Omega 88 (2019) 77–90

Contents lists available at ScienceDirect

Omega
journal homepage: www.elsevier.com/locate/omega

Managing supplier financial distress with advance payment discount

and purchase order financing
Lima Zhao a,∗, Arnd Huchzermeier b
a
Ningbo Supply Chain Innovation Institute China, MIT Global SCALE Network, Wenyuan Road 462, Ningbo 315100, China
b
WHU—Otto Beisheim School of Management, Burgplatz 2, Vallendar 56179, Germany

a r t i c l e i n f o a b s t r a c t

Article history: We examine a capital-constrained supply chain in which a small- and medium-sized enterprise (SME)
Received 20 October 2017 supplier sells to an established retailer via a wholesale price contract. To mitigate the supplier’s financial
Accepted 24 October 2018
distress, the retailer chooses between two pre-shipment finance instruments: advance payment discount
Available online 26 October 2018
(APD) and buyer-backed purchase order financing (BPOF). When either APD or BPOF can be chosen, the
Keywords: retailer prefers APD to BPOF if her internal asset level is above a certain threshold. When both APD and
Supply chain finance BPOF are adopted, the retailer prefers APD and does not initiate BPOF unless the marginal cost of financial
Capital constraint distress dominates the benefit of unit discount. We show that the financing equilibrium region of APD
Advance payment discount increases not only with the retailer’s internal capital level, but also with demand variability. The interval
Purchase order financing and magnitude of the competition penalty incentivizes collaboration between supply chain partners.
© 2018 Elsevier Ltd. All rights reserved.

1. Introduction chase order—guaranteed by the buyer—to fund the supplier be-

Financial distress of suppliers can strongly affect supply chain
efficiency. For example, in February 2008, Chrysler temporarily
closed four assembly plants and canceled one shift at a fifth plant—
citing a parts shortage after its supplier, Plastech, filed for Chapter
11 bankruptcy protection [21]. A buyer can alleviate its supplier’s
financial distress via trade finance instruments [4]. In particular, a
buyer can finance a supplier by advancing payment at a discount
prior to product shipment; this procedure is referred to as “cash in
advance”.1 Fig. 1 shows a 2008 International Monetary Fund esti-
mate on the market share of trade finance instruments, where ad-
vance payment discount (APD) represents 19–22% (i.e., 3–3.5 tril-
lion US dollars) of global trade finance. In the literature on sup-
ply chain finance, bank finance and trade credit (i.e., open account)
have been frequently addressed [11,13,3,37]. In contrast, there is a
dearth of research on advance payment discount despite its preva-
lence in practice. Hence, this paper aims to assess the value of APD
in managing supplier financial distress.
Another trade finance instrument to mitigate the supplier’s cap-
ital constraint is buyer-backed purchase order financing (BPOF),
whereby a financial institution provides a loan based on a pur-
∗
Corresponding author.
E-mail address: limazhao@mit.edu (L. Zhao).
1
The only (minor) difference between cash in advance and advance payment dis-
count is whether the supply contract includes a unit discount. Thus APD can be seen
as the special case of cash in advance in which the buyer receives such a discount.
fore order delivery [19,29]. A program of this type was launched
by Nacional Financiera (Nafin), the Mexican state-owned develop-
ment bank [12]. Moreover, the specialty lender PurchaseOrderFi-
nancing.com has secured more than $750 million since 2002 to
boost US, UK, Canadian, and Chinese business growth [23]. How-
ever, relatively less attention has been paid to buyer-backed pur-
chase order financing in spite of its emergence as a viable trade
finance instrument. Therefore, our research attempts to evaluate
the impact of BPOF and its interaction with APD in a budget-
constrained supply chain.
In sum, we seek to fill these gaps in literature by addressing the
following research questions:
(i) Which financing strategy (APD or BPOF) is more efficient and
should be chosen by the buyer?
(ii) What is the financing equilibrium of APD and BPOF when both
instruments can be adopted?
(iii) How do the buyer’s optimal sourcing and financing decisions
affect supply chain efficiency?
To answer these questions, we first analyze respectively the
mechanisms of APD and BPOF in a capital-constrained supply chain
of one small- and medium-sized enterprise (SME) supplier and
an established retailer (buyer). Moreover, we derive the financing
equilibrium between APD and BPOF and show that the equilibrium
region of APD is increasing in both the retailer’s internal capital
level and demand variability. Hence, greater demand uncertainty
increases the need for risk sharing in the supply chain. When both

https://doi.org/10.1016/j.omega.2018.10.019
0305-0483/© 2018 Elsevier Ltd. All rights reserved.
78 L. Zhao and A. Huchzermeier / Omega 88 (2019) 77–90
Fig. 1. Trade finance arrangements by market share ([4]; BU = Berne Union).
trade finance instruments are viable, the retailer prefers APD and
will initiate BPOF only under certain conditions. That is, the sourc-
ing of capital from the retailer has a higher priority than the bor-
rowing of capital from a financial institution. In addition, we find
that there are considerable costs of competition penalty when de-
mand variability and the retailer’s internal capital level (respec-
tively) are within certain intervals. Hence, incentive alignment be-
tween supply chain partners is crucial for mitigating competition
penalty.
The remainder of this paper is organized as follows.
Section 2 reviews relevant literature and compares model setup
and main results of this study with closely related papers. In
Section 3, we present our stylized model with benchmark and
base cases. Section 4 analyzes the mechanisms of APD and BPOF,
respectively. In Section 5, we characterize the equilibria be-
tween APD and BPOF in both single-financing and dual-financing
schemes. Section 6 shows numerically the effects of demand
variability and the retailer’s internal capital level on the financ-
ing equilibrium and supply chain efficiency. We conclude in
Section 7 with a brief summary and directions for future research.
2. Literature review
Three streams of literature are closely related to our paper: (i)
trade credit and bank finance to manage capital constraint of the
buyer; (ii) joint operational and financial strategies to mitigate the
financial distress of the supplier; and (iii) advance payment dis-
count or purchase order financing.
The financial instruments to alleviate the impact of the buyer’s
budget constraint on supply chain performance including trade
credit has been extensively explored in literature (See [24] for an
excellent review). For instance, Jing et al. [11] study the financ-
ing equilibrium of bank and trade credits in a supply chain with
a capital-constrained retailer. Kouvelis and Zhao [13] analyze the
role of trade credit in inventory financing for a newsvendor-type
retailer in a Stackelberg game. Cai et al. [3] investigate the re-
lationship between trade credit and bank financing when a re-
tailer faces both budget constraint and demand uncertainty. Zhong
et al. [37] propose a supply chain network design model with trade
credit to optimize system-wise location, transportation, inventory
and financing costs.
Trade credit features post-shipment finance in which the seller
offers extended payment terms to the buyer after order delivery.
Hence, the above papers focus on post-shipment finance to manage
the capital constraints of the downstream buyer. Our research dif-
fers from this stream of literature by exploring two pre-shipment
finance instruments—i.e., advance payment discount and buyer-
backed purchase order financing—to mitigate the financial distress
of the upstream supplier.
Moreover, joint operational and financial strategies can be em-
ployed to manage the supplier’s budget constraint. Babich [1] con-
siders a manufacturer’s decisions on both capacity reservation and
financial subsidies to a supplier in a dynamic and stochastic pro-
gram. The author identifies conditions under which the manu-
facturer should decouple financial and operational decisions. Vliet
et al. [32] illustrate the value of supporting suppliers financially
with reverse factoring in a base stock model with periodic review.
Lekkakos and Serrano [17] examine how the adoptions of reverse
factoring and traditional factoring affect a supplier’s inventory pol-
icy in a stochastic multi-period setting.
Likewise, reverse factoring represents a post-shipment finance
instrument in that a supplier sells accounts receivable to a finan-
cial institution after product shipment. Therefore, this paper distin-
guishes from the above research by analyzing both APD and BPOF
as pre-shipment finance instruments. A comparison of these related
trade finance instruments is provided in Table 1.
Closest to our research is the recent literature on advance pay-
ment discount or purchase order financing. Gupta et al. [8] solve
a mixed integer constrained optimization with advance payment
discount by a real-coded genetic algorithm in a finite time inven-
tory model, and demonstrate the effect of advance payment and
interest rate on inventory policy in a single firm setting. Thangam
[28] studies a supply chain of one supplier and multiple retail-
ers that offer advance payment discount to customers with two-
echelon trade credits; and compare the earned interests from ad-
vance, immediate, and delayed payment schemes. The abovemen-
tioned two papers analyze the role of advance payment discount
in different settings without capital constraint, while our study ex-
amines advance payment discount provided by one supplier to one
retailer in a capital-constrained supply chain.
Wu [33] analyzes regular and pricing support purchase order
financing in comparison with financial subsidy to a supplier, and
shows the impact of loan interest rate on the buyer’s joint sourc-
ing and financing decisions. Tang et al. [27] address buyer direct
financing and purchase order financing without the buyer’s guar-
antee in a signaling game, and focus on the effect of information
asymmetry between the buyer and bank on the supplier’s cost.
Both papers consider purchase order financing in a single-financing
scheme, in which only one financial instrument can be employed;
whereas our research assesses the relative efficiency between ad-
vance payment discount and buyer-backed purchase order financing
in both single-financing and dual-financing (i.e., when both trade
finance instruments can be adopted simultaneously) equilibria.
Our research complements to the supply chain finance litera-
ture in three key aspects. First, we explore the retailer’s adoption
of APD and its interaction with BPOF in pre-shipment financing
of the supplier’s working capital. Second, we characterize the op-
timal operational and financial decisions in both single-financing
and dual-financing schemes. Third, we focus on the impact of de-
mand variability and the retailer’s internal capital level on financ-
ing equilibrium and supply chain efficiency. Table 2 presents an
overview of the related literature and new contributions of this pa-
per.
3. A supply chain with APD and BPOF
3.1. Formulation and assumptions
We consider a supply chain in which a SME supplier sells one
product to an established retailer. All parties are assumed to be
 L. Zhao and A. Huchzermeier / Omega 88 (2019) 77–90 79

Table 1
An overview of trade finance instruments.

