Leading or benchmark companies seem to spend around double in payroll terms than many other companies. This probably has not made them 'leading'. Rather, they got to be 'leading' through a combination of factors and in their journey have developed their philosophy to the point where they (and their shareholders) really value training, and are successful enough to be able to afford to invest double the amount others do.

Do not confuse correlation with cause! This is the big mistake many practitioners make when they use control groups in their training evaluations - you know, take two groups with roughly equal performance, train one group, compare performance levels after training, hope that the trained group performs better, and if it does, compared to the control group, then assume that this is down to the training.

Having been in the kind of positions in senior management where these kinds of decisions are taken, I didn't often see an argument for more investment on training win on the grounds that "Company X spend this amount per capita, so should we." I did tend to see winning arguments along the lines of: "Investing £100K in training will deliver extra £400K to the bottom line, change the following behaviours in these areas, and contribute to staff morale, and here's the proof!"

It is quite possible to do excellent training and spend very, very little!

*Read the question that sparked this response and other replies here.