Sunteți pe pagina 1din 11

Journal of Retailing and Consumer Services 33 (2016) 143–153

Contents lists available at ScienceDirect

Journal of Retailing and Consumer Services


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

Supply and demand on crowdlending platforms: connecting small and


medium-sized enterprise borrowers and consumer investors
Erik Maier
HHL Leipzig Graduate School of Management, Jahnallee 59, Leipzig, 04109 Germany

art ic l e i nf o a b s t r a c t

Article history: Consumer crowdlending to small and medium enterprises is an increasingly relevant financial service
Received 30 November 2015 phenomenon which depends on platforms as an intermediary. To match borrowers’ supply of loan re-
Received in revised form quests and customers’ investment demand, this article proposes that crowdlending platforms need to
28 June 2016
initiate double switching behavior. On the one hand, switching of enterprise borrowers from traditional
Accepted 6 August 2016
bank financing to crowdlending is driven by greater convenience (speed, flexibility, simplicity) and
process transparency. Consumers, on the other, predominantly invest in crowdlending loans based on
Keywords: their economic performance relative to the choice set on the platform. We contribute to the literature on
Crowdlending crowdfunding, satisfaction and switching in financial services.
Crowdfunding
& 2016 Elsevier Ltd. All rights reserved.
Fintech
SME
Switching
Satisfaction

1. Introduction year after the platform had secured a $150 million investment by
Citigroup (Back, 2016)—halved the company's market value to $1.8
Driven by technological advances, new service models have billion in June 2016.
developed in the financial industry which offer additional oppor- The focus area of this article are consumer investments into
tunities to customers. Under the common denominator “fintech”, small and medium enterprises (SME) seeking financing (C2B
these new businesses aim to challenge existing financial institu- lending), rather than into requests from private borrowers (C2C
tions by using technology to deliver value to the customer in an lending). In the UK crowdlending to SME has surpassed invest-
alternative way. One such model is crowdlending: the direct fi- ments into consumer loans in June 2015 and continues to grow
nancing of credits by a group of consumers. In this, crowdlending faster. In continental Europe consumer loans are still the pre-
aims to replace traditional banks by directly connecting potential dominant investment form, but the annualized growth of crow-
borrowers and investors through a platform, where future lenders dlending to SME exceeds the increase in consumer loan invest-
can directly invest into various credit requests. ments by more than 50% (  90% vs. 150% in 2015, altfi DATA, 2016).
In recent years, crowdlending has grown out of its nascent Though a technological vanguard, one of the major challenges
phase into an established industry. While only $0.15 billion had of the industry is a classical one: matching supply and demand.
been financed via crowdlending platforms in the United States in Because crowdlending platforms work as intermediary that facil-
2010, already $10.1 billion were funded in 2015 on the two largest itate transactions between two parties (Sriram et al., 2015), char-
platforms alone—a 97% growth versus the previous year (NSR ging a fee for funded loans or made investments, their income
Invest, 2016). Also in Europe the industry has grown strongly, with stream is dependent on actual consumer investment and funded
the financed volume in the UK, for instance, having grown by 130% credits. Even if a crowdlending fintech attracted investment de-
p.a. in the last 5 years to more than d2.3 billion in 2015 (altfi DATA, mand from many consumers, these would still require a supply of
2016). Along with this overall growth, investments have soared loans as investment targets.
and major companies developed. U.S. credit marketplace SoFi, for
instance, saw a $1 billion investment in September 2015 (Cushing,
1.1. Approach and positioning
2015). However, success and failure lie close, as the recent scandals
at Lending Club show (McLannahan, 2015), which—less than a
This article investigates means to bridge the gap between
consumers’ demand for investment and the supply of loan re-
E-mail address: erik.maier@hhl.de quests from company borrowers. To ensure they provide value to

http://dx.doi.org/10.1016/j.jretconser.2016.08.004
0969-6989/& 2016 Elsevier Ltd. All rights reserved.
144 E. Maier / Journal of Retailing and Consumer Services 33 (2016) 143–153

customers, crowdfunding platforms need to offer a sufficient been out of the research focus despite its economic relevance.
number of qualified loan requests on their platform, as an in-
vestment platform without investment opportunities would be
unattractive to customers. From a different angle, potential bor- 2. The supply side: borrower loan requests
rowers need to be confident that their loans will be financed in a
reasonable amount of time, thus requiring a certain level of con- 2.1. Theoretical background
sumer investments. Thus, without the necessary diffusion of the
platform both at the borrower and lender side (Rogers, 2003), the SME have multiple options to finance their business, ranging
platform will not be attractive for either of these groups. Managing from equity to loans. To finance their operations and business
this interdependence is not only an interesting field for research projects, SME so far predominantly relied on loans from banks
(Sriram et al., 2015), but also a key challenge to this growing fi- (Mills and McCarthy, 2014). The relationship of SME to their main
nancial service industry and focus of the article. Thus, the sub- bank is characterized by stability and loyalty (e.g., 70% of U.S. SME
sequent parts will first discuss the supply of credits as investment only had a single bank in the year 2000; Trayler et al., 2000),
opportunities (i.e., the borrower perspective) and then demand though this situation is probably driven by inertia and a feeling of
side (i.e., the consumer investors) to enable a full understanding of undifferentiated offering of the competing banks (Madill et al.,
the business model (Belleflamme et al., 2015). 2002). However, satisfaction and loyalty levels with SME banks
In this context, it is necessary to distinguish SME crowdlending have been decreasing in recent years, driven by the loss of banks’
from other forms of business crowdfunding. On the one hand, role as trusted intermediary due to a higher standardization of
crowdlending can be distinguished by the form of investment. In financial products (Lundahl et al., 2009). Crowdlending platforms
crowdlending the lender only invests in specific loan requests (e.g., could capitalize on this dissatisfaction as an opportunity to over-
via platforms such as Lending Club or Funding Circle). This con- come switching inertia. However, to do so they need to generate a
trasts to other forms of crowdfinancing such as crowd equity high level of customer satisfaction with their own offering, which,
(where investors acquires a share of another company, such as via in turn, is a driver of switching (Manrai and Manrai, 2007).
Crowdcube), donation or reward-based crowdfinancing (which are The subsequent arguments will be based on (dis-)satisfaction
more concerned with altruistic motives and do not involve a legal with different elements of lending services as driver of SMEs’ in-
claim to an asset, such as Kickstarter; see Kshetri (2015)). tention to switch, and not on antecedents of technology adoption
On the other hand, within crowdlending to companies, one can (e.g., Venkatesh and Bala, 2008). This focus is based on the un-
distinguish between different stages of the lifecycle of the com- derstanding that crowdlending is not a fundamentally new service
pany in which the investment is placed. While some platforms or technology—in which case adoption models would be more
focus on financing loans from startup companies or new business appropriate—but a comparable service offered by different pro-
ideas (e.g., Seedmatch; for a research perspective, see Zhang and vider, to which an SME would need to switch. First, in SME fi-
Liu (2012)) others focus on financing projects from established nancing all utility components should be known to SME in ad-
companies, such as SME. Because the importance of SME for the vance (e.g., service quality, speed of financing, interest rates), al-
overall economy is high (e.g., constituting 46% of the GDP in the U. though they might be evaluated differently across traditional
S. or 56% in Germany; Kobe, 2012; Bundesministerium für Wirt- banks and crowdlending. Though the source of the funds in
schaft und Energie, 2015), we focus the present research on this crowdlending is innovative, the financing service itself is no in-
type of businesses. novation. This contrasts to the objects of evaluation in the most
common adoption models, such as the Theory of Reasoned Action
1.2. Contribution (Fishbein and Ajzen, 1975; Ajzen and Fishbein, 1980), the revised
Technology Adoption Model (Davis, Bagozzi, and Warshaw, 1989)
Two field studies show that successfully matching supply and or its various extensions (Venkatesh, 2000; Venkatesh and Davis,
demand in crowdlending requires double switching behavior: 2000; Venkatesh and Bala, 2008; Shih, 2004). These models focus
both from companies away from their bank, and from consumers on innovations, which require an adoption of new technology by
from other forms of investment (e.g., with their bank, on stock the user (e.g., use of a Word processing software in the 1980s,
exchange or the internet). We argue that though the business Davis et al., 1989; use of smartphones for mobile shopping today,
model is novel, the drivers of switching are not. Agrebi and Jallais, 2015). Most SME are, however, already using
To coerce SME to switch at least part of their financing to the one ore multiple forms of external financing. Consequently,
crowd, the present research finds that crowdfunding platforms crowdlending platforms position themselves as alternative to
should predominantly market their functional performance, such banks, not as technical innovation (e.g., Funding Circle U.S. even
as convenience (i.e., speed, flexibility, availability) or process advertises with an account manager, which handles the applica-
transparency (i.e., process clarity and predictability), relative to the tion process, just like a bank representative; Funding Circle, 2016).
economic (i.e., fees, interest and collateral) or relational aspects Second, extant research most frequently uses satisfaction as an
(i.e., staff relationships) of their business. Image aspects, such as antecedent for switching behavior, both in general B2B models
crowdlending innovativeness, are found to play a subordinate role. (Lam et al., 2004) and with regards to bank switching specifically
Consumer investors in crowdlending base their decisions pre- (Chakravarty et al., 2004; Ennew and Binks, 1996; Manrai and
dominantly on an evaluation of hard criteria relative to alternative Manrai, 2007), while intention to adopt has—to the best of our
offers on the platform (e.g., relative yield, interest, offers available), knowledge—not been used in this context. Thus, we will focus on
rather than on an assessment of “soft” information (e.g., use of the (dis-)satisfaction as driver of switching and not on sources of
loan, writing style). technology adoption. However, as research on adoption models
This article contributes to extant research in three dimensions: has been extensive (Marangunić and Granić, 2015), and, for in-
firstly, it extends research on SME bank switching behavior stance, perceived usefulness and ease of use (Venkatesh and Bala,
(e.g., Colgate and Hedge, 2001) to switching to crowdlending. 2008) and satisfaction drivers (Lundahl et al., 2009) are likely to
Secondly, it introduces consumer innovativeness as driver of overlap, we will refer to the adoption models in the hypothesis
switching. Thirdly, it broadens existing research on consumer in- development where it is appropriate.
vestments into private loans (e.g., Berkovich, 2011) or company SME customer satisfaction as driver of the intention to switch is
equity (e.g., Cordova et al., 2015) to SME loans, which has so far usually characterized along three dimensions, including technical,
E. Maier / Journal of Retailing and Consumer Services 33 (2016) 143–153 145