Trade finance instrument Eligibility/collateral Initiator Sum financed Discount/interest rate Payment timing

Advance payment discount Purchase order Supplier Up to 100% of purchase Depends on contract Pre-shipment
order
Purchase order financing Purchase order Supplier Part of purchase order Depends on credit Pre-shipment
rating of supplier
Buyer-backed purchase Validated purchase Buyer Part of purchase order Depends on credit Pre-shipment
order financing order rating of buyer
Trade credit Invoice Supplier 100% of invoice Depends on contract Post-shipment
Recourse factoring Invoice Supplier Part of invoice Depends on credit Post-shipment
rating of supplier
Non-recourse factoring Invoice Supplier Part of invoice Depends on credit Post-shipment
ratings of supplier
and buyer
Reverse factoring Validated invoice Buyer Part of invoice Depends on credit Post-shipment
rating of buyer

Table 2
Summary of the related literature on supply chain finance and new contributions.

Reference Capital constraint Pre-shipment finance Post-shipment finance Financing equilibrium

Advance Buyer-backed
payment Purchase order purchase order Reverse Supplier Single- Dual-
Supplier Buyer discount financing financing Trade credit factoring subsidy financing financing
√ √ √
Jing et al. [11]
√ √ √ √
Kouvelis and Zhao
[13]
√ √ √
Cai et al. [3]
√ √ √
Babich [1]
√ √ √
Vliet et al. [32]
√ √ √
Lekkakos and
Serrano [17]
√ √
Gupta et al. [8]
√ √ √
Thangam [28]
√ √ √ √
Wu [33]
√ √ √
Tang et al. [27]
√ √ √ √ √ √
This paper

risk neutral. The retailer (she) orders q units of a product from the The supplier is capital-constrained, and the retailer could also
supplier (he), who has capacity K. Customer demand D is stochas- experience financial distress when she provides financing to the
tic and not realized until the sales season. The demand distribu- supplier. For instance, both GM and Chrysler have undergone Chap-
tion function F(D) is absolutely continuous with density f(D) > 0 ter 11 reorganization in 2009 while offering financial assistance to
and support [a, b) for 0 ≤ a < b ≤ ∞; it has a finite mean and an SME suppliers [18]. We assume that the retailer’s asset and lia-
 bility are deterministic.3 The capital market is imperfect. That is,
inverse F −1 (D ). The hazard rate h(D ) = f (D )/F̄ (D ) is increasing in
 when a firm cannot repay debt obligations, it can either be liqui-
D, where F̄ (D ) = 1 − F (D ). H (D ) = Dh(D ) denotes the generalized
dated or negotiate with creditors while enduring a costly reorgani-
failure rate; then H(D) is monotonically increasing in D. Suppose
zation process. In case of liquidation, the cost of financial default is
that F has a strictly increasing generalized failure rate (IGFR), and
a proportion 1 − γ (0 < γ < 1) of firm value. In case of reorganiza-
h(D ) = D f (D )/(1 − F (D )). Without loss of generality (w.l.o.g.), the
tion, the cost of financial distress is a proportion 1 − α (0 < α < 1)
risk-free interest rate rf is normalized to zero [2,11]; this allows us
of raised capital [16,6]. Our notation and assumptions are summa-
to concentrate on the effective rate of APD and BPOF (i.e., the actual
rized in Table 3.
rate above risk-free rate).
Buyer-backed purchase order financing is the only pre-shipment
The supplier and retailer operate with limited amounts of in-
option for external financing. In this case, the retailer initiates a tri-
ternal capital before shipment, both firms have long-term capital
partite agreement with a financial institution to provide the sup-
structures financed solely by equity and have short-term debts due
plier with a BPOF loan based on the retailer’s guarantee. Fig. 2 de-
before the sales season. Thus the supplier and retailer face the risk
picts the sequence of events in BPOF. In this setting, other types
of financial distress before shipment, which occurs (or not) de-
of external financing are not accessible to the SME supplier be-
pending on whether internal assets are enough to cover loan obli-
cause of his low credit rating. Hence, the loan interest rate offered
gations. The supplier has internal asset As and short-term debt Ls .
by commercial banks to the supplier is assumed w.l.o.g. as rs = ∞.
The supplier’s asset As is assumed to be stochastic and not realized
We assume that a backup supplier is not available. In practice, this
until the sales season cf. [1,9,35]). It has the cumulative distribu-
captures two cases: (i) when there is no alternative supplier; and
tion function (CDF) (As ), the probability density function (PDF)
φ (As ), and As ∈ [As , As ] for 0 ≤ As < As ≤ ∞. The supplier’s short-
term debt Ls is deterministic and will be due before the sales sea- nal asset is subject to market variables (e.g., interest rates and commodity prices).
son.2 As an established firm, the retailer has a higher credit rating Since this leads to the stochastic internal capital level, adding stochasticity to debt
with the financial institution than the SME supplier. will not alter the structure of our main results.
3
An established firm (large corporation) typically employs financial hedging to
mitigate the impact of market dynamics (incl. commodity price, exchange and inter-
2 est rates). Hence, the value of her asset and liability can be viewed as deterministic
In practice, this refers to the case in which the supplier has been informed by
debt holders about the amount and due date of liabilities, while the value of inter- in a short term.
80 L. Zhao and A. Huchzermeier / Omega 88 (2019) 77–90

Table 3
Summary of notation and assumptions.

Symbol Description Assumptions

πi The expected profits of the supplier, retailer, and financial i = s, r, c∗

institution
Ai Asset level of supplier/retailer Exogenous, i = s, r
Li Short-term debt of the supplier/retailer due before order Exogenous, i = s, r
delivery
D Product demand Exogenous, stochastic
f( · ) The probability density function (PDF) of demand distribution Exogenous, f(D) > 0
F( · ) The cumulative distribution function (CDF) of demand Exogenous, F(D) has support [a, b)
distribution for 0 ≤ a < b ≤ ∞
F̄ (· ) The complementary cumulative distribution function (CCDF) of Exogenous
the demand distribution, F̄ (D ) = 1 − F (D )
h( · ) The failure rate of the demand distribution, Exogenous, increasing in D
h(· ) = f (· )/F̄ (· )
φ( · ) The PDF of the supplier’s asset distribution Exogenous, As ∈ [As , As ] for
0 ≤ As < As ≤ ∞
( · ) The CDF of the supplier’s asset distribution Exogenous
(
¯ ·) The CCDF of the supplier’s asset distribution, Exogenous
(
¯ As ) = 1 − (As )
K Capacity of the supplier Decision variable, K ∈ R+
q Order quantity of the retailer to the supplier Decision variable
w Unit wholesale price Decision variable, p > w > c p + ck
w−c p −ck
d Discount rate of the retailer’s advance payment to the supplier Decision variable, d ∈ [0, w
]
before order delivery
cp Unit production cost of the supplier Exogenous, deterministic, w > c p + ck
ck Unit capacity cost of the supplier Exogenous, deterministic, w > c p + ck
p Unit product price Exogenous, deterministic, p>w
α Proportional distress cost stated as a portion (1 − α ) of raised Exogenous, deterministic, α ∈ (0, 1)
capital upon reorganization
β Portion of orders at wholesale price in dual-financing scheme Decision variable, β ∈ [0, 1]
γ Proportional liquidation cost stated as a portion (1 − γ ) of Exogenous, deterministic, γ ∈ (0, 1)
firm value upon financial default
δ Portion of the financial institution’s loss compensated by the Exogenous, δ ∈ (0, 1)
retailer upon supplier liquidation
λ Portion of purchase order value received by the supplier from Decision variable, λ ∈ [0, λ̄], λ̄ ∈ (0, 1 )
the financial institution
r Interest rate of BPOF loan offered by the financial institution to r≥0
the supplier
rf Risk-free interest rate rf = 0
rs Commercial bank interest rate for the supplier rs = ∞
∗
Throughout the paper we use “s”, “r”, and “c” (for creditor) to denote the “supplier”, “retailer”, and “financial institution”, respectively.

