OH GROSS! The real problem with GDP Last week, when I wrote a column about the nature of cold hard cash from a rabid free marketeer point of view, commenter Chris Miller asked about the intricacies of GDP: is it actually a bad measure of how well-off we all are or is it used just because it's simple to calculate? Basically, the answer is “yes”. With a few minor …
Good article on the imperfections of GDP Mr W.
It's a shame all the headlines trumpeting "The economy has risen 0.2% last quarter!" don't come with the proviso " - but take it with a massive pinch of salt!".
Personally I think GDP is too flawed to be used, not because it doesn't give some indication of economic growth, but because it provides an irresistable incentive for governments to juice the figures. Your example of paying beaureacrats more increases GDP is a good one.
But also if they use the Y = C + I + G + (X − M) calculation of GDP (Y) then governments are massively incentivised to borrow huge amounts to increase government spending (G) to get that juicy headline. Sadly the GDP calculation doesn't seem to take vast debts into account.
"Back to that idea of judging a company by the P&L and not the balance sheet really"
I understand that calculating a country's 'true' balance sheet that included natural resources, human potential etc etc would be too intractable to arrive at a 'proper' answer, but surely a purely financial 'national balance sheet' would be pretty easy to work out?
Unless having an incomplete BS picture would be a worse fudge than not having any at all, giving "BS" a different meaning?
There's also the added issue that, if anything, you want per capita numbers as you can add to GDP by having ever more low paid individuals yet that doesn't equate to the nation truly becoming richer if viewed from a standard of living perspective. That's the other side to the Government juicing of the figures - can't add wealth then add headcount + debt.
It's been a while since I looked at those equations, but IIRC, when someone borrows money, investment (I) decreases. Since government (G) only increases based on what we spend on it, it's value can't increase more than investment was decreased. In contrast, I can increase when lent to someone outside the country; and when money is borrowed for consumption, the value of the consumption is expected to outweigh the cost of borrowing (or else why bother borrowing the money.)
Thanks for answering my question so fully, Tim. As GDP is (essentially) the numerator in the national productivity figures, about which the UK has so much angst at the moment, I wonder if we should really be getting so wound up about it.
National 'productivity' in France (just to take an obvious example) is measured to be significantly higher than that of the UK. This tends to confuse those (like me) who have actually worked in a French office and struggled to observe the immense efficiency that the raw figures might imply. Is it the case that if we (say) reduced the retirement age to 50 and doubled state pensions*, we'd boost GDP and close the 'gap' instantly? (The country would go broke, of course, but perhaps that would be a price worth paying :)
* Or alternatively, take half the unemployed and give them highly paid civil service jobs.
to the French story is twofold.
1) The marginal productivity of the last few hours worked will be lower than average productivity. Maybe. Thus shorter working hours will show higher average per hour productivity.
2) The least productive French workers aren't in work at all. They're unemployed instead. Thus average productivity of those in work is higher.
The bit I always try to add to that usual story is that the measurement is in fact not "national" but is of paid working hours in the marketplace. And what we really want to know is the productivity of all hours, paid and unpaid in household production. Which brings us to that division and specialisation of labour stuff in the piece. If household production hours are markedly less productive than market ones, then if there's fewer market hours and more household ones then "real" productivity will be lower than we think.
I don't actually know the French figures but one that sticks in the mind is that a German housewife actually has less leisure than an American one. Despite the American one doing more market working hours. The difference being more household hours.
I think we're largely misled by not counting those household hours properly.
BTW, the source for all of this is the Luxembourg Income Study. If you want to get your hands dirty with the data have a surf around their site. Krugman and Milanovic both work there now. It's absolutely the dog's bollocks of the subject.
And yet it is claimed that UK productivity is around 30% less that our European competitors - which is a heck of a lot - and around 17% less than would be expected for the UK had the pre-crash trends continued. Given a long period of large rises in rents, transport and (until recently) fuel and food costs you'd expect (intuitively) considerably more people queuing at soup kitches and food banks if that amount of money were actually missing from the economy. It does seem there is something the figures are not capturing.
