Allocation Under Dictatorship Research in Stalin's Archives - Paul Gregory

sp05_Article 2 8/18/05 9:39 AM Page 721
Journal of Economic Literature

Vol. XLIII (September 2005), pp. 721–761
Allocation under Dictatorship:

Research in Stalin’s Archives
Paul Gregory and Mark Harrison∗
We survey recent research on the Soviet economy in the state, party, and military
archives of the Stalin era. The archives have provided rich new evidence on the eco-
nomic arrangements of a command system under a powerful dictator including
Stalin’s role in the making of the economic system and economic policy, Stalin’s accu-
mulation objectives and the constraints that limited his power to achieve them, the
limits to administrative allocation, the information flows and incentives that governed
the behavior of economic managers, the scope and significance of corruption and
market-oriented behavior, and the prospects for economic reform.
1. Introduction: A Well-Documented These were documents never intended for

Dictatorship publication or scholarly research: secret
plans, reports, minutes, decisions, appeals,
T he Soviet planned economy was the
most important socioeconomic experi-
ment of the twentieth century, and it under-
and the official and private correspondence
of citizens from the highest authorities in the
Kremlin to the humblest provincial petition-
pinned the century’s most elaborate and
er. In Russia, these records are now held by
durable dictatorship. It was a major obstacle
the Federal Archival Service.1 The scale and
to the study of this experience, however, that
scope of its holdings are vast and intimidat-
the Soviet state also took secrecy to an
ing, running to hundreds of millions of files.
extreme.
The Soviet state intervened everywhere,
When the Soviet Union collapsed in 1991,
many historical secrets were laid bare.
1
The main federal archives of interest to economists
are the State Archive of the Russian Federation (GARF);
∗ the Russian State Economic Archive (RGAE); the
Gregory: Professor of Economics, University of
Houston, and Research Fellow, Hoover Institution on Russian State Military Archive (RGVA); and the Russian
War, Revolution, and Peace, Stanford University. State Archive of Social and Political History (RGASPI)
Harrison: Professor of Economics, University of Warwick, and Russian State Archive of Contemporary History
Distinguished Visiting Fellow, Hoover Institution, and (RGANI) which together hold the records of the Soviet
Senior Research Fellow, Centre for Russian and East communist party. The Archive of the President of the
European Studies, University of Birmingham. The authors Russian Federation (APRF) and the Central Archive of
thank Eugenia Belova, Valery Lazarev, Roger Sherman, the Federal Security Service of Russia (TsAFSBR) would
and the referees for their comments, and the Hoover also be of great interest were they not largely closed to
Institution and its archives department, the National outsiders. A gateway to web-based information on for-
Science Foundation, the Leverhulme Trust, and the mer Soviet archives is provided under the Political
British Academy for financial and other support of their Economy Research in Soviet Archives website at
research. http://www.warwick.ac.uk/go/sovietarchives/archives.
721
722 Journal of Economic Literature, Vol. XLIII (September 2005)
standing guard over state property, public This survey summarizes the first fruits of
morality, and most aspects of social life; its archival research on the political economy of
recording of decisions and outcomes was the Soviet Union in the period of “classical
meticulous. While political power and eco- socialism” (János Kornai 1992) that ended
nomic organization always rested on a with Stalin’s death in March 1953. We cover
bedrock of informal relationships, the degree topics of interest to comparative economists,
to which the exercise of power was expressed institutional economists, political econo-
in writing is nonetheless staggering. There mists, and new economic historians. As for
was no question of Stalin ever avoiding for- findings of specific interest to country spe-
mal association with difficult decisions or cialists on Russia and the former Soviet
passing them along by word of mouth. Union, we discuss them only in passing.6
Although Putin’s Russia has remained In making this survey, we are well aware of
protective of Soviet secrets of the 1960s and the debt we owe to previous generations of
since, the opening of the archives has stim- western scholarship. Despite the obstacles
ulated a true revolution in the history of the of secrecy and censorship, the quantity of
Stalinist state.2 Many volumes of documents this scholarship was large and its quality high
have been published to aid those who are enough to warrant many publications in top
themselves unable to work in the archives.3 economics journals.7 As a fresh generation
While some sensitive documents of the has grappled with new evidence, there has
period remain classified in the archives of inevitably been a certain amount of rein-
the president and the security and military venting the wheel; it has been one of our
agencies, even the most secret institu- tasks to try to discern what is genuinely new.
tions—the labor camps and the defense Following the organization of our paper, we
industry—are now being massively docu- summarize this as follows.
mented by new collections.4 Significant Part 2 deals with the economic organiza-
microfilm holdings are also available outside tion of dictatorship in a hierarchical com-
Russia, for example in the Hoover mand system. Before the archival revolution,
Institution.5 scholars had almost no direct evidence of the
political economy of centralized power; we
saw mainly what reached the public or
2
Among the landmarks of this revolution see, in partic- touched the lowest levels of economic
ular, Stephen G. Wheatcroft and R. W. Davies (1985),
Davies (1989a, 1996), Oleg Khlevnyuk (1993, 1996), E. A. administration. In politics, Kremlinologists
Rees (1997a, 2001, 2004), J. Arch Getty and Oleg Naumov made an educated guess at the shifting influ-
(1999), Yoram Gorlizki (2001a, 2002), Davies and ence of moderate and radical factions; econ-
Wheatcroft (2004), and Gorlizki and Khlevnyuk (2004).
3
Khlevnyuk, Kvashonkin, Liudmilla P. Kosheleva, and omists tended to presume a powerful role
Larisa A. Rogovaia (1995); Lars Lih, Naumov, and for technocratic central planners. We had lit-
Khlevnyuk (1995); A. Berelovich and V. P. Danilov (1998); tle idea of what Stalin really wanted or what
Berelovich (2000); Khlevnyuk, Devis [Davies], Kosheleva,
Ris [Rees], and Rogovaia (2001); Khlevnyuk, Gorlitskii constrained his ability to get it.
[Gorlizki], Kosheleva, Miniuk, Prozumenshchikov,
6
Rogovaia, and Somonova (2002); Davies, Khlevnyuk, Rees, R. W. Davies, whose own work towers over the field,
Kosheleva, and Rogovaia (2003). These include Stalin’s and others have surveyed the major findings from the
correspondence, secret police reports, state planning com- standpoint of Russian history (Davies 1989b, 1997, 2003;
mission and ministerial records, and Politburo reports. Sheila Fitzpatrick 1999; Khlevnyuk 2001a; Gábor
Another notable series covers agriculture and collectiviza- Rittersporn 2001).
7
tion (Danilov, Roberta T. Manning, and Lynne Viola 1999 The economic journals currently archived at JSTOR
continuing). published 320 articles on aspects of the Soviet economy
4
Forced labor: V. P. Kozlov (2004); the defense-indus- between 1920 and 1991; half of these appeared in just
try complex: R. S. Ganelin (2003) and V. A. Zolotarev three journals, the American Economic Review, the
(forthcoming). Review of Economics and Statistics, and the Journal of
5
Further information is available at http://www. Political Economy. A catalogue is provided under http:
hoover.org/hila/projectsarch.htm. //www.warwick.ac.uk/go/sovietarchives/before.
Gregory and Harrison: Allocation under Dictatorship 723
The archives demonstrate that Stalin’s of this surplus as long as war did not imme-
personal dictatorship was real and, from diately threaten. Stalin saw this surplus as
1932 onwards, unchallenged. Through his depending critically on worker incentives;
ten-man Politburo, Stalin was continually he managed these by giving unflagging
immersed in economic decisions. His attention to consumption. While keeping
Politburo guided the general direction of consumption low, he feared that at any time
the economy by infrequent major decisions it could fall so low as to promote worker
relating to investment and innumerable revolt. This accumulation model explains
detailed interventions relating to outputs why investment became volatile. The work-
and inputs. There were no moderates or ers’ consumption minimum was not a given
radicals; the Politburo was unified on gen- and Stalin sought to reduce it by coercive
eral strategy, but divided by the special means. The archives have recorded the con-
interests of its members, the expression of sequences of extensive coercion, many of
which Stalin particularly feared. The econ- them unintended and perverse. Having cre-
omy itself featured extreme centralization ated a surplus, Stalin had to hold rent seek-
and the priority of vertical over horizontal ers at bay. For the first time, we can witness
relationships. Given that the Politburo the rent-seeking process. Special interests
could make only a few of the decisions that formed immediately within the dictator’s
mattered, it delegated much of its power to own circle, leaving Stalin and a few associ-
lower levels in complex hierarchies to cre- ates to battle for the interests that they
ate a nested dictatorship of many smaller viewed as encompassing.
“dictators.” Since each player distrusted Part 4 deals with allocation in the economy
those below him and those below knew any as a whole. Before the archives, we already
decision could be revised by those above, knew a great deal about shortages, queues,
even trivial issues tended to be passed and corruption. The relationship between
upward. The result was a “dictator’s curse”: these was hotly debated: were shortages the
despite the intention to delegate, Stalin had unintended result of the famous “soft budget
to make many more decisions than he constraint” (Kornai 1980), for example, or
wished. were they intentionally created for corrupt
Part 3 deals with the limits on Stalin’s purposes (Andrei Shleifer and Robert W.
power to extract a surplus. Before the Vishny 1992)? By implication, if shortages led
archives, we knew the published economic to corruption, was corruption the intended or
plans, decrees, and speeches that promised unintended consequence?
an ever-brighter future for Soviet producers Archival research has shown clearly how
and consumers, but even the best informed the soft budget constraint operated and
could only speculate as to the economic pol- where it came from. It originated in Stalin’s
icy that lay beneath (Eugène Zaleski 1971, industrialization drive and was perpetuated
1980). Most basically, we did not know the by his inability to commit to financial disci-
economic fundamentals in the dictator’s pline. The “softening” process is significant
objective function nor how he would trade because the channels through which it
them against political and other objectives. operated help explain the unexpected inter-
The record of Politburo decisions, com- est of producers in higher prices and
bined with our knowledge of the informa- money, including traceable bank money.
tion that Stalin monitored most closely, The archives provide no support for the
suggests that Stalin aimed to maximize the proposition that a shortage economy was
economy’s surplus, defined as output less created to increase rent-seeking opportuni-
consumption. His horizon was distant ties. Still, despite extensive shortages,
enough to make accumulation his first use money remained a prized commodity and
producers pushed inexorably for higher 2. The Dictator, Hierarchy,

prices, contrary to previous stereotypes of and Economic Policy
the command system. There were markets
Marx, Engels, and Lenin paid almost no
or market-like institutions everywhere.
attention to how socialism would organize
These markets did not work well, were
itself politically or economically. In their pio-
severely circumscribed, and relied heavily
neering discussion of socialism, Mises (1920
on relational enforcement.
reprinted 1935), F. A. Hayek (1937), Barone
Part 5 deals, finally, with efficiency,
(1908/1935), and Oskar Lange (1964) effec-
growth, and economic reform. Before the
tively set this issue aside by assuming a tech-
archives, we already knew that the Soviet
nocratic Central Planning Board inculcated
economic system did not work that well. In
with welfare-maximizing goals that dealt
early studies of enterprise managers, David
directly with enterprises. As a result, the
Granick (1954) and Joseph S. Berliner
“socialist controversy” (Bergson 1966)
(1957) uncovered massive principal–agent
became focused on calculation and incen-
problems. Investment choices were made
tives. In terms of the real character of high-
without rational criteria (Gregory
level decision making in the Stalinist state,
Grossman 1953). Prices failed to reflect
archival documentation has given us com-
scarcity values (Abram Bergson 1964). The
pletely new knowledge. This regime was
evidence of substandard postwar growth
indifferent to calculation, preoccupied by
was overwhelming (Bergson 1978; Gur
the need to punish and deter its enemies,
Ofer 1987; William Easterly and Stanley
and bent on implementing its decisions
Fischer 1995) and increasingly preoccupied
through a complex administrative hierarchy
the country’s leadership; the system seemed
of agents motivated by threats and promises.
incapable of significant reform (Gertrude
E. Schroeder 1973, 1979, 1982; Philip
2.1 Stalin as Dictator
Hanson 1983; Morris Bornstein 1985;
Vladimir Kontorovich 1988). Early archival investigations, focused on
Research in the archives has shown how Stalin’s role in the Great Terror, showed that
decisionmakers at every level actually allo- he orchestrated it to a surprising degree
cated resources: They made it up as they (Khlevnyuk 1995), although he may not have
went along, using intuition, historical prece- intended all its consequences (Getty and
dent, and common sense. As befits a Naumov 1999; Rittersporn 2001). Stalin’s
bureaucracy, there were plenty of formal complete authority is revealed in how he was
rules, but the rules were constantly revised able to turn various major policies on and off,
or overridden. Those at higher levels made for example, stopping the Great Terror with a
formal rules, only to break them. Those at single memorandum (reproduced in Kozlov
lower levels, unable to live without rules, 2004, vol. 1). The same conclusion applies to
relied on customary norms or rules of Stalin’s role in economic decision making.
thumb. The flaws in this planned economy From approximately 1932 until his death,
became apparent to its leaders almost Stalin was a true dictator: he had his way on
immediately. Before the archives, we every matter and was not afraid to abuse and
believed that official proposals for decen- humiliate those on whom he depended most
tralizing economic reform began to circulate closely (Davies 2001a; Davies, Melanie Ilič,
after Stalin’s death; the first serious reform and Khlevnyuk 2004; Rees 2004; Gorlizki
experiment actually dates to 1932 and bears and Khlevnyuk 2004). As Khlevnyuk (2001a,
a strong resemblance to reform proposals p. 325: emphasis added) has concluded,
after Stalin’s death. But, like all those that “Stalin himself was not merely a symbol of
followed, it was quickly frustrated. the regime but the leading figure who made
the principal decisions and initiated all state appointed and scheduled, an arrangement
actions of any significance.” Stalin was per- that continued until his death in 1953 with
sonally informed about and gave time to little change. One element of formality
large numbers of minor matters. Stalin’s pre- remained: Stalin continued to bind his asso-
eminent role does not mean that he was in ciates into complicity by requiring each
control of the consequences of his decisions. Politburo member to approve his decisions
On the contrary, his correspondence is full of once he had made them (Gorlizki and
concern about “paper fulfillment” and of Khlevnyuk 2004).
angry calls for “implementation” committees The erosion of collective rule is consistent
and monitoring fulfillment by placing with Hayek’s (1944) insight that the rise of a
responsibility on designated officials sole dictator is inevitable in such an environ-
(Gregory 2004, pp. 165, 266). ment; this leaves unsolved the puzzle of its
What role did Stalin’s colleagues play? In partial restoration after Stalin’s death.
the late 1920s and at the beginning of the Stalin’s ascendancy is explained by the need
1930s, decision making in the ten-person for a tie breaker within the Politburo and the
party Politburo was still collective. Already fact that Stalin was more ambitious, brutal,
acknowledged as the senior personality, and controlled than his rivals. But why was a
Stalin still had to bargain and cajole to get tie breaker necessary? Although political sci-
his way (Khlevnyuk 1996; Rees and D. H. entists and historians had speculated about
Watson 1997). Dissenting Politburo mem- ideological divisions within the Politburo
bers were obliged to support majority deci- after 1930, the archives have revealed no
sions in public while bringing disputes to the disagreements on the basic directions of for-
full Politburo for resolution, procedures that eign and domestic policies. There was no
Stalin conveniently ignored during his power “moderate” group after the “right deviation-
struggle with the “right deviationists” ists” were eliminated; therefore, there were
Bukharin, Tomskii, and Rykov. In the course no extremists either (Rees 2004, p. 47). The
of this power struggle, Stalin also cultivated divisions that did exist within the Politburo
the support of regional leaders and other were on lines of narrow self-interest based
members of the larger party Central on departmental position.
Committee, the venue for the ultimate Stalin, who made all top-level appoint-
denouement of the power struggle in 1929 ments personally, was deeply suspicious of
and 1930. The archives confirm that Stalin professional administrators and technocrats
exercised great cunning, patience, and self- and trusted only a few old Bolsheviks. He
control in organizing his Politburo majority aimed to concentrate economic decisions in
and Central Committee support to oust his the hands of a small number of “reliable”
last rivals. people; the ethos that Stalin wanted to instill
After 1930, Stalin increasingly bypassed is summed up in the term proposed by
the formal procedures for party consultation Wheatcroft (2004): “Team Stalin.” Although
and committee decisions; this is reflected in he consulted with them regularly as individ-
the declining frequency of Politburo meet- uals, Stalin conceived of his colleagues col-
ings, which fell from weekly in 1929 to only lectively as his instrument, not a
one in the second half of 1937 (Rees and consultative body. When, for example, he
Watson 1997). While formal meetings fell saw political dominance over Gosplan
away, Stalin met increasingly with his associ- threatened, he ranted: “It is sometimes
ates in private where he could control par- worse than that: Not Gosplan but Gosplan
ticipation and agendas directly (Wheatcroft sections and their specialists are in charge
2004). He reached decisions alone or with ad [and are turning the Politburo] into a court
hoc subcommittees that he personally of appeals or a council of elders” (Belova
and Gregory 2002, p. 270). He placed hyper-centralization.” In short, interests

