Back to History

History of planning

Comparison: twenty-two answers to the same questions

The table reduces twenty-two profiles to nine questions that every planning system answers in one way or another. Read it by row to follow one author’s logic, or by column to compare answers to one question. The last row is the open planning model; the note “did not raise the question” means that the author’s profile contains no answer to it.

Who sets the goals

Who decides where the economy is heading, and how public that choice is.

Open planning

A public circuit: “one person, one vote”, citizens’ assemblies and a participatory budget; whoever sets the goals does not execute them.

Founders: the plan as forecast and the plan as directive

Gleb Krzhizhanovsky
Political leadership and expert engineers; the GOELRO plan was approved by the Congress of Soviets, where discussion meant endorsement, not a choice between alternatives.
Stanislav Strumilin
Planning bodies and party leadership in the name of the “will of the proletariat”; the goal outranks the inertial forecast and may run ahead of it.
Vladimir Bazarov and Vladimir Groman
Goals are bounded by objective trends and proportions; the choice of growth rates lay with experts, and the question of who decides was not raised.
Nikolai Kondratiev
The state sets directions, but on the basis of knowledge of trends; who chooses development goals was not specifically addressed in the model.

Industrialisation: at what cost to build

Yevgeni Preobrazhensky
Party and state; the rate of accumulation and the pace of industrialisation are described as an objective law, not a political choice.
Nikolai Bukharin
The party; industrialisation stays, but its pace should be the “highest sustained”, not a record; the dispute took place within a narrow leadership.
Nikolai Voznesensky
The state: the plan is subordinated to a goal set for the period; in wartime, concentrating all resources on defeating the enemy.

Balances and optimal planning

Wassily Leontief
Did not raise the question: final demand is given to the model from outside; the method answers “what if”, not what the goal should be.
Leonid Kantorovich
The goal is set from outside by the planning body: the model seeks the best way to produce a given product mix but does not choose the goal.
Viktor Novozhilov
Output volume is set from outside; the model seeks the minimum of total labour costs and answers “how”, not “what to produce”.
Vasily Nemchinov
The centre sets the framework, but the plan is built from below, from the enterprise, through orders and business contracts.
Nikolai Fedorenko
A single optimality criterion proposed by scientists and approved by party bodies, without open public choice.

Cybernetics and reforms: the search for feedback

Viktor Glushkov
Top leadership and planning bodies; OGAS itself sets no goals — in Glushkov’s formula it manages flows of information about the economy.
Evsei Liberman
The centre keeps planned targets for volume, product mix and delivery dates; the enterprise works out the rest itself.
Stafford Beer
The government and the organisation’s policy level; the promised worker participation stayed minimal, and the civic Cyberfolk project was never built.
Paul Cockshott and Allin Cottrell
A renewed Athenian democracy: councils and juries chosen by lot select among alternative plans, and the biggest decisions go to a direct vote.

Critics: the debate over the possibility of calculation

Ludwig von Mises
Consumer sovereignty on the market; “non-economic” goals are legitimate as human judgements but must not be passed off as calculation.
Friedrich Hayek
A comprehensive plan needs a single scale of goals that a free society lacks; the state sets general rules, not specific goals.
Oskar Lange
The central planning board; long-term investment is a conscious choice of the centre, and in one-party Poland goals were set by the apparatus.
János Kornai
Did not raise it as a central question: he studied how organisations actually behave and insisted that a mechanism cannot be judged apart from the political system.

The administrative market: how Gosplan actually worked

Vitaly Naishul
Officially the Central Committee, the Council of Ministers and Gosplan; in fact the goal is the outcome of bargaining among agencies, each holding a veto.
Simon Kordonsky
A common goal is set from above and broken down into goals for regions, sectors and socially registered groups.

Key contrasts

Geneticists versus teleologists

Vladimir Bazarov and Vladimir GromanStanislav Strumilin

In 1926–1928 Bazarov and Groman demanded that the plan be built from established trends, proportions and resources, while Strumilin wanted it built from set goals that themselves change conditions. The geneticists were right to warn of imbalances and a food crisis, the teleologists right that pre-war proportions could not be the limit for industrialisation. Neither side asked who chooses the goals, and the dispute ended with the geneticists’ arrest. From the teleologists we take the right of goals to run ahead of the forecast; from the geneticists, a public feasibility check and revision against facts.

Bukharin versus Preobrazhensky

Nikolai BukharinYevgeni Preobrazhensky

Preobrazhensky described industrialisation through hidden extraction of resources from the countryside via prices and tariffs, and called it an objective law. Bukharin replied that pumping out the maximum does not give the maximum pace, because it undermines the production from which the resources come. Preobrazhensky honestly named the price of growth and Bukharin correctly saw the links between sectors, but neither had a quantitative model, and the apparatus settled the dispute. We take explicit accounting of who pays for development and a check that the plan is backed by resources before it is approved.

