GDP, gross value added and the productivity puzzle


In the UK, GDP is stagnant. At the same time the employment and labour-hour counts are rising. Stagnant GDP with rising labour-hours means that measured productivity has fallen. Declining productivity has engaged economists who have named the phenomenon of falling productivity as the “productivity puzzle”. Two main explanations have been put forward to account for this puzzle.
Incorrect employment count
The first explanation holds that the employment statistics published by ONS mask a significant amount of disguised unemployment.  There has been a large rise in self employment over the very recent period and it is suspected that many of the newly self-employed have been pushed into this category by DWP.  Frances Coppola deals with some of the reasons for the rise in the newly self employed in her blog. Coupled to the rise in self-employment are claimants on out-of-work benefits who  have been placed on government work-for-your-benefit programmes. Claimants on these programmes are counted as employed. The productivity of those pushed into self-employment and those on work-for-your-benefit programmes is likely to be lower than for paid employees recruited via conventional channels.
Incorrect measurement of GDP
Some commentators assert that GDP cannot be stagnant if there are more people working, There is therefore an assumption that ONS’s measurement of GDP is incorrect and the productivity puzzle would be solved if GDP was measured correctly.
This strand of thought asserts that intangible fixed assets are being created as a faster rate than are tangible fixed assets. Moreover, these intangible assets are being self-constructed (made in-house), as distinct from being purchased from outside suppliers. Self-constructed fixed assets, whether tangible or intangible, would not (nor should not) feature in the calculation of gross value added and hence, so the argument goes, a significant amount of economic activity is not being captured in the GDP figure. Hence the stagnation.
An illustration
Below are value added statements for a hypothetical 2-firm economy which have been consolidated in the third column of numbers. The row shown in red represents a self-constructed fixed asset built by Firm 2. This item does not properly belong in a value added statement and has been included to illustrate  the impact its inclusion has on measured GDP.
The GDP Code column contains a letter code to denote where each item in the national value added statement would appear in the expenditure equation of national income so beloved of economics textbooks. Below the value added statements appear two tables, the first showing GDP calculated with self-constructed assets included, the second with GDP calculated under conventional principles. The figures in these tables have been taken directly from the consolidated value added statement. You will see that the method that includes self-constructed assets increases national income by the amount of the work applied in the period to the self-constructed asset

Principles of measuring GDP,  illustrated by a 2 firm economy

Value Added Statements

Sales  Firm1  Firm 2  Combined  GDP Code
   To  UK consumers £30 £170 £200 C
   To UK firms: goods & services £40 £40
   To UK firms: capital equipment £100 £100 I
   To government £20 £60 £80 G
   To foreigners £5 £35 £40 X
Change in inventories £5 £5 I
Change in self constructed fixed assets   £50 £50 I
Gross value of output £200 £315 £515
   From UK firms: goods & services £40 £40
   From foreigners £30 £50 £80 M
Gross Value of inputs £30 £90 £120
GROSS VALUE ADDED £170 £225 £395 Y
Distribution of value added
  To Labour : wages and salaries £20 £100 £120 L
  To shareholders : dividends £13 £22 £35 F
  To government : tax £15 £50 £65 T
  To the firm: retained profits £122 £53 £175 P
GROSS VALUE ADDED £170 £225 £395
GDP measured with self constructed assets included
£395 £200 £155 £80 £40 £80
GDP measured in the conventional manner
£345 £200 £105 £80 £40 £80
Although the example provided is simple, it does allow a clearer examination of the assertion that GDP is being understated due to the exclusion of  self-constructed intangible fixed assets from computations of value added and hence national income. Such exclusions may well understate GDP but then they will always have done so.
More importantly though, a value added statement is merely a recasting  of a traditional income statement. If self-constructed fixed assets were to be included in the calculation of value added and hence national income then a fundamental distinction between capital and revenue expenditure would be violated which would presumably be extended to income measures for firms.
So my message to the proponents who hold GDP measurement is wrong  because it excludes intangibles and self-constructed fixed assets is, “Nice try but the productivity puzzle is more likely to be resolved by looking at the labour market”.


    • Hi Hugh

      Thanks for your comment, which is interesting.

      I have given further thought to this matter since my original post. I am now wondering whether Value Added Statements, which are an alternative way of reporting an individual firm’s performance, report a firm’s economic performance incompletely. Currently, the practice is not to report investment expenditure in VASs because VASs are merely a re-statement of a traditional income statement.

      One of the difficulties in including expenditure on self-constructed fixed assets in Gross Value Added is the absence of a transaction with a third party to attest to such an asset’s creation. Accountants depend on objectivity for their measurement of economic activity, so that, in general, if no invoice (or other documentation) exists then no verifiable transaction took place.

      I will read the paper you have attached once I have a little more time. It looks interesting.

  1. Hi Hugh

    I have followed the link you provided but it does not appear to be working.

    Some preliminary thoughts:

    Bought-in intangible assets will count towards GVA if the correct accounting treatment of such is followed. This is because fixed assets should not be included as a bought-in service in a GVA calculation. Instead bought in capex will go straight to the purchasing firm’s balance sheet. The supplying firm will record he same value in its GVA calculation as a sale. It does not matter whether the capex is for tangible or intangible fixed assets, the accounting treatment is the same.

    Following generally accepted accounting principles, self-constructed fixed assets should be capitalised on the firm’s balance sheet in most circumstances (pure research costs are an example where this won’t happen).

    The way I see it, there may be an issue with measuring GVA when self constructed assets are capitalised. The issue arises from the operational definition of GVA (sales – cost of cost of bought-in materials and services). Clearly, there is no sale with a self-constructed asset and hence, according to the definition, there can be no value added. However, the counter point is that the capitalised labour costs will have been subsumed in the capital cost of the self-constructed asset. Effectively, some of the firm’s labour costs will have disappeared! However, the same disappearing labour cost arises in the conventional format of income reporting.

    Another way of looking at whether self-constructed fixed assets should be recorded as contributing to GVA is to ask whether a firm’s net worth is increasing as a consequence whilst they are under construction. The arithmetic answer is that it isn’t (assets – liabilities = net worth). Of course, net worth may increase in subsequent periods as the fruits of the capex are realised and this is when the impact on GVA will show.

  2. The own account asset say software is extra output. On yhe income side the firm now has an asset which attracts a rental payment. So more value added and more income, all balances. That s how software is treated. Over the recessions intangible invest has risen and tang has fallen. So bias duento the omission of intang from GDP has got worse. See haskelecon blog for some description. Yours Jonathan Haskel

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s