Overspending on smartphone purchases among Swedish young adults

Abstract

In the Nordic countries with growing markets for consumer credit, a concern is that consumption desires in conjunction with easily accessible credit make financially constrained young adults vulnerable to problem debt and over-indebtedness. In addressing this concern empirically, we investigate whether retail offers of instalment payments of discounted cash prices tempt young adults to finance purchases of more expensive premium smartphones than they would purchase by cash payment. Descriptions of smartphones ranging from budget to latest premium models are in an online experiment presented to 152 Swedish young adults between 18 and 25 years. We employ a within-group design requiring the participants in counterbalanced order to choose a preferred smartphone twice, either if the default choice is paying the regular cash price or two-year monthly instalments with a 20% discount on the cash price. Although a majority chose the same smartphone twice seemingly not influenced by the retail offer, this was not the case for about one third of the young adults who despite a negative attitude to borrowing choose instalment payments of more expensive premium smartphones, and more than half of them at a price exceeding the regular cash price they choose to pay for a cheaper smartphone. Instalment payments as well as rental contracts are penetrating many consumer markets in which young adults are large segments. These new forms of accessible credit should be particularly attractive to those who are financially constrained with potentially negative consequences for their solvency. Our results suggest that regulation policies may need to be considered.

Keywords:

• consumer credit
• Nordic loan market
• smartphone
• spending
• young adult

Introduction

In the Nordic countries with growing markets for consumer credit, it is recognized that many young adults are vulnerable to problem debts (Autio et al., 2009; Hohnen et al., 2020; Oksanen et al., 2016). In Sweden where this research is conducted, a governmental study (Swedish Government Official Reports, 2013) concludes that it is alarming that young adults are over-represented among debtors of unsecured high-interest loans. A high such debt burden in early adulthood would have negative consequences for future consumption (Elliot & Lindblom, 2019). It would also increase the risk of over-indebtedness that may result in psychological ill-being (e.g. Ahlström & Edström, 2014; Meltzer et al., 2013; Walsemann et al., 2015).

Potential determinants of consumer credit among young adults between 18 and 25 years old are shown to be related to different life transitions (Oksanen et al., 2016). In a complementary analysis focusing on individual motives, Gärling et al. (2020) highlight as a determinant urgent consumption desires evoked by aggressive product marketing and strengthened by present-biased temporal discounting (Gärling & Ranyard, 2020; Labroo & Pocheptsova, 2017). Easily accessible credit would then enable purchases of desired consumer products that are barely or not affordable. In addition, peer influence through social media strengthens both consumption desires and borrowing intentions (Sotiropoulos & d’Astous, 2012, 2013).

A major retailer in the Swedish smartphone market offers different brands with a percent discount on the cash price if payment is made in two-year monthly instalments combined with a binding subscription of telephone calls, text messaging, and internet surfing. The discounted cash price is set to not exceed the lowest market price. In addition to attracting more buyers, an increase is expected in sales of new premium models of smartphones. For many young adults to whom the cash price of a new premium model is not or barely affordable, the discounted cash price in combination with instalment payments would appear attractive. Our research question asks whether the young adults for this reason purchase more expensive smartphones? Problem debt is then a possible consequence (Autio et al., 2009; Swedish Government Official Reports, 2013).

In the remainder of the paper, we first review three strands of research with bearings on our research question. From the review we derive hypotheses that are tested in an experiment which is reported thereafter. We finally discuss the results of the experiment and their implications.

Previous research and hypotheses

Price discount

Temporary price discounts are frequently used by retailers to attract new buyers (Ailawadi et al., 2009; Bogomolova et al., 2015). In general, reliable increases in sales volumes are observed. There may still be differences depending on amount discounted and type of signaling at the point of purchase. Büyükkdag et al. (2020) investigated effects of whether a price reduction is expressed as amount or percent. Although both amount and percent increased price attractiveness and purchase intentions compared to the same price with no price discount, the former had a stronger effect than the latter. However, Saini et al. (2019) found the opposite. The results of Jha et al. (2019) suggest that information about scarcity of product quantity and time restrictions on the discount availability increases effectiveness of the discount. Still other research (see reviews in Mazumdar et al., 2005; Raghubir, 2006) has shown that consumers judge attractiveness of discounts relative to a reference price. An explicit reference price is frequently the regular price.