Team Stalin members in positions for were not less moral for being sectional.
which he would hold them personally
responsible and was dismayed when they 2.2 Principals and Agents
began to represent narrower interests. As
head of the party control commission, Team Stalin was responsible for bringing
Ordzhonikidze, a fellow Georgian and for- decisions forward and implementing them.
mer head of the Caucasian party, led the The Politburo’s own immediate staff was lim-
battle against “disloyal” managers; once ited in the early 1930s to 230 specialists;
appointed minister of heavy industry he below this layer were better-staffed ministries
became their ardent defender (Khlevnyuk and state committees that often prepared
1993). As head of the party’s transport over- decrees for the dictator. Team Stalin was
sight committee, Stalin’s first deputy assisted by a state planning commission,
Kaganovich (interviewed by F. I. Chuev Gosplan, that employed 900 specialists of
1992, p. 61) stood up for encompassing whom many were technocrats of initially
interests but began to demand more rails questionable loyalty (Khlevnyuk et al. 1995).
and investment immediately after he Team Stalin issued high-level decrees in the
became transport minister. name of the government or its subcommit-
Stalin railed against rent seeking within his tees for defense and the economy, but the
narrow circle: “It is bad when we begin to most important state decrees were issued
deceive each other” (Khlevnyuk et al. 2001, p. jointly with the Central Committee.
80). He complained bitterly about the “self- The order of issuing decrees was confus-
ishness” of the minister of heavy industry, ing even to the top leadership, and Stalin had
who pressed “on the state budget of the work- to guard against improper invocation of the
ing class, making the working class pay with party’s name by agencies seeking higher
its currency reserves for his own inadequacy” endorsements (Gregory 2004). Most decrees
and that the “use of funds must be discussed were issued to a restricted number of recipi-
in the interests of the state as a whole, not ents on a need-to-know basis. Alongside the
only in the interests of [the ministry of heavy almost 4,000 decrees published between
industry]” (Khlevnyuk et al. 2001, pp. 72, 88). 1930 and 1941 were more than 28,000 secret
He particularly loathed the deputy minister ones, of which over 5,000 were so secret that
of heavy industry for “turning our Bolshevik they were known only to a handful of people
party into a conglomerate of branch groups” (Davies 2001a). Rules on secrecy were them-
(Rees and Watson 1997, p. 16). selves especially secret, which sometimes
Examples of Politburo figures choosing complicated enforcement (Harrison 2004).
encompassing interests over their own are Standard organization charts (Gregory
rare. One case was the first secretary of the 2004) show a vertical organization with the
Ukrainian party, who, unlike other regional Politburo and central government at the top,
secretaries, did not fight for lower grain col- aided by functional and control agencies
lection targets with the result that Ukraine such as Gosplan, the committees for state
lost millions to starvation during the famine and party control, and the omnipresent inte-
of 1932–33 (Davies and Wheatcroft 2004). rior ministry known in different periods as
The practice of top party officials represent- the OGPU, NKVD, or MVD. The actual
ing narrow interests illustrates a point made “managers of production” were industrial
by Khlevnyuk (2001a, p. 325): although the ministries and regional authorities that
competition of interests “contradicted the planned and supervised production units.
principles of the dictatorship” it could also The archives dispel the pretence that princi-
limit “the destructive consequences of pals and agents at all levels of the hierarchy
united to build socialism, share information, Gosplan was established in February 1921
and forego private profit. Rather, the man- to institute and operate a “unified state plan”
agers of production who were responsible for the whole economy and harmonize the
for results engaged in disputes with those to plans of other economic departments. It was
whom they were accountable (Belova and modestly endowed for these grand tasks and
Gregory 2002; Gregory and Andrei still had only fifty specialists in 1925. In
Markevich 2002; Gregory 2004) that were 1929, Stalin made it clear that he was not
usually resolved by informal means, interested in “balances” and “limits” that
although administrative and legal remedies restricted his freedom of action. The purge
were also available (Belova 2005). of Gosplan that followed famously tested
Formal links among industrial ministries, “the civil courage of those specialists who are
regional authorities, or among enterprises, already admitting in the corridors that they
are notably absent despite the fact that hori- prefer to stand up . . . for high rates of expan-
zontal transactions are the basis of special- sion than to sit [in jail] . . . for low ones” (a
ization and exchange in all economies. plan official in 1929, cited by Edward H.
Unauthorized horizontal links represented a Carr and Davies 1969, p. 938). At the same
troubling problem for the dictator. On one time a planned economy was being laid
hand, a principal gives an order to an agent down, those who were trying to create its
precisely because the allocation that the intellectual foundations were being
principal desires differs from the one that “hound[ed] out of Moscow” in accordance
would result from the agent’s trading on her with Stalin’s instruction (cited by Belova and
own account. Hence, the obedient agent Gregory, 2002, p. 271).8
must be induced to forego opportunities to Gosplan prepared only plans that were
trade for private gain. On the other hand, highly aggregated, stating: “Gosplan is not a
frequent planning mistakes ensured that supply organization and cannot take respon-
orders issued from above could not be sibility either for centralized specification of
implemented unless supported by unautho- orders by product type or by customer or the
rized transactions, even the barter of “an regional distribution of products”(cited by
ordinary suitcase full of cigarettes” (cited by Gregory 2004, p. 139). Gosplan refused to
Davies 1996, p. 266). Such horizontal trans- plan horizontal transactions in detail, label-
actions (again to cite Khlevnyuk) “contra- ing the latter “syndicate work” (Belova and
dicted the principles of the dictatorship” but Gregory 2002). Gosplan did represent the
limited “the destructive consequences of government in interministerial conflicts and
hyper-centralization.” served as a reluctant consultant despite
pleading that “we are simply not equipped to
2.3 Planning With a Light Touch deal with such matters” (Belova and Gregory
2002, p. 271). In short, after its purge and
When key decisions were taken by Team subsequent politicization, Gosplan limited
Stalin, usually on Stalin’s personal authority, its exposure by doing as little as possible. It
what role was left for the planning profes- was not until the late 1940s that Gosplan
sionals? Textbook accounts (Gregory and
Robert C. Stuart 1974; Alec Nove 1977) sug-
8
gest that Gosplan exercised considerable Gosplan’s politicization had less of an impact on prac-
tice than one might imagine because, in the early 1930s, its
executive power over allocation. For the methods were still primitive. Gosplan’s most comprehen-
Stalin period at least, this view requires revi- sive centralized balances covered only thirty raw materials,
sion: Gosplan was important, but it was not eight energy sources, and four types of machinery allocat-
ed among large-scale industry, small-scale industry, food
powerful, and surprisingly sought to limit its processing, and exports (Wheatcroft and Davies 1985,
own power. appendices A and D).
again became an activist organization under principles only.” The archives show that
its young and influential chairman, Nikolai Hayek was right.9
Voznesenskii. While the Soviet economy was managed
The prevailing view of Gosplan before the by decree, there were few formal stable
archives exaggerated its role, at least for the rules; the rules that existed were subject to
Stalin period. Gosplan was not an all-power- override. Fresh guidelines were issued to
ful director of resources; the power plan each new year or quarter, rather than
belonged to the dictator. Stalin did not want carry over general planning rules. Ministries
a planning board with immense powers or operated without charters that spelled out
numerous staff. Why then did he keep it in corporate governance (Gregory and
being? Stalin had need of a relatively small Markevich 2002). The few accounting and
agency on which he could rely to tell him the loan administration rules were easily ignored
truth. He held most of those around him with the tacit approval of Team Stalin
accountable for results and for that reason (Gregory 2004); the enforcement of financial
did not trust them to report outcomes truth- targets and value-for-money was selective
fully. That was the specialized task of and arbitrary. Rather than let the courts
Gosplan, which became Stalin’s solution to enforce legal interagency agreements, Team
the wider principal–agent problem (Belova Stalin allowed and encouraged “administra-
and Gregory 2002. Typically, Stalin also gave tive” enforcement through appeals to vertical
overlapping responsibilities to other agen- superiors (Belova 2005).
cies such as the economic administration of Hayek’s expectation of the dictator’s aver-
the NKVD. Gosplan’s leaders did not have to sion to formal rules is exemplified by the
make a success of the economy in order to operational plans that enterprises were
retain Stalin’s trust; that was the job of the obliged to fulfill by law (Belova 2001). The
ministers for the production branches. All formal procedures were complicated, con-
Gosplan had to do was to report honestly. tradictory, and confusing for the actual par-
This did not necessarily make an independ- ticipants (Markevich 2003). In the Stalin era,
ent Gosplan leader such as Voznesenskii the operational plan for the enterprise was
popular with the other top party managers. not the unified output, input, finance, tech-
Voznesenskii was one of Stalin’s favorites nology, and labor plan [tekhpromfinplan]
while Stalin trusted him. When the others that Soviet and western planning texts
found an opportunity to sow distrust with describe. Rather, the enterprise usually
the boss, Stalin had him shot (Gorlizki and received a few output and assortment assign-
Khlevnyuk 2004). ments midway through the plan period,
while secondary targets for costs and pro-
2.4 Rules Versus Discretion
ductivity were worked out retrospectively for
Hierarchical organizations resort to con- reporting purposes.
tingent rules, customs, and conventions to
make boundedly rational decisions. Bergson 9
Contingent rules that guide decision making feature
(1966) suggested that contingent rules could in two different economic literatures, both of which are
relevant to the present survey. The rules-versus-discretion
resolve a number of problems, such as literature refers to formal stable rules, for example, those
opportunism and computation, specific to a that govern taxation. In this context, it is a problem if the
planned economy. Hayek (1944, p. 82) agent can too easily ignore or renegotiate such a rule. This
was Hayek’s point. The bounded-rationality literature dis-
asserted to the contrary that a totalitarian cusses customary rules or rules-of-thumb that are made
system “cannot tie itself down in advance to informally and evolve as agents struggle with computation
general and formal rules that prevent arbi- constraints and uncertainty. In this context, it is a problem
if such rules lead to systematic errors. Our point is that the
trariness . . . It must constantly decide ques- Soviet command system was intolerant of formal rules; on
tions which cannot be answered by formal the other hand, informal rules-of-thumb proliferated.
A surprising feature of the working too specific (Gregory 2004 gives examples).
archives of ministries and enterprises is the Delegating the right to intervene to lower
near total absence of final “approved” plans. levels was dangerous, however, since it was a
All plans were labeled “draft” or “prelimi- problem to align the interests of lower
nary.” The draft plan was no more than an agents with those of the dictatorship as
informal agreement which could be changed Bergson (1964) once pointed out.
subsequently by virtually any superior. The Plan interventions created havoc for pro-
“correcting” and “finalizing” of plans was a ducers. A meeting of industrial ministers
never-ending process; the “final” plan held in December 1946 turned into a repre-
remained always on the horizon. Searches in sentative complaint session. The minister of
the ministerial archives have located only the electronic industry: “During the quarter,
one finalized annual plan, that for light even during the month, there are a great
industry in 1939 (Markevich 2003). number of changes, modifications, and addi-
The archives provide thousands of cases of tions to the approved plan. We do not usual-
plan revisions. Ministers ordered last minute ly work according to the plan; rather we work
changes; factories were shuffled from one on the basis of supplemental decrees, admin-
authority to another; one factory was istrative decisions, and the like . . .” The min-
ordered to increase its production post haste ister of the aviation industry: “It is better to
to make up for production shortfalls in have one plan than to change it twenty times”
another factory. Even Politburo commission (Markevich 2005). The most important indus-
decisions could be changed at the last trial leader of the 1930s expressed his frustra-
minute: Although the first-quarter 1933 tion as follows: “They give us every day
vehicle distribution plan had been approved decree upon decree, each one is stronger and
by its own transport committee, the Politburo without foundation” (Khlevnyuk 1993, p. 32).
threw the plan out the window by tripling the Ministries and enterprises insured them-
Kazak party committee’s allocation and dedi- selves against interventions by holding back
cating 90 percent of the vehicles to “organs of obligatory information and submitting their
control over agricultural producers” (Valery own plan proposals at the last minute to
Lazarev and Gregory 2002, p. 332). avoid duplicating this work later; ministries
Each level received a barrage of requests often proposed relatively modest targets to
to intervene or to refrain from intervention Gosplan while quietly imposing tougher
in a highly formalized bureaucratic process. assignments on their own enterprises
In deciding how to respond, Team Stalin did (Belova and Gregory 2002). Ministries
use some implicit rules-of-thumb, such as fought for generalized plans and tried to
the priority of heavy industry and the mili- avoid divulging enterprise plans to Gosplan;
tary. Thus the minister of heavy industry: in April 1933, for example, Gosplan com-
“All orders for the Ministry of Defense must plained that ministry plans “suffered from
be fulfilled exactly according to the schedule such incompleteness that it is impossible to
not allowing any delays” (cited by Gregory use them” (Belova and Gregory 2002, p.
2004, pp. 160–61). Enterprises cited priority 274). The ministries withheld information
considerations to defend against interven- from Gosplan and financial authorities on
tions. Military shipbuilders in 1935 ignored grounds of national security (Belova and
orders from the highest state authority on Gregory 2002; Harrison 2004). For produc-
the grounds that they interfered with mili- ers, the best plan was either no plan at all or
tary objectives (Gregory 2004). Priorities, a plan so general that it left all the real deci-
however, had to be limited to be effective, sions to them. Gosplan even uncovered
and the priority statements issued by leading cases of “nonplanning”: “Enterprises [large
authorities were often either too broad or enterprises located near Moscow] declared
to our representatives that they had not seen that sentenced Stalin and his Politburo to
annual plans for a period of years” (cited by lives of toil, drudgery, and tedium (Gregory
Belova and Gregory 2002, p. 275). 2004). Threats of resignation and pleas for
Stalin (1937, p. 413) wrote: “Only bureau- lengthy vacations were commonplace. A rep-
crats can think that planning work ends with resentative Politburo meeting, held on March
the creation of the plan. The creation of the 5, 1932, had 69 participants and 171 points on
plan is only the beginning. The real direction its agenda (Khlevnyuk et al. 1995). The great-
of the plan develops only after the putting est burden fell on Stalin who, in a typical
together of the plan.” Team Stalin main- year, 1934, spent 1,700 hours in official meet-
tained the right to override plans at will ings, the equivalent of more than 200 eight-
because this provided “resource mobility.” hour days (Khlevnyuk 1996). Virtually every
In practice, such procedures served the communication requested his decision.
interests of principals at every level and On rare occasions, Stalin would explode at
could not have been better designed for the this torrent of paperwork, for example in a
exercise of political influence. Everything tirade of September 13, 1933: “I won’t read
was tentative and subject to arbitrary change drafts on educational establishments. The
by someone higher up in the chain of com- paperwork you are throwing at me is piling
mand. No one can know what went on up to my chest. Decide yourself and decide
behind the scenes when petitioners met with soon!” (Khlevnyuk 1996, p. 340). A few
superiors, but we can guess that savvy politi- weeks later the same Stalin berated the
cians like Stalin would weigh the political Politburo for not following his proposed dis-
benefits of satisfying an influential regional tribution of tractors to the letter (Khlevnyuk
or industrial leader. 1996). Stalin suffered the dictator’s curse
The aversion to explicit rules reflected the (Gregory 2004): his power to decide all gave
dynamic commitment problem of a dictator. his most trusted colleagues the incentive to
Stalin’s unwillingness to bind himself in decide as little as possible. The less they
advance cascaded down through the political decided, the less he could blame them when
system, preventing the emergence of a for- things went wrong.
mally rule-based or “law-governed” econo-
my. Kornai, Eric S. Maskin, and Gerard 3. Accumulation and Consumption
Roland (2003) suggest two kinds of commit- Before the archives, we could only guess at
ment failure: in one, a predatory principal Stalin’s real economic policy. Would the dic-
forces agents to break formal rules so as to tator foster economic growth (Mancur Olson
exploit them; in the other, agents choose to 1993; Edward Glaeser et al. 2004)? To what
break the same rules so as to exploit a weak extent would he share his rents to build loyal-
principal. In the Soviet case, producers ty or increase his political power (Ronald
could break rules citing the threat to pro- Wintrobe 1998)? The archives tell us that
duction from the rule, while superiors Stalin was obsessed with accumulation, which
reserved the right to punish hapless scape- is hardly a surprise. Between 1928 and 1937,
goats for breaking the same rules. It is some- Soviet real GNP doubled, but the fierce
times hard to detect whether such rule repression of private consumption enabled a
breaking reflected the power of officials to quadrupling of real investment (Bergson
force producers to commit violations or the 1961). More generally, investment rates in
ability of producers to commit violations and socialist countries were consistently higher
get away with them. than in capitalist countries of comparable size
Generally, to exploit power to the full and income levels (Simon Kuznets 1963).
involves encountering its limits. Specifically, Without information on high-level deci-
it was the power to live outside formal rules sion making, the prearchival literature had
little to say about aggregate investment works, blast furnaces, railways, and canals
choice.10 Bergson argued (1964, p. 241) that that were included on its itemized “title lists”
investment depends “on the output mix to of approved projects. Their purpose was to
be produced from it”; this view percolated embody the new society in cement and
through the various editions of textbooks structural steel. How were these programs
(Gregory and Stuart 1974). Hence, the pre- implemented and managed?
vailing prearchival view was that the dicta- Although Politburo meetings for the
tor’s primary target was output, from which 1930s left few formal minutes, his deputies
investment was derived. The archives show wrote to Stalin from time to time to detail
that this view was wrong: Stalin and the key Politburo meetings and seek approval of
Politburo attempted to control the basic operational decrees (Gregory 2004). They
direction of the economy through the level reveal that the Politburo consistently set
and distribution of investment. They were three targets in the 1930s: the nominal
less concerned about setting physical out- investment budget, grain collections, and
puts in operational plans. foreign exchange. These three control vari-
The literature on how a socialist economy ables all related to investment. The invest-
ought to determine aggregate investment ment budget allotted funds to industrial and
stems from Marx’s emphasis on accumula- regional agencies for construction and
tion and proceeds via the growth models of machinery. Despite an original intent to use
Preobrazhenskii and Feldman in the early physical material balances of investment
Soviet period (Alexander Erlich 1960; goods to plan investment, ministries and
Nicolas Spulber 1964) to the later models of regional authorities were simply given
Brus, Kalecki, and Dobb (Peter Rutland “investment rubles,” and no one appeared to
1985). The earlier models suggest that the know the real investment that resulted.
feasible investment rate depends on the bal- Grain collections were designed to con-
ance of power in the economy and the scope tribute to a budget surplus through the
for expropriating the “former” social classes. excess of state sale prices over purchase
The later models suggest that the optimal prices. Stalin personally directed foreign
investment rate depends on society’s dis- exchange to the import of capital goods
count rate, but that a labor-abundant market rather than the luxury goods sometimes
economy may underinvest if the realized demanded by the Bolshevik elite.
subsistence wage is above the shadow wage. If Stalin’s goal was indeed to maximize
No Politburo member was trained in eco- investment, the archives provide two types
nomics, however; nor did they feel the lack of evidence that are, at first glance, confus-
of such training. ing. First, Stalin was extremely concerned
about consumption, particularly where it
3.1 The Politburo Accumulation Model
touched upon the productivity and morale
At the core of the Politburo’s strategy to of the industrial workers. Consumer sup-
“build socialism” in the first two five-year plies were one of the most frequent items
plans, were massive programs for the hydro- on Politburo agendas; in Stalin’s words, the
electric dams, machinery complexes, vehicle “provisioning of workers” was one of “the
most contested issues” before the Politburo
10 and trade was “the most complicated min-
Bergson’s authoritative Economics of Soviet Planning
(1964) introduces investment first in chapter 11 and large- istry” (quoted by Gregory 2004, pp. 93–94).
ly puts aside the issue of its total volume, focusing instead Stalin interpreted declining labor productiv-
on rational choice of individual projects. The most widely ity as a sign that workers “were not provi-
cited paper on investment (Grossman 1953) deals exclu-
sively with the choice to be made when a given increase in sioned as well as last year,” and personally
desired output could be obtained from alternative projects. ordered the delivery of consumer goods to
Total Output, Q
Q(E)
Q∗
E(W,C)
W∗
Total Effort, E
E∗
Figure 1. The Politburo Model
cities where labor productivity was declin- Figure 1 illustrates the model that the
ing (Gregory 2004). The Politburo set itself Politburo appears to have used to set
up as the highest trading organization, investment and consumption. The figure
deciding retail trade plans, prices, assort- mirrors the Marxian concept of the surplus
ment, and even the opening of new stores. product, the gap between output and con-
Second, we have the little-known empirical sumption, as the outcome of a distributive
fact that the Politburo, which based the struggle. The model has theoretical precur-
success of its program on capital construc- sors in Wolfram Schrettl (1982, 1984) and
tion, deliberately reduced nominal invest- Leon Podkaminer (1989), and is set out
ment on two occasions, in 1933 and 1937 more fully by Gregory (2004). It belongs to
(Davies 2001b) citing the fear of inflation a general class of models in which a ruler’s
and rising construction costs. freedom of action is circumscribed by social
This evidence can be interpreted in two “tolerance limits” (Kornai 1980) or a revo-
ways. One is that that Stalin’s preferences lution or disorder constraint (Daron
were unstable; fitful humanitarian impulses Acemoglu and James A. Robinson 2000).
led him to direct more supplies to the By raising investment without limit, the
workers from time to time; periodic fits Politburo risked provoking the workers to
of financial orthodoxy led him to regret go slow, strike, or rebel.
overstraining the economy. Alternatively, The demand for labor was always enough
Stalin’s preferences were stable and he for full employment and besides all able-
reallocated consumption or reduced invest- bodied persons were required to work by
ment according to a consistent rule-of- law (Granick 1987), so the figure takes
thumb. Knowing of Stalin’s capacity for employment, N, as fixed exogenously; indi-
calculation, patience, and self-control, we vidual effort, e, was variable, so total effort,
—
reject the first explanation and investigate E = e· N, was variable although employment
further the second. was not. Total output, Q, depended on total
effort, E. Total effort varied with the real the same starting point and in the opposite
wage, w, as follows. The aggregate wage bill direction, however, not only cuts effort and
W, the consumer goods received by workers, output by more but also risks pushing the
is measured along the vertical axis in the workers into outright confrontation with the
same units as output, and is proportional to state. An investment-maximizing dictator
the real wage given that employment is must tread a fine line between the pursuit
—
fixed, i.e., W = w· N. There is a reservation of investment and the triggering of serious
wage, analogous to a tolerance limit or disor- disorder.
der constraint, below which effort is zero; Stalin managed worker morale and effort in
there is also a “fair” wage at which effort is two ways. When investment and consumption
maximized.11 As the economy moves from were about right in the aggregate, detailed
the fair wage to the reservation wage, effort plan mistakes could still leave some workers
declines as workers turn up late or drunk, or with too much and others with too little.
go slow or absent without leave; at the limit, When this happened, Stalin personally
unrest simmers and threatens to boil over ordered the reallocation of consumer goods
into local and general strikes and rebellion. to those left short. But there could also be
Thus the effort curve intersects the horizon- aggregate mistakes; when too much invest-
tal axis at the reservation wage and becomes ment threatened to disrupt the economy
vertical at the fair wage. Effort also depends and provoke the workers, the Politburo
on the level of direct coercion, C, to which preferred to sacrifice investment. This is
we turn in the next section. To maximize how we interpret the unforced investment
effort, the dictator would pay the fair wage cutbacks of 1933 and 1937. Although Stalin
and get the maximum output, but this would did not express his concern for consump-
