Introduction

Second-hand markets are experiencing rapid growth within the global circular economy, surpassing other distribution channels in terms of growth rate (Jain et al., 2022; Kwon, 2021; Weinswig, 2017). According to Future Market Insights (2022), the global second-hand apparel market was valued at 71,225.6 million USD in 2022 and is projected to reach 282,748.6 million USD by 2032, with a compound annual growth rate of 14.8%. This growth is primarily driven by the emergence of consumer-to-consumer (C2C) online second-hand marketplaces, such as Thredup, Poshmark, and Karrot (Park & Armstrong, 2019). Although these platforms offer unique advantages as two-sided markets facilitating transactions between participants with different interests (Munger, 2021), they also confront issues, such as the involvement of nonprofessional sellers or buyers and a lack of mutual trust with anonymous targets, leading to transaction disputes and fraud (Jain et al., 2022; Lee, 2021).

In this rapidly evolving market, a specific category of C2C platforms distinguishes itself through its emphasis on hyperlocal, community-centric transactions. These local second-hand marketplace platforms (LSMPs) are designed to facilitate exchanges within defined local communities or neighborhoods, leveraging mobile technology to prioritize geographical proximity and communal engagement over broader, anonymous marketplaces (Bae et al., 2022). For example, Poshmark provides regional content based on US zip codes, while Karrot designates a 6 km radius around the user’s residence as “My Neighborhood” for C2C transactions. These localized platforms promote close-distance transactions and increase the possibility of face-to-face interactions, thereby creating a sense of psychological closeness between potential transaction partners. A sense of belonging, intimacy, attachment, and homogeneity among users residing in the same area fosters trust, reduces anonymity associated with online transactions, and alleviates concerns about engaging with unknown individuals (Kim & Kim, 2022a, 2022b).

Previous research on C2C second-hand marketplaces can be broadly categorized into two streams: the first focuses on the challenges arising from seller anonymity and limited product information (Kwon, 2021; Pavlou & Fygenson, 2006; Pavlou et al., 2007; Yen & Lu, 2008), resulting in risks related to finances, time, performance, psychology, and delivery (Mao & Lyu, 2017). The second stream examines factors that encourage second-hand transactions, including value dimensions (Wu et al., 2022; Yrjölä et al., 2017; Zhang et al., 2019), perceived benefits (Liang et al., 2021; Yang et al., 2017), social contact (Guiot & Roux, 2010; Juge et al., 2022), enjoyment (Hamari et al., 2016; Tussyadiah, 2016), and trust (Möhlmann, 2015). However, there is a lack of research specifically investigating local second-hand transactions as critical factors that eliminate transaction barriers. This raises an important research question: what is the most important feature of local second-hand marketplace platform (LSMP) and how does it affect actual transaction behavior?

To address this research question, this study investigated the important features of LSMPs and their impact on transaction behavior. Drawing on transaction cost theory (TCT), we focused on the negotiation process and explored various types of negotiation costs and their influence on transaction satisfaction. Additionally, we introduced the concept of a “sense of my neighborhood,” assuming that residents’ psychological proximity in close geographic areas fostered a sense of neighborhood. We considered the sense of one’s neighborhood not only as a preceding factor influencing transaction satisfaction but also as a vital moderating factor that affected the relationship between negotiation costs and trade satisfaction.

The significance of this study lies in its exploration of the unique factors that influence consumer behavior on LSMPs. By focusing on the local service aspect and a sense of my neighborhood, we address a gap in the literature and contribute to a more comprehensive understanding of the second-hand marketplace. Furthermore, our research enriches the TCT by introducing negotiation costs as a crucial variable and uncovers the mechanisms that mitigate the negative impact of negotiation costs on platform satisfaction by highlighting the moderating role of the sense of my neighborhood. These insights advance academic understanding and offer practical implications for platform operators aiming to optimize their operations and enhance user experience.

