Consumption in an Economy


This paper analysis the determinants of consumption in an economy, it concentrates on the theory of Keynes to discuss the various factors that determine the level of income, consumption comprises the largest part in the level of income according to the income model by Keynes, therefore consumption as a component of income level in the national income level is an important factor to consider, various theories however exist to explain the determinants of consumption which are explained below.

Consumption determinant theories:

According to Keynes theory of income the income level constitutes consumption, investment, government expenditure, imports and income, the largest portion of income according to Keynes is for consumption purposes, consumption however according to Keynes was a function of income, whereby when the income level is high then the consumption level is also high.

Other theories exist that explain the determinants of consumption, example the relative income theory by James Duisenberg, the absolute income theory by Author Smith and the permanent income theory by Milton Friedman. These theories were introduced when after the marginal propensity of consumption and the average propensity to consume theory by Keynes failed to explain the status of the economy after world war two. However despite income Keynes also considered other variables that explain the consumption level, these variables include changes in the income distribution, the introduction of credit facilities in the economy and the existing stock of durable good in the economy.


Consumption in an Economy

The relative income by James Duisenberg focuses on the determinants of consumption at the individual household level, he stated that the consumption level of individual households will be affected by peer groups and neighbourhoods and also the previous peak income, in his theory he stated that the previous peak income was an important factor in determining consumption levels of an individual.

The absolute income theory was introduced by Author Smith who stated that the increase in the consumption levels after world war two was as a result of the following factors, the wealth effect whereby the lack of goods to purchase during the war made people to save their income and wealth, after the war people were in a position to purchase goods and for this reason the consumption level increased, he also considered the effect of rural urban migration whereby he explained that urban dwellers had a higher marginal propensity to consume than the rural population, for this reason therefore after world war two consumption increased as people migrated to urban centres and for this reason they experienced an increase in marginal propensity to consume. Finally he considered the importance of advertising in increasing consumption levels.

Milton Friedman introduced the permanent income theory. in his theory he considered the division of goods whereby there are those goods that are durable and those that are non durable, he stated that for the durable goods income levels were an important determinant of the consumption levels but he went ahead and stated that for the non durable goods the income levels were not important in determining the consumption levels. The permanent income theory also states that consumers make their purchase decisions through permanent income, by permanent income he meant that this was a recollection of past income levels, in this case therefore the consumption level was determined by the consumers permanent income.

From the above theories therefore it is clear that the consumption levels are determined by income, relative income and permanent income. Other theories such as the life cycle theory state that consumption is determined by the price of goods, life expectancy and the age where younger people had a higher consumption level.


Consumption in an Economy

According to the Keynesian theory Consumption is a function of income but in pour case we have to consider the disposable income, disposable income is that level of income that a consumer can use to purchase goods and services, the consumer disposable income is derived from the subtracting of tax from gross income. In this paper we consider data for the UK from 1994 to 2004 regarding disposable income and consumption levels. The data was retrieved from the UK national statistics website which is available at .


Data was retrieved from the UK national statistic website

Product.asp?vlnk=10808 , the link being data set according to September 2005, below is the data:

Household disposable income

Household consumption-ion





Consumption in an Economy













Consumption in an Economy













Consumption in an Economy









The data represents a trend of income and consumption from 1994 to 2004, this data will help us analyse Keynes theory of income and consumption, we will use the ordinary least square method whereby we will have consumption as our dependent variable and income as the independent variable, and also we will have an autonomous value.


Consumption in an Economy

Estimation model:

We specify our model as C = a + b Y, where C is consumption, a is the autonomous variable, b is the coefficient of income and Y is income level and this case we consider the disposable income. The value of b can also be described as the marginal propensity to consume.

Given our estimation model using statistical program we came up with our estimated model as follows: C = -141059.4 + 1.1742 Y

From our estimated we can conclude that if we hold all other factors constant and let the income level be equal to zero then our consumption level will be -141059.4, also if we hold all other factors constant and increase the level of income by one unit then the level of consumption will increase by 1.1742, for this reason therefore we find that as income increases then the level of consumption also increases.