Then there is the myth of US productivity
Having spent a good part of my working life either working for an American comany or clodely with companies in the US this is in reality as accurate as 'the cloud is secure'. Bollocks.
Sure they 'work' more hours than us. most Mericans get 2 weeks vacation. Many of my friends over there get fearful every two weeks. i.e. every time they get paid. This is when the pink slips get handed out. No warning, no consultation. You are then escorted from the building on the spot. Any personal belonging might get to you in a month or two. I guess this fear factor might make them work harder but this form is instant dismissal can really hit a project hard.
One such project had a key person get his 'pink slip'. He was chosen because some MBA spotted that he was lax at filling in his weekly timecard. The whole project slipped around 6 months because he had no time to handover to the rest of us. That hit Productivity and thus the GDP.
Also many professional level people in the US have other jobs.
Moonlighting is common practice. These are not people working for minimum wage but at least one has a basic salary of more than $100K in their main job. They do it to provide money for their childrens 'College Fund'. Getting their kids into the best school seems to occupy the more than anything else (and give them heart attacks/ulcers/depression)
At the moment I'm working closely with teams in the US, France, Germany, India and S. Africa. We brits get things done quicker and of better quality than any of the others. I have two people who could be sitting on their backsides waiting for the others to be ready to complete the sprint. Thankfully we have other work that will get done in the meantime. But this hits our productivity and therefore the perceived GDP.
Ah, but look at the "consumption measure" of GDP/GNI. Does any visitor to France not feel that they're better off than us. Most of the country looks much better, the infrastructure works, they think they've housing problems when the number built in a year falls below 400,000, and the food eaten by ordinary people is so much better than in Blighty.
Just to thank Tim for his article on "What is money?" that ran last week. Unfortunately real-life intervened to suck up my time (apart from the most cursory snark on one or two El Reg threads) so I didn't get the chance to say "Thank you".
Of course I have some observations, but they'll have to keep until a more appropriate article.
Given it's a figure that we almost all agree is at best a guideline, and is open to some pretty significant distortions, does it have value as a figure? It's all very well to say it's the best we've got, but when you put a figure on something people tend to go absolute on it. To an economist GDP may be a guideline with (un)known variance, but that same figure once reported will get reported by an economics journalist to a readership of vaguely informed people (us), who will then repeat such figure to completely uninformed people and then suddenly it's a rock solid measure of our economy.
Let's not even start on how politicians use it.
Is it the best measure we've got, when "no measure" is taken into account as an option?
"Given it's a figure that we almost all agree is at best a guideline, and is open to some pretty significant distortions, does it have value as a figure?"
I don't now what the business version of a luvvie alert is, but I was talking to someone on the board of Vodafone once at St James' Palace, and he said that they essentially ignore GDP, and are most interested in consumer and business confidence indices, primarily because they will be most suggestive of whether people open or close their wallets. So I don't know as GDP measures are the most important thing on a microeconomic level, based purely on what one guy at a party said.
It does have a value. At least to me as a polemicist. Those places which have been roughly capitalist and even vaguely market based for more than a couple of decades have high and rising GDPs. Those places that haven't been even approximately either do not.
Given that what GDP is the value that is available for consumption that looks like an interesting finding to me.
"Good enough" is how I'd put it. Does the trick and without taking so long that it has become irrelevant by the time the value is available.
Like the market economy itself (or democracy for that matter), a compromise. Neither of which are good enough if you want to get anally retentive about it, but changing either system to make it more controllable/predictable ends in tears sooner or later.
Yes, as Tim points out in the article if your national GDP goes up by 5% or down by 2%, that's telling you something - and probably something important (+/-0.1%, not so much*). Whatever the societal inaccuracies in calculating GDP, they probably don't change much from one year to the next.
