96. Nguyen Cam Nhung Khoa KT&KDQT 2014

96. Nguyen Cam Nhung Khoa KT&KDQT 2014 tài liệu, giáo án, bài giảng , luận văn, luận án, đồ án, bài tập lớn về tất cả cá...

iiRII .)f

~ti~O)~m~~~~~.Qi:P~*O)~a~

iI&~M~I:h.Q i:P~O)~~lb'JJl~:¥O)!/'ij~

;f / ?Cl-r\)v1tJUjjf~O)m~t~~O)~m

Is Exchange Rate Pass-Through Declining?:

Evidence from Japanese Exports to the United States and Asia

y~~y~~/t~l0)77/A~~~~~IGbm~0)~~t~~*

R iR

Cl-~-' '7-v/ (=If) rmmimllttjflJ*J

Is Exchange Rate Pass-Through Declining?:

Nguyen Cam Nhung

Abstract Unlike the previous studies, this paper reexamines the degree of exchange rate pass-through (henceforth ERPT) in-1apanesecexports=totheUnited~States and Asian countries by using the destination breakdown data of 300 export commodities at the HS 9-digit level By conducting the fixed effect panel estimation, we have found that ERPT has increased in Japanese exports to all destinations, even to the United States, at the HS 4-digit commodity level, which contrasts markedly with the Taylor's (2000) conjecture that ERPT declined in the low inflation environment.

Key words: exchange rate pass-through; pricing-to-market; Japanese exports; East Asia.

JEL Classification: F3, F4

1 INTRODUCTION Exchange rate pass-through (henceforth ERPT) has gained renewed attention in recent years in both theoretical and empirical literature Taylor (2000) conjectured that the extent of ERPT declined in the low inflation environment Recent studies, such as Campa and Goldberg (2005) and Otani, Shiratsuka and Shirota (2005), have examined the degree of ERPT in import prices using either the data on a number of source countries or the industry/commodity breakdown data.

In contrast to the previous studies, this paper employs the data on Japanese exports by commodity and by destination country to make more rigorous empirical investigation about whether the degree of ERPT has declined in the case of Japanese exports.

The highly disaggregated commodity data by destination is often used in the literature such as Knetter (1989), Takagi and Yoshida (2001), Sato (2003), Parsons and Sato (2008) and Yoshida (2009) However, except for Yoshida (2009), the above studies tend to choose only a small number of commodities at the HS 9-digit level as a "representative"

of respective industries.2)The empirical results obtained by the above empirical approach are hard to be generalized in discussing the ERPT in a broader perspective.

To overcome the weakness of the previous studies, this paper employs a panel estimation as developed by Yoshida

I) I would like to thank Kiyotaka Sato, Craig Parsons and Eiichi Tomiura for their helpful comments.An earlier version of this paper was presented at the 12th International Convention of the East Asian Economic Association in Seoul, Korea, 2-3 October 2010.

2) For example, Takagi and Yoshida(2001) analyzed only II commodities for estimating pass-through coefficientsfor Japanese exports to East Asia Sato (2003) collected 13 commodities, Parsons and Sato (2008) selected 27 commodities.

102 (398)

(2009) by pooling all 300 export commodities at the HS 9-digit level into fifty-four HS 4-digit classifications We obtain pass-through coefficients at the HS 4-digit level, which better reflects the similarity of ERPT in Japanese exports at a broader commodity category While Japanese exporting firms have built a regional production network in East Asia and the share of intra-firm trade has been growing in the region (Ito, et aI., 2010), the high ERPT is generally observed in Japanese exports to East Asia.

