Glaucia Miura Ota
Instituto Federal de São Paulo - Campus Suzano, Brazil
E-mail: mglaucia@hotmail.com
Adriano Maniçoba da Silva
Instituto Federal de São Paulo - Campus Suzano, Brazil
E-mail: adrianoms@ifsp.edu.br
Submission: 01/22/2019
Accept: 02/10/2019
ABSTRACT
Supply
contracts amongst suppliers and buyers can be used as powerful mechanisms to
manage trade-offs between risks and costs in negotiations amid the constituent
parts of the supply chain. In decentralized decision-making, as a trade
transaction between a wholesaler and a retailer, social preferences can
influence behavior of these decision makers. The objective of this study was to
verify the effect of social preferences on transactions in supply chains, using
Brazilian individuals with the purpose of comparing the results with those
obtained in other countries. The methodology proposed in this study was
quantitative, as variables already studied by previous studies were
investigated. An experiment was conducted with three handlings (normal, status
and relationship) simulating buying and selling situations in a supply chain.
The results indicated that highlighting the performance status of a supply
chain improves the efficiency of the chain when compared to normal
transactions, or when there is a relationship between the links.
1. INTRODUCTION
The
purpose of supply chains according to Chopra and Meindl (2003) is to boost the
value generated by this chain, which, according to the authors, means the
difference between the value of the final product delivered to the customer and
the effort employed by the chain to serve it.
For
superior performance to be achieved, supply chain constituencies must work
together and coordinate to pursue an overall goal aligned with the objectives
of each link (CACHON, 2003).
Transactions
in supply chains between suppliers and buyers can be carried out by a single
decision maker who has all the necessary information, which refers to a
centralized or integrated supply chain. Another way is when there are multiple
decision makers with diverse incentives and information, which is the
decentralized supply chain.
For
coordination, contracts are used in supply chains, with the objective to
optimize chain performance. Contracts also allow the risk arising from supply
chain uncertainty to be shared between the parties (HÖHN, 2010).
Social
objectives, such as justice, status and reciprocity, are social preferences
sought by decision-makers in supply chain contracts (LEE; SEO; SIEMSEN, 2018).
These social preferences influence trades in supply chain transactions (LOCH;
WU, 2008).
In
studies carried out abroad on supply contracts (LOCH; WU, 2008), it was found
that social preferences affect decision-making in negotiations of decentralized
supply chains, influencing the individual behavior of decision-makers. In this
way, this study aims to investigate the effect of social preferences on supply
chain transactions in Brazilian individuals and to compare them with studies
conducted abroad, verifying whether there is a difference in the amplitude of
the bias.
Initially,
a review of the literature on behavioral operations, an approach on contracts
and coordination in supply chains and social preferences, and transactions in
supply chains was carried out.
The
present work is organized in seven parts: introduction, where the objective of
the study is mapped out; theoretical reference; methodology used and
characteristics of the sample; analysis and discussion of results; the final
considerations, limitations and suggestions for future work;
2. LITERATURE REVIEW
Behavioral
operations, procurement and supply chain coordination, and social preferences
and supply chain transactions will be addressed in this section.
2.1.
Behavioral
operations
Behavioral
Operations is a sub-area of Production and Operations Management that has
gained prominence abroad, however, in Brazil, it is still absent from studies
(DA SILVA, 2015).
This
investigative area makes it possible to understand the relationship between the
field of operations and people, and how this interface impacts operational
performance in organizations (SILVA, 2015).
According
to the research by articles published in six renowned journals on the subject
behavioral operations carried out by Bendoly, Donohue and Schultz (2006), there
has been a record of publications since 1985.
The
influence of behavioral issues in relation to economic activities has been
researched in many fields of study, such as marketing, accounting, management
and economics, however, in Operations Management, research is still scarce
(BENDOLY; DONOHUE; SCHULTZ, 2006).
A study
carried out by Da Silva (2015) proposes suggestions of potential areas of
production management and operations where behavioral analysis can be applied,
based on previous studies.
Ribbink
and Grimm's (2014) study came to the conclusion that, as is often the case in
day-to-day supply chain operations, cultural differences also have a
significant impact on negotiations.
