Technological Forecasting & Social Change xxx (xxxx) xxx–xxx

Contents lists available at ScienceDirect

Technological Forecasting & Social Change

journal homepage: www.elsevier.com/locate/techfore

A multiple objective stochastic programming model for working capital

management
Hatem Masri⁎, Yomna Abdulla
College of Business Administration, University of Bahrain, P.O. Box 32038, Sakhir, Kingdom of Bahrain

A R T I C L E I N F O A B S T R A C T

Keywords: The level of working capital is one of the main financial decisions affecting a firm's value, shareholder wealth,
Working capital competitiveness, liquidity, and profitability. An aggressive (conservative) working capital policy has a positive
Stochastic goal programming (negative) impact on profitability but a negative (positive) impact on liquidity. In this paper, we propose a
Multiple objective stochastic programming multiple objective stochastic programming model to select an efficient working capital strategy that takes into
Regression
consideration not only the conflicting impact of working capital policy on the two firm objectives of liquidity and
Retailing industry
profitability but also the interrelationships and stochastic aspects of the components of working capital. The
proposed model was used to help a start-up retailer to determine its optimal working capital.

1. Introduction aggressive (conservative) working capital policy has a positive (nega-

Working capital concerns managing the day-to-day short-term
operations of a firm. The three main components of working capital
are accounts payable, accounts receivable and inventory. The impor-
tance of working capital is reflected by the heavy reliance of worldwide
firms on it. For instance, Summers and Wilson (2000) show that 80% of
daily trade transactions in the UK are on credit terms. Deloof (2003)
finds that in Belgium, accounts receivable, inventory, and accounts
payable represent 17%, 10%, and 13%, respectively, of total assets of
Belgian firms. Furthermore, efficient working capital management
protects firms from potential financial problems. During the recent
financial crisis of 2007–2008, the liquidity shortage highlighted the
significance of short-term financial policies such as working capital
management. In a related vein, Braun and Larrain (2005) and Abuzayed
(2012) argue that more attention be devoted to the level of working
capital during recession and crises periods. Kaddumi and Ramadan
(2012) note that U.S. finance executives became more cautious in
managing firms' working capital in the aftermath of the 2008 global
recession.
Working capital affects shareholder's wealth, firm value, competi-
tiveness, liquidity, and profitability (Smith, 1980; Shin and Soenen,
1998; Lazaridis and Tryfonidis, 2006; Baños-Caballero et al., 2014;
Aktas et al., 2015); therefore, firms usually aim to have a well-managed
working capital, as any change in the level of the working capital can be
critical. Padachi (2006) argues that managers face a trade-off between
liquidity and profitability as they seek to maximize the firm's value. The
level of working capital is directly linked to this trade-off as an
tive) impact on profitability but a negative (positive) impact on
liquidity and risk. If a firm overlooks its profitability, it cannot survive
in the long term, but on the other hand, if it overlooks liquidity, it may
face the problem of insolvency (Vahid et al., 2012). Almazari (2013)
notes that Saudi cement firms face a trade-off problem between
liquidity and profitability and recommends that a more efficient
working capital structure be developed. Hence, a firm needs to operate
with an optimal level of working capital that maximizes the two
objectives of liquidity and profitability.
The goal of this paper is to develop a model to determine the
optimal working capital structure for a start-up retailer aiming to enter
the Kingdom Bahrain market. The importance of working capital
management increases for firms in emerging markets due to their
limited external sources of finance and their heavy reliance on trade
credit. Inefficient working capital is a cause of failure in small firms and
start-ups (Lazaridis and Tryfonidis, 2006). The economic outlook
suggests that the retail industry is one of Bahrain's most promising
industries. For instance, according to the Bahrain 2014 Commercial
Market Outlook (Cluttons, 2014), Bahrain's retail sector remains one of
its best performing sectors due to strong demand. Furthermore, the BMI
Bahrain Retail 2016 report (BMI Research, 2016) forecasts that house-
hold spending in Bahrain will increase by approximately 7% over the
next five years, driven by rising incomes, tourists and positive demo-
graphic trends.
The start-up retailer seeks an efficient working capital strategy that
will have a positive impact on its profitability and liquidity, given some
specific constraints on its working capital components. The start-up

⁎
Corresponding author.
E-mail addresses: hmasri@uob.edu.bh (H. Masri), yahmed@uob.edu.bh (Y. Abdulla).

http://dx.doi.org/10.1016/j.techfore.2017.05.006
Received 30 January 2017; Received in revised form 6 April 2017; Accepted 3 May 2017
0040-1625/ © 2017 Elsevier Inc. All rights reserved.

