EHES WORKING PAPERS IN ECONOMIC HISTORY …England’s lender of last resort responsibilities which...

European Historical Economics Society

EHES WORKING PAPERS IN ECONOMIC HISTORY | NO. 117

The Bank of England as Lender of Last Resort: New historical evidence from daily transactional data

Mike Anson

David Bholat

Miao Kang

Ryland Thomas

Bank of England

NOVEMBER 2017

EHES Working Paper | No. 117 |November 2017

The Bank of England as Lender of Last Resort:

New historical evidence from daily transactional data

Mike Anson,(1) David Bholat,(2) Miao Kang(3) and Ryland Thomas(4)

Abstract We use daily transactional ledger data from the Bank of England’s Archive to test whether and to what extent the Bank of England during the mid-nineteenth century adhered to Walter Bagehot’s rule that a central bank in a financial crisis should lend cash freely at a high interest rate in exchange for ‘good’ securities. The archival data we use provides granular, loan-level insight on the price and quantity of credit, and information on its distribution to particular counterparties. We find that the Bank’s behaviour during this period broadly conforms to Bagehot’s rule, though with variation across the crises of 1847, 1857 and 1866. Using a new, higher frequency series on the Bank’s balance sheet, we find that the Bank did lend freely, with the number of discounts and advances increasing during crises. These loans were typically granted at a rate above pre-crisis levels and, in 1857 and 1866, typically at a spread above Bank Rate, though we also find some instances in the daily discount ledgers where individual loans were made below Bank rate in 1847. Another set of customer ledgers shows that the securities the Bank purchased were debts owed by a geographically and industrially diverse set of debtors. And using new data on the Bank’s income and dividends, we find the Bank and its shareholders profited from lender of last resort operations. We conclude our paper by relating our findings to contemporary debates including those regarding the provision of emergency liquidity to shadow banks. JEL classification: E58, G01, G18, G20, H12, N2, N4, N8

Keywords: Bank of England, lender of last resort, financial crises, financial history, central banking

(1) Bank of England. Email: [email protected] (2) Bank of England. Email: [email protected] (3) Bank of England. Email: [email protected] (4) Bank of England. Email: [email protected] The views expressed in this paper are those of the authors, and not necessarily those of the Bank of England or its committees. This paper is dedicated to Philip Cottrell, in memoriam. We are grateful to James Barker, Nadia Denton, Matt Everitt, Zaina Haider, and Jay Mehta for their help transcribing ledger data; Rachael Muir, Margherita Orlando and Ben White for their help with archival sources; and Shahid Nazir and staff in the Bank’s Information Centre for their help pinpointing secondary sources. We are grateful to Paul Robinson, Kilian Rieder and Rhiannon Sowerbutts for comments on the paper. We also thank Andy Haldane, Rob Elder, Sujit Kapadia, James Talbot and the Bank’s Research Steering Committee for enabling us to pursue this research. We are especially grateful to Eugene White for giving us a guided tour of what was available in the Bank’s ledgers which provided the inspiration for this paper. © Bank of England 2017

Notice The material presented in the EHES Working Paper Series is property of the author(s) and should be quoted as such.

The views expressed in this Paper are those of the author(s) and do not necessarily represent the views of the EHES or its members

mailto:[email protected]




2

Section1:Introduction

I Motivationforresearch

During the financial crisis of 2007-09, policymakers looked to the past for guidance.

According to formerBankof EnglandGovernor LordMervynKing (2016: 90), “During the

crisis, I found that thestudyofearlierperiodswasmore illuminating thananyamountof

econometricmodelling.” Inparticulartherecentcrisisbearsstrikingresemblancetocrises

thatoccurredintheUKduringthenineteenthcentury.1Thekeyinstitutionsattheheartof

thosecriseswerediscounthouses.These institutionsweresonamedbecause,whenthey

bought securities, they did not pay their full face value but instead purchased them at a

discount,equivalenttocharginganinterestrate.Likemodernmoneymarketmutualfunds

(MMMFs)orbanks’offbalancesheetspecialpurposeinvestmentvehicles(SPVs)thatlayat

the heart of the 2007-09 crisis (Brunnermeier 2009), nineteenth century discount houses

issued short-termdebt to fund a portfolio of financial assets. Thesedebtswereprimarily

held at that time by banks and other institutional investors. Financial crises in the

nineteenth century therefore occurred when these wholesale institutions demanded en

masseandatshortnoticethatthediscounthouseshonourtheirobligationstopayoutcash.

In order to meet this demand, the discount houses would either have to liquidate their

assets at fire-sale prices or default; a similar dilemma faced by MMFs and SPVs when

wholesaleinvestorswithdrewfundingforreposandasset-backedcommercialpaperduring

therecentfinancialcrisis(GortonandMetrick2012;GortonandOrdonez2014).2

Then, as now, risks to the financial system and the broader economy arguably justified

intervention. During themid-nineteenth century, the Bank of England started to act as a

‘lenderof last resort’, buying assets fromdiscounthouses in exchange for eitherBankof

Englandnotesordeposits at a timewhenothersmightnot lendbecauseofmarket-wide

uncertainty. These operations parallel those, such as quantitative easing (QE), which the

Bankundertookintherecentcrisis.Indeed,perLordKing,theyprovidedamodelforthem.

Theseearlyoperationsthereforewarrantcarefulanalysisbecauseofthelessonswemight

learnfromstudyingthem.

1QuinnandRoberds(2015)similarlyargueforaparallelwiththe1763crisisinnorthernEurope.2Salesandrepurchaseagreements(repos)are likedemanddeposits inthatthecash lender ‘deposits’ fundswith another institution, secured by collateral, with the promise to receive the funds back with interest,usually thenextdayorwithinayear. If thesetransactionsarenotrolledover, this isequivalenttothecashlender running on the institution (Gorton 2012). Asset-backed commercial paper refers to short-term debtissued by banks’ off-balance sheet special purpose investment vehicles and conduits. In the lead up to thefinancialcrisisof2007-09,thisshort-termcommercialpaperoftenfinancedthepurchaseofmortgage-backedsecuritiesfromtheconduits’sponsoringbank(Shin2009).

3

II Keyresearchquestionsandfindings

In thispaperweanalyse theBankofEngland’s lenderof last resortoperationsduring the

1847,1857and1866crises.ThosecrisesshapedWalterBagehot’sconceptionoftheBankof

England’slenderoflastresortresponsibilitieswhichhesetoutinhisseminalbookLombard

Street published in 1873. Bagehot’s conception in turn has provided a template for

managing financial crises followed by central banks around the world ever since (Sayers

1957;Fetter1965).Inparticular,Bagehotarguedthatacentralbankduringacrisisoughtto

(i) lend cash freely (ii) at ahigh/penalty rate (iii) inexchange for ‘good’ securities.3 These

threeaspectsofthe‘Bagehotrule’provideausefulorganisingframeworkandbenchmarkin

ourpaper forassessinghowand towhatextent theBankactedasa lenderof last resort

duringthemid-nineteenthcentury.

To briefly sum up, we find that the Bank did lend cash freely in that the volume of its

discountsincreasedduringcrisesrelativetotheperiodimmediatelybefore.However,these

loansweremadeunevenlytoafewkeycounterparties–typicallythetopfifthofborrowers

received over three-fourths of the amounts lent. On the issue of high or penal rates of

interest, Bank Rate increased during all three crises relative to the rate that prevailed

before. During the 1857 and 1866 crises, Bank Rate was also typically above

contemporaneousmarketrates.However,in1847,BankRateremainedbelowmarketrate.

Infact,wefindsometransactionswheretheactualrateatwhichtheBanklentwasbelow

Bankrate.Sofarasweareaware,oursisthefirstpapertodocumentthatBankRatewas

not an absoluteminimum ratebelowwhich theBankwouldnot lend (cf. Scammel 1968:

176). Coupled with evidence showing an uptick in loan application rejection rates, this

implies the Bank rationed credit in 1847. Finally, there is the issue of ‘good security.’

Perhapsthemostcompellingpieceofevidencethatthesecuritiesweregoodisthatwrite-

offsandarrearsremainedrelativelylowduringandimmediatelyaftercrises,andtheBank’s

profitsanddividendpaymentstoshareholders increasedasaresultof its lendingmoreat

higherrates.Whilethe‘goodness’ofsecuritieswasinpartdeterminedbyrulespertaining

totheirtenor,theBankalsoexerciseddiscretion,asthenamesoftheultimatedebtor,plus

all those including the discounter who had endorsed them, mattered. We find some

3 These rules have prompted an enormous debate in the literature aboutwhether Bagehot actually statedtheserulesexplicitly,whathemeantbythemifhedid,andwhethertheyarenecessaryandsufficient foralenderoflastresorttofollow(e.g.Martin2009;Bignonetal.2012).Infact,thispithytriadicformulationisnotBagehot’s but a conventional summation of Bagehot’s conception by later commentators. Here is whatBagehot (1873) actuallywrote: "Theory suggests, and experience proves, that in a panic the holders of theultimateBankreserve...shouldlendtoallthatbringgoodsecuritiesquickly,freelyandreadily."Alittlelateronhewrites:“The[goalofsuchlending]istostaythepanic…Andforthispurposetherearetworules:First.Thatthese loans should only be made at a very high rate of interest. This will operate as a heavy fine onunreasonabletimidity,andwillpreventthegreatestnumberofapplicationsbypersonswhodonotrequireit.The rate shouldbe raisedearly in thepanic, so that the finemaybepaidearly...Secondly. Thatat this ratetheseadvancesshouldbemadeonallgoodbankingsecurities,andaslargelyasthepublicasksforthem.Thereasonisplain.Theobjectistostayalarm,andnothingthereforeshouldbedonetocausealarm.Butthewaytocausealarmistorefusesomeonewhohasgoodsecuritytooffer."

4

evidencethattheBank’sdefinitionofa‘good’counterpartywasmuchmoregeographically

andindustriallyexpansivethanmosthistorianshavedescribed.

III Relatedliteratureanddata

FormerBankofEnglandDeputyGovernorSirPaulTucker(2014)recentlyopinedthat“the

relativeneglectofLOLR[lenderoflastresort]inthecoreliteratureoncentralbankingover

thepasttwentyyearsisatragedy.”Indeedmanyofthebestcontributionstotheliterature

remain thosepublishedduring the firsthalfof the twentiethcentury (Hawtrey1932;King

1936; Sayers 1936; Clapham1944).4Wehave found these sources invaluable in our own

research.However,thesehistoriestendtobeheavyonnarrativebutlightondata.Usually,

whenspecificfiguresabouttheBank’slenderoflastresortoperationsareprovidedinthese

sources, they are partial, detailing a few transactions. They do not give a holistic view.

Moreover,thedetaileddata,totheextentgiven,comemostlyfromsecondarysources.5

Ourcontribution to the literature is inexploitingprimaryarchivaldata,especially the rich

transactionalinformationavailableintheBank’sdailydiscountledgersfrom1847,1857and

1866. These ledgers include the names of all counterparties that approached the Bank’s

discountwindowduringcrises; informationonwhethertheir requests forassistancewere

acceptedorrejected;andthevalueandvolumeofassetstheyexchangedforcash,aswell

astheprice(interestrate)atwhichtheydidso.Toanalysethesedata,wetranscribedthese

ledgers into Excel files from digital images of the originals.6We are publishing the Excel

workbookcontainingthesetransactionsalongsidethispaper.7

4 Thornton (1802) also made a seminal early contribution. See Congdon (2009) for an excellent andprovocativecontributionwhich looksattherolefromamodernperspective.Goodsurveysofthekey issuesareavolumeeditedbyGoodhartandIlling(2002)andapaperbyGrossmanandRockoff(2015).5Forexample,King(1936:145),inhisanalysisofthe1847crisis,notesthatthemostprominentmoneymarketinstitution at that time, Overend Gurney, sold £80,000 worth of securities at a 9 percent discount in lateOctober. His source is a Bank official’s testimony before Parliament ten years after the 1847 crisis. Theproblemwithrelyingoninterviewdatatoestablishfinancialfacts,especiallygiventhelonglag,isthatthereisnoindependentwaytocorroboratetheclaims.Inthiscase,aninspectionoftheBank’sdailydiscountledgerson 19 October 1847 shows that, while the Bank did indeed discount securities fromOverend Gurney at 9percent, their face valuewas actually £68,460. The fact that therewere 80 securities probably caused theconfusion,leadingtotheerroneousreportingoftheirvalueas£80,000.6TheBank’sArchivehasdigitisedtheseledgersfrom1847toearly1919.TheArchive’sC28seriesrunsfrom1702to1965,thoughthereisa largegapbetween1710and1846wherethematerialhasnotsurvived.Theformatoftherecordschangedin1965.Thesehavenotsurvived.7The1847,1857and1866criseshaddifferentdurations, thereforethenumberofobservationstranscribedandanalysedvaries fromyear toyear. For1847we transcribedandanalysed thewhole calendaryear (310business days) amounting to 9,209 transactions given that therewere two phases of the crisis in April andOctober1847.For1857and1866wetranscribedandanalysedfourmonthsofdataaroundthepeakofeachcrisis:fromSeptembertoDecember(103businessdays)forthe1857crisis(3,004transactions);andfor1866,ourdatasetcontains2,842transactionsrunningfromMarchthroughJune(104businessdays).Inadditiontotherawdata,theExcelworkbookcontainsvariousmetricswehaveconstructedonadailyandmonthlybasis.These includetheproportionofdebtsacceptedandrejectedbytheBank fordiscount; thenumberof loansmade that day; the number of unique counterparties who came to the Bank to discount debt; themean,

5

Figure1.1:ExampleofapagefromtheBankofEngland’sdailydiscountledgersThispageshowstheBank’sdiscountactivitieson23October1847.Fromlefttorightthecolumnheadingsare:numberofbillsbroughtinfordiscount;thediscountrate;thenameofthediscounter;themonetaryvalueofbillsbrought;themonetaryvalueofbillsrejected;andthetotalnumberofbillsbroughtinrejected.

FewresearchershavemadeuseoftheBank’sdaily ledgersbefore,thoughwearenotthe

firsttodoso.Inparticular,ourpaperwasinspiredbyexceptionalresearchpreviouslydone

by Tessa Ogden (1988), and Vincent Bignon, Marc Flandreau and Stefano Ugolini (in a

numberofpapers).8 In someways,Ogden’sPhDthesiscomesclosest toourownproject.

Hersistheonlypriorresearchweknowofwhichproducedandanalysedanextendedseries

ofBankdiscountdata.However,ourresearchdiffersfromOgden’sthesisandmoreclosely

follows thatofBignon,FlandreauandUgolini in twoways. First, thedataweuse ismore

medianandstandarddeviationofthevolumeandmonetaryvalueofdebtbroughtinfordiscount;themean,medianandstandarddeviationofthediscountratescharged;themean,medianandstandarddeviationoftheamountslenttocounterparties.Wehavealsotranscribeddatafromafourthcrisis(1914)notdiscussedinthispaper.Weareplanningafutureresearchprojectexploitingthesedata.Weinviteexpressionsof interestviaemailfromotherscholarsinterestedinworkingonthisproject.8White(2016)andBignonandJobst(2017)havealsorecentlydoneresearchthatparallelsourown.

6

granular.Weanalysedaily loan-by-loandata.By contrast,Ogden lookedatweekly totals.

Theadditional insightweprovide fromhavinganalysed the loan-leveldata is thatweare

able to drawmoredetailed conclusions about theprofile of theBank’s customers during

crises.Second,ourpaperdiffersfromOgden’sthesisinitstimeframe.Wefocusonthemid-

nineteenth century, while Ogden focused on the period from 1870 to 1914.While both

periodsareworthstudying,oursisarguablymorerelevanttounderstandinghowtheBank

of England’s lender of last resort function first developed. As Ogden (1991: 309) herself

wrote in another publication, “the established view in the literature is that the Bank of

England accepted its role as lender of last resort sometime around 1870.” Recall, for

example,thatBagehotpublishedLombardStreetin1873.Bagehot’sargumentwasnotthat

theBankofEnglandshouldstartactingasalenderoflastresort.Rather,heclaimedthatthe

Bankwasalreadyacting thiswaybuthadnotpubliclyandpermanentlyacknowledged its

lenderof lastresortrole.BagehotadvocatedthattheBankexplicitlyacknowledgeitstacit

function,inordertolessenthelikelihoodandseverityoffinancialcrises.ForBagehot(1873:

31),therewasnothing‘constructive’inanyambiguity:

“thoughtheBankofEnglandcertainlydomakegreatadvancesintimesofpanic,yetastheydonotdosoonanydistinctprinciple…in1847,evenin1866…therewas nevertheless an instant when it was believed that the Bank would notadvance…Tolendagreatdeal,andyetnotgivethepublicconfidencethatyouwilllendsufficientlyandeffectually,istheworstofallpolicies.”

WhileBagehotassertedthattheBankmade“greatadvancesintimesofpanic,”hedidnot

empirically evidence his assertion. So by looking at the Bank’s transactional data before

1870, we are able to fact check whether Bagehot was correct in his assessment of the

Bank’sbehaviour,andanalysetheextenttowhichtheBankhadaccepteditsroleasalender

oflastresortbeforeBagehotwrotehisbook(O’Brien2003;c.f.Wood2003).

SinceOgden,therichestempiricalresearchonthehistoryoftheBank’slenderoflastresort

operationshasbeendonebyBignon,FlandreauandUgolini.Inaseriesofimportantpapers,

they examine samples of the Bank’s daily ledgers (Flandreau and Ugolini 2011; Bignon,

Flandreau and Ugolini 2012; Flandreau and Ugolini 2014). Some important findings from

their research include quantifying the predominance of non-bank recipients of Bank of

England loans;theskeweddistributionofthese loans,witha fewcounterpartiesreceiving

the bulk; and the centrality of foreign securities in these exchanges, reflecting British

imperialtradepatterns.Ourresearchfindingsfurthersupporttotheirconclusions.

Atthesametime,weextendtheirresearchinacoupleofdirections.First,thedailydiscount

data they analyse pertains to the crisis of 1866, namely one month (May 1866) when

OverendGurneydefaulted.Thatdefaultsparkedawider financialcrisis inmuchthesame

waythecollapseofLehmanBrothersdidin2008.TotheextentthatBignon,Flandreauand

Ugolinianalyseothercrises,theydosobylookingatannualfigures.Sowefollowuptheir

7

analysisbyusingdailydata, including lookingat the1847and1857 crises inmoredetail.

Dataonthesecrisesaddinsightbecause,ratherthanviewingthe1866crisisinisolation,it

mightbebetterconceptualisedastheconclusionofanearlytwentyyearjourneyoverthe

course ofwhich the British financial system and the Bank of Englandmatured, assuming

many features thatendure to this today (Kuttner2010). Indeedthedetailsofeachof the

financialcrisesof1847,1857and1866donotconcernusinthispaper.9Instead,whatwe

care to stressare the similarities in the leadup to them,and the leadup to the2007-09

financialcrisis.Forexample,asintheearly2000s,itwasbelievedbymanyinthenineteenth

century that monetary and price stability would also guarantee financial stability

(Constancio2015).However,asin2007,thisbeliefprovedmistaken.