If a percent discount is offered on the cash price paid in instalments, the young adults would save money by purchasing the same smartphone paid in this way instead of at the regular cash price. The saved money may alternatively be spent on upgrading the smartphone to a more expensive premium model. For a percent discount, a larger amount would furthermore be saved on the cash price of a more expensive premium smartphone. We propose that the young adults’ desire for a premium smartphone make them use the amount saved on the cash price to purchase the more expensive premium smartphone. The feasibility of paying for the smartphone in monthly instalments likely strengthens this intention.

Desire

Every generation of smartphones has new features (e.g., stereo sound, advanced camera) that are attractive to young adult buyers (Keaveney & Parthasarathy, 2001; Ting et al., 2019). Supplementary products (e.g. wireless charger, headset) are likewise attractive. Even though young adults consider a smartphone to be a necessity (Arif et al., 2016; Handa & Ahuja, 2020; McCasland, 2005; Ting et al., 2011), new features added to the core features making the smartphone a more expensive premium model are likely to evoke a desire to purchase it at the higher price.

Young adults would feel a financial deficit if desiring to purchase a more expensive premium smartphone at a barely affordable cash price. This feeling has the consequence that attention is focused on means of reducing the financial deficit (Cannon et al., 2019; Mani et al., 2013; Shah et al., 2012). Not having to pay the cash price at the point of purchase would do this, thus increasing the likelihood that the desired more expensive smartphone is paid in monthly instalments. In the Elaborated Intrusion theory proposed by Kavanagh et al. (2005, see also Hofmann & Van Dillen, 2012), desire is defined as an “…affectively charged cognitive event in which an object or activity that is associated with pleasure or relief of discomfort is in focal attention” (p. 447). The theory posits that a desire is represented and elaborated in working memory, and as demonstrated in previous research (Hofmann et al., 2012), a desire imposes demand on working memory that interferes with the pursuit of long-term goals. This may lead to a neglect of the future costs of monthly instalments. In indirect support, Frigerio et al. (2020) showed in a meta-analysis that over-indebtedness is associated with impulsivity.

The above reasoning leads to the proposition that if the new features of a more expensive premium smartphone evoke a desire to purchase the smartphone at a barely or not affordable cash price, easily accessible and favourable terms make instalment payments attractive.

Attitude to borrowing

An attitude is a memory representation that when activated makes people respond favourably or unfavourably to the attitude object, in thinking and expressing opinions (Albarracin & Shavitt, 2018) as well as in overt behaviour (Ajzen et al., 2019). Interest rates, marketing of loans, regulations, and cultural norms and values are factors that have been shown to influence attitudes to borrowing (Kamleitner et al., 2012). In addition, studies have found that young adults have more positive attitudes to borrowing than older people have (Gamble et al., 2019). Yet, Gärling et al. (2020) demonstrated in an online experiment that Swedish young adults between 18 and 30 years held negative attitudes that made them unwilling to borrow to fund purchases of desired consumer products. This finding was observed for several different credit types, including instalment payments. Therefore, young adults may not want to pay for a smartphone as monthly instalments. However, over time some credit types have become more accepted (Hohnen et al., 2020). For instance, payments by credit card have in the USA increased in frequency despite young adults’ negative attitudes to borrowing (Norvilitis, 2014). In Sweden, instalment payment for smartphones is likewise frequent (Larsson et al., 2016)¹. We suggest that this method of payment for smartphones is no longer perceived as a type of credit. It does not therefore activate a negative attitude to borrowing.

In summary, a desire to purchase a more expensive premium smartphone is evoked by its attractive added features. The higher cash price induces a feeling of financial deficit that in conjunction with a price discount increases the likelihood of choosing to pay for the smartphone in monthly instalments. Although the attitude to borrowing is negative, it may still not negatively influence the choice of paying in monthly instalments if it is perceived as an accepted method of payment for smartphones, not as a negatively valenced type of consumer credit.

