AL AO09 798 CENTER FOR NAVAL ANALYSES ALEXANDRIA VA INST OF NAVAL-ETC F/6 5/3EFFECTS OF TRADE RESTRICTIONS ON IMPORTS OF STEEL. (U)NOV 60 J M JONDROW

UNCLASSIFIED CNA-PP-165 NL

I fllf.f..llfffMENENDo

':~~ I IIIIDIIIIIIIII

EFFECTS OF TRADERESTRICTIONS ONIMPORTS OF STEEL

James M. Jondrow

CENTER FOR NAVAL ANALYSES

I. TMUION r ATMVE T AApprnwod for 81 2' 09 061

PROFESSIONAL PAPERt165 /November 1980

EFFECTS OF TRADERESTRICTIONS ONIMPORTS OF STEEL.

James M., Jondrow

Institute of Naval Studies

The Public Research Institute%A Division of the Center for Navel Analyses

2000 North Beauregard Street, Alexandria, Virginia 22311 DTJUIN~Af7

TABLE OF CONTENTS

Page

Introduction ..................................... 1

Introduction and Summary of Results ........... 2

The Historical Context ........................ 2An Overview of Methods Used to Estimate

The Effect of Free Trade .................... 3The Data Used for the Study ................... 3A More Detailed Description of Methods Used

to Estimate the Effect of Free Trade

on Prices and Output .......................... 4The Impact of Free Trade ........................ 6

Appendix A: Equations Used to Describethe Steel Market 9

Appendix B: Techniques Used to Estimate Gainsand Losses to Particular Groups 12

Literature Cited ................................ 16

Comment: Edmund Ayoub ........................ 17Comment: Hendrik S. Houthakker ............... 19Reply: James M. Jondrow ....................... 22

LIST OF TABLES

1 Cumulative Effect of a 1 Percent Fall

in the Price of Imported Steel on theRatio of Imports to Domestic Shipments 4

2 The Estimated Effect of Import Restrictions1969-73 5

3 Employment Estimates With and Without Free Accession ForTrade 7 NTIS GRA&I iz

DTIC T.,4 Layoff Estimates With and Without Free ' -:Inoncd / [

Trade 7 t ifi t - _

5 Gains and Losses Due to Free Trade 81 F i..

LIST OF FIGURES Av&... - >,.i

1 The Share of Imports in Apparent Consumption ,;!--t-- Actual, Predicted with Free Trade, andPredicted with Restrictions 6

IntroductionHarry Grubert

sq61he paper% in this w sioti ai (%% iii e sili ofa scries cirs Iil he riodmecd titn ouslti loaititns I or evint-of industi y studies J)% the thuicau tfit tel ii titiiiI I ibot ple. Ni I tdei th iat Is Ihe ic i kci t ii(cinanAffaiis. In aiddition to thesec it. 1iJic% has c been itand It linese sI11,11 iti e sp~nds III the 1' ititei Staite%.completed tin till. fooisscar mnd sigir, a.ndt i it .ti] Jalpinese aind ( .ei him p ititdticif nis %elI find it ciliwaiy tin ceinicails ant clectionics. Iihothesid c CitI ti cstablishi plants fit this coiinhi, .I it: dc ltipitinttaike% a c lose look at the special I eatmie s oft d pail icul ar t hat . it coni h as alt cadi star tedindustry in contrast toi sotit! of the moti c btia rti angitig Recining to d nand tot a motment. I doi not ih i t)projects reported in other session,, & '*, impl v that tic .ii A t dcrnand IM tste Is 1% iii ii -

The sleet pi otce re ported by %ir Jo1 1tdrp for (he Li ~ gilli tti t.it gh'o Idettiiit -tit't ilint lii lI inPublic Hesearcit irItIN1ile is as (he tinTs! ni.i1ov itiiitIN quilits- biseii tn etn and doicNiit- eci iarisesstudy unuiertakcii b- the BiruCait of Inelca.1 ornil I ihiti quest ion oAI osi Wtii. do forciri .iiid unsitc steelAffairs (II Al). It is as intended as a protith pr lot tile sell att diltetciii t les.1 it tnt lil . .iiu liji iici . ish%

scenes in the sctts that it% comuponent ill WIIII teInd to hc iic% filhe preic t dillci nll ii c hange 5(1 fsta.nniN aili s it

found sin the tither indii stty si iidieS. These i ndu siry dues . oits lii? till M I ond ross has e-wi tin ned this pi ohstudy cnieills ate I ).Ii ani lysis of the ic' ponsive ness te nt fi a follow -itp slt iv t hat is ltist heint! i olipic lcd aridol inipiit deninil it) rclaius prices. (') file deteCnni 1tind that foucivin and domestic sl ate nicot really

nari ofdomsti ard ftetn p ics~ t he cl~iiotsh tk~t :k detc l podutsA Ii on tile pitrchisci Pinit kit tess%

bet%%ci c hangeIl OIC1 iIi'sii illu and Citiplit) nliciti [ ViuCten steel 1.ikes IitIICli longer it) get dclii eteil ilician.1 (4) the estittatnion tif losses i lifhetime eaitrnigs binig on dered. arid the dci I a g is, miC antatic.incurred by displaced is oikers in the tttdiislv. This last( Mice c tse bti iiei %lfi peds on licigi stfeel %Isllarea, car tiitg loisses5. is, oiic in % icli 511 tHesci nch ex pcitenrce g emit % sII i.i lIy Ill S iIliIc. thtese fic totshidspioneered. Itis repoitcionnmioi c ii It ii he paper inticase invin-it his. tf tl ind i~ctsis %kti liii tcaits thtby I onis I a &thson in the session onl 1.0toi adj Us merit. - iaiiv. e ustonict, %Isll tic %k illting ito pay~ a substantial

Althuuiugli eachti i h ittdistt stildies telds ito have p 1cinnIII Cot doitiestic: steelsimilar cottipotrents. tile piper by Vitic odcr mit alil It IunaN be tisefutl Coii tile it pioint to .i fcs sliitlantiestomobsiles dicnuinst rates, t huoe ct cy iitdiistt vhasspecial tnd cm itiass Iii the utu atid steel stidicreults RoithfeaturCS AIIIJ ltiClu ,1, lesit liter ics Ii the iniA.si%. fiid that thle Pt tccelIistt sii the demiand tiil tiriplirt isIn uiutiintibilcN. tite tlfereticcts fin l~niiisis tcliii'. e loss Ili the shut ititl. but seciti it) be sets,anmong miakes of cars arid (thc di ffcu ing t.15 its tf cion - hige ii tile long iin. On i life tither hand, dinemitc pricestiet CS for thesw chaI-aIlt istic s ciCA eat L3 cstatpi oh- setting be ti ot fin tie a iiii anid steel intduitIites seemilslems. A tinec sties in;0% sis telatiug Imprt's ito the to iit. -It is estittited iti.ii ditncit steel plikes aiuerelalis e Pt ices of ittipilt ied anid ioitic ats is not tintialested hi. cliarg i iimort prices. lit that dihts.entirely adequnate .It is nit citittgit li ai icc PtteIncdi- tic atirtinuibile ptiScs ti~lliitto Iii ililit iomtpetitionrcc-, fii the e hatiginri chlaraici i cl Slit uicars, ui ing sot signifit aitki I hetic a tititobile IttidN didi not Ilt is igateculd heditnic methlods %%ish mitake adtiiStIMCei ioll [Ihe liii iinitiir of lot cigi ptic. b( b ite sissi rcsitlts arediffere nt-es iii charic te ii ls. i-or nitott 11V is ll ik ita11nd (It' t11iltCi st. \It forld!it f~ ids t hat iipii tpices are

for imports depenrd ol i clit ti prices, holdting qiial ity highly'~ re spirniis e itcapa it v li ia 1 b til-in eon-

constant. butlls tlist) is m nttu ittakes tii pat tictilat ti .st to domestic steel pi ices mis ut atipeit unafces-eitypes of eats %are ais ai laile . I ii exa.mtple. thte i eMay be . hi' diematnd cuutiil i ils. ; citltia 1.15 ith sPii'ufe ssirswitch ltodomnestics if I S. pnitduiite s s toirimaik ecairs IIlt ihik Let spec tall\ ntote s i hitis toiti til tthat are ct ieit) fiicitii atowi obiIcs fit tlills (itf si/c itcittcltisltri. I istilid like to cniipha si.c [ti lCat Ithat

and perfor ma.it lies .1i sc ott (i tic ncidt enrci i i ich tif te i bttihe itt these st idkies Is it) fi Iipi ii tilie Cs .ltnamdemand ara i.tht*c11 twoileiI pitti tect tk-eispd the tio timtf tiade putlicvy. W~ ien c hinge s in it lilt bilt rcrshedonic skmi e ttrde de ss itbed I)% NI r I ielr . arecotrnsileed eit1 is tfiki.i l ittipi i art to ik ri i hi t

In part btcatse tiC thfe ii icts oIi ctitsuimite latas . domtiestic kictitlind tesponds toi ittiil ksl Iss - os m-cconnutte% taf ss tl en toii bei ii' nlf, i,f l llt ite ki- plots tie Al istl he itcte.td itti \% hat "~.Ill I" tdw Itiirtct

toinubihe itdusiry life sttudi , thetetitte fttwi it tiecs- if [tic abuti itiket espetiisne tif' isoit s itil OiI-.

sary tot mike .u w cllev~fttliitiln til th~e titiiini Offm 0Ihuti s th le thut ittittisitv stiltit's piceetitd hiere ireClemi stale hut lItionIhtle plitis I his itiniit.ttutn Ii itt ilti t tit expand the .intint oIi as amfltblc tIttiuiturn. becoimies ipiptI.ini in pieduliin g ssIt i kinds itt tiat it il these qumesltins

Effects of Trade RestrictionsOn Imports of Steel

James M. Jondrow °

Introduction and Summary of Results would have approximately eliminated net income inthat year.

In analyzing the relative merits of free and restricted Removal of import harriers would have generated

trade, economists have tended to concentrate on the net gains to purchasers of imported steel. These gains, in

long-run effects: whether the long-run gai to one the form of lower prices, would have accrued both to

group exceed the losses to others. This paper uses an industries using steel and to consumers of steel-using

econometric model toestimate the short-term gains and products. We estimate that these gains would have

losses to particular groups from free trade in the steel more than offset the total losses to steel workers and the

industry. Because there has not been free trade, we domestic industry.studied what would have happened had there been freetrade in the period 1969-73.

During 1969-73, there were two trade barriers. One, The Historical Contexta tariff of about 7 percent. had existed for a long time.but was declining because of the Kennedy Round Before 1958. imports were not much of a challenge.agreements. The other, a voluntary quota starting in Then in the late 1950s. the import price dropped sharply1969, had been accepted by Japan and the European from well above the domestic price to about 15 percentCommon Market, the major steel exporting countries, below. The import price remained 10 percent to 20

One central finding of the study is that removal of percent below the domestic price during most of theimport barriers would not have led to a sudden shrink- next ten years, whereas the import share grew from lessage ofthe domestic industry. Import penetration would than 2 percent to almost 17 percent in 1968.have built gradually. In the first year of our period of Several reasons have been offered for the weakeningstudy, imports would have displaced less than 2 percent of the domestic steel industrys competitive position.of domestic shipments. This percentage displacement The "over-valuation" of the dollar has been suggestedwould have increased gradually to about 8 percent four by Floyd [81 as a cause for the U.S. trade deficit inyears later. general and by Thompson 1 15. p. 87[ as a cause of

Employment would have been reduced more than growth in steel imports.proportionately to output because the oldest, most In addition. American comparative advantage maylabor-intensive equipment would have been most af- have been shifting away from steel. The steel industryfected. The employment reduction would have been in the United States, relying on relatively old plant%,.gradual, and would have come about, in part, by attfi- had lost some of its technical superionty oser foreigntion. Still. there would have been some layoffs. We producers. Evidence for this position offered by Dir-estimate that removal of trade barners would have in- lan and Adams 121 is the slower I' S. adoption of theduced annual layoffs in the first years, averaging to basic oxygen furnace. In addition, foreign steel indus-roughly 3 percent of the labor force. On the average, tries had begun to recover from World War II. Thorneach displaced worker would have lost in lifetime earn- 11,11 suggests that the U.S industry also faced muchings an estimated S6 thousand fn 1960 dollars). higher wage costs than their foreign compelltors.

There would have been losses to the steel industry. In A source of import penetration not related to com-the year of greatest effect, tosses to industry would paiative advantaize is the existence of -excess'' capac-have amounted to about 3 percent of gross revenue and ity in foreign ctountrc',l I S. pp. 17 271 Another is the

policy of the domestic steel industry to hold prices*This paper n based on "'Removing Reincin, on tmrlrs of constant in the face of import competition II. pp

Steet." StAt 73-8 by J,.mc% J'indio. Ituipene tksirn I mwg,Jacol". Arnold Kait. amndtjd 0Nill,. Puhi. RswAr t nP, 627-291. Indeed., when imports precipitated a tempo-tUte. Center for NAval Anal %e1, P17% rary price %ar in 1968, the press commented that pri~c

-2-

cutting %kas a tact ic doniest fi em had loiig as tided quota begins %ah thit inailvsi% of comitielit ton hetw~een(161. Imtporte'd and doni xli'. steel . Stat stcal relaitions are

Finally. A meric an la bor isputes encttraged itt,- sittited %% h ich t clPt e xc it (he hc.ha im toM. pilt ticipantsports. Ihe first ni tir i nflos of impot Io s t I onded %% it h ill tic steelIn kit. dolie sic si ci pill, lt.kers * onles-the record I W6dai, %teel strike ttt 19 Fit epialce lost Ii'. 'iipphvrt. 11i1d lotce'ti spplir N I hei:' telations atedomestic shipnients. piihaset s Iii ed to inipott. tisct to esitethe i pitlieticil 't]Ll I i enitititmore than dJOublt ttg Im~port t ornmige AIet thei stiiket. tont ic stiecl imp o i 14t 1 ingd leaingtti, thertfthe imiport share did nti) iclin to its Pt e - tike Ic ci. oft thtrotigh IIJ 71 I liec 1ccit alle suitnia 11iid .ts difler-perhaps becatise Anitctar purtchascrs had gained ettces due it) ttice Iit tii domc-t ti' Nhipments. price.fantiliarit twith Inmports. import slipot'.-nts anld 1,topoil Pies

Throughot the sttes. the perstsi'tt Import price -1he estimated etlcct ont dontestic shipitent% is thenadvantage led to aI stead inc tease in the tImport share ti sed it) est t mat eiplo% utictt and eat rntgs . and Ito csttbttt. in addition. ittipot t petnetraition ikk ass tncreasedt mitte the efc~t (tt the a ellare of pitticititi groupswhen a strike was threaitened. IThe nex~t sectioti describes SOme tit the dlata used for

By 1968. the inipoit t hat e hid teatched at peak oft estittattit.n I his iskiolis.edbh, a ntot detailce dip-

almost 1 7 percent. I Itoutelt part otis, iA a, due to hedg - towit th le iet hodi I. tgy and results,.ing Meore at possible sit lite. it still get cimatcd coticet n tin

Congresis. ro heaid oIl pus svibfle intitdatt s quiot a. themajor exporting countries agreed ito the soltintarY The Data Used for the Studyquota.

