THE EFFECT OF SAVING FORMAT ( ” %-OFF ” VS . ” AMOUNT-OFF ” ) ON THE CHANGE OF INTERNAL REFERENCE PRICE : AN EMPIRICAL STUDY

Recently, price promotion campaigns have been used increasingly in the commercial sector. Since, for marketing specialists, price presentation has a strong impact on consumer perceptions, it is important to organize price promotions that consumers perceive positively. Purpose: The first objective of this study is to examine the effects on the consumer internal reference price of saving formats displayed in %-off as opposed to those expressed as amount-off. The second objective of the paper is to analyse whether the moderating effect of the discount level (low vs. high) and price confidence (low vs. high) will influence the main relationship between saving format and change of internal reference price. Methodology: In order to achieve the objective of the current research, both primary and secondary data were examined. Secondary data were gathered from reliable sources on price promotion, whereas primary data were collected through a survey. Findings: The results of the empirical research give suggestions to retailers as to which format of price promotion to select in order to diminish or avoid drastic reductions of the consumer internal reference price, considering the discount level and consumer characteristics such as price confidence.


Introduction
Promotional activities play a crucial role in marketing in that they offer a trade-off between consumers, producers and marketers.Therefore, on-going price promotion activities are often used by retailers to boost sales (Wierenga and Soethoudt, 2010).Due to tight competition on the market, marketers spend a huge amount of money trying to find out the best techniques to organize price promotion activities (Hardesty and Bearden, 2003) intended to positively influence consumer behaviour.Previous studies have demonstrated that price plays a crucial role for consumers' buying behaviour.Despite the positive effects of price promotion on sales (Choi and Mattila, 2014) and consumers' purchase patterns, there are also negative influences which result in lowering the consumer internal price.Hence, for retailers, the method of presentation of a price promotion is a crucial factor due to its influence on consumer behaviour.Inappropriate presentation of a price promotion leads to negative consumer evaluation, lowering sales and therefore also profits as a result (Chapman, 1993).Thus, marketing specialists should try to find the best ways to present price promotion techniques effectively.
Substantial studies exist, which analyse internal reference prices in the context of price promotion (e.g.Chandrashekaran and Grewal 2006;DelVechio et al., 2007;Bambauer-Sachse and Dupuy, 2012).These studies have found out that price promotion reduces the consumer internal reference price.To avoid such negative effects of price promotion, it is essential to identify the impact of the saving format ("%-off" vs. "amount-off") on the change (reduction) of the consumer internal reference price.It is also particularly important to determine which type of saving format ("%-off" or "amount-off") leads to a stronger change (reduction) of the internal reference price.
Therefore, the primary purpose of the following paper is to analyse the direct effect of saving formats ("%"-off vs. "amount-off") on the consumer internal reference price.
In the previous studies, price discount was linked to price expectations and saving formats (e.g.DelVechio et al., 2007;Grewal et al., 1998).In their study, Grewal, Krishan, Baker and Borin (1998) suggested that the consumer internal reference price is higher when the sale price is closer to the consumer internal reference price, and conversely.It is therefore relevant to identify which type of saving format ("%-off" or "amount-off ) has a stronger effect on the change of internal reference, considering the moderator role of the discount level (low vs. high).
Furthermore, the moderator role of price confidence was examined in the context of price promotion (e.g.Bambauer-Sachse and Dupuy, 2012).In the previous study, Bambauer-Sachse and Dupuy (2012) indicated that price confidence has a strong impact on consumer internal reference prices.The following research will also use price confidence as a moderator variable in order to identify its impact on the main relationship between the saving format ("%"-off vs. "amount-off") and the change of internal reference price.
Hence, the second aim of the paper is to determine whether the moderator role of the discount level (low vs. high) and price confidence (low vs. high) will influence the main relationship between the saving format and the change of internal reference price.