The
article elaborated by Silva (2015) pointed out that there is a strong tendency
on studies regarding decision making, where biases and heuristics make
decisions subject to errors, and that there are cognitive factors, such as the
social and individual understanding involved. It also states that motivation
may be the reason for numerous operational problems, because people are motivated
for various reasons.
The
field of study of behavioral operations has advanced in three levels:
individual, group and organizational, having relation between them. Therefore,
managers must take into account behavioral factors so that the best decisions
are made, discovering reasons for possible errors in the process and acting
effectively in operational practices.
Several
aspects of behavioral operation such as social behavior, relationships,
rewards, interaction, emotions, and motivation are similar to the issues raised
in human relations theory. Although it is a developing area of study, the
behavioral operations field addresses a questioning about the mechanistic view
that exists in operations management. In this way, the differential consists in
admitting that human behavior must be aggregated in processes and decisions,
since all actions practiced in organizations are subject to it (SILVA, 2015).
Based
on an article by Lee, Seo and Siemsen (2018), which analyzed experimental
studies in the context of behavioral laboratory operations from 2006 to 2016,
the three most researched experiments in operations were the newsvendor model,
the auction and supply chain contracts. The newsvendor model can be found in a
nationwide study by Ota and Da Silva (2017), in which the authors replicated
the experiment conducted by Bolton and Katok (2008) and Feng, Keller and Zheng
(2011), with the purpose of comparing their results with those obtained in
surveys conducted abroad.
Thus,
this study addresses the experiment of supply chain contracts.
2.2.
Supply
chain contracts and coordination
According
to Simchi-Levi, Kaminky and Simchi-Levi (2010), in a common supply chain with
two negotiators; a supplier and a buyer, the following order of events follows:
the buyer, based on a demand forecast, decides the quantity of products that
will optimize profit and makes the request to the supplier, who reacts to the
order made by the buyer. In this way, this order of events represents a
decision-making process in sequence, and thus, this chain is called a
sequential supply chain, where each subject establishes his actions, regardless
of the effects of his decisions in other parts. The authors say that this
tactic does not benefit all members in the chain, so it is not effective.
Optimum
supply chain performance can be achieved if organizations work together in a
coordinated way, although this does not always happen because of the concern of
the members in optimizing their own goals, resulting in inferior performance.
In order to achieve the best performance, it is necessary for companies to work
in a coordinated way, aligning the objectives of each chain link with the
overall desired goal (CACHON, 2003).
Supply
chain coordination, according to Giannoccaro and Pontrandolfo (2004), can be
achieved by adopting a centralized or decentralized decision-making process.
The first case occurs when there is only one decision-maker in the supply
chain, who has all the information necessary and important for the
decision-making process and also the power to implement these decisions;
however, in this case, there is the hypothesis that this control is not
realistic. According to Höhn (2010), decentralized control occurs when there
are several independent decision-makers deciding at different stages of the supply
chain, with different information and incentives. Currently, decentralized
chains are a majority, as a result of outsourcing and globalization. The
production sector, when outsourced, automatically distributes decision-making
powers among various subjects.
Giannoccaro
and Pontrandolfo (2002) affirm that contracts in the supply chain operated in
decentralized decision-making are a tool to achieve coordination and to obtain
coherent behaviors of decision-making subjects in a decentralized environment, as
if the chain were being operated centrally, and another important objective is
to optimize system performance (HÖHN, 2010).
Likewise,
Simchi-Levi, Kaminsky and Simchi-Levi (2010) reiterate the use of supply
contracts as important mechanisms to achieve global optimization and better
management of trade-offs between risk and cost and also likely benefits.
The
study presents several types of contracts, which differ based on the clauses
that are celebrated between suppliers and buyers, which include how risks
arising from uncertainty are shared between the parties in the supply chain
(HÖHN, 2010).
A
type of contract that seeks coordination in the supply chain is called
wholesale-price, which, according to Cachon (2003), occurs when the supplier
charges the retailer for a unit purchased at a fixed price. According to the
author, the wholesale price contract is not normally considered a co-ordination
contract because the supplier prefers a higher value in the price of its
products, but this type of contract is often observed in the practice.
