Access to Institutional Finance and the Use of Trade Credit

Author(s): Christina Atanasova

Source: Financial Management, Vol. 36, No. 1 (Spring, 2007), pp. 49-67
Published by: Wiley on behalf of the Financial Management Association International
Stable URL: http://www.jstor.org/stable/30133749 .
Accessed: 12/01/2015 00:31
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .
http://www.jstor.org/page/info/about/policies/terms.jsp

.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact support@jstor.org.

Wiley and Financial Management Association International are collaborating with JSTOR to digitize, preserve
and extend access to Financial Management.

http://www.jstor.org

This content downloaded from 111.68.99.250 on Mon, 12 Jan 2015 00:31:15 AM

All use subject to JSTOR Terms and Conditions

Access

to

Institutional
Use

of

Trade

Finance

and

the

Credit

ChristinaAtanasova*
I develop a conceptualframeworkfor analyzing the effect of the availability of institutional
loans on firms' demandfor supplier (trade)finance. I testfor the existence of credit constraints
and their effect on corporatefinancing policies. My empirical results support the hypothesisthat
trade credit is takenup byfirms as a substitutefor institutionalfinanceat the marginwhen theyare
credit constrained. Further, in line with studies on the credit channel of monetary policy
transmission,Ifind an increased reliance on trade credit byfinancially constrainedfirms during
periods of tight money.

For most firms,tradecreditis an essential element of their operations.In developed countries,the

majorityof firms rely heavily on tradecredit extension as a source of finance. In a FederalReserve
Boardstudy,ElliehausenandWolken(1993) note thatin 1987, accountspayableconstituted20% of
all non-bank,non-farm,small businesses' liabilities and 15% of all large firms' liabilities. On the
otherhand,accountsreceivablerepresentsone of the main assets on most corporatebalance sheets.
Therefore, an importantaspect of trade credit is the two-way nature of the transaction.Many
companies,particularlythose at intermediatepoints in the value chain, use tradecreditas customers
and provide it as suppliers.Thus, tradecreditrepresentsa substantialcomponentof both corporate
liabilities and assets.
Alongside this obvious economic importance,trade finance should be considered by policy
makersbecause of its ability to affect the outcome of policy interventions.For example, Davis and
Yeoman (1974) show evidence that large UK firms used trade credit to cushion themselves from
tight monetarypolicy in the late 1960s.
In this paper, I study the interdependencebetween the two major sources of short- to mediumterm corporatefunds: 1) institutionalloans (bank debt of short- to medium-termmaturity,lines of
credit,etc.) and2) tradecredit.Tradecreditis a more expensive financingalternativeto conventional
loans because suppliershave a higher direct cost of funds. For example, for suppliers,these higher
costs can take the form of inefficiencies in the collection of payments,but financialintermediaries
enjoy cost advantagesdue to specialization.
Previousstudies offernumerousexamples of how, for some firms,financialmarketimperfections
may create dependenceon tradecredit as a source of funds. Petersenand Rajan(1997) and Nilsen
(2002) arguethatfirmsthathave no access to marketsfor tradedlong-termsecuritiesor commercial
paperrely on tradecreditforfinancingduringeconomic downturnsandmonetarypolicy contractions.
Deloof and Jegers (1999) provide empirical evidence that the amount of trade credit used is

I am indebtedto Alexander Triantisand the anonymousrefereefor constructivecommentsand suggestions, which led to

significant improvementof this paper I would like to thankthe participants at the 34th FMA conferencein Denver, Kevin
Reilly, Keith Glaister and RobertHudsonfor helpfuldiscussions and ICCfor providing the data.
* ChristinaAtanasova is a Lecturerat the Universityof York,York,UK.

FinancialManagement* Spring2007 * pages 49 - 67

This content downloaded from 111.68.99.250 on Mon, 12 Jan 2015 00:31:15 AM

All use subject to JSTOR Terms and Conditions

50

* Spring2007
Financial
Management

determinedby the availabilityof internalfunds and is an importantalternativenot only for shortterm bank debt but also for long-termfinancialdebt.
Antov (2005) and Alphonse, Ducret, and Severin (2004), on the other hand, argue that the
availabilityof institutionalfinanceincreaseswith the level of tradecredit.Antov examines firms'
choice to use trade credit and in particularthe way in which the availability and level of
institutionalloans affect that choice. His finding that higher levels of tradecredit are associated
with higherlevels of conventionalloans suggeststhatthereexists synergyarisingfromcombining
supplierand bank credit.Alphonse et al. provide similarevidence. They arguethat tradecredit is
used as a qualitysignal thathelps firmsacquirereputationand improvetheiraccess to institutional
finance. Both of these studies,however,base their analyses on cross-sectionaldataand assume a
static environmentand the absence of macroeconomicshocks.
The two lines of researchare not mutuallyexclusive: firms use tradecredit because they are
denied access to institutionalfinance (a demandeffect), but also tradecreditgrantedby suppliers
facilitates access to institutionalloans (a supply effect). I examine the hypothesis that over time,
variationsin the use of tradecreditare affected by the availabilityof cheapersubstitutes,such as
bank loans, and that this substitutioneffect is likely to be strongerduringrecessions and periods
of tight monetarypolicy. In my analysis, I include macroeconomicvariablesthat controlfor the
prevailing monetary policy conditions and capture the supply side variation in institutional
loans.
It is difficult to disentanglethe supply and demandeffects in a cross-sectional data analysis.
However, in the context of panel data, I observe that over time, a certain class of borrowers
systematically increases its use of trade credit when the level of institutionalfinance declines.
This finding suggests a strongdemandside effect.
I hypothesizethatwhen creditconstraintsarebinding,tradecreditis a substituteform of shortterm financing to conventionalinstitutionalloans. That is, when firms are constrained,I expect
that the use of trade credit as a substitutesource of funds for institutionalloans is decreasingin
bank loans, but when firms are unconstrained,I expect this substitutioneffect to be weaker,
because firms with alternativesources of funds will avoid the more expensive tradecredit. My
empiricalfindings provide strongsupportfor this proposition.
Testingfor the effects of creditconstraintson the choice of financingalternativesraises difficult
identificationproblems. In previous studies, the basic strategy has been to divide the sample
firms into groups that may face differentdegrees of informationand agency problems.Most of
these classificationsarebased on some time-invariantpriorcriterion,andthe researcherestimates
separateequationsfor each subsample.
However, there are severalproblemswith this approach.First,it separatesfirms into groupson
the basis of a single indicatorthat may not be a good proxy for credit quality and access to
institutional finance. In general, the use of a single indicator prevents the researcher from
controllingforthe manyfactorsthatinfluencefirm'sborrowingability.Second,in the conventional
strategy, whether a firm belongs to the financially constrained or unconstrained group is
determinedexogenously and is fixed over the entire sample period. This approachis far too
restrictive,since it does not allow forfirmsto switchbetweenthe constrainedandtheunconstrained
groupsover time.
To deal with potentialmisspecificationproblemscausedby the issues discussed above, I follow
the studies on credit marketdisequilibriathatmeasurethe impact of credit constraintsdirectly.I
use an endogenousswitchingregressionmodel thatallows me to derivethe probabilitythat a firm
faces credit constraintsdirectly from the distributionof the firm's characteristics.I estimate the
model on the basis of individualfirm-level data for a period of 20 years. When the time span of
the sample is long, it is importantthatfirms are allowed to switch between groups over time.

