The main goal of this paper will be to determine which of the two is more important in case of consumption. First of all, a theoretical analysis of the possible explanations of each relationship based on the existing literature will be presented. Then the research will proceed to the empirical tests (correlation and regression using SPSS Statistics software) of a created model based on consumption, gross national income as well as length of retirement in the United States and the United Kingdom over the period of 46 years. The piece will finalise with some general conclusions that can be drawn from the conducted research.
Consumer behaviour is a subject of interest to many different players in the economy. Companies gather information on their current and potential customers to determine their selling and production strategies based on what their target customers might want, need, the timing at which to introduce new products, the prices to suit their tastes etc. Governments use the information on their citizens’ spending habits to figure out what policies to implement to receive effective results and garner political support for re-election. People themselves should be acquainted with the subject of consumer behaviour to better understand and use personal spending under different political, economic and social circumstances to lead more stable lives. Because of the immense interest in the subject of consumption many efforts have been put into research to distinguish the most important factors that influence consumer behaviour and so far much of such literature has been dedicated to the influence of current income and expectations (specifically expectations about future income).
Inhaltsverzeichnis
Theoretical analysis
Current Income Influence on Consumption
Future Income Influence on Consumption
Expectations for Future Income
Net Worth and Consumption
Uncertainty Influence on Consumption
Mix and Match
Empirical analysis methods
Empirical analysis
Empirical results from the U.S. data
Conclusions
References
What is more important in case of consumption: current income or expectations?
Consumer behaviour is a subject of interest to many different players in the economy. Companies gather information on their current and potential customers to determine their selling and production strategies based on what their target customers might want, need, the timing at which to introduce new products, the prices to suit their tastes etc. Governments use the information on their citizens’ spending habits to figure out what policies to implement to receive effective results and garner political support for re-election. People themselves should be acquainted with the subject of consumer behaviour to better understand and use personal spending under different political, economic and social circumstances to lead more stable lives. Because of the immense interest in the subject of consumption many efforts have been put into research to distinguish the most important factors that influence consumer behaviour and so far much of such literature has been dedicated to the influence of current income and expectations (specifically expectations about future income). Therefore, the main goal of this paper will be to determine which of the two is more important in case of consumption.
First of all, a theoretical analysis of the possible explanations of each relationship based on the existing literature will be presented. Then the research will proceed to the empirical tests (correlation and regression using SPSS Statistics software) of a created model based on consumption, gross national income as well as length of retirement in the United States and the United Kingdom over the period of 46 years. The piece will finalise with some general conclusions that can be drawn from the conducted research.
Theoretical analysis
Vast literature has been written about consumption and its dependence on current and expected future income. Most of it empirically challenge the ideas of the Life Cycle Model (LCM) and Permanent Income Hypothesis (PIH) yet they all come up with different findings on which resources have more impact on individuals’ consumption. To begin with, the above mentioned models should be defined. As Borella, Moscarola and Rossi (2011) point out, Life Cycle Model predicts that income fluctuations over a person’s life span should not impact their consumption and income smoothing should be done via borrowing and saving or dissaving. Thus, consumption is said to not only depend on current income, but on net worth and future income as well. Another widely discussed model, the Permanent Income Hypothesis, as defined by Carroll (1994) states that current consumption is a function of a present discounted value of future income. Therefore, this literature review discusses the past works on these hypotheses and provides a background for our own empirical study.
Current Income Influence on Consumption
First, we should discuss the literature and its arguments in favour of the notion that current income is more important to consumption than income expectations. According to Carroll (1994), there is very little evidence that would support different claims as his work with Consumer Expenditure Survey tests the LCM and finds that consumption is closely related to current income and unrelated to expected changes in income. Thus, the current income may play the major role. In addition to this, Mayer (1972) and Lavi (2003) agree that the social class one belongs to determines their consumption behaviour. More importantly, Mayer (1972) points out that permanent income is the main driver of consumption due to psychological factors since individuals in developed countries are used to fulfilling all their needs and some additional wants, therefore, increases in income do not change people’s habits and individuals only make adjustments when income decreases. This implication thus goes with above mentioned Carroll’s (1994) inference that consumption depends on income. On the other hand, Lavi’s (2003) study on consumption in Israel notices that changes in expenditure happen in both cases: when the permanent income increases as well as decreases, but only in the long run. According to his findings, consumer behaviour is not affected in the short term yet this changes when low-income individuals are affected by changes in earnings – small rises immediately lead to higher consumption (Lavi, 2003). Moreover, Baxter & Jermann (1999) in their research found an argument that consumption is excessively sensitive to income in the sense that current income still has explanatory power after accounting for the innovation in permanent income, thus, concluding that current income plays a significant role explaining impact on consumption.
Future Income Influence on Consumption
A mass of studies with different target populations have been conducted to test the above mentioned Life Cycle Model and many of them have controversial findings which argue that consumption depends rather on future income expectations than on current income. Gustman and Stafford’s (1972) study on graduate students’ finance survey concludes that consumption is more related to future income expectations than to past consumption and income experience. Although such findings could be argued due to the respondents’ young age, a 2009 study by Pounder on older households (ages from fifty-three to seventy-three years) from the Health and Retirement Study also provides empirical evidence in favour of claims that household consumption today is dependent on its future income expectations. The study uses a survey and administrative data on social security plans. Based on the findings, households with the same net worth and current income but lower expected income tend to consume less than those with higher expected future income
Expectations for Future Income
As mentioned above, a number of scientists argue that future income expectations are pivotal on consumers’ decisions. Thus, it is important to define what these expectations depend on and how one can estimate their future earnings. A few of the studies link expectations of earnings with social and economic aspects, such as occupation, education, business cycle etc. Firstly, age is positively correlated with consumption: “Age should serve as a proxy for a learned standard of living and knowledge about professional earnings, as well as for the level of composition of physical wealth” (Gustman & Stafford, 1972, p.1250). Therefore, other things being equal, older students are expected to spend more because they are more aware about their income possibilities.
