going down: the asian crisis and u.s. exports · of the east asian markets for u.s. exporters that...

1 The economic and financial cri-sis in Asia has spawned volumi-nous popular and academicliterature. An excellent sourcethat identifies much of this liter-ature is a web page producedby Nouriel Roubini that can befound at the following Internetaddress: <http://www.stern.nyu.edu/~nroubini/asia/AsiaHomepage.html>. For an ele-mentary introduction to theAsian crisis, see Neely (1999).

2 For example, Alan Greenspan inhis September 1998 testimonybefore the U.S. SenateCommittee on the Budget(1998, p. 1) said: “ ... it isjust not credible that the UnitedStates, or for that matterEurope, can remain an oasis ofprosperity unaffected by aworld that is experiencinggreatly increased stress.”

3 Since both price indexes use1960=100, the nominal andrelative price adjusted sharesfor 1960 are equal.

FEDERAL RESERVE BANK OF ST. LOU IS

33

MARCH/APR I L 1999

Going Down:The Asian Crisisand U.S.ExportsPatricia S. Pollard andCletus C. Coughlin

T he Asian economic and financial crisishas attracted attention to the trade linksamong the United States and countries

throughout Asia.1 The trade channel isundoubtedly an important mechanismthrough which this crisis affects the U.S.economy. As the effects on Asian economiesbecame more pronounced and the effects ofthe crisis spread to other countries in springand summer 1998, changes in U.S. tradebecame more visible and concerns aboutthe possibility of a U.S. recession increased.2

Nonetheless, the U.S. economy remainedrobust in 1998. Declining commodity prices(partly a result of reduced demand in Asia)and declining interest rates (partly a resultof flight-to-quality capital inflows and theFederal Reserve’s easing of monetary policyin reaction to the Asian crisis) helped spurconsumption and investment spending.This spending tended to offset the weaken-ing of export spending, especially spendingon goods destined for Asia.

Our primary goal in this paper is toprovide some rough estimates of the sizesof the export shock to the U.S. economy,generally, and the shock to specific sectorsassociated with the Asian crisis. We beginby providing some background material,such as facts about both the magnitude andchanges of U.S. exports over time. Next,we examine the flow of exports to Asia andother regions. The Asian crisis has reducedrecent growth, as well as near-term growthprospects throughout Asia, and has caused

large declines in the U.S. dollar exchangevalue of many Asian currencies. Thesechanges have caused a reduction in U.S.exports to Asia. We estimate the reductionin U.S. exports—first on an overall basis andsecond on an industry basis. In the processwe attempt to identify any recent worseningof export performance and any damage beingsuffered by specific industries.

HOW IMPORTANT ARE EXPORTS FOR THE U.S. ECONOMY?

During 1997, U.S. exports totaled $965.4billion; 71 percent were exports of goods, and29 percent were exports of services. Exportsare a relatively small, albeit increasing, shareof U.S. economic activity. A straightforwardcalculation of the current dollar value ofexports in 1997 divided by gross domesticproduct (GDP) reveals an export share of11.9 percent, as shown in Figure 1. In 1960this share was 4.8 percent. Another calcu-lation, which adjusts both exports and GDPfor their respective price changes, yields aslightly different result. Real exports of goodsand services accounted for 16.7 percent ofreal GDP in 1997, 11.9 percentage pointshigher than its level in 1960.3 Thus, theincrease in the importance of the exportsector is larger when the dollar values ofexports and output are adjusted for relativeprice changes. This is because the prices ofall goods and services included in GDP haveincreased by more than the prices of goodsand services that are exported. Figure 1 alsoindicates the sharp increase in the importanceof the export sector to the overall economysince the mid-1980s. Adjusted for relativeprice changes, the share of U.S. output thatis exported has doubled during the past 15 years.

The growth of the export sector hasbeen a driving force in the current expan-sion, as shown in Figure 2. For example,growth in exports of goods and servicesaccounted for 1.6 percentage points of the

Patricia S. Pollard is an economist and Research Officer at the Federal Reserve Bank of St. Louis. Cletus C. Coughlin is a vice president and theAssociate Director of Research at the Federal Reserve Bank of St. Louis. Heidi Beyer provided research assistance.

3.9 percentage point growth in real GDPduring 1997. During 1992–97, growth inexports accounted for more than 30 percentof the growth in real GDP.

After growing by 8.5 percent in 1996and 12.8 percent in 1997, real exports ofU.S. goods and services rose by only 1.6percent in 1998. Moreover, exports declinedduring each of the first three quarters of1998 (on an annualized basis) before risingsharply in the fourth quarter. Thus, incontrast to the 1992–97 period, exportsaccounted for only 0.2 percentage pointsof the 3.9 percentage point growth in realGDP in 1998.

Because data on U.S. exports ofservices are not available by country of

destination for all East Asian countries, theremainder of this article focuses on exportsof goods. Similar to the results associatedwith exports of goods and services, exportsof goods have been especially importantfor the current expansion, at least until1998. Between 1992 and 1997, exports ofgoods in nominal terms grew at an annualrate of 9.0 percent, increasing the ratio ofexports of goods to GDP from 7.2 percentto 8.5 percent. The growth of exports iseven more impressive when expressed inreal terms. Between 1992 and 1997, realexports of goods increased at an annualrate of 13.1 percent. Adjusted for theirrespective relative price changes, the ratioof exports of goods to GDP increased from7.2 percent to 11.5 percent.