In order to reexamine the degree of exchange rate pass-through in Japanese exports to the United States and Asian countries, nine destination countries are chosen including the United States, China, Korea, Indonesia, Malaysia, the Philippines, Singapore, Thailand, and Vietnam Since several East Asian countries were hit by the currency crisis ill 1997-1998 and the global economy has experienced a significant fall in trade transactions and economic growth on

a global scale in recent years, we exclude the crisis period from our analysis Thus, the whole sample (from January

1988 to December 2008) is divided into two sub-samples for empirically analyzing the degree of pass-through for the pre-crisis period (January 1988 to December 1996) and the post-crisis period (from January 2000 to December 2008) Comparing findings between two sub-samples, we can explore whether the Japanese exporters' pass-through behavior

in regional trade has changed in response to the dynamic development and integration in East Asian countries Our findings of the rising ERPT have.important.implications.for the price setting behavior of exporting firms in the growing production network.

The paper is organized as follows Section 2 presents the empirical model Section 3 describes the data Section 4 discusses the estimated results Finally, Section 5 concludes.

2 THE EMPIRICAL MODEL

We employ the standard regression equation used in the literature such as Knetter (1989) and Yoshida (2009) to estimate the extent of exchange rate pass-through by using a panel data.

(I)

where Ii denotes the first-difference operator; i = I, , N,j = I, , M and t = I, , T, respectively, indicate

the commodity of exports, importing countries and time; p is the export price; a is a commodity fixed effect; S is the

bilateral nominal exchange rate of the exporter's currency vis-a-vis the importers' currency; PPI is the producer price index of exporting country (a proxy for the exporter's marginal cost); IPI is the industrial production index of importing

country (a proxy for the destination market demand conditions); andE:is an error term.

Before using equation (I) to estimate the degree of exchange rate pass-through, we checked the stationary of dependent and independent variables, and found that most of variables are non-stationary in level but stationary in first differences.l)Thus, we finally use the first-differenced form ofthe ERPT equation, that is, equation (I), which has been widely employed in previous studies for analyzing the pass-through, for instance, Yoshida (2009) and Parsons and Sato (2008) However, our pass-through coefficients are different from theirs While they estimate the pass-through coefficient for a single commodity at the HS 9-digit level, we evaluate the common slope pass-through coefficients gained from pooling all commodities at the HS 9-digit level for each of HS 4-digit classifications which gives us more

accurate information on the exporters' behavior at an industry level fJ is a pass-through coefficient at the HS 4-digit level, and can be interpreted as follows If the null hypothesis of fJ = 0 cannot be rejected, this may imply that full

pass-3) We performed the augmented Dickey-Fuller(ADF) test for stationarity of variables EViews 7 is used for the ADF test.

Is Exchange Rate Pass-Through Declining? (Nguyen Cam Nhung)

Table I The Ratio of Currency Invoicing (the 2"d half of 2008)

World Currency US dollar Yen Euro GBP AUD Others

Share 49.8 39.5 6.7 1.3 0.7 2.0

US Currency US dollar Yen Euro GBP CAD Others

Share 86.8 13.0 0.1 0.0 0.0 0.1

EU Currency Euro Yen US dollar GBP SEK Other

Share 51.1 28.0 16.1 4.5 0.2 0.1 Asia Currency US dollar Yen THB TWO KRW Other

Share 50.7 47.5 0.5 0.4 0.2 0.7 Source: http://www.customs.go.jp, Ministry of Finance, Japan.

Table 2 The-Priors,for-=Tests_forcEstimating-Pass-throughCoefficients

P=l (PTM in EA) No poT (PTMin USD) NoP-T O<p<l Incomplete P-T Incomplete poT

Note: The first column shows the hypothesis of the common slope coefficientp in equations (I).

through is conducted in an aggregated industry at the HS 4-digit classification of all commodities at the HS 9-digit level If p is significant and positive, the degree of pass-through is incomplete, and exporters conduct PTM behavior to

a certain extent p =1 may suggest "no pass-through" or complete PTM, where Japanese exporters tend to stabilize the export price in the importer's currency.

