Asian Development Outlook 2014 Update

Transcription

1 1

2 Asian Development Outlook 2014 Update Shang-Jin Wei Chief Economist Asian Development Bank October 2014 The views expressed in this document are those of the author and do not necessarily reflect the views and policies of the Asian Development Bank, or its Board of Governors, or the governments they represent. 2

3 3 Key messages Developing Asia s growth forecasts kept at 6.2% in 2014 and 6.4% in 2015 Asia s domestic fundamentals offsetting slower-thanexpected advanced economy recovery Inflation moderate at 3.4% in 2014 and 3.7% in 2015 Upcoming US monetary tightening unlikely to repeat taper tantrum in Asia

4 4 Developing Asia is the fastest growing region in the world % GDP growth f 2015f f: forecast

5 5 despite slippage in advanced economies performance GDP growth (%) f 2015f Major industrial economies United States Euro area Japan f: forecast

6 6 Growth in East and South Asia remains strong f 2015 f South Asia India Pakistan Sri Lanka Southeast Asia Indonesia Malaysia Philippines Singapore Thailand Viet Nam f 2015 f Central Asia Azerbaijan Kazakhstan East Asia China, People s Rep. of Hong Kong, China Korea, Rep. of Taipei,China The Pacific Fiji Papua New Guinea f: forecast

7 7 PRC growth is driven by domestic factors Demand-side contributions to growth Trade indicators Percentage points, year to date 9 Net exports Consumption Investment GDP $ billion Trade balance (LHS) Export growth(rhs) Import growth (RHS) % Q Q2 Q3 Q4 Q Q2 Q3 Q4 Q Q Jan2012 Jul Jan2013 Jul Jan2014 Aug -40

8 PRC While a stronger stimulus package from the authorities is likely, we do not think it is necessary as long as the employment situation stays healthy. Carrying out reforms is more important than meeting the growth target. Even if one wishes to organize a stimulus package, a broad-based tax cut may be better than a massive increase in credit. Downsizing the role of the government in the economy can have a stimulus effect. 8

9 9 Confidence is bolstered in the Indian economy Industrial outlook survey Index 60 Business situation Stock price indexes (Sensex; MSCI) Index, 1Jan09 = Financial situation India Emerging Asia 0 Employment 150 Other emerging economies Q12008 Q12009 Q12010 Q12011 Q12012 Q12013 Q Jan-12 Jul-12 Jan-13 Jul-13 Jan-14 Jul-14

10 10 ASEAN-5 growth to ease this year before picking up in 2015 % GDP growth ASEAN-5 Indonesia Malaysia Philippines Thailand Viet Nam f 2015f f: forecast

11 11 Weaker commodity prices continue to temper inflation 2010= Global commodity price indexes % Inflation Energy year moving average Food Jan 2012 Jul Jan 2013 Jul Jan 2014 Jul f 2015f f: forecast

12 Demand-side price pressures remain subdued Inflation, %,yoy 6 % 6 Inflation (% change, yoy) Output gap (% of GDP) Output gap, % of GDP

13 13 Current account surplus remains steady % of GDP 6 Developing Asia Current account People's Republic of China f 2015f f: forecast

14 14 Risks to the outlook Delay in planned reforms Deterioration in PRC s real estate / shadow banking problems Commodity price spike due to geopolitical risks Weaker advanced economies performance Faster tightening in US monetary policy

15 15 Key messages Developing Asia s growth forecasts kept at 6.2% in 2014 and 6.4% in 2015 Asia s domestic fundamentals offsetting slower-thanexpected advanced economy recovery Inflation moderate at 3.4% in 2014 and 3.7% in 2015 Upcoming US monetary tightening unlikely to repeat taper tantrum in Asia Asia s participation in global value chains uneven; additional opportunities can be exploited

16 16

17 Asia in global value chains 17

18 18 What are global value chains (GVCs)? Different stages of production in different countries Production chains can be: within a country across countries within a region global

