There was an eye-catching headline on the Argus website about how ‘Salaries in affluent Brighton are £3000 below average‘.  It also says that salaries in Crawley have dropped £6,000 in the year.  And its all rubbish.

Without even knowing what the right figures are you can tell that the ones used in the story are wrong.   The clue is that it all comes from a self-promoting survey done by an Internet jobsearch site, and is based on a survey of jobs advertised rather than all jobs.

It also quotes the national average salary as being £31,969, which seems a little on the high side.  The latest ONS annual survey of hours and earnings (ASHE) for 2007 found a national median wage of £23,764  – 2.9% higher than 2006 and a full eight grand lower than what is quoted in the Argus.

That is a median weekly wage of £457 for full-time employees.  And of course, that is not the whole picture as there are many part-time employees now.

Looking at the tables for gross earnings by place of residence we see that the average earnings are:

• Brighton & Hove
  £21,795 (median – up 1.9%)
  £25,235 (mean – up 3.2%)
• Crawley
  £21,381 (median – down 4.8%)
  £25,511 (mean – no change)
• Horsham
  £25,431 (median – up 5.2%)
  £33,428 (mean – up 6.4%

But those figures are for people living in those places.  They may well be commuting elsewhere to earn those salaries.  The equivalent table for gross earnings by place of work show:

• Brighton & Hove
  £19,391 (median – up 4.6%)
  £22,311 (mean – up 4.4%)
• Crawley
  £22,378 (median – down 1.3%)
  £28421 (mean – up 1%)
• Horsham
  £19,734 (median – down 1.4%)
  £24,107 (mean – up 2.5%)

So what does all that mean?
For a start, whichever way you cut it the worthless survey showing Brighton’s average as £28,749 is actually way above the real average for Brighton & Hove.   All it proves is that the survey is based on a sample higher than the real world.

The reason for the survey’s average being higher than the real one could be down to the fact that lower-paid jobs are not advertised as much.  The reason for the survey’s average being lower than last year could be down to anything, but its not down to Brighton’s salaries going down.

There are something like 78,000 jobs in Brighton, trying to extrapolate from a sample of 3,000 which were advertised is pointless.  Many of the lower-paid jobs are not advertised as such.   Supermarkets and fast-food places are recruiting all the time and will take on thousands of new employees each year either without advertising or with a single blanket advert.  Often the advertising is an in-house poster.  So all those new jobs will have escaped the survey.

At the other end of the scale, many higher-paid jobs in the private sector are filled by headhunting or by having agencies looking through their  files of suitable candidates and may never get advertised normally. And, of course, in larger companies jobs are advertised internally and filled without ever reaching the outside world.

And many jobs will not have been advertised because they never fell vacant at all.

I just can’t understand why Tony Mernagh, executive director of Brighton and Hove Economic Partnership, was willing to be quoted so extensively without making any comment about the statistical flaws in the survey method.  When he says that salaries in the city have been below the national average for some time he is right, but not to the tune of £3,000.

The tables of salaries on the ONS website are fascinating though.  Just the few numbers quoted above can lead to all sorts of observations like:

• People who live in Horsham and Brighton and Hove earn more than people who work there.
  Presumably this is the effect of residents who commute to London and elsewhere for work.  The differential is a lot larger for Horsham.
• People who live in Crawley earn less than people who work there.
  An interesting conclusion.  Crawley contains a lot of commuters too, but also has a lot of inward commuting.  One conclusion could be that the higher-paid jobs in Crawley are filled by inward commuters
• Earnings for Horsham residents are high and going up while the salaries for jobs in Horsham are low and falling.
  Again, this will be related to commuting, but is an improvement in the overall affluence masking a decline in the local employment market?

Its hard to know where to stop though. The figures quoted above were from tables for all jobs.  The tables for full-time jobs shows much higher salaries and different relationships between just the three areas.  I’m sure you could dip in and select statistics which prove whatever you want to.  The question is, what were allthetopbananas.com trying to prove, or were they just after getting their name publicised a bit?

If that is all they were after it worked a treat because I had never heard of them before.