Basic Idea: The original idea behind the virtual water (VW) concept is that waterabundant countries will become producers of water-intensive goods and consequently net exporters of water, and this will alleviate the initial unequal distribution of hydric resources. We criticize this optimistic view by introducing theoretical insights supported by new empirical evidence that is consistent with the Heckscher-Ohlin model of international trade. A central empirical finding of our paper is that virtual water trade increases with the amount of available arable land. However, comparing countries with a similar level of available water resources, those with higher (lower) levels of physical and human capital tend to be net importers (exporters) of water. This is clearly in contrast with the promise of the VW narrative and relies on the simple intuition, that more developed countries that are relatively abundant in capital may shift away from agricultural production. In particular, developed countries that have ample access to water may end up not utilizing this precious resource that is scarce elsewhere. Thus, within the classic framework of virtual water, the unequal economic development at the global level may in itself lead to further pressure on scarce water resource of poor countries. Data and Methodology: This study takes advantage of the cross-country dataset described in Debaere (2014). We use data from Feenstra (2000) to supplement the dataset with a measure of multilateral import. This additional variable is necessary to compute the net exports from each country in each industry. The final sample consists of 68 countries (developed and developing) and 194 industries. In line with the existing literature on the Heckscher-Ohlin (H-O) theory, we consider net exports (NX) as the dependent variable (see, e.g., Deardorff, 1982; Forstner, 1985; Romalis, 2004; Nunn, 2007; Levchenko, 2007 and Debaere, 2014). Independent variables can be divided into two groups. The first group consists of four different variables measuring the country's endowment of water, land, physical and human capital; whereas, the second group refers to four variables capturing the sectors' factor intensities. We apply a Principal Component Analysis (PCA) to reduce the space of factors. This approach is justified by both theoretical and empirical considerations. From a theoretical point of view, the estimates of the H-O theorem usually suffer from the limitations of the traditional two factors-two goods-two countries setting. As proved by Bee-Yan Aw (1983), when there are more than two factors of production, the regression procedures used in most empirical works are inappropriate as a test of the factor proportions theory. From an empirical point of view, we found that a fourfactor model may lead to serious multicollinearity problems. Therefore, the PCA allows us to reconcile the analysis with the theory, avoiding inefficiencies in the estimates. Main Results: First, we confirm the H-O Theorem finding that a country will export those goods that use its abundant factors more intensively. Second, we show that the water-land/capital threshold above which a country becomes a net exporter of water decreases with water-land intensity. Finally, we show that developed countries usually import water-intensive goods not because they have a lower endowment of water but because they are relatively abundant of capital. Policy Implications: This work has important policy implications. In contrast with the virtual water theory, our results imply that trade could not make the cross-country distribution of water less uneven. Therefore, effective and coordinated policies must be designed to ensure an equal distribution of water around the World. [ABSTRACT FROM AUTHOR]