Literature Review

Transaction cost theory (TCT)

TCT, introduced by Coase (1937) and expanded by Williamson (1981), examines the costs that accompany economic transactions, extending beyond the mere price of goods to include additional costs incurred during exchange and ownership transfers (Eggertsson, 1990). TCT provides a framework for understanding the broader costs in the context of C2C negotiations for pre-owned goods. Unlike traditional economics , TCT acknowledges the real-world complexities and costs associated with market inefficiencies (Williamson, 1985). Through the lens of TCT, scholars identify and examine these inefficiencies, often arising from information asymmetries (Liang & Huang, 1998). This leads consumers to actively seek information and monitor transactions, thereby incurring various transaction costs (Coase, 1937). For instance, on LSMP, buyers assess not only the price, the quality of goods and trading methods but also seller reliability, often using reputation systems, reviews, and ratings to mitigate the transaction costs stemming from uncertainty and mistrust.

Transaction costs include search and information costs related to finding available information, negotiation and decision-making costs during the negotiation process, and enforcement and governance costs for ensuring contractual compliance (Liang & Huang, 1998). Notably, TCT posits that individuals aim to minimize these costs (Williamson, 1981), directly impacting transaction satisfaction—how well the exchange meets the involved parties' expectations (Iaccobucci et al., 1995). Enhanced satisfaction is linked to positive outcomes like improved feedback, repeat transactions, and increased referrals (Jones & Suh, 2000), highlighting the significance of managing transaction costs effectively.

The negotiation phase, in particular, garners attention for its inherent complexity and substantial contribution to overall transaction costs (Lee, 2021). Costs here escalate due to the time invested in extensive communication, potential misunderstandings, and the stress of uncertainty, directly influencing the final terms of the transaction, transaction efficiency and participants' satisfaction (Liang & Huang, 1998).

Within the C2C market for pre-owned items, the negotiation process is marked by its complexity. Parties engage in iterative discussions to reconcile differences over price, quality, and other transaction details, often entailing a series of offers and counteroffers to reach consensus. This process intensifies negotiation complexity and uncertainty, amplifying the associated costs (Liang & Huang, 1998). Therefore, a deeper understanding of transaction costs in the negotiation phase and strategies for their mitigation could boost overall transaction efficiency, and satisfaction for both consumers and platforms. In this study, the negotiation cost was defined as the monetary and non-monetary transaction costs incurred during the negotiation process on a second-hand trading platform. We systematically explored and delineated consumer behavior by focusing on the diverse dimensions of negotiation costs.

Negotiation cost

Negotiation costs are cognitive expenditures incurred when reaching an agreement between parties (Liang & Huang, 1998). These costs, pivotal in determining the transaction outcomes and efficiency, are heightened by the uncertainty inherent in decision-making processes within second-hand transactions (Liang & Huang, 1998; Pavlou & Fygenson, 2006). Reflecting the diverse sources of this uncertainty previously identified in the literature (Mavlanova et al., 2012), we systematically categorize negotiation costs into four types—pricing (Liang et al., 2021; Schiffman & Kanuk, 2000), performance (Fernando et al., 2018; Yen & Lu, 2008), time (Devaraj et al., 2002), and psychological (Teo & Yu, 2005).

Pricing uncertainty, which is largely associated with price flexibility in trading items, is a significant source of ambiguity in negotiations. Buyers aim to secure the lowest price, whereas sellers hope to maximize their returns (Lee, 2021). This diametrically opposed goal setting amplifies the uncertainty surrounding the final transaction price and increases the cognitive effort (an economic cost) in negotiations. Especially in second-hand trade, the complex quality assessment task might reveal defects only later (Dimoka et al., 2012). Thus, risk-averse buyers may fear economic loss if the product’s true value is less than its listed price (Kahneman & Tversky, 1979), leading to potential dissatisfaction with C2C platforms (Möhlmann, 2015). Based on this, we hypothesized the following:

  • H1a. Economic cost during negotiation will negatively influence transaction satisfaction.