GEOGRAPHICAL PATTERNOF U.S. EXPORTS

Coinciding with the increase in U.S.exports has been a shift in its geographicalpattern, as shown in Figure 3 and Table 1.Figure 3 shows the destination of U.S. exportsof goods by geographical area during theperiods 1970–75 and 1992–97. Table 1 liststhe 10 countries that received the most U.S.exports during these periods. The WesternHemisphere has been the most importantregion for U.S. exports during both periods,and its importance has grown slightly. Asshown in Table 1, this is a result of anincrease in the share of exports going toMexico, some of which can be attributedto the effects of the North American FreeTrade Agreement (NAFTA). In contrast,Africa and the Middle East have never beenimportant export markets for the UnitedStates. These two areas combined receiveonly 5 percent of U.S. exports.

The most notable change in the patternof U.S. exports during the past three decadeshas been the increasing importance of theAsia-Pacific area and the declining impor-tance of Europe. Over the period 1970–75,32 percent of U.S. exports were shipped toEuropean countries, whereas 22 percent ofU.S. exports were shipped to Asia-Pacificcountries. Over the period 1992–97, Europereceived 25 percent of U.S. exports and the


34

MARCH/APR I L 1999

Figure 2

Source: U.S. Department of Commerce, Bureau of Economic Analysis.

Export contributionReal GDP growth

Export Contribution to Growth1992-1998

Perce

nt

5

4

3

2

1

01992 1993 1994 1995 1996 1997 1998

Figure 1

Current $

1960 $

Perce

nt

1960 64 68 72 76 80 84 88 92 96

18

16

14

12

10

8

6

4

Exports of Goods and ServicesPercent of GDP, 1960-98

Source: U.S. Department of Commerce, Bureau of Economic Analysis.


35

4 Belgium and Luxembourg arecounted as a single country inthe trade data.

5 In this article, East Asia isdefined as China, Hong Kong,Indonesia, Japan, Korea,Malaysia, Philippines,Singapore, Taiwan andThailand.

MARCH/APR I L 1999

Asia-Pacific region received 31 percent.During the earlier period, six of the 10most important export markets for theUnited States were in Europe, whereas cur-rently only four of the top 10 markets arein Europe.4 In contrast, over the period1970–75, Japan was the only Asia-Pacificcountry in the top 10 list, whereas currentlyfour countries in the region are on the list.Moreover, the emergence of the Asia-Pacificregion as the second-most important geo-graphic area for U.S. exports is not due togrowth in exports to Japan relative to allother areas; rather, it has resulted from thegrowing importance of the markets in therest of East Asia, as shown in Figure 4.5 Infact, the share of U.S. goods exports shippedto Japan has fallen sharply during the 1990srelative to the late 1980s. Meanwhile, withthe exception of Taiwan, the share of U.S.goods shipped to other countries throughoutEast Asia rose sharply during the 1990srelative to the late 1980s.

Undoubtedly, the key factor underlyingthe increasing relative importance of marketsin East Asia for U.S. production has beenthe rapid income growth in most East Asiancountries. Table 2 shows that, excludingJapan, countries in East Asia generally grewmuch faster during 1992–97 than othereconomies—especially the major Europeantrading partners of the United States. Inaddition, the relatively slow economic growthin Japan is consistent with the fact that theincreasing importance of the Asia-Pacificregion for U.S. exports is not due to Japan.

EXPORTS TO ASIA ANDU.S. ECONOMIC GROWTH

It is precisely the increased importanceof the East Asian markets for U.S. exportersthat has made the U.S. economy more vul-nerable to the substantial income andexchange rate changes associated with theAsian crisis. Table 3 contains informationon the sharp decline in growth and, in manycases, absolute declines in overall economicactivity throughout East Asia between 1997and 1998. In seven of the 10 countrieslisted, total activity shrank during 1998.For Indonesia, Thailand, Korea, Malaysia

and Hong Kong, the declines were particu-larly large. Only China and Taiwan hadstrong economic growth in 1998. In bothcountries, however, economic growth slowedin 1998 relative to 1997.

In addition to the sharp declines ingrowth, many countries in East Asia haveexperienced sharp declines in their currenciesrelative to the U.S. dollar. Table 4 shows

Figure 3

Source: U.S. Department of Commerce, Bureau of the Census

1970-1975

WesternHemisphere

37%

Africa3%

Africa1%Western

Hemisphere39%Asia-Pacific

22%

Middle East6%

Europe32%

Europe25%

Middle East4%

Asia-Pacific31%

1992-1997

Global Distribution of U.S. Goods Export

Top Ten Destinations of U.S. Goods Exports(By Share of Total U.S. Exports)

1970-75 1992-97Share Share

Country (Percent) Country (Percent)

Canada 21.4 Canada 21.5Japan 10.2 Japan 10.4Germany 5.4 Mexico 9.3United Kingdom 4.9 United Kingdom 5.2Mexico 4.4 Germany 3.9Netherlands 3.9 Korea 3.8France 3.1 Taiwan 3.2Italy 2.9 Netherlands 2.8Brazil 2.7 France 2.6Belgium- 2.3 Singapore 2.5

Luxembourg

Sources: U.S. Department of Commerce, Bureau of the Census andInternational Monetary Fund, Direction of Trade Statistics.

