Table 3.

Coefficient stability - test for omitted variable bias.

			Total		Supply
					Total	CG	RR	DP	SIFI
			[1]		[2]	[3]	[4]	[5]	[6]
Full model	β1		0.0207		0.0343	0.2465	0.1086	0.1797	-0.1854
(Table 2)	R2		0.352		0.36	0.393	0.351	0.377	0.369
Restricted model	β1		0.0185		0.0277	0.2275	0.0657	0.1759	-0.1456
(no country-level controls)	R2		0.261		0.266	0.314	0.255	0.216	0.281
Oster Delta (Rmax=1)			-1.3		-0.8	-1.7	-0.4	-12.2	-0.6
Oster Delta (Rmax=1.3*Rfull)			-8.1		-4.5	-8.7	-2.3	-67.3	-3.7

This table reports the results of the coefficient stability test of Oster (2019). ${\mathbf{\beta }}_1$ is the coefficient of macroprudential variable, the one that is significant at the conventional level in Table 2, along with the associated R-squared, obtained by estimating Eq. (1) in a restricted version (omitting all country-level control variables) and in a full model (as presented in Table 2). The Oster Delta statistic represents the degree of selection on unobserved variables relative to that on observed variables, where we set ${\mathit{R}}^{\mathit{max}}=1.3*{\mathit{R}}^{\mathit{full}}$ or ${\mathit{R}}^{\mathit{max}}=1$. Note that ${\mathit{R}}^{\mathit{max}}$ is described as the R-squared for a speculative regression that contains unobserved confounders.