Second-hand products inherently increase condition and quality uncertainty during negotiations. Usage frequency, duration, and habits affect the condition of a product by introducing variability and unpredictable performance. This uncertainty complicates the negotiation process, making it more challenging to establish a fair price. Consequently, buyers may seek detailed information from sellers (Dimoka et al., 2012), but sellers may not provide sufficient data, leading to information asymmetry and subsequent performance costs (Hwang & Youn, 2023; Pavlou et al., 2007). Given that product uncertainty negatively affects transaction satisfaction and platform preference (Lee, 2014), we hypothesized the following:

  • H1b. Performance cost during negotiation will negatively influence transaction satisfaction.

In second-hand trading, negotiation over transaction terms requires additional communication and effort from both parties, introducing a time cost, which is defined as the perceived shopping time (Baker et al., 2002), as both parties must dedicate time and attention to the transaction process. Any delays in communication or fulfillment of transaction promises can exacerbate this time cost, particularly for parties more eager to trade (Teo & Yu, 2005). Because waiting time, whether physical or perceived, negatively affects customer satisfaction (Nie, 2000), we proposed the following hypothesis:

  • H1c. Time cost during negotiation will negatively influence transaction satisfaction.

Psychological costs refer to the mental and emotional strain consumers experience during the shopping process, often resulting from unpleasant or uncomfortable environments (Baker et al., 2002). In second-hand C2C transactions, psychological costs stem primarily from uncertainty about counterparties (Dimoka et al., 2012). This uncertainty includes the inability to verify counterparty characteristics, concerns about transaction defaults, deal breaking or reversal, and uncertainties over communication norms (Yen & Lu, 2008). Consumers must gauge the other party’s trustworthiness and deal with potentially adverse encounters during C2C trading (Pavlou et al., 2007). For instance, buyers may worry about sellers providing false or exaggerated information, whereas sellers may experience buyers requesting significant discounts, excessive communication, purchase cancellations, or refund requests. Such instances, recognized as psychological costs, can adversely affect transaction satisfaction (Kim & Kim, 2019). Thus, we proposed the following hypothesis:

  • H1d. Psychological cost during negotiation will negatively influence transaction satisfaction.

Sense of my neighborhood

On LSMPs, users engage in trading within the confines of their identified residential areas and interact with other residents in their local vicinity (Kim & Kim, 2022a, 2022b). A neighborhood is distinguished as a social unit consisting of individuals residing within specific geographical boundaries and sharing social bonds (Porteous, 1986). Residents often form collective psychological impressions and emotional attachments toward their residential locale, known as a sense of neighborhood (Nasar & Julian, 1995). This concept encapsulates emotional closeness signified by shared interests, intimacy, common values, a sense of belonging, and trust (Chun, 2004; Douglas, 2022; Gans, 1961; Long & Perkins, 2003). It also includes the practical aspects of physical proximity, such as connectivity and accessibility (Douglas, 2022).

LSMP users are likely to develop a sense of neighborhood, given the localized nature of the platforms. Research has shown that these users cultivate feelings of connection, belonging, and attachment within the neighborhood where their transactions occur (Kim & Kim, 2022a, 2022b; Park & Cheon, 2020; Yang et al., 2021). LSMPs, such as Karrot, Nextdoor, Facebook Neighborhoods, and Streetlife, require users to verify their residential locations to facilitate more connected, relevant, and enriching experiences. Users are prompted to update their location if they relocate and periodic verification requests ensure accuracy. This process of re-verification heightens neighborhood awareness, fostering unique sentiments toward one’s residential area (Yang et al., 2021).

In the context of second-hand transactions, we introduced the concept of a “sense of my neighborhood.” This term encapsulates feelings of belonging, identity, emotional intimacy and trust shared among potential transaction parties within a specific neighborhood. It provides an underpinning for local second-hand transactions. A robust sense of neighborhood is widely recognized as a critical factor that strengthens community bonds and instills regional identity (Lee, 2009). This often influences opinions on products and promotes participation in consumption activities within the community. Furthermore, it has a significantly positive impact on satisfaction, and commitment (Choi, 2005; Hagel, 1999). Accordingly, we proposed the following hypothesis:

  • H2. A sense of my neighborhood will positively influence transaction satisfaction.

