In the United States in 2012, 1,570,976 of the 3,952,841 live births, or 39.7%, were the first live birth to that mother. Among mothers listed as non-Hispanic and white (which I mention only because I think the vast majority of Jews in the States are so listed and the category is available to me), 895,171 of the 2,134,044 live births, or 41.9%, were the first live birth to that mother. However, the proportion among mothers listed as white (irrespective of being listed as Hispanic) was 1,190,207 of 2,999,820, or 39.7%, and some of the considerations that lead to a higher percentage among non-Hispanic white mothers (i.e. a lower percentage — 310,326 of 907,677, or 23.2% — among Hispanic white mothers) may also lead to a lower percentage among religious Jewish mothers.
In the United States in 2012, 2,650,744 of the 3,952,841 live births, or 67.1%, and 1,441,894 of the 2,134,044 non-Hispanic white mothers' births, or 67.6%, were vaginal.
Source: the CDC.
So, assuming that half of all babies born are male, we can use 20% as a reasonable estimate for Pfirstborn is a male live birth and 67% as a reasonable estimate for Pvaginal birth.
According to Doron Behar, et al. in a 2003 paper in the American Journal of Human Genetics, citing a 1999 paper by Neil Braman et al. in Population Specific Polymorphisms, "it is estimated that Cohanim and Levites each comprise ∼4% of the Jewish people." Therefore, we can use 92% as a reasonable estimate for PIsrael.
Therefore, we can estimate that the proportion of live births that would require a pidyon haben would be:
(PIsrael)2 * Pfirstborn is a male live birth * Pvaginal birth ≈ 11%