The Impact of Covid-19 Pandemic on Customers’ Spending Behavior in the United Arab Emirates

The Covid-19 pandemic has significantly impacted the global economy, causing consumers panic, mobility restrictions and resulting in job losses. Using survey data from 300 respondents, this study examines the effect of a Covid-19 pandemic on consumer spending behavior in the United Arab Emirates (UAE). The study finds that the spending behavior of UAE residents during the pandemic has not changed much compared to the pre-pandemic period, despite that some have lost a portion of their income. The study also reports that customers in the UAE have experienced lower panic-shopping behavior during the pandemic. In addition, the findings indicate that about 57% and 51% of the respondents lack adequate savings such as emergency fund and insurance respectively. Generally, it can be concluded that customers in the UAE were able to stabilize their spending behavior with no significant changes in their general spending behavior. The stable spending of UAE customers during the pandemic can be explained by the high level of optimism customers have on the likelihood of economic recovery in the post pandemic period.


Introduction
This paper aims to assess the relationship between the Covid-19 pandemic and consumer spending behavior in the United Arab Emirates (UAE). Understanding the spending behavior of customers during a crisis is essential for both marketers and policy makers, in order to develop crisis management strategies and maintain a smooth supply chain and inventory management of consumer goods, which could be a challenge during a crisis (Brandtner, Darbanian, Falatouri, & Udokwu, 2021). During the Covid-19 pandemic, consumers around the world have shown a panic buying and stockpiling activity, resulting in empty shelves and causing disturbance in organization and consumer behaviors (Taylor, 2021).
The Covid-19 pandemic has caused societal disruptions and has fall the whole world into high uncertainty, impacting the day-to-day activity of individuals and organizations (Ellison, McFadden, Rickard, & Wilson, 2021). The impact of the pandemic is tremendous, ranging from the discomfort of living in quarantine, wearing masks, the fear of getting infected to the significant disruption to the economic and political landscape. The mobility restrictions such as lockdown and quarantine that are imposed by governments along with the overall impact of the pandemic are believed to have caused drastic changes in the lifestyles and emotions of customers, including the way they shop (Ellison et al., 2021;Hassen, El Bilali, & Allahyari, 2020). In addition, Baker, Farrokhnia, Meyer, Pagel and Yannelis (2020) indicate that USA customers' spending had drastically increased (40% increase) in the early periods of the pandemic, but decreased by 25-30% during the lockdown periods. Ellison et al. (2021) also report that since the pandemic started more people tend to engage in online shopping. Butu et al. (2020) indicate that a pandemic tends to cause a significant disruption to the economy, leading to a change in the way businesses run and how consumers behave.
A growing number of scholarly works focus on the impact of the Covid-19 pandemic on consumer spending behaviors (Alexander & Karger, 2020;Chronopoulos, Lukas, & Wilson, 2020;Coibion, Gorodnichenko, & Weber, 2020). Previous studies have shown how any uncontrolled situation can change the way consumers spend their money. People tend to buy essentials and stockpile those items in fear of scarcity, they change the way they budget, spend, and save during crisis as they have a precautionary behavior, and they tend to invest in movable assets (Ballantine, Zafar, & Parsons, 2014;Giesing & Music, 2019;O'Neill & Xiao, 2012;). Giesing and Music (2019) argue that the level of optimism can change the way customers spend their money during the periods of crisis.