As we know that around 50% of Japanese exports to the Asia were invoiced in US dollars (see Table I) We suppose that both Japanese exporters and Asian importers have no choice but to take on the exchange rate risks against the US dollar In order to check this assumption, we need to estimate the equation (1) above by using the bilateral nominal exchange rate of the Japanese yen vis-ii-vis the-US dollar, even though export destination is Asian countries The above equation is used to analyze whether the US dollar (a third currency) is used as an invoicing currency in Japanese exports to East Asia The estimates of P in equation (I) can be interpreted in Table 2.

3 DATA

We use the highly disaggregated data of export prices at the HS 9-digit level, which is collected from the website

of the Ministry of Finance, Trade Statistics of Japan All data are monthly ranging from January 1988 to December

2008 The unit values for each commodity are calculated as the total yen value of the export divided by the total export volume The producer price index for each commodity is obtained from the website of the Bank of Japan (2005 base).

The industrial production index for each destination is drawn from the International Monetary Fund (IMF), International

Financial Statistics, CD-ROM and the CEIC Database (2005=100 for the US, Korea and Malaysia; 2000=100

104 (400)

T

!

Optical photo,

technical

Source: United Nations Statistics Division- Commodity Trade Statistics Database

for the others) All producer price indices and industrial production indices are seasonally adjusted Monthly series of

the bilateral nominal exchange rates (vis-a-vis the US dollar) are taken from the IMF, International Financial Statistics,

CD-ROM We then calculate the cross rate, i.e., the bilateral nominal exchange rate of the yen vis-a-vis the importers' currency.

All commodities are selected based on the following criteria First, we choose the top four exported categories

of Japan including general machinery; electric machinery; iron and steel; and transport equipment (see Figure I) Second, we choose all commodities having a relatively large volume of transactions and not more than 20 missing data in the whole sample The original dataset includes a larger number of commodities at the HS 9-digit level and a broader range of HS 4-digit industry classifications However, in some cases, there is only a single commodity at the HS 9-digit level for each of HS 4-digit classifications that satisfY our selection criteria or the number of observations is not large enough in the panel data model for us to get reliable results For these reasons, the final results are based on 300 commodities at the HS 9-digit level There are fifty-four HS 4-digit industry classifications used to examine the extent

of exchange rate pass-through across industries by countries We repeat this procedure and conduct panel estimation for each destination Vietnam was only estimated in the post-crisis period because before 1996 the volume and number of products that Vietnam imported from Japan were so small and would not provide reliable results The regression results are summarized in Table 3 and more detailed results for the United States and Singapore are presented in over Table A I through Table A9 in the Appendix.

4 EMPIRICAL ANALYSIS Table 3 presents the summary results of the through coefficients at the HS 4-digit classifications Fully

pass _._ - -, -

-1

(401) 105

I

i

I

LL

Industry Country WholeSYcno'EAsample Whole sampleSYmlUSD Pre-crisisSYc:n1EA Pre-crisisSYmfUSO Post-crisissy"",", Post-crisisSYmfUSO

PTM Full PT PTM Full PT PTM Full PT PTM Full PT PTM Full PT PTM Full PT Iron and steel China 0 3 0 2 0 3 I I I 2 I 2

Vietnam n.a n.a n.3 n.3 n.3 n.3 n.3 n.3 0 3 0 3 M3chinery USA n.3 n.a 4 8 n.3 n.3 2 10 n.3 n.3 I II

