Download here. - DePaul University

Bank Credit, Trade Credit or No Credit?
Evidence from the
Surveys of Small Business Finances
Rebel A. Cole
DePaul University
2011 Annual Meetings of the
Southern Finance Association
November 18, 2011
Key West, FL
Research Summary




In this study, we use data from the SSBFs to
provide new information about the use of credit
by small businesses in the U.S.
More specifically, we first analyze firms that do
and do not use credit; and then analyze why
some firms use trade credit while others use
bank credit.
We find that one in five small firms uses no
credit, one in five uses trade credit only, one in
five uses bank credit only, and two in five use
both bank credit and trade credit.
These results are consistent across the three
SSBFs we examine—1993, 1998 and 2003.
Research Summary



When compared to firms that use credit, we find
that firms using no credit are significantly
smaller, more profitable, more liquid and of
better credit quality; but hold fewer tangible
assets.
We also find that firms using no credit are more
likely to be found in the services industries and in
the wholesale and retail-trade industries.
In general, these findings are consistent with the
pecking-order theory of firm capital structure.
Research Summary



Firms that use trade credit are larger, more
liquid, of worse credit quality, and less likely to
be a firm that primarily provides services.
Among firms that use trade credit, the amount
used as a percentage of assets is positively
related to liquidity and negatively related to
credit quality and is lower at firms that primarily
provide services.
In general, these results are consistent with the
financing-advantage theory of trade credit.
Research Summary



Firms that use bank credit are larger, less
profitable, less liquid and more opaque as
measured by firm age, i.e., younger.
Among firms that use bank credit, the
amount used as a percentage of assets is
positively related to firm liquidity and to
firm opacity as measured by firm age.
Again, these results are generally
consistent with the pecking-order theory
of capital structure, but with some notable
exceptions.
Research Summary




We contribute to the literature on the availability
of credit in at least two important ways.
First, we provide the first rigorous analysis of the
differences between small U.S. firms that do and
do not use credit.
Second, for those small U.S. firms that do
participate in the credit markets, we provide new
evidence regarding factors that determine their
use of trade credit and of bank credit, and
whether these two types of credit are substitutes
(Meltzer, 1960) or complements (Burkart and
Ellingsen, 2004).
Our evidence strongly suggests that they are
complements.
Introduction



Among small businesses, who uses credit?
Among those that use credit, from where do they
obtain funding—from their suppliers, i.e., trade
credit, from their financial institutions, i.e., bank
credit, or from both?
The answers to these questions are of great
importance not only to the small firms
themselves, but also to prospective lenders to
these firms and to policymakers interested in the
financial health of these firms.
Look no farther than the Obama “Stimulus”
package--$30 billion targeted at small firms.
Introduction



The availability of credit is one of the most
fundamental issues facing a small business and
therefore, has received much attention in the
academic literature (See, for example, Petersen
and Rajan, 1994, 1997; Berger and Udell, 1995,
2006; Cole, 1998; Cole, Goldberg and White,
2004; and Cole 2008, 2009).
However, many small firms—as many as one in
four, according to data from the 2003 Survey of
Small Business Finances—indicate that they do
not use any credit whatsoever. We refer to these
firms as “non-borrowers.”
These firms have received virtually no attention
from academic researchers.
Introduction



In this study, we first analyze firms that do and do
not use credit, i.e., leveraged and unleveraged firms;
We then analyze how firms that do use credit
(leveraged firms) allocate their liabilities between
bank credit (obtained from financial institutions) and
trade credit (obtained from suppliers), in order to
shed new light upon these critically important issues.
We utilize data from the FRB’s 1993, 1998 and 2003
SSBFs to estimate a Heckman selection model, where
the manager of a firm first decides if it needs credit,
and then decides from where to obtain this credit—
from financial institutions (in the form of bank credit)
or from suppliers (in the form of trade credit).
Trade Credit

Petersen and Rajan (1997) list and
summarize three broad groupings of
theories of trade credit:
• financing advantage,
• price discrimination, and
• transaction costs.
Trade Credit: Financing Advantage



According to the financing-advantage theory, a
supplier of trade credit has an informational
advantage over a bank lender in assessing and
monitoring the creditworthiness of its customers,
which, in turn, gives the supplier a cost
advantage in lending to its customers.
The supplier also has a cost advantage in
repossessing and reselling assets of its customers
in the event of default (Mian and Smith, 1992).
Smith (1987) argues that, by delaying payment
via trade credit, customers can verify the quality
of the supplier’s product before paying for that
product.
Trade Credit: Price Discrimination




