Question

Table 2 below shows cross-sectional regression results from the study of Beck, Degryse, and Kneer (2014).1 The dependent variable is the economic growth (GDP per capita growth) of different countries averaged over the period 1980- 2007. The explanatory variables are defined as follows: intermediation is a proxy for the size of financial sector and equal to the logarithm of the credit over GDP for every country in the sample; initial GDP is the logarithm of the GDP per capita in 1980; education is the average years of schooling of the population aged 25 and above for the same period, trade openness is measured by the ratio of exports plus imports to GDP averaged over the same period, govemment consumption is the government expenditure as ratio to GDP: and inflation is the rate of inflation. Table 2. The relationship between growth and intermediation. Growth 0.810 (0.286) 0.275 (0.092) 0.577 (0.215) 0.001 (0.002) Growth 0.736 (0.282) 0.246 (0.094) 0.502 (0.221) Growth 0.858 (0.270) 0.285 (0.091) 0.571 (0.209) Panel A: Cross-section Intermediation Education Initial GDP Inflation Openness 0.003 (0.003) 0.030 (0.033) Govconsumption 5.391 Constant 4.610 (1.545) (1.669) (1.519) Observations 0.205 Adj.R2 0.191 0.199 Standard errors appear in parentheses Answer to the following questions (use the lt>2 rule of thumb in your answers to indicate significance at the 95% level): (a) What is the estimated equation that correspond to the results of the first specification of Table 2? (3 marks) (b) Is the coefficient of intermediation statistically significant in the three specifications of Table 2?

Answer 1

a)

y^ = 5.391 + 0.810 Panel A + 0.275 Education -0.577 Initial GDP -0.001 Inflation

b)

coeff	error	TS
0.81	0.286	2.832167832
0.736	0.282	2.609929078
0.858	0.27	3.177777778

since TS > 2

it is significant in all three models