Secondly, Gustman & Stafford (1972), as well as Mayer (1972), Pounder (2009) and Borella, Moscarola & Rossi (2011) all agree individuals’ education and occupational choices play a major role. Students who expect to be working more highly paid jobs are more likely to consume more while they are studying than those who expect lower wages (Gustman & Stafford, 1972). Similarly, Pounder (2009) notices that people’s employment status in terms of education, occupation and industry may have heterogeneous sizes in their pensions. Therefore, both, students and older individuals with higher education and high-paying careers tend to expect higher wages as well as retirement benefits than those who are less educated which allows the former to consume more. In addition, Borella, Moscarola and Rossi (2011) find that on its own expectedness of retirement (as proxy for future income) does not play a particularly strong role, however, different level of education does seem to produce different attitude towards consumption (higher level means more stable lifetime consumption and vice versa).
Thirdly, Brown & Taylor (2006) point out that people’s financial predictions are also influenced by the business cycle. The authors find that “financial realizations tend to fall short of expectations, which may be taken as an indicator that individuals may have a tendency to under-commit themselves financially” (Brown & Taylor, 2006 p. 18). Thus, people expecting to earn more during booms and less during recessions adjust their consumption accordingly.
Furthermore, Gustman and Stafford’s (1972) work on graduate students assess other socioeconomic variables that impact individuals’ income expectations. For example, rich family off-springs tend to consume more which can be explained either by their past habits or by future expectations of inherited wealth (ibid.). The study also finds that a person’s marital status and number of children have a significant correlation with the consumption and at the same time have higher income expectations (because they are living with an educated partner), therefore tend to consume more (ibid.). The study also finds a positive relationship between leisure time and consumption – the more leisure time you have, the more time you have to spend your money.
Net Worth and Consumption
As mentioned earlier, net worth is one of the resources individuals can finance their consumption with. Although its importance is not as crucial compared with current and future income, studies have found interesting yet unconfirmed relations with consumption. Gustman and Stafford (1972) define graduate students’ net physical wealth as a rather shady variable because it can have a positive effect on consumption due to increases in permanent income or inheritance while at the same time the effect can be negative if it is measured as a past consumption (the higher the past consumption – the lower the physical wealth). Based on the findings of the study of Pounder (2009), households with the same net worth and current income but lower expected income tend to consume less than those with higher expected future income.
Uncertainty Influence on Consumption
If individuals had perfect information on any income change they were going to experience, they could adjust their income accordingly. However, some people tend to experience income uncertainty due to various reasons, thus, it is possible to assume individuals’ consumption depends on their risk averseness. Carroll’s (1994) work finds that consumption is closely related to current income but unrelated to expected changes in income. However, the study outlines the importance of future income uncertainty and argues that consumers who are faced with a higher income uncertainty tend to consume less in order to save more (ibid.). Similarly, Pounder (2009) finds that households with higher risk aversion tend to consume less out of their future income. However, all households then are inclined to consume more out of such future income sources which are predictable, for example, Social Security benefits (ibid.). Brown and Taylor (2006) also find that individuals with a higher income risk tend to save more whereas more financially optimistic individuals save less.
The relation between uncertainty and consumption leads to the idea that long-term planning is common among low-income individuals as they are usually faced with more uncertainty and, thus, save to smooth consumption. This idea can be further extended to national level as findings on consumption in Pakistan – a low income country – made by Khan, Azam and Qureshi (2014) suggest that locals’ consumption depends on changes in permanent income (indicating a forward-looking tendency) with strongest results seen in the long term, for low-income households, especially so when the changes in permanent income are negative. Whereas, in higher-income countries like the United States (Pounder, 2009; Carroll, 1994), Israel (Lavi, 2003) or Italy (Borella, Moscarola & Rossi, 2011) current income (especially amongst higher income earners) typically has stronger influence, because of usual day to day consumption habits that are much more easily kept under stable economies making it less essential to think much about the future.
Mix and Match
What we can conclude from the literature review is that there is evidence that consumption depends on both current and expected income. As Campbell & Mankiw (1989) argue, the best way to explain consumer behaviour is to assume that some people act according to the rule of thumb therefore they spend based on their current income and others are forward-looking so they consume while thinking of permanent income an try to smooth it out to be able to keep the spending stable throughout their lifetime. This kind of view, in addition to the consideration of other social and economic variables mentioned in the previous sections, would produce the most accurate theoretical models of the reality.
Empirical analysis methods
An empirical test was conducted to determine which of the indicators – current income or expectations - is more important when determining consumption. The following section will provide an explanation of the indicators chosen, methods and countries used for testing.
First of all, the economic measures used while conducting the statistical tests will be presented. The dependent variable in the study was household final consumption expenditure per capita which represents all of the final goods and services purchased by households in the economy, but does not include housing purchases for smoothness (World Bank, n.d.a.).
To represent current income an economic indicator of gross national income (GNI) per capita (World Bank, n.d.b.) was chosen. To put it simply, GNI is all of the income made by country’s residents (both income remitted to the country by citizens living abroad and that is spent within the country by locals and foreigners) and local businesses (both domestically and from abroad) over the course of a year.
[...]