Coefficient of determination:

In our estimation the coefficient of determination was close to one, the value of the coefficient of determination was 0.99195, to describe this it is clear that 99.195% of deviations in the dependent variable are explained by the independent variable, in other word this means that 99.195% of deviations in consumption are explained by income. This is a very strong relationship between our two variables we considered, for this reason it is highly possible that our results are valid according to Keynes theory.


Consumption in an Economy

Hypothesis testing:

In any regression and economic data analysis it is always important to test the hypothesis to check the statistical significance of coefficient estimated, this test includes defining the level of test and also considering the T ratio. The standard error of the autonomous value was 23071.5 and for the income coefficient the standard error was 0.035265, we perform a two tail test at 95% level of test the results were as follows:


calculated T

critical value  at 95%







Consumption in an Economy


The autonomous value (a)

Null hypothesis:

H0: a = 0

Alternative hypothesis:

H0: a ≠ 0

Our decision rule when testing the null hypothesis is that if the critical value is greater than the t calculated then we accept the null hypothesis, if the critical value is less than the T calculated then we reject the null hypothesis:

The above diagram shows a two tail test of the null hypothesis, in our case we are considering the autonomous value whose T critical is 2.306 and the T calculated is -6.1140108, because this is a two tail test then we wil consider the t critical as – 2.306 and because the calculated value is -6.14 which lies on the rejection area we reject the null hypothesis at 95% level of test and conclude that the autonomous value is statistically significant.

The income coefficient:


Consumption in an Economy

We state the hypothesis as

Null hypothesis:

H0: b = 0

Alternative hypothesis:

H0: b ≠ 0

After calculations we find that the T calculated value is 33.2964696 and our critical value is 2.306, for this reason therefore we conclude that the critical value is less than the calculated value and for his reason we reject the null hypothesis, when we reject the null hypothesis this means that the coefficient of income is statistically significant.

From our hypothesis test it is clear that both of our estimated coefficients are statistically significant using a two tail 95% test level, for this reason therefore our model could be used to undertake forecasts regarding consumption levels, this is also supported by the strong relationship that exist between the two variables as depicted by the coefficient of determination, the coefficient of determination depict that 99% of deviation in the dependent variable is explained by the independent variable.

Confidence interval:


Consumption in an Economy

We construct confidence intervals for our estimated coefficients at 95%, in the construction of a confidence interval we consider the coefficient value, its standard deviation and T statistic value, our T statistics value is 2.306, and the following are the results:

The autonomous value:

P(-194262) ≤ a ≤ (-87856.5) = 95%

From our result this means that we are 95% confident that our autonomous value lies between -94262 and -87856.5.

The income coefficient:

P(-0.08132) ≤ b ≤ (1.255521) = 95%

This means that we are 95% confident that the value of the income coefficient lies between -0.08132 and 1.255521.


Given the Keynes theory of consumption we used data from the UK statistics website to check the validity of this theory, after using data from 1994 to 2004 we estimated a model which depict that when there is zero levels of income then the level of consumption will be negative. The model also depict that an increase in income will increase the consumption levels.


Consumption in an Economy

Further test of statistical significance show that the model estimates is statistically significant and therefore our model can be used for forecasting purposes, also the high level of the coefficient of determination depict that 99% of variations in the consumption level are explained by the levels of income.

Other theories also explain the consumption level considering other factors, this include the relative income theory by James Duisenberg, the absolute income theory by Author Smith and the permanent income theory by Milton Friedman, in our analysis however we only sought to investigate Keynes theory that is based on the assumption that only income determines the consumption level.

However the Keynes theory of consumption has been criticised for having considered the issue of determinants of income so simple, consumption is complex to be only determined by income, there are other factors that Keynes did not consider that are of much importance to the determination of income levels, such factors include the factors stated by other theorists, the price levels, nature of goods, seasons and many other factors. However income is one of the major determinants of consumption and in a theoretical point of view the Keynes theory was a breakthrough into the determinants of consumption.


Alan C. (2003) Keynesian Economics: The First Principles, Rout ledge publishers, New York

G. Douglas (1997) Macro economic Theory: A Mathematical Treatment, Macmillan publishers,


Consumption in an Economy

New York

J. Wooldridge (2006) introduction to econometrics, 3rd edition, MIT publishers, London

Schmidt P.(1996) Econometrics, Marcel Decker publishers, New York