But comparing GDP figures between countries needs a whole bigger bucketload of salt, because those societal differences can be really substantial.
* UK GDP figures are published fairly swiftly (which is why economists like them - numbers telling you what happened to your economy 12 months ago aren't much use to the Chancellor or the BofE), but they're usually adjusted (up or down) a month or two later, and often by a significant fraction of a percent. So those headlines that GDP has fallen by 01.% (woe is us) are probably commenting on noise.
One wonders whether there might be a way to total up a current national net worth.
Like start with counting the net worth of all a nation's warm bodies, including the dough they have "in the bank", though comparatively speaking, that dough might be chump change.
Add the value of "infrastructure".
Maybe subtract spending on or investment in things that, in a perfect world, might warrant zero spending. Security things: cops, bars over windows, etc, come to mind. Such subtraction might help account for "social capital".
…" what people will pay for something is only at least what they think it is worth. They might be willing to pay much more: meaning that more value is created than is being recorded."
People like to think they are getting a bargain: so what people will pay for something is often less than what they /think/ it is worth. And probably more than the vendor /thinks/ it is worth.
Whenever I'm unemployed I have plenty of time to write and publish bits of code and genealogical, historical and geographical research. Which, of course, because nobody pays me to do so adds nothing to GDP. But, provides loads of value to the people who use it, including myself. But, attempting to refuse to do any of it unless somebody paid me to do it would be like refusing to breath until somebody paid me to do it.
I'm actually *more* productive when I'm unemployed because I have the time to actually get some real work done, but of course, all available indicies will declare me as having zero productivity.
We have a huge one here in the UK. Because the efforts of the Volunteers are unpaid they don't directly count towards the UK GDP.
The preserved railways in this country contribute around £2B directly to the UK exonomy. yet a huge amount of the workforce is volunteer. I'm one of them myself. So what should it be if all this work was counted?
What would out national pridctivity be if this was counted?
That would be interesting.
It seems to me that a key characteristic of Osbrownomics is an unhealthy focus on GDP as a "good news" story to tell the electorate.
You touched on the problems with this in your final paragraph about higher pay for bureaucrats enriching us all. But surely where this corrupts all the more is in asset prices. Each time a government succeeds in pushing house prices higher, GDP rises and they claim economic success. Yet we're not richer overall: we've just transferred wealth from the productive to the rich. Rising dot-com stocks or tulip bulb prices would likewise feed through to GDP Feelgood, but house prices are the most effective of all because it's harder to opt out of housing so everyone is involved, like it or not.
I suspect a key reason for the UK productivity gap is the amount of our GDP that actually represents a zero-sum game where no value is actually produced.
But asset prices aren't a contribution to GDP. It is any one of the three of all income, all production or all consumption. Only changes in asset prices that feed into those are part of GDP.
So, selling your house and then spending the money as income upon consumption items is indeed a change in GDP. Rather offset by someone else having to defer exactly the same amount of consumption as the value of the house they've just bought.
> Assumes price = value.
Value is, by its very nature, completely subjective, so everything about it is of necessity an assumption. Price, however, you can measure - and (as Tim has pointed out) has to be something that the buyer and seller are agreed on (else there would be no sale). So price also has some validity as an objective measure, and thus is the best (or even only) proxy for value.
Only changes in asset prices that feed into those are part of GDP.
Yes, changes in asset prices. So whenever a transaction occurs, the capital gain feeds GDP. Which would be just fine and non-corrupting if capital gains reflected the actual value of an asset - so for a house that might reflect a new roof or installation of modern plumbing and wiring, but not just value-free rises.
On a related note, how do we account for stockmarket gains/losses on companies listed in London but whose actual business is predominantly not in the UK?
"So, selling your house and then spending the money as income upon consumption items is indeed a change in GDP. Rather offset by someone else having to defer exactly the same amount of consumption as the value of the house they've just bought."
