Robert Allen's monograph is a work of serious and deep scholarship that spans a century and a half of Russian economic history. Its aim is to re-evaluate the Soviet economic system under Stalin and his successors, correctly characterized by Joel Mokyr (on the back cover) as “the greatest failed experiment in the history of the human race.” Allen's work synthesizes a vast literature, not only on Russia but on economic history and economic development generally, to which he adds his own original research on comparative crop yields, Soviet living standards, and policy simulations of the 1930s. For those who wish to disagree with his findings, he provides careful documentation and explanations of methodology.

Allen's major conclusions are that Russia's (the Soviet Union's) growth was accelerated by Stalin's forced industrialization, which permitted Russia, along with Japan, to make a rare breakthrough from relative poverty to industrialization. Although imperial Russia had made considerable progress, a noncommunist Russia would not have fared nearly as well. Allen's simulations show that the industrialization success was explained by Stalin's decision to one-sidedly invest in heavy industry and by soft-budget constraints that promoted full employment in an economy with surplus rural labor. His recalculations of Soviet living standards show that the Soviet population, especially urban workers, shared in the benefits of rapid growth, thereby contradicting the famous calculation of Abram Bergson (The Real National Income of Soviet Russia Since 1928. Cambridge, MA: Harvard, 1951, p. 251), which shows stagnant or declining real wages and living standards in the 1930s. Collectivization does not perform well in Allen's simulations due to the resulting loss of livestock and output in its early years. A noncollectivized Soviet system that preserved Stalin's investment priorities and soft budgets would have yielded the same or better results. Allen's take on the Soviet climacteric after the 1960s differs from most other explanations: Although factors such as diminishing returns, and disincentives to investment likely played a role, it was poor investment decision making—namely the emphasis on modernization of old facilities—that was the primary culprit. Unlike those who blame declining Soviet performance on a collapse of discipline, Allen argues (and I tend to agree) that the problem was more that bad orders were being obeyed, not the disobedience of good orders. Allen has set, with his provocative set of propositions, a broad research agenda. Under normal circumstances and given the importance of the subject, a number of researchers would enter the fray to test and examine Allen's methodology and assumptions. I fear that this will not happen, given the limited number of younger scholars still active in the field and the language requirements for serious research.

Much of Allen's research was written prior to the first fruits of economic research in the formerly secret Soviet state and party archives, most notably the works of R. W. Davies and Stephen Wheatcroft (The Years of Hunger: Soviet Agriculture, 1931–1933. Houndsmills: Palgrave, 2004), Paul Gregory (The Political Economy of Stalinism: London: Cambridge, 2004), and Paul Gregory and Valery Lazarev (eds.) (The Economics of Forced Labor: The Soviet Gulag. Stanford, CA: Hoover Press, 2004). Allen himself can perhaps himself reexamine his own conclusions over the next five years to see how they hold up to the emerging findings of archival research.

My own substantive remarks relate to this “open” research agenda raised by Allen's provocative research, some of which can be framed in the light of this archival research.

Allen's finding of substantial growth of living standards, especially those of urban workers, between 1928 and 1937, is generally supportive of my own conclusion (Political Economy of Stalinism, chap. 4) that Stalin and the Politburo were acutely sensitive to the effects of real wages on work effort and had to balance increased investment against the reduced effort associated with falling living standards. This conclusion is buttressed by the investment cycle of the mid and late 1930s, which to a degree is the mirror image of consumption statistics. However, given the extreme care of Bergson (and his colleague Janet Chapman), it is difficult for me to decide which estimate of per capita consumption is “better.” Part of Allen's argument relates to his preference for index numbers that average 1928 and 1937 price weights, while Bergson cites primarily 1937 price weights—a dispute that involves us in the intricacies of index number theory, but I had felt earlier that Bergson's case for “later” prices was convincing in that it measured the relative ability to produce the “modern” output mix in the two periods. The actual calculational differences between Bergson and Allen relate to the relative weights of state-fixed (and low) agricultural procurement prices versus “free market” farm prices. I fear we will never get a good fix on this issue because of the growing and extreme repression (including death and gulag sentences) of “speculators”—hardly a conducive climate for the gathering of statistics on sales to organizations other than the state. Allen's firmest evidence is anthropometric; namely, the rise in life expectancy in the late 1930s, but this could have been a consequence of the educational revolution as well.

The heart of Allen's book comprises his simulations of various policy alternatives open to the Soviet leadership in 1928, where his basic policy choices are collectivization versus noncollectivized agriculture, hard or soft employment constraints, and investment allocations to light or heavy industry. Allen explains his simulation models clearly for those wishing to perform sensitivity analysis, and he clearly explains his choice of parameters, many of which are drawn from the literature. Given his intense interest in investment strategy, most simulations are reported with different investment allocations between light and heavy industry. What Allen reports are his “best” estimates without much sensitivity analysis. Assumptions (from Appendix B) that caught my eye were that labor was allocated to light and heavy industry to equalize marginal rates of return (an assumption I would question given the administrative allocation of labor), that consumer goods markets were “cleared” by prices (if I understood correctly). Moreover, it would be enlightening to see sensitivity analysis on agricultural supply elasticities. I do not know whether Allen's basic conclusions could be altered by such experiments, and I fear that other researchers will not take up this challenge.