Please cite this article as: Masri, H., Technological Forecasting & Social Change (2017), http://dx.doi.org/10.1016/j.techfore.2017.05.006
H. Masri, Y. Abdulla
retailer's total assets are constrained at BD 30 million and its debt level
at BD 7.5 million. These constraints reflect that the start-up retailer is
financially constrained and faces a high chance of failure. The start-up
retailer also specifies that its net trade cycle is 100 days, because a
longer net trade cycle might lead to liquidity problems. The main
concern of the start-up retailer is the conflicting impact of working
capital policy on profitability and liquidity.
To select an efficient working capital structure, we propose a
multiple objective stochastic programming model that determines an
optimal level of accounts receivable, accounts payable and inventory
that maximizes a firm's profitability and liquidity, taking into account
the stochastic nature of sales.
Our empirical study results show that the higher the accounts
payable and inventory levels, the higher the firm's liquidity level, while
the higher the accounts receivable level, the lower the firm's liquidity
level. With respect to profitability, the accounts payable and receivable
have a positive impact on the firm's profitability, whereas inventory has
a negative impact on the firm's profitability.
The proposed solution strategy for the start-up retailer provides a
link between two bodies of literature, working capital and operations
research, by utilizing the tools of the latter to solve a prominent
problem in the former. Prior working capital literature has highlighted
the existence of the problem of determining the optimal level of
working capital, but, as far as we are aware, we are among the first
to propose a comprehensive solution approach to this problem.
Our model differs from other models presented in the literature
(Keown and Martin, 1977; De et al., 1982; Zhang et al., 2015), as we
focus on the structure of the working capital rather than the return and
risk from such an investment. The findings of our paper also contribute
to the strand of literature on the effect of working capital on profit-
ability (e.g., Deloof, 2003; García-Teruel and Martinez-Solano, 2007) as
well as to the works on working capital and liquidity (e.g., Kim et al.,
1998). Our paper is also related to the works on the importance of
working capital management in start-ups, such as Huyghebaert and Van
de Gucht (2007). The empirical results highlight the determinants of
profitability and liquidity of retailers in Bahrain.
The remainder of the paper is organized as follows. In Section 2, we
provide a theoretical background on working capital. In Section 3, we
discuss the multiple objective stochastic programming. In Section 4, we
propose the certainty equivalent. In Section 5, we present the empirical
study, and finally we conclude the paper in Section 6.
2. Working capital management
Working capital management is important to firms of all sizes
because it grants firms financial flexibility and reduces their depen-
dence on external sources of finance (Autukaite and Molay, 2011). This
importance increases for start-ups and small firms because most of their
assets are current and they are highly reliant on current liabilities
(Petersen and Rajan, 1997; Afrifa and Padachi, 2016). The level of
aggressiveness or conservatism of working capital policies varies across
industries (Weinraub and Visscher, 1998). Retailers rely heavily on
purchasing on credit because of the nature of their goods and their
frequent interaction with suppliers (Giannetti et al., 2011). The number
of days of inventory is shorter than average in the retail industry due to
the speedy selling of goods (Koumanakos, 2008). Therefore, determin-
ing the optimal level of working capital is of great importance to start-
up retailers.
There are several reasons firms use the three working capital
components accounts receivable, accounts payable and inventory.
Firms grant accounts receivable because it smoothes the demand for
their products: firms may defer the collection of payment or extend
trade receivables during periods of low demand. Fabbri and Klapper
(2016) argue that accounts receivable is a less aggressive and more
flexible marketing instrument than price reduction, which can cause
price wars and reactions from competitors. Furthermore, firms provide
2
Technological Forecasting & Social Change xxx (xxxx) xxx–xxx
accounts receivable because they have a lending advantage over banks
in extending credit, mainly due to their frequent interactions with
customers (Petersen and Rajan, 1997).
Firms also use accounts payable for several reasons. Accounts
payable is a source of finance for firms during liquidity shocks or crisis
periods because, in the case of default, suppliers are willing to grant
firms renegotiation concessions (Wilner, 2000). Accounts payable also
reduces transaction costs by separating payment from delivery, hence
eliminating the need to hold inventories of both money and goods
(Ferris, 1981).
Similarly, inventory management is another important aspect of
working capital that helps firms avoid holding too much or too little
inventory. Holding inventory helps to increase sales as it provides
flexibility, enhances the ability to address demand swings and mitigates
manufacturing disruptions. At the same time, a high level of inventory
increases warehousing and storage costs, could cause spoilage and
damage, and could be a shield for inefficient management and poor
forecasting (Koumanakos, 2008). Several inventory management tech-
niques to aid firms in inventory decisions have been developed, such as
Material Requirements Planning (MRP) systems and Just-In-Time (JIT).
Overall, firms need to determine an optimal level of working capital
that maximizes their value. For instance, having a high level of
inventory and accounts receivable may lead to higher sales and profits.
However, locking up funds in inventory and accounts receivable may
affect a firm's liquidity and its ability to meet its day-to-day operational
needs and obligations (Eljelly, 2004).
Several studies have recognized that the level of working capital
affects both the profitability and liquidity of a firm; however, little
attention has been devoted to the idea that working capital could affect
both profitability and liquidity at the same time and that its effect could
be positive on one objective and negative on the other. Therefore, our
paper aims to fill this gap by determining the optimal level of working
capital of a start-up retailer that maximizes both liquidity and profit-
ability.
Our model also takes into account the constraints faced by the start-
up retailer. These constraints result from the fact that the start-up
retailer has no relationship history with suppliers and faces difficulties
in accessing external sources of finance. As a result, the constraints
concern the total assets, the working capital structure and the debt
level.
3. A multiple objective stochastic programming for working
capital
The main objective of the start-up retailer is to maximize its
profitability and liquidity. Eljelly (2004) and Rehman et al. (2015)
document the opposing relationship between profitability and liquidity
in Saudi listed firms. These two objectives are of great importance to
the start-up retailer because it faces limited external sources of finance
and a high possibility of failure. Therefore, it would like to pay extra
attention to the management of its internal sources of finance, such as
accounts payable, as well as maximize its benefits from accounts
receivable and inventory.
To determine the optimal level of working capital, we consider the
following three decision variables related to the components of working
capital: accounts payable (x1) is money owed by a firm to its suppliers,
accounts receivable (x2) is money owed by customers, and inventory
(x3) is goods and materials used for resale.
In addition, we consider the cash and cash equivalents (x4) along
with the fixed assets (x5) for two reasons. First, these two variables are
considered to be substitute working capital, so a low level of working
capital frees up funds to be invested in fixed assets or to be freely held
as cash (Fazzari and Petersen, 1993; Giannetti et al., 2011). Second,
both asset types are significant components of total assets, whose value
is limited for the start-up retailer. In short, the optimal level of working
capital cannot be determined without also determining these two
H. Masri, Y. Abdulla Technological Forecasting & Social Change xxx (xxxx) xxx–xxx

