One way you could do it is while you parse the input file keep a count of how many times each date is encountered then filter the holiday info based on that. Consider:

#!/usr/bin/env perl
use warnings;
use strict;

my %holidays;
my %common;

while (<DATA>) {
    chomp;
    next if /^\s*$/ || /^\#/;
    my $aref = [split /,/, $_];
    push @{$holidays{$aref->[2]}}, $aref->[5];
    ++$common{$aref->[5]};
}

my $countries = keys %holidays;

for my $country (keys %holidays) {
    @{$holidays{$country}} = grep {$common{$_} != $countries} @{$holid
+ays{$country}};
}
[download]