functional and image components (e.g., Kang and James, 2004). stressed in the context of retail banking (e.g., Arbore and Busacca,
While technical aspects pertain to the satisfaction with what is 2009; Levesque and McDougall, 1996; Naser, 2002; Molina et al.,
delivered as the core task, such as the provision of financing and 2007) and described as the main obstacle for switching (Chakra-
the related conditions. Functional aspects relate to how the fi- varty et al., 2004).
nancing process is conducted, including the personal contact to In crowdlending, however, we expect personal contact to be
the bank employees (Grönroos, 1990; Lundahl et al., 2009). Ad- less relevant than transparency about the financing process. On
ditionally, customers’ satisfaction is driven by the image generated the one hand, the relationship with existing banks should not be
from previous experiences with the institution (Kang and James, an obstacle to switching, as it is rather unlikely that SME will fully
2004). switch to online borrowing, thus ending the relationship with
Technical drivers of satisfaction relate to the financing itself and their existing bank. Rather, it is likely that only a share of the fi-
thus to economic criteria. Lundahl and colleagues (2009) identify nancing is going to be conducted online. On the other hand,
fees, interest rates and required collateral as core economic com- transparent information on the company website and automatic
ponents in offline SME financing. These technical or economic quotes are likely to replace the personal relationship to a bank
drivers of satisfaction are comparable to the perceived utility, agent. Though the borrower has direct access to potential in-
which models of technology adoption propose as one of two vestors, the actual decision on the loan request and adjacent
antecedents of intention to use (Davis et al., 1989), although the variables (e.g., fees, interest) are disentangled from personal con-
drivers of perceived utility differ by model (Venkatesh et al., 2003). tact and dependent on a set of borrowing rules, which the plat-
Though intuitively, economic aspects should function as key driver form established (Kshetri, 2015). While in offline banking it was
of switching, we propose that they play a subordinate role in necessary for the bank agent to have the “authority to satisfy most
switching to crowdlending. First, previous research in the con- SME financial needs” (Madill et al., 2002, p. 93) when the company
sumer context has shown that interest rates are a limited differ- applied for a loan, a crowdlending platform communicates con-
entiating factor between banks, because they are perceived to be venience promises (e.g., duration) and economic requirements
alike (Manrai and Manrai, 2007). Second, also in the SME context, (e.g., collateral, fees) even before the borrower applies. Thus, the
the standardization of bank products has increased (Lundahl et al., importance of the personal relationship is likely to be reduced to a
2009), thus decreasing banks’ opportunity to differentiate them- minimum in favor of communication transparency. Consequently,
selves economically. Finally, because the relationship between the transparency could be an important differentiator to banks which
bank and SME is usually long, banks can more easily estimate the is going to influence switching behavior. Additionally, borrowers
risk level of the requested loan and thus require less security will not expect a close personal relationship as the service provi-
premiums in their interest rates (Madill et al., 2002). Thus, we sion is distant and non-personal from the start. Formally, we thus
expect that economic components will be a less important pre- propose:
dictor of switching behavior relative to functional components. H3. Borrowers’ switching to crowdlending platforms is positively
Formally: influenced by the process transparency perception of the financing
H1. Borrowers’ switching to crowdlending platforms is more strongly service.
driven by functional relative to economic components. H4. Borrowers’ switching to crowdlending platforms is more strongly
Functional aspects, which are not related to the financing itself influenced by process transparency relative to relational aspects of
but rather to the process of attaining it, might be the key drivers of the financing service.
SME switching to crowdlending. On aspect of functional satisfac- Extant research proposes the importance of company image for
tion is perceived convenience, which in an offline context relates SME satisfaction (Kang and James, 2004). Because the service itself
to factors such as the location and opening hours of a bank (Arbore and crowdlending companies are comparatively new to the mar-
and Busacca, 2009). In online SME financing these physical char- ket, potential borrowers will not have formed a long-standing
acteristics of the service are replaced by how conveniently the image of both. Therefore, the role of service and company image is
financing process runs for the SME. For instance, crowdlendors as important for satisfaction as it is in the adoption of new tech-
themselves positioned their offering frequently as direct and more nology (Venkatesh and Davis, 2000). The image of crowdlending
convenient form of access to capital versus slow and indirect itself is often related to innovation and growth opportunities
banks (e.g., Funding Circle advertises with slogans such as “Stop (Mills and McCarthy, 2014; Valančienė and Jegelevičiūtė, 2014).
waiting on banks” or “Direct access to capital”). Arbore and Busacca Further, companies might expect image benefits from a positive
(2009) propose reliability, speed, functionality and accuracy of the public perceptions of financing via crowdlending. For instance,
service as functional drivers of satisfaction in the retail banking Bella flame and colleagues (2015) have suggested that companies
context, which we will henceforth describe as convenience com- may use crowdlending platforms as image and publicity device.
ponents. Also adoption models discuss convenience as “ease of Consequently, the more positive an SMEs image of crowdlending
use” (Venkatesh, 2000), although its components are strongly and the platform company, the more likely the SME is to switch
linked to the adoption of technical innovations (e.g., computer away from banking. Formally:
playfulness as motivation, or computer anxiety as emotional
measure). In sum, we posit that convenience perceptions are likely H5. Borrowers’ switching to crowdlending platforms is positively
to be a core driver of switching, formally: influenced by a positive image of crowdlending.

