Full text of Working Papers (Federal Reserve Bank of Chicago) : Contracting Innovations and the Evolution of Clearing and Settlement Methods at Futures Exchanges, Working Paper 1998-26

The full text on this page is automatically extracted from the file linked above and may contain errors and inconsistencies.

C o n t r a c t in g I n n o v a t io n s a n d t h e E v o lu t io n
o f C le a r in g a n d S e t t le m e n t M e th o d s a t
F u tu re s E x c h a n g e s
J a m e s T. M o s e r

W o rk in g P a p e rs S e rie s
R e s e a rc h D e p a rtm e n t
F e d e ra l R e s e rv e B a n k o f C h ic a g o
D e c e m b e r 1 9 9 8 (W P -9 8 -2 6 )

FEDERAL RESERVE BANK
O F CHICAGO

A u g u st 1998

P relim in ary
P lease do no t qu ote

Contracting Innovations and the Evolution of Clearing and Settlement Methods
at Futures Exchanges

by

Ja m e s T. M o se r

R esearch
Federal R eserve B a n k of C hicag o
10 S. LaS alle S tre e t
C hicago, IL 6 0 6 0 4 -1 4 1 3
(312) 3 2 2 -5 7 6 9

IN T E R N E T : J M O S E R @ W W A .C O M

Most recent draft of this paper i available a :
s
t
www.wwa.com/~mosers/papers.htm

I am ind ebte d to th e C h ica g o B oard of T rade fo r m a kin g th e ir a rch ive s a va ila b le fo r th is
research and to O w en G re g o ry fo r helping m y a cce ss o f th e se archives. T h e p a p e r has
b e n e fitte d fro m th e co m m e n ts of P e te r A lonzi, H erb Baer, B ob C lair, J e n n ie F rance , Ed
Kane, G eoff Miller, Lester Telser, Jeffrey W illiam s an d p a rticip a n ts at sem in ars he ld at the
D epartm ent of Finance U niversity of Illinois-U rbanna/C ham paign, the 1993 m e eting s of th e
Federal R eserve System C om m ittee on Financial S tru ctu re an d R egulation, th e U n ive rsity
o f S outham pton and at the London School of Econom ics. Luis F. V ila rin provide d va lu a b le
research assistance. B em ie Flores tracked down m any ob scure references. T h e an a lysis
and conclusions of this pa per the author’s and do no t su g g e st co n cu rre n ce by th e F ederal
R ese rve B ank of C hicago.




Contracting Innovations and the Evolution of Clearing and Settlement Methods
at Futures Exchanges
A bstract
D efining fu tu re s con tra cts as sub stitute s fo r a sso ciated cash tra n sa ctio n s en able s a
discussion o f the evo lution of con trols o ve r contract no n p e rfo rm a n ce risk. T h e se
controls are inco rpo rate d into exch a n g e m ethods fo r cle a rin g con tracts. T h re e clea rin g
m ethods are discu sse d : direct, ringing and com plete. T h e incide nce and op era tion of
each are de scribe d. D ire ct-cle a rin g system s fe a tu re bilateral con tracts w ith term s
spe cified by th e co u n te rp a rtie s to th e contract. E xchan ge s relying on dire ct clea ring
system s ch ie fly serve as m e d ia to rs in trade disputes. R inging is show n to fa cilita te
con tract o ffse t by incre a sin g th e nu m be r of po ten tial cou nte rpa rties. R inging
settle m en ts reduce co u n te rp a rty cre dit risk by re ducin g th e accu m u la tio n of
de p e n d e n cie s as con tracts are offset. R inging settle m en ts also low er th e co st of
m a in tain in g open co n tra ct positions, chie fly by low ering th e am oun t of re quired m argin
deposits. E xchan ge s e m p lo yin g ringing m ethods g e n e ra lly ad op te d a cle a rin g h o u se to
handle paym ents. C o m p le te clea ring interposes the cle a rin g h o u se as co u n te rp a rty to
every contract. T h is m e a su re e n sure s that con tracts are fu n g ib le w ith re spect to both
the un derlyin g co m m o d ity and cou nte rpa rty risk.




I Introduction
.
T h is p a p e r stud ies in n ovatio ns in future s co n tra ctin g b e fo re 1926. E arly tha t
ye a r th e C h ica g o B oard of T ra d e C learing C orp ora tion (B O T C C ) began inte rm e diatin g
fu tu re s con tracts. A s m o re tha n 80 pe rcen t of US fu tu re s con tracts tra ded at the C h i­
cago B oard o f T rade, this ste p e sta blishe d com p le te cle a rin g as th e standard fo r cle a r­
ing and settlin g d e riva tive s contracts. In the hu n d re d y e a rs p reced ing th e B O TC C ,
clearing and se ttle m e n t m e th o d s m oved from b ila te ra lly ne gotia te d arra n g e m e n ts to
practices, tho ug h n o w au tom a ted , ve ry sim ila r to th o s e use d today. T h is p a per
explores ho w co n tra ct term s evo lved to conform to th e w ays th e y w e re cleared.
T h e pa p e r uses a broad definition of fu tu re s con tra ctin g . T h e de finition re co g ­
nizes th e fo rce of co n tra ctu a l ob lig a tio n s in tw o d istin ct re gion s o f th e state space: con­
tra ct p e rfo rm a n ce and co n tra ct no npe rfo rm an ce. In p e rfo rm a n ce states one
counterparty, th e sh o rt position, de live rs the u n derlyin g a sse t to its counterparty, th e
long position. A t de live ry, th e long position pays th e sh o rt po sition acco rd in g to con tract
term s. T h is po rtion of th e d e fin itio n con form s to th e sta n d a rd d e fin itio n of a fu tu re s c o n ­
tra ct as an ob lig atio n be tw een cou nte rpa rties to m a ke a fu tu re -d a te d exch a n g e at a
price de te rm in e d a t th e c o n tra c t’s inception. T h e sta n d a rd de finition is insu fficient in
tw o senses. First, it o m its th e co u n te rp a rty’s ch o ice no t to pe rform th e contract. T h is
choice w ill be op tim a l to on e sid e of th e con tra ct in n o n p e rfo rm a n ce states. R ights to
exe rcise such ch o ice s are u se fu lly con strue d as n o n p e rfo rm a n ce op tions. T h e se o p ­
tions ha ve valu e. C o n tra ct co u n te rp a rtie s re cogn ize th a t th e co st o f a b solute p e rfo r­
m ance a ssu ra n ce s can e xce e d the va lu e of tra d in g b e n e fits and act as ba rriers to




1

trade. M utual provision o f n o n p e rfo rm a n ce op tio n s sub stitu te fo r a b so lu te p e rfo rm a n c e
assu ran ces o ve rco m in g th e s e ba rriers to en a b le realization of tra d in g be nefits. S e c ­
ond, th e sta n d a rd fu tu re s d e fin itio n o b scure s institu tiona l ince ntives to in n o va te co n tra ct
design. T h e se ince ntives stem from needs to re duce cre d it-risk exp o su re s b y re d u cin g
proba bilities fo r n o n p e rfo rm a n ce states and m itiga ting loss a m oun ts w hen n o n p e rfo r­
m a nce occurs. F urther, th e p a p e r d e velop s co n n e ctio n s betw een lo ss-sh a rin g a rra n g e ­
m e nts and inn ovatio ns in co n tra ct term s. E xchanges ad op tin g co m p le te c le a rin g ­
houses inte rna lize n o n p e rfo rm a n ce losses incre asing th e ir ince ntives to in n o va te in
w ays th a t reduce th e se e xp osures.
D efin itio ns of fu tu re s co n tra cts th a t om it the n o n p e rfo rm a n ce option are c o m ­
m on. E m ery (1896, p. 46) d e fin e s a fu tu re s con tract "as a con tra ct fo r the fu tu re d e liv ­
ery of som e com m odity, w ith o u t re fere nce to sp e cific lots, m ade un d e r rules of so m e
com m ercia l body, in a set form , by w hich the co n d itio n s as to th e unit of am oun t, th e
quality, and th e tim e o f d e live ry are stereotyped, and o n ly th e d e te rm in a tio n o f th e to ta l
am ount and th e p rice is left open to th e co n tractin g parties." T h is de finition is typ ic a l in
th a t it de fines fu tu re s co n tra ctin g based on con tra ctu a l d e ta ils .1 l d e fin e fu tu re s c o n ­
tra cts a s e n fo rce a b le su b stitu te s fo r tra n sa ctio n s in cash co m m o d itie s o r asse ts.
T h e definition serve s tw o purposes. First, it broad ens th e ca te g o ry o f co n tra cts
called futures. W illia m s (1982), fo r exam ple, argue s that co n tra cts tra d e d at th e B u ffa lo
B oard o f T rade du rin g th e 1840s m ig h t be classe d as fu tu re s b e ca u se th e ir te rm s w e re

1 For a recent example of the standard definition see Kolb (1991, p. 4). In contrast see Edwards
(1984, p. 225) who takes a position consistent with that taken here stating that the clearinghouse
"transforms what would otherwise be forward contracts into highly liquid futures contracts."