According to the price-discrimination theory, which dates
back to Meltzer (1960), a supplier uses trade credit to price
discriminate among its customers.
Creditworthy customers will pay promptly so as to get any
available discounts while risky customers will find the price
of trade credit to be attractive relative to other options.
The supplier also discriminates in favor of the risky firm
because the supplier holds an implicit equity stake in the
customer and wants to protect that equity position by
extending temporary short-term financing.
Meltzer (1960) concludes that trade creditors redistribute
traditional bank credit during periods of tight money, so
that trade credit serves as a substitute for bank credit when
money is tight.
Trade Credit: Transactions Cost



According to the transactions-cost theory, which
dates back to Ferris (1981), trade credit reduces
the costs of paying a supplier for multiple
deliveries by cumulating the financial obligations
from these deliveries into a single monthly or
quarterly payment.
By separating the payment from the delivery, this
arrangement enables the firm to separate the
uncertain delivery schedule from what can now
be a more predictable payment cycle.
This enables the firm to manage its inventory
more efficiently.
Trade Credit: Financing Advantage



Cuñat (2007) argues that trade creditors have an
advantage over bank creditors in collecting noncollateralized lending, in that a trade creditor can
threaten to cut off goods that it supplies to the
borrower so long as switching suppliers is costly.
This advantage enables trade creditors to lend
more than banks are willing to lend.
In this sense, trade credit is a complement rather
than a substitute for bank credit, and firms
should be expected to utilize both types of credit,
even when banking markets are competitive.
Data

We extract data from the FRB’s SSBFs:
• Four surveys:


Cross sections as of 1988, 1993, 1998, 2003
Broadly representative of 5 million privately held
firms with fewer than 500 employees.
• Stratified random samples


Oversample large and minority-owned firms.
Also stratify by census region
• Cannot use results from unweighted
descriptive statistics or from OLS to make
inferences about the population
Dependent Variables




Each SSBF includes a question asking whether or
not the firm used trade credit in the reference
year of the survey, and asking whether or not the
firm had any outstanding bank credit in the
reference year of the survey.
We use the answers to these questions to classify
a firm as using no credit, using trade credit only,
using bank credit only, or using both bank credit
and trade credit.
We calculate the amount of bank credit as the
sum of reported outstanding loans.
We calculate the amount of trade credit as the
value of accounts payable on the firm’s balance
sheet.
Explanatory Variables:


To select our explanatory variables, we rely
primarily upon the existing literature on the
availability of credit, as this may be the most
vexing issue facing small firms. These variables
are motivated by and used in Cole (1998), Cole,
Goldberg and White (2004), and/or Cole (2009).
We group these variables into four vectors:
• Firm Characteristics
• Market Characteristics
• Owner Characteristics
• Firm-Creditor Relationship Characteristics
Explanatory Variables:
Firm Characteristics





Size (Sales, Assets, Employment)
Age
Organizational form (C-Corp, S-Corp,
Partnership, Proprietorship)
Creditworthiness (Firm Delinquent
Obligations, Firm Bankruptcy, Firm
Judgments, D&B Credit Score, Paid Late
on Trade Credit)
Financial performance and condition
(profitability, leverage, liquidity)
Explanatory Variables:
Market Characteristics

Very limited information on firm location
because of confidentiality concerns. Can
only use what is available from the SSBFs.
• Banking concentration

Categorical representation with three levels,
which we convert into dummy variables for
low, medium and high concentration
• Urban/Rural Location of the Firm

Binary indicator variable
Explanatory Variables:
Owner Characteristics






Age
Experience
Education (Grad, College, Some College,
High School)
Personal Wealth
Personal Creditworthiness (Delinquent
Obligations, Judgments, Bankruptcy)
Race and Ethnicity (Asian, Black, Hispanic)
Explanatory Variables:
Firm-Creditor Relationship Characteristics