Table 1


36

6 To be precise, real exchangerate changes, which adjust thenominal exchange rates for therelative rates of inflation,should be used. The use of realexchange rates would not alterthe conclusion that many coun-tries in East Asia have experi-enced sharp declines in theircurrencies relative to the U.S.dollar.

7 Clearly, these estimates arerough. An important assump-tion is that the resources thatwould have been used to pro-duce the exports are not put toother uses immediately. Also,to the extent that the decline indemand for U.S. products bythe East Asian countries hasresulted in a rise in inventories,the negative effects on mea-sured output are reduced. Thatis, the decline in exports maybe offset by a rise in invento-ries in the GDP accounts.Unless producers find othermarkets for these inventories,however, production ultimatelywill be affected.

8 Year-over-year (rather thanquarterly annualized) compar-isons are used because thecountry-level export data arenot seasonally adjusted. Thus,fluctuations in quarterly datacould reflect seasonal patternsrather than economic factors.

MARCH/APR I L 1999

the measured changes for two periods—May 1, 1997 to January 30, 1998 and May 1, 1997 to December 31, 1998.Between May 1, 1997 and January 30,1998, seven of the 10 East Asian countries

experienced declines of more than 15 percentin the value of their currencies relative to theU.S. dollar. The declines in the Indonesianrupiah and the Thai baht were especiallysharp—both currencies fell more than 50percent against the U.S. dollar. When thelonger period is examined, one sees thatthe Asian currencies have recovered some-what. The fact remains, however, thatbetween May 1, 1997 and December 31,1998, seven of the 10 countries listed sawtheir currencies decline more than 10 per-cent against the U.S. dollar.6

Undoubtedly, these income and exchangerate changes adversely affected U.S. exportsto Asia. One rough estimate of the exporteffect of the Asian crisis on the U.S. economycan be obtained by calculating the effect ofthe decline in U.S. exports to Asia on thegrowth rate of output.7 Real merchandiseexports to East Asia fell by 11.6 percent in1998. This decline reduced the growthrate of real GDP by 0.36 percentage points.Consequently, the growth rate of real GDPin 1998 was 3.85 percent rather than 4.21 per-cent. The quarterly data show that exportsto East Asia fell by 7.4 percent on a year-over-year basis in the first quarter of 1998,by 16 percent in both second and thirdquarters, and by 7.1 percent in the fourthquarter.8 These declines had the effect oflowering the growth rate of real GDP by0.2 percentage points during the first andfourth quarters and 0.5 percentage pointsin the second and third quarters. Thesedata provide some evidence that the worsteffects of the Asian crisis may be over.

Industry EffectsThe recent declines in exports to East

Asia have not equally affected the outputof all industries. The effect of decliningexports to Asia on a particular U.S. industrydepends on the following two factors: theextent to which exports to Asia have declinedand the importance of these exports to theoutput of a particular industry. To examinethese effects, we grouped industries accordingto the two-digit Standard Industrial Classi-fication (SIC) system. Details regardingthe data are provided in the shaded insert

Real GDP Growth: Asia and Europe(Percent) 1992-1997

Compound Annual Rate

AsiaChina 11.0Hong Kong 5.1Indonesia 7.0Japan 1.4Korea 7.2Malaysia 8.7Philippines 5.3Singapore 8.8Taiwan 6.2Thailand 6.2

EuropeFrance 1.5Italy 1.2United Kingdom 2.9Germany 1.4

World 3.7

Source: International Monetary Fund, International FinancialStatistics.

Table 2

Asian Real GDP Growth(Percent)

1997 1998

China 8.8 7.8Hong Kong 5.3 -5.1Indonesia 4.6 -13.7Japan 1.4 -2.8Korea 5.5 -5.5Malaysia 7.7 -6.8Philippines 5.2 -0.5Singapore 8.0 1.5Taiwan 6.8 4.9Thailand -0.4 -8.0

Source: International Monetary Fund, World Economic Outlook(May 1999).

Table 3


37

“Details of Industry Level Export andOutput Data.”

Declining Exports. First, we look atthe change in exports by industry. As Table 5indicates, in real terms every industry studied,with the exception of the transportationequipment industry (SIC 37), saw a declinein goods exported to East Asia in 1998.Exports of transportation equipment roseby 3.5 percent during 1998. These exportsrose in all but the second quarter of 1998.Of those industries with declining exports,the food and kindred products industry(SIC 20) experienced the smallest decline(3.3 percent) during 1998. Metallic oresand concentrates industry (SIC 10)exports fell by the greatest amount (34.9percent). More than half of all the U.S.industries studied saw their exports to EastAsia fall 20 percent or more during 1998.

Though the change in exports of someindustries showed a great deal of quarterlyvariation (even on a year-over-year basis),exports in other industries fell sharply duringeach quarter. For example, crude oil andnatural gas (SIC 13) exports declined bymore than 30 percent during the first, thirdand fourth quarters but rose by 28 percentduring the second quarter. In contrast, fiveindustries—forestry and fishing (SIC 08-09),stone, clay, glass and concrete products(SIC 32), apparel and related products(SIC 23), rubber and miscellaneous plasticproducts (SIC 30), and furniture and fixtures(SIC 25)—experienced year-over-yeardeclines in exports of 15 percent or moreduring each quarter.

In most of the industries studied, the fiveAsian countries most directly associated withthe crisis—Indonesia, Malaysia, Philippines,South Korea and Thailand—experienced thelargest declines in exports. China has beenthe one bright spot in East Asia for manyindustries. More than half of the industriesstudied had increased exports to China.

