The Flaws of GDP
Every few months Australians receive another announcement about GDP. The economy has grown by 0.4 per cent, 0.6 per cent or perhaps contracted by 0.2 per cent. Politicians celebrate when the number has a plus sign in front of it and explain the result away when it does not, while financial journalists dissect the decimal places as though the national accounts were a medical diagnosis of the population. The trouble is that GDP was never designed to tell us how well people are living.
Gross domestic product measures the market value of goods and services produced within an economy over a particular period. That is useful information if the question is whether economic production is expanding or contracting. The mistake comes when a measure of activity is treated as a measure of prosperity, living standards and even national wellbeing. Adam Creighton, writing for the Institute of Public Affairs, has pointed to the absurdity this confusion now produces. Australia's latest national accounts showed GDP growing over the year, which allowed the government to advertise another expanding economy, yet output per person had barely moved and remained below its June 2022 level after contracting in ten of the previous sixteen quarters.
That distinction between total GDP and GDP per capita matters especially in a country with rapid population growth. Suppose an economy of ten million people produces $500 billion of output, then adds another million people and lifts total output to $525 billion. The government can announce five per cent GDP growth while output per person has fallen. The economy became bigger while the average slice became smaller. Headline GDP is therefore easy to misuse in a high-immigration economy. More people require more houses, food, electricity, transport, medical services and government administration, and producing those things increases GDP. The people already here may still face more congestion, dearer housing, stretched infrastructure and little or no improvement in real disposable income. GDP records the extra activity; the citizen records the traffic jam.
Even GDP per capita cannot solve the deeper problem, because GDP measures transactions rather than the quality of human life. If a natural disaster destroys thousands of homes, the rebuilding contributes to GDP: construction firms are busy, materials are bought and wages are paid. The national accounts record production, but nobody would claim that society became wealthier because houses were wrecked and then rebuilt. Crime produces a similar illusion. More burglaries generate spending on police, courts, prisons, alarms, guards, locks and insurance administration, much of which counts as output. A society in which people could leave their doors unlocked might generate less GDP from security and a far better life. Illness works the same way. Treating cancer is desirable and it contributes to GDP, yet a population that never developed the disease would be better off even if some measured medical expenditure disappeared. GDP has difficulty distinguishing activity that makes us better off from activity undertaken only to repair damage, defend ourselves or restore what we already had.
Then there is distribution. Imagine an economy in which GDP per capita rises substantially while almost the entire gain flows to a small share of the population. The average rises on paper; the median household sees little. A billionaire and a struggling worker can walk into the same statistical average and emerge looking prosperous. People do not consume GDP per capita. They consume what their actual incomes buy after tax, housing, interest, energy and other unavoidable costs. A country can therefore post respectable GDP growth while large numbers of citizens feel poorer, and they may not be imagining it.
Housing makes the point unusually clearly. Rising property prices can make existing owners wealthier on paper while making younger Australians worse off. A society in which a modest house costs eight, ten or twelve times an ordinary income is not necessarily enjoying a higher standard of living than one in which the same physical house cost four times that income. The house did not become twice as useful because its price doubled.
GDP also handles unpaid work badly. A parent who stays home caring for children performs economically valuable work that much of conventional GDP ignores. If the parent instead pays a childcare provider and takes paid employment, market transactions appear and GDP rises, even if the children receive the same amount of care. Cook your own dinner and much of the activity falls outside the accounts; pay a restaurant and GDP increases. Clean your own house and the labour is largely invisible; pay someone else and measured production rises. The boundary between "economic production" and ordinary human activity therefore depends partly on whether money changes hands.
The digital age has made that boundary still stranger. Creighton uses the smartphone well. A single inexpensive device has replaced or displaced cameras, video recorders, calculators, street directories, encyclopaedias, alarm clocks, torches, diaries, music players, newspapers and other products that earlier generations bought separately. If consumers were still buying ten devices instead of one handset, measured expenditure might be higher. Technical progress has made people dramatically better off while eliminating transactions that once padded GDP. Free digital services compound the problem. Search engines, maps, email, encyclopaedias, video calls and vast quantities of information reach consumers at little or no monetary price. Their usefulness can be enormous, but a product priced at zero is an awkward object for a system organised around market sales.
