Odd Lots Notes: A Socialist Who Says Markets Are Good

Contents

1. One word glued onto another
2. An operating agreement and a risk-priced loan
3. He names the places it breaks before you do
4. Further reading
5. One thing to take away

When the great way prevails, the world belongs to all… Goods are hated when left to rot on the ground, yet need not be hoarded for oneself; labour is hated when withheld from the body, yet need not be given for another’s sake alone.

—— “Liyun”, Book of Rites (pre-Qin to Western Han; translation mine)

In the 22 October 2026 episode of Odd Lots, Joe Weisenthal and Tracy Alloway interview Bhaskar Sunkara — co-author of The Blueprint: How Socialism Can Work in the Real World, founding editor of Jacobin, now president of The Nation. He says on air that he sits on the Mises–Hayek side of the socialist calculation debate: planning cannot out-compute a market. So his model keeps markets and consumer choice, has a new firm bring a business plan to a public bank at a risk-priced interest rate, makes every employee a member-owner paid a base wage plus residual profits, puts manager-to-worker pay differentials at four or six to one, and sets benchmark wages by occupation — his example is a line cook at $15 an hour. He also grants that matching capitalism on efficiency is “a very soft claim”, that socialising all land and housing is the most radical part of the book, and that administrative pricing creates distortions. The value of the episode sits in that blueprint; the book leaves out how you would get there from here.

One word glued onto another

Before listening, I filed “market socialism” under rhetoric. I assumed the real disagreement on the left was about speed, with everyone heading for the same destination — abolish markets, have some central body decide what gets produced and how much — and that market socialism was a compromise phrase for people who could not reach that destination.

I had a lazier assumption underneath. I thought the hard part of this project was moral persuasion, and that the technical questions could wait. Where prices come from, who funds the first shop, whether a loss-making shop is allowed to close: in my head those were the hostile questions an opponent asks, rather than the ones a proposal has to answer for itself.

Sunkara’s opening story landed on exactly that laziness. He got interested in socialism around 14, and his first question was whether it would just lead to bread lines. The answer he got from the socialists he sought out was: better a line for bread than no bread at all. He says that works if you are in 1930s China and fails in the wealthiest societies known to man. He wanted a different answer and spent decades finding a version that convinced him.

An operating agreement and a risk-priced loan

Tracy asked the most useful question of the hour: say I want to open a restaurant — where does the money come from, how do I hire, who keeps the profits.

Sunkara’s answer runs at the level of process. You take a business plan to a public bank and go through underwriting the way you would today, and the bank prices risk — a new restaurant draws a higher rate than an already profitable business borrowing to expand. Once the money lands, you and the people joining you sign a first operating agreement, which probably installs you as manager and allows a pay differential, four to one or six to one. The difference is that those people are member-owners of the firm from day one, drawing a base wage plus a share of residual profits.

Above that sits another layer: benchmark wages, a floor for broad categories of occupation, in the spirit of Australia’s old award system or Nordic centralised bargaining. A line cook at $15 an hour is not a lot of money, and it closes off one road. Sunkara uses himself to show what that road looks like. He tells the story of the first Indian Americans buying New Jersey motels — taking over a place limping along on a one or two percent profit rate, firing most of the staff, working 60 to 80 hour days, pulling in the wife, the children, the cousins, squeezing out enough profit to buy the next place, then bringing a nephew over to run it the same way. Joe remembers the New York Times headline from 4 July 1999, “A Patel Motel Cartel”. And Sunkara says: that is exactly what I would do in a socialist society, I like work, I have a large extended family of cousins I could solicit into any odd task, I would take that low road. So the benchmark wage exists to stop him. Wages cannot be pushed down any further, which leaves better management, a better product, or capital that saves labour.

Firm failure is a feature of the design. He reaches for the Hungarian economist János Kornai and the “soft budget constraint”: twentieth-century state socialism drew its legitimacy from full employment, so it kept inefficient firms alive, and the lousy factory making screws could not be shut down without jeopardising the computing sector downstream that ran on those screws. Firms going under is a sign of health — a line I enjoyed hearing from a socialist.

Why go to all this trouble rather than tax the rich and build a bigger welfare state? Tracy asked, and the answer has a moral half and a historical half. Sweden in the 1960s ran full employment without runaway inflation, and once workers felt secure they began demanding higher wages and industrial democracy. The Swedish capitalist held one decision — whether to keep investing — and past a certain squeeze he stopped, with the OPEC shock and globalisation pushing from outside. Sunkara’s reading: that system was economically possible and politically unstable. The Meidner Plan tried moving shares from capitalists to ordinary workers and went nowhere; what did happen was the third way, which kept the existing redistribution and halted the advance. He wants that veto decision out of the hands of the few who hold it.