Examining the export data at a moredisaggregated level can provide more infor-mation on what is driving the changes inexports at the two-digit industry level.9

For example, the 3.5 percent increase intransportation equipment (SIC 37) exportsoccurred as a result of rising aircraft

exports, primarily to China. In contrast,passenger car exports fell throughout 1998.

Another industry experiencing a rela-tively small decline in exports to East Asiais tobacco products (SIC 21). Overall,tobacco exports to East Asia fell by 4.9 per-cent in 1998. A slight rise in the export ofchewing and smoking tobacco was morethan offset by a decline in cigarette exports.

Turning to industries that have experi-enced sharp declines, the decline in exportsof finfish are primarily responsible for thedecline in forestry and fishing (SIC 08-09)exports. Meanwhile, declines in corn, soy-beans and cotton exports accounted fornearly all of the decline in agriculture andlivestock products (SIC 01-02) exports.

These export data suggest both thewidespread and differential effects of theAsian crisis on U.S. exporters. Obviously,certain industries have experienced muchlarger drops in exports than other industries.Understanding the full extent of the exporteffect across industries, however, requiresexamining the importance of exports to EastAsia for the industries studied.

Industry Dependence on Exports.The electrical and electronic machinery,equipment and supplies industry (SIC 36)is the only industry whose exports to EastAsia accounted for more than 10 percent

9 These data are available on request.

MARCH/APR I L 1999

Exchange Rate ChangeUS $/Foreign Currency(Percent)

May 1, 1997 - May 1, 1997 -January 30, 1998 December 31, 1998

China 0.2 0.6Hong Kong 0.1 0.0Indonesia -76.6 -68.8Japan -0.4 12.0Korea -41.4 -25.8Malaysia -40.7 -34.0Philippines -38.1 -31.9Singapore -15.7 -12.4Taiwan -19.2 -14.2Thailand -50.8 -28.5

Source: Federal Reserve Bank of New York.

Table 4


38

MARCH/APR I L 1999

U.S. Goods Exports to East AsiaShare of total exports, 1970-1997 (Percent)

13.0

12.0

11.0

10.0

9.0

8.0

JapanPercent

70 74 78 82 86 90 94

3.02.52.01.51.00.50.0

Hong KongPercent

70 74 78 82 86 90 94

5.0

4.0

3.0

2.0

1.0

KoreaPercent

70 74 78 82 86 90 94

1.2

1.0

0.8

0.6

0.4

PhilippinesPercent

70 74 78 82 86 90 94

4.03.53.02.52.01.51.0

TaiwanPercent

70 74 78 82 86 90 94

Sources: U.S. Department of Commerce, Bureau of the Census and International Monetary Fund, Direction of Trade Statistics.

2.5

2.0

1.5

1.0

0.5

0.0

ChinaPercent

70 74 78 82 86 90 94

1.21.00.80.60.40.20.0

IndonesiaPercent

70 74 78 82 86 90 94

1.81.51.20.90.60.30.0

MalaysiaPercent

70 74 78 82 86 90 94

3.02.52.01.51.00.50.0

SingaporePercent

70 74 78 82 86 90 94

1.41.21.00.80.60.40.2

ThailandPercent

70 74 78 82 86 90 94

Figure 4

of gross output during 1996, as shown inTable 6.10 For the U.S. industries studied,1996 exports to East Asia accounted for,on average, 3.7 percent of gross output.Though these numbers may seem small,they are not surprising. Exports of goodsaccounted for only 8.2 percent of U.S.GDP during 1996, with exports to Asiaaccounting for about one-third of the total.

Exports to East Asia accounted for 2.5percent of the gross output of the paperand allied products industry (SIC 26)during 1996, but only 0.7 percent of thegross output of the furniture and fixturesindustry (SIC 25). Exports to East Asia forboth industries fell by 21.3 percent during1998. As a result of the differences in theimportance of East Asian markets, this

10The most recent output-by-industry data are for 1996.

MARCH/APR I L 1999


39

U.S. Real Exports to East Asia by Industry(Percent Change From Year Ago)

1998

SIC Industry Year Q1 Q2 Q3 Q4

10 Metallic ores and concentrates -34.9 -55.6 -40.1 5.8 -66.413 Crude oil and natural gas -32.3 -33.2 27.5 -47.2 -55.624 Lumber and wood products, except furniture -32.0 -38.9 -41.6 -29.6 -10.8

08-09 Forestry and fishing -29.1 -26.8 -43.8 -25.9 -20.532 Stone, clay, glass and concrete products -28.8 -20.4 -24.7 -27.7 -41.129 Refined petroleum and coal products -27.6 -42.3 -42.0 -29.3 23.822 Textile mill products -24.9 -9.0 -27.1 -34.6 -28.333 Primary metal products -23.1 -13.8 -28.8 -27.7 -21.423 Apparel and related products -22.6 -24.5 -23.1 -24.6 -17.439 Miscellaneous manufactured goods -22.4 -23.9 -26.6 -26.5 -11.435 Nonelectrical machinery -21.6 -12.5 -21.3 -25.9 -26.130 Rubber and miscellaneous plastics products -21.4 -15.5 -23.8 -27.9 -18.325 Furniture and fixtures -21.3 -21.3 -28.7 -17.6 -16.326 Paper and allied products -21.3 -23.7 -23.4 -27.0 -10.228 Chemicals and allied products -17.1 -19.8 -18.5 -20.7 -9.112 Bituminous coal and lignite -14.5 -2.2 -15.8 -20.0 -21.2