simplicity and wide use in practice. In order to avoid the cost of

Fig. 2. Sequence of events in buyer-backed purchase order financing.
(ii) when the lead time needed to qualify a backup supplier is pro-
hibitively long. For instance, BMW sources the sun roofs for its Z4
convertible from the supplier Edscha, which filed for bankruptcy
in February 2009. Even if BMW could find an alternative supplier,
at least six months would elapse before the new company could
start producing the convertible top [25]. These considerations lead
the buyer to seek pre-shipment finance instruments to fund the
supplier’s working capital.
Since neither APD nor BPOF can be realized without the re-
tailer’s agreement, she will be able to choose between these two
financing strategies. In order to focus on financing schemes, our
model is based on a wholesale price only contract owing to its
financial distress and bankruptcy, the SME supplier may opt to in-
stall less capacity. Thus, the retailer is incentivized to assist the
supplier with financing and thereby ensure order delivery. Capac-
ity building is associated with lead time, and there is no lead time
for production. Before the sales season, the supplier installs his ca-
pacity K ∈ R+ at unit capacity cost ck . Then he manufactures at
unit production cost cp during the sales season. The salvage value
of capacity is zero.
The timeline of events is presented in Fig. 3. First, the re-
tailer bids wholesale price w to the supplier. The supplier decides
whether to accept the wholesale price anticipating his probability
of financial distress and default determined by his asset distribu-
tion. Upon acceptance, the supplier will offer the unit discount rate
d and ensure w(1 − d ) > c p + ck for his profitability in APD. Then
the retailer decides on an order quantity q and chooses the financ-
ing strategy (i.e., APD and/or BPOF). The supplier determines his
capacity K based not only on the retailer’s order quantity but also
on the extent to which his working capital can be financed. In or-
der to discourage underinvestment by a financially distressed sup-
plier, the retailer is incentivized to provide pre-shipment finance
through APD and/or BPOF. If the retailer chooses APD, she will pay
the supplier before product shipment. If the retailer chooses BPOF,
she will provide a guarantee such that a financial institution offers
a BPOF contract (λ̄, r ) to the supplier based on the retailer’s credit
 L. Zhao and A. Huchzermeier / Omega 88 (2019) 77–90
Fig. 3. Timeline of events.
rating. Here λ̄ ∈ (0, 1 ) is the upper bound on the share of the pur-
chase order that the financial institution is willing to lend, and r
is the BPOF interest rate. If the supplier accepts the BPOF contract
then he chooses a borrowing level λ ∈ [0, λ̄], after which he will
receive BPOF in the amount of λwq from the financial institution.
Then the supplier’s stochastic internal asset level is realized. Short-
term debts of both the supplier and retailer are due before demand
realization.
The sales season begins and demand is realized, then the sup-
plier produces and delivers to the retailer. In the base case, the
retailer pays w min(q, K) to the supplier. In APD, the retailer would
receive a refund from the supplier if any prepaid order quantity
is not fulfilled. In BPOF, if the supplier operates in continuation or
undergoes reorganization then the retailer pays the total purchase
order amount to the financial institution, which deducts the BPOF
loan’s principal and interest and then pays the supplier the bal-
ance. If the supplier is in liquidation, then (i) the financial institu-
tion seizes the supplier’s liquid assets and (ii) the retailer pays a
previously negotiated portion δ of the financial institution’s loss as
compensation, because the BPOF loan was backed by her guaran-
tee. The retailer sells products to customers at a unit price p that
is determined by the market. We assume p > w to ensure that the
retailer is profitable. Unmet demand is lost, and the salvage value
of unsold product is zero.
3.2. Centralized and decentralized benchmarks
To establish benchmarks for a capital-constrained supply chain,
we analyze the cases of centralized and decentralized supply
chains that do not exhibit financial distress. In a centralized supply
chain, an integrated firm with ample internal capital makes capac-
ity decisions geared to achieving optimal channel-wide expected
profit. The expected profit of an integrated firm that has made ca-
pacity decision K is4
π csc (K ) = ( p − c p )Emin(K, D ) − ck K.
The profit function is concave, and the optimal solution is K csc ∗ =
p−c p −ck
F −1 ( p−c p
).
In a decentralized supply chain, both the retailer and supplier
have sufficient internal capital. The retailer’s problem is equivalent
to (1) except that she orders inventory at wholesale price w in-
stead of producing it at cost cp and ck . Hence her expected profit
is
πrdsc (q, w ) = pE min [D, min (q, K )] − w min(q, K ).
The profit function is concave, and its optimal solution is = qdsc
F −1 ( p−w
p ). In contrast, for a centralized supply chain the optimal
4
We shall use “csc” and “dsc” to denote (respectively) the “centralized supply
chain” and “decentralized supply chain” benchmarks. For brevity we use q, w, and
K to denote (respectively) qi , wi , and Ki for i = csc, dsc, bc, bpof, apd, df, where
“bc” denotes “base case” and “df” denotes “dual financing”.
81
∗ csc ∗ = F −1 ( p−c p −ck
solution is qdsc ≤ F −1 ( p−w
p )<K p−c p ); this expres-
sion reflects the result of “double marginalization” [26].
In the absence of financial distress, the supplier sets wholesale
price and anticipates the retailer’s order quantity for any wholesale
price. In practice, this setting can be applied in two cases: (i) when
the supplier is indispensable for the retailer as a backup supplier is
not available (see Section 3.1), or (ii) when the supplier has a high
reservation profit or opportunity cost [10]. He faces demand curve
qdsc (w) and chooses optimal capacity Kdsc ∗ to maximize his profit:
πsdsc (K, w ) = (w − c p ) min(q, K ) − ck K. (3)
We follow the approach of Lariviere and Porteus [15], p. 295 and
write the supplier’s inverse demand curve as w(q ) = pF̄ (q ). We de-
fine qˆ as the least upper bound on the set of points such that
v(q) ≥ 1. Here the price elasticity of the retailer’s order is v(q ) =
−w(q )/[qdw(q )/dq] if we assume that v (q) ≤ 0 for q ∈ [a, b) and
qˆ ∈ [a, ∞ ).
Lemma 1. In a decentralized supply chain with sufficient capital, the
supplier’s first-order condition is
 ∗   ∗ 
pF̄ K dsc 1 − h K dsc = ck + c p (4)
The supplier’s profit is unimodal on [0, ∞), linear and strictly in-
creasing on [0, a), strictly concave on [a, qˆ], and strictly decreas-
∗
ing on (qˆ, ∞ ). The optimal solution qdsc to Eq. (4) is unique and
must reside on the interval [a, qˆ]. The supplier’s optimal capacity is
∗ ∗
K dsc = qdsc , his optimal sales quantity is either qdsc or a, and the
∗ ∗ ck +c p
optimal wholesale price is wdsc = pF̄ (qdsc ) = dsc ∗
.
1−h (q )
Proof. All proofs are given in the Appendix.
3.3. Base case
In this case, one established retailer orders from one SME sup-
(1) plier, where each party could be capital constrained and neither
APD nor BPOF is viable. In the presence of financial distress, the
retailer bids wholesale price w to the supplier. Then the sup-
plier chooses whether to accept the wholesale price based on his
probability of financial distress and default. If the retailer dic-
tates the wholesale price, the supplier’s expected profit is zero
as the wholesale price set by retailer equals the supplier’s ex-
pected total unit cost (cf. [22]). This increases the retailer’s or-
der quantity relative to the case where the supplier sets whole-
(2) sale price, and thereby increases the supplier’s working capital
∗ shortfall. As a result, this setting increases the supplier’s proba-
bility of underinvestment to alleviate the costs of financial dis-
tress and default. Therefore, the retailer should strategically by-
pass this scenario to ensure her supply. The retailer chooses her
order quantity upon acceptance, and the supplier determines his
capacity K by considering the retailer’s order quantity and his ex-
pected working capital level. Hence, the retailer’s expected profit
is πrbc (q, w ) = pE min[D, min(q, K )] − wbc min(q, K ). Suppose the
82 L. Zhao and A. Huchzermeier / Omega 88 (2019) 77–90
capital-constrained supplier’s capacity decision is Kbc ; then his
profit is
⎧ a BPOF loan of λwq from the financial institution. If the supplier
⎨(w − c p ) min(q, K ) − ck K
⎪ in continuation,
πs (K, w ) = (
bc w − c p ) min(q, K ) − ck K
⎪− ( 1 − α ) ( Ls − As + ck K ) in reorganization,
⎩ files for bankruptcy and chooses between reorganization and liqui-
0 in liquidation.
If the supplier’s liquid asset is able to cover current obliga-
tions (i.e., if As ≥ Ls + ck K) before order delivery, then the sup-
plier continues his operations. Otherwise, the supplier chooses be-
tween two alternatives to bankruptcy, reorganization or liquida-
tion; he chooses the latter only if the cost of financial distress ex-
ceeds the firm’s operating profit under reorganization [35]. That
is, if the supplier’s expected profit after reorganization is positive,
then he will opt to undergo a costly reorganization process. There-
fore, the supplier’s probability of continuation is Pr(c ) = ( ¯ As )

for As = Ls + ck K, the probability of liquidation is Pr(l ) = (As ) for
 (w−c p ) min(q,K )−ck K
As = Ls + ck K − (1−α ) , and the probability of reorgani-
zation is Pr(r ) = (As ) − (As ).5 In order to avoid the costs asso-
ciated with financial distress or default, a supplier may install less
capacity (i.e., underinvestment) than the retailer’s order quantity.
In case of liquidation, the supplier’s profit is zero thus his opti-
∗
mal capacity K bc = 0. In the scenarios of continuation and reorga-
∗
nization, the supplier’s optimal capacity decision Kbc satisfies the
first-order conditions (FOCs):
 ∗   ∗ 
pF̄ K bc 1 − h K bc = c p + ck in continuation,
 ∗   ∗ 
pF̄ K bc 1 − h K bc = c p + (2 − α )ck in reorganization. (5)
The supplier’s optimal capacity can be derived by taking the
weighted average of the optimal capacity in each scenario. Here
∗ (c +c ) Pr(c )+(c p +(2−α )ck ) Pr(r )
wbc = p k 1−h (K )
. The FOCs indicate that the sup-
plier trades off the marginal benefit of unit production against the
marginal cost of unit capacity. In case of reorganization, the pres-
ence of financial distress cost increases the marginal cost of unit
capacity, and thereby lowers the supplier’s optimal capacity level