Artificial intelligence could widen the gap. Within a few years an ordinary household might have cheap access to an AI that can tutor, translate, help with accounts, write code, offer preliminary medical information, plan travel and perform other services that once cost tens of thousands of dollars. If the subscription is $30 a month, conventional statistics record about $30 of expenditure. The difference between price and value becomes huge. That is one of GDP's deepest weaknesses in a technically advanced economy. Progress often consists in getting more value for less money. GDP is excellent at counting the money and much worse at measuring the value.
There are other omissions. GDP tells us little about leisure. If technology lets workers produce the same output in thirty hours rather than forty, society has gained ten hours of life per worker, yet the statistics may capture that poorly or even treat shorter hours as a disappointment. Nor does GDP directly measure security, family stability, social trust or community life. A wealthy society in which people are frightened to walk outside after dark has lost something important. A society in which people work ever longer hours to pay for housing and postpone children may be deteriorating in ways the accounts barely register. Rising loneliness, family breakdown or social disorder does not become health merely because restaurants and financial services are booking more revenue.
Environmental stocks create another accounting difficulty. Extracting and selling a non-renewable resource adds to GDP, while depletion of the underlying stock is not treated as ordinary GDP treats depreciation of a firm's capital. Activity that creates pollution can raise measured production while imposing costs elsewhere. The OECD has therefore joined a broader movement arguing that policymakers need to look "beyond GDP." It notes that GDP was designed to measure market activity, not general welfare, and recommends broader measures that take in distribution, economic security, sustainability and other dimensions of wellbeing:
https://www.oecd.org/en/topics/well-being-and-beyond-gdp.html
None of this means GDP is entirely useless. It remains the right tool for the question it was built to answer: how much market production is taking place? Trouble starts when politicians answer a different question with the same number. "Is the economy larger?" is not "Are Australians better off?" "Did total production increase?" is not "Did living standards improve?" "Did GDP rise?" is certainly not "Is this a better society in which to live?"
The obsession can distort policy. If a larger population automatically expands aggregate activity, governments acquire an easy route to headline growth. More people mean more consumption, more construction and more public spending. Aggregate GDP rises even when infrastructure per person deteriorates and existing residents see little material improvement. Government expenditure itself belongs in the accounts, because public services are real economic activity, but extra spending is not automatically extra welfare. A billion dollars spent well on useful infrastructure and a billion wasted on a disastrous project can both appear as production. GDP counts the outlay more readily than it judges the wisdom of the outlay. That is reason enough to be cautious whenever governments treat aggregate GDP as proof that their policies are working. The statistic may be accurate. The interpretation may still be nonsense.
A better picture of living standards needs a dashboard rather than a single magic number: real disposable household income per person, median incomes, housing affordability, productivity, household wealth and debt, hours worked, access to health care and education, infrastructure per person, economic security, and perhaps broader indicators of social wellbeing. No single figure can compress a civilisation into one decimal. That is the lesson of GDP's rise from specialised accounting tool to national scoreboard. Economists created a useful way to measure production. Politicians and commentators gradually asked it to measure prosperity, progress and human welfare as well. It cannot carry that weight.
The mismatch will become more obvious as technology advances. If AI gives an ordinary Australian access to services that once required a team of highly paid professionals, that person may become vastly richer in practical capability while the improvement barely registers in conventional GDP. Conversely, if Australians spend ever larger shares of their incomes on housing, security, congestion, administration and the repair of social problems, GDP can rise impressively while everyday life gets harder. The economy exists for human beings. Human beings do not exist to make the statistics look good.
GDP should be returned to its proper place. It is an important measure of production, not a measure of civilisation and not a definitive measure of how well Australians are living. When the statistical economy is booming and the people inside it insist they are going backwards, the people may not be the ones who have misunderstood reality; it is the politicians.
https://ipa.org.au/latest-news/why-gdp-is-no-longer-an-accurate-measure-of-living-standards
