Study with the several resources on Docsity
Earn points by helping other students or get them with a premium plan
Prepare for your exams
Study with the several resources on Docsity
Earn points to download
Earn points by helping other students or get them with a premium plan
The challenges in valuing consumer durables, specifically automobiles, using data from various sources. The author uses methods such as estimating annual expenditures and adjusting for price changes to determine the value of consumer assets. The document also compares different valuation procedures and discusses the limitations of various data sources.
Typology: Schemes and Mind Maps
1 / 53
This page cannot be seen from the preview
Don't miss anything!
This PDF is a selection from an out-of-print volume from the National Bureau of Economic Research
Volume Title: Studies in Income and Wealth
Volume Author/Editor: Conference on Research in Income and Wealth.
Volume Publisher: NBER
Volume ISBN: 0-870-14167-
Volume URL: http://www.nber.org/books/unkn50-
Publication Date: 1950
Chapter Title: Consumers' Tangible Assets
Chapter Author: Lenore Epstein
Chapter URL: http://www.nber.org/chapters/c
Chapter pages in book: (p. 409 - 460)
CONSUMERS'TANGIBLEASSETS 411
singlepersonslivingaloneoraslodgersandservantsinprivate householdsandinlodginghouses,hotels,etc.Theholdingsof lodgersandservantsareimportantchieflyinthecaseofmore orlesspersonalgoods,sincebydefinitiontheyownfewifany householdgoods.Forthesakeofbrevity,theciviliannonin- stitutionalpopulationwillbereferredtoasconsumers. Sincepersonslivingininstitutionsofvarioustypesandon militaryandnavalpostsarerelativelyfew,exceptduringwars, andtheirtangibleassetsarerathernegligibleontheaverage, theirinclusionorexclusionisofminorimportance.Itwould bedesirabletoexcludetheirassetswhenstudyingchangesin thevalueofconsumers'capitalinrelationtobusinesscycles, sincetheywouldnotchangewithfluctuationsineconomiccon- ditionsinthesamemannerastheholdingsoftheciviliannon- institutionalpopulation.Iftheyareexcluded,however,their valueshouldbeestimatedinsomemannerandpresentedasa separatecomponentinthenationalbalancesheet. Goodspurchasedandinthepossessionofconsumerswillbe treatedasownedbythemeventhoughtheymaynothavetitle becauseofoutstandingcreditobligations.Itwouldbeimprac- ticaltodootherwiseanditwouldbeundesirabletotryifthe valueofconsumergoodsistobecomparedwithoutstanding consumercredit.
b Consumers'tangiblea.ssets Consumers'tangibleassetsmaybedefinedasconsumerdurable goodsinaccordancewithSimonKuznets'classification,i.e., commoditiesthat,withoutmarkedchange,areordinarilyem- ployedintheirultimateuseforthreeyearsormore.Atthe otherextreme,consumers'tangibleassetsmaybedefinedto includenotonlydurablesandsemidurablesbutalsothefood 1Itwouldbeextremelydifficult,thoughnotimpossible,toobtaindirectlyfrom institutionalresidentstheinformationrequisitetoevaluatetheirassets.When estimatingnationalconsumptionfor1935-36,theNationalResourcesCommittee calculatedroughlythedisbursementsoftheinstitutionalagenciesthemselvesfor themaintenanceandcareofresidents,butdidnotincludeexpendituresbythe 2millionresidentsfromtheirpersonalfundsbecauseinformationwaslacking ontheallocationofsuchfunds.
412 PART II
onthepantryshelf,cleaningsupplies,toiletpreparations,the coalinthebin,thefueloilinthetank,etc. Theformerseemstoonarrowadefinitionforthepurpose athand.Semidurablegoods,whichMr.Kuznetsdefinesas ordinarilyemployedintheirultimateuseforsixmonthsto threeyears,compriseasubstantialtotalandshouldbeincluded inconsumers'wealth.Moreover,thedemarcationbetween durableandsemidurablegoods,sodefined,isoftentenuous. Perishablecommoditiesofthetypesillustratedaboveareeasy todistinguish.Thoughmanycanbestoredforlongperiods, theyaremoreorlesscompletelyconsumedwhenused. Wedefineconsumers'tangibleassetsasgoodsthatareordi-
months.Separateestimatesofthevalueofdurableandsemi- durablegoodsownedbyconsumerswouldbeusefulinstudy- ingchangesinassetsindifferentphasesofbusinesscycles,but theirsignificancewouldbelimitedbythedifficultyofclassi- fication.