Mises versus Lange

Ludwig von MisesOskar Lange

In 1920 Mises argued that without market prices for means of production alternative methods of production cannot be compared. Lange answered with a model in which the planning board finds prices by trial and error while keeping a consumer goods market and free choice of occupation. Formally Lange showed that calculation is possible, but Mises pointed to a deeper difficulty: without owners who risk their capital, prices remain an accounting convention. We take the rule that resource prices are discovered at auctions rather than assigned, and public comparison of every decision with the rejected alternatives.

Hayek versus Cockshott and Cottrell

Friedrich HayekPaul Cockshott and Allin Cottrell

Hayek held that the knowledge needed for economic decisions is dispersed, largely tacit and cannot be gathered at the centre. Cockshott and Cottrell answered that production information can be formalised and that labour values for the whole economy can be computed. Even Austrian critics accepted the computational complexity objection as answered, but questions about the honesty of data and the nature of knowledge remain open. From Hayek we take contracts for results instead of prescribed methods and the limits of metrics; from Cockshott and Cottrell, juries by lot and open data on public contracts — but not the abolition of money and the market.

Glushkov versus Liberman

Viktor GlushkovEvsei Liberman

In the mid-1960s the leadership was choosing between OGAS, a network of computing centres costing about 20 billion roubles, and an incentive reform that, in its supporters’ words, cost no more than the paper for a decree. The reformers feared that the network would entrench centralised planning, while Glushkov saw them as opponents of his project. Both programmes ran up against the same thing: ministries that commanded, allocated and assessed all at once. From Glushkov we take open registers and common data standards, from Liberman a few long-term norms and savings kept by the contractor, and we add the separation of roles that both of them lacked.

OGAS versus Cybersyn

Viktor GlushkovStafford Beer

OGAS was conceived as a comprehensive accounting system with a three-tier network of centres and for decades was bogged down in the agencies’ struggle over information. Beer’s Cybersyn was built in under two years on a single computer and old telexes: only significant deviations went upward, and factories were given time to correct themselves. The Chilean network proved its worth in the 1972 crisis but perished with the government; the Soviet project produced a strong engineering school but no network. In both cases worker participation remained a declaration, so we take reporting by exception and open standards, and enshrine the purpose of data and citizens’ rights in law.

Kantorovich versus Gosplan

Leonid KantorovichNikolai Voznesensky

In 1942 Kantorovich sent Gosplan a manuscript on the optimal use of resources; it was not approved, and after he appealed to Voznesensky the 1943 discussion turned hostile. Voznesensky’s Gosplan relied on material balances and the concentration of resources on a single goal, which worked in an emergency but offered no way to compare alternatives. Kantorovich was right about method, but calculation ran up against the agencies’ data and the apparatus’s unwillingness to cede decisions to a model. From Kantorovich we take shadow prices and the duty to explain rejecting a calculation; from wartime Gosplan, reserves and shortage monitoring — but not the merging of roles.

Kornai versus Lange

János KornaiOskar Lange

Lange believed that managers of state enterprises would follow the rule “price equals marginal cost” if the centre adjusted prices correctly. Drawing on the Hungarian reform, Kornai showed that an owner state cannot promise not to rescue a loss-making enterprise, and that under soft budget constraints price signals lose their force. Lange was right that adjustment against facts is possible, Kornai right that it does not work without hard accountability. We take iterative recalculation, but with a failure procedure written down in advance and every bailout of a contractor formally made public.

Naishul versus Glushkov

Vitaly NaishulViktor Glushkov

Glushkov saw the plan’s main trouble in slow and distorted paper information and proposed a network of computing centres. Naishul, who worked at an institute attached to Gosplan, showed that the plan was the outcome of bargaining among agencies with vetoes, and that bargaining distorts requests before calculation even begins; he considered the 1970s drive for automation powerless against this. The history of OGAS, which ran up against the agencies, tends to confirm Naishul’s diagnosis, although his programme of abolishing governing bodies contradicts our model. From Glushkov we take distributed registers; from Naishul, checking every procedure for tacit vetoes.

Naishul and Kordonsky: a shared diagnosis, different conclusions

Vitaly NaishulSimon Kordonsky

Naishul’s Moscow group, which studied Gosplan, and Kordonsky’s Novosibirsk group, which studied a rural district, reached the same diagnosis: the Soviet economy was bargaining over statuses and resources, not the execution of orders. Naishul saw the way out in quickly replacing the administrative market with a money market, while Kordonsky describes monetisation as a change of form without a change of essence. Naishul himself later admitted that after 1991 bargaining did not disappear but was “monetised”, which brings their assessments closer. We take the warning that auctions without independent verification will become a new form of the old bargaining, and accounting for results instead of the fact of spending.