Table 3.

Baseline: regional favoritism in the allocation of EIB loans

	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)
VARIABLE	EIB Loan Dummy					Ln Loans	Loans / GDP	Loans / Tot. Loans
Work region dummy	0.1457***	0.1768***	0.1712***	0.2064***	0.1776***	-0.0256	0.9834***	0.6050**
	(0.0487)	(0.0462)	(0.0521)	(0.0762)	(0.0513)	(0.2533)	(0.3124)	(0.2786)
Ln population			0.4608*	-0.3558	0.6064**	1.4334	-0.2290	-0.5164
			(0.2381)	(0.6956)	(0.2351)	(1.2371)	(1.0254)	(0.8516)
Ln GDP			0.1299	-0.4264***		2.1337***		2.7845***
			(0.1337)	(0.1635)		(0.7535)		(0.5775)
Ln GDP p.c. t − 6 to t − 1					0.1204
					(0.1228)
Hours worked per employee			0.4395*	-0.2512	0.4527*	0.1459	-0.0700	-2.1312*
			(0.2427)	(0.3430)	(0.2437)	(1.8445)	(1.0652)	(1.1477)
Compensation per employee			0.0009	0.0061	0.0016	0.0205	-0.0523***	-0.0305*
			(0.0045)	(0.0048)	(0.0045)	(0.0171)	(0.0192)	(0.0168)
Ln gross fixed capital formation			-0.0794	0.1253	-0.0755	0.1262	1.4807***	0.7588***
			(0.0511)	(0.0822)	(0.0532)	(0.3782)	(0.2297)	(0.2746)
Observations	16,530	7,540	6,642	6,642	6,581	2,782	6,442	6,442
R-squared	0.3844	0.2557	0.2540	0.3268	0.2526	0.2992	0.1910	0.4429
Number of regions	290	290	266	266	266	258	258	258
Mean of dependent variable	.25	.39	.42	.42	.42	4.18	.0017	.0039

*** p < 0.01, ** p < 0.05, * p < 0.1. The table presents results for the estimation of the model in equation 1. Standard errors are in parentheses and clustered on the level of NUTS 2 regions. The sample start year is 1990 for all regressions except column 1, where it is 1959. The method in all columns is OLS except for columns 7 and 8, which use PPML. Region fixed effects are included in all regressions. Country-year fixed effects are in included in all columns except 7 and 8, where only year fixed effects are included. Column 4 includes a region-specific time trend. Given the dependent variables, columns 1 to 5 analyze the extensive margin, column 6 the intensive margin, and columns 7 and 8 combine the intensive and extensive margin