Haute Lumière · The reading

The numbers say the economy is fine, and it does not feel fine

What a country counts becomes what it protects, and the count we inherited was designed in wartime to answer a different question.

The reading is slower than the room suggests. She has been on the same page for some minutes, and the page is arithmetic.

WHAT COUNTS

A measure looks like a mirror and behaves like a lever. Once a number is published on a schedule, budgets are written against it, careers are built on moving it, and the parts of life it cannot see quietly stop being argued for in rooms where money is decided. The number stops describing the economy and starts recruiting it. This is the first claim of the book, and everything else in it follows from this one.

It opens on a small scene rather than a theory. A team is building an app to optimise user productivity, chasing time saved and tasks completed per hour, and someone in the room notices that the metric has eaten the purpose. The tool is being built to extract more from people who are already overextended, and it is working. Nothing on the dashboard can register the cost, because the dashboard was designed to register throughput.

The book's recurring image for this is a forest. Try to understand one by weighing its trees. You get a real number, a defensible number, a number you can track quarterly and compare across regions. You also miss the soil, the fungal networks threading between roots, the birdsong, the decay that feeds next season's growth, and the diversity of species that determines whether the whole thing survives a dry decade.

A measurement that cannot see harm will reliably produce it, and will report growth while doing so.

The argument is not that counting is bad. It is that a narrow count, repeated with enough authority, becomes a description of what a society is for. Measure inputs and outputs and you get an economy optimised for inputs and outputs. The in-between — the collaboration, the care, the repair, the relationships that make any of it possible — does not appear in the ledger, so it does not appear in the budget, so it is the first thing cut when the ledger tightens.

So the book's project is stated plainly at the outset: build a measurement layer that can see a living system. Not a warmer story about the same numbers. A different instrument, with different columns, that registers stocks as well as flows, distribution as well as totals, and the condition of the things the economy runs on as well as the volume of what it runs through.

THE GHOST LEDGER

Gross Domestic Product is not a mistake. It is a precise answer to a question that was urgent in the 1930s and 1940s and is no longer the only question worth asking. Kuznets set out the conception in 1934; the accounts were built out in the aftermath of war to track industrial production and rebuild economies that had been flattened. For that purpose the instrument is excellent. The trouble begins when a wartime production counter is asked to stand in for prosperity.

The book works through all three ways of calculating it, because the omissions live in the arithmetic rather than in the commentary. By expenditure, GDP equals consumption plus investment plus government spending plus net exports. The worked case is a country called Econland: consumption of $1 trillion, investment of $200 billion, government spending of $350 billion, exports of $150 billion and imports of $100 billion, giving a GDP of $1.6 trillion. By income, it is wages plus profits plus interest plus rent. By production, it is value added at each stage — a farmer adds $1 of wheat, a miller adds $1 of flour, a baker adds $2 of bread, and the $4 loaf is counted once.

Each route is internally consistent and each shares the same boundary. A transaction has to happen, in money, in a market, for the thing to be visible at all. Everything outside that boundary is what the book calls the ghosts.

The sharpest illustration is a community of a hundred people. Fifty work in factories producing goods worth $1 million a year, and that sum lands in the accounts. Thirty are self-employed — carpentry, tailoring, childcare — generating an estimated $500,000 in value that is largely unrecorded and therefore excluded. Twenty raise children, care for elderly relatives and maintain community gardens, and their contribution is not merely undercounted but structurally invisible. GDP reads $1 million. The community is running on considerably more than that, and the invisible portion is the part without which the visible portion could not happen.

The ghosts are not rounding errors. They are the floor the measured economy is standing on.

There is a second omission inside the first, and it concerns quality rather than existence. A grower raising heirloom tomatoes on organic methods sells a crop that registers at its market price, and the hours spent learning the practice, building the soil and tending the plants register nowhere. A large operation using synthetic inputs produces cheaper fruit in greater volume and contributes more to the total. Both transactions are counted the same way, in the same units, with no column for what the produce is worth to the person eating it or to the ground it came out of. The measure is indifferent by construction, and indifference at scale is a preference.

Four categories of absence recur through the chapter. Unpaid work, which is most of the care a society performs. Natural capital, so that a forest felled shows as income and never as loss. Distribution, so that a country can post a strong figure while most of its people are losing ground. And wellbeing itself — health, education, social cohesion — which GDP was never designed to touch and has never claimed to. Nordhaus and Tobin asked whether growth was obsolete in 1972. The Stiglitz, Sen and Fitoussi commission asked the same question with more data in 2009. The critique is old, well-documented, and has not yet changed the instrument.