Besidesaddingmoredailydiscountobservationsintothemix,ourpaperalsoexploitsnew,

higher frequency data on the Bank’s balance sheet. The utility of these data lie in their

helpingustoidentifyhowtheBank’sbalancesheetchangedduringthecriseswestudy,in

particular, how thesemoments differed from ‘normal’ periods (cf. Ferguson, Schaab and

Schularick2015).10Wehavealsoconstructedanewseriesshowing theBank’sprofitsand

dividendpaymentstoshareholders.Thesefinancialstatementdataarealsocontainedinthe

Excelworkbookwearepublishingalongsidethispaper.

IV Outlineofthepaper

The restof thispaper isorganised into five sections,with further splitswithin them,plus

several annexes at the end. The next section gives institutional detail on how the Bank

executedlenderoflastresortoperationsinthenineteenthcentury.Thosefamiliarwiththe

Bank’shistorymaywishtoskipaheadtothefollowingsectionwhichevaluatestheevidence

astowhethertheBank’sDiscountOfficelentfreelyduringfinancialcrises.Thisisfollowed

by sections assessing whether the Bank lent at high interest rates, and purchased good

securities,respectively.Thefinalsectionbringsourfindingstobearoncontemporaryissues.

We explain how our research contributes new insights to, and to some extent recasts,

longstanding and contentious academic and policy debates about whether central banks

should lend only to illiquid, but not insolvent, institutions; about the relevance ofmoral

hazard, specifically,whethercentralbank lending should support individual institutionsor

the financial system as a whole; and the appropriate institutional perimeter, if any, for

lenderoflastresortoperations.

9ThoughwerecommendrecentblogpostsbysomeofourBankcolleagueswhichgivebriefoverviewsofthosecrisesandrefertoanumberofdetailedaccountsofthem(HuangandThomas2016a;Neumann2016;Lewis2016).ForageneraloverviewofBritishbankinginthenineteenthcenturywerecommendCollins(2012).Onthe1847crisis,seeEvans(1848).Onthe1857crisis,seeHughes(1956).On1866,seeChubb(1872).10Theweeklybalancesheetserieswehaveconstructedstartsin1844andextendstothepresent.Dailybalancesheetdataisavailableforthe1857crisis(SeptembertoDecember)and1866crisis(MarchtoJune).

8

Section2:TheinstitutionalmechanicsoftheBankasalenderoflastresort

I Thenineteenthcenturymoneymarketandthebillofexchange

A recurringtheme in thepost-crisis financialhistory literature is that therecentcrisishad

much in commonwith those in the past (Reinhart andRogoff 2009; Calomiris andHaber

2014; Morys 2014; Turner 2014; Eichengreen 2015).11 If so, this implies that there are

enduringaspectsofthefinancialsystemthatmakes itstructurally fragile.Oneofthecore

fragilitiesstems fromdebtpromising repaymentofa fixedamount,backedbyassets that

fluctuateinvalue(Goodhart1995).Thisfragility,inherentindebtcontracts(Turner2016),is

aggravatedwhendebtsareshortmaturity(oneyearor lesstenor).Thesedebtsarecalled

‘moneymarketinstruments’byfinancialanalystsor‘cashequivalents’byaccountants(Ricks

2016).Theaccountingterminologyisespeciallytelling.Itindicatesthatthesedebtsfunction

asalternativestoholdingcash.12Holdingcashcanbecostlybecauseittypicallybearslittle

orno interest.13 Short-termdebt is thereforeanattractivealternativebecause it typically

offers a higher rate of interest,while at the same time promising instant or near instant

redemption incash.14Anoftenobservedpattern is thatduring financialbooms, thevalue

andvarietyof cashequivalentsexpands,while financialbustsareoften triggeredby their

contractionandconversiontocash(Mehrling2011).

Inthenineteenthcentury,thekeymoneymarketinstrument/cashequivalentwasthebillof

exchange.Abill of exchange is awritten instructionorderingoneparty topay another.15

11AccordingtotheParliamentaryCommissiononBankingStandards(2013),“Hadthewarningsofpastfailuresbeenheeded, thisCommissionmaynothavebeenneeded.” In response toaParliamentaryCommissiononBankingStandards’recommendation,theBankofEnglandstartedaregularseriesofseminarstitled‘LearningfromPrevious Financial Crises’ organisedbyPeterBarrett. In recent years, Bank staff have alsoproducedasteadystreamoffinancialhistoryresearch(Hills,ThomasandDimsdale2010;Bholat2014;Buttonetal.2015;Sowerbuttsetal.2016).12Wecallnotes,coinsanddepositswithcentralbanks‘cash’.Cashtypicallyreferstophysicalcurrencyonlyi.e.notesandcoin.However,depositswithcentralbanksservethesamepurposei.e.theyareusedasameansoffinalsettlement.13 Accountswith the central bank sometimes bear a low rate of interest. Also, somenotes historically paidinterest(BurdekinandKeskinel2013).14Bankdepositsareagoodexampleofacashequivalent.Theyaretypically low interest-bearing,nominallyfixed, short-term debts, backed by longer-term, higher risk/reward assets that fluctuate in value. Banksorganise their books this way because it is profitable. They earn the spread between the yield on higheryielding assets, and lower yielding deposits. In the twentieth century, prior to the introduction of depositinsurance, financial crises often started if depositors came to doubt that their banks’ assetswere valuableenoughtorepaythecashtheywerepromised.Inextreme,depositorsmightthendemandredemptionoftheirdeposits en masse. Demand for cash equivalents contracted while demand for cash proper rose. In thesecrises,bankseitherhadtoliquidatetheirassetsatfire-salepricesorenterinsolvency.15ThebillofexchangeoriginatedintheArabworldintheearlyIslamicera(Geva2011). Itwassubsequentlyadopted in Continental Europe during the Middle Ages. Through trade with the Continent, it graduallyappearedasaninstrumentusedinEnglishcommerceduringthefifteenthcentury(Elliotetal.2013:4).Given

9

Whilemanyreaderstodaymayhavenopracticalexperiencewithabillofexchange,most

willbefamiliarwithcheques,whicharelegallyaspecialkindofbillofexchange(Elliotetal.

2013: 304).A cheque is ‘drawn’ (written)by apersonon their bank topay a thirdparty.

Similarly,abillofexchangeis‘drawn’byoneparty(calleda‘drawer’)onanother(calleda

‘drawee’) instructing them to pay either the drawer or a third party (called a ‘payee’).16

Unlikeacheque,abillofexchangeisnotnecessarily‘drawnon’abank.Itcanbeapayment

instruction to anybody.17 In fact, in the nineteenth century, a bill of exchangewasmost

often literallyabill followingthesaleofgoodsandservices.Forexample,amanufacturer

mightsupplygoodstoamerchantoncreditperhapsbecausethemerchantwasunableto

paycashforthegoodsuntilaftertheyhadbeensoldtoconsumers.Inthisexample,thebill

actedlikeaninvoicetangiblydocumentingthetradecreditthathadbeenextendedinthe

transactionbetween the twoparties. Themanufacturer (thedrawer)would sendabill to

the merchant (the drawee). If the merchant ‘accepted’ that they owed a debt to the

manufacturer,theywouldsigntheirnameonthebill.Legally,theywerenowreferredtoas

the‘acceptor’ofthebillinsteadofthedrawee.

Rather thanholding thebill tomaturity, themanufacturermight cash in thedebt before

maturityineitheroneoftwoways.Onewaywasforthebillofexchangetobeuseddirectly

ascurrencywhenpayingforgoodsandservices,ordischargingdebts.Indeed,insomeareas

ofBritainduringthenineteenthcentury,billsofexchangecirculatedasextensivelyasother

typesofcurrencysuchasBankofEnglandnotesandRoyalMintcoins(Ashton1953).When

apersonorinstitutionholdingabilltransferredittoanother,theyhadtosigntheirnameon

thebackofthebilljustastheacceptorhaddone.Iftheoriginalacceptordidnotpayinfull

orinpart,allendorsers(includingtheoriginaldrawerofthebill)wereliabletopaywhoever

currentlyheldit.Therewasthusa‘bandwagon’effectatplayasbillsofexchangecirculated.

Themorefrequentagivenbillcirculated,themoreendorsersithad.Sincetherewerethen

more guarantors, the bill of exchange became an increasingly safer asset, more closely

approximating cash (Santarosa 2015). The negotiable nature of bills of exchange,

that the bill of exchange arrived in England through her participation in international trade, it is perhapsunsurprisingthatforeignbillsofexchangeendedupplayingthedominantrole inLondonmoneymarkets, incontrasttodomesticor‘inland’billsofexchange,whichwereofsecondaryimportance.

16Likecheques,billsofexchangearenegotiableinstruments.Thismeanstwothings.First,itmeanstheycanbetransferredfromonepartytoanotherwithoutexplicitconsentfromthedrawee/acceptori.e.thedebtor.Second,itmeansthatanysubsequentholdersofthebills(transferees)are“capableofobtainingaperfecttitletotheinstrumentinspiteofanydefectsinthetitleofthepriorparties”(Holden1955:314).16

17Furthermore,whileachequeisawritteninstructionpayableondemand,billsofexchangeorderedpaymentonorafterawiderangeoffuturedates,though,aswenotebelow,theytendedtobemoneymarketinstrumentswithshortmaturity.Inthenineteenthcentury,itwasconventionalthatifabillorderedpaymentinthreemonths,athreedaysgraceperiodwasaddedattheendofthetermfordrawees/acceptorstomakepayment(Moxon1894:15).

10

underpinned by multiple endorsements, resulted in their emerging as the key cashequivalentinthefirsthalfofthenineteenthcentury.18

Figure2.1:Tradetransactionusingbillsofexchange

Figure2.2:Techniquesfortradingabillofexchangebeforematurity

Theotherwaytoencashthebillbeforematuritywasfortheholderofthebilltosellittoa

financial firm. Intheeighteenthandearlynineteenthcentury,holdersofbillsofexchange

oftenarrangedsalesoftheirbillsthroughbillbrokers.Billbrokersinitiallyactedasfinancial

18Furthermore,theywereafinancialsecurityappropriatetotheproductivestructureoftheeconomyatthattime.Capitalmarketsecuritiesandmulti-yearbankloanswerestillrelativelyrarefinancialinstrumentsbecausemostfirmsduringthisperiodfinancedthemselvesfromretainedearnings(Chandler1994).Whenfirmsneededexternalfinance,itwastypicallytoinvestinwhatwenowcall‘workingcapital’asopposedtolong-termrealcapitalinvestments.Sotheshort-termbillofexchangewasanappropriateinstrumentgiventheshorthorizon.Whileoverdraftswerealreadyawell-establishedfinancingmechanisminScotland,theywerelessestablishedintherestoftheUnitedKingdom.

11

intermediariesbetweenbuyersandsellersofbillsofexchange.Likebanking,billbrokering

developedinBritainduringthelateseventeenthcenturyspurredbyrealeconomicgrowth

andtheneedfornewfinancialchannelstofinanceit(Presnell1956).Indeedbillbrokering

wasan important supplement to the limitationsofbanksas theywere structuredat that

time. In this period, the vast majority of banks were single shops (unit banks) without

branches. Therefore, bill brokers acted as conduits for the buying and selling of bills of

exchange across different geographical areas. Thus banks holding bills of exchange who

wantedcashbeforematuritywouldsendtheirbills tobillbrokers,whothenarranged for

their discount by banks in other parts of the country with surpluses of cash looking for

investmentincashequivalents(Banks1999:31).Forthisservice,billbrokersearnedincome

fromcommission.

However, by the 1830s, many bill brokers transitioned from intermediaries of bills of

exchangetoinvestorsinthem.Thisshiftinbusinessmodeloccurredinresponsetothe1825

financial crisis. According to most historians, many banks, particularly in the City, felt

unnerved by the fact that, during the 1825 crisis, the Bank of Englandwas perceived as

having belatedly responded to their demands for liquidity via rediscounting of bills of

exchange (Fletcher 1976: 9). The demand from London banks for a cash equivalent

stimulated the introduction of new facilities by bill brokers, akin to how the rise of

institutionalcashpoolsintheearly2000sspurredthecreationofshadowbankingservices

(Pozsar 2014). Rather than stockpile zero yielding Bank notes, London banks began to

deposit their money ‘at call’ (on demand) with bill brokers, many of whom, spotting a

marketopportunity,startedtoofferdemanddeposits.Asaresult,manybillbrokersevolved

into so-called discount houses which financed their own portfolio of bills with funds

borrowedfrombanks.Theviabilityofdiscounthousedemanddepositstofunctionasacash

equivalentincreasedwhentheBankannouncedrediscountingfacilitiesforLondondiscount

houses in 1833 (Fletcher 1976: 99), providing assurance to banks that discount houses

wouldbeabletohonourtheircommitmentstopayBanknotesondemand.19

By the 1830s, a dense money market network had emerged structured through bills of

exchange.Atthecoreofthisnetworkwerethreekeyinstitutions:(1)‘clearing’banks20,(2)

discounthousesand(3)theBankofEngland.21Figure2.3illustrateshowtheseinstitutions

19Callloansfrombankstodiscounthousesalsobecameincreasinglypopularinthe1830sasalternativestoExchequerbills,whosemarketbecamelessliquidinthe1830sasinvestorssubstitutedthemwithrailwaysecurities(Fletcher1976:15).20Thequalifier‘clearing’infrontofbanksindicatesthattheinstitutionswearereferringtoaremostlyLondon-basedinstitutionswho‘cleared’orsettledclaimsonbehalfofcorrespondentbankslocatedelsewhereinthecountry.21Inthesecondhalfofthenineteenthcentury,thespreadofbranchbankingandoverdraftlendingledtoadeclineinbillsofexchangeasameansoffinancingdomestictrade,whileatthesametimebillsofexchange

12

were interlinked through simplified versions of their balance sheets. Banks funded

themselvesmostlybynotesanddeposits.Whiletheyusedsomeofthesefundstobuybills

directly, by the 1830s, a growing percentage of their assets were call loans to discount

houses. The discount houses used banks’ deposits to fund their portfolios of bills of

exchange.ThediscounthousesinturnmightrediscountthesebillsforcashfromtheBankof

England. The importance of these rediscounting operations became evident during the

crisesof1847,1857and1866.Inactingasalenderoflastresort,theBankarguablymade

theimpactofthosefinancialcrisesontherealeconomymuchlessseverethanthosethat

hadprecededit(Figure2.4).

Figure2.3:ConnectionsintheLondonMoneyMarketthroughsimplifiedbalancesheetsHighlightedbalancesheetitemsshowhowbanks,discounthousesandtheBankofEnglandwereconnected.

Banks DiscountHouses BankofEnglandLiabilities Assets Liabilities Assets Liabilities AssetsDepositsNotesEquity

BillsCallLoans

CallLoansEquity

BillsBanknotes

BanknotesEquity

Bills

Figure2.4:TheimpactoffinancialcrisesonrealGDPgrowth,1790-1870ThischartshowsthathitstoGDPfollowingfinancialcrises(circled)lessonedovertime.

Source:ThomasandDimsdale(2017)

becametheprimaryinstrumentfinancinginternationaltrade.TheoutbreakofWorldWar1ledtoTreasurybillsreplacingbillsofexchangeasthemainLondonmoneymarketinstrument(BankofEngland1967;Nishimura2010).

-10 -5

0

5

10

15

1790 1800 1810 1820 1830 1840 1850 1860 1870

PercentRealGDPgrowth(%pa)

13

II TheBankofEngland’sDiscountOffice

TheBankofEngland’s lenderof lastresortoperationsduringthenineteenthcenturywere

exercisedthroughtheDiscountOfficeattheBank’sLondonheadquartersonThreadneedle

Street. Though a large responsibility, the DiscountOfficewas physically small. TheOffice

was headed by the Principal of the Discount Office who, perhaps significantly we have

discovered,wasthesamepersonthroughout theperiodofouranalysis. JohnGreenElsey

startedattheBankin1830andwasinchargeoftheDiscountOfficebetweenMay1839and

July1878.ThelinkbetweentheBank’sresponsetothecrisesof1847,1857and1866was

thusaverypersonalone.Mr.Elseywasinchargeof,onaverage,sevenstaffthroughoutthe

period. Intermsofheadcount,thismadetheDiscountOfficearelativelysmallpartofthe

Bank.TheOfficewasopensixdaysaweek,orroughly305tradingdaysperyear,butwas

open only from11 am to 2 pm (Ogden 1988: 198). It is not knownwhether these hours

wereextendedduringfinancialcrises.

Figure2.5:TheBank’sDiscountOffice

ImagesabovecomefromsketchesoftheBank’sDiscountOfficedrawnbySirJohnSoane.Soane

builttheDiscountOfficein1810.ThelefthandpanelshowsthelobbyoftheDiscountOffice.The

rightpanelshowstheOfficeitself(picturedwithoutfurniture).Beforetheconstructionofthe

DiscountOffice,theBank’sdiscountbusinesshadbeendoneinthemainPayHall.However,the

Bank’sincreasingdiscountbusinessfromtheRestrictionperiod(1797-1821)on,promptedthe

constructionofaspecificofficewherethisbusinesscouldbedone.

14

The Discount Office lent in two ways: discounts and advances. Discounts involved the

DiscountOfficepurchasingbillsofexchangeatdiscountontheirfacevalue.Advanceswere

loansbytheBanksecuredondebt,akintomoderndayrepos.Thediscounthouseswould

temporarilyselldebtsecuritiestotheBankwithanagreementtobuythembackatafuture

date prior tomaturity. The debt securities used as collateral included government bonds

and railway stocks. In general, discountsweremuchmore numerous than advances. For

example,inthedatawehaveanalysed,discountsrepresented65%and62%oftransactions

in1857and1866,respectively.

Figure2.6:Proportionofdiscountsandadvancesin1857and1866

All bills brought in for discount to the Discount Office were recorded in daily discount

ledgers.Theledgersrecordedinformationonthenameofthepersonbringinginthebill,the

numberandvalueofbillsbroughtin,andtheratefordiscountonthosepurchased,orthe

number of bills rejected otherwise. Drawing Office customers (those who held deposit

accounts with the Bank) were marked as “DO.” By 1866, the ledgers also included the

amountandratechargedonadvances.Figure2.7isasnapshotofthedailydiscountledgerfrom11May1847.Eachdiscountertypicallybroughtinanumberofbillsfordiscount.These

bundles of bills were known as ‘packets’ or ‘parcels.’ For example, Cooks Sons and Co

brought inapacketof60billsonMay11(transactionboxedinblue).Thesepacketswere

likemoderndaymortgagebackedsecurities(MBS)orcollateraliseddebtobligations(CDOs)

inthattheyweredebtbundlescontainingdifferenttypesofbills.TheDiscountOfficewould

firstdecidewhich,ifany,ofthebillsinapacketitwouldaccept,andwhichitwouldreject.