This content downloaded from 111.68.99.250 on Mon, 12 Jan 2015 00:31:15 AM

All use subject to JSTOR Terms and Conditions

Atanasova* Institutional
Financeand the Use of TradeCredit

51

Since I do not know a prioriwhich firmsare subjectto creditconstraints,I estimatea switching

model with sample separationunknown.I assumethat each firmat each point in time operatesin
one of two possible regimes: a credit-constrainedregime or a credit-unconstrained
regime.'
An advantage of my method is that it allows me to analyze how substitutionin corporate
financingalternativesis likely to vary over time with the prevailingmacroeconomicenvironment
and with the monetarypolicy conditionsin particular.I use the patternsof borrowingconstraints
imposedby financialintermediariesto distinguishbetweendemandand supplyside effects. I find
that when monetarypolicy is tight, the average probabilityof being in the constrainedregime
rises and the creditconstrainedfirms increasetheir use of tradecredit.This finding suggests that
monetarypolicy may affectfirms'access to institutionalloans.Althoughcontractionarymonetary
policy has an importanteffect on the supply of institutionalcreditby limiting its availabilityto a
certain class of borrowers,it does not affect the demand for loans in the same direction. In
general,the demandfor credit will increasedue to the reducedcash flow caused by a decline in
productdemand.By observingthese patternsover time, I can begin to make inferencesaboutthe
"cycle"of corporatefinancing.
The paperis organizedas follows. Section I providesa review of the theoreticaland empirical
literatureon the motives for tradecredituse. Section II presentsmy switching regressionmodel
of the demand for tradecredit and describesthe estimationtechniquefor the model. Section III
describes the data set and presents some descriptivestatistics.Section IV reportsthe estimation
results for the model. Section V considers several checks to the robustness of the estimation
results. Section VI examines the effect of the prevailing macroeconomic conditions, and in
particularmonetarypolicy stance, on the severity of credit constraintsand corporatefinancing
choice. Section VII concludes with a summaryof the main findings.

I. MotivesforTradeCreditUse
Long, Malitz, andRavid(1993) point out thattheoreticalstudieson tradecredithave developed
in differentbut not necessarilymutuallyexclusive directions.Althoughdifferenttheoriesattempt
to explainthe existence of tradecredit(see Ng, Smith,and Smith, 1999), thereis little systematic
researchon which firmsare most relianton tradecreditand when.

A. Transaction Motive
Many firms use at least some trade credit as a normal part of their transactioncycle. The
transactiontheory of trade credit is mentioned in Schwartz (1974) and discussed in detail in
Ferris(1981). For some firms, the primarybenefit of tradecredit is as a cash managementtool.
By delaying the paymentfor purchases,a firm may be able to match the timing of cash receipts
from sales with the cash outlays for the cost of goods sold. The transactiontheory predictsthat
this cost saving motive for trade credit use will help explain a significantpartof the amountof
accountspayable on firms'balance sheets.
At the time when the productis purchased,tradecredit can also play a role in a firm's quality
control efforts. Smith (1987) and Long, Malitz, and Ravid (1993) argue that the use of trade
credit allows a firm to verify productqualitybefore paying. These types of benefits suggest that
'I use these terms loosely. The constrainedregime includesnot only the case in which firmsare literallyrationed,but also
when the premiumon externalfinanceis prohibitivelyhigh or the managersdislike the monitoringand controlproperties
of debt.

This content downloaded from 111.68.99.250 on Mon, 12 Jan 2015 00:31:15 AM

All use subject to JSTOR Terms and Conditions

FinancialManagement* Spring2007

52

the use and perceived importanceof trade credit may be related to a firm's industry.If the
complexity of the items purchasedby a firm increases, the firm's transactiondemand for trade
credit may increase as well.

B. FinancingMotive
The financemotive focuses on the customeras initiatorof the post purchase"creditextension."
Ferris(1981) arguesthattradecreditbecomes less an instrumentof tradeand more an instrument
of financeas the lengthof creditperiodincreases,with the seller actingas a financialintermediary.
Tradecredit extension then becomes a type of short-termloan between seller and buyer that is
tied to the exchange of goods in terms of value and timing (Frankset al., 1985).
Theory links the financing motive to credit marketimperfections,which may cause financial
institutions (the major source of business credit) to restrictcredit to their customers.Although
granting trade credit exposes the firm to financial risks, the supplier may be willing to offer
financing to constrainedborrowersbecause the firms have broaderinterests than the financial
transaction.The suppliermay benefit in the longer runby helping a struggling customerstay in
business and therefore make future sales. The supplier is often in a better position to obtain
informationabout the creditworthinessof a buyer than a financial intermediary.Contact from
the selling process can facilitate the monitoringof customerson an ongoing basis and, as Smith
(1987) suggests, suppliers can also use two-part terms to obtain information on credit
worthiness. Biais and Gollier (1997) and Jain (2001) argue that suppliers have private
informationabout their customersthat banks do not. Thus, savings in monitoringcosts and the
informationaladvantagemay explain why some firms provide tradecredit. Habib and Johnsen
(1999) suggest suppliers have repossession advantages when redeploying the asset sold and
Wilner (2000) argues that restructuringdebt advantages explain why trade credit is being
offered.
Previous empirical work on the financing motive of trade credit provides mixed evidence.
Long, Malitz, and Ravid (1993) find no support for the financing motive: less creditworthy
nonfinancialfirmsdo not applyto more creditworthyfirmsfor financingdue to creditconstraints.
Ng, Smith, and Smith (1999) obtain similarresults,but Antov (2005) andAlphonse, Ducret and
Severin (2004) find that firms with high levels of trade credit have high levels of institutional
loans.
In contrast,Nilsen (2002) in the case of the U.S., and Biais, Hillion, and Malecot (1995) for
France,show that small firms,which are more likely to be creditrationed,rely heavily on trade
creditwhen creditmarketconditionsdeteriorate.Also, Petersenand Rajan(1994, 1995) find that
firms that are less likely to be bank credit constrainedtend to rely less on trade credit. It is
possible thatthe ambiguousempiricalevidence on the financingmotive of tradecredituse is due
to the static and dynamic misspecificationscaused by the conventionalclassifications used to
split sample firms into constrainedand unconstrained(e.g., small versus large firms).
When financial institutionsrestrictcredit, those firms dependingon intermediatedfinanceare
forcedto use tradecredit.Given thattradecredittermsremainconstantover time (see Ng, Smith,
and Smith, 1999), during recessions and monetary contractions,trade finance may become a
relatively cheaper source of funds for some firms. It is this imperfect substitutabilityof trade
credit and institutionalfinance that allows me to observe systematic differences in the use of
tradecredit.Therefore,the increase in the demandfor tradecredit as a source of finance during
monetarycontractionsprovides a test for the existence of a credit channel of monetarypolicy
transmission.These argumentsare consistent with Laffer's (1970) suggestion that trade credit
should be considereda partof money supply.

This content downloaded from 111.68.99.250 on Mon, 12 Jan 2015 00:31:15 AM

All use subject to JSTOR Terms and Conditions

* Institutional
Financeandthe Use ofTradeCredit
Atanasova

53

II.A Switching Modelof TradeCredit

To begin, I conceptualize the link between credit constraints and trade credit demand. I
represent firm i's notional demand and supply curves for institutional loans at time t by
and Ls(r,,Z,, u2,i,),respectively.I use r1to denote the marketinterestrate. is a
LD(r,
Z,,,u,,)
Z,
vector of observablefirm characteristicsthat determinethe demandand supply of loans for firm
i at time t, and u1,,and u2,itare variablesthat representthe unobservablecharacteristicsfor the
same firm in the same time period. I denote the firm's excess demandfor loans at time t by:
L*it

(1)

LD(r,,ZitU!~ir)-LS(rrZirUZ.ir).