LOOPS AND STOCKS

Before a better instrument can be built, the thing being measured has to be described correctly. The book's description is borrowed from living systems, and it carries a specific vocabulary that does real analytical work rather than decorative work.

Flows are the movements: money circulating between producers, consumers and investors; goods travelling from farm to factory to shelf; knowledge moving through education and practice. Stocks are the accumulations those flows build and deplete: buildings and machinery, skills and health, forests and aquifers, trust and institutional memory. A flow without a stock is a number with nowhere to land. A stock without flows is a photograph of something that has stopped moving.

Feedback is how the system corrects itself, or fails to. Prices act as signals about scarcity and demand, dampening excess the way a thermostat dampens temperature. Other loops run the opposite way and amplify: a technology triggers investment, which accelerates the technology, which attracts more investment. Both kinds are real and an economy contains many of each, which is why its behaviour cannot be read off any single curve.

Coupling is the interdependence between parts — suppliers and manufacturers, wages and consumption, credit and everything. Emergence is what appears when many small interactions produce a pattern nobody designed, the way a flock produces its shape without a choreographer. Markets themselves are an emergent phenomenon in this sense, which is exactly why their outcomes resist precise prediction.

An economy is not a machine with a throttle. It is a system with stocks that remember and loops that decide.

The book insists this is arithmetic rather than metaphor, and the demonstration is the logistic growth equation, dX/dt = rX(1 − X/K), where r is the intrinsic growth rate and K the carrying capacity. Take a population of 100 butterflies with an intrinsic rate of 0.2 and a meadow that supports 500. Year one adds sixteen, bringing the population to 116. Year two adds about 17.8, bringing it to roughly 133.8. Growth accelerates while the system is far from its limit and decelerates as it approaches, without anyone deciding that it should. The same curve describes a market, a resource stock or an adoption rate, and the useful part is the K: an unstated carrying capacity is the assumption most growth projections are quietly making.

The last term the book borrows is antifragility: the property of systems that gain from disorder rather than merely surviving it. A forest fire clears deadwood and opens the canopy. An economic crisis forces models to be rewritten and reveals which arrangements were load-bearing and which were decoration. Diversity, redundancy and distributed capacity are what make that possible, and none of the three shows up as a benefit in an accounting system tuned for efficiency.

This vocabulary matters because it determines what a good measurement looks like. If the economy is a living system, then the health questions are living-system questions. Where is wealth accumulating. How fast are the stocks being drawn down. Which loops are amplifying and which are correcting. How tightly is everything coupled, and what happens when one part goes. None of those questions can be answered by a larger version of the same single number.

Nothing here is urgent. The work of changing an instrument takes longer than the work of reading one, and it is done sitting down.

BOOKS THAT BALANCE

The technical heart of the argument is an accounting discipline with an unglamorous name: stock-flow consistent accounting. The rule is simple enough to state in one line and demanding enough to break most casual economic reasoning. Every flow comes from somewhere and goes somewhere. Every pound spent is a pound received. Every asset is someone's liability. Nothing enters or leaves the system without a matching entry.

The book introduces it through a Florentine bookkeeper in 1494, trying to hold an entire city's economy in a ledger designed for single businesses. His method worked for one merchant and collapsed for a city, because it treated the economy as a snapshot rather than as movement. He needed a system that tracked both the stock of what existed — wool, gold, skilled hands — and the flows that changed those stocks over time.

The canonical worked equation is the capital stock. Change in capital equals investment minus depreciation: dK/dt = I − δK. Start a country at a capital stock of $1 trillion, invest at 5% and depreciate at 2%. Year one brings $50 billion in and takes $20 billion out, so the stock rises by $30 billion. Year two starts from $1.03 trillion, invests $51.5 billion, depreciates $20.6 billion, and adds $30.9 billion. The compounding is visible, and so is the failure mode: let investment fall below depreciation for long enough and the stock shrinks while the income statement looks unchanged.

The same logic runs through a single business. Marie's mozzarella operation opens the year with $50,000 in cash and $100,000 in equipment, against a $30,000 bank loan and $120,000 of her own money. She sells $80,000 of cheese, incurs $45,000 of costs, and buys a $20,000 machine. Traditional accounting shows a balance sheet. Stock-flow accounting shows the balance sheet moving: net income of $35,000, cash up and then down, equipment up, and a capacity decision whose returns arrive in a later period than its cost.

