Why did the planned economies fall behind, and was it inevitable?

The question the ledger leaves open

A reader who has wondered whether the way things run is the only way that works has, by this point in the volume, seen the evidence. The previous chapter put the rival bundles through their ledger: the split-country contrasts, imperfect as that chapter is careful to say, moved material outcomes by multiples; the planned side scored real wins in the columns mobilization can reach and lost decisively in income. What the ledger records, though, is only that the planned economies fell behind. It does not say why, and it does not say whether they had to. Those are different questions, and the second is the one with stakes beyond history, because it decides whether the century’s great alternative failed by nature or by choices that could have been made differently.

Two easy answers circulate, and both die on the same fact. One says people simply will not work without profit, so the system was stillborn. The other says the system was strangled from outside, by encirclement, war, and an arms race it could not refuse. The fact that kills both is the rise. The Soviet economy grew, fast, for three decades, without private profit and despite the encirclement; then, at the height of its power and peace, it stalled. An honest mechanism has to produce both halves of that record: the rise and the stall, from the same machinery. This chapter is the canonical home of that mechanism, and it runs on two shortages no planning office ever overcame, a shortage of information and a shortage of initiative.

The rise that made the fear credible

It is hard to recover now how seriously the world took Soviet growth, and the numbers explain the seriousness. In the first plan era, from 1928 to 1940, output grew about 5.8% a year in the Western scholarly estimates; through the 1950s, about 5.7%; through the 1960s, still above 5%. Rates like that double an economy every thirteen years or so, and they were posted by a country that had entered the century mostly illiterate and mostly rural. When Nikita Khrushchev told Western ambassadors in November 1956 that “we will bury you” (his interpreter’s rendering of a Russian idiom closer to a promise to outlive than a threat to dig), the line landed as it did because the arithmetic behind it looked, for a moment, plausible. Extrapolation is the oldest error in economics, and in the 1950s the extrapolators had real slopes to work with.

Then the slopes bent. Growth fell to 3.7% a year in the early 1970s, 2.6% in the late 1970s, and 2.0% in the first half of the 1980s, and those are the sympathetic numbers; the figure below shows the series, and the row beneath it shows something more damning than the slowdown itself.

How fast the Soviet economy really grew is itself contested terrain, and a chapter resting anything on the numbers owes the reader the spread. Three sets of books exist. The official statistical administration kept the first and most generous. Western estimates, rebuilt over decades from physical output series (the CIA’s among them, later tabulated in Gur Ofer’s survey of the field), run well below the official record and are what the figure above plots. Below both sits the reconstruction of Grigorii Khanin, a Soviet economist working half in the open, whose recalculations ran lowest of all: about half the official growth rate for the quarter century after 1960, and beneath the Western estimates in nearly every sub-period. The spread matters less than it first appears, for one reason: every set of books, including the state’s own, shows the same deceleration. By the official count growth fell from 7.8% a year in the late 1960s to 3.6% in the early 1980s; by Khanin’s, from 4.1% to 0.6%. The argument is about the height of the hill, never about the slope at the end.

Growth you can order, and growth you cannot

The shape in Figure 4.1 stops being mysterious once growth is split into its two possible sources. An economy can produce more by using more inputs, more workers, more machines, more land under plow, or by getting more output from the inputs it already uses. The first kind can be commanded. A state that controls investment can force the saving rate up, move millions from farm to factory, school them, electrify the rivers, and pour steel; these are exactly the tasks a plan can specify, count, and verify, and the early Soviet decades did them at scale. The second kind cannot be commanded, because it consists of millions of small discoveries about doing things better, and nobody can order a discovery.

The productivity row of Figure 4.1 is that distinction turned into a time series. Through 1970, total inputs grew around 4% a year while productivity added a further 1.5 points or so; after 1970 the input engine slowed as the reserves ran dry, the countryside had no more millions to send, the workforce was educated, the easy construction was built, and the productivity term, which by then had to carry the load, fell to zero and kept falling. The economy did not run out of orders to give. It ran out of growth that could be ordered. Why the second engine never fired is the rest of this chapter, and the answer has two parts.

The information problem

Begin with what a price does. The second chapter introduced coordination as an axis: markets discover prices, plans set targets. What that compresses is the daily miracle the planners had to replace. When tin becomes scarcer, for whatever reason anywhere on earth, its price rises, and every user of tin economizes and every producer stretches, each acting on knowledge of their own situation only. Nobody needs to know why tin is scarce, and nobody is asked for a report. Friedrich Hayek’s 1945 essay put the point in a form the debate never escaped: the knowledge an economy runs on is dispersed, particular, and perishable, and the price system is the machinery, he called it “a system of telecommunications,” that moves what matters about that knowledge without ever collecting it in one room.

The planned economy’s answer to the knowledge problem was paperwork, and the figure below traces what happened to one good’s worth of knowledge on its way to a decision. In a market, the signal path from a shopper to a shoe factory is two links long and runs in both directions every day. Under the plan, the same signal climbed a ministry, crossed the state planning committee’s materials balances, and descended a different ministry as a target, aggregated at every hop, months old on arrival, and stripped of exactly the particulars Hayek identified as the cargo. A factory could learn how many tons of footwear the district owed the year; it could not learn that the queue outside the shop was for the black pair in size seven.

