealand studies of Buckle et al. (2007) and Dungey and Fry (2009), while Sato et al. (2011) use the form of (2) when analysing 3 The model also includes an intercept in each equation, which are omitted from (1) for notational convenience. 4 The SVAR form of (1) is often obtained as a Cholesky decomposition of the covariance matrix of the disturbances in a conventional VAR that includes only lagged variables.
WP 13/17 | EMPIRICAL EVIDENCE ON GROWTH SPILLOVERS FROM CHINA TO NEW ZEALAND 11 more persistent. The peak impact occurs in this case in the quarter after the shock and the effect is statistically significant for five quarters. The persistence and the pattern of effects can be partially explained by the possible lagged effect of the mining industry on growth in Australia. In other words, the nature of Australia’s commodities means there needs to be investment to supply them and therefore longer lag times for production. This contrasts with New Zealand’s ‘soft’ commodities where supply can be ramped up relatively quickly up to a point constrained by biological processes. The implication is that much of the growth spillovers from China to New Zealand are direct effects, rather than operating indirectly through Australia, as suggested by Sun (2011) for spillovers from 'Emerging Asia'. Such a direct role is also implied by the current importance of China as a destination for New Zealand exports. With the increase in domestic output, Figure 3 shows mild but insignificant increases in domestic inflation, although with a stronger response in New Zealand interest rates and a highly significant appreciation in the exchange rate. The appreciation in the exchange rate is important, as it acts as a buffer against the inflationary impact that would otherwise apply due to the additional stimulus. Figure 3: SVAR Responses to a China GDP shock Notes: The figure depicts the impulse responses to a one-percentage point orthogonalised shock to China’s GDP. The baseline SVAR model is estimated using data for 1982Q1-2011Q4. The solid line shows the estimated mean responses, with the shaded bands indicating the 90 percent confidence interval, obtained using 2000 Monte Carlo replications. Responses are shown for the quarter of the shock (quarter 1) and 9 subsequent quarters. Source: Authors’ calculations. Using the same SVAR specification, but replacing China with the US as the international driving force, Figure 4 shows the estimated responses to a one percentage point shock to US output, and these responses differ substantially from those seen in Figure 3. The impact of the US shock (Figure 4) on both domestic New Zealand variables and also Australian GDP is relatively longer-lived and statistically more significant; however, the impact on New Zealand within the quarter is lower. In particular, after an initial estimated 12345678910 0.2 0.4 0.6 0.8 1 Percentage pts. China GDP 12345678910 0.05 0.1 0.15 0.2 0.25 Percentage pts. Australian GDP 12345678910 0 0.1 0.2 0.3 Percentage pts. NZ GDP 12345678910 -0.05 0 0.05 0.1 0.15 0.2 0.25 Percentage pts. NZ Inflation 12345678910 0 0.2 0.4 0.6 0.8 Percentage pts. Quarters NZ Interest Rate 12345678910 -0.2 0 0.2 0.4 0.6 0.8 Percentage pts. Quarters NZ Real Exchange Rate
WP 13/17 | EMPIRICAL EVIDENCE ON GROWTH SPILLOVERS FROM CHINA TO NEW ZEALAND 12 increase of 0.13 percentage points that applies in the quarter of the shock, the peak output response is at a lag of one quarter (labelled 2 in the graph) and domestic New Zealand GDP continues to increase due to positive impulse responses for a total of around 5 quarters; the cumulative response of domestic output to the shock originating in the US is estimated to be approximately 0.66. This, combined with the positive and significant response of inflation to the output increase after three or four quarters, leads to an increase in interest rates of around one percentage point. The exchange rate appreciates on impact, helping to keep inflation low, with further responses not being significant. Figure 4: Responses to a US GDP shock Notes: The figure depicts the impulse responses to a one-percentage point orthogonalised shock to US GDP. The SVAR model is estimated using data for 1982Q1-2011Q4. The solid line shows the estimated mean responses, with the shaded bands indicating the 90 percent confidence interval, obtained using 2000 Monte Carlo replications. Responses are shown for the quarter of the shock (quarter 1) and 9 subsequent quarters. Source: Authors’ calculations. Notwithstanding the importance of China in relation to New Zealand trade over the recent past, a comparison of Figure 3 with Figure 4 emphasises the key role played by the US economy for GDP growth in both New Zealand and Australia over the three decades of the sample period. Indeed, while the growth spillovers from China to New Zealand are found to have an important role for