19 19 Example: iphone production network Toshiba (Japan) Flash memory Display module Touch screen Murata (Japan) FEM Broadcom (US) Bluetooth/FM/WLAN Numonyx (US) Memory MCP Cirrus Logic (US) Audio Codec Foxconn (Taipei,China company located in Shenzhen, PRC) Assembly Infineon (Germany) Baseband Camera module RF transceiver GPS receiver Power IC RF function Dialog Semiconductor (Germany) Power IC Application Processor Function Samsung (Rep. of Korea) Application processor SDRAM-Mobile DDR Sources: Apple; Xing and Detert (2010)

20 Value-added trade map 20

21 GVCs benefit participating economies Rising GVC involvement associated with faster output, income, and employment growth in Economies doubling GVC trade had 19% faster GDP growth and 12% faster increases in per capita incomes Industries doubling GVC trade had 10% faster employment gains GVC contagion risk is real but limited Shocks to a GVC partner need not become an economy-wide shock at home Flexible labor and capital markets can contain the effects Shocks can become opportunities for nimble GVC members 21

22 22 GVCs change our understanding of trade and competitiveness measures Bilateral trade balance Revealed comparative advantage Real exchange rate

23 Gross exports decomposition Gross exports (E*) Domestic Value-added absorbed abroad (DVA) Domestic Value-Added returns home (RDV) Foreign Value-added (FVA) Pure double counting (PDC) DVA in exports E* can be at country/sector, country aggregate, bilateral /sector and bilateral aggregate; both DVA and RDV are backward-linkage based

24 24 Gross exports accounting: Domestic Value-added based on but beyond Leontief insights DVA in exports Domestic Value-added absorbed abroad (DVA) Domestic Value-Added returned home (RDV) (T1) in final goods exports (DVA_FIN) (T2) intermediates exports absorbed by direct importers (DVA_INT) (T3+T4+T5) intermediates re-exported to third countries (DVA_INTrex) (T6) intermediates that returned via final imports (RDV_FIN1) (T7) intermediates that returned in final imports via third countries (RDV_FIN2) (T8) intermediates that returned via intermediate imports (RDV_INT) (T3) Intermediate exports to third country to produce domestic final goods (DVA_INTrexI1) (T4) Produce final goods exports to third country (DVA_INTrexF) (T5) Produce intermediate exports to third country to produce exports (DVA_INTrexI2)

25 25 Gross exports accounting: Foreign Value-added and Pure Double Counting based on but beyond Leontief insights Foreign Value-added (FVA) VS Pure double counting (PDC) Foreign valueadded used in final goods exports (FVA_FIN) Foreign valueadded used in intermediate exports (FVA_INT) Pure double counting from Foreign source (FDC) Pure double counting from Domestic source (DDC) (T11) Sourced from direct importer (T13) Sourced from other countries (T12) Sourced from direct importer (T14) Sourced from other countries (T15) Due to direct importer exports production (T16) Due to other country exports production (T9) Due to final goods exports production (T10) due to intermediate exports production

26 26 Bilateral trade balance: value added terms gross terms JPN-USA (1.4) PRC-USA (0.59) Note: The first country labeled is the surplus country while the second runs a deficit. Numbers in parentheses are the ratio of value-added to gross surplus. Source: Koopman, Wang, and Wei 2014.

27 27 Traditional revealed comparative advantage (RCA) definition TTTTTTTT ii rr = ee ii rr nn rr ii=1 ee ii GG rr ee ii Traditional RCA is based on the relative share of gross exports Country C sector k s gross exports as a share of country C s total gross exports, relative to the world total exports of sector k as a share of world total gross exports If TRCA > 1, one concludes that Country C has a RCA in sector k. For example, TRCA for India s business services is 3.0, and that for Germany s business services is 0.8. One concludes that India has a strong revealed comparative advantage in business services, whereas Germany has a revealed comparative disadvantage in business services. rr nn GG rr ee ii ii rr

28 Problems with traditional RCA Does not exclude foreign value added in a country sector s gross exports More problematic for downstream sectors More problematic with increasing offshoring or international production sharing Does not take into account indirect exports of factor content through other domestic sectors More problematic for upstream sectors More problematic with rising domestic outsourcing or increasing domestic production sharing 28

