 |
| https://blogs.warwick.ac.uk/markharrison/entry/soviet_and_russian/ |
To situate the subject briefly, Cold War scholarship has left us a substantial literature on income inequality under communism. Bergson (1944), Yanowitch (1963), Wiles and Markowski (1971), Pryor (1972), Wiles (1974, 1975), Wädekin (1975), Chapman (1977), McAuley (1977), and Matthews (1978), each made valiant attempts, sometimes extending to piecemeal comparisons over countries and over time. “Considering the obscure data with which they had to work,” a survey by Schroeder (1983) remarked, “Western investigators display a large degree of agreement.” Measured by the decile ratio, the distribution of official incomes in the Soviet Union was becoming more equal over time and was substantially more equal than in the developed market economies then available as comparators. Schroeder noted, however, that Western researchers could not access data on the Soviet distribution of illegal incomes, or on privileged distribution of goods and services including accommodation and health care.
More recently, Lindert and Nafziger (2014) made an advance in another direction, examining inequality in Russia before and after the Soviet era. They concluded that pre-tax income inequality in 1997, although likely understated by official reports, was greater than in 1904.
Finally, a new paper by Allen and Khaustova (2017) examines Russian real wages in the long run. This paper does not address income inequality directly but allows inferences to be drawn from comparing real wages and productivity in industry. They find that real wages stagnated from the 1860s to 1913 (in St Petersburg, the capital, and Kursk, a provincial centre) or showed modest gains (in Moscow) but lagged everywhere behind productivity, suggesting a movement from wages to profits and income from wealth. After the troubled wartime and revolutionary period, the 1920s brought large real wage gains. These were short-lived, evaporating in the famine-led inflation of the early 1930s.
Novokmet and co-authors (NPZ) are the first to have tried to measure wealth and income inequality in Russia over the whole twentieth century. And, as many readers will be aware, their paper is part of a much larger collaborative project, the World Inequality Lab and the associated World Wealth and Income Database, one that aims to measure inequality in many countries over hundreds of years.
According to NPZ, the share of the top 10 per cent in pre-tax income distributed to adults in Russia was 47 per cent in 1905. The share fell to 22 per cent in 1928, increased modestly to 26 per cent by 1956, and began to fall gently back again, reached a low of 21 per cent in 1980. (The Soviet-era years observed are 1928, 1956, and then roughly every second, third, or fourth year to 1988, when annual observations begin.) By 1996 the top 10-per-cent share had returned to the 1905 level and remained in that vicinity through 2016. NPZ comment: “our benchmark estimates suggest that inequality levels in Tsarist and post-Soviet Russia are roughly comparable. Very top income shares seem if anything somewhat larger in post-Soviet Russia.”
Measured by the top 10-percent income share, Russia today appears in the World Inequality Lab database in the same inequality band as the United States and China. Income inequality is reported as greater in a few countries: Turkey, India, South Africa, and Brazil. All north and west European countries that are represented in the database are more equal than Russia. But all are smaller than Russia in population, and a larger population will always tend to show greater inequality, because unequal economic outcomes are promoted by heterogeneity of all kinds, and heterogeneity is inevitably increasing in population size.
In its time the Soviet Union, in contrast, was apparently one of the most equal countries in the world. This is particularly striking, considering the large size of the Soviet population, 288 million by 1991. Other countries in the WID dataset with top 10-percent shares of 26 per cent or below at any time from 1917 to 1991 are few, and they are also much smaller in population: Australia, Denmark, Mauritius, Italy, New Zealand, Norway, Portugal, Sweden, and Taiwan. Of all these countries, only Italy’s population had reached 57 million by 1991, and Taiwan’s 20 million.
These results are broadly consistent with the earlier research described above. They confirm that income inequality in Russia after the Soviet era was comparable to before the Revolution, if not greater; that the distribution of Soviet official incomes was markedly more equal than in most market economies at the time and today, and in Russia beforehand and today; and that, within the Soviet era, inequality followed a modest Kuznets curve, rising, then falling.
Seen in this light, Soviet institutions and policies appear distinctly pro-poor. Before we take that as settled, however, there are three issues that point the other way.