The recent studies on the impact of the covid-19 pandemic on consumer spending report a decline in spending due to lockdowns and loss of jobs and income (Dong, Gozgor, Lu, & Yan, 2021;Lahiri & Sinha, 2021). In addition, these studies note that many people have decided to restructure their expenditures by cutting down unnecessary expenses. However, at the early days of the pandemic customers have been observed conducting a panic buying and stockpiling their items (Chronopoulos et al., 2020). In a specific study on the impact of the lockdowns on consumer spending, Alexander and Karger (2020) and Coibion et al. (2020) indicate that the lockdown related restrictions have affected consumer spending negatively. Furthermore, Andersen, Hansen, Johannesen and Sheridan (2020a) in their study that focuses on the impact of lockdowns on different expenditure categories of customers in the Scandinavia (Denmark and Sweden) indicate that consumer spending depends on the nature of the item. For example, spending on grocery shops and pharmacies did not show a significant change, while expenses for restaurant meals, travel, retail, personal services, fuel, and entertainment have declined. Following Andersen et al. (2020a), Andersen, Hansen, Johannesen and Sheridan (2020b) examined whether age plays a role in determining the spending behavior in different categories of expenditure. Their results show that age played a significant role in spending and how it is impacted by the pandemic indicating an increase (decrease) of 5% (10%) for the old and young people respectively.
Although a growing number of studies focus on the impact of the covid-19 pandemic on different sectors of the economy, including consumers' spending, to our knowledge none of these studies focus on the UAE. The authors believe that given the diversity of the population in the UAE, understanding the spending behavior of consumers during a crisis could be of interest to businesses and policy makers. World Bank (2018) considers customers in the UAE as hard to be daunting. The UAE population is diverse with a significant proportion of foreigners ranging from low income spenders to big spenders. In addition, the World Bank (2018) reports three important features of UAE consumers that business should be aware of in order to understand their spending behavior. First, UAE customers demand high customer experience, including quality products; Second, UAE consumer composition is unique which includes, a diversified culture and income category, making it a suitable market for cheap to luxury products; finally, brands must know that gender behavior is vital as females in the UAE tend to spend more.
The UAE has reported the first Covid-19 case in January 2020 (KPMG, 2020) and has promptly taken several measures, including lockdowns and quarantine to limit the spread of the virus. Although the overall economic performance of the country recorded a sharp decline in the second quarter of the year 2020, the strong measures taken by the government seem to pay off by the third quarter, leading to a sense of economic recovery and creating optimism among UAE residents (United Arab Emirates Economic Outlook, 2020).
Despite the higher customer base in the UAE, the covid-19 related restrictions made by the government are believed to impact several sectors of the economy. KPMG (2020) report that the global and local Covid-19 mobility restrictions have impacted tourism as a result of the border closures and travel restrictions. Furthermore, Veira (2020) demonstrates that UAE customers' purchasing patterns have shifted from discretionary to essential spending during Covid-19. Despite this, Grimmelt, Iftahy, Khan and Youssef (2021) performed a poll focusing on the sentiment of UAE consumers during the corona virus crisis report that UAE customers are optimistic about the economic recovery. However, they anticipate a drop in their income in the near future, which means they must cut back on all except essential spending and strive to grow their emergency reserves.
The following is a breakdown of the paper's structure. The data source and methodology of the study are discussed in section 2, followed by the presentation of results in section 3. We report the discussion and analysis of the study in section 4. Section 5 presents the conclusions.