domestic variables beyond output, those from the US are more pervasive, with both inflation and monetary policy responding due to the strength of the growth spillovers. Interestingly, New Zealand's exchange rate responds in a similar way, irrespective of whether the growth shock originates in China or the US, hinting at the role of a common channel. Finally, Figure 5 shows the responses to a one percentage point shock to Australian GDP, estimated from the model including the US as the major international economy. These responses follow broadly similar patterns to those seen in Figure 3 and Figure 4, with 1 2 3 4 5 6 7 8 9 10 0.2 0.4 0.6 0.8 1 Percentage pts. US GDP 1 2 3 4 5 6 7 8 9 10 0.1 0.2 0.3 0.4 0.5 Percentage pts. Australian GDP 1 2 3 4 5 6 7 8 9 10 0 0.2 0.4 0.6 Percentage pts. NZ GDP 1 2 3 4 5 6 7 8 9 10 -0.1 0 0.1 0.2 0.3 Percentage pts. NZ Inflation 1 2 3 4 5 6 7 8 9 10 0.5 1 1.5 Percentage pts. Quarters NZ Interest Rate 1 2 3 4 5 6 7 8 9 10 -0.4 -0.2 0 0.2 0.4 0.6 Percentage pts. Quarters NZ Real Exchange Rate
WP 13/17 | EMPIRICAL EVIDENCE ON GROWTH SPILLOVERS FROM CHINA TO NEW ZEALAND 13 positive growth spillovers. Although the response of New Zealand to an Australian output shock (Figure 5) is stronger than to one from China (Figure 3), the patterns of a relatively quick response of domestic output alongside insignificant effects on inflation are shared by both these figures. However, domestic interest rates respond more strongly and the exchange rate appreciation is substantially more marked and significant when the output shock originates from Australia compared with China. Figure 5: Responses to Australia GDP shock Notes: The figure depicts the impulse responses to a one-percentage point orthogonalised shock to Australia’s GDP. The SVAR model is estimated using data for 1982Q1-2011Q4. The solid line shows the estimated mean responses, with the shaded bands indicating the 90 percent confidence interval, obtained using 2000 Monte Carlo replications. Responses are shown for the quarter of the shock (quarter 1) and 9 subsequent quarters. Source: Authors’ calculations. Across the three sets of impulse responses, the important role of the US in the world economy over this period is reflected in the stronger overall domestic output response to this shock than others, and also in the greater responses of domestic New Zealand inflation seen in Figure 4 compared with Figures 3 and 5. In other words, the source of the foreign growth shock matters. Also, and perhaps surprisingly, shocks originating in Australia have less impact overall on the New Zealand economy (except for the exchange rate) than those originating in the US. A further comparison of the growth spillovers from the two major foreign sources of shocks are shown in Figure 6, where the corresponding impulse responses seen in Figures 3 and 4 are aggregated over lags. In addition to those for New Zealand, the figure includes the own China or US responses and also the responses of Australia obtained from each model. While the own responses of the US or China, respectively, are greater than the spillovers, the effects on Australia are (except for the initial effect of a China 1 2 3 4 5 6 7 8 9 10 -1 -0.5 0 0.5 1 Percentage pts. China GDP 1 2 3 4 5 6 7 8 9 10 0.2 0.4 0.6 0.8 1 Percentage pts. Australian GDP 1 2 3 4 5 6 7 8 9 10 0 0.1 0.2 0.3 0.4 Percentage pts. NZ GDP 1 2 3 4 5 6 7 8 9 10 0 0.1 0.2 Percentage pts. NZ Inflation 1 2 3 4 5 6 7 8 9 10 0.2 0.4 0.6 0.8 1 Percentage pts. Quarters NZ Interest Rate 1 2 3 4 5 6 7 8 9 10 0 0.5 1 Percentage pts. Quarters NZ Real Exchange Rate
WP 13/17 | EMPIRICAL EVIDENCE ON GROWTH SPILLOVERS FROM CHINA TO NEW ZEALAND 14 shock) larger than on New Zealand. Further, according to these models, the response of output in both New Zealand and Australia to a China output shock is about half that of a US output shock. This may reflect the role of the US in the world economy over this period, with effects originating there having spillovers to many countries, which in turn influence growth in New Zealand and Australia. Furthermore, the more persistent nature of US shocks relative to China shocks may also be a contributing factor to the greater New Zealand response. The faster pattern of response of domestic New Zealand output when the shock originates in China may reflect increased demand from China for New Zealand products (such as dairy products), whereas that from the US may operate more indirectly through other countries. Figure 6: Accumulated GDP responses to growth shocks in the US and China Notes: Impulse responses are accumulated from those shown in Figure 4 and Figure 3 for US and China shocks, respectively. Source: Authors’ calculations. The overall conclusion from the baseline SVAR model, using data over the three decades from 1982 to 2011, is that growth spillovers from China to New Zealand are non-trivial, with