Another
type of contract is buyback, where the vendor charges the retailer a value per
unit bought at a fixed price, but returns the retailer a portion of the value
per unit not sold at the end.
In
the revenue-sharing contract, the vendor charges the retailer a value per unit
purchased, and the retailer gives the vendor a percentage of their revenue.
The
contract of quantity-flexibility, according to Cachon (2003), operates as
follows: the supplier charges a fixed value for units purchased; however, it
reinstates the retailer for the units that were not sold. In this way, this
type of contract offers complete protection to the retailer, while the buyback
contract does so in a partial way.
In
the sales-rebate contract, the author states that the supplier charges a fixed
value per unit purchased; on the other hand, grants the retailer a discount for
each unit sold that exceeds a limit x, in order to encourage the effort to
increase demand.
Finally,
there is the quantity-discount contract, which, according to Cachon (2003), is
a type of contract where the supplier grants the retailer a discount in the
purchase of larger quantities. Coordination per unit purchased; while granting
the retailer a discount for each unit sold that is obtained when the marginal
revenue and marginal cost curves are in the ideal quantity.
According
to their research, Cachon (2003) concluded that coordination in supply chains
can be obtained by different types of contract. According to each organization,
there is a form of contract that can suit your needs.
On
the other hand, Katok and Wu (2009) based on their experiment with some types
of contract, pointed out that the efficiency obtained was lower than the theory
predicts.
2.3.
Social
preferences and supply chain transactions
Current
research on hiring, according to Lee, Seo and Siemsen (2018), points out that
decision-makers seek to achieve reciprocity, justice, and status, which are
social goals, as well as economic factors. These social preferences, according
to Loch and Wu (2008), influence supply chain negotiations.
According
to Rabin (1993), in contemporary economics it is inferred that subjects strive
to achieve their own material advantages. But there was one exception to this
kind of behavior that attracted the attention of economics studies: people who
care about the well-being of the other, as well as their own well-being.
However, this altruism, according to psychology, is more complicated, because
it is not a continuous behavior, the subjects help when they are helped and are
cruel with the people who hurt them, thus pointing out a concern of the
subjects with justice. In his study, the author also noticed a possible
reciprocity between the participants of the game.
The
approach based on the Game Theory proposes to study problems where decision
making occurs between individuals, in a situation of interaction, in which the
decision of one affects and is affected by the other party (VASCONCELLOS,
2011).
The
study of Loch and Wu (2008) was elaborated creating a sequential game where two
players (player A and player B) choose a price. The two sums together determine
the market price of a product (p = pA + pB). Player B determines its value from
player A. The demand for this product is determined by the linear function of
the price of that product q = 16-p.
If
the cost of the product is zero, the profit of the first player in one round is
given by πA = pA (16-pA-pB), and the second player's profit is πB = pB
(16-pA-pB).
In
the experiment made by Loch and Wu (2008), they randomly selected participants
that played for 15 rounds.
At
the end of the game, for a perfect balance, player A should choose p*A=8 and
player B's preference should be p*B=4, where player A's profit would be π*A=32
and player B's profit would be π*B=16, which would result in an efficiency of
75%.
In
the experiment conducted by Loch and Wu (2008), three conducts were studied:
control, relationship and status. Each round of the game, the prices and
profits of the two participants were displayed on each player's screen. In the
control treatment, the players were chosen randomly and anonymously, and
participated in the game without having any type of communication. In the
relationship treatment, before starting the game, the participants stood face
to face, introduced themselves and shook hands. Finally, in the treatment of
status, at the end of each round the winner was announced that surpassed his
opponent in the profits.
According
to hypothesis 1 raised by Loch and Wu (2008), the decisions of the values of
the two players would be smaller in the treatment of relationship and greater
in the treatment of status, in relation to the treatment of control. On the
other hand, in hypothesis 2, player A's value decision could increase based on
the value decision determined by player B in the previous round (by
reciprocity), or player A's value may decrease if his pay in the previous
period is greater (by status), and the same condition applies to player B.