(recall that hazard rate h(D ) = f (D )/F̄ (D ) is increasing in D, and
F̄ (D ) = 1 − F (D ) is decreasing in D). Therefore, the retailer has in-
centives to offer pre-shipment finance instruments as an effort to
assure supply by mitigating the impact of financial distress on the
supplier’s capacity decision.
4. Financing with APD or BPOF
To establish the basis for analyzing the financing equilibria of
APD and BPOF, we first examine the cases where only one of the
two instruments is viable in the supply chain.
4.1. Buyer-backed purchase order financing
We start with the case in which the retailer adopts only buyer-
backed purchase order financing, ceteris paribus. Recall that the re-
tailer bids wholesale price to the supplier in the presence of finan-
cial distress. The supplier decides whether to accept the wholesale
price in accordance with his probability of financial distress and
default. If accepted, the retailer decides on an order quantity q and
initiates BPOF by providing a loan guarantee (on the supplier’s be-
half) to a financial institution based on her purchase order (see
Fig. 2). The financial institution then offers a BPOF contract (λ̄, r )
to the supplier. The supplier determines his capacity K with re-
gard to the retailer’s order quantity and how his working capital
5
The conditions for continuation, reorganization, and liquidation can be readily
derived, throughout the paper, using similar logic. So to ease the exposition, here-
after we omit the corresponding analytical expressions.
can be financed in BPOF. A supplier who accepts the BPOF contract
can choose a borrowing level λ ∈ [0, λ̄], whereafter he will receive
repays his short-term debt Ls prior to product shipment, then he
continues to operate in the sales season; otherwise, the supplier
dation. The supplier liquidates when his internal asset falls below
(w−c p ) min(q,K )−ck K−λwqr
the threshold value As = Ls + ck K − λwq − 1−α ,
thus the probability of liquidation is Pr(l ) = (As ).
The supplier ships his products after demand realization. In the
event of continuation (continued operation) or reorganization, the
retailer pays back the financial institution, which deducts the BPOF
loan principal and interest before remitting the balance to the sup-
plier. In the event of liquidation, however, the POF loan is in de-
fault; then the financial institution receives the supplier’s liquid as-
sets and the retailer—as guarantor of the supplier’s credit—pays a
previously determined portion (δ ) of the financial institution’s loss.
Any additional fees charged by the financial institution are normal-
ized to zero because they are usually small and do not affect our
structural results.
We derive optimal decisions in the supply chain as follows. In
BPOF-only scenario, the retailer’s expected profit is
πrbpof (q, w ) = pE min [D, min (q, K )] − w min (q, K )
As
− δ{λwq − γ [(w − c p ) min(q, K ) − ck K
As
−λwqr + As − Ls ]}φ (As )dAs . (6)
The retailer will agree to guarantee the supplier’s credit if her
expected profit from initiating BPOF exceeds that in the base
case—that is, if πr (q, D, K ) ≥ πrbc (q, D, K ). Hence the likelihood
bpof
of buyer participation in BPOF declines as the supplier’s liquida-
tion probability (As ) increases.
The supplier makes operational and financial decisions to max-
imize his profit:
πsbpof (λ, w, K )
⎧
⎨(w − c p ) min(q, K ) − ck K − λwqr
⎪ in continuation,
(w − c p ) min(q, K ) − ck K − (1 − α )
=
⎩(Ls − As − λwq + ck K ) − λwqr
⎪ in reorganization,
0 in liquidation.
We denote the supplier’s break-even BPOF interest rate as rˆ, and
his optimal borrowing level (λ∗ ) is defined in Proposition 1.
Proposition 1. In the case of buyer-backed purchase order financing
only, the supplier’s optimal borrowing level is
arg max πs (λ, w, K ) if r < 1 − α and r ≤ rˆ,
λ∗ = λ∈ (0,λ̄]
0 otherwise.
∗
In case of liquidation, his optimal capacity K bpof = 0. Otherwise
Kbpof ∗satisfies the FOCs
∗ ∗
pF̄ (K bpof )[1 − h(K bpof )] = c p + ck in continuation,
∗ ∗
pF̄ (K bpof )[1 − h(K bpof )] = c p + (2 − α )ck in reorganization.
(7)
Proposition 1 indicates that if capital constraints lead the sup-
plier to underinvest in capacity, then the risk of that scenario’s
transpiring can be mitigated by BPOF at the cost of an inter-
est payment. The conditions for a positive borrowing level, i.e.,
r < 1 − α and r ≤ rˆ ensure the financing cost of BPOF is lower than
proportional distress cost and is profitable for the supplier. In com-
parison with the base case, the BPOF loan decreases the sup-
plier’s probability of financial distress and default, and thereby
 L. Zhao and A. Huchzermeier / Omega 88 (2019) 77–90 83

increases the supplier’s optimal capacity level. In case that the
BPOF loan is sufficient to fund the supplier’s working capital (λ̄ ≥
ck K bpof +Ls −As
) and r ≤ rˆ, the supplier invests in capacity K bpof =
wqbpof
∗ ceteris paribus. Here we assume that the retailer purchases the
qbpof = F −1 ( p−w
p ), i.e., as if he were not capital constrained. In
other words, the supplier can make his operational decisions with-
out considering financial constraints provided the BPOF loan is suf-
ficient for his working capital needs.
Anticipating the supplier’s response, the financial institution de-
cides on (λ̄, r ) to maximize its expected profit:
  As
πc = 
¯ As λwqr + (1 − δ ){γ [(w − c p ) min(q, K ) − ck K
As
−λwqr + As − Ls ] − λwq}φ (As )dAs .
The first term on the right-hand side is the financial institu-
tion’s expected payoff when the supplier is in continuation or re-
organization, and the second term is its expected payoff when
the supplier defaults and liquidates. Although the exact number of
BPOF providers is not easy to estimate, the New York Times reports
that there are at least six major BPOF companies in the US market
[19]. We therefore assume that the BPOF lending market is com-
petitive, from which it follows that the financial institution has a
zero expected payoff:
As
λwqr[(
¯ As ) − (1 − δ )(As )γ ] + (1 − δ )γ [(w − c p ) min(q, K )
As
−ck K + As − Ls ]φ (As )dAs = (1 − δ )(As )λwq.
4.2. Advance payment discount
∗ In this case, the retailer adopts only advance payment discount,
total order quantity at a discounted price before order delivery
(this assumption is generalized in Section 5.2). Although the re-
tailer is not at risk of financial default (liquidation), she may be
subject to short-term financial distress if she finances the supplier
solely via advance payment discount. The retailer bids wholesale
price w, then the supplier decides whether to accept the whole-
sale price with regard to his probability of financial distress and
default (subject to his asset distribution). The supplier offers the
w−c p −ck
unit discount rate d ∈ [0, w ] upon acceptance. Then the re-
(8) tailer decides on order quantity q, adopts APD only by paying
wqapd (1 − d ) to the supplier before product shipment, and will re-
ceive w(1 − d )(qapd − K apd )+ after delivery in case any amount or-
dered is not fulfilled. The supplier determines his capacity K based
on the retailer’s order quantity and on the extent to which his ex-
pected working capital needs are financed in APD.6
In this APD-only scenario, the supplier’s profit
with his optimal capacity K apd∗ is π apd
( K, w) =
⎧ s
⎪ [w(1−d )−c p] min(q, K )−ck K in continuation,
⎨
[w(1 − d ) − c p] min(q, K )
⎩−ck K − (1 − α )(Ls − As + ck K − w(1 − d )q) in reorganization,
⎪
0 in liquidation.
The retailer with initial asset Ar and short-term liabil-
ity Lr (due before order delivery) might incur financial dis-
tress. Given her internal asset, the retailer chooses APD as
long as it does not lead to bankruptcy with liquidation (oth-
(9) erwise, the retailer will not adopt APD); her aim is to max-
imize the expected profit πr (K, w ) = pE min[D, min(q, K )] −
apd