The quota eovered the like sears 1909- 73. Since Ntlih oft the data rued ttt thts stttd %kas taken fromimport penetratioin wats %till t ticasitig V. hen the qutot a stIla ida, sotiesc. I t it exait pe, tonnage figiies tor it,,-was inst ituited, a stable ekluo itrti nt had not Ise'i pit and donit' 'c shipmnentsi cere as ailibte t ton thereached. The ciffec t of thle quota %aa to Ittit t further Anit.tic an l ion and Steel Inst it t e Animail 1111 tua/import penetration. Repot. I'Lese seties mte reprinted ti the Commerce

The qruota was not the tois factor limtting 1ttuPOrts In I)ep;ltt toetIt s .itlri of C tirrivtg ilb,tinvti . Sonie in-this pci tod. HIigher %&is Ild di-inant iti epecii itIsii 1 9 69 putt ati data k4 erc not as il ahle ft' nt government stfatlt%-and 1973). i ncreasintg foi cign latbor c~osts Sintd doii it tic.,! series, ats aI re sit, tcs found Itit su I t it, go ito bastcdevalutations ill vsoulld have pltied a pairt if the quota sutcs

had not been in effct. In the finial ecar of the quota, 1-it cxattple. there is no U.S. governnment price1973, these tactuirs cicated stuch a high imipot t pt ice that tII1 tles for Pi pted steel. Pethaps t cle lIte1 mieasutre I%the quota ss as not binditng. the unit value of it ipiited steel. thi ciististns %alut:

There has e been v itleir cviclical sviings Dltit tg late dts iledd ti itmptort Ilinntige t'his is not serv acuriate.1973 andi 194. there "isa itt estiei os.ie shii- tiuA,%cr. %itc a shilt titti lossr to higher valued steelage in steel. restiting ini import piies tat abose the "Isll incase the tutit %,itle. esen if ilie tcttial price isdomestic price. Then. ti 117S . %korlil itetoati fell shi P- ticlttilteil InI aidditin. (tie %a luatitot ita gis en Item1)y and thle inirt i ice till Iiltathe diomest ic Ih C,11 ti e kte tet motneil in a i ither of %a% s ani 'ftrequentlyvfall ti U.S. demtand Aa iti .tcetttited hs ciunstitner itt iluii not1 te1the l ctial trinsactiott ikie' II 1 p. 11,ventory rrindosstl ,and the tesultirug iliitp Intt iliCIK~~i to coninstict ati ittil ptit:e intdex. wc used tradenull shiptments %ai tiiieci-teftil ;in lii postivat qii's lilionttit ixport prices for Japant and theperiod. As a ot Id fil it ric iii ited ii it Isl 1

9-6, thle 'it tittin MaI at I [hi-si prices are pidiliisltei fo r par-

import price approached thec dititestic piicc, h ilt sinice ki,-l t sprs ill ste-el tith fInt Hishl iti oletirnal. liwthen, the inptt price hiaigaint fallen blii i thle iii its M 31', lt lit tint Iheic iltdjiial pi ices A cic e a ight edtic. itti itytiel ithto -I sittylt- tndex (il the ftiteign dollar

One of the most strilking lacts Ibitti the stu'i' inuts espOi t pci'.etry-'log erodofs i lte itli sis t hit ft tt rt peni- ia lot this i' pot l ie ve adided l r n spot i and ta.riff

tion was sI %lIt% A'lthtigh the peticiiation sirs little Losis ti .ijpiouisiii it' tI t, sLiid tme. I lie tians-lenting. intports neverthtcattetied tii ake:over the mta~r- I'ol '(ittitssis li iti d flititt Il( tihaurttrifle

An Overview of Methods Used to Estimate Vc- A1sontdv ,HI-,tilitt ii t ittlis oittIiiign

The Effect of Free Trade ntittls 11col su.I i i ~still i iaietlJipiti. ime ithe lip : el.-tt rit "tiulhies a1 set tit

Our estiniate (if the effec~t of tentos tug thi- imiiil anid iniput muititt pi'e ith,,s lotittisiitit.bttIr

IFurope. it was necessary to create an index of steel Another important determinant of the import-,iiput costs from price indices for separate factors. domestic quantity ratio is the expectation of a strike.it.erman factor prices were used to represent luropean When a strike is expected. consumers often turn tocosts.,1 imported steel because of limitations on domestic

capacity and because dcivenes duiing a strike can beguaranteed only by import supplie ,s. The estimated

A More Detailed Description of Methods equation indicated that a period of hedging against aUsed to Estimate the Effect of Free Trade strike. such as the one in 19%. would increase the de-on Prices and Output pendent variable by about 3 percent.

Sibsituti ,i heise 'ti %tel and other prodhcis. InThe economic model of the steel market used to esti- addition to encouraging substitution into imports frommate the impact of free trade has four behavioral equa- domestic steel. fiee trade also encourages substitutiontions. The first two describe demand: the substitution into steel from other products. The second equationbetween imported and domestic steel and the suhstitu- describes the degree of substitution into steel. estimat-tion between steel and other products. The remaining ing how consumption of .stel changesti hen the price oftwo behavioral equations describe supply: pi ice setting steel f all%.behavior by producers of domestic and imported steel. The dependent variable in this equation is a measureThe model is completed vith several identities. of the total amount of steel, domestic and imported. The

The following sections give a geneal description of measure is not the sun of the two types of steel. but athe equations and results. The equations themselves are slightly more complex magnitude based on the assump-presented in Appendix A. tion that the two are imperfect substitutes %,ith a con-

Substitlion beliieen imporled and d mar.inc sieel. stant elasticity of substitution.The first equation describes the choice between domes- This dependent variable turns out to have only atic and imported steel. The variable to be explained is limited sensitixoiy to the price of steel as a whole. A Ithe ratio of imports to domestic shipments. percent drop in the price of steel. relative tothe price of

An important determinant is the import-domestic st el-using outpu, leads to a 0.4 pe rcent rise in steelprice ratio, an index of landed import price divided by shipments. lven if free trade lowe s the price of steel,an index of the domestic steel price (represented by the steel consumption will not increase much.BLS price index for steel mill products). It is estimated Other determinants of steel demand are output inthat. if trade restrictions increase the relative import steel-using industries and lagged steel shipments. Out-price by I percent. the dependent variable wkill fall by I put has a powerful effect, indicating that steel is a highl,percent during the first year. This effect is statistically cyclical industry. lagged steel shipments have a nega-significant. its level is almost eight times its slandard ti'e effect. This probahl indicates insentory adjust-error. ments. A high value of last ,ear", shipments, holding

The relative price variable measures only the short- constant this year's desired consuniption. means thatrun effect of prices. To represent the lagged effect of inventories will need to he reduced this year. reducingprices and other variables , we include several lalgged this year's shipments.values of the dependent variable. which allows the ef- Insentory adjustment seems to hase ien pamiallfeet of a persisting price change to cumulate, responsible for the precipitous drop in steel shipments

Table I illustrates the cumulative effect ofa I percent in early 1975. Inventories built by purchasers trying todrop in import price, protect themselves against the shortage in late 1973 and

1974 and the expected 1974 coial strike were liquidatedTable 1 in early 1975 as demand dropped.

Cumulative Effect of a t Percent Fall In the Price The remaining determinant ol steel demand is a sim-

of Imported Steel on the Ratio of Imports to pie time trend ilh a Iegativle effect ofablout 1 percent

Domestic Shipments a year. The trend mai represent the discover of newapplications for competing materials such as plastics.

Percent Change cement, and alumirnum.il the kaill. oh .uhIe 'v/ pri, " ,I d m, i, i Icc' m t I lie ,ii lable to be

First year 1.0 explained is the I S W.hole.alc Price Index for steelSecond year 1.3 mill products Imports k.ould affect the domelstic priceThird year 1.9 by creating slIick in the de.inand for domestic steel.Fourth year 2.3 fhowever. we. did not find that a:.pacity utiliation ,4astLongrun efTct .9 an important d.terminant of price, wkhich suggests that

-4-_

domestic price will not fall sharply to limit import pene- posted prices, the quota variables should be significanttration. This result does not seem to stem from maccu- in the export price equation. If the quota rent is notracies in the price measure. The one major study di- included, the quota variables should le sienificant in-fected toward measuring the extent of discounting stead in the import demand equation. That is. actualfound it unimportant for steel ( 13. pp. 72- 741. Vari- import prices to the United States will be underesti-ables that did seem to affect price were input costs and mated by posted prices during the quota period and thethe price in the previous year. quota variables will pick tip this underestimation. Thus,

The supply price offioreiqn exports. This equation they %kill be significant determinants of the ratio ofexplains an index of the foreign export price. The major imports to domestic tonnage in the import demandissue is whether this export price is affected by U.S. curve.import demand. Ifso, then trade restrictions will tend to The results indicated that the rent attributable to thelower the foreign export price. U.S. quota was included in the posted price during the

U.S. demand can affect the export price either as partof world demand or through some independentmechanism. The empirical analysis suggests that our Table 2demand is important only as part of woild demand, not The Estimated Effect of Import Restrictionsindependently. Entering our demand as a determinart 1969-73of foreign export price did not yield successful results. (I) (2) (3) (4)Since our import demand is only a small part of world Effect of

Tradedemand, the export price does not depend importantly Predicted Restrictionson U.S. import demand. Hence. foreigners are unlikely Predicted without a awith T~rade Trade Percent ofto bear a large fraction of any burden imposed by trade Year Actual Rescteons Restncmons Cotumn 2irestrictions.

Though our import demand does not influence the Imports of Steel Mill Productsforeign export price, world demand does. We estimated (in thousands of net tons)that a 10 percent increase in world capacity utilization 1969 14.034 13.600 16.900 -25raised foreign export prices by almost 20 percent. 1970 11.364 12.000 16.100 -34

1971 18.304 16.8W0 26.000 -54Foreign materials prices and technical change also in- 1972 17.681 17.300 27.500 -58fluenced the foreign export price. 1973 15.15- 14.60k' 21.800 -48The effect of the quota was measured within the Domestic Shipment% of Steel Mill Products

import price equation. The presence of the quota was (in thousands of net tons)measured by dtmmy variables for the quota years 1969 196Q 93.877 98.600 97.100 1.5to 1972. A quota variable was not included for 1973, 1970 90.799 90.770 87.900 3.1since the quota was nonbinding in that year. 1971 87.038 87.800 81.800 6.8

The coefficients on the quota variables are estimates 1972 91,805 95.400 87,700 8.1of the tariff equivalent of the quota. i.e.. the tariff that 1973 111,430 111.000 103.163 6-8would restrict imports the same amount. This tariff Import Price

(index base 1967)equivalent ranges from 19 to 28 percent depending on 1969 1.27 1 33 IL04 21the year; the same order of magnrtude as those reported 1971 1.36 1.38 1.10 20by Floyd(7, p. 13-31. Ilis estimate is based on compari- 1971 1.15 1.15 0.86 26sons of prices of Japanese steel in Canada and the 1972 1.29 1.27 1.03 19United States. 1973 2.09 1.98 1.87 5

In interpreting the estimated tariff equivalent, we Import Sharenote that the dependent variable is a posted worid cx- 1969 0.14 0 13 0.16 -23port price, not necessarily specific to U.S. purchasers. 1970 0,14 o. 13 0.17 -33Foreign producers with rights to sell under the U.S. 1971 0. I 0.17 0.25 -50quota receive a monopoly rent. creating a difference 1972 0.17 0.16 0.25 -5

1973 0.12 0.12 0.18 -49between the world export price and the price to theUnited States. The question is. Which price is mca- 1969Capacity Utilizationsured by posted prices during the quota penod? 1970 0 86 0 86 0.8 3.1

We answered this question empirically. We entered 171 0 78 0.79 0.74 68the quota variables in both the export price relation (as 1972 0.87 0.90 083 8.1discused in this sectionl and in the import-domestic 1973 098 097 0.90 6.8substitution relation. If the quota rent is included in

-5,-

quota period and other purchasers received discounts The effect of trade restrictions was to limit the importflom this. This result is consistent with the fact that share to about the prequota Iescl (sec Figure I). Thequoted prices usually refer to the maximum price where restrictive effects or the tariff and quota increased overdit'erent purchasers pay different prices. Note that it is time; they are estimated to have reduced imports by 25,iiso consistent with the foreign exporters receiving the percent in the first year and by almost 60 percent inquota rent. 1972, the year of the peak effect.

Identities. The behavioral relations discussed above Trade restrictions had a major impact on import price.are supplemented with several identities. One shows as much as 25 percent in some years; yet imports are athe exchange rate conversion between the export price small fraction of U.S. consumption, so a substantialin foreign currency and the export price in dollars. restriction of imports still does not translate into a largeAnother adds transport costs and the tariff to the percentage increase in domestic shipments. In fact, weforeign dollar export price to form the landed import estimate that the largest increase in domestic shipmentsprice, the price used to estimate substitution between due to the tariff and quota was 8 percent. Had thereimported and domestic steel, been no trade restrictions, the main effect of import

penetration would have been to accentuate the cyclicaldownturn in domestic steel shipments in 1970-72: 1970

The Impact of Free Trade would have remained a mediocre year; 1971 would havebecome a very weak year; and 1972 would have been a

The model outlined above was used to estimate how mediocre rather than a good year.free trade would affect the U.S. steel market between Mke effect lt emploYmenr. Still to be determined is1969 and 1973. As shown in Table 2. we estimate that how the output changes described above would affectimports would have been much higher had trade bar- employment, and the welfare of specific groups. Toriers not existed. Import tonnage would have grown to convert output changes to employment changes, weanestimated peak of 27.5 million tons in 1972 compared used the work of Arnold Katz of the University ofwith the 17.3 million tons actually imported. Pittsburgh. Katz's research describes the way in which

.25

.20I sPredicted import shareI with free trade

•.15 ---

Importshare

•10 Predicted import sharewith trade restrictions

.05 -Actual import share

01952 5S 60 65 70 73

Year

Figure 1The Share of Imports in Apparent Consumption - Actual,Predicted with Free Trade, and Predicted with TradeRestrictions

-6-

labor requirements depend oin the age of the capital The desired employment uinder frec trade was thenstock jIlIl. Older plants tend to be more labor intensive adjusted to allow labor hoarding. Using Katz's result.than the industry's, average plant. and hence have we assumned that in the first year the firm retains one-higher variable costs. Thlese are the plint, most likely, half ofthc workers it intends finally to lay off. If. in theaffected by tire trade, hence emplo mieni losses are next yeat. demand does not return to noml.'work-more than proportioinal to output changes, Indleed. crs hoat tled from the previous year are laid off. andKatz has estimated that the older. marginal plants arc one-half of the new employment adjustment is made.about twice a% labor intensive as the inlustrs' average. We used the employment figures, after labor hoarding,so that a permanent t percent dirp in outlput l'eads to a 2 to estimate layoffs wsith and without free trade. In doingpercent drop in permanent cnmptos nient. this. we suibtracted an assumed 1.5 percent natural

The permanent emplo mierit etfect of output changes attr ition.is delayed by labor hoaidiug. which espitins why As can be seen from Table 3. the effect on employ-employment varieso less thaun output ovier the cycle. A ment cumulates because of gradual adjustment inbusiness %ill generall not know whilether an obserived domestic output arid because of labor hoarding. Thedrop in demand is permianent. If' it is judged ito be effect on layoffs (Table 4) was much smaller. Only intemporary, lai ing off usorkerN imai be costlIy; somne will two of thle five years in which the quota existed wouldfind other work. Increasing outputt hack to "nornmal"' free trade have increased estimated layoffs. In 1970 anrequires hit ing nest workers w ith attendant search costs additional 3.5 percent of the industry's labor forceand training costs. Blecause of- these costs. ritarialiernent %toulid have been laid off. In 1971 the esitimated differ-tends to delay sonme lay .offs until it becomes sure that ence is 9.4 percent of the labor force.the fall in demand is permanent. Katz estitmtes that, in Th ce ic tc!n ivelfare of. !piefic groups. Gains andthe steel industry, only 50 percent of any long-term losses, to specific groups include:employment drop is made in the first yecar of decliningdemand. The remaining half is delayed until the second 0 gain to steel purchasersyear. 9 toss to steel manufacturers

We applied Katz's model to translate ouffptrf changes 9 loss to steel workersto employment changesi. First. actual employment was 9 loss of tarlf revenue.assumed to be desired employment in thc presence oftrade restrictions. ifooote, this implicitly ignores labor The gain to consumers is estimated using a standardhoarding in the actual cniplo~rnnt figures). Fxperi- technique, coinsumer's surplus (see Appendix 11).ments with relaxing this assumiption tended ft low.er thecosts of free trade (leaving the gains unchanged). Toderive desired free trade employment, we adjusted ac- Table 4tual employment by tlie estimated percentage effect of Layoff Estimates With and Without Freefree trade on employment, tw ice the percentage output Tradeeffect.