a one percentage point shock to China's growth estimated to increase New Zealand's GDP by around 0.3 percentage points within a one year period. Further, this effect on New Zealand applies relatively quickly, and with little significant impact on other domestic variables beyond an immediate appreciation of the real exchange rate. On the other hand, shocks originating in the US or Australia have greater spillovers to New Zealand's growth, with consequent effects also on domestic interest rates in addition to the exchange rate. The implication of Figure 6 is also that growth in New Zealand has an overall slightly lower response to shocks from the major economies of the US and China than does Australia particularly in the medium term. It is important to note that the analysis so far has concentrated on the spillover impacts on the volume of domestic output. This, on the other hand, is only part of the story as the transmission channel via commodity prices has an important impact on domestic nominal GDP as well as second-round impacts on volumes via an income effect. The next subsection extends the investigation to include a commodity price channel, in order to further explain these response patterns. 0 2 4 6 8 10 12 14 16 18 20 0 0.5 1 1.5 2 2.5 percentage points (a) US shock US Australia New Zealand 0 2 4 6 8 10 12 14 16 18 20 0 0.5 1 1.5 percentage points (b) China shock China Australia New Zealand
WP 13/17 | EMPIRICAL EVIDENCE ON GROWTH SPILLOVERS FROM CHINA TO NEW ZEALAND 15 4.2 Transmission via commodity prices As noted in subsection 3.2, the evolution of commodity prices is important for the New Zealand economy. Indeed, as discussed by Bowman and Conway (2013a, 2013b), growth in China over the last decade may be particularly important as an explanation of the rise in commodity prices, which affects New Zealand as an exporter of primary products. Related to this, Roache (2012) investigates the impact of shocks to aggregate activity in China on real prices of oil and base metals (relative to the US consumer price index) for the period 2000-2011; while shocks to aggregate activity in China have a significant and persistent short-run impact on the prices of oil and copper, they are generally insignificant for other metals. However, Roache (2012) also finds that effects are larger for a demand shock originating in the United States than in China. In the Australian context, the results of Sun (2011) indicate that commodity prices are the most important channel for the transmission of shocks from emerging Asia to Australia over her sample from 2000, with some results (Sun, 2011, Figure 9) suggesting that these are important also for the transmission of shocks from that region to New Zealand. Nevertheless, this latter analysis is not entirely satisfactory in that commodity prices are treated as exogenous to even the world's largest economies. Against this background, and also the results of our baseline SVAR presented in subsection 4.1, it is useful to investigate empirically the role of commodity prices in the transmission of growth spillovers to the New Zealand economy. Since Roache (2012) finds that the US is more important for the evolution of commodity prices than China, our SVAR model includes GDP growth for both of these large economies in a single specification and, reflecting their importance for world demand, the model allows growth in both China and the US to contemporaneously influence commodity prices. In recognition of the historical role of the US for the world business cycle, contemporaneous causality between these countries runs from the US to China. However, unrestricted lags apply across these two countries and commodity prices. This block is exogenous for Australia and New Zealand, so that growth in China and the US affects these countries, but not vice versa. Due to the availability of commodity price data, the estimation period for the extended SVAR starts in 1986. A schematic representation of the model description is provided in Appendix Figure 1. Figure 7 shows the estimated responses of all eight variables to a one percentage point shock to China's growth rate in this extended model that employs the real aggregate ANZ commodity price index. Although there is virtually no effect on the US from the China growth shock, the commodity price index responds positively, increasing by about 0.7 percent within the quarter, and the effect is rather persistent and statistically significant. The pattern of response of domestic output is similar to the output response to a China shock shown in Figure 3, albeit with the estimated mean response being a little lower than in the baseline case. The real exchange rate immediately appreciates in response to the increase in China's output and the associated boost in commodity prices, while interest rates and inflation responses are small and insignificant.