A sense of my neighborhood as a moderator

The interplay between “a sense of my neighborhood” and transaction satisfaction in LSMPs embodies a complex dynamic, underscored by the socio-psychological underpinnings of community interaction. We propose that an individual’s sense of my neighborhood not only directly influences transaction satisfaction but also moderates the effect of various negotiation cost dimensions—economic, performance, time, and psychological—on transaction satisfaction.

Central to our argument is the idea that a strong sense of my neighborhood, characterized by familiarity and intimacy within a community, significantly influences communication patterns and interaction quality (Liu et al., 2010). Such enhanced communication mitigates tensions between parties (Kim & Kim, 2022a, 2022b; Srivastava & Culén, 2019) and fosters more frequent and effective transactional exchange (Huang et al., 2017), thereby reducing transactional uncertainty (Abbes et al., 2020) and the cognitive effort required for successful negotiations (Lee, 2021; Liang et al., 2021). Crucially, efficient communication expedites the finalization of transactions, promoting user flexibility in price adjustments. These adjustments are perceived as reasonable compensations for the expedited and convenient transaction process. Moreover, the often challenging negotiation process—marked by the coordination of time and location (Srivastava & Culén, 2019)— becomes considerably more efficient within the context of strong neighborhood ties. Active, direct communication allows users to swiftly reach consensus on mutually convenient trading times and locations, effectively reducing the time costs associated with second-hand transactions (Kim & Kim, 2022). Therefore, individuals with a stronger sense of my neighborhood are proposed to mitigate the negative impacts of economic and time costs on transaction satisfaction.

Additionally, the presence of a communal identity within the same geographical area bolsters trust and positive perceptions among resident (Dwyer et al., 1987; McKnight et al., 1998; Morgan & Hunt, 1994), extending these sentiments to the traded products and the parties involved (Ba, 2001). This trust acts as a deterrent against deceptive practices like overpricing or misrepresentation of product quality, thereby enhancing transaction transparency and the verification of product quality and performance. Physical meetings to verify product quality, when necessary, further highlight the advantages of this trust (Fernando et al., 2018; Kim & Kim, 2022a, 2022b; Lee et al., 2022). Furthermore, negotiation processes within local communities are often characterized by an increased sense of responsibility and trustworthiness due to the likelihood of repeated interactions with familiar individuals (Kavanaugh et al., 2005). This proximity not only raises expectations of ethical behavior but also encourages positive interactions, thereby reducing the probability of misconduct such as etiquette breaches, unfulfilled promises, and transaction failures (Lee et al., 2022). Consequently, trust and expectations of responsible behavior within a neighborhood context lessen the impact of performance and psychological costs on satisfaction.

Based on the foregoing discussion, this study posits the following hypothesis:

  • H3. A sense of my neighborhood will negatively moderate the impact of economic (a), performance (b), time (c) and psychological (d) costs on transaction satisfaction.

Transaction satisfaction as a mediator

Research across various domains has established satisfaction as a key predictor of behavioral intentions, including within e-commerce, where it influences loyalty, channel preference, and online service usage (Anderson & Sullivan, 1993; Devaraj et al., 2002; Oliver, 1980). Extending this to C2C transactions, prior studies highlight transaction satisfaction’s pivotal role in encouraging the continued use of online platforms (Kim & Kim, 2019; Tussyadiah, 2016; Yen et al., 2013). Thus, we posit that satisfaction from past LSMP transactions will positively impact users’ future platform usage intentions:

  • H4. Transaction satisfaction will positively influence the intention to use LSMP.