Eq. (9) reveals that r increases with λ when ( ¯ As ) > w(1 − d ) min(q, K ) − (1 − α )[Lr + w(1 − d )q − Ar ]+ .
(1 − δ )(As )γ . In other words, if the joint probability of continu-
ation and reorganization is greater than the liquidation probability Proposition 2. In case of liquidation, the supplier’s optimal capacity
∗ ∗
multiplied by both the bankruptcy cost ratio and the loss compen- K apd = 0. Otherwise Kapd satisfies the FOCs
sation factor—a condition frequently satisfied in practice—then the  ∗   ∗ 
F̄ K apd 1 − h K apd = c p + ck in continuation,
financial institution will charge a higher interest rate if the sup-  papd ∗   ∗ 
plier’s expected borrowing level is higher. The underlying reason is pF̄ K 1 − h K apd = c p + (2 − α )ck in reorganization.
that lending risk increases with loan size.
(10)
Buyer-backed purchase order financing enables the supplier to
receive working capital funding from a financial institution based If advance payment discount is sufficient to fund the supplier’s
∗
on the retailer’s credit rating, which creates a “win–win–win” sit- working capital, the retailer’s optimal order quantity qapd satisfies
uation by providing three benefits: (i) mitigating the retailer’s sup-  ∗
F̄ qapd = w(1 − d ) in continuation,
ply shortfall due to the supplier’s financial distress, (ii) financing
 papd ∗ (11)
the supplier’s working capital to ensure that orders are fulfilled (at pF̄ q = w(1 − d )(2 − α ) in reorganization.
the cost of BPOF interest), and (iii) yielding the financial institution
∗ ∗
the BPOF interest payment. In comparison with advance payment Moreover, qapd ≥ qdsc if and only if (1 − d )(2 − α ) ≤ 1.
discount that transfers capital within the supply chain, using a fi- This proposition7 demonstrates the retailer’s trade-off between
nancial institution’s capital via BPOF enables the supplier to fulfill marginal effect of advance payment discount and unit cost of
order requirements and the retailer to extend payment terms. financial distress when the retailer’s internal asset level is be-
Our observations are in line with empirical evidence that BPOF low a certain threshold (Ar < Lr + w(1 − d )qapd ). When the effec-
benefits not only supplier and buyer but also the financial insti- tive wholesale price (i.e., the price that determines the retailer’s
tution [12,20]. For the supplier, BPOF reduces transaction and bor- 
order quantity) we = w(1 − d ) [1 + (1 − α ) 1Lr +w(1−d )qapd >Ar ] does
rowing costs. Buyer-backed purchase order financing offers work-
not exceed w, the channel coordination benefit of APD dominates
ing capital at favorable rates, which provides instant liquidity by ∗ ∗
the financial distress effect; hence qapd ≥ qdsc . In one special case
reducing days sales outstanding (DSO) and thereby accelerates the
where a well-established retailer endowed with internal capital
cash-to-cash (C2C) cycle [5]. From the buyer’s perspective, since
reserves and thus immune to financial distress, we ≤ w. Thus the
her accounts payables are managed by a financial institution, her ∗ ∗
APD’s channel coordination effect is guaranteed: qapd ≥ qdsc . If
administrative costs of processing different payment terms with
the effective wholesale price we > w, then the channel coordina-
multiple suppliers are reduced. By financing the suppliers’ work-
tion benefit of APD is dominated by its financial distress effect and
ing capital, the buyer can enhance her reputation and relationships
with SME suppliers. For the financial institution, BPOF helps to de-
velop relationships with suppliers. For example, credit histories of 6
For the case when APD is sufficient to fund the supplier’s working capital,
SME suppliers can be assembled in BPOF. Since BPOF involves only please refer to Zhao and Huchzermeier [36].
high-quality receivables, financial institutions can expand their op- 7
Here the conditions for continuation and reorganization of the supplier and the
erations without increasing credit risk. retailer are derived respectively.
84 L. Zhao and A. Huchzermeier / Omega 88 (2019) 77–90
∗ ∗
thus qapd < qdsc . APD lowers the supplier’s probability of financial
distress and default and thereby increases the supplier’s optimal
capacity level relative to the base case. If APD is sufficient to fund
the supplier’s working capital (w(1 − d )q ≥ ck K ∗ + Ls − As ), then
∗ ∗
the supplier will invest in capacity K apd = qapd = F −1 ( p−w(p1−d ) ),
i.e., the supplier can make his capacity decisions as if he were not
capital constrained when APD fulfills his working capital needs.
The retailer benefits from APD’s channel coordination effect but
at the cost of potential financial distress. The supplier also ben-
efits from that effect as well as from the faster collection of re-
ceivables [5]. Both APD and BPOF can relieve the supplier’s capital
constraint: BPOF increases the supply chain’s overall financing ca-
pacity by borrowing capital from a financial institution, whereas
APD transfers monetary flows from voluntary to binding positions
within the supply chain. In addition to mitigating the supplier’s
budget constraint, advance payment discount reduces the whole-
sale price and thus yields a coordination benefit for the supply
chain (though at the cost of the retailer’s possible financial dis-
tress).
5. Interactions between APD and BPOF
So far we have derived the respective optimal decisions in sup-
ply chain under APD and under BPOF. If both APD and BPOF are
viable then one question naturally arises: which trade finance in-
strument(s) will be selected in the financing equilibrium? To an-
swer this question, Section 5.1 characterizes the retailer’s optimal
choice between APD and BPOF when only one of the two instru-
ments can be adopted. In Section 5.2, we present the optimal fi-
nancing and operational decisions when both APD and BPOF can
be chosen simultaneously.
5.1. Single financing equilibrium between APD and BPOF
When either APD or BPOF alone is sufficient to fund the sup-
plier’s need for working capital (i.e., when wq∗ λ̄ ≥ ck K ∗ + Ls − As
and w(1 − d )q ≥ ck K ∗ + Ls − As ), the retailer will choose one of the
instruments in single-financing scheme, ceteris paribus. We assume
w.l.o.g. that the retailer chooses APD over BPOF if she is indifferent
between them.
Theorem 1. Under single financing, there exists a unique threshold of
the retailer’s internal asset level ωr such that she prefers BPOF if and
only if Ar < ωr . Otherwise, the retailer chooses APD.
This threshold of the retailer’s internal asset implies that the
optimal financing depends: (i) under APD, on the trade-off be-
tween the benefit of channel coordination and the cost of financial
distress; and (ii) under BPOF, on the trade-off between the benefit
of a secured supply and the cost of a credit guarantee in case the
supplier liquidates. Although the retailer pays a lower price in APD
than in BPOF, she then faces increased inventory risk (because the
order quantity is greater) in addition to the risk of financial dis-
tress. When the retailer’s internal asset level is so high that finan-
cial distress is not a concern, she always prefers APD to BPOF.
Theorem 1 indicates that a more wealthy buyer prefers using
internal sources of capital to fund the supplier whereas a less
wealthy buyer—whose credit rating is higher than her supplier’s—
should leverage that interest rate spread to borrow capital from
a financial institution. The implication is that APD is most suit-
able for a supply chain consisting of a capital-constrained supplier
and an established retailer whose high internal asset level pre-
cludes financial distress; this follows because the retailer prefers
APD to BPOF provided the former’s channel coordination benefits
dominate her concerns about financial distress. These observations
conform with the empirical finding that APD accounts for 19–22%
of global trade finance [4], since established and wealthier retail-
ers with larger trade volumes are more likely to adopt APD. Be-
cause BPOF transfers credit risk from a distressed supplier to a
high-quality buyer, it is frequently used for transactions between
a buyer in a developed economy and an SME supplier located in a
developing economy [7,30,31]).
5.2. Dual financing with both APD and BPOF
If either APD or BPOF alone is not sufficient to cover the
supplier’s working capital needs, then both instruments can be
adopted in a dual-financing scheme, ceteris paribus. Under dual
financing, the retailer orders qdf = q1 + q2 from the supplier. The
terms q1 = (1 − β )qdf and q2 = β qdf (β ∈ [0, 1]) denote the respec-
tive order quantities paid at advance discount before shipment and
at wholesale price after delivery. The retailer maximizes her profit
by jointly choosing the total order quantity qdf and the portion
β of order quantity paid at wholesale price. Dual financing is vi-
able when the following two conditions hold: (i) the BPOF loan
is insufficient to cover the supplier’s working capital (i.e., wq∗2 λ̄ <
ck K ∗ + Ls − As ; this generalizes the assumption in Section 5.1); and
(ii) the retailer’s partial payment via APD is insufficient to cover the
supplier’s working capital needs (i.e., w(1 − d )q∗1 < ck K ∗ + Ls − As ;
this relaxes assumptions in Sections 4.2 and 5.1). Therefore, dual
financing represents the general case of pre-shipment finance.
Theorem 2. If dual financing is sufficient to fund the supplier’s work-
ing capital, the retailer’s optimal order quantity satisfies the FOCs
 ∗ 
pF qdf = p − w(1 − d ) − β ∗ wd in continuation,
 df ∗

pF q = p − (2 − α − β ∗ + αβ ∗ )w(1 − d ) − β ∗ wd
in reorganization. (12)
The supplier’s optimal BPOF borrowing level is
arg max πs (λ, w, K ) if r < 1 − α and r ≤ r˜,
λ∗ = λ∈ (0,λ̄]
0 otherwise.
Here r˜ denotes the supplier’s break-even BPOF interest rate un-
der dual financing. Theorem 2 shows that, in APD, the retailer
trades off the benefit of a unit discount against the possible cost
of financial distress. If APD proves insufficient, the supplier then
seeks external financing via BPOF to cover working capital needs.
In dual financing, a retailer with sufficient internal capital favors
funding of the supplier by APD whereas a retailer with insufficient
capital will first maximize her use of APD and only then use BPOF
when the cost of financial distress outweighs the benefit of a price
discount.
Lemma 2. In the case of continuation, the retailer’s total order quan-
tity is decreasing in the proportion β of order quantity paid at whole-
sale price after product shipment. In the case of reorganization, the
retailer’s total order quantity is increasing in β if and only if 1 − α >
d
.
1−d
Lemma 2 implies that the retailer strictly prefers APD to BPOF if
her internal capital is high enough to preclude financial distress. If
the retailer goes through reorganization, she may then prefer BPOF
to APD under the binding condition that the marginal cost of fi-
nancial distress is greater than the benefit of a unit discount.
Theorem 2 and Lemma 2 indicate that the choice of an opti-
mal pre-shipment finance strategy depends on the retailer’s level
of internal capital. As a risk-sharing scheme that provides channel
coordination benefits despite (possibly) incurring financial distress,
APD receives priority under dual financing when the retailer pos-
sesses ample internal capital.
 L. Zhao and A. Huchzermeier / Omega 88 (2019) 77–90
Fig. 4. Effect of demand volatility on (a) supplier’s delivered quantities; (b) retailer’s profits;
(c) supplier’s profits; and (d) supply chain profits.
6. Numerical analysis
In this section, we examine the robustness of our main re-
sults via an extensive numerical study. Examples are provided to
demonstrate how demand variability and the retailer’s internal
capital level each affects supply chain performance. Our numeri-
cal study relies on optimization via Monto Carlo simulation with
1,0 0 0,0 0 0 uncertainty scenarios (supply–demand matching states
and price–quantity decisions) and the following benchmark param-
eter values: p = 60, ck = c p = 10, α = 0.85, and γ = 0.9 (cf. [6]).
Demand D follows a normal distribution, N(10 0 0, 10 0), whose stan-
dard deviation may vary with its coefficient of variation (CV).
6.1. Impact of demand variability
We first examine how demand volatility affects the interaction
between APD and BPOF. When adjusting demand volatility, we al-
low its coefficient of variation δ (D) to vary between 0.1 and 0.5;
at the same time, we use a constant value for expected demand,
E[D] = 1, 0 0 0. We compare the channel efficiency under APD and
BPOF and plot the threshold points in Figs. 4 and 5. For each fi-
nancing scheme, we follow Jing et al. [11] in examining the per-
centage competition penalty (i.e., the portion of profit decrease in
decentralized supply chain relative to centralized supply chain)
 πs + πr 

P = 1− × 100%
π csc
85
w.r.t. demand volatility and the retailer’s internal capital level.
Fig. 4 presents the effects of increasing demand CV on supply chain
performance and on the financing equilibrium between APD and
BPOF.
Fig. 4(a) shows that the supplier’s delivered quantity is de-
creasing in demand volatility without pre-shipment finance (APD
or BPOF). Whereas the supplier’s delivered quantity is strictly in-
creasing in APD only (green curve), it is convex in BPOF only (blue
curve) as well as in the dual-financing scheme (grey curve). Ab-
sent pre-shipment finance, the supplier’s order delivery decreases
as demand volatility increases. When pre-shipment finance (APD
and/or BPOF) is employed, the supplier’s delivered quantity is less
sensitive to demand volatility because his financial distress is miti-
gated by pre-shipment finance; thus, demand risk is shared with
the retailer. Fig. 4(b) demonstrates that the retailer’s profits are
increasing in demand volatility as a result of the lower whole-
sale price. Comparing the retailer’s profits under APD only with
BPOF only reveals that the latter’s equilibrium region decreases
with demand variability. That is, in single financing, the retailer
chooses BPOF when the demand coefficient of variation is rela-
tively low (i.e., when δ (D) < 0.22); otherwise, she chooses APD in
equilibrium. When demand CV is sufficiently high, APD dominates
BPOF and becomes the only strategy adopted in dual financing; in
this case, APD only and dual financing yield the same payoff. It
is intuitive that the retailer would prefer a wholesale price with
BPOF over APD as means of reducing risk when demand variability
rises, because a lower order quantity and late payment with ex-
86 L. Zhao and A. Huchzermeier / Omega 88 (2019) 77–90

Fig. 5. Effect of retailer’s internal capital level on (a) supplier’s delivered quantities; (b) retailer’s profits;
(c) supplier’s profits; and (d) supply chain profits.