01-02 Agriculture and livestock products -13.5 -14.9 -18.5 -12.2 -8.931 Leather and leather products -12.4 -23.3 -2.4 -12.3 -11.634 Fabricated metal products except machinery -12.4 -9.5 -22.0 -24.0 6.4

and transportation equipment38 Scientific and professional instruments; -10.2 -3.1 -8.2 -12.5 -16.4

photographic and optical goods, etc.36 Electrical and electronic machinery, -7.3 -3.1 -12.5 -10.1 -3.5

equipment and supplies14 Nonmetallic minerals, except fuels -7.3 -7.0 4.9 -15.9 -10.621 Tobacco products -4.9 -5.4 0.2 -5.7 -8.527 Printing and publishing -3.6 -9.6 8.8 -12.8 -1.320 Food and kindred products -3.3 -3.0 -0.7 -7.7 -1.937 Transportation equipment 3.5 11.9 -18.8 1.3 22.5

Sources: U.S. Department of Commerce, Bureau of the Census and U.S. Department of Labor, Bureau of Labor Statistics.

Table 5

equal percentage decline in exports willhave a much larger effect on the formerindustry than on the latter.

Gross Output Effects. Given the sharpdecline in exports to East Asia during 1998,the effect on the gross output of severalindustries has been pronounced, as shownin Table 7.11 These data indicate the poten-tial effect of the decline in exports on thegrowth rate of each industry, assuming this

reduction in export sales to Asia is not offsetby additional sales in the United States orelsewhere. The average effect across indus-tries was a 0.6 percent decline in the growthrate of gross output. The industries mostseverely affected by the decline in exportsare nonelectrical machinery (SIC 35) andmiscellaneous manufactured goods (SIC39). The former industry includes suchitems as construction, metal working andgeneral industrial machinery as well ascomputers. The latter industry includessuch items as jewelry, toys, musical instru-ments, and office and art supplies.

In the absence of a decline in nonelec-trical machinery exports to East Asia, thegrowth rate of gross output in that industrywould have been 1.8 percentage points higher.Similarly, the decline in miscellaneousmanufactured goods exports to East Asiareduced gross output growth in that industryby 1.3 percentage points. The factorsaccounting for these output effects differin the two industries. The nonelectricalmachinery industry is more dependent onEast Asia for sales of its output (8.2 percentcompared with 5.8 percent), whereas themiscellaneous manufactured goods industrysuffered a slightly greater decline in exportsto the region (22.4 percent compared with21.6 percent).

The transportation equipment industry(SIC 37) received a small boost in output(0.14 percentage points) as a result ofrising exports to East Asia. Of the studiedindustries with a decline in exports, outputof the printing and publishing industry(SIC 27) has been the least affected.Declining exports lowered the growth rateby only 0.01 percentage points during 1998.The limited effect is a result of both the smallamount of production that is exported toEast Asia (0.3 percent) and the small declinein these exports (3.6 percent). Even the13 percent drop in exports of this industrythat occurred during the third quarter hadonly a 0.04 percentage point drag on growth.

The crude oil and natural gas industry(SIC 13), which had one of the largestdeclines in exports during 1998 (32.3 per-cent), was in the bottom fifth of industriesin terms of the effect of the decline in exports

11See the shaded insert,“Another Look at IndustryOutput Effects,” for a discus-sion of our cross-industry resultsrelative to a study by Noland etal. (1998) using a differentapproach.

MARCH/APR I L 1999


40

U.S. Real Exports to East Asiaas a Percent of Gross Outputby Industry (1996)

SIC Industry Percent

36 Electrical and electronic machinery, 12.2equipment and supplies

31 Leather and leather products 8.535 Nonelectrical machinery 8.238 Scientific and professional instruments; 7.5

photographic and optical goods, etc.01-02 Agriculture and livestock products 6.9

39 Miscellaneous manufactured goods 5.821 Tobacco products 5.128 Chemicals and allied products 4.5

08-09 Forestry and fishing 4.437 Transportation equipment 4.124 Lumber and wood products, except furniture 3.814 Nonmetallic minerals, except fuels 2.626 Paper and allied products 2.533 Primary metal products 2.512 Bituminous coal and lignite 2.510 Metallic ores and concentrates 2.520 Food and kindred products 2.234 Fabricated metal products except machinery 1.8

and transportation equipment23 Apparel and related products 1.732 Stone, clay, glass and concrete products 1.630 Rubber and miscellaneous plastics products 1.422 Textile mill products 1.129 Refined petroleum and coal products 1.025 Furniture and fixtures 0.713 Crude oil and natural gas 0.427 Printing and publishing 0.3

Sources: U.S. Department of Commerce, Bureau of the Census and U.S. Department ofLabor, Bureau of Labor Statistics.

Table 6

MARCH/APR I L 1999


41

Export DataExport data by industry for the group

of countries studied are available only formerchandise trade and only in currentdollars. Exports by industry are based onthe two-digit Standard Industrial Classifi-cation (SIC) system. This classification ofexports was chosen for consistency withthe output data, discussed below.