variables due to the interrelationship between these two variables and regression and a20 is the intercept and a2i are the slopes of the
working capital. We also include the long-term debt (x6) as a variable profitability regression.
due to the limited financing sources of the start-up retailer In short, the resulting program that maximizes the start-up retailer's
(Huyghebaert and Van de Gucht, 2007). liquidity and profitability, subject to the predefined constraints, is as
The start-up retailer has specific restrictions that reflect some of its follows:
financial constraints. The first constraint stems from the fact that the 6
start-up retailer cannot allocate more than TA = BD 30 million to its Max a1 + ∑ a1i xi
total assets: i =1
6
x2 + x3 + x4 + x5 = TA (1) Max a2 + ∑ a2i xi
i =1
The second constraint concerns the high chances of bankruptcy for NTC
s. t. −x1 + x2 + x3 ≤ 365
S
the start-up retailer, who consequently would prefer to limit the debt
x1 + x6 ≤ D
level to D = BD 7.5 million:
5
x1 + x6 ≤ D (2) ∑ xi = TA
1=2
The third constraint concerns the start-up retailer's aim to keep its li ≤ xi ≤ ui , i = 1, …, 6 (7)
working capital structure at the same level as that of its competitors. Let
us denote the level of sales by S. We assume that the level of sales is The program (Eq. (7)) is a multiple objective stochastic program. A
− solution strategy for a multiple objective stochastic program should be
random and follows a normal probability distribution with a mean S
and a standard deviation σS . Given this level of sales, a long Net Trade
2 based on two transformations, a multiple objective transformation and
Cycle (NTC) is considered to be a cost and could increase the chances of a stochastic transformation, that will lead to a certainty equivalent
bankruptcy (Baños-Caballero et al., 2014). A short trade cycle indicates program (Ben Abdelaziz, 2012). In the multiple objective transforma-
that the company takes a short time to sell inventory and collect tion, the program is reduced to a uni-objective model and in the
receivables and a long time to pay its suppliers. The industry level of stochastic transformation, the model is transformed into a deterministic
working capital is written in terms of the net trade cycle divided by 365 equivalent. In the next section, we propose to build a certainty
as follows: equivalent program for the multiple objective stochastic program (Eq.
(7)) based on additional hypotheses appropriately defined by the start-
NTC ∼ up retailer.
− x1 + x2 + x3 ≤ S
365 (3)
Finally, the start-up retailer has identified lower and upper values 4. The certainty equivalent program
for the problem decision variables as follows:
li ≤ x i ≤ u i (4) To build a certainty equivalent for the program (Eq. (7)), we start
with a stochastic transformation of the problem. A stochastic transfor-
where li and ui are the upper and lower limits on the decision variable mation for a multiple objective stochastic program can be based on one
xi. of the stochastic programming solution approaches, namely, the chance
Under the above constraints, the start-up retailer would like to constrained approach (Charnes and Cooper, 1963), the recourse
maximize the liquidity and the profitability of the company. The approach (Dantzig and Madansky, 1961), and the chance constrained
relationships of the problem decision variables (x1 , … , x6) with profit- recourse approach (Masmoudi and Abdelaziz, 2015). Each approach
ability and liquidity are not well defined in the literature. García-Teruel has predefined hypotheses that we need to validate before using the
and Martinez-Solano (2007) find a negative relationship between approach.
profitability and accounts payable (x1) and accounts receivable (x2) The stochastic constraint (Eq. (3)) defines the optimum structure of
in Spanish small and medium firms. The argument supporting this NTC ∼
the working capital to not exceed the random value ( 365 S ). If the
result is that low levels of accounts payable and accounts receivable working capital exceeds that threshold, then the start-up retailer will
reduce warehouse costs and the chance of bankruptcy, thus helping to face a greater liquidity risk that can be quantified as a financing
increase profitability. On the other hand, Baños-Caballero et al. (2012) recourse cost negatively affecting the liquidity of the start-up retailer
argue that a low level of working capital may result in the loss of sales, (Eljelly, 2004). We propose a recourse approach to address the
disruption in production and decreased ability to stimulate product stochastic constraint (Eq. (3)) where the excess in the value of the
demand, thus resulting in low profitability. Abuzayed (2012) shows a working capital will generate a recourse cost in the liquidity objective
positive relationship between profitability and the level of working function.
capital in Jordanian firms as the Jordanian market does not penalize To apply a recourse approach, let us first approximate the level of
profitable firms for their inefficient working capital management. Kim ∼
sales S by a discrete distribution and denote by wn, n = 1 , … , N the
et al. (1998) find that a conservative working capital policy has a possible states of nature and pn the probability of occurrence of the level
positive impact on liquidity as it avoids locking the funds in working of sales S(wn) , n = 1 , … , N. Under a recourse approach, the constraint
capital components. On the other hand, Lyroudi and Lazaridis (2000) (3) can be rewritten as follows:
observe a positive relationship between working capital and liquidity
NTC
for food firms in Greece. −x1 + x2 + x3 − y+ (wn ) + y− (wn ) = S (wn ), n = 1, …, N
365 (8)
To find the most appropriate relationship between the problem
decision variables and the start-up retailer's profitability and liquidity, + −
where y (wn) and y (wn) are the recourse variables and represent,
we propose in this paper to conduct a linear regression using Bahraini respectively, the excess and the shortage in the value of the working
retailers' data. This regression will define the impact of the decision capital when the scenario n occurs. For a given scenario n, the excess
variables on the start-up's liquidity and profitability: liquidity y+(wn) represents the opportunity loss for the start-up retailer
6 to invest that amount for a very short period at an interest rate q(wn).
Liquidity = a10 + ∑i =1 a1i xi (5) The expected opportunity loss can be defined as follows:
6 N
Profitability = a20 + ∑i =1 a2i xi (6) Q (x , w ) = ∑n =1 pn q (wn ) y+ (wn ) (9)

