A Penny Saved Is an Engine Stalled - Brink Lindsey, state capacity, and half of the argument he wouldn't make

Brink Lindsey, who used to be a vice president at Cato and now describes himself as a recovering libertarian, has an essay making the rounds again in which he says something that ought to be unremarkable and is not.

The richest, freest, most technologically advanced economies on earth all belong to countries with enormous, powerful, heavily regulated, generously insuring states. If markets were genuinely self-executing, if they sprang up wherever government failed to smother them, then the great commercial civilizations of our age would be Somalia and the Pakistani tribal areas. They are not. The market order is not a state of nature that governments interrupt. It is an artifact, and it has to be built and maintained by somebody.

From there Lindsey argues that the relevant variable was never the size of government but its quality: what the political scientists call state capacity, the ability to tax, to enforce, to deliver. The public and private sectors are not antagonists but complements. Niskanen's house term for the resulting program is the "free-market welfare state," and the case for it is that social insurance supports market function three ways: it expands what people are capable of doing, it absorbs risks no private insurer will touch, and it makes creative destruction politically survivable by ensuring the rules work for everybody rather than for the winners only.

I want to say two things about this. The first is that it is correct. The second is that it is missing the load-bearing beam, and the reason it is missing is not an oversight but a limit on how far a reformed libertarian can travel and still be recognizable to himself.

The referee and the pump

Notice the vocabulary Lindsey reaches for. Rules. Regulatory codes. Capture. Due process. Deliberation. Representation. Constitutional design. The public interest as both empowering and limiting principle. It is the vocabulary of a lawyer, which he is, and it produces a specific picture of the state: the state is a referee, and the quality of the game is a function of the quality of the officiating. Bad refs give you rent-seeking, monopoly, pollution, and a rigged contest. Good refs give you free entry, free exit, honest prices, and a contest worth watching.

I have spent a lot of my professional life in rooms where the quality of officiating is the whole ballgame, so I am sympathetic. But a referee has one indispensable property: he can, at least in principle, be neutral. He wants a fair game, not a particular winner. That neutrality is what lets Lindsey keep "limited government" as a live commitment even after he has thrown out every predetermined limit on government's size. The state gets bigger and better without getting partisan, because good officiating is in everyone's interest.

The trouble is that the referee metaphor cannot carry the argument Lindsey actually needs, which is an argument about demand.

All of his mechanisms treat the human being as an input. The welfare state builds human capital, insulates the worker against dislocation, and buys political consent for the churn. Every one of those is a story about the supply side and about legitimation. None of them is a story about the fact that capitalism, paradoxically and absolutely, requires the consuming class to consume. Somebody has to buy the output. Not as a moral matter, not as a fairness matter: as an accounting identity. Production that is not purchased is not profit. It is inventory.

And demand is not a natural resource. It does not well up out of the ground wherever people have appetites. Effective demand requires that the people with appetites also have money, and that they feel secure enough to part with it rather than hoard it against catastrophe. Both of those conditions are manufactured, and the thing that manufactures them is a set of state policies that move money down the distribution and put a floor under the fear.

Which means the state is not only a referee. It is a pump. And a pump, unlike a referee, cannot be neutral. A pump has a direction.

That is the whole difference, and it is why the argument stops where it stops.

Two folk models, both wrong

The popular imagination offers two theories of how to relate to money, and they are both, at the level of the whole economy, suicidal.

The first is the miser's model: a penny saved is a penny earned, cut the discretionary spending, build the reserve against the rainy day. As household advice this is often sound, and I have given a version of it to clients staring down a foreclosure. As macroeconomics it is a fallacy of composition, the one Keynes named the paradox of thrift. My spending is your income. If we all economize simultaneously, we do not all end up with larger reserves; we end up with smaller incomes and the same fear, having successfully starved the thing that was feeding us. The virtuous household scales up into a stalled economy. The engine has fuel and no combustion.

The second is the accumulator's model: outwork everybody, out-earn everybody, and let the gap between your pile and theirs be the scoreboard. This one fails differently, and worse, because it fails dynamically. Wealth at the top is not spent, it is placed. A family in the fiftieth percentile that gets an extra thousand dollars buys tires, dental work, a water heater, groceries... that money is back in circulation within the week. A family in the ninety-ninth percentile that gets an extra thousand buys a claim: an equity, a bond, a rental house, a share of somebody else's future income stream. The marginal propensity to consume falls as you climb, and it falls hard.

So concentration does not merely offend the sense of justice. It withdraws liquidity from the goods economy and redeploys it into the asset economy, where it bids up the price of existing claims rather than calling forth new production. Then those claims throw off returns, which are also placed rather than spent, and the process compounds. You get soaring asset prices and sluggish real output; you get an economy that looks rich on paper and feels poor on the ground; you get the entire strange texture of the last twenty years. The accumulator's model does not stall the engine by starving it. It stalls the engine by routing the fuel line into a storage tank.

Both models converge on the same failure, which is failure of circulation. The miser hoards defensively out of fear. The magnate hoards structurally out of arithmetic. Either way the money stops moving, and an economy in which money has stopped moving is not an economy, it is a museum with a gift shop.

Put that way, the case for the social democratic apparatus is not primarily humanitarian, though it is that too. Progressive taxation, transfer payments, unemployment insurance, a wage floor, sectoral bargaining, consumer protection... these are not compensation paid to the losers of the market game out of the winnings of the victors. They are the pump and valve assembly. They take money from the low-velocity end of the distribution and return it to the high-velocity end. Economists call the automatic ones "stabilizers," which is a suspiciously mild word for a mechanism that literally keeps the circulatory system pressurized when the heart stutters.

