The spring quarter at Chicago starts in a week. I will be teaching a course on the political economy of communism and the postcommunist transition. I love this class, which I taught at Wisconsin for many years, and not just because it is an opportunity to subject a captive audience to my repertoire of Soviet-era jokes. State socialism was the great social experiment of the twentieth century. Understanding why it failed, and understanding why the transition from state socialism was not always successful, teaches us much about politics and economics.
We begin the quarter with what János Kornai calls the “classical socialist system”: the central features of political and economic governance in the Soviet Union under both Stalin and Brezhnev, and also in many East European countries between 1945 and 1989. We read Francis Spufford’s superlative novel Red Plenty, which presents in the most readable fashion the underlying reasons why repeated attempts to reform the Soviet economy failed. We dig into those incentive problems, many of which have resonance in other forms of governance. And we learn of the famine and terror that accompanied a utopian vision that would not be achieved.
You see, the Soviet experiment was one of the world’s first attempts at structural transformation—at moving from an inefficient agrarian economy to an urbanized, industrialized economy. In a few short years the Soviet cityscape was populated with factories and smokestacks; Soviet industrial power helped to win the war. To make an omelette, you have to break a few eggs—or so the many copycats of the Soviet experiment undoubtedly told themselves as they launched their own tragic transformations.
This view receded into memory as evidence of Stalin’s horrors became impossible to ignore and as Soviet economic growth slowed in the 1970s and 1980s. Recently, however, there has been renewed debate about what Stalin accomplished. Two works stand out. Robert Allen’s Farm to Factory: A Reinterpretation of the Soviet Industrial Revolution recasts the Soviet Union as one of the world’s most successful developing countries. Anton Cheremukhin, Mikhail Golosov, Sergei Guriev, and Aleh Tsyvinski present a far less positive view in “The Industrialization and Economic Development of Russia through the Lens of a Neoclassical Growth Model,” published in the Review of Economic Studies.
The stakes of this debate—for how we understand history, and for what sort of policies developing countries might still be willing to adopt—are important. To help Broadstreet readers understand what is at issue, I spoke with Sergei Guriev. (Sergei, for those who don’t know, is a prominent Russian economist, currently at Sciences Po, who previously was rector of the New Economic School in Moscow and for several years chief economist at the EBRD.)
Scott: Sergei, welcome to Broadstreet! I have a number of questions about your paper, but first, why don’t you tell us a bit about what you and your coauthors do.
Sergei: Thanks! In this paper, we build and calibrate a two-sector (agriculture vs. non-agriculture) macroeconomic model of a market economy with distortions. The Tsarist economy was a market economy, but it was not a frictionless market economy. Explicitly modeling and estimating the various market frictions and barriers to reallocation from farm to factory is therefore crucial for understanding the performance—and underperformance—of the Tsarist economy. We calibrate this model on data from 1885 to 1913 and then extrapolate Tsarist economic performance for decades to come. Our analysis helps us understand the main reasons for Russia’s pre-1913 inefficiencies. We show that these were driven by the lack of competition in the industrial sector; we refer to extensive historical evidence consistent with this analysis.
We then use the same approach to understand Stalin’s industrialization from 1928 to 1940. Stalin’s economy was not a market economy, but our model helps to estimate the kind of distortions a market economy would have had to have in order to replicate Soviet economic performance. We can therefore directly compare not only the aggregate performance of the Tsarist and Stalinist economies, but also unpack the specific reasons why these differed. We show that while Stalin’s industrialization was brutally effective in moving labor from farm to factory, it greatly undermined productivity growth in both agriculture and industry, so on balance it only slightly outperformed the Tsarist trend.
Scott: You are not the first to investigate Soviet industrialization. Robert Allen, for example, notably argued that investment in heavy industry under Stalin increased growth and living standards. How is what you do different?
Sergei: Our analysis is based on the state-of-the-art methodology used in modern analyses of the macroeconomic impact of policies and market distortions. This approach was pioneered by Cole and Ohanian (2002, 2004), Chari, Kehoe and McGrattan (2007), and Hayashi and Prescott (2008) for analyses of the U.S., UK and Japanese economies. Our paper is the first to use this approach for a non-market economy. The advantage of this approach is that distortions (both in Soviet and Tsarist times) reflect policies; thus, given the policies we can project the performance of the economy. Furthermore, the methodology allows us to carry out counterfactual analyses—for example, what would have happened to the Tsarist economy if it had different frictions.
In his 2003 book Farm to Factory, Robert Allen used a more traditional simulation model. In many ways, his model is more detailed. For example, he has three sectors: agriculture, production of producer goods, and production of consumer goods; he also allows for centralized vs. decentralized procurement of agricultural production and so forth. However, as is usually done in such traditional models, certain key economic variables (for example, the allocation of capital between sectors or rural-urban mobility) are fixed by the modeler rather than determined endogenously (as in our model).