where a10 is the intercept and a1i are the slopes of the liquidity We propose to add the recourse cost (9) to the liquidity objective

3
H. Masri, Y. Abdulla Technological Forecasting & Social Change xxx (xxxx) xxx–xxx

function as follows: market. To achieve this step, we used the financial data of 10 Bahraini
6 N
retailers over the past 10 years. The data were made available to us by
∑i =1 a1i xi − ∑n =1 pn q (wn ) y+ (wn ) (10) the start-up retailer (with restrictions on publication) and the regression
results are summarized in Table 2.
The multiple objective transformation aims to reduce the number of The regression results for the liquidity objective function show that
objective functions in the program (Eq. (7).) In the literature, multiple accounts payable has a positive impact on liquidity, possibly because
objective functions have been dealt with using several approaches, accounts payable is a short-term liability that is used to finance
including the stochastic goal programming approach (Contini, 1968), liquidity. The negative relationship between accounts receivable and
where a stochastic target value is identified for the objective functions. liquidity suggests that firms should not lock a high level of funds in
In our context, the start-up retailer aims to guarantee, up to some high accounts receivable to remain liquid. Inventory has a positive relation-
level of probability, a certain profit margin π from the company's sales. ship with liquidity as inventory consists of short-term liquid assets that
Therefore, the target for the profitability objective function is equal to could easily be sold and converted into cash. Cash and cash equivalents
∼
πS . The profitability target is stochastic and cannot be achieved for all also have a positive impact on liquidity as cash is one of the most
scenarios, and the start-up retailer agrees to achieve this target with a important sources of liquidity and is used to hedge against uncertain
probability level α. A chance constrained approach is suitable for liquidity demands. On the other hand, firms with a high level of fixed
modeling the company's profitability target Eq. (10) as follows: assets are less liquid, as converting fixed assets into cash is difficult and
hence a high level of fixed assets negatively affects liquidity. Long-term
Pr (∑ 6
i =1
∼
)
a2i xi ≥ π S ≥ α
(11) debt has a positive impact on liquidity, as it is a source of finance that
could enhance liquidity. The findings of the liquidity regression are
Based on a stochastic goal programming approach and a chance consistent with the results of Kim et al. (1998) and Gao et al. (2013).
constrained approach, the profitability objective function (Eq. (6)) is The regression results for the profitability objective function
transformed into the constraint (Eq. (11)). The resulting certainty indicate that accounts payable positively impacts profitability because
equivalent program to the multiple objective stochastic program (Eq. delaying the payment to suppliers can sometimes be an inexpensive and
(7)) is as follows: a flexible source of finance. Accounts receivable has a positive effect on
6 N profitability, possibly because granting accounts receivable stimulates
Max a1 + ∑ a1i xi − ∑ pn q (wn ) y+ (wn ) sales and hence increases profits. The negative relationship between
i =1 n =1
inventory and profitability indicates that quick turnover increases
⎛ 6
∼⎞
s. t Pr ⎜a2 + ∑ a2i xi ≥ πS ⎟ ≥ α profitability. Additionally, the results indicate that the higher the level
⎝ i =1 ⎠ of cash holdings and fixed assets, the higher the level of profitability.
NTC
− x1 + x2 + x3 − y+ (wn ) + y− (wn ) = S (wn ), n = 1, …, N Finally, the negative relationship between long-term debt and profit-
365
x1 + x6 ≤ D ability suggests that higher levels of debt will cause higher borrowing
5 costs and could increase firms' chances of bankruptcy, hence decreasing
∑ xi = TA profitability. The findings of the profitability regression are in line with
i =2
the findings of Deloof (2003), García-Teruel and Martinez-Solano
li ≤ xi ≤ ui , i = 1, …, 6
(2007), and Baños-Caballero et al. (2012).
(12) The regression results reflect the opposing effect of some of the
The program (Eq. (12)) is a deterministic program with a chance working capital components on both the liquidity and the profitability
constraint (Eq. (11)). In the empirical study, we propose to linearize the objective functions; these conflicting objective functions make the start-
chance constraint and solve the resulting linear program using the data up retailer problem difficult to solve.
from the Bahraini retail market. To solve the certainty equivalent (Eq. (12)), we propose to linearize
the chance constraint (Eq. (11)) and then solve the resulting linear
program using a commercial solver.
5. The empirical study
Under the hypothesis that security returns are normally distributed
and by following Masri and Ben Abdelaziz (2010) linearization process
To find the optimal level of working capital, we use data for the
of chance constraints, the chance constraints (11) can be rewritten as
start-up retailer and its competitors in the Bahrain market. The data
follows:
reported in this section were slightly modified to protect confidential
information regarding the start-up retailer and its competitors. 6