WP 13/17 | EMPIRICAL EVIDENCE ON GROWTH SPILLOVERS FROM CHINA TO NEW ZEALAND 16 Figure 7: Responses to China GDP shock with commodity price channel Notes: The figure depicts the impulse responses to a one-percentage point orthogonalised shock to China’s GDP. The SVAR model includes the aggregate commodity price index and is estimated using data over 1986Q1 to 2011Q4. The solid line shows the estimated mean response, with the shaded bands indicating the 90 percent confidence interval, obtained using 2000 Monte Carlo replications. Responses are shown for the quarter of the shock (quarter 1) and 9 subsequent quarters. Source: Authors’ calculations. A comparison of the responses to a US shock in Figure 8 with those just discussed for a China shock reveals a number of interesting differences. Firstly, in line with the results of Roache (2012) for oil and base metals, the response of commodity prices to a US growth shock in Figure 8 is around double that seen when growth originates in China (Figure 7). Although this stronger effect is partly due to the contemporaneous causal ordering we adopt between these large economies, it also reflects the wider impact of the US on the world economy over most of the 1986-2011 period8. Secondly, the peak GDP responses of New Zealand occur in the quarter after the shock, in line with the estimates from the baseline model of subsection 4.1, but in contrast to the responses to a China shock. 8 The contemporaneous response of about 0.4 percent China to the US growth shock in Error! Reference source not found. is not sufficient to account for the different responses of commodity prices in the two figures. 1 2 3 4 5 6 7 8 9 10 -0.1 0 0.1 Percentage pts. US GDP 1 2 3 4 5 6 7 8 9 10 0 0.5 1 Percentage pts. China GDP 1 2 3 4 5 6 7 8 9 10 0 0.5 1 Percentage pts. Aggregate Com. Price 1 2 3 4 5 6 7 8 9 10 0 0.1 0.2 Percentage pts. AUS GDP 1 2 3 4 5 6 7 8 9 10 0 0.1 0.2 Percentage pts. NZ GDP 1 2 3 4 5 6 7 8 9 10 -0.1 0 0.1 Percentage pts. NZ Inflation 1 2 3 4 5 6 7 8 9 10 -0.2 0 0.2 Percentage pts. Quarters NZ Interest rate 1 2 3 4 5 6 7 8 9 10 0 0.2 0.4 0.6 0.8 Percentage pts. Quarters NZ Real Exchange Rate
WP 13/17 | EMPIRICAL EVIDENCE ON GROWTH SPILLOVERS FROM CHINA TO NEW ZEALAND 17 Figure 8: Responses to US GDP shock with commodity price channel Notes: The figure depicts the impulse responses to a one-percentage point orthogonalised shock to US GDP. The SVAR model includes the aggregate commodity price index and is estimated using data over 1986Q1 to 2011Q4. The solid line shows the estimated mean response, with the shaded bands indicating the 90 percent confidence interval, obtained using 2000 Monte Carlo replications. Responses are shown for the quarter of the shock (quarter 1) and 9 subsequent quarters. Source: Authors’ calculations. Further, using the extended model Figure 8 reveals some differences in responses to US shocks compared with the baseline case (Figure 4. In particular, in omitting both China and commodity prices when analysing the responses to US shocks, the baseline model conflates the appreciation of New Zealand's real exchange rate that operates through those sources with an anticipated negative direct effect from a US shock, with these shocks more adequately isolated in the extended model. These results suggest that commodity prices play an important role in the transmission of shocks from both China and the US to New Zealand. These effects operate partly through Australia, which itself strongly influences Australia’s domestic growth (Figure 5). The positive and significant response of the real exchange rate in Figure 7 is another manifestation of the importance of commodity prices. Although results are not shown, a one percent shock to the real commodity price index increases this variable by around 0.4 percent, which is significant and implies that this channel makes an important contribution to the impact response of the real exchange rate seen in Figure 7. These results on the role of commodity prices are in line with findings of Karagedikli and Price (2012) and Jääskelä and Smith (2011), who study the sources of terms of trade shocks for New Zealand and Australia, respectively. 