Literature review
While marketers in the commercial sector take advantage of different price promotion campaigns, the focus of this paper is on the price presentation format ("%-off" vs. "amountoff").Few studies have examined the impact of the saving format ("%-off" vs. "amount-off") on consumers' price expectations or the change of consumer internal reference price (e.g.DelVecchio et al., 2007;Chandrashekaran and Grewal, 2006;Chen et al., 1998;Bambauer-Sachse and Dupuy, 2012).
In the last decades, the importance of price promotion strategies in retail advertising has increased (Chandrashekaran and Grewal, 2006;DelVecchio et al., 2007).Previous studies have analysed reference prices and investigated their relevance with respect to consumer behaviour.
In the literature, the notion of internal reference price has been variously defined by different authors.The internal reference price is defined as a price that consumers have in mind, and which they use as a reference point in order to judge and compare current prices with the price last paid for prior purchases.Additionally, the consumer internal reference price emerges in response to external information and may shift accordingly (Biswas and Blair, 1991).In general, price reductions are quite costly and have a negative impact on the consumer internal reference price (Bambauer-Sachse and Dupuy, 2012).Nevertheless, it is very important to be aware that internal reference prices might change as a result of different price promotion strategies.Therefore, retailers are interested to have high reference prices of their products in order to enhance the consumer perception of savings in a discount condition (Chandrashekaran and Grewal, 2006).Additionally, it is crucial for retailers to be aware under which conditions consumers choose which saving format ("%-off" vs. "amount-off") (Hardesty and Bearden, 2003).
In fact, prices which are higher than this reference price are perceived negatively; therefore, consumers interest in specific products drops.On the other hand, all prices which are lower than this reference point are perceived positively, which further enhances the likelihood that consumers will require those products (Fibich et al., 2007).
Following the increased use of price reduction in the market nowadays, several research studies have analysed the change of internal reference prices after a price promotion occurs (e.g.Chandrashekara and Grewal, 2006;DelVechio et al., 2007).In order to have a deep insight into the change of internal reference prices and to find out the factors which cause the change of internal reference prices, it is relevant to include this variable in this research paper.
So far, the price discount is the most dominant method in the market and its relevance has recently increased (Darke and Chung, 2005).In general, discounts provide the consumer with economic value, also known as acquisition value and transaction value.Acquisition value refers to the benefit of paying a lower price for products, while obtaining the same gains from it, whereas the transaction value is defined as the perceived value of the offer in which an economic outcome is exceeded (Darke and Chung, 2005).
In a price discount, retailers lower the sale price and all consumers gain the same benefits.Therefore, price discounts are offered frequently in stores nowadays and their importance is increased (Darke and Chung, 2005).The depth of the discount level is a crucial characteristic of price promotion (Fibich et al., 2007).Presentation formats of the discount level appear in two patterns: frequent with a low discount level and infrequent with a high discount level (Alba et al., 1999).Therefore, it is relevant to see how consumers perceive discount level (low vs. high).
The notion of self-confidence has been used in the existing literature due to its relevance to understanding consumer behaviour.Further, the variable price confidence refers to the degree to which consumers are confident about their market decisions (Bearden et al., 2001).Indeed, two levels of price confidence can be defined (low vs. high), in which consumers can be more or less influenced by the actual price offered on the market, depending on the information processed (Babauer-Sachse and Dupuy, 2012).Price confidence has a strong impact on the consumer internal reference prices.So far, past studies have not directly investigated the role of confidence on the consumer internal reference price (Thomas and Menon, 2007).
Despite theoretical implications, the following research will also try to point out practical benefits for marketers.The study will try to find out how price promotions should be provided, in particular, to identify which type of saving formats ("%-off" or "amount-off") marketers should opt for in order to avoid the reduction of the consumer internal reference price, also by considering the discount level and price confidence.