Additionally, for the sake of simplicity, we limit our presentation to ∑ a2i xi ≥ π (S + ϕ−1 (α ) σS ) − a2
i =1
three equiprobable states of nature for the level of sales (N = 3) with S
(w1) = BD 30 million, S(w1) = BD 40 million and S(w1) = BD 50 mil- where ϕ− 1 is the inverse of the probability distribution function of a
−
lion. The average level of sales is S = BD 40 million and the standard standard normal distribution. The resulting linear program is as
deviation σS is BD 11.482 million. The start-up retailer has total assets follows:
TA of BD 30 million and a debt level D of BD 7.5 million. The lower and
6 N
upper limits, li and ui, respectively, on the decision variables are Max a1 + ∑ a1i xi − ∑ pn q (wn ) y+ (wn )
presented in Table 1. i =1 n =1
The first step in this empirical study is to run a regression analysis to 6
s. t ∑ a2i xi ≥ π (S + ϕ−1 (α ) σS ) − a2
determine the relationships between the problem decision variables i =1
(x1 , … , x6) and the profitability and liquidity of retailers in the Bahrain NTC
− x1 + x2 + x3 − y+ (wn ) + y− (wn ) = 365
S (wn ), n = 1, …, N

Table 1
x1 + x6 ≤ D
5
Limits on the decision variables.
∑ xi = TA
i =2
i 1 2 3 4 5 6
li ≤ xi ≤ ui , i = 1, …, 6
li 4.9 8.9 7.7 0.6 8.7 0.9 (13)
ui 6.4 11.1 10.1 1.5 11.4 2
The start-up retailer sets the profit margin π to 20% and the