paper.Othertangibleassetsofconsumersfallintosevenbroad groups:
Inonerespect,consumergoodsarelessdifficulttovalueat currentpricesthanareproducerdurablegoods;inanother,
shorterlifethanproducers'equipment,theproportionofthe totalconsumers'stockthatpassesthroughthemarketinayear exceedsthatofproducergoods,makingiteasiertodetermine arepresentativeprice.Abiggerdifficultyarisesfromthewide variationsinthespecificcharacteristicsofmanyconsumer goodsandcontinuousstylechanges.Thisdifficultyisavoided inpartiftheutilityandgeneralcharacteristicsofthecom- modityratherthanitsexactphysicalcharacteristicsarecon-
identicalconsumergoodsshouldbeused. Theuseofcurrentpricesofequivalentarticlesratherthan pricesatthetimeofpurchaseavoidsahybridvaluecomposite madeupofpricesdeterminedatmanydatesundervarying circumstances.Theproblemistoselectappropriatecurrent pricesforgoodspurchasedearlier.Forautomobiles,thesec- ondhandmarketissufficientlyactivetoyieldreasonableaver- agepricesforautomobilesofvariousages.3Thisistrueofonly afewotherconsumerdurables,however,certainlynotofcom- moditieswithausual'life'ofmerelyafewyears.Forgoods thatcannotbepricedinthesecondhandmarket,thepricesof equivalentnewcommodities,adjustedaswellaspossiblefor theconsumptionthathasalreadytakenplace,mustbesub- stituted.Secondhandmarketpricesarepreferablebecausethey takeintoaccountchangesinthecharacteristicsofthecom- modity. Shouldthevalueofconsumers'tangibleassetsrepresentthe valueinhand,sotospeak,ortheirreplacementvalue?Inother words,shouldsecondhandcommoditiesbevaluedattheprices theownerswouldreceiveiftheysoldthemoratpricesthe ownerswouldhavetopaytoreplacethem?Thereareargu- m.entsinfavorofbothconcepts.Wefavorreplacemen.tvalue. 3Mostsecondhandcarspassthroughdealers'hands.
CONSUMERS'TANGIBLEASSETS 415
d The (^) ofconsumergoods Onedistinguishingcharacteristicofconsumerdurabl.esisthat theirreplacementispostponable.4Asiswellknown,automo- bilepurchases,forexample,wereataverylowlevelduringthe
negligibly.Similarly,duringthewarautomobileregistrations remainedremarkablyclosetotheprewarleveldespitenearly threeyearswithoutautomobileproduction.5Acoatorasofa isseldomuseduntilitiscompletelywornout;itsusefullife, inotherwords,issocial,economic,orpsychologicalratherthan physical.Duringdepressionsoratthebeginningofadown- swing,goodscontinuetobeusedthatwouldbereplacedifper- sonswereoptimisticaboutincomeprospects.Manycommodi- tiesbecomeobsoletebecauseofstylechangeslongbeforethey arewornout.Ingoodtimes,therefore,thereplacementrateis high.True,goodsthatarereplacedremainpartofthenational stockiftheyaretradedinandboughtbyothers,butingeneral thescrappageraterisesinprosperity. Whatmeasureoflifethenismostappropriateforestimating thevalueofcommoditiesinconsumers'handsatanygiven date?Thereappeartobethreealternatives:theactualphysical life,asmeasuredintestinglaboratories;thenumberofyears duringwhichitiseconomicallyefficienttousesuchgoods,i.e., beforethecostofrepairsandmaintenancebecomeexcessive; orthecustomaryperiodofusebeforescrapping.Thethird,in partafunctionofthegenerallevelofincome,seemsthemost appropriateforthepurposeathand.Whentimesarebadthe totalstockofmostgoodsissmallerandtheiraverageagehigher thanwhentimesaregood,buttheirlifespanisalsolonger. Hence,theresidualusevalueofconsumers'tangibleassetsmay notactuallybesomuchlessinbadthaningoodtimesaswould appearifthelifespanofconsumergoodswereassumedtobe thesameatallstagesofbusinesscycles. 4SeeE.B.George,'ReplacementDemandforConsumers'DurableGoods', JournaloftheAmericanStatisticalAssociation,Vol.34,No.206,June1939, pp.239-60. (^5) Reductioninuseduetogasolinerationinghelpedtoprolongthephysicallife ofcars.
