Tag Archives: Industrial Location

Was Stalin’s Economic Policy the Root of Nazi Germany’s Defeat?

Was Stalin Necessary for Russia’s Economic Development?

By Anton Cheremukhin (Dallas Fed), Mikhail Golosov (Princeton), Sergei Guriev (SciencesPo), Aleh Tsyvinski (Yale)

Abstract: This paper studies structural transformation of Soviet Russia in 1928-1940 from an agrarian to an industrial economy through the lens of a two-sector neoclassical growth model. We construct a large dataset that covers Soviet Russia during 1928-1940 and Tsarist Russia during 1885-1913. We use a two-sector growth model to compute sectoral TFPs as well as distortions and wedges in the capital, labor and product markets. We find that most wedges substantially increased in 1928-1935 and then fell in 1936-1940 relative to their 1885-1913 levels, while TFP remained generally below pre-WWI trends. Under the neoclassical growth model, projections of these estimated wedges imply that Stalin’s economic policies led to welfare loss of -24 percent of consumption in 1928-1940, but a +16 percent welfare gain after 1941. A representative consumer born at the start of Stalin’s policies in 1928 experiences a reduction in welfare of -1 percent of consumption, a number that does not take into account additional costs of political repression during this time period. We provide three additional counterfactuals: comparison with Japan, comparison with the New Economic Policy (NEP), and assuming alternative post-1940 growth scenarios.

URL: http://EconPapers.repec.org/RePEc:nbr:nberwo:19425

Distributed by NEP-HIS on 2013-09-28

Review by Emanuele Felice

Until the late 1950s, the era of rapid Soviet growth and of Sputnik, the main question among Western scholars was: When would the Soviet Union catch up with and overtake the U.S.?*

As Cheremukhin et al. correctly emphasize, the subject of this paper – Soviet industrialization in the 1930s – is one of the most important in economic history, and in world history: Soviet Union was the country which played by far the biggest role in the defeat of Nazi Germany, standing almost alone against the land force of the Third Reich and its allies for most of the war and causing 87% of the total Axis’ military deaths (in sharp contrast with World War I, when the Tsarist empire was defeated by a German Reich fighting on two fronts). Emerging from World War II as a superpower, the victorious Soviet Union contributed to shape the next four decades of human history, boasting among its technological achievements the first voyage of a human being to the space. At the same time and during the Stalin regime (1922-1953), the scale of (politically caused) human suffering has had few parallels in world history. Furthermore, as early as the 1930s Stalin’s rule was one of the first totalitarian regimes capable of reaching levels of oppressiveness and manipulation over society unobserved before.

For these reasons Stalin’s Soviet Union should continue to be interrogated by systematic studies. At the core of that regime was industrialization, which aimed to be the material pillar of a new «civilization» (e.g. Kotkin, 1995). Regarding its impact over policy making in the twentieth century, Stalin’s forced industrialization was a source of inspiration for both economists and politicians throughout the world: its planned, top-down, implementation was widely considered to be a successful, though harsh, strategy by some contemporaries.

Joseph Stalin (b 1878 - 1953), Leader of the Soviet Union (1922-1953)

Joseph Stalin (b 1878 – 1953), Leader of the Soviet Union (1922-1953)

And yet, we still have relatively little macro-economic evidence about the Stalinist period. The article Cheremukhin et al. aims to partially fill this gap, by providing consistent figures, some new arguments and insightful counterfactuals. It builds upon a remarkable amount of original research. First, it provides a comprehensive and coherent reconstruction of data on output, consumption, investments, foreign trade and labour force. These figures are presented separately for the agricultural and non-agricultural sectors. Data begins in the last decades of Tsarist Russia (1885-1913) and for the the Soviet Union covers the launch of the first five-year plan until the Nazi’s invasion (1928-1940).

Secondly, Cheremukhin et al. propose and elaborate a growth model for the Russian economy in those two periods (i.e. Tsarist Russian and pre-Nazi invention Soviet Union). This is a multi-sector neoclassical model, which is modified to allow for the peculiarity of the economy under scrutiny; namely, due to the institutional frictions and policies that distorted household and firm decisions, three wedges are defined, corresponding to the intratemporal between-sector distortions in capital and labor allocations and to an intertemporal distortion, and price scissors in agricultural prices (between producers and consumers) − which may also be thought of as a fourth wedge − are also introduced for the Stalin’s period.