4
H. Masri, Y. Abdulla
Table 2
Regression Results of the Liquidity and Profitability Models.
Objective function Intercept Accounts payable (x1) Accounts receivable (x2)
Liquidity 3.661 0.824 −0.383
Profitability 0.087 0.498 2.683
Table 3
The optimal solution.
(millions Accounts Accounts Inventory Cash and Fixed
of BD) payable receivable cash assets
equivalents
Optimal 6.4 9.7 10.1 1.5 8.7
Soluti-
on
probability of attaining the target profit α to 0.9. The net trade cycle
(NTC) in the retail market is set to 100 days and the interest rate for
short term investment q(wn) is set to 0.05 for all scenarios.
The linear equivalent (13) was solved using the CPLEX solver
version 12.1 and Microsoft visual C++ on Intel Core 2 Duo
2.00 GHz and 4GB of RAM using Windows Vista.
The obtained optimal solution presents a liquidity level of BD 5.095
million and a profitability of BD 13.201 million (see Table 3). The
optimal accounts payable value is BD 6.4 million. The optimal accounts
receivable value is BD 9.7 million, a level that will stimulate product
demand and strengthen the supplier-customer relationship. The higher
level of accounts receivable compared to accounts payable indicates
that the start-up retailer aims to attract customers and smooth demand,
and that the start-up retailer's suppliers are cautious about selling to the
start-up on credit due to the lack of previous transaction history with
the start-up. The high level of inventory will protect the start-up retailer
from unexpected distortions in the production cycle and mitigate
unstable demand for the product. The startup retailer holds a relatively
high level of cash to act as a precaution against unfavorable liquidity
events. This result is expected due to the high failure rate of start-ups,
which is caused mainly by financial and liquidity inadequacy. The
optimal level of fixed assets is found to be at the lower limit, consistent
with the findings of Afrifa and Padachi (2016), because the nature of
the retail business is that retailers' assets are mostly short-term. Finally,
the startup should not use more than BD 1.1 million of long-term debt
to finance its business, a finding that reflects the start-up's difficulties in
borrowing money.
The overall results of the empirical study are helpful to the start-up
retailer. The proposed modeling efforts helped to identify some of the
weaknesses and financial shortcomings in the start-up retailer's invest-
ment plan. The results confirm the importance of a strategic focus on
the effects of working capital management on the start-up retailer's
profitability and liquidity. The results shed the light on the magnitude
and direction of impact of each component of working capital structure
as a determinant of profitability and liquidity. The results also docu-
ment the existence of an optimal level of working capital and open
avenues for further empirical research in this area.
6. Conclusion
In this paper, we propose a multiple objective stochastic program to
determine the optimal level of working capital for a start-up retailer.
The proposed model maximizes both liquidity and profitability under
some predefined constraints and a stochastic level of sales. A regression
analysis is used to identify the model objective functions' parameters
using data from the Bahrain retail market. The solution strategy for the
multiple objective stochastic program is based on a chance constrained
approach, a recourse approach and a stochastic goal programming
Technological Forecasting & Social Change xxx (xxxx) xxx–xxx
Inventory (x3) Cash and cash equivalents (x4) Fixed assets (x5) Long-term debt (x6)
0.298 0.098 − 0.465 0.691
− 2.333 0.097 1.039 − 1.591
approach. The empirical study guided the start-up retailer to determine
an optimal level of its working capital based on some predefined targets
and goals and accounting for the overall retail market conditions.
Long-
term
The obtained results guided the start-up retailer in determining the
debt most optimal working capital structure, taking into account some
predefined set of constraints as well as the start-up's aim to achieve a