Conceptual Framework
Identifying the major factors that might impact consumer purchasing behavior is the most important component of understanding consumer spending behavior. According to Kotler and Armstrong (2012) there are four main factors that could influence the way consumers behave. These factors include (i) situation related factors such as social situation, time, reason for purchase and the mood of the buyer or seller; (ii) personal factors related to personal view, lifestyle, gender, age; (iii) Psychological factors such as motivation, perception, attitude and (iv) Societal factors such as culture, class, family, etc. These factors are expected to influence the spending behavior of consumers.
Taking Kotler and Armstrong (2012) framework as a benchmark, this study developed the following conceptual framework to explain the factors that influence the spending behavior of residents in the UAE during the Covid-19 pandemic as shown in Figure 1. The study predicts that gender, age, employment status, income level and savings culture are likely to affect the way consumers spend their money, the type of products they spend on and the way they manage their overall finances during a crisis.

Data Source and Sampling
Data is collected from 300 respondents across the UAE using a survey monkey questionnaire distributed via social Media such as WhatsApp, Instagram and Facebook between November 1 and February 26, 2021. The survey was initially distributed to our networks, and then used a snowball sampling, asking the initial respondents to refer the survey link to their acquaintances. The snowball sampling has helped us to increase the size of the sample. The Snowball sampling procedure is one of the most commonly used sampling technique in social science studies (e.g., Markos, Alexandros, & Irini, 2021).
Before formally distributing the survey questionnaire, we have conducted a test on the survey questionnaire using a peer review and pilot study. The goal of the test was to assess the appropriateness, clarity and reliability of the survey questions.
In the first stage of the test, the draft questionnaire was sent to one colleague for feedback. The feedback received from the reviewer was used to revise the survey questionnaire, giving special emphasis to the clarity of the questions by reducing vague words or statements, repetitions and inclusion of missing items. In the second stage of the test, we conducted a pilot study with 20 survey respondents in order to determine the clarity and reliability of the questionnaire constructs. The survey test procedure has improved our confidence in the reliability of the questionnaire constructs. Thus, we conducted the final data collection using the survey questionnaire items adopted after the pilot study.
Our final sample includes 300 respondents, composed of 46% females and 54% males. Our sample also includes diverse age participants, including 18-24 years (34.67%), 25-34 years (36%), 35-45 (22.67%) and over 45 years (6.66%). The majority of respondents (81.67%) have regular income and the remaining 18.33% do not have a regular income. Of those who have indicated that they have a regular income, 54.70% and 8.57% have a full-time and par-time jobs respectively, while 28.16% of the respondents receive their regular income from their families or government. Only 8.57% of the respondents generate income through self-employment sources.
The survey questionnaire asks respondents about their gender, age, source of income, their saving habits and the proportion of income they save. Respondents were asked whether their income and expenditures have increased, decreased or remain the same compared to the pre-pandemic period. In addition, a list of twelve expenditure items was provided and asked respondents to identify the change in their monthly spending for each of the expenditure items such as grocery, rent, transportation, dining out, education etc. Lastly, we have asked respondents, eight questions such as "I have enough money each month to pay my household and utility expenses" to assess their preparedness to cope up with emergencies and unexpected events such as the pandemic.
The questionnaire covers five key socioeconomic factors which are believed to have an impact on consumer spending behavior. It focuses on assessing whether gender, age, employment, income and savings affect consumer spending during a crisis. The study mainly examines whether there is a difference in spending behavior between the pre-pandemic and during pandemic periods. We measure spending behavior using a list of expenditure items, asking respondents whether their monthly spending has increased, reduced or remains the same compared to the pre-pandemic period.

Method of Data Analysis
The data are analyzed using simple descriptive statistic measures such as averages, standard deviation and chisquare tests. The Chi-Square test is a nonparametric test that determines if categorical variables are related (Singhal & Rana, 2015). The data for this test comes from the general population; results are regarded sufficient if they are not less than 5 and there are no cells with a zero count. The Chi-square for a given data is determined using the following equation.

= ∑
(1) Where, 'O' refers to the observed frequency and E represents the expected frequency. Singhal and Rana (2015) indicate that the Ch-square result measures whether two variables are associated with each other or not. However, it is important to remember that this test only determines whether two variables are related to one another, not how they are related, which might be a drawback. The study uses SPSS statistical software to run the Chi-square tests. Table 1 reports the socio-economic characteristics of the sample respondents. In addition to identifying their gender and age group, respondents were asked whether they have a regular income and if the save regularly to put money aside for emergency spending. The gender composition of the respondents shows that 162 male (54%) and 138 female (46%) are included in the sample. Four age categories were used, ranging from 18 -24, 25 -34, 35 -45 and 46 years or older. The summary data indicate that the majority of the respondents' age fall under the 25-34 years old category (36%), followed by the 18 -24 years old class (34.7%) and only 6.7% of the respondents are older than 45 years old.