CONSUMERS'TANGIBLEASSETS 417
ifitwerenew.Again,therefore,anapproximationmustserve whenvaluingconsumergoods.
2 TypeofDataNeeded
Inessence,therequirementsforestimatingthevalueofcon- sumers'tangibleassetsarethree:aphysicalinventoryofeach good,i.e.,thenumberownedandtheiragedistribution;cur- rentpricesofnewgoods,and,forthoseforwhichthesecond- handmarketisreasonablyactive,currentpricesofgoodsof differentages;andthelifespanofeachgood.Thetotalvalue isthenthe oftheproductsofthenumberofcommodities ofvariousagesandtheappropriateprices.Obviously,itis impracticabletomakeacomprehensivephysicalinventoryof consumergoods,thentopriceallitems,particularlythosewith arelativelyshortusefullife.Toevaluatesuchgoods,informa- tiononexpendituresduringanappropriatenumberofyears
physicalinventoryanditempricedata. Sinceanyestimateoflifespanisanaverage,therearealways oldercommoditiesinuse.Thescrapvalueofsuch'over-age'
sumers'tangibleassets.Whileitmaybedifficulttoestimate scrapvalue,itispracticabletoincludeitforgoodsinvençoried andpriced(assumingotherpricesarenotobtainedfor'over- age'goods),butnotforgoodsevaluatedbycumulatingexpend- itures. Inventorydatashouldbeobtaineddirectlyfromfamilies andsinglepersons,thegroupwhosecapitalwewishtomeasure. Expendituredatamaybeobtainedeitherforconsumersorfor thenationasawhole,andforthebusiness,governmental,and institutionalsectorsoftheeconomy.Itseemslogicaltoadopt thedirectapproachwheneverpossible;iftheothermethodis adopted,theconsumers'shareisobtainedbysubtractingthe expendituresofthesethreesectorsfromthetotalforthenation. Pricedatamaybecollectedfromretailoutletsorfromcon- sumers.Thereareargumentsinfavorofeachmethod.
sumers'reportsontheageofgoodsownedandthefrequency ofreplacement.InSectionCthemethodsofobtainingthedata neededforevaluatingconsumers'tangibleassetsarediscussed. Thesesummary.commentsaredesignedtoindicatetheview- pointfromwhichpreviousestimatesandpresentlyavailable dataareappraised. BEVALUATION
KennethE.Bouldingrecentlycontendedthatheknew"ofno attempttoestimatethetotalvalueofconsumers'capital". Thiswasofcourseanexaggeration,sinceestimatesofthevalue ofconsumergoodsasacomponentofnationalwealthhave beenmadeforvariousdatesbackto1850.Ingeneral,however, lesseffortappearstohavegoneintoitsestimationthaninto theestimationofmanyothercomponents,perhapsbecauseof theobviousinadequaciesofthedata. Broadlyspeaking,datausedtoevaluateconsumers'tangible assetshavebeenofthreetypes:informationontheflowofcon- sumercommoditieswithassumedperiodsofusefulnessand depreciationrates;questionnairesonthevalueofhousehold goods,apparel,etc.;andinthecaseofpassengerautomobiles, informationonthenumberofcarsinoperation,prices, depreciationrates.Theoretically,personalpropertytaxre- turnsareafourthsourceofinformation,buttheyhavenever beenusedasabasisofestimation,presumablybecauseonly somestateslevysuchtaxesandbecausemethodsofappraisal varywidely.
1 MethodsandLimitationsofPreviousEstimates forGoodsOtherthanAutomobiles Sincetheearlyestimatesareroughguessesatbest,'°tocon- sidertheirderivationwouldcontributelittletoproblemsof 8'TheConsumption Concept in Economic Theory',AmericanEconomicRe- view,XXXV,2,May1945,p.8. 9Foradiscussionofthegeneralproblemofusingtaxationdatatoevaluate wealth,seeKuznets,StudiesinIncomeandWealth,Vol.Two,pp. 10W.I.King,WealthandIncomeofthePeopleoftheUnitedStates(Macmillan, 1915),p. 47.