To calculate the real value of exportsby industry, the data were deflated by exportprice indices. Export price index data arenot available on a SIC basis. Thus, westarted with an export price index thatgroups the data based on the StandardInternational Trade Classification (SITC)system and matched these industries withthe appropriate SIC codes. When multipleSITC codes fit one SIC category, a weightedaverage of the price indices for those cate-gories was constructed to arrive at the priceindex on an SIC basis. The weights werebased on the export share of each SITCindustry within a SIC grouping for 1995because this is the year used to weightprices in the SITC index. For example,

the table in this insert provides informationon the two SITC industries (SITCs 78 and79) that correspond to the transportationequipment industry (SIC 37). Using thisinformation, the relevant price index forthe transportation equipment industry was101.24 during the first quarter of 1997 and101.02 during the first quarter of 1998.Thus, the export price of transportationequipment fell by 0.2 percent during this period.

Output DataOutput data by industry are based on

the two-digit SIC code system. Theoutput data are gross output by industryfor 1996 (the latest year for which suchdata are available). Gross outputmeasures each industry’s total output,including the intermediate products usedand the value added during production.These output data, in both current andconstant dollars, are produced by theDepartment of Commerce, Bureau of Eco-nomic Analysis and are available

DETAILS OF INDUSTRY LEVEL EXPORTAND OUTPUT DATA

Conversion of SITC Price Index to SIC Price Index forTransportation Equipment Industry

1997:Q1

SITC Industry Price Index Weight Price Index * Weight

78 101.63 0.64 65.01

79 100.53 0.36 36.22

Sum of Weighted Indexes (SIC 37 Index) 101.24

1998:Q1

SITC Industry Price Index Weight Price Index * Weight

78 101.90 0.64 65.19

79 99.47 0.36 35.84

Sum of Weighted Indexes (SIC 37 Index) 101.02

on output (0.1 percent) because only asmall fraction of SIC 13 output is exportedto East Asia (0.4 percent). Turning to thetobacco products industry (SIC 21), thelimited effect on output (0.3 percent)results primarily from the small decline inthese exports (4.9 percent).

The final two columns in Table 7 pro-vide some perspective on the significanceof these output effects. These two columnsshow the average and range of the contri-bution of exports to East Asia to the growthof gross output over the period 1990-96.In five industries—metallic ores and con-centrates (SIC 10), bituminous coal andlignite (SIC 12), lumber and wood products(SIC 24), primary metal products (SIC 33)and tobacco products (SIC 21)—exportdeclines, on average, reduced the growth rateof gross output. Even in these industries,however, the effect of the 1998 decline in

exports to East Asia on gross output wasgreater than the average effect. Thus, forexample, the decline in exports to EastAsia in the metallic ores and concentratesindustry had a 0.86 percentage point dragon the growth of gross output during 1998.On average, over the 1990–96 period, thedecline in exports to East Asia in thisindustry lowered the growth rate of grossoutput by 0.29 percentage points.

In only nine of the 26 industries studiedis the current negative export effect ongrowth of gross output within the range ofexperience throughout the 1990–96 period.Moreover, eight of the industries neverexperienced a decline in exports to EastAsia during this period. Nonelectricalmachinery exports to East Asia, for example,contributed between 0.21 and 2.36 percentagepoints to the annual growth rate of grossoutput throughout the 1990–96 period.

MARCH/APR I L 1999


42

electronically at: <http://www.bea.doc.gov/bea/dn2.htm>. Current dollar grossoutput is roughly equivalent to an industry’ssales or receipts. Our analysis uses the1996 current dollar gross output data foreach industry.

Measuring the Effect of Changes inExports on Output

The percentage change in the realexports of each industry to East Asia isgiven by Equation 1, as follows:

where Xi, j is the real exports of industry ito country j and the subscript t refers tothe quarter in 1998 and 1997. The resultsfor each industry are given in Table 5.

Exports to East Asia as a share of thegross output of each industry in 1996 isgiven by Equation 2, as follows:where Xi, j is the real exports of industry i

to country j, and Yi is the gross output ofindustry i. The year 1996 is used tocalculate the importance of the East Asianmarket for each industry because it is thelatest year for which gross output data byindustry are available. The results of thiscalculation are given in Table 6.

Combining the percentage change inexports with the share of exports in grossoutput indicates the effect of the changein exports on output, as shown by Equa-tion 3, as follows:

The results of this calculation are given in Table 7.

( ) *,

,

,,

1 1 10098

1

10

971

10

X

X

ti j

j

ti j

j

=

=

∑

∑−

( ) *

,

2 10096

1

10

96

X

Y

i j

j

i

=∑

( ) *,

,

,,

,

3 1 10098

1

10

971

10

961

10

96

X

X

X

Y

ti j

j

ti j

j

i j

j

i

=

=

=∑

∑−

∑

Thus, the 1.77 percentage point exportdrag on growth in 1998 is particularlytroublesome for this industry.

CONCLUSIONThroughout most of the 1990s the

United States has enjoyed the benefits ofrising exports, particularly those to thegrowing economies of East Asia. Over thepast year, however, the East Asian economieshave contracted and their currencies have

fallen sharply against the dollar. As a result,demand for U.S. products has declined andU.S. exports to the region have fallen. In1998 real U.S. exports to East Asia fell by11.6 percent.

Examining the changes in exports byindustry indicates the pervasiveness of thedeclines. Only the transportation equipmentindustry sent more goods to East Asia in 1998than in 1997. This was a result of increasesin aircraft exports, many of which were orderedprior to the economic problems in East Asia.