Intriguingly, theBank kept detailed records on all bills it had rejected in a separate “Bills

Rejected”ledger(Figure2.8).Thosebillsthatwereacceptedwerediscountedtypicallyatasingle rateof interest.However,onoccasion,apacketwouldbediscountedat two rates.

Forexample,inthe11Mayledger,thefirstpacketwith7billswasdiscountedatarateof5

0

20

40

60

80

100

1857 1866

PercentDiscounts Advances

15

and6percent (transactionboxed in red). In1847,around13percentofpacketshaddual

rates. A glance through the ledger books after 1847 reveals that the Bank gradually

decreased the number of packets which were given dual discount rates. The practice

stoppedin1856.22

Ona“normal”businessday,thenumberofdiscounterstheDiscountOfficewouldservewas

small.Forexample,theaveragewas26customersduringnon-crisisweeks23in1847(Figure2.9).The listof thesediscounterswould fiteasilyon toone ledgerpage.However, in themidstof a financial crisis, the footfall couldbe tremendous, andmore than four timesas

large.Figure2.10comparesthepeakdayinacrisiswiththesamedayayearearlier.

22ItisunclearwhytheBankdidn’trecordthepacketswithbillscomingfromthesamediscounteronthesameday but at different rates as separate transactional lines. It may simply have been a bookkeeping custom.Thereisnowaytoknowtheweightedaveragerate.Inouranalysisofdiscountrates,whereapacketisgivenadualrate,weusethelowerrate.Thismeansourdiscountratesin1847aresystematicallybiaseddownward.23Wedefinecrisisweeksthroughoutthepaperasthoseweekswherethelevelofnotesanddiscountsrecordedand/orthenotereserveinBankingDepartmentaremorethantwostandarddeviationsfromthemean.

16

Figure2.7:Dailydiscountledgerfrom11May1847

17

Figure2.8Excerptfromthe“BillsRejected”ledgeron28thMarch2017The ledger recorded the details of bills that the Bank had rejected including (left column to right) thediscounter,thedrawer,theacceptor,theacceptor’saddress,whenthebillwasdue,anditsmonetaryvalue.This seems to us a costly exercise to undertake for securities the Bank wasn’t going to purchase. OnepossibilityisthattheBankusedthis informationtokeeptrackofactivityinthefinancialsysteme.g.togetasenseoftheoverallindebtednessofhighlyleveredacceptors.

18

Figure2.9Averagecustomersperdayincrisisandnon-crisisweeks

Figure2.10:Peakcrisistransactiondaycomparedwiththeyearbeforeandafter

ReturningtotheledgeronMay11,weseeitwasabusydaywith53transactions,though

thiswasbynomeans thepeakof the1847crisis.Assuming theDiscountOfficewasonly

open itsstandard3hours, itwouldhavehadtoprocessover17packetsperhour,orone

packet roughly every fiveminutes. Each of these packets containedmultiple, sometimes

hundredsof,billsofexchange.On11Maythetotalnumberofbillswas702.Againassuming

theDiscountOfficewasonlyopenitsstandard3hours, itwouldhavehadtoassesscredit

0

5

10

15

20

25

30

35

40

non-crisis crisis non-crisis crisis non-crisis crisis

1847 1857 1866

Numberofcustomers

0

10

20

30

40

50

60

70

80

90

0

10

20

30

40

50

60

70

80

90

12Oct1846

12Oct1847

12Oct1848

12Nov1856

12Nov1857

12Nov1858

11May1865

11May1866

11May1867

nodata

19

qualityatarateofcloseto8billsperminute.ThissuggeststheDiscountOfficeeitherhad

well-developed discounting rules, were operationally very efficient, or had scrutinised at

leastsomeofthesebillsinadvance(moreonthisthirdpossibilitylater).

Besides theDiscountOfficeatHeadOffice in London, it’sworthbearing inmind that the

Bank of England’s branches outside of London would have also been discounting bills.24

Unfortunately, these branch ledgers no longer exist. This means these ledgers and the

activitiestheychronicledaremissingfromouranalysis.25However,wedohaveasenseof

theaggregatevalueofthesetransactionsfromtheannualreportingoftheBank’sbranches’

activitytoCourt.26Figure2.11showstheirsignificance.Figure2.12showsdifferencesinthemonetaryvalueofthesetransactions.Duringthe1847and1857crises,roughly40%ofthe

Bank’s business discounting bills by valuewas done through its branches. In 1866 itwas

50%.

Figure2.11:MonetaryvalueofbillsdiscountedinLondon(HeadOffice)versusbranches

24TheBankwaspermittedbylawtoestablishbranchesoutsideofLondonfrom1826.25Thisisakeylimitationinouranalysis.Twootherlimitationsshouldbenoted.First,wemakenoadjustmentfor seasonality even though there is reason tobelieve that the volumeofdiscounts andadvancesprobablyvarieddependingonthetimeoftheyear(Ogden1988:201).Second,thefullextentoftheBank’screditeasingmaynotbereflectedintheledgers.Forexample,thereissomeevidencethatduringthe1847crisis,theBankchangedthemixofassetsonitsbalancesheettohelptheLondonmoneymarket,swappingoutgiltsinreturnforilliquidassetsheldbymarketcounterparties.Thesetransactionsweresimilarinformtothoseundertakenby the Bank in 2008when it swappedUK Treasury bills for banks’ loans (Domanski,Moessner and Nelson2014:70).Thesetransactionsdonotappearinourledgers.26ReportsweremadetotheSpecialDiscountCommittee,BoEC35andtheannualdatacanalsobefoundinBankofEnglandArchives,DiscountOfficeAnalysesandSummaries,BoEC30/3.

0

10

20

30

40

50

60

1844 1846 1848 1850 1852 1854 1856 1858 1860 1862 1864 1866 1868 1870

£mnLondon Branches

20

III HowtheDiscountOfficeoperated—rulesversusdiscretion

Although theDiscountOffice retained immaculate records of the bills it discounted, very

littledocumentationremainsthatshedslightonhowitoperated.Thishasledtoadebatein

the literature between those who argue that it operated by applying a few, simple,

definitiveruleswell-knowntomarketparticipants(Capie2007),andthosewhoarguethat

the Bank and the Discount Office exercised a much greater degree of discretion when

lending(FlandreauandUgolini2011).Thereexistsevidenceforbothsidesoftheargument.

Ontheonehand,mosthistoriansagreethattheBankhadsomegeneralrulesofeligibility

forbillsitdiscounted.Oneoftheseruleshadtodowiththematurityofthebill,thoughthe

exactcriteriamayhavechangedovertime.Forexample,areportbytheSpecialCommittee

on theDiscountDepartmentdated8August1844proclaimed that “noBill bediscounted

havingmorethan6monthstorun.”27OtherhistorianshaveclaimedthattheBankpreferred

to discount bills at around 65 days and did not deal in bills ofmore than 95 days’ tenor

27BankofEnglandArchive,BoEG15/62.

Figure2.12:Monetaryvalueofbillsdiscountedinbranches(inred)duringcrisisyears

£64millionpoundsdiscountedinbranchesin1847wouldbeworth£6.1billionin2016£82millionpoundsdiscountedinbranchesin1857wouldbeworth£8.5billionin2016£68millionpoundsdiscountedinbranchesin1866wouldbeworth£7.4billionin2016CalculatedusingBankofEnglandInflationCalculator:http://www.bankofengland.co.uk/education/Pages/resources/inflationtools/calculator/default.aspx

40%

41%

50%

Total£64m

Total£82m

Total£68

0

10

20

30

40

50

60

70

80

90

100

1847 1857 1866

£mn

21

(Scammel1968:176).TheDiscountOffice’sowncalculationsoftheaveragematurityofbills

it discounted corroborate this claim.28 Figure 2.13 displays a new time series we have

constructed,from1849,whenthedatastartedtoberecorded,upto1870,markingtheend

of the period we have investigated. In general, the average maturity of bills discounted

hoversjustabove60days(twomonths)andnevercomesclosetoexceeding95days’tenor.

Still,attimestheBankappearstohaveextendedthematurityofthebillsitdiscounted.For

example, in theaftermathof the1857crisis, theaveragematurityofbillsdiscountedwas

closerto70days.However,thisfollowedaperiodwhenthematurityofbillsdiscountedhad

been declining. In 1866, thematurity of bills discounted by the Bank also fell during the

crisis. This might be interpreted as the Bank tightening its lending eligibility criteria.

However,itcouldjustbeareflectionofdemand,giventhepaperprevailinginthemarket.

Whatever thecase, theaveragematurityofbillsdiscountedby theBankdidnot reach its

lowestpoint,observedinthemid-1850s.

Figure2.13:Averagematurityofbills1849-1870

At the same time, there is somearchivalevidence thatdiscretionplayeda rolewhen the

Bankdischargeditslenderoflastresortresponsibilities.Figure2.14 isanexcerptfromthe

diariesofBonamyDobree,DeputyGovernoroftheBankatthetimeofthe1857crisis.On

30OctoberherecordsameetingwithMrBarnett(ofBarnettHoareandCo.)whobroughtin

apacketofbillsforhimandthePrincipaloftheDiscountOfficer,theaforementionedMr.

Elsey,tolookover,withtheaimofsecuringapossibleadvanceshouldtheneedarise.The

wording inthediarymaybesignificant.TheDeputyGovernornotesthatbothheandMr.

Elsey gave their opinion as to the quality of the bills, implying the exercise of judgment

ratherthanmechanisticapplicationofarule.Alookattheledgersshows£194,000worthof

advances(intwobatches)wasmadeoverthenexttwoweeks.Thisepisodealsoprovidesat

28BankofEnglandArchive,DiscountOfficeAnalysesandSummaries,BoEC30/3.

0

15

30

45

60

75

90

105

1844 1846 1848 1850 1852 1854 1856 1858 1860 1862 1864 1866 1868 1870

Days

95daymax

22

leastoneexamplethattheDiscountOfficemayhaveoperatedsomesystemofinformalpre-

pledgingofcollateral inadvanceofactual loans,anantecedenttotheformalpre-pledging

of collateral that counterparties currently do in order to gain access to Bank of England

discountfacilities.

Figure2.14:Evidenceofdiscretionand“informalpre-pledging”ofcollateral“MrBarnettofLombardStreetcalledtoshowaparcelofbillsbroughtupbytheClydesdaleBankCorporation,toknowiftheBankofEnglandincaseofnecessitywouldmakeanadvancetotheClydesdaleBankcorporationof…..£150-£200,000onthesaidbills.MrElsey[acashierattheBank]lookedoverthemandgaveitashisopinionthattheBillsare,onacursoryviewofthem,ofthemofexcellentcharacterandIgaveMrBarnettmyownopinionthattheBankwouldmaketheadvance.”ExcerptfromthediariesofDeputyGovernorBonamyDobree,30thOctober1857,BoEM5/454Ledgerentryfrom9November1857

Ledgerentryfrom12November1857

IfdiscretionwasexercisedwhentheBanklent,thisraisesfurtherquestionsastothelevel

within the organisation where that discretion was exercised. Perhaps, as in the diarised

episode,itwasexercisedbothbytheDiscountOfficeandtheBank’smostseniordirectors.

For example,whilewe know that during this period seniormembers of theBank’s Court

throughtheCommitteeofDailyWaitingmonitoredtheBank’sdiscountseachday,theydid

soexpost,aftertheDiscountOfficehadmadeitslendingdecisions.29

29TheCommitteeofDailyWaitingexisteduntil1914.

23

Indeed themid-nineteenthcenturywasaperiodof increasingautonomy for theDiscount

Office. From its founding to the mid-nineteenth century, the Bank’s policy had been to

discountallbillsatasinglefixed(Bank)rateirrespectiveofthetermtomaturityofthebills,

and irrespective “the standing of the parties to them or of the applicants for

accommodation”(King1936:81).Creditlimitsexisted,buttheyweretypicallyimplemented

by restricting the quantity of loans rather than varying their price. This changed in the

1840s.WhileBankRatewasstill fixedweeklyby theCourtofDirectors, starting inMarch

1845,theDiscountOfficenolongerappliedituniformlytoallbillsbutappliedarange,aswe

showlaterinthispaper.30

The greater discretion the Discount Office had to set loan prices was an unintended

consequenceofthe1844BankCharterAct.ThatActhadsoughttopreventtheover-issueof

privatebanknotesthatmanycontemporariesfeltwasthesourceoffinancialcrisesin1825

and1837.31 To thatend, theAct gave theBankof Englandaneffectivemonopolyon the

issueofnewbanknotes.With theexceptionof an initial £14million fiduciary issue,new

Banknoteshadtobebackedone-for-onewithgold.Tofacilitatethis,the1844Actsplitthe

Bankintotwodepartmentsforaccountingpurposes.32The“IssueDepartment”oftheBank

wastolookafterthenoteissueandtheissuanceofnewnoteswastiedtotheamountof

gold it held in reserve. The framers of the Act believed thiswould ensuremonetary and

pricestability.33Meanwhile,therestoftheBank–the“BankingDepartment,”includingthe30AsKing(1936:110)notes,“Thustheapplicantfordiscountaccommodation,unlessheheldabsolutelyfirst-classandshort-datedpaper(bearingatleasttwoundoubtednames,oneofwhichhadtobeaLondonacceptor),couldneverknowinadvancepreciselywhatratehewouldbecharged,fortheBankdidnotmakeageneralpracticeoffixingandannouncingdifferentialratesforthevariouseligibleusances.”31Forexample,the1825financialcrisiswaswidelyblamedbycontemporariesonprivatecountrybanks,whowereaccusedofhavingover-issuedsmalldenominationnotes(Bagshaw1920:197).Inresponse,twoBankingActswerepassed in1826.The firstof theseprohibited the issueofnoteswith facevaluesof less than fivepounds (Fletcher 1976: 9). The second Act permitted joint stock banks to issue notes (Truptil 1936: 60).Previously, only bankswith fewer than six partnerswere allowed to issue notes,with the exception of theBank of England. This legal restriction had been designed to protect the Bank’s market position in noteissuance by limiting the equity funding options of potential competitors. However, the downside of thisrestriction manifested itself in the 1825 crisis when the banks who had issued notes were too poorlycapitalised to absorb losses. In order to improve the solvency of country banks, the Banking Act of 1826removed theupperboundonnumberofpartners,butonly forbanksoperatingoutsidea65mile radiusofLondon.Again, thiswas toprotect theBank’smarketposition innote issuance. TheAct alsopermitted theBanktoopenbranchesinthecountryside,withtheapparentgoalofgettingitsnotestoreplacethoseissuedbycountrybanks,blamedforthe1825crisis.

32ThispartitionstillexiststodayasaBankofEnglandaccountingconvention,butwithouttheoperationalimplicationsithadbackthen.33 The logic was that if gold flowed out of the country due to an overheating economy and a balance ofpaymentsdeficit,thestockofnotesincirculationwouldautomaticallydeclineaspeoplecashedintheirnotesfor gold to pay for imports. It was also envisaged that the Bankwould respond to this outflow by raisinginterestratestoattractmoregoldfromabroad,andbyreducinglendingtoprotectitsexistingreserves.Thecombination of higher interest rates and a falling money supply, it was believed, would lead to slowingeconomic activity and falling prices,whichwould improve the balance of payment, providing an automaticstabilisationmechanism.

24

Discount Office– could operate for profit like any other private bank. Importantly, the

Banking Department held part of the total stock of Issue Department notes as its own

reserve and the growth if its deposit liabilities were not tied to gold. This gave it some

flexibilitytomeetdemandsforcredit.

Withmonetary and price stability assumed to be guaranteed by the Issue Department’s

backingofBanknoteswithgold,theDiscountOfficestartedtocompetemoreaggressively

withotherbanks. Inthetwoyearsthatfollowedthepassingofthe1844BankingAct,the

BankingDepartment’sholdingsofprivatesecuritiesexpandedmarkedly, followingacut in

BankRatefrom4%to2.5%.MoregenerallythediscretionarylendingactivityoftheBanking

Departmentcouldinpartoffsetor‘sterilise’theautomaticstabilisersbuiltintotherulesfor

theIssueDepartment.34

ThediscretiontheDiscountOfficecouldexercise,however,haditslimits,whichweremost

clearlymanifestduring financial crises.Figures2.15and2.16 illustrate.While thegeneral

techniquewherebytheBankcreatednewnotestobuybillshaslongbeenunderstoodata

high-level, thesediagramsdetail theprecise institutionalmechanicstypicallyglossedover.

In a crisis, therewould be an increased demand for Bank of England notes (cash) by the

financialsystemwhichcouldbeobtainedbydiscountingshort-datedbills(cashequivalents)

with the Discount Office. Banking Department would typically pay this out of its note

reserve.Alternatively,ifthediscounterwasalsoaDrawingOfficeorbankingcustomerwith

anaccountattheBank,theDiscountOfficemightinitiallyjustcreditthecustomer’sdeposit

balance.Butinacrisisthesedepositscouldeasilygetcashedintonotes.Sotypicallyacrisis

wouldleadtoachangeintheassetmixofBankingDepartment’sbalancesheet,withmore

discountsandfewernotesheldinreserve.OntheIssueDepartment’sbalancesheet,allthat

wouldhappeninthefirstinstanceisthatnotesincirculationwouldincreaseattheexpense

ofnotesheldbytheBankingDepartment.ThesizeoftheIssueDepartment’sbalancesheet

wasunchangedand therewouldbenobreachof theBankCharterAct’sprovisionsasno

newnoteswereprinted.Therewasmerelyashiftincompositionofthetotalnotesstock.35

However, if the crisis deepened, and the demand for discounts continued, the Banking

Department’s note reserve might start to dry up. This created the possibility that the

DiscountOfficemightno longerdiscountbills.Thiscouldcauseadditionalpanic inmoney

markets. In response, the Government at this point might provide the Bank with an34Forexample, iftherewasanoutflowofgold,theBankingDepartmentcouldoffsettheimpactonnotesincirculation and themoney supply by lendingmore of its reserve notes to the private sector or by creatingadditional deposits for the borrowers in their accounts held with the Banking Department. BankingDepartmentaccountswereakintomoderndayReserveAccounts.

35ItwasofcoursepossiblethatthetotalnotestockcoulddeclineifsomeofthenotesdrawnfromtheBankingDepartmentreservewereeitherimmediatelyorsubsequentlycashedintogoldcoin.

25

indemnity allowing it to breach the 1844 Act. This would allow the Issue Department to

createadditionalnotes.ThesewouldbegiventotheBankingDepartment inexchangefor

someof itsbillsandsecurities.Thesenotescould thenbeusedby theDiscountOffice to

discountadditionalbills.In1847and1866,themereexistenceoftheindemnitystoppedthe

panicandthistransactionnevertookplace.Itdid,however,getcarriedoutin1857.