Predicted Predicted Elicit ofLayoffs tLa~offs Free Trade

Table 3 with I rade with Free as a Percent

Employment Estimates With and WithoutYer Rsicon Tad ofEpymt

Free Trade (Layoffs in thousands of workers)

(t) (2) (3) 1969 0 0 0

Effect of 1970 21.5 44 6 3.5Predicted Predicted Trade 1971 40.9 36.1 -. 1

Employment Emnpto~nreni Resiridions 1972 0 41.6 8.4with Trade w~ith tree as a Percenit 1973 a 0 0

Year Resricion Trde o Co-irin 1)*Though estimated layoff mie enerdly higher tinder freeEmptoyment in SIC 331 tradle. there is tine erceplion tree irade taviris in 1971 4re(Adjusted tfir mantroursi slightly tWo. las,41s with ir..1( e Csliction, This resuiltsfrom

thousands of workmr the fott that Lihor hoarding i% lioken into) accoitnt in the free1969 659.8 643.6 2 5trade case. huti not in the .mii( employmennt figures uscd to)

represent the %irti ion it h it Ic r-ii ions t thor hoarding1970 62&.4 58(94 6 2 under free ih 'ryci the eiiei empoloymerint droip o'er1971 578.1 544 it .8f ransyears. makingthe 191t tree Iiraic t.i-,lsnmi,11crthtn with1972 584.4 4946 154 not tradeo reitri,.tsun the delayerd latyiiis uinder free tr1973 628.4 518.8 17.4 show up in 1972

-7-

Trhe loss to steel mantitacturers is of two types. Firstili're i% the cost of labor hoarded. We assume that this Table 5It.hor is not used, so that the loss to industry is the salary,i the hoarded labor. which Undoubtedly oserstates the Gains and Losses Due to Free Trade,-t to industry since the extra labor probably has some Gains to Steelpr'oductiv ity. Purchasers

Ihere is also a further loss of profits. This is the area Year (millions)

,ise a conpetitie supply curve, or marginal cost 1969 $5339,uie, usually termed producer, surplus. We did not 1970 610,,herve an upward sloping supply curve for steel in our 1971 1048empirical work; instead we found that the price is unre- 1972 903

sponsive to market condition,. However. Kalzts work 1973 267

on costs for equipment of dilferent ages suggests a Discounted sum 2839

,hape lor the marginal cost curve, which can be used to Losses to Workersestimate profit losses. (millions)

losses to labor were estimated by comparing thelong-term earnings of displaced steel %sorkers with the 1970 170earnings of similar workers %& ho remain. Jacobson 1 9 1 1971 -36eslimates that the lifetime earnings losses per steel 1972 329worker laid off are $6.300 in 1960 dollars. To get total 1973 0costs, we adjusted this figure to 1969 dollars using the Discounted Sum 379wholesale price index (as used thioughotit the paper)and multiplied it by the number of layoffs induced by Losses to Indusiry• from Labor Hfoarding

free trade. Losses of tariff revenue are estlimated as (millions)import values (at the foreign port) multiplied by thepercentage tariff. The estimated gains and losses to 1969 S 43

1970 300specific groups are ,hown in Table 5. 1971 578

The discounted sums shown in Table 5 are present 1972 73values in 1969 dollars. Gains and losses for individual 1973 0years are first deflated to 1969 dollars using the Discounted Sum 864wholesale price index. then discounled back. to 1969,using a real interest rate of 5 percent. Losses to IndustryProducer's Surplus

As can be seen froni the table, gains to steel consum- (millions)crs are sizeable. about $2.8 billion. Much of this is atransfer from other groups. The largest losses are not to 1970 38labor, bitt to shareholders in the steel industry. They 1971 141lose about $1. 1 billion in profits. which includes about 1972 95$0.8 billion due to labor hoarding. Losses to taxpayers 1973 30in tariff revenue are another $0.7 billion, and losses to Discounted Sum 249labor are a soniew'hat more modest $0.4 billion.

Adding the gains and losses yields a net gain of about Losses of ariffRevenue

$0.02 billion. If redistribution is thought to have some (millions)cost. this net gain is not costless. The net gain necessi-tales a transler of about $2.2 billion, or over three times 1969 $141

1970 151the net effect. Part of this transfer is from taxpayers. not 1971 195a concentrated group. The transfer froit labor and in- 1972 205dustry is about $1.5 billion, more than twice the net 1973 189.effect. Discnunted Sum 726

Though the adjustment for free trade requires losses

that are large in relation It) the net gains, the losses are Total Gln% MinusTota Lossesnot sudden. nor are they large in relation to industry (millions)revenue or payroll.

In the worst year for layoffs, the expected lifetime $621

earnings loss was less than 5 percent of payroll The

-8-:

largest single year loss to industry was modest in rela- Appendix Alion to re enue, about 3 percent: however, it wouldhave exhausted net income that sear. Equations Used to Describe the Steel Market

Some of the losses dcscribed abose do nlot u',u(lly The results in the text are based on an econometricappear in calculaitions of the net gains from trade. The model. The model has four behavioral equations (dis-loss to industry in the tornn of pa!,mnt, n or hoarded cu',,cd below) and a number of identities.labor is usually not included, on the xssumption that the Bau ,4round o the demand curves. There are twoindustry moves immediately to its new equilibrium denand equations, one representing the substitutionlevel of employment. One argument for this is that labor beteen imported and domestic steel, the other repre-hoarding is a purely cyvcdi phenomenon. On the other senting the demand for steel as a whole. These eqia-hand, it seems unlikely that the ecfects of permanent lions are derived from a CES production function foroutput changes can be ditinguished ea%ily from cycli- steel-lsing commodities y. The production function iscal changes. Hence. the uncertainty and reaction to it assumed tobe ofthe following nested type, as describedmight be similar. in [4].

Also omitted from the usual gains and losses calcula-tion is the loss of industry profits and labor earnings.The rationale is that a competitive sulpply curve repre- V A b s- + b'x- I

sents the opportunity costs for both labor and capital. 0LjiHence, any losses they would suffer are completely S 1 {Cq - 0 + 2included in the steel price drop they are willing to ac-cept. But. it turns out. this lovs is exactly offset by a gain The first production function is homogeneous andto consumers of the domestic product. The gains to depends on an index of steel input x, constructed as aconsumers of the domestic product and the losses to CLS function of m (import tonnagel and q tdomesticcapital and labor cancel, leaving a net gain only to shipments to domestic purchasers). The coefficient aconsumers of the imported product. measures subsliltolmbiJily between steel aid other in-

We do not assume this cancellation for the steel in- puts, the elasticity of ,,ubstitution being hi (I + )-'.dustry because neither the steel price nor wage seems Similarly, the coellicient 13 represents substitutabil-sensitive to short-run demand conditions. Yet, the exis- itv between imported and domestic steel.tence of marginal plants alongside much more modern Profit maximi/ation by firms, taking as given theones suggests that the idnstry has an upward sloping price of output. results in the follow ing factor demandmarginal cost curve, as discussed earlier, equations:

We limited our study to specific years and to gainsand losses to specific groups. A word should he said.however, about the welfare effects that lie outside this on = (c 2 pq ) offocus. First, if import penetration causes an adjustment q C Pm/in the exchange rate, or a downward adjustment in the (I)general domestic price level, our exports will become In ( = oln + olnIpL]more competitive. If the U nited States hasn monopoly q /PI/power in export markets, import trade restrictions willhelp us exploit this and shift the terms of trade in our A)favor. Removing import restictions prevents us from s k i or

exercisingour monopoily power as effectively. We have (2)not taken account of this I Pi

On the other hand. the gains to import consumers will In(s) Ink + bin kw' ~ln(i),extend indefinitely into the future. The adtustmentcosts to industry and labor, all the cots htt the lost

wheretariff revenue. will (die oult a s adtistment becomes com-plete. Hence, limiting the study to the first few years of p., = price of impoited steelfree trade could lead to massive undercstinates of the pot price of domestic steelnet gains. py pnce of output

- . . . . . . . i . .. . . . i . . .. • . . ... • .. . .. ... .. . .. ,

A =constant "~\I\0 1)+ 0.454 i 0,265 lit -

=- (I-+ 1 (b.04)(*.

III composite price for steel. which can be Lie-fined by the ielation J1ln p, i,J lid 11, t 4 0.113 lit -, + 0.037 11.(I -m)d Ini',. whhere % , is the import value (siiS) (7.61)share

Y =output of steel-Uising toni modi tie. R2 (o2 infredd )09

N= 18For emtimat ion. both If Ha.nd (2) are assu md t(1hold in d f. 1 I2

%.tite-; of left-andi ii-h atd sit ibles ito .nIlowt for p 0 S42.gradual adustment . [hewe shOil111 t0i11 lnfrsppro~ach

(I) and (2) as adijustittent becomesc complete. whereBefore proceedinrg (t cstmmfiate% or mdi iduail eqt,

irons, it is necessary ito discuss gernci .i econometi. 111 tonnage imports of steel mill productsissue. The model 11nchides MI alota many pi edeler- q domcstic tonnage shipments of steel millmined variable% as% obsersationN. Hence, two-stage products to domestic usersleast-squares is virtually the same ats ordinars least- pm index (base 1967) of the dollar price of im-squares. Obiu leiiie t oueasubset of the pots including tariff and transport. ht is eon-predetermined variables or a set oft pintcipal infuo- structed from export quotations for the comn-nents. However, the iirst seem% arbitimr a ind involves mon mar ket and Japan. These export quota-throwing away infor mation. .1 hic klter is subject ito the tions were used to form a fixed vweightscale problem. Moreover, the endogenOUN right-hand measure of the dollar export price. To thisvariables turn out tot be not s'ers en-dogenouis. H ence. wve added a nmeasure of tariff and transportthe low variance of feast-squares may vwell outwecigh its costs. The former was based on a fixedasymptotic bias, For these reasons, ordinary least- weight average of U.S. steel tariffs. The Lit-squares is used for estimation. To correct for firs- ter was based on an index of tramp steamercrder serial correlation. generaliced least-squarecs is rates published in 15). The dollar importemployed, price. tuis formed, was then expressed its

The sibs uihiutinn C iaion.'m. The sublst ituion equation an indlex, base 1967- In estimation, the index(equation (3)) isestintiated tising a Joigenson tatiiinal lag is multiplied by 08H4. since in 1967 the median(hence the lagged vlfes of tit ) land an added s .riable import price was about l6 percent below1110o measure hedging bcfoic at possible %ifike. lfiie the domestic 115. pp. 409 - 101each major negotiatioin. steel useris bu; duse ntoiic s. jy domecstic price as measured bv the BLSdomestic shipments become Loige. domestic capiacit w% holcsale price index for steel null productsoften proves inadequate, and users often wiin fil tu -~ measuic of temporary hedge demand.poris. After the agreement is rearched, domestic ship-ments fall olf suddetnly. A ncisute of the helge's fiten- This regression suggests that imported and domesticsily is the ratio of ma xitut nritonthly shipments diii itig steel,. lie liii 1 sust i to table - esven in the shot I run. Thethe year to average monthly shiupmcntsI iir sat iabif.I II lastit i of slibstitituion after one year is about unity.is this ratio multiplied by the cst lrntellc lentgth it the I he Ioi t - List icity tit siibstit uton is Much higher.hedge. The length tucasure is proideitiit the l%8 ASti-el about ft. sinc onh 17 percent t the adjustment takeshonpori stud) ( I5. pp. 149 .S(q w4ith uci ow4n eXtensions place in tife first year.for more recent year,,. lite drltrid /,o, %"""' ("S " mu /iii he demand for

steel as a Alfmilc expiesses the steel input indes vi as aEquation (3): Substitution Equation - Annual Data function of the steel piuce index p,. the price of output1956-73 (1 values in parentheses) p,. .ind output in steel-using comnmoties v. The sari-

ables . and p, are deived uising the assumted ('ESproduct ion fuinction f or steel-using comimodities.

I litm 14 1))) In ,, The comuposite price indes is formed from the relation

q (-3.0 1 7); dfl If r, 'v d lt1,, 1 t,) lit 1,

where i,,, is the import value share. The resuiting inca- he .wipp/i pritc " djom.slh" steel. The estimatedsure of p, is expressed as an indc., hase 1967. The domesic supply curve is presented below, a% equationvariable s is formed by the relation (S). A .cpcit utilization teinm was tried. but had little

effect on the piedictisc power of the equti.Uon. To rep-i'~ + I/ resent a maintained hypothesis of complete pass-

through of cost in the long run, the coefficients on theEstimates of +,. ,,2 .nd /j are taken from the ,teady- lagged prc:e and on the cost variable were constrainedstate form of equation ( I t.to ,u to unity. Unconstrained results were similar.

The aggiegation formulas forp, and % are both basedon long-run eqiilihriibn conditions. Hence. they arejustificd only asymptotically. as the system approacheslong-run equilibrium. Equation 5: The Supply Price of Domestic Steel -

The estimates are Oiased on a longer seties than used OLS - Annual Data 1948-73 (t values in

earlier to provide needed saiation in the price ratio. parentheses)

Because of the maintained li pothesis of constant re-turns to scale. the coefficients on output in steel-usingindus. ;es and the co.fficient on lagged steel shipments In W) = 0.0053 + 0.374 Inlw) + 0.626 In JPq)-.were constrained to sum to unity. (0738) (13.0)

Equation 4: The Demand for Steel as a R' (on tiatisfrted data) 0.99

Whole-Annual Data 1948-73 (t values In N = 26

parentheses) d.f 24D-W= 1.61

InI (s) 14.44 - 0.446 hi " + .2)4 I pt p 0.440.

(14.44) (-3.63) \ (14.50)

The supply price of imports. The supply price of-0.294 In Is 1) - 0.018t. imports. like the domestic supply price, was originally(-6.0U) expressed as a function of market conditions and costs.

In addition, a measure of technical change, the percen-R

2 (on iranstorired data) 0.)5.5 tage of foreign production using the advanced basicN = 26 oxygen process, turned out to be important. To repre-

d.f. = 22 sent the effect of the quota, separate dummy variablesD-W i.9)2 were entered for the years 1969- 72. A 1973 dummy wasp = 0.045, insignificant, indicating that the import quota was not

binding in the tight 1973 market.

where The cost measure employed was a weighted averageof prices for individual steel-making materials in Ger-

3 = composite steel index many and Japan. A steel wage index did not add sig-p, = composite price of steel nificantly It) the explanatory power of the equation,

py = price of steel using output. measured here by suggesting that foreign steel is priced according to mar-the overall BLS wholesale price index ginal cost and that there is a disinclination to vary

y =index of output in steel-using commodities, employment.This is constructed as a weighted average of Market conditions were represented in two alterna-Federal Reserve Board indices of production tive ways. by the level of imports to the United States.in the auto. busincss; equipment, and inter- and by world capacity utiliration. The former wasmediate construction materials industries. dropped because of low explanatory power. This meansWeights are based on relative steel ship- that the United States does not seem to possess impor-

ments to these sources tant monopsony power in the market for imported steel.= time trend. The regression results are presented in equation (6).