First, in the Soviet era the poor might have gained relatively, but the chief factor in this was impoverishment of the rich. What the rich lost was not transferred to the poor, or was given only temporarily before the state grabbed it back, as clearly implied by Allen and Khaustova (2017). NPZ measure inequality by shares of income distributed to adults. In the Soviet era, the share of income not distributed to adults, but retained by the state, became unusually large. As a first approximation, household consumption fell from around 80 per cent of GDP in 1913 to around 50 per cent in 1940 and through the postwar period. By implication, what the rich lost was diverted into government administration and investment and defence projects; it was not passed on to the lower income strata. If there was an initial transfer to the poor, it was confined to the 1920s, and was then cancelled in the Great Breakthrough of Stalinist collectivization and industrialization.
Second, the Soviet state did not take only from the rich. It took also from the poor, including the poorest. This applied particularly in the years from 1928 to 1956, a period for which the NPZ dataset has only gaps. While I cannot find full explanation on this point, the NPZ dataset (like most Cold-War scholarship) seems to rely on reports of the distribution of official wage earnings to capture Soviet-era inequality. Wage earnings accounted for less than one third of Soviet household incomes in 1928, just over 60 per cent in 1937, and nearly 70 per cent in 1956 (Kashin and Mikov 2004: 17, 23, 34). The largest category of households excluded from reports of wage earnings were collective farmers – the great majority of Soviet farm workers – who received an uncertain dividend, not a wage. If that is the case here, then the rural poor are left out of account. (Forced labourers are also left out. There were millions of these from the 1930s to the 1950s. But they are a small omission compared with many tens of millions of collective farmers.)
Narrative accounts of rural food shortages and periodic famines indicate that rural poverty contributed substantially to Soviet-era inequality before the 1950s (e.g. Davies and Wheatcroft 2004). After that time, the compensation of collective farmers moved gradually, but never completely, towards public-sector standards.
Finally, as NPZ acknowledge, under Soviet arrangements, persistent shortages and privileged distribution decoupled consumption inequality from income inequality. In the Soviet Union everyone had an income, but not everyone could spend it on the same terms. A privileged class of insiders – the party elite and the employees of key production and service establishments – who had access to relatively high-quality goods and services at prices fixed below the market-clearing level without waiting. Others had limited access to staple goods and services, for which they either waited in line or paid a higher, sometimes illegal price. As long as the poor had money they could not spend, or faced higher prices to spend it, it is possible and even likely that consumption was distributed more unequally than income. This contrasts with the pattern that has been found to prevail in market economies, where consumption inequality is generally less than income inequality. But comprehensive data on Soviet consumption inequality would seem far more difficult to come by than income data, so this may well remain a conjecture.
Consumption inequality was important not only for ex post evaluation of economic welfare under Soviet arrangements. It was of central importance to the political economy of the time. During the 1930s, as Paul Gregory (2004: 76-109) has noted, Stalin received regular reports of discontent and falling effort among the workers in the provinces and intervened from time to time to improve their condition. When he did so, he did not order their wages to be raised because, in a supply-constrained economy, this would only have lengthened local queues. Rather, he ordered consumer goods in short supply to be redirected to the towns and factories where dissatisfaction was rising, so that the workers could more easily spend their wages.
The existence of unofficial incomes in the Soviet era only adds complexity to the problem. We guess that unofficial incomes were substantial but of time-varying size. Anecdotes on who received them are plentiful. The Soviet central bank compiled annual estimates of their aggregate size (Kashin and Mikov 2004), but we continue to lack (and may never find) data on their distribution. Thus, it is impossible to say whether their net effect was to increase or reduce the extent of inequality of different kinds.
To summarize, the extent to which Soviet institutions favoured the poorest in society is easily overstated. The impact of the Bolshevik Revolution was to flatten the distribution of wages. On that official measure income inequality fell sharply. But non-wage earnings were likely distributed more unequally than wages. Unofficial incomes also mattered; how they mattered is unclear. Consumption inequality mattered too, and arguably mattered more than income inequality. Most likely, consumption inequality did not fall to the same extent. Whereas consumption inequality in market economies is relatively stable, it is possible that Soviet consumption inequality was volatile, spiking in particular years of crisis.
Any judgement on new work must be preliminary, but my thoughts so far are as follows. NPZ (2017) is a substantial contribution. It is not the first word on the subject, and it will not be the last word either. It turns a new page and sets a new challenge.
No comments:
Post a Comment