Hypotheses

In the experiment to be presented next, a within-group design requires that the participating young adults make two choices of one of five fictitious smartphones varying from a budget model to the latest premium models. In counterbalanced order, they either choose the smartphone by paying the regular cash price or paying a 20% discounted cash price in two-year monthly instalments. In a scenario that the current smartphone has stopped working², in one between-group condition the participants are asked to think of this as an opportunity to upgrade their current smartphone when replacing it, in another between-group condition to think that it would be sufficient to replace the current smartphone with a similar one. We propose the following three hypotheses:

1. The frequency of choice of a more expensive premium smartphone is increased if the retailer offers instalment payments of a discounted cash price.

2. The frequency of choice of a more expensive premium smartphone is increased by the desire to upgrade the current smartphone compared to replace it with a similar one.

3. A negative attitude to borrowing does not reduce the frequency of choice of a more expensive premium smartphone paid for as instalments.

Method

Material

In order to prepare a choice set consisting of smartphones varying from a budget model to the latest most advanced premium models, technical information about the latest Iphone premium models 8 and 8+, the earlier premium models 7 and 7+, and the budget model SE was downloaded from Apple’s Swedish website. Descriptions with technical information (including capacity of processing memory, water resistance, size, type of camera, weight, battery time, capacity of storage memory) were prepared of the Iphone models referred to as A, B, C, D, and E without disclosing brand and model.

A pilot study was conducted to pretest the technical descriptions. A brief questionnaire was distributed to 98 undergraduates in psychology classes at University of Gothenburg. Questions were asked about their current smartphone, including brand and model, for how many months they had used it, and if they had purchased it themselves, how much and how they paid (in instalments, cash or debit card, credit card or any other type of credit). They were also asked to rate on 1-to-9 scales how satisfied they were with their current smartphone and how much they desire to purchase a new one. The technical descriptions were then presented asking the participants to indicate which information is redundant and which additional information is needed if they considered purchasing any of the smartphones. Modifications were made to the descriptions based on the replies if made by several participants and not making the descriptions importantly different from the initially down-loaded information. The participants were finally asked to guess the regular cash prices of the smartphones. Although the guesses deviated from the actual cash prices, the average rank order corresponded to the rank order of the prices.

Participants

A heterogeneous sample of Swedish young adults aged 18 to 25 years³ was randomly chosen from a web panel maintained by a market research organization (www.cint.com). The participants were invited to answer an online questionnaire programmed in Qualtrics (www.qualtrics.com). They first answered the same questions about their current smartphone as were asked in the pilot study. Last in the questionnaire questions were asked about their background. After excluding those already owning the latest Iphone models 8 and 8+, or in a few cases newer premium models launched after constructing the choice set, 152 participants remained (91 or 59.9% women, mean age 22 years). Forty-four (28.9%) had a higher education than college, 60 (39.5%) were students, 71 (46.7%) employed, and 21 (13.8%) unemployed. Sixty (39.5%) had a residence of their own, 31 (20.4%) shared with peers, and 61 (40.1%) lived with their parents or other relatives. Seventy three (48.0%) owned an Iphone (models SE, 4, 5, 6 and 7), 35 (23.0%) a Samsung, 11 (7.2%) a Sony, 9 (5.9%) a Huawei, and 24 (15.9%) other smartphones. The smartphones had been obtained on average 17 (SD = 14) months earlier. Seventy-four (48.7%) had paid an average cash price of SEK 4,540 (SD = 2,617) and the monthly cost of SEK 208 (SD = 162) for a subscription of telephone calls, text messages, and internet surfing, whereas 61 (40.1%) had paid two-year instalments combined with the subscription at the monthly cost of SEK 453 (SD = 268). The remaining participants had received the smartphone as a gift or failed to answer the question. On the 1-to-9 scales satisfaction with the current smartphone was on average rated as 7.03 (SD = 1.78) and desire to purchase a new smartphone as 5.16 (SD = 2.50).

Procedure⁴

After the initial questions about their current smartphone, the participants were presented the technical descriptions of the smartphones in the choice set. The instructions explained that the smartphones were similar to many brands as described on webpages, have a number of common features (4 G, fingerprint reader, bluetooth, USB charging, WiFi, audio, video recording), have the same warranty, the price does not include an insurance, and the purchase is not locked to a subscription of telephone calls, text messaging, and internet surfing. The instructions furthermore asked the participants to imagine that the smartphones were their favorite brand.