MARCH/APR I L 1999


43

Export Effect on Growth by Industry – Gross Output Basis (Percent)1998 Average Range

SIC Industry Year Q1 Q2 Q3 Q4 1990-96 1990-96

35 Nonelectrical machinery -1.77 -1.03 -1.75 -2.13 -2.14 0.86 0.21-2.3639 Miscellaneous manufactured goods -1.29 -1.38 -1.54 -1.53 -0.66 0.41 -0.23-1.30

08-09 Forestry and fishing -1.27 -1.17 -1.92 -1.13 -0.90 0.22 -0.70-1.9824 Lumber and wood products, except furniture -1.21 -1.47 -1.57 -1.12 -0.41 -0.12 -0.36-0.1931 Leather and leather products -1.06 -1.98 -0.20 -1.05 -0.98 0.41 -0.46-1.46

01-02 Agriculture and livestock products -0.93 -1.02 -1.27 -0.84 -0.61 0.11 -0.85-1.7236 Electrical and electronic machinery, -0.89 -0.38 -1.52 -1.23 -0.42 1.72 0.45-3.65

equipment, and supplies 10 Metallic ores and concentrates -0.86 -1.38 -0.99 0.14 -1.64 -0.29 -1.89-1.5228 Chemicals and allied products -0.78 -0.90 -0.84 -0.94 -0.42 0.19 -0.22-0.5838 Scientific and professional instruments; -0.76 -0.23 -0.61 -0.94 -1.23 0.51 -0.04-1.16

photographic and optical goods, etc33 Primary metal products -0.57 -0.34 -0.71 -0.69 -0.53 -0.01 -0.82-1.2326 Paper and allied products -0.53 -0.59 -0.58 -0.67 -0.25 0.13 -0.17-0.4532 Stone, clay, glass and concrete products -0.46 -0.33 -0.40 -0.44 -0.66 0.13 0.04-0.3223 Apparel and related products -0.39 -0.43 -0.40 -0.43 -0.30 0.18 0.03-0.3230 Rubber and miscellaneous plastics products -0.30 -0.22 -0.34 -0.40 -0.26 0.11 0.04-0.1912 Bituminous coal and lignite -0.36 -0.05 -0.39 -0.50 -0.53 -0.16 -0.38-0.1222 Textile mill products -0.27 -0.10 -0.30 -0.38 -0.31 0.03 -0.04-0.1429 Refined petroleum and coal products -0.26 -0.41 -0.40 -0.28 0.23 0.01 -0.39-0.2921 Tobacco products -0.25 -0.27 0.01 -0.29 -0.43 -0.01 -0.85-1.0134 Fabricated metal products except -0.22 -0.17 -0.39 -0.42 0.11 0.12 0.01-0.24

machinery and transportation equipment14 Nonmetallic minerals, except fuels -0.19 -0.18 0.13 -0.42 -0.28 0.12 0.04-0.2225 Furniture and fixtures -0.14 -0.14 -0.19 -0.12 -0.11 0.06 0.01-0.1113 Crude oil and natural gas -0.12 -0.13 0.11 -0.18 -0.21 0.04 -0.13-0.2620 Food and kindred products -0.07 -0.07 -0.01 -0.17 -0.04 0.11 -0.25-0.2827 Printing and publishing -0.01 -0.03 0.03 -0.04 0.00 0.01 -0.03-0.0437 Transportation equipment 0.14 0.49 -0.77 0.05 0.92 0.31 -0.23-0.89

Sources: U.S. Department of Commerce, Bureau of Economic Analysis and Bureau of the Census; Department of Labor, Bureau of LaborStatistics.

Table 7

MARCH/APR I L 1999


44

ANOTHER LOOK ATINDUSTRY OUTPUTEFFECTS

Clearly our results are roughestimates of the sectoral effects of theAsian crisis on the U.S. economy. Priorstudies, such as Noland et al. (1998) andWerling and McCarthy (1998), explorethe relationship between different macro-economic environments in Asia and theirtrade and production effects across U.S.industries. Because our approach usesactual trade changes, our results are notstrictly comparable to prior studies.Nonetheless, as a check on thereasonableness of our results, we compareour results with those generated byNoland et al. (1998).

Using a computable generalequilibrium model with 17 regions, eachwith 14 sectors and five primary factors ofproduction, Noland et al. explored theconsequences of various macroeconomicscenarios. The Asian crisis was character-ized as a combination of two types ofshocks—real exchange rate depreciationsand declines in GDP—in eight Asian

countries. These shocks generate changesin trade flows and production.

The production changes associatedwith one scenario, termed the medium-shock scenario are listed in the table inthis insert. Because the correspondencebetween the sectors used by Noland et al.and the two-digit SIC classification we useis rough and because their time horizonextends beyond 1998, it is difficult tomake strong statements.1 It appears, how-ever, that the cross-industry results arecomparable. Similar to our results thatthe nonelectrical machinery industry hasexperienced the largest relative decline,Noland et al. identified the machinerysector as experiencing the largest relativedecline. The electronics sector also wasexpected to experience relatively largedeclines, a result consistent with ourfinding for electrical and electronicmachinery, equipment and supplies. Inaddition, Noland et al. found that boththe agriculture sector and the forestry andfishery sector are likely to undergorelatively large production changes—results consistent with our findings for theindustries encompassing agriculture andlivestock (SIC 01-02) and forestry andfishing (SIC 08-09). The most obviousdifference between our results and thoseof Noland et al. concerns potentialchanges in the transportation equipmentindustry. Noland et al. estimatedrelatively large changes; to date weestimate relatively small changes.2 Someof this difference can be attributed to thefact that Noland et al. also capture importeffects that we do not.