2

s im ila r to th o se la te r a d o p te d b y th e C hicag o exchanges. T h e su cce ss of th e B uffalo
co n tra cts d e ve lo p e d fro m sh a re d com m ercia l inte rests in lessening no npe rfo rm a n ce
costs. C h ica g o m e rcha nts, h a vin g sim ilar interests, adopted s im ila r con tract term s.
Both th e N e w Y o rk m e rch a n ts and the C h ica g o e xch a n g e m em be rs fa ce d the po te n tia l
fo r n o n p e rfo rm a n ce loss and re spon ded b y a d a p tin g th e ir co n tra ct term s to con tro l loss
e xp o su re s. F o r pu rp o se s o f th is paper, th e re le van t co m m o n a lity is the e co n o m ic inte r­
est to lim it losses.
T h e seco nd p u rp o se se rve d by this definition follow s from th e first. M any fu tu re s
co n tra ct te rm s a re b e st u n d e rsto o d as e fforts to m in im ize n o n pe rfo rm an ce costs s u b ­
je c t to ava ila b le lo ss-sh a rin g a rran gem ents. T h e sp e cific m e a su re s ad opted to con trol
losse s a re d e te rm in e d b y th e e xta n t legal e n viron m en t. C on tra cts tra ded at the B uffalo
B oard o f T ra d e ra pidly d e ve lo p e d use of p e rfo rm a n ce bonds (m argins) and d e live ry
sta n d a rd s .2 H ow ever, e n fo rcin g con tra ct p e rfo rm a n ce beyond th is po int proved costly.
S u b se q u e n t c h a n g e s in co m m e rcia l law e n a b le d th e C h ica g o exch a n g e s to su rp a ss th e
B uffa lo prece d e n t, u ltim a te ly o ffe rin g p e rfo rm a n ce g u ara nte es. In th is sense, th e c o n ­
tra cts tra d e d at th e B uffa lo e xch a n g e served th e sa m e com m ercia l p u rpo ses as the
fu tu re s co n tra cts e xch a n g e d in C hicag o. D iffe re n ce s in co n tra ct de tails stem m ore
fro m d iffe re n ce s in legal e n v iro n m e n t th a n to m o re fu n d a m e n ta l diffe ren ces in eco­
n o m ic pu rpo se.
N o n p e rfo rm a n ce issu e s a re often ign ore d b e cause fa ilu re s are infrequent. An
u n d e rsta n d in g of th e e c o n o m ic prin ciples d e te rm in in g succe ss is useful. This paper

2 See Williams (1980, p. 140-145) for a discussion of early margin rules. Hill (1990) reviews the
literature on grade standards.




3

fo llo w s C o a s e (1937) In a rguin g th a t th e re co rd o f su cce ssfu lly m anag ed n o n p e rfo r­
m a n ce risk is larg e ly due to th e inte rna liza tion o f info rm a tio n and incentives o b ta in e d
w hen e x ch a n g e -a ffilia te d clea rin g h o u se s g u a ra n te e perform an ce.

I. A modern context for understanding clearinghouse operations
I
A g e n e ra l descriptio n of m o dem cle a rin g an d s e ttle m e n t o p era tions puts th e
e a rly p ro ce d u re s into a m e an in g fu l context. C le a rin g is th e process of re conciling and
re solving o b lig a tio n s be tw een cou nte rpa rties. T his sectio n d e velop s clearing an d s e t­
tle m e n t in tw o sub section s. First, clearing is exa m in e d a b se n t con sideration of n o n p e r­
form an ce. S econd, th e problem of n o n p e rfo rm a n ce is fu rth e r developed.
A. C le a rin g ab se n t n o n p e rfo rm a n ce
C le a rin g of fu tu re s con tracts is done in th re e steps. C learingh ouse s initiate th e
pro ce ss w ith re gistra tion of tra d e d contracts. R eg istratio n ide ntifies th e con tractual
co u n te rp a rtie s and records th e ir re spective liabilities. C o n tra ct stan da rdizatio n s im p li­
fie s re gistra tion. R eg istratio n o f con tracts is b y typ e o f co n tra ct and de live ry m onth. At
fu tu re s exch ang es, re gistra tion occu rs as th e b u y and sell sides of tra d e d co n tra cts are
m a tched . F u ture s cle a rin g h o u se s require n o n m e m b e r fu tu re s com m ission m e rch a n ts
(F C M s) to “ give u p “ th e ir tra d e s to m e m b e r F C M s .3 A "give up" occu rs w hen th e n o n ­
m e m b e r FC M re lin q u ish e s th e tra d e to a m e m b e r FC M . P ara lle ling the co rre sp o n d e n t
service s o ffe re d by banks, th e n o n m e m b e r "g ive s up" con tracts to its clearing m e m b e r

3 1 use the term "give up" to demonstrate the functional equivalence between present usage of the
term "give up" and the relationship between clearing members and their associated nonmember firms. In
present usage, a “give up" occurs when a thinly capitalized FCM noted for good order execution is
engaged to handle a large order. The FCM’s thin capitalization prevents him from taking the order.
Under a give-up arrangement, he executes the order, then gives it up to a better capitalized member.




4

who clears the contract through Its clearing arrangem ents and then adjusts the non­
member’s accounts.
R e g istra tio n at a ce n tra l clea rin g h o u se en a b le s offsetting, th e seco nd stag e in
th e cle a rin g process. A ggre g a tio n of th e related tra n sa ctio n s of each cle a rin g h o u se
m e m b e r id e n tifie s offse ttin g com m itm en ts. O ffset occu rs w he n the a g gre gated claim s
a g a in st a n y m e m b e r are ne tted a g a in st the a g gre gate of th e m e m be r's claim s a g a in st
all o th e r m e m be rs. T h e cu rre n t liabilities of the cle a rin g h o u s e and its m e m be rs are the
net o f th e s e ob lig atio ns. T hus, cle a rin g reduces th e n u m b e r of liabilities by relying on
th e fu n g ib ility of individual con tracts.
In th e th ird step, co n tra ct se ttle m e n t e xtin g u ish e s th e cu rre n t p a ym en t liabilities
o f co n tra c t cou n te rp a rtie s. In ba nk clea ringh ouse s, se ttle m e n t occurs w hen m e m b e r
a cco u n ts are ad ju ste d to reflect a m oun ts paid. O n paym ent, th e ob lig atio ns of all p a r­
tie s are s a tis fie d .4 In fu tu re s m arkets, ou tsta n d in g con tra cts are settled p e rio d ica lly by
m a rkin g the m to m arket. G en era lly, m arks are e ith e r th e m ost recent m a rke t-d e te r­
m in e d p rice fo r each co n tra ct or, a t th e co n tra ct's term ina tion, th e ca sh -m a rket price of
th e u n d e rlyin g a s s e t.56 A ll o u tsta n d in g con tracts are m a rked to th e settle m en t price. A s
c o n tra cts a re m a rked to m arket, pa ym e n ts are d e te rm in e d by the netted obligations.
In cre a se s in s e ttle m e n t price s p rodu ce gains fo r long p o sitio n s and losses fo r short p o ­
sitions. C o n ve rse ly, d e cre a se s in settle m en t price result in losses to long p o sitions and

4 The practice of using offsets rather than cash settlements dates to banking practice in early
Florence. See Lane and Mueller (1985, p. 81).
6
Marks are determined by settlement committees that generally follow the rules outlined here. In
exceptional circumstances, these committees can determine marks by substituting their assessed
valuations for market-determined prices.




5

ga in s to sho rt po sitions. S e ttle m e n t o ccu rs w he n p a ym e n ts ow e d to cle a rin g m e m b e rs
are m ade.
Like b a n k clearings, se ttle m e n ts set cu rre n tly p a ya b le am o u n ts be tw een
co u n te rp a rtie s to zero. U nlike th e o p e ra tio n s of b a n k clea ringh ouse s, clea red fu tu re s
co n tra cts g e n e ra lly rem ain o u tsta n d in g fo llo w in g a se ttle m en t. T h is m e ans th a t cre d it
risk, th e risk th a t one co u n te rp a rty w ill fa il to m e et its ob lig a tio n s, is not e xtin g u ish e d ;
p e rio d ic se ttle m e n t reduce s th is risk to th e u n co lla te ra lize d po rtion of th e price c h a n g e
re alized a t th e next settle m en t.
B. N o n p e rfo rm a n ce p roble m s
T h e n o n p e rfo rm a n ce problem is w ell illustra ted by th e e xp e rie n ce of seve ral
P eoria, Illinois grain e le va to rs o ffe rin g fo rw a rd co n tra cts to local fa rm ers. Q uo ting fro m
th e Federal T ra d e C o m m issio n ’s R e p o rt on the G rain Trade:
C o n tra ctin g fo r grain at a fix e d p rice has p roven an u n sa tisfa cto ry p ra ctice w ith
m a n y ele va to rs. T h e p rin cip le objection th e re to is th a t if prices are in a d va n ce of
th o se stip u la te d in th e co n tra ct w hen th e tim e of d e liv e ry arrives th e fa rm e r b e ­
co m es d issa tisfie d an d often re fuses to fu lfill th e con tra ct. If th e e le va to r th e n
a ttem p ts to e n fo rce it th e usual result is th a t th e fa rm e r tra n sfe rs his b u sin e ss to
a n o th e r ele vator. His d issa tisfa ctio n ea sily spread s to oth er fa rm ers, e s p e cia lly if
th e e le v a to r in q u estion is an ind e p e n d e n t o r o n e of a line com p a n y and m a y
re sult in se rio u s loss of b u s in e s s .6
F o rw ard co n tra ctin g is m o tivated b y e xp e cta tio n s of benefits. In th e P eoria case, th e
b e n e ficia rie s are fa rm e rs and grain ele vators. Irre sp e ctive of m o tivation s to co n tra ct
forw ard , su b s e q u e n t p e rfo rm a n ce is con d itio n a l on p rice s re alized at d e live ry dates. In
so m e states, co u n te rp a rtie s find n o n p e rfo rm a n ce is p re fe rre d to losses re alized by fu l­
fillin g co n tra ct term s. R eco gnizing this, co u n te rp a rtie s h a ve in ce n tive s to restrict th e ir 6

6 FTC (1920), Volume I, p. 113.




6

n o n p e rfo rm a n c e op portun itie s. D oing so im proves th e ir a cce ss to be nefits fro m c o n ­
tra ctin g fo rw a rd . A long sim ila r lines, S m ith and W a rn e r (1979) sho w h o w bond c o v e ­
na nts lo w e r d e b t costs by lessening the d e fau lt risk of co rp o ra tio n -issu e d bonds.
T h e s e re strictio n s lessen both th e likelihood of no np e rfo rm a n ce and th e e xte n t of
losse s sh o u ld n o n p e rfo rm a n ce becom e unavoidable. H ow ever, m o dificatio ns assu rin g
co n tra ct p e rfo rm a n ce are costly. W hen these costs exce ed benefits de rived from fu r­
th e r p e rfo rm a n ce a ssu ran ces, it becom es optim al fo r cou nte rp a rtie s to e xch a n g e n o n ­
p e rfo rm a n ce o p tio n s .7
E dw ards (1984) d istin g u ish e s betw een b a n k and fu tu re s clea ringh ouse s. B ank
c le a rin g h o u se s settle by ne tting pa ym e n ts be tw een m em bers, collectin g paym ents,
the n cre d itin g o r de biting m e m b e r accounts. T h e y are ob lig ate d on ly to th e e xte n t of a
m e m b e r’s a cco u n t balance. Futures clea rin g h o u se s g u a ra n te e p e rform an ce of
cle a re d con tracts. T h e y e xtin g u ish cu rre n t liabilities and ta ke steps to lessen e x p o su re
to fu tu re de fau lts, but p e rfo rm a n ce g u ara nte es im ply th a t som e residual e xp osure re ­
m ains. T h e con te n tio n o f th is pa p e r is th a t the evo lution of cle a rin g arra n g e m e n ts w as
im p o rta n tly influ e n ce d b y th e needs of m e m be rs to con trol th e ir risk of losses fro m
n o n p e rfo rm a n ce . T hus, e xch a n g e p o licy on con tra ct de tails like m argin and m a rking
c o n tra cts to m a rke t stem s fro m its inte rest in the cle a rin g m echanism .

