Fun with numbers

I've just stumbled across this month old article on the BBC:

Where does rent hit young people the hardest in Britain?

People in their 20s who want to rent a place for themselves face having to pay out an "unaffordable" amount in two-thirds of Britain, BBC research shows.

They face financial strain as average rents for a one-bedroom home eat up more than 30% of their typical salary in 65% of British postcode areas...

Least affordable areas outside London

Epping Forest - postcode areas IG9 (rent £1,230; 71% of income), RM4 (rent £1,126; 65% of income), and IG7 (rent £1,087; 62% of income)

I live in one of those postcodes, they're in south west Essex inside the M25 and while the rent estimates look about right to me (good enough for a low level Land Value Tax assessment, for example), the percentage figures look way too high.

If you look at outer London postcodes just across the Essex/Greater London boundary, the % spent on rent drops significantly, even though the rents are just as high.

As you might have guessed by now, this is because they used average rents down to postcode district levels, but assumed that wages are the same across whole regions. Essex is part of the East Anglia region, which has an average wage of £1,755 per month, as against an average wage of £2,275 for people living in Greater London.

1. It would be much more meaningful if they used average wages at postcode district level.

2. The percentage is actually meaningless in itself, it is a derived figure. The relevant figure is wages net of housing costs. If they bothered to do that, they'd find that this is fairly constant everywhere in Great Britain (Ricardo's Law of Rent).