Satisfaction also serves as a bridge between antecedent variables and behavioral intentions (Chen & Chou, 2012; Errajaa et al., 2022; Huang et al., 2017; Siemens, 2007; Spreng et al., 2009. This mediating effect of satisfaction is particularly significant because while satisfaction reflects past evaluative judgments, intention to use indicates future behavioral tendencies (Mittal et al., 1998). Therefore, this study predicts that various negotiation costs incurred during the transaction process will influence transaction satisfaction, which will subsequently affect the intention to continue using the online second-hand platform.

  • H5. Transaction satisfaction will mediate the relationship between negotiation costs and the intention to use LSMP.

Figure 1 shows the research model proposed in this study based on the relationship between negotiation costs, transaction satisfaction, intention to use, and sense of my neighborhood.

Fig. 1
figure 1

Research model

Full size image

Methods

Measures

We conducted an online survey targeting users who had traded fashion products on Karrot, Korea's leading online second-hand marketplace (Kwon, 2021). The survey consisted of sections on usage behavior, transaction experiences on Karrot, and demographic details. For usage behavior, we asked about the C2C marketplace platforms used and products traded over the last six months. Transaction experiences were captured by inquiries into the most recent fashion product trades, including transaction type (selling or buying), prices, the “My Neighborhood” feature usage, and travel time for the transaction.

The survey measured negotiation costs, sense of neighborhood, transaction satisfaction, and intention to use. Reflecting the cognitive effort in negotiating terms (Liang & Huang, 1998), each of the negotiation cost dimensions—economic, performance, time, and psychological—was assessed with four questions on a seven-point Likert scale, tailored for the context of LSMPs. Economic cost measures were derived from Liang and Huang (1998) and Pavlou et al. (2007). Items assessing performance costs drew upon the scales of Dimoka et al. (2012). For time costs, we  used survey items from Liang and Huang (1998) and Teo and Yu (2005). Psychological costs were assessed based on Liang and Huang (1998) and Pavlou et al. (2007).

Sense of my neighborhood was measured through questions on emotional connection, neighborhood attachment, and trust in other users, employing scales from Douglas (2022) and Nasar and Julian (1995). Transaction satisfaction was assessed by the overall positive feelings about transaction outcomes utilizing modified scales from Tussyadiah (2016). Intention to use was gauged directly through inquiries about plans to continue using Karrot, based on Lee (2021).

The questionnaire also captured demographic characteristics, including sex, age, marital status, highest level of education, occupation, average monthly household income, and average monthly clothing expenditure.

Data collection and sample characteristics

The study participants were limited to men and women in their 20 s and 50 s, who lived in the urban areas of Seoul and Gyeonggi-do in South Korea, and had experience trading fashion products on Karrot. In total, 507 survey responses were collected through an online research agency, and 460 responses (excluding 47 outliers) were used for the analysis. Frequency analysis, exploratory factor analysis, reliability analysis, and regression analysis were conducted using SPSS Statistics 26.0, and a confirmatory factor analysis was conducted using AMOS 21.0.

The study sample included 460 respondents; 36.7% (n = 169) were men and 63.3% (n = 291) were women. The age ranges of the respondents were as follows: 20–29 = 16.1% (n = 74); 30–39 = 32.2% (n = 148); 40–49 = 33.5% (n = 154); 50–59 = 18.3% (n = 84); those in their 30 s and 40 s accounted for 65.7% of the total sample. In terms of jobs, 46.5% (n = 214) of the participants were office workers, 12.6% (n = 58) were homemakers, 8.3% (n = 38) were freelancers, 7% (n = 32) were professional technical workers, and 6.5% (n = 30) were students. Regarding academic background, 322 (70%) participants were college or university graduates. The average monthly household income was as follows: 25.7% (n = 118) earned less than 3 million to 5 million won, 25.4% (n = 117) earned 5 million to 7 million won, and 21.5% (n = 99) earned 7 million to 10 million (or less) won. The participants made product transactions in the last six months at similar percentages in the following categories: women’s fashion items such as shoes and bags = 11.4% (n = 212), women’s clothing = 11.1% (n = 206), and men’s fashion items or clothing = 11.1% (n = 206). In addition to fashion products, participants made transactions for household appliances (8.2%, n = 152); books, tickets, or music (7.7%, n = 143); digital devices (7.5%, n = 139); and beauty (7.2%, n = 135).