II The evolution of contracts and clearing systems
I.
T h re e clea ring m e th o d s de ve lo p e d be fore fo rm a tio n of the BO TC C . T h e se are
cle a rin g by d ire ct settle m en t, clea ring through rings, an d co m p le te clearing. T h e s e c ­

7 This is the incompleteness referred to in Kane (1980).




7

tio n d e scrib e s th e m ethods, ide ntifies certain cre d it risk problem s and th e c o n tra c t
sp e cification s ad opted to a d d re ss th e se proble m s.
A. D irect S ettle m en t
D irect se ttle m e n t is a bilatera l re conciliatio n of co n tractual c o m m itm e n ts o b ­
ta in e d throug h de live ry or by o ffse t betw een orig in a l co u nte rpa rties. For e xa m p le , A
co n tra cts w ith B to sell 5,000 bu shels of w h e a t in M ay at $1 .00 p e r bushel. T h e re a re
th re e -categories o f possible o u tco m e s fo r a d ire ct se ttle m e n t system .
First, the spe cified te rm s of the co n tra ct can be perform ed. Thus, th e co n tra ct is
settle d w hen A de livers 5,000 bu shels of w h e a t to B in M ay and B pays $ 5 ,0 0 0 to A.
T h is is settle m en t by dire ct de live ry. S econd, pa rtie s can settle th e co n tra ct b e fo re M ay
by ag re e in g to a p rice at w hich both are w illin g to extin g u ish th e liab ilitie s of th e other.
T his is called d ire ct offset.
T he p reviou s exa m ple is exte nd ed to illu stra te d ire ct offset. Let A an d B a g re e to
a se co n d co n tra ct in M arch as follo w s: B com m its to d e live r 5 ,0 0 0 bu sh e ls in M a y to A
and A com m its to pay B $0 .9 5 p e r bushel, or $4 ,75 0. T h e tw o co n tra cts co u ld b e se t­
tle d in M ay as follo w s: Fulfilling th e initial co n tra ct A de live rs 5,0 0 0 bu shels to B an d B
pays A $5,000. T h e seco nd co n tra ct requires th a t B d e live r 5 ,0 00 bu shels to A and A
p a y B $4,750. B ecause w h e a t de live ries cancel, th e ne t from settlin g both c o n tra c ts is
a $ 2 5 0 paym ent by B to A. A ltern ate ly, both p a rtie s be nefit by re co g n izin g in M arch
th a t th e e a rlie r con tra ct has been offset on p a ym e n t by B of $250. T his is c a lle d a p a y ­
m e nt of th e difference. Both A and B be n e fit by o ffse ttin g b e cause each a vo id s title tra n s fe r costs and reduces re cord in g-ke epin g e xp ense s.




8

It m igh t be objected th a t th e presen t va lu e of $ 2 5 0 paid in M arch is g re a te r tha n
$ 2 5 0 paid in M ay, thu s B w o u ld refuse to settle on th e se term s. H ow ever, recall tha t
th e p rice fo r the M arch settle m en t is m u tually ag ree able . B ag re e s to settle e a rly p ro ­
vid e d th e diffe ren ce a m oun t paid in M arch is less tha n th e price cha nge e xp e cte d to be
realized in M ay. A is w illin g to ta ke an am oun t sm a lle r than h e r exp e cte d price ch a n g e
b e ca u se th e p a ym e n t am oun t can be invested. Thus, a m u tu a lly a g ree able se ttle m e n t
price in M arch can be based on th e presen t v a lu e of a se ttle m e n t occu rrin g in M ay.
A lternately, B cou ld be co m p e n sa te d by a pa ym e n t o f inte rest from A on th e profit re a l­
ized b y A in M arch. S om e exch a n g e s required inte rest pa ym e n ts on p ro fits .8 Both a p ­
proaches are eq u iva le n t provid e d the rate used to calcu la te inte rest pa ym e n ts eq uals
the m a rket rate o ve r the sa m e term fo r eq ual-risk investm ents.
A s a last outcom e, one party can fail to perform lea vin g co n tra ct se ttle m e n t to
stan ding e n fo rce m e n t proced ure s. A s this con tra ct is de scribe d, n o n p e rfo rm a n ce can
only o c c u r in M ay. B efo re M ay, ne ither pa rty has a duty to pe rform . A ccu ra te infe r­
en ces m igh t be m a de ab o u t co u n te rp a rty ability to pe rform ; n e verth eless, until a c o n ­
tra ct term go es unperform ed, th e contract stands. T h is a sp e ct of th e co n tra ct e le va te s
risk in tw o w ays. First, th e po ssib ility of accu m u la tin g su b sta n tia l losse s incre ase s as
tim e re m ainin g in th e con tract increases. S econd, a fa ilin g c o u n te rp a rty has ince ntives
to g a m b le in ho p e s of re surre cting net w orth. S uch ga m b lin g incre ase s credit risk.
R eco gnizing th e se risk so u rce s m otivates adoption of co n tra ct te rm s th a t im po se peri-

8
Forrester (1931) says the rules of the Liverpool Cotton Association required payment of interest on
profits. As of 1882, the General Rules of the New York Produce Exchange also specified payment of
interest on profits.




9

odic demonstrations of continued performance capability. Including these provisions
curtails loss buildups and reduces incentives to gamble for resurrection.
D irect se ttle m e n t is the old est clearing a rran gem ent. E m ery (1896, p. 3 5 -3 6 )
de scribe s tra d in g in th e w a rra n ts of the East India C o m p a n y in 1733. T h e se w e re
b e a re r instru m e n ts tra n sfe rrin g title to a w a re ho use re ce ip t fo r a q u a n tity of m e tal on a
fu tu re date. E n d o rse m e n t sig n ifie d sale of th e w arrant. Thus, tra n sfe rs w e re d ire c tly
settled at th e tim e of sale. T h e ea rlie st of the se w arra n ts w ere fo r sp e cific lots o f a
m etal, later “ g e n e ra l w a rra n ts" tra nsfe rred title fo r sp e cifie d q u antitie s and g r a d e s .9
T h e se w a rra n ts did not tra d e on exchanges. Thus, resolution of legal d is p u te s
arisin g fro m tra d in g in w a rra n ts w a s obtained in courts. T h is proved costly. O b ta in in g
a less co stly ro ute to h a n d le tra d e disputes served as an im pe tus fo r th e fo rm a tio n of
e xch ang es and tra d in g a s s o c ia tio n s .10 M ost often e xch a n g e m e m b e rsh ip b o u n d m e m ­
bers to a cce p t th e a rb itra tio n de cisio n s m ade by an ap p o in te d com m ittee .
E llison (1905, p. 15) re cords th a t trading in Liverpool began sho rtly a fte r 1781.
Ind ication s are th a t c o n tra cts w e re dire ctly settle d in th e e a rly ye a rs of th a t e xch a n g e .
E llison d e scrib e s th e m a rke t be fore 1860 as follo w s: "T he m e rch a n t sold his co tto n
throug h a se llin g broker; th e sp in n e r purchased it th ro u g h a bu ying broker. T h e re w e re
brokers w h o bo u g h t and sold, bu t th e y w ere an exce ption to th e rule, and c o m p a ra tiv e ly
fe w in n u m b e r."1 T h e Liverpool A ssociatio n active ly so u g h t to im prove co n tra ct p e rfo r1

9 Nevin and Davis (1970, p. 17-19) suggest that assignability of contracts developed much earlier.
Thus, it would not be surprising to find similar contracts trading well before 1733.
1 For similar instances, see Powell’s (1984) description of the development of arbitration in England.
0
Milgrom, North and Weingast (1990) describe earlier developments.
,1 Ellison (1905, p. 244).




10

m ance. T h e e a rlie st m e a su re s left co n tra ct n e gotia tion to cou nte rpa rties, the e xch a n g e
intervening o n ly to a rb itra te disp utes. N eg otiatio ns cou ld be com plex. Ellison
de scribe s a 1825 tra n sa ctio n invo lvin g th e pla ce m e n t of "tw o letters of credit of $ 5 0 ,0 0 0
each" on a co n tra ct fo r 6,0 0 0 ba les o f cotton estim a te d to be w orth $500,000.
T h e e ffe ctive n e ss o f Live rp o o l clearing m e thod s is sho w n by rule changes
ad op te d at th a t e xch ang e. E llison (1905, p. 29 2-2 95) de scribe s settle m en t proble m s as
"nu m e ro u s disp u te s a risin g out o f th e gig a n tic s p e c u la tiv e tra nsa ctio n s de veloped by
th e o ccu rre n ce s incide nta l to th e A m e rica n [C ivil] W ar." D um b ell (1927, p. 196) says
th a t "the con fu sio n of the w a r y e a rs fo rce d th e A ssociatio n to p rescribe fo r itself a co n ­
stitution and a g ra d u a lly incre asing n u m b e r of ru le s and bye -law s." W illia m s (1982, p.
306, fn. 2) d e scrib e s a vo te on th e ad optio n of A ssociatio n ru le s on Jun e 17, 1864.
E llison (1905, p. 32 5-3 26) states th a t lack of g ra d e sta n d a rd s created bargaining situa­
tions th a t co n trib u te d to c o n tra c t-se ttle m e n t proble m s: "A t o th e r tim e s the im po rte r
w o u ld d is c o v e r th a t his p ro p e rty had been sold ’s h o rt’, in w h ich case he w ould re fu se to
part w ith it e xce p t at a sm a rt prem ium on cu rre n t prices." T h e se com m en ts sug gest
th a t dire ct s e ttle m e n t of in d ivid u a lly a rran ged co n tra ct te rm s used in th e ea rlier yea rs
proved u n w ie ld y in th e v o la tile m a rke ts of th e 1860s. T h e m e m be rs of th e A ssociatio n
re sp o n d e d b y sta n d a rd izin g co n tra ct term s, p a rticu la rly g ra d e standards, to increase
co n tra ct fu n g ib ility. F o r tra n sfe rs of th e un derlying co m m o d ity, contracts becam e close
su b stitu te s fo r on e an oth er, in cre a sin g th e a b ility to o ffse t po sitions. By 1871, th e e a r­
lie r system of v o lu n ta ry p e rfo rm a n ce bondin g w a s re place d by a rule stipulating p e rfo r­
m a n ce bo n d s fo r all con tracts, su g gestin g th a t v o lu n ta ry b o n d in g w as proving




