APPENDIX A
Table 1: Interpretation of Regression Coefficients:
Coefficient: Interpretation:
β0 A constant - the value of the function when it crosses the Y line; indicates level of performance when there is no product diversification, no geographic diversification, 0 leverage, 0 sales growth rate, no difference in industry.
β1 The change in levels of performance when there is change in product diversification
β2 The change in levels of performance when there is change in international
diversification
β3 The change in levels of performance when there is change in firm size
β4 The change in levels of performance when there is change in sale growth rate
β5 The change in levels of performance when there is change in leverage.
Table 2. Corporate and Public Ethics Indices
Results from the empirical research conducted by Daniel Kaufmann (2004)
Ethics indices: Country averages CICC CLCC CEI PSEI JLEI CGI (Note: Theoretical range of indices 0-100%)
Italy 47.3 34.4 40.9 33.9 40.7 32.6
UK 93.2 67.4 80.3 79.7 92.1 87.9
LEGEND
CICC (Corporate Illegal Corruption Component): Percentage of firms in the country that give satisfactory ratings to the questions on corporate ethics, illegal political funding, state capture cost, average of frequency of bribery in procurement and active capture, corruption in banking (average of formal money laundering and bribery for loans).
CLCC (Corporate Legal Corruption Component): Percentage of firms in the country that give satisfactory ratings to the questions on influencing legal political funding and undue political influence.
CEI (Corporate Ethics Index): Percentage of firms in the country that give satisfactory rating to the questions on index calculated as the average of the percentage of firms’ corporate illegal corruption component and the corporate legal corruption component.
PSEI (Public Sector Ethics Index): Percentage of firms in the country that give satisfactory ratings to the questions on honesty of politicians, government favoritism in procurement, diversion of public funds, trust in postal office and the average of bribe frequencies for permits, utilities and taxes.
Table 3. Control of Corruption Measured by World Bank ITALY UK Year Percentile Rank Governance Score Standard Error Percentile Rank Governance Score Standard Error (0-100) (-2.5 to +2.5) (0-100) (-2.5 to +2.5) 2006 64.1 0.31 0.15 93.7 1.86 0.15 2005 67.5 0.41 0.15 94.7 1.94 0.15 2004 71.4 0.56 0.15 94.2 1.99 0.15 2003 75.7 0.75 0.16 95.1 2.08 0.16 2002 76.7 0.77 0.16 95.1 2.1 0.16 2000 83.5 0.98 0.17 96.1 2.13 0.17 1998 75.7 0.72 0.18 96.1 2.12 0.18 Control of Corruption 1996 72.8 0.48 0.2 96.1 2.21 0.2 LEGEND
Percentile Rank: Percentage of countries that are worse compared with Italy or UK, in term of control of corruption.
Governance Score: The score (-2.5 to +2.5) given by World Bank, in term of control of corruption
Table 4: Descriptive Statistics and Correlations Table 4.1: UK UK Mean S.D n 1 2 3 4 5 6 7 8 ROA 8.93 7.61 73 INT_DIV 8.17 6.21 73 0.15 PRO_DIV 5.41 3.29 73 -0.15* 0.11 FIRM_SIZE 15.01 0.98 73 0.07 0.29* 0.05 LEVERAGE 3.02 2.41 73 -0.17 -0.11 0.17 -0.11 SALES_GROWTH 0.08 0.17 73 0.24 -0.13 0.03 -0.25 -0.11 BANKING 0.06 0.23 73 0.05* 0.06 -0.30 -0.43** 0.00 0.05 INDUSTRIAL 0.56 0.51 73 0.08 0.35* 0.21 0.51 -0.25 -0.02 -0.38 TRANSPORTATION 0.21 0.41 73 -0.08 -0.23 0.04 -0.47 -0.01** 0.08 -0.13 -0.53 Table 4.2: Italy ITALY Mean S.D n 1 2 3 4 5 6 7 8 ROA 8.58 7.03 71 INT_DIV 21.50 16.34 71 -0.18 PRO_DIV 4.85 2.47 71 -0.05 -0.22* LEVERAGE 3.49 7.60 71 -0.29 0.33 0.21 SALES_GROWTH 0.05 0.18 71 0.03** 0.22 -0.11 0.06 INDUSTRIAL 0.64 0.49 71 -0.02 0.16 -0.37** 0.17 0.16 FIRM_SIZE 15.27 1.59 71 0.15 0.24 -0.62 -0.01 0.13 0.41 BANKING 0.09 0.29 71 -0.14 0.09* -0.46 -0.16 -0.11 -0.42* 0.13 TRANSPORTATION 0.15 0.36 71 -0.16 -0.19 0.60 0.11 -0.03 -0.27 -0.21* -0.31
Table 5. Tests for Hetroskedasticity, Autocorrelation and Multicollinearity
Model 1 Model 2
White Heteroskedasticity Test
F-statistic (Prob) 0.554(0.933) 0.447(0.929) Obs*R-squared 17.597(0.859) 18.659(0.865)
Autocorrelation
Durbin Watson Statistic 2.141 1.972
Variance Inflation Factor
Product diversification 1.594 1.286 International diversification 1.255 2.120 Firm size 2.183 2.576 Sales growth 2.203 3.530 Leverage 1.277 1.456 Banking 1.565 2.611 Industrial 1.476 1.608 Transportation 2.013 2.335
APPENDIX B
Assumptions of MR
According to the Gauss-Markov theorem, the validity of any MR model depends on 5 underlying assumptions. For the multiple regression model, if assumptions hold, then the least squares estimators are the Best Linear Unbiased Estimators (BLUE) of the parameters in a multiple regression model. This paper defines notation as xk,i representing the value of
variable k for observation i. Mathematically, the assumptions are presented as follows:
MR1. The value of y for each value of x is yi=b0 + b1 x1,i +b2 x2,i…+ε i
MR2. The expected value of the random error e is E(ε)=0
MR3. The variance of the random error e is var (ε) and is constant. MR4. The covariance between any pair of random errors is cov (εi ,εj)=0
MR5. The values of xk,i are not random and are not perfectly correlated linear functions of
