Elsevier

Knowledge-Based Systems

Volume 27, March 2012, Pages 137-151
Knowledge-Based Systems

Exploring the preference of customers between financial companies and agents based on TCA

https://doi.org/10.1016/j.knosys.2011.09.003Get rights and content

Abstract

Based on transaction cost analysis (TCA), this research explores the customers’ loyalty to either the financial companies or the company financial agents with whom they have established relationship. In the past, consumers were divided into those who rely on agents and those who do not. In this study, we use two processes (pre-process and post-process) to select suitable rules, and to explore into the relationship among attributes. In the pre-process, we utilized factor analysis (FA) to choose the variable and rough set theory (RST) that found decision table to construct the decision rules, and approach to data mining and knowledge discovery based on information flow distribution in a flow graph. The post-process applies the formal concept analysis (FCA) from these suitable rules to explore the attribute relationship and the most important factors affecting the preference of customers for deciding whether to choose companies or agents. The degree of the customers’ dependence on agents was affected by the TCA, customer satisfaction and loyalty. The principal findings were that the different degrees of dependence of customers have various characteristics. The RST and FCA were two complementary mathematical tools for data analysis. Following an empirical analysis, we use two hit testes that incorporate 30 and 36 validated sample object into the decision rule. The hitting rate of two testes, were reached 90%. The results of the empirical study indicate that the generated decision rules can cover most new objects. Consequently, we believe that the result can be fully applied in financial research.

Introduction

Identifying the trends in customer consumption behavior and understanding the relationship between customers and agents are very important issues for financial companies because loyal customers are the key drivers for increasing a company sales and profitability. If a key decision-maker can accurately predict the degree of consumers’ dependence on the agents, he will be able to effectively take preventive measures to stop customers who may otherwise follow the agents leaving the financial firms [12]. Hence, in this paper, we use the data mining techniques to generate decision rules that can provide the decision makers with information about the attribute of customers’ preference.

Many research papers have attempted to address the issue of customers’ preference in financial market; moreover, they adopted the outcome of the transaction cost analysis (TCA). The TCA was part of the “New Institutional Economics” criterion, which has replaced orthodox neoclassical economics. However, the concept of the firm has been ignored by neoclassical economics for viewing it severely as a production function. Coase’s [10] initial propositions were that the firms and markets were alternative governance structures, that several methods can effectively reduce transaction costs. Specifically, Coase [10] considers that price operation would produce the costs which were generally called transaction costs under economic system of economic of specialization and exchange. In this context, transaction costs were the “costs of running the system” and include such ex ante costs as drafting and negotiating contracts and such ex post costs as monitoring and enforcing agreements.

The works of Williamson [41], [42], [43] has augmented Coase’s initial framework by proposing that transaction costs include both the direct costs of managing relationships and the possible opportunity costs of making inferior governance decisions. Williamson emphasized that transaction costs happened due to the market failure brought by human behavioral uncertainty and environmental uncertainty. Further, he considered TCA can be discriminated between ex ante (i.e. contracting cost, negotiating cost and protecting cost) and ex post (i.e. adaptive cost, bargaining cost, constructing, and operating cost and committed cost).

In the early TCA has been applied to the manufacturing firm’s decision to the supply of materials or components or distribution were mostly focused on vertical integration [3], [16], [18], [40].

Nevertheless the TCA was widely used in the commercial field to study various staffing model such as hiring a salesperson as an independent agent or an employee of the firm [2], customer–supplier relationships [34], [5], option pricing [22], entrepreneurship.

In 1975, Williamson indicated six reasons of transaction costs described bellow: (1) Bounded rationality: decision makers have constraints on their cognitive capabilities and limits on their rationality; (2) Opportunism: decision makers may unscrupulously seek to serve their self-interests; (3) Uncertainty and complexity environment factors; (4) Small numbers: some processes of trade were too proprietary or idiosyncratic; the information and resource cannot be circulated so the small numbers which control the market, causing market failure; (5) Asymmetric information between buyers and sellers; because of the forerunner can own more information that benefits him in the market; (6) Atmosphere: distrust between buyers and sellers. Making the transactional process will be a form, and increase the unnecessary transaction cost. Dahlman [11] proposed a new classification for the transaction costs. They proposed that TCA can be classified optimally within three main contextual domains: (1) search and information costs, (2) moral crisis costs, (3) asset specificity costs.

The search and information costs were indicated when the consumers want to find the product (agent) of they need, and they give the relative cost (product feature, product positioning, place, etc.). The moral crisis costs were meant that the customers must be accepted the risk of the corporate, product and brand. If the cost was higher, the customers were more distrustful of the product (agent) or brand. The asset specificity costs were the most important item; however, it was most easily to be neglected. In other words, the asset specificity costs were meant the degree of consumers’ dependence on this product (agent). In conclusion, this was useless to consumers when the transaction cost was higher and higher.

Customer relationship management (CRM) was a broadly recognized, widely- implemented strategy for managing and nurturing a company’s interactions with customers, clients and sales prospects. It involves using technology to organize, automate, and synchronize business processes – principally sales activities, but also those for marketing, customer service, and technical support (Wikipedia). CRM was an important and popular methodology to analyze customer behavior because it can establish a complete system of customer information.