This is where my little brain ceases to compute. I see GDP as a measure computed annually.
So assume someone decides to downsize from a ₤500k UKP house to a £250k house and then spend all the released money on one year of wild living. That is £250k of additional consumption in one year fed into the economy. Meanwhile someone else has taken on additional long term debt and committed to an extra £1,000 a month (bigger mortgage) expenditure. This seems to take £12,000 out of the annual spend and put £250,000 in which doesn't seem to balance in any way. This may take money out of the long term (many years) consumption but does not seem to equate to your statement above that the purchaser defers exactly the same consumption that the vendor can now indulge in.
I suppose if you asumed that the purchaser would have otherwise borrowed an equivalent amount of money using the monthly payment and spent that money you could then try and match the two then this might sort of work but this isn't how life usually works.
Unrealistic, I know, to posit that the vendor spends the equity released over a short term but the potential is there. Is the assumption perhaps that the vendor will spend an extra amount each month roughly equivalent to the extra mortgage payments of the purchaser?
That still leaves me the problem of how savings are counted in GDP. See other post.
"But asset prices aren't a contribution to GDP. It is any one of the three of all income, all production or all consumption. Only changes in asset prices that feed into those are part of GDP."
Do savings count as consumption?
If income goes up more than prices then there is a potential surplus which can go into savings and then not immediately into direct consumption. Therefore the measures of income and consumption do not seem to be the same. Is this not the problem with "helicopter money" where giving people money to spend and thus stimulate the economy doesn't work because they use it to pay down debt or they save it for the future?
Two economists walking along the street see a pile of horse manure.
Would you eat that for $1M asks the first.
The other one considers the payback and eats it and is given a cheque for $1M
They see a 2nd pile and the nd economists says to the 1st, would you eat that for $1M - and he does
A little later one comments, "we are stupid, we are no better off - but we have both eaten horse shit"
Yes says the other, but we have grown the economy by $2M
It's a nice joke but sadly it doesn't work. Because GDP is the value added not the transactions. And the $2 million is just transactions: there's been no value added.
Well, except, perhaps, for the joy at seeing economists eat shit but we don't exactly monetise that either.
Can't find the ref but. Hurricane Katrina added a huge amount to the USA GDP because of all the rebuilding work - leading to a suggestion that the US govt should bomb cities to raise GDP.
But Hurricane Sandy hit GDP because it affected rich cities and interrupted work - so they should only bomb southern cities
"Hurricane Katrina added a huge amount to the USA GDP"
Aka the broken window fallacy.
The fallacy works depending on where you draw the boundaries of the economy.
For example, here in Christchurch New Zealand, we're still having massive rebuilding from the 2010/2011 earthquakes. That is largely being funded by insurance - mostly underwritten by overseas insurance companies.
If you draw the boundary at the local level then there has been massive inflow of money, employment and expenditure and the economy has had a boost.
If you draw the boundary to include the underwriters then it has all just been an enormous loss since resources have been consumed with no value being generated.
Both Katrina and Sandy actually pointed out idiotic land use policies in the coastal regions of the US. Large parts of NOLA is below sea level and flooded when the levees failed. Knowledgeable people had been concerned about this for years before Katrina hit. Much of the infrastructure of NYC is below grade and again is below sea level. NYC itself is a few feet above sea level. The storm surge from Sandy caused most of the problems. The surge was in many areas higher than sea level.
...but I still worry about things.
There are scenarios which (to me anyway) appear to be ready to confuse the calculation of GDP and its interpretation. Let us imagine...
I manufacture and sell a "product", and have 3 factories doing the manufacturing. Factory 1 is in the UK and employs people who quite by chance have been in the UK since William the Conqueror came over. Factory 2 is also in the UK but this time (again quite by chance) employs people who have recently entered the UK and who are remitting some of their income back to families abroad. (I must be paying too much...) Factory 3 is not in the UK, so all my manufacturing costs are outside the UK. By some quirk my profit per item is the same irrespective of where it was manufactured.