420 PART II
contentsofhousesoutsidelargecities.Totheaveragevalueof insuredfurniturewasaddedanaverageamountestimatedto representthevalueofcarriagesandmechanics'tools.'2For 1900,1904,and1912theCensusBureaureliedondataon expendituresasmeasuredbythevalueofmanufacturesand importsadjustedtoallowforthecostofdistributionandon assumptionsconcerningtheaveragelifeofsuchgoods.' Inessence,thelatterhasremainedthetechniqueexceptin the1922Census.Whiledataonconsumers'expenditureshave beenvastlyimproved,informationonthelifeofconsumer goodsisstillfragmentaryandrough.For1922theCensus Bureaubuiltupanestimateofthevalueofconsumers'tangi- bleassets(otherthanautomobiles,whichwerelistedseparately forthefirsttime)inlargepartfromanswerstoquestionnaires senttoasampleofindividualsthroughouttheUnitedStates. The1922surveywasthefirstandlastofitskind,butthe resultsservedasabasepointinthreeseriesofestimatesofthe nationalwealthpublishedsubsequently.Theconsumersam- pleapproachseemshighlydesirableifinterestattachestoa separatevaluationof.consumers'holdings.Itisnotclear,how- ever,whethertheCensusBureauchosethemethodforthat reason,sinceitaddedtotheexpandedsampleestimatesthe estimatedvalueofhorse-drawnvehiclesandofbooksinpublic libraries,orwhetheritconsideredtheholdingsofconsumer goodsbybusiness,government,andinsjitutionsofnegligible value. TheCensusestimatefor1922,asidefromanybiasduetothe samplingdesignandthetechniqueforprojectingthesample returns(about37,000returnsoutof100,000)toanational basis,hadtwoseriouslimitations.Asnotedabove,amailques- tionnairewasused;itaskedforthe"totalfairvalueofhouse-
(^12) 11thCensus,1890:ReportonWealth,DebtandTaxation,PartII,Valuation andTaxation(1895),p.8. (^13) Wealth,DebtandTaxation,1913(1915),pp.19-20. (^14) Wealth,PublicDebtandTaxation:1922,EstimatedNationalWealth(1924), pp.15-6.
CONSUMERS'TANGIBLEASSETS 421
kitchenandbedroomfurnishings,trunks,clothing,watches, jewelry,etc.".Experiencewithfieldsurveysoffamilyincomes andexpenditureshasclearlydemonstratedthatmostpersons cannotevenroughlyestimatethetotalexpenseduringayear foraparticularbroadcategoryofgoodsandmanywillnot attempttodoso,althoughtheyusuallycanandwillgive(in responsetospecificquestions)thedetailsnecessarytobuildup suchatotal.Obviously,itwouldbemuchmoredifficultfor theaveragepersontoestimatethevalueofalltheclothinghe owns,forexample,thantoestimatetheamounthespenton clothingduringtheprecedingyear.Eveniftherespondent werewillingandabletomakeanover-allestimate,however, therewouldbenoassurancethathehadincludedallthecom- moditiesassumedtobecoveredandnoindicationofhowhe valuedthem. Ifsubsequentsurveyswereconductedbypersonalinterview ratherthanmailquestionnaire,andaverydetailedschedule used,theinclusionofallgoodsmightbeassured.Thiswould not,however,solvetheproblemofvaluationthatresultsfrom
values.16Regardlessoftheinterviewer'sinstructions,some respondentswouldprobablyreportoriginalprice,othersthe originalpriceadjustedbysomeverypersonalnotionofde- preciation,andstillothersthecurrentpricetheywouldhave topaytoreplacethegoods.Manywouldprobablybeunable togiveanyestimateofthevalueofcertaincommodities.At best,ahybridvalueaggregatewouldbetheresult,withcon- sequentambiguityconcerningitsmeaning.
b Twoprojectionsofthe1922estimate TheNationalIndustrialConferenceBoardbuiltupaseries ofestimatesof"stocksofgoodsinthehandsofconsumers"for (^15) TheFederalTradeCommission,inNationalWealthandIncome,published in1926,offerednorevisionoftheCensusestimatesfor1922.However,itsug. gested(p.49)thatthefigureonchattelswasprobablytooclosetocosts,i.e., depreciationwasnotfullyallowedfor.