29 New RCA definition NNNNNNNN ii rr = vvvvvv_ff rr rr ii + rrrrrr_ff ii GG rr(vvvvvv_ff rr ii + rrrrrr_ff rr ii ) nn (vvvvvv_ff rr ii + rrrrrr_ff rr ii ) nn GG(vvvvvv_ff rr ii + rrrrrr_ff rr ii ) ii=1 ii rr The new definition looks at the share of a sector s domestic value added in exports relative to the global share It excludes foreign value added and the part of pure double counting that is originated abroad It includes indirect exports of a sector s factor content

30 Relating the new and traditional RCA measures NNNNNNNN rr ii = TTTTTTTT rr ii (ββ ii rr ββ_ff rr rr ii )αα ii GG (ββ rr ii ββ_ff rr ii )αα rr rr rr ii γγ ii nn αα rr rr ii λλ ii GG nn αα rr rr ii ηη ii ii=1 rr ii=1

31 31 Two RCA measures for electric and optical equipment exports 3.0 People s Republic of China 1.7 United States New RCA Old RCA New RCA Source: Wang, Wei, and Zhu Old RCA

32 32 Two RCA measures for business services exports 1.7 Germany 3.6 India 1.5 New RCA Old RCA New RCA 0.7 Old RCA Source: Wang, Wei, and Zhu

33 33 Typical bilateral real exchange rate (RER) RRRRRR hff = pp ff pp h Bilateral RERs are typically computed using aggregate country level price indices (ex. Chinn, 2006) This is fine if the goal is to measure changes in relative cost of living But is flawed if the goal is to measure changes in competitiveness

34 34 New bilateral RER measure mm GGGGGG RRRRRR hff = ii=1 pp ii vvv VV ii h mm jj=1 pp jj vvv VV jj h mm ww iiii hh pp jj vvv + jj=1 mm ww iiii hff pp jj vvvv jj=1 Uses GVC-REER measure that improves upon existing measures along 4 dimensions: Trade in intermediates and distinction between gross and value-added flows REER indices at the country-sector level CES production functions and preferences with elasticities estimated from the data Use of sector level prices instead of aggregate country level prices

35 Two RER measures for Germany 35

36 36 GVCs thrive where trade costs are low Keep tariffs low and predictable Trim transport costs and delays Ensure product standards do not become barriers

37 37 Low, predictable tariffs are necessary to connect to GVCs Tariff amplification due to multiple border crossings n = 5 n = % 5% 10% 20% Source: OECD 2013

38 38 (Region (% Foreign content of final goods exported) Trade costs of multi-stage production Viet Nam (48) Malaysia (46.3) Rest of Asia (63.2) Thailand (45.8) Canada (41.7) Rest of the world (27.6) XEA (33.6) SSA (26) Philippines (42.2) Taipei,China (41.6) Singapore (70.5) MEP (23) India (20.4) European Union 12 (35.3) South Africa (22.6) Russian Federation (20) Indonesia (36.6) Brazil (15.6) Republic of Korea (31.5) Mexico (12.8) Australia (21.4) CHP (53.8) ROH (37.1) Hong Kong, China (41.6) People's Republic of China (13.7) United States (14.3) European Union 15 (12.4) Japan (11.6) (Total trade cost magnification fator) East Asia pays a price for its long chains and relatively high tariffs Advanced economies have low foreign content and, hence, low costs Developing countries have large magnification ratio due to higher share of imported content in their manufacturing exports Trade Cost (% of final goods exported) Standard CIF Standard tariff Multi-stage prod CIF Multi-stage prod tariff Source: Koopman, Wang and Wei 2014.

