From the classical art to the urban art infusion effect: The effect of street art and graffiti on the consumer evaluation of products

Combinations of art and products are a classic and current topic. Examples like the collaboration of the Medici with artists in the Renaissance or the logo development of Chupa Chups by Salvador Dali are historical examples. The BMW art cars by Jeff Koons and Cao Fei or the art-based special editions of Louis Vuitton bags are current best practices. All these cases expected a positive impact of art on the product or brand evaluation. This spillover effect was coined “ art infusion effect ” by Hagtvedt and Patrick. This, as well as further studies on the art infusion effect, are predominantly concerned with classical fine arts. However, despite an observable increase of urban arts-brands collaborations, the effects of these have not been researched. Our study determines that graffiti and street art are perceived by consumers as art. To confirm the art infusion effect for urban art, a laboratory experiment was conducted. The presence of urban art has a favourable influence on the evaluation of products. These results replicate and extend the findings of Hagtvedt and Patrick. As drivers of the urban art infusion effect, we also identify two additional drivers: the fit between the art and the product and the “ lifestyle perception ” .


| INTRODUCTION
In a recent state-of-the-art article about the collaboration between art and brands, Baumgarth (2018) identified 14 avenues for further research.One major topic is the question about the robustness and the broadening of the art infusion effect, because this framework is one of the major pillars for the explanation of the impact of art and brand collaborations on the brand and product evaluation.for the intensive and fruitful collaboration between art and companies in the context of architecture and product design (e.g., Bayley, 1985).
From a behavioural point of view, it is interesting to understand the effect of collaborations between art and companies on the consumer evaluation of products, services and brands.The so-called art infusion effect by Hagtvedt and Patrick (2008a) is the main theoretical and empirical tested foundation for the analysis of such consumer effects.The main proposition of the art infusion effect is that the integration of art always has a positive effect on the product and/or brand evaluation, independent of the art content and its evaluation by consumers.A prerequisite for this positive impact of art is that consumers recognize the art integration as art (Hagtvedt & Patrick, 2008a, 2011).
To date, art infusion research focuses mainly on classical fine art (see Section 2).However, in the marketplace and the art world, an increasing number of brands collaborate with contemporary artists, in particular with graffiti and street artists.For example, the French cognac brand Hennessy offered special editions by urban artists (e.g.Shepard Fairey, Futura), the brand Unilever AXE hired graffiti artists for an outdoor mural, the German beer brand Warsteiner launched two series of special beer bottles in collaboration with twelve urban artists, and the Korean technology brand Samsung created an interactive art experience in collaboration with urban artists and augmented reality specialists in 2018 during the Art week in Berlin.
Urban art in general is often characterized by its illegality, the temporary nature and its presentation in the public space (e.g.Blanché, 2012;Reinecke, 2012).Graffiti and street art represent two different types of urban art (see Table 1).
Up to now, marketing and brand literature has almost entirely ignored the link between urban art and brands (for exceptions, see Baumgarth, 2015;Borghini, Visconti, Anderson, & Sherry, 2010;Derwanz, 2013).This article tries to close this gap by answering three main research questions: 1. Do consumers perceive graffiti and/or street art as art, which is a prerequisite of the art infusion effect?2. Is the art infusion effect also valid in the case of urban art?Can we extend the fine art infusion effect to an urban art infusion effect?
3. What are the drivers of the urban art infusion effect?