Figure 2.15:Billswould come into theBank for discountwhich itwouldpayoutof thenotereserveinBankingDepartment

Figure2.16:TheonlywayBankingDepartment(BD)couldrestorethenotereservewastoswapbillsfornoteswithIssueDepartment(ID)oncetheBankCharterActwassuspended

26

Section3:Lendingfreely

I Evidenceonaggregatelending

Thecharts that followarebasedonanewlyavailableweeklyseriesof theBank’sbalance

sheethistorically(HuangandThomas2016b).Figures3.1and3.2showstheweeklytotaloftheBank’sprivatediscounts,advancesandsecurityholdingsduringthecrisesof1847,1857

and 1866. This is based on theweekly dataset of but has been extended by the current

authors to include a breakdown of discounts and advances. The Bank appears to lend

substantialamountsduringthesecrises,atleastrelativetonon-crisisperiods.Privatesector

discounts, advances and other securities expanded between 60 and 80 percent of the

BankingDepartment’sbalancesheetduringeachcrisis.Discountsandadvancesratherthan

purchasesoflonger-termsecuritieswerethekeyformviawhichtheBanklenttotheprivate

sectorduringacrisis.36

Figure3.1:Privatediscounts,advances,andothersecurities1844-1914

36Latercrises,suchas1878and1890,exhibitednoobvious“lendingfreely”onthesamescale.Partofthathastodowiththefactthatthosecrisescentredaroundtwospecificinstitutionsi.e.theCityofGlasgowBankandBarings,respectively.However,equally importantmayhavebeenthememoryoftheBank’sactionsin1847,1857 and 1866. Once it was understood by the public that the Bank would intervene, the mere ex anteexpectationofthisresponse,togetherwithgreaterpre-emptivemovesbytheBank,weresufficienttonipamoreseriouscrisisinthebud. Itwasnotuntilthespecialcircumstancessurroundingtheoutbreakofwarin1914thattheBanksteppedinonamassivescaleagain,and,eventhen,Governmentplayedtheleadingrole.

0% 10% 20% 30% 40% 50% 60% 70% 80% 90%

1844 1849 1853 1857 1862 1866 1871 1875 1879 1884 1888 1893 1897 1902 1906 1910

Crisis DiscountsandAdvances %ofBDassetsandothersecurities

27

Figure3.2:Breakdownofprivatesectorassets1844-1870

The counterpart to the increased lending was a fall in the note reserve in the Banking

Department (Figure 3.3). Bankers’ deposit balances did not show an increase, exceptperhapsduringthe1857crisis,implyingthateveniftheextradiscountingbytheBankwas

initially credited to deposit accounts, these were soon withdrawn in notes (Figure 3.4).Figure3.5showsthat,despitethenotedrainfromtheBankingDepartment,notallofthese

noteswere thenconverted intogoldcoinandoverallbullion reserves remained relatively

adequate,atleastbytheBank’shistoriclevels.Onlyin1857didthecoinandbullionreserve

ratiotototalliabilitiesdipbelow20%

0% 10% 20% 30% 40% 50% 60% 70% 80% 90%

1844 1847 1850 1853 1856 1859 1862 1865 1868

Advances Discounts Otherprivatesecurities %ofBDassets

28

Figure3.3:NotesinBankingDepartment,1844-1914

Figure3.4:Bankers’balances(reserveaccounts)inBankingDepartment,1844-1914

0%

10%

20%

30%

40%

50%

60%

1844 1849 1853 1857 1862 1866 1871 1875 1879 1884 1888 1893 1897 1902 1906 1910

Crisis ReserveofnotesandcoininBankingDepartment %ofBDliabilities

0%

10%

20%

30%

40%

50%

60%

1844 1849 1853 1857 1862 1866 1871 1875 1879 1884 1888 1893 1897 1902 1906 1910

Crisis Bankers'balances%ofBDliabilities

29

Figure 3.5: Coin and bullion reserves of the Bank relative to consolidated liabilities,1844-1914

Information from the Bank’s daily discount ledgers corroborates the balance sheet view.

Figures 3.6 to 3.8 show a significant increase in the average number of discount

transactions, the number of bills per packet, and the monetary value of discounts,

respectively,madeincrisisweekscomparedtonon-crisisweeksin1847,1857and1866.In

allyears,theaveragesareelevatedduringcrises.ThisprovidessomeevidencethattheBank

wasatleastprovidingsignificantsupporttothemarketduringcrisisweeks.

0%

10%

20%

30%

40%

50%

60%

1844 1849 1853 1857 1862 1866 1871 1875 1879 1884 1888 1893 1897 1902 1906 1910

Crisis Coinandbullion %ofconsolidatedliabilities

30

Figure3.6:Averagenumberofdiscountsmadeperday1847,1857and1866

Figure3.7:Averagenumberofbillsperpacketinadayin1847,1857and1866

0

5

10

15

20

25

30

35

40

crisis non-crisis crisis non-crisis crisis non-crisis

1847 1857 1866

Numberofdiscounts

0

100

200

300

400

500

600


1847 1857 1866

Numberofbills

31

Figure3.8:Averagemonetaryvalueofdiscountsdoneperdayin1847,1857and1866

II Evidenceonlendingdistribution

Lookingattheledgersallowsustodigdeeperintoaggregatelendingfiguresanddescribeits

distribution.Inparticular,weextendworkbyFlandreauandUgolini(2011)onthecrisisof

1866,withanalysisoftheearliercrisesof1847and1857.Figure3.9liststhetopdiscountersineachcrisis.Discerningconnoisseursoffinancialhistorywillnotethatourfiguresfor1866

areslightlydifferent fromthoseofFlandreauandUgolini.This isbecausethey focustheir

analysis onMay 1866,whereas our analysis focuses on the period both before and after

OverendGurney’sfailureon10May.

Interestingly,weobserve that someof the topdiscounters in1847defaulted,butappear

againas topdiscounters in1857. Forexample,BruceBuxton&CoandSanderson&Reid

suspended payments to depositors in 1847. Having defaulted, these firms then

reconstitutedthemselveswithslightlydifferentpartners,becomingBruceWilkinson&Co,

andSandersonSandeman&Co,respectively.Bothfirmsthendefaultedagainin1857.Yet,

as discussed in greater detail later, such defaults did not result in financial losses for the

Bank,atleastinaggregate.SolongastheunderlyingbillspurchasedbytheBankweregood

andwererepaidbytheultimateacceptors,thesolvencyofthecounterpartieswithwhom

theBanktransacteddidnotnecessarilymatter.

0

50,000

100,000

150,000

200,000

250,000

300,000

350,000


1847 1857 1866

£

32

Thedailydiscountledgerdatarevealsacontrastbetweenthecrisesof1847and1857,on

theonehand,and the crisis in1866,on theother. First, the crisesof1847and1857are

dominatedbypurchasesfromthreetofourmajorcounterpartiesineachcase.Bycontrast,

in1866, thenumberofdiscountersmore than twostandarddeviations from themean is

more evenly spread. Second, in 1847 and 1857, the top counterparties are mostly bill

brokers, whereas in 1866 commercial and merchant banks are more prevalent. This is

becausethe1857crisiswaspartlyblamedonbillbrokersforrecklessly lending,aidedand

abettedbyresorttheyhadtotheBank’srediscountingfacilities.Asaconsequence,inMarch

1858,theBank issuedastatementannouncing itswillingnesstoshutdownre-discounting

facilities to brokers in ‘normal’ circumstances (Calomiris 2010). While bill brokers and

discounthouseswerenotdeniedemergency liquidityassistance in1866, theprovisionof

liquiditywasnowequallychannelledtobanks,symbolicoftheirgrowingascendanceinthe

hierarchyoffinancialfirmsinLondon.

Figure3.9:TopDiscounters,counterpartieswithtotalloansbyvaluemorethantwostandarddeviationsfrommean1847 1857 1866

CounterpartyName Sector

Percentofallloans CounterpartyName Sector

Percentofallloans CounterpartyName Sector

Percentofallloans

A&GNAlexander&Co BB 6.83% OverendGurney&Co BB 9.1% AlexanderCunliffes&Co BB 5.45%

BruceBuxton&Co BB 5.58% BruceWilkinson&Co BB 8.07% OrientalBankCorporation BA 5.22%

NMRothchild&Son MB 5.47% AandGWAlexanderandCo BB 6.97% TheLondonCountyBank BA 4.76%

Sanderson&Reid BB 2.52% GeorgePeabodyandCo MB 3.20% NationalDiscountCoLtd DH 3.89%

OverendGurney&Co BB 1.82% GlynandCo BA 2.80% Barclay&Co BA 3.45%

RobertLawes&Co BB 1.35% SandersonSandeman&Co BB 2.23% HarwoodKnight&Allen MI 3.01%

MorrisonD&Co MB 1.34%GlynandCoo/aUnionBkofScot. BB 1.85% TheCityBank

BA 2.23%

FHuth&Co MB 1.27% Durant&Co BA 1.69% Drake,Kleinwort&Cohen MB 2.08%

MagniacJ&Co MI 1.20% KraeutlerandMieville MB 1.65% BrightwenGilletandCo 2.03%

McCalmontBrothersandCo MB 1.58% SmithFleming&Co MB 1.99%

FHuth&Co MB 1.78%

FinlayCampbell&Co MB 1.74%

LondonAsiatic&AmericanLtd MI 1.63%

ColonialCompanyLimited MI 1.56%

SamuelMontagu&Co MB 1.50%

BankofHindustanLimited BA 1.40%Notes:BB=Privatebillbroker,BA=Bank,MB=Merchant/MerchantBank,DH=DiscountHouse,MI=Missing.Firmsinreddefaultedduringthecrisis.

33

Figure3.10givesasenseoftheoverallskewindiscountsmadebytheBank.ThesePareto

curves show that in 1847, 1857 and 1866, roughly 80% of discountswent to 20% of the

Bank’stopcounterparties.37Figure3.11showsacomparisonduringnon-crisisperiods,the

Paretocurveswerelessskewedinthesenormaltimes,especiallyafter1857when,asnoted

above,billbrokersweredeniedaccesstoregulardiscountfacilities.

Figure3.10:DistributionofDiscountsduringcrisisweeksin1847,1857and1866

37Hencethenameofthechart,whichreferstotheParetoprinciplethatroughly80%ofeffectscomeform20%ofthecauses(Kiremire2011).Again,weareindebtedtoFlandreauandUgolini(2011)fortheinspirationtocreatethesecharts.

0% 10% 20% 30% 40% 50% 60% 70% 80% 90%

100%

0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%

Num

bero

fDisc

ounters

SumsDiscounted

CumulativeDiscounts1847



34

Figure3.11:DistributionofDiscountsinnon-crisisweeks1847,1857and1866

Figure 3.12 shows similar curves for advancesmadeby theBank of England in 1857 and

1866.Thepatternisthesameasabove,withthetop20%ofcounterpartiesreceiving78%

and88%advances in1857and1866, respectively. In1857, thetopcounterpartiesarebill

brokers.However, in1866,asFigure3.13 shows, the topcounterpartiesaremorevaried.

Forexample, thecounterparty that received the largestadvancescumulativelyduring the

crisis was a bank, Agra & Mastermans. Despite receiving assistance from the Bank, it

eventuallydefaulted.

0% 10% 20% 30% 40% 50% 60% 70% 80% 90%

100%

0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%

Num

bero

fDisc

ounters

SumsDiscounted




35

Figure3.12:DistributionofAdvancesin1857and1866

0% 10% 20% 30% 40% 50% 60% 70% 80% 90%

100%

0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%

Num

bero

fBorrowers

SumsAdvanced

CumulativeAdvances1857

CumulativeAdvances1866

Figure3.13:Topadvances,withtotalloansmorethantwostandarddeviationsfrommean1857 1866

OverendGurney&Co BB 2291000 Agra&MastermansBankLimited BA 750000

AandGWAlexanderandCo BB 1662250 FrithSands&Co MB 710000

SandersonSandeman&Co BB 980000 AllianceBankLimited MI 520000

RobertLawesandCo BB 608000 SmithFleming&Co MB 490500

NationalDiscountCorporation DH 558000 HarwoodKnight&Allen MI 490000

BruceWilkinson&Co BB 413000 BankofLondonLimited MI 430000

FrithSandsandCo MB 256000 DiscountCorporationLimited MI 375000

CunliffesandCo BA 180000 LondonWestministerBank MI 300000

LondonDiscountCorporation MI 174000 RobertLowes&Co BB 256000

TheLondonCountyBank MI 200000

NationalDiscountCoofLimited DH 200000

Barclay&Co BA 200000

SheppardPelly&Co MI 169500

HaarbleichenSchumaum MB 164900

Gledstones&Co MB 163000

ColonialBank MI 160000

Barnett&Co MI 144000

R.CunliffeSon&Co BB 140000

TheBornesCoLimited MI 125500

RobertSmith&Co MI 102700

BankofHindustanLimited MI 100000

Blogg&Martin MB 94000

J.F.Pawson&Co MB 90200

MetropProvincialBankLimited MI 87000Notes:BB=Privatebillbroker,BA=Bank,MB=Merchant/MerchantBank,DH=DiscountHouse,MI=Missing.Firmsinreddefaultedduringthecrisis.

36

Section4:Interestrateevidence

I BankRateandmarketrates

We now consider how the Bank priced loans. Bagehot stated several reasons the Bank

shouldlendatahighorpenalinterestrateduringcrises.Thefirstreasonwastoensurethat

onlythosethatreallyneededloanswouldcometotheBank.Second,ahigh levelofBank

Ratewouldpenalisethosewho,with“unreasonabletimidity,”refusedtolend,raisingtheir

opportunitycost fornotdoingso.Finally,ahighBankRatewouldprotect theBank’sgold

reserveandenticegoldfromabroadatamomentwhenitmightotherwiseflowaway.

Formuch of its history, it would have been impossible for the Bank to follow Bagehot’s

prescription because of usury laws. These laws prohibited interest rate charges on short-

termbillsandpromissorynoteshigherthan6%,between1660and1714,and5%between

1714 and 1833 (Temin and Voth 2008). As a result, the Bank’s discount ratewas at this

maximum limit for almost the entire eighteenth century. This meant that Bank Rate

effectivelyoperatedasaceilingonmarketrates.Normally,whenmarketrateswerelower

thanBankRate,fewpeoplewouldcometotheBanktodiscountbills.Figure4.1compares

theBank’sdiscount ratewith twoproxiesofmarket rates in theeighteenth century– the

yield on 6-month East India Company bonds, and the yield on consols which were not

subject to usury limits. During crises, asmarket rates rose towards the usury limit,more

peoplemightcometotheBanktodiscountbills,but theBankwasrestrictedfromraising

rates to a level that might simultaneously preserve its bullion reserve, and reflect the

increasingdemandforcash.Instead,theBankrationedcreditandwouldgenerallyonlylend

toitsregularprivatemerchantcustomersinLondon.38

However, in 1833, the usury laws relevant to bills were repealed. As a result, Bank Rate

couldnowbe increasedabove5%.Figure4.2 shows that after 1833, theBank started tocharge rates above 5%. It also shows that the Bank started to change Bank rate more

frequently.Asthechartshows,duringcrises,BankRatespiked,givingsomeindicationthata

penaltyratewasbeingapplied,asBagehotwouldhaveapproved.

38 Lovell (1957) notes that, prior to the restriction period of 1797, private banks had no direct access torediscounting facilities and it was rare for merchants outside London to be given direct access the Bank’sfacilities,thoughthiscouldbearrangedindirectlyviaaLondoncounterpartyknowntotheBank.

37

Figure4.1:BankRateandmarketrates1715-1833

Figure4.2:BankRatebeforeandafter1833

ThefreedomimplicitlygrantedtotheBankingDepartmenttocompetefordiscountbusiness

bytheBankCharterActof1844meantthatBankRatewastypicallybelowmarketratesin

theleaduptothe1847crisis.Thiscarriedoverintothecrisisperiod.Figure4.3compares

Bankratestothemonthlyaverageofthreemonthbillratesinthemarket,derivedfromThe

0

1

2

3

4

5

6

7

8

1718 1726 1734 1743 1751 1759 1768 1776 1784 1793 1801 1809 1818 1826

PercentFiancialcrisis BankRate

EastIndiaBondYields(6monthrate) YieldonConsols

0

1

2

3

4

5

6

7

8

9

1694 1709 1724 1739 1754 1769 1784 1799 1814 1829 1844 1859

Percent

Usurylimit

1847

crisis

1857

crisis1866

crisis

1833

38

Economist’s Bankers’Gazette.39Market rates rose to 10% inOctober 1847whereasBank

Rateonlyincreasedto8%.

Figure4.3BankRateversusmarketratesin1847

Bignon and co-authors (2012) have argued this suggests the Bank was rationing credit

duringthe1847crisis. Thishypothesis ispartlysupportedbytherejectionrateonbills in

Figure4.4. Around10%ofbillswererejectedin1847comparedto5%in1857and4%in

1866.Also, inthe1857and1866crises,BankRatewastypicallyabovethoseprevailing in

themarket(Figure4.5and4.6).

Figure4.4Percentageofbillsrejectedbyvaluewithcrisisyearsinred

39MarketcommentaryintheEconomistindicatesthatthematurityofabillwasakeydeterminantofthediscountrate.Whetherthebillwasdrawnonabanker(abankers’bill)ordrawnonamerchant(atradebill)alsoseemstohavemadeadifference.Ingeneral,atradebillwithalongermaturitywasdiscountedatahigherrate.

2

3

4

5

6

7

8

9

10

11

Jan-47 Feb-47 Mar-47 Apr-47 May-47 Jun-47 Aug-47 Sep-47 Oct-47 Nov-47 Dec-47

PercentBankrate Marketrate

11 %

5 % 4 %

0

2

4

6

8

10

12

18471848184918501851185218531854185518561857185818591860186118621863186418651866

Percent

39



II Transactionalrates

Wenow turn from thequantityof lendingduring crises to itsprice.Asnotedearlier, the

1844Actmarkedasignificantchange inthewaytheBankofEnglandset its interestrate.

Before 1844, theBank set a uniform rate (BankRate) on all bills it discounted.However,

after1844,BankRatewasusuallytheminimumratechargedfordiscountingtheverybest

billsofshortmaturity.Forbillsoflowerqualityorlongermaturity,theBankwouldchargea

premiumontopofheadlineBankRate.

Figures4.7to4.9showthespectrumofrateschargedbytheBankduring1847,1857and

1866basedondatafromthedailydiscountledgers.Sofarasweareaware,oursisthefirst

paper to show this spectrum of transactional rates for 1847 and 1857. Crisis weeks are

shaded. In 1847we find some instanceswhere the Bank discounted at a rate below the

2

4

6

8

10

12

Sep-57 Oct-57 Nov-57 Dec-57

Percentbankrate Marketrates

2

3

4

5

6

7

8

9

10

11

Mar-66 Apr-66 May-66 Jun-66

PercentBankrate Marketrates

40

apparent‘minimum’BankRate.Theseareplottedasreddiamonds.Thishappenednotably

inOctober,atthepeakofthecrisis,whensomebillswerediscountedat5or5.5percentat

a timewhen headline Bank Rate stood at 6 percent. After 1847, transactionswhere the

Bankchargesacounterpartya ratebelowBankRatebecomerarer,andcease tooccur in

thedataby1866.ThishappensalongsideBankRatebecomingapenaltyrateabovetherate

prevailinginthemarketoncethecrisisoccurs,inaccordancewiththe‘Bagehotrule.’