So how does my contribution to GDP stack up? Factory 1 pays a labour force that spends its wages in the UK; factory 2 pays a labour force that exports some of its wages elsewhere (where it has no benefit to the UK) and factory 3 makes no contribution to the UK, at least during the manufacturing process. (I have not included marketing and distribution costs as they are common to all three sources of product.)
None of my product is sold abroad. Someone has to unscramble the above as the contribution to the country's GDP is markedly different depending on where any individual item is made, or at least I assume it does.
Years ago "balance of payments" was always a topic for worry; factory 1 doesn't cause money to go abroad other than for sourcing raw materials; factory 2 does not cause money to leave the country directly but the labour force exports it themselves. Factory 3 requries me to send money abroad to pay all the manufacturing costs, so my contribution to the country's balance of payments total is made up from different components, one of which is more or less invisible.
That brings us to "invisibility"; again years ago "invisible exports" were being trumpeted as a "good thing", but while my factories may not be exporting anything some of my employees are, and I suspect that their exporting money (very possibly a small part of the UK's total transactions and thus down in the noise) really is invisible.
At the end of all this I find myself asking "can GDP actually be calculated with any degree of accuracy", and "how much can it be used to see how the economy is doing". And, of course, "do personal exports of money actually do any damage to the UK economy, or is the total exported too small to really matter". Does anyone actually know how much is exported?
Exporting money, as your 3) workers are doing, is actually an import. Ah, the delights of GDP accounting.
1) and 2) are, from the point of view of your factory, the same thing. They're production of value in the UK. That value added is thus part of UK GDP. That the workers export some of their wages doesn't turn up in our production approach of measurement. It's still the same amount of value being added to GDP.
3) is an import: thus the value of those imports is deducted from GDP......because the value is not added the UK. Your profit margin, assuming that you're a UK company, is also not GDP. Because it's not value added in the UK. However, it is part of GNP, as it's value that accrues to someone within the country.
Wouldn't it be better to value housework at the cost of obtaining a similar service from elsewhere? A lady of the night costs x pounds per "event", so the cost of someone enjoying himself with the missus should then also be x. Doing laundry yourself costs x hours, having someone else do it costs x, so cost per hour can be determined.
If I were to value my free time the same as what I charge my boss it would suddenly become cheaper to outsource a lot of the household to someone else so I can enjoy my free time some other way.
To be honest most of this macro-economics stuff sounds to me like retrospective navelgazing with a few claims of "predicting the future". Unfortunately economists seem to have failed to accurately predict economic events over the past 10 years or so.
One cannot discuss GDP without at least mentioning this:
Wiki: Gross National Happiness is a term coined by His Majesty the Fourth King of Bhutan, Jigme Singye Wangchuck in the 1970s. The concept implies that sustainable development should take a holistic approach towards notions of progress and give equal importance to non-economic aspects of wellbeing.
"...UK GDP is of the order of £1.7 trillion a year, the turnover of the foreign exchange markets in London is around and about £1.7 trillion per banking day."
If I exchanged £1.7 trillion per banking day, the exchange fees, margins and/or commissions would have completely consumed £1.7 trillion in such fees within a year. Several times over. I'm very lucky if I can escape with only 5% in fees.
Obviously they're paying negligible fees, else the "value added" from the exchange would single-handedly generate the entire UK GDP.
Although GNP and similar figures don't include housework, and so they're not good at telling us what the net wealth of the country is... paid employment tends to change suddenly because of things like increases in interest rates, a rise in oil prices, a decline in import earnings, while the amount of work people do for themselves and each other at home isn't subject to external influences nearly as much.
So fluctuations in GNP or GDP can still tell us that something is going on that affects people's prospects for paid employment despite those limitations.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2025