ternal financing are both associated with less inventory and credit
risk. However, our numerical example shows that the established
retailer can opt to take more risk via APD—and thus generate a
higher return—when demand risk increases.
Fig. 4(c) illustrates that the supplier’s profits decline with re-
spect to increasing demand volatility. The decline is much less
in APD-only scenario owing to risk sharing between supply chain
members. The supplier’s profits are reduced at the same decreas-
ing rate in the BPOF-only and dual-financing cases (the latter fol-
lows given that, when BPOF can fully mitigate financial distress, it
is the only mechanism adopted under dual financing). Contrary to
Fig. 4(b), the supplier prefers the APD-only to the BPOF-only sce-
nario if the demand coefficient of variation is relatively high (i.e.,
δ (D) ≥ 0.48). Fig. 4(d) indicates the supply chain’s profit is concave
under APD only owing to the channel coordination effect, while
is decreasing w.r.t. demand volatility in the other three scenar-
ios. In contrast to Fig. 4(b), BPOF dominates APD unless the de-
mand CV is fairly high (i.e., δ (D) ≥ 0.44) from the supply chain’s
perspective. Therefore, supply chain members incur a competition
penalty when demand variability falls in the interval δ (D) ∈ [0.22,
0.44) with a probability of 55%. In this interval, the retailer’s op-
timal financing decision deviates from the supply chain’s optimal
choice. Thus there is performance inefficiency due to supplier–
retailer competition in a decentralized setting. For example, when
δ (D ) = 0.4, the competition penalty P = 2.4%; when δ (D ) = 0.3,
P = 8.3%. When δ (D ) = 0.22, the competition penalty is at its
maximum: P = 17.5%. When demand variability falls out of the
given interval, supply chain members’ incentives are aligned thus
channel-wide optimization can be achieved using BPOF as external
financing or APD as internal financing. These results suggest that
a channel coordination mechanism is desirable in single financing
equilibrium.
6.2. Impact of retailer’s internal capital level
In order to examine the role played by the retailer’s internal
capital level (i.e., Ar − Lr ), we vary the expected value of her asset
level Ar from 40,0 0 0 to 80,0 0 0 while keeping her short-term liabil-
ity Lr at a constant level of 30,0 0 0. Fig. 5 shows that the retailer’s
internal capital level has no significant effect except in the APD-
only scenario, because it is the only scenario where the retailer
faces the possibility of financial distress. With regard to the re-
tailer’s internal capital level, the APD-only value is increasing while
values of the other three strategies are relatively stable for supply
chain partners.
Fig. 5(a) demonstrates that the supplier’s delivered quantity is
increasing under APD only but is insensitive to the retailer’s inter-
nal capital level in the other three scenarios. The underlying reason
is that the retailer’s growing internal capital relieves financial dis-
tress and thus allows APD-based exploitation of supply chain co-
ordination; in contrast, the retailer may suffer financial distress if
she employs APD only. Similarly, Fig. 5(b) illustrates that the re-
 L. Zhao and A. Huchzermeier / Omega 88 (2019) 77–90
tailer’s profit is increasing in her internal capital level under APD
only; however, the increase is less significant when the retailer’s
internal capital exceeds a certain level (i.e., Ar − Lr > 31, 231) be-
cause financial distress is mitigated by increasing capital. A com-
parison of APD only and BPOF only (green and blue curves) ver-
ifies Theorem 1 in that BPOF dominates APD in single financ-
ing equilibrium if and only if the retailer’s internal capital is be-
low a certain threshold, i.e., Ar − Lr < ωr − Lr = 23, 218. Hence, the
BPOF-only equilibrium region is decreasing in the retailer’s inter-
nal capital level. Conversely, if the retailer’s internal capital level
is relatively high then APD dominates BPOF and could become
the only strategy adopted in the dual-financing scheme, which
verifies Lemma 2. Thus the retailer’s APD-only payoff approaches
her dual-financing payoff with increasing levels of internal
capital.
Fig. 5(c) indicates that the supplier’s profit is increasing in the
retailer’s internal capital level under APD-only as a result of the ef-
fect plotted in Fig. 5(a). Contrary to the result in Fig. 5(b), the sup-
plier prefers APD to BPOF when the retailer’s internal capital level
is relatively high, i.e., when Ar − Lr ≥ 39, 526. Fig. 5(d) presents the
net effect of the retailer’s internal capital level on supply chain
profits: the channel coordination effect of APD is increasing in the
retailer’s internal capital level. In contrast to Fig. 5(b), APD domi-
nates BPOF when the retailer’s internal capital level exceeds 36,541
from the supply chain’s perspective. Therefore, channel members
incur a competition penalty if the retailer’s internal capital level
is within the interval, i.e., when Ar − Lr ∈ [23, 218, 36, 541) with a
probability of 33.3%. In this interval, the retailer’s optimal financing
choice does not align with the supply chain’s optimal strategy. For
instance, when Ar − Lr = 30, 0 0 0 the competition penalty is P =
8.3%, whereas P = 8.3% (the maximum) when Ar − Lr = 23, 218.
This high cost associated with competition penalty makes the sup-
ply chain inefficient, thus channel coordination is crucial when
pre-shipment finance instruments are employed.
The managerial insights from Figs. 4 and 5 are as follows. First,
greater demand uncertainty increases the need for risk sharing
within the supply chain, while higher level of the retailer’s in-
ternal capital enables financing to supplier by advance payment
discount. As demand risk and the retailer’s internal capital level
increases, transferring monetary flow from voluntary to binding
parties in the supply chain becomes more desirable than raising
capital from a financial institution. Second, both the interval and
magnitude of the competition penalty imply that—even though
pre-shipment finance instruments can mitigate the supplier’s cap-
ital constraint—the incentive alignment between both channel
members is indispensable for reducing the costs of competition
penalty.
7. Conclusions
This paper considers a capital-constrained supply chain with
a SME supplier selling to an established retailer via a wholesale
price contract. The retailer chooses between two pre-shipment fi-
nance instruments—advance payment discount and buyer-backed
purchase order financing—to manage the supplier’s financial dis-
tress. We characterize the optimal supply chain decisions in APD
and BPOF respectively, and show that BPOF enables the supplier
to finance his working capital via a loan from a financial insti-
tution based on the retailer’s credit rating, while APD can bring
channel coordination benefit at the cost of the retailer’s potential
financial distress. When either APD or BPOF can be chosen, the
retailer prefers APD to BPOF if her internal asset level is above a
certain threshold. In dual-financing when both APD and BPOF are
adopted, the retailer chooses APD and will not initiate BPOF unless
87
the marginal cost of financial distress outweighs the benefit of unit
discount. We find that the equilibrium region of APD is increasing
in both the retailer’s internal capital level and demand variability.
Competition penalty can lead to considerable costs when either the
demand variability or the retailer’s internal capital level are within
certain (respective) intervals. Therefore, coordination and collab-
oration between supply chain partners is essential for mitigating
competition penalty.
7.1. Contribution to existing literature
Previous research examines a supply chain of one supplier and
multiple retailers that offer advance payment discount to customers
[28] or advance payment discount provided by a supplier in a sin-
gle firm model [8]; and analyzes the impact of discount and inter-
est rates on payment period and sourcing policies without capital
constraint. In comparison, this paper focuses on advance payment
discount offered by a supplier to a retailer in a capital-constrained
supply chain, and illustrates the effect of budget constraint and de-
mand risk on joint souring and financing decisions.
Existing literature investigates regular and pricing support pur-
chase order financing to a SME supplier in comparison with con-
signed processing and presents the effect of loan interest rate on
the optimal supply chain decisions [33], or assesses the relative
efficiency of buyer direct financing and purchase order financing
without the buyer’s guarantee in a Stackelberg game and shows
the impact of the buyer’s private information, the supplier’s asset
level, and production cost on the single-financing choice [27]. In
contrast, we concentrate on buyer-backed purchase order financing
and its efficiency relative to advance payment discount in single-
financing and dual-financing equilibria, and evaluate the optimal
sourcing and financing policies w.r.t. the retailer’s asset level and
demand variability.
In sum, this paper contributes to the existing literature on ad-
vance payment discount or purchase order financing by (i) explor-
ing how the retailer’s choice between APD and BPOF can miti-
gate the supplier’s capital constraint, (ii) deriving the optimal sup-
ply chain decisions under both single-financing and dual-financing
schemes, and (iii) demonstrating the effect of demand variability
and the retailer’s internal capital level on financing equilibrium
and supply chain efficiency.
7.2. Managerial implications
Buyer-backed purchase order financing creates a “win–win–
win” situation by providing the supplier working capital funding
from a financial institution based on the retailer’s credit rating,
in which the supplier alleviates his budget constraint and thereby
ensures order fulfillment, the retailer strengthens her relationship
with supplier and secures product quantity without incurring fi-
nancial distress, while the financial institution earns interest pay-
ment with lower credit risk (cf. [12,20]). In comparison, advance
payment discount provides channel coordination benefit to both
the supplier and retailer and accelerates the collection of the sup-
plier’s receivables thus enhances his cash conversion cycle at the
potential cost of the retailer’s financial distress. Although both APD
and BPOF can mitigate the supplier’s financial distress before prod-
uct shipment, APD transfers monetary flows from voluntary to
binding positions within the supply chain, while BPOF enables cap-
ital raising from a financial institution and thereby increases the
supply chain’s overall financing capacity.
Advance payment discount features a risk-sharing scheme that
provides a lower wholesale price to the retailer yet also exposes
her to higher inventory risk with larger order quantity and the
88 L. Zhao and A. Huchzermeier / Omega 88 (2019) 77–90

potential cost of financial distress depending on her internal capi- Proof of Proposition 1. A marginal analysis with respect to (w.r.t.)
tal level. Hence, a retailer with ample capital that precludes finan- the supplier’s optimal borrowing level λ∗ gives
cial distress strictly prefers APD, while a less wealthy retailer with
higher credit rating than her supplier should leverage on the inter- 
est rate spread by initiating BPOF with a financial institution (cf. ∂ πs (λ, D, K ) −rwq∗ i f Ls + ck K ≤ As + λwq∗ ,
=
[4,7]). Greater demand variability increases the need for risk shar- ∂λ (1 − α − r )wq∗ i f Ls + ck K > As + λwq∗
ing in the supply chain by APD, and higher level of the retailer’s
internal capital guarantees that the channel coordination benefit
(the case of liquidation leads to zero expected profit and
of APD dominates potential cost of financial distress. The interval
so does not affect the optimal borrowing rate). Here the
and magnitude of the competition penalty highlights the value of
supplier’s probability of continuation is Pr(c ) = (
¯ As ) for
incentive alignment between supply chain partners. 
As = Ls + ck K − λwq, and the supplier’s probability of reor-