The excess demandquantityL*,is not observable.I define an indicatorvariableLi.

ifotherwise.
it
0

(2)

I am interestedin the determinantsof the probabilitythat there is excess demand for loans,
i.e., Pr(L*,> 0). I assumethatcreditavailabilityis a functionof the firm'scharacteristics,andthat
for firm i at time t I can write the excess demand quantityas L*,,= Z' y + u3,i,where is a
is the innovation term that captures, unobservable
parametervector (to be estimated) and
qualities of borrowers.Then Pr(L*,,> 0) =u3,t
Pr(Z',1y + u3,it>0). (I note that if u3,itis Gaussianwith
meanzero andvarianceone, this formulationleadsto a standardprobitmodel.) The unconstrained
firms receive the desiredquality of bank credit (L, = 0), but for the constrainedfirms I observe
thatthe maximumcreditgrantedis LS,since LDi,> Li.
I denote the quantityof trade credit obtainedby the constrainedand unconstrainedfirms by
TCC,andTC' ,, respectively.In general,for firm i at time t I can write the expected level of trade
creditas:
E(TCUIII L, = 0) = X ',

/u

ULS, + E(E,

IL, = 0)

(3)

-+

or
= 1)

E(TCC, I

+c

/c

1).

(4)

In the two Equations (3) and (4), XA,is a vector that includes the two types of observed
heterogeneitythatdeterminethe level of tradecredit.One is a groupoffirm-specificcharacteristics,
and the other is a groupof variablesspecific to particularsupplierpracticescommon for firms in
the same industry.The characteristicsthat are not observable to the econometrician,but that
affect the level of tradecreditused, fall into the errorterms El,it and
tradecreditis a substituteto
Accordingto my proposition,when creditconstraintsarebinding,62,it.
conventionalinstitutionalloans. I expect thatthe amountof tradecredit used by the constrained
firms will decrease with the amount of short-terminstitutionalloans received, i.e., 6c < 0; and
< 0. If trade
that there should be no such substitutioneffect for the unconstrainedfirms, i.e., 86U
creditcomplementsconventionalloans, then I expect 6c > 0 and 36U> 0.
To obtain a measureof trade credit usage, I use the ratio of accounts payable to total assets.
(I note that all stock variablesdenote end-of-periodvalues.) This ratio gives the percentageof
total assets that is financed by trade credit, and thus representsa firm's reliance on interfirm
credit.For similarreasons,I scale institutionalloans (LS,)by total assets.
Inthe empiricalspecificationof the switchingmodel,firmi attime t operatesin theunconstrained
regime with a demandfor tradecredit equationdefinedby:
A
TA,

=__

,,

it

if
, --

<0

Z'1it
Y+u3,it

This content downloaded from 111.68.99.250 on Mon, 12 Jan 2015 00:31:15 AM

All use subject to JSTOR Terms and Conditions

(5)

FinancialManagement* Spring2007

54

or the firm operatesin the constrainedregime with a demandequationdefinedby:

=

Ai'

X 'it

TA,

6C

+
TA

E2,it

if Z

2 0.

(6)

u3,it

A. ModelEstimation
I estimate a switching regression model in which I obtain observationsfrom two different
regressionregimes, but the separationinto these regimes is unobservablea priori. I assume that
the vector of errorterms
e2,itu3,it)'is jointly iid Gaussianwith mean zero and covariance
(y,;,,
matrix1, where
} i,j = 1,2,3.
(3x3) -0ij
between E,,, E2,it and u3,it (endogenous switching) allows for possible
The nonzero covariance
correlationbetween the shocks to the demandfor tradecreditwith the shocks to firms'financial
and othercharacteristics.AlthoughI cannotobservewhich regimethe firm is in, I can specify the
probabilitywith which each regime occurs as follows:

= Z',
<0]
pr[ 'ITA,, =(AP.,/TA,,
y+u3,it
=
<-Z ', r] = (-Z 'it Y)
Pr[u3,it

(7)
(7)

Pr[A,

)c =

(AjP,/
TA,,_
Pr[Ai,/TAp,,i,
=
2 -Z ', y] =

PrZ',

u,it

(8)
0]

The likelihoodfunctionfor each observationAP, / ,_ is a weightedaverageof the conditional

TAi
density functions of l,it and
with weights Pr u3,,<-Z',, y] and Pr[u3,it -Z', y]. The
2,it
likelihood function is given by:
--O(
it

<

[/'/,i,

-Z

Y)]

Y])
'i,

'it

'it

011

(9)

where b(.)and P(.) are the normalprobabilitydensity and the cumulativedistributionfunctions,

.) denotesthe conditionaldensity,and
) denotesthe marginaldensity
respectively.
q(e,1, 0
function. q(Ei,
The second equality of Equation(9) uses the fact that a joint density equals the productof the
conditional density times the marginal density and the properties of the bivariate normal
distribution.Like the probit model, I cannot estimate 7 and
separately,so without loss of
-33
does not appear in Equation (9) and thus is not
generality I normalize 33= 1. Also,
-12
estimable.
For a panel of N firms with T observationsfor firm i, the log-likelihood function for all the
observationsis given by:
NT

L=

(10)
log(i,t).
I estimatethe parametervectors /", 3"and y and the parameters6" and 6" by maximizingthe
log-likelihood functionthroughthe EM algorithmof Dempster,Laird,and Rubin (1977).
i=1 t=l

This content downloaded from 111.68.99.250 on Mon, 12 Jan 2015 00:31:15 AM

All use subject to JSTOR Terms and Conditions

Financeand the Use of TradeCredit

Atanasova* Institutional

55

If I knew which regimegeneratedeach datapoint, the problemwould be easy to solve. To start

the iterativeprocedure,I split firmsinto two regimes accordingto their size. (I classify firmswith
total assets above/belowthe mediantotal assets for given year as large/smallin thatyear.)When
the procedure eventually converges, I overwrite the initial guess variable to contain the
probabilitiesthat a given observationadheresto the first (unconstrained)regime.
Given an initial guess of the partitionof the observationsinto the two regimes, the algorithm
estimates the classification vector, i.e., the probability that a given observation is in the
unconstrainedregime(the expectationstep). I obtainthis estimateby reweighingthe probabilities
based on the errorsof the observationsin the two regressionEquations(5) and (6). I then use the
updatedprobabilitiesto weight observations in each of the two equations (the maximization
step).As noted above, this iterativeprocedureeventuallyconvergesto the MaximumLikelihood
estimates(MLEs) of the parameters.
To ensure robustnessto heteroskedasticityand within firm (cluster) correlation,I compute
bootstrappedstandarderrorsas in Douglas (1996). I drawfirms(clusters)with replacementrather
than drawing observationswith replacement.I note that the general bootstrappingprocedure
of drawingobservationswith replacementdoes not necessarilycontrolfor heteroskedasticityand
within-firm clustering. On the other hand, bootstrappingby firms takes account of these
problems.
An importantquestion is whether the estimated endogenous switching model is statistically
significant relative to a linear single-equation model. Although the single-equation model is
nested within the switching model, under the null hypothesis of linearity there are several
unidentifiedparametersof the switching model. This problemcomplicatesthe calculationof the
degrees of freedom. In addition, the asymptotic likelihood ratio statistic does not have the
conventionalchi-squaredistribution.However,Monte Carlotests suggest thatsettingthe degrees
of freedomequal to the numberof constraintsplus the numberof unidentifiedparametersyields
a conservativetest when I use the chi-squaredistribution.
From the likelihood density function I calculate both the unconditional and conditional
probabilities of each sample firm being in the constrained regime. The unconditional
probability does not take into account the information about the amount of trade credit
used, and is simply Pr[u3,it -Z'i, 7]. The conditional probability takes into account this
information and updates the unconditional probability according to Bayes' rule. The
conditional probability that a firm faces credit constraints given the information about the
amount of trade credit used is:

' A],
Pr u3,it 2 -Z
/
l
(-2,

=
3,it

Forbrevityof exposition,I reportonly the unconditionalprobabilityof being in the constrained

regime.