Individuals and companies have a different openness to in-


H2. Borrowers’ switching to crowdlending platforms is positively
novations and new service offerings (Rogers, 2003) and crowdle-
influenced by their convenience perception of the financing service.
nding companies require a certain diffusion on both the borrower
Extant research has widely acknowledge the relational com- and lender side. The diffusion of innovations theory has frequently
ponent of functional satisfaction: the ability to connect with, un- been applied not only to radical innovations, but also to smaller
derstand and help the customer has been identified as main sa- behavioral changes, whose magnitude of innovation is comparable
tisfaction driver across the globe (e.g., Canada: Madill et al., 2002; to switching from banks to crowdlending, such as the adoption of
Hong Kong: Lam and Burton, 2006; UK: O'Donnell et al., 2002). more than one shopping channel (Lu and Rucker, 2006), or ethical
Further, the importance of relationship has also been widely vs. non-ethical tourism (Ganglmair-Wooliscroft and Wooliscroft,
146 E. Maier / Journal of Retailing and Consumer Services 33 (2016) 143–153

2016). In general, the more innovative a company, the earlier it PLS the more reliable estimator in this research context.
should adopt to changes (Rogers, 2003). SME, however, are ex- We first evaluate the measurement model and focus then on
pected to show only limited innovativeness in their bank relations, the structural relationship between switching behavior and its
as they are very used to the established forms of personal inter- drivers. The structural model investigates the impact of satisfac-
action (O'Donnell et al., 2002). However, the readiness to adjust to tion drivers directly, that is it leaves out intermediary aggregates
change differs among SME (e.g., Aboelmaged, 2014; Kurnia et al., or mediators, in line with previous research both in bank
2015). In the context of retail banking, Montazemi and Qahri- switching (Chakravarty et al., 2004) and later versions of adoption
Saremi (2015) find that innovative consumers are more likely to models (Venkatesh et al., 2003). Because the present research in-
switch to online banking. Because switching to alternative, web- vestigates initial switching and long-term use, we approached
based forms of financing is of similar nature, we formally propose: potential participants who had used both traditional banking and
H6. Borrowers’ switching to crowdlending platforms is positively crowdlending, in order to test for satisfaction differentials along
influenced by their openness to innovation. different dimensions as driver of switching. 104 borrowers parti-
cipated in the survey (86% were men, average age of 47.2 years,
2.2. Empirical results 93% managing directors) of which 48% had received financing and
the rest being rejected in the credit process. The survey items and
Study 1 investigates drivers of SME switching behavior from related constructs are listed in Appendix A. We used multi-item
traditional financing to crowdlending. We conducted a field study measurement for all but one construct (security risk perception),
among SME borrowers of a European crowdlending platform. To due to their higher reliability and validity (Szymanski and Henard,
test our hypotheses, we apply partial least square modeling (PLS) 2001). Following Drolet and Morrison (2001) recommendation, all
with Smart PLS 3. We use PLS because the focus of the present self-developed items were constructed such that each requests
research lies on an area which has not been extensively covered by additional information on the construct. Switching intention, as
theory (Sarstedt et al., 2014) and investigations thus profit from the focus endogenous construct, was measured with three items:
the higher statistical power of PLS (Hair et al., 2011). Further, its the likelihood to switch (S1), the likelihood to recommend crow-
robust results with smaller samples (Reinartz et al., 2009) render dlending to acquaintances (S2, both based on Madill et al., 2002)

Table 1
Construct validity (t¼ Outer t-statistic, 2-tailed significant t¼ 1.96 equals po 0.05; AVE ¼ Average Variance Extracted; CR ¼Composite Reliability; CA¼ Cronbach's Alpha;
VIF ¼Variance Inflation Factor).

Dimension Construct Item Mean SD Loading t¼ AVE CR CA VIF

Technical dimension Economics EC1 3.25 1.08 0.94 3.91 0.80 0.89 0.76 1.62
EC2 3.05 1.10 0.85 3.49 1.62

Functional dimension Convenience CO1 3.94 0.99 0.76 12.87 0.67 0.91 0.87 1.84
CO2 3.67 1.10 0.84 23.97 2.18
CO3 3.96 1.00 0.79 11.19 2.15
CO4 4.19 0.97 0.87 18.19 2.93
CO5 3.79 0.98 0.81 13.76 2.21

Process transparency PT1 4.06 1.18 0.84 16.93 0.56 0.86 0.79 2.10
PT2 4.47 0.89 0.74 9.50 1.62
PT3 3.95 1.30 0.78 14.33 1.55
PT4 4.07 1.23 0.69 6.15 1.67
PT5 3.97 1.15 0.64 4.76 1.57

Relational quality RQ1 3.55 0.94 0.67 7.82 0.69 0.90 0.84 1.43
RQ2 3.27 1.04 0.93 36.46 4.85
RQ3 3.34 0.95 0.93 32.87 5.04
RQ4 2.79 1.03 0.76 14.44 1.57

Staff quality SQ1 4.01 1.34 0.88 22.33 0.86 0.96 0.95 3.42
SQ2 3.85 1.34 0.95 88.01 5.25
SQ3 3.72 1.33 0.91 39.15 3.65
SQ4 3.74 1.29 0.96 85.81 5.72

Website quality WQ1 3.97 0.88 0.90 10.44 0.84 0.94 0.91 2.70
WQ2 3.89 0.99 0.91 10.37 3.45
WQ3 3.81 0.84 0.94 10.73 3.08

Image Image IM1 3.95 0.97 0.92 4.02 0.76 0.86 0.70 1.40
IM2 3.29 0.80 0.83 2.48 1.40

Innovativeness Innovativeness IN1 3.96 0.95 0.85 9.68 0.66 0.79 0.49 1.12
IN2 3.68 1.05 0.71 4.32 1.12

Online banking experience EO1 3.76 1.03 0.94 4.02 0.69 0.81 0.59 1.21
EO2 4.08 1.09 0.71 2.47 1.21

Traditional banking TB1 3.04 1.39 0.80 7.33 0.60 0.86 0.78 1.87
TB2 2.11 1.44 0.72 6.75 1.48
TB3 3.42 1.32 0.87 9.44 1.98
TB4 2.33 1.12 0.71 5.46 1.37
Security risk PSR 2.58 1.05 n/a n/a n/a n/a n/a n/a

Dependent variable Switching S1 3.99 1.03 0.86 23.93 0.81 0.93 0.88 1.75
S2 7.91 2.52 0.92 26.52 4.61
S3 7.70 2.80 0.92 30.62 4.64
E. Maier / Journal of Retailing and Consumer Services 33 (2016) 143–153 147

and the likelihood to reloan (S3). All items were measured with Table 3
five-point Likert Scales, except for S2 and S3 (likelihood of re- R2 and Q2 results (effect sizes based on Q2 assessment: 0.02 ¼ small, 0.15 ¼
medium, 0.35 ¼ large).
commendation, reloan likelihood), which were measured on a ten-
point scale similar to previous research (e.g., Madill et al., 2002). Endogenous latent construct R2 Q2 Effect size