not maximize the surplus. To maximize the tion directly, when he advocated “strength-
surplus, Q − W, he would choose the inter- ening the ruble” as a justification for less
mediate wage, effort, and output levels investment (Gregory 2004, pp. 236–42), it
denoted W ∗, E ∗, and Q∗. was the same ruble that workers had to
The shape of the effort curve is hypothet- spend in retail markets; when he feared ris-
ical; the hypothesis is Stalin’s and can be ing construction costs, these were fiscal
inferred from his observable anxiety about costs that had to be covered through taxa-
errors in the distribution of consumer tion or the inflation tax. In this sense,
goods. An effort curve of this shape makes Stalin’s behavior was stable and consistent,
the consequences of plan mistakes asym- given the constraints that he perceived.
metric: for the dictator, paying the workers Figure 1 suggests other options. Team
too little could be much worse than paying Stalin could seek to manipulate the effort
them too much. Starting from the invest- curve by offering ideological rewards in
ment optimizing position, more consump- place of material payoffs. Idealists from
tion does at least raise effort and output, Russia and abroad in fact assisted the first
and this somewhat mitigates the fall in five year plan, motivated by the idea of
investment. A mistake of the same size from building socialism. The attempt to trans-
form homo economicus into homo sovieticus
led, however, to a vicious circle of wage
11
The fair-wage concept in figure 1 can be related to equalization and declining productivity
the observation of George A. Akerlof (1984) and Akerlof
and Janet L. Yellen (1990) that workers in market (Hiroaki Kuromiya 1988; Davies 1989a,
economies give extra effort if they are paid a wage per- 1996). Subsequent mobilizations were limit-
ceived to be fair, and withdraw it if the wage is reduced ed to short lived campaigns such as for
below the psychologically determined fair level. We sup-
pose that effort has a maximum, so payments above the fair World War II, and to cultivate the “virgin
level do not elicit more effort. lands” of Kazakhstan and Siberia after the
war. The most publicized mobilization forcing the Politburo to cut investment
campaign of the 1930s, the Stakhanovite back. Stalin used the vast informant network
movement, was driven by progressive piece of the NKVD to monitor protests, strikes,
rates that permitted participating workers anti-Soviet statements, and factory-wall
to drive up their incomes by overfulfilling graffiti, and eavesdrop to gauge mass opin-
norms (Davies and Khlevnyuk 2002); the ion (Berelovich 2000). Stalin had obvious
archives suggest that Stalin abandoned it political motives to do this, but within our
because it tended to raise fair-wage aspira- framework we conclude that wages and fair-
tions among nonparticipating workers, and ness lay at the cross-hairs of politics and
also threatened inflation (Gregory 2004). economics.
Stalin could also elicit extra effort from
party “enthusiasts” in the present by prom- 3.2 Coercion: Three Experiments
ising future promotion. The problem was
that these promises eventually had to be In principle, efficient penalties are cheap-
made good. Lazarev (2005) has identified er than efficient rewards since “a promise is
the creation of new posts that carried a costly when it succeeds and a threat is costly
wage premium to reward loyal supporters when it fails” (Thomas C. Schelling 1960, p.
as an important feature of Stalin’s policy. 177). In figure 1, the fear of punishment may
Relentlessly purging the potentially disloy- cause workers to lower their reservation
al made it cheaper for him to keep his wage without reducing effort; coercion may
promises to the actually loyal. not make effort more productive but should
In an extension of the model, the make people willing to supply effort for less.
Politburo could bear down on consump- As long as coercion displaces the effort curve
tion selectively via targeted rationing. downward while leaving the production
Elena Osokina (2001) argues that Stalin curve undisturbed, the surplus is increased.
wanted to drive down consumption and In market economies, outside options, such
hence introduced a discriminatory ration. as alternate employment, leisure, or unem-
Julie Hessler (2004) argues that, on the ployment, cannot easily be limited by force
contrary, Stalin approved of nonrationed as the ubiquity of informal sector employ-
trade and a consumption oriented life- ment and illegal migration demonstrates. In
style but crises (collectivization, rearma- the Stalinist state, however, the idea of con-
ment, war, and postwar famine) allowed trolling workers’ alternatives by force proved
him to achieve this goal for only short peri- attractive.
ods. Gregory argues that Stalin saw The archives show that Stalin believed
rationing as a way to force accumulation that a wide range of problems could be
without a loss of effort of high-priority solved by force. The language with which he
workers according to the principle: “He addressed obstacles to his rule was habitual-
who does not work on industrialization ly violent. Faced with foot-dragging in the
shall not eat” (cited by Gregory 2004, p. transport ministry in 1931, he wrote: “we
98: emphasis added). Rationing carried must smash this gang . . . If you can manage
other costs, however, as Stalin became without my help, smash the gang before it’s
aware (Davies and Khlevnyuk 1999). too late.” Considering the role of speculators
Figure 1 has a further ominous extension. in a legalized urban rural market, he wrote
The fair wage was set by a mass psychology in 1932: “we must eradicate this scum . . .
that was unpredictable and hard to manipu- The OGPU [secret police] and its agencies
late. If workers concluded from the propa- must, without delay, start training its forces
ganda of economic successes that they were and studying the enemy” (Davies et al. 2003,
being cheated, the fair wage would rise, pp. 95, 102, 165).
The archives have thrown new light on relatively open discussion in the late 1920s
past controversies. Historians have argued supported a substantial literature on the col-
about the extent to which Stalin’s motives in lectivization decision, the most complete
the Great Terror of 1937–38 were primarily account of which is by Davies (1980a,
economic or political; thus Roberta T. 1980b); James W. Heinzen (2004) has now
Manning (1993) speculated that Stalin added a study of the agriculture ministry in
launched the terror in order to solve wide- the 1920s. Naum Jasny (1949), Moshe Lewin
spread economic difficulties, but the (1968), and Erich Strauss (1969) outlined
archives have not yielded any substantial evi- some general results. James R. Millar (1974)
dence to this effect (Davies 2004). Rather, and Michael Ellman (1975) concluded that
the aim of the terror was political: to isolate the investment surplus that Stalin hoped to
and remove a potential “fifth column” from gain from collective agriculture was probably
society that, Stalin believed, could endanger not forthcoming. But the detailed conse-
his regime in a time of rising international quences of collectivization and the mass
tension (Khlevnyuk 1995). Of more interest deportation or detention of peasants who
to economists, some scholars have argued were excluded or resisted were concealed
that coercion was the glue that held the behind a thick veil of secrecy.
Stalinist economy together. When coercion Mark B. Tauger (1991), Davies, Tauger,
failed, or was abandoned, the economy fell and Wheatcroft (1995), Wheatcroft and
apart (Kontorovich 1986; Olivier Blanchard Davies (2002), and Davies and Wheatcroft
and Michael Kremer 1997; Harrison 2002). (2004) have now reviewed the archives on
Others have emphasized the huge social and the immediate aftermath of collectivization,
economic costs of coercion (Robert including the famine of 1932–33. The collec-
Conquest 1987; Khlevnyuk 2001b; Davies tive farms enabled Moscow to replace local
and Wheatcroft 2004). Below we discuss the decision making with its own detailed plans,
issues of whether coercion paid for itself and instructions, and formal, but often transient
where the optimum lay. rules. Stalin was focused on what he could
Stalin conducted three notable experi- control: sown acreage and the state’s share in
ments with the use of coercion to foster what this acreage produced. But Stalin could
accumulation: the forced collectivization of not control the harvest. Acreage expanded
the peasantry, the criminalization of work- but yields collapsed; the share delivered to
place indiscipline, and the widespread use the state increased. Excessive procurements,
of forced labor. In each case, the outlines bad weather, and plan errors combined to
have been known for decades, but the full strip the countryside of grain; first the live-
consequences are only now coming to light. stock were slaughtered, then the farmers
themselves starved. They were prevented
3.2.1 Collectivizing Peasants from feeding themselves from their own
harvests by severe punishments including
Politically, collectivization aimed to death for petty theft. Davies and Wheatcroft
impose Soviet power in the countryside and dispel Conquest’s (1987) notion that Stalin
eliminate the stratum of richer peasants, manufactured the famine to kill class ene-
the kulaks. It was triggered, however, by a mies; rather they show the leadership subse-
grain marketing crisis that reflected the quently trying to ameliorate the effects of its
peasants’ perceived unwillingness to con- own bungling.
tribute sufficiently to investment-led indus- While famines usually occur in poor coun-
trialization. Collectivization began in tries with limited statistical reporting, the
earnest in December 1929, signaling Soviet archives provide good documentation
Stalin’s victory in the power struggle. The of the two peacetime famines of 1932–33 and
1946–47. Ellman (2000) has applied the enti- consequences, however, remained obscure
tlement theory of Amartya K. Sen (1981) to until the archives were opened.
the latter famine, which particularly affected The law of 26 June 1940 (Kozlov 2004,
the peasants of the Ukraine and other south- vol. 1) made absenteeism, defined as any
ern regions. Sen argued from historical expe- twenty minutes’ unauthorized absence or
rience that famines are more likely to arise even idling on the job, a criminal offense,
from entitlement or distributional failures punishable by up to six months’ corrective
than production failures, and that “no matter labor at work with a 25 percent reduction in
how a famine is caused, methods of breaking pay. Repeat offenses counted as unautho-
it call for a large supply of food in the public rized quitting, punishable by two to four
distribution system” (1981, p. 79). Ellman months’ imprisonment. Enterprises man-
makes two points from the evidence. agers were made criminally liable for failure
First, he argues, in the 1946–47 famine, to report worker violations. In August 1940,
production and entitlement failures interact- the minimum sentence for petty theft at
ed. The production failure was not such as to work and “hooliganism” was set at one year’s
leave insufficient food available to keep every- imprisonment. Wartime decrees punished
one alive. In this sense, the famine arose from defense and transport workers for unautho-
a failure of entitlement. But the famine was rized quitting with long terms in a labor
triggered by a harvest failure; if the harvest camp. After the war, a notorious decree of
had not failed, everyone would have had suf- June 1947 set the minimum sentence for any
ficient entitlements to stay alive. Therefore, theft of state or socialized property at five
the famine cannot be attributed to entitle- to seven years imprisonment. Transport
ment failures alone. Second, Ellman notes work was eventually demilitarized in
that in this famine the role of the state was March 1948 and work in the defense indus-
essentially negative: it selected those who died try two months later; otherwise, these
by denying them entitlements. Therefore, it punitive laws remained on the books until
could be argued, concentrating grain stocks in Stalin died.
the hands of the Soviet state actually More liberal governments also took pow-
increased the number of deaths. Davies and ers to direct their key workers in wartime,
Wheatcroft (2004) show that, in 1932 and but the detail, scope, and degree of enforce-
1933, the state intentionally directed food to ment of the Soviet measures went to an
those able to work in the fields and denied it extreme. A report prepared as background
to those already hospitalized by hunger. for Khushchev’s secret de-Stalinization
Accordingly, the 5.5 to 6.5 million famine speech of February 1956 (Kozlov 2004, vol.
deaths in these years far exceeded those 1, statistical appendix), shows that, from 1940
recorded in famines before the Revolution through June 1955, the regular courts and
(Davies and Wheatcroft 2004). military, transport, and labor camp tribunals
sentenced a total of 35.8 million persons for
3.2.2 Regimenting State Employees
all criminal offenses. Not allowing for repeat
As the 1940s began, Stalin redirected offenders, this would represent about one
coercion from specific class enemies to the third of the adult population of roughly 100
entire public-sector work force. A battery million. Of the 35.8 million, 15.1 million
of intimidating laws criminalized work were imprisoned and a quarter of a million
place violations which had previously been were executed. The annual rate of imprison-
managed by administrative sanctions with- ment was one million or more in most years
in the enterprise. The laws themselves up to 1950, and more than half a million at
were not secret and were described by the time of Stalin’s death. Such conviction
Conquest (1967) and Nove (1969). Their rates were about five times as large as in the
United States at about the same time.12 number of detainees ranged up to 20 million,
These totals were dominated by workplace we now know that there were approximately
offenses (Don Filtzer 2002; Andrei K. 2.3 million penal laborers at the outbreak of
Sokolov 2003, forthcoming) and by convic- World War II and about 2.5 million at the
tions for theft of socialized property (Kozlov time of Stalin’s death; similar numbers of
2004, vol. 1, statistical appendix). deportees were also confined to labor settle-
The archives show that Team Stalin, not ments in the remote interior. The forced
an independent judiciary, determined the laborers were mostly engaged in forestry,
number of convictions. When the laws were mining, and construction, where they made
enacted, Stalin had to intervene personally up substantial shares of employment, but
to force a reluctant judiciary to prosecute never more than about 3 percent of the total
tardy or lazy workers (Sokolov 2003). workforce including farm workers, and less
than this in terms of the value of national
3.2.3 The Labor Camps output (Khlevnyuk 2001b, 2003a).
Collectivization in the early 1930s, the These new facts must be set beside others,
Great Terror of 1937–38, the repression of including the very high rates of conviction
state employees in the 1940s, and the arrests and sentencing that we now know about:
of “national contingents” during and imme- although the Gulag population was smaller
diately after World War II created huge than observers had earlier guessed, it also
flows into labor camps. In the 1920s, there had much higher turnover with very large
was just one forced labor complex in the numbers entering and leaving to return to
Arctic where mainly political detainees were society. While we are confident of the stocks
held. The first major expansion came with of Gulag inmates at different points in time,
collectivization which threw hundreds of the cumulative total of persons sentenced to
thousands of peasants into camps or special the Gulag in the course of its existence,
settlements; the Gulag, the interior ministry’s probably in excess of twenty million, remains
chief administration of labor camps, was cre- the subject of debate. We now know that the
ated in 1930 to handle the sudden inflow and Gulag’s own central catalogs are inconsistent
labor camps spread across the remote interi- (Kozlov 2004, vol. 2); it appears that even
or and the far north and east to house, at the Gulag did not know the correct number.
their peak, more than 2.5 million inmates. Internal Gulag documents confirm that
Although the Gulag was shrouded in political strategy (collectivization, terror,
intense secrecy, its human side leaked out war) rather than economics dictated the
through the recollections of former inmates. Gulag’s development. Before the archives,
Estimates of its scale and scope, however, there was speculation that in the course of
could be based on little more than guesswork the 1930s economics eventually took over
until the first official figures were released in from politics as a motivation for recruiting
1989. While contemporary estimates of the forced laborers, or alternatively that the
NKVD became a lobby for forced labor (S.
12
Given the many factors that differentiated the Soviet Swianiewicz 1965). Neither of these has
judicial and penal system from others, it is hard to find
appropriate standards for detailed comparison of work- turned out to be the case, although our
place offending. It is simpler to go to the aggregate of all judgment has to be carefully shaded.
offenses. From 1940 to June 1955, the USSR imprisoned The Gulag had a consistent economic rai-
15.1 million people or, on an annual basis, roughly one per-
cent of its adult population each year. In 1958, for com- son d’être: to explore and colonize regions
parison, the United States imprisoned not more than one that were resource-rich but inhospitable,
fifth of one percent of its adult population (U.S. since forced labor could be ordered around
Department of Commerce 1960, pp. 139, 141). Thus
Stalin imprisoned his subjects at least five times more fre- the country at will (Khlevnyuk 2003a).
quently than the United States. Subsistence wages combined with the
enforcement of effort through close supervi- the Gulag’s future until its major construc-
sion were supposed to promote low-cost tion flaws became apparent. The Gulag
accumulation. leaders willingly undertook the building of
At the level of the state, the continued pri- Noril’sk (Leonid Borodkin and Simon Ertz
macy of politics over economics is shown by 2003) because they underestimated the risks
the fact that the NKVD–MVD did not lobby and difficulties that would arise. The opera-
for expansion. The NKVD projected a tion of Noril’sk (Ertz 2003) helped to expose
shrinking number of Gulag inmates for the illusions about the ease with which the
third five-year plan (1938–42), just as the first inmates could be coerced into supplying
victims of Stalin’s Great Terror began to flood effort without economic rewards.
in (Gregory 2003a). After the war, the NKVD By the postwar years, Gulag officials had
again planned for contraction and politics concluded that the camps failed to generate
again overturned this expectation a surplus. Labor productivity there was
(Khlevnyuk 2003a). In the late 1940s, Gulag extremely low relative to that of free work-
officials proposed to release all but the most ers, while guarding detainees was very
dangerous prisoners from camps (Aleksei expensive; in 1950 there was one guard to
Tikhonov 2003), but this was unacceptable to ten inmates, leading to the widespread prac-
Stalin. In 1953, within three months of tice of “unguarded” prison contingents.
Stalin’s death in March, MVD chief Lavrenty Within the Gulag, prisoners formed protec-
Beriia had released one and a half million tive networks among themselves and with
prisoners, 60 percent of the Gulag’s inmates the guards to cover for each other (Heinzen
according to a plan prepared five years earli- 2005). The arsenal of punishments was not
er. In fact, elements within the MVD were sufficient to motivate prisoners and trade-
increasingly alarmed by the Gulag’s econom- offs were complicated: prisoners placed on
ic and social costs. The economic costs were reduced rations for failing to meet work quo-
reflected in its growing financial deficits; the tas were no longer able to work effectively.
social costs were measured by high rates of One of the most effective incentive systems,
recidivism. Although the camps were sup- early release for exemplary work, deprived
posed to segregate hardened criminals from the Gulag of its best workers. Material incen-
youth offenders, the camp population was a tives played an ever larger role in motivating
mixing bowl and recidivism soared. The high penal labor (Borodkin and Ertz 2003, 2005;
turnover spread the culture and mores of Ertz 2005). In the last years of the Gulag,
camp life throughout society. there was a process of “conversion of slaves
Why did the Gulag fail? Research on the to serfs” (Khlevnyuk 2003a, p. 57); the
microeconomics of the Gulag is in its infan- camps increasingly paid prisoners civilian
cy but early case studies show a complex wages and the distinctions between penal
learning process. The Far Eastern camps and free labor became blurred.
(David Nordlander 2003) show how early
3.3 Coercion Failure?
optimism about huge surpluses in gold min-
ing was replaced by pessimism as output per The collective farms effectively ceased to
inmate fell precipitously. The Karelian exist in the mid-1960s when the farmers
camps (Christopher Joyce 2003) show the were placed on fixed wages like any other
experimental process by which the authori- employee. The Gulag was emptied of mass
ties learnt the scope and limits of the prisoners between 1953 and 1957. The dra-
exploitation of forced labor. The fact that conian labor laws of the 1940s were rescinded
the White Sea–Baltic Canal (Mikhail in the mid-1950s. These relaxations have two
Morukov 2003) was finished on time and on possible interpretations: either the post-Stalin
budget stimulated illusory expectations for leadership did not have the stomach for a
system of repression that was working or every worker became liable to prosecution
coercion was abandoned because it did not for something, including one-time and acci-
work. The new evidence supports the sec- dental violations: a broken-down commuter
ond interpretation, particularly the fact that bus could make criminals of scores of hapless
the most coercive instrument, the Gulag, workers. Rational managers might wish to
was dissolved by its own leaders. select the truly guilty for prosecution, the
In short, the initial belief that results could problem workers and repeat offenders, but
be obtained more cheaply by raising the the laws subjected even petty offenses to
penalty for failure than by increasing the harsh penalties and managers who failed to
reward for success appears to have been mis- report offenses were threatened with the
taken. Effective coercion requires that penal- same. As a result, the innocent were bundled
ties be accurately assessed and targeted, and through the courts and camps along with the
that the agents of repression are well guilty in extraordinarily large numbers.
informed about offenders and the costs of Team Stalin probably knew this but did
their crimes. It proved more difficult than not care. In March 1937, Stalin’s chief
expected to target penalties accurately; as a instrument of the Great Terror, Nikolai
result the relationship between true effort Ezhov, told his officials to expect “some
and punishment was “noisy.” The devising of innocent victims . . . Better that ten innocent
an efficient coercion system is then further people should suffer than one spy get away.
complicated if workers and managers When you chop wood, chips fly” (Simon
respond strategically to increase the noise. Sebag Montefiore 2003, p. 194). Stalin’s
The archives show that officials had little prime minister, Molotov, (interviewed by
idea whether workers were exerting full Chuev 1991, p. 416) also cared more about
effort or not; the law could do little more condemning the guilty than acquitting the
than ensure that they were physically at work innocent, “never mind if extra heads fall.”
and did not steal too much. Agricultural con- Type II errors were also clearly numerous;
trollers could order the collective farms to this is evidenced by the fact that, although
sow more land but could not assess whether penalization rates were very high, offending
the land was being farmed efficiently rates were even higher. Filtzer (2002) used
(Davies and Wheatcroft 2004). In industry, the records of eight production branch min-
attempts to pin “normal” effort down to istries in 1947 to show that almost one third
objective technological criteria proved fruit- of a million “labor desertions” gave rise to
less; attempts to maintain effort norms when fewer than 55,000 convictions, for a convic-
productivity should have been rising often tion rate of 16 percent. In other words, a
gave rise to damaging social conflicts (Davies judicial system that was supposed to “make
and Khlevnyuk 2002; Filtzer 2002). the chips fly” somehow failed to chop the
Because the state relied on indirect indica- wood. The combination of severe penaliza-
tors of effort, mistakes crept into the tion with low conviction probability for the
effort–punishment relationship. The investi- guilty is consistent with high-cost policing
gation of low effort could yield an error of and justice administration (Gary S. Becker
Type I that punished the innocent, and a 1968); the high rate of conviction of inno-
Type II error that acquitted the guilty. Errors cents, however, is more properly seen as a
of both types appear to have been present. cost of dictatorship (Simeon Djankov,
Numerous Type I errors are reflected in the Glaeser, Rafael la Porta, Florencio Lopez-
very high rates of penalization that con- de-Silanes, and Shleifer 2003), in the sense
demned hard workers along with ne’er-do- that the dictator’s efforts to achieve a lower
wells, drunks, and thieves. Such a wide range rate of offending than society was willing to
of behaviors was criminalized that virtually tolerate had highly suboptimal results.
The effectiveness of the Politburo accu- clique that fended off criticism from below
mulation model rested on the dictator’s abil- and investigation from above in the mid-
ity to create a gap between the civilian wage 1930s, but was eventually wiped out in the
as a “fair” return for effort, and low subsis- Great Terror. Belova (2001) has shown that
tence in the Gulag as the return to shirking, high-level patrons could protect the most
so that the difference between them was the egregious embezzlers.
intended punishment for shirking. Although Faced with widespread enforcement
the Gulag did not generate an internal sur- failures at lower levels, officials responded
plus for accumulation, it could still have con- in two ways. First, they aimed to under-
tributed to the surplus in the economy as a mine horizontal trust by rewarding whistle
whole. The effort curve of the accumulation blowing and informing through organized
model will be displaced downward if civilian campaigns (Belova 2001). Second, Stalin
workers expect the Gulag wage as their pun- forced the legal system, local party offices,
ishment for low effort. But the widespread and the militia to increase arrest and con-
mistakes that we have described subverted viction rates or suffer penalties themselves.
this strategy: if workers expect Type I errors The most common method of forcing
to prevail, they will be punished regardless repression was to distribute quotas by
of effort; if they expect to benefit from Type region and profession to officials at lower
II errors, they can shirk without fear. levels (Kozlov 2004, vol. 1). In the Great
Error rates, moreover, were not exoge- Terror of 1937–38, local officials had to
nous. They were fashioned by the counterac- work feverishly to achieve a set number of
tions of those threatened with punishment, confessions per day (A. I. Vatlin 2004). To
who could take steps to reduce their risks. fulfill such plans, the police officials imput-
Workers and managers diverted effort from ed individual guilt from increasingly trivial
production into mutual insurance: since the differences in behavior. Whether or not
threat was shared among them, they could these measures reduced the Type II errors,