11

inadequate.
Contracts traded at the Buffalo Board of Trade may be the first instance of ex­
tensive futures contracting in the United States. These contracts arose as grain traffic
in the Great Lakes increased. Early transactions were for cash financed by bank ad­
vances. Williams (1982, p. 310) describes the Buffalo Board of Trade organized in
1844 as having developed an extensive futures market by 1847. Williams quotes a
contemporary: "most often the forward sales considerably outnumbered sales of flour
on thespot." Settlements were direct, described as "between pairs of parties."
The Buffalo Association adopted measures to ensure performance. Contracts
were most often settled by payment of differences rather than delivery. Contract offsets
were obtained via bilateral negotiation because "individual traders were themselves
adept at enforcing the terms of contracts and keeping them comparable to others . . . " 1
2
Contract offset was helped by standardizing deliverable grain by stipulating sources of
deliverable grain or flour. Market participants understood the quality implied by these
locations, enabling them to substitute contracts for one another; i.e., contracts were
made fungible.
‘-The measures proved effective. Williams (1982, p. 313) quotes Buffalo newspa­
per descriptions of that market’s response to late-Spring ice on the Great Lakes. The
ice prevented delivery on May contracts. Prices were high "... as contracts are inter­
ested in having high prices maintained that the damages for nonfulfillment of contracts
may be corresponding." In other words, long positions expected full compensation for

1 Williams (1982, p. 315) and see Williams (1986, p. 125).
2




12

late deliveries. Apparently, this expectation was justified as a subsequent news
account stated: “All the houses which contracted to deliver breadstuffs have so far set­
tled without litigation or delay, except one, though the balances in many cases have
been very heavy..."
Futures contracts for oil traded at several exchanges before formation of Stan­
dard Oil.1 The rules of these exchanges show that settlements were exclusively direct.
3
A Petroleum A g e article describes contracting at the Oil City Exchange, stating "futures
were regulated to suit the convenience of either party."1 Lack of uniformity would have
4
limited offsets to contracts between original counterparties.
Rules of the petroleum exchanges sought to control nonperformance risk. Rules
of the Oil City exchange allowed counterparties to "call for mutual deposit on margin of
not more than ten percent of the contract price." The Titusville and New York Petro­
leum Exchange had similar provisions. Similarly, contract nonperformance could result
in suspension of trading privileges or expulsion from the exchange.
The New York Petroleum Exchange, the most active of the oil exchanges, en­
gaged Seaboard Bank as its clearinghouse agent. Its agreement with Seaboard, re­
quired exchange members to complete daily statements of "all Oil coming in and going
out." Statements were sent with a certified check "if the difference is against the sheet,
or with the Oil, if the sheet shows more going out than coming in." Members not fulfill­

1 W einer (1991) attributes the decline of petroleum futures to the development of the Standard Oil
3
Trust in the mid-1890s. I am indebted to Rob W einer who provided copies of most of the petroleumfutures references which are used here.
1 “Speculative Halls~The Oil City Oil Exchange,’
4




The Petroleum Age IV, no. 4, May, 1885.

13

ing these requirements were in default of their contracts and faced fines or expulsion
from the exchange.
Exchanges using more advanced clearing mechanisms generally permitted di­
rect offset. The New York Produce Exchange required clearing through designated
trust companies acting as its clearinghouse only for contracts held overnight. Traders
could offset directly contracts made during the day.1 Chicago Board of Trade rules
5
permitted direct settlements even after incorporation of its clearinghouse. Morris
Townley, CBOT counsel, in an opinion to the exchange’s board of directors, states "...
it is entirely lawful and proper for members to make trades, stipulating as between
themselves that such trades are not subject to clearance through the Board of Trade
Clearing Corporation. Such trades would be ex-clearing house and would be settled by
direct delivery between buyer and seller."1 Similarly, the Italian Bourse made complete
6
clearing optional. Counterparties submitting contracts to the clearinghouse within three
days of the trade obtained guaranteed performance. Otherwise, settlement was left to
the original counterparties.1
7
B. Settlement by Ringing
Direct settlements are limited to the original counterparties. This is important
because direct offsets depend on the mutual assent of counterparties. Contract fungi-

1 Emery (1896), p. 66.
5
1 Chicago Board of Trade, Board of Directors meeting 1-1-1926. Hereinafter cited BOD and the date.
6
Williams (1986, p. 307) also concludes clearing through the clearinghouse was not mandatory until the
1920s.
171 am indebted to Giorgio Szego who described these aspects of early practices at the Italian Bourse
to me.




14

bility simplifies these joint interests. The simplification means that many counterparties
can settle contracts simultaneously. Ring settlements are relatively informal arrange­
ments between three or more counterparties with interests to settle. Incentives to enter
a ring are: reduced exposure to counterparty risk and reductions in the cost of maintain­
ing open positions. To achieve these benefits, ring participants must accept substitutes
for their original counterparties. Exchanges adopted rules and practices that helped
ringing settlements to enable access to those benefits.
To illustrate settlement by ringing, consider four parties having positions in a
contract requiring delivery of 5,000 bushels of wheat in May. A sold to B at $1.00; B
sold to A at $.95; C sold to B at $.97; and D sold to C at $.93. The following table
details these positions.
Table 1
Summary of Contracts and
Counterparties
■ s ip i
1

1

1.00

A
0.95
B

1

0.05

1.00
0.95

(0.05)

0.97
C

0.97
0.93

D

0.04
0.93

Long positions are shown by entries under buy price. Entries under sell price denote




15

short positions. Profits (losses) based on direct settlements are listed for each transac­
tion reversing a previously listed transaction.
The three possibilities previously examined are available here; that is, contracts
can be settled between original counterparties through deliveries, offset, or by standing
rules for governing nonperformance. Thus, the two contracts between A and B can be
directly settled through delivery or offset. A fourth possibility, settlement by ringing, be­
comes available provided each party regards the contracts as fungible. Clearly, the
contracts are fungible with respect to the commodity delivered and the delivery date
and, as demonstrated before, settlements can accommodate differences in contract
prices. However, complete fungibility requires acceptance of substitutes for original
counterparties. Thus, B must regard the exposure stemming from a substitute contract
with D as no worse than the exposure from the original contract with C. Given equiva­
lence of nominal contract terms and credit-risk exposure, the contracts are fungible and
a ringing settlement is possible.
Standardizing contracts achieves the needed level of equivalence. The literature
on contract standardization emphasizes the early development of standard deliverable
grades. Standardization was in the interest of market participants because it created
benchmark contracts. Absent nonperformance concerns, individual commercial inter­
ests can be restated in the benchmark contract. Trades in the benchmark contract sub­
stitute for trades in specific-but-similar commodities. Transactions in these separate
and often illiquid markets for specific commodities are replaced by transactions in a liq-




16

uid benchmark commodity.1 However, like the standard definitions for futures contract­
8
ing, the contract standardization literature ignores the nonperformance option.
Working through a ringing settlement establishes the importance of counterparty
credit risk. A ring formed by the counterparties A, B, C, and D must agree to a price for
settling all contracts.1 This price must produce payoffs identical to those obtained us­
9
ing the present value of the expected settlement in May. The most recent futures price
provides this.2 Taking the trade between C and D as the most recent price, the ring’s
0
settlement price is .93. Reconstructing the previous table based on this price gives:

1 The significance of the ability to transact in futures markets is demonstrated by a 1923 letter from a
8
J.C. Wood. Wood compares transacting in the cash markets and the futures markets: “In other words,
the service performed by the broker in the execution of orders for ’seller the month’ or 'future month
delivery’ in the pit is an entirely different service than the service performed by the cash broker, whose
work is largely specialized and carries with it technical knowledge and represents an expensive activity."
BOD 11-23-1923.
1 Counterparty D, having no contracts to offset, has no incentive to enter this ring. D ’s inclusion
9
demonstrates that his interests are not damaged by the settlements arrived at within the ring.
20 This is because arbitrage restricts futures prices to the cost of acquiring and financing a position in
the underlying asset. Thus, the most recent futures price provides the current “cost of carry" for the
underlying commodity. This price gives then the present value of a May settlement.




17

Table 2
Illustration of Clearing Through a Ring

1

a93

(0.02)

1.00

0.93

(0.07)

0.95

0.02

0.97
C

0.07

0.93

B

1.00

0.95

A

0,93

(0.04)

0.97

0.04

0.93

0.00

0.04

0.93

0.00

0.00

:0.93

I

l l l l l !
0.93

D

0-93

!