2.2.1. The measurement model Switching 0.53 0.39 Large


To investigate the quality of the reflective measurement models Innovativeness 0.29 0.16 Medium

we follow suggestions by Hair and colleagues (2011) and examine


indicator loadings, composite reliability and average variance ex- switching intentions, but only marginally significantly (H6,
tracted. As shown in Table 1, all indicator loadings exceed 0.7 with
β ¼ 0.14, p¼ 0.07, f2 ¼0.04). Website (β ¼  .07, p ¼0.32, f2 ¼ 0.01)
the exception of PT4 (0.69) and PT5 (0.64). Further, all loadings are
and staff quality (β ¼ 0.01, p ¼0.17, f2 ¼0.01) did not significantly
statistically significant at the 5% level. Further, the composite re-
influence switching intention.
liability (CR) of each reflective construct exceeds 0.7. Finally, also
In sum, borrower switching to crowdlending seems to be
the average variance extracted (AVE) of each construct is higher
strongly dependent on the convenience (e.g., speed, flexibility,
than 0.5.
To assess discriminant validity, we follow Henseler et al. (2015) simplicity) of the financing, as well as on the transparency of the
in investigating the heterotrait-monotrait ratio of correlations offer (e.g., clarity, ease, sufficient information). In line with our
(HTMT) instead of the assessing Fornell-Larcker Criterion (Fornell expectations, relational aspects, which strongly influence switch-
and Larcker, 1981) and cross loading table, due to the higher re- ing in offline banking, play a significant, but far less important role
liability of the approach (Henseler et al., 2015). The HTMT criterion in the decision to switch to online banking. Borrower innova-
measures the correlations of indicators across constructs (hetero- tiveness increased the switching intention, but only marginally
trait) relative to the correlations of indicators within the same significantly. The innovativeness and publicity of crowdlending
construct (monotrait) and, thus, shows higher discriminant va- itself, however, did not induce switching.
lidity the lower the correlation coefficient. As indicated in Table 2,
discriminant validity is established as all constructs show HTMT
values well below the more rigorous threshold of 0.85 proposed by
the authors. To assess multicollinearity, we calculated variance 3. The demand side: consumer investments
inflation factors (VIFs, see Table 1) and three exceeded the critical
threshold of 5.00 (Hair et al., 2011): RQ3, SQ2, SQ4. Overall, 3.1. Theoretical background
however, reliability and validity of the constructs is sufficient to
investigate the structural model. This article aims to assess how crowdlending platforms can
bridge the gap between borrowers and lenders from a marketing
2.2.2. The structural model perspective. It does not aim and cannot contribute to a debate
Both predictive accuracy and relevance of the model are high
about the drivers or rationality of consumer investment decisions.
(see Table 3): the main structural model explains 53% of the
Rather, we aim to managerially contribute by investigating how
switching behavior with a high effect size (Q2 ¼0.39), as indicator
operators of crowdlending platforms need to structure their offer
of relevance. Further, there is no multicollinearity among the inner
paths, the highest VIF of the structural paths being 2.5. (Table 4). (available investments, framing of the descriptions, etc.) in order
As hypothesized, borrower switching to crowdlending plat- to foster consumer investments.
forms is most strongly driven by convenience (H2, β ¼0.22, It is intuitive that also consumers on crowdlending platforms
p o0.01, f2 ¼0.043, see Fig. 1) and transparency (H3, β ¼0.28, should invest more readily into assets which offer a better risk-
p o0.01, f2 ¼ 0.073) of the financing. Economics did not sig- return rate. Thus, we acknowledge and test the fundamental belief
nificantly influence switching (β ¼  0.02, p¼ 0.85, f2 ¼0.00) and that, in the absence of behavioral anomalies, expected higher re-
also influenced switching less strongly than convenience (smaller turns for less expected risk should lead to greater weight in con-
effect size, H1). Further, relational quality only had a marginally sumers’ investment portfolio (e.g., Markowitz, 1952). Further, the
significant influence on switching (β ¼ 0.21, p ¼0.07, f2 ¼0.04) and shorter the investment period, the lower the perceived risk of the
is, thus, less influential than transparency (smaller effect size, H4). investment (Sachse et al., 2012) and, hence, the propensity to in-
In contrast to H5, the image of crowdlending as innovative and
vest. Additionally, smaller loans should be seen as easier to serve
with positive publicity did not significantly influence switching
and thus less likely to default. Formally:
(β ¼0.12, p ¼0.15, f2 ¼0.02). Consumer innovativeness increased

Table 2
Discriminant validity (heterotrait-monotrait ratio of correlations, HTMT criterion, see: Henseler et al. (2015).

1 2 3 4 5 6 7 8 9 10 11

Convenience
Economics 0.32
Experience Online Banking 0.07 0.08
Image 0.67 0.39 0.16
Innovativeness 0.13 0.18 0.46 0.21
Process transparency 0.70 0.16 0.17 0.50 0.23
Relational quality 0.76 0.47 0.22 0.78 0.23 0.59
Security Risk 0.15 0.12 0.20 0.22 0.41 0.19 0.20
Staff Quality 0.41 0.11 0.30 0.27 0.08 0.64 0.47 0.22
Switching 0.67 0.17 0.16 0.59 0.32 0.70 0.66 0.28 0.44
Traditional Banking 0.21 0.14 0.35 0.19 0.63 0.32 0.23 0.09 0.16 0.34
Website Quality 0.43 0.20 0.14 0.39 0.09 0.56 0.34 0.09 0.28 0.30 0.16
148 E. Maier / Journal of Retailing and Consumer Services 33 (2016) 143–153

Table 4
Result of hypothesis testing and structural relationships.

Hypo-thesis Path Path coefficient SD t-statistic Decision

H1 Economics-Switching 0.02 0.11 0.19 Supported


(Convenience-Switching) 4 (Economics-Switching) 0.22 40.02 n/a n/a

H2 Convenience-Switching 0.22 0.10 2.30*** Supported


H3 Transparency-Switching 0.28 0.10 2.80*** Supported
H4 Relational quality-Switching 0.21 0.11 1.82* Supported
Staff quality-Switching 0.10 0.08 1.37
(Transparency-Switching) 4(Relational quality-Switching) 0.28 40.21 n/a n/a
(Transparency-Switching) 4(Staff quality-Switching) 0.28 40.10 n/a n/a

H5 Image-Switching 0.13 0.09 1.45 Not supported


H6 Innovativeness-Switching 0.14 0.08 1.82* Marginally supported

t values: **p o 0.05.


***
p o 0.01.
*
p o0.1.

Fig. 1. Structural model (*p o 0.1, **p o0.05, ***p o 0.01).