The
results of the experiment indicated that the social preferences modify the
behavior and that, between the conducts there are significant differences.
3. METHODOLOGY
In
order to reach the objectives proposed by this study, whose variables have
already been studied previously, the applied sample was non-probabilistic and
for convenience. Students from graduate level in logistics of a public higher
education institution participated in the study. A total of 58 students (29
pairs) participated in this experiment, being 13 pairs in the Control dealing,
8 pairs in the Relationship dealing and 8 pairs in the Status dealing.
An
experiment was carried out based on the methodology applied in the studies of
Loch and Wu (2008), with the objective of identifying the behavior of the
Brazilian decision makers represented here by the students of the upper level
classes.
The
design of the experiment was 3x1 in which the three treatments (control,
relationship and status) were manipulated by a student profile (graduate
level). In the Loch and Wu (2008) experiment, the randomly selected
participants played for 15 rounds and made price decisions according to a
demand function defined according to Equation 1.
q=16-p
(1) |
Where:
q =
demanded quantity
p =
price
The
tools used for the experiment (procedures, descriptions, instructions) were
similar to previous studies, so that the results could be compared. However,
unlike the original work, this experiment was performed in a classroom, by
hand, and no software was used in the application of the experiment. We used
LibreOffice Calc only for the calculations of demand, participants' profits and
graphics.
In
the application of the experiment, initially, the instructions of how the
activity would be performed were informed to the students by the instructors.
The students who participated in the control treatment were randomly divided,
obeying the criteria of balancing gender and age for all treatments, and
arranged in the classroom so that players A stayed in one extreme and players B
in the other end of the room so that they could not discover the identity of
their partner in the pair during the experiment.
The
game instructions were given to each participant, as well as a 20 rounds sheet
block (so that the study participants did not know when the experiment would
end) to player A of each pair, containing the round number, price of player A
and price of player B, and also a result sheet to each participant, for
individual control of results.
Players
A decided on their price of the product, the instructors picked up the sheets
from that pairs' round, shuffled and then handed the sheets to their
corresponding pair on the other side of the room, so they didn’t find the
identity of the other player. After player B decides on the price, based on the
price that player A decided, the sheets were collected, and then the
instructors made the prices of each player available on the board, the demand
of that product in the market and the profits of each player of that round, so
that the students could write down in their result sheets. Then the next round
started, subsequently ending in round 15.
In
the Status dealing, there was also no interaction of the pairs during the game,
and the difference was the delivery of only one results sheet, who took turns
between player A and player B of each pair during the game through the
instructors, in addition to having the winner of each round announced on the
board, in order to encourage competition between players. Already in the
relationship dealing, before the beginning of the game, the doubles were
introduced and shook hands, and it was emphasized that they (player A and
player B) cared about each other. However, during the game, there was no
interaction between the members, and each pair was assigned only a player’s
results sheet as well.
Hypothesis
1 of the theory raised by the studies of authors Loch and Wu (2008) predicts
that decisions on player A and player B's price would be lower in Relationship
and higher in Status, relative to Control. If social preference conditions do
not have an effect, the three conducts of the experiment should show results that
resemble the theories of rationality and selfishness, where pA=8 and pB=4,
during all periods. In hypothesis 2, player A's price decision increases with
player B's price decision in the previous round (reciprocity), and player A's
price decreases if his win in the previous round is higher (status). The same
goes for Player B.
4. ANALYSIS OF RESULTS AND DISCUSSION
As in
the studies of Loch and Wu (2008), the impact of social preferences in the
different treatments of this experiment was verified. Prices, decisions and
evolution during the rounds were also checked.
Table
1 shows the prices, profits and efficiency of the doubles in the three types of
treatment.