The technical descriptions (see Material) were presented on a separate page in the online questionnaire such that scrolling made possible to read each description in any order although presented in the order from A (latest most advanced premium model) at the top to E (budget model) at the bottom. Participants were asked to process the descriptions as they would do if considering to purchase one of the smartphones.

Another set of descriptions (referred to as expert evaluations shown in Table 1 translated into English) was prepared of each of the smartphones as to be similar to expert evaluations published in Swedish computer magazines⁵. The descriptions included cash prices or monthly instalments which were close to what the major retailer offered at the time of the study. As indicated, the cost of a subscription of telephone calls, text messages, and internet surfing was not included. On a page following the technical descriptions, the expert evaluations were presented in the same way as the technical descriptions had been presented.

Table 1 Fictitious expert evaluations of the smartphones.

Smartphone A New premium model that is one of the best deals in the market. Top class performance. Big but still easy to handle because of its rounded corners. New technical features are improved battery life, stereo speakers and water proof. Has dual cameras, one with wide-angle lens and one with telephoto lens allowing for optical zoom. Top image quality. Available in black, gold, silver, and pink gold. For this smartphone you pay / the cash price of SEK 7920. / SEK 260 per month in 2 years. / Costs for telephone calls, text messaging, and internet surfing are not included.

Smartphone B This new premium model’s performance is not far behind the more advanced model, but it has a smaller display with lower resolution although the same screen brightness. Has a new but less advanced camera that delivers good image quality. Battery life is the same. Not waterproof. Available in black, gold, silver and pink gold. For this smartphone you pay / the cash price of SEK 6920. / SEK 230 per month in 2 years. / Costs for telephone calls, text messaging, and internet surfing are not included.

Smartphone C Is the most advanced of earlier premium models with lower performance and shorter battery life. Has the same screen as the new premium model but lower screen brightness. Camera with acceptable image quality. Not waterproof. Available in space grey, gold, silver and pink gold. For this smartphone you pay / the cash price of SEK 6000. / SEK 200 per month in 2 years. / Costs for telephone calls, text messaging, and internet surfing are not included.

Smartphone D The less advanced of the earlier premium models with the same performance and battery life. Has the same smaller screen as the new model next to the most advanced premium model but with lower screen brightness. Camera with acceptable image quality. Not waterproof. Available in space grey, gold, silver and pink gold. For this smartphone you pay / the cash price of SEK 5040. / SEK 170 per month in 2 years. / Costs for telephone calls, text messaging, and internet surfing are not included.

Smartphone E Budget smartphone. Has all features as the more advanced smartphones but lower performance. Same battery life as the older models. Smaller display with lower resolution and screen brightness. Not waterproof. Available in space grey, gold, silver and pink gold. For this smartphone you pay / the cash price of SEK 4080. / SEK 140 per month in 2 years. / Costs for telephone calls, text messaging, and internet surfing are not included.

Note. The expert evaluations were presented twice. The slashes refer to that either the cash price or the monthly instalment was presented the first or second time in counterbalanced order.

The participants were asked to imagine that their current smartphone had stopped working, that it was fully paid, and that they would not obtain a new smartphone from the retailer. Therefore, they considered purchasing one of the smartphones. The fictitious expert evaluations were presented twice following the technical descriptions, either with a default choice of a cash price or a default choice of instalment payments. The participants were informed that if choosing the instalment payments, they were offered a 20% discount on the cash price which was the lowest price in the market.

The experimental design consisted of a within-group factor (default choice of instalment payments versus default choice of payment of the cash price) and two crossed between-groups factors, upgrade (“This is an opportunity for you to purchase a new better smartphone that you desire”) versus replace (“A new better smartphone than your old is not necessary for you to purchase”) and order between the default choices (instalment payments before versus after payment of the cash price). Between 37 and 39 participants were randomly assigned to each of the four groups.