Consistency is not a technicality. It is the discipline that stops an economic story from spending the same money twice.

Godley and Lavoie set this out at length in 2007, and it remains the reference text; Ryoo and Skott give the compact introduction. What the method buys is not prediction so much as honesty. Policy proposals that quietly assume money appears from nowhere fail the accounting before they fail in the world. Imbalances that accumulate slowly — debt building in one sector, capital eroding in another — become visible years before they become events, because the model cannot look away from where the flows are landing.

WEALTH NOT INCOME

Income is a flow. Wealth is a stock. Confusing the two is the most expensive error in public economics, and the book gives it a chapter.

The scene is a baker and an accountant. He has been through her books and found the income respectable and the assets alarming: an ancient oven, chipped mixing bowls, a building that would not impress a valuer. She points out that the oven bakes bread nothing else can match, that the bowls carry the record of a thousand mornings, and that the real capital of the place is the tradition, the customers and the skill in her hands. He has been measuring the flow and missing the stock it comes out of.

The book counts five stocks a society lives on, and the list is the practical core of the argument. Human capital: skills, education, health, the capacity to do work at all. Natural capital: soil, water, climate stability, biodiversity, the minerals and energy everything else is made from. Social capital: trust, norms, networks, the thing that decides whether a contract needs a lawyer or a handshake. Built capital: roads, grids, buildings, the physical apparatus of exchange. Intellectual capital: knowledge, patents, methods, the accumulated answers to problems already solved.

These stocks interact rather than sit side by side. Investment in health and education raises productive capacity, which generates the surplus that funds further investment in both. Environmental degradation depletes natural capital, which lowers agricultural yield, which raises food costs, which erodes household capacity. The loops run in both directions, and a system of accounts that only records annual income cannot see a single one of them.

A country can post its best year on record while consuming the stock that made the year possible.

This is the specific failure the wealth lens catches. High income drawn from a depleting stock reads identically, in the accounts, to high income drawn from a growing one. The distinction only appears when the balance sheet is kept alongside the income statement — when a felled forest, a hollowed-out training system or a bridge past its design life shows up as a reduction in the stock rather than as revenue. Dasgupta and Mäler made the case for net national product measured against wealth in 2000; Arrow and colleagues extended it. The method exists. It is not what gets read out on the evening news.

PRICING THE UNPRICED

The obvious objection to measuring five capitals is that only one of them has a price tag, and the book meets it directly rather than waving it away. Some of these stocks are genuinely hard to value, the methods available are imperfect, and the honest position is that imperfect measurement of something essential beats precise measurement of a proxy.

For natural capital, one route is shadow pricing: estimate what it would cost to replace or restore the service if it were lost. A watershed that filters water for a city has a value that can be approached by pricing the treatment plant that would be needed without it. Carbon held in standing forest can be approached by pricing the sequestration that would have to be bought instead. Costanza and colleagues ran the exercise at global scale in 1997 and the result was contentious precisely because it was large. The number is arguable. Its absence is not neutral — an absent number is treated as zero, and zero is the one value that is certainly wrong.

For human capital, present-value calculations do reasonable work: estimate the future stream of productive capacity that education and experience create, discounted to the present. It is the standard tool of finance pointed at people rather than at bonds, and it has the same weaknesses as any long-horizon discount rate, which the book does not hide.

Social capital is the hardest of the five and the book says so. There is no market in trust. What exists are proxies — survey measures of social trust, rates of volunteering, crime rates, the density of associational life — and each of them is a partial view of something that none of them measures directly. The right response is to use several, watch how they move together, and never report a single composite as though it were a reading rather than a construction.

An unmeasured stock is not treated as unknown. It is treated as free, which is how it gets spent.

There is also a structural route that avoids pricing altogether, and the book develops it early. Treat each activity as a node and each exchange of resources, knowledge or labour as an edge, and value becomes a property of position in a network rather than of a single transaction. A baker and a teacher in one community connect to a miller, a grower, a set of students, and eventually to the work those students go on to do. Network analysis makes the second-order contribution legible — who feeds whom, which nodes hold the structure together, where a removal would cascade — without requiring anyone to name a price for a conversation.