Like Allen, we find that the Tsarist economy was inefficient. Our analysis helps identify the main source of the inefficiency, which is entry barriers and monopoly power in the nonagricultural sector. Furthermore, similarly to Allen’s work, we find that the main contribution of Soviet policies to industrialization and growth was the massive movement of both capital and labour from farm to factory. Unlike Allen, however, we show that Soviet industrialization resulted in significant underperformance of both agricultural and industrial productivities relative to the Tsarist trend; this explains why our analysis finds that the Soviet economy outperformed the Tsarist counterfactual only slightly. We agree with Allen’s argument that collectivization was very costly and that the continuation of NEP [Scott: Lenin’s New Economic Policy, which reintroduced elements of a market economy after the Russian Civil War] without collectivization would have substantially outperformed the actual scenario. (We do not report our analysis of NEP in the published paper, but it is presented in the working paper that preceded it.)
Scott: The economy the Bolsheviks inherited was overwhelmingly agrarian. At least since Gerschenkron, scholars have argued that the peasant commune, with the limited labor mobility it implied, was to blame. Is that what you find?
Sergei: We discuss this issue and find evidence that the wage differential between farm and factory was substantial: urban wages were twice as high as rural ones from 1885 to 1913. This may indeed be explained by huge barriers to labor mobility from farm to factory, which can be traced back to the role of the commune. However, it may also be explained by the costs of obtaining the skills necessary for working in factories—or by other barriers to mobility. One of the disadvantages of our study is that we do not model the role of skills in pre-1917 industrialization. We have not been able to find reliable data on human capital for that period; I believe this is an important question for future research. Anyway, even if we attribute the full magnitude of mobility barriers to the commune, it is still less important quantitatively than “monopoly capitalism”: the entry barriers and market power in the industrial sector.
Scott: Stalin forced industrialization by coercively redirecting resources from the rural, agrarian economy to the urban, manufacturing economy. The human toll of these policies—mostly centrally, collectivization—was enormous, with millions killed during the Great Famine of 1932–33. But it was arguably a sort of Big Push, a top-down reallocation of resources that could not have been accomplished otherwise. Did Soviet productivity increase as a result?
Sergei: If industry is substantially more productive than agriculture then, indeed, a reallocation of labor from farm to factory would per se greatly increase the economy’s aggregate productivity. At the beginning of Soviet industrialization, 87 percent of Soviet employment was in agriculture, but agriculture accounted only for 48 percent of GDP. A simple calculation suggests that moving 20 percent of labor from farm to factory would have increased GDP by 69 percent.1 One therefore can agree with Acemoglu and Robinson’s statement in Why Nations Fail that there was “huge…economic potential from reallocating…labor from agriculture to industry.”
However, these calculations presume that productivities in both agriculture and industry would remain unchanged. Our analysis shows that this assumption is wrong: Stalin’s collectivization and industrialization greatly undermined both agricultural and industrial productivity.
Scott: One of the interesting ideas of this project (and also Allen’s work) is to explore different counterfactuals. What if, for example, the Bolsheviks had not seized power and the Tsarist regime had somehow managed to reform? What would have happened to the Russian economy?
Sergei: This is indeed a major advantage of our approach: our methodology allows us to carry out counterfactual analysis in a credible way. What would have happened to the Russian economy if it had Japanese policies and market frictions? How would Mao’s industrialization have proceeded if he had followed Stalin’s policies? We have answered these and some other questions in our papers on Soviet and Chinese industrialization. As to your particular question, in the first of these papers we find that if there had been no entry barriers or monopoly power in the non-agricultural sector, then, by 1940, GDP per capita in the Tsarist economy would have been about 50 percent higher than in the actual Soviet data.
Scott: Circling around some of the same issues, the working-paper version of this article asked (echoing Alec Nové), “Was Stalin Necessary for Russia’s Economic Development?” Was he?
Sergei: The answer is certainly “no.” Even without accounting for the huge human and social costs of Stalin’s policies, we find that Stalin’s economic performance only slightly outperformed the inefficient Tsarist trend. Both significantly lagged behind Japan’s performance. Japan is the most natural contemporaneous comparator for Russia and the Soviet Union—as mentioned by Millar (1970), who discussed Alec Nové’s thesis. Indeed, over the period from 1885 to 1913, the Russian Empire and Japan had similar levels and dynamics of GDP per capita. Before World War I, Japan also had substantial distortions and barriers to allocation; we show that these distortions were similar to those of Tsarist Russia. However, in the interwar period, Japan managed to reduce those distortions and to accelerate industrial productivity growth. Our quantitative analysis suggests that a Russia with Japan’s policies would have substantially outperformed the actual dynamics of Soviet GDP.
Scott: Stepping back a bit, there is a lot of interest among young scholars in Imperial and Soviet economic history and historical political economy. Did writing this paper highlight for you any questions about the Russian and Soviet economy that have yet to be answered?
Sergei: Andrei Markevich, Ekaterina Zhuravskaya, and I are now working on a survey of recent research on Russian and Soviet economic history that has been commissioned by the Journal of Economic Literature. In the last 15–20 years, this field has grown substantially in both quantity and quality, resulting in many publications not only in economic-history but also in general economics journals. Yet, even with this exciting growth, we see many unanswered questions. As we are not finished with our review, let me refrain for the moment from offering a list of those questions. The complete draft—hopefully available in a few months—will certainly outline an agenda for future research.
- [20 * (1 – 0.48) / (1 – 0.87)] – [20 * 0.48 / 0.87].