The Divergence

The volume has set two answers against each other — productivity, embodied in the production function and the marginal product, and power, embodied in the markup and in the cycling struggle over the share — and has watched them contend over the great fact of the falling labor share. The author's question has stood unanswered behind every plate: is the split between labor and capital set by productivity, or by power? Thomas Piketty, in the most influential work of distributional economics in a generation, supplied not merely an answer but a reframing — a long view, across two centuries, of the division between those who work and those who own, and a claim about what governs that division that cuts beneath the whole productivity-versus-power debate. The flagship draws out the reframing, and with it answers the question and closes the family of the aggregates.

Piketty's first contribution was empirical and historical. By assembling data on income and wealth stretching back two centuries, he showed that the division between labor and capital, far from being the stable thing the mid-century constancy had suggested, has varied enormously over the long run. Capital's share of income was high in the deeply unequal societies of the nineteenth century; it fell through the shocks of two World Wars, the Great Depression, and the high-growth, high-tax decades that followed; and it has risen again since around 1980 — the same decline in labor's share the previous plate described, seen now as a single swing within a centuries-long history. The mid-century constancy that Douglas had enshrined was, in this longer light, not a law at all but an interlude: a peculiar and temporary compression, the product of war and depression and deliberate policy, lying between two ages of high capital share. The division of the product, it turns out, has a history — and that history is anything but a flat line.¹

For the tendency of capital's share to rise, Piketty proposed a mechanism, captured in a now-famous inequality: r greater than g, the rate of return on capital exceeding the growth rate of the economy. When capital earns a return year upon year, and the economy as a whole grows more slowly, then capital and the income that flows to it grow faster than the economy, so capital's share of the total tends to rise, and accumulated wealth, compounding faster than income, comes to loom ever larger over the society. This is no law of iron but a tendency, and it can be overridden — by fast growth, a high g such as the postwar boom enjoyed; by the outright destruction of capital, as the wars accomplished; or by taxes and policy that hold the net return down. But absent such countervailing forces, the logic runs steadily toward a rising capital share and a concentrating stock of wealth. The forces that compressed the distribution in the mid-twentieth century have faded; the return on capital has again outpaced growth; and capital's share has accordingly climbed.²

Here is Piketty's deepest claim, and the one that answers the volume's question by rising above it. The division of the product is not governed by an immutable economic law that we merely discover — not by the production function alone, not by any natural constant fixed in the technology of making things. It is governed by forces, some of them economic, such as the return on capital and the rate of growth, and some technological; but it is decisively shaped by institutions and by politics: by taxes, by the laws of property and inheritance, by the strength or weakness of organized labor, by the wars and crises that destroy or preserve great fortunes, by the choices a society makes about how much concentration it will permit. The mid-century compression was not nature relenting but policy acting — progressive taxation, powerful unions, the regulation of capital and finance. The recent rise of capital's share was not nature reasserting itself but those same policies being unwound. Distribution, in Piketty's telling, is the running outcome of political choices, and the comfortable belief that it is fixed by impersonal economic law is itself a kind of ideology — one that conveniently absolves the winners of any responsibility for the result.³

Now the author's question can be answered with precision. Is the split between labor and capital set by productivity or by power? The truthful answer is: by both, but they play altogether different roles. The marginal-productivity theory — Clark's principle, Douglas's production function — describes a benchmark, namely what the division would be in a frictionless world of perfect competition, where each factor is paid exactly its margin and no one holds any power over price or wage. That benchmark is real and useful, and productivity genuinely shapes it. But the actual division, and above all the movements of it — the fall of labor's share since 1980, the rise of capital's — is governed by precisely the things the benchmark assumes away: market power and the markup, the balance of bargaining strength through the cycle, the cheapening of capital, and beneath them all the institutions and politics that set the rules of the game. Productivity sets the frictionless benchmark; power, operating through institutions, determines how far the actual division departs from that benchmark, and in whose favor it departs. And because the part that has moved — that has shifted, in our age, against labor — is the part governed by power, the honest weight of the answer comes down on power.⁴

