The Broken Window Fallacy: Why Destruction Doesn't Create Wealth

The parable, exactly as Bastiat told it

In his 1850 pamphlet Ce qu'on voit et ce qu'on ne voit pas ("That Which Is Seen, and That Which Is Not Seen"), Frédéric Bastiat imagines a shopkeeper, James Goodfellow (James B.), whose careless son shatters a pane of window glass. Neighbors gather to console him with a comforting theory: "Everybody must live, and what would become of the glaziers if panes of glass were never broken?" The repair costs 6 francs, that sum passes to the glazier, the glazier spends it in turn, and the town seems to hum with activity. If breaking windows makes work, the onlookers reason, then breakage is a public benefit.

Bastiat's reply is the birth of a whole method of economic thinking. "Your theory stops at what is seen," he writes. "It takes no account of that which is not seen." Yes, the glazier gains 6 francs — that is seen. But those 6 francs would otherwise have gone to a shoemaker for a new pair of shoes, or to a bookseller for a book. That sale now never happens — that is not seen. The shopkeeper ends the day with a window he already had, minus the shoes he would have owned. "Society loses the value of things which are uselessly destroyed," Bastiat concludes. The town is not richer by one glazier's wage; it is poorer by one window.

The mechanism: it is opportunity cost with a spotlight problem

The fallacy is not really about windows. It is a claim about how humans reason: we systematically over-weight vivid, concentrated, visible effects and under-weight diffuse, hypothetical, invisible ones. Bastiat's whole essay is a psychology of attention dressed as economics.

Formally, the broken window fallacy is a failure to net out an opportunity cost. Define the shopkeeper's wealth as the bundle of things he owns. Break it into cases:

Before break: wealth = window + 6 francs
No break: wealth = window + shoes (spend the 6 francs on shoes)
After break: wealth = window (spend the 6 francs replacing the window)

Compare the last two lines. The counterfactual — the world that would have existed — is the correct baseline, and against it the destruction subtracts exactly one shoes worth of value (6 francs). The naive observer instead compares "after break" to "right after break, before repair" and sees the repair as a gain. Wrong baseline, wrong conclusion.

Written as a net-benefit condition, breakage is beneficial only if:

value created by the activity > value of what was destroyed + value of the forgone alternative

For a smashed window this is impossible: the activity merely restores what existed, so its value ≤ the destruction, and the forgone alternative is pure loss. The sign is unambiguously negative. The seen is the first term; the unseen is everything on the right.

A worked example: the glazier is not a net job creator

Scale Bastiat up. Suppose a hailstorm destroys 10,000 windows in a town, each costing €300 to replace — a visible €3,000,000 of glazing work. Local newspapers report a "construction boom" and glaziers hiring. The seen effect is real and photogenic.

Now trace the unseen. Those 10,000 households each had a plan for their €300 — a phone, a weekend trip, tuition, savings that would have been lent out to a business. Those €3,000,000 of purchases evaporate. So:

• Glaziers gain €3,000,000 of revenue — jobs appear in glazing.
• Everyone else loses €3,000,000 of demand — jobs disappear in phones, travel, education, lending.
• Net jobs: roughly zero. Employment was relocated, not created.
• Net wealth: −€3,000,000. The town owns the same 10,000 windows it started with, and has forgone €3,000,000 of new goods.

The illusion survives because the €3,000,000 gain is concentrated in one visible industry, while the €3,000,000 loss is smeared thinly across a thousand transactions that simply never happen — no headline ever runs "Local phone shop sells 47 fewer units this month due to hail." Bastiat's point is that the unemployment you can photograph is not more real than the prosperity you cannot.

Why disasters seem to 'boost the economy' — GDP vs. wealth

The most consequential modern form of the fallacy is the recurring claim that natural disasters or wars are "good for GDP." After Hurricane Katrina in 2005 and Hurricane Sandy in 2012, and after most large storms, forecasters routinely predicted a growth bump from rebuilding — and were often right that measured GDP would rise.

Here is the subtle bookkeeping error that makes this true and misleading at once. GDP measures gross flow of new production, not the stock of wealth. When a hurricane destroys $100 billion of homes, roads, and factories, national accounts do not subtract that destroyed capital — GDP has no line for "wealth annihilated." But when the region rebuilds, every dollar of reconstruction is added as new output. So the ledger records only the plus side of a round trip:

ΔGDP ≈ +$100B (rebuilding) Δnet worth ≈ −$100B (destroyed) + $100B (rebuilt) − forgone alternatives

GDP goes up; the nation's balance sheet goes down or, at best, sideways, because the labor and capital used to rebuild the old could have built something new instead. This is why economists distinguish gross activity from net wealth: a country that spends 5% of its output replacing what a war destroyed is not 5% richer — it is running to stand still. Bastiat anticipated the exact confusion 155 years before Katrina.