| The art infusion effect
The "art infusion effect" has become one of the dominant theoretical pillars in the field of arts and branding (e.g., 150 entries in Google Scholar, December 2019).The concept was originated by Hagtvedt and Patrick (2008a), whose seminal paper "Art infusion: The influence of visual art on the perception and evaluation of consumer products" in the Journal of Marketing Research (citations in Google Scholar: 348, December 2019) distinguishes two variants of the effect and reports the hypotheses supported by the findings of three empirical studies.Figure 1 summarizes the results of the third of these.The third experiment tested the art infusion effect for soap dispensers and incorporates both types of the effect.
The two variants of the art infusion effect are here differentiated as "art infusion effect 1" and "art infusion effect 2".The first of those hypothesized that the integration of art has a positive effect on product and brand evaluation.A prerequisite of this positive spillover effect is that the consumer recognizes the artwork as art.Additionally, a more implicit assumption is that the consumer likes the artwork.In Figure 1, the comparison of the Monet and the photo soap dispenser treatments is the empirical evidence for the art infusion effect 1.The finding for the art infusion effect 2 is more surprising, because that variant expects a positive spillover effect from an artwork, which is evaluated negatively by the consumers, on the product or brand

Historical origin
Since the 1960s in the USA (Philadelphia and later New York), since 1975 spraying of trains; comeback of graffiti by the link to the hip-hop movement in the 2000s France (Paris) since the 1980s Primary goals of the artists increasing of the artist's awareness in the graffiti scene ("Getting-up").quantity of own tags challenging spaces increasing of the artist's awareness in the general public reach of non-arts interested target groups and trigger for political and social discussions quality of the artworks

Main elements of design
Tags and pieces; main focus: letters and numbers ("writing") Logos, styles, pictures

Pivotal techniques
Spraying, writing by markers, paint roller Spraying, stencils, poster, sticker, 3D objects, cut-outs, urban knitting, adbusting.Link of the street art pieces with photography and video evaluation.The significant differences seen in Figure 1 between the evaluation of the "Turner group" (negative evaluated art piece by the consumers) and the non-art soap dispenser treatments offers the empirical evidence for the art infusion effect 2. Hagtvedt and Patrick justify the existence of both variants of the effect by reference to the positive impact of the integration of art in generalpositive and negativeon the luxury perception of the product or brand.
A neuro-imaging study by Lacey et al. (2011) has shown that art images activate, in contrast to non-art images, reward-related regions of the brain: the ventral striatum, hypothalamus and orbitofrontal cortex.
The original paper by Hagtvedt and Patrick (2008a) stimulated many researchers to consider the art infusion effect in their own studies, with respect to the theoretical foundation, generalizability testing, and broadening of the domain.Table 2 presents an overview of the empirical art infusion studies.

The overview of the extant art infusion research presented in
Table 2 shows that all studies have focused on visual art, in particular fine art, with an emphasis on the classical masterpieces, and mainly on the art infusion effect 1.So far, 27 empirical studies with around 5,000 participants in Europe, Asia and America have tested the robustness and boundaries of the effect.All have more or less confirmed the art infusion effect.Based on the original study, others have incorporated further factors such as the type of artwork (iconic versus non-iconic; realist versus abstract), the product category (hedonic versus utilitarian; luxury; retailing) and consumer characteristics (regulatory focus; openness to art; desire to signal status or extroversion; desire for distinction).However, so far almost all studies have focused on fine art.As sketched in Section 1, one of the "newest" trends in art is urban art.Hence the goal of our study is to test the art infusion effect for this type of art.On the basis of the theoretical explanation and the empirical findings of the existing art infusion research, we formulate the following hypothesis: H1.The integration of urban art in a product, communication or brand has a positive impact on the product and/or brand evaluation.