III.Descriptionof Dataand Variables

In my empiricalwork,I use panel dataon UK firmsprovidedby ICC.The ICC databasecovers
the entirepopulationof limited-liabilityUK firms. I select a randomsample of 2,000 firms over
a 20-year period, from 1981 to 2000, as a stratifiedsample from the ICC SIC codes. The sample
firms in this study are of all sizes and industriesexcluding firms in sectors such as financial

This content downloaded from 111.68.99.250 on Mon, 12 Jan 2015 00:31:15 AM

All use subject to JSTOR Terms and Conditions

FinancialManagement* Spring2007

56

intermediationand otherfinancialservices. The sample consists of 1,839 (91.9%) privatelimited

companiesand 162 (8.1%)public limited companies.Unlike in the US, where only quotedfirms
arerequiredto file theirquarterlyor annualaccounts,UK firmsmust disclose their accountseven
if they are not tradedon the stock exchange.
A. Vector of Variables Z
In the switching function,in additionto age, industry,andyear dummies,the vector Z, includes
a set of variablesthatare indicatorsof financialstrength.The financialvariablesmeasurea firm's
ability to raise capital and offer collateral.I measureliquidityby the level of profits(PROF/TA)
andthe quickratio(currentassets over currentliabilities,QR). I use assets (logTA) andtangibility
(depreciationover assets, DEPR/TA)as my proxy for the firm's ability to collateralizeits debt.
To control for the capital structureof the firm, I include the values of the firm's leverage ratio
(LR=DEBT/TA)and of the coverageratio(operatingincome-to-interestratio,CR). I also include
in the switching functiona statusdummyvariablePD,, which equals one if ithcompanyis public
and zero if it is private. I expect that privatefirms will find it more difficult to acquireexternal
finance than will publicly held firms, both because of informationopacity and limited access to
public capitalmarkets.To investigatewhetherfinancialweakness is particularlyimportantfor the
borrowingability of privatefirms,I allow for the interactionbetween the financialratios and the
statusdummy.
Theoreticalfactorssuggest thata firm is more likely to face creditconstraintswhen its level of
profits, assets, and tangibilityare low; when its quick and coverage ratios are low; and when its
leverageratiois high (PetersenandRajan,1997). Privatefirmsaremorelikely to face information
and agency problems. Older firms tend to be more diversified,less prone to bankruptcy,have a
better track record, and therefore should suffer less severe agency costs. Thus, I expect the
coefficients of age and statusdummyto be negative.
Includingyear dummiesin the switchingfunctioncaptureschanges in the generalfinancialand
macroeconomic conditions that affect all firms in the same way. In years characterizedby
recessions or tight monetarypolicy, I expect the coefficient of the year dummy to be larger.In
additionto the observedfirm-specificvariablesexplicitly includedin the model, I may want the
probabilityof facing credit constraintsto depend on unobservablefirm-specific,time-invariant
factors. If I treatthese factors as fixed effects, I can introduce2,000 firm-specificdummiesin Z
and X. Although this methodof handlingfixed effect is possible in principle,in practiceit is not
feasible since it leads to a huge loss of degrees of freedom. To be invertedin the maximization
procedure,the matrixis of dimension(N+K), where N=2000 and K is the dimensionof either Z
or X. The inclusion of industrydummies, for each two-digit SIC code serves to capturethese
effects.
I note that the negative of the sum of the coefficients of the year and industrydummies can
be interpretedas a threshold.If a linear combinationof the financialvariablesand age exceeds
this threshold,then I expect the firm to be in the constrainedregime in the sense that Z',it~ 2 0.
I also note that the ex ante probability of being in the constrained regime is
Pru 3,, -it 'j
1- (-Z 'i,y).
B. Vector of Variables X
In the tradecreditregressions,the vector includesa set of observablefirm-specificvariables,
X,
industry-wideaverages,and industryandyear dummies.The demandfor tradecreditattributable
to the financingmotive is associatedwith the cost and availabilityof substitutesources of funds.

This content downloaded from 111.68.99.250 on Mon, 12 Jan 2015 00:31:15 AM

All use subject to JSTOR Terms and Conditions

Atanasova* Institutional
Financeand the Use of TradeCredit

57

A pecking-orderbehavior (Myers and Majluf, 1984) would suggest that internally generated
funds are higher in the pecking orderthan tradecredit. Therefore,the firm's ability to generate
cash internallywill decrease its demand for trade credit. I calculate cash flow as profits plus
depreciationand use the cash flow-to-asset ratio (CF/TA)as my proxy for the firm's reliance on
internallygeneratedfunds.
The generaldemandfor funds dependson the expected futureprofitability.I expect firmswith
higher expected futureprofitabilityto have a higher proportionof their assets financed with
accountspayable. I use the sales-to-assetsratio(SALES/TA)as my proxy for futureprofitability.
(I note thatthe resultsarenot materiallydifferentwhen I use sales growthas my proxy for future
profitabilityinstead of the SALES/TAratio.)
Another dimension of the financing motive for trade credit demand is the maturityof the
planned investment.Finance studies suggest a maturity-matchingapproach(Deloof and Jegers,
1999). Therefore,firms with more short-termassets will have a higher demand for short-term
credit in general and trade credit in particular.To capturethis effect I add three categories of
short-termassets: 1) inventory-to-assets(INV/TA), 2) accounts receivable-to-assets(AR/TA),
and 3) cash holdings-to-assets(CASH/TA)ratios.
The industry-wideaveragescontrol for the effect of common supplierpractices.I define these
averages as follows. The demandfor tradecreditattributableto the transactionmotive depends
on the firm's uncertaintyin transactionswith its suppliers. Industriesassociated with a higher
variabilityof inventorylevels should have a higher averagevolume of purchasesfrom suppliers
as their inventories are depleted, and thereforea higher demand for trade credit. The standard
deviation of inventory turnover (SD(TURN)) represents the variability or uncertainty in
transactionswith suppliers.2I also include the industry-wideaverage of accounts payable over
total assets (AP/TA), since the greaterthe industrypracticein relying on tradecredit,the more a
firm operatingin thatindustrywill use it.
Overallbusiness conditionscan have a more directeffect on short-termfinancingdecisions, so
I also include year dummiesin the tradecreditequation.Further,I include industrydummiesfor
each two-digit SIC code. The inclusion of industrydummies capturessome of the influence of
the unobservablefirm-specificeffects.
C. Descriptive Statistics
Table I reports means and medians of firm level variables for public and private companies.
The t tests show that the means are significantly different at 1% for every firm characteristic
except for profit-to-assets, quick, and leverage ratios. The results are robust to using sales
instead of assets as the scaling factor and to controlling for every two digit SIC industry
code.
TableI clarifiesseveralfacts. The firstis thatpublicfirmsare much largerthanprivatefirms,as
measuredby all three indicatorsof size: volume of sales, value of total assets, and numberof
employees. Publiccompaniesalso have lower SALES/TAratioas well as lower INV/TAandAR/
TA ratios. Public firms appearto rely less on short-termloans and trade credit than do private
firms.
The low sales SALES/TAratio of public firms may reflect a lower cost of capital, which in
equilibriummay imply a more capital-intensiveproductionprocess.
It is possible thatthe greateruncertaintymay be associatedwith a largedeviationof the sales to inventoryratio from its
"optimallevel," which has been changingover time due to the introductionof new inventorymanagementtechniques in
the early 1990s. I performtests for possible structuralbreaksin the trade credit equation.The resultsof the tests do not
suggest the presence of a break.The results are availableon request.
2