39 39 Effective tariff rate of multi-stage production Singapore (29.5) Rest of Asia (36.8) CHP (46.2) Viet Nam (52) Malaysia (53.7) Thailand (54.2) Philippines (57.8) Canada (58.3) Hong Kong, China (58.4) Taipei,China (58.4) ROH (62.9) Indonesia (63.4) European Union 12 (64.7) XEA (66.4) Republic of Korea (68.5) Rest of the world (72.4) SSA (74) MEP (77) South Africa (77.4) Australia (78.6) India (79.6) Russian Federation (80) Brazil (84.4) United States (85.7) People's Republic of China (86.3) Mexico (87.2) European Union 15 (87.6) Japan (88.4) Region(%Domestic content of final goods exported Standard tariff Effective and standard tariff difference Effective tariff magnification ratio Advanced economies have higher domestic content in their exports, hence their exporters face lower effective tariff rate Exporters in developing countries face higher effective tariff rate due to lower share of domestic content in their manufacturing exports Deep tariff cut in manufacturing sectors will benefit emerging economies more Source: Koopman, Wang and Wei 2014.

40 40 Lower transport costs may reduce trade costs even more than tariff reductions Quality of roads Index Central Asia East Asia Southeast Asia South Asia The Pacific Quality of air transport infrastructure Index Central Asia East Asia Southeast Asia South Asia The Pacific Documents required to export Number of required documents Southeast Asia East Asia The Pacific South Asia Central Asia Central Asia East Asia South Asia Southeast Asia The Pacific Index LPI score: Logistics quality and competence

41 41 East and Southeast Asia emerged as a world manufacturing hub % East Asia (inc. Japan) Southeast Asia South Asia

42 42 But few in Central Asia, South Asia, or Pacific Developing Asia average = 6.99 Hong Kong, China Singapore Georgia Timor-Leste have found their GVC niche Hong Kong, China Developing Asia average = 7.0 Singapore Republic of Korea Georgia Developing Asia average = 28.3 Singapore Hong Kong, China Republic of Korea Georgia Hong Kong, China Republic of Korea Taipei,China Developing Asia average = 4.42 Singapore People's Rep. of China People's Rep. of China Taipei,China People's Rep. of China India India India India Nepal Nepal People's Rep. of China Kyrgyz Republic Vanuatu Mongolia Tajikistan Nepal Uzbekistan Uzbekistan Uzbekistan Timor-Leste Fiji Tajikistan Afghanistan Myanmar Effective tariff rates, 2013* Simple average Number of required documents to export, 2014 Number of days to export, 2014 * latest data for India is 2009 and PRC is 2011, which are assumed equal to Sources: Doing Business, The World Bank, The Global Competitiveness Report , World Economic Forum and WITS online database Transportation Index, (Rebased 1=highest, 8=lowest)

43 43 Asia can build on its reputation as the world s workshop Region has established itself as a GVC leader Supportive policies can enhance this role further Asia can deepen, broaden, and upgrade its role in global production networks

44 44 Global Value Chains and Effective Exchange Rates at the Country-Sector Level Nikhil Patel Zhi Wang Shang-Jin-Wei

45 Introduction Real effective exchange rate (REER) Measure of competitiveness Many applications currency misalignment and manipulation vulnerability to crises (Chinn, 2000; Goldfajn and Valdés, 1999; Gagnon, 2012) This paper: New model of REER to better capture movements in competitiveness Create database of country and sector level REERs 45

46 The concept of REER RRRRRRRR JJ = VV JJ = ww JJJJ pp ii V J : value added by country J w Ji are exchange rate weights denotes log change from steady state Partial equilibrium concept Primitive shocks not modeled No restrictions on trade balance All papers (including ours) work in this setting nn ii=1 46

47 Main problems with conventional measures Ignore intermediate inputs Both offshoring and domestic outsourcing are ignored In the data, around 60 percent of world trade comprises of trade in intermediate goods Ignore sectoral heterogeneity within countries Strong evidence documenting sectoral heterogeneity (Wang, Wei and Zhu,2013) Assume elasticity of substitution in production and preferences are the same and equal to 1 47

48 Importance of Considering Sector level Flows Example: Two sector Chinese Economy Electronics Sector (*source: Koopman, Wang and Wei,2012) Fraction of value added in aggregate Chinese good >40 percent Over-predicts Chinese value added in exports Non traded goods sector % Domestic value added percent 100 percent Fraction exported High (~90 percent)* Low (~0) 48