Allen makes use of Ragnar Nurkse's and W. Arthur Lewis's famed models of surplus agricultural labor, and his simulations suggest that a key ingredient of the success of Soviet industrialization was its ability through administrative methods and willingness to take enterprise losses on a massive scale to redeploy agricultural labor from the countryside where its marginal product was zero to industry and construction where it was, at least, positive. If a “capitalist” labor market policy had been used, simulations show that the transfer would have been much slower and successes more muted. My own research (Political Economy of Stalinism, chapter 4) suggests that Allen's simulations overlook a negative side effect of soft budgets that proved to be the major labor headache of the Soviet leadership. With soft budgets and overfull employment, labor discipline deteriorated, labor turnover skyrocketed, and labor productivity collapsed. Official concern is reflected in the fact that repression was applied to the entire industrial labor force in 1940 and thereafter with laws that made absences, tardiness, and job changes criminal offenses. Given the extreme concern (and even panic) of the Soviet leadership over the discipline issue, this negative effect would somehow need to be grafted to Allen's employment simulations, but it is unclear how.

Allen concludes that the partial effect of collectivization was, at best, neutral. He obtains this result by weighing the loss of livestock and grain output during the first three years of collectivization against the increase in grain marketings. Allen also recognizes the negative effects of collectivization in the form of the famine of 1932/33 that cost 5.5 to 6.5 million lives, the deportations and camp sentences of the most able farmers, and the organizational disruption of agriculture. Allen, therefore, can be interpreted as saying that Stalinist industrialization without collectivization would have been the preferred alternative. It is hard, however, for me to envision Stalin's acceptance of a political solution that failed to impose strict Soviet rule in the countryside or to extract surpluses by terror. I realize that Allen has taken upon himself the task of economic policy alternatives, but in this case economics and politics are too closely intertwined. In my view, the imposition of command in both industry and agriculture imposed an extreme price on Soviet society that must somehow be weighed against any economic successes. Allen dismisses the fractions of labor force in camps (5 percent of industrial labor and 2 percent of the total labor force) as “not large enough to be decisive” (p. 108), but the stock of forced labor at any one point in time is not the appropriate measure of the economic and social effects of repression. According to the forthcoming seven-volume History of the Stalin Gulag (Istoriia Stalinskogo Gulaga, introduction. Moscow, Rosspen, 2004) 15 percent of the Soviet adult population had been in prison at the time of Stalin's death in March of 1953. It is hard to conclude that the disaffection associated with such extreme proportions did not have disastrous economic and social effects.

Taken in total, Allen's monograph could be interpreted as saying that relatively good economic decision making of the 1930s, backed by the clout of command and repression, allowed Russia to join the small club of advanced industrial nations but that this success was followed by bad decision making that resulted in the eventual collapse of the system. He interestingly argues that the collapse was not caused by the “failure of managers to act in accord with the plans. On the contrary the plans were implemented; the problem was that they did not make sense. The great strength of Soviet socialism was that great changes could be wrought by directives from the top…. By the 1970s, the ratio of good decisions was falling” (p. 211). I find myself in agreement with Allen's conclusions, but he fails to answer the reasons for “the failure of imagination at the top” (p. 211). In my view, the failure was explained by F. A. Hayek and Ludwig Mises in the course of the socialist controversy of the 1920s through the 1940s. Lacking markets to provide relative valuations, Soviet leaders had no checks and balances to tell them their decisions were consistently wrong. Their mistakes were therefore not the result of human error-different Soviet leaders would eventually have made the same disastrous mistakes-they were the result of the economic system.

My final point relates to Allen's discussion of my own work on the economic growth of the prerevolutionary Russian economy (Paul Gregory, Russian National Income 1885–1913. Cambridge: Cambridge University Press, 1982), the basic results of which he uses in his own calculations. Allen correctly characterizes my own work as belonging to the “optimistic” school, for it shows that Russian economic growth was rapid by the standards of the late nineteenth and early twentieth centuries. In this book I speculated that a noncommunist future would have produced a contemporary Russian economy not that far removed in affluence from its immediate European neighbors. Allen disagrees with this speculation and demonstrates that if Russia's per capita growth rate for the 1885–1913 period were projected forward to the end of the Soviet period, Russia's GDP per capita would have been only 78 percent of the USSR's actual figure (p. 33). Only if a capitalist Russia's growth had accelerated after World War I as did Germany's could my speculative case be made. As Allen points out Russia's prerevolutionary economic institutions left much to be desired (pp. 14–15). The political and institutional history of a noncommunist Russia can not be modeled, but if Russia had approached twentieth-century European institutions, its economic performance would have improved relative to its past performance. Allen emphasizes the deleterious impact of the Great Depression on a noncommunist Russia's agriculture, but contemporary living standards of affluent economies with competitive agricultures (Australia, New Zealand, Canada, and to some degree the United States) are today unaffected by the remote events of the late 1920s and early 1930s.

Even if we accept Allen's calculation that noncommunist Russia of 1989 would be 22 percent poorer than its Soviet counterpart, we could still question whether this difference was worth it. Irwin Collier (“The Measurement and Interpretation of Real Consumption and Purchasing Power Parity for a Quantity Constrained Economy: The Case of East and West Germany,” Economica 56 [February 1989]: 109–20, and “The Welfare Standard and Soviet Consumers,” Comparative Economics Studies, forthcoming June 2005) has shown that consumers in planned socialist economies would sacrifice 12–15 percent of their income to have a free market in consumer goods. Moreover, in the course of transition to a market economy, the disappearance of whole sectors of the planned economy revealed their inability to pass any type of market test in a market setting. As welfare-based measures of Russian GDP replaced Bergsonian production-potential estimates, the size of the economy shrank to a small fraction of its previous self. A noncommunist Russia would, in my opinion, not have suffered the massive waves of repression and terror of the Stalin years, which, in my view (Gregory, Political Economy of Stalinism, chap. 3) were not a historical accident of Stalin's victory in the power struggle, but an organic feature of socialist planning as argued by F. A. Hayek (The Road to Serfdom. Chicago: University of Chicago Press, 1944) 60 years ago.