Singapore I 9 2 8 2 8 2 8 0 10 I 8

Philippines 0 10 0 II 0 II I 10 0 II 0 II Thailand 2 10 I 10 0 12 0 12 0 12 0 12 Indonesi3 2 8 3 9 2 9 2 9 0 12 1 II Vietnam n.3 n.3 n.3 n.3 n.3 n.a n.3 n.3 4 3 2 5 Ele<:tric USA n.a n.3 12 17 n.a n.a 7 21 n.3 n.3 3 26

m3chinery China 3 19 3 21 2 20 I 22 3 21 4 20

Korea 0 24 2 22 2 22 2 22 0 24 1 22 Singapore 4 23 3 24 5 22 6 21 I 25 3 21 Malaysia 1 23 3 22 4 21 3 21 2 23 1 24 Philippines 0 24 2 22 3 23 2 24 0 26 2 24 Thailand 0 23 2 21 3 20 4 19 1 22 2 21 Indonesia 0 18 2 17 3 16 3 16 2 17 3 16 Vietn3m n.3 n.3 n.a n.a n.3 n.a n.a n.a 5 12 3 14 Transport USA 0.3 n.3 4 2 0.3 n.a 5 I n.3 n.3 3 3

equipment China 0 4 0 4 I 3 0 4 0 4 0 4

Philippines I 7 0 8 I 7 I 7 I 7 I 7

Vietnam o.a n.a n.a n.3 n.a n.a n.a n.3 0 4 0 4 Note: All positive and significant coefficients are counted as PTM Insignificant coefficients are counted as Full PT (pass-through) Negative and significant coefficients are not counted in this table

4) Table Al does not show the summary of the common slope coefficients of pass-through for the iron and steel imports of the United States, Thailand and the Philippines for several reasons First, there were many Anti-dumping (AD) duty activities on steel

imports from Japan by the United States in the period from 1999 to 2002 Second, like the United States, from 2003 Thailand also

attempted to protect their steel industry by imposing 36.25% AD duties on imports of cold-rolled and hot-rolled steel sheets and

plates from 14 countries including Japan Therefore, the imposition of an AD duty may lead to structural breaks in exchange rate

pass-through We need to test the effect of AD on pass-through of AD duties and exchange rates If not, we cannot conclude the

extent of pass-through Finally, for the Philippines case, we cannot find any commodities that satisfy our selected criteria for iron

and steel.

106 (402)

T

I

I

!

mechanical appliances exports to the United States.

As mentioned, the weakness of the previous studies is the use of several commodities as "representative" of

respective industries thus making it hard to generalize the results of estimation to consider the extent of pass-through

at the industry level This study used as many commodities as possible in estimating the common slope pass-through

coefficients This difference yields some opposite conclusions For instance, Parsons and Sato (2008) found PTM in a

single commodity of diodes export to the United States but full pass-through to East Asia in both the pre- and post-crisis

period This study, however, finds full pass-through in Japanese diodes exports to the United States and some East Asian

market over time but PTM in exports to Korea in the pre-crisis period; and to Singapore and Thailand in the post-crisis

period This different finding may be the result of using different data While this paper pools 14 commodities at the

HS 9-digit level for the diodes commodity [8541] at the HS 4-digit level, Parsons and Sato (2008) simply use a single

commodity of diodes for estimation.

In addition, our findings are also different from that of Takagi and Yoshida (2001) Firstly, whereas, they found that

the pass-through coefficient of piston engines [8407] in Japanese export to Philippines is observed PTM behavior in

the period from 1988Ml to 1998M12, our result shows that full pass-through behavior appears in Japanese exports of

piston engines [8407] to Philippines over time Secondly, while we found full pass-through behavior in Japanese export

of starting equipment [8511] constructing from 5 commodities at the HS 9-digit level to Malaysia, Takagi and Yoshida

(200 I) found PTM behavior.

In short, a notable finding is that Japanese exporters increase the degree of ERPT in the post-crisis period For

instance, it is well known that Japanese automobile exporters generally conduct PTM in the US market Even in exports

to the United States, 'however, the PTM behavior has declined in the post-crisis period compared to the pre-crisis one.

Japanese exporters tend to pass through exchange rate changes to East Asian importers in all industries, while PTM is

observed in some cases in the electric machinery industry However, in the post-crisis period, a decline ofPTM behavior

becomes more pronounced, and the Japanese exporters have stronger tendency to pass through the exchange rate risk to

East Asian importers, perhaps, except for Vietnam.