Existing empirical background and development of hypotheses
Marketers should choose the most effective way to provide saving formats to consumers in order to decrease the impact of price reduction on these consumers' internal reference price (Babauer-Sachse and Dupuy, 2012).Managers could set the retail price lower than the consumer internal reference price by creating a positive perception in consumers.In fact, in most of the cases, the sale price is usually higher than the consumer internal reference price (Chandrashekaran and Grewal, 2006).
Despite a small number of previous studies on the internal reference price (e.g.Babauer-Sachse and Dupuy, 2012;Chandrashekara and Grewal, 2006;DelVechio et al., 2007;Biswas/Blair, 1991), the impact of the saving presentation format on the internal reference price process still remains unclear.Chandrashekaran and Grewal (2006) found evidence that when a price reduction is presented in the "amount-off" format, consumers can more easily notice and calculate the price information than when the price is presented in the "%-off" format.Therefore, according to this study, the "amount-off format" leads to a stronger change (reduction) of the consumer internal reference price than the "% -off "format (see Table 1).The purpose of this study is to demonstrate the influence of price information such as advertised reference price and sale price on the consumer internal reference price, considering the moderator effect of the saving presentation format.
It was demonstrated that consumers pay more attention and process more information in relations to the advertised reference price than sale price in the presence of the "amount-off" saving format.Source: Rajesh Chandrashekaran; Dhruv Grewal (2006) As already mentioned, price reduction through promotion will lower the consumer internal reference price.Therefore, understanding how consumers perceive saving formats is very important in order to determine the impact of the presentation format on the change of the consumer internal reference price (Chandrashekaran and Grewal, 2006).In particular, it would be interesting to analyse and to see which saving format ("amount-off" or "%-off") has a stronger effect on the change (reduction) of the consumer internal reference price.
Furthermore, DelVecchio, Krishnan and Smith (2007) indicated that in the case of a high discount level, the "amount-off" saving format will lead to reduced future price expectations in comparison with the "%-off" saving format.However, in a case where the discount level is low, this effect is not likely to occur because there is a small difference between the actual and discounted prices.Hence, consumers will not be interested to calculate this difference.To sum up, the higher the discount level, the lower the change of the consumer internal reference price will be when the "amount-off" format is used compared to the "% off" format.On the other hand, with a low discount level, the change of the internal reference price has no effect across the two saving formats "%-off" vs. "amount-off" (DelVecchio et al., 2007), (see Table 2).Investigates the effect of promotion depth on post promotion price expectations and choice, considering the moderator role of price promotion frame (percentage off vs. cents off) The results showed that promotion depth and frame have an effect on price expectation.The higher the promotion depth, the lower the future consumer price expectations.
In addition, price expectations are also lower in the case of cents-off format than in percentage-off format.Source: Devon DelVecchio, H. Shanker Krishnan, and Daniel C. Smith (2007) In the current research, it is interesting to examine the effect of the saving format ("%off" vs. "amount-off") on the change of the internal reference price, considering the moderator role of the discount level (low vs. high).
In addition, Bambauer-Sachse and Dupuy (2012) found that price promotions in general lead to reduced consumer internal reference prices.In general, in the case of high price-confidence, the impact of price promotion is less strong, thus change in the internal reference price is less likely to occur.On the other hand, price promotion has a stronger effect on consumers who are less price-confident because they pay attention to price information such as saving information and price reduction (see Table 3).They, more so than highly price-confident customers, process this information and factor it into their internal reference price (Bearden et al., 2001;Bambauer-Sachse and Dupuy, 2012).Still, no study has examined the effect of the saving format ("%-off" vs. "amount-off") on the change of the internal reference price, considering price confidence as a moderator variable.Thus, it is compelling to analyse this relationship, too.
Despite substantial research on the internal reference price, there is only a limited number of studies that investigate the moderator role of the discount level and price confidence in the saving format ("%-off" and "amount-off") for the change of the internal reference price.Previous studies concentrated on saving formats in a slightly different and focused on different variables.Thus, they did not investigate the interaction effect with other variables, such as discount level and price confidence.
For this reason, this article is an important contribution, as it extends and supplements previous studies on the subject of price promotion, in particular on the saving format, by providing a theoretical framework for the marketing perspective.Despite theoretical implications, the following research will also try to offer practical benefits for marketers.
As seen before, according to the study of Chandrashekaran and Grewal (2006) consumers notice and pay attention much easier to price information that is presented in the "amount-off" saving format than price information that is presented in the "%-off" saving format.Relying on this argument, the first hypothesis can be raised.
Hypothesis 1: In the context of price promotion, the "amount-off "saving format leads to a stronger change (reduction) of the internal reference price compared to the "%-off" saving format.
In addition, DelVecchio, Krishnan and Smith (2007) found out that in the case of high discount levels, the "amount-off" format will reduce the future price expectations in comparison to the "%-off" format.However, in a case where the discount level is low, this effect is not likely to occur because there is a small difference between the actual price and the discounted price.Hence, in line with this argument, the following hypotheses can be raised: Hypothesis 2a: When the discount level is low, both saving formats ("%-off" and "amount-off") lead to the same change (reduction) of the internal reference price.

Hypothesis 2b:
When the discount level is high, the "amount-off" saving format leads to a stronger change (reduction) of the internal reference price compared to the "%-off" saving format.
It was found that price confidence has a strong impact on the consumer internal reference price.Consumers that are highly price-confident are less influenced by price promotion, thus a change of the internal reference price is less likely to occur.On the other hand, less priceconfident consumers are much more influenced by price promotions and pay more attention to price information such as the saving information and price reduction.This information also influences the internal reference price (Bambauer-Sachse and Dupuy, 2012).In fact, consumers more easily notice and pay more attention to the price that is presented in the "amount-off" format than to the price that is presented in the "%-off" format (Chandrashekaran and Grewal, 2006).According to this argument, the following hypotheses can be raised: Hypothesis 3a: When the price confidence is low, the "amount-off" saving format leads to a stronger change (reduction) of the internal reference price compared to the "%-off" saving format.