agree to cover up each other’s shortcomings. they seem likely to have encouraged false
The archives have added detail to the denunciation and confession and so added
prearchival literature on concealment in the to the errors of Type I.
factory (Berliner 1952). Filtzer (2002) has Team Stalin hoped, we suppose, that
shown that postwar managers tolerated late- increased coercion would induce agents to
ness and absence to maintain goodwill, and take a lower wage without withdrawing
colluded with workers to underreport such effort, making coercion effectively costless.
violations, while pursuing quitters who The evidence suggests that, in practice,
undermined morale and the factory’s capaci- those threatened with punishment raised the
ty to fulfill the plan. The rural police and chance of escaping detection (a Type II
courts pooled risks with the rural communi- error) through mutual insurance, but this
ty in sheltering the young offenders who had diverted effort from production. At the same
deserted factories or technical schools time, successful collusion against Type I
(Kozlov 2004, vol. 1). In all these ways, errors (false accusation) combined with the
mutual insurance tended to cut the individ- increased chance of a Type II error to blunt
ual risk of punishment. The archives also the effect of penalization on the reservation
show how regional party officials defied even wage. Finally, the heavy direct costs of the
the most powerful central organizations to repressive apparatus must have further
protect their own (Khlevnyuk 2004). James reduced the net payoff from coercion.
R. Harris (1999, pp. 156–63) has described We have only one empirical study of the
how the Urals regional leadership formed aggregate effects of the terror on industrial
itself into a “protected, mutually reliant” production (Barbara G. Katz 1975), which
attributes the 1937 slowdown in industrial nomenklatura into many little nomenklat-
growth to political repression. New archival uras each dependent on an intermediate
studies show the disastrous effects of the boss. With delegation and fractionalization,
Great Terror on important Gulag operations these would form natural market places for
(Ertz 2003). Finally, the archives have shown the exchange of gifts and favors for loyalty
that Soviet leaders themselves saw a direct in which distributional conflicts and deals
connection between the terror and falling took place.
productivity (Davies 2004): the terror elimi- In the 1930s, investment was the largest
nated a generation of skilled and knowledge- single use of Stalin’s rent. It is not easy to tell
able managers and technical staff, worsened whether he distributed it primarily to share
labor discipline, and damaged effort. rents or to generate growth. In principle, the
rent shared would have been the excess of
3.4 Sharing the Surplus outlays in a given use over those which a
growth-maximizing dictator would have allo-
Whatever the mechanism, Stalin’s eco- cated. Even if rent-sharing took place, how-
nomic policy certainly generated a substan- ever, it would have been rationalized under
tial increase in the excess of output over the official guise of advancing growth, and
consumption for use in investment, defense, the dictator’s most loyal agents already occu-
or other public spending. But surpluses pied the key positions most likely to receive
attract rent-seekers. To what extent did investment.
Stalin have to share the proceeds with others Recent case studies throw light on the
as the price of his power? This is a question motivation behind particular transactions. In
that could not be asked before the archives the late 1920s, the Politburo and its agent,
were available. Although the archives are Gosplan, fought against regional rent seeking,
now open, definitive answers remain elusive. accusing “irresponsible” regions of “self serv-
The archives show that Stalin presided ing projections based on local interest, lack of
over a “nested” dictatorship (Lazarev and objectivity, and inexact calculations that
Gregory 2002; Gregory and Markevich 2002; undermine the very foundations of planning”
Gregory 2004) in which power was delegat- (Gregory 2004, p. 80). Stalin resisted fierce
ed from one vertical level to the next so that regional lobbying in the second half of the
each organization duplicated the administra- 1920s, rejecting projects such as the
tive control structures of its superior. Each Dneprstroi metallurgy complex on the
principal in turn acted as the agent of a high- grounds that there were better uses of the
er principal until the top where the great dic- funds (Lih, Naumov, and Khlevnyuk 1995).
tator ruled alone. This was also a complex Gosplan’s resistance evaporated with the 1929
hierarchy with nodes from which organiza- purge of those who argued for realistic plan-
tions stemmed with partly complementary, ning. Regional leaders descended on Moscow
partly overlapping functions. The dictator with grandiose investment plans (Harris
ruled by delegation, so under him there 1999). Eventually Stalin set up institutional
ruled many smaller dictators who exercised barriers to this kind of behavior: regional
unconstrained power within their specialized party leaders required official permission to
fiefs that Stalin personally allocated to them. come to Moscow and enterprises were pro-
Recent studies of the Soviet nomenklatu- hibited from maintaining representative
ra under Stalin, which comprised up to the agents in Moscow (Khlevnyuk 2004).
order of a million posts (Lewin 2003; Stalin’s advocacy of superindustrialization
Khlevnyuk 2003b), show how power cascad- from 1928 on promoted his power as well as
ed downward through branching networks his long-run rents. He gained allies among
of agents that fractionalized the “aggregate” regional leaders by espousing a program that
promised unlimited funding. Stalin’s new factories and inflate construction” (cited
appointments diary is full of meetings with by Khlevnyuk 2000). In the early 1930s, the
regional leaders who made the pilgrimage to Politburo, in a rare show of opposition to
Stalin to seek favors. Stalin was uncharacter- Stalin, delayed the Baltic–White Sea Canal
istically concerned in 1931–32 that his home until Stalin reduced its cost by proposing the
republic Georgia was “on the verge of use of “cheap” Gulag labor (Morukov 2003).
hunger” and of “bread riots,” while he made The Red Army provides a case study in
“feigning hunger” a counterrevolutionary Stalin’s relationship with a powerful special
offense in other republics (Getty and interest. Sokolov (2005) has shown that the
Naumov 1999, p. 69). Lazarev and Gregory demands of the armed forces were signifi-
(2003) have analyzed the dictator’s distribu- cant in Stalin’s turn to forced collectivization
tion of motor vehicles, the scarcest capital and industrialization in 1929–30, and the
good of the time, from his own reserve military leaders themselves supported his
funds. The strong econometric results show resort to a command economy to bring these
that Team Stalin allocated vehicles for polit- things about. Stalin quickly moved to raise
ical gain and that economic considerations military spending in secret in 1931 (Davies
were not significant.13 1993). On the other hand, he kept profes-
Stalin needed a growing military and sional soldiers out of the Politburo. He also
industrial base and could not have allowed consistently resisted the attempts of military
rent seeking to randomize investment modernizers, such as Marshal M. N.
beyond a point. Stalin opposed the “unnec- Tukhachevskii, to use plans for a mechanized
essary spreading of investment” that resulted mass army to justify a military role in manag-
from industrial and regional authorities’ ing the defense industry (Lennart Samuelson
competing bids to start up favored projects 2000). Evidently alert to the danger that the
(cited by Davies and Khlevnyuk 1997, p. 41). armed forces could become just another spe-
A first-hand account shows Stalin advocating cial interest, Stalin encouraged rivalry and
investment rationality at a meeting in 1947: tension between the army and the industrial-
“The plan is very inflated and is not within ists who might otherwise have formed natu-
our capacity. We should give money only to ral allies (Gregory 2003b; Harrison 2003b).
projects that can be put into commission, While Stalin lived, therefore, there was no
and not spread it out among many projects. emergence of a “military–industrial com-
They are building all kinds of nonsense in plex,” whatever may be said of the Soviet
new, unpopulated areas and they are spend- Union under his successors.
ing a lot of money. It is necessary to expand Similarly, there is evidence of rent seeking
old factories. Our dear fantasists design only in military R&D but no evidence that Stalin
or his subordinates wanted it that way
13 (Harrison 2003a, forthcoming-b). He pun-
This study is based on full data on requests for motor
vehicles from the dictator’s reserve fund in 1932 and 1933, ished rent seeking where he suspected it,
a process that was quite separate from the wholesale allo- and this helped to make the intermediaries
cation of motor vehicles through ministries and regional that he charged with funding this work keen
authorities. The researchers had each request, the cases
made for the allocation, letters of support, and the eventu- to get results.
al outcome. The final decisions were made by a high-level
Politburo commission headed by Molotov. The researchers 4. Money, Prices, and the Seller’s Market
coded the cases made for the vehicle allocation as “eco-
nomic,” for example a note from Gosplan explaining why A third set of issues addressed by archival
this allocation was vital to fulfill the plan, or “political,” for research considers how plans and commands
example support from a political patron. In the regressions, are nested with markets and money, how
only the political variables were significant in explaining
approvals. Economic variables were insignificant and also markets work, and how contracts are
had the wrong signs. enforced in the context of a seller’s market.
These issues connect literatures that are not demand could be transmitted from firms to
often linked, dealing with markets and hier- the retail market and arguing that shortage
archies, contracts and enforcement (Avner could be explained more parsimoniously by
Greif 1993, 1994, 2000; Greif, Paul firms’ withholding supplies and pushing
Milgrom, and Barry R. Weingast 1995), and prices down in order to collect bribes from
the soft budget constraint (Kornai 1980; consumers.
Maskin 1996, 1999; Kornai, Maskin, and The literature thus offers a range of
Roland 2003). hypotheses for archival research. Kornai
would look for a political decision by Team
4.1 The Soft Budget Constraint
Stalin to tolerate loss-making activities so as
Kornai (1980) developed the concept of to exploit the economy for political goals.
the soft budget constraint to explain shortages Dewatripont and Maskin would look for the
and sellers’ markets in Soviet type economies; rules and incentives that made funding prin-
he attributed the soft constraint to the state’s cipals prefer to tolerate losses. Gomulka and
paternalistic domination and its desire to others would expect to see the price con-
insure enterprises against failure. The result- trols of which shortage is a by-product,
ing moral hazard damaged firms’ incentive to while Shleifer and Vishny suggest that short-
economize on inputs and created an insa- age is the intentional creation of rent-seek-
tiable hunger for resources that was eventual- ing producers who restrict output and lobby
ly transmitted to permanent shortages and for low prices.
queues in retail markets. Kornai (1980) main- Archival investigations have shown that all
tained that such shortages could not be elim- the theoretical explanations of the soft budg-
inated by price flexibility. An early critic, et constraint have some merit; they do not
Stanislaw Gomulka (1985), suggested that support the Shleifer–Vishny critique. The
price rigidity must be a necessary condition of documents of the formative years tell
shortage, specifically that input prices must Kornai’s story of a mobilizing state overrid-
be less negotiable than firms’ budgets, but ing the market to gain discretionary power
there did not appear to be a convincing expla- over resources.14 In the 1920s, the ruling
nation of why the state should cap input Bolsheviks allowed industry to go over from
prices. Others have attempted to deduce the self-financing to a regime of price controls
soft budget constraint from the formal rules and subsidies for the sake of its plans for
and incentives arising from specific institu- industrial and military mobilization; mili-
tions, rather than from Kornai’s political tary–industrial interests and military lobby-
negotiation process; for example, M. ing were more important in this process than
Dewatripont and Maskin (1995) suggested was previously recognized (Sokolov 2005).
that the softening agent is the sunk costs that The trusts that were later transformed into
arise when there is sequential monitoring of ministries were already shifting funds from
long-lived enterprises and their financing is profitable to loss-making enterprises in the
centralized. Underlying all these variants is a mid-1920s (Gregory and Tikhonov 2000). As
problem of dynamic commitment (Mark E. the state won control of agriculture through
Schaffer 1989): the state does not intend to collectivization and the first five year plan
support loss-making ventures but is unable to was enacted, soft budget constraints were
commit itself not to after the event, and those extended to industry and investment as a
responsible for the losses anticipate this whole (Harris 1999). Specific legislative
beforehand.
In contrast, Shleifer and Vishny (1992) 14
Kornai ought to have been right; he had the advan-
and Wintrobe (1998) departed fundamental- tage over others of witnessing the formative years of the
ly from Kornai, questioning how excess Hungarian planning system as an insider.
attempts to harden budgets were quickly lead to retail shortage if firms used claims on
abandoned (Gregory and Tikhonov 2000). inputs to “siphon” general-purpose com-
These episodes show that budget constraints modities intended for the retail market back
became soft through a direct exercise of into production. Kim showed that siphoning
power that changed the relationship really did drain off resources that should have
between the state and the economy. To the been available for private consumption, that
dictatorship, the loss of financial credibility actual private saving was persistently higher
was a price worth paying for the freedom to than suggested by published budget surveys,
its pursue wider goals. and that significant private saving was forced
More detailed case studies have also by firms’ siphoning activity as a result.
shown the value of institutions-based theo- Budget constraints were soft but not limit-
rizing. Long-lived R&D projects in defense less; what were the limits on softness? Kornai
industry, for example, were funded and mon- (1980, p. 211–14) defined three “tolerance
itored in installments. The resulting sunk limits” on the state’s readiness to accommo-
costs made the authorities ready to refinance date inefficient behavior: foreign currency,
projects that they would not have financed social unrest, and administrative complaints.
initially with complete foresight (Harrison The view of Stalin’s economic policy set out
2003a). Even in this case, however, the docu- above confirms the importance of all three.
ments reveal an intrinsic element of negotia- Stalin personally kept an iron grip on foreign
tion that is not captured by a model based on exchange. The unrest associated with con-
formal rules alone. Funding principals could sumption shortfalls is represented by the dis-
not or did not wish to commit to explicit order constraint in figure 1. Finally, the
rules. They understood the games that Politburo carefully monitored the flow of
unscrupulous agents could play with such complaints up the vertical hierarchy for signs
rules and they sought to regain financial con- of general worker discontent.
trol in two ways. First, they cut funding back
4.2 Money and Prices in the Seller’s Market
from time to time in ways that were to some
extent arbitrary. Second, even if establish- In theory, money was relatively unimpor-
ments were never closed for making a loss, tant in the Soviet command system. Firms’
they could be put under new management soft budget constraints created seller’s mar-
and the old management could be demoted kets, the so-called “dictatorship of the sell-
and punished individually. For self-protec- er.” Prices were ineffective in incentivizing
tion, agents built mutual insurance networks producers and signaling allocative needs
and also engaged in cutthroat rivalry to pro- (Kornai 1980). Instead, the planners’ “visible
tect funding. Thus, well after the formative hand” was supposed to direct allocation of
years, the softening of budget constraints producer goods at controlled prices (Nove
cannot be understood without close attention 1961; Ellman 1972, 1979; Gregory and
to political processes of vertical bargaining Stuart 1974). The banks supplied money and
and horizontal rivalry and collusion. credit to “follow” physical plans; under early
Does the story of the soft budget con- Soviet rule this was called “planned automa-
straint stand up as an explanation of Soviet tism” (Gregory and Tikhonov 2000). Money
retail shortage? The transmission mecha- merely enabled planners to practice “control
nism, doubted by Shleifer and Vishny (1992) by the ruble,” monitoring financial flows to
and Wintrobe (1998), has been substantiated detect departures from physical directives.
from unique access to the post-Stalin The money stock was supposed to be strictly
Gosplan archives by Byung-Yeon Kim (1997, segregated into bank money for interfirm
1999, 2002). According to Kornai (1980, pp. transactions and cash money for wage pay-
486–88), the soft budget constraint would ments. Given that firms could use money
only for planned purposes, there would be or below. Ministries also liked this arrange-
little incentive for them to acquire addition- ment because it freed them to decide how
al holdings. The state budget mainly provid- best to fulfill aggregate ministerial output in
ed a façade behind which the authorities plan rubles.
could divert resources from profit-making Final allocations of products were
enterprises to loss-makers and the military achieved through contracting between min-
(Igor Birman 1981). istries, ministry main administrations, and
A major surprise from the archives is that enterprises. Research on the late Soviet-era
money played a much larger role than we economy showed that this contracting
expected. Allocation actually began not with process was relatively decentralized (Heidi
physical supply plans but with nominal budg- Kroll 1986, 1988). In the Stalin era, supply
etary assignments to investment and other planning set limits on this process but the
government uses such as military orders. limits were broad. The ministry’s annual or
The Politburo gave much more time and quarterly supply plan was only the first salvo
energy to how rubles would be spent than to in the “battle for the plan,” in which users of
consideration of the “control figures” for out- intermediate goods entered bids with pro-
put in physical units (Davies 2001a; Gregory ducers for the contracts needed for their
2001; Davies, Ilič, and Khlevnyuk 2004). own plans (Harrison and Nikolai Simonov
Budget outlays usually came first because 2000; Gregory and Markevich 2002). At this
broad-brush supply plans could not fix the point, command-economy allocation became
detailed assortment of physical products or a market-like contracting process; official
their final uses. Plans in rubles of output retail markets that traded state goods openly
were then calculated at “fixed” plan prices. at state prices and retraded them under the
Plan targets had to be fixed in rubles because counter at a premium were just at one
most producers supplied many products and extreme of a continuum of market-oriented
output was too heterogeneous to be planned activities that even included bizarre activities
any other way. Supply quotas binding on such as home production of automobiles
individual ministries and enterprises were (Lazarev and Gregory 2002).
also denominated in rubles (Harrison 1998). Decentralized contracting generated a
While the government might announce a degree of price flexibility, and this tells us
plan target for steel tonnage, the directive much about the motivations, resources, and
plan for the enterprise was in rubles; intera- constraints of the agents involved. By devis-
gency contracts were supposed to link the ing legal and illegal ways to bid up contract
two, but were usually incomplete or hard prices, suppliers could fulfill both plans and
to enforce, even in the defense industry contracts with less effort and more financial
where compliance was monitored intently. gain. According to the stereotype, prices
Markevich and Harrison (forthcoming) were supposed to be fixed from above on the
report the case of an aircraft factory in 1934 basis of initial costs plus an allowance for
where the same managers, reprimanded for overhead and taxes. Longstanding empirical
poor-quality work, were rewarded days later concerns about official price indexes (Colin
for fulfilling the plan—in rubles, of course. Clark 1939; Alexander Gerschenkron 1947;
Stalin-era plans were too aggregated to tie Bergson 1947, 1953, 1987; Peter J. D. Wiles
producers to particular products or users to 1982; Hanson 1984), including the exploita-
particular suppliers; this suited both min- tion of new products specifically to free the
istries and planners. Planners could not dis- enterprise from fixed plan prices (Berliner
aggregate plans efficiently, and preferred 1976; Harrison 1998), suggested that the
the responsibility for disaggregation and fixed-price assumption might oversimplify.
subcontracting to lie at the ministerial level Kornai (1980, p. 363) specifically noted the
interest of the firm in higher prices. tageous if the defense ministry could be
Nonetheless, standard accounts of Soviet persuaded to settle for a smaller real quan-
allocation took it for granted that the firm tity, say Q in the figure, at some higher
was a price taker (Nove 1958; Edward Ames price level P along a unit-elastic curve
1965; Michael Manove 1971; Richard E. through Q∗, P∗. It did this by reporting high-
Ericson, 1983, 1984; Stephen M. Goldfeld er costs relative to the true effort of produc-
and Richard E. Quandt, 1988, 1990; Shleifer ing them, and by introducing new products
and Vishny, 1992). at a higher ratio of reported cost-to-effort,
The archives show that price-setting was i.e., by simulating productive effort.
one of the most important activities of Soviet Price renegotiation was costly, however.
firms (Harrison 1998); producers’ pressure on At some margin, the simulation of produc-
prices was relentlessly upward, not downward tive effort became as costly as productive
as Shleifer and Vishny (1992) predicted. effort itself. Planners monitored the prices
Actual transaction prices were negotiated of existing products although they could do
between buyers and sellers during “contract little to control the prices of new ones
campaigns” loosely managed from above (Harrison 1998). Faced with price gouging,
(Harrison and Simonov 2000; Gregory 2004). a powerful buyer like the army could com-
A memo described the consequences of fail- plain through the defense minister to the
ure to pay an “illegal” high price: “If you don’t Politburo on which he sat (Davies and
want to pay, we’ll keep this in mind when we Harrison 1997). Less powerful customers
consider your next order” (cited in Gregory could and did complain frequently to the
2004, p. 220). Official prices were supposed state arbitration commission (Gregory
to be used in more important transactions but 2004) provided they were willing to risk dis-
official price handbooks were often incom- rupting good relations with suppliers. It is
plete, lagged behind new products, or were noteworthy that the Soviet authorities
ignored. The mammoth metals administra- applied the term “signalization” to the
tion of the ministry of heavy industry scarcity information forthcoming from the
employed only three persons in its pricing arbitration courts.
department, which set the official prices of The widespread evidence on illegal price
metallurgical products (Gregory 2004). The increases raises a number of questions. If
defense ministry was particularly vulnerable enterprises had soft budget constraints, why
to inflationary pricing because of the rapidity did they make the effort to push up prices
with which its product requirements were and incur the legal risks rather than wait pas-
changing. Military buyers complained of sively for an automatic subsidy? The answer
prices based on “how much it costs whether is that subsidies were not automatic and soft-
the result of correct work or poor manage- ening budget constraints took effort. The
ment.” (Gregory 2004, p. 220). Defense sup- enterprise had to equate the effort costs of
pliers withheld information about costs on the overcoming resistance at several margins
grounds that it was too sensitive to entrust to represented by the ministry which could
the defense ministry. They would delay set- switch funding from profit to loss makers,
tlement and hold out for higher prices and the budgetary and credit authorities, and a
illegal advance payments before agreeing to buyer that had to contend with its own budg-
terms (Harrison and Simonov 2000). et constraint. Overcoming the resistance of
Figure 2 illustrates the resulting bargain. the buyer enabled one enterprise that faced
The plan of military orders obliged industry more resistance from superiors to pass its
to deliver a volume of output Q∗ at a preset problem on to another that faced less. This
price P∗. Industry’s problem was that to pro- phenomenon expressed itself in illegal but
duce Q∗ cost effort; it would be more advan- persistent interenterprise payment arrears,
Price
P Industry’s Plan
of Gross Output
P∗
Q∗
P P∗· Q
Quantity
Q Q∗
Source: for the full reasoning underlying this figure, see Gregory 2003b and Harrison and Kim 2004.
Figure 2. Price Renegotiation
ultimately made good by the state bank as personal enrichment: managers extracted
lender of last resort (Gregory and Tikhonov side payments from buyers more usually to
2000). engage in siphoning and fulfill the plan, so
Since producers had to spend effort and that the proceeds of corruption were recy-
incur risks to extract it from superiors and cled into production. Planners could shift
buyers, it follows that money was not free. the purposes of corruption from embezzle-
Despite the supposed firewall between ment to siphoning up to a point by raising
traceable bank money and anonymous cash, plan tension, although this helped the plan-
increases in official credits were in fact ner more than the consumer. It follows,
accompanied by increases in cash holdings Harrison and Kim suggest, that the lower
(Gregory and Tikhonov 2000). The impor- plan tension under Brezhnev (Schroeder
tance of money is underscored by Stalin’s 1985) may have encouraged disloyal or pri-
auditors who uncovered numerous cases of vately motivated embezzlement.
fraud and embezzlement that resulted in the
accumulation of caches of illegal money 4.3 Commands, Contracts, and
(Belova 2001). Apparently, not all scarcity Enforcement
markups were lost in lobbying costs or favors
to third parties. To persist, hierarchies and markets must
Unauthorized cash did not necessarily go both evolve ways to motivate repeated partic-
to line private pockets. Harrison and Kim ipation. According to Greif (2000), people
(forthcoming) argue that the main purpose of continue to participate in an institution when
hidden inflation and siphoning was to relieve it is in their interest to do, conditional upon
effort. Thus Soviet corruption differed from their expectation that others have made the
a conventional picture of bribe-taking for same calculation. When transactions are
sequential and the completion of or stealing. The returns to stealing, however,