0.05

(0.09)

Shaded cells denote settlement prices. An odd number of rows for a
counterparty denotes an open position after ringing. Thus, after ringing A and C have
offset their contracts, and B and D become counterparties to one open contract with B
long and D short.
Applying settlement prices to the contracts of ring participants gives the profit
(loss) on that contract and netting the profits (losses) of each counterparty gives that
individual’s net profit. Hence, A records a net profit of .05. The contract with A
accounts for 0.05 of B’s loss amount (0.09). Thus, regarding profits and losses from
their two contracts, A and B are indifferent between settling directly and settling through
the ring.2 C opts to offset its contract with B, while B remains long in the contract. Re
1

2
1
This is true despite settling the contracts at .93 rather than at .95 as in the direct-settlement
example. As long as all contracts are settled at the same price and payments are netted, the payoffs are
the same for all closed contracts regardless of settlement price. Use of the most recent price eliminates
carrying forward losses and gains for open contracts. Cox, Ingersoll and Ross (1981) show that




18

call that a direct settlement system requires that both counterparties be willing to offset.
To obtain a direct settlement, C might find it necessary to give up part of the .04 profit
per unit to obtain an offset of the contract with B. A ring settlement enables C to avoid
this bargaining problem, establishing a weak preference for settling through the ring
compared with a direct settlement.2
2
The interests of counterparty B in the contract with C deserve special attention.
Note that B loses its bargaining position with C upon agreeing to the ring’s settlement
price. In addition, B’s participation hazards greater counterparty risk. B's decision
regarding counterparty risk has two additional dimensions. First, since C has an
unrealized gain of .04, prices must rise .04 before B has any loss exposure. Replacing
C with a substitute counterparty of equivalent creditworthiness at the current settlement
price increases B's loss exposure. B will prefer a ringing settlement only if the new
counterparty poses less credit risk than presently posed by C and its unrealized gain.
This first dimension of B's decision dictates a preference for direct settlement.
The second dimension of B's decision compares counterparty risk absent C's
unrealized gain. B's present counterparty risk is subject to the performance of D in that
D's failure to perform can lead to C being unable to complete its obligations with B .2
3
Thus, the loss that B may suffer is conditional on losses that C may incur from its other
open contracts. An informed decision by B requires information on all of C’s positions,

investment of nonzero carryover amounts affects the price of a futures contract.
22 The preference is weak because at a profit of .04 from direct settlement C is indifferent.
23 Counterparty D will also be concerned with counterparty substitution. As his concerns are the same
as B’s, the discussion which follows focuses on B’s concern only.




19

especially positions potentially affecting C’s ability to perform. The exposure from
directly contracting with D may be less than the exposure from a contract with C whose
performance is dependent on the contract between C and D. To see this, consider two
alternative contracts each having no performance dependencies. In a contract
between B and C, C fails if state one occurs but performs in all other states. In a
separate contract between B and D, D fails if state two occurs but performs in all other
states. To obtain an upper bound on B’s current risk, add the condition that C’s
performance on the contract with B depends on the performance of D. With these
conditions, B suffers a loss on occurrence of states one or two. In contrast, directly
contracting with D reduces B's exposure to only state two’s occurrence. Thus, B's
upper bound is decreased from its present level. This dimension of B's decision is
determined by the extent and importance of these dependencies. As they increase, B's
preference for a ringing settlement increases.
Besides affecting counterparty risk, ringing affects costs for maintaining open
positions.2 Maintaining margin deposits is a significant part of this cost. To illustrate,
4
assume each counterparty maintains a margin deposit of .05 per bushel; that is, $250
per contract. From Table 1: A and C have two contracts each, B has three and D has
one. Margin deposits while these contracts remain open are: $500 for A and C, $750
for B and $250 for D. With direct settlements, A and B would recognize that their cross
exposures are nil so under direct settlement rules, A deposits no margin and B deposits
$250 for his open position with C. Direct settlement does not reduce the deposits of C

24
These arguments for the cost of maintaining open futures positions closely follow Baer, France,
Moser (1998).




20

because his two contracts remain open. D maintains a $250 deposit for his one open
position with C.
In a ring settlement, D is substituted for C so that margin deposits are $250 each
for B and D and zero for A and C. Thus, absent ringing, C incurs the cost for
maintaining deposits of $500. With respect to their costs of maintaining margin
deposits, A, B and D are indifferent between settling directly or through rings. C, on the
other hand, prefers ringing.
Summarizing, A is indifferent between direct and ringing settlements. B’s
preference is determined by comparing the value of bargaining power over C with an
assessment of the effect that ring entry will have on credit risk. Because C avoids a
weak bargaining position and reduces the cost of maintaining open positions, C
strongly prefers a ringing settlement. Finally, D shares counterparty substitution risk
with B. So, the interests of B and C can be in conflict. An optimal ringing rule should
enable C to improve its position without imposing costs on B or D. The exchanges
obtained this result by making ring entry voluntary, stipulating that once a ring was en­
tered, its results were binding on its participants. From the above, B’s minimum
condition for entry into the ring is reduced loss exposure after a ring settlement. On
satisfying this condition, B enters the ring and becomes bound by its settlements.
Rules enabling settlement through rings must provide finality for all offsets
arranged through rings. Referring to the above example, finality is obtained when
neither B or D can enforce a claim against C should their substituted counterparty fail to




21

perform. Exchange practice clearly intended finality.2 The courts upheld the principle
5
of offset finality in Oldershaw v. Knoles. Referring to the 1879 decision of this case in
which a commission merchant arranged for a substitute counterparty who later failed,
Bisbee and Simonds (1886, p. 158) state:
The customer was held bound by this similar transaction on the part of his
commission merchant; because, in employing the merchant, the customer was
taken as intending that the business should be done according to the custom or
usage of that market, whether or not he knew of such custom or usage.
Thus, counterparty C could offset an original contract with B assured that the liability
was terminated. The decision established that finality was independent of whether
counterparties were original or not. As ringing benefitted customers by easing contract
trading, these customers could not invalidate their contracts when ringing resulted in
losses.2
6
To facilitate ring settlements, exchanges adopted centralized mechanisms for
payments and deliveries. These arrangements performed like bank clearinghouses.2
7
A counterparty entering a ring with contract losses submitted a record of their offset
contracts along with a suitable draft to the clearing facility. Offset contracts were
confirmed by matching against the offset contracts submitted by other ring members.

25 A 1920 FTC report observed that counterparty substitution was implicitly “recognized in the rules of
various exchanges, only in Chicago and St. Louis is it set up as a right of traders executing contracts."
FTC II, p. 284.
26 Oldershaw v. Knoles (4 Bradwell 63-73 and 6 Bradwell 325-333) built on two prior cases: Horton v.
Morgan, 19 NY 170, a 1859 case in New York involving transfers of stock; and Bailey v. Bensley 87 III
556, a 1877 case involving CBOT futures.
27 Nevin and Davis (1970, p. 6) indicate that similar clearing systems were employed by the French in
the 13th century. Merchandise was bought and sold at fairs with transactions debited or credited
accordingly by an on-the-spot banker. At the close of the fair, all transactions were cleared with
settlement made in a single payment as needed between the banker and each merchant.




22

The clearing facility credited member accounts in the amount of drafts received and
debited accounts on disbursing payments to counterparties realizing gains. Deliveries
were cleared by passing warehouse receipts to the clearing facility. The receipts were
then passed to parties taking delivery. Members were charged fees for contracts
settled through the clearing facility.2
8
Clearinghouse operations originated in banking and their operations probably
were an important influence in the development of futures clearinghouses. Spahr
(1926, p. 70-71) dates the London Clearing House to 1773. Gorton (1985) and Gorton
and Mullineaux (1987) study the economic forces motivating formation of the New York
clearinghouse in 1853 as drafts on individual bank deposits replaced bank-issued
claims on specie. Timing suggests that clearing of stock transactions inspired
formation of a clearinghouse for a futures exchange. The London Stock Exchange
adopted a clearinghouse in 1874. Emery (1896, p. 69, fn. 1) says the Liverpool
clearinghouse was adopted in 1876 and describes it as the first clearinghouse for a
produce market.
Like direct settlement, ringing leaves resolution of nonperformance to individual
counterparties. A 1923 letter from the CBOT Rules Committee describes the position
of a trader with respect to the CBOT, stating:
That part of the regulation referring to the financial standing of a correspondent
should be understood to mean that the principal should keep himself well
informed, as business transactions between the two would warrant, as to the

28
A case described by Parker (1911) suggests that demand for clearing services can be substantial.
The membership of a German exchange in 1908 sought to avoid government regulation by moving the
membership to another building and dropping its clearing house arrangements. Soon after, a private firm
offered to clear settlements for those who chose to patronize it.




23

financial condition of his correspondent, so as to protect himself and the trade in
general against any losses which might occur through the correspondent
becoming insolvent.2
9
This was generally understood as illustrated in a letter by T.P. Newcomer describing
his position with the CBOT. Newcomer wrote: "We understand your Board is not a
collecting agency and do not expect you to get us our money..."3 Thus, counterparties
0
retained responsibility for monitoring the financial condition of counterparties and to
collect from them any payments due. The exchange did not take on this responsibility.
The exchanges did provide members with routes for controlling nonperformance
risk. The first of these was margin. Like direct-settlement clearing systems, exchange
rules enabled counterparties to call for margin. Two forms of margin deposit could be
required of contract counterparties. The first, original margin, was generally limited to
no more than ten per cent of the value of the contract at its most recent futures price.
This established an upper limit on the liquid assets that an exchange member could be
required to maintain. The limit curtailed calls for excessive amounts of margin to force
a counterparty into default. Margin amounts were reciprocal; members calling for
margin were required to post like amounts.3
1
?The second form of margin, sometimes called variation margin, was based on
the amount of the difference between the contract price and the current settlement

29 BOD 10-2-1923.
30 BOD 9-18-1900.
3 The CBOT attitude toward margin determination is expressed in a letter from a special committee
1
which considered exchange-determined margin: "mandatory rules are impossible and that anything else
would operate simply in the nature of a suggestion and would not only be unenforceable but ill-advised,
because of the fact that each member of this exchange governs his transactions with his customers by
his own ideas of credits." BOD 9-17-1912.




24

price. This amount applied only to the counterparty with an unrealized loss from the
contract. Amounts paid as variation margin were also kept on deposit. They were paid
out only when the contract was offset.
The right to assess margin was well recognized by the courts. The Illinois
Supreme Court in Denton et al. v. Jackson held that absent an agreement between
counterparties on margin, then the transaction was governed by the rules of the
exchange. If the rules of the exchange enabled a call for margin, a member not
fulfilling a margin call was in default.3
2
Exchange rules generally provided that margin moneys must be held in accounts
agreed to by the counterparties or with an exchange-approved bank. A 1915
amendment to CBOT rules permitted members to fulfill margin requirements with cash
or securities.3 The rules stipulated the timing of these deposits with an expectation of
3
rapid compliance. Rules often stipulated that margin amounts called for in the morning
had to be deposited by early afternoon.3 During this period the form used for
4
customer-trade confirmation by Edwards, Wood & Co. Stock Brokers and Commission
Merchants included a preprinted notice that customer positions can be closed out
"when margins are running out without giving further notice." This firm was leaving no
doubt that it could close out positions as it deemed necessary.