-11-

Equation 6: The Supply Price of Exported Foreign the area Linder the demand curse. rhi% particulair mriSteel - Annual Data 1956-73 (f values in sure is often replace:Ld tl one hasecd orii iniconme rtoii-

parentheses) pensated derntirld i r C. Ihc or rjrirma un neipersatedmeiasure is used if) i ll,' stud6 l01 secial rer1slls.

!:irst. the incior comeritd neasuic eiiiiires in-Irij) 0. 135 + 1 .84 Ini)litt-) + 1.-10 lii tw') format ion not as oailile troim otir moi del. name I the

(0.409) (3.63) (2.1 )) demand foi cicti 1inirl prrdict using steel Secoind. part

-1.09 bf + 0.089 t + 0.? 13 d,,, of the gain front a I ,rll in steel prics is toi piurducers til"(3.0s) steel-usinggoods. it is nit clear what %PC 0fconrPensA-It .2) (.0~)tion. if arny. is aippropriate loti these rent-, 'Third, the

+ 0.205 1,, + 0.276 d,, + 0.19i4 172 . cianpeflsatei arid 1iricorriperisaicd mneasures are likely(2.67) (2.46) (1.52) ito be close to irne antiher w hen file level of expenditure

on the good is small or w hen Tirice changes are small.R2 (on transforme1d data) 008Imported steel. w here at significant price change is pre-N =I1 dicted. is less than one-fifth urf I percent of GiNP.d.f. =9

D-W 1.66 The Traditional Analysis of Gains and Losses Duep 0.804, to Free Trade

The traditiOn13l arral~sis of consumner gain% and lossesis from ('or dcii Ij Ilii nal vsis assut nies thaitimpor ted

where and domestic goods are pei fctI Sui'littIttS tartassunip-

Pf index of the foreign export price tior that seem-- roughrly consistent with our results forin foreign currency units deflatted hy the long rtio but not tor tht: short run).

thefojlig wolealeprce nde Ilie Cordelri met hod treats the dhemanrd for imports asthe = orcaig wholsaepicende thre excess demiand fOr the commodity. The supply

wf =index of deflated Japanese and German curve for thre doiresc product is used to measure the

materials prices iargurral cost rf dom11estic production. F-or i good de-bf =1fraction of European and Japanese scription. see 1121.

steel made using thre basic oxygen The Modified Version of the Corden Analysisprocess For use in describing the steel miarket. we modified the

I time trend Corden arral sis ito take into accorunt the following fea-d9d2 dummy variables indicating separate tores of the stfeel market:

years in the period 1969 to 1972.

* Impor-ted and domiestic steel serm to be imperfect sui-

The positive coefficients on thre dummies indicate ltes. espctall il (lire short rumi.

that the quota had at sharp eftect on supply price in these * '1 he price nfdonesirc steel droes not seem it)ofall "hen de-inand is slack, th!s tikes place despite an rmi aard-slop-

years. Its tariff equivalent seems ito be about 201 percent. rug roarginrI oisi CUrse basd oim ileteni vintages uifThis helps explain the recent propensity to use qriritas, c.rpritaL Hence, the stippl rird margrral cost curses seemnsince a tariff of this size would certainly he difficult to to differ.obtain. 0 Ilecjirse ot rinceritrrits ahrri the permainence if any' de-

niand cha~nge . steel piridre ig fir iri do nor ;rdjusi theirwoirk forcee crirnplcielv to a rir iii demind. There is acost to themn 0 hoarding the laboir %%hen adjusting to tree

Appendix B trade.

Techniques Used to Estimate Gains and Losses to The modified analysis treats imports and domesticParticular Groups production ats imnperfr't substituteS. Hence, two dia-This appendix discusses the conversion of output. price grams are needed, one for imtpoirts and one for domesticand employment changes intor measures1T of gains and production.losses. '[he traditional method Of CAucuI~tlatrg the net ligure it-1 represents the imiport market. When again or loss makes use rifcirrsunrer suiplus. wich has tariff of I and at tonnage qliotit of tit is in effect. tariffalways been the source of cruntini ers amnrig ecurnu- reivenuesa.re cqrtil iiiarCa Aarid the quoit rent rcec dmists. The original defiiion of crnsunrer surpilus was b% forieign Prouret s sAi. r herr import restrictiuons are

-12-

Importprice

I X1

I T X 0I

I Ir

m ft 0 Imports

Figure B- I

The Gains to Trade in the Import Market

removed, imports rise from tot, to in,,. and import price not lower the domestic steel price. D~omestic outputfalls from P2 top,,. The gain it) import users is area X + falls from t/2 to (I,.

X1 + XII. Of this, there is it loss of area X in tariff Since the domnestic steel price does not fall, consum-revenue. ci sof domestic steel ito not icceisea gain. Oni the (other

The domestic steel niarket is described in Figure B1-2. hand. there is sonic liss ito producers. The loss of long-The demand for domestic steel is represented hy 1), run profits, Is shoikn by the area above the long-runwhen trade restrictions ire tit eclkct. Domnestic price is mnarginal cost cursve. areais Xiii and XIV. Itt addition.

p 3, which we assu me is jist e .q al to full ipa tyt long- there is at fiurt her short -ii loss in proti ts dire to the costrun margina cost l.RAI(I tliis, asiiptiont is cklisa- of ioiided labor. 'r'his is represented by area Nt' 4

lent to defining capaitya ;is including all %inticeN of X VI.equipment t hit woould he pi (ilt table to use at hi, ptrice. 1 [here is alIso another cost, not shoss n . This is the lossFor expositional conivenience. %%e also asslirte that, in of eairning.s tol displaced %orkers. It should be em-the presence of trade restr ictions, domest ic production phasiied that these adjutisnent costs, losses of profitsis at frill capacityv. 'I his as iniiipt ion is re lax:scd in actual and earinirgs. are a iiet loss only because the doimesticcalculations,. price does not fall. If the price always eqirniled the

Short-run marginal cost is represen ted as SA/Cl(. The nmar ginal iopportuniity 'oest oft laibor a1nd capital oils itdifference between this itid long-rm marginal ciist is would liler clmnpetitise issrrtnptiions). then thethe cost of temporarily boarding labor As bile assessing domtestic price writld fill ss leu demand wasslack. Anythe permanence of a drop tin demand. losses to capital and labor wotild be irftset h as corres-

We now turn ito the welfare eflects of free trade on the puinding gain to constnners oif domestic steel.domestic steel market. When ntiile restrictions are re-moved and the import price f.lls, the demiand for Alqebraic Formulae for Estimating Gains anddomestic steel shifts inwurd. lccarise of the laIck of Losseshistorical responsiveness of the domestic pri:e ito mar- Ibhe areas that represent gains and losses fiom feket conditions, we assume that import penetration does trade need to be expressed in algebraic form t0 allow

-1 3-

Domesticprice SRMC

// LRMC

P3

XV

I DI

I Do

go q 2

Figure B-2The Effect of Free Trade on the Domestic Market

numerical estimates of gains and losses to particular The loss of tariff revenue that would accompany freegroups. trade is estimated as the product of the value of imports

Gains to consumers, shown as areas X + XI + XII in in the foreign port multiplied by the percentage equiva-Figure B-I. are approximately equal to the following: lent of the tariff.

Ihe loss of industry profits is based on the long-run_u g marginal cost. This is shown as areas XII + XIV in

consumer gain = 2P Ll 2J Figure B-2. This area is approximately equal to

= %, p . - RM. profit loss = ,.RMC[Aql2

where

Ap = absolute value of the change in the landed whereA LRMC is the drop in marginal cot from lowerimport price due to free trade output due to free trade andAq is the drop in domestic

Am = change in the import quantity shipments.%Ap = change in the landed price of imports as a In practice, a modified version of this fomula was

percentage of the price with trade restric- used to reflect the fact that the industry would not havelions been operating at full capacity, even when tiade restric-

%m = increase in import quantity as a percentage tions are in eflect. Hence, part of the profit loss, mea-of the import quantity with trade restric- sured from full capacity, would occur in any case. Thelions new area representing profit loss is shown as the shaded

RM = import revenue including tariff and trans- area in Fsigure B-3.port. This is estimated from publihed In |igure B-3. q, is full capacity shipments, q isimport revenue in the foreign port adjuted ,hipments with trade restrictions, and q. shipmentsby our estimates of tariff and transport. with free trade. The shaded area can be evaluated as

-14-

Price

LRMC

qO q2 q, Quantity

Figure B-3

Losses of Profits for Domestic Firms

loss to producers = LRMC • Aq, We have used the assumption that full capacity long-

where iLRMC is marginal cost evaluated at full capa- run marginal cost is equal to the prevailing price. This

city minus marginal cost evaluated a[ the average of allows us to substitute price for marginal cost and then

output with and without free trade and Aq is the abso- substitute revenue for price time% shipments.

lutevale o th chngein hipent du tofre trde. Deviations in shipments from capacity are calculatedInuerenotage change form hentssd to proers isde by converting rated capacity. measured in ton% of pro-

In ercntae cang fom. he ossto rodcer isduction, to tons of shipments. Since production seldomevaluated as goes above 97 percent of rated capacity. even when

loss to producers = %ELRMC %Aq .LRMC, • q2 demand is heavy. we took 97 percent of rated capacity

= %ALRMC %Aq •R, as actual capacity.

where The calculation of the change in marginal cost wasbased on Kat's research on vintage capital I II1. He

% LRMC = ZtLRMC as a percentage of full capa- found that the elasticity of labor requirements with

city marginal cost respect to output was approximately 2. We assumed%Aq =(absolute value of the) reduction in that other input,, (which we term materials) fo~r marginal

domestic shipments expressed as it plants varied proportionally with output and that bothpercentage of shipments with trade labor and materials had fixed price%. The elasticity ofrestrictions industry cost with respect 14o output is a weighted aver-

q2 = domestic shipments in the presence age of the elasticity for individual inputs with %%eig~ht%of trade restrictions equal to (industry average) factor share%. Usitng a labor

LRMCc. = long-run marginal cost evaluated at share of 0.38, an elasticity o1 labor costs of 2. and anfull capacity elasticity of othet costs of 1, the elastlcity of total cos

R - value of domestic shipments in the with respect to output is 1.38. ('The labor share for thepresence of trade restrictions. year 1971 is taken front 13. p. 131). If this elasticity is

-15 - I

... . .. ... . . . .. IIII . .. . . .. ... II ~ ~I .. . ..I II II . . . . . l . . . . . . . . . . . . . .

tken as roughly constant when output contracts, the earnings per laid-off worker to obtain total earningsclisticity oftmarginal cost with respect to output is 0.38. losses.Hence. the percentage change in marginal cost is In this appendix we have described the calculation of

%ALR3C = .38.gains and losses to specific groups. The losses are calcu-%!LRAIC -7,11 0.38, ated for each group and foreach year between 1%9and

,Ahere %3q is shipments at full capacity mninus the aver- 1973. These annual figures are then converted to pre-age of shipments with and ssithout trade restrictions. sent values as described in the text.The difference is expressed as a percentage of ship.ments at full capacity. Literature Cited

The preceding discussion describes a loss of rent tothe industry based on the long-run vintage marginal cost I . Adams, Walter, and Joel tDirlam. "Steel imports andcurve. In addition, there is a loss due to labor hoarding. Vertical Oligopoly Power." American Economic Review.

We estimate the quantity of labor hoarded in several 2. _ . **Biemh g 9tee). 626-55 an Innovation." Tiresteps. The first is to derive desired employment. We Quarterly Journal ot Econonum3 s, o (May 1966). 167-89.take observed annual enmployment in 1969-73 as a 3. Annuial Statistic (il Report. Arnermran Iron and Steel mInsi-measure of desired employment in the presence of trade (te. 1973. Washington. D.C , 1974.restrictions. To derive desired employment without 4. Armington. Paul S. -A Theory of Demand for Products

Distinguished by Place of Produciion.~- IMFStafPaperstrade restrictions, we adjust these employment figures (March 19691. 159-76.by the percentage effect of free trade on employment. 5. Chartering Annual. New York: Maritime Research. Inc.We use Katz's estimate that the percentage effect on Various issues.desired employment is twice the output chanige. 6. Corden. W. M. "'The Calculation of the Cost of Protec-

Once the free trade employment figures are derived.,~ 7ton - Economic Ret ord. 31 (April 1957) 29-5t..Floyd. John E. "The Iffects of United States Tradethey are used to estimate free trade labor hoarding. We Restrictions on F Lonomic 1:Efficiency)" Manuscript pre-

assume that, if employment drops between year 0 and pared for the D~epartment of Labor OLABI. 1974.year 1, one-half the desired employment reduction is 8I. ____. -I ic Oservaluation of the Dollar: ANote onmade, the rest being hoarded. If demand drops farther the International P'rice Mechanism." Americanin year 2, the year I hoardees are laid off and one-half of 9 Economn Ret re%. 55 (March 1965). 95-106.

9. Jacobson. I out%. Fnrniner Loss of' Workers Displacedthe remaining employment decline is made, the renmain- ftoom the Steel Industry h% Impoirts iif Steel (PRI 197-75).der being hoarded. (If desired employment had risen Arlington. Va.: Public Reseaich Institute. 1975.sufficiently in year 2, none of the year I hoardees would 10 lighli5'htt of U'.. 1.tpirt and Imtport Trade. Augusthave been laid off.) 1974, U.S. D~epartment of C'ommerce (F! 990).

These assumptions are used to estimate hoarding in I I Katz. Arnold 'EFstimates of Labor Displacement: Athe free trade situation. 'The cost of hoarding to the Structural Appoai~ch --Retnoti nt, Re.itrictionse on Int-

prnti of Sgel. cdited by Jaitnes Jondrow ct at..- pp. 4-1 toindustry is the number of workers hoarded multiplied 4- 0 PHI 7' 2 Arlington Va : Public Research Institute.by employment costs per worker, the latter from 1 3 1. 12. Magree. Stephen ''tIhe Welfare Effects of Restrictionson

The remaining loss from free trade is the earnings loss U S tfrade.' lBriiAtne. Papersion Lionomic Achivity.to workers. The calculation of earnings losses begins 1.No. 3 (1972), 645-70t.

1.Stigler. Gecorge, and James Kindahl. The- Behavior iifwith the estimation of layoff%, with and without free lnifioirial 1rii ci New York City: National Bureau oftrade. Layoffs are the change in employment (after Economic Reseatrch. 1970.accounting for labor hoarding) minus the amount [hat 14 Thorn. Richard S' "Stee[lImports. Lab~or Productivitycan be handled by attrition. We assunie that attrition is and Cost ( ompctitivenecss.- W'estern Et ono~mic Journalequal to 1.5 percent of the previous years% labor force. lflccember 196X). 175-84.

I5 U S. Senate. Conmnittee on Finance. Staff Study. 90thFrom layoffs with and without free trade, we estimate Congress. 2d session. Steel Imnports - 1968.the change in layoffs due to free trade. This change is 16 "U S. Steel Cuts P'rices to tight Import Boom. Dismay-multiplied by Jacobson's estimate of loss in lifetime ing~ther Mills.'' Wall Street Journal. May 19,.1968, p. I.