H7. The lower the risk (higher the return) the faster the financing. of the available alternatives. Specifically, we expect that if the re-
turn (risk) of a loan exceeds competition investors will be more
H8. The more liquid the investment the faster the financing.
(less) likely to invest, increasing speed and financing volume per
H9. The smaller the loan the faster the financing. investor. Further, the more alternatives are on the platform, the
slower the financing of each loan should be, as investments are
However, crowdlending platforms differ from alternative in- likely spread. Formally, we thus propose:
vestment forms in terms of available investment alternatives. Re-
lative to the stock or bond market, the investment opportunities H10. The higher the relative yield (lower the relative risk), the faster
for consumers are more restricted, as only a limited number of the financing.
loans is offered as investment targets at the same time.
H11. The fewer loans are on the platform at one time, the faster the
The fact that investors on a crowdlending platform are con-
financing.
fronted with a limited set of investment alternatives is likely going
to influence their investments. Extant research has established, Besides the financial information on the investment, consumer
that choice depends not only on a rational evaluation, but also on investors are provided with other economic criteria on crowdle-
the available options (e.g., Tversky and Kahneman, 1991). In fi- nding platforms. Research on the P2P financing of private loans
nancial choices, Kwon and Lee (2009) have established that in- has established the key role of lender economic status in invest-
vestments appear more attractive to consumers when evaluated ment decisions (e.g., Greiner and Wang, 2011). In a C2C context,
together with a safer reference with less return, driven by the criteria such as amount requested relative to income, home
perception of investments as losses, which causes a risk seeking ownership or credit history influence investment decisions (Iyer
tendency. This should also be applicable to crowdlending, where et al., 2015). In the C2B context, we expect the ratio of debt to
each investment needs to be evaluated relative to multiple alter- revenue of the company to have a negative, and the age and
natives. We propose that consumers will not only be driven by the number of employees to have a positive effect on the perceived
absolute return, but will be influenced by the return and risk level likelihood of repayment. Formally:
E. Maier / Journal of Retailing and Consumer Services 33 (2016) 143–153 149

H12. The more positive the economic company profile (debt to loan Table 6
ratio, company age, number of employees) the faster the financing. Betas (t-values) and significance levels of Study 2.

Besides economic facts about the loan and borrower, soft in- Area Variable Model 1 Model 2
formation has been proposed to be a key driver of risk evaluation
Economic Nominal Yield  0.35 0.39 (2.25)*
(Iyer et al., 2015) and, consequently, investment (Greiner and information (Percent) (  4.60)***
Wang, 2011). Various drivers have been suggested, ranging from Relative Yield (Per-  0.67
social similarity (e.g., Ahlers et al., 2015), to esthetics (e.g., Town- centage Points) (  3.91)***
***
send and Shu, 2010), writing style (Greiner and Wang, 2011), Risk Class (Aþ ¼ 1, 0.49 (4.62)  0.98
D ¼5) (  3.35)***
spatial proximity (Agrawal et al., 2011), or characteristics of the Relative Risk (Num- 1.34 (4.67)***
borrower group (Galak et al., 2011). These non-economic criteria ber of classes)
have been found to be particularly relevant for investment deci- Loan Length (Months) 0.26 (6.31)*** 0.27 (7.06)***
sions into loan requests with either very high or low risk (Berko- Credit Volume (€) 0.26 (3.81)*** 0.22 (3.22)**
Loans on platform 0.71 (16.77)***
vich, 2011). Company Employees  0.02 (  0.80)
Frequently discussed as soft information is the role of herding. information Debt Ratio (Loan vo- 0.03 (0.90)
Studying microloans, Zhang and Liu (2012) find evidence of “ra- lume / revenue)
tional herding” where investors infer higher creditworthiness from Company Age (Years)  0.01 (  0.27)
Non-economic Number of Investors 0.29 (4.66)*** 0.31 (4.92)***
a higher number of lenders, referred to as “reinforcement model”
information (number per ´000€)
(Cordova et al., 2015). We expect the presence of other investors to Innovation 0.00 (0.09)
have a positive effect on investment volume and financing speed. Tradition  0.03 (  0.89)
Formally: Growth 0.00 (0.06)
Loyalty 0.01 (0.18)
H13. The presence of other lenders increases the speed of financing. Profit 0.04 (1.42)
Text length 0.01 (0.41)
Borrowers’ description of the loan (e.g., text length, spelling Control variables Noninst. Vol. (€) 0.16 (2.23)* 0.18 (2.64)**
mistakes) has been identified as another criterion for financing Loan Date  0.43  0.48
success (Iyer et al., 2015; Greiner and Wang, 2011). Specifically, (  12.38)*** (  12.67)***
presented information is used as a signal to infer for cred- Presence Institu- 0.07 (1.38) 0.07 (1.41)
tional Investors (0¼ o,
itworthiness (Ahlers et al., 2015). Because the present research 1 ¼yes)
focuses on C2B loans, we propose that predominantly signals of Institutional Investor 0.21 (1.97)* 0.22 (2.10)*
business success (e.g., customer relationship, profitability, in- Share in Loan
novativeness) will lead to faster financing. Formally: (Percent)
R2 0.73 0.75
H14. Signals of business success (tradition, profitability, customer AIC 2199.8 2168.2
BIC 2245.8 2252.4
loyalty, etc.) will increase the speed of financing. F 81.57 45.54
n 342 338
3.2. Empirical results
***
po 0.01
*
po 0.1
We investigate a set of 342 SME loan requests on a European **
p o 0.05
crowdlending platform between June 2014 and November 2015.
For sample details, see Table 5. To investigate the hypotheses, wo
focus on speed of financing (in days) as main dependent variable presence and volume of investments from institutional investors,
(e.g., as in Galak et al. (2011)). as these would obviously shorten the funding period. Institutional
As independent variables we used loan information (yield, risk background investments are a frequent tool of platform operators
rating, length and volume of requested credit), information on to “sweep” the marketplace clean of loans which have not been
competing investment opportunities on the platform (number of financed, by filling up the remaining investment sum with in-
loans available for investment, yield and risk class of the loan re- stitutional capital (e.g., Lending Club financed 28% of its credit
lative to competition), the relative number of other investors (per volume in 2014 from institutional investors, Lending Club, 2014).
volume to be financed) and company information (revenue, loan The empirical results are summarized in Table 7. In short, all
to revenue ratio, company age). Finally, we measured the text hypotheses on the economic drivers of consumer investments are
length of the project description and coded whether the borrower supported, while “soft” information, as company details and the
mentioned various credibility claims into the request (growth, qualitative loan description, did not seem to influence investment
loyalty of customers, profitability, tradition and innovation). We decisions.
controlled for the loan date, as we expected investment time We investigate H7–,H9 by regressing the investment period on
would be a confounding variable in such a newly developing yield, risk, loan length, credit volume and the control variables.
business as crowdlending. Further control variables were the Regression results (Table 6, Model 1) support the hypothesis that
financing speed is increased at higher return or lower risk (H7).
Table 5 Loan requests with longer runtime (H8) and larger volume (H9)
Sample characteristics. require longer to be financed. However, it is obvious that a higher
credit volume should require more time to be financed. We thus
Variable Sample average
conduct a follow-up analysis, with a relative measure (investment
Company Company age (years) 13.2 volume per day) as dependent variable and find that the non-in-
Company revenue (mn EUR p.a.) 1.4 stitutional investment volume lowers the daily investment vo-
Employees 16.5 lume, though only marginally significantly (β ¼0.07, p ¼0.079), in
Loan Yield (Percent) 7.3 further support of H9.
Average duration (Months) 33.7
Average volume (‘000 EUR) 69.9
To assess the remaining hypotheses, we regressed the invest-
ment period on a whole set of independent variables (Table 6,
150 E. Maier / Journal of Retailing and Consumer Services 33 (2016) 143–153

Table 7 considerably more than economic criteria or relationship aspects.