Scale made the compression fatal rather than inconvenient. A war economy plans well because a war economy wants few things, steel, shells, bread, and wants them identically; this is why mobilization suited the machinery, and why the machinery’s admirers kept reaching for military metaphors. A consumer economy wants millions of distinct things in changing sizes, colors, qualities, and places, and the wants shift faster than any reporting cycle. The mismatch grew with every year of success: the richer and more complicated the economy the planners built, the more knowledge their next plan needed and the smaller the fraction of it their paperwork could carry. Complexity was not a headwind the system happened to meet. The system manufactured it.

The initiative problem

The second shortage would have crippled the plan even with perfect information, because knowing what to do and having a reason to do it are different resources. Consider the position of a Soviet factory director, documented in detail by Joseph Berliner’s interviews with managers who had run the system from inside (Factory and Manager in the USSR, 1957). The director’s year was ruled by a target, and the rational moves under a target regime were these. Understate your capacity in every report upward, because the plan is bargained and honesty raises your quota. Hoard labor and materials, because supply is unreliable and the target does not care what things cost, a habit the previous chapter met from the consumer’s side, in Kornai’s account of shortage as the system’s standing condition. Overfulfill by a sliver, never by a margin, because this year’s triumph is next year’s baseline, the ratchet that punished every revealed reserve. And meet the letter of the indicator rather than its intent: where output was set in tons, Alec Nove observed, the rational factory made its goods heavy, and where it was set in units, it made them small and many.

Innovation lost that game every time. Retooling a line stops it, and a stopped line misses the quarter’s target; a better product earns the same credit as the old one if the indicator counts tons; and the upside of a successful innovation belonged to the state, on the residual-claim axis the second chapter laid out, while the downside of a failed one landed on the director personally. The invention column of the previous chapter’s ledger recorded the result from the outside: filings without follow-through, certificates without products. None of this required lazy people or true believers; it required only rational people answering the incentives in front of them. The deepest reading of the initiative problem is that the system worked exactly as designed, and the design rewarded the defense of the status quo at every desk that could have changed something.

Was it inevitable?

The mechanism explains the stall. It does not, by itself, settle the chapter’s second question, and the honest answer runs through the strongest case on the other side. Robert Allen, whose reconstruction of the Soviet record (Farm to Factory, 2003; “The Rise and Decline of the Soviet Economy,” Canadian Journal of Economics, 2001) is the most serious rehabilitation in the literature, reads the same numbers differently. Before 1970, he argues, the Soviet Union was one of the world’s fastest-growing economies, and the stall that followed traces to identifiable blunders rather than to planning as such: investment poured into retooling old rust-belt plants and into remote Siberian resource extraction just as the returns to both collapsed, and defense ministries absorbed the lion’s share of the growth in research resources, trained engineers above all, starving the civilian research that might have revived productivity. In Allen’s accounting, different choices at those forks buy the system decades more growth. The mechanism this chapter describes pushes back: the blunders were not weather, they were the machinery running normally, since a system without market tests has no instrument that says stop to a doomed investment program, and a system without residual claimants has nobody paid to say it. Where Allen sees avoidable errors, the information-and-initiative reading sees the errors the system was built to make.

The reform record is the second witness, and it testifies for both sides. The planned economies knew about the stall and tried, repeatedly, to graft the missing signals on: the Kosygin reforms of 1965 tied bonuses to sales and profit indicators, Hungary’s New Economic Mechanism of 1968 went further and abolished compulsory enterprise targets, and every such experiment ran into the same wall, which was not economic but political: prices that move and managers who decide are power surrendered by the party and the ministries, and the surrender was always taken back before it could matter. Read one way, the reforms prove the stall was curable and the cure was refused, which supports Allen. Read the other way, they prove the cure was structurally unavailable, because the monopoly on decisions was not an accessory of the system that could be reformed away; it was the system. What the record does show is that exit existed: China after 1978 and Vietnam after 1986, dated on the second chapter’s timeline, re-marketized coordination under the same party structures and grew, which is evidence that falling behind was a property of the planning bundle, not a fate attached to the countries or their politics.

Where the argument goes next

Two boundaries of this chapter are drawn on purpose. Everything here assumed a state that could actually execute a plan, collect the harvest, staff the ministries, make the targets stick; where states cannot do even that, markets fail in a different and prior way, and that machinery has its own chapter in Volume III: Countries, on why some states can tax and enforce while others cannot. And the calculation debate was fought with the technology of filing cabinets; whether modern computation and machine learning reopen it, as some now argue, is a question about futures, taken up with the other futures in Chapter 10.

For the reader’s original question, the standing result is this: the great alternative failed at a nameable task, converting dispersed knowledge and private initiative into growth, and the failure was mechanical rather than moral. That verdict cuts both ways. It retires the story in which planning lost because people are wicked or lazy, and it puts a hard question to the winner, because the market’s claim to superiority rests on the very signals, prices facing competition, that the next chapters test. The next one begins with the reader staring at a modem: if competition is the mechanism that keeps prices honest, why does half the map get exactly one seller?