This volume began, eleven plates ago, with the oldest suspicion in commerce — that the man who stands between the buyer and the seller and takes a cut of their dealing is a parasite, producing nothing, living on the labour of others — and with the function that the suspicion conceals. The question it set itself was where the truth lies between the suspicion and the function: when does the middleman earn the cut he takes, and when does he merely levy it? Eleven plates have pursued that question through the middleman's many forms — the dealer with his two prices, the matcher who brings the two sides together, the bank that watches the borrower for the saver, the arbitrageur who presses two prices into one, the platform that builds the market and prices its own indispensability, the rentier who holds a chokepoint and does nothing else, and the recurring dream of cutting the middleman out altogether. This final plate must draw the line the whole volume has been preparing — the line the table's masthead promised would fall never quite where either side claims it lies, between the middleman who creates value and the one who merely levies a toll on its passage. It is the verdict; and like most verdicts worth reaching, it is less a clean adjudication than a careful statement of how much can be settled and how much cannot.
Two parties stand ready to tell you exactly where the line lies, and both are wrong, in opposite directions. The first is the middleman's defenders — the apologists of finance, the celebrants of the platform, who would have it that the cut is all wage, every penny of it the honest price of a service rendered. To them the spread is purely the price of immediacy, the take rate purely the price of a venue, the commission purely the wage of a function, and the suspicion of the parasite a populist slander born of not understanding what the middleman does. They place the line at the far end of the field, and call the whole of the cut earned. The second party is the middleman's accusers — the moralists and populists in whom the oldest charge survives undiluted, who would have it that the cut is all toll, the middleman a leech upon honest trade who skims what others have made and adds nothing. To them every spread and fee and percentage is extraction, the financier and the platform and the broker so many tollbooths on the commerce of those who actually produce. They place the line at the near end of the field, and call the whole of the cut levied. The eleven plates of this volume have been, in effect, a sustained demonstration that both are mistaken — that the line lies at neither end, but somewhere in the contested middle, and that its precise location is the whole of the difficulty.
For the finding of plate after plate has been that the middleman's cut is not all one thing or the other but a mixture of both, in proportions the cut itself will not disclose. The dealer's spread, examined, dissolved into three components — the wage of immediacy, the premium for inventory risk, the tax of adverse selection — folded into a single number that would not say which part was which. The platform's take rate proved to be at once the wage of a venue genuinely built and the toll of a network that could not be left, the two so fused in the one percentage that they could not even in principle be prised apart. Even the toll, the volume's purest case of extraction, turned out on inspection to carry usually some sliver of buried service. The line between the earned and the extracted, that is to say, does not run between the honest middleman and the parasite, sorting the trade into good actors and bad; it runs through the cut of nearly every middleman alike, dividing the part of his take that is the wage of a real function from the part that is the toll of a mere position — and the number in which the two arrive is silent as to their proportions, wearing the same face whether it is mostly wage or mostly toll. This is the volume's central and most stubborn finding: that the line is real, and that it is hidden, and that it falls inside the cut rather than between the cutters.
The line runs not between the honest middleman and the parasite but through the cut of each, dividing the wage of his function from the toll of his position — and the number wears the same face either way.
If the cut will not disclose where the line falls within it, then the line must be found by other means, and the volume has, along the way, assembled the means for finding it — three tests, imperfect and contestable, but the best the inquiry affords. The first is the test of competition. Where a middleman faces rivals who would gladly undercut him, his cut is pressed down toward the cost of the service he performs, and whatever toll had swollen it is competed away; where he faces none — where he holds a chokepoint, or a tipped market, or a captive clientele — his cut is free to swell with levy. The presence of real competition is thus a sign that a cut is mostly wage; its absence, a warning that the cut may be mostly toll. The second is the test of removal, which the previous plate set out: try to cut the middleman out, and watch what follows. If the exchange proceeds as well without him, his cut was toll; if it falters, or if a new middleman must instantly arise to do what he did, his cut was wage. The third and most direct is the test of function: ask what, concretely, the middleman does — what gap he bridges, what risk he bears, what trust he manufactures — and measure his cut against the answer. Where a real and costly function is performed, a wage is owed; where none can be named but the holding of a position, the cut is toll, whatever its possessor may call it. None of the three is decisive alone, and all three are contested in the hard cases; but together they are how the line gets drawn — in markets, in courtrooms, in the slow judgement of the public, wherever the question of the middleman's desert is seriously pressed.