Type of Primary Financial Institution
(Commercial Bank, Savings Association,
Finance Company, or “Other”)
Distance from firm HQ to Primary FI.
Length of Relationship with Primary FI.
Total Number of FIs (also split by number
of Commercial Banks and number of NonBank FIs.
Results
No Credit
Observations (unweighted)
Observations (weighted, millions)
Percentage (weighted)
Trade Credit Only
Observations (unweighted)
Observations (weighted, millions)
Percentage (weighted)
Bank Credit Only
Observations (unweighted)
Observations (weighted, millions)
Percentage (weighted)
Trade and Bank Credit
Observations (unweighted)
Observations (weighted, millions)
Percentage (weighted)
Total
Observations (unweighted)
Observations (weighted, millions)
Percentage (weighted)
1993
1998
2003
656
0.911
18.6%
622
1.087
21.5%
614
1.239
21.1%
828
1.116
22.8%
662
1.171
23.1%
619
1.074
18.3%
761
0.866
17.7%
506
0.811
16.0%
570
1.083
18.5%
1,917
1.991
40.8%
1,395
1.993
39.4%
1,765
2.466
42.1%
4,162
4.885
100.0%
3,185
5.062
100.0%
3,568
5.862
100.0%
Results:
Firm Uses Credit
Variable
log of Sales
ROA
Cash
Tangible Assets
Sales Growth is Positive
Corporation
log of Firm Age
Firm Delinquency
MSA
1993
Marginal
Effect t-stat
0.048
-0.016
-0.153
-0.062
-0.003
-0.010
-0.010
0.057
-0.041
13.56
-4.57
-7.14
-3.54
-0.27
-0.82
-1.02
3.31
-3.20
1998
Marginal
Effect t-stat
a 0.041 12.76
a -0.025 -5.41
a -0.212 -9.22
a -0.072 -3.60
0.007 0.58
0.042 2.81
-0.014 -1.34
a 0.135 4.86
a -0.018 -1.14
2003
Marginal
Effect t-stat
a 0.032 12.41
a -0.030 -4.64
a -0.219 -11.54
a -0.050 -2.79
0.024 2.20
a 0.029 2.39
-0.012 -1.60
a 0.008 2.05
-0.022 -1.62
Firms that use credit are larger, less profitable, less liquid, have
fewer tangible assets and are less creditworthy.
a
a
a
a
b
b
b
Results:
Firm Uses Credit
Variable
Primary Mfg
Secondary Mfg
Transportation
Wholesale
Retail
Finance/Real Estate
Business Services
Professional Services
1993
Marginal
Effect t-stat
0.155
0.013
-0.095
-0.064
-0.082
-0.118
-0.095
-0.098
2.70
0.34
-2.89
-2.52
-4.22
-5.15
-5.23
-4.69
1998
Marginal
Effect t-stat
a -0.138
-0.102
a -0.256
b -0.176
a -0.196
a -0.216
a -0.150
a -0.219
-3.19
-2.35
-6.49
-4.67
-6.60
-6.44
-5.33
-7.27
2003
Marginal
Effect t-stat
a
b
a
a
a
a
a
a
-0.008
-0.017
0.001
-0.070
-0.087
-0.085
-0.078
-0.099
-0.19
-0.44
0.02
-2.18
-3.69
-3.09
-3.55
-4.40
b
a
a
a
a
Firms that use credit are less likely to be in wholesale trade, retail
trade, finance/real estate, business services and professional
services.
Results:
Firm Uses Credit
Variable
1993
Marginal
Effect t-stat
log of Owner Age
-0.112
log of Owner Experience
0.013
Owner is Female
0.011
Owner is Asian
-0.044
Owner is Black
-0.016
Owner is Hispanic
-0.001
Owner has College Degree 0.026
Owner has Graduate Degree -0.004
Owner Percentage Ownership
-0.0006
Owner Delinquency
0.033
-4.06
1.13
0.90
-1.86
-0.57
-0.04
2.06
-0.26
-2.44
1.89
1998
Marginal
Effect t-stat
a
c
b
b
c
-0.075
0.042
-0.007
-0.022
-0.024
-0.057
0.004
0.011
-0.0004
-0.023
-2.34
3.46
-0.52
-0.79
-0.87
-2.46
0.27
0.58
-1.36
-1.07
2003
Marginal
Effect t-stat
b
a
-0.075
0.044
-0.037
-0.047
-0.030
b -0.025
-0.022
0.004
-0.0005
0.030
-2.70
4.34
-3.32
-1.95
-1.27
-1.04
-1.74
0.24
-2.03
1.78
a
a
a
c
c
b
c
Owners of firms that use credit are younger but more experienced.
Results:
Firm Uses Trade Credit
Variable
Intercept
log of Sales
ROA
Cash
Current Assets
Sales Growth is Positive
Corporation
log of Firm Age
Firm Delinquency
MSA
1993
Marginal
Effect t-stat
0.48
0.008 1.71
-0.001 -0.17
0.126 3.55
0.160 5.98
-0.023 -1.55
-0.043 -2.65
-0.004 -0.29
0.099 4.63
-0.003 -0.21
c
a
a
a
a
1998
Marginal
Effect t-stat
-0.99
0.012 3.09
-0.001 -0.08
0.143 4.25
0.162 5.97
0.004 0.26
0.062 3.50
-0.007 -0.52
0.136 4.92
0.014 0.78
a
a
a
a
a
2003
Marginal
Effect t-stat
-0.58
0.013 3.62
-0.014 -1.50
0.180 6.13
0.075 3.11
0.035 2.38
0.072 4.64
0.032 3.17
0.007 1.34
-0.035 -1.97
Firms that use trade credit are larger, more liquid, more likely to