a contract rests on advances by one agent are reduced if what is stolen cannot be sold,
to another, the institutional equilibrium and this is more likely when market transac-
requires an enforcement mechanism to pre- tions cannot rely on the law and depend on
vent cheating by some that makes abstention private enforcement alone. Therefore the
by others their best option. enforceability of vertical commands is likely
The command system combined hierar- to vary inversely with that of private con-
chy with elements of horizontal dealing; tracts. Stalin made use of this trade-off by
the archives show that the dictator gave his privileging socialist property over private
tacit approval to this. When hierarchies and property and enacting harsher penalties for
markets coexist, they must find a joint theft from the state than from fellow citizens
equilibrium in the sense that changes in (Gorlizki 2001b).
the marginal returns are likely to induce When agents resorted to unauthorized
agents to reallocate effort from one to the business contacts, the ability to complete
other. An increase in market returns, or a transactions depended on a business reputa-
reduction in the power of principals, would tion for plain dealing and keeping one’s
shift resources out of the vertical hierarchy word. Belova (2001, 2005) has analyzed the
into unofficial horizontal dealing. Thus ver- “relational” contracting system that arose
tical enforcement relies on the ability of where unauthorized contracts were costly to
principals to reward loyalty and punish dis- enforce or unenforceable in the courts:
loyalty and so offset the potential gains agreements rested on a handshake and were
from horizontal exchange. enabled by personal contact and friendship
The problem of horizontal relationships which alone could overcome the culture of
was particularly acute given the persistence low trust.
of outside options for agents in the hierarchy. Even loyal agents, however, had to engage
Stalin may have aspired to monopolize the in some unauthorized market trading to ful-
economy into a corporation that was integrat- fill plans that were incompletely provided
ed on both horizontal and vertical lines, but for. Despite the threat to dictatorial power of
this aspiration was not and could not be real- unauthorized horizontal trade, the clear
ized. Bureaucrats could choose to deal pri- message of the industrial archives is that a
vately with each other rather than through “good” manager was expected to get the job
superiors. There was a thriving labor market done by all means necessary and at any price
that moved workers from enterprise to enter- [liuboi tsenoi]. The minister of heavy indus-
prise, even when coercion was at its most try, for example, bluntly relayed this message
intense. Everyone could turn either to legal to his managers: “We will not listen to those
or illegal commodity markets. Outside the people who say our materials have not been
country, citizens could sell secrets to a foreign delivered, but we say that a good manager, a
business or power or defect to a foreign coun- good shop director, a good master technician
try. In this general sense, the final collapse knows how to organize things and produce
of the Soviet command system was a failure the required results” (cited by Gregory 2004,
of vertical enforcement and compliance p. 164). In short, the command system relied
(Harrison 2002). on loyal agents’ unauthorized horizontal
According to Greif (2000), the “funda- dealing on behalf of the plan, despite the
mental problem of exchange” is the possibil- fact that disloyal trading detracted from the
ity of horizontal cheating in markets. In dictator’s goals.
hierarchies, there is a parallel fundamental Principals could not necessarily distinguish
problem of command (Harrison 2005): an the unofficial deals that agents made to fulfill
agent may exploit the principal by shirking the plan from those that lined their pockets.
A study of the last Soviet-era anticorruption process the information required to make
drive, based on rare access to the post-Stalin efficient decisions. The information and
judicial archives (Luc Duhamel 2004), sug- decision overload came in part from the
gests that disloyal agents would always claim Politburo’s inability to distinguish the impor-
loyal motivation, and sometimes the investi- tant from the trivial. Decision-making
gators did not want to know because they resources were scarce; in the 1930s Stalin
had a political agenda. Belova (2001) argues and the Politburo made between 2,300 and
that, in the Stalin era, principals insured 3,500 recorded decisions per year (Gregory
against disloyal corruption by tolerating hor- 2004). Much time was spent on trivia, from
izontal networks while restricting their individual permits for foreign travel to loca-
scope. As a result, horizontal reputation tions for monuments and vegetable and
could only accumulate within small groups. metro prices in Moscow. Issues defined as
This limited private exchange by reducing major were delegated to ad hoc commis-
the efficiency of relational contracting. sions with very high transactions costs: for
example, a special commission made up of
5. Efficiency, Growth, and Reform
the chairman of Gosplan, who doubled as
The debate over the feasibility of central deputy prime minister, the head of
planning occupied leading economic Gosplan’s fuel commission, timber ministry
thinkers for a century. Barone’s “Ministry of officials, and regional officials had to meet to
Production in the Collectivist State” (1908, decide on a requested cut in timber supplies
reprinted 1935) was followed by Mises’ equal to a fraction of one percent of total
“Economic Calculation in the Socialist output (Rees 1997b).
Commonwealth” (1920, reprinted 1935) and While Team Stalin labored over details
Hayek’s many writings on the information that were often trivial, it delegated major
problems of a nonmarket economy (1935, decisions of project choice by default. While
1937, 1940, 1945). Bergson’s two surveys the Politburo required major projects to be
(1966, 1967) remain authoritative accounts approved on official “title lists,” the produc-
of the controversy. Following the Great ers themselves largely determined how to
Depression and World War II, some econo- spend their investment rubles. The title lists
mists argued that socialized investment lacked cost estimates in many cases despite
could offset high private discount rates, the efforts of Gosplan, Gosbank, and the
relieve bottlenecks, or eliminate the busi- finance ministry to enforce formal rules on
ness cycle. Even Bergson (1948, reprinted cost discipline (Belova and Gregory 2002).
1966, p. 230) suggested that “in a highly When information was not simply lacking,
dynamic economy, a centralist allocation of it was limited by opportunism. Investment
investment might lead to fewer and smaller contractors opportunistically refused to pro-
errors than a competitive allocation.” vide information; they complained, for
Despite the collapse of the Soviet Union and example, that cost audits would delay urgent
China’s gradual conversion to a market econ- tasks or compromise state security (Harrison
omy, the debate over the merits of socialism and Simonov 2000; Harrison 2004). More
continues (Don Lavoie 1985; James Junker generally, those who possessed valuable
1992; Pranab K. Bardhan and John E. information tried to monopolize it and
Roemer 1993; Shleifer and Vishny 1994; shared it only when they could extract a rent
Peter J. Boettke 2001). in exchange.
Information problems led to catastrophic
5.1 Efficient Choices
investment blunders such as the
The archives illustrate the Hayekian prob- Baltic–White Sea Canal, which was too shal-
lem: the decisionmakers could not gather or low for effective use, and the Baikal–Amur
Mainline which remained unfinished for it is true that invention was not typically
forty years (Khlevnyuk 2001b; Joyce 2003). matched by innovation (Amann 1986), the
Investment projects initiated without feasi- record of invention was still remarkable for
bility studies and budgets proved unexpect- a relatively poor country.
edly difficult to finish. Increases in Archives and interviews have unraveled
unfinished construction rose sharply in crisis the decisions that established new atomic
years such as 1931–32 and 1936–37 when and aerospace industries after World War II
the proportion of gross investment that (David Holloway 1994; Simonov 2000;
merely added to work in progress rose above Harrison 2000). Notably, decisionmakers
15 percent of the total (Harrison 1994). tended to reserve their scarce attention and
Market economies are not immune to available funding for military priorities, and
investment blunders, but no one has yet the decisive information that signaled
compared unfinished construction in Soviet- where resources should go came from the
type and market systems. One reason, we monitoring of progress abroad.
speculate, is that statistical offices do not Previous insider accounts of Soviet inven-
see the problem of unfinished construc- tions were highly selective; they concentrat-
tion under market arrangements as worth ed on success stories and gave self-serving
monitoring. interpretations of what motivated success-
Not all high-level decisions were ineffi- ful inventors and designers. The archives
cient or irrational. Had this been the case, record the failures as well as the successes,
the economy could hardly have grown, sur- and Harrison (2000, 2003a, forthcoming)
vived the onslaught of Germany in World formulates a framework for Soviet inven-
War II, or created modern nuclear and tion as an economic activity: the initial offer
aerospace industries. In some respects, of funding in a field such as aviation creat-
Stalin’s economic decision making did bet- ed a “market for inventions” that attracted
ter than would have been predicted by designers and projects in large numbers. As
those who regarded socialist planning as one would expect, the designers tended to
infeasible. This mixed evaluation was be heterogeneous in talent and motivation;
already present in the literature before the on a first pass, it would seem that they were
archives. Studies confirmed low rates of motivated by a varying mix of intellectual
technological improvement and diffusion curiosity, the expected reputational rewards
(Antony C. Sutton 1968, 1971, 1973; for breakthrough projects, and the funding
Berliner 1976; Amann, Julian Cooper, and that could be consumed before a project
Davies 1977; Bergson 1978; Ronald that failed was terminated. Notably, the dis-
Amann and Cooper 1982, 1986; Hanson tribution of scientific reputation did not
and Keith Pavitt 1987) due to the lack of rely on the state for enforcement; it was
incentives for producers to adopt new conferred, at least in part, by the communi-
technologies and planning “from the ty of specialists. At the same time, the pro-
achieved level.”15 But the same system longed refinancing of unsuccessful projects,
also gave rise to some spectacular firsts for example for steam-powered bombers,
and near firsts, especially in the technolo- suggests that the uncertainty and informa-
gies of defense and heavy industry. While tion biases surrounding R&D permitted
adverse selection and provided fertile soil
15
Russians used this expression to convey a rule-of- for rent seeking.
thumb that sets the next target equal to the previous level In contrast, there was no reputation to be
achieved plus an arbitrary increment. Birman (1978) gained from replicating a technology “not
described the practice and Martin L. Weitzman (1980) and
Michael Keren (1982) formalized it in the concept of the invented here.” A variety of studies
ratchet effect. (Holloway 1994; Nataliia Lebina 2000;
Harrison 2000) have shown that, while Stalin basis of feel; there is no explanation”
regarded the fact that a technology had (Gregory and Markevich 2002, pp. 805–06).
already been developed elsewhere as a posi- According to another: “Our problem is that
tive signal, Soviet nuclear, missile, and air- we can’t really check orders and are not
craft designers disliked being ordered to able to check them. . . . We operate partial-
follow in others’ footsteps without imposing ly on the basis of historical material—we
their own stamp; this sometimes resulted in are supposed to give so and so much in this
backward steps. A case study of aeroengine quarter, and at the same time you are sup-
development (Harrison forthcoming) shows posed to give us this much.” (cited by
that, when a breakthrough was still antici- Gregory 2004, p. 172). Workers’ piece-rate
pated, the inventors were willing to work for norms were also set from the achieved level
relatively modest material rewards that were or just “by eye” (Davies and Khlevnyuk
spread around fairly evenly because the 2002, p. 877). Mises (1949, reprinted 1998,
major prize for which they competed was p. 696) predicted that in the absence of
the reputation for priority in an invention. economic calculation planning would be
After the breakthrough, when a reputation reduced to “groping in the dark.” This, then
for priority could no longer be won, the is how they groped.
same people required monetary rewards Ministry and supply officials understood
that were many times higher than before that enterprises, being “greedy oppor-
and much more regressive to engage in the tunists,” demanded “too much.” When the
development work. minister of heavy industry asked: “Tell us
please how our enterprises received 50 per-
5.2 Planning by Feel
cent of supplies they requested and fulfilled
The concept of planning by feel is illustrat- their production programs 100 percent?”
ed by an anecdote: in Moscow in 1981, a the response was: “In July, I told my people
Soviet historian told one of the authors that to prepare a report about the fulfillment of
the great skill of Gosplan director, Nikolai plans for individual branches. I then began
Voznesenskii, as an economic planner was to edit these reports and saw that the pro-
that “he could feel the disproportions in the duction programs had been fulfilled 102
economy through his fingertips.” The percent but only 40 percent of supplies had
archives reveal that this was not just a clever been received. I believed there was some-
remark. Overwhelmed by unreliable informa- thing left over from the previous year. I
tion and computational constraints, the plan- looked into the report for the previous
ners had to turn to intuition and year—again 103 percent and supplies only
rules-of-thumb based on experience. Some, 40 percent. I couldn’t look at the year previ-
like Voznesenskii, appeared to have better ous to that because I could not find it” (cited
intuition than others. by Gregory 2004, p. 172). When the first
Planners were supposed to distribute vehicles began rolling off assembly lines in
materials according to engineering norms, 1933, the producer demanded the entire
but the first allocations took place before year’s output for its own use (Lazarev and
norms were compiled (Gregory and Gregory 2002, pp. 329–30); Gosplan retali-
Markevich 2002; Gregory 2004). Supply ated by ordering customers to “liquidate
agencies used intuition, trial and error, and sloppy and unjustified requests.”
“historical experience.” According to one Without norms and with “sloppy” material
supply official: “We give 100 units to one requests, initial allocations of materials and
branch administration, 90 to another. In the equipment were arbitrary but reached equi-
next quarter we’ll do the reverse and see librium with surprising speed. As the distri-
what happens. You see, we do this on the bution of domestically produced vehicles
began in 1932, only fifteen percent of orders worked well while the Soviet economy was
were filled, rising to fifty percent by the less developed, and needed reform only after
fourth quarter of 1934 and to almost ninety industrial modernization.
percent by the first quarter of 1937 (Lazarev The story that Davies (1996) tells is rough-
and Gregory 2002). This rapid convergence ly as follows. The difficulties of supply plan-
is not explained by the increase in supply ning in 1929 and 1930 quickly convinced
alone. Enterprises, government agencies, Stalin’s industry chief, Ordzhonikidze, that
and individuals had a virtually unlimited detailed interplant transactions should be
notional demand. Their effective demand, decentralized. By 1931 he had become a
however, was limited by the expectation that keen advocate of cost accounting and the
this year’s allocation would start from last idea that, given harder budget constraints,
year’s; this illustrates planning from the enterprises could subcontract for supplies in
achieved level. Once initial allocations were a decentralized way without planners. His
fixed, a basic distributional consensus quick- economic mechanism to enforce budget
ly emerged.16 Given the difficulty of balanc- constraints and make managers automatical-
ing supplies with notional demands for ly accountable for their own choices was
commodities in short supply, transaction clearly enunciated: “If you supply [the
costs were limited by adapting effective buyer] everything on time in accordance
demands to the existing distribution. While with the contract, you will get the appropri-
limiting transaction costs, therefore, planning ate payment . . . If you don’t meet the obli-
from the achieved level also limited structur- gations that you have taken on, [the buyer]
al adaptation. Even worse, and contrary to won’t pay you, the bank won’t pay you on
what has been argued (Manove 1971), it did [the buyer’s] behalf any more, and you, dear
not necessarily eliminate large initial errors comrade, will have to have a very hard think
since it induced expectations to converge on about how to pay your wages, how to carry
initial allocations however inefficient. on the work at your factory” (cited by Davies
1996, p. 12: emphasis added). Some of
5.3 Obstacles to Reform
Orzhonikidze’s officials went still further,
In the early 1960s, Soviet officials began to advocating the hardening of investment
discuss openly whether it was possible to nest budget constraints by financing it on the
the informational and incentive advantages basis of repayable loans, and a considerable
of markets within the hierarchical structures degree of price liberalization.
of the command economy under the banner Three barriers to reform reinforced each
of socialist economic “reforms.” The archives other. First, Stalin and Molotov regarded
show, however, that they had been pursuing the retention of money and the return to
this quest behind closed doors from the early cost accounting after the chaos of 1929–30
1930s. The need to reform became evident as essentially temporary expedients. They
to insiders from the start, refuting the text- could have no intellectual sympathy for hard
book stereotype of a planning system that budget constraints, particularly for invest-
ment. Second, these reservations appeared
16
From a starting point of violent disagreement over justified by events. At the end of 1932,
the distribution of vehicles, the correlation coefficients Ordzhonikidze unexpectedly cancelled cen-
among the distribution plans prepared by the three agen- tralized equipment supply plans for the iron,
cies involved—producer, Gosplan, and the responsible
Politburo commission—rose to above 95 percent by the steel, coal, and oil industries for 1933, and
third quarter of 1933 despite their different agendas told producers and users to sort it out them-
(Lazarev and Gregory 2002, p. 336). The initial endow- selves. “This sudden freedom caused panic”
ment was essentially random; planning from the achieved
level then locked users into an “endowment” that drifted (Davies 1996, p. 269). On the one side,
incrementally in subsequent periods. equipment suppliers had no instructions on
how to ration supplies; in law they were First, the archives show a powerful but fal-
criminally liable if they refused an order. On lible dictatorship comprised of Stalin and his
the other hand, the equipment purchasers, Politburo immersed in the detail of econom-
freed from all restraint, tried to place orders ic decisions. They guided the general direc-
that were vastly inflated, but succeeded in tion of the economy by infrequent major
