The thing is, there's almost no way to stop spoofing. It would require a complete redesign of the telephone system.
There's currently no easy way to confirm that a number
arriving into the UK PSTN network is from an originator who has the right to use it. A new certificate verification system is in place in the USA and Ofcom are looking, very slowly, into a similar system here, probably not to be implemented until the PSTN network is discontinued after 2025 - which is an opportunity for a redesign of this aspect of the telephone system.
However, it's relatively easy for an originating network to block invalid caller IDs from being
transmitted, because they have a register of their customers and the numbers that they own: for example, BT certainly used to limit the caller ID on ISDN lines to number ranges owned by the customer, which would otherwise be rejected by the exchange, and VOIP provider Sipgate limit the CID to numbers they provide.
So it's possible, but highly unlikely without legislation and enforcement in
every country.
...given that most of the presentation numbers used by the fraudsters are real numbers, surely at some point they must get a number that is actually in the recipient's phone as it is a contact's number? ...
Yup: I got a scam call purporting to come from my neighbour's number a few days ago, which happens to be an anagram of my number. The recent trend to base the spoofed caller ID on a number close to the number dialled makes them both more likely to be active phone numbers and more likely (thouh still unlikely) to be someone you know in your local area, particularly on smaller exchanges where the first two or three digits are the same for all subscribers' numbers.