This content downloaded from 111.68.99.250 on Mon, 12 Jan 2015 00:31:15 AM

All use subject to JSTOR Terms and Conditions

FinancialManagement* Spring2007

58

Table I: Descriptive Statistics for Public and Private Firms

The samplecontains2,000 UK firmsfrom 1981to 2000. Themeanandmediannumbersare in British
deviationsarereportedin parenthesis.
pounds.Standard
Public
Variable
log(SALES)
log(TA)
SALES/TA
Employees
Short-term
Liabilities
AP/TA
L/TA
Short-term
assets
CASH/TA
AR/TA
INV/TA
Liquidity
PROF/TA
QR
Capitalstructure
LR
CR

Private

Mean

Median

Mean

Median

17.6(2.3)
17.8(2.3)
1.2(1.2)
6934.8(19913)

17.3
17.6
1.2
418

15.9(1.6)
15.5(1.8)
2.3 (15.4)
467.7(2582)

15.7
15.3
1.6
85

12.6%(12.2)
11.9%(17.4)

10.2%
6.9%

16.6%(17.2)
21.3%(97.9)

12.5%
11.8%

6.2%(9.5)
16.9%(15.3)
17.9%(16.4)

2.5%
14.2%
16.0%

8.8%(14.4)
22.4%(18.2)
20.4%(18.9)

2.1%
20.2%
17.0%

0.051(0.16)
3.8 (98.8)

0.058
1.3

0.077(2.79)
3.3 (50.5)

0.052
1.3

50.7%(40.5)

44.3%

56.7%(251.6)

44.4%

46.8 (674)

3.9

23.7 (258)

3.0

Second, public firms hold lower stocks of inventories relative to assets, which may reflect
differences in factor costs related to capital market imperfections.Stocks of inventoriesrelax
financial constraints, since they are the preferred form of collateral for bank loans. Also,
inventories (particularlyraw materials, which constitute the bulk of inventories) are easily
appraisedand liquidated.Thus,firmsthatdependon institutionalcreditmay find it advantageous
to select material-intensiveproductiontechniques,or they may be requiredby lending covenants
to maintaina minimumamountof working capital.

IV.EstimationResults
In Table II, I report the estimated parametersof the conventional linear equation model. I
report robust standard errors to control for heteroskedasticity and within cluster (firm)
correlation.
Column (1) in TableII containsthe estimatedparametersfor the whole sample. I estimatethe
paneldatamodel with randomeffects. The resultsshow thatthe demandfor tradecreditdecreases
with institutionalloans. The estimation results supportboth the maturity-matchinghypothesis
and the pecking-ordertheory.
The coefficients of cash holdings, inventory and accounts receivable are positive and highly
significant. In contrast, internally generated funds have a negative effect on the demand for
trade credit. The coefficient of the standarddeviation of the industry-wideinventory turnover
ratio is positive and highly significant, which suggests that greateruncertaintyof transactions

This content downloaded from 111.68.99.250 on Mon, 12 Jan 2015 00:31:15 AM

All use subject to JSTOR Terms and Conditions

Financeand the Use of TradeCredit

Atanasova* Institutional

59

Table II:Estimation of Linear Regression Model

Thetableshowsthelinearregressionresults.Theregressionis:

=
TA.,_

+ Ei.
TAi.,_

(T.2.1)

errorsrobustto heteroskedasticity
and
Thesamplecontains2,000UK firmsfrom1981to 2000. Standard
withinfirmclusteringarereportedin parentheses.
Whole sample
L/TA
X variables
CF/TA
SALES/TA
AR/TA
INV/TA
CASH/TA
Industryaverages
SD(TURN)
AP/TA
Log-likelihood

Large firms

Small firms

-0.06298 (0.0023)***

-0.10592 (0.0056)***

-0.04883 (0.00246)***

-0.0003 (0.0002)
0.0058(0.0005)***
0.1102(0.0046)***
0.0331(0.00512)***
0.024(0.0059)***

-0.11545(0.0099)***
0.00581(0.0005)***
0.144879(0.0097)***
0.08697(0.01046)***
0.00976(0.0117)

-0.00066(0.00246)
0.01225(0.00144)***
0.10033(0.005)***
0.0164(.005615)***
0.02772(0.0063)***

0.0002(5.89e-06)***
0.0036(0.0005)***
12591.31

0.0001(7.00e-06)
0.003(0.00977)
4834.57

0.0003(1.73e-06)***
0.003(0.00098)***
8712.10

atthe0.01level.
***Significant

with suppliers is associated with greater demand for accounts payable. The coefficient of the
industry-wide average of accounts payable to assets is also positive and significant.
The second part of Table II reports the estimated parametersof the linear model for the
subsamples of small and large firms. I find that for small firms, trade credit is decreasingwith
institutionalfinance. The coefficient of institutionalloans is largerthan the coefficient for the
whole sample and is highly significant. For large firms, trade credit is also decreasing with
institutionalfinance.Althoughthe coefficient of loans is smaller,it is still highly significant.This
finding is at odds with the fact that researchersassume that large firms have wide access to
institutionalfinance.
Table III presentsthe estimatedparametersfor the switching model. The constrained-regime
equation resembles the linear model, but with some differences. Most notably, the effect of
institutionalloans on the demandfor tradecreditis much stronger,which is also trueof the effect
of internallygeneratedfunds, albeit theireffect is only weakly significant.On the otherhand,the
unconstrained-regimeparameterestimatefor institutionalloans contrastssharplywith the linear
model estimates.Thecoefficienton institutionalloans is not statisticallysignificantatconventional
levels and is much smaller in magnitude.The rest of the results are similar to the estimates in
TableII.
Table III also reports the estimated parameters of the switching function. These results
are consistent with the findings of previous studies. In particular, the coefficients on the
profits, assets, and tangibility, quick, and coverage ratios are significant and negative,
and the coefficient on the leverage ratio is significant and positive. The age coefficient is
negative.
In line with the discussion in the previous subsection, the estimated coefficient of the status
dummy has a negative sign. Of the interactionterms, the coefficient of the leverage ratio is

This content downloaded from 111.68.99.250 on Mon, 12 Jan 2015 00:31:15 AM

All use subject to JSTOR Terms and Conditions

FinancialManagement* Spring2007

60

Table IIl:Estimation of Switching Regression Model

Thetableshowstheswitchingregressionresults.Theregressionequationsare:

Ai'

TAT
AT

',,

u+

X 'it c

8" ,

TA1

if Z'it Y+

+
Elit

< 0.