Results

Measurement model validation

Confirmatory factor analysis (CFA) was conducted on measurement items for the seven latent variables. To improve the goodness-of-fit of the measurement model, we checked whether items hindered validity based on the standardization coefficient and modification indices. The overall fit index of our structural model was confirmed (χ2 = 963.678, df = 410, p = 0.000, GFI = 0.871, NFI = 0.908, CFI = 0.944, Standardized RMR = 0.088, RMSEA = 0.054). Cronbach’s α coefficient for all latent variables was above 0.861, and the CR (composite reliability) was above 0.979. The reliability of the measurement items was satisfactory. The factor loadings of the measurement items for the latent variable were significant at the 0.001 level, and the average variance extracted value was above 0.50. Thus, the convergence validity was confirmed, as shown in Table 1. In addition, the AVE (average variance extracted) values of all the components were larger than the square of the correlation coefficients between the constructs (Table 2). Thus, all measurements satisfied discriminant validity (Hair & Tatham, 2010).

Table 1 Results of the confirmatory factor analysis of measurements
Full size table
Table 2 Discriminant validity of measures: squared correlation and AVE of variables
Full size table

Hypothesis testing

Result of moderated hierarchical regression analysis

To test hypotheses 1–4, a three-step moderated hierarchical regression analysis was conducted. In the first step (Model 1), the control variables (sex, year of birth, transaction type, and transaction frequency) were included. In the second step (Model 2), sense of my neighborhood and negotiation costs, including economic, performance, time, and psychological costs, were included as independent variables. In the final step (Model 3), we investigated the possible interaction variables between negotiation costs and sense of my neighborhood. As shown in Table 3, the regression model was significant. Additionally, the variance inflation factor value between all variables was small (≤ 2.5), and the tolerance limit was large (≥ 0.8), showing no problem of multi-collinearity.

Table 3 The result of the three-step moderated hierarchical regression analysis
Full size table

The first step of the moderated hierarchical regression analysis (Model 1) revealed a significant model (F = 2.719, R2 = 0.023, p < 0.05). Model 2 incorporated four negotiation cost dimensions (economic, performance, time, and psychological) alongside sense of my neighborhood. The addition of these variables significantly improved the fit of the model (F = 50.640, R2 = 0.503, ∆R2 = 0.480, p < 0.001). Economic (ß = − 0.493, p < 0.001) and time (ß = − 0.182, p < 0.001) costs had significantly negative effects on transaction satisfaction, supporting H1a and H1c. However, performance (ß = − 0.073, p > 0.05) and psychological costs (ß = 0.052, p > 0.05) had no effect on transaction satisfaction, leading to the rejection of H1b and H1d. Model 2 also found that sense of my neighborhood had significant effect on transaction satisfaction (ß = 0.308, p < 0.001). Thus, H2 was supported.

In Model 3, the moderating effect of a sense of my neighborhood between negotiation cost and transaction satisfaction was tested, and the model fit was significantly increased (F = 41.747, R2 = 0.549, ∆R2 = 0.046, p < 0.001). Of the four interaction effects, three were statistically significant: economic cost*sense of my neighborhood (ß = 0.147, p < 0.001), performance cost*sense of my neighborhood (ß = 0.098, p < 0.05), time cost*sense of my neighborhood (ß = 0.097, p < 0.05). Thus, H3a, H3b, and H3c were supported. However, the interaction between psychological cost and a sense of my neighborhood was not statistically significant (ß = − 0.074, p > 0.05); therefore, H3d was rejected.

In addition, transaction satisfaction had a positive effect on the intention to use LSMPs (ß = 0.679, p < 0.001), supporting H4.