In the order instalment payments before payment of the cash price, the additional instructions before the first choice read: “If you pay for the smartphone in two-year monthly instalments, the cash price is right now 20% lower which is the lowest market price. Which of the smartphones A to E would you choose?” and before the second choice: “If you instead pay for the smartphone at the regular cash price, which smartphone would you then choose?” In the reversed order, before the first choice the participants were asked to choose one of the smartphones A to E at the regular cash price, and before the second choice: “If you are offered to pay for the smartphone in two-year monthly instalments of a cash price that right now is 20% lower than that you just paid and this price is the lowest market price, which smartphone would you then choose?” Participants made the choice by clicking on one of the alternatives. Thereafter they made a rating on a 1-to-9 scale of how much more attractive the chosen smartphone was than the next most attractive.

After having made the choices of smartphone, the participants first rated on a 1-to-9 scale how much more or less they would prefer the smartphone they chose last (either by paying in cash or in instalments depending on condition), how satisfied they would be if actually purchasing the most preferred smartphone of those chosen, how important it was that the smartphone was a new model, had a low price, and was liked by peers. Thereafter, they rated on another four 1-to-9 scales how economically strenuous it was to pay for the smartphone (“It would be economically strenuous to pay for the smartphone”, “I would feel that I need to save on other expenses if paying for the smartphone”, “On second thought, I feel I would not afford paying for the smartphone”, “I would actually need a better personal economy to purchase the smartphone”). A reliable index of payment strain was constructed by averaging the ratings, M = 5.68, SD = 1.97, α=.84.

Attitude to borrowing was then assessed by averaging 1-to-9 ratings of statements developed for this study based on a more extensive measure used in previous research (Gärling et al., 2020) (“I do not think it is ethical to borrow to consumption”, “It does not feel good to borrow to consumption”,” It’s more convenient to pay in cash or with debit card than borrow to consumption”,” Nobody can influence me to borrow to consumption”), M = 3.74, SD = 0.60, α=.73, after reverse coding.

A number of required additional ratings made possible to construct indexes of financial management, self-assessed financial resources, and frequency of previous borrowing. The questions were identical or similar to those used by Gärling et al. (2020). None of the indexes had significant correlations (r<|.20|, p>.10) with how much the participants spent on the smartphone purchase and are therefore not reported further.

Results

Descriptives

The results are presented in Table 2. It can be seen that when the smartphone is paid in instalments instead of in cash, the more expensive latest premium models (A, B) are chosen more frequently than the earlier premium models (C, D) and the budget model (E). The average preference ratings are also higher. With the discount on the cash price, the participants would still on average pay less for the latest premium models than the cash price for the other models. Further examining the differences between choice of smartphone paid in instalment and choice of smartphone paid in cash, it was found that across individuals the distributions of the differences were virtually identical for the two orders in which the choices were made (see also footnote 6). Combining the choices in the different orders revealed that 53 (34.9%) participants chose to upgrade to a more expensive smartphone when paying in instalments compared to paying in cash, 74 (48.7%) chose the same smartphone, and 25 (16.4%) chose a cheaper smartphone. Of those who choose the same smartphone twice, 41 (55.4%) chose the most expensive smartphone, 15 (9.9%) the next most expensive, 5 (3.2%) each the third and fourth most expensive, and 8 (10.8%) the cheapest. Of those who chose to upgrade to a more expensive smartphone, about equally many (25 or 47.2%) would pay a lower price (SEK 5462 instead of SEK 5923) as those (28 or 52.8%) who would pay a higher price (SEK 5931 instead of SEK 4937).

Table 2 Percentage and mean choice of smartphone, mean total cost, and mean preference rating.