Demographic Characteristics
Out of the 300 respondents, 245 (81.7%) respondents have responded that they have a regular income, while 55 (18.3%) respondents indicated that they do not have a regular income. Those participants with a regular income were asked to describe their source of income. Table 2 reports that 134 (54.7%) respondents receive regular income from their full-time job, while 21 (8.57%) of the respondents receive a regular income from their par-time job. On the other hand, 55 (22.45%), 14 (5.7%) and 21 (8.57%) of the respondents receive regular income from family support, government subsidy and other sources respectively. When asked about their saving culture, 101 (33.7%) of the respondents indicated that they do not save regularly.
The 199 (66.3%) respondents who have indicated that they make regular savings, were asked the proportion of savings they make every month. Table 3 presents the proportion of savings made by the respondents of the survey. The majority of the respondents 60 (30.15%) save between 5 -10% of their monthly income, followed by 42 (21.11%) respondents who save less than 5% of their monthly income. Only 28 (14.07%) of the respondents save more than 40% of their monthly income.

UAE Residents' Income before and during the Pandemic
The respondents were also asked to rate their pre Covid-19 period income compared with the income during the Covid-19 period using five Likert scale options, from 'Significantly lower' to 'Significantly higher', as shown in table 4. The majority of the respondents (57.3%) indicate that their income remains the same. However about 33 (11.1%) and 16 (5.3%) of the respondents had a slightly lower and significantly lower income before the Covid-19 period, respectively, indicating that their income has increased during the Covid-19 period. On the other hand, 36 (12.0%) and 43 (14.30%) of the respondents indicated that their income before the Covid-19 period was slightly higher and significantly higher respectively, confirming a decline in their income during the Covid-19 period.

UAE Residents' Spending before and during the Pandemic
This section reports the results related to UAE residents' monthly household expenditures during the Covid pandemic compared with the pre-pandemic period. The paper examines consumers' spending over twelve categories of expenditures using a five scale Likert scale ranging from "significantly lower" to "significantly higher." Figure 3 reports the respondents' spending during the pandemic compared with their monthly spending before the pandemic. Overall, figure 3 shows that, the majority of the respondents have indicated that their monthly household spending during the pandemic remained the same as the pre-pandemic period.   The respondents who have indicated either slight or significant decline in their expenditure during the pandemic period were asked a follow up question. Figure 4 reports the reasons for a decline in the respondents' monthly household expenditure.

Figure 4. Reasons of lower spending
This section reports the factors that have implications on the spending behavior of consumers. The respondents were asked eight questions that assess their financial and risk management ability to cope up with the difficulties during a crisis. The following questions were asked and the respondents were requested to respond as 'Yes' or 'No'.
• I have enough money each month to pay my household and utility expenses; • I have enough money to pay for an emergency; • I save regularly for travel expenses and future plans; • I have at least three months' expenses set aside in a readily accessible account; • I have adequate insurance to cover "big" unexpected expenses, such as a hospital bill, disability, or liability for damages to others; • I use less than 20 percent of my monthly income to pay my credit cards, entertainment, and car service payments; • I pay credit card bills in full to avoid interest charges; • I avoid impulse purchases and do not use shopping as a form of recreation.  Table 5 focuses on the respondents whose income have declined during the pandemic classified by their age category. The matching between the age categories and the income indicate that the 18 to 24 years old age group was the most affected in terms of income decline during the pandemic. Figure 6 shows that the income for the 18-24 years old age category was slightly or significantly higher before the pandemic. The study also used the Chisquare test using SPSS software to test the independence and correlation between the age and income before and during the pandemic and the results are reported in Table 7.