Results of the hypotheses-testing. This could be an indication that the financial terms of crowdle-
nding platforms are perceived as only on par, and not better than
Area Hypothesis Independent variable Financing
duration offers by the SMEs’ banks, potentially driven by favorable offers
based on a long-standing relationship and an exact knowledge of
Economic H7 Yield – ✓ the SMEs’ risk profile (Madill et al., 2002). The low relevance of
Risk þ ✓ economic criteria for the switching of financial institutions mat-
H8 Loan Length þ ✓
H9 Loan Volume þ ✓
ches the findings of previous research in the retail banking context
H10 Relative yield – ✓ (Manrai and Manrai, 2007).
Relative risk þ ✓ The importance of the functional aspects convenience and
H11 Presence of other loans þ ✓ process transparency in our model contrasts, however, to the de-
Company H12 Debt to revenue ratio þ ✗ creasing role of perceived ease of use in technology adoption
information Company age – ✗ models (Davis et al., 1989) and their empirical evaluations (Szajna,
Employees – ✗ 1996). Venkatesh and Bala (2008, p. 282) argue that “once in-
Non-economic H13 Number of other len- – ✗ (rev.) dividuals get accustomed to the system and gain hands-on ex-
ders per credit volume perience with the system, the effect of perceived ease of use on
H14 Signals (customers, tra- – ✗
behavioral intention will recede into the background.” The present
dition, profit, etc.)
Textlength – ✗ findings suggest for SME bank switching, however, that con-
venience or ease of use might be a key differentiator between
different financing options for SME and thus play an increased
Model 2). We calculated relative yields and risk levels for each loan role, because the utility derived from the financing itself does not
compared to the average yield and risk levels which were pre- differ between banks and crowdlendors.
sented on the platform at the same time as the loan in question. As The relationship between borrower and financial institution,
expected, consumers evaluate the financial performance of loans key driver of offline satisfaction and switching, still matters, but
on the platform relative to other offers, with relatively higher re- far less than in traditional forms of financing. It appears that
turns or lower risks increasing the financing speed (H10). The borrowers on crowdlending platforms are not looking to replace
standardized betas and significance of the relative measures are their bank relationships because they are dissatisfied (Colgate and
higher than of the absolute variables, indicating a stronger im- Hedge, 2001), but predominantly for an alternative, flexible way of
portance of the relative perspective. As expected, the more loans financing. Further, while borrower innovativeness increases
were on the platform, the longer the financing of each loan took switching propensity, image beliefs in the innovativeness and
(H11). In summary, all variables linked directly to the credit or the publicity of crowdlending did not. In sum, this research finds that
situation in the market place were supported. though comparable service quality criteria matter for borrowers
Non-credit information on the company, from which the in- on crowdlending platforms as in offline financial services, crow-
vestors could infer creditworthiness, were not significant, how- dlending needs to exceed banking in convenience and transpar-
ever: neither the debt to revenue ratio, nor company age, nor the ence to induce switching. To attract an increased supply of bor-
number of employees significantly influenced the speed of finan- rowers to their platforms, crowdlending companies should
cing. We thus reject H11. therefore focus on further improving their convenience: finding
To assess H12, we calculated the number of bids per 1000€ ways to ensure speed and flexibility, while efficiently managing
investment volume, as the absolute number of bids would have the required documentation and securities will be key for success.
been mainly dependent on the credit volume. Contrary to our To match a growing supply of SME loan requests with sufficient
expectations, the relative number of bids, as a potential indicator investment demand, crowdlending platforms need to focus on
of positive evaluation by other investors and potential cause of how they present investment opportunities to potential investors.
herding behavior (H12), significantly decreased the speed of fi- While company information and a “soft” assessment of the loan
nancing. Loans which were financed by many lenders did not only profiles appeared to have only a limited influence on consumer
require more time to be fully funded, they also attracted sig- investment decisions, the comparison to other loans had a con-
nificantly less capital per investor (β ¼  1.2, p o0.001, regression siderable impact. Crowdlending platform operators cannot influ-
of average investment volume on the same independent variables ence the risk-return relationship, as it is influence by the company
as in Model 2). A potential explanation could be reversed caus- and loan characteristics, nor the loan length the borrower re-
ality: unsuccessful loans attract only small investments and thus quests. However, operators could facilitate investments by influ-
require many investors and a long time to be funded. encing the investment options consumers see. On the one hand,
Further, none of the trust-building signaling terms (H13: they might present fewer options to increase the financing speed
growth, loyalty of customers, profitability, tradition and innova- for the presented loans. On the other, they might offer choice sets
tion) had a significant impact on the financing speed, what could with dominant options: for instance, higher returns relative to
either have been caused by the fact that investors do not care for other available options increased financing speed, even when
the loan and company descriptions, or by our coding (we simply controlling for the higher risk. The same could work for different
checked for the presence of the above-mentioned terms). Finally, combinations of loan durations or risk levels.
also the length of the text did not matter. In sum, we reject H14. Future research on SME borrowing on crowdlending platforms
should tackle the limitations of this research. Specifically, our
borrower model contains no endogenous constructs (e.g., for sa-
4. Summary, implications, and further research tisfaction, Manrai and Manrai, 2007) between the drivers (e.g.,
convenience) and switching behavior. Further, the present work
This article investigated how to attract both borrowers and only uses intentional variables (switching, recommendation and
lenders to crowdlending platforms. On the supply side, having reloan intention) as measures of switching. Future research should
focussed on (dis-)satisfaction drivers rather than on sources of investigate the impact of short term switching intentions on the
technological adoption, we find that for inducing borrower long term SME financing portfolio. Further, it is to be investigated
switching, convenience and process transparency matter whether personal relationships become more important over time
E. Maier / Journal of Retailing and Consumer Services 33 (2016) 143–153 151

in crowdlending as in most other business relations (Madill et al., crowdlending to SME. Finally, the present research is limited to
2002), or whether the borrower motivation remains as transac- assessing drivers of investments in specific loans, and does not
tional as today. And finally, it is to be clarified whether the re- assess drivers of overall spending per consumer. Future research
placement of a trusted relationship between borrower and lender on consumer investment behavior should move beyond this focus,
by a mere platform generates borrower behavior which tries to as the total investment volume is another key driver of crowdle-
exploit investors (Llewellyn, 2005). nding growth.
On the investor side, future research should, firstly, apply re-
cent findings on the role of personality in investment decisions
(e.g., Disatnik and Steinhart (2015), Zhou and Tuan Pham (2004))
to crowdlending. Second, a more sophisticated analysis of the Appendix A
choice comparisons seems indicated, as the present research only
worked with an average of choice alternatives and did not analyze See Appendix Table A1.
the role of single alternatives in driving choice (e.g., Tversky and
Kahneman (1974)). Third, the negative effect of the presence of
other lenders on financing speed calls for research on herding in

Table A1
Survey items and related constructs.

Dimension Construct Items Source

Technical dimension Economics EC1: How do you evaluate the following criteria relative to your Self-developed
offline bank… Interest
EC2: Fees

Functional dimension Convenience CO1: How do you evaluate the following criteria relative to your Self-developed
offline bank (Bank better vs. Crowdlending Platform better)…
Speed
CO2: Flexibility
CO3: Required securities
CO4: Simplicity of the process
CO5: Simplicity of the legal requirements
Process transparency PT1: The credit process was transparent at all times. Self-developed
PT2: The application process was easy.
PT3: I had sufficient support when I needed it.
PT4: I felt lost at times (reversed).
PT5: I frequently had questions which I could not answer myself
(reversed).
Relational Quality RQ1: How do you evaluate the following criteria relative to your Self-developed
offline bank … Ease to reach
RQ2: Trust
RQ3: Reliability
RQ4: Personal relationship
Staff quality SQ1: How satisfied were you with the staff… Friendliness? Self-developed
SQ2: Competence
SQ3: Speed
SQ4: Trustworthiness
Website quality WQ1: I like the way the internet site looks. Adapted from Mathwick et al., 2001; Parasura-
WQ2: This site makes it easy to find what I need. man et al., 2005; Wolfinbarger and Gilly, 2003
WQ3: The website has comprehensive information.

