Cowgill & Dorobantu (2012) use domestic and cross-border data from Google AdWords that tracks clicks on ads and various types of conversion of this click into a purchase or other valuable online actions. They also find a reduction in distance-related trade costs by about two thirds. However, in line with Blum & Goldfarb (2006), their results also show that cultural distance, measured by (the absence of) a shared language and religion, matters. The trade costs related to cultural distance may increase six to tenfold when moving from offline to online trade. To assess the net impact of these increases and decreases in online trade costs (compared to offline costs), the authors estimate the "border effect" (McCallum, 1995) or home bias: to what extent do consumers have a natural preference for domestic over cross-border purchases? They find that the border
effect is very much alive in online trade and can in some estimates turn out to be even stronger than offline.