The last stretch turns to AI, where Joe offers an elegant reversal: early AI tried to write intelligence down as rules and failed, then succeeded by collecting data; if AI keeps getting better at predicting the result of an experiment, you stop having to run it — and capitalism looks like entrepreneurs forming hypotheses and testing them with firms, so AI could remove the need for the firm as a hypothesis machine. Sunkara declines. People’s wants and needs change constantly, so there is no data set to collect; AI could compute the most efficient way for the state to lay a railroad and produce a static perfect plan, then fail in execution, and any change means redrawing the plan. The two of them agree to revisit it in ten years.

He names the places it breaks before you do

You may hold another version of my old assumption: someone whose politics sit far from yours is talking institutional design, and listening is a waste of an hour. This episode changed my test to one question — does the person proposing it name where their own proposal breaks.

Sunkara does it throughout. He flags the housing chapter as the most radical in the book and the hardest sell on himself: he is from the New York suburbs, the dream was a nice place on Long Island, and on air he takes apart what he actually wants from a home. The answer is security — a landlord who cannot raise him a thousand dollars a month next month. So his version socialises all land and housing while keeping a market for rents, where someone who wants Manhattan Beach pays more; construction is commissioned by the public sector while the building firms stay worker-owned, and they build wherever rents are climbing, with no zoning and no community board to stop them. Then he adds the price himself: protecting secure tenure means the state smoothing rents when an area runs up 20, 30, 40 percent, which is administrative pricing, which distorts, and which adds layers of bureaucracy. His bet is that unleashing the state on construction outweighs those inefficiencies. He does not say by how much.

That turns out to be more useful to me than any conclusion, and I use it on research. When a bullish write-up lands, I look first for the failure conditions the author concedes, and if there are none I treat the thing as an advertisement. The hole I have fallen into runs the other way: for a company I like, I can find an explanation for every piece of bad news, and what I am holding then is a belief rather than a judgement.

A second layer follows. Technically workable and politically survivable are two separate gates, and the Swedish story cleared the first and died at the second. Tracy closes the episode on what the book leaves out — people own their houses today, they own publicly traded stocks, so how does the transition happen and how much opposition meets it — and Joe asks the harder version: you move to socialism, and four years later the voters dislike it, do you move back? In a country with elections every two years, staying is harder than arriving. The same gate shows up in my own holdings: a company’s model can work and still depend on a subsidy, a particular rule, or cheap money, and then the thing to track is whether that condition still holds rather than the gross margin.

The third layer is where I am unresolved. A benchmark wage is also a price floor, and the low road it closes is the exact road the family in his motel story walked; facing a $15 floor back then, that second motel probably never happens. He says capital improvements should be requested, approved and reimbursed — who approves, how long it takes, what happens on refusal, none of that is in the episode. Nor is who sets the degree of rent smoothing, or on what basis. These are not reasons to dismiss the model. They are the pages I will turn to next time someone tells me a system can work.

Further reading

• Odd Lots, 22 October 2026, the Bhaskar Sunkara interview, hosted by Joe Weisenthal and Tracy Alloway
• The Blueprint: How Socialism Can Work in the Real World, by Bhaskar Sunkara with Ben Burgis and Mike Beggs
• János Kornai on the soft budget constraint, the concept behind the episode’s account of why state socialist firms never closed
• Public material on the Mondragón cooperative group in Spain, Sunkara’s working example of labour-managed firms at scale
• Research on Sweden’s Meidner Plan, against the episode’s “economically possible, politically unstable” passage

One thing to take away

How much a proposal is worth depends on whether the person proposing it can name where it breaks. In one hour Sunkara named the distortions of administrative pricing, the extra layers of bureaucracy, and called the efficiency case a very soft claim; those admissions are what made me take the rest seriously, and a proposal with nothing but upside in it reads to me as a sales pitch.

Here is something I have tried. Take one thing you are currently selling to someone else — it need not involve investing; the holiday plan you pitched at home, the new process you took to your manager, the exercise habit you announced. Write two lines: line one, the condition under which this turns ugly; line two, how far that condition sits from you right now. The first time I tried it, half my list stalled on line one, and that half never happened.

Raw Markdown

This article is an educational discussion of investment method. It is not advice to buy or sell any individual security, offers no target prices, and does not analyze any current holding. Investing carries risk; make your own decisions or consult a qualified professional.