GDP - methodology Flashcards
Issues?
- Can we use GDP per capita as a welfare measure?
- How do Facebook and the digital economy fit into GDP, and are we measuring it correctly?
- Transfer pricing problem – international chains of value added
- Durables, housing, and investment. Are we double counting?
- Does the current production boundary make any sense? Should we move it?
GDP and welfare?
- Economists tend to have a utilitarian perspective
- Household utility depends on consumption
- Positive marginal utility, diminishing marginal utility
- GDP per capita or equivalised household, is a proxy for consumption
- But GDP includes non-consumption item? Government consumption is for the population, investment is saving for future consumption?
- 2 steps – utility evaluated at average consumption, average consumption approximated by average incomes – both can be bad approximations
Jensens inequality?
- Jensens inequality
- If not linear, approximation will be incorrect
- Strict inequality unless all households consume the same
- Bc of diminishing marginal utility
- High income people have a lower marginal utility than poorer people
- The distribution of income (consumption) matters – the same average consumption will yield less total utility if it is more unequally distributed
- Billionaire black holes – marginal utility of billionaires is zero – is a rare case – Billionaires suck in resources and create little to no utility
Methods to address issues?
A.) Use median household
Median less effected by outliers – median is 50% in cumulative distribution
B.) Adjust mean – inequality adjusted mean income (HDI)
UN uses Atkinson measure – multiply mean income by
* If income equally distributed, Geometric equals arithmetic – as inequality increases, geometric becomes smaller relative to arithmetic
* Or use Gini coefficient – inequality adjusted income (1-Gini).income – Gini equals zero for most equal distribution
Both A and B need some measures household income and distribution
Other problems with GDP as welfare?
- Leaves out other welfare enhancing/reducing aspects – public goods, externalities
- Leaves out leisure – hours not worked have zero value
- Broader measure is HDI, includes health and education
Is per capita GDP a good measure of welfare?
- Wasn’t designed as such and so at best is a rule of thumb guide
- Median households a better measure
- Household happiness and wellbeing not affected much by income as long as have the basics – family, friends, enjoyable work, time for hobbies
The digital economy?
- Big increase in internet use since 2000 and computing since 1990
- Pre pandemic – 20% of expenditure online
- Continuing developments – cloud computing, AI, 5G
- Digital economy seems to replicate existing issues in older technologies but more extreme – main difference being the scale achieved made possible
Example - Amazon?
- Replaces existing distribution network of warehouses, shops and delivery
- Pre amazon – GVA would have included retail and more centralised distribution
- Amazon -get rid of shops, wage bills – more efficient distribution infrastructure due to scale (fewer larger warehouses)
- Total value added relatively similar however, what’s happening is less going onto wages more into operating surplus
- Difference to shops such as Argos is its size, scope and ability to innovate using internet technology
- Principle of amazon quite similar to Argos with a catalogue mail order
Free goods?
- Free goods – value of Facebook posts, tweets, free or cheap music streaming, google maps etc.
- Not included in GVA but have clear welfare benefit
Transfer pricing?
- Wage income is ‘located’ – people work and live at particular places, however, operating surplus is not so clear
- Multinational companies can move operating surplus around by its internal pricing policy
- E.g., Amazon buys CD for £3 and sells for £5, delivery cost is £1, operating surplus is £1. Now Amazon Luxembourg could sell to Amazon UK for £4, operating surplus is £1 in Luxembourg and £0 in UK
- In practice, multinational corporations can move profits to low tax jurisdictions – can destroy GVA in larger countries
- The UK loses GVA – Amazon, Starbucks, Google pay almost no corporation tax – especially important for internet operations; little physical presence
- Unfair – shops pay taxes and tax on profits whilst providing jobs and externalities (footfall)
Consumer durables and double counting?
- Made at time t and provide services in subsequent periods
- Should we count them when they are made and sold or add services over their lifetime
- Most measured just when they are sold
- New house construction is included in GCFM, investment – the GVA of the new housing enters at this period – however, imputed (or actual) rent is also included in subsequent periods as consumption – double counting?
Production boundary?
- Widen to include home production?
- In practice don’t have information for lots of things
- Have produced household satellite accounts
- Home production – e-GDP (extended GDP
- Estimate hours spent into these things
- Could value at opportunity cost with household income, instead look at replacement where what is the market value of that activity? – may overestimate it
- Wide range in leisure between countries
Conceptual problem?
- If leisure is a final output, GVA is incorrect
- Wage bill would become intermediate and only operating surplus is GVA
- In a competitive economy, the real wage equals the MRS between leisure and consumption
- The 60% of national income made up by wages is compensation for leisure lost, not GVA
- Leisure matters for welfare but is a non-market good without a market price (although does have opp cost)
- GVA calculation assumes leisure has no value – extreme assumption
Back to housing. Housing is a necessity
. Is it a final consumption good, or (at least in part) an intermediate?
Should we include all of imputed rent or just “surplus” (rent over and above basic).
People have to live somewhere…
GDP assumes all consumption expenditure on food, housing etc is GVA,
makes no distinction between necessary expenditures and discretionary.
Also, is welfare (happiness, utility) the final “good”. GDP is simply an “intermediate”:goods and services that we consume are simply a means to an end, an intermediate output.
- If trying to measure happiness, should examine a more direct measure than GDP
- Edgeworth’s Hedonometer?
Conclusions?
- GDP is not primarily a measure of welfare. Need to look at median household as a better measure. From a utilitarian perspective, inequality matters a lot (Billionaire Black holes). Is GDP just an intermediate input to happiness?
- The digital economy raises issues, but these are more of extent rather than totally new.Catalogues and mail order made digital and much more extensive due to internet. Also, information has become more valuable (marketing and propaganda can be targeted more effectively)
- Transfer pricing. The prices charged along the production chain allocate GVA. With international supply chains can influence GVA registered in UK. Multinational firms can “move” GVA to low tax locations.
- Double counting (Housing, investment)? Currently, we count the production of housing(and investment) and also the stream of services (investment incomes). Misleading?
- Production boundary: home production and leisure. Important – large magnitude. Important when activities move across boundary over time (Childcare).