Determining The Effect That Democratic and Republican Parties Have on The Unemployment Rate

Introduction

One of the main subjects that is discussed prior to elections is how each candidate will use monetary and fiscal policy to affect such variables as unemployment and inflation. Since, as the Phillips Curve shows us, the change in unemployment and the change in inflation are negatively correlated, and it is generally considered favorable for each to have low growth, each candidate has to choose whether a low unemployment rate or a low inflation rate is more favorable. It is generally assumed that the decision will be made according to the candidates’ political party: Democrats aim to lower unemployment, and Republicans aim to lower inflation. It therefore seems probable that unemployment will on average be lower during the terms of democratic presidents than during terms of republican presidents. However, the data and regressions presented here show that political party has no significant effect on the unemployment rate.

Model

To determine the factors that have an effect on unemployment, I used the economic model that shows that unemployment is a shortage in the demand for labor. Therefore to reduce unemployment the demand for labor has to be increased, which is done through increasing the equilibrium quantity of aggregate demand. In the aggregate supply - aggregate demand model (AS-AD model), aggregate demand is the sum of consumption, investment, government expenditure and the net exports (exports - imports). Anything that has an effect on these factors, including monetary and fiscal policy, will in turn have an effect on unemployment. For example, as demand for goods and services increase, firms have to increase supply to meet the higher demand. To increase output requires an increase in inputs, specifically, labor and materials. This increase produces a shift in the demand curve in the labor market. A shift in the demand, which is an increase in demand for labor at any given wage results in a reduction in unemployment. To create an equation I used unemployment as the dependent variable, and the independent variables of government deficit, change in money supply, the T - bill interest rate and a dummy variable called Democrat that has a value of one if the president for that period was a democrat and a value of zero if the president was a republican. Together these variables account for the affects that the government can have on the unemployment rate. The equation is as follows:

Unemployment = b₀ + b_1*Democrat + b_2*Govdef + b_3*Changemoneysup + b_4*Tbillinterstrate + e

Data

The data for the regression was taken from a DRI database from Eviews. The observations used are quarterly from the second quarter of 1959 to the third quarter of 2002. Where appropriate, the variables are in 2002 dollars. The change in money supply variable measures the percent difference in the money supply in the current period compared to the last period. The T-Bill interest rate is the rate on six month U.S. Treasury Bills. Since interest rates tend to rise and fall together, I used this as a representative of interest rates in general. Essentially it is measuring the cost of borrowing money. The variable government deficit is measured as the percent of real GDP. Although taxation is fiscal tool that affects the unemployment rate, it is not represented as its own variable in this regression. The form that the government deficit variable takes accounts for the effects of taxation. This is because the government deficit is the difference between the amount of tax collected by the government and the total government expenditure. For this regression the variable represents the effect that tax has on unemployment as well as the effect that government expenditure has. The Democrat variable is assigned a value of one during the quarters there are Democratic presidents and a value of zero during the quarters there are Republican presidents.

Results

The regression of the equation:

Unemployment = b₀ + b_1*Democrat + b_2*Govdef + b_3*Changemoneysup + b_4*Tbillinterstrate + e

produced estimates for the coefficients for each of the variables. The results of the regression showed the coefficient of the Democrat variable to be -0.25. This means that the unemployment rate is only a quarter of a point lower when Democrats are in office. However, the results show the standard error to be 0.17. Using this standard error, I tested to see whether the coefficient is significantly different from zero. The results from a T-test gave the variable a T-statistic of -1.46. The absolute value of the T-statistic is less than the critical value of 1.98, therefore it cannot be determined that the coefficient for the Democrat variable does not equal zero. Therefore no difference in the unemployment rate can be determined for either political party. The coefficients and standard errors for all of the parameters of the equation are in the table below. All of the other variables have T-statistics that are higher than the critical value of 1.98, they are therefore deemed significantly different from zero.

Sample (adjusted): 1959:2 2002:3
Included observations: 174 after adjusting endpoints
Variable	Coefficient	Std. Error	t-Statistic	Prob.
DEMOCRAT	-0.250848	0.171432	-1.463250	0.1453
TBILLINTERSTRATE	0.243937	0.034300	7.111757	0.0000
CHANGEMONEYSUP	0.267808	0.043144	6.207244	0.0000
GOVDEF	-0.007915	0.000797	-9.928842	0.0000
C	3.729840	0.290841	12.82433	0.0000
R-squared	0.531478	Mean dependent var	5.918199
Adjusted R-squared	0.520389	S.D. dependent var	1.480551
S.E. of regression	1.025340	Akaike info criterion	2.916241
Sum squared resid	177.6735	Schwarz criterion	3.007018
Log likelihood	-248.7130	F-statistic	47.92718
Durbin-Watson stat	0.271623	Prob(F-statistic)	0.000000

Conclusion

The variables for t-bill interest rate, change in money supply and government deficit all have a significant effect on the unemployment rate. When these variables change, the unemployment rate will also change. The Democrat variable is not significant, which means that the change in the unemployment rate will on average be the same when there is a Democratic president and when there is a Republican president. This means that although the two parties are assumed to have different goals the end result for the unemployment rate is the same.

Site Author: Traci Kinney

Last modified: 11/5/03