7.3. Future research ganization is Pr(r ) = (As ) − (As ). If r < 1 − α and r ≤ rˆ =

A
(w−c p ) min(q, K )−ck q−∫s (1−α )(Ls −As −λwq+ck K )φ (As )dAs
As
Our study could motivate several directions for future research. λ∗ wq , then the supplier
First, our single-currency supply chain model of pre-shipment fi- chooses his optimal borrowing level λ∗ = arg maxλ∈(0,λ̄] πs (λ, w, K )
nance could be extended to a global supply chain setting where so as to cover—as much as possible—his working capi-
exchange rate uncertainty is incorporated and can be managed by tal needs. Otherwise, BPOF is not profitable for the sup-
various hedging mechanisms. Moreover, our finding of high com- plier and is therefore not used (i.e., λ∗ = 0). The retailer’s
petition penalty costs advocates the exploration of coordination profit is πrbpof (q, w) = pE min[D, min(q, K )] − w min(q, K ) −
contracts (e.g., buyback, revenue-sharing, and quantity discount) in As
∫ δ{λwq − γ [(w − c p ) min(q, K ) − ck K − λwqr + As − Ls ]}φ (As )dAs .
the presence of trade finance instrument, see Kouvelis and Zhao As
[14] and Xiao et al. [34]. Third, since we assume that supply chain ∗
In case of liquidation, the supplier’s optimal capacity K bpof = 0.
partners are risk neutral, incorporating risk-aversion in decision
Otherwise, the FOCs of the supplier’s profit,
making could be considered with diverse objective formulations.
In addition, the role of pre-shipment finance can be examined in
   ∗  ∗ 
a competitive setting including supplier completion, buyer compe- pF̄ qbpof 1 − h qbpof = c p + ck in continuation,
tition, and competing supply chains. Fifth, empirical research could bpof ∗
  bpof ∗ 
pF̄ q 1−h q = c p + (2 − α )ck in reorganization,
be conducted to test the relative effectiveness of trade finance in-
struments in mitigating capital constraints and streamlining supply
chains in both developing and developed economies. ck K bpof +Ls −As
establish that if λ̄ ≥ and r ≤ rˆ, the optimal capac-
wqbpof
∗ ∗ ∗
APPENDIX: Proofs ity K bpof = qbpof , the optimal sales quantity is either qbpof
or a, and the optimal wholesale price is wbpof∗ = pF̄ (K bpof∗ ) =
(c p +ck ) Pr(c )+(c p +(2−α )ck ) Pr(r )
1−h (K )
.
Proof of Lemma 1. In a decentralized supply chain, the sup-

plier’s revenue function is defined as R(q ) = (w(q ) − c p )q [15];
the supplier therefore sets q to maximize his profit πsdsc (K ) = Proof of Proposition 2. In case of liquidation, the supplier’s op-
∗
(w(q ) − c p )q − ck q = R(q ) − ck q. The first- and second-order dif- timal capacity K apd = 0. Otherwise, the FOCs of the supplier’s
ferentiations of the revenue function yield, respectively, R (q ) = profit,
w(q )[1 − 1/v(q )] and R (q ) = w (q )[1 − 1/v(q )] + v (q )w(q )/v(q )2 .
The supplier’s first-order condition (4) can be derived by setting  ∗   ∗ 
pF̄ K apd 1 − h K apd = c p + ck in continuation,
marginal revenue R(q) equal to marginal cost ck . Since F(D) has  ∗   ∗ 
support [a, b) for 0 ≤ a < b ≤ ∞ and since w(q ) = pF̄ (q ), it follows pF̄ K apd 1 − h K apd = c p + (2 − α )ck in reorganization.
that w(q) is strictly decreasing for q ∈ [a, b). In addition, v (q) ≤ 0
for q ∈ [a, b) and so the revenue function is strictly concave for
q ∈ [a, qˆ]; this result is derived as follows. Here the supplier’s probability of continuation is Pr(c ) =

Because R (q ) = p − c p for q ∈ [0, a), profit is linear and strictly (
¯ As ) for As = Ls + ck K − w(1 − d )q, the probability of liq-
increasing. For q ∈ [a, qˆ ) we know that F̄ (q ) is strictly decreasing 
uidation is Pr(l ) = (As ) for As = Ls + ck K − w ( 1 − d )q −
and also that (1 − h(q ) ) is positive (by construction) and weakly [w(1−d )−c p ] min(q,K )−ck K
, and the probability of reorganization
(1−α )
decreasing (by the IGFR assumption). Hence R(q) is strictly de-
is Pr(r ) = (As ) − (As ). Therefore, the optimal wholesale
creasing, which makes revenues and profits strictly concave. For ∗ (c +c ) Pr(c )+(c +(2−α )c ) Pr(r )
q > qˆ, marginal revenue R (q) is negative and costs are increas- price is wapd = p k (1−h(K p))(1−d ) k . If APD is suf-
∗
ing; hence profits must fall. The uniqueness of qdsc follows from ficient to cover the supplier’s working capital needs, i.e.,
∗ ∗

its strict concavity over [a, qˆ]. Solving R (q ) = 0 establishes that w(1 − d )q ≥ ck K ∗ + Ls − As , the optimal capacity K apd = qapd =
qˆ must maximize the supplier’s revenue. Hence the optimal order F ( p ). In this case, the retailer’s optimal order quantity
−1 p−we
∗
quantity qdsc ∈ [a, qˆ], and the supplier will choose to sell a only ∂π apd (q,D,K )
satisfies the condition r
∂q = 0 if and only if pF̄ (qapd ) =
if a > 0 and limq→a+ pF̄ (q )[1 − h(q )] < ck + c p . Since v(q ) = 1/h(q )
w(1 − d )[1 + (1 − α ) 1Lr +w(1−d )qapd >Ar ]. Here the effective whole-
and since F is IGFR, it follows that v(D) must be decreasing. There-
fore, the optimal wholesale price is wdsc∗ = pF̄ (qdsc∗ ) =
ck +c p
= sale price is we = w(1 − d ) [1 + (1 − α ) 1Lr +w(1−d )qapd >Ar ]. Hence
1−h (q
dsc∗ ) ∗ ∗
v(qdsc∗ )(ck +c p ) qapd ≥ qdsc if (1 − d )[1 + (1 − α ) 1Lr +w(1−d )qapd >Ar ] ≤ 1 and
. ∗ ∗
v(qdsc∗ )−1 qapd < qdsc otherwise. 
 L. Zhao and A. Huchzermeier / Omega 88 (2019) 77–90 89

bpof∗ apd∗
Proof of Theorem 1. The retailer chooses the financing strategy that leads to higher expected profit: πr > πr if and only
A
if pEmin(qbpof∗ , D ) − wqbpof∗ − Ass δ{λwq − γ [(w − c p − ck )K bpof∗ − λwqr + As − Ls ]}φ (As )dAs > pEmin(qapd∗ , D ) − w(1 − d )qapd∗ −
(1 − α )[Lr + w(1 − d )qapd∗ − Ar ]+ , from which the threshold value of the retailer’s internal asset is ωr = Lr + w(1 − d )qapd∗ −

A
p Emin(qapd∗ , D )−w(1−d )qapd∗ −p Emin(qbpof∗ , D )+wqbpof∗ +∫As δ{λwq−γ [ (w−c p −ck )K bpof∗ −λwqr+As −Ls ]}φ (As )dAs
s
1−α .

 ∗ df ∗
Proof of Theorem 2. Under dual financing, the effective wholesale price is wdf = (1 − β )w(1 − d ) + β w. Hence qdf = F −1 ( p−wp )∈
[F −1 ( p−w −1 ( p−w(1−d ) )] provided that wdf ∈ [w (1 − d ), w]. That is, the order quantity in dual financing lies between the respective
p ), F p
quantities in the decentralized benchmark and APD-only cases.
⎧ In this scenario, the supplier’s profit along with his capacity and financing decisions is πsdf (K, w) =
+ df , K ) − c K − λwq r in continuation,
⎪ w ( 1 − d ) min ( q , K ) + w min[ ( K − q ) , q ] − c min ( q
⎨ 1 1
+
2 p k 2
w(1 − d ) min(q1 , K ) + w min[(K − q1 ) , q2 ] − c p min(qdf , K ) − ck K − λwq2 r
⎩−(1 − α )(Ls + ck K − As − wq1 (1 − d ) − λwq2 ) in reorganization,
⎪
0 in liquidation.