32 Bisbee and Simonds, p. 150. Common law also provided a right to call for margin. Under common
law, a reasonable period had to be provided to meet the margin requirement before the contract could be
regarded in default.
33 Determinating securities allowable for margin purposes was left to the counterparties.
34 The CBOT adopted a one-hour rule in 1887. It required members to meet calls for margin within
one banking hour. Prior to that date three banking hours were allowed.




25

Failure to make a required margin deposit was a contract default. Margin rules
enabled members to determine financial ability by calling counterparties for margin.
Failure to post the margin was nonperformance, enabling members to curtail the
accumulation of further losses and prevent gambles to resurrect net worth.
Rules requiring margin deposits were also facilitated by the clearinghouse. The
ability to offset contracts and, by that, substitute counterparties required notification
rules. The clearinghouse kept track of contract counterparties, enabling their
identification. However, because principals often confidentiality, commission merchants
obligated themselves to fulfill contract terms.3 Thus, exchange rules generally
5
regulated calls for margin between commission merchants and not their customers, the
actual principals. The commission merchants, in turn, arranged for margin deposits
from actual principals. Margin called from actual principals who were not members of
the exchange were not subject to exchange limits on margin requirements.
Periodic contract settlement, today called marking to market, was not generally
adopted by the exchanges. An exception was the Liverpool Cotton Association. Ellison
(1905, p. 354-356) suggests that periodic settlements were adopted by that exchange
in 1883 because of heavy broker losses incurred during a corner. Forrester (1931, p.
196-207) states: “Liverpool has weekly settlements; all outstanding contracts are
reduced to a weekly settlement price and all differences must be cleared." Liverpool’s
periodic settlement adoption followed that of the London Stock Exchange. Forrester
(1931, p. 196-207) says the motivation for both organizations was the same: "to

35 S ee Bisbee and Simonds (1886, p. 182-183).




26

prevent plungers without capital and unduly optimistic speculators from proceeding so
far as to hurt the market before a check is applied." Periodic settlement curtailed
nonperformance losses in two ways. First, it lessened the probability of incurring a loss
by imposing repeated demonstrations of financial ability to perform. Second, it curtailed
loss accumulations.
Standards for the financial integrity of exchange members give another route for
controlling nonperformance risk. The CBOT took early steps to lessen credit risk by
regulating membership based on financial ability. On March 27,1863, its membership
adopted a rule stating
Any member of the association making contracts either written or verbal, ar\d
failing to comply with the terms of such contract, shall, upon representation of an
aggrieved member to the Board of Directors, accompanied by satisfactory
evidence of the facts, be suspended by them from all privileges of membership
in the association until such contract is equitably or satisfactorily arranged and
settled.3
6
Thus, failure to comply with the terms of a contract could result in loss of a membership
in the association, the value of this membership stemming from the right to trade
contracts "on ’Change." The relevance of this rule was illustrated a year later during a
debate over initiation fees: "The amount of initiation fee is not one of the questions
taken into account when a man is proposed for membership. The character and
standing of the applicant is the only matter for consideration."3
7

36 Quoted from Andreas (1894), Volume III, p. 351. The following communication illustrates a typical
settlement between counterparties. "We beg to advise you that a private settlement has been arranged
on the Sept. Barley on which we yesterday reported default. This settlement is satisfactory to all parties
concerned; consequently we ask that our request for the appointment of a Committee to determine a
settlement price be withdrawn." BOD meeting 10-7-1919.
37 FTC II, p. 72.




27

In 1873, the CBOT extended its efforts by making nonperformance of any
contract, on or off the exchange, grounds for requiring a demonstration of financial
ability. The rule stated:
Any member of this Association who fails to comply with and meet any business
obligation or contract, may, on complaint of any member of this Association, be
required to make an exhibit of his financial condition on oath to the Directory of
this Board, which shall be open to any aggrieved member; and should such
member, failing as aforesaid, refuse to make such statement, he shall be
expelled from the Association.3
8
This measure amplified rules protecting members from nonperformance losses.
The usefulness of earlier measures depended on the ability to anticipate counterparty
failures. This dependence weakened their effectiveness. In particular, failure of a
0

counterparty could result from failure on a contract made by that counterparty with yet
another member. Like dominoes, contract failures could cause a string of seemingly
unrelated counterparties to fail. Ringing, because it left risk assessments to individual
members who lacked the ability to detect the exposures of their counterparties, was
susceptible to these systemic failures. The 1902 bankruptcy of George Phillips
illustrates this problem. Losses from the Phillips bankruptcy reached the accounts of
748 members, more than 42% of the CBOT membership.
Arrangements for payments from counterparties ranged from banking
arrangements to handle payments to adoption of a clearinghouse within the
association. The CBOT developed a clearinghouse for handling difference payments in
1883. Describing the forthcoming clearinghouse the CBOT Annual Report said it
"meets a want which has long been felt by the trade." A contemporary described its

38 Quoted from Andreas (1894), Volume III, p. 299.




28

operations saying "It takes no cognizance of the transactions on the board, but simply
plays the part of a common fund, to which each member pays the excess of his daily
debit over his daily credits, or receives the excess in case the later aggregate be
greater than the former."3 This clearinghouse began operating on September 24,
9
1883. Summarizing its operations during its first fourteen weeks, Chicago’s Tribune
reported that 26,986 checks had been processed. That newspaper reported that under
the previous system 260,000 checks would have been required.4 Contract registration
0
initially occurred only at contract settlement. Introduction of contract registration in
1884 extended clearing operations beyond handling of difference payments and
beyond those performed by bank clearinghouses. After this date, traders intending to
settle through the clearinghouse were required to turn in Check Slips at the inception of
these contracts. The registration requirement recorded contractual obligations.
The extent of clearing operations in the years before BOTCC cannot be directly
ascertained. The importance of the CBOT clearinghouse does come through in various
communications. A 1906 letter describes the consequences to a member on being
denied access to the clearinghouse: "It started a run on my firm. Those who had
credits with us wanted their money at once, while those who owed us refused to pay,
waiting, as some of them said, till they could settle for 10%."4 Other evidence confirms
1
an increasing role for the CBOT clearinghouse. A 1903 change to the rules required
39 Chicago

Tribune, 1-1-1884, p. 9.

40 A newspaper report of check clearing by the Chicago Clearing Association indicates that growth in
check clearings had declined. This was explained as due to the operations of the clearing house.
Chicago Tribune, Decem ber 3 ,1 8 8 3 , p. 6.
4 BOD 10-17-1906.
1




29

that loss or damages from defaulted contracts be paid through the clearinghouse.
While this left determination of these amounts to the affected parties, their processing
through the clearinghouse gave the exchange a record of member performance. In
1917 the exchange began requiring members to notify the clearinghouse when calling
for margin. Members making their deposits were required to notify the clearinghouse of
their compliance. On failure to make the required margin deposit, the calling party had
the right to offset the contract on the exchange floor. The defaulted contract then
became “due and shall be payable through the Clearing House,..., the same as though
the said contract had fully matured."
Ringing, generally with a facilitating clearinghouse or bank, was the predominate
method of clearing at the US exchanges. Emery (1896, p. 66) quotes from a copy of
the clearing sheet used at the New York Produce Exchange: The clearing bank is
intended “to facilitate the payment of differences on the deliveries, direct settlements
and rings of the previous day."4 Rules for Lard and Provisions contracts traded at the
2
New York Produce Exchange stipulated that on agreement to form a ring, the parties
were compelled "to settle their differences on said contract with each other, on the
basis of the settlement price." This rule bound exchange members to the settlements
reached by the ring. The New York Cotton Exchange, established on April 8, 1871,
used direct and ringing settlements with no clearinghouse until 1896.4 Boyle (1931)
3

42 It is not clear when the New York Produce Exchange adopted a clearing bank. Trading at that
exchange began in 1877. A Chicago Tribune article summarizing business developments in 1883
reported that the New York exchange formed an in-house clearinghouse on October 2 9 ,1 8 8 3 on a threeweek trial basis. The operation halted after the trial period.
43 Emery (1896), p. 68




30

describes clearing at the New Orleans Cotton Exchange. The exchange organized on
September 19, 1871 allowed contracts to "be settled through the Clearing House of the
Cotton Exchange, or by offset between members, or by ringing out and paying only the
balance due."
C. Settlement with Complete Clearing
Complete clearing interposes the clearinghouse as counterparty to each side of
exchange-traded contracts. Contracts agreed to on the floor of the exchange and
accepted for clearing require the clearinghouse to take the buy side of every contract to
sell and the sell side of every contract to buy. This role substitutes the credit risk of the
clearinghouse for the credit risk of individual counterparties. Thus, contracts exchanged
in a complete clearing system are completely fungible: grade standards imply that
commodities underlying contracts are the same and complete clearing implies that all
contracts have equivalent credit risks.
The following table adds the clearinghouse as counterparty to the contracts used
to illustrate the ringing system.