placing only a small fraction. Still committed decisions that set aggregate investment,
to the reform, Ordzhonikidze turned the which they poorly controlled, while reserving
balancing of supply and demand over to a the right to make unlimited detailed inter-
conference of industry representatives in ventions in current operations. The com-
Moscow. But, in the spring of 1933, mand system that resulted featured extreme
the Politburo instigated the dismissal of centralization and the priority of vertical over
the most radical reformists, leaving horizontal relationships. Minor decisions
Ordzhonikidze frustrated and humiliated. were delegated from top to bottom through a
A third reason for reform failure lay under- hierarchy of “nested” dictatorship. Formal
neath the other two: “With the high level of rules were avoided in favor of ad hoc decision
investment which obtained in 1932, such a making. A compliant planning board was dis-
reformed system was quite impracticable, engaged from responsibility for detailed allo-
and it may have been entirely incompatible cation to guarantee its loyalty; the dictator
with Soviet economic objectives” (Davies was particularly loathe to delegate economic
1996, p. 267). If so, then Ordzhonikidze was decisions to politically unreliable tech-
at fault in failing to understand this. nocrats. As a result, delegation did not work
The archives reveal that the reform stale- well: subordinates funneled even trivial deci-
mate of the post-Stalin years dates to the sions upwards to limit their own exposure,
beginning of the Stalin era. Agents could not placing a “dictator’s curse” of excessive
trust principals not to make mistakes, point- administrative burdens on their superiors at
ed to stupid, contradictory, changing orders each levels, most heavily on Stalin himself.
and excessive meddling, and demanded Second, in the context of the great increase
greater autonomy. But when they won in investment in the 1930s, the record of
autonomy they behaved opportunistically decision making in the Politburo supports
and sought rents rather than profits. the hypothesis of a dictator interested in
Principals could not trust agents to behave investment, growth, and efficiency, while bal-
altruistically and eventually had to intervene ancing these against other objectives such as
to curb rent-seeking and restore order. retaining loyalty and avoiding revolts. His
Agents behaved the way they did because management of investment required unflag-
they rationally expected principals to inter- ging attention to the effect of consumption
vene. Living in a nested dictatorship, they on worker incentives. The fear of consump-
saw no role for the dictators at every level tion falling to a point where worker unrest
other than to dictate. As long as they expect- would spread contributes to explaining the
ed this there was little point in looking for cyclical behavior of investment.
financial savings that were could be confis- Third, the archival literature sheds consid-
cated at any time. Better to go on hoarding, erable light on limits on the power of the
concealing, and bargaining. dictator to extract a surplus through large-
scale coercion. Frightened and intimidated
people still did not do as they were told;
6. What’s New?
rather, they invested all the more in horizon-
What do recent studies of the Soviet tal transactions that protected them from
archives offer to the political economy of repression and diverted effort from planned
command systems and dictatorship? goals. One of the most surprising results of
research on the Gulag is that its own admin- planning from the achieved level, a custom-
istrators regarded it as an economic and ary rule which introduced order at the
social failure, and that its supreme adminis- expense of freezing initial allocations and
trator was the major force pushing for its liq- inhibiting adaptation.
uidation shortly after Stalin’s death. Sixth, the dictator struck a deal with func-
Coercion was scaled down not because new, tional agencies, such as the State Planning
more principled leaders came to power but Commission, various control commissions,
because the old unprincipled leaders knew it and, above all, the secret police that he
did not work. This insight raises an interest- would not hold them responsible for final
ing problem: if, as many have argued, the results if they provided honest information.
Soviet-type system cannot function without But the information available to the
coercion, what should it do when coercion Politburo, although far greater in quantity,
also does not work? suffered from the same flaws as that which
Fourth, having created a surplus, Stalin was published. Most information was pro-
had to prevent vigorous rent seeking from vided by the producers themselves, who had
politicizing its distribution. Special branch incentives to exaggerate inputs needs and to
and regional interests formed immediately understate capacity. Even the secret police
within the dictator’s own circle, leaving had an incentive to exaggerate dangers to
Stalin and a few associates to battle for what enhance their claim on resources. The dicta-
they viewed as encompassing interests. Even tor’s honest information brokers could at
with access to virtually all official docu- best perform infrequent audits. This infor-
ments, we cannot measure the degree to mation flowed with difficulty because pro-
which the top leadership allocated invest- cessing capacity was limited, and because
ment to purchase loyalty rather than for eco- information was monopolized and traded for
nomic rationality. Stalin may have been private gain rather than freely shared. Good
more tolerant of rent seeking in earlier peri- information was so hard to get that planning
ods when his regime was still fragile. relied more on intuition than on the “scien-
Political motives may help explain the alloca- tific” methodology of Soviet handbooks.
tion of regional investment and motor vehi- Huge blunders resulted. Where decisions
cles in the early 1930s, but Stalin seems to have appeared wise in retrospect, for exam-
have stamped on rent seekers when he ple, in defense technology, they often relied
detected them in military matters. on information from abroad and mecha-
Fifth, the industrial archives contribute to nisms such as scientific reputation that did
debate about whether there existed anything not depend on state enforcement.
that we should call “planning.” All so-called Seventh, the soft budget constraint originat-
operational plans of the Stalin era were pro- ed historically in the Stalinist political commit-
visional and subject to change by any superi- ment to mobilization and was perpetuated by
or. Faced with provisional plans and the centralized institutions that prevented ex ante
prospect of endless interventions, enterprises commitments to financial discipline from
and ministries sometimes refused to commit being implemented after the event. Notably,
to plans; some enterprises operated without subsidies were not extended automatically to
“plans” for years. Resource allocation by loss makers without lobbying investments,
intervention rather than by plan is consistent which helps explain the unexpected interest of
with the dictator’s aversion to formal rules. A producers in higher prices and money, includ-
“final” plan represents a formal rule that ing traceable bank money. The archives pro-
could prevent superiors from exercising vide no support for the proposition that Soviet
“resource mobility.” The chaos of ad hoc enterprises sought lower prices to maximize
interventions then explains the attraction of bribes; rather enterprises took advantage of
the seller’s markets to extract higher prices Haven: Yale University Press.
Amann, Ronald, and Julian Cooper, eds. 1986.
even from influential buyers such as the Technical Progress and Soviet Economic
defense ministry. Corruption existed but the Development. Oxford: Blackwell.
archives leave the impression that most ille- Amann, Ronald, Julian Cooper, and R. W. Davies, eds.
1977. The Technological Level of Soviet Industry.
gal siphoning was done to fulfill the plan, not New Haven: Yale University Press.
for private gain. Ames, Edward. 1965. Soviet Economic Processes.
Eighth, the pervasiveness of markets in Homewood, Ill.: Irwin.
Bardhan, Pranab K., and John E. Roemer. 1993.
which even high-level administrators (such Market Socialism: The Current Debate. New York:
as ministers) participated reflects the dilem- Oxford University Press.
ma of the dictator who requires subordi- Barone, Enrico. 1908/1935. “Il Ministro della
Produzione nello Stato Collettivista,” translated as
nates to solve their own problems “at any “The Ministry of Production in the Collectivistic
price,” including by recourse to informal State,” in Collectivist Economic Planning: Critical
markets, while the networks that formed as Studies on the Possibility of Socialism. F. A. Hayek,
ed. London: Routledge, 245–90.
a result had the potential to undermine dic- Becker, Gary S. 1968. “Crime and Punishment: An
tatorial power. These markets did not work Economic Approach.” Journal of Political Economy,
well, were severely circumscribed, and 76(2): 169–217.
Belova, Eugenia. 2001. “Economic Crime and
relied heavily on relational enforcement; Punishment,” in Behind the Façade of Stalin’s
private contracts were not legally enforce- Command Economy: Evidence from the State and
able and reputations could spread only Party Archives. Paul R. Gregory, ed. Stanford:
Hoover Institution Press, 131–58.
within narrow circles. Belova, Eugenia. 2005. “Legal Contract Enforcement
Ninth and finally, the flaws of the planned in the Soviet Economy.” Comparative Economic
economy became immediately apparent to Studies, 47(2): 387–401.
Belova, Eugenia, and Paul Gregory. 2002. “Dictator,
its leaders, prompting reform proposals to Loyal, and Opportunistic Agents: The Soviet
increase enterprise independence. But the Archives on Creating the Soviet Economic System.”
results of partial reforms were if anything Public Choice, 113(3–4): 265–86.
Berelovich, A., ed. 2000. Sovetskaia derevnia glazami
worse than those of the unreformed system. VChK-OGPU-NKVD, 2: 1923–1929. Moscow:
Enterprises might prefer to circumvent the ROSSPEN.
apparently arbitrary, even stupid interven- Berelovich, A., and V. P. Danilov, eds. 1998. Sovetskaia
derevnia glazami VChK-OGPU-NKVD, 1:
tions of superiors; but the same enterprises, 1918–1922. Moscow: ROSSPEN.
unconstrained by the discipline of market Bergson, Abram. 1947. “A Problem in Soviet Statistics.”
forces, could not be trusted with independ- Review of Economics and Statistics, 29(4): 234–42.
Bergson, Abram. 1953. “Reliability and Usability of
ence. The resulting stalemate continued to Soviet Statistics: A Summary Appraisal.” American
the last days of the Soviet Union. Statistician, 7(3): 13–16.
Bergson, Abram. 1961. The Real National Income of
REFERENCES Soviet Russia Since 1928. Cambridge: Harvard
Acemoglu, Daron, and James A. Robinson. 2000. University Press.
“Political Losers as a Barrier to Economic Bergson, Abram. 1964. The Economics of Soviet
Development.” American Economic Review, 90(2): Planning. New Haven: Yale University Press.
126–30. Bergson, Abram. 1966. Essays in Normative
Akerlof, George A. 1984. “Gift Exchange and Economics. Cambridge: Harvard University Press.
Efficiency-Wage Theory: Four Views.” American Bergson, Abram. 1978. Productivity and the Social
Economic Review, 74(2): 79–83. System: The USSR and the West. Cambridge:
Akerlof, George A., and Janet L. Yellen. 1990. “The Harvard University Press.
Fair Wage–Effort Hypothesis and Unemployment.” Bergson, Abram. 1987. “On Soviet Real Investment
Quarterly Journal of Economics, 105(2): 255–83. Growth.” Soviet Studies, 39(3): 406–24.
Amann, Ronald. 1986. “Technical Progress and Soviet Berliner, Joseph S. 1952. “The Informal Organization
Economic Development: Setting the Scene,” in of the Soviet Firm.” Quarterly Journal of Economics,
Technical Progress and Soviet Economic 66(3): 342–65.
Development. Ronald Amann and Julian Cooper, Berliner, Joseph S. 1957. Factory and Manager in the
eds. Oxford: Blackwell, 5–30. USSR. Cambridge: Harvard University Press.
Amann, Ronald, and Julian Cooper, eds. 1982. Berliner, Joseph S. 1976. The Innovation Decision in
Industrial Innovation in the Soviet Union. New Soviet Industry. Cambridge: MIT Press.
Birman, Igor. 1978. “From the Achieved Level.” Soviet Gregory, ed. Stanford: Hoover Institution Press,
Studies, 30(2): 153–72. 61–80.
Birman, Igor. 1981. Secret Incomes of the Soviet State Davies, R. W. 2001b. “Why Was There a Soviet
Budget. London: Nijhoff. Investment Cycle in 1933–37?” University of
Blanchard, Olivier, and Michael Kremer. 1997. Warwick. PERSA Working Paper No. 16.
“Disorganization.” Quarterly Journal of Economics, Davies, R. W. 2003. “The Archives and the Stalinist
112(4): 1091–1126. Economy.” University of Warwick. PERSA Working
Boettke, Peter J. 2001. Calculation and Coordination: Paper No. 27.
Essays on Socialism and Transitional Political Davies, R. W. 2004. “The Soviet Economy and the
Economy. London: Routledge. Launching of the Great Terror.” University of
Bornstein, Morris. 1985. “Improving the Soviet Warwick. PERSA Working Paper No. 29.
Economic Mechanism.” Soviet Studies, 37(1): 1–30. Davies, R. W., and Mark Harrison. 1997. “The Soviet
Borodkin, Leonid, and Simon Ertz. 2003. “Coercion Military–Economic Effort under the Second Five-
versus Motivation: Forced Labor in Norilsk,” in The Year Plan 1933–1937.” Europe–Asia Studies, 49(3):
Economics of Forced Labor: The Soviet Gulag. Paul 369–406.
R. Gregory and Valery Lazarev, eds. Stanford: Davies, R. W., Melanie Ilič, and Oleg Khlevnyuk. 2004.
Hoover Institution Press, 75–104. “The Politburo and Economic Policy-Making,” in
Borodkin, Leonid, and Simon Ertz. 2005. “Forced The Nature of Stalin’s Dictatorship: The Politburo,
Labor and the Need for Motivation: Wages and 1924–1953. E. A. Rees, ed. Basingstoke: Palgrave,
Bonuses in the Stalinist Camp System.” Comparative 108–34.
Economic Studies, 47(2): 418–36. Davies, R. W., and Oleg Khlevnyuk. 1997. “Gosplan,”
Carr, Edward H., and R. W. Davies. 1969. A History of in Decision-Making in the Stalinist Command
Soviet Russia 3: Foundations of a Planned Economy, Economy, 1932–37. E. A. Rees, ed. Basingstoke and
1926–1929, Part 1. London: Macmillan. London: Macmillan, 32–66.
Chuev, F. I. 1991. Sto sorok besed s Molotovym. Davies, R. W., and Oleg Khlevnyuk. 1999. “The End of
Moscow: Terra. Rationing in the Soviet Union, 1934–1935.”
Chuev, F. I. 1992. Tak govoril Kaganovich. Moscow: Europe–Asia Studies, 51(4): 557–610.
Otechestvo. Davies, R. W., and Oleg Khlevnyuk. 2002.
Clark, Colin. 1939. A Critique of Russian Statistics. “Stakhanovism and the Soviet Economy.”
London: Macmillan. Europe–Asia Studies, 54(6): 867–903.
Conquest, Robert. 1967. Industrial Workers in the Davies, R. W., Oleg V. Khlevnyuk, E. A. Rees,
USSR. London: The Bodley Head. Liudmilla P. Kosheleva, and Larisa A. Rogovaia, eds.
Conquest, Robert. 1987. The Harvest of Sorrow: Soviet 2003. The Stalin–Kaganovich Correspondence,
Collectivization and the Terror-Famine. Oxford: 1931–36. New Haven: Yale University Press.
Oxford University Press. Davies, R. W., Mark B. Tauger, and Stephen G.
Danilov, Viktor P., Roberta T. Manning, and Lynne Wheatcroft. 1995. “Stalin, Grain Stocks and the
Viola, eds. 1999. Tragediia sovetskoi derevni. Famine of 1932–1933.” Slavic Review, 54(3): 642–57.
Kollektivizatsiia i raskulachivanie. 1927–1939. Davies, R. W. and Stephen G. Wheatcroft. 2004. The
Dokumenty i materialy. Moscow: ROSSPEN. Industrialization of Soviet Russia, 5: The Years of
Davies, R. W. 1980a. The Industrialization of Soviet Hunger: Soviet Agriculture, 1931–1933. Basingstoke:
Russia, 1: The Socialist Offensive: The Palgrave.
Collectivization of Soviet Agriculture, 1929–1930. Dewatripont, Mathias, and Eric Maskin. 1995. “Credit
Basingstoke: Macmillan. and Efficiency in Centralized and Decentralized
Davies, R. W. 1980b. The Industrialization of Soviet Economies.” Review of Economic Studies, 62(4):
Russia, 2: The Soviet Collective Farm, 1929–1930. 541–55.
Basingstoke: Macmillan. Djankov, Simeon, Edward Glaeser, Rafael la Porta,
Davies, R. W. 1989a. The Industrialization of Soviet Florencio Lopez-de-Silanes, and Andrei Shleifer.
Russia, 3: The Soviet Economy in Turmoil, 2003. “The New Comparative Economics.” Journal
1929–1930. Basingstoke: Macmillan. of Comparative Economics, 31(4): 595–619.
Davies, R. W. 1989b. Soviet History in the Gorbachev Duhamel, Luc. 2004. “The Last Campaign Against
Revolution. Basingstoke: Macmillan. Corruption in Soviet Moscow.” Europe–Asia Studies,
Davies, R. W. 1993. “Soviet Military Expenditure and 56(2): 187–212.
the Armaments Industry, 1929–33: A Easterly, William, and Stanley Fischer. 1995. “The
Reconsideration.” Europe–Asia Studies, 45(4): Soviet Economic Decline.” World Bank Economic
577–608. Review, 9(3): 341–71.
Davies, R. W. 1996. The Industrialization of Soviet Ellman, Michael. 1972. Soviet Planning Today.
Russia, 4: Crisis and Progress in the Soviet Economy, Cambridge: Cambridge University Press.
1931–1933. Basingstoke: Macmillan. Ellman, Michael. 1975. “Did the Agricultural Surplus
Davies, R. W. 1997. Soviet History in the Yeltsin Era. Provide the Resources for the Increase in
Basingstoke: Macmillan. Investment in the USSR during the First Five Year
Davies, R. W. 2001a. “Making Economic Policy,” in Plan?” Economic Journal, 85(340): 844–63.
Behind the Façade of Stalin’s Command Economy: Ellman, Michael. 1979. Socialist Planning. Cambridge:
Evidence from the State and Party Archives. Paul R. Cambridge University Press.
Ellman, Michael. 2000. “The 1947 Soviet Famine and Council of Ministers and the Soviet Neopatrimonial
the Entitlement Approach to Famines.” Cambridge State, 1946–1953.” Journal of Modern History, 74(4):
Journal of Economics, 24(5): 603–30. 699–736.
Ericson, Richard E. 1983. “On an Allocative Role of Gorlizki, Yoram, and Oleg Khlevnyuk. 2004. Cold
the Soviet Second Economy,” in Marxism, Central Peace: Stalin and the Soviet Ruling Circle,
Planning, and the Soviet Economy: Economic Essays 1945–1953. New York: Oxford University Press.
in Honor of Alexander Erlich. Padma Desai, ed. Granick, David. 1954. Management of the Industrial
Cambridge: MIT Press, 110–32. Firm in the USSR. New York: Columbia University
Ericson, Richard E. 1984. “The ‘Second Economy’ and Press.
Resource Allocation under Central Planning.” Granick, David. 1987. Job Rights in the Soviet Union:
Journal of Comparative Economics, 8(1): 1–24. Their Consequences. New York: Cambridge
Erlich, Alexander. 1960. The Soviet Industrialization University Press.
Debate, 1924–1928. Cambridge: Harvard University Gregory, Paul R. 2001. “The Dictator’s Orders,” in
Press. Behind the Façade of Stalin’s Command Economy:
Ertz, Simon, 2003. “Building Norilsk,” in The Evidence from the State and Party Archives. Paul
Economics of Forced Labor: The Soviet Gulag. Paul Gregory, ed. Stanford: Hoover Institution Press,
R. Gregory and Valary Lazarev, eds. Stanford: 11–33.
Hoover Institution Press, 127–50. Gregory, Paul R. 2003a. “An Introduction to the
Ertz, Simon. 2005. “Trading Effort for Freedom: Economics of the Gulag,” in The Economics of
Workday Credits in the Soviet Camp System.” Forced Labor: The Soviet Gulag. Paul R. Gregory
Comparative Economic Studies, 47(2): 476–91. and Valery Lazarev, eds. Stanford: Hoover
Filtzer, Don. 2002. Soviet Workers and Late Stalinism: Institution Press, 1–21.
Labour and the Restoration of the Stalinist System Gregory, Paul R. 2003b. “Soviet Defence Puzzles:
after World War II. Cambridge: Cambridge Archives, Strategy, and Underfulfillment.” Europe–
University Press. Asia Studies, 55(6): 923–38.
Fitzpatrick, Sheila. 1999. Everyday Stalinism: Gregory, Paul R. 2004. The Political Economy of
Ordinary Life in Extraordinary Times: Soviet Russia Stalinism: Evidence from the Soviet Secret Archives.
in the 1930s. Oxford: Oxford University Press. New York: Cambridge.
Ganelin, R. S., ed. 2004. Istoriia sozdaniia i razvitiia Gregory, Paul R., and Andrei Markevich. 2002.
oboronnopromyshlennogo kompleksa Rossii i SSSR. “Creating Soviet Industry: The House That Stalin
1901–1963 gg. Dokumenty i materialy, 1: Voennaia Built.” Slavic Review, 61(4): 787–814.