C uiie tsteel, bilt fiign siihsidics anid trad estitoshvprevented if frotm being exploited itt world trade. tIn

Edmud Aoubsuch ai situation it is tiot at all clviir that "'free tiade'' inEdmundee Ayoub ti l tj Sw naticonal iteice t. let alone that oft he

indu sttrvJortdit identifies his time-frante (1969-73) ats the

period bets'. ect, imtport penictritiirn (the 19600' (and theperiod Of **teCLiLCed threat tf om imptris" (after 1973).Perhaps he is right f r the former period. but it is notclear thait he is right foi the litter period. F-or one thing.the decline of inmport penetration since 1973 resulted

I find much to agree and ito disagree with in Mr. largelv front an "institittioril- factor in collectiveJondrow's study. His time-fraitel%) -73)allows hint hargiiing -the 1 sperinrental Negotiating Agreementa little complacency about ,feel inipoi t penrtlrationi andl (FN At. -tic F NA prevented an industry-wide strikeour ability to adjust ii) its -graidiihoi I re he faicisitso Cr ecoincomic iSstis, thus% comnir ca ling Io) steel-show a decline in import penet rai nf nt i l9i 1071 t L'3 it iiing itdu 51rie% that stokkpiling was rinneces-However, the peauk penetration tie in 19)71 '.5 as itei saurv. I N A etleei els alteted the buyving behaviorquadruple what it was in 19N), tUnder somite cir- of clnititning industries. Withoiit ENA. we wouldeumastances one might consider iniport penetration surely hasve had ittother surge if imports accompanyinggrowth from 4.7 percenti t) 17.9 percent in eleven year% the 1974 steel ne ,iloliations. It ow ever, Imports rose onlyas reasonably easily adiusted ito But tile ads cisclot) nioicet arely. teflectirig dittestic short ages of part icuitlarimpact of steel inipoi t pettertatiott compou tied the priod uctrs and so me a ppreie n sum osver a possible coalproblem of the secular decliire in emtploynment causecd strike in late 1974, -1hus, by liiiing, the period of studyby such factors as technological deselopnientS and thle tire atiiior isdenied hiinsc If a -01chient length of timeU.S. industry's significanl% declining share of the tiir art approttriat perspectis on, the import prohlemworld market. All told, the job I mpact 11.1kes on a mag - :nd he has utinilted an il 'ciii t a iiisal fauctor in (flienitude that justifies our concern about the steel Iitport questrott oif inmport penetation that a broader time-problem. frame would has e provided

Though neither appears it, be on thie horiuin. two Thvre are two cycles in steel that affect output anddevelopments could trultigale ourcitncern: the achics - crmpltty tent -the "eciionic cycle* and thement and mairlenince of it fulil emplioyment ciinorrv. rieginttring c~clc -iTy ii )ci lajp. but their separatteand a strengtherting of "fair" trade in "odtrd steel efltects 1% t l'oje itCfti ified . pat tiCtil.11 Il% with respect it) the

The author of tIs paper seeks. to ic le mine %%hait t he in itt hi t es iiicbIheittt.t Ipiseotle egltiieffect of free trade xsoild ha% c beert dutring the 1069c)- 71igc c spttcia vprritrtliitprs inttvperiod, had sic not hid the restratints iftt;inffs anit the ickiucicilvdges tle ellect of citiact negotiaitiions kiilVoluntary Resti ait Agreemtent t%'.N H lriclN . I he ittpiti rpertinititiiliwec .i silteae nsetllirve that the periiid mif the ins cstigattitn is toi~ n o"rt to wallk as laststrti ltr i the caulsal relatiornshiip.allow sufficient perspective on the problem tit steel lhre diati sui:'ct fihe ivise Ihc hte k shtcrhit iriportimports. This results in a (itlt t ion (it the impact iif free penctietir itt pikcit in ne ct ailtek evar Anrd the peakstrade on the U.S. steel industry - were higier mr cii Ii sticcessis e riilting sear oft thle

The structure of worrld steel trade Itas, changed in the 19N)% anid in 11)71 -the last negirtiatirtg year precedingpost -World War 11 piod. In most coiniries . steel %i as F NA 'I he sic i ge itf steel hitis inrg . hirad ccas frioiSi pro-an industry that produced mimIl I ir domitt st cin- nirikied itt hie cwii liti n inre mni rrhs preceding contractsumption. 'rhe postwar rebuiliding of new. elficient in- termntia~tion. I tis, %is isll.ciei lN a sharp drop in (lit-

dustries abroad %oirn resuilled in file surfeit oif dotiestic putllimiil ertpli itiit Aiter the k ottiicr terrmintion dlatesupply in other colintrres. part of'%v hich wus expiitled it) as irisentiiics %%ere wcnr ked cul lDuting the interimthe U.S. market. I hie U.5 Share of the world miarket sears het cc cn nep'ctuiatiotts. steel irnipitr %%ere sis-continued to decline long after the rehihilitation peiit il rainedl it relutickl hih lesels Ieciuse iif ciinsiringabroad, when decline was expected and inevitable Re- iridiitucls' I.iitnlitv with iinpiir ptiirchaing. anrdstrictive practices of U S. trading partnters, t hit i're long ret cirtt i on t It eqire mc itis A it h liirellen siipplic rsestablished ring this% rebriliting period hive butt oin. toi piotct steel huusers tritir shoirt supplies dicting theand now constitute an uinfair advantage in ni de to the nest ptesettlencmie cli' I Iris cs clicu pattern did nolt

U.S. indtustry. In tother wuircls. the U~nited States tiiht oiccut itt the 1974 ricit'.'tuirrns nuder I-NA.well have a natural co0mparative advantage in produding I tit) oit mrean Itrsrugpest thiat inrpirtdcrnestrc pnce

-17-

differential is not a factor; but it is not at stifficierrtly suiccessful. %%c catn espect at rise in 1:uropcan imports of

,,.ripfete explanation. It may tiggzest th1A steel Import Steel it) the I. irlite SIMCSe. prolxI.1hs by Means if dump-

poichasing. under certain conditions. is not ats pi ice ing.I

senlsitive as one might think it piori. Steel-consumning The rising level and propoi tin of imports into the

,ndustries that chose to stockpile in the eai y stage% tif United St~ic it uel the past de-cade and a half have cost

the negotiating c~cle prior ito the 1974 iiegititioiis \%miild file 'CCl IindustrN dcaii IN in )ohs. A methodology based

p iobaly have purchased imports even vs iihou t an a p- upon tinceit dii .sisstipt ion aiid questionable data

pt piae price differential. That is at h} pishest thaiit wkhich leids Ito olic lii ons abo it oh loss and the ability

w,'sild warrnnt investig-ation. by studying price ditlceei- to absorb sich losses isosie I find difficult to accept. We

gsls during the high import peiiods preceding 1965, has e lost tens ofthous~ands of steel jobs to imports. The

v~'if. and 1971 steel negotiations, loss (itf income to rndis duial Ceiiiployees is great. In ant

I wish to take exception to another one of the author's arrid.e published in this vsoiime, Louis Jacobson evii-

.i'surnptions. The Voluntary Restraint Agreemelit mies in earnings loss, for displaced steelworkers oVer a

A RAl is assumed by Jondros to has e rest aied trakde six-veir periid of about 24 pcitccnt. with ani average

,hiring most of the period 199 73. IVRA %kit% in effe.f loss for the fiist two years of over 46 percent. Manyror the period ji96 9 -74.) 1 think that the reverse V. is cities ti this country are largely dependetnt upon steel

true. Without the "restraint'' of the quota, less 'steel for ernplsi~inent. TIherefore, etre communities are ad-probably would have entered our market. VRA pro- versely uffecteul wihen %teel joibs ate lost. 1I he tax base

slded for an arnual quota increase of 5 percent. w.hich 1 deterioiates. retail estiiblishiieits suffer. .and the entire

douible the average long-run late oi increase ii treel infrasrit ictsire of the commnirnty is threatened. Adjust-dleimand in the United States. l-oreign steel pioduicers nierits toloh losses of such pervasive influence take asought to garner their quota shares despite shortages of long r inic and are ver y costly.sreel in their own couintries during part of this pet iid. 'Ihe aiuthor rakes a ease for trade liberalization. The

Ihe long-run benefits of cstaiblislring a larger base in the effort ti these comments is not to argue against tradeworlds% largest market mte obviokis. Without the en libeiadalaion It is lo suggest that trade must be a two-.riiragernn of the VRA, the foreign pioducers might wa5 street to be effective, and it must assure reasonable

%tell have supplied their own domestic needs mte stibji it it Id maikts.l 'I hat foreign imports did$ notfuilly, and the U.S. market rniight have been spared the "take over" the 1U.S. market during the industty'salleged benefits of V'RA "protection." Voluntairy c "long peiiiid of vulnerabilitN - is no assurance that%traint agreements can be useful on at temporary basis growiiig inpirt pe.netratiurn could not resume and dofor circumventing market disruption: but without a further ufaniage tio U.S. markets and jobs. The least wefniilf-in penetration factor. should be able to) expect is a saifeguard against market

Is the U.S. industry's period of vulnerability to rrn'- disruption, and some correspoinding effort by foreignport penetration over? Jondrow scems to sklggest t1hur it pioducers to eliminate trade restrictions in thseir own

is. The time-frame forthis study coiincided with whit he markets.believed to be the peak period of import penctration, at We are faced with a prirl lem of growing nationalleast for several years ahead. Alte~rnaitively I base :onccin. ft is inadvisable to allow excessive deterrora'suggested that the ENA played atdecisive role in liiriting tion of ouir steel base tin this country through furtherfurther penetration in the periord after 1973. iport pecrrerrutuon We A. ll., it some point, jeopardize

Market penetration is an important goal of foreign national security interests in excessive import dis-%teel producers: once they achieve if. they tend it) harg piern fdmsi te rdcin te hr

On. Price differentials are a mechanism for penetratlion. rlagemdento expcei tel productIfigh .imoteetr-

Mus, even if rapidly nsing cost factis are wor king ton m di xpoitried sien t cap% Iital ivsmenrt in new

IRtiflst the comparative advantage iof foreign steel pro- lion soira es oulfwefind caursalivesteneto supply

ducers, I am not fully satisfied that penetratiirn V. 5 uld our ow i. dunecl welds filly. thekled lnabor it supply

Itahilize or decline. rhe recent as.'reenient between the that iow doeineed hevuif.Iheg %ipdortsulyel

ILEC and Japan to reduce the Litter's exports tor the bcinine anotheti bottleneck to expansion later on.I'urope-an market has shifted those exports to thre U .S. 'Ihcs are matters thait must also be addressed innhirket. Japanese imports rose tV~o illiorn tonis in 1976. evaluting the impact of steel trade in the U.S. market.represienting nearly 36 percent of ltral unirnprts.MOreover. E:uropeatn producers havc expressed a dc- ~ i nre nJniis57Tenna aesffI

%tic to restrict steel shipments with oin aniother. -%%,title ,mr-v1% bhs'd ai rihe lii is i i mrintis ti 1977. n.iounted io 4.8maintaining output levels in the individual countrics, If millsi-e. wn, 4')2 rx~ rtrsis isitc i car 1976 imiwirri

Mr. Jondrow has made an effective contribution to Commentthe study of trade liberalization in steel: but we must becautious about drawing premature conclusions aboutpolicy implications. A much broader perspective on theissues is essential at this stage.

The following comments on the paper hy Jondrow etal. deal mostly iith chapter 3 on thc econometricmodel. 1 his chapter is of particular interest not onlybecause of its central position in the study and its gener-ally high quality, but also because of the analytical andpolicy issues it raises Since many of my comments will

ine, itably be critical. let mc preface them by expressingbroad agreement with the approach taken in the studyand with the methods by which they were derived.More specifically I endorse the principal finding thatremoval of steel import restriction% would not lead to asignificant decline in domestic employment and that thebenefits to the public at large from such removal wouldoutweigh the lose Indecd I shall argue that the studydid not go far enough in identifying the benefits of freetrade in steel to the U.S economy.

The basic model formilated in the beginning ofchap-ter 3 presents no great problems, though disaggregationwith respect to product cli ses or steel-consuming in-dustries might have added tw its u lulnc, (and, to besure, to the difficully of estimating it). The first majorquestions atisc on pages 3 9. , het it is stated withoutfurther explanation that "The estimates are con-structed using annual data (1956 - 71). Iquitions werealso estimated based on quarterly data (1964 -- 1973. II,but they were much less saisfactory and hence are notrepored."

Why did not the annual data cover a longer period?Some of the graphs indicate thait earlier data %% ere avail.able: since no list of data sources is included, it isdifficult to check further into this question It is impor-tant because an adequate number of observations isessential to confidence in any economcint: exercise.Considering the laige number of vanable% that have tobe considered, eighteen or nineteen observations aredecidedly skimpy

It would also have been illuminating t o knoA moreabout the problems with qu.iiterlv da t (aid whv theywere available for less thin ten years) Did the quarterlyresults contradict the underlying economic theory, orwere too niany estimltes siatistically insignificant' I heauthor may well have been correct in concentrating on

-19-

the annual result, hut if quartetk im aa gas e a slilfetnii mulde to the extensise litertuetiirn industnial price equa-picture the reade r shoul d bie tol tioir nh %ih is siress5 on labout product isitv tout put per

Although the ,inal% sis of utibtituttioni 'tt fianhour). this fauit oi is not adequattely captured by thehome -produtc ed aind tinpitteild Itel is c.'nipeicitils f'.i -' gen penettition ra.tio used by londross as adone, further insight cou0ld hiveC [LCcii pol kit !iv look- iS t tecfio.kiil stige. As is trule in) the othering at the composition ot imtp.'its li f .tl~l ta' st's. Pi ii.. liw- ptisc definition titfile daititblcs andwhich was rather different Itom hie s'ttI"l1siii''i Ot tc~if reIA11S Ill the JIA sv..Uld hitse been tin order.domestic outpuit Intpott penicii (ion %%,t, gre..i'.t tin he itithot niaks's onkv modest claim% for the remain-simple products such as reiiilorcinui hai, Avtle the 11n0 1%Wo VL11.11t0i.. . his model. those for domesticdomestic industry . at least .ittil the impitprstn ,t miiusad f.'icn suippl . and his mnodesty is jus-quotas, was not seriousls threatened tit such bt L.. i- ttiid PI1tthe pul'hshcf equiations are the result of aand- hut ter productIs a% C Old -rot led sheet I h c e t hut ed Ioii pr . c, it t ri.d aid cruim that is not de scrnbed inelasticity of substitution bet vs een itipoti ts:iinl hoisin at iIc 1 \ ppuclik Ishbe pr t ioble m was ap-produicts may in large patrt reflect suhsttuiii d.itt 11 ptr cc cd iifo ful t a-ccm clttratther than throughdifferent product classes. a hether on the stippl sidecw pitits, igitit thur liet.intirc ott ins estment functions istin the demand side. noit re tcitc , If itt iilu.itc t hat a more sat is lactory

I-he overall demiand eqi tation for steel is niich less C 5 a i e tt a id nor ithe obtained, for the de-cianvincing than the sri stitut ion eq it ion. Itt s ttot clear vel, tc it 11 itt1M ap I .1 a or ptiIl Since thethat Q as defined hy J undrow is a suti tisnt coin - 9Sfs onk Is iic umiatoi tie" frit has been huilIt in theprehensive measure of a,:tivitv in steel-conuningi M. L nitel Sii ind ntne t under construction at prey-dustries: where do can% and electric appliances coime cilt o.ii [i he e sustitg mtill, uie now obsolete byin. for instance' Moreos er one vs tnders fhitbe (_ is tue i nitionif taind~mi ( .,rid otifen also by enivironmen-really exogenous. conasideting thati at, ifete~d by im till stniiudi~st. iand lhi, iias tkell be the main reason forports of steel-intensive products such ats autortiobles%. It Oiii steel indtiLtry s vujln abilits, ito imports. As far asisalso diffieult to believe that tionte rroius nietil a i c the one cti tl. c a ptc is has actutally> been dec finmg de -MaJOT Substitute fot steel, tol the eit assUrited Ill theC spii. sktis its estiteit in the moiidernization and expan-equation on p. 3-- 28) that it chige itt tonleracus pi ices sin ofl em Sing plaitswouild have as muc h impact tin steel en iitipt ioni as ait Phie fite91 ign IN Sup l i cv piti price)I equat ion featuresequal perce itige change in st eel pi ices. tnt this equa t ion iia Iitr.'e it iite t c pat .i tile icr, (im e.v %ith only eighteenthe elaisicity estimate is not %cry significaiit and at fien cii bsers attonS) hut even so ilies, not lit the data veryterm wiith asurpristngky large catsecelce does ofie Snc. these v artahlcs could advantageouslynearly all the vs ork, coisting doutbt oil ithe astimtpt ott hive hee i ontiti ied or sintintid. I he most in te restingthat the coefficient of' In Q is unity Vcofl Imot fro fi his equtiotn is "the strong effect of