Assessment of the mediating effect of transaction satisfaction

To verify H5, we tested the indirect effects of transaction satisfaction. We employed a bootstrapping approach with 5,000 samples, as suggested by Hayes (2017), using the PROCESS macro for SPSS to examine the direct, indirect, and total effects. Direct effects of negotiation costs on intention to use were initially assessed, followed by an analysis of indirect effects via transaction satisfaction, employing bootstrapping for significance testing of mediation to use. Confidence intervals were computed to ascertain the significance of these effects (Table 4).

Table 4 Indirect effects of transaction satisfaction
Full size table

Economic cost not only had a direct effect on the intention to use LSMPs (CI [confidence interval]=[ − 0.1334, − 0.2997], effect size=0.2166) but also an indirect effect through transaction satisfaction (CI=[− 0.1529, − 0.2842], effect size=0.2134). Performance cost had a direct effect on intention to use LSMPs (CI = [− 0.1481, − 0.0055], effect size = − 0.0768), but no indirect effect through satisfaction (CI = [− 0.0766, 0.0187], effect size = 0.0241). Time cost (CI = [− 0.0995, − 0.0227], effect size = − 0.0597) only had an indirect effect on intention to use LSMPs through transaction satisfaction. However, psychological cost had neither a direct effect on intention to use nor an indirect effect through transaction satisfaction. Thus, H5 was partially supported.

Discussion and conclusion

Platform-based services, coupled with evolving consumer behaviors, have spurred the growth of C2C second-hand transactions within specific geographic boundaries, showcasing unique patterns and decision-making processes distinct from traditional marketplaces. This study delves into neighborhood-based second-hand transactions, shedding light on negotiation cost dimensions, their impact on transaction satisfaction, and the subsequent intention to use LSMPs, with a spotlight on the dual role of “sense of my neighborhood.” The insights and academic implications of this study are as follows:

Primarily, this research contributes to a deeper understanding of negotiation costs in C2C transactions, positioning them as multi-dimensional rather than singular, comprising economic, performance, time, and psychological aspects. This refined and multi-faceted view not only addresses gaps from previous literature but also advances TCT, offering a comprehensive lens to examine consumer challenges in second-hand transactions. The study uniquely explores how these cost dimensions influence consumer decision-making on local C2C platforms.

Second, this study significantly advanced our understanding by demonstrating both the direct and indirect impacts of various negotiation cost dimensions on the intention to use the platform, with the validation of the mediating effect of transaction satisfaction. Economic costs notably drive both direct motivation for platform use and indirect decision-making via transaction satisfaction, emphasizing the critical influence of economic factors in C2C transactions. This aligns with and expands upon existing literature, such as Liang et al. (2021), Liang and Huang (1998) and Hwang and Youn (2023), underscoring how uncertainties and the cognitive burden of price negotiations can deter participation in C2C transactions. Time costs indirectly influence platform use through transaction satisfaction, supporting Devaraj et al. (2002), who observed time costs' negative effect on channel satisfaction and selection. Performance costs demonstrated a direct bypassing effect on intention to use LSMPs without the need for transaction satisfaction. Conversely, psychological costs have no significant effect, aligning with Park et al. (2017), who found product, not seller, uncertainty significantly influences intentions to engage in second-hand transactions. This doesn't suggest consumers ignore counterpart uncertainty but indicates a willingness to embrace a certain level of it. Such acceptance underscores the limited impact of perceived costs associated with transaction parties on the decision to utilize second-hand marketplace platforms. With these insights, this study presents a more nuanced understanding of the unique characteristics of negotiation on LSMPs, augmenting the knowledge base established by previous studies that primarily focused on the motivation, risks and benefits associated with second-hand trading.

