Volume 3, Issue 1, June (2016)

  Cluster Sample Inference with Very Few Groups                    1-10  
  Johan Vikström [1] 

[1] IFAU Uppsala, Box 513, S-752 20 Uppsala, Sweden and UCLS, Department of Economics University, johan.vikstrom@ifau.uu.se

 
  Abstract  
  If the number of clusters equals the number of cluster-invariant regressors the usual clusters-specific standard errors cannot be computed. One important example is the difference in differences setting with two groups and two time periods. This paper introduces a new method to calculate how large the within cluster correlation of the model errors can be yet still enable OLS coefficients to retain statistical significance at a desired level. This sensitivity approach is applied to previous results regarding the effects of workers compensation benefits.  
  Keywords:  Cluster effect; cluster robust; difference in difference; sensitivity; workers compensation.

JEL Codes: C12; C21; C23.

doi: 10.5455/Elet.2016.3.1.1

 
 

Do Regional Differentials in Economic Freedom Yield Regional Unemployment Rate Differentials in the U.S.? A Brief Exploratory Note Adopting Panel Data Analysis    11-25

 
  Richard J. Cebula[1]

[1] Davis College of Business, Jacksonville University, 2800 University Boulevard N, Jacksonville, FL 32211 U. S. A; dr.richardcebula@gmail.com

 
  Abstract  
  This study empirically investigates the hypothesis that the higher the degree of economic freedom in a state, the lower the unemployment rate in that state, other thing held the same. This hypothesis is based on the principle that greater economic freedom leads to greater real economic growth, which in turn reduces the unemployment rate. The framework studied consists of a panel dataset for the 50 U.S. states representing the period 2000 to 2012, a period which includes the Great Recession, as well as years prior to and following the Great Recession. The model estimated in this study includes for each state, in addition to a measure/index of overall economic freedom, control variables for the percentage of the population that is black, the percentage of the population that is Hispanic/Latin, the female labor force participation rate, and the percent of the population (age 25 and older) with at least a high school diploma. A number of state fixed-effects and dynamic panel data estimations are provided, all of which yield the finding that the greater the level of overall economic freedom, the lower the unemployment rate.  
  Keywords:  State unemployment rate; economic freedom

JEL Codes: J10; J70; C23.

doi: 10.5455/Elet.2016.3.1.2