There's a number of ways they can divide up the revenue. Here's some I can think of:

- Per boarding: All boardings are recorded, so dividing up the proceeds per boarding would, IMO, be relatively simple. Attempts by individual operators to 'skew' the boardings by forcing changes would be counter-productive, as people don't like changing buses or modes - it's only worth doing to 'aggregate' two half-bus loads together to make a full bus load (and I'd guess that's the theory at places such as Washington). M1 feeds in to Metro because it can't compete journey-speed wise with Metro; but most GNE buses at Gateshead go through to Newcastle, where the interchange time penalty is greater than the time of the extra bus ride.
- Per boarding weighted by average fare: GNE and Arriva run many longer, inter-urban services, whereas SNE's routes are mainly town or city centre services. GNE and Arriva might argue that they forego a higher fare every time they accept an NTL ticket when compared with SNE. Sounds fairer, but gets more difficult to administer.
- Proportion of mileage operated: forget trying to count boardings, simply divide the takings by the number of miles each operators' buses operate. This assumes that there costs/revenues are broadly the same across the county, which I doubt is true; but I throw this one in for completeness.

That's just three potential models - I've no idea which one they use, but I'd guess the second might work.

BTW, did anyone notice that NTL holds 27 of its own shares? I'd guess that this enabled NTL to be paid enough of the revenue (not quite three per cent?) to cover its own costs... clever!

Re the GNE 'share' being less than the PTE's: surely this just reflects how the NTL market works? Multi-modal tickets, where it's possible to take advantage of the faster speed of rail and the higher frequency of buses, surely suggest that combined bus/Metro journeys is the main use that these tickets are put to?