Image IM IM1: How do you evaluate the following criteria relative to your Self-developed
offline bank … Innovation
IM2: Publicity
Innovativeness Innovativeness IN1: In general, you are among the first in your circle of friends to Adapted from Parasuraman, 2000; Goldsmith
acquire new technology when it appears. and Hofacker, 1991
IN2: I would never buy something new, before many people are
already using it. (reversed)
Experience with online EO1: I have a lot of experience with online banking. Self-developed
banking EO2: For my business I frequently conduct financial transfers
online.
Traditional banking TB1: Personal contact is very important to me when applying for Self-developed
financing.
TB2: A purely online credit application process cannot work.
TB3: I need a contact person to discuss my questions and
problems.
TB4: In financing, I preferably trust my established business
partners.
Perceived security risk PSR: If you send information over the internet, you never know Self-developed
where they end up.

Dependent variable Switching S1: How likely is it that you will prefer crowdlending to a tradi- Adapted from (Madill et al., 2002)
tional form of financing in the future?
S2: How likely is it that you would recommend the company to
a friend, family member or business partner?
S3: How likely is it that you will apply for another credit with
the company?
152 E. Maier / Journal of Retailing and Consumer Services 33 (2016) 143–153

References propensity on the evaluation of financial products. J. Bus. Res. 62 (7), 719–725.
Lam, R., Burton, S., 2006. SME banking loyalty (and disloyalty): a qualitative study
in Hong Kong. Int. J. Bank Mark. 24 (1), 37–52.
Aboelmaged, M.G., 2014. Predicting e-readiness at firm-level: an analysis of tech- Lam, S.Y., Shankar, V., Bvsan Murthy, M.K.E., 2004. Customer value, satisfaction,
nological, organizational and environmental (TOE) effects on e-maintenance loyalty, and switching costs: an illustration from a business-to-business service
readiness in manufacturing firms. Int. J. Inf. Manag. 34, 639–651. context. J. Acad. Mark. Sci. 32 (3), 293–311.
Agrawal, A.K., Catalini,C., Goldfarb, A., 2011. The Geography of Crowdfunding (NBER Lending Club, 2014. How has Lending Club's investor base changed? Retrieved June
Working Paper No. 16820). Retrieved June 27, 2016, from 〈http://www.nber.org/ 27, 2016, from kb.lendingclub.com/investor/articles/Investor/How-has-Lend-
papers/w16820〉. ing-Club-s-investor-base-changed/?l¼en_US&fs ¼RelatedArticle.
Agrebi, S., Jallais, J., 2015. Explain the intention to use smartphones for mobile Levesque, T., McDougall, G.H.G., 1996. Determinants of customer satisfaction in
shopping. J. Retail. Consum. Serv. 22, 16–23. retail banking. Int. J. Bank Mark. 14 (7), 12–20.
Ahlers, G.K.C., Cumming, D., Günther, C., Schweizer, D., 2015. Signaling in equity Llewellyn, D.T., 2005. Trust and confidence in financial services a strategic chal-
crowdfunding. Entrep.: Theory Pract. 39 (4), 955–980. lenge. J. Financ. Regul. Compliance 13 (4), 333–346.
Ajzen, I., Fishbein, M., 1980. Understanding Attitudes and Predicting Social Beha- Lu, Y., Rucker, M., 2006. Apparel acquisition via single vs. multiple channels: college
viour. Prentice Hall, Upper Saddle River NJ. students’ perspectives in the US and China. J. Retail. Consum. Serv. 13 (1),
altfi DATA, 2016. Liberum AltFi Volume Index UK and Index Continental Europe. 35–50.
Retrieved June 27, 2016, from 〈www.altfi.com/data/indices〉. Lundahl, N., Vegholm, F., Silver, L., 2009. Technical and functional determinants of
Arbore, A., Busacca, B., 2009. Customer satisfaction and dissatisfaction in retail customer satisfaction in the bank-SME relationship. Manag. Serv. Qual. 19 (5),
banking: exploring the asymmetric impact of attribute performances. J. Retail. 581–594.
Consum. Serv. 16, 271–280. Madill, J.J., Feeney, L., Riding, A., Haines Jr., G.H., 2002. Determinants of SME owners'
Back, A., 2016, May 9. Lending Club: When the Disrupter Gets Disrupted. The Wall satisfaction with their banking relationships: a Canadian study. Int. J. Bank
Street Journal. Retrieved from 〈www.wsj.com/articles/lendingclub-when-the- Mark. 20 (2), 86.
disrupter-gets-disrupted-1462813571〉. Manrai, L.A., Manrai, A.K., 2007. A field study of customers’ switching behavior for
Belleflamme, P., Omrami, N., Peitz, M., 2015. The Economics of Crowdfunding bank services. J. Retail. Consum. Serv. 14 (3), 208–215.
Platforms (CORE Discussion Paper No. 2015/15). Retrieved June 27, 2016, from Marangunić, N., Granić, A., 2015. Technology acceptance model: a literature review
〈www.uclouvain.be/cps/ucl/doc/core/documents/coredp2015_15web.pdf〉. from 1986 to 2013. Univers. Access Inf. Soc. 14 (1), 81–95.
Berkovich, E., 2011. Search and herding effects in peer-to-peer lending: evidence Markowitz, H., 1952. Portfolio selection. J. Financ. 7 (1), 77–91.
from prosper.com. Ann. Financ. 7 (3), 389–405. Mathwick, C., Malhotra, N., Rigdon, E., 2001. Experiential value: conceptualization,
Bundesministerium für Wirtschaft und Energie, 2015. Wirtschaftsmotor Mittel- measurement and application in the catalog and Internet shopping environ-
stand – Zahlen und Fakten zu den deutschen KMU. Retrieved June 27, 2016, ment☆1. J. Retail. 77 (1), 39–56.
from 〈www.bmwi.de/BMWi/Redaktion/PDF/W/wirtschaftsmotor-mittelstand- McLannahan, B., 2015, April 15. Lending Club strikes $150m deal with Citigroup.
zahlen-und-fakten-zu-den-deutschen-kmu〉. The Financial Times, Retrieved from 〈www.ft.com/cms/s/0/8602eb62-e29c-
Chakravarty, S., Feinberg, R., Rhee, E.-Y., 2004. Relationships and individuals' bank 11e4-ba33–00144feab7de.html#axzz4CUndZKU2〉.
switching behavior. J. Econ. Psychol. 25, 507–527. Mills, K.G., McCarthy, B., 2014. The State of Small Business Lending: Credit Access
Colgate, M., Hedge, R., 2001. An investigation into the switching process in retail during the Recovery and How Technology May Change the Game (Harvard
banking services. Int. J. Bank Mark. 19 (5), 201–212. Business School Working Paper No. 15-004). Retrieved June 27, 2016, from
Cordova, A., Dolci, J., Gianfrate, G., 2015. The determinants of crowdfunding suc- 〈www.hbs.edu/faculty/Publication%20Files/15–004_09b1bf8b-eb2a-4e63–
cess: evidence from technology projects. Procedia - Soc. Behav. Sci. 181, 9c4e-0374f770856f.pdf〉.
115–124. Molina, A., Martín-Consuegra, D., Esteban, Á., 2007. Relational benefits and custo-
Cushing, C., 2015, September 30. SoftBank leads $1 billion investment in U.S. fintech mer satisfaction in retail banking. Int. J. Bank Mark. 25 (4), 253–271.
startup SoFi. Reuters. Retrieved from 〈www.reuters.com/article/2015/10/01/us- Montazemi, A.R., Qahri-Saremi, H., 2015. Factors affecting adoption of online