1.1 predefined level of profitability while maximizing liquidity. Another
scenario that should be investigated is to maintain a certain level of
liquidity while maximizing the profitability.
In addition to addressing a real life problem, this paper has the
merit of combining two quantitative approaches, a regression analysis
to determine the model objective functions and then multiple objective
stochastic programming models and approaches to reach an optimal
solution. The proposed solution strategy can be extended and used to
solve other problems in finance, investment, and supply chain manage-
ment, as well as problems in other fields where descriptive analysis and
decision theory are used.
References
Abuzayed, B., 2012. Working capital management and firms' performance in emerging
markets: the case of Jordan. Int. J. Manag. Financ. 8 (2), 155–179.
Afrifa, G., Padachi, K., 2016. Working capital level influence on SME profitability. J.
Small Bus. Enterp. Dev. 23 (1), 44–63.
Aktas, N., Croci, E., Petmezas, D., 2015. Is working capital management value-enhancing?
Evidence from firm performance and investments. J. Corp. Finan. 30, 98–113.
Almazari, A., 2013. The relationship between working capital management and
profitability: evidence from Saudi cement companies. Br. J. Econ. Manag. Trade 4 (1),
146–157.
Autukaite, R., Molay, E., 2011. Cash holdings, working capital and firm value: Evidence
from France. In: International Conference of the French Finance Association (AFFI).
Baños-Caballero, S., García-Teruel, P.J., Martínez-Solano, P., 2012. How does working
capital management affect the profitability of Spanish SMEs? Small Bus. Econ. 39 (2),
517–529.
Baños-Caballero, S., García-Teruel, P.J., Martínez-Solano, P., 2014. Working capital
management, corporate performance, and financial constraints. J. Bus. Res. 67 (3),
332–338.
Ben Abdelaziz, F., 2012. Solution approaches for the multiobjective stochastic
programming. Eur. J. Oper. Res. 216 (1), 1–16.
BMI Research, 2016. Bahrain Retail Report. BMI Research Fitch Company, U.K.
Braun, M., Larrain, B., 2005. Finance and the business cycle: international, inter-industry
evidence. J. Financ. 60 (3), 1097–1128.
Charnes, A., Cooper, W.W., 1963. Deterministic equivalents for optimizing and satisficing
under chance constraints. Oper. Res. 11 (1), 18–39.
Cluttons, 2014. Bahrain Commercial Market Outlook. Cluttons LLC.
Contini, B., 1968. A stochastic approach to goal programming. Oper. Res. 16 (3),
576–586.
Dantzig, G.B., Madansky, A., 1961. On the solution of two-stage linear programs under
uncertainty. In: Proceedings of the fourth berkeley symposium on mathematical
statistics and probability. Vol. 1. University of California Press Berkeley, CA, pp.
165–176.
De, P.K., Acharya, D., Sahu, K.C., 1982. A chance-constrained goal programming model
for capital budgeting. J. Oper. Res. Soc. 33 (7), 635–638.
Deloof, M., 2003. Does working capital management affect profitability of Belgian firms?
J. Bus. Financ. Account. 30, 573–587.
Eljelly, A.M., 2004. Liquidity-profitability tradeoff: an empirical investigation in an
emerging market. Int. J. Commer. Manag. 14 (2), 48–61.
Fabbri, D., Klapper, L., 2016. Bargaining power and trade credit. J. Corp. Finan. 41,
66–80.
Fazzari, S.M., Petersen, B.C., 1993. Working capital and fixed investment: new evidence
on financing constraints. RAND J. Econ. 328–342.
Ferris, J., 1981. A transactions theory of trade credit use. Q. J. Econ. 243–270.
Gao, H., Harford, J., Li, K., 2013. Determinants of corporate cash policy: insights from
private firms. J. Financ. Econ. 109 (3), 623–639.
García-Teruel, P., Martinez-Solano, P., 2007. Effects of working capital management on
SME profitability. Int. J. Manag. Financ. 3 (2), 164–177.
Giannetti, M., Burkart, M., Ellingsen, T., 2011. What you sell is what you lend? Explaining
trade credit contracts. Rev. Financ. Stud. 24 (4), 1261–1298.
Huyghebaert, N., Van de Gucht, L.M., 2007. The determinants of financial structure: new
insights from business start-ups. Eur. Financ. Manag. 13 (1), 101–133.
5
H. Masri, Y. Abdulla