4.3 Historical decomposition Historical decompositions measure the contributions of different structural shocks to realisations of endogenous variables. In order to shed further light on the importance of the different shocks, Figures 9 and 10 show the estimated historical decompositions of New Zealand's GDP and real exchange rate growth in terms of the contributions of each foreign shock, as identified using the extended SVAR model of the preceding subsection. 12345678910 0 0.5 1 Percentage pts. US GDP 12345678910 0 0.2 0.4 Percentage pts. China GDP 12345678910 0 0.5 1 1.5 Percentage pts. Aggregate Com. Price 12345678910 0 0.2 0.4 Percentage pts. AUS GDP 12345678910 0 0.2 0.4 Percentage pts. NZ GDP 12345678910 -0.1 0 0.1 0.2 Percentage pts. NZ Inflation 12345678910 0 0.5 1 Percentage pts. Quarters NZ Interest rate 12345678910 0 0.2 0.4 0.6 Percentage pts. Quarters NZ Real Exchange Rate
WP 13/17 | EMPIRICAL EVIDENCE ON GROWTH SPILLOVERS FROM CHINA TO NEW ZEALAND 18 In each case, growth rates of quarterly actual GDP and the real exchange rate from 1986 to 2011 are shown as the first panels of the figures, with these growth rates then decomposed into the effects as sourced by the three of the eight separate shocks of the model. Figure 9: Contributions of foreign shocks to New Zealand's economic growth Notes: The historical decomposition relates to the sources of shocks as identified in the SVAR model of subsection 4.2. Source: Authors’ calculations. In line with the impulse responses seen in Figure 7 and Figure 8, shocks originating in US have been relatively more important overall for New Zealand GDP than those from the China. The positive contribution of shocks from China’s growth after 2005 and the negative impact of US growth shocks during the global financial crisis are evident. Shocks to Australia’s GDP have had a sizeable contribution throughout the sample period. Turning now to the real exchange rate, Figure 10 shows that foreign shocks play a major role in the dynamics of New Zealand’s real exchange rate while domestic shocks are relatively less important (not shown but are available upon request). 1990 1995 2000 2005 2010 -3 -2 -1 0 1 2 3 Percentage pts. Actual NZ GDP (qpc,%) 1990 1995 2000 2005 2010 -1 -0.5 0 0.5 1 Percentage pts. US GDP Shock 1990 1995 2000 2005 2010 -1 -0.5 0 0.5 1 Percentage pts. China GDP Shock 1990 1995 2000 2005 2010 -1 -0.5 0 0.5 1 Percentage pts. AUS GDP Shock
WP 13/17 | EMPIRICAL EVIDENCE ON GROWTH SPILLOVERS FROM CHINA TO NEW ZEALAND 19 Figure 10: Contributions of foreign shocks to New Zealand's real exchange rate Notes: See Figure 9. Source: Authors’ calculations. 4.4 Further commodity price analysis The analysis of subsections 4.2 and 4.3 illustrates the importance of world commodity prices for the New Zealand economy, particularly for movements in the real exchange rate. However, it is plausible that different sub-categories of commodity prices may affect the New Zealand economy in different ways. In particular, Bowman and Conway (2013a, 2013b) highlight changes associated with China's demand for dairy and forestry products from New Zealand. Commodity prices, particularly those relating to mineral products, are important also for the Australian economy, which in turn affects New Zealand. The present subsection therefore analyses sub-groups of the ANZ commodity price index, specifically dairy, forestry, meat and wool, and aluminium. Table 1 shows the accumulated impact of shocks to the growth rates of both China and US on these commodity prices; those shown for the aggregate index in Table 1 are accumulated from the responses shown in Figures 7 and 8, with those for commodity categories obtained in the same way, but using the respective sub-group world price index in the SVAR model of subsection 4.2. The effects of demand shocks originating in both countries have statistically significant impacts on all commodity price series. On the other hand, the impact varies across sectors, with the biggest responses being for aluminium and dairy products. A one percentage point shock to growth rates in the US and China leads to accumulated increases in the real price of aluminium within four quarters of 12.6 and 5.5 percentage points respectively. The corresponding increases in dairy products are 6.7 and 5.6 percentage points respectively. The relatively high elasticity for aluminium products may be due to the rapid growth of China’s manufacturing sector during the period studied here. While the response of forestry product prices is also sizeable in both cases, the impacts on the meat and wool category are relatively muted. As discussed in 1990 1995 2000 2005 2010 -10 -5 0 5 Percentage pts. Actual NZ Real Exchange Rate (qpc,%) 1990 1995 2000 2005 2010 -3 -2 -1 0 1 2 3 Percentage pts. US GDP Shock 1990 1995 2000 2005 2010 -3 -2 -1 0 1 2 3 Percentage pts. China GDP Shock 1990 1995 2000 2005 2010 -3 -2 -1 0 1 2 3 Percentage pts. AUS GDP Shock