| Drivers of the (urban) art infusion effect
Also, this study explores possible explanations for the urban art infusion effect.The study by Hagtvedt and Patrick (2008a) justifies the art infusion effect through the increase in the "luxury perception".
The authors assume that art in itself leads to an increased luxury perception of the product and that this increased luxury perception improves the product evaluation.The following hypotheses can be derived from this: H2a.The art perception of a product, communication or brand improves the luxury perception of a product and/or brand.
H2b.The level of the luxury perception impacts the evaluation of a product and/or brand.
The original paper by Hagtvedt and Patrick (2008a) and following studies (e.g.Lee, Chen, & Wang, 2015;Peluso, Pino, Amatulli, & Guido, 2017) explained the art infusion effect mainly by the impact of the art on the luxury perception.In addition to this "luxury" effect of art, classical theories on image spillover (e.g.Dancin & Smith, 1994;Magnusson, Krishnan, Westjohn, & Zdravkovic, 2014) proposed a direct transfer from (urban) art associations on brand image (e.g., Masè, Cedrola, & Cohen-Cheminet, 2018).In addition to luxury, art in general and urban art in particular are connected with further associations such as creativity, youth and urban culture (e.g., Borghini et al., 2010).Therefore, we proposed the following hypotheses: H3a.The art perception of a product, communication or brand improves the lifestyle perception of a product and/or brand.
H3b.The level of the lifestyle perception impacts the evaluation of a product and/or brand.
Particularly in related fields like testimonials (e.g.Choi & Rifon, 2012;Till & Busler, 2000) or brand alliances (e.g., Park, Jun, & Shocker, 1996;Simonin & Ruth, 1998), the fit between the two objects (e.g., brand A and B) are pivotal explanations for the evaluation of the joint object.This positive effect of the fit was also analysed and partly supported in the art infusion research (e.g., Hagtvedt & Patrick, 2011;Kim, Ko, & Lee, 2012).In addition to that, a study by Hagtvedt and Patrick (2008b) has shown that the integration of art in the context of brand extension can improve the fit.Considering these arguments, we suppose that the art perception increases the fit and the fit has a positive impact on the product evaluation.Hence: H4a.The art perception of a product, communication or brand improves the fit between the visual and the product and/or brand.
H4b.The fit between the visual and the product and/or brand impacts the evaluation of a product and/or brand.
Figure 2 summarizes the six hypotheses in a causal model.

| METHOD AND RESULTS
Two empirical studies were conducted to answer the postulated research questions and the hypothesis.In a first qualitative study, it was investigated whether the consumer even recognizes urban art as art at all.This perception is a prerequisite of the urban art effect.
Additionally, the exploratory qualitative study selected suitable product categories and art pieces for the main study.The main study tested the formulated hypotheses in a laboratory study in conjunction with a causal modelling approach.

| Design, sample and measurements
The first study was an exploratory study with the main goals of (a) answering the first research question and (b) checking whether the manipulations of the independent variables had been effective.
Twenty-two respondents were confronted with seven images in a randomized order.Three of the images were fine art pieces (van Gogh, Miró, Mondrian), two street art pieces (both by Banksy), one graffiti artwork (The Dixons) and one non-art photo as control stimuli.We asked participants to look at the images and report on separate 7-point Likert scales whether each image was a work of art (1 = "definitely" and 7 = "definitely not").Additionally, we tested the familiarity with the art pieces and the fit between the art pieces and different product categories.

| Results
The exploratory study shows that urban art (graffiti and street art) was perceived as art (see Figure 3).However, the graffiti art piece

| Design, sample and measurements
The main study was a laboratory experiment with a 2 (product cate-

| Results
In a first step, the arts perception as a manipulation check was The formal hypothesis test is based on a series of t-tests with a comparison of one "art packaging" with the "non-art packaging" for both product categories.In addition to the t-test, we conducted an ANOVA for the overall test.The results confirm for both product categories (F beer = 5.901**; F household detergent = 3.299*) a significant impact of the art type on the product evaluation.See Table 3.
Firstly, the results replicate the findings of the study by Hagtvedt and Patrick (2008a) for the fine arts.This close replication increases the generalizability of the findings for a different country (United States vs. Germany) and for additional products (original study: set of silverware, soap dispenser vs. our study: beer, household detergent).
This finding is further evidence of the classical "fine art infusion effect".
Additionally, the results confirm the art infusion effect also for urban art.For both product categories, the product evaluation of the street art version is significantly more positive than the non-art version.The graffiti version is only significantly better for the beer category.Hence, we find empirical support for the extensions of the art infusion effect.We coined this effect the "urban art infusion effect".

