Horsepower measured at the wheel will be lower than what's measured at the crank for one simple reason. Drivetrain losses. Losses from friction, inertia, etc.

Typical losses for manual trans FWD cars is from about 10-15%, and RWD generally 5% more than that. It largely depends on how far the drive wheels are from the crankshaft, where transmission types, clutch, flywheel, driveline and axle weight's and design will have effects. The closer the drive wheel to the crank, theoretically the less loss due to friction. Looking at it another way, the more bearings there are supporting gears, shafts, etc, the more losses there are.

Also, these losses are a percentage of power, not an absolute. Just because a car loses 5hp between the crank and the drive wheel, doesn't mean if you build this motor to 300rpm you will still lose only 5hp. Much of these losses are friction. Friction increases with the square of speed. So if it takes 5hp to drive a shaft 4000rpm, then it takes 25hp to run the same shaft 8000rpm.

Also the ratings in the 1960's appear to be higher than today. This is true because before 1973-ish, the rating were SAE Gross, which means the engine on the dyno was set for max HP numbers, and not running through any real exhaust system, no alternators or other accessories were attached, and tuned for big numbers which is not necessarily the same it would be tuned leaving the factory. SAE net, which is what automakers publish today, net, which is still HP at the crank, but with all accessories attached as it would be in a car, as well as "normal" tune and exhaust. SAE NET has proven to be about 20-25% lower than the GROSS numbers, so an old 327HP Chevy 327cid would rate at about 260hp today (coincidentally the same as a 1995 Impala SS w/ a 350).

Some enthusiast magazines will state the car has xxxHP at the drive wheel, which means they drove the car onto what's called a chassis dyno, (a DynoJet or something similar) and the car's wheels actually rotate on rollers and HP is calculated from this.