Lastly, our study reveals the “sense of my neighborhood” as both a precursor and a moderator within second-hand transactions, enhancing transaction satisfaction. Users experiencing enhanced levels of intimacy, homogeneity, bonds, and trust within their local trading community exhibited higher transaction satisfaction, leading to an increased intention to use the platform. Simultaneously, it alleviates the adverse effects of economic, performance, and time-based negotiation costs on transaction satisfaction, resonating with Pavlou and Gefen (2004) and Lee (2021), who underscore trust’s role in reducing cognitive transaction costs. Trust toward a trading counterpart can mitigates risk concerns about costs, time, and product performance (Song, 2020). Highlighting social cues' pivotal role, our study significantly advances online second-hand trading research by detailing how the 'sense of my neighborhood' enriches transaction satisfaction and mitigates negotiation costs' effects. This dual capacity as both an antecedent and a moderator, fueled by community bonds, provides a deeper insight into the dynamics of second-hand transactions.

The practical implications for LSMP management from this study are substantial. Economic and time costs significantly impact transaction satisfaction, indicating that platform enhancements to streamline negotiations could greatly improve user experience. Specifically, integrating transparent pricing mechanisms and benchmarking features using historical data can reduce economic costs by clarifying price expectations. Similarly, features that expedite the negotiation process, like a standardized negotiation interface with preset options, can lower time costs and enhance efficiency.

Ensuring product quality is crucial because it can directly influence users' intentions to use LSMPs. Standardized forms for detailed product descriptions and high-quality photos, along with algorithms highlighting listings that meet specific quality criteria, can support this goal by providing buyers with clear, accessible information and helping sellers to quickly sell their products.

The dual role of a “sense of my neighborhood” underscores the need for strategies to foster a sense of community among users. Developing features that promote local events, highlight local sellers, and facilitate interactions can foster a stronger community feel. Gamification strategies, such as badges, leaderboards, and rewards, can further engage users in community-building efforts, offering social recognition and motivating continued platform interaction.

While psychological costs were not found to significantly impact transaction satisfaction or intention to use the platform, addressing potential concerns through secure payment methods, privacy measures, and efficient dispute resolution remains important because it can help mitigate users’ potential psychological concerns. Leveraging technology, such as artificial intelligence, to manage excessive communication (Park et al., 2017), filter inappropriate language, ensure friendly interactions, and accurately calculate user reputation, can also support a trustworthy and user-friendly environment. Implementing legal or institutional safeguards to address any transaction issues is advisable. These strategies, derived from the study’s findings, suggest that LSMP operators have several avenues to enhance user satisfaction and engagement, underscoring the importance of user-centric platform design and community-building initiatives.

Although this study offers considerable insight, it is essential to identify its limitations. First, it does not differentiate between the experiences of buying and selling within the negotiation process. This approach may not entirely encapsulate the distinct negotiation costs and experiences of buyers and sellers. In the context of second-hand transactions, buyers often experience information asymmetry (Pavlou et al., 2007), which can amplify performance costs. However, sellers may endure greater communication load and potential conflicts when dealing with multiple prospective buyers, resulting in higher psychological costs. Future studies should separately scrutinize negotiation costs related to sales and purchases across various contexts.

Second, this study was constrained to a single local second-hand platform, disregarding the wider spectrum of nonregional platforms on which second-hand transactions also occur. Differences in the cognitive effort and negotiation costs may exist across the platforms and future research should compare the findings across different types of second-hand marketplaces to better understand their unique characteristics.

Finally, this study focused solely on second-hand fashion product transactions, which bring about specific challenges owing to factors such as fit and evaluation criteria. In C2C second-hand transactions, the non-returnable nature of products can be a major deterrent for potential buyers. This issue is particularly relevant for fashion items, where product attributes like size, fit, and style can increase the level of uncertainty and hesitation for the buyer. High-value fashion items, such as luxury brands, can lead to more complex negotiations. In contrast, general products, like everyday necessities and books, may come with less uncertainty. As such, future research should consider the nature of the products being traded and examine how this impacts negotiation costs and transaction outcomes. These suggested directions for future research could lead to a deeper understanding of negotiation costs in different transaction contexts, variations among different types of second-hand platforms, and the role of product characteristics in shaping transaction dynamics.