moo-vs-pop5.pl

#!/usr/bin/env perl

use strict;
use warnings;
use Benchmark qw(cmpthese);
use Class::XSAccessor qw();

{
	package Foo;
	use Moo;
	has id => (is => "ro", required => 1);
}

{
	package Bar;
	sub new {
		my $class = shift;
		my $self = bless {@_} => $class;
		exists $self->{id} or die;
		return $self;
	}
	sub id {
		my $self = shift;
		return $self->{id};
	}
}

cmpthese(-1, {
	UseMoo    => q{ my $o = Foo->new(id => 42); $o->id for 1..100 },
	UsePOP5   => q{ my $o = Bar->new(id => 42); $o->id for 1..100 },
});

__END__
           Rate UsePOP5  UseMoo
UsePOP5  5023/s      --    -54%
UseMoo  10882/s    117%      --

Except that this benchmark compares a pure Perl accessor with an XS one, as Moo uses Class::XSAccessor to generate its getter, but that's not the case for your "plain old Perl 5" getter. Replace Bar::id by :

use Class::XSAccessor getters => { id => 'id' }

and see what happens.