With the short -termyutppl y equnat ion Mii epfrec isely., st i[ dciii tid orttire e spci t price,'' i Ilustrat ing thethe price equation) %ke conito the crux of the stel tatc lllcintit iic ee lebility between the worldindustry's problems. Steel executive% hive loing becen rm!' kit inds the I.i S mtarket ft1, is lso estimated that inemphatic in their rejection of price competition is inap POW)1 - ltie espisri qtioit-ts hell iport% into the Unitedpropniate for i ndu stilnifate ri ak Is scept for the i egui SI. C aucs Afi ti 's a&li t 20 )1 ie ic it. a plausible figure thatlated sector, there is probtrbly no maloitndustry \%hlure itt. Ks cite v ikdtt i s l the tquuut.i scheime was pursuedprices are less responsive ito demand tflictuatiotis itin vs ii h sutch vigil, sitiecii )tit iirket couild have absorbedin steel. though some discunt ing tiom li st prices tic. te lv ttiill ciintonail to ns vs it hi ut sen.t~l uaiftll ectingcurs intermittently. 1~his inflexibility ofI steel price%. to the kt'titstic iflttiltrewhich I shall return be loss . is conlia niek] by i. mitruiv 's ' s tlist ii c ifigIlffck:is oft pairticul ar equnat ions,analysis where the ceffic ient of ciapacify ut ili/aition is the 1"iCiiil) iOf s hIpC ci us tsstP..1bC as% a first uppro i -small and barely in excess oaf its standard error. mt.tionitlic %hole is fict thtan the pat ts I shill not

In other respect,,. however, the discuissiion of short- coiiiint ont the piiie ins (or 1i74 78, since theterm price determinaition leaves siomethting tot be de- tsiuritiors tin \010t1 ORi t u'i e bsed ippciar to havesired. The observed leveling-ff tuf steel prikite itt the beeti muirc.alisit .. ppmttil the niieuicknic fore-1960s may indeed have been title It imliuti. butl stild i tisi ot tis putrpose slit tnot inltikipate tile sever-also have reflected President Kennedy'% 1962 unt(ervtn uN ,I the I f4 ' -oinsii In 19"S. for instance.tion and later jawboning by the Couincil kit Icoirtia doirtitei hilirticri tkucie cuts X0 million tints ciim'Advisers, for which steel as ithe putncuipat iTi'Ct pacil lo thue pcii,1 : litilioni tit's and nipoti%Furthermore the eairly and middlle lhtts A ere a. ptiiid a ecc s kcli h, I- 'lie piiues i\ Art updaing ofof relative price stability generally. Noi tefeteit is t ticsc tt11,it tionits . Ilutu1, cla rlIs V be \lrir Ibl

"1 0

Space does not pernmit deltiled coinment on t he oIltier cait Ie. %&hile (,it ill-. (tinless proftih iti 1% encou age pricechapters. In chapter 4 of the udetIlsing I Cset I ICh 1eporT e1Csibi IIt Vthat was sumnmarized in Joridrovss pitper -%rnold Kill., I hie fcdei il gov ermnent tinder Presidents Johnsongrapples imaginatively, bill In (tie end 111c,.nCtusivelV, and Nixon .i s onts too tvillir, ito negotlte **votun-with the well-knowin conuindiun 0l incresitg ctrus tisv citoit quii . folt ing aI paitterni atieatY laidto labor in the short ruin. Sori of his wili o n the Itt tmestlsadnetfrign steel industries w&ereproduction fu nction could be integi a td i11tofilie ox Ci all pre j-.t red It' C:o pce hema ,,ccix port quotais permnitiedmodel of chapter I Ihle akriat%s it o displaicicirnt losses iheni to charge higher prices In the United States. As f -arin chapter 5 is extrenmely detailed and tppml entlv com- as revenuie is concerned. an export qUOrt IN equivalentplete; whether it will con vince thle nifon..ii I, an. t her to a d LIt CoUlee td by the expt~orting Industry (not. be itmatter. Chapter 6. a sophisticated discutisson ot sini plus noted. h, (tie exporting Countri t )f iiaking the quotameasurement, is somiewhait out olt plaice in (his tenciiafly ' oltuntil conipetisatot titidet thre 6cneral Agree-down-to-earth studsy ment on I Fade ird Itaril titlrie, xi as avoided.

Chapter 7 deals mostly with the suitability of basic- Were these arraiigemeiits tit the puiblic interesi" Al-oxygen penetration as an indicator oLit Cost diftI ic C'. tioUgli the s olunft i qiot.. s h.. s lapsed. the questtionnot surprisinglyv it does not comne ito a deftiniteI concl u- is not lucreN O f academc 111 tC e Ls t . ir arty ftttie prol-sion. The fact is that there have been Imiportant tecrion giv en to .rii steel irdostri may %&ell be on sitti-technological change% unrelated to lIOF. but these are lar lines. Ii my~ opinion i e ansi m'carlv tin). Inlexi-never mentioned. Examples are continuous Casting, ble prices ate conirary to ecotonrict ciliiency birth in awhich dispenses with !he labor- intensive bloomigmni I. muicoorit airnld ailactoecoitonic senuse. icro-and computerization of rolling mills: the ever- ecornieAllv., because caritel-like pricing practicesincreasing size ofihlast furnaces atid the pretreatment of tend ito keep inetficnt tiiitix in orpertion., there haveores can also have a significant effect cin cost. [he been very few batnk. tipics itt this stagnant ittdusirtendency throughout the report to identity technirlogi- vrhere the effict t ni tiaiv appi crrtv retrainedcall change with the basic-oxygen furnace detracts fromt from expansioin lcesuecrl thi otigl new planit for fearits usecutisess, as -a contributio" to) stel InclussTY of uipsetting the eslaHiishcdl utaket pattern As aI resulteconomics. both the pt ofitabillfit and the produlCtix t% pertormance

In conclusion I return to the questinr oifprice flexibil- of the steel indusrix have been below average.ity as it is affected by impiorts. I[he tidustry's abhor- Evetn nmoire weitous ts [lie macroectontiirc etiect ofrence of price competition goes back many years,. at intlexible prices i more precisetv ,ot prics that .ire de-least to Judge FHbert Gary I 1846 192) %s ho, fnistrated terintitlis prriliativ Is isost and hear little relaiiin toin his plans to turn U.S. Stcl inrto a inotipol y, enjitioined dc iand coiid itI111 inir 5she pirices ate Intile xible.the successor firms toi "livse and let livxe." In etiecti. the chances itt deii. d are miiiedi.tte s t lims kited Init osteel industry behaves like a catel, thbough this does not c hang~e,, in out1put - and ii tiliii e Ini ciii p1o~riete thus iinecessarily mean that its; practices are illegal. ( I lie his- if demand in iii ii n'- it aiit indolit i sl isietas s teel)itorical argument for this industry', adhetence ito full- weakens. -its cuslolTnIet s C 1t tin gixoI cilllri incentivie tocost pricing, namely large fixed cost. is incidentally buty' ninie. CspecOAll, lor ins enrorv I Ie fill in outputcontradicted by the report, which puts variable cost at and emplo)N nent brings -1i ittil c liscal un't mtonetary87 percent of total cost.-) policies In.to action.- xxiitt their amteii-~nat intlaitionary

This harmony aintg potential nvals, was threatenedl piessures. Iniadequiate pric tlexibility in majorwhen import-, becanie competitive in the W0%(l. 'Ihe Industries --- steel is not art exiception, -just an evitemeexpoilers did not belting to the U.S. inuformal cai tel and case -theref'rre ciirlutes puss cfully to the corn-they were not used to inflexible prices.~ Although the bined Inflation and utrertiplovis nrt that has increasinglyEuropean and Japanese steel indlustnieis are hardly plagued its it)[ccciii sears h-i ietitore. the low% levelsexamples of perfect competition. siilst.rntial price tiuc- of itivesttinctit irirpuit in :aiti-ilike pitting deprive thetuations in response to changes, in demtand dir (occur. ccorrony Of tiCeded stiIruits tlr rthe piate sector.The American industry wias conerned not only he- I luc rernuval of mlloit rextiiitions on steel. ( here-cause of the lossxof bus iness. hut also because its pncing forie, has ,In tirio tii~c siell fe inst rtet indiistry in-practices becamte v ulne rable. - tls is title of the reat. ins nic that cly atfet t eu As t he i epsi I thik s. thte di sli tiv cwhy imports were ultimately contolled by- qtiotals etiets sit such thibet ition would be minor in anyrather than tariffs; import quotas reifoirce a domestic casec Arriong its benef-its. Lodirsttabte wecight should

he giuen to the invigorating force of price competition Replahich could gisc our steel industry the dynamic role in,or economy it had many years ago. James M Jondrow

Though agreeing broadly with the general approachand findings of the study. Professor Houthakker makesa number of specific comments on the estimation of themodel.

One is on the limited number of annual observations.This comment is applied ,pecifically to the import sup-ply equation. Though more ohserations are clearlybetter than fewer, a large number of observations is notnecessary for precision of estimation. Most of theparameters of interest in the model have significantcoefficients, even %%hen degrees of fieedom are takeninto account. A necessity lot reliable estimation is vari-ation in the independent variable,, and this variationwas available for most of our time scrie, Furthermore.by using a short sample period one avoids some of thedifficulties associated with qructural change that mayaffect long-term estimates.

One possibility for increasing the sample size. asProfessor Houthakker mentions, is quarterly data. Oursources for quarterly data did not extend back as far asfor the atnual data. Further more. though quarterly dataincreased the number of obscrs.ations. there was actu-ally less variation in the independent variables over theshorter number of years and there was apparentlyrandom variation in the dependent %arrables. In thequarterly model, these problems led to imprecise esti-mates of substitution between domestic and importedsteel. Estimates of the elaticily of substitution hadhigh standard errors and were not robust to even minorchanges in the specification of the equation. In con-trast, results based on annual data over a longer span ofyears were more precise and robust.

Professor Houthakker finds the demand for steelequation less convincing than the substiution equation.The latter is much more important in evaluating theeffect of imports. lie questions the use of the non-ferrous price its the price of substitute materials in the

'In this connection it is unfortunate that the report has little or demand for steel eqialion Yet results using thenoingto say about steel esports. which have been sgnitiant even wholesale price index for all commodities as re )rted ini tome recent years (in 1970 and 1973 exrpns were ,iumint h.lf( the the prper wcre much the ,ime. This robustness is de-in of impons). A less hidehound prilnir and invesltnt ikllicy

UIi well make our industry into a sizable exporter o cenain sI rable since there is no obvious best measure of theItrodt. price of substitutes.

-22-

His main comment on the short-term price equationis that there are other explanainn besides imports forthe leveling off of steel ptice, in the 1900s. We agree.

steel price during the I960,; could be expLained hy aleveling of input prices. But othcr factors might also beinvolvsed.

An estimated capacity equation vas reported in theresearch report on wkhich our conference paper wasbased. The equation turned out to he unimportant, andwas dropped. Still, as Professor Houthakker com-ments, an equation explaining capacity formationwould be very useful. We agree. In fact. our findinc--s onthe subject seem plausJible. Output is a Str On'a dCtCrndn-ant of capacity %% hen corrected for cyclical variationand substitution between capacity and other inputs islimited.

-23-

CNA Professional Papers - 1976 to Presentt

PP 141 p 155 PP 167

Mflanhs. blurice M, "Gneralleed Hermit. Poio Squires Michael L . 'Soviet Foreign Poicy and Kleinman Samuel ROTC Sern,ce Commitments a

nomils." 5 pp. Feb 1976 (Reprinted from the Third World Nations, 26 pp Ju 1976 tPrepared Comment." 4 pp.. Nov 1976. (Published in Public

Jounrnal of Computational and Applied Mathe for pevenaion a; the lyt,diorst Political Science Choice, Vol XXIV. Fall 1976) AD A033 305

maetci. Vol 1. No 4 11975) 273277) Assi.ciaton meetings Ap 30, 1976) AD A028 388Relearch supported by the National Science PP 168

Foundation PP 15o Lccfnan. Robert F , 'Revelrdaton of CNA

drailige AilIho 'Approaches 1,1 Ch i.haniu Support Personnel Selection Measures,' 36 pp, Nov

PP 143 ret Recrigti.'n 12 pp. Ju 1976 Reptitd 1976

Morowitz. Stanley and Sherman, Allen lLCdr Iron, Pairei. Recognitin iirPergfon PFesti. Vol 8.

USN. "Mintenance Personre Efllctienes in the pp. 8 911 1976) AD A028 692 PP 169

Nany." 33 pp. Jan 1976 (Presented at the RAND Jacobson. Louis S., "Earnis Lowses of Workers

Conference on Defense Manpower. Feb 19761 PP 157 Displaced from Manufacturing Industries," 38 pp.

AD A021 581 Morgan Wihirn F 'liinamployment and the Pena Nov 1976. (Delivered at ILAB Conference in Dec

9cn Budget r, There Anythiuig i the Empty Pork 1976i. AD A039 809

PP 144 Bartelr" 20 pp Aug 1976 AD A030 455

DOrch. William J.. -The Navy of the Republic of PP 170

Chine - History, Problems, and Prospects,' 66 pp, pp 1i Brechlilng, Frank P . "'A Time Series Analysis of

Aug 1976 (Published in -A Guide to Asiatic Haskell. LCdr Richard 1 (USN) "'Eprimenrtal Labor turnover." Nov 1976 (Oelivered at (LAB

Fleets.' ed, by Barry M. Blechman and Robert Velidation of Probability Predicron,r

25 pp.. Aug Conference in Dec 1976)

Berman. Navol Institute Pressl AD A030 460 1976 (Presented at the Military Operations Re

vearch Society Meeting fall 11976) AD A030 458 Pp 171

PP 145 Jordan A S * and Ralston. J M , 'A Diffusion

Kelly. Anne M.. "Port Visits and the "Inter 1 159 Model to, GaP Red LED Degradation, 10 pp. Non

nationalist Mission" of the Soviet Navy. 36 pp.. McCoi irll J.i 'r lh5 "le corhio Artcle, The 1976. (Published in Journal of Applied Physics.

