The statement about the effect of FB use isn't really true for Philippines (as it is for Korea).
Philippines does not get extra approvals in EB because of low approvals in FB.
EB3-P does get higher approvals than the prorated amount in EB would suggest, but that's because Philippines has few approvals in EB1-EB2 and EB4-EB5.
In FY2019, Philippines used 0.86% of the EB1 allocation, 3.34% of the EB2 allocation, 12.65% of the EB3 allocation, 2.17% of the EB4 allocation and 0.16% of the EB5 allocation.
Against the prorated EB limit, Philippines used 4.98% of the 7% EB limit.
Only once in the last 10 years has Philippines used more than the 7% limit in EB (and that was within the range of error - not CO's finest).
Philippines EB Total Approvals
FY --------- No. ---- %
FY2019 --- 7,074 --- 4.98%
FY2018 --- 8,985 --- 6.40%
FY2017 --- 8,824 --- 6.30%
FY2016 --- 8,917 --- 6.35%
FY2015 -- 10,363 --- 7.16%
FY2014 --- 8,172 --- 5.44%
FY2013 -- 10,537 --- 6.65%
FY2012 --- 9,504 --- 6.56%
FY2011 --- 7,628 --- 5.45%
FY2010 --- 6,786 --- 4.50%
Average -- 8,679 --- 5.98%
Nor is it really an overallocation. The interpretation of how the 7% limit is calculated (i.e. equals FB + EB) has been consistent for a very long time.
The practice of combining EB and FB for purposes of determining the 7% limit dates back to at least as far as FY2000 (the oldest data I could find) but probably has been in force since at least when the Immigration system was last revamped in 1990.
https://travel.state.gov/content/dam...%20app%20A.pdf
https://travel.state.gov/content/dam...%20app%20B.pdf
The same also appears to be true of using more in some subcategories (e.g. EB3), when other subcategories have insufficient demand, as long as the total for the category (EB) is not exceeded.
In FB, FB3 & FB4 India have regularly benefited from this interpretation.
For example, India used 18.4% of the FB3 allocation in FY2019 and 11.5% of the FB4 allocation in FY2019, while using 6.2% of the total FB allocation because use in FB1 (1.2%), FB2A (1.8%) and FB2B (1.0%) was relatively low.