Hypothesis 3b:
When the price confidence is high, both saving formats ("%off" and "amount-off") lead to the same change (reduction) of the internal reference price.

Main study: Experimental design, sample size and data collection procedure
The experiment was based on a 2 (saving format: "%-off" vs. "amount-off") x 2 (product: jeans vs. t-shirts) x 2 (discount level: low, 10% vs. high, 50%) design.The variable "discount level" was manipulated either as "amount-off" saving format or "%-off" saving format.In this experiment, a convenience sample was used, consisting of a total of 240 respondents, whome 53.3% were females and 46.7% males.In addition, the research contained 8 experimental groups, with 30 respondents needed in each experimental group.The respondents' age ranged from 17 to 50, with the average being 24.38.
The primary data of the study were collected through face-to-face interviews.The respondents were assigned to one of the eight groups and asked to fill in the respective questionnaire.Two types of questionnaires were created for two product types: jeans and tshirts, in which the respondents had to answer identical questions.Half of the respondents received the questionnaire related to jeans and the other half received the questionnaire related to t-shirts.The stimulus material contained 9 ads, in which each group of respondents were exposed with 9 ads associated with one type of product (jeans or t-shirts), one type of the saving format ("%-off" or "amount-off") and one discount level (low or high).In addition, the ads also contained information about the regular price and the reduced price.
The discount amount in the case of the "%-off" saving format was set to 10% for the low discount and 50% for the high discount (Chen et al., 1998).Likewise, in the case of the "amount-off" saving format, the discount level was manipulated with the same amount of discount but expressed in the "amount off" format.The survey included a number of questions regarding consumers reference prices.Afterwards, the respondents were exposed to the ads and asked questions to evaluate these ads that contained information regarding the discount level, price confidence and price presentation formats.Finally, demographic questions were included to ask about the respondents' age and gender.

1 Data analysis and results of the research
Before discussing the main results of the research, the results of the manipulation check for the discount level will be presented.The results of t-test show that respondents evaluate the low discount level (M=2.71)much lower than high discount level (M=4.72);(t = -10.186,p = 0.000 < .0.05).In other words, these results demonstrate that the manipulation of the discount level has been successful and measured what was supposed to be measured.
To test the hypotheses of the research, it was appropriate to implement the independent sample t-test, because the mean values should be compared between two groups (Zikmund and Babin, 2007).
According to t-test results of the first hypothesis (t=0.062,p=0.951, >0.05), there is no statistical evidence that the "amount-off" saving format leads to a stronger change (reduction) of the internal reference price.Even though the descriptive results of the mean scores are consistent with the assumption of hypothesis 1, this difference is not statistically significant.
Hence, it can be concluded that both saving formats ("%-off" and "amount-off") lead to the same change (reduction) of the internal reference price.Based on these results, the first hypothesis was not confirmed.Table 4 shows the results of the mean values.Furthermore, t-test results (t = 0.726, p = 0.469 > 0.05) of the H2a show that there are no significant differences across the two groups of saving formats.Even with the descriptive results of mean values in the case of the low discount level, the "amount-off" saving format leads to a stronger change (reduction) of the internal reference price than the "%-off" saving format.However, statistically, there is evidence that in the low discount level both saving formats ("amount-off" and "%-off") lead to the same change (reduction) of the internal reference price.This leads to confirm the H2a.Table 5 shows the results of the mean values.The t-test results of the H2b (t=-0.540;p= 0.590 >0.05) show that there is no significant difference across the mean values of the two saving formats.The descriptive results of the mean scores show that the "%-off" saving format leads to a stronger change (reduction) of the internal reference price than the "amount-off" saving format.Even though these outcomes show inverse direction with the prediction of H2b, this difference based on t-test outcomes is not statistically significant.Thus, H2b is not confirmed.It can be concluded that in the case of a high discount level, both saving formats "%-off" and "amount off" lead to the same change (reduction) of the internal reference price.The results of the mean values are shown in the table below (Table 6).The mean scores are inversely correlated with the supposition of the hypothesis 3a.Even though there is an inverse direction, there is no significant difference between the mean values.The results of the t-test (t = -.171,p = 0.864 > 0.05) of the hypothesis 3a show that the differences between the mean values are not significant.Therefore, the supposition of the H3a is not confirmed.Thus, when the price confidence is low, both saving formats ("%-off" and "amount-off") lead to the same change (reduction) of the internal reference price.Table 7 shows the mean values.Although there is a difference across the mean values of the two saving formats, the t-test results of the hypothesis 3b (t = .134,p = 0.894 > 0.05) show that there is no significant difference across the mean values.Therefore, hypothesis 3b can be confirmed: "When the price confidence is high, both saving formats ("%-off" and "amount-off") lead to the same change (reduction) of the internal reference price".Table 8 shows the mean values.