Two disciplines keep this from becoming arithmetic theatre. First, every valuation carries its method and its uncertainty on the same page as its figure; a shadow price presented without its assumptions is an opinion in a numeral's clothing. Second, the accounts track change over time as well as level, because depreciation and appreciation are where the decisions live. A building wears out. A skill compounds. Infrastructure investment adds to one stock while drawing on another. The flow-fund structure exists to hold all of that in one set of books that balances.

The plant was here before the desk and will outlast it. That asymmetry is the subject.

THE VITAL SIGN

An average is a summary of a distribution, and a distribution is where the information is. The book treats inequality not as a moral supplement to economic measurement but as a diagnostic reading, in the clinical sense — a vital sign that tells you something about system function that no total can tell you.

The chapter opens at a farmers' market: stalls full, band playing, children running, and a woman on a bench doing arithmetic on a grocery list and finding the apples out of reach. Both facts are true of the same square at the same moment. Any measure that reports only the first has not simplified the scene; it has edited it.

The Gini coefficient is the standard tool and the book works it end to end. Rank everyone from poorest to richest, plot cumulative share of population against cumulative share of income, and you get the Lorenz curve. Perfect equality is the diagonal. The coefficient is the area between the curve and that diagonal, divided by the total area beneath the diagonal, running from 0 to 1. The worked example takes five people earning $10,000, $15,000, $20,000, $25,000 and $30,000, and arrives at 0.4 — a moderate reading for a small and well-behaved distribution.

The Theil index does something Gini cannot: it decomposes. Because it sums individual income shares against the mean of the subgroup each person belongs to, the total can be split into inequality within groups and inequality between them. The book's example divides five people into two groups of three and two, with subgroup means of $15,000 and $27,500, and shows the arithmetic. That decomposition is the difference between knowing a country is unequal and knowing where the inequality sits — between regions, between sectors, between demographics, or inside every group at once.

Concentration is a structural reading, not a complaint. A pond with deep pockets and shallow edges supports less life than a pond of even depth.

The living-systems framing sharpens what these numbers are for. Concentration of wealth is also concentration of political influence, which shapes the rules that govern concentration — a reinforcing loop that cannot be broken by treating the symptom. Interdependence means the capacity of the least-resourced participants is a property of the whole system rather than a fact about them. And resilience depends on the breadth of the base, which is why distribution belongs on the dashboard next to the totals rather than in a footnote after them.

IN THE MARKETS

The argument would be academic if capital did not read the same accounts. It does, and the book's market chapters show what happens at the point where the measurement meets the money.

Two companies compete for investment. FossilFuel Inc. holds proven reserves, tangible assets on a balance sheet, and a market capitalisation that reflects them. SolarSpark has spent years on research and holds harder things to price: intellectual property, engineers, a customer base and a technology whose value is mostly in the future. Under an accounting system that reads only financial metrics, the first company looks stronger, and capital moves accordingly.

The book's proposal is an added column rather than a replacement — an Environmental Impact Score, calculated from scientific data and modelling, that reflects each firm's environmental footprint alongside the financial figures. Put both in front of the same investor and the ranking can change, because the second column carries information the first was never capable of holding. It does not make anyone virtuous. It makes a previously invisible variable available to a decision that was already being made.

The apparel example is the sharper one because the gap is a wage. EcoThreads pays garment workers $15 an hour and sources organic cotton from smallholders at fair prices. FastFashion pays the legal minimum of $7.25 and sources from large-scale monoculture. In the income statement, the difference is a cost line and FastFashion wins it. In a living-systems account, the difference is a distributional flow: one firm is building household capacity in the places it operates, and the other is drawing on it.

Two metrics make that visible, and both are constructible from data firms already hold. The living wage gap compares wages actually paid against a calculated living wage for the region, which turns a philosophical dispute into a figure. The supplier income share reports what percentage of revenue flows back to suppliers — particularly smallholders at the base of a chain — which turns the structure of a supply chain into a number you can track quarterly.

Every supply chain is already a distribution mechanism. The only question is whether anyone is reading it.

The pattern under all of these cases is the same one from the first chapter, arriving where it does the most damage or the most good. Investors allocate against what they can see. Widen the visible set and the allocation changes without anybody being persuaded of anything. Keep it narrow and a company can produce a decade of reported growth while consuming the workforce, the watershed and the supplier base that produced it, with every quarterly report accurate and every number true.