Figure4.7:Spreadofdiscountratesin1847


2

3

4

5

6

7

8

9

10

11

Jan-47 Feb-47 Mar-47 Apr-47 May-47 Jun-47 Jul-47 Aug-47 Sep-47 Oct-47 Nov-47 Dec-47

Percent

rangeofledgerrates crisisweeksBankrate ratebelowBankrate

2

4

6

8

10

12

14


Percent

rangeofledgerrates crisisweeks Bankrate ratebelowBankrate

41


The1847 crisis standsout from the1857and1866 crises in another respect too.Figures4.10 to 4.12 compare Bank Rate to the average discount rate calculated from the daily

discountledgersweightedbythevalueofdiscounts.In1847,theweightedaveragediscount

rate series is volatile and does not neatly coincidewith advertised Bank Rate.Moreover,

duringcrisisweeks inMay,AugustandOctober thatyear, theeffectiveweightedaverage

discountrateisaround1to2percentaboveBankRate.Bycontrast,in1857and1866,the

weightedaveragediscountrateseries issmoother. ItmoreneatlyconformstoBankRate,

withnodiscernibledifferencesbetweencrisisandnon-crisisweeks.

Figure4.10:Dailyaverageratevs.Bankrateandmarketratefor1847

2

4

6

8

10

12

14


Percent

rangeofledgerrates crisisweeks Bankrate ratebelowBankrate

2

3

4

5

6

7

8

9

10

11

Jan-47 Feb-47 Mar-47 Apr-47 May-47 Jun-47 Jul-47 Aug-47 Sep-47 Oct-47 Nov-47 Dec-47

Percent

crisisweeks Bankrate weighteddailyaverage

42

Figure4.11:Dailyaverageratevs.Bankrateandmarketratein1857

Figure4.12:DailyaverageratevsBankrateandmarketratein1866

2

4

6

8

10

12

14


Percent

crisisweeks Bankrate weighteddailyaveragerate

2

4

6

8

10

12

14


Percent

crisisweeks Bankrate weighteddailyaveragerate

43

Section5:Goodsecurity

I Eligibility

Thebills of exchangepurchasedby theBankof Englandwereunsecureddebts.Although

these billsmight document the exchange of economic goods, they did not give the Bank

recourse to them.40 Therefore, the quality of a bill of exchangewas based on the Bank’s

judgment of the creditworthiness of the acceptor and subsequent endorsers of the bill.

Accordingtomosthistorians,inorderforabillofexchangetobeeligiblefordiscountatthe

Bank’sDiscountOffice, ithadtobeendorsedby two ‘good’Britishcounterparties,oneof

which had to be the acceptor (Ogden 1988: 185; Scammel 1968: 87). Getting an

endorsement on a bill of exchange from a ‘good name’ in the nineteenth century was

somewhat akin to receiving a tripleA rating froma credit rating agency today. In fact, it

meantevenmore,insofarasthosewhoendorsedbillsofexchangewereliableforpayment

of thedebt if theacceptordefaulted.However,whoexactlyqualifiedasa ‘goodname’ is

unclear as thiswas not codified in any systematicway, at least as far aswe know.Most

historians have assumed that ‘good names’meant prominentmerchant banks in London

(annex E). Their knowledge of particular trades meant that their judgment about the

financial health and business prospects of other firms could be trusted by the Bank.

However,aswediscussindetailbelow,wechallengethat‘goodnames’andtherefore‘good

security’wasdefinedsonarrowly.

Thedailydiscountledgersdonotprovideanydetailaboutthenamesunderwritingthebills

ofexchangepurchasedbytheBank.However,thisdetailcanbefoundinadifferentsetof

customer“withanduponledgers.”These ledgerscontaintwokindsof informationonthe

Bank’scounterparties.First, the ledgersdetaileachof the individualbills thataparticular

counterpartydiscounted“with”theBank.Second,theledgersdetailallbillsdiscountedby

theBankwhichweredrawn“upon”thatparticularcounterparty.Thesewerebillswherethe

counterpartywastheacceptorofthebill.AnnexEdescribesthecustomer“withandupon”

ledgersinmoredetail.Figure5.1isanexampleofanentryfromtheseledgers.Itrelatesto

BieberandCo.TheentryshowsallthebillswhichBieberandCoeitherdiscountedwiththe

Bank,orwhichwerepurchasedbytheBankfromanothercounterpartythatnamedBieber

andCo.astheacceptor.Whilethefirmwasfirstaddedtothecustomerledgerson5January

40 Bills of exchange could be collateralised with real property if they circulated alongside additional legaldocumentation.Thisoftenoccurredinthemarketforinternationalbillsofexchange.Forexample,anexportermightdrawabillonaforeignimporter.Whentheexportershippedtheirgoods,theywouldreceiveabilloflading from the shipping company.Theexportermight thendiscount thebill drawnon the importerwithabanker,handingoverthebillofexchange,thebillofladingandotherdocumentsassociatedwiththeexportofthegoods.Thebillofladingandassociateddocumentsgavethebankertitletothegoodsshipped(Leaf1926;Banks1999).

44

1865,thispageshowsbillsofexchangediscountedwithanduponBieber&Co.in1866.In

particular,thebillsintheboxedareasarethosepurchasedbytheBankfromBieber&Coon

11May1866,thedayafterOverend&Gurneyfailed.Thecustomer“withandupon”ledgers

canthereforebe linkedto thedailydiscount ledgers. In thecustomer ledger,weseethat

Bieber&Cobroughtin20billscollectivelyvaluedat£13,921on11May.Thiscorrespondsto

thecircledentryinthedailydiscountledgerinFigure5.2,showingBieberandCobroughtinapacketofbills,whichwerediscountedatarateof10%.

Figure5.1:AnexcerptfromBieber&Co’sledger—transactionson11May1866boxed

45

Figure5.2:Bieber&Co’scorrespondingdailydiscountledgertransactionon11May

Thevalueaddfromlookingatthecustomer“withandupon”ledgersisthattheyallowusto

‘unpack the packet’ of bills to understand, in granular detail, each bill of exchanged

containedwithin.Inthisonepacketalone,weseethatthedrawersofthesebillsarelocated

in a multitude of places including Singapore, Hamburg, New York, Madras and Rio. This

speakstothefactthattheforeignbillofexchangewasthedebtdejure inan increasingly

globalisedLondonmoneymarket.Equallydiversewasthenumberofacceptorsofthebills.

This diversity suggests that ‘good acceptors,’ and therefore ‘good names’ and ‘good

security,’ for the Bank was not narrowly limited to meaning a handful of City firms but

encompassedawidersetofeligiblecounterparties.

46

Figure5.3onthefollowingpagegivesasenseoftheoccupationaldiversityoftheacceptorsacceptabletotheBank.Thisinformationcomesfromanotherarchivalsource,theDiscount

Office’s ‘list of acceptors.’41 The figure lists firms which first appear in the Bank’s list of

approvedacceptorsduringthecrisisyearsof1847,1857and1866.Inorderforadebttobe

discounted,anacceptorhadtobe‘introduced’(thediscountoftheirdebtsupported)bya

Bank director or member of Court. In the original archival materials, the names of the

sponsoringseniormemberoftheBankappearnexttofirm’sname,occupationandaddress.

It is worth stressing that the list of acceptors in Figure 5.3 is only a subset of all the

acceptorsin1847,1857and1866whosedebtwasdiscounted.Thelistislimitedonlytonew

acceptors ‘introduced’ to theBank for the first timethatyear.Acceptorswhosedebthad

beendiscountedbytheBankpreviously,andwhomayhavehadmoreoftheirdebtbought

bytheBankduringcrisisyears,arenotlisted.Evenso,thisselectlistrevealsthatfirmsina

widevarietyofoccupationsbeyondmerchantbankscouldbeconsideredgoodcreditworthy

names.Theacceptors list includescabinetmakers, flax spinners,publishers,andumbrella

manufacturers.ThisevidencecorroboratesastatementmadebyJohnClapham,theBank’s

firstofficialhistorian.ReflectingontheBank’s lendingactivity inthefirst fewyearsofthe

nineteenthcentury,heobservedthat“themostremarkablefeaturesaboutthediscountsat

this time are the great number of the Bank’s discounting clients and the great range of

Londonbusinessrepresented” includingbakers,chinadealers,druggists,shipbuildersand

toymerchants(Clapham1944:205).ThisobservationholdstruewhenwelookattheBank’s

discountactivitiesatmid-century.

Thecustomer“withandupon”ledgersarevast.Becausethefocusofourresearchwason

the Bank’s daily discount ledgers, we have only scratched the surface of the wealth of

informationcontainedwithinthem. Inthispaper itwasthereforenotpossibletoquantify

theBank’srelativeexposuretomerchantbanksversusothertypesoffirms.However,inthe

spirit of Flandreau and Ugolini (2011), we did scan the “with and upon” ledgers and

identifiedcustomerswhoappearedtobethe largestacceptorsduringthe1847,1857and

1866crises.ThesearelistedinFigure5.4.Lookingacrosstime,thereappearstohavebeena

shift inthenatureof the largestacceptors. In1847, there isaroughlyevensplitbetween

merchantbanksandother typesof financial firms.However,by1866,weseecommercial

banksbecomingmore important acceptorsofbills. Thesearebolded. This is a somewhat

surprisingfindinggiventhepreviousliterature,thoughitaddstotheevidencepresentedin

inthissectionthat‘goodsecurity’wasnotasynonymformerchantbanks.

41BankofEnglandArchive,DiscountOffice,ListofAcceptors,1809-72,BoEC29/7-9.

47

Figure5.3:Selectlistofacceptorsandtheiroccupationsincrisisyears

1847 1857 1866

Name Occupation Name Occupation Name OccupationAbelBros GreekMerchants Bell&SonsAlex SpanishMerchants Abderden IronCompanyArbuthnotLatham&Co Merchants Bridgette&CoJ Silkmen Anderson&Sons RussiaMerchantsBegbie,Young&Begbies CornFactors BoydJohn IronMerchant AinsworthThos FlaxSpinnerBell&Hughes WineMerchant CoventryShephardCo CornFactors BeneckeDouchay&Co ConcessionMerchants

Betts&CoJ.J.RectifyingDistillersandMetallicCapsuleManufacturers

EngelhardJ Merchant Bell&Grant Merchants

BrooksRobert Merchant&Ship-owner FowlerH&R SlaveMerchant BlythGreen&Co Merchants

BolithoJ&W Merchants&TinSmeltersHansonSmiths&Stephens

Warehousemen BurtBolton&Haywood Timbercontractors

CheswrightSheffieldCo TimberMerchants Henderson&Constable WholesaleSugarDealers CatelliBrothers Merchants

ChildColes&Co CoalMerchants JoachimHenry WoolBroker CaldecottSons&Co Warehousemen

Cockburn&Co Merchants LairdW&H CoalMerchants CorryEdwards CopperMerchant

ComberRichard SilkBroker&Agent LockeLancaster&Co LeadMerchants Crosier&Pettigrew Warehousemen

CondellGeo.Smith W.J.Merchants MorganBrothers WineMerchantsDickson’sFerguson&Co

Manufacturers

Drakeford&CoD SilkBrokers Neville&Co Who.Hosiers Duncan&Co UmbrellaManufacturers

Forster&Smith Merchants Nicholls&Sons Warehousemen Evans&CoRichard TrimmingManufacturers

HenkelDuBuissonCo Merchants OrientalBank

ElderA.L. Merchant

LeGrosThompson&Bird GrapeManufacturers Palmer&Co PatentCandleManufacturerEnstrom,Browning&Co

Merchants

LuptonHooton&CoManchesterWarehousemen

RichardHy. AmericaMerchant GrantSon&CoAlex GravelManufacturers

Poland&Son Fur&SkinMerchants Pirie&Co Ship&InsuranceBroker Gibbs&SonsAntony Merchants

WalkersParker&co E.J.Merchants Powell&SonsRoanMakersandManufacturers

GoddardJ&J RussiaBrokers

RalliPaulaleonGouslantine

Merchant Hakim&CoA Merchants

Reid&Co WineMerchants Keeling&Sons WineandSpiritBrokers SadlerSand LinenFactor Kemp&Sons SilkManufacturers Spartali&Co Merchants KiplingPain&Co SilkManufacturers Smee&Sons CabinetMakers Keill&CoG.M. CapeMerchants Swonnell&Son Maltsters KoebelJameson&Co Merchants Stuart&Sharp Warehousemen Langton&Birkwells OilMerchants ScrutonSon&Co Ship&InsuranceBroker Lefevre&Co Merchants

VandesWilligenSimon Merchant&Comm'AgentMactaggartTidman&Co

E.J.Merchants

WhiteSon&Co Warehousemen MelasBrothers Merchants

MewsJohn TimberMerchants

Milne&Co Merchants

Mills&Halls ProvisionMerchants

Morata&Co Merchants

Noakes&Son HopFactors

Pearce&Co BlackwellHallFactors

PyeField&Co WineMerchants

Routledge&Sons Publishers

RobertHoar&Co MahoganyBrokers

RossGustier CommissionMerchant

Ross&Ash Merchants

Spalding&Hodge Who.Stationers

SatowH&J.T. Merchants

SaundersLindsay&Co AustraliaMerchants

Snellgrove&Leech Merchants

SpeyerBrothers Merchants

Sescan&Co Merchants

SimmondsHunt&Co Merchants

Silber&Fleming Warehousemen

Spotten&Co LinenManufacturers

TagartBryson&Slee TimberMerchants

WhaleyF.R. ColonialBrokers

William’sOverbury&Co

WoolBrokers

Wilson&CoImportersandManufacturers

WormsG&A Merchants YoungEhlers&Co Merchants

48

Figure5.4:TopAcceptorsinthe1847,57and66crises

October1847 November1857 May1866

BaringsBrothersandCo BaringBrothersandCo LondonJointStockBank

FruhlingandGoschen BieberandCo UnionBankofLondon

GlynandCo CHambroandSons TheNationalBank

HeathFurseandCo DraperPietroniandCo FruhlingandGoschen

FHuthandCo FinlayHodgsonandCo Agra&Masterman'sBank

JonesLoydandCo FHuthandCo TheCityBank

GLoder FruhlingandGoschen NorthWesternBank

MastermanandCo HSillemandSon London&CountyBank

NMRothschilds HavaandCo BaringBrothers&Co

SmithPayneandSmiths HoareBuxtonandCo RoyalBankofLiverpool

SchroderandCo JHSchroder DrakeKleinwort&Cohen

NMRothschild FHuth&Co

SievkingandSon FinlayHodgson&Co

CityofGlasgowBank

49

II Performance

Perhaps the best evidence that the Bank purchased good securities was their ex post

performance.Figure5.5showsbaddebtontheBank’sbooks.Write-offswereminimal. In

1847, the number of non-performing (late payment) loans spiked close to 6 percent.

However,in1857and1866,inspiteoftheuptick,theoveralllevelremainedcloseto1%.42

Relatedly,whilethenumberofdiscounterswhoseaccountsweresuspendedduetounpaid

billsincreasedduringeachcrisis,theywereafractionoftheoverallnumberofdiscounters.

Figure5.5:BaddebtontheBank’sbooks

42Bycomparison,duringthenadiroftherecentfinancialcrisis,theaverageNPLrateacrossUKbankswas3.5percent(Bholatetal.2016).

0% 1% 2% 3% 4% 5% 6%

1844 1846 1848 1850 1852 1854 1856 1858 1860 1862 1864 1866 1868 1870

Write-offsasapercentageoftotaldiscounts

Latepaymentasapercentageoftotaldiscounts

Figure5.6:Discounterswithsuspendedaccounts,crisisyearsinred

0

2

4

6

8

10

12

14

1844 1846 1848 1850 1852 1854 1856 1858 1860 1862 1864 1866 1868 1870

Percent

50

Infact,itappearsthattheBankreapedrewardsfromthegreaterrisksitincurredbyacting

asalenderoflastresortduringfinancialcrises.Figure5.7reveals,forthefirsttimepublicly,

the Bank’s profits and interest income during this period. Interest income from bills

discounted increased in the reporting periods immediately following each of the three

crises, reflecting a greater quantity of bills discounted at a higher rate of interest.43

Increases in interest income in turn increased profits. The Bank recorded profits semi-

annually, inFebruaryandAugusteachyear,aheadofpayingdividends to shareholders in

April and September, respectively. During this period, the Bank paid all of its profits to

shareholderseachperiod; thedividendpay-out ratiowas100percent, irrespectiveof the

stateoftheeconomy.44 InFebruary1848,followingthe1847crisis,and inFebruary1858,

followingthe1857crisis,profitsandthereforedividendsincreasednearly18%year-on-year.

InAugust1866,afterOverendGurney’scrisisinMaythatyear,profitsanddividendswent

up nearly 38% year-on-year. Letters exempting the Bank from note issuance restrictions

imposedby the 1844Bank CharterAct stipulated that theGovernment did notwant the

Banktoprofitfromhighratesofinterestchargedonloansduringcrises.Indeedtheletters

setouttheexpectationthattheBankshouldrecompensetheTreasuryifthehigherinterest

incomeexceededlossesonunpaidbills.However,AnsonandCapie(n.d.)findnoevidence

that theBankever transferredtheseprofitsacross to theTreasury. Instead, theyseemto

havebeentransferredtoBankshareholders.Likegoldtoday,BankofEnglandstockwasa

countercyclicalasset,thoughitremainsanopenquestionastowhetherinvestorsbackthen

perceivedthistobethecase.

43ThereportingperiodsafterthecrisesareFebruary1848,February1858andAugust1866.44RecallthatatthisstagetheBankwasprivatelyowned.TheBankwasnationalisedin1946.

Figure5.7Profitsandinterestincomefrombillsdiscounted,1845-1870

0

200,000

400,000

600,000

800,000

1,000,000

1,200,000

Feb1845

Feb1847

Feb1849

Feb1851

Feb1853

Feb1855

Feb1857

Feb1859

Feb1861

Feb1863

Feb1865

Feb1867

Feb1869

£Reportingperiodafterfinancialcrises Profits Interestonbillsdiscounted

51

Section6:Conclusion

I Summaryofempiricalfindings

OurpaperhasprovidednewempiricalevidenceontheoriginsandmechanicsoftheBankof

England’s role as a lender of last resort historically.We set out to establishwhether the

Bank acted in theway Bagehot prescribed a central bank should act i.e. (i) to lend cash

freely (ii) at a high or penalty rate (iii) in exchange for ‘good’ security, during the mid-

nineteenth century crises leadingup toBagehot’spublicationof LombardStreet.We find

thattheBank’sbehaviourevolvedtowardstheBagehotruleoverthisperiod.