Smartphone (Cash price/Monthly cost in SEK^a)	Replace (n = 74)		Upgrade (n = 78)
	Cash price	Instalment	Cash price	Instalment
A (coded 5) (7920/260)	35.1%	47.3%	35.9%	43.6%
B (coded 4) (6960/230)	21.6%	21.6%	16.7%	20.5%
C (coded 3) (6000/200)	14.9%	12.2%	20.5%	20.5%
D (coded 2) (5040/170)	12.2%	6.8%	14.1%	9.0%
E (coded 1) (4080/140)	16.2%	12.2%	12.8%	6.4%
Choice of smartphone [M, (SD)]	3.47 (1.48)	3.85 (1.40)	3.49 (1.43)	3.89 (1.25)
Total cost SEKa [M, (SD)]	6454 (1423)	5412 (1009)	6468 (1373)	5418 (904)
Preference rating (1-9) [M, (SD)]	5.28 (2.31)	5.49 (2.28)	5.35 (2.36)	5.54 (1.97)

Note. M = Arithmetic mean; SD = Standard deviation.

a 1 SEK approximately equal to 0.12 USD or 0.10 Euro.

Statistical tests

In two OLS linear regression analyses testing the hypotheses, the dependent variable was in one analysis the difference (ranging from 4 to −4) between the choice of smartphone (on the scale from 1 [budget model] to 5 [most advanced premium model]) paid in instalments and the choice of smartphone paid in cash, and in the other the difference (ranging from 8 to −8) between the preference ratings of the chosen smartphone paid in instalments and the preference ratings of the chosen smartphone paid in cash⁶. As can be seen in Table 3, in accordance with hypothesis H1, the choices of more expensive smartphones are significantly more frequent for instalment payments than for payments of the cash price. Refuting hypothesis H2 but supporting hypothesis H3, the choices in the upgrade condition did not differ from the choices in the replace condition and the attitude to borrowing had a negative coefficient which was not significant⁷. The results for the preference ratings were consistent with these results although the differences between the instalment payments and payment of the cash price did not reach significance, whereas the negative coefficient associated with attitude was significant at p<.05.

Table 3 Unstandardized coefficients (b), standard errors (SE), t-statistics and associated p values from OLS linear regression analyses.

	Dependent variable
	Difference in choice between instalments and cash				Difference in preference ratings between instalments and cash				Preference for smartphone paid in instalments
Independent variable	b	SE	t	p	b	SE	t	p	b	SE	t	p
Installments versus Cash price	0.38	0.12	3.16	.002	0.20	0.16	1.23	.221	0.31	0.18	1.66	.100
Upgrade (1) versus Replace (-1)	−0.01	0.12	−0.02	.986	0.00	0.16	0.00	.993	0.04	0.18	0.21	.837
Attitude to borrowing (1-9)^a	−0.07	0.07	−0.94	.350	−0.21	0.10	−2.09	.040	−0.12	0.12	−1.00	.321

a Mean centered.

Another OLS regression analysis was performed on the ratings of how much the participants preferred the smartphone chosen in the different conditions. The ratings (ranging from 1 to 9) revealed a higher preference for the smartphone paid in instalments (M = 5.20, SD = 2.28) than paid in cash (M = 4.58, SD = 2.26). Table 3 shows that this difference was not large enough to reach significance. The coefficients associated with upgrade versus replace and attitude to borrowing were not significant.

To investigate how the post-choice ratings were related to the choices, in Table 4 participants who chose instalment payments but did not upgrade the smartphone are compared to those who chose instalment payments and upgraded to the latest premium models (A or B) or the earlier premium models (C or D). To keep the comparison between models balanced, 14 participants were excluded who chose to pay for the budget model in instalments. Eight of these participants also chose the budget model when offered at the cash price, whereas the remaining six participants choose a cheaper smartphone than they choose at the cash price. All the average ratings, satisfaction with the purchase (M = 5.60, SD = 2.05), payment strain (M = 5.75, SD = 1.88), importance to choose latest model (M = 5.73, SD = 2.23), and importance to choose low price (M = 5.85, SD = 1.92) were above the midpoint of the 1-to-9 scale. The exception was the average ratings of importance of liked by peers which was below the midpoint (M = 3.46, SD = 2.47). As may be seen in Table 4, there are differences depending on upgrade versus no upgrade and latest models versus earlier models. Satisfaction tends to be higher for choices that are not upgraded than upgraded and higher for choices of latest models than earlier models. Payment strain is also rated higher for the more expensive latest models. The participants who indicated that the latest model was important for their choices chose one of the latest models and indicated that a low price was not important.