CONSUMERS'TANGIBLEASSETS 423
"meaningfurniture,musicalinstruments,clothing,jewelry, privatelibrariesandpersonaleffectsgenerally",hemaintained thattherecouldbenorealaccounting.Henotedthatatthe sameratepercapitaasreportedbytheCensusfor1922,their aggregatevaluewouldhavebeenabout.$44billionin1929. Ontheassumptionthatthepercapitavaluehadincreased— "thepeoplehadmorenewthings,radios,electricrefrigerators, etc."—heguessedthatthevaluein1929was$48billion,point- ingoutasconfirmationthattheChamberofCommerceesti- mated$49.6billionfor1930.
d Doane'stwoestimates InhisMeasurementofAmencanWealth,R.R.Doanepre- sentedaseriesofestimatesofthevalueofchattels(including automobiles),fordecennialdates1860-1900,for1904,andan- nually1909-32,withoutexplaininghismethodbutmerely acknowledgingdatafromtheBureauoftheCensus,theCen- susofManufactures,estimatesoftheUnitedStatesChamber ofCommerce,andvarioustradejournalstudies.'9Hisfigure for1921isidenticalwiththe1922Censusestimateofthevalue ofconsumers'tangibleassetsexclusiveofautomobiles;his estimateisnearly$3billionsmallerthantheCensustotal.
Doanebasedhisestimatesofthevalueofchattelgoodsother than inth.epossessionofindividualsin1930and 1938onthe1922Censusfigure.2°Heassumedthattheper capitavalueofclothingwasthesamein1930asmn1922and thatthevalueofhousefurnishingsincreasedinthesamepro- portionasthatofnewresidentialbuilding.Thisyieldedan estimateof$59billionforchattelgoodsin1930.For Doaneusedadifferentmethod,"duetotheunprecedented situationexistingbetween1930 1938".Firsthedivided the1922Censusaggregateforchattelgoodsbetweenclothing andpersonalarticles,ontheonehand,andfurniture,etc.,on theother,inaccordancewiththeCensusallocationfor1912. (^19) (Harper,1983),pp.10-1,208. (^20) (Harper,1940),pp.238and256-7.
424 PART II
Thenheappliedindexesofchangesinannualconsumers'pur- chasesofthesegoods(extrapolatingNBERdata)totheesti- matedpercapitaholdingsofeachclassofsuchgoodsin1922. Frompercapitaholdings,soestimated,hearrivedat$39. billionforchattelgoodsotherthanautomobilesin1938.This procedureyieldedanestimatefor1930of$54.6billioninclud- ingthevalueofpassengercars;theestimatehepresentedas hisbestapproximation,excludingcars,was$59billion.
eNRPBestimatefor TheNationalResourcesPlanningBoardestimatedthevalue ofpersonalpropertyasacomponentofnationalwealthin bycumulatingexpendituresforaperiodofyears.21Itsesti.
Kuznets'CommodityFlowandCapitalFormation,withas- sumedratesofdepreciation,andadjustedforpricemovements. Inthecaseofconsumerdurablegoodsotherthanautomobiles, aconstantannualrateofdepreciation,10percent,wasassumed. Thus,thevaluein1935wasestimatedbycomputing10percent ofthe1926value,20percentofthe1927value,etc.,adjusting thesevaluesto1935pricesandaddingtheadjustedvalues, yielding$20.8billionfor1935in1935prices.Forconsumer semidurablegoods(exclusiveofdrygoodsandnotionswhich Kuznetsincludes),cumulativesurvivalrateswereassumed: 30percentofvaluelost(thereportsays"remains")afterthe firstyear,50,70,90,and100percentafterthesecond,third, fourth,andfifthyears.Thisyielded$17.1billion,oranesti- mateof$37.9billionforconsumergoodsotherthanauto- mobiles.ThedifferencebetweenitandtheNICBestimate, $30.8billion,reflectsinpartthefactthat,bythenatureofthe sourcematerial,theNRPBestimateincludesholdingsofcon- sumergoodsbybusiness,government,andprivateinstitutions, aswellashouseholds,whiletheNICBfigures,extrapolated fromtheCensusestimate,morenearlyrepresenttheholdings ofconsumersasdefinedinthispaper.Theinclusionofreplace- (^21) TheStructureoftheAmericanEconomy(NationalResourcesCommittee, 1939),PartI,BasicCharacteristics,p.376.