On the first criterion,we find the Bank increased its propensity to lend freely across the

crisesof1847,1857and1866. InspectionoftheBank’sweeklybalancesheetrevealsthat

discountsandadvances increasedmarkedlyduringcrises. In termsof thepricingof these

loans,specificallywhethertheBankchargedhighorpenalrates,theevidencesuggeststhe

Bank’sattitudeevolvedtowardsBagehot’sprescription.Duringeachofthe1847,1857and

1866 crises, theBank increased its rate relative to thatwhichprevailedbefore.However,

whilein1857and1866Bankratewastypicallyhigherthancontemporaneousmarketrates,

in1847itwaslower.AnalysisoftheBank’sdailyledgerdatafromthe1847crisisalsoshows

that theBank lent at rates below its publicly advertised rate in someof its dealingswith

counterparties.Coupledwiththehigherloanrejectionratesobservedin1847,thissuggests

theBankwasstillrationingcreditatthispoint.Sotheperiodbetween1847and1866does

appeartomarkanevolutionintheBank’sbehaviourduringcrises.

Thefinalcriterionrelatedto‘goodsecurity’isthemostdifficultforustoassess.Noneofthe

discountedbillsofexchangesurvivesoweare forcedtorelyonthe informationtheBank

keptinitscustomerledgers.AndthesecuritiesonwhichtheBankmadeadvancesandthe

haircuts that were applied to that collateral are also unknown. However, we have some

indirect evidence that their quality was high.While bad debts did spike following crises,

particularlyin1847,theiroveralllevelremainedreasonable.Withtheexceptionofthe1857

crisis,theaveragematurityofthebillsofexchangepurchasedremainedlargelyunchanged,

andwellwithintheBank’sriskappetiteofupto95daysmaturity.Andfinancialcriseswere

generallyprofitablefortheBanksuggestingits‘collateral’managementwasgood.

Whileonbalance theempiricalevidencewehaveamassedsuggests theBank’sbehaviour

during these crises evolved towards the actions Bagehotwould have recommended, this

52

conclusion comes with a significant caveat. For Bagehot, it was crucial that the Bank

preannounce that it would behave in the way he prescribed before crises happened. By

doing so, Bagehot hoped that crises could be prevented, since the conversion of cash

equivalentsintocashwouldbepreordained,nullifyingtherationalefordepositorstorunin

thefirstplace(Humphry1989).However,likescholarsbefore,wefindnoevidencethatthe

Bank clearly communicated how it would act during financial crises at this stage in its

history.Thatsaid,throughitsrepeatedactions,theBankmayhavedoneenoughtosignalto

marketparticipantssuchthattheyknewexanteroughlyhowtheBankwouldrespondeven

if not exactly how. Inmodern terms, the central bank reaction functionwas clear if not

precisely defined. This might explain why the Bank was not required to increase its

discounts to thesamedegree in latercrisessuchas1878,1890and1907,althoughthere

arenaturallyotherreasonsforthisthatgobeyondthescopeofthispaper.

II Currentimplicationsofourfindings

Inhisrecentbook,formerBankofEnglandGovernorMervynKing(2016:202)haswritten

thatBagehot’s “maxim ‘lend freelyagainstgoodcollateralatapenalty rate’ isoutdated.”

AccordingtoKing,thisisbecausebankingassetsarenowmorecomplexandvariousthanin

Bagehot’s day, and because current disclosure requirements makes borrowing from the

centralbankatapenalrateprohibitiveforbanksbecauseofthestigmaattached.Inlieuof

Bagehot’sconceptionofalenderoflastresort,Kinginsteadproposesa‘pawnbrokerforall

seasons’willingtolendagainst‘poor’i.e.riskier,illiquidassetsbutwherebankswouldpre-

position thesewith the central bank so that their quality could be evaluated outside the

heatof a crisis andhaircutson thoseassetsestablishedwell in advance.Whiledeparting

from Bagehot, King’s proposal inmany respects alignswith recent central bank practice.

Duringthefinancialcrisisof2007-09,centralbanksacrosstheworldlentagainstasuiteof

securitiesbeyondthoseconventionallyclassifiedascentralbankeligibleandatrateslower

than those that prevailed before the crisis or in the market (Bholat 2014). As Ian

Plenderleith (2012: 14) observed in his review of the Bank of England’s provision of

emergency liquidityassistanceduring the recent financial crisis, “thenatureof theBank’s

lenderoflastresortfunctionhasbeenfundamentallytransformedsince2008.”And,itcould

beadded,sincethefinancialcrisesof1847,1857and1866.Yetinspiteofmanychangesto

bankingsystemsandhowlenderoflastresortoperationsworktoday,wethinkourpaper

offersinsightonseveralacademicandpolicyissuesstillcurrent(LeMauxandScialom2013).

The first issue relates to the types of institutions who can access lender of last resort

facilities. Before the 2007 financial crisis, direct access to Bank of England liquidity was

limited to fewer than twenty banks, on the premise that these banks could on-provide

53

liquiditytootherinstitutionsinacrisisifnecessary(Winters2012:83).However,duringthe

crisis, the interbank market ceased to operate normally, underscoring that, although in

theory,theprivatesectormightbeabletochannelcashtoilliquidbutsolventinstitutions,in

practiceitmightnotdosoinconditionsofheighteneduncertainty(Freixasetal.1999).Asa

result,theinstitutionsabletoaccessregularlystandingBankofEnglandliquidityfacilities45

is now much broader, and includes a range of non-bank counterparties such as broker-

dealers and central counterparty clearing houses (Bank of England 2015; Hauser 2016).

Indeed, in exceptional circumstances, subject to approval by the Chancellor of the

Exchequer,theBankcanalsoprovideemergencyliquidityassistancetoinstitutionsoutside

itsregulatoryperimeter(BankofEngland2012).

The broadening of access to lender of last resort loans during and after the crisis has

occurredacrosstheglobe.Butithasnotbeenwithoutitscritics.Somehavearguedthatthe

extension of lender of last resort facilities to non-banks is both undesirable and

unprecedented (Bordo2014).Whether it is desirableor not is a normative issueopen to

debate.Butwithrespecttoprecedenttherecanbenodoubt(cf.BuiterandSibert2007).On

thecontrary,mostoftheBank’slendingintheperiodwehavestudiedwasnottobanksbut

todiscounthouses,whosebalancesheetsmostcloselyresembledassetmanagersandother

so-called ‘shadow’banks.Aswehavedocumented, theBankpurchaseddebtdrawnona

rangeofcounterparties,includingmerchants,manufacturersandothernon-financialfirms.

Indeed it isworthrecallingthegeneralpreferenceexpressedbymonetarytheorists inthe

nineteenth century for the Bank to lend against bills of exchange documenting real

economic transactionsasopposedtoclaimssecuredon landor topurecash flows (Mints

1945).Insum,BankofEnglanddiscountsandadvanceswereneitherlimitedtobanksnorto

thefinancialsectorbutencompassedabroaderswatheoftheeconomy.

Asecondcontemporaryissueourpaperbearsonisthecreditqualityofthecounterparties

dealingwiththeBank.Ithaslongbeentheconventionalwisdomthatacentralbankshould

only lend to illiquidbutnot insolvent institutions. Indeed this is theBank’s currentpolicy

(Plenderleith 2012). However, as has often been noted, distinguishing illiquidity from

insolvency in crisis conditions can be difficult because of asymmetric information. For

example, a run on an institution might reflect underlying insolvency, or it might not.

Furthermore, fluctuations in the fair value of financial assets during a financial crisis can

temporarily exaggerate losses in amanner out of proportionwith themedium to longer

45TheBankprovidesthreeregularlystandingliquidityinsurancefacilities,eachofwhichallowmembersoftheSterlingMonetaryFrameworktoexchangelessliquidcollateralformoreliquidassets.Ofthese,theDiscountWindowFacilityprovidesaccesstobilateralliquiditysupporttofirmsexperiencinganidiosyncraticshock.Itallowsparticipantstoborrowhighlyliquidassetsinreturnforlessliquidcollateralatscaleandonvariableterms(BankofEngland2015).

54

termvalueof futurecash flows fromthem (Haldane2011). Illiquidity canquicklybecome

insolvency,andvice-versa(RochetandVives2004).

At thispoint it isuseful to recall thatBagehotnever specified counterparty solvencyas a

prerequisite for the Bank of England to grant loans (cf. Castiglionesi andWagner 2012).

Rather,as formerBankofEnglandMonetaryPolicyCommitteememberCharlesGoodhart

(1999)oncepointedout,Bagehotrequiredonlythatthecounterpartyhave‘goodsecurity.’

Thefocusonthe‘security’orcollateraloftheapplicantasopposedtotheircreditworthiness

probably had to do with the special nature of bills of exchange. While bills were

underwrittenbythefirmsapplyingtotheBankfortheirdiscount,theywerenotpayableby

that firm in the first instance. Instead these were debts owed by third-party acceptors.

Moreover, thebillswereguaranteedbyotherpartiesbesides theapplicantdiscounter, in

most cases requiring that theywere underwritten by ‘good names’—firmswith specialist

knowledgeofthegeographyorproductspaceinwhichtheultimatedebtoroperated.Given

thesafeguardsinplacebackthen,distinguishingbetweensolventandinsolventapplicants

may not have been the Bank’s central preoccupation, though it still may have been an

importantfactor.Instead,perhapslikeMervynKing’sproposedpawnbroker,whatmattered

mostwasacounterparty’s‘collateral’,nottheirnetworth.

Athirdissueonwhichourpaperoffersfreshinsightrelatestothepurposeoflenderoflast

resort operations. There is a long anddistinguished line of thinking dating back toHenry

Thornton(1803)whichargues for lenderof lastresortoperationsonessentiallymonetary

(pricestability)grounds.Theargumentisthatifcashequivalentsareconvertedtocashand

hoarded,thenpriceswillfall,leadingtodeclinesinoutput,employmentandrealeconomic

activity,withcontractionsofbank lendingpromptedby theseconditions furthershrinking

the money supply in an iterative fashion, leading to a downward deflationary spiral

(FriedmanandSchwartz1963).Accordingtothisschoolofthought,thisjustifiesthecentral

bankinterveningtopropupbroadmoneytomaintainpricestability.Sincethejustification

forlenderoflastresortfacilitiesisessentiallymacro-economic,itfollowsthatthereislittle

enthusiasmamongthinkersinthisschoolforprovidingbilateralassistancetoafirmfacing

an idiosyncratic liquidity shock. On the contrary, it is argued that to do sowould induce

moralhazard.Bankswillbehavelessprudentlye.g.holdinglesscashandreadilyrealisable

assetsiftheyknowtheycantapliquidityfromthecentralbankondemand.46Asaresult,so

theargumentgoes,bilateralliquiditysupportwouldinducetheveryriski.e.bankilliquidity

46Instead,itisarguedthatfirmsshouldself-insureagainstfirm-specificliquidityshocksbyholdingcashandotherreadilyrealisableassets,or,failingthose,borrowingfromotherbanks.

55

that it is meant to mitigate (Selgin 1989).47 Instead, liquidity should be provided to the

‘market’throughchannelslikeanonymousauctions.

Howeversoundmanyaspectsofthis lineofthinkingare,theactionsweobservetheBank

taking in the mid-nineteenth century lead us to conclude that the finer points of these

argumentswerestillunclear tohistorical contemporaries.Whenpushcame to shove ina

crisis,pricestabilityrankedsecondtopreservingfinancialstability.Thisisclearfromthefact

thatduringmid-nineteenthcenturycrises,the1844BankCharterAct,whichanchoredthe

Bank’snote issueto itsgoldreservesandwasbelievedbycontemporariestobethebasis

forensuringmonetarystability,wasrepeatedlysuspended.

Moreover, the Bank appears to have been more willing to provide liquidity support to

particularfirmsthan latercentralbanktheoristswouldcountenance.Thiscanbededuced

fromtheskeweddistributionindiscount loans.TheBankwasnot lendinganonymouslyto

the ‘market’ (cf. Capie 2007; Wood 2007). On the contrary, throughout the nineteenth

century, it was developing an increasingly elaborate set of ledgers for recording and

monitoring its loans to various counterparties, including several large single-name

exposures. Coupled with the knowledge and experience of seasoned staff such as the

Principal of theDiscountOffice,Mr. Elsey, andwell-connected directors, the Bankwould

havehadsomeinformationtodrawconclusionsonwhowas‘goodsecurity.’Still,thisistrue

onlyuptoapoint.Forexample,thereisnolistwehavebeenabletofindsummarisingto

which acceptors the Bank was most exposed. Nor do we have evidence that these

exposuresweremonitoredtokeepcreditexposuretoparticulargeographiesor industries

within certain limits. We have also found no evidence that the Bank used the ledger

informationtonetoffobligations,whichcouldhavereducedtheircumulativesizeandthe

risk of default by the Bank’s counterparties. In the round, this raises doubts about how

muchmonitoringofitscreditexposurestheBankactuallydid.

Finally,ata timewhenour currentphaseofglobalization is subject to increasing scrutiny

(Carney 2016), it is worth emphasising that financial globalization is hardly new. Casual

inspection of the Bank’s customer ledgers confirms that the Bank discounted debts

originating from all corners of the globe, from Birmingham to Bombay, Canton to Cape

47Thereisanevenmoreradicallineofthinkingopposedtolenderoflastresortlendingbycentralbanks,fullstop.Fromthisperspective,lenderoflastresortfacilitiesareasortofsubsidythatbenefitbanksandtheirclaimholders;centralbanksupportdiminishesthelikelihoodofdefault,resultinginlowerfundingcostsandhigherprofitsforbanksthanwouldotherwisebethecase(Ricks2016).Andlikeallsubsidiestheexistenceoflenderoflastresortfacilitiesispurportedtofueltheover-productionofthegoodsubsidised(inthiscase,bankingassets),makingthebankingsystembiggerandmorerisky.

56

Town. Theglobaldiversificationof theBank’sdiscountsmayhave contributed to the low

levels of bad debtweobserve in the ledger data, thoughdrawing a definitive conclusion

would require testing against other candidate explanations. So, viewed from a longer

historicalstandpoint, theBank’sdecisionsduringthemostrecentcrisis to lendagainstUS

agency debt, or to lend in dollars, for example, look less like departures from central

bankingnormsthantherenewalofanoldertraditioncharacterisedbytheBankactingnot

onlyasalenderoflastresortintheUKbutasastabilisingforceinthewiderglobalfinancial

system.

57

References

Anson,MikeandForrestCapie.n.d.ProfitsandprofitabilityintheBankofEngland,1694to1980.

Unpublishedmanuscript.

Ashton,T.S.1953.TheBillofExchangeandPrivateBanksinLancashire,1790-1830.InPapersin

EnglishMonetaryHistory,T.S.AshtonandR.S.Sayers,eds.Oxford:ClarendonPress.

Bagehot,Walter.1873[2006].LombardStreet:ADescriptionoftheLondonMoneyMarket.New

York:Cosimo.

Bagshaw,JohnFrancisGomm.1920.PracticalBanking.London:SirIsaacPitman&SonsLimited.

Banks,Erik.1999.TheRiseandFalloftheMerchantBanks.London:KoganPage.

BankofEngland.1967.TheLondondiscountmarket:somehistoricalnotes.BankofEngland

QuarterlyBulletin7:144-156.

BankofEngland.2012.MemorandumofUnderstandingonfinancialcrisismanagement.Availableat

http://www.bankofengland.co.uk/about/Documents/mous/moufincrisis.pdf

BankofEngland.2015.TheBankofEngland’sSterlingMonetaryFramework.London:Bankof

England.

Bholat,David.2014.Theinterwargoldstandardinlightofthepresent.JournalofCommonMarket

Studies52(4):711-712.

Bholat,David.2014.Money,bankdebtandbusinesscycles:betweeneconomicdevelopmentand

financialcrises.Pp.11-32inTheAvailabilityofCreditandSecuredTransactionsinatimeofcrisis.

Cambridge:CambridgeUniversityPress.

Bholat,David,Lastra,Rosa,Markose,Sheri,Miglionico,AndreaandKallolSen.2016.Non-performing

loans:regulatoryandaccountingtreatmentsofassets.BankofEnglandStaffWorkingPaper594.

Bignon,Vincent,Flandreau,MarcandStefanoUgolini.2012.Bagehotforbeginners:themakingof

lender-of-last-resortoperationsinthemid-nineteenthcentury.EconomicHistoryReview65(2):580-

608.

Bignon,VincentandClemensJobst.2017.EconomicCrisesandtheEligibilityfortheLenderofLast

Resort:Evidencefrom19thcenturyFrance.BanquedeFranceWorkingPaper618.

58

Bordo,Michael.2014.Rulesforalenderoflastresort:Anhistoricalperspective.Paperpresentedat

theHooverInstitution,Stanford,California,29-30May.

Bordo,MichaelandEugeneWhite.1991.Ataleoftwocurrencies:BritishandFrenchfinanceduring

theNapoleonicWars.TheJournalofEconomicHistory,51(02):pp.303-316.

Buiter,WillemandAnneSibert.2007.TheCentralBankasMarketMakeroflastresort1.Financial

Times12August.

Brunnermeier,Markus.2009.DecipheringtheLiquidityandCreditCrunch2007-08.Journalof

EconomicPerspectives23(1)Winter:77-100.

Burdekin,RichardandMericKeskinel.2013.Liquiditypreferenceandinterest-bearingmoney:the

OttomanEmpire,1840-1851.FinancialHistoryReview20(1):91-102.

Button,Richard,Knott,Samuel,McManus,ConorandMatthewWillison.2015.Desperate

adventurersandmenofstraw:thefailureoftheCityofGlasgowBankanditsenduringimpacton

theUKbankingsystem.BankofEnglandQuarterlyBulletin55(1):23-35.

Calomiris,Charles.2010.Bankingcrisesyesterdayandtoday.FinancialHistoryReview17:3-12.

Calomiris,CharlesandStephenHaber.2015.FragilebyDesign:ThePoliticalOriginsofBankingCrises

andScarceCredit.Princeton:PrincetonUniversityPress.

Capie,Forrest.2007.TheemergenceoftheBankofEnglandasamaturecentralbank..Pp.297-316

inTheLenderofLastResort,ForrestCapieandGeoffreyWood,eds.London:Routledge.

Carney,Mark.2016.TheSpectreofMonetarism.SpeechgivenatLiverpoolJohnMooresUniversity

onDecember5.Availableat:

http://www.bankofengland.co.uk/publications/Pages/speeches/2016/946.aspx

Castiglionesi,FabioandWolfWagner.2012.TurningBagehotonhishead:Lendingatpenaltyrates

whenbankscanbecomeinsolvent.JournalofMoney,CreditandBanking44(1):201-219.