Lindsey's welfare state makes creative destruction politically survivable. Mine makes it economically possible, because there is no point destroying the old firm to make room for the new one if nobody has the money to buy what the new one sells.

Why the pump argument is unwelcome even when it wins

In making this argument, I am not blind to its limitations.

The demand-side story is not a universal law of development. Economies in the middle of catch-up industrialization are constrained by capital formation, not by consumption, and the East Asian developmental states got where they got by suppressing household consumption in favor of forced saving and directed investment. The circulation argument is an argument about mature, demand-constrained economies, which is to say about the ones Lindsey is actually talking about. The literature on whether growth in such economies is wage-led or profit-led is genuinely contested and I am not going to pretend otherwise. Fine. Stipulate all of it. Within the rich world the case still holds, and it holds for the ordinary reason that a country full of people who can afford things is a better place to sell things than a country full of people who can't.

So why has it lost for fifty years?

Michał Kalecki answered this in 1943, in an essay on the political aspects of full employment, and I have never encountered a better piece of political economy in ten pages. His observation was that business opposition to full-employment policy could not be explained on economic grounds, because full employment is good for profits: the shopkeeper wants customers with money. The opposition is about something else. Under permanent full employment, the sack loses its terror. The disciplinary mechanism that makes an employee compliant is the plausible threat of being replaced, and that threat is only plausible when there are people waiting to replace him. Take away the reserve of the unemployed and the boss's authority inside the workplace becomes negotiable. Class position, and specifically the everyday experience of being obeyed, turns out to be worth more to a certain kind of man than the marginal profit he would earn from a booming consumer market.

This is the thing Lindsey's framework cannot metabolize. He diagnoses the small-government dogma as an intellectual failure, a blind spot, a flawed ideology, an economics profession that professionalized itself into taking the market order for granted. Correct the error and the coalition reassembles around effective government in the public interest. But if Kalecki is right, the resistance was never an error. It was an interest, correctly perceived and rationally defended, and you do not talk a man out of a position he did not reason himself into.

Which is exactly the point at which a diagnosis of the state as referee becomes indispensable and a diagnosis of the state as pump becomes intolerable. Everyone can profess to want good officiating. Nobody can profess to be neutral about which direction the money flows.

A note on the word "neoliberalism"

I want to flag a terminological trap, because none of you use this word correctly.

It is tempting to describe the circulation-maximizing order I have sketched as neoliberalism's own best case, and to enjoy the irony of the Scandinavians beating the libertarians at the libertarians' game. But this gets the intellectual history backwards. The neoliberal project was never about a small state, and its founders would have found Lindsey's discovery banal. Röpke and the ordoliberals wanted a strong state, strong precisely so that it could impose and defend market discipline against the pressure of organized labor and mass democracy. As Quinn Slobodian has argued at length, the Geneva school program was not deregulation but encasement: build an armature of law, treaty, central bank independence, and investor protection that puts the market beyond the reach of ordinary politics. Hayek did not want the referee abolished. He wanted the referee insulated from the crowd.

So the real quarrel was never big state against small state. It was over what the strong state is for, and against whom it is insulated. Karl Polanyi named this in 1944 in terms nobody has improved on: laissez-faire was planned, the self-regulating market was a state project, and the protective countermovement that followed was the spontaneous part. What Lindsey has done, I think without quite meaning to, is rediscover The Great Transformation with the politics filed off, the double movement rendered as a bipartisan governance-quality problem. And the postwar settlement he is implicitly nostalgic for already has a name that fits better than neoliberalism: John Ruggie called it embedded liberalism, markets left free to allocate but bound inside a social compact that guaranteed the losses would be socialized. Neoliberalism is the name for the project of disembedding them again.

If we are going to fight about this, and we are, we should at least fight in the right vocabulary.

The uncomfortable coda

The orthodox Marxist reply to everything above is that I have described a maintenance schedule rather than a politics. Eliminate the medium of exchange, produce for use and distribute according to need, and the whole problem of circulation dissolves along with the circuit that generated it. Any argument that the welfare state is good for capitalism is, from that angle, an argument for capitalism, and the more persuasive it is the worse it is, because it makes a doomed system survivable and postpones the reckoning.

I do not think this objection can be waved off. It is true that if you defend social insurance on the ground that it keeps the engine running, you have volunteered to be the engine's mechanic. Rosa Luxemburg saw this a century ago and the tension has not aged. There is a partial answer in André Gorz's idea of non-reformist reforms: demands that are winnable inside the system but that shift the balance of power in ways the system cannot comfortably absorb, sectoral bargaining and de-commodified housing and a genuine right to refuse work being the obvious candidates. That answer is real and I lean on it.

But my actual answer is plainer: the stateless, classless, moneyless society is not arriving this quarter. Between here and there, actual people are being evicted, garnished, denied insulin, and worked to injury, and I meet them in my office on Fridays. Immiseration has never been an effective accelerant; heightening the contradictions is a theory with a long empirical record and it is not a good one. Whatever the eventual settlement looks like, it will be built by people who had enough slack in their lives to imagine one, and slack is precisely what a strong safety net, real labor protection, and enforceable consumer rights actually purchase.

So: yes to everything Lindsey says about state capacity, and then one step further than he can go. The state is not merely the referee that makes the game fair. It is the pump that keeps the blood moving, and a pump has a direction, and choosing the direction is not a governance-quality problem to be solved by better congressional staffing. It is the fight. It has always been the fight. The most a recovering libertarian can do is arrive at the edge of it and hand us the map.