| Causal model test
For the analysis of the structural model we used the variance-based software ADANCO (Henseler & Dijkstra, 2015).The software solution employs a partial least square path modelling approach to test our hypotheses.Among variance-based SEM methods, partial least square (PLS) modelling is the most developed and general model (McDonald, 1996) and has been coined as the "silver bullet" (Hair, Ringle, & Sarstedt, 2011).In order to test our model, we followed the guidelines and recommendations by Henseler, Hubona, and Ray (2016) and Benitez, Henseler, Castillo, and Schuberth (2020).We firstly conducted the analysis for the beer category and the household detergent separately and secondly for the united data set (full model).For the calculation of the model we considered the relationships of H2a-H4b.In addition to that, we integrated for the test of the mediation effect a direct link between art perception and product evaluation as well as the correlations between the three constructs luxury perception, lifestyle perception and fit.
Table 4 summarizes the results of steps 1 and 2. In general the findings of steps 1 and 2 confirmed the quality of the models and their measurements.
Hence, we can use the empirical data and model specification for the third step.The third step focuses on the structural model and the test of the hypotheses.We consider and report the explained variance (R 2 and adjusted R 2 ) and the significance of the path coefficients.
Table 5 summarizes the findings.
The findings support the reasoning by Hagtvedt and Patrick (2008a).The luxury perception is an important explanation for the art infusion effect (H2a/2b).However, the results show that lifestyle perception (H3a/3b) is an additional important driver of the product evaluation.Particularly for the beer category, the art perception has a stronger impact on the lifestyle factor than on the luxury factor.
Finally, the art infusion effect also depends on the fit construct.Only in the beer category does the art perception have a significant impact on the fit evaluation (H4a) .That means that the art perception does not have a very strong effect on the fit evaluation.However, in all three models the fit has a significant impact on the product evaluation (H4b).Also, the findings for the hypotheses in conjunction with the non-significant link between art perception and product evaluation in all three models confirm that the art infusion effect is a fully mediated effect via the three constructs luxury perception, lifestyle perception and fit (e.g.In summary, the proposed model broadens the explanations for the art infusion effect and reaches, with an explained variance of 51-60%, satisfying levels of explanatory power.

| General findings and management implications
Urban art is in the art scene one of the most important and "hottest" art trends in recent years.Furthermore, an increasing number of companies collaborate with art and in particular with street art and graffiti.
Therefore, a scientific and empirical analysis of the art infusion effect for urban art is a missing, logical, necessary and important step.Additionally, the study pinpoints the relevance and the value of replications.Finally, the "lifestyle perception" and the "fit dimension" broaden the spectrum of possible drivers of the art infusion effect.In particular, a deeper understanding of the reasons behind the "(urban) art infusion effect" could be an interesting field for further research.
By establishing the urban art infusion effect, we emphasize that also in the case of urban art, the mere presence of art has a positive effect on the product evaluation.Hence, we support the robustness of the art infusion effect and thereby the cooperation between companies and art.Also, we show that it could be interesting for brand managers to actively join forces with artists to promote co-branding initiatives or related measures (e.g.special arts editions) independent of whether the brand seeks a "relation" with fine art or urban art.
Indeed, urban art might have a higher fit to specific types of brands than fine art.Particularly, this might be the case for brands aiming to attain a wild, rocking, youthful and fancy image.However, a lot of management-relevant questions are unanswered so far.

| Theoretical implications
Firstly, the study is a replication of the study by Hagtvedt and Patrick.
The findings support the results of the original study for a different country and additional product categories.Hence, the study responds to the call for more replications (e.g., Evanschitzky, Baumgarth, Hubbard, & Armstrong, 2007;Hubbard, 2017) and extends the existing perspective.Also, the study incorporates with the "lifestyle perception" and "fit" alternative explanations for the art infusion effect.