Why does the answer matter so greatly? Because the division of the yearly product hardens, over time, into the distribution of accumulated wealth, and wealth is power of a deeper and more durable kind. A high and rising capital share, sustained by a return that exceeds growth, flows to the owners of capital, who save and reinvest much of it, so that the stock of wealth grows and concentrates — and concentrated wealth, especially once it is inherited rather than earned, shapes not only consumption but politics, opportunity, and the very rules of the game, in a self-reinforcing spiral. The functional division between labor and capital that this volume has studied is thus the headwater of the personal inequality that so preoccupies our politics: a falling labor share today becomes a more unequal and more dynastic society tomorrow. This is why the question of whether the split is natural or chosen is no academic nicety. If the division is natural, it must simply be endured; but if it is chosen — set by institutions and politics, as Piketty argues — then it can be changed, and the concentration it produces is a decision a society is making, not a fate that it is merely suffering.⁵

Piketty's thesis has drawn vigorous challenge, and honesty requires that the limits be stated. Critics question whether the return on capital reliably exceeds growth, and by a margin wide enough to drive the dynamics he claims; whether the return would not fall as capital accumulates, diminishing returns pulling it back down toward the growth rate; and whether his capital, which lumps together productive machinery, housing, and financial claims, is even the right object — much of the measured rise in wealth, some argue, is the soaring value of urban land and housing rather than the productive capital his mechanism assumes. Others contend that the recent rise in capital's share reflects market power and fattening markups, the Kaleckian story, more than the patient accumulation that r exceeding g describes — which would, ironically, strengthen rather than weaken the power answer while changing its mechanism. The magnitude of the gap between the return and growth, the behavior of the return as capital swells, and the proper measurement of capital all remain genuinely contested. What survives the criticism, and commands wide assent even among the critics, is the central reframing: that the long-run division between labor and capital is not fixed by any law of nature but moves with institutions and politics, and has moved, in our own age, toward capital.⁶

With the share, the family of the aggregates is complete, and it is worth seeing the four motions whole. Growth is the long rise of the economy's real output — the swelling of the product across the generations. The cycle is the fluctuation of that output around its trend — the recurring lurch between boom and idleness. The level is the value of the money in which the product is counted — the changing worth of the very unit of measure. And the share is the division of the product — how the whole that grows, fluctuates, and is measured comes finally to be split between those who made it with their labor and those who owned the capital with which it was made. Together the four are the economics of the whole: how much a people produces, how steadily, in money of what worth, and divided in what proportion. And the share is, in a sense, the family's moral center, for it asks not how large the product is or how steady or how soundly measured, but to whom it goes — the question on which all the others finally bear, and the one a society cannot escape by growing richer.⁷

And the share closes a circle that the very first volume opened. The Wage took up the price of a single hour of labor and found, beneath the marginal product, the bargain — the relation of power between the worker who must sell his labor to live and the employer who buys it. The Share takes up that same relation at the scale of the whole economy: not the price of one hour but the portion of the entire national product that labor, in the aggregate, commands against capital. The two are one question asked at two scales, and they return the same answer — that productivity sets a benchmark but power sets the outcome, whether for the single wage or for the aggregate share. The worker bargaining over his pay and the working class commanding its share of the national product are the same figure seen near and seen far, and the forces that hold down the one — weak bargaining power, the dominance of the employer, the absence of countervailing institutions — are precisely the forces that hold down the other. From the price of an hour to the division of a nation's product, the thread of power runs unbroken.⁸

The division of the product between labor and capital is not, in the end, a law of nature that economics discovers, but an arrangement that societies make. Productivity sets a benchmark — the frictionless split each factor would receive if paid exactly its margin — and that benchmark is real. But the actual division, and the great movements of it, are governed by power: by the markups that market dominance permits, by the bargaining strength that employment and organization confer, by the cheapening of capital, and beneath all of these by the institutions and the politics that write the rules — the taxes, the property and inheritance laws, the protections of labor, the tolerance of concentration. Piketty's long view shows the share swinging across two centuries with these forces, compressed by the policy of the mid-century and widening again in our own time as that policy was undone, while the return on capital outran the growth of the economy and capital's portion climbed. Is the split set by productivity or by power? By productivity at the benchmark, but by power at the margin that matters — and since power is exercised through institutions that a society chooses, the division is, in the last analysis, a choice. With that, the aggregates are complete: the economy's product, having grown and fluctuated and been measured, is at last divided — and the dividing is the work not of nature but of us.