Apr 1976, AD A023 436 New Gorhkov Book and Their Relation to Policy, ' Vol 47. pp 45184527, Oct 1976)

93 pp., Jul 1976 Published in Soniet Naval In 'Bell Laboratories

PP 147 fluence' Domestic and Foreign Dimensions, ed by

Kessler. J. Christian, "Legal Issues in Protecting M MccGwmre and J. MtcDonnell. New York, PP 172Offshore Structures," 33 pp., Jun 1976 (Prepared Praeger, 1971) AD A029 227 Claser, Kathleen P. "Unemployment Insurance

under task order N0001448SAOQ91-0023 for and the Length of Unemployment.' Dec 1976.DNA) ADA028 389 PP 160 (Presented at the Unrversity of Rochester Labor

Wilson, Desmond P. Jr . "The V.S Sixth Fleet and Workshop on 16 Nor 19761

PP 149 the Conventional Defense of Europe." 50 pp , SepSquires, Michael L., "Counterforce Effectiveness A 1976, AD A0105 457Comparison of the Tsipis "K" Measure and a Com PP 173

uses Simulation," 24 pp.. Mar 1976 (Presented at PP 161 Klem.n. Samuel 12. "A Note on Recll Oil

the International Study Association Meetings. 27 Meich. tichael E and Pees Vice Adm Ray (USN. ferences in the AddedWorker/DiscouragedWorker

Feb 1976) AD A022 591 Reriredi, "Flee Comnianders Afloat or Ashore" Controversy." 2 pp., Dec 1976. (Published in the

9 p., Aug 1976 (Reprinted from US Naval In American Economit, Vol. XX. No. 1, Spring 1976)

PP 150 stiute PFroceedgs. J 197b) AD A030 456Kelly, Anne M. and Petersen. Charles, "Recent PP 174Changes in Soviet Naval Policy. Prospects for Arms PP 162 Mahoney, Robert B . Jr "A Comparison of theLimitations in the Mediterranean and Indian Friedhem. Robert L. Padian-enar, 1)iplomacy,- Btooklngi and International Incidents Pitaerir.r 12

Ocean," 28 pp.. Ape 1976, AD A 023 723 106 pp Sep 1976 AD A033 306 p k en Irao Id Pocs 12pp. Feb 1977 AD A037 206

PP 151 PP 163Horowitz, Stanley A.. "The Economic Con Lockman. Robert F A "0 Mod-r for Predicting PP 175

sequencres of Political Philosophy.' 8 pp. Apt 1976 Recruit Losses " 9 pp . Sep 1976 Presented at the Levine. Daniel, Stoloff. Peter and Sprull. Nancy.

(Repinted from Economic Inquiry. Vol. XIV. No 94th annual convention of the American Psycho "Pubic Drug Treatment and Addict Crim," Jose

1, Mar 1976) Inlgial Arrocratlon, Washintn D C. 4 Sep 1976) 1976, (Published in Journal of Legal Studts. Vol

(Published in Defense Manpower Policy (Richard V. 5. No 21

PP 152 L Cooper od I, The Rand Corporation, 1979).Mizrahi. Maurice M., "On Path Integral Solutions of AD A030 459 PP 176the Scheodinger Equation, Without Limiting Pro' Felix, Wend, "Correlates of Retention and Pfomo

cidure." 10 pp., Ape 1976 IReprinted from PP 164 lion for USNA Graduates." 38 pp., Mar 1977.Journal of Mathematical Physics, Vol. 17. No. 4 Mahoney. Robert B- Jr , An Asssesmen of Public AD A039 040(Ape 1976), 566-576. and Elise Perceptlonis in Fiance. 'Ihe United Kng

"Research supported by the National Science dom, and the Federal Rlepublic of Germany," 31 PP 177

Foundation pp , Feb 1977 IPresenred at Conference "PPep Lockman Robert F end Warner, John T., Prericn of the S - Soviet Balance and the Pnlitical dicting Atrition A Test of Alternatie Ap-

PP 163 Iter of Military Power tponined by Director, pronches." 33 pp Mar 1977 (Presented at the

azrah. Maurice M , 'WKB Espanserns by Path Advanced Research Polrct Agency, April 19761 OSD/ONR Conference on Enlisted Attrition. XeroxIntesels. With Applications so the Anhammone AD A036 599 International Training Centel. Leesbueg. Virgeta,Osrcillero." 137 pp. May 1976, AD Al25 440 4 7 April 19771. AD A039 047*Ra eflch eutvioeted by the Nertonal Scence pp 165

Foundatnon Joridrow James M Ftfecti it Trade Restrictionson Imports of Steel" 67 PP . November 1976 PP 178

PP 154 Del.eed at ILAR rnnference in rec 19761 K.iten. Samuel D- "An Evaluation of Navy UnMaeatth, Maurice M . "On the Semi Classical restcted Line Officer Accession Propgams." 23 pp

fEptiarose in Ossrstamu Mechanics toe Adhiteery PP 166 ais-ed April 1977, (Presented i the NATO Conference on

1dtonfett.e" 19 pp, May 1976 (Published in Feldman Pau. "Wh It's Dificul to Change Manpowr Planning and Org nization Drses,

Joraenl of Magtheemcal Physrcs. Vol I8 No 4, Revulton (let 1976 AD A037 682 Strt., Italy. 20 June 1977), AD A036 048

pp, i-.M Ape 19771. AD A025441

tCNA Profeeleonal Papes with an AD numbe, may be obtained from the National Technical Information Service, U.S. Departmentof CofpnnqWce. Springf eld, Virginia 22151. Oher papers are available from th Management Information Office Centr for NavalAnallysea 2000 North Bluregard Street, Alexarndria. Virginia 22311. An Index of Selected Publications is elgo available on request.The index includes a Listing of Profeaaonal P.;rs: with abstracts; islfed from 1969 to June 1980.

PPi 179pp189 pp 202

Stofoff. Pet"r H. and Sefut. Stephen 4.. "Voat. A Coils. Russell C. "Error Detect-on in Computerized Feldman, Paul, 'Why Regulation Doiesn't Work"-Moat for Perrisonnel Invntsory Planning Under Information RtevfData Baesn." Joly, 1977, 13 (Reprinted from Technoogoial Chsange and Welter*Chagng Manragaemnt Policy," 14 pp. April 1977. pp. (Presentedl at the Sitsh Crainfsefd International in the Regulated i ndustries, Brooltrsges Reprst 219,tPreasented at the NATO Confereatli On ManPocrar Conrference on Mchanized information Storage and 1971. and Review of Sotrait Economy. Vol. XXIX,Planning and Organszation Design. Strmn. ItAlN. 207 Rutrat Sysleess. Ceanseld lisstitifl of Tuncs March. 1971. Nlo 1.) Saep 1977, 8fppJune 1977). AD A039 049 noly. Cranfseld, Bedford, England, 26-29 July

197 1f. AD A"13 580 PP 203pp 180 Feldman. Paul. "Efficiaoncy. Distribution, and tf.

Horowitz, Stanley A and Sharman. Alflan "The pp 190 Role of Government in a Market Economry.'Chantrtm.. of Na...l Pernn.l antI Personnel Mahoney. Robert? , Jr., "furopeani Percupson' (Reprinted from Thce Journal of Poltcal Economy.Perfonrmance." 16 pp April 1977. l~srournted at the and EanivWei Competttton[ 96 pp., July 1977 Vol. 79. No, 3 May/Jurs 1971 I Sep 1977. 19 pp,NATO Conference on Manpsower Plasning andO (Prepasred for prerentotion at the annual meeting of AD A045 675Organizationr Design. Stress. Italy. 20 June 1977) the Infernational Studios Association. St. Louis.AD A039 060 Mo.. March. 19771. AD A043 661 pp 204

Well,, Anthony R.. "The 1967 June Whan Sovietpp 181 pp 191 Naval Diplomacy and The Sixth Flows '- A Ae

Sait. Stephen J. and Stoloff. Peter. "An Inventory Sawyer. Ronald. "The Independent Field Assign appraisal." Ott 1977. 36 pp . AD A047 236Planning Model for Nany Enlisted Perronr." 35 mains One Mon's Vuwo." August 1977. 26 pp.pg.. May 1977 Prepared far presentation at the PP 205Joint National Meeting of the Dpeations Raesarch pp 192 Code., Runoff C.. "A Biblsrnsrc Exnstmmation ofSociety of America and The Insttute for Manage Holan. Arlene. "Effects of Unemploynment In. the Squae Roof Theory of Useanttin Publicationmenir Science. 9 May 1977. San Francinsco. Calt curentce Entitlement on Duration and Job Search Productivity." Prsesented at the annual meeting offern.). AD A042 221 Outcome." August 1977. 6 pp., (Reprinted from the Amercan Society, for Information Scteece.

Industrial and Labor Relations Reviem, Vol . 30, Chicao", Ilrntos, 29 September 1977 1 Oct 1977, 6PP 182 No. 4. Jul 19171 pp,. AD A047 237

Murray. Russell. 2nd, "The Gluest for the PerfectStudy oe My Firest 113B Days at CNA." 57 PP. PP 193 PP 206Apedl 1977 Horowtz, Stanley A.. "A Model of Unemployment McConnell. James M. 'Strategy and Marsron, of the

Insurance" and the Work Test." August 1977. 7 pp. Soniet "an in the year 2000"- 48 pg, Non 1977PP 183 1(Reprinted haom Industrial and Labor Relatinons (Presented at a Conference on Problems of Sea

Kesn. David, "Chvin n Soviet Naval Foces." Renrewy. Vol. 30, No. 40, Jul 19711 Penmmo a me Approach the 21cr Century. Ipoesrd33 Pp.. November, 1976. (Publinshed as pert nf by the American Enterprise Instintute forl PublicoChaplt 3, "Generall Purpose Force, Many end pp 194 Poicty Research. 6 October 1977. ancd subeoguarrtyMarine Corps." in Arms, Mass, and Miltry Budgets. CGasten. Kathleent P., "The Effect, of Ursemploy Publinshed in a collection of pawse by the les,Francis P. Hisabor and Wfilliamv Schneider, Jr latn . meet Insurance on the Duratioin of Unemployment total. AD A047 244(Cranes. Russetk & Company. inc New Ynrkl. and Subsqent arn.'August 1977. 7 pp

1977), AD A040 106 (Reprinted from Industriael and Labor Relations pp 207Reniew,. Vol. 30. No. 40. Jul 19771 Gotdbevgo. terence. 'Cost Effectie-es of P0ic

pp 184 lentuaf Federal Poft-e Affecting Reearch 6Locksman, Robert F., "An Overiew of the OSDI pp 195 Denelopmrent Expenditures an the Auto Steel andOMR Conference ors First Term Enlisted Attritionv,- Beechling. Frank. "Unrerployment Insurance Tunes Food Iedustrial - 38 ppg Oct 1877 Pfresented an22 pp.. June 1977. (Presented to the 39th MORS and Labor Turnover Summary of Theoretical Southern Ecceontmic Association Manins p gesWorking Group on Manpower end Personnel Findings." 12 pp. lRePt-nted from Indutrial and 2 Norvsber, 19771) AD A046 29Planning. Annapolis. Md.. 28,30 Jun 19771, Labor Relations Rectory Vol. 30. No. 40. Jul 1977)AD A043 618 PP9206

pp 196 Robertc. Stephen S . "The Decline of the Ocerseatepp 185 Ralston. J. M. and Lorinmur. 0, G., "Degradation of Statin Fleets The Unitred Staves Aisr Fleer and

Kaming. David. "Nary Technology and Naval Foeces Bulk Electeoliutnecenf Effinciency in Zn, 0ODoped the Shansghi Crisis. 1932 18 pp Nov 1977 1R.in the South Atlantic." 22 pp. (This, paper me, the GaP LED',," July 1977. 2 tpp. (Reprnted from Printed front The Amertcon Nteptune, Vel IXXEVItbasis for a Presentation Made at the Institute fo at IEEE Teonvactns on Electro Dence. Volt No 3. July 19771, AD A047 245Foretgn Policy Analyses, Cambridge. Most.. 29 April ED 24. No 7, July 197711977), AD A043 619 PP 209l - Classified

PP 197pp 186 Wells. Anthony R.. "The Centre for Navel Anal PP 210

Miahi, Maurince M., "Phase Space Integrals. Withb yles." 14 pp.. Dec 1977. AD A049 107 Kassing. David. 'Protactrsg The Fleer.' 40 pp Dec,out Limiting Procedure," 31 pp., May 1977. (in 1977 (Prepared foe the Anmerican Entererse Incited paPr esented at the 1977 NATO tnstitute pp 198 slrtuts Conference on Prohlents of See Pome a W9eon Path Integrals and Thetr Apptlication it Clamion. Kathleeon P.. "The Dinrhiutional Effacts of Approach the 21st Century, October 8.7 111711),Quantum Statistical. and Solid State Phicsn,c tUnemrployment Insurance." 25 pp.. Sept. 1977 AD A049 109Antwerp, Belgium, July 17 30. 19771 (Published in (Preranted at a Hiontr Instintution Conference onsJournal of Maftbematicaf Phystcs 191(111. PP9 Income Distribution, Oct 78B. 1977). AD A054 423 pp9211298.307. Jan 19781). AD A040 107 Munrahin. Maurcine M. 'On Ailponinsatteg the

pp 199 Circular Coverage Punction.' 14 p Feb 1978.PP 187 Dutrch, Willinam J4. "Revolution From A F .A.R. - AD A054 429

Cote, Reneff C., "Phomography, for Opeations Re The Coben Armed Forces in Afrinca and the Middlesearch." 35 pp.. Aprtt 1977 (Preeted at the Jotnt Eart.' Sep 1977. 16 pp.. AD A044 28 PP 212National Meeting of the Operations Research Mangevl. Marc. "On Surse Charmrrmt lssrtSociety of America and The Insitute tot Manage PP 200 Value Problems mtrh Unique Sofattrns" 20 pg..meet Services. San Prancico. California, 9 May Froware. Bruce F,. "The tutecf Scenes Navy." 40 Junt 1979. AD AON 535197711, AD A043 620 ppi. Dec 1977 i~suhlched a a chapter in The U.S

War Machine by Salamander Books. Engiaed.19781. PP 213pp ISO AD A049 108 M~engel. Marc. "Fltasns an Shieleme sth Mallis

Dutch. Wlliacc J.. "Informalton, Processing and Pie Steadyh Stares. Applscrun to LaoleterO Egmac.Outcome Forecasting Ice Multilateral PPotatoc P201 teon." 12 pp.. Feb 78. (Presented at the, FirmTesting One Approach." 53 pp.. May 1977 (Pro Dutcht. Wtiltam J. "The Cuban Milfttery in Africa Annual Werkshep on thle Iefcemation Leakfage be-pared foe llt p fessieton to the 111th Anneal Con anld The Middle East, Front Algeria en Angrola," pean Apantd Mathemattostle and Irnte. Naccf PGsenaen of the leternantvolsl Studies Associations Sep 1977.807 pp.. AD A045 876 School. Feb 23 26. 1978), AD A071 472hose-Park Poe. HOl St, Lousn. blirnocurt. March

14-20. 187,7). AD A042 222

PP 214 PP 226 PP 241Womlarsd. Robert G.. "A Somewast Different View Rolm.o. J M. 00d J1 W, hiktol "~Tmperature and Mangol, Marc. "Fisicsiat.on at Charrsrca lotla1ss

of Tire Opimal Ndaval Pt te.r, 37 pp.. Jun 1978 Current Deipendence of Degradations to Fled. .t..' 24 pp., De. 1978 (Published in Journal oflPtesisntd at the 1976 Corseentron of its. Arrnassn Emtting GaP L.ED,.- 34 pp.. Jon 1978 (Pulished Chemical Physics. VoL 69. No 8. Oct 15. 1978),Political Science Asssciattoo (APSA/IUS Panel 00 or Journal of Applied Phystcs. 50,. 3630, May 19791 AD A063 767"Chnegiog Straegc ifeaaan..nants and Military AD A058 538Postuie"). Chrstge. Ill. Septenmber 2, 19761. - Belt Telephsone Laboratories, Inc.

AD A056 228 PP 242PP 227 Sinpson, Wilam Rf. -The Analysis of Dynamnically

PP 215 Mangrel. Marc. "Uniformr Treatment of Fluctuations interactive Systems IA~r Combat by the Numbers)."Cotta. Russell C. "Cottnsants oat. Ptrnctptes of In at Critical Ponts..' 50 pp,. May 1978, 160 pp. Doc 1979. AD AG63 760forarattion Retieval by Maenfted Ktchren.~ 10 pp. AD A058 539Mar, 78. fIsublinhadus a .Letter so the Editr PP24Journal of Ocsumenntton. Vol. 31. No 4, pages PP 2233poWlia9. APoailsi omlto299-301. December 1975). AD A054 426 Mongol, Marc. "Rfelaxation at Critical Points Spo.Wltt . APaasrtcFraatr

Deterministic and Stochstic Theorty.' 54 pp.. Jun of Murphy Dynamrc, os Applied to thre Analyts of

PP 716 1978. AD A058 540 Dperatronal Research Problems."* 19 pp., Dec 1978.