(T.3.1)

u3,it

if

Y+

(T.3.2)
u3,it20.
standard
errorsarereported
Thesamplecontains2,000UKfirmsfrom1981to 2000.Robustbootstrapped
in parentheses.
Switching Model
C regime
L/TA
X variables
CF/TA
SALES/TA
AR/TA
INV/TA
CASH/TA
Industryaverages
SD(TURN)
AP/TA

UC regime

-0.1251 (0.0114)***

-0.0158 (0.034)

-0.0369 (0.028)
0.0189(0.003)***
0.497(0.012)***
0.3346(0.012)***
0.284(0.019)***

-0.046 (0.0084)***
0.003(0.001)***
0.189(.0066)***
0.099(0.005)***
0.0001(0.006)
0.00002(7.31e-06)***
0.051(0.0103)***

0.00003(0.00001)***
0.0433(0.021)***

Z variables

Switch Function

Age
PROF/TA
QR
logTA
DEPR/TA
LR
CR
PD
PD*QR
PD*LR
PD*CR

-0.0121 (0.0002)***
-0.449 (0.022)***
-2.799 (0.0036)***
-0.088 (0.002)***
-7.741 (0.113)***
0.0001(4.42e-06)***
-4.071 (0.009)***
-1.581 (0.099)***
1.551(0.0096)***
-0.001 (0.00003)***
0.80 (0.02)***
12875.42

Log-likelihood
atthe0.01level.
***Significant

significantand negative; the quick and coverage ratios are significantand positive. This result
impliesthatfor privatefirms,high levels of indebtednessandlow liquidityincreasethe probability
of facing credit constraints.But for public firms, a high level of debt increasesthe probabilityof
being in the unconstrainedregime. For these firms,having obtaineddebt in the past is perceived
as a good signal. Overall, my results confirmthat indebtednessplays a differentrole for the two
types of firms.
Using the likelihood ratio test, the null hypothesis of linearityis rejectedat any conventional
level of significance. The nonlinearswitching regression model fits the data much better than

This content downloaded from 111.68.99.250 on Mon, 12 Jan 2015 00:31:15 AM

All use subject to JSTOR Terms and Conditions

Financeand the Use of TradeCredit

Atanasova* Institutional

61

does the linear model. This is the case when I estimate the model with the whole sample, and
when I estimate it with the two subsamples of small and large firms. The switching model
indicatesthatwhen facing constraintson institutionalfinance,firmsuse tradecreditas a substitute
to mitigate these constraints.The overall results supportthe importantrole of credit market
imperfectionsin corporatefinancingbehavior.

V.FurtherInvestigation
The major caveats of our estimation procedure concern the use of industry dummies to
control for firm-specific fixed effects, and that some of the determinants of the demand
for trade credit are endogenously determined. For instance, since the variable institutional
loans is a choice variable for the firm, it is correlated with the innovation term of the trade
credit equation. Here, I evaluate the size and significance of the possible bias due to these
caveats.
I split the sample into firm-yearsfor which Z ', < 0 (unconstrained)and firm-yearsfor which
Z',, 2 0 (constrained).I estimate a random-effectunbalancedpanel model for each subsample,
using a 2SLS IV procedure,in which the instrumentsarethe exogenous variables(industry-wide
averages)and the lagged value of the endogenousregressors(observedfirm-specificvariables).
Table IV reportsthe estimationresults. There is no materialdifference between the results in
TableIV andthose in TableIII, althoughsome of the mainresultsappearto be even stronger.This
finding implies that the endogeneity of short-termloans and the inclusion of industrydummies
does not generatesubstantialbias in my originalresults.
Althoughthisprocedureis a robustnesscheckon my mainresults,it is not completelysatisfactory.
To providefurtherresultsfor the effect of the unobservablefirm-specificheterogeneity,I demean
the data by using the within transformation,and re-estimatethe switching model. The withintransformationremovesthe statusdummyPD and reducesthe variationof the industryaverages.
TableV reportsthe results, which are clearly similarto the estimatedparametersin Table III.
The unobservedfirm specific effects do not appearto have a significanteffect on the parameter
estimates.This evidence suggests that in contrastto the case in which firm effects are important
and so larger firms are generally less constrained,as firms grow larger over time, access to
institutionalfinance for them becomes easier.

VI.CreditConstraintsand the Business Cycle

In this section, I provideevidence based on a balancedsubsampleof firmswhere the data are
availablecontinuouslyfrom 1983 to 1999. I choose a balancedsamplebecausemy objectivehere
is to assess how macroeconomic conditions, and monetary policy in particular,affect the
probability(acrossfirms)thatfirmsface creditconstraints.Choosinga balancedsampleeliminates
the possibility of the average probability changing over time because of a changing sample
composition.I choose the lengthof the new panel to strikea balancebetweenhavinga periodthat
is long enoughto coversubstantialbusinesscycle changes,butnot so long as to reduceexcessively
the numberof firms in the sample, and thereforethe cross-sectional variationin the data. The
final sample comprises676 firms. I estimate the switching regressionmodel using the balanced
sample. The estimationresults are not materiallydifferent.3
3The estimationresultsare availableon request.

This content downloaded from 111.68.99.250 on Mon, 12 Jan 2015 00:31:15 AM

All use subject to JSTOR Terms and Conditions

FinancialManagement' Spring2007

62

Table IV:2SLS Random-Effects IVEstimation

Thetableshowstheinstrumental
variablesregressionresults.I splitthesampleonthebasisof theestimated
Theregressionequationsare:
switchingfunctionandperformpaneldataanalysisforeachsubsample.
AgP

TA, ,

X
',,fu

Ls
+ Su TA +

TA,

(,

if Z ', J < 0.

(T.4.1)

A,

The samplecontains2,000 UK firmsfrom 1981 to 2000. The instruments

arethe exogenousvariables
(industry-wide
averages)and the lagged value of the endogenousregressors(observedfirm-specific
and within firm clusteringare reportedin
variables).Standarderrorsrobustto heteroskedasticity
parentheses.
Constrained firms
L/TA
X variables
CF/TA
SALES/TA
AR/TA
INV/TA
CASH/TA
Industryaverages
SD(TURN)
AP/TA

Unconstrained firms

-0.063 (0.0023)***

-0.0536 (0.112)

-0.0005 (0.0025)
0.0058(0.00051)***
0.1094(0.0047)***
0.03437(0.0053)***
0.0221(0.0061)***

-0.0523 (0.01745)***
0.0023(0.0056)
0.3059(0.0354)***
0.03773(0.0113)***
0.07222(0.0217)***