Table
1: Descriptive statistics in 15 rounds
Treatment |
Average |
Standard
deviation |
Median |
Control |
|
|
|
Price of A |
5,4974 |
2,9574 |
5 |
Price of B |
5,7333 |
3,2413 |
5 |
Profit of A |
21,2 |
15,5483 |
24 |
Profit of B |
20,2872 |
14,8951 |
22 |
Efficiency |
64,82% |
|
|
Relationship |
|
|
|
Price of A |
6,1083 |
3,5142 |
5 |
Price of B |
5,2 |
3,3895 |
5 |
Profit of A |
21,5583 |
14,8606 |
24 |
Profit of B |
17,8583 |
13,5301 |
16 |
Efficiency |
61,59% |
|
|
Status |
|
|
|
Price of A |
5,3750 |
2,8845 |
5 |
Price of B |
5,4583 |
2,8606 |
5 |
Profit of A |
21,3333 |
14,7108 |
20,5 |
Profit of B |
21,9500 |
14,5035 |
23 |
Efficiency |
67,63% |
|
|
Source:
Authors (2018)
The
results indicate that different types of treatment are effective and change
behavior in decision making, that is, people behave according to the
environment in which they are.
This
study showed that profits in the treatment of Status were slightly higher than
in the Control and Relationship treatment, unlike the study conducted by Loch
and Wu, where the profits were higher in the Relationship treatment and lower
in the Treatment status, than in the Control treatment.
Next,
player A's average price decisions in the 15 rounds are shown in Figure 1 along
with the best responses (which would be the rational decision that would
maximize the profit).
Graph 1: Price
of player A for 15 rounds in the different treatments
Source:
Authors (2018)
According
to the revised theory, rational and selfish players should make the same price
decision throughout the game, and this is not what happened in this study. An
interesting fact is that even in the Control treatment, where the participants
did not have any type of interaction, their decisions were lower than their
best answer (Graph 3), as well as the decisions of player A (Graph 1).
It
can also be noticed that the average decisions of the subjects do not change
considerably over periods, which reinforces that the effects of the treatments
remain stable throughout the game.
Still
in the analysis of Table 1, it can be noted that the prices of the Status
treatment are lower than in the Control treatment, and higher in the
Relationship Treatment.
Graph 2: Price
of player B and comparisons with the best price response
Source:
Authors (2018)
Graph
3 shows that in the Control treatment, player B's price falls below the best
rational response, contrary to the results of Loch and Wu (2008).
Graph 3: Price
vs. Best Response in Control Treatment
Source:
Authors (2018)
In
the Status treatment, decisions of player B almost reach the best response
(Graph 4), and in the experiment of Loch and Wu the response of player B is
higher than the best answer.
Graph 4: Price
vs. Best response in Status Treatment
Source:
Authors (2018)
In
the Relationship treatment, shown in Figure 5, player B's decisions fluctuate
strongly around the best answer, unlike the work of Loch and Wu (2008), where
decisions remained below the best answer.
Graph 5: Price
vs. Best response in Relationship Treatment
Source:
Authors (2018)
Graph
6 presents the accumulated frequencies of the price decisions of player A
(Graph 2) and player B (Chart 3) over the 15 rounds, in the three conducts. It
can be seen from the graph that the social preferences in the different
treatments change the behavior of the decision-making subjects.
In
both graphics, the Control Treatment has its distribution on the left side,
while the Status treatment has its distribution in the middle for player A, and
to the right on player B. Already in the Relationship treatment, in the
decisions of player A, he remains on the right, and decisions of player B's stay
in the middle.
Graph 6:
Cumulative frequency of price decisions in all rounds
Source:
Authors (2018)
Graph 7 shows the profit of players A and B
respectively during the rounds.
Graph 7: Profit
of players A and B in the 15 rounds
Source:
Authors (2018)
The
efficiency achieved in the treatments shows that in Status an efficiency of
67.63% was reached, in Control 64.82% and in Relationship efficiency was
61.59%.
5. CONCLUSION
Most
of the studies on supply chain contracting converge in negotiations with
rational subjects and disregard the impact of behavior in this type of scenario
(LOCH, WU, 2008). The present study aimed to verify the effect of social
preferences in decision making in negotiations in a supply chain.