Code. Russell C.. "Latka's Frequenty DitrrbutionADA676of Snertrfrc Productitry," 19 pp. Feb 1979. PP 229)Pultsshd to tire Journal of the Amrerican Socrety Mangol, Marc. "Diflsso Theory of Reacton PP 244for Infoensaton Scasnes. Vol. 29. Nlo. 6. pp. Ratesr I formulatron and Einstein-Smoluchowski Sherman. Allan and Horoeitz, Stanley A. "Maim36370. Novneber 19771). AD A054 425 Approximation." 50 pp. Jan 1978, AD A058 541 tenante Costs of Complex Eqipment." 20 pp., Dec

1979 iPublished by The Ameritan Society of Nanalpp 217 PP 230 Engineers, Natai Engineers Journal, Vol 91. No 6.

Costae. Russell C . "Bsbtsountsrc Studies of Scienifict Mangel. Mast. "Diffusion Theory of Reactron Dec 1979) AD A071 473

Produactivity." 17 pp.. Mar 78. ifteantrd at the Rates. 11 Ornstein Uhlenbeck Approximation," 34Annual meetinsg of the Aeriecan Society for In pp.. Feb 1978. AD A0418 542 PP 245teetnararon Soanat held in San Prancisco. L.ifornsa. Srmpson, William R.. "The Actelerometer MethodsOctaober 1976l, AD A05A 442 PP 231 of Obtaining Aircraft Performance Irons Flight Test

Wison. Desmorrd P. Jr.. "Naval Protection Fortes. Data (Dynamic Performance T'dstsq)." 403 pp..PP 718 - Classelsad. The Case los a Resontree MAP." Aug 1978. Jun 1979, AD A075 226

AD A059 543PP 219 PP 346

Nuetasogar. R. LeVer, "Markert Analysis wrti. PP 233 Beecirg. Frank. "Layoffs and UnamploymensRationsal Espactations: Theory and Estimation." 80 Jatobson. Lous, "Can Poicy Changes Be Made insurance," 35 pp.. Feb 1979 (Presented at thepp., Ape 78. AD AGM4 422 Acceptable so Labor?" Aug 1978 ( Submitted los NBE A Conrference on "Low Income Labor

pueblication in Industrral and Labor Relations R~e Markets."* Chicago, Jun 1978)PP 220 renw), AD Al061 528

Maurer, Donald E., "Diagaonalization by Group PP 248Srsces." 26 pp.. Ape 78. AD A054 443 PP 23 Thomas. Janms A.. Jr.. "The Transport Properties

Jacobsons. tous. "An Alternative Explanation of of Dile Gases so Applied Fields," 163 pp., MaePP 221 the Cyclical Pattern of Gsses'" 23 pp.. Sep 1979 1979

Wotniand. Robert G. "Superpowver Natal Dsploroacv an the October 1973 Arab-Isreli Was." 76 PP 234 - Reisedpp.' Jue 1979 (Publirshed so Seapowrai in the Jondrow. James and Levy Rtobert A.. 'Does PP 249

AereaseePolirtical Utrftty ansd Military Coo Federal Expenditure Displate Slte and Local Glasses. Kenneth S.. "A Secretary Problem wisth a

steaintr. Tha Washington Papers No. 61, Beverly Enpenditure: The Case of Comstruction Grunts." 25 fiandom, Nurmber of Choetes." 23 pp.. Mar 1979H4lls and Lonsdon Sag. Pulications, 1979) pp., Oct 1979. AD A061i 529AD A055 5641 PP 250

PP 335 hfagnvl. Matc. "Modeessg Fluctuations in MattePP 222 Misam. Ma-r i "Tihe Semr,cla,..ucall Expansion scopic System." 36 pa.. Jan 1979

Iflkerait. Maurica M.. "Confespsondenca Rules end of she Anfrarmoost Oscillator Propagator.' 41 pp.,Path Inteprair." 30 pp.. Jun 1979 (i nvited pape Ocr 1978 tPublished io Journal of Mathenmatical pp 351peesensted at the CNRS mneeting on "Matharotitall Phrysrcs 30 (1978). pp. 844 855). AD A0I)B 538 Trost. Robert P.. "The Estimation and Interpreta

Probtess.. in Foepoman's Path fooepas." Mrraslle. Iron of Seneral Selectivity Modals.." 37 pp., Jun

Prance. May 22.26, 19781 )Ptssdrehd so Speineges PP 337 1879. AD A075 541Vaea Lectrr Motortso Ph~ysics, 108. 01979). Maurer, Donald. "A Matrix Criterion Ifnr Normal234-2531 AD A055 536 Integral Barnt." 10 pp.. Jan 1878 (Pulitshed in tha pp 32

Illinois Journal of Mathematics. Vnt 22 189781. op NenWatrR "arsnPedn atiPrr

pp 23 72 91Knowledge of Conersaoncao Prarameters." 5 pp.. JanMonrgol. Marc. "Stochrastic Macharcs of Molecule 1879 (Published on IEEE Transactions an AerospaceIon Molecule Reactioan" 21 pp.' Jun 1878. PP 238 & Electronic System,. Vol AES 15. No. 3. MarchAD A056 227 Urgoff. Kathleen Classen. "Unemployment in 17

suranca and The Employmsrent Rate." 30 pp. OctPP 224 1878 Presented at the Cofeenc an Economic PP 25

Mongol, Marc. "Aggegaf ton. toltca~eon. and Eax indictators and Parformance The Current Dilemma Gotr ont . Ts hieSceayPotrictian In Eaploigad Aitrral Proputlsn." 48 Feting Goteomoent and Businass Leaders. presantad Gass ant . Th .hsgSceayPepp.. Mar 1978. AD A058 636 by Indrna Univrsitssy Graduate Schtool of Suttons lsm." 32 pp.. Jun 1879. AD A075 225

apllsorton of this week ware sarted at the Ittute AD A061 527of Applied Megfeengrc and Statistics. Univerity of PP 254Biish Catuarie.s Vancounear. D.C.. Canadali PP 239 Mongol. Mart and Oulanbask. Daerd B., "Intelloato

Trott. R P end Warner. J T . "*The Effects of of a Bsefrote Normal Ovr an Offset Cece." 14PP 225 Military Occupetional Training on Crnilian Earnings pp. Jan 1979

11nga. Marc. "oedfttone. Flt uafions, and the An Income Selectivity Appeoacfh." 38 pp. NovHoof Bifurcation," 43 pp.. Jan 1978. 1979. AD A077 831 PP 255 - Classinfied. AD 8061 441LAD AN6837"atonse of "hi work, rie coorplatad at the PP 240 pp 258Iensities of Applied Mihhesrse ond Statistics, Powers, Breece. "Goals of thre Cester lot Nal Marer, Donald E., "Using Personnal DntrsbeatonUrafeerlgyV of Selfish Colramrbus. Veowa. Canada Analyse.s," 13 pp. Dec 198. AD A053 758 Models." 27 pp.. Fob I99. AD A02 218

PP 267 PP 270 PP 286Throw. . "0 oanon and Fieatl Crament nts R.,nned Richard. 'fonie Policy in thle Honn of furisail. Mauric. M- "On the WKB Approit.ion

Why Ormcaearo ts Avinteyt heqlI' 10 pp. A. 9 Africa The Decision to Inltnten, 52 pp, ).n to lt. Propagator to, Arbtrary Hatttettomet~. 2b

1079. AD A015 224 1980 I To be Publtihed in "Te S-tn.t Union itn Ili. pp. A.9 19Sf0

Third W.oIld Succoe 0t lad.o," ad by Robert I

pp MDon.Idsan IIIt.,.n Prowe.0 od., Co ,S.-te., PP 287

Mengalf. Marc S, and Tholsa. James. A. I 191101 AD AD8l 106 Cop., Pool, "Lint Cycle Solutions of Rlection

"Antalytical Methods in Seamrch ftooe, .- 6 PP Diffuion Equation.- 3251 pp.. .1tte 19600

Nfo. 109, AD A077 832 pp 271 AD A087 114McConnell James, 'Soyre and Ante,,,. Sutratgic

pp Mg D-,,ri-, (In. More Tir. 43 pp, Jan 1990 pp 288

Glass. D-nd V Ne.. ItI Ching Nain, Wlt. R AD A08 1 192 Golnmn. Waiter, "Dcn't Let Yvp, Sles FP You

antd Penn. DOid A-. 'A Clats of Cottoe~eA Poeleet Guide to Visual, That Really Aid.- 29Matilon Moormae.- 17 pp.. No. 1919, AD A077 833 PP 21,1 Patt Oct 1980

We.,t Kenn..th G1 r Itt. A ....t .n Oplotnany andipp 296 Sttt.f'i. 1940 1945. 46 pp. Met 1980 PP 29

Motngl. Mat S a" Cope. Don,, K. "Detention AD A085 094 Robinton. Jack. "*Adeqet. Clatefieor Gidatce

Rlate and Soires Wtdth it Veial Seatrch.' 14 pp - A Solutton end a Ptoblowt." 7 pp.. August 1980

Not 1979. AD 407? 634 PP 273Nakadil, Michael I *Labor Sornel ii1 Wlvee with PP 290

PP 261 Hustbandt Eniployed kItc foll Timt. or Part Watson, Giegoip Hi 'Enltton of Compuatet Soft

Vtal, Catl.$ L Zella., Danid I aitd Rote Joltn 1t-. 39 pp Me, 1961) AD AUB2 220 mete inan Operational Environmnt,." 17? pp.. AuS

"lianch Cordon Thleory of Cheical Dynattics VI 1990

Angidat Deibatio of Reaction Ptod.itle 14 pp 11b

pp. face 1919 (Repinod lion, Jonutii Cheni l.Ildtory I on-eni.... oi. Adt,titing an PP 291

Pltys 701121 lb Jun 1921). AD AGI16 7 Rotcei~ tnet 34 op Mei 1960, Mtddela. G S - and Trost, R P.. "Some E.AD A1062 III towttetoen 0 the NortIon. Pree% Model." 12 pp.. 0.1

pp 262 1980Pottes. Chattet C. 'Thud Worldt Mttitaii Filtet in PP 2711 "Untetvi of lorte.

Scales Petebec lt.-.. e so p op 19. 9 AgGldtti9 totrci Delaying .n TOtwilarl andAD A077935 Slip.I flc-pnint 410 Pp Man 1960 PP 292

ADl A08, CAO, Th...n. Ai. Jamas A, "Tihe Trateport Propotioen

pp 263 of Itiierv Goo Mixltes to Applied Me~elt

Robitnson. Kathy t . Usinmg Commetnol loork., pp I7 ,ild,' 10 pp.. Sept 1960 (Pu~blithed to Journalt ofand Coanetehlh lot Navy Utnderway tiaotih hfttrne

Miut %.1 Titf locictio inSttit Chtetmicl hics 72 (101. 16 Mary 1980

ment." 26 Pp Noe 1679. AD A077 836 Multiple Ln-i Cyclot. 19 pp Me, 190 (PubIed '. SIAM J Appi ath Viil 18 No I Feb pp n93

pp 1890, AD A086 229 Thtomas. Jt Jamn. A. "kEaheetton ol Ktttatt111nt-fand, Roblert (3 TheIS Rary in Alt lTheory Coflision Integral, thin~g tile Genteralibedj

Pacific Poet Pan-ont oitd Gloipen of the futue pp )16 Phrase Shit Appnsocit." 1? pp. Sept 1380 Iftoetled

31 pp. No.1 1926 ttiMtoted at thInternational Micia, M--.,~ A 1 agettig P,,bln,- Ill Ie .i Journal of Ceemocol Pity.., 12 1101. 16 Mary

Symnponeom n - th US. so nt..... It tvhe tIt., t.leetd Votl. Apwoi,t ' 2 3 pp APt 1980 19B0

no.ongl fnlo o lo Stretegen Studte Tih, All AQ 005 096

8toeltittg Inttton end tile Vonto,- Sttirib-o pp ?94

Tokyo. 16 20 Oct 19781 AD A066 831 7pp 219 Roberts Staphenr S ,"French hlant Pointy De~ed.lelali Stitihrt Vi C.a 10-i.iii."ei't, , n f Euoo,* 30 pp. Sapt 1900 (Prenentd at tile

PP 2%1 til In,. A u~iito.i t,u oll~o Wie ii tontavntee of few. $action on Mehtttt Studies.

eteland. Rfobert G . "Wat cud ea. ,, lt, Notiti op M,v 1980 AD AO1111 091 lItteeeite.tel Srodtee Aesculetepe, Iltwalt IlandtO

Same Poleinat Iineinto the Ctarig. Ntl, S C Itaty Situation in Nortlo turope IS PPe Noe tpp 280

1I26 (Prepared fnn petoirut ti he C,.fi-ene CGoidti klo-it ir, ..... i.t

i lb tirtir iil pp 7t

of ite aotifi Sata.. on Pettpact-e ril Changfi A Stell. Ite,, on tn. p., hat. A. 1~ q9., Rabaul.. Stepthen S ,'Anoitneat of Iformtial

Military and Politinel Situation Cet-,e ho, anuii An Anpoiatn, if Itiuinnt, Anoictit Empire Paritains of U'S Nanty Cristing on Desolate

Stiatagi, endt lIntnational Studies Gnoirptitnt IS pp Ap,. 1990 ADI A005 090 Stat-in. 1860 1897." 40 pp. Sept 1980 (Presetead

Unnietity, Jan 15 16, 1978! AD A1077 828a osl eve"Soystpssas.U 9

pp 261 Anadenty. 260October 1979pp 266 Mtoreh, hMat-e M0. 'Cimmsrt on 'Onenettoetuon

tlgoft. Kathy Clano, end Btenhlingl Fltankt Probilem of Panctiotnell Integrele , oPhtae Speci, p

"Taxies and Inflatio,, 25i PP Rot 1979 pp. May 180 Oetele.Bradford ated P nont ~nee C., Mr

AD A081 194 limit Factor* Affecting tboeae Security:* hector..spp 233 Matit,,oe Owe Amcte La Socuroed fItcal 14

pp NY1 Dtttttiee 1fodloid, 'lopecteod Demanld hat the pp. Oct 100troll. Robert P. ati Vogel Robiet C T, 12 S iaN.Y to Same se An inttrumet of U S

Respon of Stole (3n-rmeni ffennptsto F oreign Policy Tinitng Abonut Plfutmel aed 0411 PP -9 (Ietif edPeeomino Fluctuatins end the All-tn if IfteT snenoneatlal hto,'30 pp" April IO8,Coutst, Cyclical Rovn.n. Sharing Giaents I? IPp AD A0M 00Den 1979) lfeoptitted ftomt the Review ofEconomtics and Statistict Vol I.( RnI N 3 Augnuet PP 2941119 I Keiot.* Nun,, and U Suemtla. "Th.

l"agaerr Ttamtfo-m' 119 pp May 198pp 2M "Th.e Greduate Scool .1 Mantagemnt. Ulniversity

Thlmmnn Jantevie S, Seapor Deptend.-itcm of Rochter and tile Cente hot Noval Anealpeeletter Sctt Cooperation Til. Cate.1n Sob Setata -e ra tduate School of Management. UnersityAfrica, 141pp, Jan 1980 ADOAOI 193 of Floteetat AD AO45 100

Wt, Kenethli G Tile S-eut novi in ii ie Roeitotat Richard 8. '8,eeepttnter Secuity InOpiate Wat' 42 pp Join 1680 10losveen.if Of the tott t" the Ittianr 0eae Ate." 26 pp. Jun

Sotalln Cloifert,o en Slan Stell.,e in 00.1o., 101 AD Aft?11137") AD A0182219

DATE

ILMED