Monday, May 18, 2009
2008 Earnings Survey
Office for National Statistics: 2008 Annual Survey of Hours and Earnings
Following on from the earlier discussion about the house price to earnings ratio, here's the ONS annual survey from last year. I've looked and I still can't see where they get that £35,122 from. If you look at "Employees on adult rates who have been in the same job for at
least 12 months" (which is what the Halifax say's it uses), the figure for males was £27,500 in April 08 and I doubt it's gone up much in the last 12 months.
Maybe someone can enlighten me?
Posted by becky @ 05:42 PM (1490 views) Add Comment
14 Comments
- If you do not have an admin password leave the password field blank.
- If you would like to request a password allowing you to add comments and blog news articles without needing each one approved manually, send an e-mail to the webmaster.
- Your email address is required so we can verify that the comment is genuine. It will not be posted anywhere on the site, will be stored confidentially by us and never given out to any third party.
- Please note that any viewpoints published here as comments are user's views and not the views of HousePriceCrash.co.uk.
- Please adhere to the Guidelines
1. timmy t said...
Becky - When the ONS calculated inflation they understated it by about 50%. In the interest of balance they obviously needed to overstate something by a similar amount, and overstating salaries makes people feel good so they opted for that.
2. paul said...
The ONS has always overstated wage data whenever possible because it is used for all kinds of international surveys on the quality of living.
The modal income is much more accurate because it is 'what most people earn'. This strips out the outlying top 1% that earn millions and ignores the resulting median.
The modal income in the UK is around £16,000. Welcome to the real Britain - enjoy your stay.
3. denzil said...
A rule of thumb is to always allow for an error of -/+50% to all ONS figures.
4. little professor said...
Becky - they use the mean, not the median income. The mean is much higher (skewed due to the inequality in wealth distribution in this society and the number to super-high earners in the City etc.) They also only use the figures for males in full-time employment for the past 12 months, excluding everyone who works part-time, recent recruits, and indeed all women.
From the 2008 ONS figures, mean income for males in full-time employment was indeed £32,838. The median was £25,997
For all employees, the mean was £26,020 and the median £20,801. I would argue the last of these figures is the most meaningful.
Figures taken from table 1.7a here
5. little professor said...
Sorry, should read:
From the 2008 ONS figures, mean income for males in full-time employment was indeed £35,122. The median was £27,500
Interestingly the mean income for males in full time employment is £35k, while the equivalent figure for females in full time employment is a whole ten grand less at £25k. The median figures are much more evenly matched, confirming the effect of male superbankers on skewing the mean upwards.
6. peter_2008 said...
I thought the statistical definition of median is the halfway split point of earning, which means that half of the UK earners earn more than £35,122, and the other half earn less. If ONS is using this definition, then the numbers should be less scewed (compared with Mean or Average) by the higher earners.
For example. if three chaps earn £15k, £35k and £40k, the median is £35k (the earning that split the group in the middle) and the mean is £30k (i.e. (15+35+40)/3=30).
As shown above, the thing concerns me here, is the difference between the Median and the Mean. To have a significant difference like the one we see in the UK, you need some very very poor people to drag the Mean further down away from the Median. In this case the lowest earner in the group will only earn £7.5k to have a mean of £27.5K, as mentioned in ONS.
I guess Paul's figure of £16,000 modal income reflects this as well.
7. drewster said...
Becky -
The £35,000 figure is the mean for full-time workers. I believe it includes all overtime and bonuses.
If you look at the median for full-time workers then the figure is closer to £25,000.
The only reference I can find to back this up is the Halifax historical house price data (Excel spreadsheet) - the last tab states the National Average Earnings Full-Time for Men as £36,324.
When comparing house prices to incomes, the higher figure (mean) is actually more meaningful. If Roman Abramovich and all his rich pals move to the UK, the mean income increases whereas the median barely budges. Abramovich can afford to pour £10m into the property market which cascades into higher prices all down the chain.
However comparing average incomes doesn't take into account the level of employment. We could easily have 15%+ unemployment (Spain has already hit 17.4%) yet no change in average income, because it only includes people who are working. A far better metric would be average (mean) income multiplied by the number of workers - i.e. total worker income.
8. peter_2008 said...
Sorry! Didn't read the data carefully! LP is right. ONS does use Mean, NOT Median!! Well, my re-written example should be
If three chaps earn £15k, £30k and £60k, the mean is £35k ((15+30+60)/3=35) and the Median is £30k (i.e. the earning that split the group in the middle).
And yes, LP is also right that obviously, high earners scew Mean.
9. drewster said...
Little Prof,
Much of the gender gap comes from women failing to progress up the career path because of child-rearing breaks in their careers. Once you adjust for that, the difference is negligible.
10. paul said...
@peter_2008
The modal figure is by far the most accurate. The problem with the median is that it can still be skewed by the outlying top 2% at each end. You'll have to imagine this graph, because I've never seen it published anywhere where Y-Axis is gross income, X-Axis is number of working age people in UK. What would the graph look like?
Sounds like a job for Wolfram Alpha!
Oh. Wolfram|Alpha isn't sure what to do with your input.
(that seems to be their answer for a lot of things ... )
Well, the graph would have a small bump (no-one earns zero, unless like me they are currently a net drain on the national income), and then a long almost flat gradient right up until the last 5% where it would shoot up exponentially.
Of course, the bottom 1% is of less concern because there is a minimum wage, but the top 1% massively skews the mean and median.
The modal is the overwhelmingly flat part of the curve in the middle, and that is the best indication of "what most people earn" - which is after all what we need to know. Unfortunately most people think the mean will tell you that (or maybe the median) but they won't.
Modal is definitely the important one.
11. paul said...
Actually the top percentages won't influence the median hugely, but they will still have a positive effect.
12. will said...
Could be the household income.
13. matt_the_hat said...
Mode just tells us what the majority civil servants at grade x earn
14. paul said...
@matt_the-hat
If the majority of people in the UK work in the civil service on grade x, then bingo you've found a winner.
It's still the most commonly earned salary - which is the important number.