I\’m getting confused about a number which is being bandied about.
For example, the wage share of GDP as at page 6 of this.
Which seems remarkably different from the labour share of income here.
And what sparks my interest is this definition of calculating up (or dividing out if you prefer) GDP via the income approach.
And what particulalry sparks my interest is that, well, I\’m not entirely sure that those wage share of GDP figures are telling us quite what the TUC (and others) say it is telling us.
So0, what I\’d be interested in is someone who actually knows this stuff to aid me in stumbling through it.
Now, what I think is true is that labour compensation is one of those bits that go to make up GDP. And the difference between labour compensation and labour wages (which is I think what the TUC is using) is in part the national insurance taxes paid by employers.
compensation of employees: the total remuneration, in cash or in kind, payable by an employer to an employee in return for work done by the latter during the accounting period; the compensation of employees is broken down into: wages and salaries (in cash and in kind); employers\’ social contributions (employers\’ actual social contributions and employers\’ imputed social contributions);
So, a change in the employers\’ national insurance paid is going to lead to a change in the labour wages share of GDP.
Am I right so far?
If I am, then we\’ve had, since 1975 (just the historical date I could find) a rise in employers\’ NI from 8.5% of wages to 13.8% of wages. I think, but am not certain, that the amount at which you start paying has also fallen in real terms, meaning that more wages are now caught in the net. And we\’ve also had the abolition of the cap on employers\’ NI which seems to have taken place in 1985.
Now, this does not mean that we suddenly move 5% of GDP from labour wage share to some other part of the accounting. Because there is that NI free part of wages. Which will reduce the effect: just as the uncapping will increase it.
But, if it is true that the wage share of GDP does not include employers\’ NI, then increases in employers\’ NI are going to reduce the wage share of GDP.
And there\’s a second part too. Mixed income, essentially the incomes of the self-employed, are not included in wage share of GDP either.
mixed income: this is the remuneration for the work carried out by the owner (or by members of his/her family) of an unincorporated enterprise; this is referred to as \’mixed income\’ since it cannot be distinguished from the entrepreneurial profit of the owner;
So, a rise in self-employment will lead to a reduction in the wage share of income as well.
So, and now to the question. The important one that is.
We\’ve most definitely had a change in the wage share of GDP. Yup, we have. From 58% or so in 1955, to 65% or so in 1975, to 53 % or so today.
How much of this is simply due to being an artefact of how the statistic is constructed? How much of it is due to increases in employers\’ national insurance and how much due to an increase in self-employment?
I don\’t actually know that there has been an increase in self-employment but I rather fancy there has.
So, is there anyone out there who can guide me through this? Perhaps it\’s already been done, neatly wrapped up in a nice paper? If not, shouldn\’t it be?
For wouldn\’t it be interesting if the changes in the wage share of GDP were just statistical artefacts resulting from changes in taxation and employment classification, rather than some dastardly plot to increase the returns to either capital or the rich bastards?
I’d like to know this too. You know, the ONS people at Newport some times answer emails.
Ritchie – if you’re reading this, the question was for a REAL economist.
The latter document shows the labour share of domestic income, not of GDP. 75% of domestic income being from wages – the rest being from pensions, benefits, investment income and so sounds about right – on the low side if anything. Rather a lot of GDP does not end up as anyone’s domestic income (most obviously dividends that get paid to overseas residents), and I suspect that accounts for the gap.
This of course only strengthens my view that GDP is not a very useful measure of anything much. But I’ll stop there on the GDP issue before getting too OT.
If you look closer at the ONS definition of “domestic income”, it’s actually an impenetrable morass of additions and subtractions of four-letter acronyms, a lot of which seem to do with commercial property rental, not “household income” (as I assumed) at all. So who knows what the hell the statistics given in their forecast mean? Nothing to do with GDP – that’s the short answer.
So you still need a real economist.
You are complaining below of those who” deny the very basis of statistics itself” but here despite a plethora of statistics, you cannot make nor tail of them.
I think what you want is indirect taxes (VAT etc), not national insurance.
The 55% figures is for ‘total compensation of employees’, as this in 2010 was 799bn, and GDP was 1,458bn, so that makes 55% of GDP.
The remainder was corporations surplus (307), ‘other’ (168) & taxes on products and production (179).
Total compensation of employees is ‘wages and salaries’ (661bn) and ’employers social contributions’ (138bn)
VAT and other sales taxes now account for large part of income.
See A12 p.27 of PDF
Over time, e.g. in 1976, wages & NI were 62%, taxes 10%, corp profits 17%. In 2010 55%, 14%, 21%.
So the 8% fall was split equally in VAT, corp profits (note ‘other’ was stable at 11%).
But ’76 was a special year, by 1981, the ratios were 58%, 12%, 19%, and by 1984, 55%, 12%, 22%
Tim adds: That is a good answer thank you. So 50% or so by your reckoning is as statistical artefact, not a switch from labour to capital income.
It might be more for the TUC figures do state wage income, not labour compensation, so those changes in NI would also have an effect. And would be interesting #to see what change is any ther has been in self-emp.
But thanks for this, my original surmise seems to have something to it at least.
One other piece mentioned above is “property income”.
Definition The property income attributed to insurance policyholders is the investment income receivable by insurance enterprises on insurance technical reserves; it is shown in the accounts as being paid by the insurance enterprises to the insurance policyholders because the technical reserves are assets of the policyholders.
In other words, investment income within insurance companies for with-profits and pension products for the large part which are primarily for policy holders and only in minor part for shareholders.
Yeah the [email protected] alias at the ONS has real stats bods behind it, they do respond to mail, albeit sometimes slowly.
I always wonder about demographics here; if the baby boomers have been building up capital, wouldn’t you expect that returns to capital would grow as a proportion of GDP?
just loves the way that DBC Reed comes and magisterially declares complete ignorance of the topic under discussion…sheer class. Most of the time, he is content to be wrong.