And this is the deepest thing the volume has to say about its own question: that the line between the wage and the toll is not a fact waiting to be read but a task waiting to be performed. Because the cut hides its own composition, no one can simply look at a spread or a fee and pronounce it earned or extracted; the proportions must be inferred, by the tests of competition and removal and function, and inference is the work of judgement, and judgement is contestable, and so the line is perpetually in dispute and perpetually being redrawn. The regulator who must decide whether a platform's cut reflects its value or its dominance, the court that must decide whether a patent-holder's royalty is a reward or a holdup, the saver who must decide whether his banker's margin is the price of a service or the toll of a captive — each is drawing the line afresh, in a particular case, against resistance, and none can appeal to a number that settles it. The volume's project, then — the distinguishing of the wage from the toll — is not a problem that can be solved once and filed away, but a labour that every market and every age must take up again, because the middleman is permanent, and his cut is permanently ambiguous, and the contest over how much of it he has earned is built into the very form of the price. The line is not given; it is drawn, and drawn again, and never quite agreed.
The line between wage and toll is not a fact to be read from the number but a task to be performed upon it — drawn by judgement, redrawn in every age, and never quite agreed.
Step back, at the end, to the claim from which the whole table proceeds: that every ordinary price is a verdict on a hidden contest, commonly mistaken for a fact of nature. The spread — the middleman's cut, in all its forms — is among the clearest vindications of that claim the table can offer. To the eye that does not look, it is the most natural thing in the world, simply what middlemen charge, a friction of doing business no more to be questioned than the weather. But it is nothing of the kind. It is a verdict, rendered anew in every transaction, on a contest between the middleman's service and the middleman's power — between the value he creates by standing between and the toll he can exact for being unavoidable there — and the size of the cut is the running tally of that contest, the measure of how far, in this market and this moment, the service has been disciplined toward its cost or the power allowed to swell toward its limit. The volume has not dissolved that contest, for it cannot be dissolved; the middleman will always be partly a servant and partly a rentier, and the proportion will always be fought over. What the volume has done is to show that the contest is there, beneath the placid surface of the number, and to furnish the means of reading the number for it. The spread is not a fact of nature. It is a verdict on a hidden contest — and now, perhaps, a little less hidden.
So the verdict, plainly stated, is a divided one, as the honest verdict on a divided thing must be. The middleman is neither the parasite his accusers see nor the pure servant his defenders claim. He is, in the main and in most of his forms, a genuine worker — the bearer of functions that exchange between distant or mistrustful or mismatched parties truly cannot do without, and that someone must perform and be paid for — and the long failure of every attempt to abolish him is the proof of it. But he is also, in part, and in some of his forms wholly, a rentier — the holder of positions that let him charge beyond the worth of any service, the keeper of chokepoints and networks and captive markets from which he draws a toll owed to nothing but his place. The oldest suspicion was therefore half right and half wrong: right that the toll is real, and present, and sometimes the whole of the cut; wrong that it is the whole story, that the middleman produces nothing and merely levies. His cut is a contested boundary between the earned and the extracted, falling at no fixed point but somewhere that must be found case by case, and found anew as the cases change. That is as far as the question can be carried, and it is, the volume would insist, a good deal further than either the celebrants or the accusers were willing to go — for both preferred a settled answer to a true one, and the true one is that the line is real but unfixed, and that the finding of it is never finished.
The Spread is finished, then, its twelve plates drawn, and the price of standing between is anatomized as well as this work can manage: the cut that is sometimes the wage of a service and sometimes the toll of a position, divided by a line that is real, and hidden, and never quite where either side would place it. And with this plate something larger closes. The Spread is the fifth of the table's first five volumes, and the last of them; with it the first of the table's six families is complete. The Wage, the price of labour; The Rent, the price of place; The Rate, the price of time; The Premium, the price of risk; and The Spread, the price of standing between — these are The Prices, the five fundamental returns, and each, the family has laboured to show, is the price not of a thing but of a position, and each rides the same unsettled seam between the earned and the extracted that this last plate has tried to name. Five volumes, sixty plates: the first family of the table, whole. What the middleman's cut claims a share of — the gain that comes into being when two sides who could not meet are brought together, the surplus the meeting creates, which the parties and the one between them then divide — is the threshold of another family, where the gains of exchange and the contest over their division are taken up in their own right; and the line this plate has drawn, between the service that earns a share of that surplus and the toll that merely seizes one, is the first cut into that larger question. As the table says of every price, first to last: the argument remains open.