5 CONCLUDING REMARKS

In contrast to the previous studies, our panel estimation approach enables us to investigate the destination specific

ERPT at a broader commodity category It is found that the degree of ERPT has increased over the sample period in

exports to all destinations, even to the United States, which differs markedly from the findings of the previous studies.

The results of estimation at an industry level also show the rising ERPT in all industries in exports to all countries.

While Japanese automobile exports to the United States are considered as a typical example of the PTM behavior, our

results reveal that ERPT has prevailed even in Japanese automobile exports to the US market Itis often_pointed.out:that

Japanese exporters have suffered from an appreciation of the yen Our findings imply that Japanese exporting firms have

changed their pricing behavior in the 2000s, which may reflect their growing production network established in East

Asia After the global financial crisis, however, pricing behavior of Japanese firms may be changed in the face of large

fluctuations in the exchange rate Further empirical analysis will be necessary in the future research.

I

I

~

Appendix

Table AI: Estimates of 13 across commodities for USA

code

1988M 1-2008M 12 YenlUSD

1988MI-1996M12 YenlUSD

2000M 1-2008M 12 YenlUSD

(a) Iron and steel

(b) Machmery and mechanIcal appliances

8407 II Spark.ignition reciprocating or rotary internal 0.504'" 0.210 0.747" 0.395 0.655" 0.290

8408 7 Compression ignition internal combustion piston 0.511'" 0.140 0.766'" 0.245 0.181 0.173

8414 12 Air or vacuum pumps, air or other gas compressors 0.288 0.366 -0.259 0.539 0.597 0.572

8422 6 Dish washing machines, machinery for cleaning -0.159 0.368 0.022 0.623 -0.074 0.523 (c ) Electric machinery

8502 2 Electric generating sets and rotary converters 0.403' 0.234 0.367 0.480 0.621" 0.268

8504 5 Electrical transformers static converters 0.533' 0.318 0.406 0.420 1.045' 0.641

8511 9 Electrical ignition or starting equipment 0.514'" 0.156 0.483' 0.286 0.321 0.225

8516 5 Electric instantaneous or storage water heaters 0.533' 0.321 0.159 0.345 0.791 0.691

8521 2 Video recording or reproducing apparatus 0.861'" 0.255 1.252'" 0.286 0.338 0.564

8526 3 Radar apparatus, radio navigational aid apparatus 0.600 0.491 0.707 0.837 0.098 0.410

8527 2 Reception apparatus for radio-broadcasting 1.274'" 0.406 0.997' 0.552 0.958 0.694

8531 4 Electric sound or visual signaling apparatus 0.093 0.401 -0.014 0.599 -0.025 0.747

8532 10 Electrical capacitors, fixed, variable or adjustable 0.229 0.351 0.438 0.466 0.227 0.735

8535 2 Electrical apparatus for switching circuits -2.980 1.970 -3.982' 2.463 -8.049" 4.008

8536 9 Electrical apparatus for electrical circuits 0.314 0.298 0.846" 0.380 -0.430 0.576

8541 14 Diodes, transistors and simii::lr semi-conductor 0.238 0.444 -0.097 0.808 0.394 0.630

8544 7 Insulated wire, cable, other electric conductor 0.084 0.742 0.161 1.035 0.451 1.404

8546 3 Electrical insulators of any material -0.100 1.872 -3.414 2.840 3.086 3.674

8547 2 Insulating fittings for machines, aooliances 1.305'" 0.388 1.440*** 0.450 0.689 0.750 (d) Transport equipment

8703 4 Motor cars and other motor vehicles 1.208*** 0.308 0.541'" 0.098 2.324'" 0.768

8716 2 Trailers and semi-trailers, other vehicles 0.643 1.455 1.298 1.347 0.807 3.434

Note: *.* at 1%, * at 5%, * at 10%, P denotes the common slope pass-through coefficient for each HS 4-dlglt level 's.e.' the standard errors.