Kaddumi, T.A., Ramadan, I.Z., 2012. Profitability and working capital management: the
Jordanian case. Int. J. Econ. Financ. 4 (4), 217.
Keown, A.J., Martin, J.D., 1977. A chance constrained goal programming model for
working capital management. Eng. Econ. 22 (3), 153–174.
Kim, C.S., Mauer, D.C., Sherman, A.E., 1998. The determinants of corporate liquidity:
theory and evidence. J. Financ. Quant. Anal. 33 (3), 335–359.
Koumanakos, D., 2008. The effect of inventory management on firm performance. Int. J.
Product. Perform. Manag. 57 (5), 355–369.
Lazaridis, I., Tryfonidis, D., 2006. Relationship between working capital management and
profitability of listed companies in the Athens stock exchange. J. Financ. Manag.
Anal. 19 (1), 26–35.
Lyroudi, K., Lazaridis, Y., 2000. The cash conversion cycle and liquidity analysis of the
food industry in Greece. In: Working Paper. University of Macedonia.
Masmoudi, M., Abdelaziz, F.B., 2015. A chance constrained recourse approach for the
portfolio selection problem. Ann. Oper. Res. 1–12.
Masri, H., Ben Abdelaziz, F., 2010. Belief linear programming. Int. J. Approx. Reason. 51
(8), 973–983.
Padachi, K., 2006. Trends in working capital management and its impact on firms'
performance: an analysis of Mauritian small manufacturing firms. Int. Rev. Bus. Res.
Pap. 2 (2), 45–58.
Petersen, M.A., Rajan, R.G., 1997. Trade credit: theories and evidence. Rev. Financ. Stud.
10 (3), 661–691.
Rehman, M.Z., Khan, M.N., Khokhar, I., 2015. Investigating liquidity-profitability
relationship: evidence from companies listed in Saudi stock exchange (Tadawul). J.
Appl. Financ. Bank. 5 (3), 159–173.
Shin, H., Soenen, L., 1998. Efficiency of working capital management and corporate
profitability. Financ. Pract. Educ. 8 (2), 37–45.
Smith, K., 1980. Profitability versus liquidity tradeoffs in working capital management.
In: Smith, K.V. (Ed.), Readings on the Management of Working Capital. West
Publishing Company, St Paul, MN, pp. 549–562.
Summers, B., Wilson, N., 2000. Trade credit management and the decision to use
Technological Forecasting & Social Change xxx (xxxx) xxx–xxx
factoring: an empirical study. J. Bus. Financ. Account. 27 (1–2), 37–68.
Vahid, T.K., Mohsen, A.K., Mohammadreza, E., 2012. The impact of working capital
management policies on firm's profitability and value: evidence from Iranian
companies. Int. Res. J. Financ. Econ. 88, 155–162.
Weinraub, H.J., Visscher, S., 1998. Industry practice relating to aggressive conservative
working capital policies. J. Financ. Strateg. Decis. 11 (2), 11–18.
Wilner, B., 2000. The exploitation of relationship in financial distress: the case of trade
credit. J. Financ. 55, 153–178.
Zhang, Q., Huang, X., Zhang, C., 2015. A mean-risk index model for uncertain capital
budgeting. J. Oper. Res. Soc. 66 (5), 761–770.
Hatem Masri, is an Associate Professor and Director of the Quality Assurance and
Accreditation office at the College of Business Administration, in the University of
Bahrain, Kingdom of Bahrain. He received a PhD in Management in 2004 and a master in
Operations research in 1999 from the University of Tunis, Tunisia. His research interests
include stochastic programming, multiple objective stochastic programming, supply
chain management, financial engineering, vehicle routing problems. He published in
several international journals (EJOR, FSS, IJAR …) and his research was funded by the
University of Tunis, University of Victoria, the University of Nizwa and the University of
Bahrain. Dr. Hatem is member of the International Society on Multiple Criteria Decision
Making, IEEE and the Tunisian Decision Aid Society.
Yomna Abdulla, is an Assistant Professor in the College of Business Administration at the
University of Bahrain, Kingdom of Bahrain. She obtained a PhD in Finance from the
University of Manchester and an MSc from the University of Glasgow. She did several
research works on empirical issues in corporate finance including trade credit, working
capital, debt maturity, capital structure and financial policies in public versus private
firms. She is a member of the Financial Management Association and the European
Finance Association.