Conclusion, limitations and suggestions for future research
These findings do not only provide a theoretical framework for a discussion of the impact of price promotions on internal reference prices, but also offer practical benefits for marketers.
In addition, the results provide suggestions for marketers as to how they should present saving formats to consumers in order not to reduce the consumer internal reference price, considering the discount level and price confidence.
The objective of the study was to determine which saving format ("%-off" or "amountoff") has a greater influence on the change of the consumer internal reference price.Further, it was relevant to analyse the effect of the moderator variables, the discount level and price confidence.
Regarding the first hypothesis (H1), the results did not show any difference between the two saving formats ("%-off" and "amount-off") in terms of the "change of internal reference price" There was no statistical support for either the "%-off" or "amount-off" saving format having a stronger impact on the change of the internal reference price.In line with these results, it can be concluded that both saving formats have the same influence on the change of the internal reference price.Therefore, in the context of price promotion, marketers could use both saving formats because there is no evidence that either the "%-off" or the "amount-off" saving format would lead to a greater change (reduction) of the internal reference price.
In the case of the moderating role of the discount level, the results demonstrated that there is no significant effect in either of the saving formats.It was statistically confirmed that both saving formats ("%-off" and "amount-off") lead to the same change (reduction) of the internal reference price.Hence, it can be said that when the discount level is low, marketers could choose to use either "%-off" or "amount-off" saving format.On the other hand, it was not statistically confirmed that in the case of a high discount level the "amount-off" saving format leads to a stronger change (reduction) of the internal reference price.Thus, also in this case, retailers are free to decide and choose between two saving formats ("%-off" or "amountoff") due to the fact that there are no significant differences.
Surprisingly, in the case of low price confidence, the results indicate that there is no statistical difference across two saving formats.Thus, it was not confirmed that when the price confidence is low the "amount-off" saving format leads to a stronger change (reduction) of the internal reference price.According to the results, both saving formats ("%-off" and "amount-off") lead to the same change (reduction) of the internal reference price.Also, in the case of high price confidence, the same situation occurs.It was statistically confirmed that both saving formats lead to the same change (reduction) of the internal reference price.Hence, in conclusion, it can be said that marketers in both cases of price confidence (low and high) could choose between the two saving formats due to the same effect on the internal reference price.
A few limitations of the presented research have to be mentioned.The first limitation is associated with the study sample.In the current research, the majority of the respondents were students.In fact, students have a lack of monetary outcome.This factor was not considered in the sample.Perhaps this target population was not appropriate for inclusion in the survey.Choosing a quota sample taking into account the age and income might perhaps have resulted in more accurate findings.As such, future studies could target a different population.
Another limitation could be related to the product category.For the presented research, two products within the same category (clothes) were used: jeans and t-shirts.Perhaps the use of different product categories could have shown other results.So, for future research it might be interesting to use products that are not within the same category.Finally, the fact that this research was not carried out in a natural setting is considered as a limitation.If the face-toface interview were to take place in a real buying situation (e.g.shopping mall, stores) the results would be possibly different.Thus, future researchers should conduct a field experiment instead of a laboratory experiment in order to obtain different results.

Table 1 :
Anchoring effects of advertised reference price and sale price: The moderating role of saving presentation format

Table 2 :
Cents or percents?The effects of promotion framing on price expectations and choice

Table 3 :
Do price promotions lead to a reduction of the consumer internal reference price and if so, when is this effect less strong?

Table 4 :
Mean values with "CIRP" as dependent variable

Table 5 :
Mean values with "CIRP" as dependent variable

Table 6 :
Mean values with "CIRP" as dependent variable

Table 7 :
Mean values with "CIRP" as dependent variable

Table 8 :
Mean values with "CIRP" as dependent variable Source: Author's own contribution