J988MI.2008M12 1988MI.~008MI2 1988MI.I996MI2

HS4

,ad,

HS9 Code-Description

1988MI.I996MI2 YenlUSD

2000MI-2008MI2 Yen/KWR

2000MI.2008MI2 YenllJSD

(a) Iron and sleel

(b) MlKhlncry and mcchllnlclllllpphanccs

(c) Electnc machinery

8501

8541

" Diodes, transistors and similarsemi-eonductor 0.150 0.217 0.689 0.315 0.241 ••• 0.499 \.224" 0.415 -0.258 0.434 -0.277 0.591

(d) Trans nequipment

NOlO:: ••• al 1°/•••• at 5~~•• at 10% j\ denotesIh.: COlllnl(>n slope pass-Ihrough coefficienl for each HS 4-digitl.::vel, 's.e.' the stllndord crrors

I

F . -HS4

code

HS9 Code Description

Table A3: Estimates of P across commodities for China

1988MI-2008M12 1988MI-2008M12 1988MI-1996M12 YenIRMB YenlUSD YenIRMB

1988M 1.1996M 12 YenlUSD

2000MI-2008M12 YenIRMB

2000M 1-2008M 12 YeniUSD

(a) Iron and steel

7205 2 Granules and powders, ofpigiron, iron or steel

7210 2 Flat-rolled products of iron, a width of 600 mm

7212 4 Flat-rolled products of iron, less than 600 mm (b) Machinery and mechanical appliances

s.c.

3.540 1.656 0.882

8407 3 Spark-ignition reciprocating or rotary internal 1.257"- 0.043 0.415 0.547 1.731 ••• 0.691 0.880 1.270 0.375 0.300 0.324 0.300

8408 3 Compression ignition internal combustion piston 0.582 0.414 0.716 0.541 1.010 0.547 2.157-" 0.899 -0.025 0.745 -0.085 0.735

8409 6 Parts suitable for use solely 8407,8408 0.173 0.449 0.224 0.616 0.201 0.633 0.391 1.112 -0.287 0.871 -0.186 0.871

8413 8 Pumps for liquids, liquid elevators 0.098 0.280 0.453 0.393 0.112 0.417 0.910 0.738 0.336 0.477 0.298 0.476

8418 7 Refrigerators, freezers, electric, heat pumps 0.365 0.494 0.129 0.710 0.277 0.643 -0.424 1.204 -0.355 1.101 -0.360 1.096

8419 5 Machinery plant or laboratory equipment 0.305 0.502 0.904 0.705 0.542 0.689 2.426" 1.221 0.471 1.031 -0.353 1.028

8421 7 Centrifuges, filtering or purifying machinery 0.062 0.360 -0.022 0.504 -0.245 0.489 -0.921 0.862 0.628 0.692 0.684 0.691

8422 2 Dish washing machines, machinery for cleaning 0.015 0.637 -0.677 0.897 0.458 0.764 0.046 1.357 0.003 1.602 0.129 1.600 (c )Electric machinery

8501 7 Electric motors and generators -0.659- 0.383 -0.045 0.539 -1.007- 0.557 -0.132 1.001 0.749 0.694 0,611 0.693

8502 3 Electric generating sets and rotary converters 1.86' ••• 0.563 0.895 0.753 2.854-.- 1.031 2.310 1.806 -0.081 0.712 -0.140 0.709

8504 6 Electrical transformers, static converters 0.034 0.451 0.618 0.611 -0.305 0.617 0.645 1.047 1.449 0.966 1.346 0.971

8505 4 Electro-magnets, permanent magnets, clamps, 0.147 0.468 0.235 0.657 0.122 0.799 0.315 0.407 0.390 0.586 0.375 0.591

8511 6 Electrical ignition or starting equipment 0.579 0.401 0.597 0.562 0.530 0.461 0.408 0.815 0.945 1.544 0.956 1.042