promyshlennost’ Rossii v nachale XX veka Gregory, Paul R., and Robert C. Stuart. 1974. Soviet
(1901–1918 gg.). Moscow: Novyi Khronograf. Economic Structure and Performance. New York:
Gerschenkron, Alexander. 1947. “The Soviet Indices of Harper and Row.
Industrial Production.” Review of Economics and Gregory, Paul R., and Aleksei Tikhonov. 2000. “Central
Statistics, 29(4): 217–26. Planning and Unintended Consequences: Creating
Getty, J. Arch, and Oleg Naumov. 1999. The Road to the Soviet Financial System, 1930–1939.” Journal of
Terror: Stalin and the Self-Destruction of the Economic History, 60(4): 1017–40.
Bolsheviks, 1932–1939. New Haven: Yale University Greif, Avner. 1993. “Contract Enforceability and
Press. Economic Institutions in Early Trade: The Maghribi
Glaeser, Edward L., Rafael La Porta, Florencio Lopez- Traders’ Coalition.” American Economic Review,
de-Silanes, and Andrei Shleifer. 2004. “Do 83(3): 525–48.
Institutions Cause Growth?” NBER Working Paper Greif, Avner. 1994. “Cultural Beliefs and the
10568. Organization of Society: A Historical and Theoretical
Goldfeld, Stephen M., and Richard E. Quandt. 1988. Reflection on Collectivist and Individualist Societies.”
“Budget Constraints, Bailouts, and the Firm under Journal of Political Economy, 102(5): 912–50.
Central Planning.” Journal of Comparative Greif, Avner. 2000. “The Fundamental Problem of
Economics, 12(4): 502–20. Exchange: A Research Agenda in Historical
Goldfeld, Stephen M., and Richard E. Quandt. 1990. Institutional Analysis.” European Review of
“Output Targets, the Soft Budget Constraint and the Economic History, 4(3): 251–84.
Firm under Central Planning.” Journal of Economic Greif, Avner, Paul Milgrom, and Barry R. Weingast.
Behavior and Organization, 14(2): 205–22. 1995. “Coordination, Commitment, and Enforce-
Gomulka, Stanislaw. 1985. “Kornai’s Soft Budget ment: The Case of the Merchant Guild.” Journal of
Constraint and the Shortage Phenemenon: A Political Economy, 102(4): 745–76.
Criticism and Restatement.” Economics of Planning, Grossman, Gregory. 1953. “Scarce Capital and Soviet
19(1): 1–11. Doctrine.” Quarterly Journal of Economics, 67(3):
Gorlizki, Yoram. 2001a. “Stalin’s Cabinet: The 311–43.
Politburo and Decision-Making in the Post-War Hanson, Philip. 1983. “Success Indicators Revisited: The
Years.” Europe–Asia Studies, 53(2): 291–312. July 1979 Decree on Planning and Management.”
Gorlizki, Yoram. 2001b. “Theft Under Stalin: A Soviet Studies, 35(1): 1–13.
Property Rights Analysis.” University of Warwick. Hanson, Philip. 1984. “The CIA, the TsSU, and the
PERSA Working Paper No. 10. Real Growth of Soviet Investment.” Soviet Studies,
Gorlizki, Yoram. 2002. “Ordinary Stalinism: The 86(4): 571–81.
Hanson, Philip, and Keith Pavitt. 1987. The Economic Studies, 47(2): 456–75.
Comparative Economics of Research, Development Hessler, Julie. 2004. A Social History of Soviet Trade:
and Innovation in East and West: A Survey. Chur, Trade Policy, Retail Practices, and Consumption,
Switzerland: Harwood. 1917–1953. Princeton: Princeton University Press.
Harris, James R. 1999. The Great Urals: Regionalism Holloway, David. 1994. Stalin and the Bomb: The
and the Evolution of the Soviet System. Ithaca: Soviet Union and Atomic Energy, 1939–1956. New
Cornell University Press. Haven: Yale University Press.
Harrison, Mark. 1994. “National Income,” in The Jasny, Naum. 1949. The Socialized Agriculture of the
Economic Transformation of the Soviet Union, USSR: Plans and Performance. Stanford: Stanford
1913–45. R. W. Davies, Mark Harrison, and S. G. University Press.
Wheatcroft, eds. Cambridge: Cambridge University Joyce, Christopher. 2003. “The Gulag in Karelia:
Press, 38–56. 1929–1941,” in The Economics of Forced Labor: The
Harrison, Mark. 1998. “Prices, Planners, and Producers: Soviet Gulag. Paul R. Gregory and Valery Lazarev,
An Agency Problem in Soviet Industry, 1928–1950.” eds. Stanford: Hoover Institution Press, 163–88.
Journal of Economic History, 58(4): 1032–62. Junker, James. 1992. Socialism Revised and
Harrison, Mark. 2000. “New Postwar Branches (1): Modernized: The Case for Pragmatic Market
Rocketry,” in The Soviet Defence-Industry Complex Socialism. New York: Praeger.
from Stalin to Khruschev. John Barber and Mark Katz, Barbara G. 1975. “Purges and Production: Soviet
Harrison, eds. Basingstoke: Macmillan, 118–49. Economic Growth, 1928–1940.” Journal of
Harrison, Mark. 2002. “Coercion, Compliance, and the Economic History, 35(3): 567–90.
Collapse of the Soviet Command Economy.” Keren, Michael. 1982. “The Ministry, Plan Changes,
Economic History Review, 55(3): 397–433. and the Ratchet in Planning.” Journal of
Harrison, Mark. 2003a. “The Political Economy of a Comparative Economics, 6(4): 327–42.
Soviet Military R&D Failure: Steam Power for Khlevnyuk, Oleg V. 1993. Stalin i Ordzhonikidze.
Aviation, 1932 to 1939.” Journal of Economic Konflikty v Politburo v 30-e gody. Moscow: Rossiia
History, 63(1): 178–212. molodaia.
Harrison, Mark. 2003b. “Soviet Industry and the Red Khlevnyuk, Oleg V. 1995. “The Objectives of the Great
Army under Stalin: A Military–Industrial Complex?” Terror, 1937–38,” in Soviet History, 1917–1953:
Les Cahiers du Monde Russe, 44(2–3): 323–42. Essays in Honour of R. W. Davies. J. M. Cooper,
Harrison, Mark. 2004. “Why Secrets? The Uses of Maureen Perrie, and E. A. Rees, eds. New York: St.
Secrecy in Stalin’s Command Economy.” University Martin’s Press, 158–76.
of Warwick. PERSA Working Paper No. 34. Khlevnyuk, Oleg V. 1996. Politburo. Mekhanizmy
Harrison, Mark. 2005. “The Fundamental Problem of politchecheskoi vlasti v 1930-e gody. Moscow:
Command: Plan and Compliance in a Partially ROSSPEN.
Centralized Economy.” Comparative Economic Khlevnyuk, Oleg V. 2000. “Sovetskaia ekonomicheskaia
Studies, 47(2): 296–314. politika na rubezhe 40–50 godov i delo Gosplana.”
Harrison, Mark. Forthcoming. “A Soviet Quasi-Market Paper presented at a conference on “Stalin’s
for Inventions: Jet Propulsion, 1932 to 1946.” Politburo,” Florence, Italy, March 29–31, 2000.
Research in Economic History. Khlevnyuk, Oleg V. 2001a. “Stalinism and the Stalin
Harrison, Mark, and Byung-Yeon Kim. Forthcoming. Period after the ‘Archival Revolution’.” Kritika, 2(2):
“Plans, Prices, and Corruption: The Soviet Firm 319–28.
Under Partial Centralization, 1930 to 1990.” Journal Khlevnyuk, Oleg V. 2001b. “The Economy of the
of Economic History. Gulag,” in Behind the Façade of Stalin’s Command
Harrison, Mark, and Nikolai Simonov. 2000. Economy: Evidence from the State and Party
“Voenpriemka: Prices, Costs, and Quality Assurance Archives. Paul R. Gregory, ed. Stanford: Hoover
in Interwar Defence Industry,” in The Soviet Institution Press, 111–30.
Defence-Industry Complex From Stalin to Khlevnyuk, Oleg V. 2003a. “The Economy of the
Khrushchev. John Barber and Mark Harrison, eds. OGPU, NKVD, and MVD of the USSR, 1930–1953:
Basingstoke: Macmillan, 223–45. The Scale, Structure, and Trends of Development,”
Hayek, F. A. 1937. “Economics and Knowledge.” in The Economics of Forced Labor: The Soviet
Economica, 4(13): 33–54. Gulag. Paul R. Gregory and Valery Lazarev, eds.
Hayek, F. A. 1940. “Socialist Calculation: The Stanford: Hoover Institution Press, 43–66.
Competitive ‘Solution’.” Economica, 7(26): 125–49. Khlevnyuk, Oleg V. 2003b. “Sistema tsentr-regiony v
Hayek, F. A. 1944. The Road to Serfdom. Chicago: 1930–1950-e gody. Predposylki politizatsii nomen-
University of Chicago Press. klatury.” Les Cahiers du Monde Russe, 44(2–3):
Hayek, F. A. 1945. “The Use of Knowledge in Society,” 253–68.
American Economic Review, 35(4): 519–30. Khlevnyuk, Oleg V. 2004. “Vzaimootnesheniia tsentra i
Heinzen, James W. 2004. Inventing a Soviet regionov v 30-e gg.” Vestnik Moskovskogo
Countryside: State Power and the Transformation of Universiteta, 21(3): 79–98.
Rural Russia, 1917–1929. Pittsburgh: University of Khlevnyuk, Oleg V., R. W. Davies, L. P. Kosheleva, E.
Pittsburgh Press. A. Rees, and L. A. Rogovaia, eds. 2001. Stalin i
Heinzen, James W. 2005. “Corruption in the Gulag: Kaganovich. Perepiska. 1931–1936 gg. Moscow:
Dilemmas of Officials and Prisoners.” Comparative ROSSPEN.
Khlevnyuk, Oleg V., Yoram Gorlizki, L. P. Kosheleva, A. from Stalin to Khruschev. John Barber and Mark
I. Miniuk, M. I. Prozumenshchikov, L. A. Rogovaia, Harrison, eds. Basingstoke: Macmillan, 184–94.
and S. V. Somonova, eds. 2002. Politburo TsK VPK(b) Lewin, Moshe. 1968. Russian Peasants and Soviet
i Sovet Ministrov SSSR. 1945–1953. Moscow: Power: A Study of Collectivization. London: Allen
ROSSPEN. and Unwin.
Khlevnyuk, Oleg V., A. V. Kvashonkin, L. P. Kosheleva, Lewin, Moshe. 2003. “Rebuilding the Soviet
and L. A. Rogovaia, eds. 1995. Stalinskoe Politburo v Nomenklaturas, 1945–1948.” Les Cahiers du Monde
30-e gody. Sbornik dokumentov. Moscow: AIRO-XX. Russe, 44(2–3): 219–52.
Kim, Byung-Yeon. 1997. “Soviet Household Saving Lih, Lars, Oleg Naumov, and Oleg Khlevnyuk. 1995.
Function.” Economics of Planning, 30(2–3): 181–203. Stalin’s Letters to Molotov, 1925–1936. New Haven:
Kim, Byung-Yeon. 1999. “The Income, Savings, and Yale University Press.
Monetary Overhang of Soviet Households.” Journal Manning, Roberta T. 1993. “The Soviet Economic
of Comparative Economics, 27(4): 644–68. Crisis of 1936–1940 and the Great Purges,” in
Kim, Byung-Yeon. 2002. “Causes of Repressed Inflation Stalinist Terror: New Perspectives. J. Arch Getty and
in the Soviet Consumer Market, 1965–1989: Retail Roberta T. Manning, eds. Cambridge: Cambridge
Price Subsidies, the Siphoning Effect, and the Budget University Press, 116–41.
Deficit.” Economic History Review, 55(1): 105–27. Manove, Michael. 1971. “A Model of Soviet-Type
Kontorovich, Vladimir. 1986. “Discipline and Growth Economic Planning.” American Economic Review,
in the Soviet Economy.” Problems of Communism, 61(3): 390–406.
34(6): 18–31. Markevich, Andrei. 2003. “Was the Soviet Economy
Kontorovich, Vladimir. 1988. “Lessons of the 1965 Planned? Planning in the People’s Commissariats in
Soviet Economic Reform.” Soviet Studies, 40(2): the 1930s.” University of Warwick. PERSA Working
308–16. Paper No. 25.
Kornai, János. 1980. The Economics of Shortage. Markevich, Andrei. 2005. “Soviet Planning Archives:
Amsterdam: North-Holland. The Files that Bergson Could Not See.”
Kornai, János. 1992. The Socialist System: The Political Comparative Economic Studies, 47(2): 364–86.
Economy of Communism. Princeton: Princeton Markevich, Andrei, and Mark Harrison. Forthcoming.
University Press. “Quality, Experience, and Monopoly: The Soviet
Kornai, János, Eric Maskin, and Gerard Roland. 2003. Market for Weapons under Stalin.” Economic
“Understanding the Soft Budget Constraint.” History Review.
Journal of Economic Literature, 41(4): 1095–1136. Maskin, Eric S. 1996. “Theories of the Soft Budget-
Kozlov, V. P., ed. 2004. Istoriia Stalinskogo Gulaga. Constraint.” Japan and the World Economy, 8(2):
Moscow: ROSSPEN. 125–33.
Kroll, Heidi. 1986. “Decentralization and Precontract Maskin, Eric S. 1999. “Recent Theoretical Work on the
Disputes in Soviet Industry.” Soviet Economy, 2(1): Soft Budget Constraint.” American Economic
51–71. Review, 89(2): 421–25.
Kroll, Heidi. 1988. “The Role of Contracts in the Soviet Millar, James R. 1974. “Mass Collectivization and the
Economy.” Soviet Studies, 40(3): 349–66. Contribution of Soviet Agriculture to the First Five-
Kuromiya, Hiroaki. 1988. Stalin’s Industrial Year Plan: A Review Article.” Slavic Review, 33(4):
Revolution: Politics and Workers, 1928–1932. 750–66.
Cambridge: Cambridge University Press. von Mises, Ludwig. 1920/1935. “Die Wirtschaftsrechnung
Kuznets, Simon. 1963. “A Comparative Appraisal,” in im sozialistischen Gemeinwesen,” translated in
Economic Trends in the Soviet Union. Abram Collectivist Economic Planning: Critical Studies on the
Bergson and Simon Kuznets, eds. Cambridge: Possibility of Socialism. F. A. Hayek, ed. London:
Harvard University Press, 333–82. Routledge, 87–130.
Lange, Oskar. 1964. On the Economic Theory of von Mises, Ludwig. 1949/1998. Human Action: A
Socialism. New York: McGraw-Hill. Treatise on Economics. Auburn, Ala.: Ludwig von
Lavoie, Don. 1985. Rivalry and Central Planning: The Mises Institute.
Socialist Calculation Debate Reconsidered. Montefiore, Simon Sebag. 2003. Stalin: The Court of
Cambridge: Cambridge University Press. the Red Tsar. London: Weidenfeld and Nicolson.
Lazarev, Valery. 2005. “Promotion Incentives and Morukov, Mikhail. 2003. “The White Sea–Baltic
Support for the Soviet Regime.” Comparative Canal,” in The Economics of Forced Labor: The
Economic Studies, 47(2): 346–63. Soviet Gulag. Paul R. Gregory and Valery Lazarev,
Lazarev, Valery, and Paul R. Gregory. 2002. “The eds. Stanford: Hoover Institution Press, 151–62.
Wheels of a Command Economy: Allocating Soviet Nordlander, David. 2003. “Magadan and the Economic
Vehicles.” Economic History Review, 55(2): 324–48. History of Dalstroi in the 1930s,” in The Economics
Lazarev, Valery, and Paul R. Gregory. 2003. of Forced Labor: The Soviet Gulag. Paul R. Gregory
“Commissars and Cars: A Case Study in the Political and Valery Lazarev, eds. Stanford: Hoover
Economy of Dictatorship.” Journal of Comparative Institution Press, 105–25.
Economics, 31(1): 1–19. Nove, Alec. 1958. “The Problem of ‘Success Indicators’
Lebina, Nataliia. 2000. “The Defence-Industry in Soviet Industry.” Economica, 25(97): 1–13.
Complex in Leningrad (2): The Postwar Uranium Nove, Alec. 1961. The Soviet Economy. London: Allen
Industry,” in The Soviet Defence-Industry Complex and Unwin.
Nove, Alec. 1969. An Economic History of the USSR. a Treadmill of ‘Reforms’,” in Soviet Economy in a
London: Allen and Unwin. Time of Change. U.S. Congress Joint Economic
Nove, Alec. 1977. The Soviet Economic System. Committee. Washington, D.C.: U.S. Government
London: Allen and Unwin. Printing Office, 312–40.
Ofer, Gur. 1987. “Soviet Economic Growth: Schroeder, Gertrude E. 1982. “Soviet Economic
1928–1985.” Journal of Economic Literature 25(4): ‘Reform’ Decrees: More Steps on the Treadmill,” in
1767–1833. Soviet Economy in the 1980s: Problems and
Olson, Mancur. 1993. “Dictatorship, Democracy, and Prospects. U.S. Congress Joint Economic
Development.” American Political Science Review, Committee. Washington, D.C.: U.S. Government
87(3): 567–76. Printing Office, 65–88.
Osokina, Elena. 2001. Our Daily Bread: Socialist Schroeder, Gertrude E. 1985. “The Slowdown in Soviet
Distribution and the Art of Survival in Stalin’s Industry, 1976–1982.” Soviet Economy, 1(1): 42–74.
Russia, 1927–1941. Armonk, NY: M. E. Sharpe. Sen, Amartya K. 1981. Poverty and Famines: An Essay
Podkaminer, Leon. 1989. “Theory of Investment Cycles on Entitlement and Deprivation. Oxford: Oxford
in the Centrally Planned Economies: An Economic University Press.
Interpretation.” Forschungberichte no. 158. Wiener Shleifer, Andrei, and Robert W. Vishny. 1992.
Institut für Internationaler Wirtschaftsvergleiche. “Pervasive Shortages under Socialism.” RAND
Rees, E. A., ed. 1997a. Decision-Making in the Stalinist Journal of Economics, 23(2): 237–46.
Economy, 1932–37. Basingstoke: Macmillan. Shleifer, Andrei, and Robert W. Vishny. 1994. “The
Rees, E. A. 1997b. “The People’s Commissariat of the Politics of Market Socialism.” Journal of Economic
Timber Industry,” in Decision-Making in the Perspectives, 8(2): 165–76.
Stalinist Economy, 1932–37. E. A. Rees, ed. Simonov, Nikolai S. 2000. “New Postwar Branches (2):
Basingstoke: Macmillan, 124–49. The Nuclear Industry,” in The Soviet Defence-
Rees, E. A. 2001. “Leaders and their Institutions,” in Industry Complex from Stalin to Khrushchev. John
Behind the Façade of Stalin’s Command Economy: Barber and Mark Harrison eds. Basingstoke:
Evidence from the State and Party Archives. Paul R. Macmillan, 150–72.
Gregory, ed. Stanford: Hoover Institution Press, Sokolov, Andrei K. 2003. “Forced Labor in Soviet
35–60. Industry: The End of the 1930s to the Mid-1950s: An
Rees, E. A., ed. 2004. The Nature of Stalin’s Overview,” in The Economics of Forced Labor: The
Dictatorship: The Politburo, 1924–1953. Basing- Soviet Gulag. Paul R. Gregory and Valery Lazarev,
stoke: Palgrave. eds. Stanford: Hoover Institution Press, 23–42.
Rees, E. A., and D. H. Watson. 1997. “Politburo and Sokolov, Andrei K. 2005. “Before Stalinism: The
Sovnarkom,” in Decision-Making in the Stalinist Defense Industry of Soviet Russia in the 1920s.”
Command Economy, 1932–37. E. A. Rees, ed. Comparative Economic Studies, 47(2): 437–55.
Basingstoke: Macmillan, 9–31. Spulber, Nicolas. 1964. Soviet Strategy for Economic
Rittersporn, Gábor T. 2001. “New Horizons: Growth. Bloomington, IN: Indiana University Press.
Conceptualizing the Soviet 1930s.” Kritika, 2(2): Stalin, I. V. 1937. Voprosy Leninizma. Moscow:
307–18. Gospolitizdat.
Rutland, Peter. 1985. The Myth of the Plan. LaSalle, Strauss, Erich. 1969. Soviet Agriculture in Perspective:
IL: Open Court. A Study of its Successes and Failures. London: Allen
Samuelson, Lennart. 2000. Plans for Stalin’s War and Unwin.
Machine: Tukhachevskii and Military–Economic Sutton, Antony C. 1968. Western Technology and
Planning, 1925–41. London: Macmillan. Soviet Economic Development, 1: 1917–1930.
Schaffer, Mark E. 1989. “The Credible-Commitment Stanford: Hoover Institution Press.
Problem in the Center–Enterprise Relationship.” Sutton, Antony C. 1971. Western Technology and
Journal of Comparative Economics, 13(3): 359–82. Soviet Economic Development, 2: 1930–1945.
Schelling, Thomas C. 1960. The Strategy of Conflict. Stanford: Hoover Institution Press.
Oxford: Oxford University Press. Sutton, Antony C. 1973. Western Technology and
Schrettl, Wolfram. 1982. Consumption, Effort, and Soviet Economic Development, 3: 1945–1960.
Growth in Soviet-Type Economies: A Theoretical Stanford: Hoover Institution Press.
Analysis. Ann Arbor: University Microfilms Swianiewicz, S. 1965. Forced Labour and Economic
International. Development: An Enquiry into the Experience of
Schrettl, Wolfram. 1984. “Anspruchdenken, Soviet Industrialization. London: Oxford University
Leistungbereitschaft und Wirtschaftszyklen,” in Press.
Wachstumsverlangsamung und Konjunkturzyklen in Tauger, Mark B. 1991. “The 1932 Harvest and the Soviet
Unterschiedlichen Wirtschaftssystemen. A. Schuller, Famine of 1932–1933.” Slavic Review, 50(1): 70–89.
ed. Berlin: Duncker and Humblot, 153–91. Tikhonov, Aleksei. 2003. “The End of the Gulag,” in
Schroeder, Gertrude E. 1973. “Recent Developments The Economics of Forced Labor: The Soviet Gulag.
in Soviet Planning and Incentives,” in Soviet Paul R. Gregory and Valery Lazarev, eds. Stanford:
Economic Prospects for the Seventies. U.S. Congress Hoover Institution Press, 67–74.
Joint Economic Committee. Washington, D.C.: U.S. United States Department of Commerce. 1960.
Government Printing Office, 11–38. Statistical Abstract of the United States. Washington,
Schroeder, Gertrude E. 1979. “The Soviet Economy on D.C.: U.S. Government Printing Office.
Vatlin, A. I. 2004. Terror raionnogo masshtaba. Russian History. S. G. Wheatcroft, ed. Basingstoke:
“Massovye operatsii” NKVD v Kuntsevskom raione Palgrave, 112–46.
Moskovskoi oblasti. 1937–1938 gg. Moscow: Wiles, Peter J. D. 1982. “The Soviet Investment
ROSSPEN. Deflator Again.” Soviet Studies, 34(4): 616–17.
Weitzman, Martin L. 1980. “The ‘Ratchet Principle’ Wintrobe, Ronald. 1998. The Political Economy of
and Performance Incentives.” Bell Journal of Dictatorship. Cambridge: Cambridge University
Economics, 11(1):302–08. Press.
Wheatcroft, Stephen G. 2004. “From Team Stalin to Zaleski, Eugène. 1971. Planning for Economic Growth
Degenerate Tyranny,” in The Nature of Stalin’s in the Soviet Union, 1918–1932. Chapel Hill:
Dictatorship: The Politburo, 1924–1953. E. A. Rees, University of North Carolina Press.
ed. Basingstoke: Palgrave, 79–107. Zaleski, Eugène. 1980. Stalinist Planning for Economic
Wheatcroft, Stephen G., and R. W. Davies, eds. 1985. Growth, 1933–1952. Chapel Hill: University of
Materials for a Balance of the Soviet National North Carolina Press.
Economy, 1928–1930. Cambridge: Cambridge Zolotarev, V. A., ed. Forthcoming. Istoriia sozdaniia i
University Press. razvitiia oboronno-promyshlennogo kompleksa
Wheatcroft, Stephen G., and R. W. Davies. 2002. “The Rossii i SSSR. 1901–1963 gg. Dokumenty i materi-
Soviet Famine of 1932–3 and the Crisis in aly, 2: Sovetskoe voenno-promyshlennoe proizvodst-
Agriculture,” in Challenging Traditional Views of vo (1918–1926). Moscow: Novyi Khronograt.