According
to the results of the experiment, it was verified that the subjects deviate
from the quantity that maximizes the profit due to the effects of the social
preferences. The greatest difference found was in the average player B's profit
in the Status and Relationship treatments, where the average profit difference
between these two conditions was 18.63%, affirming the best efficiency of the
Status treatment in our study, with 67.63 %, unlike the work done by Loch and
Wu (2008), where it was found that the Relationship treatment had the highest
efficiency in relation to the other conditions.
Loch
and Wu (2008) reiterated the importance of social preferences, which can both
encourage cooperative behavior between the parties, but also produce the
opposite effect by undermining profits. They also affirmed that another study
on this theme pointed to social preferences as being as or more effective than
contracts.
This
study can be considered innovative in the national context, because it was the
first to analyze social preferences in contracts and compare it with other work
done abroad.
As a
limitation to the study, the fact that the participants are not completely
foreign to each other is highlighted, although there was no contact during the
experiment, they study in the same teaching institution; another limiting
factor was the number of students recruited in the research, as well as the
fact that the experiment was applied manually.
Thus,
as a suggestion for future research, it is recommended to replicate this study
with a larger number of participants.
REFERENCES
BENDOLY, E.; DONOHUE, K.; SCHULTZ, K. L. (2006)
Behavior in operations management: Assessing recent findings and revisiting old
assumptions. Journal of operations management, v. 24, n. 6, p.
737-752.
CACHON, G. P.
(2003) Supply chain coordination with contracts. Handbooks in operations research and management science, v. 11, p.
227-339.
CHOPRA, S.; MEINDL, P. (2003) Gerenciamento da cadeia de suprimentos: Estratégia, planejamento e operação. 1. ed.
São Paulo: Prentice Hall.
DA SILVA, A. M. (2015) Operações Comportamentais:
Oportunidades de Pesquisa no Contexto Brasileiro. Anais do XVIII SemeAd - Seminários em Administração. São
Paulo-SP: Nov.
GIANNOCCARO, I.; PONTRANDOLFO, P. (2004) Supply chain
coordination by revenue sharing contracts. International journal of
production economics, v. 89, n. 2, p. 131-139.
GIL, A. C. (2008) Métodos e técnicas
de pesquisa social.
6. ed. Editora Atlas SA.
HÖHN, M. I. (2010) Literature review on supply chain contracts. In: Relational supply
contracts. Springer, Berlin, Heidelberg, p. 19-34.
KATOK, E.; WU, D.
Y. (2009) Contracting in supply chains: A laboratory investigation. Management Science, v. 55, n. 12, p.
1953-1968.
LEE, Y. S.; SEO,
Y. W.; SIEMSEN, E. (2018) Running Behavioral Operations Experiments Using
Amazon's Mechanical Turk. Production and
Operations Management, v. 27, n. 5, p. 973-989.
LOCH, C. H.; WU,
Y. (2008) Social preferences and supply chain performance: An experimental
study. Management Science, v. 54, n.
11, p. 1835-1849.
OTA,G. M.; DA SILVA, A. M. (2017) Efeito centralização no
comportamento de decisores brasileiros. III Congresso de Logística (IFSP –
Campus Suzano 2017). Anais do
IFLOG.. Disponível em: <http://szn.ifsp.edu.br/portal2/arquivos/static/anais_completos_iflog2017_2.pdf>
RABIN, M. (1993)
Incorporating Fairness into Game Theory and Economics. The American Economic Review, v. 83, n. 5, p. 1281-1302.
RIBBINK, D.;
GRIMM, C. M. (2014) The impact of cultural differences on buyer–supplier
negotiations: An experimental study. Journal
of Operations Management, v. 32, n. 3, p. 114-126.
VASCONCELLOS, M. A. S. (2011) Economia: micro e macro. 5. ed. São Paulo: Atlas.
SIMCHI-LEVI, D; KAMINSKY, P.; SIMCHI-LEVI, E. (2010) Cadeia de suprimentos: projeto e gestão.
3. ed. Porto Alegre: Bookman.
SILVA, G. (2015) OPERAÇÕES COMPORTAMENTAIS: PERSPECTIVAS
TEÓRICAS. Anais do XVIII SemeAd -
Seminários em Administração. São Paulo-SP: Nov.