It may be worth adding that when connecting wedges to policies, the Cheremukhin et al. appear to be adequately aware of the historical context and of the differences between a planned economy and a free-market one: for instance, the response of the Stalinist economy to a drop in agricultural output is likely to be the opposite − because of the price scissors policy which kept producer’s agricultural prices artificially low − to the predictions of a frictionless neoclassical growth model: it will probably lead to a further reallocation of labour from agriculture to industry and services and, therefore, to an additional reduction of agricultural output; such a distortion is here acknowledged and reasonably calibrated.

 “Smoke of chimneys is the breath of Soviet Russia”, early Soviet poster promoting industrialization, 1917-1921

“Smoke of chimneys is the breath of Soviet Russia”, early Soviet poster promoting industrialization, 1917-1921

Thirdly, the paper by Cheremukhin et al. further elaborates on data and models, by providing a number of counterfactuals. Comparisons are made with the Tsarist economy by extrapolating Tsarist wedges for 1885-1913 to the 1928-1940 years. Also by comparing the performance of both economies (Tsarist and Stalinist), for the years following 1940 under the assumption that World War II never happened.

Another comparison takes place with Japan, a country similar to Russia before World War I in terms of GDP levels and growth rates. Early in the twentieth century Japan suffered similar distortions as Russia but during the interwar period Japan undertook an economic transformation which provided Cheremukhin et al. an alternative scenario to both the Tsarist and the Stalin policies (the Japanese projections are based upon previous reconstructions of the Japanese macro-economic figures, which happen to be available for the same period as for Russia, 1885-1940).

Japanese assault on the entrenched Russian forces, 1904

Japanese assault on the entrenched Russian forces, 1904

And what is probably the most intriguing counterfactual, at least in actual historical terms, is yet one more alternative scenario, constructed by assuming that Lenin’s New Economic Policy or NEP (launched in 1921 and outliving Lenin until 1927) would have continued even after 1927: such a counterfactual requires elaborating a model for the NEP economy as well, but unfortunately the lack of reliable data for the years 1921 to 1927 makes the discussion for this scenario «particularly tentative». Furthermore, it is worth mentioning that two more alternative scenarios are provided for the Stalin economy based on alternative growth rates for the years 1940 to 1960 and again under the assumption that World War II never happened; and that robustness exercises are also performed (with further details provided in the appendix).

Broadly speaking, the results are not favourable to Stalin. According to Cheremukhin et al., Stalin was not necessary for Russian industrialization − neither, it could be consequently argued, to the defeat of Nazism and to the Russia’s rise to a superpower status. Actually, by 1940 the Tsarist economy would probably have reached levels of production and a structure of the economy similar to the Stalinist one, but which far less short-term human costs. This result may not be irreconcilable to Gerschenkron’s (1962) theses about substitute factor − in Russia this was the State, already exerting such a role in late Tzarist times − and the advantages of backwardness: these latter would have permitted to backward Russia, once its industrialization had been set in motion at the end of the nineteenth century, to see its distance to the industrialized West reduced by the time of World War II more than in World War I, in any case – that is, also under the Tzarist regime. It does contrast, however, with other findings from pioneering cliometric articles on the issue, such as the one by Robert Allen published almost twenty years ago, according to which Stalin’s planned system brought about rapid industrialization and even a significant increase of the standard of living (Allen, 1998). Similarly, but from a different perspective, long-run reconstructions of Soviet labour productivity tend to emphasize as a problem the slow-down in the period following post World War II, rather than the performance the 1930s (Harrison, 1998) – both Allen and Harrison are cited in this paper, but not these specific articles.

The Dnieper Hydroelectric Station under construction, South-Eastern Ukraine (the work was begun in 1927 and inaugurated in 1932)

The Dnieper Hydroelectric Station under construction, South-Eastern Ukraine (the work was begun in 1927 and inaugurated in 1932)

Now, at the core of the results by Cheremukhin et al. is the finding that, according to their estimates, total factor productivity of the USSR in the non-agricultural sector did not grow from 1928 to 1940. Maybe it is worth discussing this point in a little more detail. Is such a finding plausible? At a first sight it seems puzzling, given the technological advance of that period especially in the heavy sectors. And yet, at a closer inspection it may turn out to be entirely logical: the growth of output was a consequence of massive inflows of inputs, both machinery (capital) and labour. But all considered these were not used in a more efficient way.