The purpose of this study was using the points of TCA and CRM to discuss the customers’ preference; then rough set theory (RST) was applied to identify attributes/characteristics of customers’ preference. CRM regard both customer loyalty and customer satisfaction. This paper utilizes the customer loyalty and customer satisfaction to help us to understand the relation between the agents and customers.

The customer loyalty was affected by their satisfaction, though the structure of the relationship was not totally symmetric and linear [1], [23], [24]. The measure of behavioral loyalty was on the basis of attitudinal loyalty statement that was to say, actual repurchase was recommended behavior rather than intention [7]. According to Bandyopadhyay and Martell [4], the presence of such situational factors as stock being not accessible, such personal or intrinsic factors as opposition to vary or such communal and cultural factors as communal restraint intensifies the demand to discriminate customer loyalty from repurchase behavior.

A number of study focuses on the quantification of the problem by streamlining all parameters and applying statistical tools to analyze the data. Pawlak [32] proposed RST as a rule-based decision-making instrument. It can handle both crisp and fuzzy datasets. In this study, RST was used to analyze data contents and data features.

For this paper, we use two steps to perform the data analysis. The first step was pre-process, which use factor analysis (FA) to choose the variable and then utilize RST to find decision rules [17]. The second step was post-process, which creates additional values on those rules by the formal concept analysis (FCA) in order to gather the decision rules to construct the concept and to explore the relationship among attributes. This information provides prior knowledge for decision makers [37], [25]. The FCA provides the mathematical theory, which belongs to algebra and was a branch of lattice theory.

This study adopts RST combined flow graphs and Formal Concept Analysis to analyze customers’ preference/characteristic, and the results demonstrate that the combined approaches were well suited to find the characteristic relationship of customers between financial companies and agents. Furthermore, we applied a hit test to check the feasibility of the decision rules. It would be clear that new data matches the decision classes reaching 90%. The results of this research showed that the decision rules can effectively predict the degree of customers’ dependence to agents. The analytical process was shown in Fig. A.1 of Appendix A.

The rest of this paper was organized as follows: in Section 2, concepts to be used in this study were outlined and described. Section 3 shows an empirical study to explore the customers’ preference of company or agent in the financial market. Finally, in Section 4, was presented the conclusions.

Section snippets

Overview of this research

In this section, we briefly review RST, flow graphs and FCA, which were used in analyzing the customers’ preference/characteristic. The theory of RST was described in Section 2.1. In Section 2.2 was narrated the flow graphs. The FCA theory was presented in Section 2.3.

Empirical study: a preference of customers between financial companies and agents

The questionnaires were distributed to customers in the North and Northeast districts of Taiwan. The respondents were of two categories: one set contains people who are active investors or have financial interests; and the other contains people with little or no financial interests. Data were collected based on nominal and ordinal scales. There were 107 valid questionnaires from a total of 118 received. The percentage of valid questionnaires was 90%. Among the valid respondents, there were 55

Conclusions

In this study, RST generated 44 rules. These rules can be explored further to gain additional information through FCA, such as the most important factors/attributes affecting the relationship between personal preference for agents and its result was same as flow graphs. Therefore this attributes relationship can give decision makers a priori prediction.

The main characteristics of the low dependence on agents were with medium attitudinal loyalty (c62), college (c82), and lower behavioral loyalty

Acknowledgements

Part of this paper has been presented on the 6th International Symposium on Management Engineering, August 5–7, Dalian, China. We expanded and re-wrote this paper from 6 pages in our conference paper.

References (49)

  • Z. Pawlak

    Rough sets, decision algorithms and Bayes’ theorem

    European Journal of Operational Research

    (2002)
  • Z. Pawlak

    Rough classification

    International Journal of Human–Computer Studies

    (1999)
  • Z. Pawlak

    Rough classification

    International Journal of Man–Machine Studies

    (1984)
  • J.Y. Shyng et al.

    Rough set theory in analyzing the attributes of combination values for the Insurance Market

    Expert Systems with Applications

    (2007)
  • J.Y. Shyng et al.

    An integration method combining rough set theory with formal concept analysis for personal investment portfolios

    Knowledge-Based Systems

    (2010)
  • C.H. Wang et al.

    Mining the R&D innovation performance processes for high-tech firms based on rough set theory

    Technovation

    (2010)
  • B. Walczak et al.

    Tutorial rough sets theory

    Chemometrics and Intelligent Laboratory Systems

    (1999)
  • L.A. Zadeh

    Fuzzy sets

    Information and Control

    (1965)
  • L.Y. Zhai et al.

    Feature extraction using rough set theory and genetic algorithms an application for the simplification of product quality evaluation

    Computers & Industrial Engineering

    (2002)
  • E.W. Anderson et al.

    The antecedents and consequences of customer satisfaction for firms

    Marketing Science

    (1993)
  • E. Anderson

    The salesperson as outside agent or employee: a transaction cost analysis

    Marketing Science

    (1985)
  • S. Balakrishnan et al.

    Technical change, competition and vertical integration

    Strategic Management Journal

    (1986)
  • M. Bensaou et al.

    Buyer-supplier relations in industrial markets: when do buyers risk making idiosyncratic investments?

    Organization Science

    (1999)
  • R.H. Coase

    The nature of the firm

    Economica

    (1937)
  • Cited by (0)

    View full text