8.89e-06(5.97e-06)
0.602(0.046)***

2.79e-06(0.00001)
0.393(0.106)***

atthe0.01level.
***Significant
In the switching regression model, through both the balance sheet variables and the year
dummies included in the switching function, general financial macroeconomic conditions
will affect the probability of being in one or the other regime. For example, deteriorating
macroeconomic conditions can be reflected in the decreasing collateral value of the firm's
assets. Similarly, less cash flow will be available to cover interestpayments.Includingthe year
dummies in the switching function captureschanges in general macroeconomicconditionsthat
affect all firms in the same way, and that are not accounted for by changes in firm-specific
variables. In years characterizedby recession or contractionarymonetary policy, the theory
predicts a largervalue of the coefficient of the year dummy,since firms are more likely to face
financial constraints.
Table VI reports the coefficients of the year dummies in the switching function of the
estimated balanced switching model. The table also presents, for each year, the probabilityof
being in the constrained regime, which I obtain by averaging individual probabilities across
firms in that year. I interpretthis average probabilityas a more complete summaryindicatorof
the effects of macro shocks, since it incorporatesboth the effect of year dummies and changes
in firm-specific variables. The results in Table VI show that the year-dummycoefficients and
the average probability of facing financial constraints are the highest in the recession years
1990, 1991, and 1992.
I investigate whether the time variationin the coefficients of the year dummies capturesany
changes in the stance of monetarypolicy. To addressthis question, I use the Bank of England

This content downloaded from 111.68.99.250 on Mon, 12 Jan 2015 00:31:15 AM

All use subject to JSTOR Terms and Conditions

Financeand the Use of TradeCredit

Atanasova* Institutional

63

TableV: Estimation of the Switching Model with Firm Specific Effects

Thetableshowsthe switchingregressionresultswithfirmspecificeffects.I demeanthe observations
for
each variablein Z (zi, - 2i.) andX (x, - xi.) andre-estimatethe switchingmodel. The within-transformation

removesthe statusdummyPD andreducesthe variationof the industryaverages.The samplecontains

errorsrobustto heteroskedasticity
andwithinfirmclustering
2,000UKfirmsfrom1981to 2000. Standard
in parentheses.
arereported
Switching Model
C regime

AP/TA
L/TA
X variables
CF/TA
SALES/TA
AR/TA
INV/TA
CASH/TA
Industryaverages

UC regime

-0.1609 (0.038)***

-0.0158 (0.0389)

-0.04379 (0.0057)***
0.01696(0.00597)***

-0.0521 (0.0058)***
0.00172(0.0006)***

0.16272 (0.00581)***
0.0682 (0.00655)***
0.1883 (0.00713)***

0.16057 (0.00593)***
0.067267 (0.0067)***
0.0189 (0.00735)***

SD(TURN)

1.07e-06(3.15e-06)

1.19e-06(3.19e-06)

AP/TA

0.0065 (0.0056)

0.00787 (0.0058)

Z variables

Switch Function
-0.0117 (0.0002)***
-0.2405 (0.01635)***
-4.1845 (0.016)***
-0.09596 (0.0021)***
-9.4926 (0.1102)***
0.00019 (0.00004)***
-3.8241 (0.0534)***
1.8109 (0.01643)***
-0.0006 (0.00004)***
0.00052 (0.00003)***

Age
PROF/TA
QR
logTA
DEPR/TA
LR
CR
PD*QR
PD*LR
PD*CR
*** Significant
atthe0.01level.

base rateas an indicatorof how restrictivemonetarypolicy is (see Bernankeand Blinder, 1992).

I note that the resultsare not materiallydifferentif I use the spreadbetween the T-bill and base
rateto measurethe stance of monetarypolicy.
I includethe averagevalueof the monthlylevel of the base rateforthe previousyearas a regressor
in the switchingfunctioninsteadof the yeardummies.TableVII reportsthe estimationresults.
The table indicatesthatthe coefficientof the lagged base rateis positive andhighly significant.
Keeping in mind that the yearly nature of the data does not allow me to capturea complete
picture of the dynamic effects of changes in base rate in an adequate manner, the results
neverthelesssuggestthattightmonetarypolicy increasesthe probabilitythatfirmsfindthemselves
in the constrainedregime.
Next, I analyze the time patternof the averageprobabilityof facing financialconstraintsand
examine the implicationof the results for the existence of a credit channel of monetarypolicy
transmission.Figure 1 shows that there is a clear comovement between the average probability
of being in the constrainedregime and the base rate, with the formerfollowing the latterwith a
lag. This finding suggests that restrictivemonetarypolicy may work throughthe effect it has on
firms' access to institutionalfinance.

This content downloaded from 111.68.99.250 on Mon, 12 Jan 2015 00:31:15 AM

All use subject to JSTOR Terms and Conditions

FinancialManagement* Spring2007

64

Table VI:Time Profile of Coefficients on Year Dummies and Average Probability

of Facing Financial Constraints
Column(2) reportsthecoefficientsof theyeardummiesin theswitchingfunctionof theestimatedbalanced
of beingin the constrained
switchingmodel.Column(3) presents,for eachyear,the probability
regime,
whichI obtainby averagingindividualprobabilities
acrossfirmsin thatyear.Thesamplecontains676 UK
firmsfrom1983to 1999.Robustbootstrapped
standard
errorsarereported
in parentheses.
Year

Year Dummy
-1.01 (0.03)***
-1.04(0.03)***
-1.05 (0.03)***
-1.05 (0.03)***
-1.06(0.04)***
-1.04(0.04)***
-0.09 (0.02)***
-0.10 (0.02)***
-0.12(0.02)***
-0.96 (0.02)***
-1.02 (0.02)***
-1.02(0.02)***
-1.02 (1.66)
-1.01 (0.17)***
-1.00(0.02)***
-1.01 (1.07)

1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999

Average Probability
0.397886
0.398021
0.400711
0.399459
0.395344
0.396031
0.400324
0.405676
0.400720
0.399459
0.395874
0.391206
0.392374
0.391074
0.391704
0.392451

*** Significantat the 0.01 level.

Overall,my evidence providessupportfor the existence of a creditchannelof monetarypolicy

transmission.

VII.Conclusion
In this paper I analyze quantitativelythe effects of credit marketimperfectionson corporate
short-termfinancing behavior.The asymmetric informationtheories predict that these effects
depend on various firm characteristics,the overall macroeconomicconditions,and the stance of
monetarypolicy. Firmswith weak balancesheet positions and for which asymmetricinformation
problemsare more severe will rely on tradecreditas a source of funds,despite tradecreditbeing
an unattractivesubstituteto conventionalinstitutionalloans.
To test these predictions, I develop a switching model in which I assume that firms operate
in eitheran unconstrainedor a constrainedregime. I then determinethe probabilityendogenously
by a switching function. The endogenous determinationof each regime attemptsto overcome
misspecification problems that arise when I use an a priori criterion to classify firms into
constrainedand unconstrainedregimes.
My empiricalformulationallows me to controland test for the multiplicityof factorsthat may
affect firms' access to institutionalcredit. It also recognizes that the significanceof such factors
may vary over the business cycle.
I estimate the model using panel data for UK firms. The empiricalevidence strongly suggests
that trade credit is an importantsource of funds for constrainedfirms, but that unconstrained

This content downloaded from 111.68.99.250 on Mon, 12 Jan 2015 00:31:15 AM

All use subject to JSTOR Terms and Conditions

Atanasova* Institutional
Financeand the Use of TradeCredit

65

TableVII:Estimation of the Switching Model with Monetary Policy Variable

The table shows the switchingregressionresults.The sample contains 2,000 UK firms from 1981 to 2000.
I include the average value of the monthly level of the base rate for the previous year as a regressorin the
switchingfunctioninsteadof the year dummies.BR representsthe level of lag base rate.Robustbootstrapped
standarderrorsare reportedin parentheses.