Here the supplier’s probability of continuation is Pr(c ) = (
¯ As ) for As = Ls + ck K − wq1 (1 − d ) − λwq2 , the probabil-
 w(1−d ) min(q1 ,K )+w min[ (K−q1 )+ ,q2 ]−c p min(qdf ,K )−ck K−λwq2 r
ity of liquidation is Pr(l ) = (As ) for As = Ls + ck K − wq1 (1 − d ) − λwq2 − 1−α ,
and the probability of reorganization is Pr(r ) = (As ) − (As ). Therefore, the supplier’s break-even BPOF interest rate un-
+ As 
 w(1−d ) min(q1 ,K )+w min[(K−q1 ) ,q2 ]−c p min(q ,K )−ck K−∫As (1−α )(Ls +ck K−As −wq1 (1−d )−λwq2 )φ (As )dAs
df
der dual financing is r˜ = λw q 2 . The retailer’s ex-
pected πrdf (q, w ) = pE min[D, min(qdf , K )] − w(1 − d ) min(q1 , K ) − w min[(K − q1 )+ , q2 ] − (1 − α )[Lr + w(1 − d )q1 − Ar ]+ −
profit is
 As
As δ{λwq2 − γ [w(1 − d ) min(q1 , K ) + w min( (K − q1 )+ , q2 ) − c p min(qdf , K ) − ck K − λwq2 r + As − Ls]}φ (As )dAs . The financial institution’s
 As
expected profit is πc = (
¯ As )λwq2 r + (1 − δ ){γ [w(1 − d ) min(q1 , K ) + w min( (K − q1 )+ , q2 ) − c p min(qdf , K ) − ck K − λwq2 r + As − Ls ] −
As
λwq2 }φ (As )dAs . Since the BPOF lending market is assumed as perfectly competitive, πc = 0.
∗
In case of liquidation, the supplier’s optimal capacity K df = 0. Otherwise, the FOCs of the supplier’s profit,

 
∗  ∗ 
pF̄ K df 1 − h K df = c p + ck in continuation,
 ∗   ∗ 
pF̄ K df 1 − h K df = c p + (2 − α )ck in reorganization,

show that if dual financing is sufficient to fulfill the supplier’s working capital needs, i.e., w(1 − d )q1 + λwq2 ≥ ck K ∗ +
∗ ∗ df
Ls − As , the optimal capacity K df = qdf = F −1 ( p−w
p ), then the retailer’s expected profit is πrdf (q, w ) = pEmin(D, qdf ) −
∞  qdf
wβ qdf − w(1 − d )(1 − β )qdf − (1 − α )[Lr + w(1 − d )(1 − β )qdf − Ar ]+ = qdf
( p − (1 − d + β d )w)qdf f (D )dD − 0
(1 − d + β d )
w(qdf − D ) f (D ) dD − (1 − α )[Lr + w(1 − d )(1 − β )qdf − Ar ]+ . Therefore, if the retailer experiences financial distress then
the FOC is: ( p − (1 − d + β d )w )(1 − F (qdf )) − F (qdf )(1 − d + β d )w − (1 − α )w(1 − d )(1 − β ) = 0 if and only if pF (qdf ) = p −
w(1 − d )(2 − α − β + αβ ) − β wd. When the retailer is free from financial distress, the FOC is simply pF (qdf ) = p − w(1 − d ) − β wd.
Recall that the retailer’s partial purchase via APD can not cover the supplier’s working capital shortfall, i.e., w(1 − d )q∗1 < ck K ∗ + Ls − As ,
thus the supplier employs BPOF in addition to APD as determined by the following marginal analysis w.r.t. λ:


∂ πs (λ, w, K ) (1 − α − r )wq∗2 i f Ls + ck K > As + wq∗1 (1 − d ) + λwq∗2 ,
=
∂λ −rwq∗2 i f Ls + ck K ≤ As + wq∗1 (1 − d ) + λwq∗2 .

In this case, if r < 1 − α and r ≤ r˜ then the supplier chooses his optimal borrowing level λ∗ = arg max πs (λ, w, K ). Otherwise, the
λ∈(0,λ̄]
supplier does not use BPOF because it is not profitable (λ∗ = 0). 

Proof of Lemma 2. The FOCs of retailer profit can be rewritten as

 ∗ 
pF qdf = p − w(1 − d ) − β ∗ wd in continuation,
 ∗
pF qdf = p − (2 − α )w(1 − d ) + (1 − α − 2d + α d )wβ ∗ in reorganization.

In the case of continuation, we can readily derive that qdf is decreasing in β (i.e., in the portion of order quantity paid at wholesale
price after product delivery). In the case of reorganization, we can establish that: qdf is increasing in β if and only if 1 − α − 2d + α d >
0 ⇔ 1 − α > 1−dd
.
90 L. Zhao and A. Huchzermeier / Omega 88 (2019) 77–90

References
2010. January 30, 2010. URL http://www.nytimes.com/2010/01/31/business/
smallbusiness/31order.html Retrieved on December 7, 2016.
[1] Babich V. Independence of capacity ordering and financial subsidies to risky
[20] Navas-Alemán L, Pietrobelli C, Kamiya M. Inter-firm linkages and finance in
suppliers. Manuf Serv Oper Manag 2010;12:583–607.
value chains. Working paper, Inter-American Development Bank; 2012.
[2] Brennan M, Maksimovic V, Zechner J. Vendor financing. J Financ
[21] Nussel P, Sherefkin R. Plastech bankruptcy could shut all of Chrysler. Auto-
1988;43:1127–41.
motive News; 2008. URL http://www.autonews.com/apps/pbcs.dll/article?AID=
[3] Cai G, Chen X, Xiao Z. The roles of bank and trade credits: theoretical analysis
/20080204/OEM/302049890 Retrieved on October 2, 2016.
and empirical evidence. Prod Oper Manag 2014;23:583–93.
[22] Perakis G, Roels G. The price of anarchy in supply chains: quantifying the effi-
[4] Chauffour JP, Malouche M. Trade finance during the great trade collapse.
ciency of price-only contracts. Manag Sci 2007;53:1249–68.
Washington, D.C.: World Bank; 2011.
[23] PurchaseOrderFinancing.com. 2018. PO Financing rates made affordable. URL:
[5] Farris MT II, Hutchison PD. Cash-to-cash: the new supply chain management
https://www.purchaseorderfinancing.com Retrieved on March 14, 2018.
metric. Int J Phys Distrib Logist Manag 2002;32:288–98.
[24] Seifert D, Seifert RW, Protopappa-Sieke M. A review of trade credit literature:
[6] Gamba A, Triantis AJ. Corporate risk management: integrating liquidity, hedg-
opportunities for research in operations. Eur J Oper Res 2013;231:245–56.
ing, and operating policies. Manag Sci 2014;60:246–64.
[25] Sodhi M, Tang C. Tactical approaches for mitigating supply chain risks: Finan-
[7] Gold M, Jacobs G. Analysis of the potential for development of SME buyer-
cial and operational hedging. In: Sodhi M, Tang C, editors. Managing supply
backed purchase order financing products. United States Agency for Interna-
chain risk. Springer; 2012. p. 109–33.
tional Development; 2007. URL http://pdf.usaid.gov/pdf_docs/Pnadm257.pdf
[26] Spengler J. Vertical integration and antitrust policy. J Polit Econ
Retrieved on September 24, 2013.
1950;58:347–52.
[8] Gupta RK, Bhunia AK, Goyal SK. An application of Genetic Algorithm in solving
[27] Tang CS, Wu J, Yang SA. Sourcing from suppliers with financial constraints and
an inventory model with advance payment and interval valued inventory costs.
performance risk. Manuf Serv Oper Manag 2018;20:70–84.
Math Comput Model 2009;49:893–905.
[28] Thangam A. Optimal price discounting and lot-sizing policies for perishable
[9] Hortaçsu A, Matvos G, Shin C, Syverson C, Venkataraman S. Is an au-
items in a supply chain under advance payment scheme and two-echelon
tomaker’s road to bankruptcy paved with customers’ beliefs. Am Econ Rev
trade credits. Int J Prod Econ 2012;139:459–72.
2011;101:93–7.
[29] Tice, C. 2010. Can a purchase order loan keep your business growing? June 17,
[10] Hwang I, Radhakrishnan S, Su L. Vendor certification and appraisal: implica-
2010. URL: http://www.entrepreneur.com/article/207058. Retrieved on Decem-
tions for supplier quality. Manag Sci 2006;52:1472–82.
ber 8, 2013.
[11] Jing B, Chen X, Cai G. Equilibrium financing in a distribution channel with cap-
[30] United States Agency for International Development [USAID]. Case studies in
ital constraint. Prod Oper Manag 2012;21:1090–101.
buyer-backed purchase order financing. Crimson Capital Corp; 2011.
[12] Klapper L. The role of factoring for financing small and medium enterprises. J
[31] United States Agency for International Development [USAID]. Launching buy-
Bank Financ 2006;30:3111–30.
er-backed purchase order financing in east Africa. Crimson Capital Corp; 2011.
[13] Kouvelis P, Zhao W. Financing the newsvendor: supplier vs. bank, and the
[32] K van der Vliet, Reindorp MJ, Fransoo JC. The price of reverse factoring: financ-
structure of optimal trade credit contracts. Oper Res 2012;60:566–80.
ing rates vs. payment delays. Eur J Oper Res 2015;242:842–53.
[14] Kouvelis P, Zhao WH. Supply chain contract design under financial constraints
[33] Wu A. A supportive pricing model for suppliers with purchase order financing.
and bankruptcy costs. Manag Sci 2016;62:2341–57.
Int J Prod Res 2017;55:5632–46.
[15] Lariviere MA, Porteus EL. Selling to the newsvendor: an analysis of price-only
[34] Xiao S, Sethi SP, Liu M, Ma S. Coordinating contracts for a financially con-
contracts. Manuf Serv Oper Manag 2001;3:293–305.
strained supply chain. Omega 2017;72:71–86.
[16] Leland HE. Corporate debt value, bond covenants, and optimal capital struc-
[35] Yang SA, Birge JR, Parker RP. The supply chain effects of bankruptcy. Manag Sci
ture. J Financ 1994;49:1213–52.
2015;61:2320–38.
[17] Lekkakos SD, Serrano A. Supply chain finance for small and medium sized
[36] Zhao L, Huchzermeier A. Managing supplier financial risk with pre-shipment
enterprises: the case of reverse factoring. Int J Phys Distrib Logist Manag
finance instruments. In: Zhao L, Huchzermeier A, editors. Supply chain finance
2016;46:367–92.
– integrating operations and finance in global supply chains. Springer; 2018.
[18] Marr K. Auto suppliers seek help in wake of Chrysler, GM woes. Washing-
p. 121–42.
ton Post; 2009. May 14, 2009. URL http://www.washingtonpost.com/wp-dyn/
[37] Zhong Y, Shu J, Xie W, Y Zhou. Optimal trade credit and replenishment policies
content/article/20 09/05/13/AR20 09051303382.html Retrieved on October 10,
for supply chain network design. Omega 2018;81:26–37.
2013.
[19] Martin A. The places they go when banks say no. The New York Times;