Figure 6. The relationship between age and income during the pandemic
Note. This graph depicts the age group that has shown the highest reduction in income during the pandemic.  The chi-square test of independence is used to determine whether age has a statistically significant relationship with income during the Covid pandemic period. The statistical results are presented in the following contingency table. The P-values for the Pearson chi-square and likelihood ratios are lower than the significance level (5%), indicating that the association between age and income during the pandemic is significant. Note. In these results, 5 cells (25.0%) have expected count less than 5 and the minimum expected count is 1.07.  Vol. 17, No. 9;2022 As it is seen in table 7, 5 cells (25.0%) have an anticipated count less than 5, which should ideally be less than or equal to 5 in order to reject the null hypothesis. The predicted minimum count is 1.07. Therefore, based on the chisquare result, it can be concluded that younger consumers were impacted the most by the income change due to the pandemic. Table 8 reports the Spearman's rho correlation coefficients. Our data is measured in nominal and ordinal scales, and the data distribution is not normal. Thus, the Spearman's rho correlation analysis is the appropriate tool to assess the strength and direction of the relationship between the variables of the study. We have computed the Pvalues of the t-scores for each correlation coefficient to identify how significant the relationship is. For P ≤ 0.01, the relationship is very strong (***), 0.01˂ P ≤ 0.05, the relationship is strong (**) and 0.05 ˂P ≤ 0.10, the relationship will be considered marginally significant (*). Notes. *** significance level at 1%, ** significance level at 5%, *significance level at 10%.

Correlation Analysis
The results reported in Table 8 indicate that there is a marginal relationship between the savings and, gender and age of the respondents. On the other hand, having a job (employment) is positively and strongly related to gender and age. Similarly, income is positively and significantly related to gender, age and savings. The spending variable is negatively and significantly related to gender. Table 8 also provides important results related to the available funds for emergency (Emergency funds) and being economical in spending (economical). It is not surprising that there is a positive and strongly significant relationship between emergency funds and savings. The emergency funds variable is also positively and significantly related to employment. Likewise, being economical in spending is positively and significantly related to savings, employment and emergency funds. There is also a positive and significant relationship between economical and gender. Table 4 reports that the majority of the respondents (57.3%) indicate that their income remains the same, while 26.3% of the respondents reported a decline in their income during the pandemic. Nevertheless, 16.4% of the respondents have indicated that their income has increased during the pandemic. The decline in income can be explained by several factors such as pay decrease, job loss, lockdown/quarantine etc. Figure 3 reports whether the monthly household expenditures of UAE residents have changed during the pandemic.