| Limitations and further research
To the best of our knowledge, this is the first study on the art infusion effect with urban art.By confirming our hypotheses, we contribute to the literature by expanding the art infusion effect to the "urban art infusion effect" and thereby paving the way for more research.For example, the danger of a wear-out effect by using art-brand collaborations over a long period of time and/or for a high number of brands or products in one sector is not addressed in the research.Also, the relevance of the fit between art and product category or brand is not analysed in detail (Kim, Vaidyanathan, Chang, & Stoel, 2018).In addition to that, the possible danger of art-brand collaborations on brand constructs like brand trust and reputation is unclear.Furthermore, we only consider "luxury" and "lifestyle perception" as potential mediators from the art to the product and/or brand.The perception of art in general and the perception of an art-infused product or brand is a more complex perception.The research on aesthetic emotions regarding art and other objects (Hager, Hagemann, Danner, & Schankin, 2012;Schindler et al., 2017) could be a good starting points for a deeper understanding of the "secrets" of the art infusion effect.
Finally, the management and organizational challenges of an artistbrand-collaboration are not considered in our model.

ORCID
At first sight, art and companies are two different and antagonistic worlds.However, in the past and present, companies actively collaborate with art and artists.Historic posters by the French artist Henri de Toulouse-Lautrec (e.g.Moulin Rouge), the BMW Art Cars by the pop artist Andy Warhol or the collaboration of the Japanese comic artist Takashi Murakami with the luxury brand Louis Vuitton are only a few examples.Artistic movements like the Arts and Crafts in the UK or the Werkbund and Bauhaus in Germany are prominent historical examples reaches a lower level of "art perception" than the two street art pieces.These results answer the first research question: Consumers perceive all examples of urban art as art, whereas graffiti (The Dixons) reaches a lower art level than street art (Banksy).F I U R E 2 Drivers of the urban art infusion effectAdditionally, the exploratory study measures the fit between eight product categories and urban art.On this basis, for the main study a product category with a high fit (M beer = 5.0) and low fit (M household detergent = 2.8) was selected.
gory: beer, household detergent) × 4 (art type: fine art, graffiti, street art, non-art) design.The aim was to provide answers to the second and third research questions and to test H1.Each participant evaluated one beer and one household detergent in a randomized order.The 255 respondents (Germany, summer 2017, online survey) saw the manipulated images (see Appendix 1) and the whole questionnaire on screen.After the elimination of incomplete questionnaires and answers with a very short response time (< 2.5 minutes), 216 questionnaires (n = 432 cases) were considered for data analysis.The sample covers the German population (sex: 49% female, 51% male; age: < 31 years, 38%; 31-40 years, 34%; > 40 years, 28%).After the presentation of the manipulated product packaging with different types of art or without art (control situation), the participants evaluated (1) the product (four items, α = 0.94), (2) the "luxury perception" (three items, α = 0.85), (3) the "lifestyle perception" (three items, α = 0.89), (4) the fit (one item), and (5) the art perception (one item) (see Appendix 2).To test the formal hypothesis, t-tests and ANOVAs were conducted.
analysed.The results (M fine art = 5.21; M street art = 5.30; M graffiti = 4.62; M non-art = 2.38) confirm the manipulation and support the results of the exploratory study regarding the art character of urban art.

F
I G U R E 3 Art perception of classical and urban art [Colour figure can be viewed at wileyonlinelibrary.com] and consider the cut-off value of 0.08.All three models fulfil this threshold = 0.037).In addition to that, we consider the two goodness of overall model fit criteria d ULS and d G(Benitez et al., 2020).The goodness of fit of all the models violates slightly the recommended thresholds of d ULS d G , but the values are acceptable particularly for the two product-specific models with smaller sample sizes (n = 216) in comparison to the overall model (n = 432).