Allocation Under Dictatorship Research in Stalin's Archives - Paul Gregory

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Allocation Under Dictatorship Research in Stalin's Archives - Paul Gregory

Uploaded by

Copyright:

Available Formats

sp05_Article 2 8/18/05 9:39 AM Page 721

Journal of Economic Literature

Allocation under Dictatorship:

1. Introduction: A Well-Documented These were documents never intended for

722 Journal of Economic Literature, Vol. XLIII (September 2005)

Gregory and Harrison: Allocation under Dictatorship 723

724 Journal of Economic Literature, Vol. XLIII (September 2005)

producers pushed inexorably for higher 2. The Dictator, Hierarchy,

Gregory and Harrison: Allocation under Dictatorship 725

726 Journal of Economic Literature, Vol. XLIII (September 2005)

and Gregory 2002, p. 270). He placed hyper-centralization.” In short, interests

Gregory and Harrison: Allocation under Dictatorship 727

728 Journal of Economic Literature, Vol. XLIII (September 2005)

Gregory and Harrison: Allocation under Dictatorship 729

730 Journal of Economic Literature, Vol. XLIII (September 2005)

Gregory and Harrison: Allocation under Dictatorship 731

732 Journal of Economic Literature, Vol. XLIII (September 2005)

Gregory and Harrison: Allocation under Dictatorship 733

734 Journal of Economic Literature, Vol. XLIII (September 2005)

Gregory and Harrison: Allocation under Dictatorship 735

736 Journal of Economic Literature, Vol. XLIII (September 2005)

Gregory and Harrison: Allocation under Dictatorship 737

738 Journal of Economic Literature, Vol. XLIII (September 2005)

Gregory and Harrison: Allocation under Dictatorship 739

740 Journal of Economic Literature, Vol. XLIII (September 2005)

Gregory and Harrison: Allocation under Dictatorship 741

742 Journal of Economic Literature, Vol. XLIII (September 2005)

Gregory and Harrison: Allocation under Dictatorship 743

744 Journal of Economic Literature, Vol. XLIII (September 2005)

Gregory and Harrison: Allocation under Dictatorship 745

746 Journal of Economic Literature, Vol. XLIII (September 2005)

Gregory and Harrison: Allocation under Dictatorship 747

Figure 2. Price Renegotiation

748 Journal of Economic Literature, Vol. XLIII (September 2005)

sequential and the completion of or stealing. The returns to stealing, however,

Gregory and Harrison: Allocation under Dictatorship 749

750 Journal of Economic Literature, Vol. XLIII (September 2005)

Gregory and Harrison: Allocation under Dictatorship 751

752 Journal of Economic Literature, Vol. XLIII (September 2005)

Gregory and Harrison: Allocation under Dictatorship 753

754 Journal of Economic Literature, Vol. XLIII (September 2005)

Gregory and Harrison: Allocation under Dictatorship 755

756 Journal of Economic Literature, Vol. XLIII (September 2005)

Gregory and Harrison: Allocation under Dictatorship 757

758 Journal of Economic Literature, Vol. XLIII (September 2005)

Gregory and Harrison: Allocation under Dictatorship 759

760 Journal of Economic Literature, Vol. XLIII (September 2005)

Gregory and Harrison: Allocation under Dictatorship 761

You might also like