In the model by Cheremukhin et al., capital and labour are computed through a Cobb-Douglas production function, with constant elasticity coefficients for labour and capital (0.7 and 0.3 respectively in the non-agricultural sector; 0.55 and 0.14 in the agricultural one, thus assuming a land’s elasticity of 0.31). The authors make a point that the new labour force entering the non-agricultural sector was largely unskilled and, often, was not even usefully employed. Actually exceeding the real needs of that sector: this politically induced distortion could hardly have increased TFP (although, under different assumptions, it could be alternatively modeled through a decreasing elasticity of labour: but the results in terms of total output would not change). This may also explain the good performance of Soviet Union during World War II, when due to manpower shortage the exceeding labour force finally could be profitably employed. The capital stock is calculated by the authors at 1937 prices, for the years 1928-1940.

Anti-Nazi propaganda poster, 1945

Anti-Nazi propaganda poster, 1945

We do not have enough information in order to judge whether a bias can be caused by the use of constant prices based on a late-year of the series. But this possible bias should lead to an underestimation of capital growth in that period  − given that quantities are probably weighted with relative prices lower in 1937 for the heavy sectors, than in 1928 − which would then produce an overestimation in the TFP growth proposed by the authors: in actual terms, therefore, the growth of TFP may be even lower than what estimated; in more general terms – and although caution is warranted for the lack of detailed figures – their results look realistic in this respect.

The most interesting finding, however, is the one relative to the NEP counterfactual. It is the most interesting because, in genuine historical terms, the Tzarist model was no longer a viable option to Stalin, while NEP’s strategy was. But of course, data for the NEP years are much more precarious and thus this counterfactual can only be a particularly tentative one. Nonetheless, the authors build two scenarios for the NEP policy: a lower-bound one, where a growth rate of TFP in manufacturing after 1928 similar to the average Tsarist 0.5% is tested; and an upper-bound one, with a growth rate of 2% similar to the one experienced by Japan in the same interwar period. In the first scenario the results for the Soviet economy would have been slightly worse, but in the second one much better. Given that the two scenarios correspond to the boundaries of the possibility frontier, we may conclude that probably, under the NEP, the performance of the Soviet economy would have been better than both the one observed under the Stalin and that predictable under the Tzar. This may confirm the view that the 1920s were somehow the “golden age” of Soviet communism, as well as the favourable assessment of Lenin’s and later of the collective Soviet leadership in that decade (although, admittedly, Lenin intended the NEP only as a temporary policy). After all, a more inclusive leadership – as opposed to the harshness of Stalinist autocracy in the 1930s, as well as to Hitler despotic conduct of war since the winter of 1941 – was also the one which helped the Red Army to win World War II.

“The victory of socialism in the USSR is guaranteed”, 1932

“The victory of socialism in the USSR is guaranteed”, 1932


Allen,  Robert C., Capital accumulation, the soft budget constraint and Soviet industrialization, in «European Review of Economic History», 1998, 2(1), pp. 1-24.

Gerschenkron, Alexander, Economic backwardness in historical perspective, Cambridge, Mass., The Belknap Press of Harvard University Press, 1962.

Harrison, Mark, Trends in Soviet Labour Productivity, 1928−85: War, postwar recovery, and slowdown, in «European Review of Economic History», 1998, 2(2), pp. 171-200.

Kotkin, Stephen, Magnetic Mountain: Stalinism as a Civilization, University of California Press, Berkeley, Los Angeles, and London, 1995.

Source of quote:
Gur Ofer (1987) “Soviet Economic Growth: 1928-1985,” Journal of Economic Literature, Vol. 25, No. 4, pp. 1767-1833 (cited in this paper, p. 2).


The postman always rings twice: measuring market access and endowment in the German Empire through postal data

It’s all in the Mail: The Economic Geography of the German Empire


Florian PLOECKL (florian.ploeckl@adelaide.edu.au) University of Adelaide


Information exchange is a necessary prerequisite for economic exchange over space. This relationship implies that information exchange data corresponds to the location of economic activity and therefore also of population. Building on this relationship we use postal data to analyse the spatial structure of the population distribution in the German Empire of 1871. In particular we utilize local volume data of a number of postal information transmission services and a New Economic Geography model to create two index measures, Information Intensity and Amenity. These variables respectively influence the two mechanisms behind the urban population distribution, namely agglomeration forces and location endowments. By testing the influence of actual location characteristics on these indices we identify which location factors mattered for the population distribution and show that a number of characteristics worked through both mechanisms. The model is then used to determine counterfactual population distributions, which demonstrate the relative importance of particular factors, most notably the railroad whose removal shows a 34% lower urban population. A data set of large locations for the years 1877 to 1895 shows that market access increases drove the magnitude of the increase in urban population, while endowment changes shaped their relative pattern.