49 49 More on sector level heterogeneity within countries China 2002 IO Industry % value added % processing exports GVC position Electronic computer Downstream Textiles production Upstream Source: Koopman, Wang and Wei, 2012 Germany 2011 WIOD sector Intermediate exports (% of total) GVC position Basic metals (C12) 93 Upstream Machinery (C13) 45 Downstream Source: Wang, Wei and Zhu 2013

50 The Need to Go Beyond Country to Individual Sector REERs World economy increasingly marked by specialization and global value chains Different sectors within a country can show very different behavior with regard to competitiveness 50

51 This paper New model of REER Distinguish intermediate and final goods Construct sector and country level REER indices Construct and incorporate sector level price indices Estimate and incorporate different elasticities of substitution in production and consumption Focus on short run changes in competitiveness Take GVC and production outsourcing pattern as given 51

52 Literature REER IMF FED BIS BJ BST BZ This paper Value added competitiveness Sector level exchange rates / prices Trade in intermediate goods Heterogeneous elasticities Key: BJ: Bems and Johnson (2012); BST: Bayoumi et al. (2013); BZ: Bennett and Zarnic (2009) Global Value Chains / Export Accounting / Vertical Specialization: Koopman, Wang and Wei (AER 2014, NBER wp 2014, and JIE 2001), Wang, Wei and Zhu (2014), Hummels, Ishii and Yi (2001) etc. 52

53 Roadmap Illustrative examples with stylized three country global value chain General model Results Elasticity estimation REER indices for countries and sectors Application: Bilateral Exchange rates Conclusion 53

54 Weighting matrix for 3 by 2 GVC 54

55 55 General Model: Production n countries, m sectors

56 56 General Model: Final Demand and Market Clearing

57 General Model: Solution 57

58 Data Main Source: World Input-Output Database (WIOD) Inter Country Input-Output tables at the country-sector level 40 countries and 35 sectors within each country Available in both current and previous year prices Annual: ( for previous year prices) Detailed description in Timmer et. al (2012) Other sources: IMF International financial statistics 58

59 Estimation of elasticities Basic framework: Feenstra (1994) Subsequently developed by Broda and Weinstein (2006) and Soderbery (2013) We modify the framework to allow for elasticities<1 Use hybrid grid search constrained LIML estimator Consistency relies on T Right now we have T=14 Take steps to deal with small sample LIML instead of 2SLS Winsorize data at 10 and 90 percentile Report medians from panel bootstrap distributions 59

60 60 Elasticity Estimates: Medians across different country and sector groups

61 61 Heterogeneity Across Sectors Within A Country

62 62 Application: Bilateral Real Exchange Rates (RER) for Competitiveness Bilateral RERs are typically computed using aggregate country level price indices (ex. Chinn, 2006) RRRRRR hff = pp ff pp h This is fine if the goal is to measure changes in relative cost of living But is flawed if goal is to measure changes in competitiveness

63 63 Example: Bilateral RER Given the price changes, RRRRRR UUUU indicates an increase in competitiveness of the US (RRRRRR UUUU =0.008) This is misleading since US competes only with C1 and p V (c 1 ) The conceptually correct should indicate a fall in competitiveness of the US

64 64 We define bilateral RER using GVC- REER weights as follows: mm GGGGGG RRRRRR hff = ii=1 pp ii vvv VV ii h mm jj=1 pp jj vvv VV jj h mm ww iiii hh pp jj vvv + jj=1 mm ww hff vvvv iiii pp jj jj=1

65 Standard vs GVC-RER: Germany 65

66 Conclusion Main contribution: REER measure improving upon existing measures along four dimensions Trade in intermediates and distinction between gross and value added flows REER indices at the country-sector level CES production functions and preferences with elasticities estimated from the data Use of sector level prices instead of aggregate country level prices Other contributions Inter country model of sector level trade: can easily be extended to a DSGE setting First distinct estimates of production and consumption elasticities 66