Looking through some old notes, I found these passages from Ronald Wintrobe’s book The Political Economy of Dictatorship. I think he offers a very clear explanation of why the history of the Soviet Union unfolded as it did.
These considerations reveal the basic contradiction of Communist rule. The ideological basis of communism is solidarity. In order to promote that solidarity, markets and private ownership are suppressed. But in order to make the system work, it has to function as a bureaucracy that is under political control. But in any bureaucratic system, vertical control is paramount, and solidarity among the work force interrupts this control and lowers output and productivity. This result is especially likely when the whole society is organized as a single bureaucratic system, such as in the former Soviet Union and other Communist countries. The more the system operated as any bureaucracy must, the more the contradiction between its reality and its promises, as embedded in its ideology, became apparent (p. 214).
[T]he classic Soviet system, like any bureaucracy, did not run primarily on orders or commands but on exchange. The basic difference between a bureaucratic and a market system is that exchanges within bureaucracy are based not on laws but on trust or loyalty. Under communism, loyalty to the Party combined with the Party’s capacity to repress opposition became the source of its power. Consequently, when the Party was strong, either because it was ruthless in its use of repression or because it was believed to be capable of fulfilling its promises, the system was capable of good economic performance. The fundamental prediction of this model is therefore that in a Soviet-style system, there is a positive correlation between the power of the Party and measures of economic performance such as economic growth.
The basic problem with such a system as an economic system lie in the conditions for running any large bureaucracy efficiently; bureaucracies require vertical or hierarchical loyalty and not horizontal solidarity among co-workers, which can be used to escape Party control and therefore tend to lower productivity. In turn, this implies that there is a fundamental contradiction between the promises of communism — essentially, equality and solidarity — and efficiency. . . . over time, this contradiction became more and more apparent, and the system could only maintain itself, Stalin-style, through the use of the purge and other techniques for breaking up the horizontal networks and other nonsanctioned alliances which tended to grow up within it (p. 217).
The central problem of any bureaucratic system . . . is that over time, horizontal trust (as well as vertical trust) tends to accumulate and the accumulation of horizontal trust is ultimately very damaging to the efficiency of the system from the point of view of the leaders. We would predict that this problem was particularly acute in the Soviet system, with the intertwining of the Communist party and the state, as well as the consequent absence of an institutionalized takeover mechanism (such as general elections in politics or hostile takeovers in business) or any other mechanism which could “shake up” the loyalties which tend to accumulate within such a system. Consequently, the only weapon available for this purpose was the purge (p. 225).