1 Noland et al. also do not include China and Hong Kong in theirstudy. Excluding these from our sample does not change the compa-rability of the results.

2 Our transportation equipment industry results are consistent withthose of Werling and McCarthy (1998). They estimated below-average production declines for motor vehicles and parts.Meanwhile, for aerospace, which accounts for a relatively smallershare of the transportation equipment industry than motor vehicles,they did find above-average production declines.

Medium Shock Scenario —Output Effects (Percent)

Industry Change in Production

Machinery -2.63Electronics -2.38Forestry and fishery -1.39Motor vehicles and parts -1.21Other transportation equipment -1.13Agriculture -0.89Light manufacturers -0.82Intermediate goods -0.71Textile and apparel -0.63Mining -0.49Energy -0.01Processed food 0.02Services 0.39Housing and construction 0.71

Total -0.07

Source: Noland et al. (1998), Table 3.1.

Moreover, more than half the industries studiedexperienced declines in exports to East Asiaof more than 15 percent. Such declines arehighly unusual. Between 1990–96 onlynine of the 26 industries experienced a year-over-year decline in exports as large as theones they experienced during 1998. Becausethe effects differ across industries and withinthe two-digit SIC industries, these exportdata also suggest that certain regions of theUnited States may be more affected by theexport declines than other regions.12

Focusing solely on the export dataoverstates the relevance of these declinesto the industries in question and presentsan inaccurate picture of the industries thathave been hardest hit by the export effectsof the Asian crisis. The effect of the declinein exports on the output of a particularindustry depends on the extent to which thatindustry is dependent on the East Asianmarket to sell its output. Incorporating theexport declines with the market share dataindicates the extent to which each industryhas been affected. For most of the industriesstudied, the decline in exports has loweredgrowth by 0.4 percentage points or less. Formany industries, however, output declinedby 1 percent or more. Generally speaking, ourcross-industry results are consistent with thesimulation results of Noland et al. (1998).

These results, however, need to beinterpreted with care. They may be biaseddownward because they capture only thedirect effects of a decline in exports to EastAsia on industry output. They do notincorporate any secondary effects. Thus,for example, the stone, clay, glass and con-crete products industry may be both directlyaffected by declining exports to East Asiaand indirectly affected by reduced ordersfrom other U.S. firms as a result of decliningdemand in East Asia. That is, as the amountof transportation equipment sold to EastAsia declines, the transportation equipmentindustry’s demand for window glass declines.On the other hand, increases in demand byU.S. and other foreign consumers andbusinesses may mitigate the effects of thedecline in demand in East Asia.

Our focus on exports also ignores thepotentially negative effects on some U.S.

producers-decreased demand for theiroutput stemming from increased imports.The steel industry is a specific example.13

Imports of steel rose by 24 percent during1998. This surge in imports can be connectedto the Asian crisis in that reduced steeldemand throughout Asia has reduced theworld price of steel and has lead producers,especially those from Russia, Japan andBrazil, to ship their excess steel to theUnited States.14

REFERENCESGazel, Ricardo C., and Russell L. Lamb. “Will the Tenth District Catch the

Asian Flu?” Federal Reserve Bank of Kansas City Economic Review(second quarter 1998), pp. 9–26.

Greenspan, Alan. “The Crisis in Emerging Market Economies,” testimonybefore the U.S. Senate Committee on the Budget, September 23,1998, <http://www.bog.frb.fed.us/boarddocs/testimony/19980923.htm>.

Lucentini, Jack. “Commerce Could Soon Levy Steel Penalties,” TheJournal of Commerce (November 25, 1998).

Neely, Michelle Clark. “Paper Tigers? How the Asian Economies LostTheir Bite,” Federal Reserve Bank of St. Louis Regional Economist(January 1999), pp. 5–9.

Noland, Marcus, Li-Gang Liu, Sherman Robinson, and Zhi Wang. GlobalEconomic Effects of the Asian Currency Devaluations. Institute forInternational Economics, 1998.

Roubini, Nouriel. “What Caused Asia’s Economic and Currency Crisis andIts Global Contagion?” <http://www.stern.nyu.edu/~nroubini/asia/AsiaHomepage.html>.

Valletta, Rob. “East Asia’s Impact on Twelfth District Exports,” FederalReserve Bank of San Francisco Economic Letter (November 20, 1998).

Werling, Jeffrey F., and Margaret McCarthy. The Asian Crisis and U.S.Industry: An Industry Perspective. Manufacturers Alliance, 1998.

12See Valetta (1998) and Gazeland Lamb (1998) for an analy-sis of the regional effects of thedecline in exports to East Asia.

13Though increased imports ofsteel might have adverseeffects on U.S. producers ofsteel, the lower priced importswill benefit consumers and pro-ducers of goods requiring steelas an input.

14As a result of the increase inU.S. steel imports, 12 U.S.steel companies in conjunctionwith the United Steelworkers ofAmerica, have filed dumpingcharges alleging that steel pro-ducers from Russia, Japan, andBrazil were selling steel in theUnited States below their pro-duction costs. See Lucentini(1998) for additional details.

MARCH/APR I L 1999


45

MARCH/APR I L 1999


46

going down: the asian crisis and u.s. exports · of the east asian markets for u.s. exporters that...

Documents