8512 3 EJectricallighting or signaling equipment 0.282 0.286 QAOI 0.400 0.348 0.359 0.727 0.634 0.468 0.638 0.484 0.638

8515 7 Electric laser or photon beam, ultrasonic 0.452 0.531 0.155 0.745 0.951 0.739 1.159 1.320 0.106 1.107 -0.047 1.105

8516 2 Electric instantaneous or storage water heaters 0.153 0.893 -0.735 1.265 0.603 1.104 -0.283 1.973 -1.768 2.438 -1.764 2.431

8518 4 Microphones and stands therefor; loudspeakers -0.213 0.816 -0.354 1.160 -0.558 1.011 -1.556 1.831 1.501 2.029 1.614 2.026

8526 3 Radar apparatus navigational aid apparatus 0.987" 0.469 1.262" 0.658 0.744 0.595 0.754 1.065 0.711 1.128 0.769 1.127

8529 3 Parts suitable for use with 8525 to 8528 0.368 0.499 0.258 0.700 -0.783 0.687 -0.172 1.220 0.041 1.160 0,078 1.159

8531 3 Electric sound or visual signaling apparatus 0.111 0.680 -0.321 0.947 0.137 1.082 -0.936 1.899 0.081 1.077 0.008 1.076

8532 6 Electrical capacitors, variable or adjustable 0.155 0.504 -0.144 0.679 0.106 0.903 -0.780 1.512 0.372 0.476 0.404 0.476

8533 6 Electrical resistors -0.088 0.294 0.242 0.412 -0.393 0.450 -0.329 0.801 0.943" 0.490 0.984" 0.490

8536 8 Electrical apparatus not exceeding IOOOvolts 0.562" 0.252 0.258 0.352 0.763" 0.391 0.559 0.688 0.137 0.418 0.087 0.418

8537 2 Boards, panels consoles, desks 0.113 1.024 2.394- 1.443 -1.320 1.545 0.942 2.791 3.238" 1.654 3.371" 1.651

8538 4 Parts suitable for use with 8535 to 8537 0.787 0.639 -0.236 0.894 0.176 1.013 -3.259- 1.796 1.797- 1.027 1.773- 1.025

8539 5 Electric filament or discharge lamps -1.141- 0.669 -0.089 0.944 -2.297" 1.140 -1.154 2.059 0.388 0.743 0.418 0.741

8540 2 Thermionic, cold cathode or valves 0.310 0.552 0.183 0.781 0.190 0.734 -0.157 1.364 0.946 1.336 0.898 1.336

8541 12 Diodes, transistors and similar semi-conductor -0.140 0.247 0.194 0.333 -0.157 0.480 0.485 0.706 -0.109 0.322 -0.101 0.322

8543 2 Electrical machines and apparatus 1.506 1.205 1.223 1.689 2.045 1.528 2.560 2.696 -1.433 1.848 -1.625 1.846

8547 2 Insulating fittings for electrical machines, 0.466 0.675 2.872 ••• 0.936 -0.111 1.040 3.945" 1.809 1.699 1.073 1.777- 1.071

(d) Transport equipment

8708 II Parts and accessories of motorvehicles 0.329 0.206 0.219 0.289 0.508- 0.306 0.630 0.539 -0.101 0.277 -0.114 0.277

8714 5 Parts and accessories of vehicles of871 tto 8713 -0.087 0.275 0.033 0.388 0.000 0.307 0.227 0.543 0.572 0.729 0.528 0.732

8716 2 Trailers and semi-trailers, other vehicles 0.735 1.120 0.798 1.582 0.692 1.573 0.545 2.983 0.100 2.139 0.057 2.134

Note: ••• at 1%, •• at 5%, • at 10%, P denotes the common slope pass-through coefficient for each HS 4-digitlevel 's.e.' the standard errors.

f(")

8

1