Discussion
Respondents were asked to identify whether their expenditures on twelve items has changed or not, based on a five-point Likert scale, ranging from 'significantly lower' to 'significantly higher'. As it is shown in figure 3, respondents have indicated that their spending has remained the same on nine expenditure items out of the twelve expenditure categories provided. On the other hand, their spending on three items including travel, dining out, and gym or sport seems to decline during the pandemic period compared with the pre-pandemic period. The decline in spending in these three categories is mainly due to the government mobility restriction policies that include shutting down service providers and travel ban. In addition, many people still fear visiting public areas such as gyms in fear of getting infected by the virus. Currently dining out at restaurants seems to get back to its prepandemic period level, because the precautionary measures taken by the service providers install confidence in consumers. Figure 4 shows that 36% of the participants chose the option 'other' factors other than job loss, budget management, pay decrease and lockdown/quarantine contributed to their decline in their spending during the Covid pandemic period. On the other hand, about 35% of the respondents have indicated that their spending during the pandemic period has reduced mainly due to the government imposed lockdown/quarantine. About 17% of the respondents have cut down their expenditures during the pandemic for the purpose of budget management. Nearly 12% of the ijbm.ccsenet.org International Journal of Business and Management Vol. 17, No. 9;2022 respondents have indicated that they have reduced their spending due to job loss/pay decrease. Figure 6, table 5, 6, and 7 shows the result of the comparison between the age intervals and the income change after the pandemic. The result of the Chi-square test indicated that the age category of 18 to 24 is the most affected category between the other three intervals. This can be due to the job instability. People at this age can start their work life and since it's the beginning of their work life, it means that their job is not secure. It's known in UAE's labor law that to start working they must be 18 or older. On the other hand, it was also proved in the last question that most of the people who live in the UAE didn't do any impulse buying as 67% of the participants mentioned that they do not use shopping as a form of recreation, and they avoided the impulse purchases during the pandemic.
It is also important to note that the UAE's customers seem not to give enough emphasis to risk management. The survey result indicates that 57% and 51% of the respondents do not have adequate savings and insurance to cover future an expected expenses. In addition, 49% of the respondents have indicated that they do not save regularly for travel expenses and future plans despite their plan to travel as it is reported in Figure 6. The lower savings and less preparations for future unexpected expenses could be explained by the nature of consumers in the UAE. The majority of the UAE residents are expatriates, and they tend to transfer part of their monthly incomes to their home country, which is more likely to impact their savings. In addition, the high optimism on the recovery of the UAE economy during the post pandemic might have increased the confidence of consumers, leading into lower household level risk management activities.
Finally, the correlation coefficients reported in table 8 report important findings on the relationship between the variables of the study. Having a job (employment) is positively and strongly associated with gender and age, indicating that female and younger people tend to be unemployed. Our survey data indicate that those who do not have a job, receive financial supports from their families and government. A positive and significant relationship between income and, gender, age and savings suggest that male, older person and a person who regularly save are more likely to earn a higher income. On the other hand, spending is negatively and significantly related to gender, suggesting that females tend to spend more than males. The availability of emergency fund (emergency funds) is associated with savings and employment, suggesting that a person who has a job and saves regularly tends to maintain emergency funds to meet unexpected expenditures in the future. Although marginally significant, the association between emergency funds and spending is negative, indicating that a person who is able to lower his/her periodic spending is likely to maintain an emergency fund. Likewise, being economical in spending is positively associated with gender, savings, employment and emergency funds. This result depicts that a person who prioritizes his/her spending and avoids loans and penalties on credit cards is more likely to regularly save and maintain emergency funds. In addition, male and a person who has a job (employment) tend to be more economical in their spending. Overall, our results indicate that gender, employment, savings and maintenance of emergency funds are important factors in shaping the spending behavior of customers.

Conclusion
The Covid-19 pandemic continued to be the major uncertainty across the globe, impacting the livelihoods of billions of people. Using a survey data from 300 respondents across the UAE, this study examines the impact of Covid-19 pandemic on customer spending behavior. This study presents three key findings. First, the results demonstrate that the overall customers' spending in the UAE tend to be stable during the pandemic, in contrast to the customer spending in other countries such as the United States (e.g., Dong et al., 2021). This finding suggests that the decline in customer spending can be attributed to loss of income, mobility restrictions and uncertainty about the economic recovery. The stable spending of UAE customers during the pandemic can be explained by the high level of optimism customers have on the likelihood of economic recovery in the post pandemic period. A study conducted by McKinsey & Company (2021) indicates that customers in the UAE are very optimistic about the economic recovery despite the existing covid-19 related restrictions. The customers in the UAE were able to maintain stable spending and income during the pandemic as compared with the pre-pandemic period. However, customers in the UAE revealed that their expenses related to travelling, gym or sport, and dining-out were lower during the pandemic compared to the pre-pandemic period. A decline in the aforementioned expenses is mainly due to the covid-19 related restrictions; second, the study also reports that customers in the UAE have experienced lower panic-shopping behavior and avoided impulse buying during the pandemic. The study also indicates that about 57% and 51% of the respondents lack adequate savings such as emergency fund and insurance respectively; the findings indicate that UAE customers lack the required fund to meet future unexpected expenses despite they are able to maintain their spending. Most importantly, young people (18 -24 years old) tend to face a decline in income due to job loss, salary reduction or a reduction in family support due to the uncertainty during a pandemic. Overall, our results indicate that gender, employment, savings and maintenance of emergency funds are important factors in shaping the spending behavior of customers.
Lastly, the authors believe that it would be interesting if future studies examine the risk management skills of the UAE consumers and their implication to future crisis management. Future research could also investigate whether spending behavior of customers vary by nationality (Locals versus expatriates) and whether the spending behavior of customers has changed after the announcements of vaccine rollouts.