Reversibility as a Factor of Production: A General Theory of the Economics of Error

What the Article proposes. All growth is born of trials — projects launched, firms founded, techniques attempted — and most trials fail. What each society does with those failures is, this Article argues, the most decisive and least measured variable in economics. Where honest failure leaves its author with his capabilities, his assets and his reputation, people try often, learn fast and prosper; where failure destroys — perpetual debt, stigma, disqualification, personal ruin — the most rational agents stop trying, and the economy dies out for want of experiments. The Article gives this intuition a complete theoretical body: it introduces the “reversibility coefficient ρ,” the fraction of his capabilities an agent retains after an honest failure, elevates it to a factor of production on a par with capital and labor, and endows it with a law of formation, a geometry, theorems and an instrument of measurement. A pharmaceutical laboratory screens ten thousand molecules to retain a single one for a drug; the 9,999 failures are not losses but precious information — each closes a path, sharpens the map and brings the medicine closer, to the point that laboratories parse their competitors’ failures line by line. The Article asks that the economy as a whole treat error as the laboratory treats it — a raw material of discovery — and not as a punitive society treats it, where a mistake closes the future to whoever makes it. The thesis holds that development begins neither with savings nor with capital goods, but when falling ceases to mean being destroyed. Why reversibility truly is a factor of production. Economic science admits as a factor of production any magnitude that meets three criteria: it is scarce and costly to produce; its increase raises output, other factors held constant; and it accumulates and depreciates like a stock. Reversibility passes all three tests. It is productive: with capital, labor and talent held identical, ρ governs both the number of trials a society dares to make and the yield at which it converts its failures into reusable knowledge. It is costly and can be built: swift commercial courts, discharge of debts, social safety nets, secondary markets on which to resell assets, published case law — all real investments, and the stock thus constituted depreciates if maintenance stops, stigma returning like rust. It therefore has a law of motion, which sets it apart from the “institutions” invoked wholesale in the literature; it can be measured (the Reversibility Index, RI), decomposed and steered. It is complementary to the other factors and commands them: without it, capital and labor run below capacity — capital injected into an irreversible economy finances machines with no experimenters to discover their uses. The discipline has already widened the list of factors three times: human capital with Becker, knowledge with Romer, institutions with North. Reversibility is the next enlargement — and the first of these enlargements to arrive equipped, from the outset, with its measuring instrument and its theorems. What the Article brings to economic theory: completing the great theories. The deepest contribution is architectural. The received grand theories — from Smith to Keynes, from Solow to endogenous growth — were all written, often unknowingly, under one implicit assumption: that of a world in which failure is nearly reversible, in which bankruptcies are settled, debts erased, careers rebuilt. The Article’s “correspondence theorem” establishes that these theories are exact in that world and wrong by a growing margin as reversibility collapses. They are not refuted; they are incomplete — the special case ρ = 1 of a more general theory, just as Newtonian mechanics is the low-velocity case of relativity. Nine theoretical bodies are re-examined and amended on this basis: employment theory above all, where the Article demonstrates the existence of involuntary unemployment without any wage rigidity — Keynes had the right conclusion and the wrong premise. To the family of poverty traps (savings, coordination, nutrition, human capital), the Article adds a new species, the “conformity trap”: a loop in which growth depends on reversibility, which in turn depends on income, because the shock absorbers of failure — courts, safety nets, markets — cost money. Two stable equilibria result: lenient prosperity above, punitive poverty below. This trap explains why sixty years of capital injections failed where the diagnosis of capital shortage seemed correct: capital was missing downstream, but the right to be wrong was missing upstream. The geometry of the trap (catastrophe theory) yields four refutable predictions — transitions by discrete jumps; an exit cost far exceeding the cost of prevention; the divergence of twins (South Korea and Ghana, comparable in 1960, separated by a factor of ten fifty years later); and an early-warning signal legible in the statistical series fifteen years before the tipping point. Finally, the “under-supplied reversibility theorem” grounds the normative dimension: failure produces information of which its author captures only a fraction; the market therefore structurally supplies fewer second chances than the social optimum requires, and the institutions of clemency — bankruptcy law, safety nets — cease to be acts of generosity and become the correction of a market failure, on the same footing as the funding of basic science. What the Article brings to policymakers: measuring in order to govern. Because what gets measured gets governed, the Article constructs a Reversibility Index (RI) and pairs it with an operational guide enabling any country, whatever the sophistication of its statistical apparatus, to compute its full index within six months and on a modest budget. The index aggregates five components, each instrumented by existing or readily constructible indicators: financial (creditor recovery rates, duration of insolvency proceedings, share of credit extended without the entrepreneur’s personal guarantee, and so on); legal (time to debt discharge, existence and length of directorship bans, decriminalization of good-faith failure, personal fresh start); social (values surveys on the fear of failure, share of founders reporting a prior failure, safety nets open to the self-employed); cognitive (publication of case law, open business registers, firm-demography statistics); and political (peaceful exits from power over the long run, the actual fate of former leaders). The guide fixes the protocol: the orientation of each sub-indicator, normalization on bounds set a priori and identical for all countries, mandatory publication of the decomposition. The choice of aggregation is itself a matter of theorem: because the failed entrepreneur passes in succession through the court, the bank, the social gaze and the tax authority, and each stage can crush him on its own, the components multiply rather than add — a zero on a single link drags the whole index to the floor, as a bridge collapses the moment one pier gives way. From this structure follows a computable rule of government, the link-equalization rule: the marginal return to reform is highest on the lowest component, and effort should be allocated until the return per unit of cost is equalized everywhere. Weakest-link diagnosis ceases to be a consultant’s intuition and becomes a derivative, computable in a spreadsheet cell; in the same stroke, the rule proscribes showcase reforms, which polish the link that already shines. Sectoral instruments complete the toolkit: symmetric taxation of gains and losses, protection of the person rather than the job, institutional sorting of honest failure from fraud, honorable exits from power — and a theorem of reform strategy, the “clean leap,” which establishes that gradualism is strictly dominated, repression propagating through beliefs ten to fifteen times faster than clemency. This Article turns the second chance into a measurable magnitude, a variable of economic policy and the missing link of the great theories: the nations that dominate are not those that avoid mistakes, but those that have learned to lose without destroying their capabilities — and it hands the others the instruction manual.