31

Table 3
Illustration of Complete Clearing
llp p p ll

W

■ w ill!

illi

iillillili

If lB ii

B

0.93

1.00

0.07

0.95

A

0-93

(0.02)

1.00

0.93

illiliili
1.00
0.93
0.05

OM

0.95

0.02

(0.07)

B P l

1.00

0.07

0.02

0.95

0.93

(0.02)

0.93

0.97

0.04

0.97

0,93

(0.04)

0.93

0.93

0.00

0.93

0.93 ’

0.00

--

0.93 | 0.95

V

0.97

0.93

..(•04)

: 0.93

0.97

0.04

0.93

0.93

0.00

0.93 | 0.93

0.00

c

D

( 0 .0 7 ^

(0.09)

0.04

(0.05)

0.09

(0.04)

As before shaded cells are the settlement prices obtained from D’s trade. The first five
columns repeat Table 2. The four columns on the right side give the clearinghouse
position in each of these contracts. The clearinghouse can be seen to have the
opposite side of each contract. Naturally, these contracts offset one another in
performance states. Thus, in states of the world where nonperformance options expire
unexercised, the final outcome from a complete clearing system is identical to that of a
ringing system. The routes taken by payments do differ in a complete clearing system.
In complete clearing, the cash payment made by A goes to the clearinghouse. The
clearinghouse, in turn, makes a cash payment to B. However, since the cash flows of
counterparties A and B remain unchanged, A and B are indifferent between a ringing
settlement and complete clearing. Thus, from the perspective of A and B, the complete




32

clearing system operates like a ringing system augmented by a clearinghouse to
facilitate payments between counterparties.
Contrasting this offset with that made through the ringing settlement, B’s original
and substituted counterparty is the clearinghouse. Recall that B’s minimum condition
for entering a ring settlement was a reduction in counterparty loss exposure. Provided
a choice between ringing settlements and settlements through a complete clearing
house, B again requires lower loss exposures as the minimum condition for accepting
complete clearing. With complete clearing, B is assured of no contract dependencies.
A result that, in itself, frequently can reduce loss exposure. Given B’s participation in a
ringing settlement, counterparty C’s ability to offset the contract with B is unchanged by
the adoption of complete clearing. C prefers complete clearing when it reduces
dependence on B’s decision to participate. Otherwise, C is indifferent between ringing
and complete clearing. Finally, like B, D’s preferences are determined after comparing
the loss exposures obtained from each settlement arrangement.
Treating the adoption of a clearing system as a permanent choice, B prefers a
complete clearing system if loss exposure from the clearinghouse is expected to be
less than that typically obtained from ringing arrangements. While some counterparties
may pose less exposure than the clearinghouse, the avoidance of contract
dependencies obtained by complete clearing implies lower loss exposures than those
realized from ringing arrangements. Electing to not adopt a complete clearing system
implies that B regards the proposed complete clearinghouse as a greater risk than a
typical counterparty.




33

C's consideration of these alternative clearing systems focuses on cost con­
cerns. C can anticipate the increased costs incurred each time that B elects to not
participate in ringing settlements. C is indifferent between clearing arrangements
provided B’s participation is assured. Complete clearing makes C’s offset automatic
rather than at B’s option, thus C’s vote is entirely based on higher expected costs under
a ringing system.
-As all traders can expect to find themselves on occasion in B or C shoes, ballots
for the adoption of complete clearing express a consensus of the members’ assess­
ments of the net values of these future loss exposures and cost savings. Traders
favoring adoption of complete clearing will be those who can either expect substantial
savings on their margin deposits or generally less counterparty loss exposure. Traders
having modest required margin deposits will expect little savings from adoption of
complete clearing. If these traders can also sufficiently manage their counterparty loss
exposures through a ringing system, they will lack incentives to vote for adoption of a
complete clearinghouse. Once adopted, complete clearing shifts the realization of
nonperformance losses to the clearinghouse. The loss-sharing arrangements made
between members of the clearinghouse serve to allocate any losses among these
members. Their need to control loss exposures from nonperformance motivates the
clearinghouse to adopt measures which reduce its exposure to nonperformance risk.
The adoption of rules by the clearinghouse is constrained in two ways. First, the
additional cost of adhering to clearinghouse rules cannot exceed the value added by
adopting complete clearing. Thus, members exert an external influence on




34

clearinghouse decisions. Second, clearinghouse loss-sharing arrangements imply loss
exposures from all open contracts. Members will avoid any loss-sharing arrangement
that inadequately compensates for this risk. As lack of participation in the loss-sharing
arrangement implies the clearinghouse is nonviable, this results in internal pressures
on clearinghouse decisions.
The internal pressure motivates pricing of clearing services to compensate
participants in the loss-sharing arrangement for their exposure to losses from contract
nonperformance. However, compensatory payments reduce the value of benefits
obtained from complete clearing creating external pressure to limit rates of
compensation. The compensation rate for the complete clearinghouse is its contract­
clearing charge per unit of loss exposure. Pricing its services at a level which just
satisfices its external pressures, its rate of compensation can be controlled by adopting
trading rules which limit loss exposures. However, as trading rules impose explicit and
implicit costs, these costs when added to the clearing fees would invoke external
pressure against the clearinghouse. Hence, adoption of contract rules requires clearing
fees at or below the indifference point of a majority of members. The clearinghouse
satisying this externally imposed boundary price will find its marginal condition is met
when reducing clearing fee revenues just equals the incremental cost stemming from a
rule change.4 Parties to the clearinghouse arrangement will be those members who
4
find this rate of compensation adequate for the level of exposure which results. Risk

44
This marginal condition presumes that most members are indifferent to an addition or loss in the
number of members. If additional members are valued, the sum of clearing fees and the cost of rules
may need further reduction to offset the value lost when members dissatisfied with the cost structure
leave the exchange.




35

neutral clearinghouse members will demand compensation just equal to expected
losses.
The analysis here suggests that adoption of complete clearing systems would
stem from needs to reduce counterparty exposure and to lower the cost of maintaining
open positions. The following comment provides direct evidence of the importance
attached to reductions in margin costs. After the Kansas City exchange adopted
complete clearing an observer contrasted its complete clearing with its former arrange­
ments} stating:
Under the system the tying up of large sums of money in margins, in event that a
long or short on the other end refuses to ’ring out,' is avoided. Thus, an evil
which tends to concentrate future trading into the hands of the stronger firms is
eliminated.4
5
At times, exchange members avoided the cost of carrying these balances by simply not
conforming to rules requiring quick responses to calls for margin. The following 1920
complaint from the Rogers Grain Co. to the Board of Directors of the CBOT illustrates:
This rule has practically become a dead letter with many. Very few members put
up margins until after the close even though they are called at Nine O’clock in
the morning, while the rules provide that margins shall be put up and evidence is
submitted of same within one hour,...4
6
This response reduced the cost of carrying margin balances, but increased
nonperformance risk and, notably, further elevated margin requirements to safeguard
against this risk.
Complete clearing systems originated in Europe. Emery (1896, p. 71-72)

45 “The Exchanges of Minneapolis, Duluth, Kansas City, Mo., Omaha, Buffalo, Philadelphia,
Milwaukee and Toledo," Annals of the American Academy of Social Science, 1911, p. 227-252.
46 BOD 10-26-1920.




36

indicates that European coffee exchanges featured complete systems. In France, a
complete clearinghouse was referred to as a Caisse de Liquidation and in Germany as
a Liquidationskasse. Hirschstein (1931, p. 213) indicates that after 1924, the produce
exchanges of Germany adopted complete clearing systems modeling them on a
system used for many years by a futures exchange operating in Hamburg. De
Lavergne’s (1931, p. 218) description of French clearing is consistent with the system
described by Emery in the late 19th century:
To replace the collateral security which might theoretically be required, but in fact
is not, there has been set up at some of the exchanges a bureau of settlement
(Caisse de Liquidation) in the form of an independent corporation. Such
bureaus are attached to the exchanges at Havre, Lille, and Roubaix, as well as
the sugar exchange at Paris.
When a contract is entered into, the function of the bureau of settlement is to
substitute for the original contract between buyer and seller, two new and distinct
contracts--one between the bureau as buyer and the original seller, and the
other between the buyer and the bureau as seller. As a result of this operation,
the individual buyers and sellers have no direct relations with one another, but
each has a contract with the bureau.
In the United States, officials of the New York Coffee exchange proposed to
copy the clearing system used in the European coffee exchanges. That proposal was
rejected-apparently more than once.4 Complete clearing was first adopted in the
7
United States by the Minneapolis Chamber of Commerce in 1891.4 That exchange,
8
later renamed the Minneapolis Grain Exchange, was first organized in October 1881.
The adoption of Rule VI in 1891 implemented its complete clearinghouse:
Section 1. All transactions made in grain during the day shall be cleared through

47 Emery (1896), p.72.
48 See FTC II, p. 17. I am indebted to Lester Telser who provided this reference.




37

the clearing association, unless otherwise agreed upon by the parties to the
transaction. Upon acceptance by the manager of such transactions, the clearing
association assumes the position of buyer to the seller and seller to the buyer in
respect to such transactions and the last settling price shall be considered as the
contract price.4
9
The operation of the Association’s clearinghouse is described as follows:
Most of the larger firms own memberships in it and it has been found to be
almost a vital necessity to the trade. Certainly it insures less friction than the old
way of trading and also facilitates business generally. When the trades are
checked at the close of the session the member gives a check to the clearing
house for margins or in case the market has fluctuated in favor of their
customers they receive a check. It does away with a great deal of trouble. To
•settle with the clearing house at a certain time every day is a far different matter
than calling each other for margins.5
0
Rule VII provided the Chamber of Commerce with the right to control its exposure to
nonperformance risk by requiring margins. The rule states:
Section 1. The manager of this Association may call from purchasers below the
market and from sellers above the market such reasonable margins as in his
judgement may be necessary for the protection of the association. Such
margins to be placed to the credit of the party paying the same and to be
retained by the manager, in whole or in part, as he may deem necessary until
the trades for which such margins have been paid have been settled.5
1
The Board of Trade of Kansas City (Missouri) was organized in 1869, but by the
mid-1890s was regarded as unsuccessful. In March 1899, the Kansas City exchange
organized a Board of Trade Clearing Company modeled after the clearing corporation
of the Minneapolis Chamber of Commerce. Referring to control over its exposure to
nonperformance risk, George G. Lee, clearinghouse manager said: "As the

49 FTC II, p .1 4 6 .
5 Annals (1911), p. 232.
0
5 FTC II, p. 146.
1




38

clearinghouse is responsible for all trades put through it, close tab must be kept on the
position of each member." The clearing manager of the Kansas City exchange had
wide latitude in setting margin requirements.5 The Duluth Board of Trade, first
2
organized in 1881, incorporated its clearing association in 1909 with rules patterned
after those adopted by the Minneapolis Association.5
3
The complete clearing systems established in the United States prior to 1925 all
followed the Minneapolis loss-sharing model. Each was incorporated with share
purchases limited to exchange members. Shares entitled members to clear the trades
of the exchange and to charge for this service. No other members were permitted to
clear trades. Incorporation limited the loss to the owners of the clearinghouse to the
value of their shares and the sum of their deposits held by the clearinghouse.
D. Evolution of clearing system?
This section summarizes the three methods of clearing. Markets employing
these methods have certain commonalities. Each market sought to increase the ability
of members to obtain contract offset. This ability increased the benefits that could be
derived from transacting in benchmark commodities rather than making similar but
more costly transactions in the actual commodities.
Participants in direct settlement markets were able to offset contracts provided
they could be assured of counterparty performance. The rules developed by markets
relying on direct settlement systems were responses to the problem of contract

