Expensive, yes I suppose. Also in support of your statement, here is an article from ECMWEB
Who cares about 25 ohms or less?
However, my concern is not so much power quality, but its more safety, so if NEC doesn't require it, I always make it a point to put in my Specifications. At many instances, concrete foundations provide lesser grounding resistance than the electrical ground (thanks to this don't care policy) which in my view is a safety hazard to people and is some cases failure of devices during lightning surges.
You will have to explain to me how a resistance of 25-ohms or less provides additional safety. Unless you are able to get the resistance of a ground rod to around 2 ohms or less it will not trip a 20 amp breaker with a 120 volt fault to the ground rod in a reasonable amount of time. In the event of lightning, the voltage is so high that a resistance difference of a few (or a few hundred) ohms would be negligible. The code making panels understand that requiring a low resistance on ground rods would serve no purpose so they do not require it.
Just use Ohm's law. If we apply 120 volts to a 25 ohm resistance there will be 4.8 amps flowing, so you will not trip a 20-amp breaker.
Lower the resistance by half to 10 ohms and there will be 12-amps flowing, so still no breaker trip.
Lower the resistance to 5 ohms and you will see 24 amps flowing which MAY eventually trip a 20-amp breaker, but I wouldn't want to hold my breath waiting for it!
Ground rods serve a purpose, but it is NOT tripping overcurrent devices!
If you have 1,000,000 volts in a lightning strike and a 25-ohm resistance you would measure 40,000 amps. If you had a resistance of 100 ohms you would only measure 10,000 amps. I don't see where that would make a big difference in the trip time of a breaker.
Grounding does what it does, it is not magic. Requiring low resistance on a ground rod is a waste of time and money and does not add to the safety of an installation.