WP 13/17 | EMPIRICAL EVIDENCE ON GROWTH SPILLOVERS FROM CHINA TO NEW ZEALAND 20 Bowman and Conway (2013a), wool would have been influenced by China in late 1980s but more stable since then even though they are the dominant market. China, on the other hand, is becoming more important for meat exports but the effect might not have shown up yet. Overall, the results in Table 1 show that US demand is generally more important than China in driving global commodity prices, whether these prices are examined through the aggregate ANZ index or through sub-indices relevant to New Zealand. This partly reflects the larger international spillovers generated by US growth to other major economies, indicated in Figure 8 for China, and is in line with the findings of Roache (2012). It is important to note here that our analysis is silent about the supply side responses that may have an impact on the movements in commodity prices. Nevertheless, the effects on commodity prices of growth from these major economies are only part of the story, since the focus of interest of this study is the effects on the New Zealand economy itself. Therefore, Table 2 provides the comparable accumulated responses of domestic GDP growth to US and China shocks in the SVAR model with commodity prices, both when the aggregate commodity price series is used (as in subsection 4.2) and when sub-group indices are employed in place of the aggregate. However, the overall pattern is unaffected by which index is used. To be specific, the US shocks are found to have a substantially greater impact on New Zealand GDP than those originating from China. Although the confidence intervals are relatively wide, the response to China peaks one quarter after the shock while the response to the US is longer-lived, with a one percentage point rise in US and China’s GDP estimated to result in accumulated increases of approximately 0.5 and 0.2 percentage points, respectively, in that of New Zealand within four quarters.
WP 13/17 | EMPIRICAL EVIDENCE ON GROWTH SPILLOVERS FROM CHINA TO NEW ZEALAND 27 Appendix Appendix Table 1 – Data description Series Mean Std. Deviation Skewnes s Source US GDP (SAAR, Bil.Chn.2005$) 0.71 0.66 -1.02 Haver Analytics China GDP, (SA, 1997 price rmb mn) 2.43 1.04 -0.27 Haver Analytics & National University of Singapore* Australia GDP (SA, Mil.Chn.Q3:09-Q2:10.A$) 0.83 0.72 0.08 Haver Analytics New Zealand GDP (SA, Mil.Chn.Q3/1995-Q2/1996.NZ$) 0.60 1.15 -0.61 Haver Analytics (1987Q2- 2011Q4). Prior to 1987Q2 the series is backdated as was used in Dungey and Fry (2009). New Zealand Consumer Price Index (SA, Q2-06=1000) 0.99 1.10 2.09 Haver Analytics New Zealand 90-Day Bank Bill Yield (Average, %)** 9.37 5.36 1.20 Haver Analytics New Zealand Real Exchange Rate 0.10 4.20 -1.07 Haver Analytics ANZ World Commodity Price Index, 1986=100 0.52 4.51 -0.17 Haver Analytics*** ANZ Dairy Products Index, 1986=100 0.50 8.69 0.30 Haver Analytics*** ANZ Forestry Products Index, 1986=100 -0.12 5.73 0.22 Haver Analytics*** ANZ Meat, Skins and Wool Index, 1986=100 0.50 4.30 -0.22 Haver Analytics*** ANZ Aluminium Prices Index, 1986=100 0.08 9.70 -0.67 Haver Analytics*** Notes: Unless indicated otherwise, the sample period is 1982Q4 to 2011Q4. Data are expressed as growth rates by multiplying the difference of the log series by 100 with the exception of 90-day rates that are in levels. * Link: http://www.fas.nus.edu.sg/ecs/esu/data.html ** Percentage points *** The sample period is 1986Q1 to 2011Q4. the available values are seasonally adjusted using X-11 and deflated by US CPI
WP 13/17 | EMPIRICAL EVIDENCE ON GROWTH SPILLOVERS FROM CHINA TO NEW ZEALAND 28 Appendix Figure 1 – Schematic description of the extended model Notes: The arrows indicate the direction of feedback across the three countries. New Zealand’s macroeconomic variables comprise real GDP, inflation, interest rate and the real exchange rate.