Chadha,JagjitandElisaNewby.2013.‘Midas,transmutingall,intopaper:’theBankofEnglandand

theBanquedeFranceduringtheNapoleonicWars.CambridgeWorkingPapersinEconomics1330.

Chandler,Alfred.1994.ScaleandScope:DynamicsofIndustrialCapitalism.Cambridge:Harvard

UniversityPress.

Chapman,Stanley.1984.TheRiseofMerchantBanking.London:GeorgeAllenandUnwin.

59

Chubb,Hammond.1872.TheBankActandtheCrisisof1866.JournaloftheStatisticalSocietyof

London35(2):171-195.

Clapham,John.1944.TheBankofEngland:AHistoryVolume2,1797-1914.Cambridge:Cambridge

UniversityPress.

Collins,Michael.2012.MoneyandbankingintheUK:Ahistory.London:Routledge.

Congdon,Tim.2009.Centralbankinginafreesociety.London:InstituteofEconomicAffairs.

Constancio,Vitor.2015Financialstabilityrisks,monetarypolicyandtheneedformacroprudential

policy.SpeechatWarwickEconomicsSummiton15February.

Dimsdale,NicholasandAnthonyHotson.2014.BritishFinancialCrisessince1825.Oxford:Oxford

UniversityPress.

Domanski,Dietrich,Moesnner,RichhildandWilliamNelson.2014.Centralbanksaslendersoflast

resort:experiencesduringthe2007-10crisisandlessonsforthefuture.BISPapersNo.79Re-

thinkingthelenderoflastresort.Basel:BankforInternationalSettlements.

Eichengreen,Barry.2015.HallofMirrors:TheGreatDepression,TheGreatRecessionandtheUses-

and-MisusesofHistory.Oxford:OxfordUniversityPress.

Elliot,Nicholas,Odgers,John,andJonathanPhillips.2013.BylesonBillsofExchangeandCheques

29thEdition.London:Sweet&Maxwell.

Evans,D.Morier.1848.TheCommercialCrisis1847-1848.London:Letts,SonandSteer.

Ferguson,Niall,Schaab,Andreas,andMoritzSchularick.2015.Centralbankbalancesheets:

Expansionandreductionsince1900.CentreforEconomicPolicyResearchNo.10635.

Fetter,Frank.1965.DevelopmentofBritishMonetaryOrthodoxy1797-1875.Harvard:Harvard

UniversityPress.

Flandreau,MarcandStefanoUgolini.2011.WhereItAllBegan:LendingofLastResortandtheBank

ofEnglandDuringtheOverend,GurneyPanicof1866.EuropeanHistoricalEconomicsSociety

WorkingPapersinEconomicHistoryNo.7.February2011.

Flandreau,MarcandStefanoUgolini.2014.TheCrisisof1866.Pp.76-93inBritishFinancialCrises

since1825,Dimsdale,NandHotson,A,eds.Oxford:OxfordUniversityPress.

60

Fletcher,Gordon.1976.TheDiscountHousesinLondon:Principles,OperationsandChange.London:

MacmillanPress.

Freidman,MiltonandAnnaSchwartz.1963.AMonetaryHistoryoftheUnitedStates1867-1960.

Princeton:PrincetonUniversityPress.

Freixas,Xavier,Curzio,Giannini,Hoggarth,GlennandFaroukSoussa.1999.Lenderoflastresort:a

reviewoftheliterature.BankofEnglandFinancialStabilityReviewNovember:151-167.

Geva,Benjamin.2011.ThePaymentOrderofAntiquityandtheMiddleAges:ALegalHistory.Oxford:

HartPublishing.

Goodhart,Charles.1995.TheCentralBankandtheFinancialSystem.London:MacmillanPress.

Goodhart,Charles.1999.MythsabouttheLenderofLastResort.InternationalFinance2(3):339-360.

Goodhart,CharlesandGerhardIlling.2002.FinancialCrises,ContagionandtheLenderofLastResort:

AReader.Oxford:OxfordUniversityPress

Gorton,Gary.2012.MisunderstandingFinancialCrises:WhyWeDon’tSeeThemComing.Oxford:

OxfordUniversityPress.

Gorton,GaryandAndrewMetrick.2012.SecuritisedBankingandtherunonrepo.Journalof

FinancialEconomics104(3):425-451.Gorton,GaryandOrdonez,Guillermo.2014.CollateralCrises.

AmericanEconomicReview104(2):343–378.

Grossman,RichardandHughRockoff.2015.Fightingthelastwar:economistsonthelenderoflast

resort.CentreforEconomicPolicyResearchWorkingPaper10361.

Haldane,Andrew.2011.Accountingforbankuncertainty.SpeechattheInformationforBetter

MarketConferences,InstituteofCharteredAccountantsinEnglandandWales,London,19

December.

Hauser,Andrew.2016.Betweenfeastandfamine:transparency,accountabilityandtheLenderof

LastResort.SpeechgivenattheCommitteeonCapitalMarketsRegulationconference,‘TheLender

ofLastResort:aninternationalperspective,’10February.

Hawtrey,Ralph.1932(2007).Theartofcentralbanking:thelenderoflastresort.Pp.183-198inThe

LenderofLastResort,ForrestCapieandGeoffreyWood,eds.London:Routledge.

61

Hills,Sally,Thomas,RylandandNicholasDimsdale.2010.TheUKrecessionincontext-whatdothree

centuriesofdatatellus?BankofEnglandQuarterlyBulletin50(4):271-291.

Huang,HuaxiangandRylandThomas.2016a.Theghostofcrisespast,presentandfuture:TheBank

CharterActgoesontrialin1847.BankUnderground,19December.Availableat:

https://bankunderground.co.uk/2016/12/19/the-ghost-of-crises-past-present-and-future-the-bank-

charter-act-goes-on-trial-in-1847/

Huang,HuaxiangandRylandThomas.2016b.TheBankofEnglandWeeklyBalancesheet.OBRA

dataset.Availableat

http://www.bankofengland.co.uk/research/Pages/datasets/default.aspx#balancesheet

Hughes,J.R.T.1956.TheCommercialCrisisof1857.OxfordEconomicPapers8(2):194-222.

Humphry,Thomas.1989.TheLenderofLastResort:Theconceptinhistory.FederalReserveBankof

RichmondEconomicReviewMarch/April:8-16.

Holden,JamesMilnes.1955.AHistoryofNegotiableInstrumentsinEnglishLaw.London:Athlone

Press.

King,WilfredThomasCousins.1936.HistoryoftheLondonDiscountMarket.London:George

RoutledgeandSons.

King,Mervyn.2016.TheEndofAlchemy:Money,BankingandtheFutureoftheGlobalEconomy.

London:Little,BrownBookGroup.

Kiremire,Ankunda.2011.TheapplicationoftheParetoprincipleinsoftwareengineering.Louisiana

TechUniversity:Ruston,LA.

Knyaston,David.1994.TheCityofLondonVolume1:AWorldofItsOwn1815-1890.London:Chatto

&Windus.

Kosmetatos,Paul.2014.Financialcontagionandmarketinterventioninthe1772-73creditcrisis.

CambridgeWorkingPapersinEconomicandSocialHistory21.

Kuttner,Kenneth.2010.Victorianfinancialcrisesandtheirimplicationsforthefuture.Business

Economics45(2):102-109.

Lamoreaux,Naomi.1994.InsiderLending:Banks,personalconnections,andeconomicdevelopment

inindustrialNewEngland.Cambridge:CambridgeUniversityPress.

62

Leaf,Walter.1926(1999).Banking.ReprintedinBankingTheory1870-1930Volume7,ForrestCapie

andGeoffreyWoods,eds.London:Routledge.

LeMaux,LaurentandLaurenceScialom.2013.Centralbanksandfinancialstability:Rediscovering

thelender-of-last-resortpracticeinafinanceeconomy.CambridgeJournalofEconomics37:1-16.

Lewis,John.2016.Untousalenderoflastresortisborn:OverendGurneygoesbustin1866.Bank

Underground,21December.Availableat:https://bankunderground.co.uk/2016/12/21/unto-us-a-

lender-of-last-resort-is-born-overend-gurney-goes-bust-in-1866/

Lovell,Michael.1957.TheRoleoftheBankofEnglandasLenderofLastResortintheCrisesofthe

EighteenthCentury.ExplorationsinEntrepreneurialHistory10October:8-21.

Martin,Antoine.2009.ReconcilingBagehotwiththeFed'sresponsetoSeptember11.StaffReports

217FederalReserveBankofNewYork.

Mehrling,Perry.2011.TheNewLombardStreet:HowtheFedbecametheDealerofLastResort.


Mints,Lloyd.1945.AHistoryofbankingtheoryinGreatBritainandtheUnitedStates.Chicago:

UniversityofChicagoPress.

Morys,Matthias.2014.GoldstandardlessonsfortheEurozone.JournalofCommonMarketStudies

52(4)June:728-741.

Moxon,Thomas.1894.EnglishPracticalBanking.London:Routledge.

Neal,Larry.1990.TheRiseofFinancialCapitalism:InternationalCapitalMarketsintheAgeof

Reason.Cambridge:CambridgeUniversityPress.

Neumann,Tobias.2016.TheNightmarebeforeChristmas:Financialcrisesgoglobalin1857.Bank

Underground,20December.Availableat:https://bankunderground.co.uk/2016/12/20/the-

nightmare-before-christmas-1857-financial-crises-go-global/

Nishimura,Shizuya.2010.TheDeclineofInlandBillsofExchangeintheLondonMoneyMarket1855-

1913.Cambridge:CambridgeUniversityPress.

O’Brien,Denis.2003.TheLender-of-Last-ResortConceptinBritain.HistoryofPoliticalEconomy

35(1):1-19.

O’Brien,PatrickandNunoPalma.2016.DangertotheOldLadyofThreadneedleStreet?TheBank

RestrictionActandtheregimeshifttopapermoney,1797-1821.EuropeanHistoricalEconomics

SocietyWorkingPapersinEconomicHistory100.

Ogden,E.M.1988.ThedevelopmentoftheroleoftheBankofEnglandasaLenderofLastResort,

1870-1914.UnpublishedDoctoralThesis:CityUniversityLondon.

63

Ogden,Tessa.1991.AnanalysisoftheBankofEngland’sdiscountandadvancebehaviour1870-

1914.InNewperspectivesonthelateVictorianeconomy:essaysinquantitativeeconomichistory

1860-1914.JamesForeman-Peck,editor.Cambridge:CambridgeUniversityPress.

ParliamentaryCommissiononBankingStandards.2013.Changingbankingforgood.Volume1:

Summary,andConclusionsandrecommendations.London:HouseofCommons.

Plenderleith,Ian.2012.ReviewoftheBankofEngland’sProvisionofEmergencyLiquidityAssistance

in2008-09.London:BankofEngland.

Pozsar,Zoltan.2014.ShadowBanking:TheMoneyView.OfficeofFinancialResearchWorkingPaper

14.

Presnell,Leslie.1956.CountrybankingintheIndustrialRevolution.Oxford:OxfordUniversityPress.

Quinn,StephenandWilliamRoberds.2015.RespondingtoaShadowBankingCrisis:TheLessonsof

1763.JournalofMoney,CreditandBanking47(6)September:1149-1176.

Reinhart,CarmenandKenRogoff.2009.ThisTimeisDifferent:EightCenturiesofFinancialFollies.


Ricks,Morgan.2016.TheMoneyProblem:RethinkingFinancialRegulation.Chicago:Universityof

Chicago.

Roberts,Richard.1993.What’sinaName?Merchants,MerchantBankers,AcceptingHouses,Issuing

Houses,IndustrialBankersandInvestmentBanks.BusinessHistory35(3)July:22-38.

Rochet,Jean-CharlesandXavierVives.2004.Coordinationfailuresandthelenderoflastresort:Was

Bagehotrightafterall?JournaloftheEuropeanEconomicAssociation2(6):1116-1147.

Santarosa,Veronica.2015.FinancingLong-DistanceTrade:TheJointLiabilityRuleandBillsof

ExchangeinEighteenth-CenturyFrance.TheJournalofEconomicHistory75(3):690-719.

Sayers,Richard.1936.BankofEnglandOperations1890-1914.London:KingandSons.

Sayers,Richard.1957.CentralBankingafterBagehot.Oxford:OxfordUniversityPress.

Scammell,W.M.1968.TheLondonDiscountMarket.NewYork:St.Martin’sPress.

Selgin,George.1989.Legalrestrictions,financialweakeningandthelenderoflastresort.Cato

Journal9(2):429-469.

Shin,Hyun.2009.ReflectionsonNorthernRock:TheBankRunthatHeraldedtheGlobalFinancial

Crisis.JournalofEconomicPerspectives23(1)Winter:101-119.

Sowerbutts,Rhiannon,Schneelberg,Marco,andFlorenceHubert.2016.ThedemiseofOverend

Gurney.BankofEnglandQuarterlyBulletinQ2:94-106.

Taylor,James.2006.Creatingcapitalism:JointstockenterpriseinBritishpoliticsandculture,1800-

1870.Woodbridge:RoyalHistoricalSociety.

64

Temin,PeterandHans-JoachimVoth.2008.InterestRateRestrictionsinaNaturalExperiment:Loan

AllocationandtheChangeintheUsuryLawsin1714.TheEconomicJournal118(528),pp.743-758.

Thomas,RylandandNicholas,Dimsdale(2017).AmillenniumofUKmacroeconomicdata.Bankof

EnglandOBRAdataset.

http://www.bankofengland.co.uk/research/Pages/onebank/threecenturies.aspx

Thornton,Henry.1965(1802).AnEnquiryintothenatureandeffectsofthepapercreditofGreat

Britain.NewYork:AugustusKelley.

Truptil,RogerJean.1936.BritishBanksandtheLondonMoneyMarket.London:AldenPress.

Turner,Adair.2016.Betweendebtandthedevil:Money,creditandfixingglobalfinance.Princeton:

PrincetonUniversityPress.

Turner,John.2014.Bankingincrisis:TheriseandfallofBritishfinancialstability,1800tothe

present.Cambridge:CambridgeUniversityPress.

Tucker,Paul.2014.Thelenderoflastresortandmoderncentralbanking:principlesand

reconstruction.BISPapersNo.79Re-thinkingthelenderoflastresort.Basel:BankforInternational

Settlements.

White,Eugene.2016.RescuingaSIFI,HaltingaPanic:theBaringsCrisisof1890.BankUnderground,

10February.Availableat:https://bankunderground.co.uk/2016/02/10/rescuing-a-sifi-halting-a-

panic-the-barings-crisis-of-1890/

Winters,Bill.2012.ReviewoftheBankofEngland’sFrameworkforprovidingliquiditytothebanking

system.London:BankofEngland.

Wood,Geoffrey.2007.Thelenderoflastresortreconsidered.Pp.367-395inTheLenderofLast

Resort,ForrestCapieandGeoffreyWood,eds.London:Routledge.

Wood,John.2003.Bagehot’slenderoflastresort:Ahollowhallowedtradition.IndependentReview

VII(3):343-351.

65

AnnexA:TheoriginsoftheBankasalenderoflastresort

ItishardtosayexactlywhentheBankofEnglandstartedtoperformitslenderoflastresort

role.Originally,theBankwassetupin1694tomanagethegovernment’sdebt.Lendingto

theprivatesectorwasnot itsprimarypurpose.Evenso,Lovell (1952)arguedthatthere is

evidencethattheBankofEnglandactedasa lenderof lastresortduringsomeeighteenth

centurycrises(seealsoKosmetatos2014).Whilethereislittlehighfrequencydatafromthis

periodtovalidateLovell’sclaim,annualdataontheBank’sholdingofprivatesecuritiesand

advances, coupled with Lovell’s own calculations based on the average revenue from

discountsandadvances,drawingonClapham(1944),doessuggestthattheBankincreased

itslendingduringsomefinancialcrisespriortotheFrenchwars(1793to1815).FigureA.1chartstheempiricalevidence.

FigureA.1:TheBankofEngland’slendingtotheprivatesector,1696-1797

However,theBank’sabilitytofullyactasa lenderof lastresortatthisstage in itshistory

was constrained because trust in Bank of England noteswas still developing. Bank notes

circulatedalongsideothercompetingcurrenciesincludingcoin,billsofexchange,andnotes

issued by other banks. Consequently, the conversion of Bank notes into gold happened

morefrequentlythaninlaterperiods.ThisconstrainedtheamountofnewnotestheBank

could possibly issue to support financial firms in distress because of the commitment to

redeemall notes in bullion.FigureA.2 shows that during several late eighteenth centurycrises,theBank’sbullionreservefelltodangerouslylowlevelscomparedtoitsliabilities.

0

5

10

15

20

25

30

35

40

1696 1706 1716 1726 1736 1746 1756 1766 1776 1786 1796

Datesoffinancialcrises SecuritiesotherthanGovernment

Lovell(1952)averagevolumeofdiscounts%oftotalassets

66

FigureA.2:TheBank’sbullionreserve,1696-1797

Note:Nobalancesheetdataareavailablein1765and1774.

Akeyturningpointoccurredin1797.FollowingadrainofbullionfromtheBankdrivenby

fearsof an impendingFrench invasion, thegovernmentordered theBank to suspend the

convertibilityofitsnotesintogold.48Thereafter,thelinkbetweenBanknotesandgoldwas

broken.Althoughitwasre-establishedin1821,astructuralbreakseemedtohaveoccurred

from1797,withtheBankrespondingtofinancialcrisesbylendingmorefreelythereafter.49

FigureA.3,basedonmonthlydata,showsthattheBankincreaseditsdiscountingofbillsin

eachofthecrisesof1797,1810,1825and1837.50

48Therationaleforthesuspensionwasdebatedatthattimeandhasbeendebatedsincebyscholars(BordoandWhite1991;ChadhaandNewby2013;O’BrienandPalma2016).49 The note issue of the Bank increased in the immediate aftermath of the restriction. Lending to thegovernment was entirely short-term and modest compared to the overall increase in government debt.Althoughtherewasalsoanincreaseinlendingtotheprivatesector,muchofthiswasreignedinafter1810,inresponsetoBullionistcriticswhoarguedforareturntothegoldstandard.Atonelevel,therestrictionperiod(1797-1821) could be viewed as a successful fiat money experiment. The price level and interest ratesremainedrelativelystable,fiatmoneyremainedincirculationandthegoldstandardwasresumedin1821atthe pre-war parity. The government also ran primary surpluses that helped reduce the government debtburdeninthepost-warperiod.50Themost seriousof thesecrisesoccurred in1825 (Turner2014).Aquote fromJeremiahHarman,aBankDirector,cited inBagehot’sLombardStreet, isoftenadducedtoshowhowfreely theBankwaspreparedtolend:“Welentit[thepublic]byeverypossiblemeansandinmodeswehadneveradoptedbefore;wetookinstockonsecurity,wepurchasedExchequerbills,wemadeadvancesonExchequerbills,wenotonlydiscountedoutright,butwemadeadvancesonthedepositofbillsofexchangetoanimmenseamount,inshort,byeverypossiblemeansconsistentwiththesafetyoftheBank,andwewerenotonsomeoccasionsover-nice.Seeingthe dreadful state in which the public were, we rendered every assistance in our power…” Even so, laterscholarshavecriticisedtheBankforbeingtoosluggishinitsinitialreluctancetoexpandlending(DimsdaleandHotson2014).