Table 4 Means and standard deviations (within parentheses) of post-choice ratings related to upgraded versus not upgraded choice of earlier versus latest models of smartphone paid in instalments of discounted cash prices.

	Not upgraded		Upgraded
Dependent variable	Earlier models (C, D) (n = 25)	Latest models (A, B) (n = 60)	Earlier models (C, D) (n = 12)	Latest models (A, B) (n = 41)
Satisfaction with purchase (1-9)	5.56a (2.18)	6.33a (1.88)	4.25b (1.66)	4.95ab (1.94)
Payment strain (1-9)	5.52a (2.06)	5.76a (1.95)	5.56a (1.74)	5.94a (1.73)
Latest model (1-9)	5.68a (2.34)	6.30a (2.16)	4.50b (2.24)	5.29ab (2.08)
Low price (1-9)	5.92a (2.02)	5.50ab (1.73)	6.42ac (1.73)	6.15abc (1.57)
Peer liking (1-9)	3.24a (2.65)	3.57a (2.61)	3.25a (2.67)	3.39a (2.46)

Note. Means in each row sharing subscripts do not differ significantly in independent t-tests at p<.05.

Discussion

Findings and contributions

Our results show that on average more expensive premium smartphones are chosen by the young adults when paying two-year monthly instalments of a 20% discounted cash price instead of paying the regular cash price. It seems unlikely that these choices resulted from explicit calculations. It is more plausible that the choices were based on an intuitive judgment that a desired more expensive premium smartphone is affordable since the price would be the same or lower than the regular cash price of a cheaper smartphone (Kahneman & Frederick, 2005). Of those about one third of the participants who chose a more expensive premium smartphone, close to half of them would also pay a higher price. Presumably, they had higher aspirations or were less cautious in spending their money. Almost half of the participants chose the same smartphone twice. For a majority of them the discount could not have had the hypothesized effect since at the regular cash price they chose the most expensive smartphone. These participants would save money if choosing instalment payments. A few chose a cheaper smartphone at the discounted cash price. They probably wanted to minimize the two-year monthly instalment payments, as Gärling et al. (2020) found that participants did if they held negative attitudes to borrowing.

The discount on the cash price increased choices of more expensive smartphones even though the young adults were instructed to consider replacing their old smartphone with a similar one. In order to investigate whether the results were due to the instructions failing to have the hypothesized effect, we conducted another online experiment asking 145 undergraduates sampled from different study programs at University of Gothenburg to choose a preferred smartphone from the choice set. The results showed that a more expensive premium smartphone was chosen by those for whom the instructions emphasized the opportunity to upgrade the old smartphone than by those for whom the instructions did not emphasize this opportunity but to replace the old smartphone with a similar one (t(143)=2.32, p=.022; d = 0.39). No choices were made of payment method although information was given of both the cash price and the monthly instalments. A percent discount on the cash price was not offered. The results suggest that the discounted cash price combined with instalment payments was important for our observation that more expensive smartphones were chosen by the participants although not explicitly framed as an opportunity. The discounted cash price thus appears to have induced a stronger temptation to choose a desired premium smartphone at the same (or higher) price instead of saving money on the cash price. Although it cannot be concluded with certainty, the offered instalment payments are likely to have contributed to this choice if participants would barely or not afford to pay the cash price.

Gärling et al. (2020) proposed that a negative attitude to borrowing acts as a heuristic (Pratkanis, 1989), resulting in a choice of not borrowing after limited deliberation. Our results showed, similarly as Gärling et al. did, that on average the attitude to borrowing is negative. It did still not have a negative effect on the choice of a more expensive smartphone paid in instalments. We suggest that the reason is that paying for a smartphone in instalments has become an accepted payment method. This would make people unaware of that they are borrowing when paying in instalments (Gamble et al., 2019; Hohnen et al., 2020; Norvilitis, 2014). Therefore, a negative attitude to borrowing is not activated.