URL: http://econpapers.repec.org/scripts/search.pf?aus=Florian%20Ploeckl

Review by Anna Missiaia

This paper was distributed by NEP-HIS on 2015-04-11. The work by Florian Ploeckl lays in the expanding branch of historical economic geography, which looks at, broadly speaking, the role of geographical factors in regional development. In particular, the author looks at the effect of actual location characteristics on the information exchange and endowment (calculated through two indices) in the German Empire between 1877 and 1895. The empirical model used in the paper uses the indices that describe market access and endowments effects as dependent variables and test which geographic, institutional and cultural characteristics shaped them.


Otto Von Bismarck (1815-1898), First Chancellor of Germany

The paper relies on detailed data on the postal system to measure the diffusion of information across 41 districts in the Empire. The creation, after the German unification, of a common and homogeneous postal system with the same rates across locations allows the author to use postal flows as proxy for “information intensity”.   This measure tells us the level of information exchange for each location considered. The author meticulously identifies business related correspondence for each location by selecting specific types of mail for the analysis and relating it to the general mail. The empirical exercise appears very well engineered and executed.


 Kaiserliches Postamt sign, about 1900

The next step is to relate this indirect measure of economic activity to the access to markets for any given location. Following a well-established practice in the discipline, Ploeckl relies on the concept of market potential. Market potential is a measure of the centrality of a given location and can be constructed in two main ways. The first option, when trade volumes among locations are available, is a gravity model. This is the method used nowadays by economic geographers but also economic historians lucky enough to have access to internal trade flows (see Redding and Venables, 2004 for the former and Wolf, 2007 for the latter). This method basically looks at actual levels of trade and derives from these the potential for a location. The second option, used when trade flows are unknown, relies on the methodology proposed by Harris (1954) which uses GDP of the locations weighted by the inverse of distance to calculate the potential levels of trade across the locations given the size of their economies. Examples of this estimation procedure are Crafts (2005), Schulze (2007) and more recently Crafts and Klein (2012). This paper approaches the issue in a very innovative way, escaping the dichotomy that normally characterizes the calculation of market potential. As we understand, neither trade volumes nor regional GDP are available for Germany in this period. Therefore the author relies on the assumptions that “market potential translates in commercial transactions” and that “each transaction causes the same amount of mail” to claim that the measure from step 1 is able to capture the access to markets of the locations. The first assumption is shared with the broader group of scholars that use gravity models for market access and is perfectly reasonable when dealing with trade volumes. The use of quantitative evidence on correspondence to proxy for economic activity is not new in the literature: Crafts (1983) provided GDP estimates based, among the others, on letters per capita. The method proved to be quite misleading applied for instance to the Italian case (Esposto, 1997). Because of the indirect measure used in the paper, the relationship between information flows, market potential and actual exchange is of course much more questionable. However, it must be pointed out that the empirical effort in this paper makes its use of postal data more convincing compared to other more dated attempts.

The paper is also very interesting in that it finds a way to split market access into firm market access and consumer market access. This is a crucial point in the analysis of market forces as the two measures could well be following very different trajectories.

The last step is to calculate an endowment index based on real wages and the trade cost matrix across locations (the details on the methodology are explained in Ploeckl, 2012).

The bottom line results of the paper are that important factors like railroads and coal were important in the location of population (and therefore economic activity) both through the market channel and the endowment channel. The impact of these channels is quantified through counterfactual analysis, leading for instance to a 30% impact of the removal of the railroads on the population level.

Summing up, this paper contributes to a very hot debate on the determinants of the location of economic activity. It does so by finding an innovative empirical method to overcome the chronic lack of data in historical research. The limitations of these indirect methods should not, as usual, be neglected. However, the exercise appears more than reasonable and some features of these papers could find fruitful applications in a variety of other lines of research in historical economic geography.


Crafts, N., 1983, Gross National Product in Europe 1970-1910: Some New Estimates, Explorations in Economic History, Vol. 20, No. 4, 387-401.

Crafts, N., 2005, Market Potential in British regions, 1871-1931, Regional Studies, Vol. 39, pp. 1159-1166.

Esposto, A., 1997, Estimate Regional Per Capita Income: Italy, 1861-1914, Journal of European Economic History, Vol. 26, No. 3, p.585-604.

Ploeckl, F., 2012, Endowments and Market Access; the Size of Towns in Historical Perspective: Saxony 1550-1834, Vol. 42, p. 607-618.

Redding, S. and A. Venables, 2004, Economic Geography and International Inequality, Journal of International Economics, Vol. 62, No. 1, pp. 53-82.

Schulze, M. S., 2007, Regional Income Dispersion and Market Potential in the Late Nineteenth Century Hapsburg Empire, LSE Working Papers no. 106/07.