The key assumption — and where the fallacy legitimately breaks down

The parable's power depends on one crucial assumption: resources are fully employed. The 6 francs, the glazier's hours, the town's glass — all have valuable alternative uses. That is why diverting them is a loss. The forgone alternative is only a cost if it was really available.

Relax that assumption and the argument changes. Suppose the glazier was unemployed, the shopkeeper's 6 francs were sitting idle under a mattress, and no shoes would have been bought either way — a world of idle resources and deficient demand, as in a deep depression. Then the repair employs a man who would otherwise have produced nothing, using money that would otherwise have circulated nowhere. The opportunity cost of the unseen is close to zero, and spending really can raise output. This is the Keynesian critique, sharpened by Henry Hazlitt's opponents: in a slump, the "unseen" alternative may be nothing at all.

So the honest statement is conditional. When the economy is at or near capacity, the broken window fallacy holds ironclad: destruction is pure loss. When the economy has deep involuntary idleness, breaking the window can raise measured output — though even then, note the deeper point: you never needed to break anything. If idle resources are the problem, hire the glazier to install a new window in a new building, and you get the employment benefit plus a net addition to wealth. Destruction is never the efficient way to employ idle hands; it is, at best, a costly excuse.

The fallacy in the wild: policy, war, and 'make-work'

Bastiat's essay was not really about vandalism — the window is a teaching device for a hundred policy arguments that all commit the same accounting sin:

• "War is good for the economy." WWII is often credited with ending the Great Depression via spending. But wartime output was tanks and shells that were then destroyed; consumer living standards were rationed downward. The recovery in living standards came after 1945 when resources returned to civilian goods — the unseen made seen.
• Disaster and "rebuilding booms." The Katrina/Sandy narratives above; also the recurring line that earthquakes "stimulate" construction. Rebuilding restores; it does not enrich.
• Make-work and destruction-for-jobs. Bastiat elsewhere skewers "dig a hole and fill it in" logic and even proposes a mock "candlemakers' petition" to block out the sun so lamp-makers can prosper — a reductio of the same error. Modern echoes: protectionist tariffs that "save" visible jobs while raising costs on unseen millions; "cash for clunkers" (US, 2009) which destroyed ~690,000 serviceable used cars, aiding new-car dealers while raising used-car prices for poorer buyers — a broken window with a subsidy attached.
• Planned obsolescence and "buy to help the economy." The intuition that throwing away working goods "keeps money moving" is the fallacy in a shopping bag.

Henry Hazlitt built his 1946 bestseller Economics in One Lesson on this single move, distilling it into a maxim: "The art of economics consists in looking not merely at the immediate but at the longer effects of any act or policy; it consists in tracing the consequences of that policy not merely for one group but for all groups."

The subtle mistake people still make

The most common misreading is to think the fallacy claims "repairing the window has no benefit." It does — for the glazier. Bastiat never denies the seen benefit; he denies that it is a net benefit for society. The glazier's gain is exactly offset by the shoemaker's loss. The fallacy is not "spending is bad" but "gross spending is not net gain."

A second, deeper confusion is treating money as wealth. The onlookers see 6 francs circulating and mistake the motion of money for the creation of value. But money is a claim on goods, not a good; shuffling it faster does not multiply the underlying loaves, shoes, and windows. What was destroyed was real — a physical window — and no velocity of francs restores it without using up real labor and materials that had another use.

Finally, a subtle correct point often lost on both sides: the fallacy is fundamentally about counterfactuals, which is why good economics is hard. The "unseen" is invisible precisely because it is the world that didn't happen. Sound reasoning requires you to imagine and price the road not taken — the same discipline behind opportunity cost, cost-benefit analysis, and every serious policy evaluation. Bastiat's lasting gift was not a slogan about windows but a habit of mind: always ask what else that money, that labor, that hour would have done.

The seen transaction vs. the unseen alternative — same money, very different bookkeeping

Dimension	The SEEN (naive view)	The UNSEEN (full view)
What is counted	Glazier earns 6 francs; "jobs created"	Shoemaker's lost 6-franc sale; job merely moved
Effect on wealth	Appears to add activity	Society is poorer by exactly one window
What the shopkeeper owns after	A window (that he already had)	A window instead of a window AND shoes
Measured in GDP	Repair spending shows up, boosts GDP	Destroyed capital never subtracted from GDP
Economic logic	Spending = prosperity	Spending on replacement = restoring, not gaining