"I
426 PART II
tothesamelimitationasKeller'sfigures.Atthesametime, becauseofthesharpincreasesinpricesafter1941,thisserved
periodinventoriesthatwouldotherwisehaveresultedfrom themethodofcomputingthebaseperiodinventory.Limita- tionsduetotheuseofthesamplesurveydataarediscussed below.
2 MethodsofPreviousEstimatesforAutomobiles Thevalueofmotorvehiclesasaseparatecategoryofnational wealthwasfirstpresentedin1922.Noattemptwasmadeto valueseparatelypassengerautomobilesownedbyconsumers forpersonaluse.However,inthisasinmostsubsequentesti- matesofthevalueofmotorvehicles,thegeneralmethodology isofinterestbecauseitapproximateswhatweproposefor evaluatingconsumerdurablegoods. TheCensusbaseditsestimatefor1922onreportedoutput duringaperiodcoveringtheestimatedaveragelifeofauto- mobiles,trucks,motorcycles,andtrailers,andonstatisticsof registration,withallowancefordepreciation,valuedinterms ofpricesprevailingin1922.24Forpassengercars,forexample, theaveragelifewasassumedtobesevenyears;hencecarsthat wentintoservicein1916wereregardedasoutofuseatthe closeof1922,havinglostone-seventhoftheirvalueeachyear. Thosethatwentintousein1915wereassignedajunkvalue of$25.Onthisbasis,thenumberofcarsavailableforuseon December31,1922wascalculatedtobeequivalenttoslightly morethan5millionnewcars.Theaveragepricefornewcars in1922wasestimatedtobe$770,yieldingatotalvalueof $3,942,026,000forpassengerautomobiles.Forallmotorvehi- cles,theestimatedvaluewas$4,567,407,000. IntheNICBestimatesofwealthfor1922thevalueofmotor vehiclesisaseparatecomponent,butpassengercarsarecom- binedwithothervehicles.25Themethodisnotoutlinedin
currentpricewithallowancefordepreciationwasused:"From 24EstimatedNationalWealth,p.11. 25Op.cit.,pp.120and130.
CONSUMERS'TANGIBLEASSETS 427 registrationfiguresthenumberofvehiclesinusewasobtained; fromproductionandmortalityfigures,thenumberofeach ageineachyear;fromtheoriginalpriceofeachyear'sproduct andtheamountofdepreciation,thevalueofcarsofeachagein eachyearwasdetermined,andthesumofthelattergavethe totalvalueofallcarsinuse."For1922theNICBestimateis slightlyhigherthantheCensus.
valueofpassengercars,whichhetreatsasaseparatecategory, for ApparentlyverysimilartothatoftheNICB,
PetroleumInstitutefiguresonthepercentagesofcarssurviv- ingeachyearKellerestimatedthenumberofcarssurviving inanygivenyear.Heapplieditsannualdepreciationratesto theaverageretailpriceforeachyeartodeterminetheaverage valueinanygivenyearofcarssoldinprecedingyears.Toget theaverageretailvalueofnewcarsineachyear,heapplieda standard33.3percentmarkuptotheaveragewholesalevalue, calculatedbydividingthevalueofproductminusthevalueof exportsbythetotalnumberoffactorysalesminusexports. Thesecomputationsalsoyieldedanaggregateslightlyhigher thantheCensusestimateforpassengerautomobilesfor1922. Ingallsassumedanaveragepriceof$200perregistrationat theendof1929—26,500,000,ofwhichabout3,500,000were trucks—andsoderived$5,300millionwhich,hecomments, "isnotunreasonable".27Hecitesinconfirmationanestimate of$5,461millionforallmotorvehiclesin1930preparedbya committeeoftheAmericanAutomobileAssociation. Doane,inhisfirststudy,didnotestimatethevalueofpas- -sengercarsorofallmotorvehiclesseparately.Inthesecond, hederivedatotalfor1930byapplyinganaveragevalueof$ percar(derivationnotexplained)tothenumberofcarsregis- tered;for1938hetookoveranestimateofaggregatevalue preparedbytheAutomobileManufacturersAssociation.
26Op.cit.,p.135. 27Op.cit.,p.668. 28TheAnatomyofAmericanWealth,pp.235and256.