offer trade credit, more likely to be corporations and less
creditworthy.
a
a
a
b
a
a
b
Results:
Firm Uses Trade Credit
Variable
Primary Mfg
Secondary Mfg
Transportation
Wholesale
Retail
Finance/Real Estate
Business Services
Professional Services
1993
Marginal
Effect t-stat
-0.005
0.003
-0.166
-0.023
-0.052
-0.205
-0.071
-0.154
-0.13
0.07
-4.17
-0.68
-2.07
-6.79
-2.93
-5.62
1998
Marginal
Effect t-stat
a
b
a
a
a
0.024
0.085
-0.191
-0.101
-0.071
-0.258
-0.070
-0.131
0.45
1.50
-4.58
-2.68
-2.40
-7.54
-2.60
-4.27
2003
Marginal
Effect t-stat
a
a
b
a
a
a
-0.132
-0.087
-0.217
-0.138
-0.114
-0.264
-0.171
-0.239
-2.77
-1.88
-5.57
-3.29
-3.62
-7.38
-5.90
-7.58
Firms that use trade credit are less likely to be in transportation,
retail trade, finance/real estate and business and professional
services.
a
c
a
a
a
a
a
a
Results:
Firm Uses Trade Credit
Variable
1993
Marginal
Effect t-stat
1998
Marginal
Effect t-stat
2003
Marginal
Effect t-stat
log of Owner Age
0.011 0.29
0.025 0.60
0.012
log of Owner Experience
0.027 1.67 c 0.023 1.47
0.007
Owner is Female
0.005 0.26
-0.017 -0.98
-0.004
Owner is Asian
0.001 0.02
-0.014 -0.37
-0.004
Owner is Black
-0.011 -0.28
-0.028 -0.75
-0.110
Owner is Hispanic
-0.084 -2.80 a -0.088 -2.91 a -0.050
Owner has College Degree
0.024 1.43
0.023 1.23
0.027
Owner has Graduate Degree 0.037 1.76 c 0.020 0.85
0.051
Owner Pctg. Ownership
-0.0009 -2.98 a 0.0003 0.93
-0.0001
Owner Delinquency
0.006 0.28
-0.072 -2.87 a
0.012
0.30
0.50
-0.25
-0.12
-3.26 a
-1.58
1.59
2.39 b
-0.41
0.55
There are no consistently significant differences in the owners of
firms that do and do not use trade credit.
Results:
Firm Uses Bank Credit
Variable
log of Sales
ROA
Cash
Tangible Assets
Sales Growth is Positive
Corporation
log of Firm Age
Firm Delinquency
MSA
1993
Marginal
Effect
t-stat
0.053 10.03 a
-0.031 -5.70 a
-0.366 -10.42 a
0.029
1.19
0.074
4.97 a
0.013
0.79
-0.031 -2.31 b
0.062
3.10 a
-0.029 -1.69 c
1998
Marginal
Effect
t-stat
0.038
7.92 a
-0.029 -4.00 a
-0.383 -10.42 a
-0.067 -2.45 b
0.020
1.18
-0.022 -1.15
-0.016 -1.10
0.006
0.23
-0.049 -2.30 b
2003
Marginal
Effect
t-stat
0.027
7.14 a
-0.037 -3.77 a
-0.354 -11.32 a
-0.021 -0.81
0.024
1.55
0.001
0.08
-0.062 -5.28 a
-0.005 -0.90
0.026
1.50
Firms that use bank credit are larger, less profitable, younger and
have less financial slack as proxied by cash.
Results:
Firm Uses Bank Credit
Variable
Primary Mfg
Secondary Mfg
Transportation
Wholesale
Retail
Finance/Real Estate
Business Services
Professional Services
1993
Marginal
Effect
t-stat
-0.024
-0.045
0.133
-0.056
-0.033
0.080
-0.008
0.065
-0.67
-1.28
2.45
-1.93
-1.44
2.33
-0.36
2.37
1998
Marginal
Effect
t-stat
b
c
b
b
-0.109
-0.102
0.094
-0.055
-0.105
0.182
-0.060
-0.025
-2.38
-2.44
1.51
-1.38
-3.49
3.59
-2.13
-0.76
2003
Marginal
Effect
t-stat
b
b
a
a
b
0.043
-0.049
0.119
-0.078
-0.084
0.101
0.007
0.055
0.90
-1.26
2.37
-2.23
-3.20
2.67
0.29
1.87
Firms that use bank credit are more likely to be in transportation
and finance/real estate, and less likely to be in wholesale trade,
retail trade.
b
b
a
a
c
Results:
Firm Uses Bank Credit
Variable
log of Owner Age
log of Owner Experience
Owner is Female
Owner is Asian
Owner is Black
Owner is Hispanic
Owner has College Degree
Owner has Graduate Degree
Owner Pctg. Ownership
Owner Delinquency
1993
Marginal
Effect
t-stat
-0.155
0.005
-0.016
-0.058
0.040
0.007
-0.034
-0.050
0.0012
-0.064
-3.97
0.28
-0.94
-1.65
0.93
0.20
-2.05
-2.41
4.14
-2.96
1998
Marginal
Effect
t-stat
a
b
b
a
a
-0.080
-0.012
-0.037
-0.069
0.028
0.094
0.003
-0.013
-0.0008
0.022
-1.68
-0.69
-1.88
-1.76
0.64
2.34
0.16
-0.49
-1.96
0.77
2003
Marginal
Effect
t-stat
c
c
c
b
c
-0.052
0.001
-0.032
-0.046
-0.050
0.004
-0.063
-0.045
0.0003
-0.041
-1.22
0.06
-1.86
-1.31
-1.30
0.10
-3.68
-2.04
0.91
-1.82
Owners of firms that use bank credit are younger, less educated and
less likely to be female or Asian.
c
a
b
c
Results:
Amount of Trade Credit