Switching Model
C regime
L/TA
X variables
CF/TA
SALES/TA
AR/TA
INV/TA
CASH/TA
Industryaverages
SD(TURN)

AP/TA
Z variables

UC regime

-0.1257 (0.0114)***

-0.0157 (0.034)

-0.0374 (0.028)
0.0188 (0.003)***
0.496 (0.012)***
0.3326 (0.012)***
0.283 (0.019)***

-0.046 (0.008)***
0.0025 (0.0069)***
0.187 (.0066)***
0.098 (0.005)***
0.0003 (0.006)

0.0000231 (0.00001)***

0.000058 (7.33e-06)***

0.05299(0.021)***

0.0542(0.0101)***
Switch Function
-0.0121 (0.0002)***
-0.431 (0.022)***
-2.774 (0.0036)***
-0.091 (0.002)***
-7.781 (0.113)***
0.0001(4.44e-06)***
-4.031 (0.009)***
-1.68 (0.0992)***
1.538(0.0096)***
-0.001 (0.00003)***
2.77 (0.02)***
0.0169(0.002)***

Age
PROF/TA
QR
logTA
DEPR/TA
LR
CR
PD
PD*QR
PD*LR
PD*CR
BR
at the0.01level.
*** Significant

firmsavoid using it. Such evidence suggests a stronginfluenceof creditmarketimperfectionson

corporatefinancingbehavior.
I also show that low levels of liquidity,profitability,and values of financialratios increasethe
probabilityof facing creditconstraints.I also find some evidence that high leverageis associated
with greaterrelianceon tradecredit for privatefirms, but not for public firms. I also find thatthe
likelihood of being in the constrained regime varies over the business cycle with general
macroeconomicconditionsand the stance of monetarypolicy.
My study extends previous research on trade credit by considering explicitly the effects of
business-cycle dynamics on firms' financing decisions. The credit channel of monetarypolicy
transmissionpredicts that when monetarypolicy tightens, the reduction of institutionalloans
causes some firmsto cut spendingindependentlyof changes in the cost of capital.I demonstrate
that credit-constrainedfirms increase their use of trade credit, which is an unattractiveand
expensive substitutefor bank credit.This identificationfrom the supply side providessupportfor
a potent channelof monetarytransmission.E

This content downloaded from 111.68.99.250 on Mon, 12 Jan 2015 00:31:15 AM

All use subject to JSTOR Terms and Conditions

66

FinancialManagement*Spring2007

1.Average
ofConstrained
andtheBaseRate
Figure
Probability
Regime
The figure provides informationabout the comovement between the average probabilityof being in the
constrainedregime andthe Bankof Englandbase rate.The left axis of the plot shows the averageprobability
scale. The right axis shows the base rate scale.
Base rate

Averageprobability
CD

CD
S

liI

I!

Average probability
2---- Base rate

~.
o
cO
o

6
cC

o0
0
00

6
c3)
o,

CI)
0)

CD

0
c\J

0
Co

1985

1987

1989

1991

1993

1995

1997

1999

References
Alphonse, P., J. Ducret, and E. Severin, 2004, "When Trade Credit FacilitatesAccess to Bank Finance:
Evidence from US Small Business Data,"University of Lille WorkingPaper.
Antov, D., 2005, "TradeCreditand InstitutionalFinancing:Theoryand Evidence,"NorthwesternUniversity
WorkingPaper.
Bernanke,B. andA. Blinder, 1992, "TheFederalFundsRate and the Channelof MonetaryTransmission,"
AmericanEconomic Review 82, 901-921.
Biais, B. andC. Gollier, 1997, "TradeCreditandCreditRationing,"Reviewof FinancialStudies10, 903-937.
Biais, B., P. Hillion, and J. Malecot, 1995, "La StructureFinanciere des Entreprises:Une Investigation
Empiriquesur Donnees Francaises,"Economie et Prevision 120, 15-28.
Davis. E and K. Yeoman, 1974, "CompanyFinance and the CapitalMarket:A Study of the Effects of Firm
Size," University of CambridgeOccasionalPaper.
Deloof, M. and M. Jegers, 1999, "TradeCredit,CorporateGroups,and the Financing of Belgian Firms,"
Journal of Business Finance andAccounting26, 945-966.

This content downloaded from 111.68.99.250 on Mon, 12 Jan 2015 00:31:15 AM

All use subject to JSTOR Terms and Conditions

Financeand the Use of TradeCredit

Atanasova* Institutional

67

Dempster,A., N. Laird, and D. Rubin, 1977, "MaximumLikelihood from Incomplete Data via the EM
Algorithm,"Journal of the Royal StatisticalSociety, Series B 39, 1-38.
Douglas, S., 1996, "BootstrapConfidence Intervalsin a Switching Regression Model," Economic Letters
53, 7-15.
Elliehausen, G. and J. Wolken, 1993, "An Empirical Investigation into Motives for Demand for Trade
Credit,"FederalReserveBoard Staff Study.
Ferris,S., 1981, "A TransactionTheoryof TradeCreditUse," QuarterlyJournalof Economics96, 243-270.
FranksJ., J. Broyles, andW. Carlton, 1985, CorporateFinance: ConceptsandApplications, Boston, MA,
Kent PublishingCompany.
Habib,M. and D. Johnson,1999, "TheFinancingandRedeploymentof SpecificAssets,"Journalof Finance
54, 693-720.
Jain,N., 2001, "MonitoringCostsandTradeCredit,"QuarterlyReviewof EconomicsandFinance41, 89-110.
Laffer,A., 1970, "TradeCreditand the Money Market,"Journal ofPolitical Economy78, 239-267.
Long, M., I. Malitz, and S. Ravid, 1993, "TradeCredit, Quality Guaranteesand ProductMarketability,"
Financial Management22, 117-127.
Myers S. and N. Majluf, 1984, "CorporateFinancing and Investment Decisions When Firms Have
InformationThat InvestorsDo Not Have,"Journal ofFinancial Economics 13, 187-221.
Ng C., J. Smith,and R. Smith, 1999, "Evidenceon the Determinantsof TradeCreditTermsUsed in Interfirm
Credit,"Journal of Finance 54, 1109-1130.
Nilsen, J., 2002, "TradeCredit and the Bank Lending Channel,"Journal of Money, Creditand Banking
34, 226-253.
Petersen, M. and R. Rajan, 1994, "The Benefit of Lending Relationships:Evidence from Small Business
Data,"Journal ofFinance 49, 3-36.
Petersen,M. andR. Rajan,1995, "TheEffectof CreditMarketCompetitionon Firms-CreditorRelationships,"
QuarterlyJournal of Economics 110, 407-443.
Petersen,M. and R. Rajan, 1997, "TradeCredit:Theories and Evidence,"Review of Financial Studies 10,
661-91.
Schwartz,R., 1974, "AnEconomic Model of TradeCredit,"Journal ofFinancial and QuantitativeAnalysis
9, 643-657.
Smith,J., 1987, "TradeCreditand InformationAsymmetry,"Journal ofFinance 42, 863-873.
Wilner, B., 2000, "The Exploitation of Relationships in Financial Distress: The Case of Trade Credit,"
Journal ofFinance 55, 153-178.

This content downloaded from 111.68.99.250 on Mon, 12 Jan 2015 00:31:15 AM

All use subject to JSTOR Terms and Conditions

68

FinancialManagement* Spring2007

This content downloaded from 111.68.99.250 on Mon, 12 Jan 2015 00:31:15 AM

All use subject to JSTOR Terms and Conditions