0

20

40

60

80

100

1696 1706 1716 1726 1736 1746 1756 1766 1776 1786 1796

Financialcrises %oftotalliabilties

67

FigureA.3:TheBankofEngland’slendingtotheprivatesector,monthlyseries,1794-1844

0

20

40

60

80

100

1794 1799 1804 1809 1814 1819 1824 1829 1834 1839 1844

Notesandbillsdiscountedasa%oftotalshorttermassets

Financialcrises

68

AnnexB:Definitionsofkeynineteenthcenturymoneymarketterms51

Billofexchange Awrittenorderrequiringthepersontowhomitisaddressedtopaya

specificpersonorthebearer(holder)ofthebillofexchange

Orderbill Abillofexchangeorderingpaymenttoaspecificpayee

Bearerbill Abillofexchangeorderingpaymenttowhoeverbearsthebill

Drawer Thepersonorderingpayment

Drawee Thepersonrequiredtopay

Acceptor Thelegalnameforadraweeaftertheysigntheirnametoabill

Payee Apersonnamedbythedraweronthebillwhothedraweemustpay

Transferee Apersonreceivingabille.g.aspaymentforgoodsandservices

Holder Thepersoninpossessionofthebillakathebearer

Endorser Apersonwhouses thebill forpaymentmust sign (endorse) thebill

therebyincurringliabilityforitsrepaymentatmaturity

Usance Thelengthoftimeuntilabillofexchangematures

Currency SeeUsance

Inlandbill Abillofexchangeinvolvingdomesticcounterparties

Foreignbill Abillofexchangeinvolvingforeigncounterparties

Realbill Abillofexchangearisingfromarealeconomictransaction

Accommodationbill Abillofexchangecreatedforfinancingpurposes

Financebill SeeAccommodationbill

Cheque Abillofexchangedrawnonabankerpayableondemand

51SourcesincludeBagshaw(1920),Elliotetal.(2013)andScammell(1968).

69

AnnexC:Accommodationbills

Bills of exchange could increase the quantity of money in circulation both directly as a

supplementtocash,andindirectlywhen,forexample,banksissuednewnotesinthecourse

of purchasing bills of exchange. This concerned many nineteenth century monetary

theorists.Theprevailing‘realbillsdoctrine’heldthatnoeconomicharm,suchasinflation,

emanated from bills of exchange, so long as they were trade bills representing real

economictransactions(Mints1945).However,contemporarieswereconcernedthatnotall

billswere‘real.’Thismightbesoforoneoftworeasons.First,somebillswere‘fictitious’in

thatthedocumentshadbeenforged.Forexample,apersonmightdrawabillonafictitious

draweeinordertousethebilltoobtaincashthroughitsoutrightsaletoadiscounthouseor

bank,orthroughtheuseofthefictitiousbillascollateralforanadvance.Alternatively,even

iftheallthecounterpartiesinthetransactionwerereal,notallbillsofexchangereflected

realeconomytransactions.Instead,theymightbewhatwereknownas‘accommodation’or

financebills.Forexample,adraweemightacceptabilldrawnonthem,evenifthatacceptor

didnotreceivegoodsandservicesfromthedrawer.Thedrawercouldthenusethebillof

exchangeasameans forobtainingcash froma financial firm.Thedrawer could thenuse

thatcashforoperationsandinvestmentsoverthetermofthebillofexchange,remittingan

amountequaltothefacevalueofthebillbacktotheacceptorjustbeforematuritysothey

could repay the financial firm. The acceptor might ‘accommodate’ this transaction in

exchangeforanacceptancefeepaidbythedrawer.Thediagrambelowillustratesthis.

FigureC.1:Billsofexchange

70

AnnexD:Internationalbillsofexchange

Bills of exchange were often used in international finance. In this case, a typical

arrangement would involve two prominent merchants in two different locations. The

transaction might be structured so that the drawer of the bill would be a prominent

merchant inthesamelocationasthebuyer(say in India).Themerchantwouldthendraw

thebillonacorrespondentmerchant inanother locationwherethesellerwasbased(say

London).Thisensuredthesellerreceivedpaymentintheirowncountryandcurrency.This

fourpartytransactionisshowninFigure1,basedonNeal(1990).

FigureD.1:Afourwaytransaction

71

AnnexE:LoanstoDrawingOfficecustomers

Section 3.2 showed the skewed distribution of discounts and advances towards a small

number of customers receiving large amounts ofmoney. This begs thequestionwhether

therewasanypreferentiallendingbytheBankofEnglandduringcrisestoparticulartypesof

borrowers. For example, some studies of bank lending in the United States during the

nineteenthcenturyhavefoundevidenceofinsiderlending(Lamoreaux1994).52Inourcase,

wewere interestedtosee ifcustomersof theBank’sDrawingOfficereceivedpreferential

treatmentfromtheBank’sDiscountOffice.RecallthatDrawingOfficecustomerswerethose

whoheldacurrentaccountwiththeBank.Giventheselectiveeligibilitycriteriaforacurrent

account, it would not be surprising if Drawing Office customers received loans from the

Bank on preferential terms, in much the same way that today a bank’s depositor may

receiveabetterdealonaloanfromthatbankthanfromanotherlenderbecausethebank

hasmore informationabout theabilityof thatdepositor to repay,alongsidemechanisms,

suchascompensatingbalances,tomitigatethebank’scounterpartycreditrisk.53

In the daily discount ledgers, Drawing Office customers are typically identifiable by an

acronym ‘D.O.’ scribbled in themargins next to their name. The precise reasonwhy the

Discount Office kept track of this information is unknown, though we surmise that this

helpedtheBankkeeproughtabsonitsnetexposuretoparticularcounterparties.FigureE.1showstheledgeron15May1866,afewdaysafterthefailureofOverendGurney.Sixofthe

63DiscountOfficecustomersthatday(boxedasanexample)haveD.O.writtenintheleft-

most column in the row pertaining to them, implying they were also Drawing Office

customers. However, like any record-keeping procedure, the system of tracking Discount

Office customers in the daily discount ledgerswas not free from error. For example, the

secondD.O.entryrelatestoPhilipLevi&Co.All£19,403.68worthofbillsbythisapparent

DrawingOfficecustomerarerecordedasrejectedbytheDiscountOffice—thelargestvalue

of bills rejected among all customers in 1866.However, the rightmost ‘Remarks’ heading

thenstatesthatthereasonforrejectingwas“A/Cnotopened,”meaningthatPhilipLevi&

Cowas, in fact, not really a Drawing Office customer.We have corrected errors such as

theseinthestatisticsthatfollow.

52Insiderlendingreferstothegrantingofloansbybankstotheirstafforcloseassociatesonpreferentialterms.53Acompensatingbalanceisanamountofmoneythatmustbeheldondepositduringthetermofaloan.

72

FigureE.1:Thedailydiscountledgeron15May1866

Figure E.2 shows the volume of discount transactions split by Drawing Office and non-

Drawing Office customers during crisis and non-crisis periods. Figure E.3 gives the same

splitsbythevalueofdiscount loans. Inbothcases,weseethatDrawingOfficecustomers

were a relatively small segment within the overall business conducted by the Discount

Office. In fact,during crises, their already small shareof totaldiscount loans shrunkeven

further in both volume and value terms, as the Bank extended loans to non-regular

customers facing financialdifficulties.FigureE.4charts theaverage interest rateschargedby the Bank of England to Drawing Office customers and non-Drawing Office customers,

splitoncemorebetweencrisisandnon-crisisweeks.Thereislittlediscernibledifferencein

interest rates and therefore no evidence of price discrimination. In all cases the average

interestratedifferentialislessthantenbasispoints,withtheexceptionofnon-crisisweeks

in1857,whennon-DrawingOfficecustomersattheBank’sDiscountOfficeonaveragepaid

a 32 basis point premium over the rate obtained by Drawing Office customers. At other

times, Drawing Office customers were actually charged a higher rate of interest. For

example, during the crisis of 1847,DrawingOffice customers on averagewere charged 5

basispointsmorethannon-DrawingOfficecustomers.

73

FigureE.2:VolumeofdiscounttransactionssplitbyDrawingOfficeandnon-DrawingOfficecustomers

FigureE.3:Valueofdiscountloanssplitbydrawingofficeandnon-drawingofficecustomers

10.3 % 8.5 % 13.7 % 9.1 % 15.2 % 14.1 % 0

10

20

30

40

50

60

70

80

90

100


1847 1857 1866

Totaltransactions

Percentdrawingoffice non-drawingoffice

2.8 % 2.1 % 7.5 % 4.8 6.8 % 5.2 % 0

10

20

30

40

50

60

70

80

90

100


1847 1857 1866

Totalloan


74

FigureE.4:AverageinterestratepaidbyDrawingOfficeandnon-DrawingOfficecustomers

FigureE.5isbasedoninformationintheledgersabouttheamountofbillsrejected.Itshows

the relative success of Drawing Office and non-Drawing Office customers in getting bills

discountedbytheBank.Inbothcrisisandnon-crisisweeks,DrawingOfficeandnon-Drawing

Officecustomersapplyingforloansaremostlysuccessful.Acrossallthreeyears,thesuccess

rateisneverlessthan80%foranysegment,and,inalmostcases,isover90%.Thereisno

obviouspreferencegiventoDrawingOfficecustomers.Infact,duringcrises,thedatashows

non-Drawing Office customers were more successful than Drawing Office customers in

receivingloans,thoughthedifferenceinpercentagetermsisrelativelyminor.Forexample,

duringthe1857crisis,88%ofthebillsbroughtinfordiscountbyDrawingOfficeCustomers

werediscountedbytheBank.Putdifferently12%oftheirbillswererejected.Bycontrast,

94%ofbillsbroughtinfordiscountbynon-DrawingOfficeCustomerswereacceptedbythe

Bank.Putdifferently,6%oftheirbillswererejected.

Lookingat thedifferenceofmeans in interest ratecharged (0.07% indifference)andbills

rejected rate (3.42% in difference) between Drawing Office customers and non-Drawing

Office customers, it is obvious that the Drawing Office customers were not treated

preferentiallybytheBank’sDiscountOffice.Invalueandvolumeterms,theyaccountedfor

a small fraction of the overall business conducted by theDiscountOffice. DrawingOffice

customers were charged rates of interest roughly equivalent to non-Drawing Office

4

5

6

7

8

9

10

11


1847 1857 1866

Averagerate


75

customers.Moreover,theywerenottypicallymoresuccessfulingettingbillsdiscountedby

theBank.Infact,incrisisperiods,theywereslightlylesssuccessful.

FigureE.5:SuccessratebetweenDrawingOfficeandnon-DrawingOfficecustomers

70

75

80

85

90

95

100


1847 1857 1866

Successrate


76

AnnexF:Merchantbanks

The list of ‘good names’ acceptable by the Bankwas nevermade public.Most historians

have assumed that ‘good names’ conventionally referred to prominent international

merchants who had specialist knowledge of the creditworthiness of themajor importers

andexportersoperatingwithinparticulargeographicandproductspaces(Leaf1926:189).

ThesemerchantsincludedBarings(1763),whichstartedoutaswoolmerchants;Rothschilds

(1808),whichstartedascottongoodmerchants;Schroders(1818),whichstartedtradingas

sugar merchants; and Morgan Grenfell (1838), which began life as dry good merchants

(Roberts 1993: 23). Over time, these firms’ role accepting (giving guarantees to) bills of

exchange in return for commission became their primary business. As they evolved from

distributorsofrealeconomicgoodstounderwritersoffinancialassets,theywerereferred

tosuccessivelyas ‘merchantbanks,’ ‘acceptinghouses,’andeventually ‘investmentbanks’

(Chapman1984;Knyaston1994).

The useful function fulfilled by accepting houses in underwriting bills was in helping

overcomeasymmetric informationthatcouldhaveotherwisehamperedmarketexchange.

For example, a bilateral bills of exchange transaction could be structured instead as a

trilateralarrangement involvingthebuyerandsellerofgoods,plusaprominentaccepting

house.Inthisscenario,thebuyerofgoodswouldbecomethedrawerofabillorderingthe

acceptinghousetopaytheseller.Theacceptinghousewouldacceptthebillinexchangefor

afeefromthebuyer,agreeingtopaytheselleronthebuyer’sbehalfin,say,threemonths’

time.Thebuyerofgoodswouldremitthebillofexchangetotheseller.Thesellermightfind

thisarrangementmoresuitablebecause it feltmoreconfidentaboutthecreditworthiness

oftheacceptinghousethanthebuyerofthegoods.

FigureF.1:Atrilateralbillsofexchange

77

AnnexG:Thecustomer“withandupon”ledgers

FigureG.1:Discounters’ledgers Billbrokers’andDrawingOfficeledgers

TheBanksegmenteditscustomersintofivecategoriescorrespondingtofivedifferent“with

anduponledgers:”

1. “Discounters” (pictured top left). These were reputable City firms with access to

Discount Office facilities. To be on the Bank’s list of recognised discount houses

required that the firm be ‘introduced’ (their application supported) by a senior

member of theBank.Discounters had a daily discount limit, though the extent to

whichthiswasenforcedisunclear.

2. “Billbrokers”(picturedtopright).3. “Bankers.” These were largely private partnership and joint stock banks

headquartered in London. Initially, the joint stock banks were included in the bill

brokers’ ledgers.After1864,however,theyhadtheirownsetof ledgers,reflecting

the growing importance of joint stock banks following the extension of limited

liabilityprivilegestobankshareholdersin1862(Taylor2006).BanksoutsideLondon

had indirect access to the Discount Office through their correspondent banks in

London,ortheydiscountedbillsatBankofEnglandbranches.

4. “Drawing office or ‘DO’ customers (pictured top right). As detailed earlier, thesewerecustomerswithaDrawingOffice(deposit)accountwiththeBank.

78

5. Non-customer acceptors of bills. These were companies without a discount or

DrawingOfficeaccountbutwhowereacceptorsofthebillsdiscountedbytheBank.

The ledger for thesecustomerswerecalledsimply the“upon ledgers”as they just

recordedacceptances.

Like thedailydiscount ledgers, the customer “withandupon” ledgersevolvedover time.

The entry above comes from an 1847 ledger. It relates to N M Rothschild & Sons, the

famousmerchantbank,between8and12October1847,inthemidstofthecrisis.Although

there are no titles for the columns in this ledger, we have been able to decipher their

meaningbylookingatlaterledgers.Thefirstcolumncapturesthegeographiclocationofthe

drawerofthebillonRothschild.Thesecondcolumngivesthenameofthedrawer.Thethird

columngives thedateonwhich thebillwasdiscountedby theBank. The fourth column,

whereapplicable,containsa folionumber thatallowsyou togo to the relevantcustomer

ledgerpageoftheacceptorofthebill.Thecolouroftheinksignifieswhethertheacceptoris

alreadyaBankcustomer.Aredentrysignifiestheyare.Ablackentrysignifiestheyarenot.

Thefifthcolumncontainedboth“with”and“on”counterparties.A“with”entrymeantthat

thenamedcounterpartyhadbroughtthebill infordiscountandwasdrawnonRothschild

whowas thus theacceptorof thebill.An“on”entrymeant that thenamedcounterparty

wastheacceptorandthatRothschild’shadbroughtthebillinfordiscount.Thepenultimate

columncontainedthedateonwhichthebillneededtobepaid.Ifitwaseventuallysettled,

thiswassignifiedbycrossingouttheamountinthefinalcolumn.

FigureG.2:Excerptfrom1847customer“withanduponledgers,”Rothschild&Sons

79

FigureG.3:Excerptfrom1857customer“withanduponledgers”pertainingtoSchroders

By 1857, the customer ledger information is organised differently andmore clearly. Theimageabove isanexample relating toShroders.A fewnew featuresareworthnoting. Inadditiontodiscounts,advancesarenowshown(boxedinblue).Therearealsonowseparate“with” and “upon” columns, enabling easier identification ofwhether Schroder’swas thediscounteroracceptor(boxedinred).Ratherconfusingly,however,the“with”and“upon”columnsrefertoSchrodersratherthanthecounterpartynamedinthe“Discounter”column.This is in contrast to the 1847 ledgers, where the “with” and “on” terms refer to thecounterparty.Sointhiscaseallthebillsinthe“Discountswith”columnwerebroughtinfordiscountbySchroders,whileallthebillsinthe“Discountsupon”columnwerebillsacceptedbySchroders.Asacorollary,thismeantthecounterpartynamesinthe“Discounter”columncould be either an acceptor (for a “with” entrywhen Schroderswas the discounter) or adiscounter (for an “upon” entry when Schroders was the acceptor). This potential forconfusionintheledgerswasremovedby1866,ascanbeseenintheledgerentryforBieber& Co for May 1866 on the following page. The word “Discounter” was replaced in thecolumntitleby“AcceptororDiscounter”.Moreimportantly,itappearstheBankhadstartedmonitoringitsexposurebysummarisingitscumulativeexposuretothekeyacceptorsofthebillsbroughtinfordiscount(boxedinblue).

80

FigureG.4:Excerptfrom1857customer“withanduponledgers”pertainingtoBieberCo

Cumulative exposure to Bieber

European Historical Economics Society

EHES Working Paper Series Recent EHES Working Papers 2017 EHES 116 EHES 115 EHES 114 EHES.113 EHES.112 EHES.111 EHES.110 EHES.109

Harbingers of Modernity: Monetary Injections and European Economic Growth, 1492-1790 Nuno Palma ‘Rational’ Farmers and the Emergence of Modern Accounting in Danish Dairying Markus Lampe, Paul Sharp Missed opportunities? The development of human welfare in Western Europe, 1913-1950 Daniel Gallardo Albarran Portugal’s wine globalization waves, 1750-2015 Pedro Lains Effects of reforms and supervisory organizations: Evidence from the Ottoman Empire and the İstanbul bourse Elmas Yaldız Hanedar, Avni Önder Hanedar and Ferdi Çelikay Resource endowments and agricultural commercialization in colonial Africa: Did labour seasonality and food security drive Uganda’s cotton revolution? Michiel de Haas and Kostadis J. Papaioannou Currency Valuations, Retaliation and Trade Conflicts: Evidence from Interwar France Thilo Albers Measuring Extractive Institutions: Colonial Trade and Price Gaps in French Africa Federico Tadei

All papers may be downloaded free of charge from: www.ehes.org The European Historical Economics Society is concerned with advancing education in European economic history through study of European economies and economic history. The society is registered with the Charity Commissioners of England and Wales number: 1052680

http://iibf.ogu.edu.tr/Akademisyen.aspx?ID=52

http://www.ehes.org/

Date post:	10-Jul-2020
Category:	Documents
Upload:	others
View:	0 times
Download:	0 times