Wolf, N., 2007, Endowments vs. Market Potential: What Explains the Relocation of Industry after the Polish Reunification in 1918?, Explorations in Economic History, Vol. 44, (2007), 22-42.



Industrial Location and Path Dependency during the British Industrial Revolution

The Location of the UK Cotton Textiles Industry in 1838: a Quantitative Analysis


Nicholas CRAFTS (n.crafts@warwick.ac.uk)  University of Warwick

Nikolaus WOLF (nikolaus.wolf@wiwi.hu-berlin.de) Humboldt University


We examine the geography of cotton textiles in Britain in 1838 to test claims about why the industry came to be so heavily concentrated in Lancashire. Our analysis considers both first and second nature aspects of geography including the availability of water power, humidity, coal prices, market access and sunk costs. We show that some of these characteristics have substantial explanatory power. Moreover, we exploit the change from water to steam power to show that the persistent effect of first nature characteristics on industry location can be explained by a combination of sunk costs and agglomeration effects.

URL:  http://econpapers.repec.org/paper/heswpaper/0045.htm

Review by Anna Missiaia

This paper was distributed by NEP-HIS on 2013-09-13. Nick Crafts and Nikolaus Wolf, who have both provided significant contributions to the literature on industrial location, engage here in the analysis of the UK cotton textile industry. In particular cotton production during the Industrial Revolution was heavily concentrated in Lancashire, the region just north of Manchester. Moreover, this concentration persisted over the 19th century. The two authors are therefore interested in explaining both the original concentration and its persistence throughout time.

The paper presents a solid statistical work. The dataset comprises 1823 cotton mills and covers 148 locations in all of the UK. To explain the employment in textiles across these locations, the authors use information on coal prices, geography, climate and access to markets. All these measures are specific to each location and region fixed effects are included to avoid the omitted variable bias. Firms are assumed to be profit maximizers in their location decisions. The factors that influence the location decisions are considered into two broad groups: the “first nature” characteristics, which are considered exogenous to earlier location choices (i.e. climate) and the “second nature” characteristics which are endogenous (i.e. access to markets). Crafts and Wolf separate these two elements because they are interested in identifying the case of location choices that eventually modify the characteristic of the location itself (in particular market access).

They reckon that access to market can be so important that the cotton industry remained in a location in spite of higher variable costs because these were outweighed by better access to markets. Another way in which past choices can affect current choices is through sunk costs: once an investment in energy production was made in one location, it could hardly be moved to another location. However, it could often be adapted to new technology. The example provided is the switch from water to steam power, during which waterwheels were adapted to steam. This allowed some location that at that point did not have a fist nature advantage, to maintain their industries through path dependence.


Hibert, Platt & Son’s cotton machines.  Illustrated London News, 23 August 1851. 

On the empirical side, the paper uses a Poisson model to estimate the expected number of cotton mills and employed persons for each location as a function of the characteristics of the locations. The main findings are that water power production and number of patents registered increase the likelihood of location; coal prices had a surprisingly weak effect; agglomeration forces had a positive effect on the number of persons employed but a negative effect of the average size of the mills, suggesting that cotton industry was organized in a network of small specialized mills. This is confirmed by anecdotal evidence on Lancashire’s cotton industry. The authors also provide several robustness checks on their data to support their claims.

The paper then moves on to discussing why Lancashire achieved such a high concentration of cotton industries. The two authors explain that the high concentration was the result of a combination of first and second nature geography. To prove this statistically, Crafts and Wolf perform a counterfactual analysis in which they replace each characteristic with the average value of the UK and then impose a 10% change in the variables to compare their effect individually. Doing so, they come up with a “conversion table” that tells us what variation of the x variable is needed to offset a 10% variation of the y variable. The main results are that the location choices were driven both by first nature characteristics such as water power and second nature characteristics such as market access. The persistence of the location is liked to sunk costs and agglomeration economies, which allow some regions to maintain their industries in spite of the original advantage being vanished.


An image of the Lewis Textile Museum in Blackburn, Lancashire.

To conclude, the contributions of this paper are several. First, it makes for the first time use of statistical techniques to explain the location of cotton industries, which were crucial during the British Industrial revolution. Doing so, it contributes to the wider debate about the determinants of the location of industries in general, proposing a methodology based on counterfactuals which allows to compare the relative strenghts of the different factors. Finally, the paper adresses the always ‘hot issue’ of path dependency in location choices, which is faced by any researcher in this particular field. The next step in the research, to which we look forward,  will be to estimate the model as a panel in order to cast more light on the persistence of location through time.