6 Annals (1911), p .2 2 7 .
2
5 FTC II, p. 158.
3




39

nonperformance. The intent was to lessen the frequency and extent of non­
performance problems. Successfully avoiding these problems moved the members of
these exchanges closer to obtaining full contract offset. The two most common rules
adopted were exchange arbitration of disputes and the right to collect margin deposits.
Arbitration of disputes through the exchange avoided the slower and more costly
resolutions available in civil courts. Provision of member rights to collect margin moved
this source of protection from a right under common law where adequate time had to be
allowed for the posting of the required margin. Making the right to require margin a
right of membership, enabled quicker access to this protection as well as establishing
margin nonperformance as a contract default. This later feature enabled members to
protect themselves against further build up of nonperformance exposure.
Ringing systems took advantage of the increased contract fungibility obtained
from uniform grade standards. When the nominal terms of all contracts entering the
ring were identical, offsets could be obtained by ring participants These participants
must accept an externally imposed settlement price and must be willing to accept the
loss exposures implied by substituted counterparties. The benefits obtained from
ringingwere the reduction in contract dependencies and lower costs for maintaining
open positions. This later reduction coming principally through lower margining
requirements as contracts were offset. These features enabled more cost efficient
controls over nonperformance risks. Exchange rules adapted by binding the
participants to every ringing settlement to the agreements made by the ring.
Complete clearing established contracts as completely fungible. As every




40

contract accepted by the clearinghouse involved the clearinghouse as a counterparty,
the credit risk of every contract was identical. Exchanges created clearinghouses as
synthetic members of the exchange. Members of the clearinghouse became
participants in a loss-sharing arrangement. Faced with nonperformance risk,
clearinghouses adopted rules to limit their exposures to nonperformance losses and
sought compensation for exposures remaining.
Thus, clearing evolved from arrangements negotiated between individuals to
rules imposed by the exchanges. Flexibility lost during the course of this evolution
resulted in lower risks of contract nonperformance and lower costs of maintaining open
positions. The memberships of these acted in a manner consistent with Smith and
Warner (1979); that is, by binding themselves to externally determined rules they
achieved lower rates of contract default and lower costs of operation.

V. Summary
Futures contracts are defined as substitutes for associated cash transactions.
This definition enables a discussion of the evolution of controls over nonperformance
risk. Three clearing methods are discussed: direct, ringing and complete. The inci­
dence and operation of each is described. Direct clearing systems feature bilateral
contracts with terms specified by the counterparties to the contract. Exchanges relying
on direct clearing systems serve chiefly as mediators in trade disputes. Ringing is
shown to facilitate contract offset by increasing the number of potential counterparties.
Increased ability to obtain contract offset is valuable because counterparties can
reduce the number of dependencies of their outstanding contracts and can reduce the




41

costs incurred by maintaining open positions. Entry into a ring settlement was
voluntary; but on joining the ring, exchange rules bound the participants of the ring to its
settlements. Exchanges which cleared through ringing methods generally adopted a
clearinghouse to handle payments.
Complete clearing interposes the clearinghouse as counterparty to every
contract. This measure ensures that contracts are fungible with respect to both the
underlying commodity and counterparty risk. Exchange members benefit from
complete clearing because contract offset is automatic rather than dependent on
counterparty interest in offset. The loss-sharing arrangements of the complete
clearinghouse produces exposure to loss from every open contract. Members of the
clearing house respond by requiring compensation for this risk. As the amount of this
compensation reduces the value of complete clearing, the amount of this compensation
is bounded. Participants of the loss-sharing arrangement will substitute rules for pricing
up to the point where the marginal value of risk reduction obtained from rules equals
the marginal benefit from compensation.




42

Bibliography
Andreas, A.T. (1894): History of Chicago from the Earliest Time to the Present Time.
Baer, Herbert, Virginia Grace France and James T. Moser (1998): "Opportunity Cost
and Prudentiality: A Model of Futures Clearinghouse Behavior," Federal Reserve Bank
of Chicago working paper.
Bisbee, Lewis H. and John C. Simonds (1884): The Board of Trade and the Produce
Chicago: Callaghan.

Exchange: Their History, Methods and Law.

Boyle, James E. (1931): "Cotton Seed Oil Exchanges," Annals of the American
A c a d e m y of Social Sciences, 1931, p. 171.
Chicago Board of Trade, Annual Report, various issues.
Coase, Ronald H. (1937): "The Nature of the Firm," Economica 4, p. 386-405.
Cox, J., J. Ingersoll and S. Ross (1981): "The Relation Between Forward Prices and
Futures Prices," Journal of Financial Economics 9, pp. 321-346.
De Lavergne, A. (1931): "Commodity Exchanges in France," Annals of the American
A c a d e m y of Social Sciences, 1931, p. 218-222.
Dumbell, Stanley (1927): "The Origin of Cotton Futures." Economic History1, pp. 193201.
Edwards, Franklin (1984): "The Clearing Association in Futures Markets: Guarantor and
Regulator." The Industrial Organization of Futures Markets, editor Ronald W.
Anderson, Lexington Books, pp. 225-254.
Ellison, Thomas (1886): The Cotton Trade of Great Britain. (London: Effingham Wilson,
Royal Exchange).
Ellison, Thomas (1905): Gleenings and Reminiscences (Liverpool: Henry Young.)
Emery, Henry Crosby (1896): Speculation on the Stock and Produce Exchanges of the
United States. (New York: Greenwood Press, Publishers).
Federal Trade Commission (1920): Report on the Grain Trade (six volumes),
Washington DC.
Forrester, R. B. (1931): "Commodity Exchanges in England," Annals of the American
A c a d e m y of Social Sciences 151, pp. 196-207.




43

Gorton, Gary (1985): "Clearinghouses and the Origins of Central Banking in the U.S.,"
J o u r n a l o f E c o n o m i c H i s t o r y 42, p. 277-284.
Gorton, Gary and Donald J. Mullineaux (1987): "The Joint Production of Confidence:
Endogenous Regulation and Nineteenth Century Commercial-Bank Clearinghouses,"
J o u r n a l o f M o n e y , C r e d i t , a n d B a n k i n g 19, p. 457-468.
Hill, Lowell D. (1990): G r a i n G r a d e s a n d S t a n d a r d
(Champaign: University of Illinois Press.

s :

H is to r ic a l I s s u e s

S h a p in g

th e

F u tu r e .

Hirschstein, Hans (1931): "Commodity Exchanges in Germany,"
A m e r i c a n A c a d e m y o f S o c i a l S c i e n c e s 151, pp. 208-217.

A n n a ls

Hoffman, G. Wright (1931): "Governmental Regulation of Exchanges,"
A c a d e m y o f S o c i a l S c i e n c e s 151, pp. 39-55.

o f th e

A n n a ls

o f th e

A m e r ic a n

Kane, Edward J. (1980):"Market Incompleteness and Divergences Between Forward
and Futures Interest Rates" J o u r n a l o f F i n a n c e 35, pp. 221-234.
Kolb, Robert W. (1991):
Co.).

U n d e r s ta n d in g

F u tu r e s

M a r k e ts

(Miami, FL: Kolb Publishing

Lane, Frederick C. and Reinhold C. Mueller (1985): M o n e y a n d B a n k i n
a n d R e n a i s s a n c e V e n i c e , Volume 1. Johns Hopkins University Press.
Loman, H.J. (1931): "Commodity Exchange Clearing System,"
o f S o c i a l S c i e n c e s 151, pp. 100-109.

A n n a ls

g

in

M

o f th e

e d ie v a l

A m e r ic a n

A c a d e m y

Lurie, Jonathan (1979): T h e C h i c a g o B o a r d o f T r a d e
University of Illinois Press, Urbanna IL.

1 8 5 9 -1 9 0 5 :

T h e

D y n a m ic s

o f S e lf-

R e g u la tio n .

Milgrom, Paul R., Douglass C. North and Barry R. Weingast (1990): "The Role of
Institutios in the Revival of Trade: The Law Merchant, Private Judges and the
Champagne Fairs" E c o n o m i c s a n d P o l i t i c s 2, pp. 1-23.
Nevin, Edward and E. W. Davis (1970):
London.

T h e

L o n d o n

C le a r in g

B a n k s .

Elek Books,

Odle, Thomas (1964): "Entrepreneurial Cooperation on the Great Lakes: The Origin of
the Methods of American Grain Marketing," B u s i n e s s H i s t o r y R e v i e w 38, p. 439-455.
Parker, Carl (1911): "Governmental Regulation of Speculation," in
A m e r i c a n A c a d e m y o f S o c i a l S c i e n c e s pp. 126-154.




44

A n n a ls

o f th e

Powell, Edward (1984): "Settlement of Disputes by Arbitration in Fifteenth-Century
England," L a w a n d H i s t o r y R e v i e w 2 , pp. 21-43.
Spahr, Walter Earl (1926): T h e C l e a r i n g
Publishing Corporation, New York.

a n d

C o lle c tio n

o f C h e c k s

The Banker’s

Smith Jr., Clifford W. and Jerold B. Warner (1979): “On Financial Contracting: An
Analysis of Bond Covenants," J o u r n a l o f F i n a n c i a l E c o n o m i c s 7, p. 117-161.
Taylor, Charles H. (1917):
cago.

H is to r y o f th e

B o a r d

o f T r a d e

o f th e

C ity o f C h ic a g o ,

Telser, Lester (1986): “Futures and Actual Markets: How They are Related,"
B u s i n e s s S5-S20.

Chi­

J o u r n a l o f

Weiner, Robert J. (1991): "Origins of Futures Trading: The Oil Exchanges in the 19th
Century," Brandeis University Working Paper No. 301.
Williams, Jeffery C. (1980): T h e
dissertation, Yale University).

E c o n o m ic

F u n c tio n

o f F u tu r e s

Williams, Jeffrey C. (1982): "The Origin of Futures Markets."
306-316.
Williams, Jeffery C. (1986):




T h e

E c o n o m ic

F u n c tio n

45

M a r k e ts .

(Ph.D.

A g r ic u ltu r a l H is to r y

o f F u tu r e s

M a r k e ts .

56, p.