The question of weighting in multilateral comparison divides practice. On the one hand the very plausible method of Geary-Khamis uses international prices with proportional weights for countries of different size. Therefore, a big country's prices may dominate the international pattern. On the other hand Gerardi's method is neutral, as it most practically uses geometrical means. Thus, it makes for every product a purely individual international price, independent of other products or size differences of countries. Both methods lead to additivity of international values, simply as a consequence of their using international prices. Of course, this is very important in practice. It is embarrassing, however, to get different results.