Amount of trade credit decreases with firm age
and increases with amount of financial slack and
the amount of current assets.
It is greater when firm credit quality is worse and
when owner credit quality is worse.
It is less when the firm has more tangible assets
that can be pledged as loan collateral.
It is smaller among firm in business or
professional services, finance/real estate and
retail trade.
Results:
Amount of Bank Credit



Amount of bank credit decreases with firm
size, the amount of financial slack and the
amount of tangible assets, and increases
with profitability and firm opaqueness as
proxied by firm age.
By industry, it is greater for transportation
firms and lesser for retail trade firms.
None of the owner characteristics are
consistently significant.
Summary and Conclusions



In this study, we have analyzed data from three
nationally representative surveys of privately
held U.S. firms for evidence on the use of credit
by small firms.
We contribute to the literature on the availability
of credit in at least two important ways.
First, we document that one in five small U.S.
firms uses no bank credit or trade credit, and
provide the first rigorous analysis of the
differences in these firms and other small U.S.
firms that do use credit.
Summary and Conclusions


Second, for those small U.S. firms that do
participate in the credit markets, we provide new
evidence regarding factors that determine their
use of trade credit and bank credit, and whether
these two types of credit are substitutes (Meltzer,
1960) or complements (Burkart and Ellingsen,
2004).
Our evidence strongly suggests that they are
complements, as two in five small U.S. firms
consistently use credit of both types. This is not
surprising because trade credit is primarily shortterm whereas bank credit is typically longerterm.
Summary and Conclusions



Our evidence also has important implications for
fiscal policy, as the administration and Congress
look for ways to stimulate credit provided to
small business lending.
Existing proposals focus exclusively on bank
lending while totally ignoring trade credit, which
is an equally important source of capital for small
businesses.
Complementary proposals should explore how to
expand trade credit offered by supplier as well as
how to expand bank credit offered by financial
institutions.