With the aim of providing additional information about why the participants chose a more expensive premium smartphone, questions were asked after the choices had been made. In the sample of young adults, payment strain was generally high and higher the more expensive the smartphone they chose. Satisfaction with the choice was still high and higher for the more expensive premium smartphones. Satisfaction was however less for the choices of the upgraded smartphones. Since most choices of not upgraded smartphones were made by participants making the same choice twice, it is conceivable that the higher satisfaction primarily reflected their stronger preference for the chosen smartphone. It should also be noted that obtaining the latest model was an important stated motive for choosing the more expensive premium smartphones. A low price was more important for upgrading choices, thus suggesting an effect of the price discount. Peer likings were not an important stated motive for the choice of smartphone. This is not consistent with the findings of Sotiropoulos and d’Astous (2012, 2013). Unawareness of or unwillingness to admit peer influences is a possible limitation of our results.

In summary, we have shown that instalment payments combined with a price discount make young adults on average to choose more expensive premium smartphones. About one sixth of them would also pay a higher price than they would do otherwise. This is an outcome that is worth taking seriously, in particular since payment in instalments usually costs more than cash payments. Yet, the experiment was designed in a way that instalment payments were not more expensive. This might have made the payment method more attractive. Thus, the price discount seemed to have worked, as probably was expected by the Swedish retailer who adopted it. One may still question whether combining the two-year installment plan with a binding subscription of telephone calls, text messaging, and internet surfing would have had the same positive effect. Currently, the competition in the Swedish smartphone market has increased due to other retailers marketing less costly subscriptions, thus making it more attractive for buyers to have the option to choose smartphone and subscription plan from different retailers. Even though the current results may not in these respects accurately correspond to the real-life case today, our contribution is still to show that young adults may choose to purchase more expensive premium smartphones if the cash price is discounted and they are offered instalment payments to fund the purchase.

Limitations and future research

A limitation of this study is that the experimental design confounded the effects of the discount on the cash price with offering monthly instalments. Although the additional study reported above and the replies to the post-choice questions suggested a role of the price discount, the choice of instalment payments likely also played an important role since it did not require available cash. For financially constrained young adults, this would in itself be an attractive opportunity to acquire a more expensive smartphone. It is thus plausible to conclude that both factors had effects. Additional research is therefore motivated to investigate the strength of their separate effects, taking our results as a point of departure. Some independent measure of preference for instalment payments versus cash price would be needed.

It seems worthwhile to extend this research to other product categories (e.g. computers, TV sets) to investigate the extent to which young adults are willing to choose instalment payments. In the Nordic countries following an international trend, this form of credit as well as rental contracts are penetrating several markets. In many of these markets young adults constitute large segments. Since young adults tend to be more financially constrained than many other consumers, these new forms of accessible credit should be particularly attractive to them with potentially negative consequences for their solvency. If this turns out to be true, as our results suggest it may, regulation policies need to be considered.

Acknowledgements

We thank Jeanette Carlsson Hauff, Viktor Elliot, Ted Lindblom, and Jonas Nilsson for valuable comments on earlier versions of the manuscript, and Foisal Ahmani for research assistance. The study was in 2018 presented at the IAREP/SABE conference at University of Middlesex, London.

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

This research was financially supported by grants to Centre for Finance, School of Business, Economics, and Law, University of Gothenburg, Göteborg, Sweden, from Handelsbanken Research Foundations [grant #P2015-0265:1] and from Swedish Agency for Innovation Systems (Vinnova) [grant #2010‐02449].

Notes

1 In the proportion of our sample of young adults who reported having paid themselves for their current smartphone, 45.1% paid in instalments.

2 Even though smartphones may not frequently stop working, the scenario represents the typical features of the retail environment, including marketing of new models of smartphones, customers desiring to immediately purchase the new models, and accessible payments in instalments.

3 Eighteen years is the lowest legal age for making instalment payments. Since a majority of young adults up to approximately 25 years earn low or no salaries from paid work, they are unlikely to afford to pay in cash for expensive premium smartphones.

4 The purpose of the expert evaluations was to provide non-technical information such that less knowledgeable participants would consider also other features than price.

5 Additional regression analyses did not reveal a significant main effect of whether instalment payments were presented before payment of the cash price or the reverse, neither that this factor interacted significantly with the other factors.

6 Note that a negative coefficient implies that a positive attitude to borrowing would decrease instalment payments, thus opposite to what Gärling et al. (2020) observed.