softbank-group-investment-sofi-idUSKCN0RV3CV20151001〉. banking: a meta-analytic structural equation modeling study. Inf. Manag. 52
Davis, F.D., Bagozzi, R.P., Warshaw, P.R., 1989. User acceptance of computer tech- (2), 210–226.
nology: a comparison of two theoretical models. Manag. Sci. 35 (8), 982–1003. Naser, A.J.K., 2002. Customer satisfaction and retail banking: an assessment of some
Disatnik, D., Steinhart, Y., 2015. Need for cognitive closure, risk aversion, un- of the key antecedents of customer satisfaction in retail banking. Int. J. Bank
certainty changes, and their effects on investment decisions. J. Mark. Res. 52 (3), Mark. 20 (4), 146–160.
349–359. NSR Invest., 2016. NSR Platform Market Summary, from 〈www.nsrplatform.com/
Drolet, A.L., Morrison, D.G., 2001. Do we really need multiple-item measures in #!/〉.
service research? J. Serv. Res. 3 (3), 196. O'Donnell, A., Durkin, M.G., McCartan-Quinn, D., 2002. Corporate banking in the
Ennew, C.T., Binks, M.R., 1996. The impact of service quality and service char- UK: personal vs remote interaction. Int. J. Bank Mark. 20 (6), 273–284.
acteristics on customer retention: small businesses and their banks in the UK1. Parasuraman, A., 2000. Technology Readiness Index (Tri): a multiple-item scale to
Br. J. Manag. 7 (3), 219. measure readiness to embrace new technologies. J. Serv. Res. 2 (4), 307–320.
Fishbein, M., Ajzen, I., 1975. Belief, Attitude, Intention and Behavior: An Introduc- Parasuraman, A., Zeithaml, V.A., Malhotra, A., 2005. E-S-QUAL: a multiple-item
tion to Theory and Research. Addison-Wesley, Reading, MA. scale for assessing electronic service quality. J. Serv. Res. 7 (3), 213–233.
Fornell, C., Larcker, D.F., 1981. Evaluating structural equation models with un- Reinartz, W., Haenlein, M., Henseler, J., 2009. An empirical comparison of the ef-
observable variables and measurement error. J. Mark. Res. 18 (1), 39–50. ficacy of covariance-based and variance-based SEM. Int. J. Res. Mark. 26 (4),
Funding Circle, 2016. Company landing page., Retrieved June 27, 2016, from 〈www. 332–344.
fundingcircle.com/us/〉. Rogers, E.M., 2003. Diffusion of Innovations 5th ed.. Free Press, New York, NY.
Galak, J., Small, D., Stephen, A.T., 2011. Microfinance decision making: a field study Sachse, K., Jungermann, H., Belting, J.M., 2012. Investment risk – the perspective of
of prosocial lending. J. Mark. Res. 48, S130–S137. individual investors. J. Econ. Psychol. 33 (3), 437–447.
Ganglmair-Wooliscroft, A., Wooliscroft, B., 2016. Diffusion of innovation: the case of Sarstedt, M., Ringle, C.M., Smith, D., Reams, R., Hair, J.F., 2014. Partial least squares
ethical tourism behavior. J. Bus. Res. 69 (8), 2711–2720. structural equation modeling (PLS-SEM): a useful tool for family business re-
Goldsmith, R.E., Hofacker, C.F., 1991. Measuring consumer innovativeness. J. Acad. searchers. J. Fam. Bus. Strategy 5 (1), 105–115.
Mark. Sci. 19 (3), 209–221. Shih, H.-P., 2004. Extended technology acceptance model of Internet utilization
Greiner, M.E., Wang, H., 2011. Building Consumer-to-Consumer Trust in E-Finance behavior. Inf. Manag. 41 (6), 719–729.
Marketplaces: An Empirical Analysis. Int. J. Electron. Commer. 15 (2), 105–136. Sriram, S., Manchanda, P., Bravo, M.E., Chu, J., Ma, L., Song, M., Subramanian, U.,
Grönroos, C., 1990. Service Management and Marketing : Managing the Moments 2015. Platforms: a multiplicity of research opportunities. Mark. Lett. 26 (2),
of Truth in Service Competition. Lexington Books, Lexington, MA. 141–152.
Hair, J.F., Ringle, C.M., Sarstedt, M., 2011. PLS-SEM: indeed a silver bullet. J. Mark. Szajna, B., 1996. Empirical evaluation of the revised technology acceptance model.
Theory Pract. 19 (2), 139–152. Manag. Sci. 42 (1), 85–92.
Henseler, J., Ringle, C., Sarstedt, M., 2015. A new criterion for assessing discriminant Szymanski, D., Henard, D., 2001. Customer satisfaction: a meta-analysis of the
validity in variance-based structural equation modeling. J. Acad. Mark. Sci. 43 empirical evidence. Acad. Mark. Sci. 29 (1), 16–35.
(1), 115–135. Townsend, C., Shu, S.B., 2010. When and how aesthetics influences financial deci-
Iyer, R., Khwaja, A.I., Luttmer, E.F.P., Shue, K., 2015. Screening peers softly: inferring sions. J. Consum. Psychol. 20 (4), 452–458.
the quality of small borrowers. Manag. Sci. 62, 1554–1577. Trayler, R., Nielson, J., Jones, R., 2000. How small business firms select a bank:
Kang, G., James, J., 2004. Service quality dimensions: an examination of Grönroos's comparisons between the United States and Australia. J. Financ. Serv. Mark. 5
service quality model. Manag. Serv. Qual. 14 (4), 266–277. (1), 73–85.
Kobe, K., 2012. Small Business GDP: Update 2002–2010, Retrieved June 27, 2016, Tversky, A., Kahneman, D., 1974. Judgment under uncertainty: heuristics and biases.
from United States Small Business Administration: 〈www.sba.gov/sites/default/ Science 185 (4157), 1124–1131.
files/rs390tot_0.pdf〉. Tversky, A., Kahneman, D., 1991. Loss aversion in riskless choice: a reference-de-
Kshetri, N., 2015. Success of crowd-based online technology in fundraising: an in- pendent model. Q. J. Econ. 106 (4), 1039–1061.
stitutional perspective. J. Int. Manag. 21 (2), 100–116. Valančienė, L., Jegelevičiūtė, S., 2014. Crowdfunding for creating value: stakeholder
Kurnia, S., Choudrie, J., Mahbubur, R.M., Alzougool, B., 2015. E-commerce technol- approach. Procedia - Soc. Behav. Sci. 156 (26), 599–604.
ogy adoption: a Malaysian grocery SME retail sector study. J. Bus. Res. 68 (9), Venkatesh, V., 2000. Determinants of perceived ease of use: integrating control,
1906–1918. intrinsic motivation, and emotion into the technology acceptance model. Inf.
Kwon, K.-N., Lee, J., 2009. The effects of reference point, knowledge, and risk Syst. Res. 11 (4), 342–365.
E. Maier / Journal of Retailing and Consumer Services 33 (2016) 143–153 153

Venkatesh, V., Bala, H., 2008. Technology acceptance model 3 and a research predicting etail quality. J. Retail. 79 (3), 183–198.
agenda on interventions. Decis. Sci. 39 (2), 273–315. Zhang, J., Liu, P., 2012. Rational herding in microloan markets. Manag. Sci. 58 (5),
Venkatesh, V., Davis, F.D., 2000. A theoretical extension of the technology accep- 892–912.
tance model: four longitudinal field studies. Manag. Sci. 46 (2), 186–204. Zhou, R., Tuan Pham, M., 2004. Promotion and prevention across mental accounts:
Venkatesh, V., Morris, M.G., Davis, G.B., Davis, F.D., 2003. User acceptance of in- when financial products dictate consumers' investment goals. J. Consum. Res.
formation technology: toward a unified view. MIS Q. 27 (3), 425–478. 31 (1), 125–135.
Wolfinbarger, M., Gilly, M.C., 2003. eTailQ: dimensionalizing, measuring and

S-ar putea să vă placă și