Table 2.

Value-added ROO and GSP exports to the EU.^a.

	Linear		Logarithmic		Square root		Geometric
	$\alpha =br$		$\alpha =b\ln \left(1+r\right)$		$\alpha =b\sqrt{r}$		$\alpha =b\left(r+r^2-r^3\right)$
	Pref. exp.	Non-pref.	Pref. exp.	Non-pref.	Pref. exp.	Non-pref.	Pref. exp.	Non-pref.
		exports		exports		exports		exports
	[1]	[2]	[3]	[4]	[5]	[6]	[7]	[8]
Preference margin	−2.720	−0.943	−2.720	−0.943	−2.720	−0.943	−2.720	−0.943
	(2.082)	(3.307)	(2.082)	(3.307)	(2.082)	(3.307)	(2.082)	(3.307)
ROO	−2.679∗	3.813∗	−3.745∗	5.330∗	−3.362∗	4.785∗	−2.046∗	2.910∗
	(0.816)	(0.928)	(1.140)	(1.297)	(1.024)	(1.165)	(0.623)	(0.709)
Observations	22632	116193	22632	116193	22632	116193	22632	116193
Pseudo R2	0.984	0.979	0.984	0.979	0.984	0.979	0.984	0.979

^aNote: All regressions use a pseudo Poisson maximum likelihood estimator and contain exporter $\times$ year, product $\times$ exporter, and product $\times$ year fixed effects. Statistical significance is indicated with superscript $*$ for $p$-value $<0.001$, $⊛$ for $p$-value $<0.01$, and $†$ for $p$-value $<0.05$. In all regressions, standard errors are clustered at the 4-digit of the HS times year level to account for the fact that this is the level at which ROO are set in the EU. The first two columns provide results assuming a linear relationship between $\alpha$ and $r$, the third and fourth columns assume a logarithmic relationship, the fifth and sixth columns a square root relationship, and the last two columns a geometric relationship.