THE FIRST LEDGER

The book refuses to leave this at the level of national statistics, and its operational chapters run the same framework at three scales, because the arithmetic does not change when the numbers get smaller.

Start with the inventory. For a person, that is financial assets, physical assets, and human capital — skills, experience, health — alongside the social connections and reputation that determine what any of it can actually do. For a business, tangible assets and intangible ones: brand, patents, methods, and the accumulated skill of the people who work there. For a country, national wealth accounts that reach past income to natural capital, infrastructure, education levels and health outcomes. Nothing is measured until it is listed.

Then valuation, using whatever instrument fits the asset and saying which one was used. Market prices where a market exists. Professional appraisal for unique holdings. Replacement cost for infrastructure and for natural capital, where no market price is available and the replacement estimate is the most defensible number on offer.

Then the accounting framework, which is the part most often skipped. Wealth is not static; it moves with markets, investment, depreciation and discovery, so the accounts need regular updating on a fixed cadence rather than an occasional audit. And they need to be legible. Published national wealth accounts let anyone check whether the stock a country lives on is growing or being drawn down, which is a different question from whether the annual figure went up.

Then the part that makes it worth doing: the accounts have to enter decisions. For an individual, a clear picture of stocks rather than flows changes how debt, saving and training get weighed. For a business, tracking change in its capital stocks reveals what is quietly eroding while revenue holds steady. For a state, wealth accounts inform infrastructure, environmental and social spending in a way an income figure cannot, because they show what the spending is maintaining.

Measure the stock you are living on before you celebrate the income you took out of it.

Institutions have existing scaffolding to build against — the Genuine Progress Indicator, the Happy Planet Index, the UN's System of Environmental-Economic Accounting, B Corp reporting, the Global Reporting Initiative. None is complete and the book does not present any of them as finished. They are working attempts at the same problem, and the practical move is to adopt one, publish the result alongside the conventional figures rather than instead of them, and let the two columns be compared in public.

THE LUMINOUS LENS

The closing argument is not a new metric. It is a change in what the metrics are understood to be for.

Nobody would assess a person's life by their bank balance alone, and nobody would assess a redwood by counting the acorns it dropped this year. The tree's condition is in its rings, its roots, the fungal network it feeds and is fed by, and the centuries it has left. The acorns are the flow. Everything else is the stock, and the stock is what determines whether there will be acorns in fifty years. The book's whole case is that an economy has the same structure and deserves the same reading.

That reading changes the questions. Not how much was produced, but what condition the productive base is in. Not what the total was, but how it landed. Not whether the number rose, but whether the rise was drawn from a stock that is growing or one that is being spent. Are the natural systems holding. Are the communities connected enough to absorb a shock. Is knowledge accumulating. Are the loops that correct this system still working, or have the amplifying ones taken over.

Prosperity is a condition of a system, not a quantity extracted from it.

None of this requires abandoning GDP, and the book never argues for that. GDP is a good answer to the question it was built for, and that question still matters. What it argues is that a single-column account of a living system is an instrument reading one channel, and that the other channels exist, are measurable by methods that are already documented, and are currently reported as zero because nobody publishes them.

This is one volume of thirteen chapters in the Living Systems Economics shelf, and it runs to 47,803 words: the accounting, the equations, the worked numbers, the market cases and the operational frameworks at all three scales. The whole thing is free to read, start to finish, with no account and nothing to sign. Buying a copy is for keeping it — the EPUB, the PDF, the press file — and that is the only thing a price is ever for here. The reading is open either way.

Free to read

Free to read, and free to hear. Every chapter of every book in this house, and every narration of it, is open to anybody. No account, no card, nothing to cancel.

What is in it

• What We Count Becomes What We Valueopen this in the house search
• National Accounts and Their Ghosts: What GDP Was Built to Missopen this in the house search
• Stock-Flow Consistent Accounting: Making the Books Balance Across a Whole Economyopen this in the house search
• Wealth, Not Just Income: Measuring the Capital Stocks a Society Lives Onopen this in the house search
• Distribution as a Vital Sign: Inequality Metrics That Actually Diagnoseopen this in the house search

What a society refuses to count, it eventually refuses to fund.

An unpriced stock is never treated as unknown. It is treated as free.

Income tells you what came out this year. Wealth tells you whether there is a next year.

The average is the part of the distribution you are allowed to forget.

Keep looking

Every phrase on this page opens into the house search. The shelf holds Living Systems Economics and six other shelves, and the reading is free.