Consumption Expenditure and Gasoline Price
An Application of Regression Technique
Introduction
Econometrics is a subject that applies statistical tools to analyze economic data. We study relationships between economic variables or interpret economic phenomenon through econometric analysis of the relevant data. Regression is a tool of econometrics. Regression analysis is done to establish the relationship between two or more variables. This statistical analysis helps us to find whether there is any significant statistical relationship between two variables. More specifically, it tells us how a variable is affected by other variables. When we want to understand the impact of one variable on the other we use regression technique. For example, we want to understand the effect of the use of a particular fertilizer on the production of a crop. In that case crop production is the dependent variable and the amount of fertilizer is the independent variable. We can also use another variable here which affects crop yield that is, rainfall. Thus, in this analysis, there will be two independent variables and one dependent variable. Once we run the regression, we can understand whether the specified fertilizer does have any effect on the crop yield. The producers of the crop can then use the fertilizer according to the result. In this article we demonstrate the use of regression analysis taking an example from real life. In the next section we discuss the steps involved in making a regression analysis. In section III we present the objective of our study along with a brief review of the issues. In section IV we frame the regression model and discuss the data source. In section V we present the result and its analysis. In section VI we conclude our paper.
Steps Involved in Regression Analysis
The first step in regression analysis is to specify the objective of the study. We have to understand what relation we want to establish or hypothesize. So we have to frame a hypothesis that we have to test through our empirical analysis. Once we have framed the hypothesis the next step is to specify the model. We have to create a function to understand the relationship between the variables that our hypothesis requires . We have to identify the variables.
The parameters of the specified model have to be estimated through the empirical study. We have identified the variables clearly and find data on the variables. We can now use the data to run the regression exercise and find the coefficients of our model.
The final stage of the regression analysis is to analyze and interpret the results. We can go a step further after this to make meaningful predictions about future movements of the variables, based on our estimates. Now let us follow these steps and complete our regression analysis. We begin by specifying the objective of our study.
Objective of Study and Review of Issues
Gasoline is a necessity item in our daily life. Thus the demand for gasoline is price inelastic. That is as the price of gasoline falls the demand does not increase to any large extent. Similarly, as the price of gasoline rises the demand is not cut drastically. It has been observed in the past that a rise in the gasoline prices makes the household to save up from their grocery bills to compensate for the increased expenditure on gasoline . In the same way as the price of gasoline falls the demand for gasoline does not increase much/ So it saves consumers’ income spent on gasoline. This saved expenditure is spent on consumption . Thus the fall in the gasoline prices will be manifested in an increase in the consumer spending. The increased consumer spending will in turn increase the aggregate demand in the economy and lead to growth. Thus gasoline price fall can lead to rise in the growth rate.
In our analysis, we try to find the relationship between gasoline prices and the consumer spending in the US. We present the methodology and data in the next section.
Model Specification and Data Source
We want to find out whether the change in the gasoline prices has any effect on the consumer spending in the US. Let us frame a linear regression model as given below:
PC_Cons_Exp = b0 + b1gasoline_price + b2per-capita income + e
Where:
PC_Cons_Exp: is the per capita consumer spending
gasoline_price: is the price of gasoline
per_capita income: is the per capita income in the US
b0, b1 and b2 are the coefficients of the equation that we intend to estimate.
e is the error term
Since per capita expenditure is greatly influenced by the income level, we have kept income as an explanatory variable in the regression model.
We have collected data on the per capita income, gasoline prices and per capita consumer expenditure from the Federal Reserve Economic Data . We have taken the data from 1990 to 2015. This is a 26 year period. This period has seen ups and downs in the gasoline prices. So this can be a period from which we can understand the relationship between gasoline price and per capita consumption expenditure.
Result
The data that we have obtained from the Fred is presented in table 1 below. We present the regression result in table 2.
The regression result shows that both per capita income and gasoline price has a positive effect on the per capita consumption. The coefficient for per capita income is 1.451 and the coefficient for gasoline price is 430.06. The t stat for per capita income is 17.21 and the t stat for gasoline price is 3.6. Thus, both the coefficients are significant. The value of R square is 0.98 that means the model has a high degree of predictability. So we can rely on the results of the model.
The coefficient 1.451 suggests that if per capita income increases by 1 per cent per capita consumption will increase by 1.451 per cent. In the same way when gasoline price increases by 1 percent the consumer spending increases by 430.06 per cent. The positive values of the coefficient suggest that both the variables influence the per capita consumption in a positive way. Thus, contrary to popular notion the gasoline price does not influence consumer spending in a negative way. In fact, as gasoline prices rise the expenditure rises. This result supports microeconomic theory. When the price of a good that is price inelastic, rises the expenditure on the good increases as the demand cannot be reduced and when the price falls the falls as the demand is not increased to any significant extent. Thus the recent rise in the consumer spending has not been triggered by the fall in the fuel price, but there are other factors that are responsible for it.
Conclusion
In this paper, we presented a brief idea about how to conduct a regression analysis. We have framed a model and analyzed it empirically. But it should be remembered that a regression analysis has more details into it. Thus we need to modify the regression model time and again to get the best estimates.
Bibliography
Farrell, Diana, and Fiona Greig. How Falling Gas Prices Fule the COnsumer. Evidence From 25 Million People. Survey Report, JP MorganChase & Company, 2015.
Fred. Gasoline Prices, Per capita Income, Personal COnsumption Expenditur Per Capita. Economic Survey Report, St. Louis: Federal Reserve Bank of St. Louis, 2016.
Gicheva, Dora, Justine Hastings, and Sofia Villas-Boas. Revisiting the Income Effect: gasoline Prices and Grocery Purchases. NBER Working Paper No. 13614, Cambridge: NBER, 2007.
Woolridge, Jeffrey M. Econometrics. Cengage Learning, 2009.
